Rapid synthesis of flavone-based monoamine oxidase (MAO) inhibitors targeting two active sites using click chemistry

Article abstract of DOI:10.1111/cbdd.12841

A new library of flavone derivatives targeting two active sites of monoamine oxidases (“aromatic cage” and substrate cavity) were designed and synthesized using click chemistry (CuAAC reaction) between 6-N₃-2-phenyl chromones (Az1–Az2) and a series of alkynes (k1–k20). Their inhibitory activities against MAO isoforms (MAO-A and MAO-B) are evaluated. Compounds with fluorine, amide bonds, or amino bonds have shown better inhibition. The most potent flavone MAO inhibitor studied is Az2k19 (1.6?μm for MAO-A, 2.1?μm for MAO-B), while Az1k15 and Az2k15 displayed better selectivity toward MAO-B (SI?>?10). Docking studies are in accordance with our hypothesis that these inhibitors are most likely located at both the substrate cavity and the “aromatic cage”. Our results show that it is considerable to develop new MAO inhibitors from C6 substitution of flavone derivatives and that these compounds are also potential for the treatment of diseases associated with MAOs.

Full text of DOI:10.1111/cbdd.12841

Received: 5 December 2015
DOI: 10.1111/cbdd.12841
Revised: 29 May 2016
Accepted: 13 August 2016
|
|
R E S E A R C H A R T I C L E
Rapid synthesis of flavone-based monoamine oxidase (MAO)
inhibitors targeting two active sites using click chemistry
Wei Zhen Jia1
Feng Cheng1
Yin Jun Zhang1
Jin Yan Ge1
Shao Q. Yao2
|
|
|
|
|
1
Qing Zhu
1Institute of Bioengineering, Zhejiang
A new library of flavone derivatives targeting two active sites of monoamine oxi-
dases (“aromatic cage” and substrate cavity) were designed and synthesized using
University of Technology, Zhejiang, China
2Department of Chemistry, National University
of Singapore, Singapore, Singapore
click chemistry (CuAAC reaction) between 6-N -2-phenyl chromones (Az1–Az2)
3
Correspondence
and a series of alkynes (k1–k20). Their inhibitory activities against MAO isoforms
Shao Q. Yao, Department of Chemistry,
National University of Singapore, Singapore,
Singapore.
(
MAO-A and MAO-B) are evaluated. Compounds with fluorine, amide bonds, or
amino bonds have shown better inhibition. The most potent flavone MAO inhibitor
studied is Az2k19 (1.6 μm for MAO-A, 2.1 μm for MAO-B), while Az1k15 and
Az2k15 displayed better selectivity toward MAO-B (SI > 10). Docking studies are
in accordance with our hypothesis that these inhibitors are most likely located at both
the substrate cavity and the “aromatic cage”. Our results show that it is considerable
to develop new MAO inhibitors from C6 substitution of flavone derivatives and that
these compounds are also potential for the treatment of diseases associated with
MAOs.
Email: chmyaosq@nus.edu.sg
and
Qing Zhu, Institute of Bioengineering, Zhejiang
University of Technology, Zhejiang, China.
Email: zhuq@zjut.edu.cn
K E Y W O R D S
click chemistry, flavones, low cytotoxicity, molecular docking, monoamine oxidase inhibitors
Monoamine oxidases (MAOs) are a family of mitochondria-
bound flavoenzymes, which contain the flavin adenine di-
nucleotide (FAD), and could catalyze the deamination of a
range of monoamines producing H O and the correspond-
neurons, metabolizes bulky endogenous amines, and its in-
hibitors, such as clorgyline, are used in the treatment of de-
[13]
pressive illness.
In contrast, MAO-B is mainly found in
astrocytes acting on dopamine and small exogenous amines
such as phenylethylamine. Inhibitors of MAO-B, such as par-
gyline, rasagiline, and L-deprenyl selegiline, are utilized as
therapeutics for neurodegenerative disorders.[
2
2
[1–3]
ing amines.
