On the prodrug potential of novel aldose reductase inhibitors with diphenylmethyleneaminooxycarboxylic acid structure

Article abstract of DOI:10.1016/S0928-0987(01)00192-0

Diphenylmethyleneaminooxycarboxylic acids were found to represent novel type inhibitors of the enzyme aldose reductase. Ester derivatives of the most active compound (3c) (IC₅₀ = 33 μM) were prepared as potential prodrugs and the rate of degradation was studied by treatment with buffers, plasma, and various hydrolytic enzymes. Whereas all compounds were not hydrolysed at physiological pH, incubation in the presence of enzyme led to hydrolysis. The rate of enzymatic degradation, however, depended on the nature of the ester function. Whereas the isopropyl ester (4) turned out to be the most stable compound, the ethyl ester (2c) could be cleaved in the presence of esterase and lipase, respectively. The benzylic and aromatic esters were found to be hydrolysed rapidly in the presence of lipase (benzyl ester, 7), or in plasma, by cholinesterase and esterase (phenyl ester, 6), respectively.

Full text of DOI:10.1016/S0928-0987(01)00192-0

European Journal of Pharmaceutical Sciences 15 (2002) 11–20
www.elsevier.nl/locate/ejps
On the prodrug potential of novel aldose reductase inhibitors with
diphenylmethyleneaminooxycarboxylic acid structure
Dietmar Rakowitza , , Barbara Matuszczak , Stefan Gritsch , Peter Hofbauer , Andreas Krassnigg ,
a
a
a
a
*
Luca Costantinob
aDepartment of Pharmaceutical Chemistry, Institute of Pharmacy, University of Innsbruck, Innrain 52a, A-6020 Innsbruck, Austria
b
`
Dipartimento di Scienze Farmaceutiche, Universita di Modena, Via Campi 183, I-41100 Modena, Italy
Received 6 March 2001; received in revised form 28 June 2001; accepted 16 August 2001
Abstract
Diphenylmethyleneaminooxycarboxylic acids were found to represent novel type inhibitors of the enzyme aldose reductase. Ester
derivatives of the most active compound (3c) (IC50533 mM) were prepared as potential prodrugs and the rate of degradation was studied
by treatment with buffers, plasma, and various hydrolytic enzymes. Whereas all compounds were not hydrolysed at physiological pH,
incubation in the presence of enzyme led to hydrolysis. The rate of enzymatic degradation, however, depended on the nature of the ester
function. Whereas the isopropyl ester (4) turned out to be the most stable compound, the ethyl ester (2c) could be cleaved in the presence
of esterase and lipase, respectively. The benzylic and aromatic esters were found to be hydrolysed rapidly in the presence of lipase
(benzyl ester, 7), or in plasma, by cholinesterase and esterase (phenyl ester, 6), respectively.
2002 Elsevier Science B.V. All rights
reserved.
Keywords: Aldose reductase inhibitors; Diphenylmethyleneaminooxycarboxylic acids; Oxime ethers; Prodrugs; Enzymatic hydrolysis
1. Introduction
aldose reductase, which is the rate limiting enzyme of the
polyol pathway, should therefore provide a pharmaco-
logical approach to the prevention and treatment of these
complications.
Most diabetic patients suffer from so-called long-term
complications such as neuropathy, nephropathy, re-
tinopathy and cataracts. These complications arise from
chronic hyperglycaemia, which causes damage to blood
vessels and peripheral nerves, greatly increasing the risk of
heart attack, stroke, blindness, amputation, and kidney
failure (Porte and Schwartz, 1996). Several metabolic
pathways have been implicated in the aetiology of sec-
ondary complications, such as excess flux through the
polyol pathway, elevated nonenzymatic glycation and
increased glycolytic metabolism (i.e. abnormally activated
intracellular messenger systems, particularly diacylglycerol
and protein kinase C) (Sarges and Oates, 1993). To date,
several studies in diabetic animals and man have clearly
shown a link between glucose metabolism via the polyol
pathway and diabetic complications (Kador et al., 1985;
Kador, 1988; Lipinski and Hutson, 1984). Inhibition of
Although numerous compounds have been selected as
inhibitors of aldose reductase (for a selection of the most
potent substances, see Anonymous, 1984, 1987a,b, 1989,
1995; Ashizawa and Aotsuka, 1998; Brittain and Wood,
1980; Itoh et al., 1992; Mylari et al., 1991; Oku et al.,
1987; Sestanj et al., 1984; Terashima et al., 1984), the
most potent compounds are mostly limited to cyclic imides
and carboxylic acids with the acidic proton being neces-
sary for inhibition of the enzyme.
Recently, we have prepared a series of diazinyl substi-
tuted methanonoximethers of type A (Scheme 1) as potent
inhibitors of aldose reductase (Rakowitz et al., 2000). For
this novel class of aldose reductase inhibitors, the follow-
ing structure–activity relationships can be established: the
carboxylic acid function seems to be essential for enzyme
inhibition since compounds lacking the terminal carboxylic
group were found to be inactive. On the contrary, the
geometry around the C=N double bond did not affect the
biological properties. The length of the alkyl spacer
between the oxime ether and the carboxylic acid function
*
Corresponding author. Tel.: 143-512-507-5247; fax: 143-512-507-
2940.
