Carboxymethylated pyridoindole antioxidants as aldose reductase inhibitors: Synthesis, activity, partitioning, and molecular modeling

Article abstract of DOI:10.1016/j.bmc.2008.03.039

Starting from the efficient hexahydropyridoindole antioxidant stobadine, a series of carboxymethylated tetrahydro- and hexahydropyridoindole derivatives was synthesized and tested for the inhibition of aldose reductase, an enzyme involved in the etiology of diabetic complications. In vitro inhibiton of rat lens aldose reductase was determined by a conventional method. Kinetic analysis of (2-benzyl-2,3,4,5-tetrahydro-1H-pyrido[4,3-b]indole-8-yl)-acetic acid (5b) and (2-phenethyl-2,3,4,5-tetrahydro-1H-pyrido[4,3-b]indole-8-yl)-acetic acid (5c), the most potent compounds in this series with activities in micromolar range, showed uncompetitive inhibition. In addition to the importance of the acidic function, the inhibition efficacy was highly influenced by the steric conformation of the lipophilic aromatic backbone when comparing tetrahydro- and hexahydropyridoindole congeners. Selectivity with respect to the closely related aldehyde reductase was determined by measuring the corresponding inhibitory activities. Antioxidant action of the novel compounds was documented in a DPPH test and in a liposomal membrane model, oxidatively stressed by peroxyl radicals. The presence of a basicity center at the tertiary nitrogen, in addition to the acidic carboxylic function, predisposes these compounds to form double charged zwitterionic species, a characteristic which may remarkably affect their pH-lipophilicity profile. For compounds 5b and 5c, a maximal distribution ratio in a system comprised of 1-octanol/phosphate buffer was recorded near the neutral physiological pH, the region where the isoelectric point lies. Molecular docking simulations into the ALR2 active site performed for the zwitterionic species provided an explanation for the observed structure-activity relationships and the calculated parameters were in agreement with characteristic differences in the stereoelectronic profiles of the tetrahydro- versus hexahydropyridoindoles. 'Drug-likeness' of the novel aldose reductase inhibitors was assessed by applying the criteria of Lipinski's 'rule of five'.

Full text of DOI:10.1016/j.bmc.2008.03.039

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 4908–4920
Carboxymethylated pyridoindole antioxidants as aldose
reductase inhibitors: Synthesis, activity, partitioning,
and molecular modeling
Milan Stefek,a,* Vladimir Snirc,a Paul-Omer Djoubissie,a Magdalena Majekova,a
Vassilis Demopoulos,b Lucia Rackova,a Zelmira Bezakova,c Cimen Karasu,d
Vincenzo Carbonee and Ossama El-Kabbanie
aInstitute of Experimental Pharmacology, Slovak Academy of Sciences, Dubravska cesta 9, 841 04 Bratislava, Slovakia
bAristotle University of Thessaloniki, School of Pharmacy, Department of Pharmaceutical Chemistry, Thessaloniki 54124, Greece
cDepartment of Pharmaceutical Chemistry, Faculty of Pharmacy of Comenius University, Odboja´rov 10,
832 32 Bratislava, Slovak Republic
dFaculty of Medicine, Gazi University, Department of Medical Pharmacology, 06510 Besevler, Ankara, Turkey
eDepartment of Medicinal Chemistry, Victorian College of Pharmacy, Monash University, (Parkville Campus),
381 Royal Parade, Parkville, Vic. 3052, Australia
Received 28 January 2008; revised 5 March 2008; accepted 14 March 2008
Available online 17 March 2008
Abstract—Starting from the efficient hexahydropyridoindole antioxidant stobadine, a series of carboxymethylated tetrahydro- and
hexahydropyridoindole derivatives was synthesized and tested for the inhibition of aldose reductase, an enzyme involved in the eti-
ology of diabetic complications. In vitro inhibiton of rat lens aldose reductase was determined by a conventional method. Kinetic
analysis of (2-benzyl-2,3,4,5-tetrahydro-1H-pyrido[4,3-b]indole-8-yl)-acetic acid (5b) and (2-phenethyl-2,3,4,5-tetrahydro-1H-pyr-
ido[4,3-b]indole-8-yl)-acetic acid (5c), the most potent compounds in this series with activities in micromolar range, showed uncom-
petitive inhibition. In addition to the importance of the acidic function, the inhibition efficacy was highly influenced by the steric
conformation of the lipophilic aromatic backbone when comparing tetrahydro- and hexahydropyridoindole congeners. Selectivity
with respect to the closely related aldehyde reductase was determined by measuring the corresponding inhibitory activities. Antiox-
idant action of the novel compounds was documented in a DPPH test and in a liposomal membrane model, oxidatively stressed by
peroxyl radicals. The presence of a basicity center at the tertiary nitrogen, in addition to the acidic carboxylic function, predisposes
these compounds to form double charged zwitterionic species, a characteristic which may remarkably affect their pH-lipophilicity
profile. For compounds 5b and 5c, a maximal distribution ratio in a system comprised of 1-octanol/phosphate buffer was recorded
near the neutral physiological pH, the region where the isoelectric point lies. Molecular docking simulations into the ALR2 active
site performed for the zwitterionic species provided an explanation for the observed structure–activity relationships and the calcu-
lated parameters were in agreement with characteristic differences in the stereoelectronic profiles of the tetrahydro- versus hexahyd-
ropyridoindoles. ‘Drug-likeness’ of the novel aldose reductase inhibitors was assessed by applying the criteria of Lipinski’s ‘rule of
five’.
ꢀ 2008 Elsevier Ltd. All rights reserved.
1. Introduction
tions, substantial evidence suggests a key role for the
polyol pathway.1–4 In tissues that do not require insulin
for glucose uptake, the high systemic level of glucose
developing during diabetes readily translates into high
tissue levels of glucose. Aldose reductase (ALR2,
E.C.1.1.1.21), the first enzyme of the polyol pathway, re-
duces some of this excess glucose to the organic osmo-
lyte sorbitol in an NADPH-dependent manner. Due to
its poor membrane penetration and slow metabolism
by sorbitol dehydrogenase, sorbitol accumulates
Although multiple biochemical pathways are likely to be
responsible for the pathogenesis of diabetic complica-
Abbreviation: Ph, Phenyl.
Keywords: Aldose reductase inhibition; Pyridoindole; Antioxidant;
Zwitterion; Diabetic complications.
*
Corresponding author. Tel.: +421 2 59410667; fax: +421 2 554775928;
e-mail: exfastfk@savba.sk
0968-0896/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.03.039

M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
4909
processes, compounds such as pyridazines,22 benzopyr-
anes,23 and pyridopyrimidines24 were synthesized and
were shown to display antioxidant as well as aldose
reductase inhibitory activities under in vitro conditions.
Based on the premise that a bifunctional compound with
joint antioxidant/aldose reductase inhibitory (AO/ARI)
activities could be multifactorially beneficial, in the
current study we synthesized and tested new carboxyme-
thylated pyridoindoles, structurally based on the antiox-
idant drug stobadine15 (Chart 1). Under conditions of an
experimental glycation model in vitro, the parent drug
stobadine was found to protect bovine serum albumin
against glyco-oxidative damage.25 Using a model of
streptozotocin-diabetic rats in vivo, stobadine was found
to attenuate pathological changes in diabetic myocar-
dium,26,27 kidneys,27,28 eye lens29 and retina,30 and vas
deferens,31 to decrease matrix collagen cross-linking,28
and to reduce plasma cholesterol26 and triglyceride26,32
levels in diabetic animals. Stobadine treatment normal-
ized calcium homeostasis in diabetic rat heart and liver32
and produced a beneficial effect on leukocyte function.33
The newly synthesized (Scheme 1) and investigated car-
boxymethylated pyridoindoles were two hexahydropy-
ridoindoles (7a, b) and three tetrahydropyridoindoles
(5a–c), with the R2-substituents being -methyl (5a, 7a),
-benzyl (5b, 7b), or -phenethyl (5c).
CH3
N
H3C
N
H
Stobadine
Chart 1.
intracellularly, resulting in disruption and eventually
death of the cells. Depletion of NADPH due to aldose
reductase activity reduces intracellular glutathione
(GSH), an endogenous antioxidant, thereby inducing
oxidative stress.5 In this way, the polyol pathway in cells
is believed to contribute to the etiology of long-term
diabetic complications such as cataract, retinopathy,
nephropathy, neuropathy, micro-, and macroangio-
pathy. Under physiological conditions, ALR2 serves as
an extrahepatic detoxifying enzyme against endogenous
and xenobiotic aldehydes.1
Implicated in diabetes complications, aldose reductase is
a potential target of drug action,6 and the search for al-
dose reductase inhibitors (ARIs) offers an attractive
strategy for preventing the onset or the progression of
the complications.7–10 A large variety of structurally di-
verse compounds have been identified as potent in vitro
aldose reductase inhibitors. However, in humans, with
few exceptions, these compounds have produced little
evidence of clinical benefit, and some even produced del-
eterious side effects, prompting efforts to search more
potent and specific inhibitors. Owing to ALR-2 pharma-
cophore requirements11,12 for an acidic proton, most
ARIs contain an acetic acid moiety or N-unsubstituted
cyclic imides. Carboxylic acids are ionized at physiolog-
ical pH resulting in their poor biological availability.
Substitution of the acetic acid chain with the hydantoin
or succinimide moiety partially improved the yield in the
target tissues.7,8 The bioisosteric principle may offer an-
other way how to overcome unfavorable partitioning of
acidic ARIs.13,14
2.2. Chemistry
We describe (Scheme 1) the synthesis of putative
pharmacologically active tetrahydropyridoindoles, 2-
methyl- (5a), 2-benzyl- (5b) and 2-phenethyl- (5c)
2,3,4,5-tetrahydro-1H-pyrido[4,3-b]indole-8-yl acetates,
and hexahydropyridoindoles, ( )-2-methyl- (7a) and
( )-2-benzyl- (7b) (4a,9b)-cis-1,2,3,4,4a,9b-hexahydro-
1H-pyrido[4,3-b]indole-8-yl acetates. The synthesis of
the target compounds started from the previously
reported16,17 4-cyanomethylphenylhydrazine hydro-
chloride (1) or 4-carboxymethylphenylhydrazine hydro-
chloride (2) and commercially available N-alkylsubsti-
tuted-4-piperidones 3a–c. The initial step was carried
out under Fischer’s indole synthesis conditions18
(Scheme 1).
The implications of oxidative stress and the polyol path-
way in the etiology of diabetic complications are widely
accepted. Since the inhibition of both the processes is
desirable, a bifunctional compound with joint antioxi-
dant/aldose reductase inhibitory (AO/ARI) activities
could be multifactorially beneficial. In this study, we
present carboxymethylated hexahydropyridoindoles
and tetrahydropyridoindoles, structurally based on the
hexahydropyridoindole antioxidant drug stobadine15
(Chart 1) as aldose reductase inhibitors endowed with
antioxidant activity.
Specifically, acetonitrile 4a was synthesized by the treat-
ment of 1 with 3a in a refluxing 10% ethanolic HCl solu-
tion to give the product in a fair yield. A similar
procedure was used to synthesize the methyl ester 4c
started from 2 and 3c. However, in this case, best results
were obtained by utilizing a 10% methanolic HCl solu-
tion. The condensation of 1 with 3b in the above-men-
tioned acidified alcohols did not result in cyclization to
4b. Alternatively, the corresponding hydrazone was first
formed and then cyclized to the desired product 4b with
the action of acetic acid containing 6% HCl under con-
trol heating conditions. This two-step procedure was
found to result consistently in high yields of 4b.
2. Results and discussion
2.1. Design
The reduction of 4a and 4b with NaBH3CN in trifluoro-
acetic acid (TFA) gave hexahydropyridoindole com-
pounds 6a and 6b. The reaction was found to be
stereoselective, yielding the cis-configuration of the
hydrogens in the positions 4a and 9b of the correspond-
Oxidative stress and the involvement of the polyol path-
way in the etiology of diabetic complications have been
widely reported.1,5,21 With the aim of inhibiting these

