Antihyperglycemic activity of new 1,2,4-oxadiazolidine-3,5-diones

Article abstract of DOI:10.1016/S0223-5234(00)01191-0

A series of 1,2,4-oxadiazolidine-3,5-diones was synthesized and evaluated as oral antihyperglycemic agents in the obese insulin resistant db/db and ob/ob mouse - the two models for Type 2 diabetes mellitus. The majority of the prepared methoxy- and ethoxy-linked oxazole 1,2,4-oxadiazolidine-3,5-diones normalized plasma glucose levels at the 100 mg kg^-1 oral dose in the db/db diabetic mouse model, and several amongst them reduced the glucose levels at the 20 mg kg^-1 oral dose. The most potent compounds in the db/db mouse model were also active in the ob/ob mouse model normalizing the plasma glucose levels at the 20 mg kg^-1 oral dose. The trifluoromethoxy analog 32 was the most active compound of the series, reducing significantly the plasma glucose levels at the 5 mg kg^-1 oral dose. Oxadiazole-tailed 1,2,4-oxadiazolidine-3,5-diones were also active in both the db/db and ob/ob diabetic mouse models normalizing plasma glucose levels at the 100 mg kg^-1 oral dose.

Full text of DOI:10.1016/S0223-5234(00)01191-0

Eur. J. Med. Chem. 36 (2001) 31–42
´
© 2001 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved
PII: S0223-5234(00)01191-0/FLA
Original article
Antihyperglycemic activity of new 1,2,4-oxadiazolidine-3,5-diones
Michael S. Malamasa*, Janet Sredyb, Michael McCalebc, Iwan Gunawana, Brenda Mihana,
Donald Sullivana
aWyeth–Ayerst Research Inc., CN 8000, Princeton, NJ 08543-8000, USA
bInstitute for Diabetes Discovery, 23 Business Park Drive, Branford, CT 06405, USA
cAmgen Inc., 1840 DeHavilland Drive, Thousand Oaks, CA 91320-1789, USA
Received 28 May 2000; revised 20 August 2000; accepted 5 September 2000
Abstract – A series of 1,2,4-oxadiazolidine-3,5-diones was synthesized and evaluated as oral antihyperglycemic agents in the obese
insulin resistant db/db and ob/ob mouse — the two models for Type 2 diabetes mellitus. The majority of the prepared methoxy-
and ethoxy-linked oxazole 1,2,4-oxadiazolidine-3,5-diones normalized plasma glucose levels at the 100 mg kg−1 oral dose in the
db/db diabetic mouse model, and several amongst them reduced the glucose levels at the 20 mg kg−1 oral dose. The most potent
compounds in the db/db mouse model were also active in the ob/ob mouse model normalizing the plasma glucose levels at the 20
mg kg−1 oral dose. The trifluoromethoxy analog 32 was the most active compound of the series, reducing significantly the plasma
glucose levels at the 5 mg kg−1 oral dose. Oxadiazole-tailed 1,2,4-oxadiazolidine-3,5-diones were also active in both the db/db and
ob/ob diabetic mouse models normalizing plasma glucose levels at the 100 mg kg−1 oral dose. © 2001 Editions scientifiques et
´
me´dicales Elsevier SAS
insulin resistance / antihyperglycemic agents / oxadiazolidinediones / diabetic animal models
1. Introduction
are the most widely used antidiabetic agents. These
agents are acting on pancreatic b-cells stimulating
Insulin resistance in the liver and peripheral tissues,
insulin secretion. However, these agents are often
together with a pancreatic b-cell defect is the common
known to induce severe hypoglycemia and weight
cause of Type 2 diabetes [1]. The prevalence of insulin
gain, and become ineffective after 5 years of treat-
resistance in pre-diabetic or glucose intolerant sub-
ment [4–7]. Another group of drugs is the thiazo-
jects has long been recognized [2]. The failure of
lidinediones, which enhance insulin action. The
insulin to suppress adequately the hepatic glucose
PPARg receptor, a member of the nuclear hormone
output postprandially, combined with the reduced
receptor superfamily, has been identified recently as
glucose disposal by the peripheral tissues, leads to
the major functional receptor for the thiazolidine-
abnormal glycemic control after feeding. The in-
diones [8]. Since the pioneering discovery of ciglita-
creased and sustained plasma glucose levels pro-
zone 1 (figure 1) by Takeda scientists [9], which
gresses gradually into a number of debilitating
reduced insulin resistance and normalized plasma glu-
diabetic complications as retinopathy, neuropathy,
cose levels in genetically diabetic and/or obese animal
nephropathy, atherosclerosis, and coronary artery
models, a plethora of new thiazolidinediones has been
disease [3]. Therefore, it is essential to control blood
developed [10–18]. Troglitazone 2, a thiazolidine-
glucose at the early stages of the disease.
dione-type pharmacological agent, has been used ef-
Treatment of Type 2 diabetes consists usually of
fectively as an insulin enhancing agent in a large
diet, exercise, and hypoglycemic agents. Sulfonylureas
number of Type 2 patients [19]1. However, troglita-
Abbreviations: PPARg, peroxisome proliferator activated receptor g.
* Correspondence and reprints.
1 Troglitazone is being marketed in the United States as
Rezulin and in Japan as Noscal.
E-mail address: malamam@war.wyeth.com (M.S. Malamas).