E-mail addresses: dietmar.rakowitz@uibk.ac.at (D. Rakowitz),
barbara.matuszczak@uibk.ac.at (B. Matuszczak).
0928-0987/02/$ – see front matter
PII: S0928-0987(01)00192-0
2002 Elsevier Science B.V. All rights reserved.

12
D. Rakowitz et al. / European Journal of Pharmaceutical Sciences 15 (2002) 11–20
Scheme 1.
has an effect on the activity, the hexanoic acid derivatives
being the most potent (IC50510–110 mM). However, up
to the present time the influence of the heterocyclic moiety
is not known since only compounds bearing either the
pyrimidine or pyridazine ring system have been investi-
gated. In order to gain additional insight into the structure–
activity relationships, further structural modifications of
compounds of type A became the object of our interest. In
the course of these studies, congeners characterised by
isosteric replacement of the diazine moiety by a benzene
nucleus were prepared. It should be noted that this
structural modification leads to simplification with respect
to synthesis and purification.
Due to the fact that aldose reductase inhibitors bearing a
carboxylic acid function are generally less active in vivo
than in vitro (Costantino et al., 1997), ester derivatives of
the target compounds are also of interest. It is known that
esterification of drugs bearing a carboxylic acid function
leads to increased bioavailability (Bundgaard, 1985). In
order to investigate the prodrug potential of these ester
derivatives we have studied whether they could be trans-
formed, under physiological conditions, into the corre-
sponding carboxylic acid. It should be emphasised that the
enzymatic hydrolysis has to take place primarily at the
ester group and not at the oxime function.
thin layer chromatography using Polygram SIL G/UV
254
(Macherey-Nagel) plastic-backed plates (0.25 mm layer
thickness) and visualised using an UV lamp. Column
chromatography was conducted on Merck silica gel 60
(230–400 mesh). Elemental analyses were performed by
Mag. J. Theiner, Institute of Physical Chemistry, Universi-
ty of Vienna, Austria.
Incubation and mixing (1400 revolutions per minute)
were performed using an Eppendorf Thermomixer com-
fort. HPLC was performed with a Merck–Hitachi system,
which consisted of an L-6200A pump, an AS-2000 auto-
sampler, an L-4250 UV detector operated at 250 nm and a
D-6000 interface. Chromatographic separation was
achieved on LiChrospher 100 RP-18 (12534.6 mm, 5
mm particle size). The mobile phase consisted of a mixture
of acetonitrile/water (80:20, v/v) at a flow-rate of 0.5
ml/min.
2.1. General procedure for the synthesis of ethyl v-
[[(diphenylmethylene)amino]oxy]alkanoates (2a–c,
method A)
Powdered potassium hydroxide (1.42 g, 25.3 mmol, 2
equiv.) was added to a solution of benzophenone oxime
(2.50 g, 12.7 mmol, 1 equiv.) in 20 ml of dry dimethyl
sulfoxide under a nitrogen atmosphere. After stirring for
30 min at room temperature, ethyl v-bromoalkanoate (25.3
mmol, 2 equiv.) was added and stirring was continued for
an additional 30 min. The mixture was then poured into 1
N HCl and extracted with dichloromethane (3325 ml).
The organic layer was washed with 2 N NaOH, water, and
brine, dried over anhydrous sodium sulphate and evapo-
rated to dryness in vacuo. The oily residue thus obtained
was used for further reactions without purification. For
analytical purposes a small amount of the compound was
purified by column chromatography (silical gel, dichloro-
methane). For yield, analytical and spectroscopic data, see
Tables 1 and 2.
2. Materials and methods
Melting points were determined with a Kofler hot-stage
microscope (Reichert) and are uncorrected. Solvents were
purified by distillation and stored over molecular sieves (3
˚
A). Light petroleum refers to the fraction with b.p. 40–
608C. Infrared spectra (KBr pellets) were recorded on a
Mattson Galaxy Series FTIR 3000 spectrophotometer.
Mass spectra were obtained on a Finnigan MAT SSQ 7000
spectrometer (EI, 70 eV). All NMR spectra were recorded
in CDCl3 solution in 5 mm tubes at 308C on a Varian
1
Gemini 200 spectrometer (199.98 MHz for H, 50.29 MHz
2.2. General procedure for the synthesis of v-
[[(diphenylmethylene)amino]oxy]carboxylic acids (3a–c,
method B)
for 13C) with the deuterium signal of the solvent as the
lock and TMS as internal standard. Chemical shifts are
expressed in parts per million (ppm). DEPT spectra were
run in a standard manner. Reactions were monitored by
A
solution
of
the
appropriate
ethyl
v-

D. Rakowitz et al. / European Journal of Pharmaceutical Sciences 15 (2002) 11–20
13
Table 1
Yields and analytical data for compounds 2a–c, 3a–c, 4, 6 and 7
Compound
R
n
Yield
(%)
M.p.
