Benzimidazolones: A new class of selective peroxisome proliferator- activated receptor γ (PPARγ) modulators

Article abstract of DOI:10.1021/jm201061j

A series of benzimidazolone carboxylic acids and oxazolidinediones were designed and synthesized in search of selective PPARγ modulators (SPPARγMs) as potential therapeutic agents for the treatment of type II diabetes mellitus (T2DM) with improved safety profiles relative to rosiglitazone and pioglitazone, the currently marketed PPARγ full agonist drugs. Structure-activity relationships of these potent and highly selective SPPARγMs were studied with a focus on their unique profiles as partial agonists or modulators. A variety of methods, such as X-ray crystallographic analysis, PPARγ transactivation coactivator profiling, gene expression profiling, and mutagenesis studies, were employed to reveal the differential interactions of these new analogues with PPARγ receptor in comparison to full agonists. In rodent models of T2DM, benzimidazolone analogues such as (5R)-5-(3-{[3-(5-methoxybenzisoxazol-3-yl)benzimidazol-1-yl]methyl}phenyl) -5-methylox-azolidinedione (51) demonstrated efficacy equivalent to that of rosiglitazone. Side effects, such as fluid retention and heart weight gain associated with PPARγ full agonists, were diminished with 51 in comparison to rosiglitazone based on studies in two independent animal models. (Figure presented)

Full text of DOI:10.1021/jm201061j

Article
pubs.acs.org/jmc
Benzimidazolones: A New Class of Selective Peroxisome Proliferator-
Activated Receptor γ (PPARγ) Modulators†
Weiguo Liu,* Fiona Lau, Kun Liu, Harold B. Wood, Gaochao Zhou, Yuli Chen, Ying Li, Taro E. Akiyama,
Gino Castriota, Monica Einstein, Chualin Wang, Margaret E. McCann, Thomas W. Doebber,
Margaret Wu, Ching H. Chang, Lesley McNamara, Brian McKeever, Ralph T. Mosley, Joel P. Berger,
and Peter T. Meinke
Merck Research Laboratories, P.O. Box 2000, Rahway, New Jersey 07065, United States
S
* Supporting Information
ABSTRACT: A series of benzimidazolone carboxylic acids and oxazolidinediones were designed
and synthesized in search of selective PPARγ modulators (SPPARγMs) as potential therapeutic
agents for the treatment of type II diabetes mellitus (T2DM) with improved safety profiles
relative to rosiglitazone and pioglitazone, the currently marketed PPARγ full agonist drugs.
Structure−activity relationships of these potent and highly selective SPPARγMs were studied
with a focus on their unique profiles as partial agonists or modulators. A variety of methods, such
as X-ray crystallographic analysis, PPARγ transactivation coactivator profiling, gene expression
profiling, and mutagenesis studies, were employed to reveal the differential interactions of these
new analogues with PPARγ receptor in comparison to full agonists. In rodent models of T2DM,
benzimidazolone analogues such as (5R)-5-(3-{[3-(5-methoxybenzisoxazol-3-yl)benzimidazol-1-yl]methyl}phenyl)-5-methylox-
azolidinedione (51) demonstrated efficacy equivalent to that of rosiglitazone. Side effects, such as fluid retention and heart
weight gain associated with PPARγ full agonists, were diminished with 51 in comparison to rosiglitazone based on studies in two
independent animal models.
have demonstrated that selective PPARγ modulators
(SPPARγMs) bind to their receptors in a distinct mode
relative to full agonists, providing a physical basis for differential
pharmacological effects. Such ligands act as partial agonists in
cell-based transcriptional activity assays and induce an
attenuated adipocyte gene signature. In preclinical species,
SPPARγMs were shown to retain antidiabetic efficacy
comparable to that of full PPARγ agonists while displaying
reduced PPARγ mechanism-based adverse effects.11 Recently,
we reported the discovery of a series of indole-based carboxylic
acids. These selective PPARγ modulators12 showed potent
PPARγ activity in binding assays, partial agonism in cell-based
transactivation assays, and robust effects in lowering hyper-
glycemia in rodent models of diabetes. While pursuing the
SARs of this series of analogues, we discovered that the indole
core structures could be replaced by benzimidazolones,
resulting in increased selectivity toward PPARγ receptors and
a unique interaction profile with a panel of PPAR coactivators.
In this report, we describe X-ray crystallographic analyses,
mutagenesis studies, and coactivator recruiting patterns of this
new class of benzimidazolone SPPARγMs, as well as their SARs
and in vivo antidiabetic efficacy vs propensity to induce side
effects in preclinical species.
INTRODUCTION
■
Diabetes is a metabolic disorder escalating in the world
population at an alarming rate.1 Non-insulin-dependent type 2
diabetes mellitus (T2DM) accounts for more than 90% of the
cases. It is characterized by insulin resistance in the liver and
peripheral tissues. To achieve the physiologically required level
of glucose metabolism, diabetic patients require more insulin by
either increased endogenous production or direct injection of
the recombinant peptide. Both processes disrupt the normal
functioning of the insulin producing organ and often result in
pancreatic β-cell dysfunction. One way to tackle the underlying
causes for T2DM would be to increase the sensitivity of tissues
and therefore the body’s response to insulin.
The peroxisome proliferator-activated receptors (PPARs)
comprise a subfamily of ligand-activated nuclear hormone
receptors2 that are involved in regulating nutrient storage and
catabolism.3 There are three PPAR subtypes, commonly
designated as PPARα, PPARγ, and PPARδ(β).4 Agonists of
the γ subtype have been studied extensively for their role in
regulating glucose metabolism and insulin sensitivity.5 PPARγ
full agonists, such as rosiglitazone and pioglitazone, have been
developed and marketed for the treatment of T2DM.6
However, mechanism-based side effects including weight gain,
edema, congestive heart failure, and the recently reported
increased risk for bone fracture following rosiglitazone or
pioglitazone treatment are major untoward effects associated
with the use of PPARγ full agonists.7,8 Lately, several reports in
the literature,9 including work from our own laboratories,10
Received: August 7, 2011
Published: November 9, 2011
© 2011 American Chemical Society
8541
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
Chart 1. Benzylindole Carboxylic Acid vs Benzimidazolone Carboxylic Acid
resulted in double alkylation products and poor yields. Boc
protection was later found to be more stable for N-alkylation
and was commonly used for scale-up chemistry. The two
regioisomers resulting from Boc protection were separated by
flash chromatography and confirmed by comparison with
products derived from 8. The first N-substitution by acylation,
alkylation, or boronic acid coupling followed by deprotection
gave intermediate 9 with diverse R2 groups. Introduction of a
variety of R3 bearing acidic moieties as the second N-
substitution affords a targeted combinatorial library of
benzimidazolones analogues (10).
CHEMISTRY
■
In the previous reports,12c we described a series of N-
benzoylindole carboxylic acid analogues (1, Chart 1) as
SPPARγMs. Although potent, this class of compounds suffered
a few drawbacks such as modest inhibition in a panel of
cytochrome P450 enzymes, multiple off-target hits in a
commercial in vitro screen of common enzymes and receptors
(Panlabs), and some residual PPARα activity which complicates
interpretation of efficacy results.10b Conversion of the indole
core into an azaindole decreased the number of off-target hits
and P450 inhibition. However, these azaindole analogues
suffered from poor PK characteristics and thus diminished in
vivo efficacy.13 In the context of expanding SAR and improving
properties, we sought to replace the indole with a non-indole
moiety while retaining the overall structural topology and the
general SAR trend. The benzimidazolone design (2) fits this
criteria and allows the introduction of substitutions at both
imidazolone nitrogens, mimicking the 1,3-indole substitution. A
series of such 1,3-disubstituted benzimidazolone analogues
were prepared following procedures as depicted in Scheme 1.
Several key R2 and R3 moieties were prepared as depicted in
Schemes 2−4. Starting with properly substituted salicylic acids
(11), esterification followed by treatment with hydroxylamine
gave the corresponding hydroxamates (12). Cyclization with
carbodiimide followed by heating with phosphorus oxychloride
gave the substituted 3-chlorobenzisoxazoles (14).14 The chiral
phenoxycarboxylic acids or esters were prepared following
procedures as described in previous reports.12c Mitsunobu
coupling between a substituted anisole (15) and a chiral ethyl
lactate (16) followed by free radical bromination yielded the
desired benzyl bromides (18). Synthesis of the chiral
oxazolidinedione moieties involved multistep chemical and
asymmetric transformations as shown in Scheme 4. Again,
starting with a properly substituted anisole, formation of the
trifluoromethanesulfonyl ester (OTf) with subsequent palla-
dium (dppp) catalyzed coupling using vinyl butyl ether gave
intermediate 1915 which was hydrolyzed in HCl to yield the
acetophenone (20). Wittig reaction of the acetophenone gave
the corresponding α-methylstyrene (21) which was subjected
to Sharpless’s asymmetric dihydroxylation conditions to
produce the corresponding chiral diols (22).16 Oxidation of
this diol with Pt/C followed by esterification with diazo-
methane afforded the carboxylic ester (24). Cyclization of the
α-hydroxycarboxylic ester with urea in sodium ethoxide
produced the chial oxazolidinedione (25), which was protected
by a trityl group followed by bromination with NBS and AIBN
in carbon tetrachloride to give the benzyl bromide moiety (27)
ready for attachment to the benzimidazole core as the R3
substitution.
a
Scheme 1
a
Reagents and conditions: (a) AcOH/Ac2O, 0 °C; (b) AcOH/HNO3,
Systematic variation of R1, R2, and R3 groups in conjunction
with a convergent assembly process led to the syntheses of large
numbers of benzimidazolone analogues (10). In this
communication, only selected examples of these compounds
are presented to illustrate SAR.
room temp; (c) Pd/C/H2, MeOH; (d) NaOH, MeOH; (e) CDI,
THF; (f) NaH/(Boc)2O, followed by column separation of two
regioisomers; (g) R2-Cl/Cs2CO3/DMF, R2 = benzisoxazole, benzo-
thiazole, acyl, or benzyl; R2-B(OH)2, dppf/Pd(OAc)2/dioxane, R2 =
Ar; (h) TFA/MeOH; (i) R3-Br, Cs2CO3/DMF, room temp.
Starting from properly substituted anilines (3) or phenylenedi-
amines (6), protected benzimidazolone cores were prepared by
routine chemical manipulations. The acetyl protected benzimi-
dazolones (8) with position of R1 clearly defined were initially
used for subsequent transformations. However, the acetyl
group was labile under many alkylation conditions, which often
RESULTS AND DISCUSSION
■
In several previous reports, we have described extensive SAR
studies of indole-12 and azaindole-containing13 SPPARγMs. For
the benzimidazolone series, our initial efforts were to compare
their PPARγ activity with similar analogues in the indole series.