32
M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
zone has produced severe liver toxic effects in a small
number of patients, and has been withdrawn from the
market. Recently, thiazolidinediones of SmithKline
Beecham (rosiglitazone; 3a) and Takeda (pioglita-
zone; 3b) have been approved for marketing. Patients
on these new agents will also be monitored for poten-
tial liver toxicity. Biguanides — a third class of
oral-lowering agents — were associated with lactic
acidosis, and only metformin (3c), offers a useful
treatment for insulin-resistant overweight Type-2 dia-
betes patients [20, 21]. The glucose-lowering effect of
metformin occurs without stimulation of insulin se-
cretion and results mainly from increased glucose
utilization. Metformin enhances insulin action at the
postreceptor level in peripheral tissues such as muscle.
In the present study, we disclose a detailed and
systematic structure–activity relationship (SAR)
study of new 1,2,4-oxadiazolidine-3,5-diones as po-
tent antihyperglycemic agents. Some limited informa-
tion on oxadiazolidinediones as antihyperglycemic
agents has been reported previously [15] in a series of
oxazolylethoxy-hydroxyureas, which exhibited hypo-
glycemic properties.
ence of potassium carbonate afforded azoles 7 (n=1).
Coupling of alcohols 8 with either hydroxybenzalde-
hyde or hydroxyacetophenone, using the Mitsunobu
protocol [27], produced azoles 7 (n=2). Azoles 7
were converted to the hydroxylamines 10 in a two-
step process. First, azoles 7 were treated with hydrox-
ylamine hydrochloride in the presence of sodium
acetate to produce the corresponding oximes 9, which
were reduced further to the hydroxylamines 10 with
sodium cyanoborohydride under acidic conditions.
The hydroxylamines 10 were converted to the 1,2,4-
oxadiazolidine-3,5-diones 11 with N-(chlorocar-
bonyl)isocyanate at low temperatures (0–5°C).
The benzoxazole 63 [28] (table IV) was prepared
from 2-(2-benzoxazolylmethylamino)ethanol [17] and
4-hydroxybenzaldehyde substantially in the same
manner as described above.
The benzofurans 68–72 (table V) were prepared
according to figure 3. Condensation of oxazole 12
with 3-bromo-2-hydroxybenzaldehyde in the presence
of sodium methoxide afforded benzofuran 13. Reduc-
tion of the ketone functionality, first with sodium
borohydride to the corresponding secondary alcohol,
and second with triethylsilane/trifluoroacetic acid,
produced benzofuran 14. Conversion of the aromatic
bromide of 14 to a formyl group (16) was achieved
through a two-step process. First, the generation of
the nitrile 15 with copper cyanide in N,N-dimethyl-
formamide, and second, the reduction of 15 with
AlꢀNi in formic acid furnished 16. Benzofuran 16 was
converted to the final product 18 (R3 =H) according
to figure 2. Compound 18, when R3=CH3, was pre-
2. Chemistry
The 1,2,4-oxadiazolidine-3,5-diones (tables I–IV
and VI) were prepared according to figure 2. Treat-
ment of either hydroxybenzaldehyde or hydroxyace-
tophenone with either oxa(thia)zoles 4 [22–24],
oxadiazoles 5 [22, 25] or triazoles 6 [26] in the pres-
Figure 1. Known antihyperglycemic agents

M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
33
2

34
M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
Table II. Chemical and biological data of ethoxy-linked oxazole 1,2,4-oxadiazolidine-3,5-diones.a
c
Compound
R1
R2
R3
POA
m.p. (°C)
% Decrease in plasma glucose b
db/db mouse
db/db mouse
100 mg kg−1
20 mg kg−1
36
37
38
39
40
41
42
43
44
45
46
47
48
49
Ph
Ph
Ph
H
H
H
H
H
H
H
H
H
H
H
H
H
F
H
H
CH3
H
H
H
H
H
H
H
H
H
H
H
4
3
4
4
4
4
4
4
4
4
4
4
4
4
205–206
173–174
169–170
185–186
201–202
196–197
130–132
205–206
200–202
190–192
185–186
198–199
195–196
207–208
43*
45*
40*
18*
30*
i
45*
i
d
Naphthyl
4-OCH3ꢀPh
4-CH3ꢀPh
Cyclohexyl
Thienyl
Furyl
4-CF3ꢀPh
4-FꢀPh
3-FꢀPh
2-FꢀPh
Ph
i
i
57*
i
43*
75*
71*
63*
63*
38*
a All compounds were prepared according to figure 2.
b All values for the drug-treated groups, other than i, are significant vs. vehicle-treated mice, PB0.05.
c POA, Point of attachment to the phenyl ring.
d i, Inactive, generally less than −15% change, P\0.05.
pared similarly from methyl-ketone 17. Ketone 17
(R3=CH3) was generated during a two-step process
from 16. First, the formyl group was converted to the
corresponding secondary alcohol with methylmagne-
sium chloride, and second, it was oxidized with Jones
reagent to furnish 17.
18]. The in vivo activity was assessed at a daily dose of
5–100 mg kg−1 (p.o.) for 4 days, and measured as a
specified decrease in plasma glucose and insulin levels
of the drug-treated group relative to a vehicle-treated
control group. Decreases of 50–60 (db/db) and 30–
40% (ob/ob) in plasma glucose levels normalize gener-
ally the glucose levels that are similar to the
non-diabetic controls. A compound considered active,
at the specific dosage administered, if the difference of
the plasma glucose level had a P<0.05. All com-
pounds with P>0.05 were reported as inactive. Cigli-
tazone was the reference standard in all of the assays.
Results with the methoxy-linked oxazole 1,2,4-oxa-
diazolidine-3,5-diones are shown in table I. All the
compounds, with the exception of the methyl-substi-
tuted analogs 21 and 22, were active in the db/db
diabetic mouse. The most potent compounds were
substituted at the 2-phenyl moiety of the oxazole tail
with electron withdrawing groups. The para-substi-
tuted (trifluoromethyl)- and (trifluoromethoxy)-
3. Results and discussion
The test compounds were evaluated in two classical
insulin-resistant hypoglycemic obese mouse models,
the genetically obese C57 BL/6J ob/ob, and diabetic
C57BL/KsJ db/db, for their ability to decrease
plasma glucose and insulin levels. The ob/ob animal
model is severely insulin resistant, hyperinsulinemic,
and glucose intolerant, and the db/db model is also
glucose intolerant with fasting hyperglycemia and oc-
casional hyperinsulinemia. These animal models [29]
have been validated well by several laboratories for
the assessment of novel glucose lowering agents [10–

