The effect of 1,3-diaryl-[1H]-pyrazole-4-acetamides on glucose utilization in ob/ob mice

Article abstract of DOI:10.1021/jm010032c

This article provides evidence of a new class of compounds, 1,3-diaryl-[1H]-pyrazole-4-acetamides, initially identified from their ability to increase glucose transport in an adipocyte and muscle cell line and ultimately demonstrating dramatic glucose lowering in ob/ob mice, a diabetic animal model. The lead compound, 1, possessed some behavioral-like effects which were removed by structural variation during the course of this investigation. Specifically, 11g (R1 = meta-CF₃, Ar2 = 4′biphenyl, R3 = diethylamide) illustrated the potency of this series with ED₅₀ values for glucose lowering in ob/ob mice of 3.0 mg/kg/day. Concomitant with its effect on glucose lowering, 11g also caused a 50% reduction in insulin levels consistent with an agent that increases whole body insulin sensitivity. 11g showed favorable pharmacokinetic data with acceptable absorption, negligible metabolism, and good duration of action. 11g demonstrated no appreciable adipogenic effect through PPARγ agonism, a characteristic of the thiazolidinediones (TZD), and so represents a potentially new class of agents for the treatment of diabetes.

Full text of DOI:10.1021/jm010032c

J . Med. Chem. 2001, 44, 2601-2611
2601
Th e Effect of 1,3-Dia r yl-[1H]-p yr a zole-4-a ceta m id es on Glu cose Utiliza tion in
ob/ob Mice
Gregory R. Bebernitz,*,† Greg Argentieri,| Beverly Battle,‡ Christine Brennan,‡ Bork Balkan,‡,§
Bryan F. Burkey,| Michele Eckhardt,| J iaping Gao,‡ Prasad Kapa,# Robert J . Strohschein,† Herbert F. Schuster,†
Mary Wilson, and David D. Xu#
Metabolic and Cardiovascular Diseases, Novartis Institute for Biomedical Research, Novartis Pharmaceuticals Corporation,
556 Morris Avenue, Summit, New J ersey 07901
Received J anuary 24, 2001
This article provides evidence of a new class of compounds, 1,3-diaryl-[1H]-pyrazole-4-
acetamides, initially identified from their ability to increase glucose transport in an adipocyte
and muscle cell line and ultimately demonstrating dramatic glucose lowering in ob/ob mice, a
diabetic animal model. The lead compound, 1, possessed some behavioral-like effects which
were removed by structural variation during the course of this investigation. Specifically, 11g
(R1 ) meta-CF3, Ar2 ) 4′biphenyl, R3 ) diethylamide) illustrated the potency of this series
with ED50 values for glucose lowering in ob/ob mice of 3.0 mg/kg/day. Concomitant with its
effect on glucose lowering, 11g also caused a 50% reduction in insulin levels consistent with
an agent that increases whole body insulin sensitivity. 11g showed favorable pharmacokinetic
data with acceptable absorption, negligible metabolism, and good duration of action. 11g
demonstrated no appreciable adipogenic effect through PPARγ agonism, a characteristic of
the thiazolidinediones (TZD), and so represents a potentially new class of agents for the
treatment of diabetes.
Diabetes mellitus may be categorized into two sub-
classes: type I, known as insulin dependent diabetes
mellitus (IDDM), and type II, noninsulin dependent
diabetes mellitus (NIDDM). IDDM accounts for about
10% of all diabetes and results from autoimmune-
mediated destruction of insulin-secreting â-cells of the
pancreas. In contrast, NIDDM is a chronic and progres-
sive metabolic disorder of carbohydrate and lipid me-
tabolism and accounts for the remaining 90% of diabetes
mellitus.1 It is the fourth leading cause of death in
developed countries and affects more than 5% of the
world’s population (and one in four people over the age
of 60). Fewer than half of all diabetics receive treatment
and of these only a very small proportion achieve a level
of glucose control that is sufficient to avoid the morbidity
associated with the disease, namely, macrovascular
(coronary artery disease, stroke) and microvascular
(retinopathy, neuropathy, nephropathy, and other mi-
croangiopathies) complications. In the 10-year diabetes
control and complication trial (DCCT)2 study, it was
found that tight control of blood glucose levels reduced
the incidence and progression of neuropathy, retinopa-
thy, and nephropathy, each by more than 50% in IDDM
patients. Because therapeutic improvements were con-
tinuously related to incremental reductions in glycemia
(toward normal levels), a new treatment goal of nor-
moglycemia has been established for both IDDM and
NIDDM patients. The initial therapy for newly diag-
nosed NIDDM patients has conventionally been diet and
exercise. Although this leads to a marginal improvement
in insulin sensitivity and a corresponding reduction in
hyperglycemia, this treatment usually fails within a few
months due to lack of compliance and further progres-
sion of the disease, and an oral antidiabetic agent is
prescribed. The standard regimen has consisted of
treatment with a member of the class of sulfonylurea
drugs that reduce glycemia by inducing the â-cell to
release more insulin. However, undesired consequences
of prolonged use of sulfonylureas include hypoglycemic
episodes, ultimate exhaustion of the â-cell, as well as
the long-term angiogenic side-effects that are a result
of chronic 24-h exposure to increased insulin levels.3-5
Agents that reduce the glycemia without these above
side effects are highly desirable and have been the aim
of many research groups over the past 20 years.
Recently, a new class of compounds, the thiazolidinedi-
ones (TZD), have reached the market whose effects on
glucose levels in the diabetic are manifested through
effects on insulin sensitivity in peripheral tissues with-
out directly effecting the output of insulin from the
pancreas. Although successful at reducing glucose levels
in diabetic patients, troglitazone, the leader in this class,
has been withdrawn from the market due to significant
and sometimes lethal liver toxicity in a small portion
of the population. Other TZDs, such as rosiglitazone and
pioglitazone, have been successful in the treatment of
diabetes, although there are still some concerns with
this class of drugs with respect to a potential for weight
gain, edema, and hepatotoxicity.6 Since the initial
discovery of the TZDs, renewed efforts have been
directed toward discovering other novel templates with
mechanisms that would provide similar or preferably
superior overall effects on insulin sensitivity and glucose
homeostasis without any accompanying toxicity.7
* Address correspondence to Gregory Bebernitz, phone 908-277-
7089, fax 908-277-2405, E-mail greg.bebernitz@pharma.novartis.com.
† Department of Medicinal Chemistry.
‡ Department of Pharmacology.
§ Current Address: IDD, 23 Business PK Dr., Branford, CT 06405.
| Department of Biochemistry.
#
Chemical Development.
10.1021/jm010032c CCC: $20.00 © 2001 American Chemical Society
Published on Web 07/06/2001

2602 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16
Bebernitz et al.
macokinetic properties including bioavailability, dura-
tion of action up to 8 h, and a general readout of overt
toxicities, an effect of major concern.
A number of pyrazole compounds have been sited in
the literature that elicit anti-hyperglycemic effects
including: 1,3-disubstituted pyrazoles,12 1,3,5-trisub-
stituted pyrazoles,13 and pyrazolones.14 Structurally
though, the compounds described herein have much
closer similarity to a body of work performed around
nonsteroidal antiinflammatory drugs (NSAID) as ex-
emplified by lonazolac15 (Figure 1). Amid concerns that
the anti-hyperglycemic activity of this series could be
related to their antiinflammatory activity, a number of
NSAIDs were evaluated in vitro in L6 myocytes for their
effects on glucose utilization and showed reduced rather
than increased glucose transport and utilization16 in this
cell line.
F igu r e 1. Diabetic agents (troglitazone and metformin), lead
pyrazole structure (1), Lonazolac.
Ch em istr y
In an attempt to identify compounds that would
increase peripheral glucose utilization, a high-through-
put adipocyte screen was used to evaluate our internal
compound collection as to their effect on glucose trans-
port. A compound of particular interest, SAH 57-749 (1)
(Figure 1), was shown to increase [3H] glucose incorpo-
ration into lipids in adipocytes in the presence of
submaximal concentrations of insulin, suggesting that
the compound might increase the sensitivity of the
adipocyte to insulin.8 Furthermore, this compound also
demonstrated an increase in glucose transport into L6
myocytes, a cultured muscle cell line. Administration
of 1 to ob/ob mice provided a significant reduction in
glucose levels over 3 days dosing without any indication
of toxicity (see experimental). However, when 1 was
administered to normal rats, it was observed to elicit
some behavioral-like side-effects. To further examine
this, 1 was evaluated both in a CNS receptor panel9 as
well as in a primary observation test (POT)10 prior to
consideration as a lead. No effects were observed in the
receptor panel at concentrations up to 10-6 M. However,
pronounced effects were demonstrated in the POT in
both rats and mice at doses of 10, 60, and 360 mg/kg,
the most notable effect being a reduction in motor
activity and locomotion. These effects eliminated 1 as
a potential lead for the glucose utilization project, and
the goal became to both separate the antihyperglycemic
activity of 1 from its undesired behavioral-like effects
as well as improve the potency of the compound as
compared to 1. Early on in this evaluation, it was found
that the magnitude of the response in the in vitro assays
did not always directly correlate with the magnitude of
the response obtained in the in vivo assay, an observa-
tion others have previously noted.11 For this reason,
compounds were evaluated based initially on lack of any
behavioral-like effects using an abbreviated POT in
normal rats followed then by evaluation of their poten-
tial for glucose lowering in a diabetic animal model, ob/
ob mice.