4910
M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
CN
COOH
NHNH2.HCl
NHNH2.HCl
(1)
(2)
HCl / CH3OH
R2
N
R2
N
HCl / EtOH
(HCl / AcOH)
R2 = CH3 (3a), CH2C6H5 (3b)
R2 = (CH2)2C6H5 (3c)
O
O
R2
N
R2
N
R2
N
CN
COOK
COOCH3
KOH
KOH
i-PrOH
i-PrOH
N
N
N
H
(4c)
H
H
(4a), (4b)
(5a) R2 = CH3, (5b) R2 = CH2C6H5
(5c) R2 = (CH2)2C6H5
NaBH3CN / TFA
R2
N
R2
N
COOK
CN
H
H
KOH
i-PrOH
N
H
N
H
H
H
6a) R2 = CH3
(
(
7a) R2 = CH3
7b) R2 = CH2C6H5
(
(
6b) R2 = CH2C6H5
Scheme 1.
ing hexahydropyridoindoles in more than 98% rate,19
this finding being in agreement with relevant literary
data.20 The products were isolated as racemic mixtures
and no attempt was made to separate the enantiomers.
Finally, the hydrolysis of the intermediate products
4a–4c and 6a, 6b in refluxing KOH/i-PrOH yielded the
target products 5a–5c and 7a, b, which were isolated
as potassium salts. Compound 7b was characterized in
the form of its sodium salt.
form, dehydrostobadine, were active as inhibitors of
ALR2. However, the introduction of the carboxymethyl
group into the molecule of stobadine resulted in hexa-
hydropyridoindole compounds 7a, 7b with a mild inhibi-
tion of ALR2. Activities in micromolar range were
recorded for the unsaturated tetrahydropyridoindole
congeners 5a–5c.
In the next step, we analyzed the enzyme kinetics for
compounds 5b and 5c. Uncompetitive inhibition was ob-
served in relation to the substrate D,L-glyceraldehyde
and the cofactor NADPH (for 5b see Figs. 1 and 2,
respectively). Table 2 shows values of the corresponding
kinetic parameters. The uncompetitive inhibition of
ALR2 indicates that the glucose substrate would not
compete with the inhibitor for the enzyme. Thus, under
hyperglycemia conditions in diabetics, excessive glucose
would not presumably decrease the inhibitory efficiency
of the drug. The experimentally obtained Km values for
2.3. Enzyme inhibition
The compounds 5a–5c and 7a, 7b were evaluated for
their ability to inhibit the in vitro reduction of D,L-glyc-
eraldehyde by partially purified ALR2 from rat lens,
using zopolrestat as a reference (Table 1). In the primary
screening studies, the three following questions were ad-
dressed: (i) the effect of introducing the carboxymethyl
group into the molecule of stobadine, (ii) the effect of
switching between hexahydropyridoindole and tetrahyd-
ropyridoindole type structure, and (iii) the effect of lipo-
philicity of the substituent –R2. As shown, neither
stobadine nor its unsaturated tetrahydropyridoindole
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Table 1. Inhibition of rat lens ALR2
Compound
IC50 (lM)
5a
5b
54.0 12.1
18.2 1.2
16.7 1.2
380.7
5c
7a
-10
-5
0
5
10
15
20
Glyceraldehyde (mM) 1/c
7b
266.0 4.1
>1000
>1000
Stobadine
Dehydrostobadine
Zopolrestat
Figure 1. Inhibitory effect of compound 5b on rat lens aldose
reductase. Typical double reciprocal plot of the initial enzyme velocity
versus the concentration of substrate (D,L-glyceraldehyde) in the
presence or absence of 5b: (j) no inhibitor; (d) 20 lM 5b (uncom-
petitive type of inhibition).
0.005 0.001
Experimental results are mean values from two or mean values SD
from at least three experiments.

M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
4911
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
2.4. Antioxidant activity
According to the published data, stobadine has been
postulated as a chain-breaking antioxidant character-
ized by its ability to scavenge chain-propagating peroxyl
radicals.39,40 As shown in Table 4, in agreement with our
previously published results,41 stobadine and its carbo-
xymethylated hexahydropyridoindole derivatives 7a
and 7b rapidly reacted with a,a0-diphenyl-b-pic-
rylhydrazyl (DPPH), their antiradical activity being
from 4 to 8 times higher than that of their tetrahydropy-
ridoindole analogues 5a–5c. An absorbance decrease at
518 nm at time intervals of 5 min was used as a measure
of the antiradical activity of the compounds tested.
DPPH, as a weak hydrogen atom abstractor, is consid-
ered a good kinetic model for peroxyl ROOÆ radicals.42
In a homogeneous solution system of DPPH in ethanol,
antioxidant activity stems from an intrinsic chemical
reactivity toward radicals. In membranes, however, the
relative reactivity may be different since it is determined
also by additional factors, such as location of the antiox-
idant and radicals, ruled predominantly by their parti-
tion ratios between water and lipophilic compartments.
-40
-20
0
20
40
60
80
NADH (mM) 1/cP
Figure 2. Inhibitory effect of compound 5b on rat lens aldose
reductase. Typical double reciprocal plot of the initial enzyme velocity
versus the concentration of cofactor (NADPH) in the presence or
absence of 5b: (j) no inhibitor; (d) 20 lM 5b (uncompetitive type of
inhibition).
Table 2. Kinetic parameters of rat lens ALR2 and the inhibitory effect
of compounds 5b and 5c
Varied substrate/
cofactor
Km (lM)
Vmax (10,000 Kia (lM)
OD/s/mg)
D,L-Glyceraldehyde 585.0 90.0 4.3 1.3
17.9 3.0 (5b)
17.7 2.1 (5c)
In our further experiments, unilamellar dioleoyl L-a-
phosphatidylcholine (DOPC) liposomes were used as
model membranes.41,43 Peroxidation of liposomes was
induced by a water-soluble radical generator, 2,20-azo-
bis(2-amidinopropane)hydrochloride (AAPH), which
simulates an attack by free radicals from the aqueous re-
gion. The monoene DOPC was used in order to mini-
mize adventitious heat-induced autooxidation of
liposomes. Indeed, no measurable accumulation of
hydroperoxides was observed in the reaction system of
liposomes heated to 50 ꢁC when the initiator AAPH
was omitted. In a complete reaction system, DOPC lip-
osomes/AAPH/buffer, lipid peroxidation proceeded at a
constant rate, with a near linear time-dependent increase
of lipid hydroperoxides observed, without an induction
period (Fig. 3). The carboxymethylated tetrahydropy-
ridoindole 5b effectively suppressed oxidation and gave
a distinct lag phase in the lipid peroxide accumulation
curve, indicating a much higher reaction rate of the
compound with peroxyl radicals compared to the rate
of chain propagation of phosphatidylcholine peroxida-
tion. However, the presence of polar acidic function in
NADPH
54.0 12.0 6.8 1.4
11.0 1.9 (5b)
10.5 1.9 (5c)
a Uncompetitive type of inhibition. Results are mean values SD from
at least three experiments.
the aldose reductase, (Km)Glyceraldehyde = 0.585 mM and
(Km)NADPH = 0.054 mM, were in the range of those
determined by other authors for partially purified rat
lens ALR2.34–36
An important feature of pharmacologically applicable
ARIs is their selectivity of action. The co-inhibition of
structurally related physiological oxidoreductases might
have unwanted side effects. In testing for selectivity we
used the comparison to an enzyme with the highest
homology, aldehyde reductase (ALR1).37,38 The IC50
values of compounds 5a–5c and 7b, for their inhibition
of the reduction of glucuronide substrate by partially
purified ALR1 from rat kidney, in comparison with
the standard valproate, are shown in Table 3. The corre-
sponding selectivity factors calculated for the tetrahyd-
ropyridoindole congeners 5a–5c ranged from approx.
20 for 5b through 40 for 5c to about 60 for 5a. These
compounds are expected to have even less affinity for
other enzymes that are not as closely related to ALR2.
Table 4. Antiradical activities of substituted pyridoindoles in a DPPH
testa
Compound
Absorbance decrease
(ꢀDA/5 min)
5a
5b
0.031 0.007
0.030 0.006
0.031 0.009
0.132 0.014
0.187 0.020
0.239 0.019
0.033 0.008
Table 3. Inhibition of rat kidney ALR1
5c
7a
Compound
IC50 (lM)
5a
5b
3081.3
7b
Stobadine
Dehydrostobadine
328.5 21.9
603.0
>3000
5c
7b
Results are mean values SD from at least three measurements.
a The ethanolic solution of DPPH radical (50 lM) was incubated in the
presence of the compound tested (50 lM). Absorbance decrease at
518 nm during the first 5 min interval was determined.
Valproic acid
56.1 2.7
Results are mean values from two or mean values SD from at least
three experiments.