Although retaining their binding affinity, selectivity, and partial
8542
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
a
Scheme 2
a
Reagents and conditions: (a) TMS−CH2N2, benzene/MeOH (4:1), room temp; (b) NH2OH, NaOH, dioxane, room temp; (c) CDI/THF, 75 °C;
(d) POCl3, Et3N, 90 °C.
a
Scheme 3
a
Reagents and conditions: (a) PPh3/DIAD, CH2Cl2, room temp; (b) NBS/AIBN, CCl4, reflux.
a
Scheme 4
a
Reagents and conditions: (a) Tf2O/Py; (b) vinyl butyl ether, dppp/NEt3/Pd(OAc)2, DMF, 80 °C; (c) HCl, MeOH; (d) Ph3PCH3Br/n-BuLi,
ether; (e) AD-mix β, tBuOH/H2O, 0 °C; (f) 10% Pt/C, NaHCO3, H2O, 70 °C; (g) CH2N2, Et2O; (h) urea/NaOEt, EtOH, reflux; (i) TrCl/Et3N,
CH2Cl2; (j) NBS/AIBN, CCl4, reflux.
agonist activity, these new SPPARγMs exhibited divergent SAR
and certain unique profiles. The selected analogues described
herein are summarized in Tables 1−7. Their biological activities
are evaluated on two criteria: their affinity for human PPARγ
and PPARα (IC50)17 and their agonist activity in a human
PPAR-GAL4 transactivation assay in COS-1 cells in terms of
potency (EC50) and maximum activation (Max) relative to
rosiglitazone for PPARγ agonism.10 Initial SAR was focused on
systematic variations of R2 with constant R1 and R3
substitutions (Table 1). Direct analogues of the most potent
agonists in the indole series such as 28 and 29 were only
moderately active. The tert-butylphenyl substituted analogue 30
showed good binding activity. However, activation level was
very low in the transactivation assay. The benzisoxazole
substitution as R2 yielded the most potent analogue (32),
with a 68 nM binding potency and a typical partial agonist
profile in the transactivation assay. Next, the effects of different
substitutions on the benzisoxazoles (Table 2) were examined.
In general, most types and positions of substitution on the
benzisoxazole ring were well tolerated and yielded potent
PPARγ partial agonists. Among them, methoxy substitution at
the 6-position of the benzisoxazole produced the most potent
analogue (37) in PPARγ binding assay with an IC50 of 4.1 nM.
With 6-methoxybenzisoxazole as R2 and the original acid
moiety as R3, types and positions of R1 were explored. Unlike
the SAR in the indole series, both 6- and 5-substitutions on the
benzimidazolone are equipotent, including the unsubstituted
analogues. Trifluoromethyl and trifluoromethoxy groups
remain the best substituents at either the 5- or 6-position
(Table 3). With the rest of the structure optimized, the acidic
moiety R3 was extensively explored to maximize potency and
optimize PK profiles, as well as to minimize off-target activity
(Table 4). In the carboxylic acid series, meta-positioned
phenoxy-(2R)-propanoic acid analogues with a p- or an o-
chloro substitution (37, 42) continued to be the most potent
PPARγ partial agonists. The direct analogue with a (S)-
configuration (41) at the chiral center is about 10 times less
active in binding and exhibits no activity in the functional assay.
Analogues with no substitutions on the aryl (43) or larger
substitutions on the α-carbon (44) were significantly less active.
The same SAR can be transferred to the corresponding
analogues where the chiral phenoxycarboxylic acid was replaced
by a chiral oxazolidinedione with the same configuration. Here
again, the meta-substitution pattern was crucial for activity.
8543
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
Table 1. SAR at N3-Substitution of the Benzimidazolone
Table 3. SAR for Substitutions on the Benzimidazolones
It is important to note that all the compounds in this
benzimidazolone series demonstrated a high degree of
selectivity for PPARγ vs the PPARα and PPARδ subtypes.
No activity (>50 μM) was observed at these two receptors in
binding assays. Mouse PPAR activities were measured for
selected compounds and were found to be similar.
From the collection of potent SPPARγMs synthesized in this
series, the in vivo efficacy of selected analogues was evaluated
using db/db mice by daily oral administration of the test
compounds for 10 days.18 On the 11th day, blood glucose (glu)
levels from test animals were measured vs those from the
vehicle control, with rosiglitazone at 10 mg/kg (10 mpk) as the
positive control. Total exposures of test compounds in the
animals were also measured at the end of the study. As shown
in Table 5, despite being potent PPARγ ligands in vitro with
excellent pharmacokinetic profiles and moderate to high
exposure levels, most of the analogues failed to generate robust
glucose corrections in db/db mice except for compounds 51
and 52. Many factors including plasma protein binding and
tissue distribution could affect this outcome, and it has been
difficult to fully evaluate these effects. Less protein binding may
not transfer to increased efficacy for analogues with poor
permeability. On the contrary, highly protein bound ligands
could be more efficacious because of their preferential
distribution to key tissues such as adipocytes. Since 51 and
52 are similar in structure to the rest of the analogues with
comparable exposure levels and pharmacokinetic profiles, we
speculate that their enhanced efficacies may be derived from
their relatively higher activation levels in the transactivation
assay. In repeated assays, the two compounds consistently
produced PPARγ activation of 70−101% relative to rosiglita-
zone while other benzimidazolone analogues were in the range
of 18−50%. In the past, we have relied on reduced maximal
agonism relative to that of rosiglitazone in the PPARγ
transactivation assay as a criterion for selecting partial agonists
for in vivo evaluation. With activation levels almost as high as
rosiglitazone, are compounds 51 and 52 still SPPARγMs? To
answer this question, a series of experiments were carried out to
compare the activities of compound 51 and rosiglitazone head-
to-head.
a
SPA = scintillation proximity assay. Mean value of three
b
determinations. TA = transactivation assay. Rosiglitazone TA
response set to 100%. Mean value of three determinations.
Table 2. SAR for Substitutions on the Benzisoxazole
Analogues with an ortho or a para substitution (45 and 46)
were essentially inactive, while its meta-substituted counterpart
(47) was a potent PPARγ partial agonist. Removing the CH2
linker between the benzimidazolone core and the phenyl group
bearing the acid moiety resulted in a loss of activity (48).
Similar to the carboxylic acid series, introducing chloro
substitution on the aromatic ring bearing the acid moiety
afforded the most potent analogue (49) with about 10-fold
increase in potency.
Compound 51 was cocrystallized with human PPARγ ligand
binding domain (LDB), and its structure was compared with
those of rosiglitazone and pioglitazone, two prototypical
thiazolidinedione-based PPARγ full agonists. As shown in
8544
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
Table 4. SAR of N1-Substitution at the Benzimidazolone
Figure 1, the two full agonists bind across helix 3 with the acidic
thiazolidinedione moiety interacting through hydrogen binding
with Tyr-473 in helix 12, thereby stabilizing this activated
receptor−ligand complex.19 However, compound 51, like other
indole-based SPPARγMs reported earlier,20 binds along the axis
of helix 3 without interaction with Tyr-473 and helix 12. Its
acidic oxazolidinedione moiety actually hydrogen-binds with
Ser-342 far removed from Try-473 and helix 12. It is clear from
the crystal structures that compound 51 binds to the human
PPARγ receptor in a different mode than rosiglitazone. Its
binding mode is actually very similar to that of other reported
PPARγ partial agonists.21
Tyr-473 is believed to play an important role in the
stabilization of the LBD of the receptor by the full agonists.
This amino acid residue lies within the helix 12 transcriptional
activation function-2 domain that comprises a section of the
transcriptional coactivator binding pocket of the LBD.22,23
Many SPPARγMs lacking a direct interaction with Tyr-473
have been recently reported24−26 to result in improved
tolerability in preclinical animal testing. The apparent absence
of an interaction between compound 51 and Tyr-473 based on
the crystallographic data described above provides a physical
basis for a differential biological response.
composed of the GAL4 DNA-binding domain and either the
hPPARγWT or hPPARγY473A LBD. As shown in Figure 2,
rosiglitazone activity was dramatically reduced with hPPAR-
γY473A relative to hPPARγWT. In contrast, almost no changes
in potency with compounds 51 and 52 were observed between
the mutant and wild-type receptors. In our previous studies,
many structurally diverse PPARγ full agonists and SPPARγMs
were evaluated with hPPARγWT and hPPARγY473A, and
those data demonstrated that there were marked reductions in
activation maxima for full agonists but not with SPPARγMs
with the mutant receptor.20 This result provides further
physical evidence to support the assignment of compounds
51 and 52 as SPPARγMs.
Ligand-dependent protein−protein interactions between
nuclear receptors and their coactivators is a critical step in
the regulation of transcription.27 Agonist-induced conforma-
tional changes of the PPARγ LBD involves rearrangement of
helices 3 and 12 that modulate the binding surface for various
coactivators and co-repressors. Helices 3 and 12 are also
directly involved in ligand binding.28 Therefore, the binding of
unique small molecule ligands to different amino acid residues
of the PPARγ LBD could result in distinct conformational
changes and PPARγ-co-activator affinities.29 By use of a
homogeneous time-resolved fluorescence (HTRF) assay,30
PPARγ-co-activator affinity can be measured and quantified in
terms of binding EC50 and percent of activation. A panel of
known nuclear receptor coactivators, including cAMP response
element-binding protein (CBP),31 peroxisome proliferator-
activated receptor γ coactivator 1 (PGC-1),32 steroid receptor
coactivator 1 (Src-1),33 and PPARγ binding protein (PBP),34
were profiled in HTRF assays with a variety of PPARγ ligands.
In these assays, compound 51 and other partial agonists were
To further confirm the role that Tyr-473 plays in the binding
between PPARγ full and partial agonists and the functional
importance of this differential receptor−ligand interaction,
recombinant mutagenesis techniques were employed to replace
the Tyr-473 residue of PPARγ with an alanine.20 The activities
of compound 51 and rosiglitazone were then examined in
transcriptional activity assays performed as described pre-
viously10 in COS-1 cells transiently co-transfected with a
GAL4-responsive reporter construct and a chimeric receptor
8545
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
Table 5. In Vitro and in Vivo Profiles of Selected Benzimidazolone Analogues
a
Blood glucose levels of db/db mice orally administered with the test compounds were measured vs those from the vehicle control and expressed as
b
c
% of corrections (glu%). Total exposures of test compounds in the test animals were measured at the end of the study. 2 mpk po dosing in 5%
methylcellulose; 0.5 mpk iv dosing in ethanol/PEG400/water (1/4/5).
Figure 1. PPARγ ligand binding domain showing ligand−protein interactions of (a) full PPARγ agonists rosiglitazone (green) and pioglitazone
(yellow) (PDB access code 2PRG) and (b) PPARγ selective modulator, compound 51 (gray) (PDB access code 3TY0). The helix 3 region is
marked, and the tyrosine 473 residue is shown in orange.
compared with PPARγ full agonists including rosiglitazone. As
shown in Table 6, the full agonists displayed potent binding
affinities and a high level of activation on all four coactivators.