MAOs have two isoforms, namely MAO-A
and MAO-B, which share approximately 70% amino acid
sequence homology but differ in their cellular localization,
14]
[
4–7]
substrate preference, and inhibitor sensitivity.
These two
The applications discussed above have led to intensive
research into the development of novel MAO inhibitors
(MAOIs) over the past few decades. Current inhibitors can be
divided into three main generations. Compounds belonging
to the first two generations are irreversible and were aban-
doned after clinical trials due to their negative side effects
such as hypertensive response, hepatotoxicity, and “cheese
enzyme classes play very important roles in brain develop-
ment and function by regulating neurotransmitter metabolism
in the brain, thereby maintaining an appropriate concentra-
[8]
tion of intracellular amines. However, overactivation of
MAOs is reported to generate high concentrations of neuro-
toxic by-products (H O , ammonia, aldehyde, etc.), resulting
2
2
[
15–17]
in neuronal abnormalities and causing affective disorders,
stroke, and neurodegenerative diseases, such as Parkinson’s
effect”.
Currently, scientists are developing the third
generation of MAOIs, which are specific towards MAO sub-
types and in some cases are reversible in their action. Several
natural extracts, such as flavonoids, β-carboline, alkaloids,
[9–12]
disease (PD), and Alzheimer’s disease (AD).
MAO-A
is primarily expressed in catecholaminergic and serotonergic
Chem Biol Drug Des 2016; 1–11
wileyonlinelibrary.com/journal/cbdd
© 2016 John Wiley & Sons A/S.
|
1

2
Jia et al.
|
potent MAOs inhibitors.[
32–35]
Meanwhile, it is also reported
tannins, coumarins, xanthones, proantho-cyanidins, iridoid
glucosides, curcumin, and cannabinoids were reported to be
that the flavone and its derivatives could inhibit activity of
MAOs because they have hydrophobic interaction with resi-
dues Phe168, Ile199, Gln206, Cys172 for MAO-B, Phe208,
Asn181, Gln215, Phe352 for MAO-A in the substrate cav-
[
18–20]
potent MAOIs.
These compounds display low cytotox-
icity and have attracted our interests in an endeavor to de-
velop new MAOIs.
[27,30,36]
Flavone, distributed widely in fruits and vegetables, is
ity.
So we hypothesized that the compounds gener-
[21,22]
a class of 2-phenyl-1-benzopyran-4-one (Figure 1A).
ated by the integration of the aromatic ring and 1,2,3-triazoles
with flavone will be able to occupy two active sites of MAO
enzyme and thus increase the inhibitory activity toward
MAOs. Huisgen Cu(I)-catalyzed alkyne–azide cycloaddition
(CuAAC) reaction is highly modular and efficient and such is
Flavones and their analogues display a wide spectrum
of physiological and biochemical roles in disease treat-
ment, including antioxidant, antitumor, antiproliferative,
and anti-inflammatory activities. They are also used in the
treatment of cardiovascular diseases and neurodegenera-
[37]
widely used in drug design.
Click chemistry is based on a
[
23–27]
tive disorders.
Several years ago, flavone attracted our
simple reaction, carried out under mild conditions, produces
few by-products, and provides easy isolation and purification
of products.
interest due to its ability to inhibit MAOs. Some naturally
occurring flavones have been reported to act as effective
MAOIs. These include apigenin (IC : MAO-A = 1.7 μm;
5
0
[28]
MAO-B = 12.8 μm)andluteolin(22.6 μm), whilequercetin
1
1
|
METHODS AND MATERIALS
General information
also shows a selectivity for MAO-A (IC : MAO-A = 2.8 μm;
5
0
MAO-B = 90.0 μm). Esculetin-6-methylether and scopole-
.1
|
tin were reported by Sang-Jun Lee to exhibit MAO-A and
MAO-B inhibitory characteristics with IC values of 32.2
All raw materials, commercially available solvents, and rea-
gents were bought from Energy-Chemical (China) and Xiya
reagent cooperation. Reaction processes were monitored by
thin-layer chromatography (TLC) using silicagel glassplates
5
0
[29]
and 45.0 μm, respectively.
fied the core structure flavones at positions 7 and 4′ using –H,
F, –CH , and –OCH , successfully achieving high inhibition
Franco Chimenti group modi-
–
3
3
[30]
1
with an IC of 1.34 μm.
and observed by UV light (254 nm) and LC/MS. H NMR
5
0
MAOs were reported to have three functional domains
in the active center, named the substrate cavity, the entrance
spectra were recorded using a Bruker instrument (500 MHz).
