Design, synthesis and structure-activity relationships of azole acids as novel, potent dual PPAR α/γ agonists

Article abstract of DOI:10.1016/j.bmcl.2009.01.030

The design, synthesis and structure-activity relationships of a novel series of N-phenyl-substituted pyrrole, 1,2-pyrazole and 1,2,3-triazole acid analogs as PPAR ligands are outlined. The triazole acid analogs 3f and 4f were identified as potent dual PPARα/γ agonists both in binding and functional assays in vitro. The 3-oxybenzyl triazole acetic acid analog 3f showed excellent glucose and triglyceride lowering in diabetic db/db mice.

Full text of DOI:10.1016/j.bmcl.2009.01.030

Bioorganic & Medicinal Chemistry Letters 19 (2009) 1451–1456
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis and structure–activity relationships of azole acids
as novel, potent dual PPAR a/c agonists
Hao Zhang a, Denis E. Ryono a, Pratik Devasthale a, Wei Wang a, Kevin O’Malley c, Dennis Farrelly b,
Liqun Gu b, Thomas Harrity b, Michael Cap b, Cuixia Chu b, Kenneth Locke c, Litao Zhang c, Jonathan Lippy c,
Lori Kunselman b, Nathan Morgan b, Neil Flynn b, Lisa Moore b, Vinayak Hosagrahara d, Lisa Zhang d,
Pathanjali Kadiyala e, Carrie Xu e, Arthur M. Doweyko g, Aneka Bell h, Chiehying Chang f, Jodi Muckelbauer f,
Robert Zahler a, Narayanan Hariharan b, Peter T. W. Cheng a,
*
a Metabolic Diseases Chemistry, Bristol-Myers Squibb Research and Development, Princeton, NJ 08543-5400, USA
b Metabolic Diseases Biology, Bristol-Myers Squibb Research and Development, Princeton, NJ 08543-5400, USA
c Lead Evaluation, Bristol-Myers Squibb Research and Development, Princeton, NJ 08543-5400, USA
d Metabolism and Pharmacokinetics, Bristol-Myers Squibb Research and Development, Princeton, NJ 08543-5400, USA
e Bio-Analytical Research, Bristol-Myers Squibb Research and Development, Princeton, NJ 08543-5400, USA
f Macromolecular Crystallography, Bristol-Myers Squibb Research and Development, Princeton, NJ 08543-4000, USA
g Computer-Assisted Drug Design, Bristol-Myers Squibb Research and Development, Princeton, NJ 08543-4000, USA
h Gene Expression and Protein Biochemistry, Bristol-Myers Squibb Research and Development, Princeton, NJ 08543-4000, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The design, synthesis and structure–activity relationships of a novel series of N-phenyl-substituted pyr-
Received 10 November 2008
Revised 8 January 2009
Accepted 12 January 2009
Available online 15 January 2009
role, 1,2-pyrazole and 1,2,3-triazole acid analogs as PPAR ligands are outlined. The triazole acid analogs 3f
and 4f were identified as potent dual PPARa/c agonists both in binding and functional assays in vitro. The
3-oxybenzyl triazole acetic acid analog 3f showed excellent glucose and triglyceride lowering in diabetic
db/db mice.
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Dual PPAR
Azole
a/c agonists
Pyrrole
Pyrazole
Triazole
Conformational constraint
Peroxisome proliferator-activated receptors (PPARs) are a group
of nuclear hormone receptors which function as transcription fac-
tors in the regulation of genes involved in glucose and lipid fatty
acid metabolism and vessel wall function.1 Three PPAR subtypes
agonist activities in a single compound would result in synergistic
improvements in insulin sensitivity and normalization of glucose
metabolism as well as amelioration of the dyslipidemia associated
with type 2 diabetes.
have been identified: PPAR
a,
c
and d. PPAR
a
is predominantly ex-
We have previously reported on the discovery and exploration
pressed in catabolically active tissues such as liver, heart, kidney,
and muscle. It is involved in the uptake and oxidation of fatty acids
of the SAR of the novel oxybenzylglycine series of dual PPARa/c
agonists, which culminated in the discovery of the clinical candi-
as well as in lipoprotein metabolism.2 The clinically used PPAR
a
date Muraglitazar 5.7a,7b Compound 5 (Fig. 1) shows potent func-
agonists are the fibrate class of drugs (including fenofibrate3 and
gemfibrozil4), which elevate HDL cholesterol levels and lower tri-
tional activity in vitro at both human PPARa and PPARc and
excellent efficacy in animal models of type 2 diabetes and the asso-
glyceride and LDL cholesterol levels. PPARc is mainly expressed
ciated dyslipidemia.7c–e As part of our ongoing efforts to discover
in adipose tissue and regulates insulin sensitivity, glucose and fatty
additional novel, potent, highly efficacious and safe dual PPARa/c
acid utilization as well as adipocyte differentiation. The clinically
agonists with differentiated profiles from muraglitazar, we decided
to further explore the SAR of the oxybenzylglycine series by exam-
ining the effects of constraining the carbamate N-acid moiety. One
of the possible modes of constraint is shown in Figure 1, where the
used PPAR
of anti-diabetic drugs such as rosiglitazone5 and pioglitazone.6 It
has been hypothesized that the combination of PPAR and PPAR
c agonists comprise the thiazolidinedione (TZD) class
c
a
carbamate oxygen is cyclized onto the glycine a-carbon to form a
ring. This could be effected through a structure such as an azole,
which could be envisioned to be various regioisomers of pyrroles
* Corresponding author. Tel.: +1 609 818 4952; fax: +1 609 818 3460.