The SPPARγMs, however, showed diverse binding potencies
and activation levels toward these different coactivators, and all
SPPARγMs produced attenuated levels of activation compared
to full agonists. In general, SPPARγMs tended to have a
relatively higher level of activation on PGC-1 and PBP but
8546
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
Figure 2. SPA binding potencies of PPARγ ligands with hPPARγY473A versus hPPARγWT: (A) binding of rosiglitazone to hPPARγY473A and
hPPARgWT; (B) binding of 51 to hPPARγY473A and hPPARγWT; (C) binding of 52 to hPPARγY473A and hPPARγWT.
a
Table 6. Coactivator Recruiting Patterns between Full and Partial Agonists
CBP1-453
EC50 (μM)
PGC1
Src1
PBP
compd
Max (%)
EC50 (μM)
Max (%)
EC50 (μM)
Max (%)
EC50 (μM)
Max (%)
rosiglitazone
0.046
0.005
0.001
0.006
0.007
0.48
105
112
101
50
54
31
9
0.11
82
124
77
52
67
48
63
44
71
19
29
0.2
84
167
90
16
18
0
0.03
104
100
107
74
89
54
8
farglitazar37
0.004
0.001
0.005
0.005
0.46
0.006
0.002
>50
0.009
>50
>50
>50
0.13
>50
>50
0.003
0.001
0.005
0.006
0.44
SB21999438
1a
1b
30
32
50
51
53
54
>50
3.24
0
>50
>50
15
67
14
10
0.002
0.029
0.008
0.002
0
0.003
0.01
40
96
21
55
0.06
23
0
0.011
>50
0.011
1.82
0
a
Homogeneous time-resolved fluorescence (HTRF) assay of PPAR ligands with cAMP response element-binding protein (CBP), peroxisome
proliferator-activated receptor γ coactivator 1 (PGC-1), steroid receptor coactivator 1 (Src-1), and PPARγ binding protein (PBP). Full titrations
were conducted for all compounds from 15 μM down with WT hPPAR LBD. The PPARγ full agonist, 3-chloro-4-(3-(3-phenyl-7-propylbenzofuran-
6-yloxy)propylthio)phenylacetic acid (L-796449)39 was used as the 100% response control for each of the four coactivators. Rosiglitazone, farglitazar,
and 4-[2-(2-benzoxazolylmethylamino)ethoxy]-α-(2,2,2-trifluoroethoxy)- (αS)-benzenepropanoic acid (SB219994) were used as full agonist control.
a
moderate to no activation on CBP-1 and Src-1. The coactivator
recruiting pattern of compound 51 was in general agreement
with the other partial agonists. It recruited PGC-1 with a 71%
efficiency compared with 83% of rosiglitazone at a final
concentration of 15 μM but was only 23% efficient in binding
to Src-1 relative to rosiglitazone’s 84%. It is worth noting that
the activation level of compound 51 on PGC-1 was among the
highest in partial agonists tested, and it was also one of the
most efficacious partial agonists in db/db mice study for
glucose correction. This result is consistent with findings in
several recent publications that PGC-1 is preferably recruited
by SPPARγMs vs other coactivators and has been shown to be
more directly associated with insulin sensitization efficacy.35
This coactivator was also found to be critical in preventing
triglyceride accumulation and potentiating insulin signaling.36
Compound 51 was also evaluated in microarray gene
expression studies using differentiated 3T3-L1 adipocytes. A
collection of PPARγ full and partial agonists were tested for
effects on expression levels of 4619 reporter genes believed to
be directly or indirectly regulated by PPARγ activation.40 The
gene signature of 51 was similar to those of other SPPARγMs,
demonstrating a γ activation index (GAI)12d of 0.71 relative to
1.0 for rosiglitazone and most other full agonists.
Table 7. Multidosing db/db Studies of Compound 51
compd 51
parameter
rosiglitazone
dose (mpk)
3
49
70
10
74
20
77
25
84
30
87
10
b
glucose (%)
60−80
220−720
b
c
AUC (μM·h)
129
450
a
Male db/db mice and lean mice were dosed daily for 11 days by
gavage with vehicle or the indicated doses of test compounds. Glucose
corrections were calculated as percent correction of vehicle control.
b
c
Mean value (n = 7). P < 0.01 vs vehicle control. Mean value (n = 4).
better than rosiglitazone. At 10 mpk and an exposure of 129
μM, compound 51 produced a glucose correction of 74%
compared to the 60−80% glucose lowering by 10 mpk
rosiglitazone with exposures ranging from 220 to 720 μM
(based on historical data). At 20, 25, and 30 mpk, compound
51 continued to show improved glucose correction of 77%,
84%, and 87%, respectively. In a parallel experiment, the water
content of the epididymal fat pad (EWAT) of the db/db mice
treated with compounds 51 and 52 and rosiglitazone were
measured vs that of the untreated group. In contrast to
rosiglitazone which caused significant dose dependent water
content increase in EWAT at 10, 30, and 100 mpk, compounds
51 and 52 generated almost no change of water content at the
equally efficacious doses of 10 and 30 mpk, and 10 and 50 mpk,
respectively (Figure 3). These results suggest a diminished
propensity of these SPPARγMs to induce edema in comparison
Multidosing studies of 51 in db/db mice were carried out
with simultaneous monitoring of blood glucose and drug
exposure levels. As shown in Table 7, compound 51 produced
dose- and exposure-dependent glucose corrections in this
diabetic mouse model with efficacy equivalent to or slightly
8547
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
IC50 values of 91, 22, 10, and 2 μM, respectively, for 2D6, 3A4,
2C8, and 2C9, an improved profile over some of the indole
analogues. In addition, a receptor panel screen (140 assays,
Panlabs) of 51 revealed only one assay (serine/threonine
kinase, 0.951 μM) with activity under 1 μM and no ion channel
(IKr and PXR) activities at concentrations of up to 10 μM.
CONCLUSION
■
A new class of potent and highly selective PPARγ partial
agonists based on the benzimidazolone platform was discovered
by rational design. Large numbers of analogues were prepared
through a convergent synthetic strategy, and their SARs were
extensively explored. Replacement of the indole core with
benzimidazolone resulted in a significant increase in PPARγ
selectivity over PPARα and PPARδ with moderate improve-
ments in off-target activities. Substituted benzisoxazoles are
preferred over other aryl, alkyl, and acyl groups as R2 for
occupying the hydrophobic region of the receptor. Use of chiral
oxazolidinediones in place of phenoxycarboxylic acids as the
acidic moiety (R3) generally enhances pharmacokinetic profiles.
The lead compound, 51, displayed robust glucose and insulin
lowering activity similar or superior to that of rosiglitazone in
rodent models of T2DM. Unlike thiazolidinediones and other
carboxylic acid based PPARγ full agonists, this new class of
compounds behaves as selective PPARγ modulators
(SPPARγM). This designation is supported by X-ray crystallo-
graphic analysis and mutagenesis studies as well as their unique
profiles in recruiting a panel of nuclear receptor coactivators.
Preferential recruitment of PGC-1 and PBP over other
coactivators by 51 may contribute to its enhanced in vivo
efficacy compared to other similar partial agonists in its class.
Plasma volume expansion and cardiac hypertrophy were
diminished with 51 vs that caused by rosiglitazone in obese,
insulin resistant fa/fa rats. These unique biological properties
displayed by SPPARγM 51 make it an attractive candidate for
further evaluation as a potential therapeutic agent in the
treatment of T2DM.
Figure 3. PPARγ full agonist rosiglitazone vs SPPARγMs 51 and 52 in
EWAT water content in db/db mice. Male db/db mice were dosed as
described in the in vivo db/db mouse studies. At the end of day 11
dosing, epididymal fat pad tissue (EWAT) of test animal was obtained
and water content was determined by vacuum drying. Data are
displayed as mean value of seven determinations.
with PPARγ full agonists at comparable levels of antihypergly-
cemic efficacy.
To further evaluate compound 51 in vivo as a SPPARγM, a
7-day study was conducted in obese, insulin resistant Zucker fa/
fa rats,40 a preclinical model we used to directly compare the
insulin sensitizing efficacy vs the mechanism-based side effects
of PPARγ ligands. The plasma volume (PV) and heart weight
of test animals were measured as well as their insulin levels and
lipid profiles. At the conclusion of the study, compound 51
(0.1, 1.0, 10, 100 mpk) was compared to rosiglitazone (1.0, 10,
and 100 mpk). Both compounds substantially lowered elevated
levels of free fatty acids, insulin, and triglycerides, indicating
significant insulin sensitizing activity. However, efficacy was
achieved with a slightly lower exposure of compound 51
(insulin lowering ED50 = 281 μM·h) than rosiglitazone (insulin
lowering ED50 = 312 μM·h). Importantly, while rosiglitazone
produced clear dose responsive increases in plasma volume and
heart weight at dosages of ≥10 mpk, no significant increase in
heart weight was seen with compound 51. Although a modest
increase in plasma volume was observed with compound 51 at
the highest dose tested, it was much diminished in comparison
with rosiglitazone, especially at the higher doses (Figure 4).
Another attractive feature of the benzimidazolones was
revealed in counterscreening assays. For instance, in a screen of
cytochrome P450 isozymes for potential inhibition, 51 showed
EXPERIMENTAL SECTION
■
In Vitro Assays. Activities of compounds were evaluated for both
binding affinity and functional activity. Binding affinities for the PPARs
were measured in a scintillation proximity assay (SPA) as previously
described.10a Selected compounds from the binding assay were then
evaluated for their potencies of PPAR gene activation in cell-based
Figure 4. Evaluation of 51 in fa/fa Zucker rat for plasma volume and cardiac effects. Animals (n = 8) were treated daily for 7 days with vehicle,
rosiglitazone (1, 10, and 100 mpk), or 51 (0.1, 1.0, 10, 100 mpk). Circulating drug and insulin levels, plasma volume, and heart weight were
determined as described in the Experimental Section. The drug exposure levels associated with a 50% reduction in plasma insulin (ED50) for
rosiglitazone and 51 were 312 and 281 μM·h, respectively.
8548
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
transcription assays (TA) using GAL4-PPAR chimeric receptors as
previously described.20 Homogeneous time-resolved fluorescence
(HTRF) assays of PPAR ligands with selected nuclear receptor
coactivators were carried out according to literature methods.40 All
results were produced in triplicate, and mean values are reported.
In Vivo db/db Mouse Studies. Male db/db (C57BLKS/J-m
+/+Leprdb) mice at 12−13 weeks of age and nondiabetic db/+ (lean)
mice from Jackson Laboratories were housed seven mice per cage and
fed a rodent chow with free access to water. Mice (seven per group)
received a once daily dose of test compounds with vehicle (0.25%
methylcellulose) by oral gavage for 11 days. Blood was collected from
the tail immediately prior to the next dosing at days 0, 4, 7, and 11 for
measurement of the plasma glucose levels.
Evaluation of Insulin Sensitization. Tail nick blood samples were
collected 1 day before the start of dosing and 1 day after the seventh
dose. Plasma glucose (Sigma glucose Trinder assay kit, St. Louis, MO,
U.S.), insulin (rat insulin RIA from American Laboratory Products
Company, Windham, NH, U.S.), free fatty acids (FFA), and
triglyceride (both from Roche Diagnostics, Basel, Switzerland)
concentrations were determined following the methods provided by
the manufacturers.
Tissue Water Content Measurements. Immediately after eutha-
nasia, epididymal fat pads were removed and weighed and placed in
individual glass scintillation vials. Weighed tissues were subjected to a
vacuum-applied Speed Vac centrifugation/drying process over a 24 h
period. After this drying process, tissues were weighed again to
calculate individual tissue water content.
Plasma Volume Measurement. After determination of extracel-
lular fluid volume, plasma volume was measured in anesthetized
animals using a dye dilution technique following methods described by
Belcher and Harriss38 with minor modifications. Evans blue dye
solution (25 mg/mL in physiological saline) was filtered through a
0.22 μm filter prior to injection into a femoral vein. Twenty minutes
after injection, a heparinized blood sample (2 mL) was withdrawn
from the descending aorta. Plasma was separated by centrifugation of
the blood at 1100g for 15 min; samples were kept at −80 °C until
assayed. Absorbance of the thawed plasma was read at 620 nm, and
plasma Evans blue dye concentrations were calculated according to a
standard curve generated by a serial dilution of the 25 mg/mL Evans
blue dye saline solution. Plasma volume was calculated by using the
dilution factors of Evans blue as shown below.
of compound 51 was resolved to 2.35 Å. The refined coordinates have
been deposited at the Protein Data Bank with access code of 3TY0.
Synthetic Materials and Methods. Reagents and solvents were
obtained from commercial suppliers and were used without further
purification. Flash chromatography was performed using E. Merck
1
silica gel (230−400 mesh). H NMR spectra were recorded in the
deuterated solvents specified on a Varian Unity INOVA 500 MHz
instrument. Chemical shifts are reported in ppm in the indicated
solvent with TMS as internal standard. LC−MS was conducted using
HP1100 and Micromass ZQ instruments. Elemental analysis results
were obtained from Robertson Microlit Laboratories. Purities of all
compounds reported were above 95% determined by high perform-
ance liquid chromatography (HPLC) with UV detection at two
wavelengths of 220 and 254 nm. Purities of key compounds were also
confirmed by elemental analyses.