(8C)
Formula
(MW)
Elemental analysis
(calc. %, found %)
C
H
N
2a
2b
2c
3a
3b
3c
4
C2H5
C2H5
C2H5
H
3
4
5
3
4
5
5
5
5
61
93
94
89
75
81
52
47
46
Oil
C19H21NO3
(311.38)
C20H23NO3
(325.41)
C21H25NO3
(339.44)
C17H17NO3?0.1H2O
(285.13)
C18H19NO3?0.2H2O
(300.96)
C19H21NO3
(311.38)
C22H27NO3
(353.47)
C25H25NO3
(387.48)
73.29
73.47
73.82
73.74
74.31
74.52
71.61
71.57
71.84
72.01
73.29
73.01
74.76
74.95
77.49
77.22
77.78
77.56
6.80
6.72
7.12
6.96
7.42
7.30
6.08
6.15
6.50
6.35
6.80
7.06
7.70
7.54
6.50
6.33
6.78
6.96
4.50
4.49
4.30
4.28
4.13
4.22
4.91
5.04
4.65
4.78
4.50
4.54
3.96
3.94
3.61
3.57
3.49
3.45
Oil
Oil
78–82
77–81
88–90
Oil
H
H
i-C3H7
C6H5
CH2C6H5
6
44–47
57–59
7
C26H27NO3
(401.51)
[[(diphenylmethylene)amino]oxy]alkanoate
(7.90–13.2
2.4. General method for the synthesis of esters 6 and 7
mmol, 1 equiv.) in 40–60 ml of ethanol was treated with 2
N NaOH (8.7–14.5 mmol, 1.1 equiv.) and stirred overnight
at room temperature. The solvent was then evaporated, the
residue treated with a small amount of water and the pH
adjusted to 5 with 2 N HCl. The reaction mixture was
extracted with ethyl acetate, the organic layer washed with
water and brine, dried over anhydrous sodium sulphate and
evaporated to dryness in vacuo. The crystals thus obtained
were subsequently recrystallised from diisopropyl ether/
light petroleum. For yield, analytical and spectroscopic
data, see Tables 1 and 2.
Thionyl chloride (9.83 g, 83.0 mmol) and catalytic
amounts of N,N-dimethylformamide were added to a
solution of 3c (0.60 g, 1.93 mmol) in 30 ml of absolute
dichloromethane. The solution was stirred and refluxed for
30 min and stirring was continued at 408C for 5 h. After
removal of excess thionyl chloride, the remaining acid
chloride (5) was used for the subsequent reaction step
without further purification. Thus a solution of 5 (0.21 g,
0.7 mmol) in 10 ml of absolute dichloromethane was
added dropwise to a mixture of triethylamine (0.14 g, 1.4
mmol). The appropriate alcohol (1.4 mmol) and the
solution thus obtained was stirred overnight at room
temperature, then diluted with water and extracted exhaus-
tively with dichloromethane. The combined organic layers
were washed with water and brine, dried over anhydrous
sodium sulphate and evaporated to dryness in vacuo. The
resulting oil was subjected to column chromatography
(silica gel, dichloromethane/light petroleum, 9:1) followed
by recrystallisation from diisopropyl ether.
2.3. Synthesis of ester 4
A solution of 3c (0.20 g, 0.71 mmol) in 10 ml of
absolute isopropanol was treated with 0.5 ml of concen-
trated sulfuric acid and stirred for 2 h. After removal of
excess alcohol under reduced pressure, the residue was
poured on ice and the pH adjusted to 5 with saturated
sodium hydrogencarbonate solution. The mixture was
extracted with dichloromethane, washed with water and
brine, dried over anhydrous sodium sulphate and evapo-
rated to dryness in vacuo. The oily residue thus obtained
was purified by column chromatography (silica gel, di-
chloromethane). For yield, analytical and spectroscopic
data, see Tables 1 and 2.
2.5. Aldose reductase inhibitory assay
Sorbinil was a gift from Pfizer (Groton, CT, USA).
Partially purified aldose reductase (ALR2, alditol: NADP1
oxidoreductase, EC 1.1.1.21) was used for the inhibitory

14
D. Rakowitz et al. / European Journal of Pharmaceutical Sciences 15 (2002) 11–20
Table 2
Spectroscopic data for compounds 2a–c, 3a–c, 4, 6 and 7
Comp.
