PASS-predicted design, synthesis and biological evaluation of cyclic nitrones as nootropics

Article abstract of DOI:10.1016/j.bmcl.2007.06.071

Out of 400 virtually designed imidazoline N-oxides, five cyclic nitrones were selected on the basis of PASS prediction as potent nootropics and were evaluated for their biological activities in albino mice. The selected N-alkyl and aryl-substituted nitron

Full text of DOI:10.1016/j.bmcl.2007.06.071

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5251–5255
PASS-predicted design, synthesis and biological evaluation
of cyclic nitrones as nootropics
Alka Marwaha,a R. K. Goelb,* and Mohinder P. Mahajana,*
aDepartment of Applied Chemistry, Guru Nanak Dev University, Amritsar 143005, Punjab, India
bDepartment of Pharmaceutical Sciences and Drug Research, Punjabi University, Patiala, India
Received 13 December 2006; revised 25 June 2007; accepted 27 June 2007
Available online 30 June 2007
Abstract—Out of 400 virtually designed imidazoline N-oxides, five cyclic nitrones were selected on the basis of PASS prediction as
potent nootropics and were evaluated for their biological activities in albino mice. The selected N-alkyl and aryl-substituted nitrones
were found to be excellent nootropics. A series of lead compounds acting as cognition enhancers have been provided, which can be
further exploited in search of such New Chemical Entities (NCEs).
ꢀ 2007 Elsevier Ltd. All rights reserved.
One of the major challenges of the third millennium will
be to restore normal brain functions in individuals suffer-
ing from neurodegenerative diseases like Alzheimer’s dis-
ease, Parkinson’s disease or from cognitive dysfunction,
one of the main symptoms accompanying ageing, stroke,
acute as well as chronic cerebrovascular diseases, inflam-
mation, traumatic brain injury, intoxication and oth-
ers.1–4 Nootropics, also known as smart drugs or
cognition activators, are the drugs that enhance mental
functions especially under conditions of disturbed neural
metabolism resulting from the lack of oxygen, electro-
shock or age-related induced changes and facilitate infor-
mation flow between cerebral hemispheres. This category
of drugs is becoming especially critical on account of the
rapidly growing segment of the victimized population.
involve inhibition of Nitric Oxide Synthase (iNOS)
expression, as well as cytokine production, modulation
of calcium homeostasis, glutamate regulation and apop-
tosis.5a The piracetam group of antiamnesic compounds
works by several mechanisms to revitalise neural func-
tion. By supplying glutamic acid analogues to the Kreb’s
cycle they enhance glucose utilization in aerobic respira-
tion, the major means by which animal cells extract
chemical energy from sugars via ATP formation. This,
in turn, raises phospholipid cAMP levels, thereby
enhancing the function of dopamine and acetylcholine
neurons.
O
N
NH2
Antioxidants are gaining a paramount significance as a
panacea for a large number of life-style diseases like age-
ing, cancer, diabetes, cardiovascular and other degener-
ative diseases, etc. Nitrones, by acting as excellent spin
traps, have been known to oppose the oxidative damage
and the associated challenges.5 However, the mechanism
with which nitrones operate as antioxidants does not lie
solely in their spin-trapping tendency. Other mecha-
nisms of action, well documented in the literature,
O
Piracetam - the original smart drug
2-Oxo-1-pyrrolidine
Piracetam like nootropics revert amnesia induced by
scopolamine and other amnesic drugs, electroconvulsant
shocks and hypoxia by an unknown mechanism. They
also function as antioxidants (structure comparable to
vitamin C) and retard lipofuscin formation. But none
of the mechanisms has, so far, gathered general consen-
sus. Thus, years after the identification of piracetam, a
unifying hypothesis on the mechanism of action is till
lacking and this class of cognition enhancers, despite
Keywords: Nootropics; ANOVA; Imidazoline N-oxides; Lipinski’s rule
of five; PASS.