1-Boc-5-Trifluoromethylbenzimidazole-2-one and 1-Boc-6-
Trifluoromethylbenzimidazole-2-one (7a, R1 = 5-CF3; 7b, R1 =
6-CF3). To a solution of 5-trifluoromethylbenzimidazole-2-one (25 g,
123 mmol) in DMF (150 mL) was added NaH (95%, 3.5 g, 148
mmol). The reaction mixture was stirred at room temperature for 2 h
followed by addition of di-tert-butyl dicarbonate (27 g, 123 mmol).
The mixture was stirred at room temperature overnight, diluted with
water (400 mL), and extracted with ethyl acetate (2 × 150 mL). The
organic extract was washed with water (2 × 100 mL) and brine (100
mL), dried over anhydrous MgSO4, and concentrated under vacuum.
The residue was chromatographed on silica gel with hexane/ethyl
acetate (3:1) as the solvent system to obtain 9.2 g (25% yield) of 1-
Boc-5-trifluoromethylbenzimidazole-2-one (7a) as an earlier eluted
product. 1H NMR (CDCl3, 500 MHz) δ 9.79 (brs, 1H), 7.87 (d, 1H),
7.45 (d, 1H), 7.43 (s, 1H), 1.72 (s, 9H). LC−MS (ESI): >95% purity
at λ 220 and 254 nm. MS: m/z 303.09 [M + H]+. Fractions containing
the later eluted product were combined and concentrated to dryness
to obtain 17.5 g (47% yield) of 1-Boc-6-trifluoromethylbenzimidazole-
1
2-one (7b). H NMR (CDCl3, 500 MHz) δ 9.98 (brs, 1H), 8.09 (s,
1H), 7.49 (d, 1H), 7.23 (d, 1H), 1.76 (s, 9H). LC−MS (ESI): >95%
purity at λ 220 and 254 nm. MS: m/z 303.09 [M + H]+.
1-(4-Chlorobenzyl)-5-trifluoromethylbenzimidazole-2-one
(9a, R1 = 5-CF3, R2 = 4-Chlorobenzyl). To the solution of 1-Boc-6-
trifluoromethylbenzimidazole-2-one (7b, 100 mg, 0.33 mmol) in DMF
(2 mL) were added 4-chlorobenzyl bromide (81 mg, 0.39 mmol) and
Cs2CO3 (300 mg, 0.92 mmol). The mixture was stirred at room
temperature for 2 h, diluted with water (4 mL), and extracted with
ethyl acetate (2 × 3 mL). The organic extracts were combined and
concentrated to obtain a solid. The solid was dissolved in TFA (1.5
mL). The mixture was stirred at room temperature for 2 h, and
concentrated to dryness. The residue was purified by silica gel
chromatography with hexane/ethyl acetate (4:1) as solvent system to
Heart Weight Measurements. Animals were then euthanized by
pneumothorax and exsanguination, and the heart was excised, blotted,
and weighed.
1
obtain 90 mg (84% yield) of a white solid. H NMR (DMSO-d6, 500
MHz) δ 9.35 (brs, 1H), 7.38 (s, 1H), 7.34 (d, 2H), 7.28 (d, 2H), 7.27
(d, 1H), 6.93 (d, 1H), 5.10 (s. 2H). LC−MS (ESI): >95% purity at λ
220 and 254 nm. MS: m/z 327.0 [M + H]+.
Pharmacokinetic Studies. For the determination of test
compound exposures in animals, blood samples were taken at
specified intervals. The plasma was separated, acidified by the addition
of 0.5 M formate buffer, pH 3.0 (0.3 mL of formate buffer per mL
plasma), and stored at −70 °C prior to extraction and analysis.
Concentrations of the test compounds were determined following
protein precipitation with acetonitrile. Calibration curves were
constructed using plasma of control animals by addition of known
quantities of the testing compounds as an internal standard.
Quantitative analysis were carried out by LC−MS−MS using a PE
Sciex API 3000 triple quadrupole mass spectrometer. Plasma
concentrations and PK parameters were determined using Watson
software.
X-ray Crystallography Studies. Purified PPARγ LBD (residues
Gln-203 to Tyr-477) was mixed with compound 51 at 1.1:1 M ratio of
compound/protein on ice and allowed to stand at 4 °C overnight.
Crystals were grown by vapor diffusion at room temperature in a
Cryschem MVD-24 crystallization tray (Charles Supper, Natick, MA).
A suitable crystal was selected, and its X-ray diffraction data were
collected at the ID-17 beamline of the Industrial Macromolecular
Crystallography Association Collaborative Access Team. The structure
1-(4-Chlorobenzoyl)-5-trifluoromethylbenzimidazole-2-one
(9b, R1 = 5-CF3, R2 = 4-Chlorobenzoyl). To the solution of 1-Boc-
6-trifluoromethylbenzimidazole-2-one (7b, 200 mg, 0.66 mmol) in
anhydrous pyridine (4 mL) was added 4-chlorobenzoyl chloride (115
mg, 0.66 mmol). The mixture was stirred at room temperature
overnight and concentrated to dryness. The residue was dissolved in
TFA (2 mL) and stirred at room temperature for 2 h. The mixture was
concentrated to dryness and purified by silica gel column
1
chromatography to obtain 200 mg (90% yield) of a solid. H NMR
(DMSO-d6, 500 MHz) δ 7.92 (d, 1H), 7.81 (d, 2H), 7.56 (d, 2H),
7.66 (brs, 1H), 7.48 (d, 1H), 7.32 (brs, 1H). LC−MS (ESI): >95%
purity at λ 220 and 254 nm. MS: m/z 341.0 [M + H]+.
1-(4-tert-Butylphenyl)-5-trifluoromethylbenzimidazole-2-
one (9c, R1 = 5-CF3, R2 = 4-tert-Butylphenyl). To the solution of
1-Boc-6-trifluoromethylbenzimidazole-2-one (7b, 870 mg, 2.88 mmol)
in methylene chloride (100 mL) were added 4-tert-butylphenylboranic
acid (960 mg, 2.88 mmol), Cu(OAc)2 (521 mg, 2.88 mmol),
triethylamine (0.94 mL, 14 mmol), and molecular sieves (4 Å, 200
mg). The mixture was stirred at room temperature under open air
8549
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
overnight and filtered. The filtrate was concentrated to dryness. The
residue was dissolved in TFA (5 mL), stirred at room temperature for
3 h, and concentrated to dryness. The residue was redissolved in ethyl
acetate (5 mL) and passed through a silica gel column and washed
with hexane/ethyl acetate (3:1) (500 mL). The filtrate was
concentrated to obtain 800 mg (84 yield) of a solid. LC−MS (ESI):
>95% purity at λ 220 and 254 nm. MS: m/z 334.0 [M + H]+.
1-[(6-Chloro)benzisoxazol-3-yl]-5-trifluoromethylbenzimi-
dazole-2-one (9d, R1 = 5-CF3, R2 = 6-Chlorobenzisoxazol-3-yl)
and 1-[(6-Chloro)benzisoxazol-3-yl]-6-trifluoromethylbenzimi-
dazole-2-one (9e, R1 = 6-CF3, R2 = 6-Chlorobenzisoxazol-3-yl).
To a solution of 1-Boc-6-trifluoromethylbenzimidazole-2-one (7b, 800
mg, 264 mmol) in DMF (4 mL) was added 3,6-dichlorobenzoisox-
azole (500 mg, 264 mmol) and Cs2CO3 (1.5 g, 570 mmol). The
suspension was heated to 150 °C in an oil bath and stirred overnight.
The mixture was then cooled to room temperature, diluted with water
(5 mL), and extracted with ethyl acetate (3 × 5 mL). The organic
extracts were combined, dried over anhydrous MgSO4, and
concentrated to dryness. The residue was purified by silica gel column
chromatography using hexane/ethyl acetate (4:1) as solvent system.
Fractions containing earlier eluted product were combined and
concentrated to obtain 140 mg (15% yield) of 1-[3-(6-chloro)-
benzisoxazoyl]-5-trifluoromethylbenzimidazole-2-one (9d) as a white
1-(6-Chlorobenzisoxazol-3-yl)benzimidazole-2-one (9i, R1 =
H, R2 = 6-Chlorobenzisoxazol-3-yl) and 1-(6-Methoxybenzi-
soxazol-3-yl)]benzimidazole-2-one (9j, R1 = H, R2 = 6-
Methoxybenzisoxazol-3-yl). To a solution of benzimidazole-2-
one (Aldrich, 1.1 g, 7.5 mmol) in DMF (20 mL) was added 3,6-
dichlorobenzoisoxazole (1.5 g, 7.5 mmol) and Cs2CO3 (4.8 g, 15
mmol). The suspension was heated to 150 °C in an oil bath and
stirred overnight. The mixture was then cooled to room temperature,
diluted with water (30 mL), and stirred in an ice bath for 2 h. The
mixture was then filtered. The solid was washed with water and dried
under high vacuum to obtain 2.0 g (93% yield) of a yellow solid as 1-
(6-chlorobenzisoxazol-3-yl)benzimidazole-2-one (9i). LC−MS (ESI):
>95% purity at λ 220 and 254 nm. MS: m/z 286.1 [M + H]+.
To the solution of the above product (1.0 g, 3.5 mmol) in DMF (15
mL) was added NaOMe solution in methanol (30 wt %/wt, 10 mL).
The mixture was heated to 80 °C under vacuum to remove residual
methanol and then stirred at 110 °C under nitrogen overnight. The
mixture was then cooled to room temperature, diluted with water (50
mL), and adjusted to pH 6 with 10% aqueous HCl. The resulting
suspension was stirred in an ice bath for 2 h and filtered. The solid was
washed with water and dried in an oven (90 °C) for 2 h and then
under high vacuum at room temperature for 3 h to obtain 1.1 g (93%
yield) of a slightly yellow solid (9j). LC−MS (ESI): >95% purity at λ
220 and 254 nm. MS: m/z 282.1.0 [M + H]+.
1
solid. H NMR (CDCl3, 500 MHz) δ 8.72 (brs, 1H), 8.29 (s, 1H),
7.96 (d, 1H), 7.64 (m, 2H), 7.52 (d, 1H), 7.44 (s. 1H). LC−MS
(ESI): >95% purity at λ 220 and 254 nm. MS: m/z 370.1 [M + H]+.
Fractions containing later eluted product were combined and
concentrated to obtain 46 mg (5% yield) of 1-[3-(6-chloro)-
benzisoxazoyl]-6-trifluoromethylbenzimidazole-2-one (9e) as a white
2-(3-Methylphenyl)propene (21a, Y = H). To a suspension of
methyltriphenylphosphonium bromide (33.2 g, 92.9 mmol) in diethyl
ether (200 mL) at room temperature was added slowly n-BuLi (2.5 M
in hexanes, 37 mL). The resulting yellow solution was stirred for 30
min before 3-methylacetophenone (12 mL, 90 mmol) was introduced.
The mixture was stirred at room temperature overnight. The
precipitate was removed by filtration. The solvent was removed in
vacuum. Purification by flash chromatography gave 8 g (67% yield) of
the 2-(3-methylphenyl)propene product.
(2R)-(3-Methylphenyl)-1,2-propane-diol (22a, Y = H). To a
suspension of AD-mix-β (7 g) in H2O/tBuOH (25 mL/25 mL) at 0
°C was added 2-(3-methylphenyl)propene (21a) (0.66 g, 5 mmol).
