Design and synthesis of a novel class of dual PPARγ/δ agonists

Article abstract of DOI:10.1016/j.bmcl.2006.11.029

The design and synthesis of dual PPAR γ/δ agonist (R)-3-{2-ethyl-4-[3-(4-ethyl-2-pyridin-2-yl-phenoxy)-butoxy]-phenyl}propionic acid is described. This compound dose-dependently lowered plasma glucose in hyperglycemic male Zucker diabetic fatty (ZDF) rats

Full text of DOI:10.1016/j.bmcl.2006.11.029

Bioorganic & Medicinal Chemistry Letters 17 (2007) 1052–1055
Design and synthesis of a novel class of dual PPARc/d agonists
*
Isabel C. Gonzalez, Jason Lamar, Fatima Iradier, Yanping Xu, Leonard L. Winneroski,
Jeremy York, Nathan Yumibe, Richard Zink, Chahrzad Montrose-Rafizadeh,
Gary J. Etgen, Carol L. Broderick, Brian A. Oldham and Nathan Mantlo
Lilly Research Laboratories, A Division of Eli Lilly & Company, Lilly Corporate center, Indianapolis, IN 46285, USA
Received 2 October 2006; revised 7 November 2006; accepted 10 November 2006
Available online 15 November 2006
Abstract—The design and synthesis of dual PPAR c/d agonist (R)-3-{2-ethyl-4-[3-(4-ethyl-2-pyridin-2-yl-phenoxy)-butoxy]-phen-
yl}propionic acid is described. This compound dose-dependently lowered plasma glucose in hyperglycemic male Zucker diabetic
fatty (ZDF) rats and produced less weight gain relative to rosiglitazone at an equivalent level of glucose control.
Ó 2006 Elsevier Ltd. All rights reserved.
5
Type 2 diabetes is a metabolic disorder characterized by
hyperglycemia and/or insulin resistance. It is often asso-
ciated with obesity, hypertension, and dyslipidemia.
Current therapies for reducing plasma glucose include
the use of sulfonylureas, metformin, acarbose, and thia-
to the liver. Recently, we reported the design and syn-
thesis of a novel class of PPARc/d dual agonists, analogs
of compound 1 (Fig. 1), which had potent PPARc/d
6
activity and good selectivity versus PPARa. We showed
that a PPARc/d dual agonist with a properly controlled
c/d ratio lowered glucose and caused less weight gain
than a marketed PPARc-selective agent, rosiglitazone,
at an equivalent level of glucose control.This reduced
side effect profile was attributed to the propensity of
1
zolidinones (TZDs). Statins and fibrates are used to
2
treat dyslipidemia. The antidiabetic effects of TZDs
3
and the hyperlipidemic effects of the fibrate drugs are
due to the activation of the peroxisome proliferator-ac-
5
tivated receptors (PPARs) c and a, respectively.
PPARd to improve insulin sensitivity and stimulate fat-
ty acid oxidation. Subsequent SAR efforts were under-
7
The PPARs constitute a highly conserved set of ligand-
activated transcription factors in the nuclear hormone
receptor subfamily. Three distinct PPAR subtypes
(PPARc, PPARa, and PPARd) have been identified in
most mammalian species, each forming a functional het-
taken to explore alternative ways to address the multiple
metabolic pathways of 1, namely reduction of the ben-
zoyl group and oxidations of the ethyl group and the
dihydrocinnamate. Herein, we report the design, synthe-
sis and preclinical biological evaluation of 3-{4-[3-(2-ar-
yl-phenoxy)butoxy]-phenyl}propionic acids as novel
PPAR c/d dual agonists.
erodimer complex with the 9-cis retinoic acid receptor
(
RXR). A potent selective PPARd agonist, {2-methyl-
-[4-methyl-2-(4-trifluoromethylphenyl)-thiazol-5-yl-
4
methylsulfanyl]-phenoxy}-acetic acid (GW501516), has
been reported to increase serum high-density lipoprotein
cholesterol while lowering the level of low-density lipo-
protein cholesterol, fasting triglicerides, and fasting
The chemistry used to prepare these analogs is illustrat-
ed in Scheme 1. Treatment of (S)-butane-1,3-diol (2)
4
insulin in a dose-dependent manner. The cholesterol
mediating effects were attributed to the ability of
PPARd to up-regulate the expression of the ABC-A1 re-
verse cholesterol transporter, which promotes apolipo-
protein A1-specific cholesterol transport of cholesterol
Keywords: PPAR; Diabetes; Dyslipidemia.
*
Corresponding author. Tel.: +1 317 277 9192; fax: +1 317 276
545; e-mail: gonzalez_isabel_c@lilly.com
6
Figure 1. Compound 1.
0
960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.029