*
Corresponding authors. Tel.: +91 1832258802 09 (M.P.M.); e-mail:
mahajanmohinderp@yahoo.co.in
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.06.071

5252
A. Marwaha et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5251–5255
chemical similarity and close pharmacological profile,
uses an amazing variety of molecular mechanisms to
produce the final nootropic effect. The excellent cogni-
tive properties6 of piracetam, its neuropatholobiology
and the known antioxidant properties of nitrones pro-
vided us an impetus for the design, synthesis and biolog-
ical evaluation of the nitrones (structurally related to
piracetam), endowed with a similar pharmacological
profile. The chemical and biochemical properties of nit-
rones as well as their brain penetration and toxicity are
tremendously influenced by the substitution pattern of
these molecules.
used in a flexible manner—(i) only activities with Pa > Pi
are considered as possible for a particular compound.
(ii) If Pa > 0.7—the chance to find the activity experi-
mentally is high. But, in many cases the compound
may occur to be a close analogue of known pharmaceu-
tical agents. (iii) If 0.5 < Pa < 0.7—the chance to find the
activity experimentally is less, but the compound is
probably not so similar to known pharmaceutical
agents. (iv) If Pa < 0.5—the chance to find the activity
experimentally is even less, but the chance to find a
structurally new compound, that is, NCEs increases.
Recently, an Internet version of PASS7c has been made
available at the PASS developer’s web site. The user can
submit the MOL-file of the molecule under study and
obtain the predicted biological activity spectrum on
their computer immediately. This new internet version
of PASS provides access to prediction of 783 kinds of
biological activity, in contrast to an earlier version that
predicted 319 activities.
In an effort to optimize their biological profile, a wide
structural diversity of nitrones typified by 3, possessing
p-nitro phenyl moiety as a silent feature of their chemi-
cal structure, have been virtually designed, synthesized
and investigated for the foresaid pharmacological activ-
ity. To the best of our knowledge, there are no reports
concerning the cognition-enhancing activity of such nit-
rones. The compounds chosen for activity determination
were selected on the basis of their best prediction values
for crossing the blood–brain barrier from each set of the
virtually screened compounds as well as ease of their
synthesis. The design was initiated with unsubstituted
nitrones followed by the incorporation of various sub-
stituents on the nitrogen as well as on the carbon atom
of the Schiff’s bases and also on the carbon atom of
nitrosoalkenes from which they are synthesized. In order
to accelerate search for potent New Chemical Entities
(NCEs), the assistance of computer-aided drug discov-
ery program PASS (Prediction of Biological Activity
Spectra)7 was used to predict the cognition-enhancing
action for different imidazoline N-oxides.
Selection of possible cognition enhancers. Prediction of
biological activity spectra was made for about 400 virtu-
ally designed variedly substituted structures from the
theoretical calculations on the basis of PASS prediction.
The compounds screened were the structures formed
from the various combinations of R (H, CH3, C2H5,
C6H5, p-CH3AC6H4, p-OCH3AC6H4, C6H5CH2, cyclo-
hexyl, iso-propyl, n-butyl, furyl), R1 (H, CH3, C2H5,
C6H5, p-NMe2AC6H4) and R2 (H, C6H5 p-CH3AC6H4,
p-NO2AC6H4) groups in the nitrone moiety. On the
basis of prediction results from the database analysis,
potential cognition enhancers were selected. The follow-
ing criteria were used for selection:
Contrary to many other existing methods of SAR/
QSAR/QSPR/molecular modelling methods focused on
predicting a single type of biological activity within the
same chemical series, computer-aided program PASS
predicts not only for the desirable pharmacological ef-
fect but also for molecular mechanisms of action and
different unwanted side effects like mutagenicity, carcin-
ogenicity, teratogenicity and embryotoxicity. Such anal-
ysis of heterogeneous sets increases considerably the
chance of discovering NCEs (e.g., cognition enhancers).