IR
MS
(m/z)
1H-NMR
13C-NMR
(cm21
)
2a
1733
(n C=O)
311.1
(M1)
7.53–7.32 (m, 10H, phenyl H),
4.24 (t, J 5 6.2 Hz, 2H, OCH2),
173.2 (C=O), 156.6 (C=N),
136.5, 133.3 (C-1, C-19),
4.13 (q, J 5 7.1 Hz, 2H, OCH2CH3),
2.40 (t, J 5 7.4 Hz, 2H CH2COO),
2.13–199 (m, 2H, CH2),
129.1, 128.1, 128.0, 127.8
(C-3/5, C-2/6, C-39/59, C-29/69),
128.7 (C-4, C-49), 73.3 (OCH2),
]
1.26 (t, J 5 7.1 Hz, 3H, CH3)
60.2 (OCH2CH3),
30.8, 24.6 (23CH2), 14.1 (CH3)
]
2b
2c
1733
(n C=O)
325.1
(M1)
7.50–7.27 (m, 10H, phenyl H),
4.22–4.06 (m, 4H, 23CH2),
2.34–2.27 (m, 2H, CH2COO),
1.73–1.64 (m, 4H, 23CH2),
1.23 (t, J 5 7.1 Hz, 3H, CH3)
173.5 (C=O), 156.5 (C=N),
136.6, 133.5 (C-1, C-19),
129.2, 128.2, 128.0, 127.8
(C-3/5, C-2/6, C-39/59, C-29/69),
129.1, 128.7 (C-4, C-49),
74.0 (OCH2), 60.1 (OCH2CH3),
]
34.0, 28.6, 21.5 (33CH2), 14.2 (CH3)
1733
(n C=O)
339.1
(M1)
7.50–7.28 (m, 10H, phenyl H),
4.20–4.04 (m, 4H, 23CH2),
2.27 (t, J 5 7.3 Hz, 2H, CH2COO),
1.78–1.57 (m, 4H, 23CH2),
1.45–1.29 (m, 2H, CH2),
173.4 (C=O), 156.2 (C=N),
136.6, 133.4 (C-1, C-19),
129.1, 128.0, 127.9, 127.7
(C-3/5, C-2/6, C-39/59, C-29/69),
129.0, 128.5 (C-4, C-49), 74.2 (OCH2),
1.22 (t, J 5 7.1 Hz, 3H, CH3)
60.0 (OCH2CH3), 34.1, 28.7,
25.4, 24.6 (43CH2), 14.1 (CH3)
]
3a
3b
3c
4
1704
(n C=O)
283.1
(M1)
7.50–7.28 (m, 10H, phenyl H),
4.22 (t, J 5 6.2 Hz, 2H, OCH2),
2.43 (t, J 5 7.3 Hz, 2H, CH2COO),
2.10–2.00 (m, 2H, CH2)
178.7 (C=O), 157.0 (C=N),
136.5, 133.4 (C-1, C-19),
129.3, 128.8 (C-4, C-49),
129.2, 128.2, 128.1, 127.9
(C-3/5, C-2/6, C-39/59, C-29/69),
73.1 (OCH2), 30.5, 24.4 (23CH2)
1704
(n C=O)
297.1
(M1)
7.50–7.28 (m, 10H, phenyl H),
4.19 (t, J 5 6.2 Hz, 2H, OCH2),
2.35 (t, J 5 7.1 Hz, 2H, CH2COO),
1.81–1.61 (m, 4H, 23CH2)
178.6 (C=O), 156.6 (C=N),
136.6, 133.4 (C-1, C-19),
129.2, 128.2, 128.0, 127.9
(C-3/5, C-2/6, C-39/59, C-29/69),
129.2, 128.7 (C-4, C-49), 74.0 (OCH2),
33.7, 28.5, 21.3 (33 CH2)
1700
(n C=O)
311.1
(M1)
7.50–7.31 (m, 10H, phenyl H),
4.18 (t, J 5 6.5 Hz, 2H, OCH2),
2.34 (t, J 5 7.4 Hz, 2H, CH2COO),
1.80–1.57 (m, 4H, 23CH2),
1.47–1.33 (m, 2H, CH2)
179.1 (C=O), 156.5 (C=N),
136.7, 133.5 (C-1, C-19),
129.3, 128.2, 128.0, 127.9
(C-3/5, C-2/6, C-39/59, C-29/69),
129.1, 128.7 (C-4, C-49), 74.3 (OCH2),
33.8, 28.8, 25.5, 24.4 (43CH2)
1729
(n C=O)
353.1
(M1)
7.50–7.27 (m, 10H, phenyl H),
173.1 (C=O), 156.3 (C=N),
136.7, 133.5 (C-1, C-19),
129.3, 128.1, 128.0, 127.8
5.09–4.90 [m, 1H, CH(CH3)2],
]
4.17 (t, J 5 6.6 Hz, 2H, OCH2),
2.25 (t, J 5 7.4 Hz, 2H, CH2COO),
1.79–1.57 (m, 4H, 23CH2),
1.45–1.33 (m, 2H, CH2),
(C-3/5, C-2/6, C-39/59, C-29/69),
129.1, 128.6 (C-4, C-49), 74.3 (OCH2),
67.3 [OCH(CH3)2], 34.6, 28.8,
]
1.21 [d, J 5 6.2 Hz, 6H, CH(CH3)2]
25.5, 24.8 (43CH2), 21.8 [OCH(CH3)2]
]
]
6
7
1725
(n C=O)
387.1
(M1)
7.54–7.07 (m, 15H, phenyl H),
4.25 (t, J 5 6.5 Hz, 2H, OCH2),
2.58 (t, J 5 7.3 Hz, 2H, CH2COO),
1.85–1.74 (m, 4H, 23CH2),
1.59–1.47 (m, 2H, CH2)
172.0 (C=O), 156.4 (C=N), 150.7 (C-19),
136.7, 133.5 (C-1, C-19), 129.3, 129.2,
128.2, 128.0, 127.8, 121.5
(C-3/5, C-2/6, C-39/59, C-29/69, C-30/50,
C-20/60), 129.1, 128.7 (C-4, C-49),
125.7 (C-49), 74.3 (OCH2), 34.3, 28.8,
25.5, 24.7 (43CH2)
1722
(n C=O)
401.1
(M1)
7.53–7.36 (m, 15H, phenyl H),
173.4 (C=O), 156.4 (C=N),
5.13 (s, 2H, CH2C6H5),
136.7, 133.5 (C-1, C-19), 136.1 (C-10),
129.2, 128.5, 128.2, 128.1, 128.0, 127.8
(C-3/5, C-2/6, C-39/59, C-29/69, C-30/50,
C-20/60, C-40), 129.1, 128.6 (C-4, C-49),
]
4.19 (t, J 5 6.5 Hz, 2H, OCH2),
2.37 (t, J 5 7.4 Hz, 2H, CH2COO),
1.78–1.63 (m, 4H, 23CH2),
1.48–1.30 (m, 2H, CH2)
74.3 (OCH2), 66.0 (CH2C6H5), 34.2, 28.8,
]
25.5, 24.7 (43CH2)

