Development of selective and reversible pyrazoline based MAO-B inhibitors: Virtual screening, synthesis and biological evaluation

Article abstract of DOI:10.1016/j.bmcl.2011.02.030

In an effort to develop selective MAO (monoamine oxidase) B inhibitors, structure based virtual screening was initiated on an in-house library. Top 10 HITS were synthesized and evaluated for MAO (A and B) inhibitory activity, both against human and rat en

Full text of DOI:10.1016/j.bmcl.2011.02.030

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1969–1973
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Development of selective and reversible pyrazoline based MAO-B
inhibitors: Virtual screening, synthesis and biological evaluation
⇑
Nibha Mishra , D. Sasmal
Department of Pharmaceutical Sciences, Birla Institute of Technology, Mesra, Ranchi 835 215, India
a r t i c l e i n f o
a b s t r a c t
Article history:
In an effort to develop selective MAO (monoamine oxidase) B inhibitors, structure based virtual screening
was initiated on an in-house library. Top 10 HITS were synthesized and evaluated for MAO (A and B)
inhibitory activity, both against human and rat enzymes. All the compounds were found selective,
reversible and active in nM range (100 times more potent than selegeline) towards MAO-B. Outstanding
co-relation between predicted and experimental Ki values were observed.
Received 19 December 2010
Accepted 9 February 2011
Available online 13 February 2011
Keywords:
Pyrazoline
Ó 2011 Elsevier Ltd. All rights reserved.
Monoamine oxidase
Virtual screening
Human monoamine oxidases A and B (MAO-A and MAO-B)
modulate the intracellular levels of arylalkylamines such as dopa-
mine and serotonin by catalyzing their oxidative deamination with
the concomitant production of hydrogen peroxide.1 Impairment in
the catabolism of neurotransmitters and oxidative damage are
important factors in the physiology of neurodegenerative disorders
such as Alzheimer’s and Parkinson’s diseases.2 For this reason,
MAO-B inhibition represents one of the strategies to alleviate
symptoms of patients suffering from these pathological conditions.
There have been considerable efforts in the last few years to devel-
op reversible and selective MAO inhibitors to be used as neuropro-
tective agent. Selective inhibitors are also required to avoid
unwanted side effect so called cheese effect.3
A focused combinatorial library of pyrazolines was enumerated
with SmiLib. The library was constructed as follows: Twenty differ-
ent acetophenones and 25 different benzaldehydes were initially
considered, to generate 20 Â 25 = 500 chalcones. These chalcones
were connected with as many as 36 linkers in all possible combi-
nations to generate 500 Â 36 = 18000 molecules. Altogether we
had 18000 + 500 = 18500 molecules (Fig. 1). The generated smiles
structures of the molecules were converted to sdf through SmiLib
v 2.0 program.7 They were then processed with the LigPrep
program to assign protonation states appropriate for pH 7.0.
Conformer generation was carried out with the MacroModel.
Potentials were computed using the OPLS2005 force field. After
these ligands were prepared they were saved individually in pdb
format.
In our earlier studies we employed structure based docking
with AUTODOCK 4.0 for analyzing the binding mode. One of the inter-
esting observations from those reports was, an outstanding corre-
lation was observed between the predicted and experimental
activities. More interestingly, this correlation was equally good
for both the isotypes (MAO-A and MAO-B). Therefore, we have a
strong basis to select the same computational method for VS
purpose in the current work.
Recently from our laboratory we have reported4–6 a few pyraz-
olines with selective and reversible MAO inhibitory property. How-
ever, majority of those reported compounds are selective MAO-A
inhibitors. So far, we were unsuccessful in developing any promis-
ing selective MAO-B inhibitors. The current study is therefore, ini-
tiated with the objective to develop potent, selective and reversible
MAO-B inhibitors. Our study is designed as follows:
1. A focused combinatorial library was constructed with the avail-
able synthetically feasible fragments.