E-mail address: peter.t.cheng@bms.com (P.T.W. Cheng).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.01.030

1452
H. Zhang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1451–1456
CO2H
CO2H
O
N
n
Ph
Conformational
O
N
N
Y
m
Ph
O
X
R
N
O
O
O
Constraint
3- & 4- oxyphenyl or oxybenzyl analogs
5 (Muraglitazar)
γEC50 = 0.035 μM (107%)
αEC50 = 1.41 μM (139%)
OMe
CO2H
CO2H
N2
CO2H
CO2H
n
N
N
R
N3
m
O
O
O
O
N1
R
N1
R
N2
N
R
2a: R=N2-Ph; 3-
2b: R=N1-Ph; 3-
2d: R=N2-Ph; 4-
2e: R=N1-Ph; 4-
3a: R=N2-Ph; m=1; n=0; 3-
3b: R=N1-Ph; m=1; n=0; 3-
3c: R=N3-Ph; m=1; n=0; 3-
3d: R=N2-Ph; m=0; n=0; 3-
3e: R=N2-Ph; m=0; n=1; 3-
3f: R=N2-Ph; m=1; n=1; 3-
4a: R=N2-Ph; m=1; n=0; 4-
4b: R=N1-Ph; m=1; n=0; 4-
4c: R=N3-Ph; m=1; n=0; 4-
4d: R=N2-Ph; m=0; n=0; 4-
4e: R=N2-Ph; m=0; n=1; 4-
4f: R=N2-Ph; m=1; n=1; 4-
1a: R=CH3; 3-
1b: R=Ph; 3-
1c: R=CH3; 4-
1d: R=Ph; 4-
2c: R=Ph; 3-
Figure 1. Design of azole acid analogs from oxybenzylglycines.
O
1a–1d, pyrazoles 2a–2e, and triazoles 3a–3f and 4a–4f. We report
here a systematic SAR study of this azole acid series, which re-
O
O
CN
b
CO2H
CN
a
Cl
sulted in the discovery of analogs with distinct PPAR
functional activities, of which a number proved to be potent dual
PPAR and PPAR agonists in binding and functional assays
in vitro.
a and PPARc
N
NH
Ph
9
10
11
a
c
CO2CH3
CO2H
O
O
R
An exemplary route to N-substituted pyrrole-3-carboxylic acids
d
c
1a–1d is shown in Scheme 1. Wittig reaction of aldehyde 67b gave
N
N
N
N
Ph
Ph
the a, b-unsaturated ester 7, which was reacted with tosylmethyl
12
isocyanide to provide the pyrrole ester 8.8 Pyrrole 8 underwent a
Cu(I)-mediated coupling reaction9 with an appropriate arylboronic
acid to provide the corresponding N-aryl pyrrole ester, which was
subsequently deprotected to furnish the N-substituted pyrrole
acids 1a–1d.
2a: 3-
2d: 4-
O
R =
Ph
N
Scheme 2. The synthesis of 2a and 2d: (a) i—NCCH2CO2CH3, CH3ONa, CH3OH, 52%;
ii—NaOH, 95%; (b) PhNþ2 Cl-, 11%; (c) i—methyl acrylate, NaH, rt, (C2H5)3 N, 33%; (d)
i—BBr3, CH3OH, 25–30%; ii—2-(5-methyl-2-phenyloxazol-4-yl)ethyl methanesulfo-
nate, K2CO3, CH3CN, 90 °C, 12 h, 90%; iii—NaOH, THF, H2O, 90%.
The synthesis of 5-(3-oxybenzyl)pyrazole-4-carboxylic acid 2a
and 5-(4-oxybenzyl)pyrazole-4-carboxylic acid 2d is described in
Scheme 2. Methyl cyanoacetate was alkylated with benzyl chloride
9, followed by ester hydrolysis to provide cyanoacetic acid 10.10
Treatment of 10 with a phenyl diazonium salt (generated in situ)
provided the corresponding cyano-hydrazone 11. Reaction of 11
with methyl acrylate under basic conditions11 generated the key
pyrazole ester intermediate 12. Finally, a three-step sequence
involving: (1) demethylation of anisole 12, (2) alkylation of phenol
with 2-[5-methyl-2-phenyloxazol-4-yl] ethyl mesylate and (3)
deprotection of the carboxylic acid, furnished 2a and 2d.
in situ) furnished the chlorohydrazone 16.13 Base-mediated ther-
mal intramolecular cycloaddition of 16 then furnished the pyra-
zole-lactone 17. Concomitant ring-opening/deoxygenation of 17
could be achieved with TMSI14 to furnish the pyrazole ester 18.