Mass spectrum was recorded using a Bruker Esquire 6000
mass spectrometer. Synthetic procedure and analytic and
spectroscopic data of compounds 1, 2a-b, 3a-b, 4a-b, and
[4,31]
cavity, and the “aromatic cage”.
“Aromatic cage” is
formed by Tyr398 and Tyr435 and flavin adenine dinucle-
otide (FAD), and a few aromatic or heterocycle compounds
targeting this binding site, such as pyrazoline, oxazolidinone,
[
38]
k1–k20 have been already reported in the literature.
MAO-A (M7316) and MAO-B (M7441) were purchased
from Sigma-Aldrich.
1
,2,3-triazoles, and their analogues, have been designed as
FIGURE 1 (A) Some known flavone-based MAO inhibitors; (B) our strategy of inhibitors assemble using click chemistry

Jia et al.
3
|
SCHEME 1 The synthetic scheme of three flavone azides
1
.2
|
Chemistry
1.2.2
|
Preparation of alkyne blocks
The synthetic routes of flavone derivatives are shown in
Figure 1B and Scheme 1. The characterization of all com-
pounds is shown in supporting information.
Alkynes (k1–k20) were prepared from commercial avail-
[39–42]
able phenol and aniline as described in the literature.
To generate diversity, the phenyl ring attached to the alkyne
was modified with a variety of functional groups (Figure 2B).
1
.2.1
|
Synthesis of 6-azide-substituted
flavones (Az1,2)
1.2.3
|
Click assemble small flavones-based
inhibitor library
3
-Phenyl flavone (3a-b) I (0.15 g) was added to a solution
2
of 2a (b)(10 mmol) in 10 mL dimethylsulfoxide, and the re-
A solution (THF: H O = 1:1) of azide Az1 (Az2) (1 equiv)
2
sulting mixture was refluxed with vigorous stirring for 3 h
was dispensed into 1-mL vial containing a corresponding
at 130 °C. Afterwards, CH Cl (50 mL) was added into the
alkyne (1.2 equiv) and catalytic amount of CuSO (1 equiv)
2
2
4
mixture, and then, organic layer was washed with water,
dried over Na SO , purified by chromatography (EtOAc–
and sodium ascorbate (1.5 equiv). The reactions were moni-
tored using LC/MS, and yields ranged from 83% to 94%.
Pure flavone derivatives (Az1k1–20; Az2k1–20) were
2
4
petrol, 5:1) to obtain 3a (b) as buff acicular solids (yield
[
35]
8
5–90%).
-N -flavone (Az1-2) To a solution of 3a (b) (1 mmol) in
separated by preparative HPLC as white/yellow solids.
6
Typical compounds (Az2k19 and Az2k15) were charac-
3
1
13
1
2
0 mL CCl , NBS (178 mg) and BPO (242 mg) were added,
terized by H-NMR, C-NMR, and MS. The others were
4
and refluxed with continuous stirring for 8 h. The solvent was
evaporated under reduced pressure to give the crude, which
was dissolved in CH Cl (100 mL) and washed with H O
only characterized by H NMR and MS due to the limited
amount.