Variation at the 1- and 3-positions of the pyrazole
nucleus generally required an individual synthesis for
each desired template. Two somewhat overlapping
methods (methods A and B) were utilized to assemble
the pyrazole nucleus as shown in Scheme 1. For method
A, the synthesis commonly began with commercially
available ketones 2 and hydrazines 3 as a starting point.
Condensation between a ketone and hydrazine provided
a hydrazone 4 in high yields in most cases. The
Vilsmeier-Haack reaction using 2.5 equiv of reagent
performed a double addition of reagent to afford,
ultimately after hydrolysis, the desired cyclized alde-
hydes, 5 in 80-90% yield.17 Ortho-substituted phenyl-
hydrazines, however, were not applicable via this route
as the condensations either did not lead to product or
proceeded in poor yields. It was crucial in this reaction
to allow aqueous hydrolysis of the Vilsmeier adduct to
proceed to completion, or partial hydrolysis products
would complicate the workup and significantly reduce
the overall yield. Condensation of 5 with malonic acid
provided the unsaturated â-substituted acrylic acid 6
in high yield. Treatment of 6 with sulfur and ammonium
hydroxide in a pressure vessel under the Willgerodt-
Kindler reaction proceeded with loss of a carbon as a
result of these conditions to afford acetamide, 7. In
general the Willgerodt-Kindler reaction conditions
allow for carbonyl walk and in the specific case of aryl
alkyl ketones afford ultimately terminal carboxamides
without any loss of carbon atoms.18 However, in the case
of â-substituted acrylic acids such as 6, aryl acetamides
are efficiently produced.19 Acetamide 7 was hydrolyzed
under acidic conditions to provide the acid, 8. Compound
8 is then converted in two steps to the desired amide
products 10-12. Overall yields over seven steps ranged
from 47 to 90%. Although the reaction sequence is
somewhat long and linear with few common intermedi-
ates, each reaction step is worked up by simply quench-
ing into water and filtering off the product. Method B
allowed incorporation of the acetic acid side chain into
the template prior to the Vilsmeier-Haack reaction and
so shortened the reactions scheme by 2-3 synthetic
steps. Whereas the reaction sequence was streamlined
in comparison to method A, its usefulness was depend-
ent on the ease of purification of ester product, 14.
The various hydrazines and methyl ketones are all
commercially available with the exception of 2k which
The initial strategy comprised a traditional medicinal
chemistry approach involving variations in both the
electronic and steric properties of the appendages off
the central pyrazole nucleus of 1. This evaluation
protocol allowed an understanding of not only the
compound’s ability to lower glucose levels in diabetic
animals, but also provided an initial measure of phar-

1,3-Diaryl-[1H]-Pyrazole-4-Acetamides and Glucose Utilization
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16 2603
Sch em e 1. Synthetic Methods for the Preparation of Pyrazolesa
a
The methods utilized here for the preparation of the pyrazole products are general. Please refer to the Tables 1, 2, and 3 for the
specific substitution pattern of the various starting materials as well as the method used for preparation. Conditions: (a) AcOH, water;
(b) POCl3, DMF; (c) malonic acid, piperidine, pyridine, reflux; (d) sulfur, p-dioxane, NH4OH, bomb; (e) AcOH, H2SO4, water, reflux; (f)
SOCl2, DMF(cat), CH2Cl2; (g) R3-NH2 (or other nucleophile), CH2Cl2; (h) AlCl3, 1,1,2,2-tetrachloroethane (i) R3NH2, AlCl3, CH2Cl2.
was prepared under standard Friedel Crafts acylation
conditions. 13o was prepared from fluorene by exhaus-
tive alkylation using methyl iodide and KHMDS. In
addition, the extended esters 14j,o,s,t,u ,x, a n d y used
in method B were prepared by the direct Friedel Crafts
acylation of the corresponding aromatic precursor, 13,
under standard conditions using methyl 4-chloro-4-
oxobutyrate. Oxidation of the methylene position of the
fluorenyl side chain of 11n was carried out using
KMnO4 to afford 11r . Further reduction of the carbonyl
of 11r with NaBH4 and subsequent acylation with acetic
anhydride afforded 11p . Alternatively, addition of me-
thylmagnesium chloride to 11r provided 11q.
rate, and temperature. Although we expected the be-
havioral-like effects to eliminate a majority of the
compounds, we were surprised to find these effects in
only a few analogues. In short, the only compounds that
exhibited the behavioral-like side effects were various
amides of 1 and the phenol ester of 8. In general,
substitution of the aromatic of Ar2 with any substituent
resulted in elimination of the behavioral-like activity.20
Although glucose levels were not expected to change in
this normoglycemic rat model, they were also recorded
at 3 h postdose to verify that any noted behavioral
effects were not the result of significant aberrations in
glucose levels either hypoglycemic or hyperglycemic. No
statistically significant changes in glucose levels were
found for any of the compounds tested in this model
including the lead compound, 1, even considering the
high dose of compound administered (1 mmol/kg).
Compounds that exhibited no behavioral changes in
normal animals were then evaluated for their effects
on glucose lowering in ob/ob mice, a diabetic animal
model. This entailed dosing the animals once a day for
Resu lts a n d Discu ssion
Each compound was evaluated first in a normal
behavioral rat model under the standard protocols of a
CNS - primary observation test (POT).10 Observations
were recorded at 0.5, 1, 2, and 3 h after oral dosing and
were based on 10 indices of behavioral activity includ-
ing: mortality, diarrhea, lacrimation, locomotion, pilo-
erection, twitches, tremors, body position, respiratory

2604 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16
Ta ble 1. Substitution at the R1 Position
Bebernitz et al.
ob/ob mice % efficacyc
cmpd #
R1 position
Syn method
MP (°C)
empirical formulaa
dose mg/kg
day 3 4 h
day 3 8 h
1
3-CF3
A
A
A
A
A
A
A
A
A
A
A
A
A
A
110-112
108-109
102-105
194-195
100-102
109.5-111
138-139
141-143
141-144
128-129
144-145
145-146
120.5-122
127-128
C20H18N3F3O
300
300
300
100
100
300
300
300
100
300
300
300
300
300
71d
NS
74d
-72d
NS
39d
47d
84d
NS
45d
34d
52d
NS
NS
36d
63d
75d
-88d
NS
35d
52d
39d
NS
37d
53d
NS
NS
NS
10a
10b
10c
10d
10e
10f
10g
10h
10i
10j
10k
10l
10m
3-CF3
C20H16Cl2F3N3O
C21H18Cl2FN3O
C19H16Cl2N4O3
C20H19Cl2N3O
3-Fb
3-NO2
3-CH3
4-OMe
4-Cl
4-F
4-tert-butyl
4-CF3
3,4-Cl2
3,5-Cl2
3,5-(CF3)2
3,5-(CH3)2
C
20H19Cl2N3O2
C19H16Cl3N3O
C19H16Cl2FN3O
C23H25Cl2N3O
C20H16F3Cl2N3O
C19H15Cl4N3O
C19H15Cl4N3O
C21H15Cl2F6N3O
C21H21Cl2FN3O
a
b
Analytical results were within 0.4% of the theoretical value. Results shown are for the pyrrolidine amide rather than the
dimethylamide. Ref 42. c Values are presented as % efficacy where 100% efficacy results in normalization of the blood glucose levels to
d
that of normal mice. NS ) nonsignificant. p < 0.05.
3 days and obtaining glucose levels at 2 and 4 h on day
1 and 2, 4, and 8 h postdose on day 3 (see Experimental
Section).
that might account for the noted difference between 11g
and 11n are electronics and conformational flexibility.