I. C. Gonzalez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1052–1055
1053
First, we focused our efforts on the replacement of the
benzoyl group to address the reduction of the carbonyl
group we observed with human and rat liver slices.
Removal of the carbonyl group and direct attachment
of the two phenyl rings in the tailpiece gave 7 with the
desired in vitro PPARa/d selectivity but with significant-
ly decreased PPARc potency (Table 1).
3
Replacement of the benzoyl substituent R with thio-
phene (8, 9), furan (10), thiazole (14, 15) or oxazole
(
16 and 17) rings was detrimental to PPAR c and d affin-
ities. Interestingly, 2-pyridyl (11) and 3-pyridyl (12)
replacements led to compounds with good selectivity
over PPARa and potent PPAR c/d affinity and function-
4
al activity. Replacement of the ethyl substituent R of 11
with Cl (18) attenuated the PPARc and PPARd transac-
tivation EC ’s, but with CF (19) the desired in vitro
profile was maintained. To protect the dihydrocinna-
5
0
3
Scheme 1. Reagents and conditions: (a) p-TosCl, NEt
CH Cl , 72%; (b) Cs CO
= Et); (c) MsCl, NEt , CH
5 °C; (e) R OH, NaOH aq, 7 (55%), 8 (63%), 9 (53%), 10 (65%), 11
3
, (Bu
= Me), 82%
CO , DMF,
3
Sn)
2
O,
mate headpiece from b-oxidation, we replaced the meth-
yl group R with a sterically more demanding ethyl
2
2
2
3
, DMF, 55 °C, 64% (R
Cl
, 0 °C, 98%; (d) Cs
1
1
(
R
1
3
2
2
2
3
2
6
5
group (20 and 21). Compounds 20 and 21, with potent
(
63%), 12 (59%), 13 (26%), 14 (71%), 15 (71%), 16 (53%), 17 (35%), 18
54%), 19 (46%), 20 (39%), 21 (59%) both steps.
PPARc/d dual agonist activity, were tested in db/db
mice for 7 days at 30 mg/kg oral gavage dosing. Com-
pound 20 showed 89 ± 4% glucose normalization, com-
parable to 21 that exhibited 81 ± 19% glucose
normalization. The concentration in blood of com-
pound 20 after 1 h of the last dose was 3.2 ± 1.3 lg/
mL, comparable to 21 that exhibited a concentration
of 4.2 ± 1.7 lg/mL.
(
with p-TosCl, NEt , and (Bu Sn) O provided 3. Nucle-
3
3
2
ophilic substitution of 3 with phenols 4 followed by
mesylation provided intermediates 5. Treatment of
mesylates 5 with phenols 6 and subsequent hydrolyses
8
afforded carboxylic acids 7–21.
a
Table 1. Binding IC50 and transactivation EC50 data on human PPAR receptor subtypes
b,c
d
3
1
R
4
R
Compound
R
hPPARa
hPPARa
hPPARc
hPPARc
hPPARd
hPPARd
IC50 (nM)
EC50 (nM)
IC50 (nM)
EC50 (nM)
IC50 (nM)
EC50 (nM)
1
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
Benzoyl
Phenyl
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Cl
CF
Et
CF
4257
nb
877
2880
na
5
215
568
1090
940
46
4
836
1670
1990
2300
182
140
576
837
2270
1980
696
304
67
3
5
1
6
3-Thiophenyl
2-Thiophenyl
2-Furanyl
2-Pyridyl
3-Pyridyl
4-Pyridyl
4-Thiazolyl
2-Thiazolyl
4-Oxazolyl
2-Oxazolyl
2-Pyridyl
2-Pyridyl
2-Pyridyl
2-Pyridyl
5950
9110
4500
8290
nb
8
42
41
399
9
2690
2800
2870
2900
na
9
1
14
3
1
1
1
1
1
1
1
1
1
2
2
60
5
15
5
nb
520
346
1530
979
483
88
5
4100
nb
2680
na
3
22
25
274
50
24
6
12
13
5
4990
10,400
nb
2810
2890
2980
2340
2850
2630
na
2
3
3
5880
7350
6090
nb
20
35
4
167
111
308
3
7
Et
16
67
3
nb
6
Rosiglitazone
na
a
Concentration of test compound required to displace 50% of tritiated ligand, PPARa/PPARd agonist 2-(4-{2-[3-(2,4-difluoro-phenyl)-1-heptyl-
ureido]-ethyl}-phenoxy)-2-methyl-butyric acid, and PPARc agonist, 2-methyl-2-(4-{3-[propyl-(5-pyridin-2-yl-thiophene-2-sulfonyl)-amino]-pro-
pyl}-phenoxy)-propionic acid, nb = no binding.
b
c
Concentration of test compound which produced 50% of the maximal reported activity, na = efficacy relative to control was less than 20% at 10 lM.
Gal4-hPPARa and Gal4-hPPARd were used to eliminate interference by endogenous PPARc receptors in CV-1 cells.
d
9
Minimum significant ratio (MSR), hPPRa IC50 MSR = 1.53, hPPARa EC50 MSR = 2.48, hPPRc IC50 MSR = 2.6, hPPARc EC50 MSR = 1.79,
hPPRd IC50 MSR = 1.26, hPPARd EC50 MSR = 2.89.