The same three-step sequence: (a) removal of the phenolic protect-
A synthesis of the regioisomeric carboxypyrazole 2c is shown in
Scheme 3. Treatment of aldehyde 13 with ethynylmagnesium bro-
mide furnished the acetylenic alcohol adduct 14. Thermolysis of 14
with ketene dimer12 provided the corresponding acetoacetate es-
O
O
O
OH
Cl
O
CHO
O
O
c
b
a
ter, which was chlorinated to provide the
15. Heating of 15 with a phenyl diazonium salt (generated
a-chloro-b-ketoester
13
14
15
Ph
O
O
O
N
CO2CH3
N
O
O
O
a
H
b
d
e
CHO
CO2-tBu
Cl
N
N
Ph
O
N
Ph
O
O
N
N
Ph
7
6
17
16
18
O
N
CO2H
N
N
Ph
CO2H
O
Ph
f
R =
O
CO2-tBu
R
Ph
O
c
N
N
R
N
2c
H
8
1a ~ 1d(R = Me or Ph)
Scheme 3. The synthesis of 2c: (a) HC„CMgBr, 79%; (b) i—ketene dimer, Et3N; ii—
Scheme 1. The synthesis of 1a–1d: (a) Ph3PCHCO2tBu, toluene, reflux, 80%; (b) 4-
Me-C6H4-SO2CH2NC, NaH, DMSO, 70%; (c) i—RB(OH)2, Cu(I) salt, base or RX, base;
ii—TFA 50-70%.
SO2Cl2, 40%; (c) PhNþCl-, 94%; (d) Et3N, toluene, reflux, 69%; (e) TMSCI, NaI, CH3CN,
reflux, 81%; (f) i—BBr3, CH3OH; ii—2-(5-methyl-2-phenyl-oxazol-4-yl)ethyl meth-
anesulfonate, K2CO3, CH3CN, 90 °C, 12 h; iii—NaOH, 70%.
2

H. Zhang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1451–1456
1453
N(Me)2
CO2CH3
N2
O
O
NHPh
R
O
R
NHPh
O
b
O
a
b
COCl
a
CO2CH3
m
O
O
O
O
O
O
19
20
21
24(m=1)
28
CO2CH3
N
CO2H
O
O
d
O
c
R
CO2H
CONHPh
O
O
R =
Ph
N
N
R
R
N
N
Ph
22
O
N
Ph
2b
c
N
N
N
R =
Ph
N
N
N
Ph
Ph
29
Scheme 4. The synthesis of 2b: (a) i—Meldrum’s acid, pyridine, 0 °C; ii—CH3OH,
reflux, 80%; (b) (CH3)2NCH(OCH3)2, 0 °C, 32%; (c) PhNHNH2, 100 °C, 77%; (d) i—
BBr3/MeOH; ii—2-(5-methyl-2-phenyloxazol-4-yl)ethyl methanesulfonate, K2CO3,
CH3CN, 90 °C, 12 h; iii—NaOH, 75%.
3b: 3-
4b: 4-
Scheme 6. The synthesis of 3b and 4b: (a) TsN3, Et3N, rt, 2.5 h, 77%; (b) PhNH2,
TiCl4, 80 °C, 2 h, 90%; (c) i—BBr3, 85–90%; ii—2-(5-methyl-2-phenyloxazol-4-yl)
ethylmethanesulfonate, K2CO3, CH3CN, 90 °C, 12 h, 90%; iii—KOH, EtOH, 90 °C, 24 h,
30–70%.
ing group of 18, (b) alkylation of the resulting phenol with 2-(5-
methyl-2-phenyloxazol-4-yl)ethyl mesylate and (c) ester hydroly-
sis, furnished the N-substituted pyrazole-acid 2c.
Scheme 4 illustrates the synthesis of the regioisomeric 1,3-alk-
oxybenzyl 1-phenyl-3-carboxy-pyrazole analog 2b. Reaction of
phenylacetyl chloride 19 with Meldrum’s acid under basic condi-
tions, followed by methanolysis, furnished the b-ketoester 20.
Treatment of 20 with dimethylformamide dimethyl acetal15 gave
oxime-hydrazone 25 followed immediately by reduction (Zn/HCl)
furnished the desired 1,2,3-triazole 26.18 The similar three-step se-
quence: a) demethylation of 26, (b) alkylation of phenol with 2-(5-
methyl-2-phenyloxazol-4-yl) ethyl mesylate and (c) deprotection
of the triazole-amide 27 furnished 5-oxybenzyl-2-phenyl-2H-
1,2,3-triazole-4-carboxylic acids 3a, 3d, 4a, and 4d.19
the
a-enamino-b-keto-ester 21. The reaction of 21 with phen-
The synthesis of the ‘‘N-1” aryl triazole isomer 3b and 4b is
shown in Scheme 6. Treatment of b-keto anilide 24 with p-toluene-
ylhydrazine followed by intramolecular cyclization, provided the
N-phenyl-pyrazole ester 22. The same three-step sequence: (a)
demethylation of anisole 22, (b) alkylation of phenol with 2-(5-
methyl-2-phenyloxazol-4-yl) ethyl mesylate and (c) deprotection
of the carboxylic acid, furnished the 5-benzyl-1-phenyl-1H-pyra-
zole-4-carboxylic acid 2b.
sulfonyl azide20 furnished the corresponding b-keto
a-diazo-ani-
lide 28. Lewis acid-mediated reaction of 28 with aniline provided
the corresponding 5-alkoxybenzyl-1-phenyl-1H-1,2,3-triazole-4-
carboxylic acid 29.21 A three-step sequence: (a) BBr3-mediated
demethylation of 29; (b) alkylation of the resultant phenol-triazole
with 2-(5-methyl-2-phenyloxazol-4-yl)ethyl mesylate and, (c) ba-
sic hydrolysis of the anilide, provided the 5-oxybenzyl-1-phenyl-
1H-1,2,3-triazole-4-carboxylic acid 3b or 4b.