The reaction mixture was stirred at 0 °C overnight. Na2SO3 (7.5 g)
was added. After 1 h at room temperature, the mixture was extracted
with EtOAc (3×). The combined organic layers were washed with
brine, dried over MgSO4, filtered, and concentrated in vacuum.
Purification by flash chromatography gave 0.71 g (82% yield) of (2R)-
(3-methylphenyl)-1,2-propane-diol. LC−MS (ESI): >95% purity at λ
220 and 254 nm. MS: m/z 167.1 [M + H]+.
1
solid. H NMR (CDCl3, 500 MHz) δ 8.44 (brs, 1H), 8.30 (s, 1H),
8.18 (s, 1H), 7.64 (m, 2H), 7.55 (d, 1H), 7.25 (d. 1H). LC−MS
(ESI): >95% purity at λ 220 and 254 nm. MS: m/z 370.1 [M + H]+.
1-[(6-Chloro)benzisoxazol-3-yl]-5-trifluoromethoxylbenzi-
midazole-2-one (9f, R1 = 5-OCF3, R2 = 6-Chlorobenzisoxazol-3-
yl) and 1-[(6-Chloro)benzisoxazol-3-yl]-6-trifluoromethoxyl-
benzimidazole-2-one (9g, R1 = 6-OCF3, R2 = 6-Chlorobenzisox-
azol-3-yl). To a solution of 1-Boc-6-trifluoromethoxybenzimidazole-
2-one (7d, R1 = 6-OCF3, 5 g, 15.7 mmol) in DMF (20 mL) were
added 3,6-dichlorobenzoisoxazole (3.0 g, 15.7 mmol) and Cs2CO3 (11
g, 31.4 mmol). The suspension was heated to 150 °C in an oil bath
and stirred overnight. The mixture was then cooled to room
temperature, diluted with water (30 mL), and extracted with ethyl
acetate (2 × 20 mL). The organic extracts were combined, dried over
anhydrous MgSO4, and concentrated to dryness. The residue was
purified by silica gel column chromatography using hexane/ethyl
acetate (4:1) as solvent system. Fractions containing earlier eluted
product were combined and concentrated to obtain 2.4 g (40% yield)
of 1-[3-(6-chloro)benzisoxazoyl]-5-trifluoromethoxylbenzimidazole-2-
(2R)-2-Hydroxy-2,3-dimethylbenzeneacetic Acid (23a, Y =
H). To a solution of (2R)-(3-methylphenyl)-1,2-propane-diol (22a)
(0.7 g, 4.2 mmol) in water (50 mL) were added NaHCO3 (0.4 g, 5.5
mmol) and 10% Pt/C (0.7 g). Air was bubbled through the reaction
mixture via a gas dispenser at 70 °C overnight. The mixture was cooled
to room temperature and filtered through Celite. The filtrate was
acidified with aqueous H2SO4 (1.0 N) to pH 2 and extracted with
EtOAc (3×). The combined organic layers were washed with brine,
dried over MgSO4, filtered, and concentrated in vacuum to give 0.66 g
1
one (9f) as a white solid. H NMR (DMSO-d6, 500 MHz) δ 8.28 (d,
1H), 8.12 (s, 1H), 7.74 (s, 1H), 7.54 (d, 1H), 7.14 (m, 2H). LC−MS
(ESI): >95% purity at λ 220 and 254 nm. MS: m/z 370.1 [M + H]+.
Fractions containing later eluted product were combined and
concentrated to obtain 1.3 g (19.1% yield) of 1-[3-(6-chloro)-
benzisoxazoyl]-6-trifluoromethoxylbenzimidazole-2-one (9g) as a
1
(88% yield) of the acid. H NMR (DMSO-d6, 500 MHz) δ: 7.32 (s,
1
1H), 7.28 (d, 1H), 7.19 (t, 1H), 7.02 (d, 1H), 2.25 (s, 3H), 1.59 (s,
3H). LC−MS (ESI): >95% purity at λ 220 and 254 nm. MS: m/z
181.1 [M + H]+.
white solid. H NMR (DMSO-d6, 500 MHz) δ 8.28 (d, 1H), 8.12
(s, 1H), 7.66 (s, 1H), 7.54 (d, 1H), 7.22 (m, 2H). LC−MS (ESI):
>95% purity at λ 220 and 254 nm. MS: m/z 370.1 [M + H]+.
1-[(6-Methoxy)benzisoxazol-3-yl]-6-trifluoromethoxylbenzi-
midazole-2-one (9h, R1 = 6-OCF3, R2 = 6-Methoxybenzisox-
azol-3-yl). To a solution of 1-[3-(6-chloro)benzisoxazoyl]-6-trifluor-
omethoxylbenzimidazole-2-one (9g, 1.2 g, 3.25 mmol) in DMF (5
mL) was added NaOMe solution in methanol (30 wt %/wt, 20 mL).
The mixture was heated to 80 °C under vacuum to remove residual
methanol and then stirred at 110 °C under nitrogen overnight. The
mixture was then cooled to room temperature, diluted with water (50
mL), and adjusted to pH 6 with 10% aqueous HCl. The resulting
suspension was stirred in an ice bath for 2 h and filtered. The solid was
washed with water and dried in an oven (90 °C) for 2 h and then
under high vacuum at room temperature for 3 h to obtain 1.1 g (93%
yield) of a slightly yellow solid. LC−MS (ESI): >95% purity at λ 220
and 254 nm. MS: m/z 366.0 [M + H]+.
(2R)-2-Hydroxy-2,3-dimethylbenzeneacetic Acid Methyl
Ester (24a, Y = H). To a solution of (2R)-2-hydroxy-2,3-
dimethylbenzeneacetic acid (23a, Y = H) (3.64 g, 20.2 mmol) in
anhydrous diethyl ether (100 mL) was added diazomethane ether
solution (freshly produced following procedures in Aldrich Technical
Bulletin AL-180) until a bright yellow color is produced or no more
bubbles evolved. The solution was then concentrated to dryness to
obtain 3.9 g (99% yield) of a white solid. 1H NMR (CDCl3, 500 MHz)
δ: 7.39 (s, 1H), 7.35 (d, 1H), 7.25 (t, 1H), 7.13 (d, 1H), 3.79 (s, 3H),
2.40 (s, 3H), 1.80 (s, 3H). LC−MS (ESI): >95% purity at λ 220 and
254 nm. MS: m/z 195.1 [M + H]+.
(5R)-5-Methyl-5-(3-methyphenyl)-2,4-oxazolidinedione
(25a, Y = H). To a solution of methyl (2R)-2-hydroxy-2,3-
dimethylbenzeneacetic acid methyl ester (24a, Y = H) (3.9 g, 20.1
8550
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
mmol) in anhydrous ethanol (50 mL) were added NaOEt in ethanol
(21 wt %/wt, 9.8 mL, 30 mmol), and urea (1.5 g, 24.2 mmol). The
mixture was heated to 95 °C and refluxed overnight. The solution was
then cooled to room temperature, acidified with 1 N HCl,
concentrated to small volume, and diluted with water (100 mL).
The aqueous mixture was extracted with ethyl acetate (3 × 50 mL).
The organic extracts were combined, washed with brine, dried over
anhydrous MgSO4, and concentrated to dryness to obtain 4.3 g (100%
yield) of an oil which was used in the next step without further
purification. 1H NMR (CDCl3, 500 MHz) δ: 8.70 (s, broad, 1H), 7.4−
7.2 (overlapping signals, 4H), 2.40 (s, 3H), 1.96 (s, 3H). MS: m/z
206.1 [M + H]+.
(5R)-5-Methyl-5-(3-methyphenyl)-N-trityl-2,4-oxazolidine-
dione (26a, Y = H). To a solution of (R)-5-methyl-5-(3-
methyphenyl)oxazolindione (25a, 4.3 g, 20.9 mmol) in anhydrous
methylene chloride (50 mL) were added triethylamine (2.1 g, 23
mmol) and trityl chloride (6.4 g, 23 mmol). The mixture was stirred
under nitrogen at room temperature for 1 h, washed with water (20
mL), brine (20 mL), dried over anhydrous MgSO41, and concentrated
to dryness to obtain 8.9 g (95% yield) of a solid. H NMR (CDCl3,
500 MHz) δ: 7.40−7.20 (multiple overlapping peaks, 19H), 2.40 (s,
3H), 1.76 (s, 3H). MS: m/z 448.1 [M + H]+.
(5R)-5-(3-Bromomethyphenyl)-5-methyl-N-trityl-2,4-oxazoli-
dinedione (27a, Y = H). To a solution of (R)-5-methyl-5-(3-
methyphenyl)-N-trityloxazolidinedione (26a, 3.0 g, 6.7 mmol) in
carbon tetrachloride (100 mL) were added N-bromosuccinimide (1.1
g, 6.7 mmol) and AIBN (catalytic). The mixture was heated to 80 °C
and refluxed overnight. The solution was cooled to room temperature,
washed with saturated NaHCO3 solution (20 mL), water (20 mL),
and brine (20 mL), and concentrated to dryness. The residue was
purified by silica gel column chromatography with hexane/ethyl
acetate (9:1) as a solvent to obtain 1.35 g (38% yield) of a white solid.
1H NMR (CDCl3, 500 MHz) δ: 7.5−7.2 (multiple overlapping peaks,
19H), 4.56 (s, 2H), 1.76 (s, 3H). LC−MS (ESI): >95% purity at λ 220
and 254 nm. MS: m/z 526.1 [M + H]+. Enantiomeric purity: >96% ee.
For analysis of enantiomeric purities of the products, a 10 μL
sample solution of approximately 1.0 mg/mL in concentration was
injected onto a Chiralcel OD analytical column (4.6 mm × 50 mm, 10
μm). Products were then eluted with an isocratic solvent system
consisting of 10% ethanol in heptane at a flow rate of 0.75 mL/min.
Peaks were recorded at a wavelength of 220 μm with an UV detector.
Under these conditions, the retention time of the S enantiomer is 6.96
min while the retention time of the R enantiomer is 8.98 min.
Enantiomeric excess (% ee) is calculated as area under curve of the S
enantiomer minus area under curve of the R enantiomer and divided
by the sum of the two areas.
(2R)-2-[2-Chloro-5-[3-(4-tert-butylphenyl)-6-trifluoromethyl-
benzimidazol-1-yl]methyl]phenoxy]acetic Acid (30). 1H NMR
(CDCl3, 500 MHz) δ: 7.62 (m, 2H), 7.61 (s, 1H), 7.52 (d, 2H), 7.39
(m, 2H), 7.17 (d, 1H), 7.12 (s, 1H), 6.91 (d, 1H), 5.15 (dd, 2H), 4.77
(dd, 1H), 1.49 (d, 3H), 1.35 (s, 9H). LC−MS (ESI): >95% purity at λ
220 and 254 nm. MS: m/z 547.1 [M + H]+.
(2R)-2-[2-Chloro-5-[3-(6-chlorobenzothiozol-2-yl)-6-trifluor-
omethylbenzimidazol-1-yl]methyl]phenoxy]acetic Acid
(31). 1H NMR (CDCl3, 500 MHz) δ: 8.89 (d, 1H), 7.94 (d, 1H),
7.91 (s, 1H), 7.59 (d, 1H), 7.50 (dd, 1H), 7.40 (d, 1H), 7.23 (s, 1H),
7.01 (d, 1H), 6.91 (s, 1H), 5.12 (dd, 2H), 4.74 (dd, 1H), 1.67 (d, 3H).
LC−MS (ESI): >95% purity at λ 220 and 254 nm. MS: m/z 582.0 [M
+ H]+.