2. The generated library of compound was virtually screened.
3. Top 10 hits were identified and synthesized.
4. The synthesized compounds were evaluated for MAO (A and B)
inhibitory activity using both human and rat enzymes.
For VS we have used AUTODOCK 4.0,8 the entire VS process was
automated by an in house program recently reported by our
group.9 Docking Parameters used, are as follows: Number of
Genetic Algorithm (GA) runs: 10, Population size: 150, Maximum
number of evaluation: 2500000, Maximum number of generation:
27000.
After structure based virtual screening top 10 HITS with
superior selectivity towards MAO-B were identified. The structures
of the identified HITS are given in Table 1. Surprisingly, all the
identified HITS were 3,5-diaryl pyrazolines of 9-anthracene
⇑
Corresponding author.
E-mail address: nmishra@bitmesra.ac.in (N. Mishra).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.02.030

1970
N. Mishra, D. Sasmal / Bioorg. Med. Chem. Lett. 21 (2011) 1969–1973
H
O
O
R1
H3C
R2
25 benzaldehhydes
20 acetophenones
O
R2
R1
500 chalcones
36 LINKERS
R2
R2
R2
R2
NH
H
N
N
N
N
R1
N
N
R4
R1
N
N
R3
R1
X
SO2
O
R1
R4 = Aryl,Alkyl
X=S,O,NH
R3=Aryl,Alkyl
R5
R1,R2 and R5 are different substitution on the benzene ring.
500x36 = 18000 molecules
Figure 1. Construction of the in-house library for structure based virtual screening.
carboxaldehyde. In-order to gain more insight into their binding
modes molecular docking results was scrupulously analyzed.
The active site architecture of MAO-B10 is characterized by two
large pockets. Pocket1, large aromatic cage made up of FAD, PHE
343 and four tyrosine residues TYR 188,189,435 and 398. Pocket2,
a hydrophobic pocket characterized by LEU 171, TYR 326, PHE 168,
ILE 198 and 199. All the top 10 HITS with a bulkier anthracene ring
perfectly fit to Pocket1 (Fig. 2a and b). The other aromatic ring also
accommodates well in the other hydrophobic pocket, Pocket2.
Surprisingly these top scoring molecules does not show any hydro-
For MAO-A this loop deviates to almost 3 Å, making the active site
more accommodative. For MAO-B, however, this loop is com-
pressed, spanning more inside the active site, making it sterically
impossible for any other group to accommodate itself.
The identified HITS were synthesized and characterized.12 The
compounds were synthesized according to previously reported
method,4–6 shown in Scheme 1. Briefly chalcones 1–10 were pre-
pared through Claisen–Schmidt condensation of different aceto-
phenone with anthracene-9-carboxaldehyde. Compounds 11–20
were synthesized by the reaction of excess hydrazine hydrate
(99%) with 1–10, respectively in ethanol. Reactions were moni-
tored by thin-layer chromatography on silica gel plates in either
iodine or UV chambers. The intermediates were characterized by
elemental analysis for CHN. In the elemental analysis, the observed
values were within 0.4% of the calculated values. Final com-
pounds were characterized by 1H NMR and FAB mass spectrometry
(MS). The structure, physicochemical and spectral data of these
analogs were given in Table 1.
gen bonding or
p stacking interactions with the receptor. However,
their docking scores are superior >À12.5 with predicted Ki in nano-
molar range.
The identified HITS not only showed a superior affinity towards
MAO-B, but also provided a high MAO-B/A selectivity index.
In order to understand this superior selectivity, structures of
these two subtypes (MAO-A and B) along with their docked ligands
were overlaid (Fig. 2c). This overlaid structure revealed a few
important results. Human MAO-A and MAO-B share
a
high
The synthesized compounds were assayed for MAO (A and B)
inhibitory activity using both rat and human enzymes. For assay
employing rat enzymes, MAO was purified from the rat liver13
and for assay involving human enzymes, recombinant human
MAO-A and B were procured from Sigma–Aldrich Chemical Co.
sequence similarity 72%, however, a significant structural differ-
ence is observed in the active site loop conformation, consisting
of residues 210–216 in hMAO B.10,11 This loop in the active site
was found intrinsic for selectivity towards either of these isotypes.

