Synthesis and aldose reductase inhibitory activity of 5-arylidene-2,4-thiazolidinediones

Article abstract of DOI:10.1016/S0968-0896(01)00366-2

Several (Z)-5-arylidene-2,4- hiazolidinediones were synthesized and tested as aldose reductase inhibitors (ARIs). The most active of the N-unsubstituted derivatives (2) exerted the same inhibitory activity of Sorbinil. The introduction of an acetic side chain on N-3 of the thiazolidinedione moiety led to a marked increase in lending inhibitory activi y, conducting to the discovery of a very potent ARI (4c), whose activity level (IC₅₀ =0.1 μM) was in the same range of Tolrestat . Moreover, the corresponding methyl esters (3), devoid of any acidic functionality, showed appreciable inhibitory activity similar to that of the N-unsubstituted compounds. It was also found that the substitution pattern on the 5-benzylidene moiety markedly influenced the activity of N-unsubstituted 2,4-thiazolidinediones 2, compounds with substituents at the meta position being generally more effective than the para-substituted ones; however, this SAR was not evidenced in acetates 3 and acids 4. Copyright

Full text of DOI:10.1016/S0968-0896(01)00366-2

Bioorganic & Medicinal Chemistry 10 (2002) 1077–1084
Synthesis and Aldose Reductase Inhibitory Activity of
5-Arylidene-2,4-thiazolidinedionesy
G. Bruno,a L. Costantino,b C. Curinga,c R. Maccari,c,* F. Monforte,c
F. Nicolo,a R. Ottanac and M.G. Vigoritac
a
` `
Dipartimento Ch. Inorg., Chim. Anal. e Ch.-Fis, Facolta Scienze MMFFNN, Universita di Messina,
Salita Sperone 31, 98166 Messina, Italy
`
b
`
Dipartimento Sc. Farmaceutiche, Facolta Farmacia, Universita di Modena e Reggio Emilia,
Via G. Campi 183, 41100 Modena, Italy
` `
Dipartimento Farmaco-chimico, Facolta Farmacia, Universita di Messina, Vl. SS. Annunziata, 98168 Messina, Italy
c
Received 23July 2001; accepted 12 October 2001
Abstract—Several (Z)-5-arylidene-2,4-thiazolidinediones were synthesized and tested as aldose reductase inhibitors (ARIs). The
most active of the N-unsubstituted derivatives (2) exerted the same inhibitory activity of Sorbinil. The introduction of an acetic side
chain on N-3of the thiazolidinedione moiety led to a marked increase in lending inhibitory activity, conducting to the discovery of
a very potent ARI (4c), whose activity level (IC50=0.13 mM) was in the same range of Tolrestat. Moreover, the corresponding
methyl esters (3), devoid of any acidic functionality, showed appreciable inhibitory activity similar to that of the N-unsubstituted
compounds. It was also found that the substitution pattern on the 5-benzylidene moiety markedly influenced the activity of N-
unsubstituted 2,4-thiazolidinediones 2, compounds with substituents at the meta position being generally more effective than the
para-substituted ones; however, this SAR was not evidenced in acetates 3 and acids 4. # 2002 Elsevier Science Ltd. All rights
reserved.
Introduction
to prevent or delay the onset and to minimize the serious-
ness of chronic diabetic complications.
Aldose reductase (EC 1.1.1.21, ALR2), a member of the
aldo-keto reductase superfamily, is the first enzyme of the
polyol pathway: it catalyzes the NADPH-dependent
reduction of glucose to sorbitol, which in turn is oxidized
by sorbitol dehydrogenase to fructose. ALR2 has low affi-
nity for glucose, so there is little flux of glucose through the
polyol pathway under normal conditions. However, under
conditions of hyperglycaemia such as in diabetes, there is a
marked increase of glucose metabolized by the polyol
pathway in tissues possessing insulin-independent glu-
cose transport. A large amount of evidence has demon-
strated a link between enhanced metabolism of glucose
through the polyol pathway and the onset and progression
of long-term diabetic complications (retinopathy, catar-
acts, neuropathy and nephropathy).1À5 Therefore, ALR2
inhibition has received attention as an attractive strategy
Over the last 20 years, numerous structurally different
compounds have been reported to inhibit ALR2, but
many of them have produced little clinical benefit in
human studies, because of unfavourable pharmacoki-
netic properties, toxic side-effects or poor potency.1,3,6À11
Two main classes of orally active aldose reductase inhi-
bitors (ARIs) have been clinically tested: cyclic imides
(mostly hydantoins, for example Sorbinil) and car-
boxylic acids (for example Tolrestat). The carboxylic
acid derivatives in vitro are very active; however, in vivo
they are generally less active than imides, probably
owing to their lower pKa values that can result in less
favourable pharmacokinetics. Currently, Epalrestat
(ONO-2235) is the only ARI available on the market.8,12
There is a great interest in 2,4-thiazolidinedione deriva-
tives as ARIs,3,8,13À15 since they can be viewed as
hydantoin bioisosters potentially free of the hypersensi-
tivity reactions which are linked to the presence of the
hydantoin system. Moreover, they are useful both in the
treatment of non insulin-dependent diabetes mellitus
*Corresponding author. Tel.:+39-090-676-6406; fax:+39-090-355613;
e-mail: rmaccari@pharma.unime.it
A preterninary account of this work was presented at Italian/Hungar-
ian/Polish Joint Meeting on Medicinal Chemistry, Giardini Naxos,
Italy, 28 September–1 October 1999.