The synthesis of the 1-phenyl-1H-1,2,3-triazole-5-carboxylic
acid 3c and 4c was accomplished using the route described in
Scheme 7. Ethyl propiolate was added to aldehyde 30 under basic
conditions22 to furnish the corresponding acetylenic alcohol ad-
duct 31. Deoxygenation of alcohol 31 (Et3SiH/BF3-OEt2)23 provided
the acetylenic ester 32. Dipolar cycloaddition of the acetylenic es-
ter 32 with phenyl azide under thermal conditions24 furnished,
after deprotection of the carboxylic acid, a ꢀ1:2 mixture of the de-
sired 1-phenyl-1H-1,2,3-triazole-5-carboxylic acids 3c and 4c as
well as the regioisomeric 1-phenyl-1H-1,2,3-triazole-4-carboxylic
acid 3b and 4b. The regioisomers, 3c/3b and 4c/4b, were separated
by reverse phase preparative HPLC.
The synthesis of the 5-oxybenzyl-2-phenyl-2H-1,2,3-triazole-4-
carboxylic acids 3a, 3d and 4a, 4d is described in Scheme 5; only
the desired ‘‘N-2”-phenyl regioisomer should be obtainable by this
route. Reaction of a suitably protected oxybenzoic or oxyphenyl-
acetic acid chloride 23 with Meldrum’s acid in the presence of base
provided the corresponding crude acylation product which was
immediately reacted with aniline to give the b-keto anilide 24.16
The keto-amide 24 was reacted with nitrous acid (generated
in situ from acid/sodium nitrite), then with acid to furnish the cor-
responding
condensed with phenylhydrazine to provide the corresponding b-
a a-oxime-ketone was then
-oxime b-keto amide.17 The
hydrazone-
a
-oxime-ketone 25. CuSO4-mediated cyclization of
O
O
a
NHPh
b
COCl
m
m
O
O
The homologated ‘‘N-2”-triazole acetic acids 3f and 4f were syn-
thesized as shown in Scheme 8. The 4-carboxy-2-phenyl-2H-1,2,3-
triazole 3a or 4a was reacted with oxalyl chloride. The resultant
acid chloride was converted to the homologated methyl ester of
23
24
O
NHPh
CONHPh
O
O
c
d
OH
N
m
N
m
N
N
N
Ph
NH
Ph
26
25
27
OH
O
O
O
b
a
R
R
R
CHO
CO2H
CONHPh
CO2Et
CO2Et
O
O
R
e
R
31
30
32
m
N
m
N
N
N
Ph
N
N
Ph
CO2H
CO2H
O
O
R
c
R
O
N
Ph
N
N
N
+
R =
Ph
O
3a: m=1, 3-; 4a: m=1, 4-
3d: m=0, 3-; 4d: m=0, 4-
N
N
N
R =
Ph
Ph
N
3c: 3-
4c: 4-
3b: 3-
4b: 4-
Scheme 5. The synthesis of 3a, 3d, 4a and 4d: (a) i—Meldrum’s acid, pyridine, 0 °C,
2 h, rt, 2 h; ii—aniline, reflux, 3 h, 80%; (b) i—NaNO2, H2SO4, 0 °C, 30 min; ii—
PhNHNH2, EtOH, reflux, 2 h, 56%; (c) i—CuSO4, pyridine; ii—Zn, HCl, 1 h, 89%; (d) i—
BBr3; ii—2-(5-methyl-2-phenyloxazol-4-yl)ethyl methanesulfonate, K2CO3, MeCN,
90 °C, 12 h; iii—KOH, EtOH, 90 °C, 24 h, 70%.
Scheme 7. The synthesis of regioisomeric 1-Ph-substituted-5-carboxy triazoles 3c/
4c and 1-Ph-substituted-4-carboxy triazoles 3b/4b: (a) ethyl propiolate, nBuLi/THF,
ꢁ78 °C, 1 h; (b) (C2H5)3SiH/BF3-Et2O, 0 °C–rt, 2 h, 50% from 30; (c) i—PhN3, 130 °C,
18 h; ii—1 M LiOH/THF, rt, 18 h, 13% for 3c and 4c, 26% for 3b and 4b.

1454
H. Zhang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1451–1456
Table 3
In vitro PPAR
CO2H
N
COCHN2
O
O
a and c activities of 3- and 4-oxybenzyl triazole acid analogs
R
a
R
N
CO2H
O
N
N
N
N
N3
Ph
Ph
N
O
N1
3a & 4a
N2
33
R
CO2H
Compound Oxybenzyl
substitution
R
c
IC50
M)
c
EC50
(
l
M)
a
IC50
M)
a
EC50
(lM)
O
R
b
(l
IA (%)
(
l
IA (%)
O
N
N
Ph
R =
N2-Ph
N1-Ph
N3-Ph
N2-Ph
0.069
0.75
1.48
3.79 (70%) 0.069 1.90 (30%)
N
N
3ª
3-
3-
3-
4-
4-
4-
Ph
3b
3c
4a
4b
4c
7.90 (19%) >25
9.27 (49%) 7.29
9.27 (87%) 2.77
13.6 (21%)
5.80 (21%) >7.5
>15
>15
6.15 (52%)
>15
>15
3f: 3-
4f: 4-
0.712
N1-Ph 10
>25
N3-Ph
0.88
Scheme 8. The synthesis of homologated 3f and 4f: (a) i—(COCl)2; ii—CH2N2, rt, 2 h,
67%; (b) i—PhCO2Ag, Et3N, rt, 1 h; ii—1 M LiOH/THF, rt, 18 h, 40%.