N. Mishra, D. Sasmal / Bioorg. Med. Chem. Lett. 21 (2011) 1969–1973
1971
R
O
i.
R
H3C
O
H
O
1-10
ii.
R
N
NH
11-20
Scheme 1. Synthetic scheme for the newly synthesized 3-(anthracene-9-yl)-5-aryl
pyrazolines Reagents and conditions: (i) EtOH, NaOH (60%), 0 °C, stirring 2 h, kept at
room temp for 2–4 days; (ii) EtOH, hydrazine hydrate (99%), reflux 1–3 h.
À1 was used to calculate the initial velocity of the reaction. Results
were expressed as nmol hÀ1 mgÀ1. For selective measurement of
rat MAO-A and MAO-B activities, homogenates were incubated
with the substrate p-tyramine following the inhibition of one of
the MAO isoform with selective inhibitors. Human MAO-A and
MAO-B were procured as pure isoforms thus no pre-incubation
with selective inhibitor was required. The compounds were
dissolved in DMSO and used in the concentration range of
1–1000 lM. The compounds were then incubated with purified
MAO at 37 °C for 0–60 min prior to adding to the assay mixture.
Reversibility of the inhibition of MAO by these compounds was
a
10
9
8
MAO-A(Experimental)
7
MAO-A(Predicted)
MAO-B(Experimental)
MAO-B(Predicted)
PKi
6
5
4
3
1
2
3
4
5
6
7
8
9
10
Compounds
b
10
9
MAO-A(Rat)
MAO-A(Human)
MAO-B(Rat)
8
MAO-B(Human)
Figure 2. Structural screenshots of the newly synthesized compounds. (a) Most
active compound 9 docked in MAO-B active site. (b) Top 10 HITS docked in the
MAO-B active site. (c) MAO-A (yellow ribbons) and MAO-B (red ribbons) superim-
posed, most active MAO-B inhibitor 9 docked to the active site of both the subtypes.
7
Pki
6
5
4
3
The MAO activity was measured spectrophotometrically according
to the method of Holt.13 Briefly, assay mixture (vanillinic acid,
4-aminoantipyrine and peroxidase type II in potassium phosphate
buffer (pH 7.6)) was pre-incubated with substrate p-tyramine be-
fore addition of enzyme. The reaction was initiated by the addition
of enzymes and increase in absorbance was monitored at 498 nm
at 37 °C for 60 min. Molar absorption coefficient of 4654 MÀ1 cm
1
2
3
4
5
6
7
8
9
10
Compounds
Figure 3. (a) Plot comparing the experimental and predicted pKi values against
both the isoforms of MAO. (b) Plot comparing the pKi values against both the
isoforms of MAO for human and rat enzymes. ⁄pKi is calculated as negative
logarithm of Ki expressed in moles.