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00366-2

1078
G. Bruno et al. / Bioorg. Med. Chem. 10 (2002) 1077–1084
1
(NIDDM)16À24 and in the prevention of its late compli-
cations. In fact, to date, several 2,4-thiazolidinediones
have been patented as antihyperglycaemic and ALR2
inhibitory agents.3,8
The H NMR spectra of compounds 2 show only one
kind of 5-methylidene proton in the range 7.72–7.97 ppm
(Table 2). In the 13C NMR spectra, the signals of 5-meth-
ylidene carbon (at 130.1–132.8 ppm) and thiazolidine-
dione C5 (in the range 122.0–126.7 ppm), as well as the
absence of the 5-CH2 resonance of thiazolidinedione 1,
are diagnostic for the structural assigment of 2 (Table 4).
On these considerations, in a search for new ARIs, we
have synthesized and tested 5-arylidene-2,4-thiazolidi-
nediones 2 and 4. They appear to possess the essential
structural requisites (an acidic proton, hydrogen-bond
acceptor groups and an aromatic moiety) for ALR2
inhibitory effect, in accordance with known pharmaco-
phoric requirements.25À31 In particular it is known that
the presence of an acidic functionality is an important
requirement for all ARIs, since they interact, in their
ionized form, with the active site of the enzyme.3,26À28
Compounds of series 2 possess the acidic hydrogen of
the thiazolidinedione imidic moiety, with pKa higher
than those of carboxylic acids.18,32 Thus, the corre-
sponding carboxylic acids 4 have been synthesized and
their activity compared with that of compounds 2.
Compounds of structure 3, which do not possess acidic
protons, have also been tested.
Compounds (Z)-2 appear to be thermodynamically
stable. Attempts to isomerize them by irradiation with a
high pressure mercury-vapour lamp in acetonitrile for
1
24 h at room temperature failed. H NMR monitoring
of the experiment did not reveal any change, contrary to
what has been reported for some analogous com-
pounds.19,35,36
1H and 13C NMR spectra of compounds 3 and 4 show a
single signal due to the resonance of 5-methylidene
group (the proton in the range 7.75–8.01 ppm and the
carbon at 130.1–134.9 ppm) and a singlet attributable to
N–CH2 resonance (the proton at 4.22–4.56 ppm and the
carbon at 40.9–42.5 ppm) (Tables 3and 5).
Chemistry
Table 1. Chemical-physical data of 2,4-thiazolidinediones 2–4
5-Arylidene-2,4-thiazolidinediones 2 were synthesized
by the condensation, according to Knoevenagel, of
commercially available 2,4-thiazolidinedione 1 with
various meta- and para-substituted benzaldehydes,
using piperidine as base, in refluxing ethanol, according
to a known procedure.19
Compd
Mp (ꢀC)a
Yield (%)
Formula
2a
2b
2c
2d
2e
2f
2g
2h
2i
169–172
178–180
146–148
187–190
180–182
263–265
210–212
183–186
216–218
206–208
288–289
112–115
92–95
52
56
65
52
57
88
65
55
70
54
90
96
78
76
98
C
C10H6FNO2S
C11H9NO2S
C16H11NO3S
C11H9NO3S
C11H6F3NO2S
C17H12N2O3S
C10H6FNO2S
C16H11NO3S
C11H9NO3S
C11H6F3NO2S
C17H12N2O3S
C13H10FNO4S
C14H13NO4S
C19H15NO5S
C14H13NO5S
14H10F3NO4S
13H10FNO4S
C14H10F3NO4S
C12H8FNO4S
C13H11NO4S
C18H13NO5S
C13H11NO5S
C13H8F3NO4S
C12H8FNO4S
The treatment of 2 with methyl bromoacetate provided
esters 3 which in turn were hydrolyzed in acidic medium
providing acids 4 in high yields (Table 1, Scheme 1).
2j
2k
3a
3b
3c
3d
3e
3g
3j
4a
4b
4c
4d
4e
4g
The structures of all synthesized compounds were
1
assigned on the basis of elemental analyses, IR, H and
13C NMR data (Tables 2–5). In particular, the Z config-
uration of the exocyclic double bond was unambigously
determined on the basis of X-ray diffractometric analy-
sis of compound 2e (Fig. 1), which confirmed what we
117–120
48–50
170–17388
221–22378
192–195
215–216
230–231
216–217
153–155
213–215
240–245
(dec.)