It has been shown by various techniques22 that whereas
the twist angle between phenyl groups in the fluorenyl
group is expectedly 0°, the twist in the biphenyl is
approximately 26-30° indicating that these two sub-
stituents have a somewhat different orientation in
space. In addition, substituent stability studies have
demonstrated23 that more electronic character of the
substituents is transferred to the adjacent aromatic
when the two aromatics are planar as is the case for
the fluorenyl system than would be the case for the
biphenyl. As variation in electronic character at the R2
position had no preferential effect on activity, this may
suggest that the planar orientation of the fluorenyl
moiety24 might be the primary determinant of enhanced
activity perhaps through some preferential effect on pi
stacking. We therefore examined variations in Ar2 that
would challenge the ability of this substituent to ef-
ficiently pi stack. Functionalization of the 9-position of
the fluorenyl group with hydrogen bond donors and
acceptors (11o-11r ) did not compromise their activity,
suggesting that the binding pocket does allow some
flexibility. However, addition of a third aromatic group
as in 11j resulted in loss of activity. As it had been
demonstrated (11g-11i) that para-substitution of the
second aromatic group of the biphenyl was necessary
for activity, we further investigated what effect changes
in orientation would have on activity. Substitution of a
phenanthrene for Ar2 with attachment at either the 2-
or 9-position (11v vs 11w ) provided additional support
for this preferred para-aromatic orientation. Although
both compounds (11v and 11w ) possessed similar
physical and pharmacokinetic properties,25 only 11v was
found to be active. On the basis of further substitutions
Variations at the 1, 3, and amide positions of the
pyrazole (R1, Ar2, and R3, respectively) were each done
independently due in part to the linearity of the
synthesis (Tables 1, 2, and 3, respectively). Although
the specific target for these compounds is unknown, the
potency and efficacy of these compounds in vivo would
argue for a specific target, and this premise forms the
basis for the discussion to follow. For variation of
substituents on the aromatic group at the 1-position (R1,
Table 1), both electron withdrawing and releasing
groups appeared beneficial without any apparent pref-
erence in terms of 3- or 4-position on the aromatic ring.
Simple alkyl substitution proved generally inadequate,21
while substitution with the 3-nitro (10c) actually caused
the opposite and undesired effect of hyperglycemia.
As compared to substitution at R1, substitution at Ar2
resulted in more substantial improvement in activity
(Ar2, Table 2). Although both electron withdrawing
(10a , 11a ) and electron releasing groups (11b, 11f)
appeared beneficial, bulky and flexible substituents
placed at the 4-position of the aromatic (11c-11e)
resulted in a loss of activity. In light of these findings,
it was interesting that substitution of an aromatic group
at the 4-position to afford a biphenyl moiety (11g)
resulted in an improvement in activity. The position of
the aromatic substitution was shown to be important
with the para-phenyl substitution providing much better
activity than the corresponding meta-substituted (11i)
or ortho-substituted (11h ) adducts. Immobilization of
the two aromatic groups of biphenyl by means of a cyclic
fluorenyl analogue (11n ) proved especially interesting
with an increase in potency of 26-fold. Two parameters

1,3-Diaryl-[1H]-Pyrazole-4-Acetamides and Glucose Utilization
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16 2605
Ta ble 2. Substitution at the Ar2 Position

2606 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16
Ta ble 2 (Continued)
Bebernitz et al.
a
b
Analytical results were within 0.4% of the theoretical value. Values are presented as % efficacy where 100% efficacy results in
normalization of the blood glucose levels to that of normal mice. c Parent compound is prepared by the indicated method. However further
chemistry is required to prepare the shown product as described in the Experimental Section. NS) nonsignificant. *p < 0.05
around Ar2, it was concluded that, in general, com-
pounds that could adopt a para-biphenyl orientation
(11g, 11k , 11n , 11o-11r , 11v, 11x) would be active
while compounds that could not adopt this orientation
(11h , 11i, 11l, 11m , 11s, 11w ) would be inactive.
However, two compounds, 11t and 11y, which did not
conform to this para-biphenyl paradigm, were also
active, suggesting ultimately that some pi stacking motif
or pharmacokinetic property may be responsible for the
activity profile observed for substitution at Ar2.
Variation of the amide portion (R3) suggested that
tertiary amides were generally preferred, with second-
ary amides (12h ) and esters or acids (12j and 8) being
inactive. Interestingly, whereas the pyrrolidine and
dimethyl amides were found to have similar activity in
most cases (i.e., 12a and 12c) other structurally similar
amides (12d and 12b, respectively) were not active.
Substituted piperazine and dibenzyl amides (12g and
12m ) also exhibited some interesting activity although
these substitutions dramatically increased their lipo-
philic character (cLogP 7-9).
The highly lipophilic character of this series of
compounds would be considered a significant develop-
ment hurdle. Even though 11g and 11n had high
estimated cLogP′ values26 (5.9 and 5.8, respectively) and
in vitro penetration values,27 which would indicate poor
bioavailability, absolute oral bioavailability in normal
rats was determined on average to be quite good (i.e.,
25-30% from normal aqueous CMC suspensions). Ad-
dition of formulation adjuvants28 used typically for
water insoluble compounds allowed for a substantial
increase in bioavailability (50-60% range) for 11g with
a concomitant 2.5-fold improvement in EC50 to 0.56 mg/
kg (Figure 3). Although 11n already demonstrated

1,3-Diaryl-[1H]-Pyrazole-4-Acetamides and Glucose Utilization
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16 2607
Ta ble 3. Substitution at the R3 Position
ob/ob mice % efficacyb
cmpd #
R3 position
diethylamine
piperidine
pyrrolidine
Syn method
MP (°C)
empirical formulaa
dose mg/kg
day 3 4 h
day 3 8 h
8
A
A
A
A
A
A
A
A
194-197
92-93
C
C
18H11Cl2F3N2O
22H20Cl2F3N3O
300
300
300
300
300
300
100
100
30
300
300
300
300
300
100
NS
45c
NS
72c
NS
54c
NS
NS
NS
NS
57c
NS
48c
NS
52c
NS
50c
NS
76c
NS
NS
42c
45c
48c
NS
NS
NS
35c
NS
49c
12a
12b
12c
12d
12e
12f
12g
137-138
124-125
178-179
158-159
171-173
123-125
C23H20Cl2F3N3O
C22H18Cl2F3N3O
C22H18Cl2F3N3O2
morpholine
4-methyl piperazine
4-phenylpiperazine
4-diphenylmethyl piperazine
C
C
23H21Cl2F3N4O
28H23Cl2F3N4O
C35H29Cl2F3N4O
12h
12i
12j
12k
12l
12m
aniline
N-methylaniline
ethanol
methoxyamine
hydroxylamine
dibenzylamine
A
A
A
A
A
A
161-166
104-105
66-67
212-214
200(D)
C24H16Cl2F3N3O
C
C
C
25H18Cl2F3N3O
20H15Cl2F3N2O2
19H14Cl2F3N3O2
C18H12Cl2F3N3O2
32H24Cl2F3N3O
102-103
C
a
b
Analytical results were within 0.4% of the theoretical value. Values are presented as % efficacy where 100% efficacy results in
normalization of the blood glucose levels to that of normal mice. NS) nonsignificant. c p < 0.05.
remarkable potency (ED50 ) 0.05 mg/kg), a correspond-
ing increase in bioavailability was not seen however for
11n under similar dosing conditions due to its substan-
tially reduced solubility in the microemulsion vehicle
itself.29 Although SAR studies suggested that some
measure of lipophilicity was necessary for the desired
in vivo activity, a plot of activity versus cLogP did not
provide any meaningful correlation.30
It has been suggested that the thiazolidinedione’s
effect on moderating glucose levels and insulin sensitiv-
ity is through effects on adipocyte differentiation which
F igu r e 2. Stimulation of preadipocyte differentiation by
are mediated through binding to the PPAR-γ receptor,
antidiabetic agents effects of above listed compounds on 3T3-
a key adipogenic transcription factor.31 It has also been
L1 preadipocyte conversion. Two-day postconfluent 3T3-L1
preadipocytes were placed in DMEM containing 10% FBS and
160 nM insulin. Cells were fed fresh medium and compound
every 2 days. On day 7, cells were washed once and fixed in
shown that the relative affinity of these ligands to the
PPAR-γ receptor is predictive of their corresponding
anti-hyperglycemic effects in vivo.32 This mechanism
has resulted chronically in increased adiposity in animal
models33 as well as in patients.34 For this reason, we
evaluated 11g for its effects on adipocyte differentiation.
As evident from Figure 2, whereas troglitazone exhib-
ited enhanced adipocyte differentiation, metformin and
11g had no effect. This was a very positive finding for
this series and supported that these compounds were
providing glucose lowering through some mechanism
other than through PPAR-γ receptor activation.