1972
N. Mishra, D. Sasmal / Bioorg. Med. Chem. Lett. 21 (2011) 1969–1973
Table 1
Structure, physicochemical and spectral properties of the newly synthesized pyrazoline derivatives
A
R
B
N
NH
Compound
R
MF
MW
Y (%) MP (°C) 1H NMR (d ppm)
FAB-MS
(m/z)
1
2
3
4
5
6
4-NO2
4-OH
4-Cl
C23H17N3O2 367.4
C23H18N2O 338.4
23H17N2Cl
76
72
178.5
148.3
151.2
158.8
125.6
130.8
3.38 (HA, t), 3.87 (HM, dd, JMA = 17.05 Hz, JMX = 11.4 Hz), 6.32 (HX, dd,
JMX = 9.6 Hz), 7.03–8.6 (13H, m, Ar-H), 10.10 (1H, s, –NH–)
3.63 (HA, t), 3.79 (HM, dd, JMA = 16.95 Hz, JMX = 11.4 Hz), 6.30 (HX, dd,
JMX = 9.5 Hz), 6.83–8.77 (13H, m, Ar-H), 9.99 (1H, s, Ar-OH) 10.12 (1H, s, –NH–)
3.71 (HA, t), 3.77 (HM, dd, JMA = 16.85 Hz, JMX = 11.4 Hz), 5.87 (HX, dd,
JMX = 9.6 Hz), 7.03–8.6 (13H, m, Ar-H), 9.95 (1H, s, –NH–)
2.34 (3H, –CH3, s) 3.66 (HA, t), 3.71 (HM, dd, JMA = 16.65 Hz, JMX = 12.4 Hz),
6.39 (HX, dd, JMX = 9.7 Hz), 7.00–8.74 (13H, m, Ar-H), 8.95 (1H, s, –NH–)
3.67 (HA, t), 3.79 (HM, dd, JMA = 16.66 Hz, JMX = 11.7 Hz), 5.96 (HX,
dd, JMX = 9.5 Hz), 7.06–8.75 (14H, m, Ar-H), 10.23 (1H, s, –NH–)
3.55 (HA, t), 3.91 (HM, dd, JMA = 16.92 Hz, JMX = 11.34 Hz), 6.16 (HX, dd,
JMX = 10.03 Hz), 7.26–8.55 (13H, m, Ar-H), 9.05 (1H, s, Ar-OH), 10.01
(1H, s, Ar-OH), 11.13 (1H, s, –NH–)
368 (M+H)
337 (M+H)
357 (M+H)
337 (M+H)
333 (M+H)
339 (M+H)
C
356.85 90
336.43 80
4-CH3
–H
C24H20N2
C23H18N2
C23H18N2O
332.4
338.4
77
81
2-OH
7
8
2,4 Di-OH
C
23H19N2O2 354.4
73
132.6
100.0
109.7
107.4
3.45 (HA, t), 3.95 (HM, dd, JMA = 16.85 Hz, JMX = 11.4 Hz), 6.16 (HX, dd, JMX = 9.6 Hz), 355 (M+H)
7.26–8.75 (12H, m, Ar-H), 9.03 (1H, s, Ar-OH), 10.33 (1H, s, –NH–)
Pyridine-
2yl
3-NO2
C22H17N3
324.39 70
85
352.43 85
3.50 (HA, t), 3.96 (HM, dd, JMA = 16.66 Hz, JMX = 11.7 Hz), 5.97 (HX, dd,
JMX = 9.41 Hz), 7.05–8.85 (13H, m, Ar-H), 10.53 (1H, s, –NH–)
3.41 (HA, t), 3.77 (HM, dd, JMA = 17.03 Hz, JMX = 11.54 Hz), 6.38 (HX,
dd, JMX = 9.9 Hz), 7.03–8.72 (13H, m, Ar-H), 10.14 (1H, s, –NH–)
2.54 (3H, –OCH3, s) 3.76 (HA, t), 3.87 (HM, dd, JMA = 16.84 Hz, JMX = 11.23 Hz),
6.42 (HX, dd, JMX = 10.07 Hz), 7.04–8.81 (13H, m, Ar-H), 9.98 (1H, s, –NH–)
325 (M+H)
368 (M+H)
353 (M+H)
9
C23H17N3O2 367.4
C24H20N2O
10
4-OCH3
Table 2
Calculated and experimental Ki values corresponding to the inhibition of MAO isoforms by the newly synthesized pyrazoline derivatives
Code
Experimentala (Ki)
MAO-A (nM) MAO-B (nM)
Human Human
17.08
Calculatedb (Ki)
MAO-A (nM) MAO-B (nM)
Inhibition type
Reversibility
MAO inhibitory selectivity
Rat
Rat
1
2
3
4
5
6
7
8
9
20.14
2.30
6.75
2.10
3.80
3.43
5.91
4.77
9.91
0.45
0.60
1.15
7.51
1.80
3.56
4.21
7.35
5.54
5.75
0.31
1.70
11.12
203.19
17.30
167.08
269.97
63.26
77.49
183.90
20.49
268.14
ND
1.71
5.97
1.72
4.45
3.31
3.10
3.18
6.41
0.33
2.51
ND
Competitive
Competitive
Competitive
Competitive
Competitive
Competitive
Competitive
Competitive
Competitive
Competitive
Competitive
Reversible
Reversible
Reversible
Reversible