The technique of PASS is based on the analysis of SARs
for the training set currently including about 46,000
drugs, drug candidates and lead compounds whose bio-
logical activities are determined experimentally. The set
of MNA (Multilevel Neighbourhood atoms) descriptors
is generated on the basis of structural formulas pre-
sented in the MOL-file (SDF-file) form. Since MNA
descriptors are generated for each compound de novo,
new descriptors can be obtained upon presentation of
a novel structural feature in the compound under study.
Based on the statistics of MNA descriptors for active
and inactive compounds from the training set, two prob-
abilities are calculated for each activity: Pa—probability
of compound being active and Pi—probability of com-
pound being inactive. Being probabilities, the Pa and
Pi values vary from 0.000 to 1.000 and in general
Pa + Pi < 1, since these probabilities are calculated inde-
pendently. The PASS predictions can be interpreted and
1. Those compounds were selected, which had cognition
enhancing activity with Pa > 0.4 (intended to find
NCEs) in their predicted activity spectra.
2. In case a number of structures in a certain chemical
series were predicted to be active, only a few represen-
tatives were selected.
The relationship between the number of compounds
predicted as probable nootropics and representative val-
ues of probability to be active Pa is shown in Figure 1.
Chart showing the predicted nootropic activity of various
compounds
450
390
372
400
350
318
305
300
250
182
200
146
150
100
50
22
0
Pa>Pi
Pa>30
Pa>40
Pa>50
Pa>60
Pa>70
Pa>80
Predicted nootropic activity (Pa)
Figure 1. The number of compounds predicted as probable cognition
enhancers versus calculated probability to be active Pa.

A. Marwaha et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5251–5255
5253
Out of the 400 compounds chosen for PASS prediction,
R
N
Hc
Ha
R1
H
318 from were predicted as cognition enhancers with
Pa > 40%. On the basis of the above-mentioned criteria,
five potential nitrones were selected for testing their noo-
tropic activities. The compounds screened above were
further subjected to ‘Lipinski rules of five’.8 The MIPC
(Molnspiration Property Calculator) program has been
utilized (www.molinspiration.com) for calculating the
Lipinski descriptors. The logP values and the associated
parameters for the five compounds under consideration
have been provided in Table 1 and the probabilities of
the test compounds for being active (Pa) are provided
in Table 2. On the basis of above-mentioned criteria
and with a goal to provide the diversity of potent cogni-
tion enhancers, the chosen compounds were synthesized
and experimentally tested as nootropics.
H
N
H
dry DCM
stir, r.t.
Hb
R1
+
N
N
R2
O
R2
R
O
3
2
1
A. R = C6H4-OCH3(p); R1 = C6H4- NMe2(p)
B. R = Ph; R1 = Ph
C. R = Cyclohexyl; R1 = Ph
D. R = Furyl; R1 = Ph
E. R = Benzyl; R1 = C6H4- NMe2(p)
R2 = C6H4-NO2(p)
Scheme 1.
of Guru Nanak Dev University, had free access to food
and water and maintained under 12:12 h light and dark
cycles. All experiments were carried out during day time
from 09:00 to 14:00 h. Institutional Animals Ethics
Committee (IAEC) approved the experimental protocol
and care of animals was taken as per guidelines of CPC-
SEA, Dept. of Animal Welfare, Govt. of India.
General procedure for the synthesis of nitrones 3. The
acyclic aryl and alkyl imines (Schiff’s bases) 1 derived
from the condensation of aryl/alkyl amines and alde-
hyde, reacted with a-nitrosoalkenes 2 (1.2 equiv), gener-
ated in situ from a-bromooxime and sodium
bicarbonate, in dry dichloromethane at room tempera-
ture under stirring conditions for 24–28 h to give the de-
sired nitrones 3 (Scheme 1), characterized as 1-alkyl/
aryl-4-(4-nitrophenyl)-2-aryl-2,5-dihydro-1H-imidazole
3-oxide on the basis of their spectral and analytical
data.9a
Drugs and chemicals. Scopolamine hydrobromide (Sig-
ma–Aldrich, USA) piracetam (Nootropilꢁ, UCB India
Pvt. Ltd, Vapi, Gujarat) were suspended in 1% CMC
solution. Volume of oral and ip administration was
1 ml/100 g.