C
89
96
98
94
98
98
96
1
had presumed by means of H NMR experiments: in
fact, in the spectra of 2 recorded in the presence of
Eu(fod)3, a significant downfield shift of the methylidene
proton resonance had been observed, without any effect
on the aromatic protons signals. (E)-2 isomers were
never obtained, also under different experimental con-
ditions (i.e., performing the condensation at room tem-
perature, or in acetic acid and in the presence of sodium
acetate,33 or in toluene and in presence of piperidine34).
4j
251–252
97
C13H8F3NO4S
aRecrystallization solvent: MeOH for compounds 2 (EtOH/H2O for
2f); EtOH/H2O for 3; EtOH for 4.

G. Bruno et al. / Bioorg. Med. Chem. 10 (2002) 1077–1084
1079
Scheme 1. (a) C5H11N, C2H5OH, reflux; then, p-aminophenol for 2f and 2k; (b) NaH, DMF, 80ꢀC; (c) BrCH2COOCH3, 80ꢀC; (d) AcOH, HCl,
reflux.
1
Table 2.
Compd
2a
H NMR data of 5-arylidene-2,4-thiazolidinediones 2a
CHb arylidene
7.78
Aromatic
NHc
Others
—
7.30–7.35
7.40–7.45
7.54–7.61
7.27–7.44
7.07–7.2312.64
7.34–7.56
7.05-7.16
12.71
2b
2c
7.72
7.76
12.62
12.63
2.35 (s, 3H, CH3)
—
2d
7.78
7.97
3.81 (s, 3H, OCH3)
7.43–7.49
7.76–7.91
2e
2f
12.36
12.20
—
8.67 (s, 1H, CH¼N), 9.56
(s, 1H, OH, exchangeable with D2O)
—
7.836.81, 7.26, 7.68, 7.72,
7.96, 8.07
7.39 (dd, J=10.1, 8.6)
7.68 (dd, J=7.8, 5.5)
7.08–7.25
2g
2h
7.80
7.77
12.62
12.56
—
7.43, 7.61
7.09, 7.55
7.80, 7.88
6.84, 7.29, 7.72, 8.0312.28
2i
2j
2k
7.75
7.86
7.84
12.44
12.68
3.82 (s, 3H, OCH3)
—
8.69 (s, 1H, CH
¼N), 9.63
(s, 1H, OH, exchangeable with D2O)
J are expressed in Hz.
aDMSO-d6 solution.
bSinglet.
cBroad signal (exchangeable with D2O).
Results and Discussion
The structure of acids 4 was also assigned on the basis
of their IR spectra, which show a very broad band at
3400–2450 cmÀ1 attributable to the stretching of car-
boxylic OH. In addition, the treatment of 4 with
methanol and catalytic amount of H2SO4 provides par-
ent esters 3.
The ALR2 inhibitory activity of compounds 2–4 was
evaluated in vitro by measuring their ability to inhibit
the reduction of d,l-glyceraldehyde, using partially
purified ALR2 from bovine lenses (Tables 6 and 7).

1080
G. Bruno et al. / Bioorg. Med. Chem. 10 (2002) 1077–1084
Table 3. 1H NMR data of methyl 5-arylidene-2,4-thiazolidinedione-3-acetates 3a and 5-arylidene-2,4-thiazolidinedione-3-acetic acids 4b
d
Compd
CHc arylidene
Aromatic
CH2
Others
3a
3b
7.93
7.88
7.34–7.66
7.21–7.40
4.53
4.48
3.72 (s, 3H, COOCH3)
2.40 (s, 3H, CH3),
3.77 (s, 3H, COOCH3)
3.81 (s, 3H, COOCH3)
3.79 (s, 3H, COOCH3),
3.86 (s, 3H, OCH3)
3.80 (s, 3H, COOCH3)
3.80 (s, 3H, COOCH3)
3.80 (s, 3H, COOCH3)
—
3c
3d
7.90
7.91
7.08–7.43
6.98–7.42
4.51
4.50
3e
3g
3j
4a
4b
4c
4d
4e
4g
4j
7.96
7.91
7.95
7.90
7.75
7.82
7.93
7.95
7.80
8.01
7.60–7.77
7.19, 7.52
7.63, 7.74
7.14–7.48
7.04–7.21
7.05–7.46
6.99–7.41
7.77–7.92
7.01, 7.32
7.84–7.92
4.51
4.50
4.51
4.56
H, CH 4.232.21 (s, 3
)
3
4.35
4.56
4.41
4.22
4.41
—
3.87 (s, 3H, OCH3)
—
—
—
J are expressed in Hz.
aCDCl3 solution.
bDMSO-d6 solution.
c,dSinglets.
