Synthesis and structure-activity relationship studies of cinnamic acid-based novel thiazolidinedione antihyperglycemic agents

Article abstract of DOI:10.1016/S0968-0896(03)00393-6

A number of 2,4-thiazolidinedione derivatives of -phenyl substituted cinnamic acid were synthesized and studied for their PPAR agonist activity. The E-isomer of cinnamic acid, 11, showed moderate PPAR transactivation. The corresponding Z-isomer, 23, and d

Full text of DOI:10.1016/S0968-0896(03)00393-6

Bioorganic & Medicinal Chemistry 11 (2003) 4059–4067
Synthesis and Structure–Activity Relationship Studies of Cinnamic
Acid-based Novel Thiazolidinedione Antihyperglycemic Agents
Partha Neogi,a,* Fredrick J. Lakner,a Satyanarayana Medicherla,b
Jin Cheng,b Debendranath Dey,c Maya Gowri,c Bishwajit Nag,
Somesh D. Sharma, Lesley B. Pickfordb and Coleman Grossd
aDepartment of Chemistry, Calyx Therapeutics Inc., 3513 Breakwater Avenue, Hayward, CA 94545, USA
bDepartment of Physiology, Calyx Therapeutics Inc., 3513 Breakwater Avenue, Hayward, CA 94545, USA
cDepartment of Biochemistry, Calyx Therapeutics Inc., 3513 Breakwater Avenue, Hayward, CA 94545, USA
dDepartment of Clinical Development, Calyx Therapeutics Inc., 3513 Breakwater Avenue, Hayward, CA 94545, USA
Received 11 February 2003; accepted 15 May 2003
Abstract—A number of 2,4-thiazolidinedione derivatives of -phenyl substituted cinnamic acid were synthesized and studied for their
PPAR agonist activity. The E-isomer of cinnamic acid, 11, showed moderate PPAR transactivation. The corresponding Z-isomer,
23, and double bond reduced derivative, 15, were found to be much less potent. Although the E-isomer showed a moderate PPARg
transactivation, it demonstrated a strong glucose-lowering effect in a genetic rodent model of diabetes. Results of pharmacokinetic,
metabolism and permeability studies are consistent with 11 being an active prodrug with an active metabolite, 14, that has similar
glucose lowering and PPARg agonist properties.
# 2003 Elsevier Ltd. All rights reserved.
Introduction
directly result from the altered metabolic milieu in
DM2.
Type 2 diabetes (DM2) is having an increasingly adverse
impact on morbidity, mortality and overall health care
costs in North American and Western European coun-
tries.1,2 The incidence of the disease, currently under-
diagnosed, is expected to reach 215 million by the year
2010.3 Early stages of the disease are characterized by
tissue resistance to the effects of insulin secreted by
pancreatic beta cells. As the disease progresses, the
pancreas’ ability to continue increased production of
insulin diminishes over time. As insulin production
declines in the face of insulin resistance, glucose dis-
posal from the circulation into muscle is diminished
and suppression of hepatic glucose output is decreased.
DM2 is the result of these events and is manifested
clinically by hyperglycemia. The increased and sustained
blood glucose levels eventually give rise to retinopathy,
neuropathy, nephropathy.4 The macrovascular (athero-
sclerotic) complications of DM2 are less closely linked to
hyperglycemia but nonetheless contribute to substantial
morbidity in DM2. The large vessel complications
When recommended dietary modification and exercise
fail to control elevated blood glucose levels, pharmaco-
logical therapy5 is required to restore normoglycemia,
reduce the attendant sequelae and delay the need for
insulin therapy. Pharmacological agents, available or
under investigation, include drugs that improve insulin
resistance or increase insulin secretition. Metformin, a
biguanide, acts primarily by decreasing hepatic glucose
output and increasing peripheral glucose utilization.6 It
is a first line therapeutic option for DM2. Sulfonylureas
stimulate insulin secretion by blocking ATP-dependent
potassium channels7 but are associated with a sig-
nificant risk of hypoglycemia.8 The meglitinides are a
different class of drug but are similar in action to sulfo-
nylureas. Recently introduced thiazolidinediones
(TZDs)9 act by improving peripheral insulin sensitivity.
These agents act as agonists of the peroxisome pro-
liferator-activated receptor gamma (PPARg)10 but their
exact mechanism of action in reducing insulin resistance
is unclear.11 During the last decade, an exhaustive search
for novel agents based on the thiazolidinedione ciglita-
zone12 has yielded several development candidates, as
*Corresponding author. Tel.: +1-510-656-6480; fax: +1-510-780-
1025; e-mail: p.neogi@att.net
0968-0896/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00393-6

4060
P. Neogi et al. / Bioorg. Med. Chem. 11 (2003) 4059–4067
well as the marketed drugs pioglitazone, 1,13 rosiglita-
zone, 2,14 and troglitazone (recently withdrawn from the
market), 3.
We have previously shown that the a-phenyl substituted
cinnamic acid derivative, 4, possessed a weak anti-
hyperglycemic effect, possibly by an interaction with
insulin receptor15 in various animal models.16 We pos-
tulated that linking this molecule with a 2,4-thiazolidi-
nedione (TZD) moiety could provide compounds
retaining the original glucose-lowering activity and,
additionally, an affinity for PPAR, and that such com-
pounds would be useful for treating Type 2 diabetes.
Chart 1.
yielded the a-phenyl substituted cinnamic acid 7 exclu-
sively as the E-isomer. The geometry of the double bond
was confirmed by 1H NMR comparison with the
reported compound.17 Esterification of 7 followed by
condensation with 4-fluorobenzaldehyde yielded 9.
Knovenagel condensation of aldehyde 9 with 2,4-thia-
zolidinedione in the presence of piperidinium benzoate
with azeotropic removal of water gave a good yield of
10.
This paper reports synthesis and biological activities of
2,4-thiazolidinedione analogues of 4 and related cin-
namic and phenylpropionic acid and esters. Compound
11 has shown moderate in vitro PPARg transactivation
and strong glucose-lowering activity in animal model of
Type 2 diabetes (Chart 1).
Results and Discussion
Chemistry
A major challenge was selective hydrogenation of one of
the double bonds in order to produce compounds 11, 17
and 18 (Scheme 2). Reduction of 10 with magnesium/
methanol was nonselective and yielded a mixture of
products. Zinc-acetic acid reduction gave a mixture of
polar products. Hydrogenation with 10% palladium on
The procedure used for synthesis of thiazolidinediones is
shown in Scheme 1. Perkin condensation of 3,5-dime-
thoxybenzaldehyde 5 with 4-hydroxyphenylacetic acid 6
Scheme 1. Reagents and conditions: (a) acetic anhydride, Et3N, 6 h, 130 ꢀC, 47%; (b) MeOH, H2SO4, 20 h, reflux, 97%; (c) 4-fluorobenzaldehyde,
NaH, DMF, 18 h, 80 ꢀC, 77%; (d) 2,4-thiazolidinedione, piperidine, benzoic acid, toluene, 5 h, reflux, 86%; (e) Pt/C(10%), AcOH–HCOONH4, 15 h,
125 ꢀC, 64%; (f) NaBH4, EtOH, 1 h, 25 ꢀC, quantitative; (g) PBr3, CH2Cl2, 25 ꢀC, 1 h, 99%; (h) BuLi, 2,4-thiazolidinedione, THF, 0 ꢀC, 45 min,
15%; (i) aqueous NaOH, MeOH, 15 h, 25 ꢀC, 73%.