(2R)-2-[2-Chloro-5-[3-(6-chlorobenzisoxazol-3-yl)-6-trifluor-
omethylbenzimidazol-1-yl]methyl]phenoxy]acetic Acid
(32). 1H NMR (CDCl3, 500 MHz) δ: 8.25 (d, 1H), 8.15 (s, 1H),
7.88 (d, 1H), 7.72 (s, 1H), 7.56 (d, 1H), 7.54 (d, 1H), 7.42 (d, 1H),
7.16 (s, 1H), 7.06 (d, 1H), 5.12 (s, 2H), 4.93 (dd, 1H), 1.50 (d, 3H).
LC−MS (ESI): >95% purity at λ 220 and 254 nm. MS: m/z 566.1 [M
+ H]+.
(2R)-2-[2-Chloro-5-[3-(7-methylbenzisoxazol-3-yl)-6-trifluor-
omethylbenzimidazol-1-yl]methyl]phenoxy]acetic Acid
(33). 1H NMR (CDCl3, 500 MHz) δ: 8.09 (d, 1H), 7.90 (d, 1H),
7.58 (d, 1H), 7.43 (d, 1H), 7.39 (d, 1H), 7.25 (m, 2H), 6.98 (m, 2H),
5.19 (s, 2H), 4.78 (dd, 1H), 2.62 (s, 3H), 1.62 (d, 3H). LC−MS
(ESI): >95% purity at λ 220 and 254 nm. MS: m/z 546.0 [M + H]+.
(2R)-2-[2-Chloro-5-[3-(5-methylbenzisoxazol-3-yl)-6-trifluor-
omethylbenzimidazol-1-yl]methyl]phenoxy]acetic Acid
(34). 1H NMR (CDCl3, 500 MHz) δ: 8.01 (d, 1H), 7.96 (d, 1H),
7.79 (d, 1H), 7.51 (m, 2H), 7.45 (t, 2H), 7.10 (d, 1H), 7.03 (d, 1H),
5.19 (dd, 2H), 4.80 (dd, 1H), 2.56 (s, 3H), 1.66 (d, 3H). LC−MS
(ESI): >95% purity at λ 220 and 254 nm. MS: m/z 546.0 [M + H]+.
(2R)-2-[2-Chloro-5-[3-(6-methylbenzisoxazol-3-yl)-6-trifluor-
omethylbenzimidazol-1-yl]methyl]phenoxy]acetic Acid
(35). 1H NMR (CDCl3, 500 MHz) δ: 8.01 (m, 2H), 7.72 (m, 2H),
7.20 (m, 2H), 6.98 (m, 3H), 5.14 (s, 2H), 4.82 (m, 1H), 2.58 (s, 3H),
1.72 (d, 3H). LC−MS (ESI): >95% purity at λ 220 and 254 nm. MS:
m/z 546.0 [M + H]+.
(2R)-2-[2-Chloro-5-[3-(5-chlorobenzisoxazol-3-yl)-6-trifluor-
omethylbenzimidazol-1-yl]methyl]phenoxy]acetic Acid
(36). 1H NMR (CDCl3, 500 MHz) δ: 8.30 (d, 1H), 8.14 (s, 1H),
7.99 (s, 1H), 7.60 (d, 1H), 7.54 (d, 1H), 7.45 (d, 1H), 7.39 (d, 1H),
7.16 (s, 1H), 7.02 (d, 1H), 5.17 (dd, 2H), 4.91 (m, 1H), 1.50 (d, 3H).
LC−MS (ESI): >95% purity at λ 220 and 254 nm. MS: m/z 566.1 [M
+ H]+.
(2R)- and (2S)-2-[2-Chloro-5-[3-(6-methoxybenzisoxazol-3-
yl)-6-trifluoromethylbenzimidazol-1-yl]methyl]phenoxy]acetic
Acid (37 and 41). 1H NMR (CDCl3, 500 MHz) δ: 8.07 (d, 1H),
7.95 (d, 1H), 7.50 (d, 1H), 7.42 (d, 1H), 7.28 (s, 1H), 7.01−7.08 (m,
4H), 5.15 (dd, 2H), 4.79 (dd, 1H), 3.96 (s, 3H), 1.71 (d, 3H). LC−
MS (ESI): >95% purity at λ 220 and 254 nm. MS: m/z 562.0 [M +
H]+.
General Procedures for Assembly of Most of the Final
Target Compounds. To a solution of intermediate 9 (0.1 mmol) in
DMF (1 mL) were added the benzyl bromide intermediate such as 18
or 27 (0.1 mmol) and Cs2CO3 (0.2 mmol). The mixture was stirred at
room temperature overnight, diluted with water (1.0 mL), and
extracted with ethyl acetate (2 × 1.0 mL). The organic extracts were
combined, dried over anhydrous MgSO4, and concentrated to dryness
to obtain a solid that was hydrolyzed either by NaOH in MeOH (in
the case of 18) or by neat trifluoroacetic acid (in case of 27) followed
by preparative HPLC purification to obtain final products. The
following benzimidazolone analogues were prepared by this procedure
unless described separately.
(2R)-2-[2-Chloro-5-[3-(4-chlorobenzyl)-6-trifluoromethyl-
benzimidazol-1-yl]methyl]phenoxy]acetic Acid (28). 1H NMR
(CDCl3, 500 MHz) δ: 7.28−7.34 (m, 5H), 7.13 (s, 1H), 6.99 (d, 1H),
6.94 (d, 1H), 6.88 (s, 1H), 6.85 (dd, 1H), 5.12 (dd, 2H), 5.06 (s, 2H),
4.78 (dd, 1H), 1.65 (d, 3H). LC−MS (ESI): >95% purity at λ 220 and
254 nm. MS: m/z 539.1 [M + H]+.
(2R)-2-[2-Chloro-5-[3-(6-methoxybenzisoxazol-3-yl)-6-tri-
fluoromethoxybenzimidazol-1-yl]methyl]phenoxy]acetic Acid
(38). 1H NMR (CDCl3, 500 MHz) δ: 8.13 (d, 1H), 7.76 (s, 1H), 7.32
(d, 1H), 7.23 (s, 3H), 7.10 (s, 1H), 7.03 (t, 2H), 5.27 (dd, 2H), 4.45
(m, 1H), 3.95 (s, 3H), 1.57 (d, 3H). LC−MS (ESI): >95% purity at λ
220 and 254 nm. MS: m/z 578.0 [M + H]+.
(2R)-2-[2-Chloro-5-[3-(6-methoxybenzisoxazol-3-yl)-5-tri-
fluoromethylbenzimidazol-1-yl]methyl]phenoxy]acetic Acid
(39). 1H NMR (CDCl3, 500 MHz) δ: 8.09 (d, 1H), 7.85 (s, 1H),
7.47 (d, 1H), 7.38 (m, 3H), 7.26 (s, 1H), 7.5 (t, 2H), 5.16 (dd, 2H),
4.81 (m, 1H), 3.96 (s, 3H), 1.68 (d, 3H). LC−MS (ESI): >95% purity
at λ 220 and 254 nm. MS: m/z 562.0 [M + H]+.
(2R)-2-[2-Chloro-5-[3-(6-methoxybenzisoxazol-3-yl)-5-tri-
fluoromethoxybenzimidazol-1-yl]methyl]phenoxy]acetic Acid
(40). 1H NMR (CDCl3, 500 MHz) δ: 8.13 (d, 1H), 7.75 (s, 1H), 7.37
(d, 1H), 7.13 (m, 3H), 7.06 (s, 1H), 7.01 (t, 2H), 5.16 (dd, 2H), 4.82
(m, 1H), 3.94 (s, 3H), 1.61 (d, 3H). LC−MS (ESI): >95% purity at λ
220 and 254 nm. MS: m/z 578.0 [M + H]+.
(2R)-2-[2-Chloro-5-[3-(4-chlorobenzoyl)-6-trifluoromethyl-
benzimidazol-1-yl]methyl]phenoxy]acetic Acid (29). 1H NMR
(CDCl3, 500 MHz) δ: 8.04 (d, 1H), 7.84 (dd, 2H), 7.57 (dd, 2H),
7.51 (m, 1H), 7.37 (m, 2H), 6.97 (m, 2H), 5.05 (dd, 2H), 4.97 (dd,
1H), 1.61 (d, 3H). LC−MS (ESI): >95% purity at λ 220 and 254 nm.
MS: m/z 553.0 [M + H]+.
(2R)-2-[4-Chloro-3-[3-(6-methoxybenzisoxazol-3-yl)-6-tri-
fluoromethylbenzimidazol-1-yl]methyl]phenoxy]acetic Acid
8551
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
(42). 1H NMR (CDCl3, 500 MHz) δ: 8.26 (d, 1H), 7.97 (d, 1H), 7.82
(s, 1H), 7.51 (d, 1H), 7.44 (d, 1H), 7.40 (s, 1H), 7.38 (m, 1H), 6.87
(m, 2H), 5.31 (s, 2H), 4.73 (dd, 1H), 3.94 (s, 3H), 1.49 (d, 3H). LC−
MS (ESI): >95% purity at λ 220 and 254 nm. MS: m/z 562.0 [M +
H]+.
(2R)-2-[3-[3-(6-Methoxybenzisoxazol-3-yl)-6-trifluorome-
thylbenzimidazol-1-yl]methyl]phenoxy]acetic Acid (43). 1H
NMR (CDCl3, 500 MHz) δ: 8.25 (d, 1H), 8.15 (s, 1H), 7.89 (d,
1H), 7.70 (s, 1H), 7.56 (t, 2H), 7.27 (t, 1H), 7.04 (d, 1H), 6.96 (s,
1H), 6.78 (d, 1H), 5.23 (s, 2H), 4.80 (dd, 1H), 3.98 (s, 3H), 1.46 (d,
3H). LC−MS (ESI): >95% purity at λ 220 and 254 nm. MS: m/z
528.1 [M + H]+.
(2R)-2-[3-[3-(6-Methoxybenzisoxazol-3-yl)-6-trifluorome-
thylbenzimidazol-1-yl]methyl]phenoxy]propanoic Acid
(44). 1H NMR (CDCl3, 500 MHz) δ: 8.25 (d, 1H), 8.15 (s, 1H),
7.88 (d, 1H), 7.72 (s, 1H), 7.55 (t, 2H), 7.26 (t, 1H), 7.01 (s, 2H),
6.79 (d, 1H), 5.22 (s, 2H), 4.63 (m, 1H), 3.92 (s, 3H), 1.85 (m, 2H),
0.95 (t, 3H). LC−MS (ESI): >95% purity at λ 220 and 254 nm. MS:
m/z 541.1 [M + H]+.
(5R)-5-(2-{[3-(6-Methoxy-1,2-benzisoxazol-3-yl)-6-trifluoro-
methylbenzimidazol-1-yl]methyl}phenyl)-5-methyloxazolidi-
nedione (45). 1H NMR (CDCl3, 500 MHz) δ: 8.12 (d, 1H), 8.01 (d,
1H), 7.62 (s, 1H), 7.50 (s, 1H), 7.40−7.49 (m, 3H), 7.23 (s, 1H), 7.10
(s, 1H), 7.05 (d, 1H), 5.20 (s, 2H), 3.96 (s, 3H), 1.98 (s, 3H). LC−
MS (ESI): >95% purity at λ 220 and 254 nm. MS: m/z 553.1 [M +
H]+.
(5R)-5-(4-{[3-(6-Methoxy-1,2-benzisoxazol-3-yl)-6-trifluoro-
methylbenzimidazol-1-yl]methyl}phenyl)-5-methyloxazolidi-
nedione (46). 1H NMR (CDCl3, 500 MHz) δ: 8.19 (d, 1H), 7.98 (d,
1H), 7.65 (d, 2H), 7.52 (d, 2H), 7.13 (d, 1H), 7.11 (s, 1H), 7.02 (d,
1H), 6.83 (s, 1H), 5.20 (s, 2H), 3.98 (s, 3H), 1.99 (s, 3H). LC−MS
(ESI): >95% purity at λ 220 and 254 nm. MS: m/z 553.1 [M + H]+.