M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
35
tions on the central aromatic ring with either electron
withdrawing or donating groups (i.e. halogen,
methoxy) resulted in small reductions of the in vivo
activity (23, 24 vs. 19; 29, 30 vs. 27).
phenyl analogs 27 and 32 were the best compounds of
the series. Both compounds reduced plasma glucose
levels by 37 and 46%, respectively, at a 20 mg kg−1
oral dose. The analogous meta-substituted analogs 28
and 33 were weaker in vivo. Substitutions at the
2-phenyl ring may play a critical role in the metabolic
stabilization of the phenyl group. The attachment
location of the 1,2,4-oxadiazolidine-3,5-dione head-
piece to the central aromatic ring of the molecule was
also important to the activity. The meta-substituted
phenyl (central aromatic group) analog 27 was about
five times more potent in vivo (db/db) than the para-
substituted compound 26. The most active com-
pounds of the methoxy-linked analogs were also
evaluated in the ob/ob diabetic mouse model. The
trifluoromethoxy analog 32 was the most potent com-
pound in the ob/ob mouse model. It normalized
plasma glucose levels at an oral dose of 20 mg kg−1,
and also reduced the plasma glucose levels signifi-
cantly by 27% at the 5 mg kg−1 oral dose. Substitu-
The ethoxy-linked 1,2,4-oxadiazolidine-3,5-diones
were also active in vivo. Similarly to the methoxy-
linked analogs, substitutions on the 2-phenyl moiety
of the oxazole tail produced potent compounds in
vivo. Electronegative groups (trifluoromethyl, halo-
gen) were the best substituents. Compounds 45 and
46 lowered plasma glucose in the db/db mouse by 75,
and 71%, respectively, as compared with 36 [15],
which lowered plasma glucose by 43% at the oral dose
of 100 mg kg−1. The 4-methoxyphenyl analog 40 was
active in vivo, while the 4-methylphenyl analog 41 was
inactive. Metabolic oxidation of the methyl group of
41 might have been a factor to the loss of the in vivo
activity. Replacement of the 2-phenyl moiety of the
oxazole tail with aromatic, hetero-aromatic, and cy-
cloalkyl groups resulted in various decreases in
Table III. Chemical and biological data of oxadiazole 1,2,4-oxadiazolidine-3,5-diones [31].a
c
Compound
R1
n
POA
m.p. (°C)
% Decrease in plasma glucose b
% Decrease in
plasma insulin b
db/db mouse
ob/ob mouse
ob/ob mouse
100 mg kg−1
100 mg kg−1
100 mg kg−1
50
51
52
53
54
55
56
57
58
59
60
61
62
Ph
4-CF3ꢀPh
Ph
2
2
1
1
1
1
1
1
1
1
1
1
1
4
4
4
3
4
3
4
4
4
3
4
3
3
152–153
128–129
150–152
133–135
160–162
90–92
156–157
147–149
168–169
159–160
144–145
120–121
93–95
41*
i
30*
18*
40*
19*
d
(20)
47**
87**
Ph
36* (75 mg kg−1
20*
)
53** (75 mg kg−1
73**
)
4-CF3ꢀPh
4-CF3ꢀPh
4-FꢀPh
4-F-Ph
4-ClꢀPh
4-ClꢀPh
4-OCF3ꢀPh
4-OCF3-Ph
3,5
46** (75 mg kg−1
31*
)
45* (75 mg kg−1
88**
)
i
25
i
38*
48*
i
di-CH3ꢀisoxazole
a All compounds were prepared according to figure 2.
b All values for the drug-treated groups, other than i, are significant vs. vehicle-treated mice; * PB0.05; ** PB0.01.
c POA, Point of attachment to the phenyl ring.
d i, Inactive, generally less than −15% change, P\0.05.

36
M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
Table IV. Chemical and biological data of 1,2,4-oxadiazo-
or weaker to the parent compound (table III). The
trifluoromethoxy analog 61 normalized plasma glu-
cose levels in the db/db mouse at the oral dose of 100
mg kg−1. Compounds 52, 53, 55 and 56 normalized
plasma glucose levels in the ob/ob mouse at the oral
doses of 75–100 mg kg−1.
lidine-3,5-diones.a
Replacements of the oxazole tail with various
groups are shown in table IV. The benzoxazole
analog [17] 63 and the 2-phenyl thiazoles 64 and 65
were active in vivo in the db/db mouse, while the
2-methyl-thiazole 66 and 3-phenyl triazole 67 were
inactive.
Replacement of the central aromatic region with a
benzofuran nucleus produced good in vivo com-
pounds (table V). The methyl-substituted phenyl
analogs 69 and 70 normalized the plasma glucose
levels in the db/db mouse at the oral dose of 20 mg
kg−1. The naphthyl analog 71 was inactive.
Further replacement of the central aromatic region
(table VI) with either a naphthalene moiety 73–75 or
a tetrahydronaphthalene 76 nucleus produced either
weakly active (75) or inactive in vivo (db/db mouse
model) compounds. The indane analog 77 decreased
plasma glucose levels by 26% at the 100 mg kg−1 oral
dose.
There have been several reports that the thiazo-
lidinediones are high-affinity peroxisome proliferator
activated receptor-g (PPARg) agonists, and there is a
direct correlation between the in vitro potency at the
PPARg receptor and the in vivo antihyperglycemic
potency in ob/ob mice [30]. Since our compounds
possess some structural similarities with the thiazo-
lidinediones-type antihyperglycemic agents, a PPARg
agonistic effect cannot be ruled out as a possible
mechanism for the antihyperglycemic properties of
these compounds. While our present studies are out-
side the scope of addressing such a possible mecha-
nism of our compounds, future work will be necessary
to elucidate this hypothesis.
c
Compound R
POA m.p. (°C) % Decrease
plasma
glucoseb
db/db
Mouse
100 mg kg−1
63
4
155–157 34*
67–168 28*
64
65
66
67
4
3
4
4
139–140 27*
d
165–166
175–177
i
i
a All compounds were prepared according to figure 2.
b All values for the drug-treated groups, other than i, are
significant vs. vehicle-treated mice; *, PB0.05; **, PB0.01.
c POA, Point of attachment to the phenyl ring.
d i, Inactive, generally less than −15% change, P\0.05.
plasma glucose levels. The naphthyl and thienyl
analogs 39 and 43 were either weakly active or inac-
tive in vivo, while the cyclohexyl and furyl analogs 42
and 44 maintained their in vivo activity and were
similar to 36.
Replacement of the oxazole tail with an oxadiazole
tail (table III) produced several compounds that nor-
malized plasma glucose levels in both, the db/db and
ob/ob diabetic mouse models at a 100 mg kg−1 oral
dose. The phenyl analog 50 was similar to the
analogous oxazole analog 36. Substitutions on the
phenyl group of the oxadiazole-tailed compounds
were in variance when compared with the oxazole-
tailed compounds. Electron withdrawing or donating
groups produced compounds that were either similar
4. Conclusions
In summary, we have prepared a series of 1,2,4-
oxadiazolidine-3,5-diones and evaluated in two differ-
ent diabetic animal models of insulin resistance as
antihyperglycemic agents. The majority of the
methoxy- and ethoxy-linked oxazole 1,2,4-oxadiazo-
lidine-3,5-diones normalized plasma glucose levels at
the 100 mg kg−1 oral dose in the db/db diabetic
mouse model, and several amongst them reduced the