4912
M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
Table 6. Distribution ratios of the substituted pyridoindoles in the
system 1-octanol/phosphate buffer
0.6
0.5
0.4
0.3
0.2
0.1
0
Compound
Distribution ratio (D)a
5a
5b
0.13 0.05
0.35 0.02
0.51 0.03
<0.10
5c
7a
7b
Stobadine
Dehydrostobadine
0.20 0.04
2.50 0.08
6.00 0.42
Results are mean values SD from at least three independent
experiments.
a Distribution ratio in the system of 1-octanol/0.1 M phosphate buffer
(pH 7.4) determined by a shake-flask method.
0
20
40
60
80
Time (min)
Figure 3. Typical kinetic curve of AAPH-induced peroxidation of
DOPC liposomes (d) and the appearance of inhibition period in the
presence of compound 5b (250 lM, m). DOPC liposomes (0.8 mM)
were incubated in the presence of AAPH (10 mM) in phosphate buffer
(pH 7.4, 20 mM) at 50 ꢁC.
0.6
0.5
0.4
0.3
0.2
0.1
0
Table 5. Antioxidant activity of compounds 5b and 7b in comparison
with stobadine, dehydrostobadine, and standard trolox
a
Compound
IC50 (lM)
5b
7b
168.1 14.5
75.6 2.5
25.3 14.6
72.7 19.8
93.5 8.5
Stobadine
Dehydrostobadine
Trolox
0
2
4
6
8
10
12
14
pH
Figure 4. pH-distribution profile of compounds 5b ( ) and 5c (s) in
1-octanol/buffer system. Experimental points represent mean values
from two measurements or a mean value SD from four measurements.
Results are mean values SD from at least three independent
incubations.
a Values of IC50 determined for the inhibition of AAPH-induced per-
oxidation of DOPC liposomes at the time interval of 80 min. DOPC
liposomes (0.8 mM) were incubated in the presence of AAPH
(10 mM) in phosphate buffer (20 mM; pH 7.4) at 50 ꢁC.
acidic ARIs whose carboxylic acid function is ionized
at neutral pH resulting in a sharp drop in distribution
ratios and poor biological availability under physiolog-
ical conditions. Owing to the presence of a basicity cen-
ter at the tertiary nitrogen, the compounds are typical
ampholytes.44 Dissociation constants were determined
for compound 5b as follows: pKAa cidic ¼ 4:34 ꢁ 0:07 and
pKBa asic ¼ 8:65 ꢁ 0:07. Since pKaAcidic < pKaBasic, the acidic
and basic functionalities can be ionized at the same time
resulting in double charged zwitterionic species.44 Zwit-
terionic ampholytes often exhibit incompletely under-
stood effects of intramolecular neutralization that may
render the zwitterionic species markedly more lipophilic
than the single charged cations and anions, giving typi-
cally a lipophilicity profile of a characteristic bell-shape
with the maximum around the isoelectric point.44 The
parabolic distribution profiles shown for 5b and 5c are
those of typical zwitterionic ampholytes.
the molecule of 5b is expected to hinder its passage into
the lipid compartment, reflected by correspondingly
lower distribution ratio of 5b compared with its more
lipophilic congener dehydrostobadine (Table 6). This
lipophilicity drop eventually may account for lower
overall antioxidant efficiency of 5b and 7b, based on
IC50 values (see Table 5), observed in the liposomes in
spite of the fact that the intrinsic DPPH antiradical
activity of 5b and 7b are similar to those of parent dehy-
drostobadine and stobadine, respectively (see Table 4).
Still, the antioxidant activity of hexahydropyridoindole
7b is comparable to that of the standard trolox.
2.5. Partitioning
Distribution ratios (D), defined by the total concentra-
tion of a solute in organic phase divided by that in aque-
ous phase, were determined for the compounds studied
by a standard shake-flask technique. Experimental D
values obtained at the physiological pH are shown in
Table 6. The pH-lipophilicity profile experimentally
determined for compounds 5b and 5c in the system of
1-octanol/buffer was characterized by a bell-shaped
curve, with a maximal distribution ratio near neutral
pH (Fig. 4). This behavior is in contrast with that of
Switch from the methyl substituent in compound 5a to
more lipophilic benzyl- or phenethyl-substituents in
compounds 5b and 5c, respectively, resulted in signifi-
cant increase in the distribution ratios (Table 6) with ex-
pected consequences in their potentially increased
biological availability to lipid compartments under
in vivo conditions. On the other hand, the lipophilicity
of the –R2 substituent affected the aldose reductase inhi-
bition just marginally compared with the effect of reduc-

M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
4913
Table 7. Individual interaction energies between the inhibitors and the
specific amino acid residues of ALR2
ing the tetrahydropyridoindole moiety to that of hexa-
hydropyridoindole, as discussed below.
Compound
Eint (I ꢀ R) (kcal/mol)
2.6. Molecular modeling
NADP+
Tyr48
His110
Trp111
5b
(ꢀ)-7b
Zopolrestat
ꢀ12.5
ꢀ9.2
0.0
ꢀ19.4
ꢀ9.4
ꢀ3.4
ꢀ3.8
ꢀ2.6
ꢀ3.8
Based on the comparison of the appropriate IC50 val-
ues listed in Table 1, the unsaturated tetrahydropyrid-
oindole type compound 5b was found to be an over
one order more efficient inhibitor of ALR2 than the
saturated hexahydropyridoindole congener 7b. To re-
late the inhibitory efficiencies of compounds 5b and
7b to their structural features, we performed several
computational studies. The lower inhibitory activity
of 7b in relation to 5b may be due to the specific dis-
tinction of the stereoelectronic profiles of these com-
pounds. Visualization of low energy conformations
showed almost planar tricyclic moiety of the tetrahyd-
ropyridoindole 5b, contrasting with severe space dis-
tortion of the lipophilic heterocyclic backbone of
hexahydropyridoindole 7b (see Fig. 5). The presence
of an extended aromatic planar region in the majority
of potent ARIs is well documented as a crucial phar-
macophoric element.11,12,45–48
ꢀ19.3
ꢀ12.3
ꢀ14.1
tive site of ALR2 (Tyr48, His110, Trp111)49 and NADP+,
calculated for 5b and for the theoretically more efficient
(ꢀ)-enantiomer of 7b, are summarized in Table 7.
Interaction with Tyr48 seems to be crucial for the inhibi-
tory activity of the compounds since comparison of the
Eint with Tyr48 gives values of ꢀ19.4 and ꢀ9.4 kcal/mol
for 5b and (ꢀ)-7b, respectively. This effect is reflected also
by the geometric parameters of the minimized complexes:
the mutual distance between the carboxylic oxygen of the
inhibitor and the hydrogen of Tyr48 was found to be
˚
˚
1.6 A for 5b, but 2.7 A for (ꢀ)-7b (Fig. 6). Molecular
electrostatic potential of ALR2 complexed with (ꢀ)-7b
and 5b derivatives after total optimization is shown in
Figure 7.
The values of total interaction energies Eint between
ALR2 residues and the individual compounds studied
were calculated by docking followed by total optimiza-
tion of the inhibitor-ALR2-NADP+ ternary complex.
The calculated results of Eint obtained for double-
charged zwitterionic species, ꢀ90.7 kcal/mol for 5b,
ꢀ77.3 and ꢀ74.5 kcal/mol for (ꢀ)-7b and (+)-7b, respec-
tively, and that of ꢀ107.0 kcal/mol for standard zopolre-
stat agreed well with the order of the inhibitory
activities. For structure 5b a much lower value of Eint,
ꢀ59.3 kcal/mol, was calculated for the uncharged
molecular species. The neutral molecule is practically
the only form expected to be in equilibrium with
corresponding zwitterions around the physiological
pH, being in this case close to pHiso = 6.5, calculated
on the basis of the macroscopic dissociation constants.
2.7. ‘Drug-likeness’
The ‘drug-likeness’ of the novel aldose reductase inhibi-
tors was assessed on the basis of their structural proper-
ties by applying the Lipinski’s ‘rule of five’, and
moreover the polar surface area and the number of
rotatable bonds as two additional molecular descrip-
tors50–52 (Table 8). For all the drug candidates studied,
none of the criteria was violated, thus predicting their
good oral bioavailability.
3. Conclusions
The carboxymethylated hexahydropyridoindoles 7a, 7b
and tetrahydropyridoindoles 5a–5c structurally based
on the antioxidant drug stobadine are presented as novel
aldose reductase inhibitors endowed with antioxidant
activity. The computed stereoelectronic properties of
compounds 7b and 5b along with a modeling study of
the ALR2-binding site could provide an explanation
for the higher inhibitory efficiencies of the tetrahydropy-
ridoindoles in comparison to the hexahydropyridoin-
doles. The distribution profile of the representative
inhibitors 5b and 5c showed maximal extraction yield
around pH 7 in a 1-octanol/water system. This behavior
can obviously be accounted for by the zwitterionic
ampholyte nature of the compounds and by higher lipo-
philicity of the doubly charged zwitterionic form when
compared to a singly charged species. Properties implicit
to ‘drug-likeness’, inferred from simple molecular
descriptors, rendered novel aldose reductase inhibitors
potential drugs with expected good bioavailability and
with a prospect of further optimization that could lead
to compounds of even higher potency. The zwitterionic
principle thus offers an interesting way how to improve
potentially the bioavailability of aldose reductase inhib-
itors bearing an acidic function.
Individual interaction energies between the inhibitor
molecules and the specific amino acid residues of the ac-
Figure 5. Stereoview of hexahydropyridoindole (ꢀ)-7b (left) versus
tetrahydropyridoindole 5b (right).