Reversible
Reversible
Reversible
Reversible
Reversible
Reversible
Reversible
Selective for MAO-B
Selective for MAO-B
Selective for MAO-B
Selective for MAO-B
Selective for MAO-B
Selective for MAO-B
Selective for MAO-B
Selective for MAO-B
Selective for MAO-B
Selective for MAO-B
Selective for MAO-B
385.26
320.13
389.49
396.57
77.61
225.13
281.19
250.27
332.41
69.91
95.65
81.54
241.54
43.90
205.45
32.16
10
SEL
230.3
10566.0
301.11
67250.0
1350
1960
SEL-selegeline, ND-not done.
a
Values were determined from the kinetic experiments in which p-tyramine (substrate) was used at 500
lM to measure MAO-A and 2.5 mM to measure MAO-B. Pargyline
or clorgyline were added at 0.50 M to determine the isoenzymes A and B. Newly synthesized compounds and the known inhibitors were preincubated with the homog-
l
anates for 60 min at 37 °C. Each value represents the mean SEM of three independent experiments.
b
Values obtained through AUTODOCK program.
assessed by dialysis. Kinetic data for interaction of the enzyme
with these compounds were determined using Microsoft Excel
package program. IC50 values were determined from plots of
residual activity percentage, calculated in relation to a sample of
the enzyme treated under the same conditions without inhibitor,
versus inhibitor [I] concentration.
All the compounds were found as extremely potent and selec-
tive towards MAO-B, (Table 2) with at least 100 times more potent
than the positive control selegiline. Compounds 9 and 10 were
most promising amongst all, with Ki equal to 0.31 nM and
1.7 nM, respectively for human MAO-B, and 32.16 nM and
301.11 nM for MAO-A, respectively. Both the compounds 9 and
10 were >100 times more selective for MAO-B.
with the predicted one and how much rat and human enzyme
activity correlates. For analyzing the results, Ki values were con-
verted to pKi (negative logarithm of Ki expressed in per moles).
These values were then plotted against each compound. In the first
plot (Fig. 3a) experimental (human) and predicted values for both
MAO-A and MAO-B were compared. Where as in the second
(Fig. 3b), experimental values of human and rat enzymes (both
MAO-A and MAO-B) was compared.
A superior almost perfect agreement (Fig. 3a) between the
experimental and predicted values was obtained. The way, struc-
ture based docking with AUTODOCK corresponded the experimental
Ki values is truly exceptional. We have been consistently report-
ing4–6 and emphasizing on these outstanding correlations. These
employed protocols were not only successful in identifying the
resulting HITS through a reasonable VS procedure but also,
We have made some scrupulous analysis of the observed re-
sults. Primarily to check the congruence of the observed results

N. Mishra, D. Sasmal / Bioorg. Med. Chem. Lett. 21 (2011) 1969–1973
1973
consolidated the model, giving us enough support, to be used in fu-
References and notes
ture as a query model.