M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
37
glucose levels at the 20 mg kg−1 oral dose. The most
potent compounds in the db/db mouse model were also
active in the ob/ob diabetic mouse model at the oral
dose of 20 mg kg−1. The trifluoromethoxy analog 32
was the most active compound, reducing significantly
the plasma glucose levels in the ob/ob mouse model at
the 5 mg kg−1 oral dose. Oxadiazole-tailed 1,2,4-oxadi-
azolidine-3,5-diones were also active in both the db/db
and ob/ob diabetic mouse models normalizing plasma
glucose levels at the 100 mg kg−1 oral dose.
Figure 2. Reagents: (a) K2CO3, DMF; (b) Diethyl azodicarboxylate, Ph3P, THF; (c) HONH2·HCl, NaOAc, EtOH, H2O; (d)
NaCNBH3, HCl, MeOH, THF; (e) ClCONCO, THF.
Table V. Chemical and biological data of benzofuran 1,2,4-oxadiazolidine-3,5-diones.a
.
Compound R1
R2
m.p. (°C) % Decrease in plasma glucose b
% Decrease in plasma insulin b
db/db mouse db/db mouse ob/ob mouse ob/ob mouse
100 mg kg−1 20 mg kg−1
20 mg kg−1
20 mg kg−1
68
69
70
71
72
Ph
H
H
H
H
189–190
194–196
180–181
228–229
62*
26*
42*
45*
52**
83**
4-CH3ꢀPh
3-CH3ꢀPh
Naphthyl
Ph
c
i
CH3 174–175
46*
a All compounds were prepared according to figure 3.
b All values for the drug-treated groups, other than i, are significant vs. vehicle-treated mice; *PB0.05; **PB0.01.
c i, Inactive, generally less than −15% change, P\0.05.

38
M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
Figure 3. Reagents: (a) NaOCH3, EtOH, 3-Br-2-OHꢀC6H3CHO; (b) NaBH4 MeOH, THF; (c) Et3SiH, TFA, CH2Cl2; (d) CuCN,
DMF; (e) AlꢀNi, 70 % HCO2H; (f) CH3MgCl, Et2O; (g) Jones Reagent, acetone.
5. Experimental protocols
zoles 6, and alcohols 8 by the following representative
procedures. All the required intermediates were pre-
pared as described previously [19–24].
5.1. General chemistry
5.1.1. 4-(5-Methyl-2-phenyl-oxazol-4-ylmethoxy)-
Melting points were determined in open capillary
tubes on a Thomas–Hoover apparatus, and are re-
ported uncorrected 1H-NMR spectra were determined in
the cited solvent on a Bruker AM 400 (400 MHz), or a
Varian XL-300 (300 MHz) instrument, with te-
tramethylsilane as an internal standard. Chemical shifts
are given in parts per million (ppm) and coupling con-
stants are in hertz. Splitting patterns are designated as
follows — s, singlet; br s, broad singlet; d, doublet; t,
triplet; q, quartet; and m, multiplet. The infrared spectra
were recorded on a Perkin–Elmer 781 spectrophotome-
ter as KBr pellets or as solutions in chloroform. Mass
spectra were recorded on either a Finnigan model 8230
or a Hewlett–Packard model 5995A spectrometer. Ele-
mental analyses (C, H, N) were performed on a Perkin–
Elmer 240 analyzer and all compounds are within
0.4% of theory unless indicated otherwise. All prod-
ucts, unless noted otherwise, were purified by ‘flash
chromatography’ with use of 220–400 mesh silica gel.
Thin-layer chromatography was done on silica gel 60
F-254 (0.25-mm thickness) plates. Visualization was ac-
complished with UV light and/or 10% phosphomolybdic
acid in ethanol. Unless otherwise noted, all materials
were obtained commercially and used without further
purification. All reactions were carried out under an
atmosphere of dried nitrogen.
benzaldehyde (7, R1, R3=H, R2=CH3
n=1, para-substituted)
, X=O, Y=C,
A mixture of 4-chloromethyl-5-methyl-2-phenyl-oxa-
zole [22, 23] (5.5 g, 26.5 mmol), 4-hydroxybenzaldehyde
(3.23 g, 26.5 mmol), potassium carbonate (3.66 g, 26.5
mL) and N,N-dimethylformamide (80 mL) was stirred
at 80°C for 8 h. The mixture was poured into H2O and
extracted with EtOAc. The organic extracts were dried
over MgSO4. Evaporation of the volatiles and purifica-
tion by flash chromatography on silica gel (eluting sol-
vent hexane/EtOAc 4/1) gave a yellow solid (6.8 g, 86%
yield) — m.p. 103–105°C; 1H-NMR (DMSO-d6, 400
MHz) l 2.46 (s, 3H, CH3), 5.1 (s, 2H, CH2), 7.25 (m,
2H, ArꢀH), 7.51 (m, 3H, ArꢀH), 7.86–7.93 (m, 4H,
ArꢀH), 9.9 (s, 1H, CHO); IR (KBr, cm−1) 1690 (CO);
MS m/e 294 (M+H)+. Anal. (C18H15NO3) C, H, N.
5.1.2. 4-(5-Methyl-2-phenyl-oxazol-4-ylmethoxy)-
benzaldehyde oxime (9, R1, R3=H, R2=CH3, X=O,
Y=C, n=1, para-substituted)
A solution of sodium acetate (7.27 g, 82.7 mmol) in
H2O (40 mL) was added into a solution of 4-(5-methyl-
2-phenyl-oxazol-4-ylmethoxy)-benzaldehyde (6.5 g, 22.2
mmol), hydroxylamine hydrochloride (4.62 g, 66.55
mmol) and ethanol (300 mL). The mixture was stirred at
room temperature for 12 h, and then poured into H2O,
acidified with HCl (1N) and extracted with EtOAc. The
The oxadiazolidinediones described in this paper were
synthesized from oxa(thia)zoles 4, oxadiazoles 5, tria-