4914
M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
Figure 6. Geometry of the active site for (ꢀ)-7b (left) and 5b (right) derivative optimized with the whole enzyme. Dashed line denotes the hydrogen
bonds between the carboxylic oxygen and the hydrogen of Tyr48.
Figure 7. Molecular electrostatic potential of ALR2 complexed with (ꢀ)-7b (up) and 5b (below) derivatives after total optimization.73 Due to sterical
restrictions, the saturated hexahydroderivative (ꢀ)-7b is not able to insert into the cavity of the active site sufficiently and thus to contact the specific
amino acid residues in proper interaction distances as the unsaturated tetrahydropyridoindole derivative 5b.
4. Experimental
300.13 MHz for 1H and 75.46 MHz for 13C. 1H and
13C NMR spectra were acquired using internal acetone
(1H d = 2.225; 13C d = 31.07) as a reference standard in
CDCl3 or D2O used as solvents. To assign the signals,
the COSY, HSQC, TOCSY, NOESY, and DEPT tech-
niques were used. Elemental analyses were performed
4.1. Chemistry
4.1.1. General. NMR spectra were recorded on Bruker
Avance DPX 300 spectrometer operating at

M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
Table 8. ‘Drug-likeness’ of the carboxymethylated pyridoindoles
4915
2
˚
PSA < 120 A
Compound
MW (<500)
NH/OH (<5)
N/O (<10)
ClogP (<5)
nROTB (<8)
5a
5b
5c
7a
7b
244.32
320.42
334.45
246.34
322.44
2
2
2
2
2
4
4
4
4
4
1.57
2.97
3.38
1.13
2.60
56.33
56.33
56.33
52.56
52.56
3
4
5
3
4
MW, molecular weight; OH/NH, number of H-bond donors; O/N, number of H-bond acceptors; ClogP, calculated decimal logarithm of octanol/
water partition coefficient71; PSA, polar surface area72; nROTB, number of rotatable bonds.
on EA 1108 Carlo Erba (for 5b also by Galbraith Lab-
oratories, Knoxville, TN, USA) and were within 0.4%
of the theoretical values. Melting points were deter-
mined on a Kofler melting point apparatus with digital
thermometer (DT012C) and are uncorrected. The purity
of the intermediate products (4a–4c) and (6a, 6b) was
determined by GC–MS on Hewlett–Packard GC HP
5980 using the column HP-5 (12 m · 0.22 mm ·
0.33 lm) and mass detector Hewlett–Packard MSD
HP 5970B. Molecuar masses of the final potassium salts
(5a–5c) and (7a, 7b) were determined on MALDI TOF
IV (Shimadzu, Kratos Analytical, Manchester, UK)
instrument. 2,5-Dihydroxy benzoic acid was used as a
matrix. Samples were irradiated by 337 nm photons
from the nitrogen laser. Typically, 100 shots were
summed into a single mass spectrum. Stobadine, (ꢀ)-
cis-2,8-dimethyl-2,3,4,4a,5,9b-hexahydro-1H-pyrido[4,
3-b]-indole, and its unsaturated analogue dehydrostoba-
dine, 2,8-dimethyl-2,3,4,5-tetrahydro-1H-pyrido[4,3-b]-
indole were supplied by the Institute of Experimental
Pharmacology, Slovak Academy of Sciences as
hydrochloride salts.53,54 The starting materials 4-cya-
nomethylphenylhydrazine hydrochloride (1) and 4-car-
boxymethylphenylhydrazine hydrochloride (2) were
synthesized by published procedures in yields compara-
ble with literary data.16,17 All solvents and reactants
were purified before use according to standard proce-
dures.55 Commercially available 1-methyl-4-piperidone
(3a) and 1-benzyl-4-piperidone (3b) were purified by
vacuum distillation and 1-(2-phenethyl)-4-piperidone
(3c) was crystalized from n-hexane. Column chromatog-
raphy was done by using the flash chromatography tech-
nique and was carried out on silica gel 60 (230–400
mesh, Merck). L-a-Phosphatidylcholine dioleoyl
(C18:1, [cis]-9) (DOPC) (99% grade), 2,6-di-tert-butyl-
p-cresol (BHT), and a,a0-diphenyl-b-picrylhydrazyl
(DPPH) radical were obtained from Sigma Chemical
Co. (St. Louis, MO, USA). 2,20-azobis (2-amidinopro-
pane)hydrochloride (AAPH) was obtained from Fluka
Chemie GmbH (Buchs, Switzerland). Other chemicals
were purchased from local commercial sources and were
of analytical grade quality. All solvents used for lipid
peroxidation studies were deaereated under nitrogen.
cooled to 5–8 ꢁC and stirred vigorously for 30 min. The
solid material formed was filtered, washed with water
and dried. The crude product was dissolved in a hot
mixture of 3% triethanolamine in toluene (600 mL),
cooled and filtered through a short (5 cm high and
3 cm diameter) column of silica gel. The column was
washed several times with the same solvent as used for
dissolving the product, and the filtrates were concen-
trated under reduced pressure to afford crude 4a, which
was crystalized at ꢀ5 ꢁC from methanol, filtered and
washed with a small volume of cold methanol to yield
pure 4a (12.97 g, 57.6%): mp: 164–166 ꢁC; 1H NMR
(CDCl3, 25ꢁ): 2.54 (s, 3H, CH3N), 2,78 (t, 2H, H-4),
2,84 (t, 2H, H-3), 3,68 (s, 2H, H-1), 3.80 (s, 2H,
CH2CN), 6.93 (d, 1H, H-6), 7.10 (d, 1H, H-7), 7.31 (s,
1H, H-9), 8.81 (s, 1H NH); 13C NMR (CDCl3, 25 ꢁC):
23.68 (C-4), 23.74 (CH2CN), 45.79 (CH3N), 51.63 (C-
3), 52.41 (C-1), 108.40 (C-9b), 111.24 (C-6), 116.84
(C-9), 119.04 (CN), 120.42 (C-8), 120.82 (C-7), 126.46
(C-9a), 133.30 (C-5a), 135.68 (C-4a); MS m/z
(%) = 225 (M+, 46), 182 (100), 154 (14), 140 (7), 128
(7) 99 (11).
4.1.3. 2-Benzyl-2,3,4,5-tetrahydro-1H-pyrido[4,3-
b]indole-8-yl acetonitrile (4b). The mixture of
1
(18.36 g, 0.1 mol), 3b (8.93 g, 0.1 mol) and methanol
(500 mL) was stirred and refluxed for 1.5 h in argon
atmosphere and then the solution was evaporated under
reduced pressure to dryness. The solution of 6% HCl in
acetic acid (600 mL) was added to the residue and the
reaction mixture was stirred and heated up to reflux un-
der argon atmosphere for 10 s, and then was stirred
without heating for 15 min. The reaction mixture was
heated again to a short reflux, and the further procedure
was as in the previous step. This procedure was repeated
six times, then the mixture was concentrated under re-
duced pressure, diluted with water (300 mL), alkalized
with 40% aqueous NaOH to pH 10 and extracted several
times with CH2Cl2. The combined organic phases were
washed with brine, dried (Na2SO4), and concentrated
under reduced pressure. The crude product was dis-
solved in a hot mixture of 3% triethanolamine in toluene
and treated as previously described for 4a, crystalized
from methanol at ꢀ5 ꢁC, filtered and washed with a
small volume of cold methanol to give pure 4b
(25.47 g, 84.5%): mp: 120–122 ꢁC; 1H NMR (CDCl3,
25 ꢁC): 2.78 (t, 2H, H-4), 2.84 (t, 2H, H-3), 3.72 (s,
2H, H-1), 3.79 (s, 2H, CH2CN), 6.93 (d, 1H, H-6),
7.14 (d, 1H, H-7), 7.20–7.42 (m, 6H, H-9, Ph-H), 8.23
(s, 1H, NH); 13C NMR (CDCl3, 25 ꢁC): 23.7 (C-4,
CH2CN), 49.59 (C-3), 50.14 (C-1), 62.41 (CH2Ph),
108.43 (C-9b), 111.24 (C-6), 116.97 (C-9), 119.05 (CN),
4.1.2. 2-Methyl-2,3,4,5-tetrahydro-1H-pyrido[4,3-
b]indole-8-yl acetonitrile (4a). The mixture of
1
(18.36 g, 0.1 mol), 3a (11.31 g, 0.1 mol), and 500 ml of
anhydrous 10% HCl ethanolic solution was stirred and
refluxed for 5 h under argon atmosphere. After evapora-
tion of the solvent to dryness, water (300 mL) was
added, stirred for a few minutes, then the solution was
carefully made basic to pH 10 with 40% aqueous NaOH,