2
2
2
(
2 × 100 mL), dried with Na SO . Then, the solvent was
2-((4-chlorophenyl))-6-((((4-(4-fluorophenyl) amino)
methyl)-1H-1,3-triazol-1-yl)methyl)-4H-chromen-4-
one (Az2k19)
2
4
evaporated to give the crude 4a (b). Without further treat-
ment, 4a (b) was dissolved in DMF (5 mL), NaN (130 mg,
3
1
2
1
mmol) was added, and the reaction mixture was refluxed for
2 h at room temperature, poured into 50 mL water, extracted
Light yellow solid, yield 91.5%; m.p. 192.3~194.6 °C; H
NMR (500 MHz, CDCl ) δ 8.18 (d, J = 1.9 Hz, 1H), 7.93–
3
with CH Cl (3 × 100 mL), and the organic layers were com-
7.85 (m, 2H), 7.63–7.50 (m, 4H), 7.44 (s, 1H), 7.40–7.31 (m,
2
2
bined and dried over Na SO . The residue was purified by
1H), 6.89 (t, J = 8.7 Hz, 2H), 6.83 (s, 1H), 6.62–6.57 (m,
2
4
1
3
silica column chromatography to obtain pure compounds
2H), 5.63 (s, 2H), 4.42 (s, 2H); C NMR (125 MHz, CDCl )
3
Az1–2 as white solids.
δ 167.8, 161.5, 155.2, 144.8, 133.8, 129.0, 128.0, 124.2,

4
Jia et al.
|
FIGURE 2 The structures of azide and alkyne building blocks and target compounds
1
6
23.2, 122.8, 118.97, 115.0, 114.9, 113.0, 113.0, 107.2, 83.6,
the results were expressed by IC and the selectivity index
50
(SI, IC of MAO-B/MAO-A).
50
+
6.0, 51.9, 46.0, 20.3; HRMS (M + H) m/z calculated for
C H ClFN O 461.1102; found 461.1165.
2
5
18
4 2
1
.3.2
|
Cell cytotoxicity
2
-((4-chlorophenyl))-6-((((4-((((3-hydroxyphenoxy))
methyl))-1H-1,2,3-triazol-1-yl))methyl))-4H-chromen-
RTE (rat tracheal epithelial) cell line was cultured at 37 °C
4
-one (Az2k15)
and 5% CO in DMEM medium supplemented with 10%
2
Light yellow solid, yield 87.9%; m.p. 194.5~195.7 °C;
fetal bovine serum. Cytotoxicity of tested compounds was
determined by MTT assay according to the general procedure
1
H NMR (500 MHz, CDCl ) δ 8.12 (d, J = 2.2 Hz, 1H),
3
[44]
7
.89–7.85 (m, 2H), 7.66–7.57 (m, 3H), 7.55–7.51 (m, 2H),
.14–7.08 (m, 1H), 6.83 (s, 1H), 6.53–6.47 (m, 3H), 5.64
previously described by us.
Exponentially growing cells
7
were collected using trypsin and plated in 96-well plates at
5000 cells/well. Cells were exposed to the drugs (eight con-
centration gradients, six parallel control) for 24 h and subse-
quently added 50 μL MTT (1 mg/mL) for another 4 h. Then,
culture media were removed and 150 μL DMSO was used to
dissolve the precipitate, and the absorption was measured at
1
3
(
s, 2H), 5.18 (s, 2H); C NMR (125 MHz, CDCl ) δ178.0,
3
1
1
1
62.9, 159.3, 157.7, 156.0, 138.3, 133.8, 132.2, 130.2, 129.6,
29.4, 127.6, 125.1, 123.7, 123.2, 119.3, 108.9, 107.5, 106.8,
+
02.7, 61.9, 53.4, 53.3; HRMS (M + H) m/z calculated for
C H ClN O 460.0986; found 460.1061.
2
5
18
3 4
5
70 nm. The cytotoxicity was expressed by IC .
50
1
.3
.3.1
|
Biological activities evaluation
Detection of enzyme inhibition
1
.3.3
|
Molecular modeling studies
1
|
To explore the binding modes of target compounds into
the active center of MAO, docking study was carried out
using the Autodock4.2.6 software (AutoDockTools-1.5.6).
in vitro
The inhibitory activities of flavones were determined in vitro
by using fluorescent probe method following the general pro-
To prepare the receptor, the crystal structures of harmine/
[43]
[45]
cedure described in the literature.