M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
39
Table VI. Chemical and biological data of 1,2,4-oxadiazo-
lidine-3,5-diones.a
5.1.3. N-[4-(5-Methyl-2-phenyl-oxazol-4-ylmethoxy)-
benzyl]-hydroxylamine (10, R1, R3=H, R2=CH3,
X=O, Y=C, n=1, para-substituted)
A solution of HCl (4 N, in dioxane) was added
dropwise into a solution of 4-(5-methoxy-2-phenyl-oxa-
zol-4-ylmethoxy)-benzaldehyde oxime (6.0 g, 19.42
mmol) in MeOH (300 mL), THF (60 mL) and methyl
orange (indicator, 20 mg) in a rate the reaction pH was
maintained at a range 3–4. When a persistent red color
was observed, the reaction mixture was poured into
H2O, basified with NaOH (1 N) to pH 9 and extracted
with EtOAc. The organic extracts were dried over
MgSO4. Evaporation of the volatiles and purification by
flash chromatography, on silica gel (eluting solvent
EtOAc/MeOH 10/1) gave a yellow solid (5.2 g, 86%
yield) — m.p. 109–110°C; 1H-NMR (DMSO-d6, 400
MHz) l 2.43 (s, 3H, CH3), 3.8 (s, 2H, CH2), 4.97 (s, 2H,
CH2), 5. 88 (brs, 1H, NH), 6.98 (d, J=8.7 Hz, 2H,
ArꢀH), 7.2 (s, 1H, OH), 7.24 (d, J=8.7 Hz, ArꢀH),
7.54 (m, 3H, ArꢀH), 7.95 (m, 2H, ArꢀH); IR (KBr,
cm−1) 3400 (NH), 3250 (OH); MS m/e 311 (M+H)+.
Anal. (C18H18N2O3) C, H, N.
.
Compound
X
m.p. (°C)
Decrease
plasma glucose
(%) b
db/db
100 mg kg−1
c
73
74
210–211
207–208
i
i
75
207–209
17*
76
77
158–159
162–163
i
5.1.4. 2-{4-[(5-Methyl-2-phenyl-1,3-oxazol-4-yl)-
methoxy]benzyl}-1,2,4-oxadiazolidine-3,5-dione (19)
N-(Chlorocarbonyl)isocyanate (0.41 mL, 5.16 mmol)
was added dropwise into a cold (−5°C) solution of
N-[4-(5-methyl-2-phenyl-oxazol-4-ylmethoxy)-benzyl]-
hydroxylamine (1.6 g, 5.16 mmol), and THF (20 mL).
The reaction mixture was stirred for 30 min, poured into
HCl (1 N), and extracted with ethyl acetate. The organic
extracts were dried over MgSO4. Evaporation of the
volatiles and purification by flash chromatography on
acid washed (5% H3PO4/MeOH) silica gel (hexane/ethyl
acetate 2/1) gave a white solid (1.4 g, 72% yield) — m.p.
26*
a All compounds were prepared according to figure 2.
b All values for the drug-treated groups, other than i, are
significant vs. vehicle-treated mice; *, PB0.05.
c i, Inactive, generally less than −15% change, P\0.05.
1
170–172°C; H-NMR (DMSO-d6, 400 MHz) l 2.44 (s,
3H, CH3), 4.7 (s, 2H, CH2), 5.0 (s, 2H, CH2), 7.04 (d,
J=8.7 Hz, 2H, ArꢀH), 7.27 (d, J=8.7 Hz, ArꢀH),
7.49–7.53 (m, 3H, ArꢀH), 7.93–7.96 (m, 2H, ArꢀH),
12.36 (s, brs, 1H, NH); IR (KBr, cm−1) 3450 (NH),
1730 (CO); MS m/e 380 (M+H)+. Anal. (C20H17N3O5)
C, H, N.
organic extracts were dried over MgSO4. Evaporation of
the volatiles and crystallization from acetone/ethyl
ether/hexane, gave a white solid (6.1 g, 89% yield) —
m.p. 192–193°C; 1H-NMR (DMSO-d6, 400 MHz) l
2.45 (s, 3H, CH3), 5.03 (s, 2H, CH2), 7.06 (m, 2H,
ArꢀH), 7.49–7.55 (m, 5H, ArꢀH), 7.92–7.96 (m, 2H,
ArꢀH), 8.07 (s, 1H, CH), 10.08 (s, 1H, OH); IR (KBr,
5.1.5. (5-bromo-1-Benzofuran-2-yl)[5-methyl-2-
(4-methylphenyl)-1,3-oxazol-4-yl]methanone (13,
R1=CH3)
Sodium methoxide (25% in MeOH, 42.6 g, 197.3
mmol) was added into a mixture of 4-bromo-2-hydroxy-
benzaldehyde (39.6 g, 197.3 mmol) and ethyl alcohol
(500 mL). The reaction mixture was stirred for 20 min
cm−1
)
3200 (OH); MS m/e 309 (M+H)+. Anal.
(C18H16N2O3) C, H, N.