Table 4. 13C NMR data of 5-arylidene-2,4-thiazolidinediones 2a
Compd
CH
Aromatic
C5
C¼O
Others
—
2a
130.4
116.7 (d, JCF=22.2), 117.2
(d, JCF=21.1), 125.5 (d, JCF=2.25),
131.3 (d, JCF=8.55), 135.4
125.2
167.2, 167.6
(d, JCF=7.95), 162.3(d, JCF=243.6)
126.4, 128.6, 130.0, 131.2, 132.4, 138.1
118.8, 119.4, 120.1, 124.3,
2b
2c
130.5
131.1
122.7
124.6
166.7, 167.3
167.3, 167.8
20.3 (CH3)
—
124.9, 130.3, 135.0, 155.9, 157.6
115.3, 116.2, 121.9, 130.4, 134.3, 159.5
126.5 (q, JCF=3.75), 126.8
2d
2e
131.7
132.8
123.8
125.8
167.2, 167.8
167.1, 167.4
55.3 (OCH3)
127.4 (q, JCF=270.0, CF3)
(q, JCF=3.75), 130.0 (q, JCF=32.2),
130.2, 130.4, 134.2
116.5, 123.5, 128.9, 131.8,
133.0, 134.3, 138.1, 142.8, 157.3
116.5 (d, JCF=21.75),129.8
(d, JCF=3.0),132.6 (d, JCF=9),
163.0 (d, JCF=255)
2f
130.5
130.9
125.3168.1, 168.6
156.8 (CH
¼N)
2g
123.4
122.0
167.5, 168.0
—
—
2h
131.3
118.4, 119.8, 124.6, 130.4,
132.2, 127.8, 155.5, 158.9
114.9, 120.3,132.2, 161.0
167.5, 168.1
2i
2j
2k
131.9
130.1
130.4
125.5
126.7
124.7
167.5, 168.1
167.3, 167.8
167.5, 167.9
55.5 (OCH3)
124.0 (q, JCF=270.0, CF3)
155.9 (CH¼N)
126.2, 129.6 (q, JCF=32.2), 129.7, 137.1
115.8, 122.9, 128.9, 130.9,
135.1, 137.8, 142.2, 156.8
J are expressed in Hz.
aDMSO-d6 solution.
5-Arylidene-2,4-thiazolidinediones 2 with various sub-
stitution patterns on the benzylidene moiety were tested
and displayed IC50 values ranging from 1.86 to 40.83
mM (Table 6). The most active of them (2f) exerted inhi-
bitory activity (IC50=1.86 mM) lying in the same range
of Sorbinil, followed by 2c and 2g with IC50=6.14 and
8.21 mM, respectively.37 Meta-substitution patterns on
the 5-benzylidene moiety generally improved activity,
independently of the nature of the substituent group.
effective as its counterpart 2c, displayed the most sig-
nificant inhibitory properties (IC50=0.13 mM), nearly
reaching those of Tolrestat. All acids 4 showed to be at
least threefold (from 3- to 23-fold) more active than
Sorbinil.
On the other hand, esters 3, devoid of any acidic proton,
in general proved to have ALR2 inhibitory properties
similar to those of 2,4-thiazolidinediones 2 (Table 7),
with the exception of 3c and 3j which were 5 and 10
times, respectively, more effective than their counter-
parts 2c and 2j. In particular, 3c proved to be the most
active among these acetates (IC50=1.32 mM) and also
twice as active as Sorbinil, confirming the positive
influence exerted by meta-phenoxy substitution on
activity. Thus, in this series of compounds, the presence
The insertion on N-3of an acetic chain led to 2,4-thia-
zolidinedione-3-acetic acids 4, which were significantly
more potent than N-unsubstituted analogues 2 (Table
7). The increase in the inhibitory activity varied from
about 10 times (4g) to almost 100 times (4j). The 3-
phenoxy substituted derivative (4c), almost 50 times as

G. Bruno et al. / Bioorg. Med. Chem. 10 (2002) 1077–1084
1081
Table 5. 13C NMR data of methyl 5-arylidene-2,4-thiazolidinedione-3-acetates 3a and 5-arylidene-2,4- thiazolidinedione-3-acetic acids 4b
Compd
CH
Aromatic
C5
CH2
41.9
C¼O
Others
3a
133.0
116.5 (d, JCF=22.5), 117.5
(d, JCF=21.3), 125.8 (d, JCF=2.85),
130.8 (d, JCF=8.25), 135.0
122.5
165.0, 166.5, 166.6
52.8 (COOCH3)
(d, JCF=7.65), 162.8 (d, JCF=247)
127.3, 129.1, 130.9, 131.6, 132.9, 139.0
3b
3c
3d
3e
3g
134.9
133.8
130.2
132.6
133.5
120.5
121.6
120.6
121.7
120.6
41.8
41.7
41.8
42.0
41.9
164.2, 166.2
165.1, 166.5, 168.1
164.4, 166.7
36.5 (CH3),
52.8 (COOCH3)
52.6 (COOCH3)
119.0, 119.3, 120.4, 124.0, 124.5,
129.8, 130.4, 134.4, 155.9, 158.1
116.7, 117.1, 119.2, 122.6, 134.2, 159.9
52.8 (COOCH3),
55.3(OCH 3)
52.9 (COOCH3),
126.9, 127.0, 127.2, 129.9,
131.9 (q, JCF=37.5), 133.8
116.6 (d, JCF=22.2), 129.3
166.5, 166.7
123.5 (q, JCF=270.0, CF3)
52.9 (COOCH3)
164.1, 165.5, 166.6
(d, JCF=3.45), 132.3 (d, JCF=8.85),
163.7 (d, JCF=253.3)