The MAO inhibitory
MAO-A (PDB code: 2Z5X)
and farnesol/MAO-B (PDB
[
46]
potency of test compounds was evaluated by measuring fluo-
rescent intensity using fluorescence spectrophotometer, and
code: 2KB3)
were retrieved from the PDB. The pro-
tein structures were preparation for docking simulation by

Jia et al.
5
|
extracting the enzyme code using Uedit32 software. The
visualizations of simulation results were achieved by PyMol
software.
two series of compounds was performed in the SYBYL using
“align database” (Fig S9).
The CoMFA method was performed according to the
[48,49]
reported publications.
A partial least-squares (PLS)
approach was used to derive the 3D-QSAR, in which the
CoMFA descriptors were used as independent variables, and
1
.3.4
|
3D-QSAR analysis
Molecular modeling of the forty compounds (inhibitors)
was performed using SYBYL-X 2.0 software (Tripos, Inc.).
Each structure was fully geometry optimized using a con-
jugate gradient procedure based on the TRIPOS force field
PIC values were used as dependent variables. The cross-
50
validation with the leave-one-out (LOO) option and the
SAMPLS program, rather than column filtering, was carried
out to obtain the optimal number of components to be used
in the final analysis. After optimization of the numbers of
components, a non-cross-validated analysis was performed
without column filtering. The modeling capability (goodness
[47]
and Gasteiger and Hückel charges.
The IC values of
50
all compounds towards MAO-A and MAO-B used to derive
[48]
the CoMFA analyses model are listed in Tables 1 and 2.
The activity was expressed in terms of PIC by the formula
2
of fit) was judged by the correlation coefficient squared (r ).
5
0
PIC = −log (IC ). The compounds Az1k1 (Table 1) and
50
10
50
Az1k11 (Table 2) were used as templates to build the other
molecular structures. Because these compounds share a com-
mon skeleton, 24 atoms marked with an asterisk were used
for RMS fitting onto the corresponding atoms of the template
structure (Fig S8). The compounds in Tables 1 and 2 are di-
vided into two databases, and then, the superimposition of the
2
2
|
RESULTS AND DISCUSSION
Chemistry
.1
|
Flavones (Az1–Az2) containing an azide at the C6 posi-
tions were synthesized in several steps from commercially
TABLE 1 Inhibition of MAO-A and
X
MAO-B by C6-position modification inhibitors
O
O
(
Class I)
N
N
N
MAO-A IC
MAO-B IC
50
5
0
Class
Inhibitors
Alkyne
X
(μm)
(μm)
SI
1.7
I
Az1k1
Az1k2
Az1k3
Az1k4
Az1k5
Az1k6
Az1k7
Az1k8
Az1k9
Az1k10
Az2k1
Az2k2
Az2k3
Az2k4
Az2k5
Az2k6
Az2k7
Az2k8
Az2k9
Az2k10
k1
k2
k3
k4
k5
k6
k7
k8
k9
k10
k1
k2
k3
k4
k5
k6
k7
k8
k9
k10
H
183.6 ± 5.4
110.9 ± 3.8
98 ± 4.3
H
727 ± 81.4
84.6 ± 6.7
265.7 ± 13.8
25.3 ± 1.7
737.5 ± 79.8
ND
7.4
5.4
2.1
0.9
5.5
H
15.6 ± 2.1
123.8 ± 5.5
28.5 ± 1.0
134.4 ± 4.8
ND
H
H
H
H
H
326.6 ± 12.5
ND
138.6 ± 6.2
2.4
H
413 ± 43.4
151.1 ± 6.3
32.91 ± 2.6
86 ± 7.7
H
ND
>6.6
3.1
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
103.1 ± 5.7
541 ± 49.2
372 ± 19.8
ND
6.3
47 ± 3.3
7.9
321 ± 13.1
287 ± 14.7
121 ± 6.6
ND
ND
678 ± 61.1
ND
5.6
ND
626 ± 43.9
513 ± 49.2
377 ± 19.3
ND
ND
(
1) All IC50 values shown in this table are the mean ± SD from three experiments. SI: selectivity
index = IC50(MAO-A)/IC50(MAO-B).