As the mechanistic target for 11g is unknown, a
further evaluation of its pharmacological characteristics
was undertaken. 11g exhibited increased 2-DG uptake
into 3T3-L1 adipocytes both in the presence and absence
of submaximal insulin.35 11g had no effect on fatty acid
oxidation, glucose production, or ATP levels in isolated
hepatocytes.36 In addition to the significant anti-hyper-
glycemic effect 11g demonstrated in ob/ob mice, a
corresponding 50% reduction in insulin levels37 was
observed consistent with an improvement in overall
10% formalin, and lipids were stained with old red O. Cellular
old red O content was measured by image analysis and
adipocyte differentiation was expressed as a percent maximal
adipocyte differentiation stimulated using standard differen-
tiation medium (160 nM insulin, 0.5 mM IBMX, 0.25 µM DEX).
Data are represented as mean ( SEM of three independent
experiments.
insulin sensitivity. Both body weights and food intake
were reduced in comparison to vehicle matched con-
trols38 for 11g. In addition, as these compounds had
structural similarities to lonazolac, 11g was evaluated
in a model of acute inflammation (p.o. in rat carrag-
eenan-induced paw oedema) and was weak as compared
to the activity of lonazolac and other NSAIDs.39 The
behavioral-like side effects characteristic of 1 were not
observed with 11g. Taken together with its bioavail-
ability (vide supra), duration of action (up to 8 h) and
lack of tachyphylaxis,40 11g was considered an attrac-
tive candidate for glucose lowering in diabetics.

2608 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16
Bebernitz et al.
Microphot.fxa microscope with an attached Panasonic CB50
video camera. Data are expressed as a percent of maximal
adipocyte differentiation obtained using a standard differen-
tiation medium and are the mean ( SEM of three independent
experiments.
(b) P r im a r y Obser va tion Test (P OT). Normal male
Sprague Dawley rats were housed in groups under normal
light cycle (lights on at 6 am) with ad lib access to chow (Purina
rodent chow) and tap water. The animals (n ) 3 per group)
were orally dosed with 1000 µmol/kg in vehicle (carboxymeth-
ylcellulose (CMC, 0.5%) with Tween-80 (0.2%)) in the postab-
sorptive state (food removed at 6 am dosing at 10 am). SDZ
57-749(1) was included in each experiment as a positive
control. Independent behavioral observations were made by
two investigators at 0.5, 1.2, and 3 h postdose. These observa-
tions were based on 10 indices of behavior including: mortal-
ity, diarrhea, lacrimation, locomotion, piloerection, twitches,
tremors, body position, respiratory rat,e and temperature. One
blood sample was obtained via tail nick for the determination
of blood glucose concentrations at 3 h postdose.
(c) Ch r on ic Glu cose Low er in g in ob/ob Mice. Adult
male C57BL ob/ob mice were housed four per cage in hanging
wire bottom cages with standard laboratory conditions and a
12:12 h light/dark cycle. Food (Purina rodent chow) and water
were available ad libitum. For blood sampling, the mice were
placed in a restrainer and a drop of blood was obtained through
a nick in the tail. Blood glucose concentrations were deter-
mined in 10 mL of blood using a YSI27 analyzer (Yellow
Springs Instruments). The protocol for compound evaluation
was as follows: The mice were distributed into groups (n ) 6
per group) matched for blood glucose levels on day 0. The
animals were dosed (p.o.) in the mornings of days 1-3 with
vehicle (CMC, 0.5% with Tween-80 (0.2%)) or compound in
vehicle. Blood glucose concentrations were monitored at 2 and
4 h post-dose on day 1 and 2, 4, and 8 h post-dose on day 3.
Behavioral observations were made before dosing on day 2 and
on day 3 before the 4 h sampling. Results are expressed as
percent efficacy, that is the ability of the compound to
normalize blood glucose levels to 100 mg/dL and calculated
F igu r e 3. Dose response curves of glucose lowering in ob/ob
mice for 11g, 11g, (microemulsion) and 11n . Dose response
curves were generated using standard ob/ob mouse protocol
as found in Experimental Section at selected doses. Key
indicates compound and vehicle in parentheses (CMC )
carboxymethylcellulose, ME ) microemulsion vehicle as de-
scribed in text). ED50 values were calculated using a sigmoidal
fit within Origin 6.1.
In conclusion, 11g is a potent and efficacious agent
that significantly lowers glucose and insulin levels in
an animal model of diabetes (ob/ob mouse) without
showing any effect on glucose levels in normal rats. The
behavioral effects apparent in the lead structure, 1, were
readily eliminated and this effect seems to have been
an aberration not generally observed with the majority
of analogues described herein. Finally, these compounds
compared favorably in ob/ob mice with marketed drug
agents such as metformin (ED50 ) 128 mg/kg/day) and
troglitazone (ED50 ) 45 mg/kg/day)41 and thus have the
potential to provide a valuable addition to the repertoire
of antidiabetic therapies.
by the formula {1 - ([glucose]compound - 100)/([glucose]vehicle
-
100)} × 100. Animals that displayed glycemia of less than 150
mg/dL on day 0 (basal) were considered not diabetic and were
excluded from the study. Statistical significance (p < 0.05) was
evaluated by a t-test comparison to the appropriate sample in
vehicle-treated animals.
Exp er im en ta l Section
Biologica l Meth od s. (a ) P r ea d ip ocyte Differ en tia tion
Assa y. Preadipocytes were seeded in 48-well plates at a
density of 5000 cells/0.5 mL of predifferentiation medium (high
glucose (25 mM) DMEM (Gibco, Gaithersburg, MD), containing
10% newborn calf serum (Gibco), 50 units/mL penicillin
(Gibco), and 50 mg/mL streptomycin sulfate (Gibco)) and
For clarity, only data from day 3 @4 and 8 h are shown. If
the results were not significant, NS is displayed in the column.
Ch em ica l Meth od s
o
maintained in a 7% CO2 humidified atmosphere at 37 C. On
Melting points were determined on a Thomas-Hoover capil-
lary melting point apparatus and are uncorrected. 1H NMR
were recorded on a Bruker AC300 MHz NMR using tetram-
ethylsilane (TMS) as an internal standard and are reported
in ppm (δ). Mass spectra were performed on a Finnigan Mat
4600 spectrometer. Elemental analysis were performed with
a Carlo Erba CHNS-O EA 1108 elemental analyzer. Data are
within 0.4% of theoretical values unless otherwise indicated.
All compounds were routinely checked by TLC with Macherey-
Nagel Polygram SIL G/UV254 plates. Yields are of purified
products and were not optimized.
Meth od A: Gen er a l P r oced u r es for th e P r ep a r a tion
of P yr a zole. P r ep a r a tion of 3-[1,1′-bip h en yl]-4-yl-N,N-
d im et h yl-1-[3-(t r iflu or om et h yl)p h en yl]-1H -p yr a zole-4-
a ceta m id e (11 g): (a ) 1-[1,1′-Bip h en yl]-4-yl-, [3-(tr iflu o-
r om et h yl)p h en yl]h yd r a zon e, (1E ) et h a n on e (4g). To
4-phenylacetophenone (0.1 mol, 19.89 g) in 200 mL of acetic
acid and 10 mL of water was added 3-trifluoromethylphenyl-
hydrazine (0.1 mol, 19.4 g). Shortly after addition, a solid mass
formsed and acetic acid was added to maintain stirring for 15
h. The reaction mixture was diluted with an additional 50 mL
of water and subsequently filtered and the solid was dried in
vacuo to afford 33.1 g of crystalline product 4g (93% yield):
mp 130-133 °C; 1H NMR (CDCl3) δ 2.28 (s, 3H), 7.38 (m, 3H),
day 7, confluent preadipocytes were changed to postdifferen-
tiation medium (high glucose DMEM containing 10% fetal
bovine serum (FBS, Gibco), 50 units/mL penicillin, 50 mg/mL
streptomycin sulfate, 10 mg/mL D-biotin (Sigma, St Louis,
MO), and 10 mg/mL pantothenic acid (Sigma)). To induce
differentiation, test compounds were added on day 7 to
postdifferentiation medium or postdifferentiation medium
containing 1 mg/mL porcine pancreas insulin (Sigma, #I-3505),
where the final DMSO concentration in the medium was 0.2%.
Cells were refed fresh test medium every 2 days. As a control,
full differentiation was induced on day 7 by treating cells with
postdifferentiation medium containing 0.5 mM 3-isobutyl-1-
methylxanthine (IBMX, Sigma), 1 mg/mL porcine pancreas
insulin (Sigma, #I-3505), and 0.25 mM dexamethasone (Sigma).