40
M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
and then 2-bromo-1-[5-methyl-2-(4-methylphenyl)-1,3-
oxazol-4-yl]-1-ethanone (58.0 g, 197.3 mmol) was added.
The new mixture was refluxed for 12 h, cooled to 0°C,
and the precipitated solid filtered and washed with ethyl
alcohol to yield a yellow solid (60.6 g, 78% yield) —
with ethyl acetate. Evaporation and crystallization from
acetone/ethyl ether/hexane gave a yellow solid (15 g).
The solid was taken in formic acid (70%, 170 mL) and
AlꢀNi (15 g) was added. The mixture was stirred for 4 h,
after which, the precipitated solids were filtered and
discarded. The filtrate was diluted with ethyl acetate and
washed with water, NaOH (2 N), and brine. The or-
ganic extracts were dried over MgSO4. Evaporation of
the volatiles and purification by chromatography on
silica gel (eluting solvent EtOAc/MeOH 10/1) gave a
yellow solid (12.1 g, 46% yield) — m.p. 125–126°C;
1H-NMR (DMSO-d6, 400 MHz) l 2.33 (s, 3H, CH3),
2.39 (s, 3H, CH3), 4.1 (s, 2H, CH2), 6.8 (s, 1H, CH),
7.29 (d, J=8.09 Hz, 2H, ArꢀH), 7.7 (d, J=8.3 Hz, 1H,
ArꢀH), 7.77–7.81 (m, 3H, ArꢀH), 8.15 (s, 1Hz, 1H,
ArꢀH), 10.02 (s, 1H, CHO); IR (KBr, cm−1) 1695 (CO);
MS m/e 332 (M+H)+. Anal. (C21H17NO3) C, H, N.
1
m.p. 210–211°C; H-NMR (DMSO-d6, 300 MHz) l 2.4
(s, 3H, CH3), 2.8 (s, 3H, CH3), 7.4 (d, J=8.6 Hz, 2H,
ArꢀH), 7.8 (m, 2H, ArꢀH), 8.01 (t, J= 8.6 Hz, 2H,
ArꢀH), 8.2 (s, 1H, ArꢀH), 8.64 (s, 1H, CH); IR (KBr,
cm−1
)
1635 (CO); MS m/e 395 (M+). Anal.
(C20H14BrNO3) C, H, N.
5.1.6. 4-[(5-bromo-1-Benzofuran-2-yl)methyl]-
5-methyl-2-(4-methylphenyl)-1,3-oxazole (14,
R1=CH3)
Sodium borohydride (10.4 g, 277 mmol) was added
portionwise into a cold (0°C) mixture of (5-bromo-
1-benzofuran-2-yl)[5-methyl-2-(4-methylphenyl)-1,3-
oxazol-4-yl]methanone (55.0 g, 138.9 mmol), THF (400
mL) and MeOH (500 mL). The mixture was stirred for
2 h, and then poured into water and the precipitated
solid filtered and dried to afford a brownish solid (53.0
g). The solid was then taken in dichloromethane (500
mL), and triethylsilane (42.5 mL, 266.3 mmol) and
cooled to 0°C. Trifluoroacetic acid (60 mL) was added
to the reaction mixture. The mixture was stirred at 0°C
for 1 h, and then was allowed to warm up to room
temperature where it was stirred for 3 h. The volatiles
were removed in vacuo and the residue was taken in
ethyl acetate (1000 mL) and washed with water, NaOH
(1 N), and brine. The organic extracts were dried over
MgSO4. Evaporation of the volatiles and crystallization
from acetone/ethyl ether/haxane gave a yellow solid
(45.2 g, 85% yield) — m.p. 143–144°C; 1H-NMR
(DMSO-d6, 400 MHz) l 2.33 (s, 3H, CH3), 2.37 (s, 3H,
CH3), 4.07 (s, 2H, CH2), 6.6 (s, 1H, CH), 7.29 (d,
J=8.09 Hz, 2H, ArꢀH), 7.35 (dd, J=8.7, 2.07 Hz, 1H,
ArꢀH), 7.49 (d, J=8.7 Hz, 1H, ArꢀH), 7.75 (d, J=2.07
Hz, 1H, ArꢀH), 7.79 (d, J=8.09 Hz, 2H, ArꢀH); MS
m/e 381 (M+). Anal. (C20H16BrNO2) C, H, N.
5.2. Biological methods
5.2.1. Determination of plasma glucose lowering in
db/db mice
On the morning of day 1, 35 mice male diabetic db/db
(C57BL/KsJ) mice (Jackson Laboratories), 2–7 months
of age and bodyweight of 50–70 g, were fasted for 4 h,
and weighed, and a baseline blood sample (15–30 ml)
was collected from the tail-tip of each mouse without
anesthesia, and placed directly into a fluoride-containing
tube, mixed and maintained on ice (in each individual
experiment, ages of the mice were identical between
treatment groups (i.e. groups were always matched in
any given experiment) and the range of ages as indicated
above was actually much narrower than indicated).
Food was then returned to the mice. The plasma was
separated and levels of glucose in plasma determined by
the Abbott VP analyzer. Because of the variable plasma
glucose levels of the db/db mice, the five mice having the
most extreme (i.e. highest or lowest) plasma glucose
levels were excluded and the remaining 30 mice were
assigned randomly into seven groups of equivalent mean
plasma glucose level (vehicle control, ciglitazone, and
five drug groups). On the afternoon of days 1, 2 and 3,
vehicle (0.2 mL of 2% Tween 80/saline w/v), control or
test drugs were administered (p.o.) to the ad libitum fed
mice. On the morning of day 4, the mice were weighed
and food removed, but water was available ad libitum.
After 3 h, a blood sample was collected and, then, the
mice were administered the fourth dose of drug or
vehicle. Blood samples were collected again from the
5.1.7. 2-{[5-Methyl-2-(4-methylphenyl)-1,3-oxazol-
4-yl]methyl}-1-benzofuran-5-carbaldehyde (16,
R1=CH3)
A mixture of 4-[(5-bromo-1-benzofuran-2-yl)methyl]-
5-methyl-2-(4-methylphenyl)-1,3-oxazole (30.0 g, 78.5
mmol), copper cyanide (14 g, 157 mmol) and N,N-
dimethylformamide (500 mL) was stirred at 160°C for
24 h. The mixture was then cooled to room temperature,
poured into ammonium hydroxide (30%), and extracted