P. Neogi et al. / Bioorg. Med. Chem. 11 (2003) 4059–4067
4061
poisoning of the catalyst by the 2,4-thiazolidinedione
moiety in the molecule; the resulting mixtures contained
18 as a minor product. To solve this problem (as shown
in Scheme 2), 8 was first reduced, by using 10% palla-
dium on carbon as catalyst, to 15 quantitatively followed
by coupling with 4-fluorobenzaldehyde and 2,4-thiazoli-
dinedione to furnish 17 in good yield. Reduction of 17
with palladium on carbon catalyst for a longer period of
time and catalyst renewal halfway through the reaction
followed by chromatographic purification over C-18
reverse-phase silica gel produced 18 in moderate yield.
The synthetic strategy adopted to prepare 23, the cor-
responding Z-isomer of 11, is outlined in Scheme 3.
Prolonged heating of 7 with acetic anhydride and tri-
ethylamine19 yielded the corresponding Z-isomer 19 in
13% yield. Compound 19 was converted to the corre-
sponding methyl ester 20 by acid catalyzed esterifica-
tion. Coupling of 20 with 4-fluorobenzaldehyde in
presence of sodium hydride resulted in compound 21.
No significant racemization was observed in afore men-
tioned two steps. Interestingly, the reaction of 2,4-thi-
azolidinedione with 21, in order to produce 22, showed
minimal isomerization of the cinnamic acid double
bond and resulted in a mixture of E- and Z-isomers in a
ratio of 1:7, respectively. Reduction was carried out
without further purification and the final product was
purified by preparative HPLC to yield 23.
Scheme 2. Reagents and conditions: (a) Pd/C (10%), H2, 18 h, 25 ꢀC,
quantitative; (b) 4-fluorobenzaldehyde, NaH, DMF, 18 h, 80 ꢀC, 69%;
(c) 2,4-thiazolidinedione, piperidine, benzoic acid, toluene, 2 h, reflux,
81%; (d) Pd/C (10%), H2 (60 psi), 48 h, 25 ꢀC, 38%.
carbon as catalyst in 1,4-dioxane yielded a mixture of 11
and 18 in a ratio of 6:4. Separation of the compounds
from this mixture was only possible by reverse-phase
chromatography on C-18 silica. These problems were
overcome in several ways. Hydrogenation of 10 using
ammonium formate as hydrogen donor in the presence
of palladium catalyst18 produced minimal amounts of
the over-reduced product 18, and isolation of 11 in high
purity was possible by repeated crystallization from
methanol. Changing the catalyst to 10% platinum on
carbon further reduced byproduct formation and
increased the isolated yield of 11 from 50 to 64%. In
another attempt to make 11, the aldehyde 9 was reduced
to alcohol 12 which upon treatment with PBr3 yielded
the bromo compound 13 in high yield. The bromo
compound was condensed with 2,4-thiazolidinedione
anion generated by BuLi to produce 11 in low yield.
Pharmacological evaluation
Antidiabetic compounds of the TZD class increase per-
ipheral tissue sensitivity to insulin via PPARg activation.
Compound 4 was tested for PPARg agonist activity in a
reporter gene assay at five different concentrations
(0.063, 0.25, 1.0, 4.0, and 16 mM) and did not show any
effect. Analogues of 4 were made by introducing the
TZD moiety into the molecule, with the goal of intro-
ducing PPARg agonist activity. The three possible
methyl ester analogues with double bond(s) at different
positions (10, 11, and 17), the methyl ester without any
double bond 18, the free acid analogue 14 of compound
11, and the Z-isomer 23 of 11 were made and tested.
It was difficult to synthesize 18 in good yield from either
10 or 11 by palladium catalyzed hydrogenation due to
Scheme 3. Reagents and conditions: (a) acetic anhydride, Et3N, 24 h, 125 ꢀC, 13%; (b) MeOH, H2SO4, 18 h, reflux, 35%; (c) 4-fluorobenzaldehyde,
NaH, DMF, 18 h, 80 ꢀC, 74%; (d) 2,4-thiazolidinedione, piperidine, benzoic acid, toluene, 5 h, reflux, 91%; (e) Pd/C (10%), ammonium formate,
acetic acid, 20 h, 120 ꢀC, 15%.

4062
P. Neogi et al. / Bioorg. Med. Chem. 11 (2003) 4059–4067
Agonist activity on PPARg was explored in an in vitro
system using HEK293 cells transfected with a human
PPARg2 expression vector and a PPRE-luciferase
reporter gene (PPAR responsive element-minimal pro-
moter-firefly luciferase chimeric cDNA). Cells were
cotransfected with renilla luciferase to control for
transfection efficiency. Compounds were added to the
transfected cells and over a dilution series of at least
seven concentrations. Cells were cultured for 24 h, then
lysed and activation determined using a luciferase assay
system (Promega, WI) and read in a luminometer. The
results from this in vitro transactivation assay for the
tested compounds are summarized in Table 1. The most
potent compound in this series was found to be 11
(EC50 of 0.28 mM) which had approximately one-thir-
tieth the activity of rosiglitazone, 2 (EC50 of 0.009 mM)
in the same assay. The maximal activity of compound
11 was equivalent in magnitude to rosiglitazone. The
potency for compounds 10, containing two double
bonds and 14, the corresponding acid analogue of 11
were less than that of compound 11. Both the com-
pounds 17 and 18 were devoid of the cinnamic acid
double bond and essentially inactive. The Z-isomer 23
showed less than one-tenth the potency of compound 11
in this assay. This indicates that the geometry of the
double bond is critical to activity.
Based on these in vitro results, compound 11 was eval-
uated in the ob/ob mouse model of Type 2 diabetes for
its oral antihyperglyemic activity. The effects of 11 at a
dose of 10 mg/kg body weight were compared to vehi-
cle, as well as to comparators of the same class, piogli-
tazone, 1 (20 mg/kg), rosiglitazone, 2 (10 mg/kg), and
troglitazone, 3 (100 mg/kg), used at dose levels known
to be maximally effective. Compounds were adminis-
tered by oral gavage for nine days to groups of five
young male ob/ob diabetic mice (Fig. 1). Compound 11
at 10 mg/kg rapidly lowered blood glucose from
284Æ57 mg/dL to a low of 151Æ43 mg/dL on Day 6,
and maintained stable levels in the normal range
throughout the study. Compounds 1, 2, 3, and 11
showed statistically and physiologically significant glu-
cose decreases at the end of the treatment period of 25,
42, 26 and 29% from their Day 0 value, respectively.