3f or 4f via a silver-mediated Wolff rearrangement of the corre-
should be noted that all five analogs in these pyrazole-acid series
show only partial functional activity at PPAR (in spite of good
binding affinity at PPAR for 2a and 2c).
The next azole to be explored as a carbamate bioisostere was
the 1,2,3-triazole. We systematically examined the effect of all
three possible (N1, N2 and N3) regioisomeric phenyl-substituted
triazole acids in the context of both the 3- and 4-oxybenzyl series.
As in the pyrrole and pyrazole series, the different N-phenyl substi-
sponding
The PPAR
a
-diazo-ketone 33.25
a
a
and PPARc
in vitro activities (binding affinity26 and
a
transactivation activity27) of the different series of azole acid ana-
logs are shown in Tables 1–4. From the initial SAR with pyrrole acid
analogs 1a–1d (Table 1), we were able to determine that in both
the 3- and 4-oxybenzyl series, pyrrole showed some promise as a
bioisostere for the carbamate moiety of muraglitazar 5. Analogs
with an N-phenyl substituent (1b and 1d) were more potent ago-
tution patterns on the triazole ring had a major effect on the PPAR
and PPAR agonist activity of these analogs. In both the 3- and 4-
oxybenzyl series, for instance, analogs with N1 or N3-phenyl sub-
stituents (e.g., 3b, 3c, 4b and 4c; Table 3), did not show any PPAR
agonist activity. Analogs 3a and 4a, with an N2-phenyl group, dis-
play relatively more balanced and potent dual PPAR and PPAR
agonist activities. Overall, however, these series of triazole acids
displayed relatively poor PPAR functional activities.
a
nists at PPARc than the corresponding analogs (1a and 1c) contain-
ing an N-methyl group. In addition, among these pyrrole-acid
analogs, the 3-oxybenzyl substituted pyrrole analog 1b, which is
c
a
a relatively selective PPAR
PPAR ), shows more potent PPAR
4-oxybenzyl analog 1d.
c
agonist (with only partial activity at
a
c
activity than the corresponding
a
c
In the pyrazole series, the regiochemistry of the N-phenyl sub-
stituent clearly plays an important role in the functional activity. A
phenyl group at N2 appears to be optimal for PPAR activity, as ana-
a/c
Based on the results of this initial SAR survey of the oxybenzyl
triazole acids (3a–c, 4a–c, m = 1; n = 0; Table 3), we then systemat-
ically examined the effects of inserting a methylene linker between
the oxybenzyl group, the triazole and the acid. The results of this
study are shown in Table 4 (using the optimized N2-phenyl substi-
tuent). The relatively constrained 3-oxyphenyl triazole acid 3d
(m = n = 0) displayed weak agonist activities at PPARa and PPARc.
Insertion of a methylene linker from the oxyphenyl group to the
triazole (i.e., oxybenzyl vs oxyphenyl; 3a vs 3d) resulted in a
logs 2a, 2c and 2d are more potent PPARc agonists than the corre-
sponding N1-aryl substituted pyrazole analogs (e.g., 2b and 2e). It
Table 1
In vitro PPAR
a and c activities of pyrrole-acid analogs
CO2H
O
N
Ph
ꢀ10-fold improvement in the binding affinities to PPAR
a and
O
PPAR , but surprisingly there was no corresponding improvement
c
N
R
in functional activity. Insertion of a methylene between the tria-
1,3- & 1,4- series
zole and the acid in the oxyphenyl series (3e vs 3d, n = 0 vs 1) re-
Compound
Oxybenzyl
substitution
R
c
IC50
M)
c
EC50
M) IA (%)
a
IC50
M)
a
EC50
M) IA (%)
sulted in not only improved binding affinity (20-fold at PPAR
6-fold at PPAR ), but also more potent functional activity at both
receptors (4-fold at PPAR and 10-fold at PPAR. Moreover, when
c and
(l
(l
(l
(l
a
1a
1b
1c
1d
3-
3-
4-
4-
–CH3
–Ph
–CH3
–Ph
1.67
0.16
3.82
0.89
2.37 (72%)
0.55 (82%)
1.57 (84%)
4.20 (72%)
5.92
0.09
>10
1.47 (26%)
0.12 (18%)
>15
c
the triazole acetic acid is combined with the oxybenzyl moiety
(3f; m = 1 and n = 1), this additional structural adjustment resulted
in >100-fold increases in both the binding affinity and functional
0.89
0.66 (13%)
activities at PPAR
EC50 = 6 nM (83%), which is a potent dual PPAR
An X-ray co-crystal structure of 3f with PPAR
a
and PPAR
c
for analog 3f (
c
EC50 = 4nM (138%);
agonist.