(5R)-5-(3-{[3-(6-Methoxy-1,2-benzisoxazol-3-yl)-6-trifluoro-
methylbenzimidazol-1-yl]methyl}phenyl)-5-methyloxazolidi-
nedione (47). 1H NMR (CDCl3, 500 MHz) δ: 8.14 (d, 1H), 8.01 (d,
1H), 7.72 (s, 1H), 7.58 (s, 1H), 7.40−7.49 (m, 3H), 7.21 (s, 1H), 7.07
(s, 1H), 7.03 (d, 1H), 5.25 (s, 2H), 3.96 (s, 3H), 1.96 (s, 3H). LC−
MS (ESI): >95% purity at λ 220 and 254 nm. MS: m/z 553.1 [M +
H]+.
(5R)-5-(3-{[3-(6-Methoxy-1,2-benzisoxazol-3-yl)-6-trifluoro-
methylbenzimidazol-1-yl]}phenyl)-5-methyloxazolidinedione
(48). 1H NMR (CDCl3, 500 MHz) δ: 8.18 (d, 1H), 8.05 (d, 1H), 7.70
(s, 1H), 7.62 (s, 1H), 7.38−7.52 (m, 3H), 7.25 (s, 1H), 7.17 (s, 1H),
7.05 (d, 1H), 3.98 (s, 3H), 1.98 (s, 3H). LC−MS (ESI): >95% purity
at λ 220 and 254 nm. MS: m/z 539.0 [M + H]+.
1
as solvent system to obtain 358 mg (78% yield) of a white solid. H
NMR (500 MHz, CDCl3) δ 8.18 (d, 1H), 7.92 (d, 1H), 7.71 (s, 1H),
7.56 (brs, 1H), 7.43 (brs, 2H), 7.23 (m, 2H), 7.08 (s, 1H), 7.02 (m,
2H), 5.28 (dd, 2H), 3.98 (s, 3H), 1.99 (s, 3H). LC−MS (ESI): >95%
purity at λ 220 and 254 nm. MS: m/z 485.1 [M + H]+. Enantiomeric
purity: >96% ee. Enantiomeric purities of the product were
determined by chiral HPLC on a Chiralcel OD analytical column
(4.6 mm × 50 mm, 10 μm). The product was eluted with an isocratic
solvent system consisting of 10% ethanol in heptane at a flow rate of
0.75 mL/min. Under these conditions, the retention time of the S
enantiomer is approximately 14 min while the retention time of the R
enantiomer is about 17 min. Enantiomeric purity for other analogues
should be similar, since their chiralities were derived from the same
intermediate (27a).
(5R)-5-(3-{[3-(6-Chlorobenzisoxazol-3-yl)benzimidazol-1-yl]-
methyl}phenyl)-5-methyloxazolidinedione (52). 1H NMR
(CDCl3, 500 MHz) δ: 8.39 (s, 1H), 8.33 (brs, 1H), 7.92 (d, 1H),
7.71 (s, 1H), 7.61 (s, 2H), 7.58 (d, 1H), 7.42 (m, 2H), 7.21 (d, 2H),
7.02 (d, 1H), 5.23 (dd, 2H), 1.99 (s, 3H). LC−MS (ESI): >95% purity
at λ 220 and 254 nm. MS: m/z 489.0 [M + H]+.
(2R)-2-[4-Chloro-3-[3-(6-methoxybenzisoxazol-3-yl)-
benzimidazol-1-yl]methyl]phenoxy]acetic Acid (53)). 1H NMR
(CDCl3, 500 MHz) δ: 8.12 (d, 1H), 7.98 (dd, 1H), 7.37 (d, 1H), 7.15
(m, 2H), 7.12 (d, 1H), 7.05 (dd, 2H), 6.78 (dd, 1H), 6.87 (s, 1H),
5.29 (d, 2H), 4.64 (dd, 1H), 3.97 (s, 3H), 1.59 (d, 3H). LC−MS
(ESI): >95% purity at λ 220 and 254 nm. MS: m/z 494.1 [M + H]+.
(5R)-5-(3-{[3-(6-Methoxy-1,2-benzisoxazol-3-yl)-
benzimidazol-1-yl]methyl}-4-chlorophenyl)-5-methyloxazoli-
dinedione (54). 1H NMR (CDCl3, 500 MHz) δ: 8.18 (d, 1H), 7.93
(d, 1H), 7.60 (s, 1H), 7.58 (s, 1H), 7.23 (m, 2H), 7.08 (m, 4H), 5.26
(s, 2H), 3.94 (s, 3H), 1.86 (s, 3H). LC−MS (ESI): >95% purity at λ
220 and 254 nm. MS: m/z 519.1 [M + H]+.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional experimental procedures for synthesis of 4−7 and
11−18, elemental analysis results for 51 and 52, structures and
spectra data of other benzimidaxolones analogues, and chiral
HPLC traces of selected compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
Accession Codes
†Protein Data Bank (PDB) access code for X-ray crystallo-
graphic data of compound 51: 3TY0.
(5R)-5-(3-{[3-(6-Methoxy-1,2-benzisoxazol-3-yl)-6-trifluoro-
methylbenzimidazol-1-yl]methyl}-4-chlorophenyl)-5-methyl-
oxazolidinedione (49). 1H NMR (CDCl3, 500 MHz) δ: 8.14 (d,
1H), 8.02 (d, 1H), 7.67 (s, 1H), 7.57 (s, 1H), 7.52 (d, 1H), 7.26 (m,
2H), 7.09 (s, 1H), 7.06 (d, 1H), 5.38 (s, 2H), 3.98 (s, 3H), 1.83 (s,
3H). LC−MS (ESI): >95% purity at λ 220 and 254 nm. MS: m/z
587.0 [M + H]+.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: 732-594-2782. Fax: 732-594-9556. E-mail: weiguo_
liu@merck.com.
(2R)-2-[2-Chloro-5-[3-(6-methoxybenzisoxazol-3-yl)-
benzimidazol-1-yl]methyl]phenoxy]acetic Acid (50). 1H NMR
(CDCl3, 500 MHz) δ: 8.13 (d, 1H), 7.88 (dd, 1H), 7.37 (d, 1H), 7.25
(m, 2H), 7.08 (d, 1H), 7.02 (dd, 2H), 6.98 (dd, 1H), 6.77 (s, 1H),
5.29 (d, 2H), 4.64 (dd, 1H), 3.97 (s, 3H), 1.59 (d, 3H). LC−MS
(ESI): >95% purity at λ 220 and 254 nm. MS: m/z 494.1 [M + H]+.
(5R)-5-(3-{[3-(6-Methoxybenzisoxazol-3-yl)benzimidazol-1-
yl]methyl}phenyl)-5-methyloxazolidinedione (51). To a solu-
tion of 1-[3-(6-methoxy)-benzisoxazoyl]benzimidazole-2-one (9j, 266
mg, 0.95 mmol) in DMF (3 mL) were added (R)-5-methyl-5-(3-
bromomethylphenyl)-N-trityloxazolidinedione (27a, 500 mg, 0.95
mmol) and Cs2CO3 (620 mg, 1.9 mmol). The mixture was stirred
at room temperature overnight, diluted with water (10 mL), and
extracted with ethyl acetate (2 × 10 mL). The organic extracts were
combined, dried over anhydrous MgSO4, and concentrated to obtain a
solid. The solid was dissolved in neat TFA (5 mL), and the mixture
was stirred at room temperature for 6 h. The mixture was then
concentrated to dryness under vacuum and purified by silica gel
column chromatography with hexane/ethyl acetate/TFA (3/1/0.01)
ACKNOWLEDGMENTS
■
We gratefully acknowledge the technical assistance from
Neelam Sharma and Roger Meurer. We also thank Dr. Stephen
Soisson for depositing the X-ray coordinates and Dr. Gino
Salituro for PK studies.
ABBREVIATIONS USED
■
T2DM, type 2 diabetes mellitus; PPAR, peroxisome prolifer-
ator-activated receptor; SPPARγM, selective peroxisome
proliferator-activated receptor γ modulator; CYP, cytochrome
P450; TZD, thiazolidinedione; PV, plasma volume; mpk, mg/
kg
REFERENCES
■
(1) Albrecht, S. S.; Kuklina, E. V.; Bansil, P.; Jamieson, D. J.;
Whiteman, M. K.; Kourtis, A. P.; Posner, S. F.; Callaghan, W.
8552
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
Diabetes: trends among delivery hospitalizations in the U.S., 1994−
2004. Diabetes Care 2010, 33, 768−773.
(13) Debenham, S. D.; Chan, A.; Lau, F. W.; Liu, W.; Wood, H. B.;
Lemme, K.; Colwell, L.; Habulihaz, B.; Akiyama, T. E.; Einstein, M.;
Doebber, T. W.; Sharma, N.; Wang, C. F.; Wu, M.; Berger, J. P.;
Meinke, P. T. Highly functionalized 7-azaindoles as selective PPARγ
modulators. Bioorg. Med. Chem. Lett. 2008, 18, 4798−4801.
(14) Geffken, D. Cyclization of salicylamides and salicylohydroxamic
acids with 1,1′-carbonyldiimidazole. Liebigs Ann. Chem. 1981, 1513−
1514.
(15) McConville, M.; Blacker, J.; Xiao, J. Heck reaction in diols and
cascade formation of cyclic ketals. Synthesis 2010, 349−360.
(16) Wang, Z.-M.; Sharpless, K B. A solid to solid asymmetric
dihydroxylation procedure for kilogram-scale preparation of
enantiopure hydrobenzoin. J. Org. Chem. 1994, 59, 8302−8303.
(17) For the binding assay procedure, see the following: 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-
propylbenzisoxazoles; human PPARγ, δ and α agonists. Bioorg. Med.
Chem. Lett. 2003, 13, 931−936.
(18) Berger, J. B.; Bailey, P.; Biswas, C.; Cillinana, C. A.; Doebber, T.
W.; Hayes, N. S.; Saperstein, R.; Smith, R. G.; Liebowitz, M. D.
Thiazolidinediones produce a conformational change in peroxisomal
proliferator-activated receptor-gamma: binding and activation correlate
with antidiabetic actions in db/db mice. Endocrinology 1996, 137,
4189−4195.
(19) Gani, O. A. B. S. M.; Sylte, I. Ligand-induced stabilization and
activation of peroxisome proliferator-activated receptor γ. Chem. Biol.
Drug Des. 2008, 72, 50−57.
(20) Einstein, M.; Akiyama, T. E.; Castriota, G. A.; Wang, C. F.;
McKeever, B.; Mosley, R. T.; Becker, J. W.; Moller, D. E.; Meinke, P.
T.; Wood, H. B.; Berger, J. P. The differential interactions of
peroxisome proliferator-activator receptor γ ligands with Tyr473 is a
physical basis for their unique biological activities. Mol. Pharmacol.
2008, 73, 62−74.
(21) Lamotte, Y.; Martres, P.; Faucher, N.; Laroze, A.; Grillot, D.;
Ancellin, N.; Saintillan, Y.; Beneton, V.; Gampe, R. T. Jr. Synthesis and
biological activities of novel indole derivatives as potent and selective
PPARγ modulators. Bioorg. Med. Chem. Lett. 2010, 20, 1399−1404.
(22) 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. Ligand binding and co-activator assembly of the
peroxisome proliferator-activated receptor-gamma. Nature 1998, 395,
137−143.