Untreated animals showed a considerable increase in
blood glucose levels during this study. Thus compound
11 exhibited marked antihyperglycemic activity when
administered orally for 9 days to the ob/ob mice and was
active in the same range as other compounds of this
class. In comparison, oral administration of aqueous
solution of compound 4 at a dose of 20 mg/kg for 9
days to ob/ob mice produced only a 19% decrease in
glucose from the Day 0 value (Fig. 2).
Table 1. Induction of PPARg mediated luciferase activity by thiazo-
lidinedionesa
Compd
EC50 (mM+SD)
1 (rosiglitazone)
0.009Æ0.007
1.136
10
11
14
17
18
23
0.284Æ0.036 (n=5)b
0.690Æ0.038 (n=2)
23.9
57.7
3.686Æ1.454 (n=2)
In dose range-finding study, ob/ob diabetic mice were
orally administered 11 at three different dose levels (3.1,
6.3, and 12.5 mg/kg body weight) or troglitazone 3 at a
dose of 12.5 mg/kg for 19 days. All dose levels of
aResults of several independent experiments. Each experiment con-
tains at least seven different concentrations of drug treatment (between
0.1 to 30 mM) and each concentration was tested in triplicate. EC50
values were calculated by non-linear regression analysis using Graph-
Pad Prism (San Diego, CA, USA) software.
bn, number of independent experiments.
Figure 1. Blood glucose: group mean blood glucose levels in ob/ob
mice administered vehicle, 11, or comparators pioglitazone, 1, rosigli-
tazone, 2 and troglitazone, 3. ***p<0.001, *p<0.05.
Figure 2. Group meanÆSD blood glucose concentration in ob/ob
mice after administration of 4 or vehicle for 8 days.

P. Neogi et al. / Bioorg. Med. Chem. 11 (2003) 4059–4067
4063
Table 2. Percent change in group mean blood glucose compared with
control group
proposal that 11 is a prodrug of 14 in vivo. The simi-
larity of activity of 11 and 14 as PPARg agonists and in
lowering glucose when administered directly in vitro
assays, where there is no conversion of the compounds,
together with the metabolism data, suggests that com-
pound 14 is an active metabolite of 11, which can be
therefore be considered an active prodrug. Further, in
vitro permeability studies in Caco-2 cells have demon-
strated that compound 11 is about 17-fold higher in
absorption potential than 14. Thus compound 11 would
be the preferred form for oral administration, compared
with 14, but the net pharmacologic effect would appear
to be derived from a combination of both compounds.
Group
Percent decrease time (days)
0a
3
7
10
15
19
Compound 11 (3.1 mg/kg)
Compound 11 (6.3 mg/kg)
Compound 11 (12.5 mg/kg)
Compound 3 (12.5 mg/kg)
À0.77
16
37
49
44
3
21
43
42
34
25
40
40
30
35
55
48
23
1
1
1
30
21
23
aOn day 0 all group mean blood glucose wereÆ1% of vehicle.
Conclusion
We have identified a new series of thiazolidine-1,4-dione
substituted a-phenyl cinnamic acids with moderate
PPARg agonist activity showing strong oral glucose
lowering effects in animal models of type 2 diabetes. The
data suggests that the presence of the cinnamic acid
double bond as well as its geometry is very important
for PPAR agonism. Thus cinnamic acid based TZDs
can provide lead compounds to develop new anti-
hyperglycemic agents.
Experimental
Chemistry
Figure 3. Group meanÆSD blood glucose concentration in ob/ob
mice after administration of 11, 14 or vehicle for 11 days.
General methods. Melting points were measured on a
Mel-Temp melting point apparatus and are uncor-
rected. The 1H NMR and 13C NMR spectra were
recorded on a JEOL Eclipse (400 MHz) or Nicolet NT
36 (360 MHz) spectrometer and are reported as parts
per million (ppm) downfield from TMS. The infrared
spectra were recorded on a Nicolet Impact 410 FT-IR
spectrophotometer. The mass spectra were recorded on
a Fison VG Platform II of HP 1100 MDS 1964A mass
spectrophotometer. UV spectra were recorded on a
Beckman DU650 spectrophotometer. TLC was per-
formed on Merck silica gel F254 precoated plates. The
silica gel used for column chromatography was ‘Baker’
silica gel (40 mm) for flash chromatography.
compound 11 were effective in reducing blood glucose
at Day 19 by 35, 55, and 47%, from the vehicle group
(pꢁ0.01, 0.001, 0.001, respectively). A dose–time rela-
tionship was evident, and overall blood glucose con-
centrations were maximally reduced at 6.3 mg/kg.
Troglitazone decreased blood glucose by 23–44% over
the course of the study. The data is summarized in
Table 2. Details of this study and additional biological
activities of compound 11 are presented elsewhere.20
When compound 11 and its corresponding free acid 14
were orally administered to ob/ob mice at a dose of 5
mg/kg per day for 11 days, a comparable glucose low-
ering and normalizing effect was observed for both the
groups compared with vehicle-treated controls (Fig. 3).
There were no differences in body weights between the
two groups over this period. Both the ester 11 and the
acid 14 showed similar PPARg agonist activity and
potency (EC50 of 0.28Æ0.14 mM and 0.59Æ0.17 mM,
respectively), and both the compounds showed similar
glucose lowering effect. Compound 11 is extensively and
rapidly metabolized in mice to compound 14. In a PK
study in male mice dosed orally with compound 11 both
11 and 14 were detected in plasma from 2 min onwards.
Terminal half lives of 11 and 14 were 6.5Æ0.7 and
3-(3,5-Dimethoxyphenyl)-2-(4-hydroxyphenyl)-acrylic
acid (7). To a mixture of 3,5-dimethoxybenzaldehyde,
5, (500 g, 3.0 mol) and 4-hydroxyphenylacetic acid, 6,
(457 g, 3.0 mol) was added acetic anhydride (1.0 L, 10.6
mol) and triethylamine (420 mL, 3.0 mol). After stirring
at 130–140 ꢀC for 6 h, the mixture was cooled to room
temperature. Concentrated HCl (1 L) was added to the
reaction mixture slowly over 50 min while keeping the
temperature between 20 and 30 ꢀC. The light-yellow
precipitate obtained was filtered and washed with water.
The solid was dissolved in 3 N NaOH (5 L) and stirred
for 1 h and filtered. The filtrate was acidified to pH 1
with concentrated HCl while maintaining a temperature
at 25–30 ꢀC. The precipitated product was filtered and
washed with water to give crude product that was
recrystallized from MeOH–H2O and dried at 40 ꢀC for
11.1Æ4.8 h and plasma AUC0À24
of 5.9Æ1.3 and
h
85.8Æ7.5 mg h/mL, respectively. Only 7% of the total
mass of compound 11 and 14 was detectable as com-
pound 11 in the blood of male mice.21 These results
indicate rapid conversion of 11 to 14 and support the

4064
P. Neogi et al. / Bioorg. Med. Chem. 11 (2003) 4059–4067
6 h to yield 7 (428 g, 47%): mp 225–227 ꢀC (lit.