4916
M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
120.51 (C-8), 120.92 (C-7), 126.55 (C-9a), 127.43,
128.5, 129.35, 138.00 (C-Ph), 133.48 (C-5a), 135.58
(C-4a); MS m/z (%) = 301 (M+, 18), 182 (100), 154 (4),
91 (27).
4.1.5.1. ( )-2-Methyl-(4a,9b)-cis-1,2,3,4,4a,9b-hexa-
hydro-1H-pyrido[4,3-b]indole-8-yl acetonitrileÆ2 HCl
(6a). Isolated 10.3 g, (68.6% yield), mp: 172–175 ꢁC;
1H NMR (CDCl3, 25 ꢁC): 1.78, 1.92 (m, m, 1H, 1H,
H-4,40), 2.27 (s, 3H, NCH3), 2.3–2.39 (m, 3H, H-1, H-
3,30), 2.63 (dd, 1H, H-10), 3.21 (dd, 1H, H-9b), 3.63 (s,
2H, CH2CN), 3.72 (br s, 1H, NH), 3.81 (dd, 1H, H-
4a), 6,63 (d, 1H, H-6), 6.95 (d, 1H, H-7), 7.05 (s, 1H,
H-9); 13C NMR (CDCl3, 25 ꢁC): 22.96 (CH2CN), 28.62
(C-4), 40.88 (C-9b), 46.36 (CH3N), 51.04 (C-3), 56.58
(C-1), 57.04 (C-4a), 110.11 (C-6), 118.52 (CN), 123.03
(C-7), 119.72 (C-9), 127.21 (C-9a), 132.77 (C-5a),
150.32 (C-8); MS m/z (%) = 227 (M+, 64), 184 (18),
155 (52), 71 (100), 58 (96).
4.1.4. Methyl ester of 2-phenethyl-2,3,4,5-tetrahydro-1H-
pyrido[4,3-b]indole-8-yl acetic acid (4c). The mixture of 2
(17.06 g, 0.1 mol) and 3c (20.33 g, 0.1 mol) in a solution
of 10% HCl in anhydrous CH3OH (500 mL) was stirred
and refluxed for 6 h under argon atmosphere. After
evaporation of the solvent, water (300 mL) was added,
and carefully made basic to pH 10 under cooling to 5–
8 ꢁC and vigorous stirring for 30 min by addition of
40% aqueous NaOH solution. The solid material was
extracted by the addition of several portions of CH2Cl2.
The combined organic phases were washed with brine,
dried (Na2SO4), and concentrated under reduced pres-
sure. The crude product was subjected to flash chroma-
tography procedure as described for 4a followed by
crystalization from ether. Recrystalization from metha-
nol/ether yielded 4c (15.82 g, 45.4%): mp: 127–129 ꢁC;
1H NMR (CDCl3, 25 ꢁC): 2.80–3.08 (m, 8H, H-3, H-4,
CH2CH2Ph), 3.58 (s, 2H, H-1), 3.68 (s, 3H, CH3O),
3.95 (s, 2H, CH2CO2CH3), 6.96–7.32 (m, 8H, H-6, H-
7, H-9, Ph-H), 8.48 (br s, 1H, NH); 13C NMR (CDCl3,
25 ꢁC): 22.72 (C-4), 33.14 (CH2CH2Ph), 41.32 (CH3O),
45.76 (CH2CO2CH3), 49.29 (C-3), 50.55 (C-1), 59.13
(NCH2CH2Ph), 110.94 (C-6, C-9b), 117.91 (C-9),
122.81 (C-7), 124.97 (C-4a), 125.95, 126.47, 128.61,
128.74 (C-Ph), 131.75 (C-9a), 135.35 (C-5a), 138.94 (C-
8), 172.96 (CO); MS m/z (%) = 348 (M+, 1), 334(18),
215 (100), 156 (54), 91 (24).
4.1.5.2. ( )-2-Benzyl-(4a,9b)-cis-1,2,3,4,4a,9b-hexahy-
dro-1H-pyrido[4,3-b]indole-8-yl acetonitrileÆ2 HCl (6b).
Isolated 13.41 g (71.3% yield), mp: 197–201 ꢁC; 1H
NMR (CDCl3, 25 ꢁC) 1.72, 1.90 (m, m, 1H, 1H, H-
4,40), 2.23 (dd, 1H, H-1), 2.43 (m, 2H, H-3,30), 2.62
(dd, 1H, H-10), 3.13 (dd, 1H, H-9b), 3.47 (s, 2H,
NCH2Ph), 3.52 (s, 2H, CH2CN), 3.76 (dd, 1H, H-4a),
4.11 (br s, 1H, NH), 6.57 (d, 1H, H-6), 6.85 (s, 1H, H-
9), 6.89 (d, H-7), 7.31 (m, 5H, Ph-H); 13C NMR (CDCl3,
25 ꢁC): 22.79 (CH2CN), 28.35 (C-4), 40.69 (C-9b), 48.96
(C-3), 54.17 (C-1), 57.58 (C-4a), 62.82 (CH2Ph), 109.90
(C-6), 118.48 (CN), 119.42 (C-7), 123.15 (C-9), 126.86,
127.05, 128.03, 128.87 (C-Ph), 132.98 (C-9a), 138.01
(C-5a), 150.31 (C-8); MS MALDI m/z (%): 303 (M+H).
4.1.6. General procedure for hydrolysis of 4a–c and 6a–b.
The mixture of KOH (0.15 mol, 3.0 equiv), isopropanol
(100 mL) and water (4 mL) was flushed off the air by
bubbling with argon for 10 min. Then, the appropriate
compounds 4a–c or 6a, b (0.05 mol, 1.0 equiv) were
added and the reaction mixture was heated and stirred
under argon atmosphere on reflux. In the case of com-
pound 4c, a 1-day reaction time was used. Elimination
of free ammonia indicated progress of the reaction in
the case of the acetonitriles 4a, b and 6a, b. After 5 days
of refluxing, the release of ammonia stopped, the reac-
tion mixture was concentrated under reduced pressure
to dryness and the residue was dissolved in air-free water
(150 mL), warmed with active charcoal and filtered. Un-
der intensive stirring in ice/NaCl bath, an ice-cold solu-
tion of 20% aq HCl was slowly added to the filtrate
(temperature kept in the range 3–7 ꢁC) in argon atmo-
sphere to adjust pH to approx. 6.5, when the solid prod-
uct started to crystallize. The intensive stirring
continued for further 4 h at the above temperature, the
mixture was filtered and the solid was washed with a
small volume of cold air-free water to afford an amino
acid product, which was dried over P2O5 and argon
atmosphere to constant weight. Potassium salts of
the appropriate amino acids were prepared by careful
treatment of their suspensions in dry methanol
(150 mL) with solid KHCO3. Then the mixtures were
filtered through a short (5 cm high and 3 cm diameter)
column of silica gel. The column was washed with
methanol, the filtrates were concentrated under reduced
pressure to dryness and the crude products were recrys-
talized as potassium (sodium) salts at ꢀ5 ꢁC from the
solvents listed in Table 9.
4.1.5. General procedure for the reduction of tetrahydro-
pyridoindoles (4a and 4b). To the magnetically stirred
cold solution (ꢀ5 to 0 ꢁC) of 4a or 4b (0.05 mol) in
TFA (125 mL), 95% NaBH3CN (6.61 g, 0.1 mol,
2.0 equiv) was gradually added during a 40-min period.
Then after removing the cooling bath, the reaction mix-
ture was stirred at room temperature for 2.5 h. The reac-
tion mixture was carefully poured into an ice-cold
solution of NaOH (70 g) in water (105 mL), with the fi-
nal pH being approx. 10. The target bases 6a or 6b were
taken up in CH2Cl2 by repeated extraction. The com-
bined organic phases were washed with brine, dried
(anhydrous Na2SO4), and concentrated under reduced
pressure. The residue was dissolved in dry methanol
(15 mL) at 40–50 ꢁC, acidified by the addition of a
15% methanolic HCl solution (15 mL), well mixed,
and then crystalized at ꢀ5 ꢁC overnight. The crystals
of the dihydrochloride salts of 6a and 6b were filtered,
washed with a small volume of cold methanol and ace-
tone, and dried at 90 ꢁC during a 1-h period to give pure
compound 6a or 6b. For NMR analysis and the final
hydrolysis step, free bases were isolated as follows: the
dihydrochloride salts were dissolved in water and after
the addition of 40% aqueous NaOH to reach pH 10,
the mixture was extracted several times with CH2Cl2.
The combined organic phases were washed with brine,
dried with anhydrous Na2SO4, filtered and concentrated
under reduced pressure to dryness to produce the bases
of hexahydropyridoindole 6a or 6b for further
applications.