This methodology is a facile and direct itinerary towards
the synthesis of novel and biologically potent heterocy-
clic N-oxides containing an a-hetero atom.9b
Behavioural paradigm to evaluate learning and memory—
Elevated plus maze. The elevated plus maze served as the
exteroceptive behavioural model (wherein the stimulus
existed outside the body) to evaluate learning and mem-
ory in mice. The apparatus consisted of two open arms
(16 · 5 cm) and two covered arms (16 · 5 · 12 cm). The
arms extended from a central platform (5 · 5 cm), and
maze was elevated to a height of 25 cm from the floor.
On the first day, each mouse was placed at the end of
open arm, facing away from the central platform. Trans-
fer latency (TL) was taken as the time taken by mouse to
move into one of the covered arms with all its four legs.
TL was recorded on the first day. If the animal did not
enter into one of the covered arms within 90 s, it was
gently pushed into one of the two covered arms and
the TL was assigned as 90 s. The mouse was allowed
to explore the maze for 10 s and then returned to its
home cage. Memory retention was examined 24 h after
the first day trial on the second day.10
Animals. Albino mice of either sex weighing around 18–
25 g (older ones, aged 28 weeks) were used in the present
study. The animals were housed in central animal house
Table 1. LogP values and the associated parameters for the com-
pounds under consideration
MlogP
TPSA
N atoms
MW
A
B
C
D
E
2.32
84.767
72.295
72.295
85.435
75.533
32
27
27
27
31
433.49
360.39
366.44
364.38
417.49
2.449
2.411
1.341
2.019
nON
nOHNH
nViolations
Nrotb
A
B
C
D
E
8
6
6
7
7
1
1
1
1
1
0
0
0
0
0
6
4
4
5
6
Group I: represented control group for young mice
(n = 6). 10 ml/kg of distilled water (DW), po, was
administered. TL was noted after 30 min of administra-
tion on and after 24 h, that is, on next day.
Table 2. The probabilities of being active Pa of the test compounds on
the basis of PASS prediction
Group II: piracetam, 200 mg/kg, ip, was injected to both
young and aged mice, respectively. TL was noted after
30 min of injection and after 24 h.
Compound
Pa
Pi
A
B
C
D
E
50.1
69.2
51.8
40.9
56.7
11.4
3.3
9.9
Groups III–VII: test compounds (A–E) were adminis-
tered orally at a dose of 75 mg/kg, except test compound
D at a dose of 90 mg/kg, 30 min before subjecting the
19.3
7.0

5254
A. Marwaha et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5251–5255
Table 3. Effect of transfer latency (TL)
results of this chemical series. The compound E, the only
compound with non-hydrogen substituents at R1 and
R2, shows significantly less activity compared to other
compounds A–D. This may be perhaps because the
addition makes the region too bulky for efficient
scavenging to occur. Although the activity under study
is whole-animal behaviour, so many other properties
of the compounds could also be responsible for differ-
ences in observed activity, for example, susceptibility
of particular functional groups to metabolism, second-
ary effects from certain groups, etc. Since, the mecha-
nism of action (MOA) of nitrones for their nootropic
activity is well documented in their abilities to scavenge
free radicals, the underlying phenomena of mechanism
are thought to be influenced by electronic nature of
the substituents on the nitrone moiety. However, the
detailed studies are still underway to establish the exact
MOA.
S. No.
Treatment
group
Dose
(mg/kg)
n
% Age
decrease in TL
1
2
3
4
5
6
7
Control
Piracetam
—
200
75
75
75
90
75
6
6
6
6
6
6
6
11.2 + 2.78
51.3 + 6.82*
40.4 + 8.38*
36.4 + 7.77*
51.6 + 9.28*
64.2 + 6.66*
10.8 + 5.66
A
B
C
D
E
* P < 0.05 as compared to control.
animals to elevated plus maze test. TL was noted again
after 24 h.