226–228 ꢀC);17 1H NMR (360 MHz, DMSO-d6) d 12.48
(br s, 1H), 9.42 (s, 1H), 7.59 (s, 1H), 6.95 (d, J=8.0 Hz,
2H), 6.76 (d, J=8.0 Hz, 2H), 6.35 (t, J=2.2 Hz, 1H),
6.27 (d, J=2.2 Hz, 2H), 3.56 (s, 6H); MS (EI) m/z 299
[M]À.
NMR (360 MHz, DMSO-d6) d 12.53 (br s, 1H), 7.78 (s,
1H), 7.73 (s, 1H), 7.63 (d, J=9.2 Hz, 2H), 7.25 (d,
J=9.2 Hz, 2H), 7.13 (overlapped d, J=8.3 Hz, 2H),
7.11 (overlapped d, J=8.6 Hz, 2H), 6.42 (t, J=2.2 Hz,
1H), 6.27 (d, J=2.2 Hz, 2H), 3.73 (s, 3H), 3.59 (s, 6H);
MS (EI) m/z 518 [M]+. Anal. (C28H23NO7S) C, H, N.
3-(3,5-Dimethoxyphenyl)-2-(4-hydroxyphenyl)-acrylic
acid methyl ester (8). Methanol (3.0 L) was added to a
thoroughly dried 7 (427.5 g, 1.42 mol) under argon. To
this stirred suspension concentrated sulfuric acid (100
mL) was added and heated at reflux for 20 h under
nitrogen. Methanol was evaporated under reduced
pressure at 30 ꢀC. The residue was taken up in ethyl
acetate (3.0 L) and washed with water (2Â1.0 L), satu-
rated aqueous NaHCO3 (2Â1.0 L), brine (2Â1.0 L). The
organic layer was dried on anhydrous magnesium sul-
fate, filtered and the solvent was evaporated. Com-
pound 8 was obtained after drying under high vacuum
as white solid, (433.6 g, 97%): mp 106–108 ꢀC; 1H NMR
(360 MHz, CDCl3) d 7.72 (s, 1H), 7.06 (d, J=7.9 Hz,
2H), 6.77 (d, J=7.9 Hz, 2H), 6.33 (t, J=2.2 Hz, 1H),
6.26 (d, J=2.2 Hz, 2H), 5.74 (s, 1H), 3.81 (s, 3H), 3.60
(s, 6H); MS (EI) m/z 315 [M]+. Anal. (C18H18O5) C, H.
3-(3,5-Dimethoxyphenyl)-2-{4-[4-(2,4-dioxothiazolidin-5-
ylmethyl)phenoxy]-phenyl}-acrylic acid methyl ester (11).
To a solution of 10 (599 g, 1.16 mol) in glacial acetic
acid (11.5 L), ammonium formate (4.0 kg, 62.9 mol) was
added and stirred for 30 min. A slurry of Pd on carbon
(10%, dry, 300 g) in glacial acetic acid (500 mL) was
added to the flask and heated at 120 ꢀC for 24 h fol-
lowed by stirring at room temperature for 48 h. The
resulting mixture was filtered through a bed of Celite1.
The filtrate was poured slowly into vigorously stirred
water (12 L) and the solid that separated solid was fil-
tered and dried. The resulting solid was purified by
slurring twice in hot methanol and once in hot ethanol
to yield 11 as white solid (296 g, 49.2%): mp
126–128 ꢀC; IR (KBr) nmax 3200, 2950, 2850, 1700,
1
1600, 1500, 1350, 1150, 850 cmÀ1. H NMR (400 MHz,
DMSO-d6) d 12.01 (br s, 1H), 7.73 (s, 1H), 7.28 (d,
J=8.6 Hz, 2H), 7.19 (d, J=8.6 Hz, 2H), 7.02 (d, J=8.6
Hz, 2H), 6.96 (d, J=8.6 Hz, 2H), 6.40 (t, J=2.2 Hz,
1H), 6.27 (d, J=2.2 Hz, 2H), 4.92 (dd, J=9.2 and 4.4
Hz, 1H), 3.73 (s, 3H), 3.57 (s, 6H), 3.37 (dd, J=14.8 and
4.3 Hz, 1H), 3.12 (dd, J=14.8 and 9.4 Hz, 1H); MS (EI)
m/z 518 [MÀH]À, 265, 249, 113. Anal. (C28H25NO7S) C,
H, N.
3 - (3,5 - Dimethoxyphenyl) - 2- [4 - (4 - formylphenoxy) -
phenyl]-acrylic acid methyl ester (9). Under argon, com-
pound 8 (433.0 g, 1.37 mol) was dissolved in dry DMF
(1.6 L) and sodium hydride (60.4 g, 1.51 mol) was
added. To the resulting orange solution, 4-fluoro-
benzaldehyde (185.0 mL, 1.71 mol) was added and
heated at 80 ꢀC for 18 h. The reaction mixture was
cooled to room temperature, diluted with ethyl acetate
(3.0 L) and extracted with water (3Â1.0 L), then brine
(1Â1.0 L). The organic layer was dried over anhydrous
sodium sulfate, filtered and solvent was evaporated. The
residue was suspended in methanol (3.0 L), stirred
overnight, solid was filtered and dried under vacuum at
40 ꢀC to yield 9 as pale-yellow solid (445 g, 77%): mp
108–110 ꢀC; 1H NMR (360 MHz, CDCl3) d 9.94 (s, 1H),
7.86 (d, J=8.6 Hz, 2H), 7.80 (s, 1H), 7.28 (d, J=8.6 Hz,
2H), 7.11 (overlapped d, J=9.0 Hz, 2H), 7.08 (over-
lapped d, J=9.0 Hz, 2H), 6.36 (t, J=2.2 Hz, 1H), 6.25
(d, J=2.2 Hz, 2H), 3.83 (s, 3H), 3.63 (s, 6H). Anal.
(C25H22O6) C, H.
3-(3,5-Dimethoxyphenyl)-2-{4-[4-(2,4-dioxothiazolidin-5-
ylmethyl)-phenoxy]-phenyl}-acrylic acid methyl ester
(11). Compound 10 (20 g, 38.6 mmol), ammonium for-
mate (150 g, 2.38 mol), 10% Pt/C (dry, 4 g) and acetic
acid (660 mL) were combined into a round-bottom flask
equipped with reflux condenser, thermometer and
mechanical stirrer. The reactor was evacuated and
purged three times with nitrogen then, heated to a
steady reflux (ca. 124 ꢀC). Reaction was completed
within 15 h and allowed to cool with stirring to ambient
room temperature. After cooling, the mixture was fil-
tered though a pad of Celite1 (5 g) and the filter pad
washed with fresh acetic acid (2Â100 mL). The mother
liquor and washes were combined and concentrated.
The residue was then diluted with dichloromethane (400
mL), and the combined organics were extracted twice
with water (400 mL) and 5% bicarbonate (400 mL). The
organic portion was then dried and poured through
silica gel (30 g) and washed with dichloromethane
(2Â100 mL). The washes were combined and con-
centrated. The residue was diluted with ethanol, allowed
to cool to 60 ꢀC and seed crystals were added. This
slurry was stirred at 50 ꢀC for about 30 min then
allowed to cool to ambient room temperature to yield
compound 11 (12.85 g, 64%) with an HPLC assay of
98.1%.