126.1, 130.2, 131.5 (q, JCF=34.5), 136.3
3j
132.6
132.1
121.7
122.2
42.1
41.7
164.1, 166.5, 166.6
163.8, 166.3, 167.4
53.0 (COOCH3),
123.5 (q, JCF=277.5, CF3)
—
4a
115.9 (d, JCF=22.2), 117.0
(d, JCF=21.1), 125.3(d, JCF=2.85),
130.4 (d, JCF=9.3), 134.5
(d, JCF=7.65), 162.2 (d, JCF=253.5)
126.7, 128.7, 131.0, 134.0, 132.5, 138.5
117.8, 118.1, 119.1, 122.9, 123.5,
128.7, 129.4, 133.3, 154.7, 156.7
114.6, 116.0, 122.1, 133.7, 159.4
127.0 (q, JCF=4.5), 127.3
4b
4c
130.4
132.1
120.3
120.6
41.6
40.9
166.5, 167.6
163.8, 165.6, 166.6
20.3 (CH3)
—
4d
4e
130.1
132.8
120.9
122.9
41.6
42.4
167.3, 167.5
164.8, 166.5, 167.9
54.8 (OCH3)
127.7 (q, JCF=271.5, CF3)
(q, JCF=4.5), 130.0 (q, JCF=31.9),
130.6, 132.3, 133.9
116.0 (d, JCF=21.9), 128.9,
4g
4j
132.5
130.8
123.0
123.8
41.6
42.5
165.1, 167.3, 167.5
164.9, 166.7, 168.1
—
131.8 (d, JCF=8.85), 160.0 (d, JCF=248.2)
126.3, 130.2 (q, JCF=31.5),132.2, 136.8
123.9 (q, JCF=270.75, CF3)
J are expressed in Hz.
aCDCl3 solution.
bDMSO-d6 solution.
Figure 1. View of 2e on the molecule mean plane showing the atom numbering scheme. Empty atoms are of the centrosymmetric molecule at (1-x, 1-
y, -1-z) evidencing the strong H-bonds of the dimer. Thermal ellipsoids are drawn at 50% of probability, while H size is arbitrary.
of an acidic proton did not appear to be an essential
structural requisite. Instead, it could be hypothesized that
the N-3 acetate chain was able to get and bind (through H-
bonds) the polar positively charged recognition region of
the ALR2 active site formed by Tyr48, His110, Trp111
residues and the nicotinamide ring of NADP+.3,27,28
most active ones. Their ALR2 inhibitory potency is lar-
gely independent of the position of substitutents on the
benzylidene moiety, whereas in series 2 substituents in
meta position are generally more favourable than para-
substitution. Since the most active compounds in all
series 2–4 bear an additional aromatic ring (see com-
pounds c and f), it can be hypothesized that a larger
lipophilic moiety increases the enzyme–inhibitor com-
plex stability through interactions with the hydrophobic
region of the ALR2 active site. A lot of observations
were reported about the hydrophobic nature of this
active site which greatly favours the binding with lipo-
philic compounds.3,27,28 In addition, in 2f the p-amino-
phenol moiety could be involved in further interactions
through H-bonds.
In any case, acids 4 proved to be more efficacious inhi-
bitors than esters 3, clearly due to the presence of the
carboxylic anionic head of the N-3acetic chain.
In conclusion, the in vitro results here reported indi-
cated 5-arylidene-2,4-thiazolidinediones 2–4 as poten-
tially promising ARIs: among them, 3-acetic acids 4,
showing IC50 values lower than 1 mM, proved to be the

1082
G. Bruno et al. / Bioorg. Med. Chem. 10 (2002) 1077–1084
The importance of the carboxylic function on N-3was
stressed by the finding that all esters 3 were less active
than acids 4. However, this activity order (4>3) might
be reversed in vivo, since the ester function could
improve pharmacokinetics, unless it undergoes fast
hydrolysis. Analogously, N-unsubstituted derivatives 2,
possessing pKa values higher than carboxylic acids,
might display more favourable pharmacokinetic pro-
files, without the adverse effects related to the hydantoin
moiety of Sorbinil and analogues.
of the data can be obtained, free of charge, on application
to CCDC, 12 Union Road, Cambridge CB2 lEZ, UK
(fax:+44-1223-336033; e-mail: deposit@ccdc.cam.ac.uk).
General method for the synthesis of 5-arylidene-2,4-
thiazolidinediones (2)
A mixture of 2,4-thiazolidinedione 1 (2.4 g, 20 mmol),
aldehyde (20 mmol), piperidine (1.4 g, 16 mmol) and EtOH
(150 mL) was refluxed for 16–24 h. The reaction mixture
was poured into H2O and acidified with AcOH to give 2a–
e,g–j as solids which were recrystallized from methanol.