1054
I. C. Gonzalez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1052–1055
1
0
Compound 20 was selected for further studies and
exhibited an improved pharmacokinetic profile versus
compound 1. In Sprague–Dawley rats 20 was 71% bio-
available with a half life of 3.0 h and AUC of
7
5.5 ± 5.7 lg h/mL when dosed orally at 12 mg/kg as a
suspension in sodium carboxymethyl cellulose (NaC-
MC)/sodium lauryl sulfate (SLS)/povidone. The clear-
ance of 20 was 1.9 ± 0.1 mL/min/kg when dosed
intravenous at 1 mg/kg in 6% solutol/EtOH (1:1 w/v)
in saline. In contrast, 1 was 27% bioavailable with a half
life of 1.4 h and AUC of 3.7 ± 0.1 lg h/mL when dosed
orally at 10 mg/kg. The clearance of 1 was 11 ± 2 mL/
min/kg when dosed intravenous at 1 mg/kg. These
favorable features made 20 a good candidate for further
1
1
evaluation using in vivo models.
Compound 20 was tested over a range of oral dose levels
in male ZDF rats. As shown in Figure 2, 20 dose-depen-
dently lowered plasma glucose levels in ZDF rats.
Figure 3. Dose–response for effects of compound 20 on body weight in
ZDF rats after 7 days of oral gavage dosing.
At the oral dose of 1 mg/kg 20 exhibited 87 ± 10% glu-
cose normalization, comparable to rosiglitazone with
9
1
± 1 g above control, significantly less than the
9 ± 2 g increase noted with rosiglitazone (P < 0.05)
8
1
9 ± 14% glucose normalization at the same dose of
1
2
(Fig. 3). At the 1 mg/kg oral dose, 20 lowered plasma
triglycerides by 53% and plasma free fatty acids by
mg/kg.
Compound 20 increased body weight
2
3% compared to control animals.
A 2-week female ZDF rat study was conducted with 20
to determine if the effects observed in glucose normaliza-
tion were related to an improvement in insulin sensitiv-
ity. Compound 20 dosed orally at 2 mg/kg/d lowered the
glucose response to an oral glucose challenge. Further-
more, both fasting insulin levels and the insulin response
to the glucose challenge were significantly reduced with
20. These effects were comparable to those observed
with rosiglitazone at 1 mg/kg/d (Fig. 4).
In summary, we identified compound 20 as a PPARc/d
dual agonist, with high affinity to hPPARc and
hPPARd, and potent agonistic activity in cell-based
transactivation assay. In male ZDF rats, compound 20
exhibited potent glucose lowering activity with less in-
crease in body weight above control levels than rosiglit-
azone at the same dose of 1 mg/kg. In female ZDF rats,
compound 20 significantly lowered both glucose and
insulin responses to a glucose challenge. These studies
corroborate our initial hypothesis that a PPARc/d dual
Figure 2. Dose–response for effects of compound 20 on glucose
normalization in ZDF rats after 7 days of oral gavage dosing.
Figure 4. Glucose and insulin responses to a glucose challenge (2.5 g/kg/BW) in female ZDF rats dosed by oral gavage for 14 days with compound 20
or rosiglitazone.