After 2 days, control cells were washed once with Dulbecco’s
PBS (D-PBS) and exposed to the above medium without IBMX
and dexamethasone for two additional days. Thereafter, fully
differentiated control cells were maintained in postdifferen-
tiation media in the absence of insulin, dexamethasone, or
IBMX. On day 14, cell monolayers were fixed for 1 h in 10%
buffered formalin, and cellular lipid was stained with oil red
O. Accumulation of oil red O was measured using the IBAS
image analysis software (Kontron, Munich, Germany) as the
mean gray density of light transmission using a Nikon

1,3-Diaryl-[1H]-Pyrazole-4-Acetamides and Glucose Utilization
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16 2609
7.47 (m, 4H), 7.63 (d, J ) 9.7 Hz, 4H), 7.87 (d, J ) 9.7 Hz,
2H); MS (DCI, isobutane) m/z (rel intensity) 355 (100).
7.96 (d, J ) 8 Hz, 1H), 8.07 (s, 1H), 8.12 (s, 1H); MS (DCI,
isobutane) m/z (rel intensity) 450 (100). Anal. (C26H22F3N3O),
C,H, N.
(b) 3-[1,1′-Bip h en yl]-4-yl-1-[3-(tr iflu or om eth yl)p h en yl]-
1H-p yr a zole-4-ca r boxa ld eh yd e (5g). Phosphorus oxychlo-
ride (0.21 mol, 20 mL) was added to 150 mL of DMF at 0 °C
and stirred for 30 min. 4g (0.094 mol, 33.1 g) was added as a
solid slowly to this mixture and stirred for 15 h. The crude
reaction was then quenched into 1 L of water and stirred for
an additional 15 h. The resulting solid was filtered, dissolved
in methylene chloride, and dried over MgSO4, filtered, and
evaporated to a crude solid that was crystallized from ether/
hexane (5:1) to afford 30.4 g of 5g in 83% yield: mp 190-191
9,9-Dim et h ylflu or en e (13o). Fluorene (50 mmol, 8.3 g)
was dissolved in THF (150 mL) and cooled to 0 °C. KHMDS
(50 mmol, 100 mL of 0.5 M solution/toluene) was added
dropwise at 0 °C, and then the reaction mixture was allowed
to warm to RT for 30 min. The mixture was then cooled to 0
°C and methyl iodide (50 mmol, 3.2 mL) was added and stirred
for 30 min. This cycle was repeated with an additional amount
of KHMDS (50 mmol, 100 mL of 0.5 M solution/toluene) and
subsequently methyl iodide (50 mmol, 3.2 mL) and the reaction
mixture was allowed to warm to RT overnight. The reaction
mixture was quenched into aqueous NaH2PO4 and extracted
with MTBE. The organic extracts were dried and evaporated
to afford 13o as a waxy solid (9.8 g, 100%): 1H NMR (CDCl3)
δ 1.50 (s, 6H), 7.3 (m, 4H), 7.44 (m, 2H), 7.70 (m, 2H); MS
(DCI, isobutane) m/z (rel intensity) 195 (100).
Meth od B: Gen er a l P r oced u r e for th e P r ep a r a tion of
P yr a zoles. (a ) Step A - P r ep a r a tion of th e Acyla ted
P r ecu r sor 14: 9,9-Dim eth yl-γ-oxo-9H-flu or en e-2-bu ta n o-
ic a cid , Meth yl Ester (14o). To 9,9-dimethylfluorene (13o)
(50 mmol, 9.7 g) and methyl 4-chloro-4-oxobutyrate (51 mmol,
6.3 mL) in 1,1,2,2-tetrachloroethane (100 mL) at 0 °C was
added AlCl3 (105 mmol, 14 g) portionwise over 10 min. The
resulting mixture was stirred for 1 h at 0 °C and 2 h at ambient
temperature and then quenched into cold 6 N HCl and
extracted 3× with methylene chloride. The combined organic
layer was dried over MgSO4 and then evaporated to a crude
oil that was chromatographed over silica (20% EtOAc/hexane)
to afford 14o as a thick oil (14 g, 91%): 1H NMR (CDCl3) δ
1.52 (s, 6H), 2.80 (t, J ) 7.5 Hz, 2H), 3.40 (t, J ) 7.5 Hz, 2H),
3.74 (s, 3H), 7.37 (m, 2H), 7.47 (m, 1H), 7.77 (m, 2H), 8.00 (d,
J ) 7.5 Hz, 1H), 8.06 (s, 1H); MS (DCI, isobutane) m/z (rel
intensity) 309 (100).
(b) Step B - P r ep a r a tion of th e P yr a zole P r od u ct. (1)
3-(9,9-Dim eth yl-9H-flu or en -2-yl)-N,N-d im eth yl-1-[3-(tr i-
flu or om eth yl)ph en yl]-1H-pyr azole-4-acetam ide (11o). 14o
(16.2 mmol, 5 g) was directly treated with 3-trifluorometh-
ylphenylhydrazine (16.4 mmol, 3.21 g) in propionic acid (50
mL) for 4 h before being evaporated to a crude oil. The residue
was partitioned between MTBE and water, washed with brine,
dried, and evaporated to afford 15o as a thick oil (7.6 g, 99%).
1H NMR (CDCl3) δ 1.55 (s, 6H), 2.7-2.8 (m, 2H), 2.9-3.15 (m,
2H, E/Z isomers), 3.73 (s, 3H), 6.9-7.9 (m, 11H), 9.1 (bs, 1H)
MS (DCI, isobutane) m/z (rel intensity) 467 (100). The Vils-
meier reagent was prepared by adding POCl3 (86 mmol, 8 mL)
to DMF (50 mL) at 0 °C and allowing the mixture to stir and
warm to room temperature over 2 h). Crude 15o (16 mmol 7.6
g) was dissolved in DMF (25 mL), chilled to 0 °C, and added
to the Vilsmeier reagent, and the mixture was stirred to 20 h
warming to room temperature. The reaction mixture was
quenched into water (1 L) and stirred for 15 h. The aqueous
layer was decanted and the residue was dissolved in MTBE
and washed with brine, dried, and evaporated to an oil.
Chromatography on silica (50-80% CH2Cl2/pentane) afforded
3-(9,9-dimethyl-9H-fluoren)-2-yl-1-[3-(trifluoromethyl)phenyl]-
1H-pyrazole-4-acetic acid, methyl ester (16o) (4.6 g, 61%): 1H
NMR (CDCl3) δ 1.55 (s, 6H), 3.74 (s, 3H), 3.84 (s, 2H), 6.9-8.2
(m, 12H); MS (DCI, isobutane) m/z (rel intensity) 477 (100).
Dimethylamine (69 mmol, 3.1 g) was dissolved in toluene
at 0 °C before addition of AlCl3 (20 mmol, 2.8 g) portionwise
over 10 min. The mixture was stirred for 15 min and then 16o,
dissolved in 15 mL of toluene, was added and the reaction
mixture was allowed to warm to RT and stirred for 15 h. The
reaction was evaporated, quenched into ice water (300 mL),
extracted with EtOAc, and washed with aqueous 1 N HCl and
brine. The organic layer was dried and evaporated to a crude
oil. The oil was crystallized from MTBE/heptane to afford 11o
(3.05 g, 65%): mp 98-101 °C; 1H NMR (CDCl3) δ 1.54 (s, 6H),
2.95 (s, 3H), 3.01 (s, 3H), 3.77 (s, 2H), 7.29-7.41 (m, 2H), 7.42-
7.64 (m, 4H), 7.68-7.86 (m, 3H), 7.97 (d, J ) 8 Hz, 1H), 8.08
(s, 1H), 8.14 (s, 1H); MS (DCI, isobutane) m/z (rel intensity)
490 (100); Anal. (C29H26F3N3O), C,H,N,F.
1
°C; H NMR (CDCl3) δ 7.39 (m, 1H), 7.45 (t, J ) 7.5 Hz, 2H),
7.66 (d, J ) 6.75 Hz, 4H), 7.73 (d, J ) 8.25 Hz, 2H), 7.93 (d,
J ) 8.25 Hz, 2H), 8.02 (m, 1H), 8.17 (s, 1H), 8.63 (s, 1H), 10.15
(s, 1H), MS (DCI, isobutane) m/z (rel intensity) 393 (100).