D. Rakowitz et al. / European Journal of Pharmaceutical Sciences 15 (2002) 11–20
15
assays. Calf lenses were obtained locally from freshly
2.6.2. Enzymatic hydrolysis
2.6.2.1. Plasma
Fifteen microliters of the stock solution of the appro-
priate compound (10% in DMSO) was mixed with 500 ml
of phosphate buffer pH 7.2 containing 10% Tween 20,
and 500 ml of porcine plasma (Sigma) at 378C and 1400
rpm. At selected time intervals, 100 ml aliquots were
treated as described above.
slaughtered animals. The capsule was incised, and the
frozen lenses were suspended in 10 mM sodium phosphate
buffer pH 7.00 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 22 000 g at 48C for 40 min, and ALR2
present in the supernatant was partially purified by ion-
exchange chromatography on DE52 as previously de-
scribed (Costantino et al., 1996). Enzyme activity was
measured by monitoring the change in absorbance at 340
nm which accompanies the oxidation of NADPH catalysed
by ALR2. The assays were performed as previously
described (Constantino et al., 1996) using 4.7 mM D,L-
glyceraldehyde as substrate in 0.25 M sodium phosphate
buffer, pH 6.80, containing 0.38 M ammonium sulfate and
0.11 mM NADPH (378C). The sensitivity of the enzyme to
inhibition by the compounds under study was tested under
the above assay conditions by including the inhibitor
dissolved in DMSO at the desired concentrations in the
reaction mixture. DMSO in the assay mixture was kept at a
constant concentration of 1%. A reference blank con-
taining all the above reagents except the substrate was used
to correct for the nonenzymatic oxidation of NADPH. IC50
values (the concentration of the inhibitor required to
produce 50% inhibition of the enzyme catalysed 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 inhibition between 20 and 80%, with two repli-
cates at each concentration. The 95% confidence limits
(95% CL) were calculated from T values for n 2 2, where
n is the total number of determinations (Tallarida and
Murray, 1987). Sorbinil, used as a standard, possesses an
IC50 value of 1.8 (60.3) mM.
2.6.2.2. Cholinesterase
Twelve microliters of the stock solution of the appro-
priate compound (10% in DMSO) was mixed with 400 ml
of phosphate buffer (pH 7.2) containing 10% Tween 20,
and 5 ml human cholinesterase (E.C. 3.1.1.8; Boehringer
Mannheim, 50 U/ml) at 378C and 1400 rpm. At selected
time intervals, 100 ml aliquots were treated as described
above.
2.6.2.3. Esterase
Fifteen microliters of the stock solution of the appro-
priate compound (10% in DMSO) was mixed with 1500 ml
of phosphate buffer pH 7.2 containing 10% Tween 20,
and 3 ml of porcine esterase suspension (E.C. 3.1.1.1;
Boehringer Mannheim, 130 U/mg, 30 mg/3 ml) at 378C
and 1400 rpm. At selected time intervals, 100 ml aliquots
were treated as described above.
2.6.2.4. Lipase
Fifteen microliters of the stock solution of the appro-
priate compound (10% in DMSO) was mixed with 500 ml
of PIPES buffer (pH 6.5, 0.1 M, containing 25 mM
CaCl2), 500 ml of sodium cholate solution (25 mM), and
30 ml of porcine lipase (E.C. 3.1.1.3; Boehringer Mann-
heim, 4500 U/ml) at 378C and 1400 rpm. At selected time
intervals, 100 ml aliquots were treated as described above.
2.6. In vitro incubations
2.6.1. Chemical hydrolysis
3. Results and discussion
Chemical hydrolysis was studied at 378C in 50 mM
sodium citrate/HCl (pH 1.2, 2.0, 4.0), phosphate (pH 6.0,
7.4, 8.0), and borate (pH 10.0) buffers adjusted to an ionic
strength of m 5 0.5 by addition of a calculated amount of
sodium chloride.