The dose of the test compounds was selected as molecu-
lar weight equivalents to piracetam.
Statistical analysis. All the results were expressed as
means ( SEM). The data from elevated plus maze and
passive avoidance tasks were analyzed using ANOVA
followed by Student’s (Unpaired) ‘t’ test.10 Utility of
an elevated plus maze for the evaluation of nootropics,
scopolamine and electroconvulsive shock,11 Ascorbic
acid: a promising memory enhancer in mice.12
Thus, innovative computer-assisted approaches have
been applied in a search for new cognition enhancers.
We have used virtual combinatorial design of highly di-
verse chemical compounds to increase the probability of
finding new chemical entities. Different types of imidaz-
oline N-oxides were analyzed during the project. The
most likely compounds (presumably NCEs) were se-
lected, synthesized and tested as potential cognition
enhancers. Five compounds from the library of 400
compounds (predicted as nootropics) were selected as
potential cognition enhancers, out of which, three com-
pounds have comparable or greater effect in consider-
ably less concentration in comparison with the classic
cognition enhancer, piracetam. Thus we have provided
a novel series of lead compounds acting as cognition
enhancers, which can further be expanded a lot to ex-
ploit such NCEs. It is anticipated that these drugs when
taken in large concentrations can further improve their
nootropic activities. Thus, our investigation has shown
an increased probability of compounds to be biologi-
cally active in the subset of compounds selected on the
basis of PASS prediction.
Effect of test compounds on facilitation of passive behav-
iour. The mice showed higher transfer latency (TL) val-
ues on first day as compared to second day (after 24 h),
piracetam (200 mg/kg, ip) pretreatment significantly de-
creased transfer latency on as compared to control
group, indicating improvement in both learning and
memory. The pretreatment with test compounds A–E
significantly reduced the TL as compared to control
group. Piracetam, A, B, C, and D are all significantly
different than control. Compounds A and C are within
the standard deviations of piracetam. However A
showed lower % TL as compared to C, although still
within the SEM. The compound C showed comparable
to and compound D exhibited a better effect than pirace-
tam (Table 3).
Structure–activity relationship. The resulting biological
activities of the compounds studied are the consequence
of both pharmacokinetic and pharmacodynamic proper-
ties that may be differently affected by structural
modifications indicating that there is a strict correlation
between potency and structural as well as conforma-
tional characteristics of the molecules. Comparison of
cognitive activities of the test compounds A–E (Table
3) shows that the electronic nature of the substituents
present on N1 (i.e., alkyl and aryl) and C2-atoms could
be responsible for the potency of the compounds and
this is substantiated by the known scavenging mecha-
nism of the nootropics. Electron-withdrawing nature
on C-2 position has been thought to be responsible for
the better activities of C and D compared to A and E
(p-methoxyphenyl group in A and E is electron-donating
relative to bare phenyl in C and D). Further N-alkyl
substituted nitrones show much better activities in our
results compared to N-aryl substituted ones. Thus A,
C and D compounds represent the most significant
References and notes
1. Cacabelos, R.; Alvavrezi, A.; Lombardi, V.; Fernandes-
Novoa, L.; Corzo, L.; Perez, P.; Laredo, M.; Picheli, V.;
Hernandez, A.; Varela, M.; Figueroa, J.; Prous, J., Jr.;
Windisch, M.; Vigo, C. Drug Today 2000, 5, 415.
2. Castro, A.; Conde, S.; Rodriguez-Franco, I.; Martinez, A.
Mini. Rev. Med. Chem. 2002, 2, 37.
3. Voronica, T. A. In Alzheimer’s Disease: Therapeutic
Strategies; Birkhauser: Boston, 1994; p 265.