Experimental
Synthesis of 5-[3-(4-hydroxyphenylaminomethylidene)-
phenyl]-methylidene-2,4-thiazolidinedione (2f) and 5-[4-
(4-hydroxyphenylaminomethylidene)phenyl]-methylidene-
2,4-thiazolidinedione (2k). A mixture of 2,4-thiazolidi-
nedione 1 (2.4 g, 20 mmol), piperidine (1.4 g, 16 mmol)
and EtOH (120 mL) was added dropwise to a solution
of isophthalaldehyde or terephthalaldehyde (2.68 g, 20
mmol) in EtOH. The reaction mixture was refluxed for
24 h and then poured into H2O and acidified with
AcOH to give 5-(3-formylphenyl)-methylidene-2,4-thia-
zolidinedione 5 and 5-(4-formylphenyl)-methylidene-
2,4-thiazolidinedione 6. The solid products were pur-
ified by column chromatography on silica gel, using
CHCl3 (for 5) or diethyl ether/light petroleum 40–
60 ꢀC=7:3(for 6) as eluant.
Melting points were recorded on a Kofler hot-stage
apparatus and are uncorrected. TLC controls were car-
ried out on precoated silica gel plates (F 254 Merck). IR
spectra were obtained with a Perkin-Elmer 683spectro-
photometer as Nujol or esachlorobutadiene mulls. 1H and
13C NMR spectra were recorded on a Varian 300 MHz
spectrometer; chemical shifts are given in d units (ppm)
relative to internal standard Me4Si and refer to CDCl3
or DMSO-d6 solutions. 13C NMR spectra were deter-
mined by Attached Proton Test (APT) experiments and
the resonances were always attributed by proton-carbon
heteronuclear chemical shift correlation. Europium
tris(6,6,7,7,8,8,8-heptafluoro-2,2-dimethyl-3,5-octane-
dionate), Eu(fod)3, was used as lanthanide shift reagent.
Mass spectra were recorded on a Shimadzu GC17A/
MSQP5050 spectrometer. Elemental analyses (C, H, N)
were performed at Redox Labs, Cologno Monzese,
Milan (Italy). Irradiation of compounds 2 were carried
out by using a high pressure mercury-vapour lamp (250
W) in acetonitrile for 24 h at room temperature. Crystals
suitable for X-ray analysis were obtained by recrystalliza-
tion of compound 2e from methanol solution. Diffraction
data were collected at room temperature from a color-
less 0.92ꢁ0.75ꢁ0.36 mm3 prismatic crystal sample by
using a Siemens P4 automated four-circle single-crystal
diffractometer with graphite-monochromated MoKa
radiation (l=0.71073A ). Crystallographic data (exclud-
ing structure factors) for compound 2e have been depos-
ited with the Cambridge Crystallographic Data Centre
as supplementary publication no. CCDC 167055. Copies
A solution of p-aminophenol (1.1 g, 10 mmol) in EtOH
was added to a solution of 5 or 6 (2.8 g, 12 mmol) in
EtOH and the reaction mixture was stirred at room
temperature for 24 h. The solid product (2f or 2k) was
isolated by filtration and recrystallized from methanol
(2k) or EtOH/H2O (2f).
General method for the synthesis of methyl 5-arylidene-
2,4-thiazolidinedione-3-acetates (3)
Sodium hydride (0.576 g, 24 mmol) was added
portionwise to a solution of 5-arylidene-2,4-thiazol-
idinedione 2 (20 mmol) in dry DMF and the mixture was
stirred at 80 ꢀC for 1.5 h. The mixture was cooled to
room temperature and a solution of methyl bromoace-
tate (3.7 g, 24 mmol) in dry DMF was added dropwise.
After being stirred at 80 ꢀC for 15–20 h, the reaction
Table 6. In vitro bovine lenses ALR2 inhibitory activity of 5-aryli-
dene-2,4-thiazolidinediones 2
a
Table 7. In vitro bovine lenses ALR2 inhibitory activity of methyl 5-
arylidene-2,4-thiazolidinedione-3-acetates 3 and 5-arylidene-2,4-thia-
zolidinedione-3-acetic acids 4
Compd
IC50
2a
2b
9.10 (8.60—9.63)
9.23(8.58—9.93)
a
a
2c
2d
2e
2f
2g
2h
2i
2j
2k
Sorbinil
Tolrestat
6.14 (5.54—6.81)
13.28 (11.16–15.81)
12.81 (10.41–15.76)
1.86 (1.46—2.37)
8.21 (7.44—9.06)
41% (37 mM)
40.83(45.92–41.86)
31.49 (29.18–33.99)
22.93(19.20–27.38)
3.04 (2.91–3.52)
Compd
IC50
Compd
IC50
3a
3b
3c
3d
3e
3g
3j
40% (76 mM)
4a
4b
4c
4d
4e
4g
4j
0.74 (0.61–0.90)
0.65 (0.50–0.84)
0.13(0.11–0.15)
0.48 (0.36–0.63)
0.47 (0.41–0.54)
1.14 (0.93–1.40)
0.46 (0.41–0.52)
10.10 (8.70–11.73)
1.32 (1.00–1.74)
8.87 (6.28–12.52)
28.67 (24.63–33.37)
12.90 (11.50–14.47)
3.44 (2.93–4.04)
3.04 (2.91–3.52)
0.074 (0.064–0.084)
Sorbinil
Tolrestat
0.074 (0.064–0.084)
aIC50 (mM) (95% C.L.) or % inhibition at the given concentration.