(c) 3-[3-[1,1′-Bip h en yl]-4-yl-1-[3-(tr iflu or om eth yl)p h e-
n yl]-1H-p yr a zol-4-yl]-, (2E)-2-p r op en oic a cid (6g). To 5g
(0.077 mol, 30.35 g) in 175 mL of pyridine and 1 mL of
piperidine was added malonic acid (0.16 mol, 16.9 g), and the
mixture heated at 90 °C for 12 h. An additional portion of
malonic acid (0.028 mol, 4 g) was added and heated at 90 °C
for 3 h. The temperature was then raised to 135 °C and heating
was continued for 3.5 h. The reaction was cooled to ambient
temperature and then quenched into a mixture of HCl (conc)
and ice. After stirring the sample for 2 h, the resulting solid
was filtered, washed 3 times with water, and dried in vacuo
to afford 6g (33.4 g, 99%): mp >250 °C; 1H NMR (d6DMSO) δ
6.51 (J ) 16.5 Hz, 1H), 7.39 (m, 1H), 7.45 (t, J ) 7.5 Hz, 2H),
7.59 (d, J ) 16.5 Hz, 1H), 7.75-7.9 (m, 8H), 8.30 (m, 2H), 9.41
(s, 1H); MS (DCI, isobutane) m/z (rel intensity) 435 (100).
(d ) 3-[1,1′-Bip h en yl]-4-yl-1-[3-(tr iflu or om eth yl)p h en yl]-
1H-p yr a zole-4-a ceta m id e (7g). To 6g (0.077 mol, 33.4 g) in
25 mL of ammonium hydroxide (conc) and 75 mL of 1,4-
dioxane was added sulfur (0.5 mol, 16 g) and the resulting
slurry was heated in a pressure vessel to 175 °C for 15 h. The
reaction vessel was cooled, the pressure was released, and the
reaction mixture was quenched into water and stirred for 2 h.
Filtration afforded 7g as an off-white solid (36.5 g) that was
1
used directly in the next reaction: mp 241-243 °C; H NMR
(d6DMSO) δ 3.57 (s, 2H), 7.04 (s, 1H), 7.40 (t, J ) 7.4 Hz, 1H),
7.50 (m, 3H), 7.69 (d, J ) 8 Hz, 1H), 7.75 (m, 3H), 7.78 (d, J
) 8 Hz, 2H), 7.86 (d, J ) 8 Hz, 2H), 8.22 (d, J ) 8 Hz, 1H),
8.24 (s, 1H), 8.64 (s, 1H); MS (DCI, isobutane) m/z (rel
intensity) 422 (100).
(e) 3-[1,1′-Bip h en yl]-4-yl-1-[3-(tr iflu or om eth yl)p h en yl]-
1H-p yr a zole-4-a cetic a cid (7g). The crude amide 7g (0.077
mol, 36.5 g) was suspended in glacial acetic acid (200 mL) and
50% sulfuric acid (140 mL) was refluxed for 2 h and then
poured into 2 L of water. The resulting solids were filtered,
washed with water until the filtrate was neutral, and subse-
quently dried in vacuo to afford acid 8g (31.3 g, 97%): 1H NMR
(CD3OD) δ 3.76 (s, 2H), 7.37 (t, J ) 7 Hz, 1H), 7.48 (t, J ) 7
Hz, 2H), 7.6-7.8 (m, 8H), 7.96 (d, J ) 7 Hz, 1H), 8.06 (s, 1H),
8.13 (s, 1H); MS (DCI, isobutane) m/z (rel intensity) 423 (100).
(f) 3-[1,1′-Biph en yl]-4-yl-N,N-d im eth yl-1-[3-(tr iflu or om -
eth yl)p h en yl]- 1H-p yr a zole-4-a ceta m id e (11g). 8g (0.035
mol, 15 g) was dissolved in methylene chloride (300 mL) with
a catalytic amount of DMF (0.5 mL) and thionyl chloride (0.35
mol, 26 mL) was added and stirred for 4 h before the reaction
was evaporated and pumped on in vacuo for 12 h. The
resulting acid chloride, 9g was dissolved in methylene chloride
(200 mL) and cooled to 0 °C. Dimethylamine (0.1 mol, 4.5 g)
was condensed into cold methylene chloride (20 mL) and added
to the acid chloride and stirred warming to ambient temper-
ature over 15 h. The reaction mixture was evaporated and then
partitioned between ethyl acetate and water. The combined
organic layer was dried over MgSO4, filtered, and evaporated
to give the amide as a tan solid. The crude material was
chromatographed on silica eluting with hexane to 50% EtOAc/
hexane to afford 11g (12.7 g, 81%): mp 125-126 °C; 1H NMR
(CDCl3) δ 2.98 (s, 3H), 3.01 (s, 3H), 3.78 (s, 2H), 7.37 (t, J ) 7
Hz, 1H), 7.47 (t, J ) 8 Hz, 2H), 7.52 (d, J ) 7 Hz, 1H), 7.57 (t,
J ) 7 Hz, 1H), 7.65 (d, J ) 7 Hz, 2H), 7.72 (q, J ) 9 Hz, 4H),

2610 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16
Bebernitz et al.
(5) Nestler, J . E.; Clore, J . N.; Blackard, W. G. Dehydroepiandros-
terone: The “Missing Link” Between Hyperinsulinemia and
Atherosclerosis? FASEB J . 1992, 6, 3073-3075.
(6) (a) King, A. B. A Comparison in a clinical setting of the efficacy
and side effects of three thiazolidinediones Diabetes Care 2000,
23, 557. (b) Sood, V.; Colleran, K.; Burge, M. R. Thiazolidinedi-
ones: A Comparitive review of approved uses. Diabetes Technol.
Ther. 2000, 2, 429-440.
(7) Zhang, B. B.; Moller, D. E., New Approaches in the Treatment
of Type 2 Diabetes. Curr. Opin. Chem. Biol. 2000, 4, 461-467.
(8) 1 increased glucose transport into 3T3-L1 cells by 159% of control
(@100 µM), in adipocytes by 131% of control (@ 300 µM) and in
L6 muscle cells by 146% of control (@300 µM).
(9) 1 was tested in the following receptor binding assays: alpha 1,
alpha 2, D1, D2, and opiate with PIC50 < 6.
(10) Irwin, S. Comprehensive obervational assessment: Ia. A sys-
tematic, quantitative procedure for assessing the behavioral and
physiologic state of the mouse. Psychopharmacologia 1968, 13,
222-257.
(11) Sarges, R.; Hank R. F.; Blake, J . F.; Bordner, J .; Bussolotti, D.
L.; Hargrove, D. M.; Treadway, J . L.; Gibbs, E. M. Glucose
Transport-Enhancing and Hypoglycemic Activity of 2-Methyl-
2-phenoxy-3-phenylpropanoic Acids. J . Med. Chem. 1996, 39,
4783-4803.
(12) Wright, J . B.; Dulin, W. E.; Markillie, J . H. The Antidiabetic
Activity of 3,5-Dimethylpyrazoles. J . Org. Chem. 1964, 7, 102-
105.
(13) (a) Soliman, R.; Faid-Allah, H. M.; El Sadany, S. K. Preparation
and Antidiabetic Activity of New Substituted 3,5-Diarylpyra-
zolesulfonylurea Derivatives II: Structure-Activity Relation-
ship. J . Pharm. Sci. 1987, 76, 626-632. (b) Masataka, F.;
Shimago, K.; Okano, S.; Nakatani, H.; Hesaka, K. Esters of
3-Pyrazolemethanols and Process for their Preparations. 1972
United States Patent # 3,666,756.
(14) Kees, K. L.; Fitzgerald, J . J . J r.; Steiner, K. E.; Mattes, J . F.;
Mihan, B.; tosi, T.; Mondoro, D.; McCaleb, M. L. New Potent
Antihyperglyceimc Agents in db/db Mice: Synthesis and Struc-
ture-Activity Relationship Studies of (4-substitutedbenzyl)-
(trifluoromethyl)pyrazoles and -pyrazolones. J . Med. Chem.
1996, 39, 3920-3928.
3-[9-(Acet yloxy)-9H -flu or en -2-yl]-N,N-d im et h yl-1-[3-
(tr iflu or om eth yl)ph en yl]-1H-pyr a zole-4-a ceta m id e (11p ).