Twenty-seven microliters of the stock solution of the
appropriate compound (10% in DMSO) was mixed with
900 ml of the appropriate buffer solution and 900 ml of a
Tween 20 solution (10% in the appropriate buffer),
followed by brief sonication. The mixture was incubated at
378C using an Eppendorf Thermomixer comfort. At select-
ed time intervals, 100 ml aliquots were quenched by
addition to 50 ml ice-cold 0.5 M HClO4 followed by
centrifugation at 5500 rpm for 10 min, and aliquots were
assayed by HPLC.
The target compounds (type B) are characterised by
isosteric replacement of the diazine core by benzene. In
order to investigate preliminary structure–activity relation-
ships, the length of the spacer group was varied, and
compounds with butyric acid to hexanoic acid (n53–5)
were studied.
The compounds were prepared as shown in Scheme 2.
Direct reaction of benzophenone oxime (1) with 6-bromo-
caproic acid following a procedure given in the literature
(Barth et al., 1996) was unsuccessful (Scheme 2), probably
due to deprotonation of the carboxylic acid. Therefore, we
envisaged an alternative synthetic strategy. Reaction of 1
with ethyl v-bromoalkanoates afforded the ester deriva-
tives 2a–c (Scheme 2, method A), which could be

16
D. Rakowitz et al. / European Journal of Pharmaceutical Sciences 15 (2002) 11–20
Scheme 2.
results are in accordance with our previous observations
for the pyridazinyl substituted analogues A (Rakowitz et
al., 2000).
smoothly transformed into the corresponding carboxylic
acids 3a–c by alkaline hydrolysis (Scheme 2, method B).
Compounds 3a–c have been described in the literature
(Gilon et al., 1967). In the course of the synthesis of
v-aminooxycarboxylic acids, compounds 3a–c were pre-
pared as intermediates by reaction of the sodium salt of
benzophenone oxime and the corresponding lactone. How-
ever, no spectroscopic or biological data were given.
The in vitro assay of compounds 3a–c clearly showed
that, within this series, the hexanoic acid derivative 3c
exhibits the highest inhibitory activity (Table 3). These
The most active compound 3c was chosen as a model
substance for our investigations of the prodrug potential of
ester derivatives. Preparation of the esters could be
achieved by direct reaction of the carboxylic acid 3c with
the appropriate alcohol or via the corresponding acid
chloride (5) (Scheme 3). A series of derivatives of
compounds 3a–c was prepared by our group. Full ana-
lytical data are described in a diploma thesis (Krassnigg,
2000). A selection of the new compounds was subjected to
in vitro degradation under various reaction conditions. In
order to determine the effect of the alcoholic part of the
ester moiety on hydrolysis, the following compounds were
selected: 2c derived from a linear alcohol, 4 derived from a
branched alcohol, 6 derived from a phenol, and 7 derived
from benzyl alcohol.
Table 3
Biological data for compounds 3a–c
Compound
Percent inhibition
(concentration)
IC50
(95 CL, mM)
3a
3b
3c
32% (64.0 mM)
n.d.
64 (75–54)
33 (40–27)
In order to determine the chemical stability of the novel
ester derivatives, pH–rate profiles were determined with
n.d., not determined
Scheme 3.

D. Rakowitz et al. / European Journal of Pharmaceutical Sciences 15 (2002) 11–20
17
buffers ranging from pH 1.2 to 10.0. Due to the very low
solubility of the compounds in the aqueous test medium,
the investigations were performed in the presence of
Tween 20 as solubilizer (,5%). It should be noted that it
cannot be excluded that Tween 20 leads to partial
enzyme inhibition. Although this could be a contributory
factor in rate of hydrolysis studies, this effect was not
considered because we are presently only interested in
structure–hydrolysis relationships.
The degree of degradation was observed over a period
of 14 days at 378C; the final values are presented in Fig. 1.
Both 4 (isopropyl ester) and 7 (benzyl ester) are very
stable compounds, in that no significant amounts of
hydrolytic products were detected in the strong acidic or
basic media (pH 1.2, less than 20%; pH 10.0, less than
10%). Whereas the ethyl ester 2c was transformed into the
corresponding carboxylic acid by incubation in buffer
solutions below pH 4 (pH 1.2, ca. 42%; pH 2.0, ca. 19%),
almost no hydrolysis could be observed in the neutral
range and, surprisingly, also in the basic range (less than
5%). In contrast, the phenyl ester (6) showed a high
hydrolysis rate at pH 10.0 (more than 50%); however, it
was very stable in the neutral and even in the strong acidic
range (pH 1.2, less than 18%) (Fig. 1).
esters studied rather than concentrations present in physio-
logical activities [e.g., cholinesterase: physiological, 3–8
U/ml (Merck, 1974); tested at 0.3 U/ml].