3-(3,5-Dimethoxyphenyl)-2-{4-[4-(2,4-dioxothiazolidin-5-
ylidenmethyl)-phenoxy]-phenyl}-acrylic acid methyl ester
(10). To a stirred suspension of 9 (352 g, 0.82 mol) in
anhydrous toluene (2.5 L), 2,4-thiazolidinedione (98.6 g,
0.84 mol), benzoic acid (134 g, 1.10 mol) and piperidine
(107.4 g, 1.26 mol) was added sequentially and heated at
reflux temperature with continuous removal of water
with the help of Dean–Stark apparatus for 5 h. Toluene
(1.0 L) was removed from the reaction mixture and
cooled overnight at 4ꢀC. The solid that separated was
filtered and mother liquor was evaporated to dryness
under reduced pressure. The residue obtained was re-
dissolved in a mixture of MeOH–diethyl ether (1 : 1, 3.0
L). On standing overnight at 4 ꢀC, the solution yielded
more solids. The solid from both lots were combined
and dried overnight in vacuum oven at 40 ꢀC to give 10
as yellow solid (362.5 g, 86%): mp 106–108 ꢀC; 1H
3 - (3,5 - Dimethoxyphenyl) - 2[4 - (4 - hydroxymethylphe-
noxy)-phenyl]-acrylic acid methyl ester (12). Compound
9 (5.0 g, 11.9 mmol) was suspended in anhydrous etha-
nol (60 mL) at room temperature and sodium borohy-

P. Neogi et al. / Bioorg. Med. Chem. 11 (2003) 4059–4067
4065
dride (0.23 g, 6.1 mmol) was added with efficient stir-
ring. The reaction was followed by TLC and was com-
plete in 1 h, solvent was evaporated and the residue was
dissolved in ethyl acetate (100 mL). The organic layer
was extracted with water (50 mL), washed with brine
(25 mL) brine, dried on anhydrous magnesium sulfate,
filtered and solvent was evaporated yield the title com-
pound 12 as white solid (5.1 g, 100%): mp 93–95 ꢀC. 1H
NMR (400 MHz, DMSO-d6) d 7.72 (s, 1H), 7.35 (d,
J=8.8 Hz, 2H), 7.19 (d, J=8.8 Hz, 2H), 7.02 (d, J=8.4
Hz, 2H), 7.00 (d, J=8.4 Hz, 2H), 6.41 (t, J=2.4 Hz,
1H), 6.29 (d, J=2.0 Hz, 2H), 5.18 (t, J=6.4 Hz, 1H),
4.49 (d, J=4.8 Hz, 2H), 3.73 (s, 3H), 3.57 (s, 6H); MS
(EI) m/z 315 [M]+. Anal. (C25H24O6) C, H.
recrystallized from ethanol to give 14 as white solid
(7.14 g, 73%): mp 138–140 ꢀC. 1H NMR (400 MHz,
DMSO-d6) d 7.69 (s, 1H), 7.28 (d, J=8.8 Hz, 2H), 7.19
(d, J=8.8 Hz, 2H), 7.02 (d, J=8.8 Hz, 2H), 6.97 (d,
J=8.8 Hz, 2H), 6.41 (t, J=2.4 Hz, 1H), 6.28 (d, J=2.4
Hz, 2H), 4.92 (dd, J=9.2 and 4.4 Hz, 1H), 3.58 (s, 6H),
3.38 (dd, J=14.0 and 4.0 Hz, 1H), 3.13 (dd, J=14.4 and
9.2 Hz, 1H); MS (EI) m/z 506 [M]+. Anal.
(C27H23NO7S) C, H.
3-(3,5-Dimethoxyphenyl)-2-(4-hydroxyphenyl)-propionic
acid methyl ester (15). To a suspension of 8 (6.28 g,
20.0 mmol) in ethanol (200 mL), palladium on carbon
(10%, wet, 0.63 g) was added and the mixture was stir-
red under H2 at atmospheric pressure at room tem-
perature for 18 h. The catalyst was filtered through a
bed of Celite1 and solvent was evaporated under
reduced pressure ꢀto yield 15 as white solid (6.32 g,
100%): mp 63–65 C. 1H NMR (400 MHz, DMSO-d6) d
7.15 (d, J=8.7 Hz, 2H), 6.74 (d, J=8.7 Hz, 2H), 6.29 (t,
J=2.4 Hz, 1H), 6.25 (d, J=2.4 Hz, 2H), 3.78 (t, J=8.7
Hz, 1H), 3.72 (s, 6H), 3.62 (s, 3H), 3.31 (dd, J=13.5 and
8.4 Hz, 1H), 2.93 (dd, J=13.5 and 6.9 Hz, 1H). MS (EI)
m/z 317 [M]+. Anal. (C18H20O5) C, H.
2-[4-(4-Bromomethylphenoxy)-phenyl]- 3-(3,5-dimethoxy-
phenyl)-acrylic acid methyl ester (13). A solution of
PBr3 (4.8 mL of 1.0 M in CH2Cl2) was added dropwise
to 12 (5.0 g, 11.9 mmol) dissolved in CH2Cl2 (20 mL) at
room temperature with good stirring. After 1 h, the
solution was extracted with water (2Â60 mL) and brine
(20 mL). The organic phase was dried over anhydrous
magnesium sulfate, filtered through a small bed of silica
gel (20 g) and solvent was evaporated. The resulting
tacky syrup was dried under high vacuum for 48 h at
room temperature to yield compound 13 (5.7 g, 99.0%):
mp 79–81 ꢀC. 1H NMR (400 MHz, DMSO-d6) d 7.73 (s,
1H), 7.49 (d, J=8.4 Hz, 2H), 7.22 (d, J=8.4 Hz, 2H),
7.07 (d, J=8.4 Hz, 2H), 7.00 (d, J=8.4 Hz, 2H), 6.42 (t,
J=2.4 Hz, 1H), 6.28 (d, J=2.0 Hz, 2H), 4.73 (d, J=4.8
Hz, 2H), 3.68 (s, 3H), 3.58 (s, 6H). Anal. (C25H23BrO5)
C: calcd 61.12; found 62.26.
3 - (3,5 - Dimethoxyphenyl) - 2- [4 - (4 - formylphenoxy) -
phenyl]-propionic acid methyl ester (16). To a suspen-
sion of sodium hydride (60% in oil, 0.25 g, 6.3 mmol) in
DMF (2 mL) under argon, 15 (2.0 g, 6.3 mmol) in dry
DMF (3 mL) was added. To the resulting yellow solu-
tion, 4-fluorobenzaldehyde (0.68 mL, 6.3 mmol) was
added and heated at 80 ꢀC for 18 h. The reaction mix-
ture was cooled to room temperature and water (20 mL)
was added and extracted with ethyl acetate (3Â50 mL).
The organic layer was dried over anhydrous sodium
sulfate, filtered and the solvent was evaporated. The
ethyl acetate solution of crude product was filtered
through a small bed of silica gel to yield 16 (1.83 g,
3-(3,5-Dimethoxyphenyl)-2-{4-[4-(2,4-dioxothiazolidin-5-
ylmethyl)phenoxy] phenyl}-acrylic acid methyl ester (11).