4. Le Merrer, J.; Nogues, X. Pharmacol. Res. 2000, 41, 503.
5. (a) Dhainaut, A.; Tizot, A.; Raimbaud, E.; Lockhart, B.;
Lestage, P.; Goldstein, S. J. Med. Chem. 2000, 43, 2165;
(b) Green, A. R.; Ashwood, T.; Odergren, T.; Jackson, D.
M. Pharmacol. Therap. 2003, 100, 195, and references
cited therein; (c) Khakh, B. S.; Henderson, G. J. Autonom.
Nerv. Syst. 2000, 81, 110; (d) Meir, A.; Ginsburg, S.;
Butkevich, A.; Kachalsky, S. G.; Kaiserman, T.
J. Pharmacol. Exp. Ther. 1998, 284, 858.
6. (a) Buccafusco, J. J.; Terry, A. V. J. Pharmacol. Exp. Ther.
2000, 295, 438; (b) Guirgea, C.; Moeyersoons, F. E.;

A. Marwaha et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5251–5255
5255
Evraerd, A. C. Arch. Int. Pharmacodyn. Ther. 1967, 166,
238.
OCH3), 4.79 (dd, J = 14.1 and 3.1 Hz, 1H, ACH2), 5.15
(dd, J = 14.1 and 5.4 Hz, 1H, ACH2), 6.08 (dd, J = 3.1
and 5.4 Hz, 1H, methine), 6.59 (d, J = 8.8 Hz, 2H, ArH),
6.70 (d, J = 8.7 Hz, 2H, ArH), 6.81 (d, J = 8.8 Hz, 2H,
ArH), 7.44 (d, J = 8.7 Hz, 2H, ArH), 8.29 (d, J = 8.8 Hz,
2H, ArH), 8.49 (d, J = 8.8 Hz, 2H, ArH); 13C NMR: 40.3
(AN(CH3)2), 52.7 (ACH2), 55.5 (AOCH3), 90.7 (methine),
112.1, 113.7, 114.9, 122.2, 123.9, 127.2, 128.8, 132.9, 135.2,
138.3, 147.8, 151.7 and 152.8; MS: m/z 432 (M+) and 416
(M+À16); Anal. Calcd for C24H24N4O4: C, 66.69; H, 5.59;
N, 12.96; found: C, 66.63; H, 5.52; N, 12.99; (b) Marwaha,
A.; Singh, P.; Mahajan, M. P. Tetrahedron 2006, 62, 5477.
10. Itoh, J.; Nabeshima, T. and Kameyama, T.; 1990, 12.
11. Milind, P.; Dhingra, D. Psychopharmacology 2003, 101,
27.
7. (a) Poroikov, V. V.; Filimonov, D. A.; Borodina, Yu. V.;
Lagunin, A. A.; Kos, A. J. Chem. Inf. Comput. Sci. 2000,
40, 1349; (b) Poroikov, V. V.; Filimonov, D. A.; Ihlenfeld,
W.-D.; Gloriozova, T. A.; Lagunin, A. A.; Borodina, Yu.
V.; Stepanchikova, A. V.; Nicklaus, M. C. J. Chem. Inf.
Comput. Sci. 2003, 43, 228; (c) Website: http://
www.ibmh.msk.su/PASS.
8. Lipinski, C. A.; Lombardo, F.; Dominy, D. W.; Feeney,
P. J. Adv. Drug Deliv. Rev. 2001, 46, 3.
9. (a) {4-[1-(4-Methoxy-phenyl)-4-(4-nitrophenyl)-3-oxy-2,5-
dihydro-1H-imidazol-2-yl]-phenyl}-dimethyl-amine (3a).
Red crystalline solid; yield 88%; mp: 166–167 ꢂC; mmax
/
cmÀ1 (KBr): 1220 (NAO), 1542 and 1591 (C@N); 1H
NMR (200 MHz): d 2.96 [(s, 6H, AN(CH3)2], 3.73 (s, 3H,
12. Milind, P.; Dhingra, D. J. Pharmacol. Sci. 2003, 93, 129.