1. Shih, J. C.; Chen, K.; Ridd, M. J. Annu. Rev. Neurosci. 1999, 22, 197.
2. Riederer, P.; Lachenmayer, L.; Laux, G. Curr. Med. Chem. 2004, 11, 2033.
3. Strolin Benedetti, M.; Dostert, P. Adv. Drug Res. 1992, 23, 65.
4. Karuppasamy, M.; Mahapatra, M.; Yabanoglu, S.; Ucar, G.; Sinha, B. N.; Basu, A.;
Mishra, N.; Sharon, A.; Kulandaivelu, U.; Jayaprakash, V. Bioorg. Med. Chem.
2010, 18, 1875.
5. Sahoo, A.; Yabanoglu, S.; Sinha, B. N.; Ucar, G.; Basu, A.; Jayaprakash, V. Bioorg.
Med. Chem. Lett. 2010, 20, 132.
6. Stirrett, K. L.; Ferreras, J. A.; Jayaprakash, V.; Sinha, B. N.; Ren, T.; Quadri, L. E.
Bioorg. Med. Chem. Lett. 2008, 18, 2662.
7. Schüller, A.; Schneider, G.; Byvatov, E. QSAR Comb. Sci. 2003, 22, 719.
8. Docking methods.
9. Sharma, V.; Pattanaik, K. K.; Jayprakash, V.; Basu, A.; Mishra, N. Bioinformation
2009, 4, 84.
10. De Colibus, L.; Li, M.; Binda, C.; Lustig, A.; Edmondson, D. E.; Mattevi, A. Proc.
Natl. Acad. Sci. U.S.A. 2005, 102, 12684.
Surprisingly, we have obtained a superior correlation (Fig. 3b)
between the Ki values obtained from rat and human enzymes.
Earlier reports from Son et al.,11 suggested a close structural
match between the human and rat enzymes. However, lot of sup-
port is still required to infer whether rat MAO enzyme can re-
place human enzymes in assays, which can be explored further
in future.
From the current work, we have successfully employed a struc-
ture based VS protocol and identified 10 selective and extremely
potent MAO-B inhibitors. We have made some significant infer-
ences in the process, which can be summarized as follows:
1. HITS were successfully identified through VS with AUTODOCK 4.0.
Superior correlation between experimental and predicted Ki
values was observed.
11. Son, S. Y.; Ma, J.; Kondou, Y.; Yoshimura, M.; Yamashita, E.; Tsukihara, T. Proc.
Natl. Acad. Sci. U.S.A. 2008, 105, 5739.
12. General procedure for the synthesis of chalcones (1–10).
To
a solution of different acetophenone (0.01 M) and anthracene-9-
2. All the synthesized molecules were reversible and selective
inhibitors of MAO-B. They were active in nM range, almost
100 times more potent than selegiline.
3. A superior correlation was observed between the Ki values
obtained from rat and human enzymes.
carboxaldehyde (0.01 M) in ethanol (10 ml) was added aqueous solution of
Sodium hydroxide (60%) drop wise with continuous stirring at 0 °C over a
period of 2 h. The reaction mixture was kept at room temperature for about
48 h with occasional shaking. After 48 h it was poured into ice-cold water. In
case hydroxy chalcone pH was adjusted to 2 using 6 N hydrochloric acid. The
yellow precipitate obtained was filtered, washed, dried and recrystallized from
dry methanol. The intermediates 1–10 were obtained.
General procedure for the synthesis of 3-anthracene-5-aryl-4, 5-dihhydro-1H-
pyrazole (11–20).
Appropriate chalcone (1–2) was treated with 10 times excess of hydrazine
hydrate (80%) in dry ethanol and refluxed for 3–6 h. The hot reaction mixture
was then poured into ice-cold water. The solid separated out was filtered,
washed, dried and recrystallized from ethanol to afford respective pyrazoline
(3–4).
Acknowledgement
This work was financially supported by University Grants Com-
mission, India. The authors are grateful to Mr. Arijit Basu for his
contributions in the manuscript preparation.
13. Holt, A.; Sharman, D. F.; Baker, G. B.; Palcic, M. M. Anal. Biochem. 1997, 244, 384.