receptor ligand-
a
a/c
c
Table 2
In vitro PPAR
a
and
c
activities of pyrazole-acid analogs
binding domain (LBD) was determined to 2.25 Å resolution
(Fig. 2).28 Of particular interest to the functional SAR is the H-bond-
ing network observed between ligand carboxylic acid, water and
neighboring residues: Ser289 (H3), His323 (H5), His449 (H11)
and Tyr473 (H12). In addition to the carboxylate inter-actions, an
indirect H-bond is also observed between the oxazole nitrogen,
water and Ser342 backbone NH (6–7 loop). The H-bond between
Tyr473 on H12 and the carboxylate anion would be expected to
be strong and is consistent with the observed superior agonist
behavior exhibited by 3f when compared to its seco analog, the tri-
azole acid 3a, where the expected trajectory of its carboxylate
would not be compatible with a strong H-bond to Tyr473.
CO2H
N
CO2H
O
O
N
Ph
Ph
N
N
N1
R
O
O
N2
2c
R
2a, 2b, 2d, 2e
Compound Oxybenzyl
substitution
R
c
IC50
M)
c
EC50 IA (%)
M)
a
IC50
M)
a
(%)
EC50
(lM)A
(
l
(l
(l
2a
2b
2c
2d
2e
3-
3-
3-
4-
4-
N2-Ph 0.14
0.023 (76%)
0.437 (19%)
0.156 (65%)
0.752 (90%)
0.490 (39%)
0.07
>25
0.14
2.17
>25
0.011 (26%)
0.648 (17%)
0.311 (24%)
0.126 (21%)
>25
N1-Ph 4.74
N-Ph
0.07
N2-Ph 2.16
N1-Ph 1.67

H. Zhang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1451–1456
1455
Table 4
In vitro PPAR
a
and c activities of homologated 3- and 4-triazole analogs
CO2H
n
CO2H
n
O
N
O
m
N
m
N
N
N
Ph
Ph
N
N
O
N
O
3- series
4- series
Compound
m/n
c
IC50
(lM)
c
EC50
(
lM)
a
IC50
(l
M)
a
EC50
(lM)
Compound
m/n
c
IC50
(
lM)
c
EC50
(lM)
a
IC50
(lM)
a
EC50
(lM)
IA (%)
IA (%)
IA (%)
IA (%)
3a
3d
3e
3f
1/0
0/0
0/1
1/1
0.069
0.67
0.034
0.005
3.79 (70%)
2.19 (39%)
0.56 (85%)
0.004
0.069
1.61
0.28
1.90 (30%)
4.02 (45%)
0.39 (72%)
0.006 (83%)
4a
4d
4e
4f
1/0
0/0
0/1
1/1
0.712
1.35
0.517
0.007
9.27 (87%)
7.33 (23%)
3.86 (20%)
0.013 (86%)
2.77
>50
3.35
0.054
6.15 (52%)
>15
4.14 (29%)
0.011 (118%
0.085
thus enabling these oxybenzyl triazole acetic acid analogs to better
form key interactions with key residues such as Tyr473 of PPAR
c
than the more rigid triazole acids. However, it should be noted that
higher binding affinity in the ligand-binding domains (LBDs) of
both PPAR
a
and PPAR
c
does not always necessarily correlate with
).
improved functional agonist activity (e.g., 2a and 2c at PPAR
a
From these in vitro data, several conclusions can be drawn.
First, analogs from the 3-oxybenzyl azole-acid series generally dis-
play more equivalent, potent PPARc and PPARa agonist activity
than their 4-oxybenzyl azole-acid counterparts. Second, among
the various azole-acid series, analogs with an N2-phenyl substitu-
ent (vs N1 and N3) have demonstrated the most potent PPAR
c
and/or PPARa agonist activities in vitro. Finally, within the triazole
acid series, the installation of a methylene linker (m = 1, n = 1) on
either the C-4 or C-5 position of the triazole is preferred for optimal
PPAR
among these triazoles is 3f, which has increased potency at PPAR
(>10-fold) and at PPAR (>200-fold) relative to muraglitazar
(PPAR EC50 = 35 nM; IA = 107% and PPAR EC50 = 1.41 M;
IA = 139%) in vitro.
c and PPARa functional activity. The most potent compound
c
a
c
a
l
Figure 2. The X-ray structure of 3f bound to PPARc LBD is illustrated with ligand
electron density shown as a mesh. Residues having significant H-bonding contact
with the ligand are highlighted along with two water molecules (red spheres).
The in vivo anti-diabetic and lipid-lowering activities of 3f were
further characterized in a 14-day study in female db/db mice29; the
data are shown in Table 5. Compound 3f (administered orally at a
10 mg/kg/day dose) showed excellent efficacy in significantly
reducing levels of plasma glucose (ꢁ36%) and triglycerides
(ꢁ50%) in this diabetic animal model.
0
Distances between heavy atoms are given in ÅAngstroms.