2,4-Thiazolidinedione (2.83 g, 24.2 mmol) was dissolved
in dry THF (170 mL) and cooled to 0 ꢀC under argon.
Butyllithium (1.6 M in hexanes, 30 mL, 48.0 mmol) was
added dropwise and stirring was continued for 0.5 h at
0 ꢀC. Under argon, a solution of 13 (5.7 g, 11.8 mmol) in
dry THF (30 mL) was added rapidly via syringe to the
reaction mixture with rapid stirring. The temperature
was maintained at 0 ꢀC for 45 min before quenching
with aqueous HCl (5%, 40 mL). Additional H2O (40
mL) was added and the mixture was extracted with
ethyl acetate (3Â30 mL). The organic layers were com-
bined, washed with brine, dried over anhydrous mag-
nesium sulfate, filtered and the solvent was evaporated.
Flash chromatography over silica gel using hexanes-
ethyl acetate (3:2) as eluting solvent yielded the title
compound, 11, (0.93 g, 15%). Melting point and 1H
NMR of compound 11 made by this method was iden-
tical with 11 produced from 10.
1
69%) as oil. H NMR (400 MHz, DMSO-d6) d 9.91 (s,
1H), 7.84 (d, J=8.7 Hz, 2H), 7.33 (d, J=8.7 Hz, 2H),
7.04 (d, J=5.4 Hz, 2H), 7.01 (d, J=5.4 Hz, 2H), 6.30 (t,
J=2.1 Hz, 1H), 6.25 (d, J=2.1 Hz, 2H), 3.86 (t, J=7.8
Hz, 1 Hz), 3.76 (s, 6H), 3.66 (s, 3H), 3.36 (dd, J=12.6
and 8.1 Hz, 1H), 2.97 (dd, J=13.5 and 7.5 Hz, 1H). MS
(EI) m/z 421 [M]+.
3-(3,5-Dimethoxyphenyl)-2-{4-[4-(2,4-dioxothiazolidin-5-
ylidenmethyl)-phenoxy]-phenyl}-propionic acid methyl es-
ter (17). To a stirred suspension of 16 (1.81 g, 4.3 mmol)
in anhydrous toluene (25 mL), 2,4-thiazolidinedione
(0.56 g, 4.74 mmol), benzoic acid (0.68 g, 5.60 mmol)
and piperidine (0.60 mL, 6.03 mmol) were added
sequentially and heated at reflux temperature with con-
tinuous removal of water (Dean–Stark apparatus) for 2
h. The solvent was evaporated to dryness under reduced
pressure the residue was purified by silica gel chroma-
tography eluting with hexane–ethyl acetate (1:1) to yield
3-(3,5-Dimethoxyphenyl)-2-{4-[4-(2,4-dioxothiazolidin-5-
ylmethyl)-phenoxy]-phenyl acrylic acid (14). To a stirred,
cooled below 10 ꢀC, suspension of 11 (10 g, 19.27 mmol)
in methanol (50 mL), aqueous sodium hydroxide (2 N,
33.7 mL, 67.4 mmol) was added and stirring was con-
tinued for 15 h at room temperature. The resulting pale-
yellow solution was cooled to 10 ꢀC and acidified with
aqueous HCl (5%, 115 mL). The solid that separated
was filtered and washed with water (3Â30 mL), and
17 (1.82 g, 81%): mp 104–106 ꢀC. H NMR (400 MHz,
1
DMSO-d6) d 12.53 (br s, 1H), 7.76 (s, 1H), 7.60 (d,
J=8.7 Hz, 2H), 7.35 (d, J=8.7 Hz, 2H), 7.07 (d, J=4.8
Hz, 2H), 7.03 (d, J=4.8 Hz, 2H), 6.33–6.28 (m, 3H),
4.01 (t, J=7.5 Hz, 1 Hz), 3.66 (s, 6H), 3.56 (s, 3H), 3.22
(dd, J=13.8 and 8.4 Hz, 1H), 2.90 (dd, J=13.5 and 7.2

4066
P. Neogi et al. / Bioorg. Med. Chem. 11 (2003) 4059–4067
Hz, 1H). MS (EI) m/z 520 [M]+. Anal. (C28H25NO7S)
C: calcd 64.73; found 65.89.
yield 20 as white solid (0.24 g, 35%). 1H NMR
(400 MHz, CDCl3) d 7.26 (d, J=8.4 Hz, 2H), 6.82 (s,
1H), 6.76 (d, J=8.4 Hz, 2H), 6.45 (d, J=2.0 Hz, 2H),
6.34 (t, J=2.0 Hz, 1H), 4.97 (s, 1H), 3.73 (s, 3H), 3.72
(s, 6H).
3-(3,5-Dimethoxyphenyl)-2-{4-[4-(2,4-dioxothiazolidin-5-
ylmethyl)-phenoxy]-phenyl}-propionic acid methyl ester
(18). Compound 17 (1.6 g, 3.08 mmol) was dissolved in
dioxane (45 mL) and transferred in a hydrogenation
bottle and Pd on carbon (10%, 1.0 g) was added.
Hydrogenation was done at 65 psi for 34 h. Following
this period, additional Pd on carbon (10%, 0.6 g) was
added and hydrogenation was allowed to continue for
another 18 h. Catalyst was filtered through a bed of
Celite1 and solvent was evaporated. The residue was
purified by column chromatography on reverse phase
silica gel (C-18) using acetonitrile–water (1:1) mixture to
elute 18 as white solid (0.60 g, 38%): mp 125–128 ꢀC. 1H
NMR (400 MHz, DMSO-d6) d 12.04 (br s, 1H), 7.31 (d,
J=8.4 Hz, 2H), 7.25 (d, J=8.4 Hz, 2H), 6.92 (d, J=8.6
Hz, 4H), 6.30 (d, J=2.0 Hz, 2H), 6.29 (t, J=2.0 Hz,
1H), 4.90 (dd, J=9.2 and 4.4 Hz, 1H), 3.98 (t, J=8.0
Hz, 1H), 3.67 (s, 6H), 3.56 (s, 3H), 3.37 (dd, J=13.6 and
4.0 Hz, 1H), 3.21 (dd, J=14.0 and 8.8 Hz, 1H); 3.11
(dd, J=14.0 and 9.2 Hz, 1H), 2.90 (dd, J=13.6 and 7.6
Hz, 1H). MS (EI) m/z 522 [M]+. Anal. (C28H27NO7S)
C, H, N.