M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
41
unanesthetized mice 2 and 4 h after drug administration.
The plasma was separated and level of glucose in
plasma was determined by the Abbott VP analyzer.
To assess the drug activity (considering the well-
known variation in plasma glucose values from animal
to animal within a litter), the percent change of the
animal’s plasma glucose level on day 4 (mean of two
consecutive bleeds after compound administration; 2-
and 4-h samples) compared with the pretreatment
plasma glucose level for each individual mouse (from
respective level before drug administration (day 1 base-
line sample) as follows — (mean of 2- and 4-h samples
(day 4)/baseline sample (day 1) multiplied by 100 to give
the percentage change).
Analysis of variance (ANOVA) followed by Dunnett’s
multiple comparison (one-sided) used to estimate the
degree of statistical significance of the difference be-
tween the vehicle control group and the individual drug-
treated groups. A drug considered active, at the specific
dosage administered, if the difference of the plasma
glucose level had a P<0.05. All compounds with P>
0.05 were reported as inactive.
of the active moiety of the compound. Control mice
received vehicle only.
On the morning of day 4, 4 h after drug administra-
tion, blood was collected into the tubes containing
sodium fluoride after decapitation under anesthesia. The
plasma was isolated by centrifugation and the concen-
tration of glucose was measured enzymatically on an
Abbott V.P. analyzer and the plasma concentration of
insulin was determined by radioimmunoassay [32]. For
each mouse, the percentage change in plasma glucose on
day 4 was calculated relative to the mean plasma glu-
cose of the vehicle treated mice. ANOVA, followed by
Dunnett’s comparison test (one tailed) was used to
estimate the significant difference between the plasma
glucose values from the control group and the individual
compound treated groups. A drug considered active, at
the specific dosage administered, if the difference of the
plasma glucose level had a P<0.05. All compounds with
P>0.05 were reported as inactive.
The positive control, ciglitazone produces a 18–34%
decrease in plasma glucose levels at 100 mg kg−1 per
day×4 days, p.o. (P<0.05).
References
[1] DeFronzo R.A., Bonadonna R.C., Ferrannini E., Pathogenesis
of NIDDM, Diabetes Care 15 (1992) 318–368.
[2] Reaven G., Bernstein R., Davis B., Olefsky J.M., Nonketonic
diabetes mellitus: insulin deficiency or insulin resistance, Am. J.
Med. 60 (1976) 80.
5.2.2. Determination of plasma glucose lowering effect
in ob/ob mice
Male or female ob/ob mice (C57BL/6J), ages 2–3
months (40–50 g), were used to assess the effect of
compounds on both glucose and insulin lowering. In
each study, the ob/ob mice were from the same litter, so
they were age-matched and had similar bodyweights and
levels of glycemia. Since the degree of glycemia does
vary somewhat within the litter, the mice from the litter
were randomized into different treatment groups (four
groups of ten mice) by bodyweight (we used bodyweight
instead of glycemia as a parameter since we found that
bodyweight in this age of ob/ob mice is well correlated
with glycemic state. From earlier studies, we know that
randomization by bodyweight yields tight plasma glu-
cose values for each group). The mice were housed five
per cage and were maintained on normal rodent chow
with water ad libitum. The mice received the compound
daily (morning) by gavage (suspended in 0.5 mL of 0.5%
methyl cellulose) for 4 days. The dose of compound
administered ranged from 5 to 100 mg kg per day.
Bodyweight of animals fed was measured at the begin-
ning of each week and doses for the entire week were
calculated using this weight and were expressed in terms
[3] Reaven G., Role of insulin resistance in human disease, Dia-
betes 37 (1988) 1595–1607.
[4] Goldman J.M., Oral hypoglycemic agents: an update of sul-
fonylureas, Drugs Today 25 (1989) 689–695.
[5] Kolterman O.G., Prince M.J., Olefsky J.M., Insulin resistance
in non-insulin-dependent diabetes mellitus. Impact of sulfony-
lureas agents in vivo and in vitro, Am. J. Med. 74 (Suppl. 1A)
(1983) 82–101.
[6] Ferrannini E., The insulin resistance syndrome, Curr. Opin.
Nephrol. Hypertens. 1 (1992) 291–298.
[7] Bailey C.J., Hypoglycemic and anti-hyperglycemic drugs for the
control of diabetic, Proc. Nutr. Soc. 50 (1991) 619–630.
[8] Spiegelman B.M., PPAR-g: adipogenic regulator and thiazo-
lidine receptor, Diabetes 47 (1998) 507–514.
[9] Sohda T., Mizuno K., Tawada H., Sugiyama Y., Fujita T.,
Kawanatsu Y., Studies on antidiabetic agents II. Synthesis of
5[4-(1-methylcyclohexylmethoxy)benzyl]thiazolidine-2,4-dione
and its derivatives, Chem. Pharm. Bull. 30 (1982) 3563–3573
also pp. 3580–3600.
[10] Fujiwara T., Yoshioka S., Yoshioka Y., Ushiyama I., Horikishi
H., Characterization of new oral antidiabetic agent CS-045,
Diabetes 37 (1988) 1549–1558.
[11] Momose Y., Meguro K., Ikeda H., Hatanaka C., Oi S., Sohda
T., Studies on antidiabetic agents: synthesis and biological
activities of pioglitazone and related compounds, Chem.
Pharm. Bull. 39 (1991) 1440–1445.