Table 5
In vivo activity profile of Compound 3f in db/db mice dosed p.o, q.d. for 14 days,
overnight fasted
In summary, the SAR of a series of N-phenyl-substituted azole
Treatment (10 mg/kg/day for
14 days)
Plasma triglycerides
(mg/dL)
Plasma glucose (mg/
mL)
acid analogs was explored. These studies showed that the PPAR
c
and activities in this series can be modulated by: (a) varying
a
Vehicle
3f
156 10
497 32
77 7 (ꢁ50%)*
316 34 (ꢁ36%)*
the oxybenzyl substitution pattern (3-oxybenzyl series vs 4-oxyb-
enzyl series); (b) the presence of a methylene linker at either the C-
4 or C-5 position of the triazole, and (c) the regiochemistry of the
preferred N-phenyl substituent on the azole ring. In particular,
*
p < 0.05.
analogs 3f and 4f exhibit very potent dual PPARa and PPARc ago-
In the 4-oxyphenyl series, triazole acid analog 4d (m = 0; n = 0)
showed weak agonist activity only at PPAR . Addition of a methy-
lene linker to the oxyphenyl ring of 4d (oxybenzyl analog 4a) re-
sulted in moderately improved PPAR (now a full agonist) and
PPAR agonist activity versus 4d. Insertion of a methylene linker
between the triazole and carboxylic acid of oxyphenyl analog 4d
(n = 1; 4e) resulted in comparable dual PPAR and PPAR in vitro
nist activity in vitro and analog 3f also showed excellent anti-dia-
betic and anti-dyslipidemic activity in the diabetic db/db mouse
model. In conclusion, we have established a novel series of N-phe-
nyl-substituted 1,2,3-triazole acid analogs as a new template/
c
c
a
chemotype for the design of potent dual PPARa/c agonists.
a
c
Acknowledgment
activities. Similar to the result from the 1,3-oxyphenyl series,
insertion of methylene linkers at both C-4 and C-5 of the triazole
(m = n = 1; 4f) also resulted in very substantial improvements in
We thank the BMS Discovery Analytical Sciences Department
for analytical support.
both the PPAR
oxybenzyl triazole-acetic acid 4f (now a potent dual PPAR
agonist) versus the parent oxyphenyl triazole acid 4d. As shown
from the X-ray crystal structure of 3f bound to PPAR , the insertion
of the two methylene linkers in 3f and 4f apparently allows for in-
creased configurational flexibility of the N-phenyl azole template,
c
and PPAR
a
agonist activities (>100-fold) of the
a/c
dual
References and notes
1. For general reviews on PPARs (a) Gervois, P.; Fruchart, J.-C.; Staels, B. Nat. Clin.
Practice Endocrinol. Metab. 2007, 3, 145; (b) Pershadsingh, H. A. Treatments
Endocrinol. 2006, 5, 89; (c) Cheng, X.-C.; Xu, W.-F. Drugs of the Future 2006, 31,
875; (d) Cheng, P. T. W.; Mukherjee, R. Mini Rev. Med. Chem. 2005, 5, 741.
c

1456
H. Zhang et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1451–1456
2. (a) Isseman, I.; Green, S. Nature 1990, 347, 645; (b) Lee, S. S.; Pineau, T.; Drago,
J.; Lee, E. L.; Owens, J. W.; Kroetz, D. L.; Fernandez-Salguero, P. M.; Westphal,
H.; Gonzalez, F. Mol. Cell. Biol. 1995, 15, 3012.
3. Staels, B.; Dallongeville, J.; Auwerx, J.; Schoonjans, K.; Leitersdorf, E.; Fruchart,
J.-C. Circulation 1998, 98, 2088.
15. Almansa, C.; Gómez, L. A.; Cavalcanti, F. L.; Arriba, A. F.; García-Rafanell, J.;
Forn, J. J. Med. Chem. 1997, 40, 547.
16. Pak, C. S.; Yang, H. C.; Choi, E. B. Synthesis 1992, 16, 1213.
17. Hamanaka, E. S.; Guzman-Perez, A.; Ruggeri, R. B.; Wester, R. T.; Mularski, C. J.
Patent application WO9943663.
4. Todd, P. A.; Ward, A. Drugs 1988, 36, 314.
5. Malinowski, J. M.; Bolesta, S. Clin. Ther. 2000, 22, 1151.
18. Armani, V.; Dell’Erba, C.; Novi, M.; Petrillo, G.; Tavani, C. Tetrahedron. 1997
1997, 53, 1751.
6. Gillies, P. S.; Dunn, C. J. Drugs 2000, 60, 333.
19. Cheng, P. T. W.; Devasthale, P.; Jeon, Y. T.; Chen, S.; Zhang, H. Patent application
WO200121602.