Incubation of derivatives 2c, 4, and 7 in the presence of
plasma (50%) showed that the ethyl, isopropyl, and benzyl
esters were stable under these conditions (generally less
than 5% hydrolysis in 24 h). However, the phenyl ester 6
was found to undergo hydrolysis to yield 3c (about 40%
after 24 h). Since cholinesterases represent some of the
most important enzymes in plasma, it was of interest to
verify whether the cholinesterases (E.C. 3.1.1.8; also called
pseudocholinesterase) are responsible for the cleavage of
the phenyl ester function. Almost no degradation (less than
5%) was observed in the presence of human cholinesterase
for 2c, 4, and 7. By contrast, the hydrolysis rate of the
phenyl ester (6) under the same conditions was about 40%
(see Fig. 2). Remarkably, these results indicate that, in this
series examining the cleavage of ester functions by
cholinesterases, no structural similarity to choline esters is
necessary.
In view of the fact that hydrolysis of ester functions
takes place both in plasma and by means of cholinester-
ases, the use of unspecific liver esterase and lipase was
tested. Whereas only slow degradation was observed for 4
and 7 (between 20 and 40%) in the presence of porcine
esterase, more than 85% of the carboxylic acid 3c was
found to be released using 2c and 6 as substrates. The
results are summarised in Fig. 3.
Since our compounds exhibited high chemical stability
at physiological pH, we became interested in enzyme-
mediated hydrolysis and examined whether cleavage takes
place selectively at the ester group as the basic requirement
for the derivatives to be prodrugs. Moreover, the influence
of the alcoholic moiety on enzymatic degradation was
determined.
Upon incubation with lipase (in the presence of sodium
cholate), almost no hydrolysis (less than 5%) was observed
with the isopropyl ester (4), whereas the ethyl ester
derivative 2c again afforded a high hydrolysis rate (ca.
70% after a maximum incubation period of 24 h). Under
The enzymes were used in concentrations which pro-
vided comparable data over a period of 24 h for all the
Fig. 1. pH–rate profiles for the formation of 6-[[(diphenylmethylene)amino]oxy]hexanoic acid (3c) from 2c, 4, 6, or 7 in aqueous solution at 378C after 14
days.

18
D. Rakowitz et al. / European Journal of Pharmaceutical Sciences 15 (2002) 11–20
Fig. 2. Plots showing the time course for the formation of 6-[[(diphenylmethylene)amino]oxy]hexanoic acid (3c) from 2c, 4, 6, or 7 by treatment with
human cholinesterase at 378C for 24 h.
these conditions, the phenyl ester 6 was found to be more
stable than in the presence of esterase, in that only 45% of
the ester function was hydrolysed after 24 h. The formal
introduction of a methylene spacer between the oxygen
atom and the aromatic moiety results in an increase of
enzymatic cleavage, the benzyl ester (7) representing the
Fig. 3. Plots showing the time course for the formation of 6-[[(diphenylmethylene)amino]oxy]hexanoic acid (3c) from 2c, 4, 6, or 7 by treatment with
porcine esterase at 378C for 24 h.

D. Rakowitz et al. / European Journal of Pharmaceutical Sciences 15 (2002) 11–20
19
Fig. 4. Plots showing the time course for the formation of 6-[[(diphenylmethylene)amino]oxy]hexanoic acid (3c) from 2c, 4, 6, or 7 by treatment with
porcine lipase at 378C for 24 h.
best substrate for lipase (complete hydrolysis after 12 h).
The results are summarised in Fig. 4.
drugs for the oral delivery of the aldose reductase inhibitor
3c.
References
4. Conclusions
Anonymous, 1984. Tolrestat. Drugs Future 9, 670–672.
Anonymous, 1987a. Epalrestat. Drugs Future 12, 336–340.
Anonymous, 1987b. ICI-128436. Drugs Future 12, 347–348.
Anonymous, 1989. FR-74366. Drugs Future 14, 26–28.
Anonymous, 1995. Zopolrestat. Drugs Future 20, 33–36.
Ashizawa, N., Aotsuka, T., 1998. Benzothiazole aldose reductase in-
hibitors. Drugs Future 23, 521–529.
In conclusion, esters 2c, 6, and 7 were shown to be
useful biolabile prodrugs for the carboxylic acid 3c since
hydrolysis only takes place at the ester function. As
demonstrated above, all esters studied (2c, 4, 6, and 7)
exhibit a high degree of stability in aqueous solution
(almost no degradation was observed in the neutral pH
range after the maximum incubation period of 14 days).
Moreover, we found that the alcohol component of the
ester function influences the chemical stability and en-
zymatic reactivity. Of the series tested, the isopropyl ester
4 was found to be the most stable compound; even
incubation in buffer solution (pH 1.2 to 10) or treatment
with various enzymes did not lead to the corresponding
carboxylic acid (3c). This observation may be attributed to
the steric hindrance of the ester function. In contrast, the
ethyl ester (2c) turned out to be a compound which could
be cleaved in the presence of esterase and lipase. Interest-
ingly, the benzyl ester was only hydrolysed by porcine
lipase. In our opinion, the most interesting compound in
the series is the phenyl ester (6), which can be cleaved in
the presence of plasma, cholinesterase, and unspecific
esterase. Thus, in vivo studies will be necessary in order to
unequivocally ascertain if 2c, 6, and 7 are suitable pro-
Barth, B., Dierich, M., Heinisch, G., Jenny, V., Matuszczak, B., Mereiter,
¨
K., Planer, R., Schopf, I., Stoiber, H., Traugott, T., v. Aufschnaiter, P.,
1996. Pyridazino[3,4-b][1,5]benzoxazepin-5(6H)-ones: synthesis and
biological evaluation. Antiviral Chem. Chemother. 7, 300–312.