(23) Sheu, S. H.; Kaya, T.; Waxman, D. J.; Vajda, S. Exploring the
binding site structure of the PPAR gamma ligand-binding domain by
computational solvent mapping. Biochemistry 2005, 44, 1193−1209.
(24) Schupp, M; Clemenz, M.; Gineste, R.; Witt, H.; Janke, J.;
Helleboid, S.; Hennuyer, N.; Ruiz, P.; Unger, T.; Staels, B.; Kintscher,
U. Molecular characterization of new selective peroxisome proliferator-
activated receptor gamma modulators with angiotensin receptor
blocking activity. Diabetes 2005, 54, 3442−3452.
(25) Thieme, T. M.; Steri, R.; Proschak, E.; Paulke, A.; Schneider, G.;
Schubert-Zsilavecz, M. Rational design of a pirinixic acid derivative
that acts as subtype-selective PPAR.γ modulator. Bioorg. Med. Chem.
Lett. 2010, 20, 2469−2473.
(2) Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B. R. The
PPARs: from orphan receptor to drug discovery. J. Med. Chem. 2000,
43, 527−550.
(3) Kersten, S.; Desvergne, B.; Wahli, W. Roles of PPARs in health
and disease. Nature 2000, 405, 421−424.
(4) Nuclear Receptors Nomenclature Committee, 1999.. A unified
nomenclature system for the nuclear receptor superfamily. Cell 1999,
97, 161−163.
(5) Berger, J.; Bailey, P.; Biswas, C.; Cullinan, C. A.; Doebber, T. W.;
Hayes, N. S.; Saperstein, R.; Smith, R. G.; Leibowitz, M. D.
Thiazolidinediones produce a conformational change in peroxisomal
proliferator-activated receptor-γ: binding and activation correlates with
antidiabetic actions in db/db mice. Endocrinology 1996, 137, 4189−
4195.
(6) Oakes, N. D.; Kennedy, C. J.; Jenkins, A. B.; Laybutt, D. R.;
Chisholm, D. J.; Kraegen, E. W. A new anti-diabetic agent, BRL 49653,
reduces lipid availability and improves insulin action and glucose
regulation in the rat. Diabetes 1994, 43, 1203−1210.
(7) Nesto, R. W.; Bell, D.; Bonow, R. O.; Fonseca, V.; Grundy, S. M.;
Horton, E. S.; Winter, M. L.; Porte, D.; Semenkovich, C. F.; Smith, S.;
Young, L. H.; Kahn, R. Thiazolidinedione use, fluid retention, and
congestive heart failure: a consensus statement from the American
Heart Association and American Diabetes Association. Diabetes Care
2004, 27, 256−263.
(8) Kahn, S. E.; Zinman, B.; Lachin, J. M.; Haffner, S. M.; Herman,
W. H.; Holman, R. R.; Kravitz, B. G.; Yu, D.; Heise, M. A.; Aftring, R.
P.; Viberti, G. Rosiglitazone-associated fractures in type 2 diabetes: an
analysis from A Diabetes Outcome Progression Trial (ADOPT).
Diabetes Care 2008, 31, 845−851.
(9) Burgermeister, E.; Schnoebelen, A.; Flament, A.; Benz, J.; Stihle,
M.; Gsell, B.; Rufer, A.; Kuhn, B.; Marki, H. P. A novel partial agonist
of peroxisome proliferator-activated receptor-γ (PPARγ) recruits
PPARγ-coactivator-1α, prevents triglyceride accumulation, and
potentiates insulin signaling in vitro. Mol. Endocrinol. 2006, 20,
809−830.
(10) (a) Berger, J. P.; Petro, A. E.; McNaul, K. L.; Kelly, L. J.; Zhang,
B. B.; Richards, K.; Elbrecht, A.; Johnson, B. A.; Zhou, G.; Doebber, T.
W.; Biswas, C.; Parikh, M.; Sharma, N.; Tanen, M. R.; Thompson, G.
M.; Ventre, J.; Adams, A. D.; Mosley, R.; Surwit, R. S.; Moller, D. E.
Distinct properties and advantages of a novel peroxisome proliferator-
activated protein-γ selective modulator. Mol. Endocrinol. 2003, 17,
662−676. (b) Meinke, P. T.; Wood, H. B.; Szewczyk, J. W. Nuclear
hormone receptor modulators for the treatment of diabetes and
dyslipidemia. Annu. Rep. Med. Chem. 2006, 41, 99−126.
(11) Elbrecht, A.; Chen, Y.; Adams, A; Berger, J. P.; Griffin, P.; Klatt,
T.; Zhang, B. B.; Menke, J.; Zhou, G.; Smith, R. G.; Moller, D. E. L-
764406 is a partial agonist of human peroxisome proliferator-activated
receptor-γ. The role of CYS313 in ligand binding. J. Biol. Chem. 1999,
274, 7913−7922.
(12) (a) Acton, J. J.; Black, R. M.; Jones, A. B.; Moller, D. E.; Colwell,
L.; Doebber, T. W.; MacNaul, K. L.; Berger, J.; Wood, H. B. Benzoyl
2-methyl indoles as selective PPARγ modulators. Bioorg. Med. Chem.
Lett. 2005, 15, 357−362. (b) Liu, K.; Black, R. M.; Acton, J. A. III;
Mosley, R.; Debenham, S.; Abola, R.; Yang, M; Tschirret-Guth, R.;
Colwell, L.; Liu, C.; Wu, M.; Wang, C. F.; MacNaul, K. L.; McCann,
M. E.; Moller, D. E.; Berger, J. P.; Meinke, P. T.; Jones, A. B.; Wood,
H. B. Selective PPARγ modulators with improved pharmacological
profiles. Bioorg. Med. Chem. Lett. 2005, 15, 2437−2440. (c) Liu, W.;
Liu, K.; Wood, H. B.; Mccann, M. E.; Doebber, T. W.; Chang, C. H.;
Akiyama, T. E.; Einstein, M.; Berger, J. P.; Meinke, P. T. Discovery of a
peroxisome proliferator activated receptor γ (PPARγ) modulator with
balanced PPARα activity for the treatment of type 2 diabetes and
dyslipidemia. J. Med. Chem. 2009, 52, 4443−4453. (d) Acton, J. J. III;
Akiyama, T. E.; Chang, C. H.; Colwell, L.; Debenham, S.; Doebber, T.;
Einstein, M.; Liu, K.; McCann, M. E.; Moller, D. E.; Muise, E. S.; Tan,
Y.; Thompson, J. R.; Wong, K. K.; Wu, M.; Xu, L.; Meinke, P. T.;
Berger, J. P.; Wood, H. B. J. Med. Chem. 2009, 52, 3846−3854.
(26) Furukawa, A.; Arita, T; Satoh, S.; Wakabayashi, K.; Hayashi, S.;
Matsui, Y.; Araki, K.; Kuroha, M.; Ohsumi, J. Discovery of a novel
selective PPARγ modulator from (−)-cercosporamide derivatives.
Bioorg. Med. Chem. Lett. 2010, 20, 2095−2098.
(27) McKenna, N. J.; O’Malley, B. W. Minireview: nuclear receptor
coactivatorsan update. Endocrinology 2002, 143, 2461−2465.
(28) Dreijerink, K. M. A.; Varier, R. A.; van Beekum, O.; Jeninga, E.
H.; Hoeppener, J. W. M.; Lips, C. J. M.; Kummer, J. A.; Kalkhoven, E.;
Timmers, H. T. M. The multiple endocrine neoplasia type 1 (MEN1)
tumor suppressor regulates peroxisome proliferator-activated receptor
γ-dependent adipocyte differentiation. Mol. Cell. Biol. 2009, 29, 5060−
5069.
(29) Pochetti, G.; Mitro, N.; Lavecchia, A.; Gilardi, F.; Besker, N.;
Scotti, E.; Aschi, M.; Re, N.; Fracchiolla, G.; Laghezza, A.; Tortorella,
P.; Montanari, R.; Novellino, E.; Mazza, F.; Crestani, M.; Loiodice, F.
8553
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554

Journal of Medicinal Chemistry
Article
Structural insight into peroxisome proliferator-activated receptor γ
binding of two ureidofibrate-like enantiomers by molecular dynamics,
cofactor interaction analysis, and site-directed mutagenesis. J. Med.
Chem. 2010, 53, 4354−4366.
(30) Zhou, G.; Cummings, R.; Hermes, J.; Moller, D. E. Use of
homogeneous time-resolved fluorescence energy transfer in the
measurement of nuclear receptor activation. Methods 2001, 26, 54−61.
(31) Torchia, J.; Rose, D. W.; Inostroza, J.; Kamei, Y.; Westin, S.;
Glass, C. K.; Rosenfeld, M. G. The transcriptional co-activator p/CIP
binds CBP and mediates nuclear-receptor function. Nature 1997, 387,
677−684.
(32) Sadana, P.; Park, E. A. Characterization of the transactivation
domain in the peroxisome-proliferator-activated receptor γ co-activator
(PGC-1). Biochem. J. 2007, 403, 511−518.
(33) Leo, C.; Chen, J. D. The SRC family of nuclear receptor
coactivators. Gene 2000, 245, 1−11.
(34) Zhu, Y.; Qi, C.; Jain, S.; Rao, M. S.; Reddy, J. K. Isolation and
characterization of PBP, a protein that interacts with peroxisome
proliferator-activated receptor. J. Biol. Chem. 1997, 272, 25500−25506.
(35) Li, Y.; Kovach, A.; Suino-Powell, K.; Martynowski, D.; Xu, H. E.
Structural and biochemical basis for the binding selectivity of
peroxisome proliferator-activated receptor γ to PGC-1.alpha. J. Biol.
Chem. 2008, 283, 19132−19139.
(36) Burgermeister, E.; Schnoebelen, A.; Flament, A.; Benz, J.; Stihle,
M.; Gsell, B.; Rifer, A.; Ruf, A.; Kuhn, B.; Marki, H. P.; Mizrahi, J.;
Sebokova, E.; Niesor, E.; Meyer, M. A novel partial agonist of
peroxisome proliferator activated receptor-γ (PPARγ) recruits PPARγ-
coactivator-1a, prevents triglyceride accumulation, and potentiates
insulin signaling in vitro. Mol. Endocrinol. 2006, 20, 809−830.
(37) Sorbera, L. A.; Leeson, P. A.; Martin, L.; Castaner, J. Farglitazar.
Antidiabetic PPARγ agonist. Drugs Future 2001, 26, 354−363.
(38) White, I. R.; Man, W. J.; Bryant, D.; Bugelski, P.; Camilleri, P.;
Cutler, P.; Hayes, W.; Holbrook, J. D.; Kramer, K.; Lord, P. G.; Wood,
J. Protein expression changes in the Sprague Dawley rat liver proteome
following administration of peroxisome proliferator activated receptor
alpha and gamma ligands. Proteomics 2003, 3, 505−512.
(39) 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 peroxisome proliferator-activated receptor (PPAR) γ and
PPARδ ligands produce distinct biological effects. J. Biol. Chem.
1999, 274, 6718−6725.
(40) Chang, C. H.; McNamara, L. A.; Wu, M. S.; Muise, E. S.; Tan,
Y.; Wood, H. B.; Meinke, P. T.; Thompson, J. R.; Doebber, T. W.;
Berger, J. P.; McCann, M. E. A novel selective peroxisome proliferator-
activator receptor-gamma modulator-SPPARγM5 improves insulin
sensitivity with diminished adverse cardiovascular effects. Eur. J.
Pharmacol. 2008, 584, 192−201.
8554
dx.doi.org/10.1021/jm201061j | J. Med. Chem. 2011, 54, 8541−8554