To 11n (0.44 mmol, 0.21 g) in ethanol (10 mL) cooled to 5 °C
was added NaBH4. After 1 h, the reaction was evaporated and
then partitioned between pH 4 buffer and CH2Cl2, extracted
with CH2Cl2, dried, and evaporated to afford the alcohol as a
white foam. This material was subsequently dissolved in CH2-
Cl2 and treated with Et3N (0.72 mmol, 0.10 mL) and acetic
anhydride (0.5 mmol, 0.04 mL) at 5 °C. After 1 h, the reaction
was quenched into pH 9 buffer and extracted with CH2Cl2,
dried, and concentrated to afford crude acetate. Recrystalli-
zation from EtOAc afforded 11p (131 mg, 60%): mp 191-192
1
°C; H NMR (CDCl3) δ 2.20 (s, 3H), 3.01 (s, 6H), 3.76 (s, 2H),
6.86 (s, 1H), 7.33 (t, J ) 8 Hz, 1H), 7.44 (t, J ) 7 Hz, 1H),
7.49-7.63 (m, 3H), 7.66-7.80 (m, 3H), 7.85 (s, 1H), 7.97 (d, J
) 8 Hz, 1H), 8.05 (s, 1H), 8.13 (s, 1H); MS (DCI, isobutane)
m/z (rel intensity) 460 (100) [-HOAc], 520 (40.5); Anal.
(C29H24F3N3O3), C,H,N.
N,N-Dim et h yl-3-(9-oxo-9H -flu or en -2-yl)-1-[3-(t r iflu o-
r om eth yl)p h en yl]-1H-p yr a zole-4-a ceta m id e (11r ). KMnO4
(89.2 mmol, 14.1 g) was dissolved in 400 mL of water and
added to a basic solution of nBuNSO4 (4.1 mmol, 1.42 g) in
KOH (26 mmol, 1.67 g in 210 mL of water). This mixture was
added to 11n (43 mmol, 20 g) in CH2Cl2 (400 mL), and the
mixture was stirred overnight at RT. Solid KH2PO4 was added
until the reaction was neutral, and then solid NaHSO3 was
added until the MnO2 was reduced. The organic layer was
separated and extracted with CH2Cl2, dried, and concentrated
to an oil that was chromatographed over silica (EtOAc) to
1
afford 11r (16 g, 78%): mp 144-146 °C; H NMR (CDCl3) δ
3.03 (s, 6H), 3.76 (s, 2H), 7.29-7.37 (dt; J ) 7,1 Hz, 2H), 7.48-
7.72 (m, 6H), 7.84-7.93 (m, 1H), 7.96 (d, J ) 7 Hz, 1H), 8.05
(s, 1H), 8.12 (s, 1H); MS (DCI, isobutane) m/z (rel intensity)
476 (100). Anal. (C27H20F3N3O2), C,H,N,F. A second material
was also obtained (3 g, 12%) that corresponded to an additional
oxidation at the methylene of the acetamide side chain to
afford a ketoamide.
(15) (a) Rainer, G. Pyrazol-4-Acetic Acid Compounds. Byk Gulden
Lomberg Chemische Fabrik GmbH. 1979 U.S. Patent # 4,146,-
721. (b) Newberry, R. A. Pyrazole Compounds 1970 London
Patent # 1373212.
3-(9-Hyd r oxy-9-m eth yl-9H-flu or en -2-yl)-N,N-d im eth yl-
1-[3-(tr iflu or om eth yl)p h en yl]-1H-p yr a zole-4-a ceta m id e
(11q). To 11r (3.4 mmol, 1.62 g) in THF (50 mL) at -60 °C
was added MeMgCl (3.6 mmol, 1.2 mL in THF). The reaction
was stirred for 30 min and then allowed to warm to RT.
Workup entailed quenching into aq. NaH2PO4 and extracting
with MTBE, drying, and evaporating to a crude oil. Chroma-
tography over silica (EtOAc to 10% ethanol/EtOAc) afforded
11q as a solid (1.41 g, 84%): mp 170-172 °C; 1H NMR (CDCl3)
δ 1.80 (s, 3H), 1.84 (s, 1H), 2.99 (s, 6H), 3.76 (s, 2H), 7.31-
7.44 (dp J ) 7.5,1 Hz, 2H), 7.50-7.74 (m, 6H), 7.87 (s, 1H),
7.96 (d, J ) 8 Hz, 1H), 8.06 (s, 1H), 8.11 (s, 1H); MS (DCI,
isobutane) m/z (rel intensity) 474 (100) [-H2O], 492 (33.6); Anal.
(C28H24F3N3O2), C,H,N,F.
(16) Various NSAIDs were evaluated in L6 myocytes for their
potential effects on glucose transport. The following effects on
glucose uptake were observed in L6myocytes with compound @
300 µM and represented as % of control: acetominophen 83%;
ibuprofen 68%; indomethacin 110%; lonazolac 81%; 1 146%;
insulin (1µM) 161%; for a description of the L6 myocytes assay
see Aicher, T. D. et al. J . Med. Chem. 1998, 41, 4556.
(17) Kira, M. A.; Abdel-Rahman, M. O.; Gadalla, K. Z. The Vilsmeier-
Haack Reaction III Cyclization of Hydrazones to Pyrazoles.
Tetrahedron Lett. 1969, 109-110.
(18) (a) For a review of the various transformations of the Willgerodt
reaction see Brown, E. V. The Willgerodt Reaction Synthesis
1975, 358-375. (b) For discussion of mechanisms see (1) Maier,
L. Helv. Chim. Acta 1970, 53, 1216 or (2) Carmack, M. the
Willgerodt-Kindler Reactions. 7. The Mechanisms J . Heterocycl.
Chem. 1989, 26, 1319-1323
(19) Carmack, M.; Davis, C. The Willgerodt Reaction. V. Substituted
Acetamides from â-substituted Acrylic Acids. J . Org. Chem.
1947, 12, 76-78.
Ack n ow led gm en t. The authors thank Dr. Michael
J . Shapiro and Bertha Owens for NMR and MS analysis.
We also thank Dr. Edwin Villhauer for helpful com-
ments while reviewing this manuscript.
(20) Generally behavioral effects were only noted with compounds
where Ar2 was unsubstituted aromatic as in 1. Therefore, the
pyrrolidine amide of 1 also demonstrated marked behavioral
effects. With Ar2 other than unsubstituted phenyl no other
substitutions at either R1 or R3 had any behavioral-like effects
except for the phenol ester of 8. This activity was not observed
in any other compound from Table 3 including the acid itself
(8).
Su p p or tin g In for m a tion Ava ila ble: Additional spectral
and analytical data for analogues. This material is available
free of charge via the Internet at http://pubs.acs.org.
(21) Only tested up to 100 mg/kg.
(22) Roberts, R. M. G. Conformational Analysis of Biphenyls using
13C NMR Spectroscopy. Magn. Reson. Chem. 1985, 23, 52-54.
(23) Bolton, R.; Burley, R. E. M. Stability of Carbonium Ions. Part
3. The Transmission of Substituent Effects Across the Fluorene
and Biphenyl Systems. J . Chem. Soc., Perkin Trans. 2 1977, 4,
426-429.
(24) (a) Duncia, J . V.; Santella, J . B. III;, Chiu, A. T.; Wong, P. C.;
Wexler, R. R. Dibenzobiclyclo [2.2.2]octane Angiotensin II
Receptor Antagonists. Bioorg. Med. Chem. Lett. 1995, 5, 214-
246. (b) Burnelli, S.; Varioli, L.; Guarnieri, A.; Lumachi, B.
Nonsteroidal Antiinflammatories. XXII. Chiral Arylhydroxy-
fluorenylpropionic acids. Bull. Chim. Farm. 1991, 130, 133-135.
(c) Grunewald, G. L.; Carter, A. E.; Sall, D. J .; Monn, J . A.
Conformational Preference for the Binding of Biaryl Substrates
and Inhibitors to the Active Site of Phenylehtanolamine N-
Methyltransferase. J . Med. Chem. 1988, 31, 60-65.
Refer en ces
(1) Cowie, C. C., Eberhardt, M. S. Diabetes 1996 Vital Statistics;
American Diabetes Association: Alexandria, VA.
(2) Diabetes Control and Complications Trial Research Group. The
effect of intensive treatment of diabetes on the development and
progression of long-germ complications in insulin-dependent
diabetes mellitus. New Eng. J . Med. 1993, 329, 977-986.
(3) Sowers, J . R.; Standley, P. R.; Ram, J . L.; J acober, S.; Simpson,
L.; Rose, K. Hypinsulinemia, Insulin Resistance, and Hyperg-
lycemia: Contributing Factors in the Pathogenesis of Hyperten-
sion and Atherosclerosis. Am. J . Hypertens. 1993, 6, 260S-270S.
(4) Depres, J . P.; Lamarche, B.; Mauriege, P.; Cantin, B.; Dagenais,
G. R.; Moorjani, S.; Lupien, P. J . Hyperinsulinemia as an
Independent Risk Factor for Ischemic Heart Disease. New Engl.
J . Med. 1996, 334, 952-957.