M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
4917
4.1.6.1. Potassium 2-methyl-2,3,4,5-tetrahydro-1H-
pyrido[4,3-b]indole-8-yl acetate (5a). 1H NMR (D2O,
25 ꢁC): 2.35 (s, 3H, CH3N), 2,73 (br s, 4H, H-3, H-4),
3,50 (s, 4H, H-1, CH2COOK), 7.00 (d, 1H, H-7), 7.25
(s, 1H, H-9), 7.25 (d, 1H, H-6); 13C NMR (D2O, 25
ꢁC): 23.30 (C-4), 44.86 (N–CH3), 45.48 (CH2COOK),
51.21 (C-1), 52.15 (C-3), 107.24 (C-9b), 112.00 (C-6),
118.31 (C-9), 123.30 (C-7), 126.42 (C-9a), 128.88 (C-
4a), 133.91 (C-5a), 135.62 (C-8), 182.82 (COOK); MS
MALDI m/z = 283 (M+H).
4.1.6.5. Sodium salt of ( )-2-benzyl-(4a,9b)-cis-
1,2,3,4,4a,9b-hexahydro-1H-pyrido[4,3-b]indole-8-yl ace-
tic acid (7b). H NMR (D2O, 25 ꢁC): 1.80, 1.94 (m, m,
1
1H, 1H, H-4,40), 2.26 (dd, 1H, H-1), 2.51, 2.65 (m, m,
1H, 1H, H-3,30), 2.78 (dd, 1H, H-10), 3.20 (dd, 1H, H-
9b), 3.39 (s, 2H, CH2COOK), 3.59 (m, AB quartet,
2H, NCH2Ph), 3.77 (dd, 1H, H-4a), 6.77 (d, 1H, H-6),
6.96 (s, 1H, H-9), 6.99 (d, 1H, H-7), 7.38 (m, 5H, Ph-
H); 13C NMR (D2O, 25 ꢁC): 27.55 (C-4), 40.91 (C-9b),
44.98 (CH2COONa), 49.50 (C-3), 53.97 (C-1), 57.57
(C-4a), 62.93 (NCH2Ph), 112.67 (C-6), 125.74 (C-7),
129.48 (C-9), 128.80, 129.97, 131.15, 133.61 (C-Ph),
136.48 (C-9a), 149.40 (C-5a), 163.70 (C-8), 182.36
(COONa); MS MALDI m/z = 345 (M+H).
4.1.6.2. Potassium 2-benzyl-2,3,4,5-tetrahydro-1H-
pyrido[4,3-b]indole-8-yl acetate (5b). 1H NMR (D2O,
25 ꢁC) 2.35 (br s, 2H, H-4), 2.44 (br s, 2H, H-3), 3.37
(s, 2H, NCH2Ph), 3.43 (s, 2H, CH2CO2K), 3.48 (s,
2H, H-1), 6.93–7.23 (m, 8H, H-6, H-7, H-9, Ph-H);
13C NMR (D2O, 25 ꢁC) 23.09 (C-4), 45.61 (CH2CO2K),
49.13 (C-3), 49.71 (C-1), 61.74 (NCH2Ph), 106.96 (C-
9b), 111.97 (C-6), 118.42 (C-9), 123.31 (C-7), 126.55
(C-4a), 128.50, 128.71 129.30, 130.85, (C-Ph), 133.52
(C-9a), 135.63 (C-5a), 137.22 (C-8), 182.74 (COOK);
MS MALDI m/z = 359 (M+H).
4.1.7. Partitioning. The distribution ratios D in 1-octa-
nol/buffer systems, defined by total concentration of a
solute in organic phase divided by that in aqueous
phase, were measured using the shake-flask technique56
at room temperature. The organic and aqueous phases
were mutually saturated. Compounds were dissolved
in aqueous buffer solutions (0.1 M phosphate or citrate
buffers) with KCl (0.15 M) in final concentration of
100 lM; the solutions were shaken with 1-octanol for
3 h. Both aqueous and organic phase volumes were
3 mL. The phases were separated by centrifugation for
1 h. The organic layer was removed with a Pasteur pip-
ette. The concentration of the solute was determined in
both phases by UV spectrophotometry.
4.1.6.3. Potassium 2-phenethyl-2,3,4,5-tetrahydro-1H-
pyrido[4,3-b]indole-8-yl acetate (5c). 1H NMR (D2O,
25 ꢁC): 2.45–2.71 (m, 8H, H-3, H-4, CH2CH2Ph), 3.44
(s, 2H, CH2CO2K), 3.51 (s, 2H, H-1), 6.94–7.29 (m,
8H, H-6, H-7, H-9, Ph-H); 13C NMR (D2O, 25 ꢁC):
23.19 (C-4), 33.46 (CH2Ph), 45.59 (CH2CO2K), 49.30
(C-3), 50.31 (C-1), 59.46 (NCH2CH2Ph), 107.03 (C-
9b), 111.99 (C-6), 118.37 (C-9), 123.32 (C-7), 126.57
(C-4a), 127.18, 128.79, 129.55, 129.65 (C-Ph), 133.88
(C-9a), 135.66 (C-5a), 141.10 (C-8), 182.72 (COOK);
MS MALDI m/z = 373 (M+H).
4.1.8. DPPH test. To investigate the antiradical activity
of the pyridoindole derivatives, the ethanolic solution of
DPPH (50 lM) was incubated in the presence of a com-
pound tested (50 lM) at laboratory temperature. The
absorbance decrease, recorded at kax = 518 nm, during
the first 5-min interval was taken as a marker of the anti-
radical activity.
4.1.6.4. Potassium salt of ( )-2-methyl-(4a,9b)-cis-
1,2,3,4,4a,9b-hexahydro-1H-pyrido[4,3-b]indole-8-yl ace-
1
tic acid (7a). H NMR (D2O, 25 ꢁC): 1.78, 1.83 (m, m,
1H, 1H, H-4,40), 2.30 (s, 3H, CH3-N), 2.32 (um, 1H,
H-1), 2.52, 2.82 (um, m, 1H, 1H, H-3,30), 2.95 (um,
1H, H-1), 3.30 (dd, 1H, H-9b), 3.44 (s, 2H, CH2COOK),
3.80 (dd, 1H, H-4a), 6.81 (d, 1H, H-6), 7.02 (d, 1H, H-
7), 7.12 (s, 1H, H-9); 13C NMR (D2O, 25 ꢁC): 27.73
(C-4), 41.14 (C-9b), 44.83 (CH2COOK), 45.64 (CH3N),
51.43 (C-3), 56.99 (C-1, C-4a), 113.02 (C-6), 125.61 (C-
7), 129.36 (C-9), 133.45 (C-9a), 149.22 (C-5a), 164.62
(C-8), 172.01 (COOK); MS MALDI m/z = 285 (M+H).
4.1.9. Liposome peroxidation. The methods were de-
scribed previously.41,43 L-a-phosphatidylcholine dioleoyl
(DOPC, C18:1, [cis]-9, 15.7 mg) was dissolved in chloro-
form (5 mL) and placed into a round-bottom flask. The
solvent was subsequently removed under nitrogen and
the resulting thin film on the walls was dispersed in
phosphate buffer (20 mL, 20 mM, pH 7.4) by vigorous
stirring for 2 min followed by sonification for the same
period of time. A suspension of unilamellar liposomes
(1 mM DOPC) was thus obtained. The liposomes (final
concentration of 0.8 mM DOPC) were incubated in the
presence of different concentrations of the compounds
tested and of the water-soluble initiator AAPH (final
concentration 10 mM) at 50 ꢁC for different periods of
time. Aliquots (1 mL) of the incubation mixtures were
extracted by 2 mL portions of ice-cold mixture CHCl3/
MeOH (2:1, v/v) containing BHT (0.05%). Lipid hydro-
peroxide content was determined by thiocyanate method
according to Mihaljevic57 by sequentially adding CHCl3/
MeOH (2:1, v/v) mixture (1.4 mL) and the thiocyanate
reagent (0.1 mL) to 1-mL aliquots of the liposome ex-
tracts. The reagent was prepared by mixing equivalent
volumes of methanolic solution of KSCN (3%) and fer-
rous-ammonium sulfate solution (45 mM in 0.2 mM
HCl). After the mixture was left at the ambient
Table 9. Chemical data for the carboxymethylated tetrahydropyrido-
indoles 5a–5c and hexahydropyridoindoles 7a, 7b
Compound Recrystalization Formulaa
Yieldb
(%)
solvent
(Molecular mass)
5a
5b
5c
7a
7b
i-PrOH/CH3OH C14H15N2O2K (282.39)
CH3OH/acetone C20H19N2O2K (358.48)
50.5
67.3
48.6
38.7
i-PrOH
CH3OH/acetone C14H17N2O2K (284.40)
i-PrOH/acetone
C21H21N2O2K (372.51)
C20H21N2O2Na (344.38) 41.5
a Elemental analyses for C, H, N were within 0.4% of the calculated
value (for details see Supporting information).
b The yields are based on the corresponding acetonitriles 4a, 4b or 6a, b
and the methylester 4c.