Z-3-(3,5-Dimethoxyphenyl)-2-[4-(4-formylphenoxy)-
phenyl]-acrylic acid methyl ester, (21). Under argon,
compound 20 (0.60 g, 1.9 mmol) was dissolved in dry
DMF (4 mL) and sodium hydride (60% in oil, 0.09 g,
2.28 mmol) was added. To the resulting orange solution,
4-fluorobenzaldehyde (0.25 mL, 2.28 mmol) was added
and heated at 80 ꢀC for 18 h. The reaction mixture was
cooled to room temperature and water (10 mL) was
added. The mixture was extracted with ethyl acetate
(3Â20 mL) and the crude product obtained after eva-
poration was purified by chromatography over silica gel
and elution with a mixture of hexane–ethyl acetate (4:1)
to yield 21 as white solid (0.59 g, 74%). 1H NMR
(400 MHz, CDCl3) d 9.93 (s, 1H), 7.86 (d, J=8.8 Hz,
2H), 7.49 (d, J=8.8 Hz, 2H), 7.10 (overlapped d, J=8.8
Hz, 2H), 7.08 (overlapped d, J=8.8 Hz, 2H), 6.96 (s,
1H), 6.53 (dd, J=2.8 Hz, 2H), 6.43 (t, J=2.0 Hz, 1H),
3.80 (s, 3H), 3.79 (s, 6H).
Z-3-(3,5-Dimethoxyphenyl)-2-{4-[4-(2,4-dioxothiazoli-
din-5-ylidenmethyl)-phenoxy]-phenyl}-acrylic acid methyl
ester (22). To a stirred suspension of 21 (0.53 g, 1.3
mmol) in anhydrous toluene (10 mL), 2,4-thiazolidine-
dione (0.15 g, 1.30 mmol), benzoic acid (0.21 g, 1.69
mmol) and piperidine (0.19 g, 1.26 mol) were added
sequentially and the mixture was heated at reflux tem-
perature with continuous removal of water (Dean–Stark
apparatus) for 5 h. Toluene was evaporated and the
residue was chromatographed over silica gel eluting
with hexane–ethyl acetate (1:1) to yield a mixture of 22
and 10 (0.60 g, 91%) in a ratio of 7:1 on the basis of
Z-3-(3,5-Dimethoxyphenyl)-2-(4-hydroxyphenyl)-acrylic
acid (19). E-3-(3,5-Dimethoxyphenyl)-2-(4-hydroxy-
phenyl)-acrylic acid, 7 (10.0 g, 0.33 mmol) was dissolved
in a mixture of acetic anhydride (40 mL, 0.42 mol) and
triethylamine (40 mL, 0.29 mol) and heated at 125 ꢀC
for 24 h.19 The mixture was cooled to room temperature
and ethyl acetate (150 mL) was added, the mixture was
cooled to 5 ꢀC, acidified with concentrated HCl (30 mL)
and stirred for 90 min. The organic layer was separated
and the aqueous layer was extracted with ethyl acetate
(3Â100 mL). The combined organic layers were washed
with water (2Â50 mL) and extracted with aqueous
NaOH (5M, 3Â70 mL). The aqueous alkaline layer was
acidified with glacial acetic acid (65 mL) to pH 5.2 and
stirred at 0 ꢀC for 30 min. The solid that separated was
filtered and mother liquor was acidified with con-
centrated HCl (90 mL) and stirred at 5ꢀC for 1 h. The
solid that separated was filtered, washed with cold water
(2 ꢀ50 mL) and drieꢀd at 45 ꢀC for 6 h to yield, 19, (1.3 g,
1
proton NMR analysis. H NMR (400 MHz, DMSO-d6)
d 12.53 (br s, 1H), 7.79 (s, 1H), 7.65 (d, J=8.8 Hz, 2H),
7.54 (d, J=8.8 Hz, 2H), 7.18 (overlapped d, J=8.8 Hz,
2H), 7.16 (overlapped d, J=8.8 Hz, 2H), 7.17 (over-
lapped s, 1H), 6.57 (d, J=2.0 Hz, 2H), 6.50 (t, J=2.0
Hz, 1H), 3.79 (s, 3H), 3.75 (s, 6H).
Z-3-(3,5-Dimethoxyphenyl)-2-{4-[4-(2,4-dioxothiazoli-
din-5-ylmethyl)-phenoxy]-phenyl}-acrylic acid methyl es-
ter (23). To a solution of 22 (0.60 g, 1.6 mmol) in acetic
acid (15 mL) was added Pd on carbon (10%, 300 mg)
and ammonium formate (4.3 g, 55.8 mmol) and the
mixture was heated at 120 ꢀC for 20 h. Catalyst was fil-
tered through a bed of Celite1 and acetic acid was eva-
porated under reduced pressure. Water (50 mL) was
added to the residue and solid separated was filtered.
Pure Z-isomer 23 was isolated by preparative HPLC
using Intersil ODS-3 preparative column (250Â4.6 mm,
5 mm) running at a rate of 15 mL per min using metha-
nol/acetonitrile/water (3:3:2) containing formic acid
(0.05%), purity 99.0%, (0.09 g, 15.0%): mp 65–66 ꢀC.
1H NMR (400 MHz, DMSO-d6) d 12.05 (br s, 1H), 7.48
(d, J=9.2 Hz, 2H), 7.29 (d, J=8.4 Hz, 2H), 7.13 (s,
1H), 7.03 (overlapped d, J=8.8 Hz, 2H), 7.01 (over-
lapped d, J=8.4 Hz, 2H), 6.56 (d, J=2.0 Hz, 2H), 6.49
(t, J=2.0 Hz, 1H), 4.90 (dd, J=9.2 and 4.4 Hz, 1H),
1
13%): mp 135–137 C. H NMR (400 MHz, DMSO-d6)
d 13.28 (br, 1H), 9.70 (br, 1H), 7.32 (d, J=10.4 Hz, 2H),
6.81(s, 1H, overlapped), 6.79 (d, J=9.7 Hz, 2H), 6.67
(d, J=2.5 Hz, 2H), 6.64 (t, J=2.5 Hz, 1H), 3.73 (s, 6H).
MS (EI) m/z 299 [M]À.
Z-3-(3,5-Dimethoxyphenyl)-2-(4-hydroxyphenyl)-acrylic
acid methyl ester (20). Concentrated sulfuric acid (10
drops) was added to a stirred methanol suspension of
thoroughly dried 19 (0.60 g, 2.0 mmol) under argon and
heated at reflux for 18 h. Methanol was evaporated
under reduced pressure, the residue was taken up in
ethyl acetate (20 mL) and washed with water (20 mL),
saturated aq NaHCO3 (10 mL) and brine (10 mL). The
organic layer was dried on anhydrous magnesium sul-
fate, filtered and the solvent was evaporated. The crude
product obtained was purified by chromatography over
silica gel and eluted with hexane–ethyl acetate (7:3) to

P. Neogi et al. / Bioorg. Med. Chem. 11 (2003) 4059–4067
4067
3.77 (s, 3H), 3.75 (s, 6H), 3.38 (dd, J=14.8 and 4.8 Hz,
1H), 3.13 (dd, J=14.4 and 9.2 Hz, 1H). MS (EI) m/z
518 [MÀH]À.
the Calyx Therapeutics Institutional Animal Care and
Use Committee. Animals were housed at 22 ꢀC and 50%
relative humidity, with a 12-h light and dark cycle, and
received a regular rodent diet (Harlan Teklad, Madison,
WI, USA) ad libitum with free access to water. Male
C57BL/6J-ob/ob mice were obtained from Jackson
Laboratories (Bar Harbor, ME, USA) when their age
was 5 weeks. Groups of 5–10 7–8-week-old animals were
orally dosed with suspensions of test compounds, or
vehicle, 0.5% carboxymethyl cellulose (CMC, Sigma, St.