aIC50 (mM) (95% CL) or % inhibition at the given concentration.

G. Bruno et al. / Bioorg. Med. Chem. 10 (2002) 1077–1084
1083
mixture was poured into H2O and the solid product fil-
tered and recrystallized from EtOH/H2O.
4. Wolffenbuttel, B. H.; van Haeften, T. W. Drugs 1995, 50, 263 .
5. Porte, D., Jr.; Schwartz, M. W. Science 1996, 272, 699.
6. Sarges, R. In Advances in Drug Research Vol. 18; Testa, B.,
Ed.; Academic: London, 1989; pp 139–175.
General method for the synthesis of 5-arylidene-2,4-
thiazolidinedione-3-acetic acids (4)
7. Costantino, L.; Rastelli, G.; Vianello, P.; Cignarella, G.;
Barlocco, D. Med. Res. Rev. 1999, 19, 3 .
8. Costantino, L.; Rastelli, G.; Gamberoni, M. C.; Barlocco,
D. Exp. Opin. Ther. Pat. 2000, 10, 1245.
9. Pfeifer, M. A.; Schumer, M. P.; Gelber, D. A. Diabetes
1997, 46 (Suppl. 2), S82.
10. Inoue, J.; Cui, Y.; Sakai, O.; Nakamura, Y.; Kogiso, H.;
Kador, P. F. Bioorg. Med. Chem. 2000, 8, 2167.
11. Oka, M.; Matsumoto, Y.; Sugiyama, S.; Tsuruta, N.;
Matsushima, M. J. Med. Chem. 2000, 43, 2479.
12. Castaner, J.; Prous, J. Drugs Future, 1987, 12, 336.
13. Larson, E. R.; Lipinski, C. A.; Sarges, R. Med. Res. Rev.
1988, 8, 159.
A mixture of acetate 3 (10 mmol), glacial AcOH (40
mL) and HCl 12 N (10 mL) was refluxed for 2 h. After
evaporation in vacuo, the residue was refluxed again
with AcOH (40 mL) and HCl (10 mL) for 2 h. After
evaporation to dryness in vacuo, the crude solid was
washed with H2O and recrystallized from EtOH pro-
viding pure carboxylic acid 4.
Enzyme section
14. Castaner, J.; Prous, J. Drugs Future, 1991, 16, 895
15. Fresneau, P.; Cussac, M.; Morand, J.; Szymonski, B.;
Tranqui, D.; Leclerc, G. J. Med. Chem. 1998, 41, 4706.
16. Sohda, T.; Mizuno, K.; Imamiya, E.; Sugiyama, Y.;
Fujita, T.; Kawmatsu, Y. Chem. Pharm. Bull. 1982, 30, 3580.
17. (a) Drugs Future 1989, 14, 846. (b) Drugs Future 1997, 22,
1055.
18. Zask, A.; Jirkovsky, I.; Nowicki, J. W.; McCaleb, M. L. J.
Med. Chem. 1990, 33, 1418.
19. Momose, Y.; Meguro, K.; Ikeda, H.; Hatanaka, C.; Oi, S.;
Sohda, T. Chem. Pharm. Bull. 1991, 39, 1440.
20. Cantello, B. C. C.; Cawthorne, M. A.; Cottam, G. P.;
Duff, P. T.; Haigh, D.; Hindley, R. M.; Lister, C. A.; Smith,
S. A.; Thurlby, P. L. J. Med. Chem. 1994, 37, 3977.
21. 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. J. Med. Chem. 1996,
39, 665.
22. 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. S. M.; Chakrabarti, R.; Dileepkumar, T.;
Rajagopalan, R. J. Med. Chem. 1999, 42, 1927.
23. Lohray, B. B.; Bhushan, V. Drugs Future 1999, 24, 751.
24. Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke,
B. R. J. Med. Chem. 2000, 43, 527.
25. Kador, P. F.; Sharpless, N. E. Mol. Pharmacol. 1983, 24, 521.
26. Lee, Y. S.; Pearlstein, R.; Kador, P. F. J. Med. Chem.
1994, 37, 787.
27. Urzhumtsev, A.; Tete-Favier, F.; Mitschler, A.; Barban-
ton, J.; Barth, P.; Urzhumtseva, L.; Biellmann, J. F.; Pod-
jarny, A.; Moras, D. Structure 1997, 5, 601.