4918
M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
temperature for at least 5 min, the absorbance at 500 nm
was recorded by Hewlett–Packard Diode Array Spectro-
photometer 8452A.
sodium phosphate buffer, pH 7.2, containing 0.25 M
sucrose, 2.0 mM EDTA dipotassium salt, and 2.5 mM
b-mercaptoethanol. The homogenate was centrifuged
at 16,000g at 0–4 ꢁC for 30 min and the supernatant
was subjected to ammonium sulfate fractional precipita-
tion at 40%, 50%, and 75% salt saturation. The pellet
obtained from the last step, possessing ALR1 activity,
was redissolved in 10 mM sodium phosphate buffer,
pH 7.2, containing 2.0 mM EDTA dipotassium salt
and 2.0 mM b-mercaptoethanol to achieve total protein
concentration of approx. 20 mg/mL. DEAE DE 52 resin
was added to the solution (33 mg/mL) and after gentle
mixing for 15 min removed by centrifugation. The
supernatant containing ALR1 was then stored in
smaller aliquots in liquid nitrogen. No appreciable
contamination by ALR2 in ALR1 preparations was de-
tected since no activity in terms of NADPH consump-
tion was observed in the presence of glucose substrate
up to 150 mM.
4.1.10. Determination of the proton dissociation constant.
The potentiometric determination58–60 was performed
with Precision Digital pH Meter OP-208/1 equipped
with glass electrode OP-0718 P and saturated calomel
electrode OP-0830 P (Radelkis, Budapest, Hu). Auto-
matic byuret OP 930 (Radelkis, Budapest, Hu) was used
for titrations. The compound 5b was dissolved in dis-
tilled water free of carbon monoxide at a concentration
of about 2 mM. The solution was acidified by 0.1 M
HCl to pH of about 1.8 and titrated with fresh solution
of NaOH (0.1 M) at 22 ꢁC 1 ꢁC and at a constant io-
nic strength I = 0.1 M set by KCl. Three parallel titra-
tions were performed.
4.2. Enzyme section
4.2.1. General. NADPH, b-mercaptoethanol, D,L-glycer-
aldehyde, D-glucuronate, and sodium valproate were
obtained from Sigma Chemical Co. (St. Louis, MO,
USA). Diethylaminoethyl cellulose DEAE DE 52 was
from Whatman International Ltd (Maidstone, Eng-
land). Zopolrestat (lot # 43668-12-7F) was supplied as
gift samples by Pfizer (Groton, CT, USA). Other chem-
icals were purchased from local commercial sources and
were of analytical grade quality. Total protein in enzyme
preparations was determined according to Geiger and
Bessman.61
4.2.5. Enzyme assays. ALR1 and ALR2 activities were
determining
assayed
spectrophotometrically
by
NADPH consumption at 340 nm and were expressed
as decrease of the optical density (OD)/s/mg protein.
To determine ALR2 activity,63 the reaction mixture con-
tained 4.67 mM D,L-glyceraldehyde as a substrate,
0.11 mM NADPH, 0.067 M phosphate buffer, pH 6.2
and 0.05 mL of the enzyme preparation in a total vol-
ume of 1.5 mL. The reference blank contained all the
above reagents except the substrate D,L-glyceraldehyde
to correct for oxidation of NADPH not associated with
reduction of the substrate. The enzyme reaction was ini-
tiated by the addition of D,L-glyceraldehyde and was
monitored for 4 min after an initial period of 1 min at
30 ꢁC. ALR1 activity23 was assayed analogically using
20 mM D-glucuronate as a substrate in the presence of
0.12 mM NADPH in 0.1 M phosphate buffer pH 7.2
at 37 ꢁC. Enzyme activities were adjusted by diluting
the enzyme preparations with distilled water so that
0.05 mL of the preparation gave an average reaction
rate for the control sample of 0.020 0.005 absorbance
units/min. The effect of inhibitors on the enzyme activity
was determined by including in the reaction mixture
each inhibitor at required concentrations. The inhibitor
at the same concentration was included in the reference
blank. IC50 values (the concentration of the inhibitor re-
quired to produce 50% inhibition of the enzyme reac-
tion) were determined from the least-square analysis of
the linear portion of the semilogarthmic inhibition
curves. Each curve was generated using at least four
concentrations of inhibitor causing an inhibition in the
range from at least 25% to 75%.
4.2.2. Animals. Male Wistar rats 8–9 weeks old, weigh-
ing 200–230 g, were used as organ donors. The animals
came from the Breeding Facility of the Institute of
Experimental Pharmacology Dobra Voda (Slovak
Republic). The study was approved by the Ethics Com-
mittee of the Institute and performed in accordance with
the Principles of Laboratory Animal Care (NIH publi-
cation 83-25, revised 1985) and the Slovak law regulat-
ing animal experiments (Decree 289, Part 139, July 9th
2003).
4.2.3. Preparation of ALR2. ALR2 from rat lens was
partially purified using a procedure adapted from Hay-
man and Kinoshita62 as follows: lenses were quickly re-
moved from rats following euthanasia and homogenized
in a glass homogenizer with a teflon pestle in 5 vol of
cold distilled water. The homogenate was centrifuged
at 10,000g at 0–4 ꢁC for 20 min. The supernatant was
precipitated with saturated ammonium sulfate at 40%,
50%, and then at 75% salt saturation. The supernatant
was retained after the first two precipitations. The pellet
from the last step, possessing ALR2 activity, was dis-
persed in 75% ammonium sulfate and stored in smaller
aliquots in liquid nitrogen container.
4.3. Computational methods
Low energy conformations (starting geometries) of the
compounds were obtained by semi-empirical (AM1)
geometry optimization64 followed by Monte Carlo
(MMFF94) conformational search.65 The compounds
were docked into the active site of the enzyme (PDB
structure 1PWM—human ALR2 complexed with Fida-
restat)66 by the program DOCK (version 3.5)67 utilizing
the flexible ligands option. Candidate configurations of
4.2.4. Preparation of ALR1. ALR1 from rat kidney was
partially purified according to the reported procedure of
Costantino et al.23 as follows: kidneys were quickly re-
moved from rats following euthanasia and homogenized
in a knife homogenizer followed by processing in a glass
homogenizer with a teflon pestle in 3 vol of 10 mM