Louis, MO, USA) in water or phosphate-buffered saline
(PBS), in some cases containing PEG-400 (10%), once
daily by gavage. Body weights were recorded periodically,
and animals were observed daily. Blood glucose measure-
ments were made with One Touch Glucose Meter (Life
Scan, Inc., Milpitas, CA, USA) prior to administering the
next dose or 24 h after the last dose and in the fed state.
Pioglitazone 1, rosiglitazone, 2, and troglitazone, 3,
used were isolated by purification of commercially
available Actose1, Avandia1 and Rezulin1 respec-
tively. Purity and identity of all the three compounds
were checked by HPLC and 1H NMR.
Biology
Transfection and luciferase activity assay. Human
PPARg2 expression vector was constructed by inserting
PPARg2 encoding region into pcDNA3.1+ vector
(Invitrogen, Carlsbad, CA, USA). Luciferase reporter
vector was constructed by ligating PPRE response ele-
ment upstream of the Firefly luciferase coding region.
Control vector pRL-SV40 expressing Renilla luciferase
was purchased from Promega (Madison, WI, USA).
About 2.7Â104 HEK293 cells (ATCC, Manassas, VA,
USA) were plated into a 35 mm culture well and main-
tained in Eagle’s Minimal Essential Medium (EMEM,
ATCC, Manassas, VA, USA) supplemented with 10%
heat inactivated horse serum (ATCC, Manassas, VA,
USA) for 24 h. Expression, reporter and control vectors
(2.5 ng control and 100 ng others per culture well) were
transfected by Lipofectamin PlusTM Reagent (Invitro-
gen, Carlsbad, CA, USA). Transfection reagent and
DNA were prepared according to manufacture’s
recommendations and incubated with cells for 3 h fol-
lowed by addition of an equal volume of EMEM sup-
plemented with 20% horse serum. Twenty-four h after
transfection, cells were treated with vehicle or com-
pounds at indicated final concentration for 24 h. Final
concentration of DMSO was 0.01% in medium. Vehicle
and compound treatment were all conducted in tripli-
cate. Each culture well was then assayed for a response
characterized by increased Firefly luciferase activity
normalized with Renilla luciferase activity.
References and Notes
1. Harris, M. I.; Flegal, K. M.; Cowie, C. C.; Eberhardt,
M. S.; Goldstein, D. E.; Little, R. R.; Weideyer, H. M.; Byrd-
Holt, D. D. Diabetes Care 1998, 21, 518.
2. Howard, B. V. Int. Congr. Ser. 1995, 1100, 446.
3. Bennet, P. H. Diabet. Metab. Rev. 1997, 13, 583.
4. (a) Reaven, G. M. Diabetes 1988, 37, 1595. (b) Reaven,
G. M. Diabetes 2000, 2 (Suppl.), S274.
5. Gerich, J. E. N. Engl. J. Med. 1989, 321, 1231.
6. Wiernsperger, N. F.; Baily, C. J. Drugs 1999, 58, 31.
7. Schmid-Antomarchi, H. S.; De Weille, J.; Fosset, M.; Laz-
dunski, M. J. Biol. Chem. 1987, 262, 15840.
8. (a) Ferner, R. E.; Neil, H. A. W. Br. Med. J. 1988, 269, 949.
(b) Burge, M. R.; Sood, V.; Sobhy, T. A.; Rassam, A. G.;
Schade, D. S. Diabetes, Obes. Metab. 1999, 1, 199.
9. Schoonjans, K.; Auwerx, J. The Lancet 2000, 355, 1008.
10. Nuclear Receptors Nomenclature Commitee. Cell 1999,
97, 97–161.
11. Lehmann, J. M.; Moore, L. B.; Smith-Oliver, T. A.;
Wilkinson, W. O.; Wilson, T. M.; Kliewer, S. A. J. Biol.
Chem. 1995, 270, 12953.
12. Shoda, T.; Mizuno, K.; Imamiya, E.; Sugiyama, Y.;
Fujita, T.; Kawamatsu, Y. Chem. Pharm. Bull. 1982, 30, 3580.
13. Mosome, Y.; Meguro, K.; Ikeda, H.; Hatanaka, C.; Ot,
S.; Shoda, T. Chem. Pharm. Bull. 1991, 39, 1440.
14. (a) Cantello, B. C. C.; Cawthorne, M. A.; Haig, D.; Hind-
ley, R. M.; Smith, S. A.; Thurlby, P. L. Bioorg. Med. Chem.
Lett. 1994, 4, 1181. (b) Cantello, B. C. C.; Cawthorne, M. A.;
Cottam, G. P.; Duff, P. T.; Haig, D.; Hindley, R. M.; Lister,
C. A.; Smith, S. A.; Thurlby, P. L. J. Med. Chem. 1994, 37, 3977.
15. Dey, D.; Neogi, P.; Nag, B. FASEB J. 2001, 15, A526.
16. Dey, D.; Medicherla, S.; Neogi, P.; Nag, B. Diabetes 1999,
48, A119.
17. Pettit, G. R.; Singh, S. B.; Schmidt, J. M.; Niven, M. L.;
Hamel, E.; Lin, C. M. J. Nat. Prod. 1998, 51, 517.
18. (a) Hudlicky, M. ACS Monograph 188; ACS: Washington,
DC, 1996; p 46. (b) Ram, S.; Ehrenkaufer, R. E. Synthesis
1988, 91.
Assays for Firefly luciferase activity and Renilla luci-
ferase activity followed the standard protocol of Dual-
luciferase Reporter1 Assay System (Promega, Madison,
WI, USA). Briefly, 400 mL Passive Lysis Buffer was
added into each culture well and all wells were placed on
a shaker for 15 min. Cell lysate (5 mL) of each well was
added to a reaction tube. Luciferase reagent II and Stop
& Glo1 were injected into the reaction tube sequentially
by Sirius Luminometer (Berthold Detection Systems,
Pforzheim, Germany). Final reporter activity was cal-
culated as the ratio of Firefly luciferase activity over
Renilla luciferase activity.
Caco-2permeability. Caco-2 permeability studies were
done by Absorptions System, Exton following standard
protocol and analyzing samples by mass spectrometer with
electrospray interface and single ion monitoring mode.
19. Kessar, S. V.; Nadir, U. K.; Gupta, Y. P.; Pahwa, P. S.;
Singh, P. Indian J. Chem. 1981, 20B, 1.
20. Dey, D.; Medicherla, S.; Neogi, P.; Gowri, M.; Cheng, J.;
Gross, C.; Sharma, S. D.; Reaven, G. M.; Nag, B. Metabo-
lism. In press.
In vivo studies. All procedures performed were in com-
pliance with the Animal Welfare Act and US Depart-
ment of Agriculture regulations and were approved by
21. Sen, A.; Rydzewski, J.; Subramaniam, K.; Gross, C.
AAPS Pharm. Sci. 2002, 4, M1290.