Sorbinil was a gift from Pfizer. Tolrestat was synthe-
sized according to a published procedure.38
Calf lenses for the purification of ALR2 were obtained
locally from freshly slaughtered animals. The capsule
was incized and the frozen lenses were suspended in
sodium potassium phosphate buffer pH 7 containing 5
mM DTT (1 g tissue/3.5 mL) and stirred in an ice-cold
bath for 1 h. The suspension was then centrifuged at
22,000g at 4 ꢀC for 40 min and the supernatant was
subjected to ion exchange chromatography on DE52.
Enzyme activity was measured by monitoring the
change in absorbance at 340 nm which accompanies the
oxidation of b-NADPH catalyzed by ALR2. The assay
was performed at 37 ꢀC as previously described,39 using
4.7 mM d,l-glyceraldehyde as substrate in 0.25 M
sodium phosphate buffer pH 6.8 containing 0.38 M
ammonium sulphate and 0.11 mM NADPH. The sensi-
tivity of ALR2 to inhibition by different ARIs was tes-
ted in the above assay conditions by including the
inhibitors dissolved in DMSO at the desired concentra-
tions in the reaction mixture. The DMSO in the assay
mixture was kept at constant concentration of 1%. A
reference blank containing all the above reagents
except the substrate was used to correct for the non-
enzymatic oxidation of NADPH. IC50 values (the con-
centration of the inhibitor required to produce 50%
inhibition of the enzyme catalyzed reaction) were
determined from least squares analyses of the linear
portion of the log dose–inhibition curves. Each curve
was generated using at least three concentrations of
inhibitor causing an inhibition between 20 and 80%
with two replicates at each concentration. The 95%
confidence limits (95% CL) were calculated from T
values for n-2, where n is the total number of determi-
nations.40
28. El-Kabbani, O.; Wilson, D. K.; Petrash, J. M.; Quiocho,
F. A. Mol. Vis. 1998, 4, 19.
29. Lee, Y. S.; Chen, Z.; Kador, P. F. Bioorg. Med. Chem.
1998, 6, 1811.
30. Hohman, T. C.; El-Kabbani, O.; Malamas, M. S.; Lai, K.;
Putilina, T.; McGowan, M. H.; Wane, Y. Q.; Carper, D. A.
Eur. J. Biochem. 1998, 256, 310.
31. Singh, S. B.; Malamas, M. S.; Hohman, T. C.; Nila-
kantan, R.; Carper, D. A.; Kitchen, D. J. Med. Chem. 2000,
43, 1062.
32. Lipinski, C. A.; Fiese, E. F.; Korst, R. J. Quant. Struct.-
Act. Relat 1991, 10, 109.
33. Unangst, P. C.; Connor, D. T.; Cetenko, W. A.; Sorenson,
R. J.; Kostlan, C. R.; Sircar, J. C.; Wright, C. D.; Schrier,
D. J.; Dyer, R. D. J. Med. Chem. 1994, 37, 322.
34. Lohray, B. B.; Bhushan, V.; Reddy, A. S.; Rao, P. B.;
Reddy, N. J.; Harikishore, P.; Haritha, N.; Vikramadityan,
R. K.; Chakrabarti, R.; Rajagopalan, R.; Katneni, K. J. Med.
Chem. 1999, 42, 2569.
References and Notes
1. Kador, P. F.; Kinoshita, J. H.; Sharpless, N. E. J. Med.
Chem. 1985, 28, 841.
2. Kador, P. F. Med. Res. Rev. 1988, 8, 325.
3. Costantino, L.; Rastelli, G.; Cignarella, G.; Vianello, P.;
Barlocco, D. Exp. Opin. Ther. Pat. 1997, 7, 843.
35. Murata, M.; Fujitani, B.; Minuta, H. Eur. J. Med. Chem.
1999, 34, 1061.

1084
G. Bruno et al. / Bioorg. Med. Chem. 10 (2002) 1077–1084
36. Ispida, T.; In, Y.; Inoue, M.; Tanaka, C.; Hamanaka, N.
J. Chem. Soc., Perkin Trans. 2 1990, 1085.
37. 2c and 2g also showed low in vivo toxicity (LD50>500
mg/kg po, mouse)
38. Sestany, K.; Bellini, F.; Fung, S.; Abraham, N.; Treas-
uywala, A.; Humber, L.; Duquesne, N. S.; Dvornik, D. J.
Med. Chem. 1984, 27, 255.
39. Costantino, L.; Rastelli, G.; Vescovini, K.; Cignarella, G.;
Vianello, P.; Del Corso, A.; Cappiello, M.; Mura, U.; Bar-
locco, D. J. Med. Chem. 1996, 39, 4396.
40. Tallarida, R.J.; Murray, R.B. Manual of Pharmacological
Calculations with Computer Programs, 2nd ed.; Springer: New
York, 1987.