42
M.S. Malamas et al. / European Journal of Medicinal Chemistry 36 (2001) 31–42
[12] Dow R.L., Bechle B.M., Chou T.T., Clark D.A., Hulin B.,
Stevenson R.W., Benzyloxazolidine-2,4-diones as potent hypo-
glycemic agents, J. Med. Chem. 34 (1991) 1538–1544.
[20] Bailey C.J., Biguanides and NIDDM, Diabetes Care 15 (1992)
755–772.
[21] DeFronzo R.A., Diabetes Rev. 6 (1998) 89–131.
[13] Clark D.A., Goldstein S.W., Volkmann R.A., Eggler J.F.,
Holland G.F., Hulin B., Stevenson R.W., Kreutter D.K., Gibbs
E.M., Krupp M.N., Merrigan P., Kelbaugh P.L., Andrews
E.G., Tickner D.L., Suleske R.T., Lamphere C.H., Rajeckas
F.J., Kappeler W.H., McDermott R.E., Hutson N.J., Johnson
M.R., Substituted dihydrobenzopyran and dihydrobenzofuran
thiazolidine-2,4-diones as hypoglycemic agents, J. Med. Chem.
34 (1991) 319–325.
[22] Malamas M.S., Carlson R.P., Grimes D., Howell R., Glaser
K., Gunawan I., Nelson J.A., Kanzelberger M., Shah U.,
Hratman D.A., J. Med. Chem. 39 (1996) 237–245.
[23] Turchi I.J., Oxazoles. Heterocyclic compounds, Wiley, New
York, 1986, pp. 1–341.
[24] Meguro K., Fugita T., Thiazolidinedione derivatives; their pro-
duction and use, Eur. Patent 177353, 1986.
[14] Hulin B., Clark D.A., Goldstein S.W., McDermott R.E.,
Dambek P.J., Kappeler W.H., Lamphere C.H., Lewis D.M.,
Rizzi J.P., Novel thiazolidine-2,4-diones as potent euglycemic
agents, J. Med. Chem. 35 (1992) 1853–1864.
[25] Tawada H., Nakata F., Sugiyama Y., Fujita T., Meguro K.,
Syntheses and antidiabetic activity of 1,2,4-oxadiazole deriva-
tives, J. Takeda Res. Lab. 43 (1984) 1–11.
[26] Browne E.J., Nunn E.E., Polya J.B., 2-Oxazoli-4-ones and their
conversion into 3-hydroxymethyl-1,2,4-triazoles, J. Chem. Soc.
(C) (1970) 1515–1970.
[15] Goldstein S.W., McDermott R.E., Gibbs E.M., Stevenson
R.W., Hydroxyurea derivatives as hypoglycemic agents, J.
Med. Chem. 36 (1993) 2238–2240.
[27] Hughes D.L., The Mitsunobu reaction. Organic reactions, Wi-
ley, New York, 1992, pp. 335–656.
[16] Cantello B.C., Cawthorne M.A., Haigh D., Hindley R.M.,
Smith S.A., Thurlby P.L., The synthesis of BRL-49653 — a
novel and potent antihyperglycemic agent, Biorg. Med. Chem.
Lett. 4 (1994) 1181–1184.
[28] Malamas M.S., Gunawan I., Palka C., Aralkyl-1,2,4-oxadiazo-
lidine-3,5-diones as antihyperglycemic agents, US Patent
5480896 (1996).
[17] Cantello B.C., Cawthorne M.A., Cottam G.O., Duff P.T.,
Haigh D., Hindley R.M, Lister C.A., Smith S.A., Thurrlby
[29] Shafrir E., Animal models of non-insulin-dependent diabetes,
P.L.,
[[ꢀ-(Heterocycloamino)alkoxy]banzyl]-2,4-thiazolidine-
Diabetes/Metab. Rev. 8 (1992) 179–208.
diones as potent antihypergl; ycemic agents, J. Med. Chem. 37
(1994) 3977–3985.
[30] Lehman J.M., Moore L.B., Smith-Oliver T.A., Wilkinson
W.O., Kliewer S.A., An antidiabetic thiazolidinedione is a high
affinity ligands for peroxisome proliferator-activated receptor g
(PPAR-g), J. Biol. Chem. 270 (1995) 12953–12956.
[31] Malamas M.S., Gunawan I., Palka C., di-Oxadiazolidine
derivatives as antihyperglycemic agents, US Patent 5420146
(1995).
[18] Reddy K.A., Lohray B.B., Bhushan V., Bajji A.C., Reddy
K.V., Reddy P.R., Krishna T.H., Rao I.N., Jajoo H.K., Rao
N.V., Chakrabarti R., Dileepkumar T., Rajagopalan R., Novel
and hypolipidemic agents. 3. Benzofuran-containing thiazo-
lidinediones, J. Med. Chem. 42 (1992) 1927–1940.
[19] Matsumoto K., Miyake S., Yano M., Tominaga Y., Increase of
lipoprotein (a) with troglitazone, Lancet 350 (1997) 1748–1749
additional information in Scrips 2282 (1997) 21, and Scrips
2292 (1997) 20.
[32] Heing L.G., Determination of total serum insulin (IRI) in
insulin-treated diabetic patients, Diabetologia 8 (1972) 260–
266.