Brittain, D.R., Wood, R. (ICI), 1979. Enzyme inhibitory phthalazin-4-
ylacetic acid derivatives, pharmaceutical compositions thereof, and
process for their manufacture. Eur. Pat. Appl. EP 2, 895, Chem.
Abstr., 1980, 92, 76533w.
Bundgaard, H., 1985. Design of prodrugs: bioreversible derivatives for
various functional groups and chemical entities. In: Bundgaard, H.
(Ed.), Design of Prodrugs. Elsevier, Amsterdam, pp. 1–92.
Costantino, L., Rastelli, G., Cignarella, G., Vianello, P., Barlocco, D.,
1997. New aldose reductase inhibitors as potential agents for the
prevention of long-term diabetic complications. Exp. Opin. Ther.
Patents 7, 843–858.
Costantino, L., Rastelli, G.,Vescovini, K., Cignarella, G.,Vianello, P., Del
Corso, A., Cappiello, M., Mura, U., Barlocco, D., 1996. Synthesis,
activity and molecular modeling of
a new series of tricyclic
pyridazinones as selective aldose reductase inhibitors. J. Med. Chem.
39, 4396–4405.
Gilon, C., Knobler, Y., Sheradsky, T., 1967. Synthesis of v-aminooxy
acids by oxygen-alkyl fission of lactones. Tetrahedron 23, 4441–4447.

20
D. Rakowitz et al. / European Journal of Pharmaceutical Sciences 15 (2002) 11–20
Itoh, Y., Ao, S., Notsu, Y., Hashimoto, M., 1992. Novel aldose reductase
inhibitors: synthesis and structure–activity studies of (3-benzyl-2,4-
dioxo-1,2,3,4-tetrahydroquinazolin-1-yl)acetic acids. Drugs Future 17,
31–37.
production and pharmaceutical compositions comprising them. Eur.
Pat. Appl. EP 218999 A2; Chem. Abstr., 1987, 108, 75411s.
Porte, Jr. D., Schwartz, M.W., 1996. Diabetes complications: why is
glucose potentially toxic? Science 272, 699–700.
Kador, P.F., 1988. The role of aldose reductase in the development of
diabetic complications. Med. Res. Rev. 8, 325–352.
Kador, P.F., Kinoshita, J.H., Sharpless, N.E., 1985. Aldose reductase
inhibitors: a potential new class of agents for the pharmacological
control of certain diabetic complications. J. Med. Chem. 28, 841–849.
Krassnigg, A., 2000. Diploma thesis, Innsbruck.
Rakowitz, D., Heinisch, G., Lukavsky, P., Kiendler, S., Trenkwalder, C.,
Barlocco, D., Rastelli, G., Costantino, L., 2000. A series of diarylsub-
stituted oximes as potential substrate for new aldose reductase
inhibitors. J. Heterocycl. Chem. 37, 1089–1096.
Sarges, R., Oates, P.J., 1993. Aldose reductase inhibitors: recent develop-
ments. Prog. Drug Res. 40, 99–161.
Lipinski, C.A., Hutson, N.J., 1984. Aldose reductase inhibitors as a new
approach to the treatment of diabetic complications. Annu. Rep. Med.
Chem. 19, 169–177.
Merck, E., 1974. Cholinesterase im Serum. In: Klinisches Labor, 12th
Edition. E. Merck, Darmstadt, pp. 178–184.
Sestanj, K., Bellini, F., Fung, S., Abraham, N., Treasurywala, A.,
Humber, L., Simard-Duquesne, N., Dvornik, D., 1984. N-[[5-
(Trifluoromethyl)-6-methoxy-1-naphthalenyl]thioxomethyl]-N-
methylglycine (tolrestat), a potent, orally active aldose reductase
inhibitor. J. Med. Chem. 27, 255–256.
Mylari, B.L., Larson, E.R., Beyer, T.A., Zembrowski, W.J., Aldinger,
C.E., Dee, M.F., Siegel, T.W., Singleton, D.H., 1991. Novel, potent
aldose reductase inhibitors: 3,4-dihydro-4-oxo-3-[[5-(trifluoromethyl)-
2-benzothiazolyl]methyl]-1-phthalazineacetic acid (zopolrestat) and
congeners. J. Med. Chem. 34, 108–122.
Tallarida, R.J., Murray, R.B., 1987. Manual of Pharmacological Calcula-
tions with Computer Programs, 2nd Edition. Springer, New York.
Terashima, H., Hama, K., Yamamoto, R., Tsuboshima, M., Kikkawa, R.,
Hatanaka, I., Shigeta, Y., 1984. Effects of a new aldose reductase
inhibitor on various tissues in vitro. J. Pharmacol. Exp. Ther. 229,
226–230.
Oku, T., Ito, Y., Namiki, T., Sawada, K., Kasahara, C., Baba, Y. (Fujisawa
Pharmaceutical), 1987. New quinazoline derivatives, process for their