M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
4919
12. Lee, Y. S.; Chen, Z.; Kador, P. F. Bioorg. Med. Chem.
1998, 6, 1811.
13. Nicolaou, I.; Zika, C.; Demopoulos, V. J. J. Med. Chem.
2004, 47, 2706.
14. Rakowitz, D.; Muigg, P.; Schroder, N.; Matuszczak, B.
Pharmazie (Weinheim) 2005, 338, 419.
the molecules were visualized using the InsightII 2.1
package (Biosym Technologies Inc., San Diego, CA
USA) and selected based on active site interactions with
the C4 of the nicotinamide ring for the coenzyme as well
as with the active site residues Tyr48, His110, and
Trp111. Individual structures were assigned correct
bonds, hybridization states, and charges prior to mini-
mization. The ternary complex (enzyme, NADP+ and
the inhibitor) was fully energy minimized using the Dis-
cover 2.7 package (Biosym Technologies, San Diego,
CA, USA) on an O2 (R12000) workstation (Silicon
Graphics, Mountain View, CA, USA) according to pro-
cedures described previously.68–70 Calculations were per-
15. Horakova, L.; Stolc, S. Gen. Pharmacol. 1998, 30, 627.
16. Betzemeier, B.; Brandl, T.; Breitfelder, S.; Brueckner, R.;
Gerstberger, T.; Gmachl, M.; Grauert, M.; Hilberg, F.;
Hoenke, C.; Hoffmann, M.; Impagnatiello, M.; Kessler,
D.; Klein, C.; Krist, B.; Senden, M. U.S. Patent
20060100254, 2006; Thiazolyl-dihydro-indazole.
17. Glen, R. C.; Selwood, D. L.; Martin, G. R.; Foster, C. J.
U.S. Patent 5922748, 1999; Indole derivatives as 5-HT
agonists.
18. Robinson, B. The Fischer Indole Synthesis; John Wiley &
Sons: New York, 1982.
19. Snirc, V. Ph.D. thesis. Institute of Experimental Pharma-
cology, Slovak Academy of Sciences, Bratislava, 2006.
20. Maryanoff, B. E.; McComsey, D. F.; Nortey, S. O. J. Org.
Chem. 1981, 46, 355.
21. Baynes, J. W.; Thorpe, S. R. Diabetes 1999, 48, 1.
22. Coudert, P.; Albuisson, E.; Boire, J. Y.; Duroux, E.;
Bastide, P.; Couquelet, J. Eur. J. Med. Chem. 1994, 29, 471.
23. Costantino, L.; Rastelli, G.; Gamberini, M. C.; Vinson, J.
A.; Bose, P.; Iannone, A.; Staffieri, M.; Antolini, L.; Del
Corso, A.; Mura, U.; Albasini, A. J. Med. Chem. 1999, 42,
1881.
24. La Motta, C.; Sartini, S.; Mugnaini, L.; Simorini, F.;
Taliani, S.; Salerno, S.; Marini, A. M.; Da Settimo, F.;
Lavecchia, A.; Novellino, E.; Cantore, M.; Failli, P.;
Ciuffi, M. J. Med. Chem. 2007, 50, 4917.
formed including
a constant valence force field
incorporating the simple harmonic function for bond
stretching and excluding all nondiagonal terms (cut off
˚
distance between 26 and 33 A) using the algorithms
steepest descents and conjugate gradients (down to a
maximum atomic root-mean-square derivative of
˚
˚
10.0 kcal/A and 0.1 kcal/A, respectively). Individual
contributions of the interactions between the residues
(Tyr48, His110, and Trp111) or the cofactor NADP+
and the inhibitor were calculated by Discover. Predic-
tions of partition coefficients (ClogP) and polar surface
areas (PSA) were made by relevant fragment-based
methods.71,72
Acknowledgments
25. Stefek, M.; Krizanova, L.; Trnkova, Z. Life Sci. 1999, 65,
1995.
26. Stefek, M.; Sotnikova, R.; Okruhlicova, L.; Volkovova,
K.; Kucharska, J.; Gajdosik, A.; Gajdosikova, A.; Mihal-
ova, D.; Hozova, R.; Tribulova, N.; Gvozdjakova, A.
Acta Diabetol. 2000, 37, 111.
27. Ulusu, N. N.; Sahilli, M.; Avci, A.; Canbolat, O.;
Ozansoy, G.; Ari, N.; Bali, M.; Stefek, M.; Stolc, S.;
Gajdosik, A.; Karasu, C. Neurochem. Res. 2003, 28, 815.
28. Stefek, M.; Gajdosik, A.; Tribulova, N.; Navarova, J.;
Volkovova, K.; Weismann, P.; Gajdosikova, A.; Drimal,
J.; Mihalova, D. Methods Find. Exp. Clin. Pharmacol.
2002, 24, 565.
This work was supported by VEGA Grants No. 2/5005/
25, 2/7074/27, VEGA M.S. SR No. 4293/07, APVT
Grant No. 20-020802, APVV No. 51-017905, COST-
B35 and 6th FP-NoE (Dr. Durisova). We are thankful
to Dr. Anna Tsantili-Kakoulidou, Ph.D. for the permis-
sion to use the Pallas program.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmc.
2008.03.039.
29. Kyselova, Z.; Gajdosik, A.; Gajdosikova, A.; Ulicna, O.;
Mihalova, D.; Karasu, C.; Stefek, M. Mol. Vis. 2005, 11, 56.
¨
30. Yulek, F.; Karasu, C.; Or, M.; Ozog˘ul, C.; Ceylan, A.;
¨
Stefek, M. Arch. Med. Res. 2007, 38, 503.
31. Gunes, A.; Ceylan, A.; Sarioglu, Y.; Stefek, M.; Bauer, V.;
Karasu, C. Fundam. Clin. Pharmacol. 2005, 19, 73.
32. Pekiner, B.; Ulusu, N. N.; Das-Evcimen, N.; Sahilli, M.;
Aktan, F.; Stefek, M.; Stolc, S.; Karasu, C. Biochim.
Biophys. Acta 2002, 1588, 71.
33. Demiryurek, A. T.; Karasu, C.; Stefek, M.; Stolc, S.
Pharmacology 2004, 70, 1.
34. DeRuiter, J.; Borne, R. F.; Mayfield, C. A. J. Med. Chem.
1989, 32, 145.
References and notes
1. Yabe-Nishimura, C. Pharmacol. Rev. 1998, 50, 21.
2. Oates, P. J. Int. Rev. Neurobiol. 2002, 50, 325.
3. Kador, P. F.; Lee, J. W.; Fujisawa, S.; Blessing, K.; Lou,
M. F. J. Ocul. Pharmacol. Ther. 2000, 16, 149.
4. El-Kabbani, O.; Ruiz, F.; Darmanin, C.; Chung, R. P.
Cell. Mol. Life Sci. 2004, 61, 750.
35. DeRuiter, J.; Mayfield, C. A. Biochem. Pharmacol. 1990,
40, 2219.
36. Haraguchi, H.; Hayashi, R.; Ishizu, T.; Yagi, A. Planta
Med. 2003, 69, 853.
37. Barski, O. A.; Gabbay, K. H.; Grimshaw, C. E.; Bohren,
K. M. Biochemistry 1995, 34, 11264.
38. Rees-Milton, K. J.; Jia, Z.; Green, N. C.; Bhatia, M.; El-
Kabbani, O.; Flynn, T. G. Arch. Biochem. Biophys. 1998,
355, 137.
5. Obrosova, I. G. Antioxid. Redox Signal. 2005, 7, 1543.
6. Hotta, N. Biomed. Pharmacother. 1995, 5, 244.
7. Miyamoto, S. Expert Opin. Ther. Patents 2002, 12, 621.
8. Costantino, L.; Rastelli, G.; Gamberini, M. C.; Barlocco,
D. Expert Opin. Ther. Patents 2000, 10, 1245.
9. Srivastava, S. K.; Ramana, K. V.; Bhatnagar, A. Endocr.
Rev. 2005, 26, 380.
10. Costantino, L.; Rastelli, G.; Vianello, P.; Cignarella, G.;
Barlocco, D. Med. Res. Rev. 1999, 9, 3.
39. Steenken, S.; Sunquist, A. R.; Jovanovic, S. V.; Crockett,
R.; Sies, H. Chem. Res. Toxicol. 1992, 5, 355.
11. Schlitzer, M.; Rodriguez, L.; Kador, P. F. J. Pharm.
Pharmacol. 2001, 53, 831.

4920
M. Stefek et al. / Bioorg. Med. Chem. 16 (2008) 4908–4920
40. Stefek, M.; Masarykova, M.; Benes, L. Pharmacol. Tox-
icol. 1992, 70, 407.
57. Mihaljevic, B.; Katusin-Razem, B.; Razem, B. Free Radic.
Biol. Med. 1996, 21, 53.
41. Rackova, L.; Stefek, M.; Majekova, M. Redox Rep. 2002,
7, 207.
42. Blois, M. S. Nature 1958, 181, 1199.
43. Rackova, L.; Snirc, V.; Majekova, M.; Majek, P.; Stefek,
M. J. Med. Chem. 2006, 49, 2543.
44. Pagliara, A.; Carrupt, P. A.; Caron, G.; Gaillard, P.;
Testa, B. Chem. Rev. 1997, 97, 3385.
45. Kador, P. F.; Kinoshita, J. H.; Sharpless, N. E. J. Med.
Chem. 1985, 28, 841.
46. Buyukbingol, E. J. Math. Chem. 1991, 8, 195.
47. Ishida, T.; In, Y.; Tanaka, C.; Inoue, M. Acta Cryst. 1991,
B47, 806.
48. Nicolaou, I.; Demopoulos, V. J. J. Med. Chem. 2003, 46,
417.
49. El-Kabbani, O.; Podjarny, A. Cell. Mol. Life Sci. 2007, 64,
1970.
50. Kelder, J.; Grootenhuis, P. D.; Bayada, D. M.; Delbres-
sine, L. P.; Ploemen, J. P. Pharm. Res. 1999, 16, 1514.
51. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P.
J. Adv. Drug Deliv. Rev. 2001, 46, 3.
58. Bates, R. G. Determination of pH: Theory and Practice,
2nd ed.; John Wiley & Sons: New York, 1973.
59. Benet, L. Z.; Goyan, J. E. J. Pharm. Sci. 1967, 56,
665.
60. Takacs-Novak, K.; Box, K. J.; Avdeef, A. Int. J. Pharm.
1997, 151, 235.
61. Geiger, P. J.; Bessman, S. P. Anal. Biochem. 1972, 49,
467.
62. Hayman, S.; Kinoshita, J. H. J. Biol. Chem. 1965, 240,
877.
63. Da Settimo, F.; Primofiore, G.; La Motta, C.; Salerno, S.;
Novellino, E.; Greco, G.; Lavecchia, A.; Laneri, S.;
Boldrini, E. Bioorg. Med. Chem. 2005, 13, 491.
64. Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J.
J. P. J. Am. Chem. Soc. 1985, 107, 3902.
65. SPARTAN SGI Version 5.1.3 OpenGL, Wavefunction,
Inc., 18401 Von Karman Avenue, Suite 370, Irvine, CA
92612, U.S.A.
66. El-Kabbani, O.; Darmanin, C.; Schneider, T. R.; Haze-
mann, I.; Ruiz, F.; Oka, M.; Joachimiak, A.; Schulze-
Briese, C.; Tomizaki, T.; Mitschler, A.; Podjarny, A.
Proteins 2004, 55, 805.
52. Remko, M.; Swart, M.; Bickelhaupt, F. M. Bioorg. Med.
Chem. 2006, 14, 1715.
53. Stolc, S.; Bauer, V.; Benes, L.; Tichy, M. Czech. Patent
229067, 1983.
67. DOCK 3.5, http://dock.compbio.ucsf.edu.
68. El-Kabbani, O.; Carbone, V.; Darmanin, C.; Mitzuru, O.;
Mitschler, A.; Podjarny, A.; Schulze-Briese, C.; Chung, R.
P.-T. J. Med. Chem. 2005, 48, 5536.
69. Darmanin, C.; El-Kabbani, O. Bioorg. Med. Chem. Lett.
2000, 10, 1101.
70. Darmanin, C.; El-Kabbani, O. Bioorg. Med. Chem. Lett.
2001, 11, 3133.
71. Pallas software. Version 3.1.1.2, CompuDrug Interna-
tional, 115 Morgan Dr., Sedona, AZ 86351, U.S.A.
72. http://www.molinspiration.com.
54. Stolc, S.; Povazanec, F.; Bauer, V.; Majekova, M.; Wilcox,
A.L.; Snirc, V.; Rackova, L.; Sotnikova, R.; Stefek, M.;
Gasparova-Kvaltinova, Z.; Gajdosikova, A.; Mihalova,
D. Slovak Patent Registration PP 1321, 2003.
55. Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R.
Purification of Laboratory Chemicals; Pergamon Press:
Oxford, 1980.
56. Sangster, J.. In Octanol–Water Partition Coefficients:
Fundamentals and Physical Chemistry; John Willey &
Sons: England, 1997; vol. 2.
73. VEGA ZZ 2.1.0 software, http://www.ddl.unimi.it/vega.