7. (a) Devasthale, P. V.; Chen, S.; Jeon, Y.; Qu, F.; Shao, C.; Wang, W.; Zhang, H.;
Farrelly, D.; Golla, R.; Grover, G.; Harrity, T.; Ma, Z.; Moore, L.; Ren, J.; Seethala,
R.; Cheng, L.; Sleph, P.; Sun, W.; Tieman, A.; Wetterau, J. R.; Doweyko, A.;
Chandrasena, G.; Chang, S. Y.; Humphreys, W. G.; Sasseville, V. G.; Biller, S. A.;
Ryono, D. E.; Selan, F.; Hariharan, N.; Cheng, P. T. W. J. Med. Chem. 2005, 48,
2248; (b) Devasthale, P. V.; Chen, S.; Jeon, Y.; Qu, F.; Ryono, D. E.; Wang, W.;
Zhang, H.; Cheng, L.; Farrelly, D.; Golla, R.; Grover, G.; Ma, Z.; Moore, L.;
Seethala, R.; Sun, W.; Doweyko, A. M.; Chandrasena, G.; Sleph, P.; Hariharan, N.;
Cheng, P. T. W. Bioorg. Med. Chem. Lett. 2007, 17, 2312; (c) Harrity, T.; Farrelly,
D.; Tieman, A.; Chu, C.; Kunselman, L.; Gu, L.; Ponticiello, R.; Cap, M.; Qu, F.;
Shao, C.; Wang, W.; Zhang, H.; Chen, S.; Devasthale, P.; Jeon, Y.; Seethala, R.;
Ren, J.; Zhou, M.; Ryono, D.; Biller, S.; Mookhtiar, K.; Wetterau, J.; Gregg, R.;
Cheng, P. T. W.; Hariharan, N. Diabetes 2006, 55, 240; (d) Tozzo, E.; Ponticiello,
R.; Swartz, J.; Farrelly, D.; Zebo, R.; Welzel, G.; Egan, D.; Kunselman, L.; Peters,
A.; Gu, L.; French, M.; Devasthale, P.; Janovitz, E.; Staal, A.; Harrity, T.; Belder,
R.; Cheng, P. T.; Whaley, J.; Taylor, S.; Hariharan, N. J. Pharmacol. Exp. Ther. 2007,
321, 107; (e) Harrity, T.; Chandrasena, G.; Chen, S.; Chu, C.; Devasthale, P.;
Farrelly, D.; Jeon, Y.; Kunselman, L.; Qu, F.; Ryono, D.; Sasseville, V.; Selan, F.;
Shao, C.; Wang, W.; Wetterau, J.; Zhang, H.; Cheng, P. T. W.; Hariharan, N.
Diabetes 2002, 51, 407.
20. Padwa, A.; Prein, M. J. Org. Chem. 1997, 62, 6842.
21. Ohno, M.; Itoh, M.; Ohashi, T.; Eguchi, S. Synthesis 1993, 08, 793.
22. Midland, M. M.; Tramontano, A.; Cable; John, R. J. Org. Chem. 1980, 45, 28.
23. Czernecki, S.; Ville, G. J. Org. Chem. 1989, 54, 610.
24. Earl, A. R.; Townsend, B. L. Can. J. Chem. 1980, 58, 2550.
25. Bachmann, W. E.; Struve, W. S. Org. React. 1942, 1, 38.
26. Seethala, R.; Golla, R.; Ma, Z.; Cheng, L.; Zhang, H.; O’Malley, K.; Lippy, J.; Zhang,
L.; Hariharan, N.; Cheng, P. T. W. Anal. Biochem. 2007, 363, 263.
27. In vitro PPAR agonist functional assays were performed by transiently
transfecting
respectively, into HEK293 cells stably expressing
Luciferase. Data were normalized for efficacy at 1 mM to known agonists
(rosiglitazone for hPPAR and GW-2331 for hPPAR ). Agonist binding results
GAL4-hPPAR
a
-LBD
or
GAL4-hPPAR
c
-LBD
constructs,
5
copies of GAL4RE-
c
a
in an increase in luciferase enzyme activity which can be monitored by
measuring luminescence upon cell lysing and the addition of luciferin
substrate. EC50 values (
as the concentration of the test ligand (
induction of HEK293 cells. The ‘‘intrinsic activity” of a test ligand is defined as
its activity at 1 M (expressed as a percentage) relative to the activity of the
primary standards (GW2331 for PPAR and rosiglitazone for PPAR
respectively, both tested at 1 M).
28. PDB deposition number is 3BC5.
lM) for PPAR
a
l
or
c agonist activity were calculated
M) required for the half-maximal fold
l
8. Williams, Johnathan M. J. Ed., Preparation of Alkenes, A Practical Approach,
Chapter 2, The Wittig reaction and related methods, Lawrence, N. J. Oxford
University Press, 1996.
a
c,
l
9. Lam, P. Y. S.; Clark, G. C.; Saubern, S.; Adams, J.; Winters, P. M.; Chan, M. T. D.;
Combs, A. Tetrahedron Lett. 1998, 39, 2941.
10. Skorcz, J. A.; Skorcz, J. A.; Suh, J. T.; Judd, C. I.; Finkelstein, M.; Conway, A. C. J.
Med. Chem. 1966, 9, 656.
11. Kim, Y. H.; Choi, J. Y. Tetrahedron Lett. 1996, 37, 8771.
12. Kato, T.; Chiba, T. Chem. Pharm. Bull. 1975, 20, 2203.
29. db/db mice, 8–10 weeks old, fed normal chow ad-lib, were orally dosed with
compound or vehicle (5% NMP, 20% PEG-400, 20 mM aqueous sodium
phosphate buffer, pH 8) once daily for 14 days. Blood samples were drawn
from the tail vein on the 7th day after overnight fasting, and plasma analysis
was performed on a COBAS automated analyzer. After 14 days, fasted animals
were sacrificed. Blood and tissue samples were collected for clinical
biochemistry analyses.
13. Garanti, L.; Sala, A.; Zecchi, G. Synthesis 1975, 666.
14. Sabitha, G. Synth. Commun. 1998, 28, 3065.