I. C. Gonzalez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1052–1055
1055
agonist with a properly controlled c/d ratio can attenu-
ate the weight gain side effect associated with marketed
TZDs. Interestingly, a recent report indicated that
GW501516, a potent and selective PPARd agonist, did
not influence the weight in db/db mice. The evaluation
of the in vivo side effect profile with respect to PPAR c
and d potency ratio will be reported in due course.
Auwerx, J.; Yanagisawa, M.; Kodama, T.; Sakai, J. Proc.
Natl. Acad. Sci. U.S.A. 2003, 100, 15924.
. Gonzalez Valcarcel, I. C.; Mantlo, N. b.; Shi, Q.; Wang,
M.; Winneroski, L. L.; Xu, Y.; York, J. S.
WO2005019151.
. Minimum significant ratio (MSR) = Smallest ratio of
potencies expected to be statistically significant Eastwood,
B. J.; Farmen, M. W.; Iversen, P. W.; Craft, T. J.;
Smallwood, J. K.; Garbison, K. E.; Delapp, N. W.; Smith,
G. F. J. Biomol. Screen 2006, 11, 253.
8
9
1
3
References and notes
28 33 4
10. Compound 20: HRMS calculated for C H NO
1
4
47.5717, found 447.2409; H NMR (CD
d 8.58 (1H, d, J = 4.3 Hz), 7.79 (2H, m), 7.39 (1H, d,
J = 1.8 Hz), 7.34 (1H, m), 7.22 (1H, dd, J = 8.3 Hz,
= 2.8 Hz), 7.07 (2H, m), 6.67 (1H, d, J = 2.4 Hz), 6.61
(1H, dd, J = 8.4 Hz, J = 2.3 Hz), 4.68 (1H, m), 3.98 (2H,
3
OD, 400 MHz)
1
. (a) Chipkin, S. R. Am. J. Med. 2005, 118, 4S; (b) Riddle,
M. C. Endocrinol. Metab. Clin. North Am. 2005, 3, 77.
. (a) Hauner, H. Diabetes Metab. Res. Rev. 2002, 18, S10;
(b) Smith, U. Int. J. Clin. Pract. Suppl. 2001, 13; (c) Bailey,
C. J. Br. J. Diabetes Vas. Dis. 2001, 1, 7.
1
2
J
2
1
2
t, J = 6.6 Hz), 2.89 (2H, t, J = 8.2 Hz), 2.65 (4H, m), 2.49
(2H, J = 8.2 Hz), 2.03 (2H, m), 1.19–1.32 (9H, m).
3
4
. Issemann, I.; Green, S. Nature 1990, 347, 645.
. Oliver, W. R., Jr.; Shenk, J. L.; Snaith, M. R.; Russell, C.
S.; Plunket, K. D.; Bodkin, N. L.; Lewis, M. C.; Winegar,
D. A.; Sznaidman, M. L.; Lambert, M. H.; Xu, E.;
Sternbach, D. D.; Kliewer, S. A.; Hansen, B. C.; Willson,
T. M. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 5306.
. Lee, C.-H.; Chawla, A.; Urbiztondo, N.; Liao, D.;
Boisvert, W. A.; Evans, R. M. Science 2003, 302, 453.
. Xu, Y.; Etgen, G. J.; Broderick, C. L.; Canada, E.;
Gonzalez, I.; Lamar, J.; Montrose-Rafizadeh, C.; Oldham,
B. A.; Xie, C.; Shi, Q.; Winneroski, L. L.; York, J.; Yumibe,
N.; Zink, R.; Mantlo, N. J. Med. Chem. 2006, 49, 5649.
. (a) Wang, Y.-X.; Lee, C.-H.; Tiep, S.; Yu, R. T.; Ham, J.;
Kang, H.; Evans, R. M. Cell 2003, 113, 159; (b) Tanaka,
T.; Yamamoto, J.; Iwasaki, S.; Asaba, H.; Hamura, H.;
Ikeda, Y.; Watanabe, M.; Magoori, K.; Ioka, R. X.;
Tachibana, K.; Watanabe, Y.; Uchiyama, Y.; Sumi, K.;
Iguchi, H.; Ito, S.; Doi, T.; Hamakubo, T.; Naito, M.;
11. Male and female ZDF rat in vivo models: ZDF male rats
were divided into groups: control, different dose groups
(0.3, 1, 3, and 10 mg/kg/d) with compound 20, and one
dose group (1 mg/kg/d) with rosiglitazone. Vehicle 1% wt/
v CMC, 0.25% Tween 80 was used in all in vivo studies.
ZDF rats (8 weeks of age) were dosed once daily by oral
gavage in the morning for 7 or 14 days. The ZDF rats were
weighted before dosing and plasma glucose levels were
determined one hour after dosing.
5
6
12. Historical data for rosiglitazone tested in a dose–response
ZDF male rat study for 7 days show that a 1 mg/kg/d dose
is the dose that consistently produced a >80% glucose
7
normalization to approximate the ED80
.
13. Lee, C.-H.; Olson, P.; Hevener, A.; Mehl, I.; Chong, L.-
W.; Olefsky, J. M.; Gonzalez, F. J.; Jungyeob Ham, J.;
Kang, H.; Peters, J. M.; Evans, R. M. Proc. Natl. Acad.
Sci. U.S.A. 2006, 103, 3444.