1,3-Diaryl-[1H]-Pyrazole-4-Acetamides and Glucose Utilization
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16 2611
(36) Hepatocytes were isolated from the livers of 18 fasted
(25) Values for 11v/11w cLogP: 6.4/6.4; caco-2 penetration (Papp
×
h
10-5) 1.1/2.2; microsomes (stability) (rat-T1/2 - min) stable/18.2.
(26) cLogP values were determined by the method of Biobyte using
CLOGP3. The traditional shake-flask method for determining
LogP values was not possible for this series of compounds due
to their insolubility in water. As these compounds are not
ionizable, LogD would be of no additional value.
Sprague-Dawley rats (weighing 280-290 g) by the method of
Berry and Friend. (Berry, M. N., Friend, D. S. High-yield
preparation of isolated rat liver parenchymal cells: a biochemical
and fine structure study. J . Cell Biol. 1969; 43, 506-520.) Cells
were incubated in the presence or absence of compounds at the
indicated concentrations for 30 min at 37 °C before the addition
of 10 mM lactate, 1 mM pyruvate, and 0.5 mM [1-14C]oleate.
The specific activity of [1-14C]oleate (Du Pont-NEN, Boston, MA)
was 1 mCi/mmol. After 30 min additional incubation at 37 °C,
aliquots of the cell suspensions were immediately placed in an
ice bath and were deproteinized with an equal volume of ice-
cold 12% trichloroacetic acid (TCA). Supernatants from the
centrifuged hepatocyte suspensions were assayed for glucose
content with a spectroscopic (A505) automated glucose oxidase
method and for acid-soluble products (∼95% ketone bodies in
control incubations) from the 14C radioactivity via liquid scintil-
lation counting. Total cellular ATP concentrations were deter-
mined enzymatically from the TCA supernatants with phospho-
glyceratephosphokinaseandglyceraldehydephosphatedehydrogenase
(Sigma kit 366-UV). 11g had no effect up to 300 µM. Troglitazone
(@300 µM) exhibited effects on fatty acid oxidation (increased
25%) and glucose production (decreased 60%), and ATP levels
were reduced to 25% of control.
(27) Caco-2 penetration (Papp × 10-5) 1.7/2.1 - Apparent permeabili-
ties (Papp) across caco-2 cells apical to basolateral at 120 min.
(28) The dosed compound was dissolved in a 40% microemulsion
vehicle in water. Microemulsion vehicle: 43% Cremophor RH40,
35.75% corn oil-monoglyceride, 10.62% propylene glycol, and
10.62% ethanol. Vehicle itself had no significant effects on
metabolic parameters in ob/ob mice.
(29) A comparison of 11g and 11n , which have similar cLogP values
(5.9 and 5.8, respectively) and are both insoluble in water (<0.04
mg/mL, pH 6.8) yet had a 10-fold difference in ethanol solubility
(24 and 2 mg/mL, respectively), illustrates the problems associ-
ated with formulating these compounds as drug substances
(30) cLogP values were determined by the method of Biobyte using
CLOGP3. A plot of cLogP versus activity in vivo (ob/ob mice)
provided no correlation (data not shown).
(31) (a) Tontonoz, P.; Hu, E.; Spiegelman, B. M. Stimulation of
Adipogenesis in Fibroblasts by PPARγ2,
a Lipid-activated
Transcription Factor. Cell 1994, 79, 1147-1156. (b) Hu, E.;
Tontonoz, P.; Spiegelman, B. M. Transdifferentiation of Myo-
blasts by the Adipogenic Transcription Factors PPARγ and
C/EBPR. Proc. Nat. Acad. Sci. U.S.A. 1995, 92, 9856-9580.
(32) (a) Lehman, J . M.; Moore, L. B.; Smith-Oliver, T. A.; Wilkinson,
W. O.; Willson, T. M.; Kliewer, S. A. An Antidiabetic Thiazo-
lidinedione is a High Affinity Ligand for Peroxisome Proliferator-
Activated Receptor γ (PPAR-γ) J . Biol. Chem. 1995, 270, 12953-
12956. (b) Willson, T. M.; Cobb, J . E.; Cowan, D. J .; Wiethe, R.
W.; Correa, I. D.; Prakash, S. R.; Beck, K. D., Moore, L. B.,
Kliewer, S. A., Lehmann, J . M. The Structure Activity Relation-
(37) Sufficient plasma was taken at 4 h to obtain both glucose and
insulin levels. Day 3-4 h insulin levels were 820 µU/mL (p <
0.05) vs 1995 µU/mL vehicle controls.
(38) Procedure same as for glucose lowering in ob/ob mouse described
in experimental study except duration for 7 days and n ) 10
per group. Control animals gained weight over the time period
(47 g, day 1; 51 g, day 7). On day 7, body weights were
significantly decreased as compared to control in the metformin
(350 mg/kg/day: 48 g, +1 g, p < 0.05) and 11g (15 mg/kg/day:
47 g, no change, p < 0.05) groups but were unchanged compared
to controls in the troglitazone (200 mg/kg/day: 51 g, +4 g)
treated animals. Food intake was reduced modestly by 10% for
11g as compared to troglitazone and vehicle matched control.
(39) Carrageenan oedema model: Compounds were administered
orally in a vehicle of Tween 80/tragacanth 1 h prior to carrag-
eenan. 0.1 mL of 1% carrageenin suspended in physiological
saline was given by sub-plantar injection into a hind paw. The
control reading was taken immediately after the injection (0-h
value), and the swelling was measured after 3 h by means of an
antiphlogometer. Four rats (Tuttlingen, 150-170 g) were used
per group. The mean values of the 3-h readings after deduction
of the corresponding 0-h readings were calculated. Results are
presented as % inhibition of paw swelling in treated animals
versus controls. Results: diclofenac (3 mg/kg) 79%; lonazolac (5
mg/kg) 47%; 11g (5 mg/kg) 36%.
(40) Study performed as per normal 3 day dosing of ob/ob mice as
explained in Experimental Section except study was run for 7
days. Efficacy values @4 h postdose were day 1, 37.8%; day 2,
45.8%; day 3, 71.1%; and day 7, 74.4% (All values were
statistically significant (p < 0.05)
(41) These results are from in-house investigations run parallel to
lead compounds 11g and 11n in ob/ob mice under the normal
protocol.
(42) These data are for the pyrrolidine amide. Pyrrolidine and
dimethyl amides were found to be equivalent in activity in all
other cases as indicated in text. The lack of activity of the
dimethyl amide in this case is considered an aberration.
ship Between Peroxisome Proliferator-Activated Receptor
Agonism and the Antihyperglycemic Activity of Thiazolidinedi-
ones. J . Med. Chem. 1996, 39, 665-668.
γ
(33) (a) Yoshioka, S.; Nishino, H.; Shiraka, T.; Ikeda, K.; Koike, H.;
Okuno, A.; Wada, M.; Fujiware, T.; Horikoshi, H. Antihyper-
tensive Effects of CS-045 Treatment in Obese Zucker Rats.
Metabolism 1993, 42, 75-80. (b) Buchanan, T. A.; Meehan, W.
P.; J eng, Y. Y.; Yang, D.; Chan, T. M.; Nadler, J . L.; Scott, S.;
Rude, R. K.; Hsueh, W. A. Blood Pressure Lowering by Piogli-
tazone: Evidence for a Direct Vascular Effect. J . Clin. Invest.
1995, 96, 354-360. (c) DeSouza, C. J .; Yu, J . H.; Robinson, D.
D.; Ulrich, R. G.; Meglasson, M. D. Insulin Secretory Defect in
Zucker fa/fa rats is Improved by Ameliorating Insulin Resis-
tance. Diabetes 1995, 44, 984-991. (d) Sugiyama, Y.; Taketomi,
S.; Shimura, Y.; Ikeda, H.; Fujita, T. Effects of Pioglitazone in
Glucose and Lipid Metabolism in Wistar Fatty Rats. Arzmeim-
Forsch. 1990, 40, 263-267.
(34) Mori, Y., Murakawa, Y., Okada, K., Horikoshi, K., Yokoyama,
J ., Tajima, N., Ikeda, Y.; Effects of Troglitazone on Body Fat
Distribution in Type 2 Diabetic Patients. Diabetes Care 1999,
22, 908-912.
(35) Compounds were evaluated as to 2-deoxyglucose uptake and
phosphorylation in fully differentiated 3T3-L1 adipocytes (as per
Experimental Section) in the presence or absence of 1 nM
submax insulin. Cells were treated with compounds for 18 h in
serum-free DMEM, washed and incubated for 30 min in the
presence or absence of 1 nM insulin, followed by 2-DG uptake
measurement. Both 11g and troglitazone increased 2-DG uptake
by 200-300% over control values @ 100 µM.
J M010032C