2) ND means the enzyme inhibition rate is lower than 50% at the concentration of 1.0 mm.
(

6
Jia et al.
|
TABLE 2 Inhibition of MAO-A and
MAO-B by C6-position modification inhibitors
X
O
O
N
(Class II)
N
N
MAO-A IC
(μm)
MAO-B IC
50
(μm)
5
0
Class
Inhibitors
Alkyne
X
SI
II
Az1k11
Az1k12
Az1k13
Az1k14
Az1k15
Az1k16
Az2k11
Az2k12
Az2k13
Az2k14
Az2k15
Az2k16
Az1k17
Az1k18
Az1k19
Az1k20
Az2k17
Az2k18
Az2k19
Az2k20
k11
k12
k13
k14
k15
k16
k11
k12
k13
k14
k15
k16
k17
k18
k19
k20
k17
k18
k19
k20
H
350 ± 25.2
ND
84.7 ± 7.6
ND
4.1
H
H
ND
ND
ND
ND
322 ± 27.6
527 ± 42.9
75.8 ± 9.4
97.4 ± 5.0
336.1 ± 18.9
97.4 ± 10.8
291 ± 23.6
505 ± 52.2
71.3 ± 7.6
93.9 ± 9.6
149.8 ± 16.9
77.4 ± 3.7
38.4 ± 1.1
176 ± 7.4
>3.1
H
H
>13.2
>10.3
1.0
H
Cl
Cl
Cl
Cl
Cl
Cl
H
346 ± 16.2
151.9 ± 23.1
ND
1.6
>3.4
ND
ND
>14.0
>10.7
2.1
ND
317 ± 34.4
108.9 ± 8.0
56.5 ± 3.2
333.6 ± 32.4
283 ± 43.9
192.5 ± 15.4
1.6 ± 0.2
165.8 ± 8.5
H
1.4
H
1.5
H
1.9
Cl
Cl
Cl
Cl
350 ± 30.8
202.7 ± 9.7
2.1 ± 0.7
0.8
0.9
0.7
132.3 ± 12.1
1.3
(
1) All IC50 values shown in this table are the mean ± SD from three experiments. SI: selectivity
index = IC50(MAO-A)/IC50(MAO-B).
2) ND means the enzyme inhibition rate is lower than 50% at the concentration of 1.0 mm.
(
available compounds as shown in Scheme 1. Alkynes (k1–
k20) were prepared from commercially available phenol and
for IC of MAO-A over IC of MAO-B) of all compounds
5
0
50
are summarized in Tables 1 and 2. The test molecules can be
divided into two classes: Class I contains 1,2,4-triazole at-
tached to an aromatic group via an amide linker at the C6 po-
sition (entries Az1k1–10, Az2k1–10), and Class II contains
1,2,4-triazole-O-aryl or 1,2,4-triazole-NH-aryl groups at the
C6 position (entries Az1k11–20, Az2k11–20).
[39–42]
aniline.
To generate diversity, the phenyl ring attached
to the alkyne was modified with a variety of functional groups
(
Figure 2B). The reaction condition was optimized to obtain
the quantitative triazole products when the reaction was car-
ried out in a THF:H O (1:1) solvent mixture with CuSO /
2
4
ascorbate acid as catalysts. Although the click products could
be used directly without further purification, a few were sub-
jected to preparative HPLC for further analysis. Ultimately,
Focusing on Class I, twenty compounds (Az1k1–10,
Az2k1–10) were synthesized through the reaction of azide
Az1 (or azide Az2) and different-length alkynes containing
6
-triazole-substituted flavonoid derivatives (Az1k1–20,
p-CH (Cl, F, or H)-substituted benzamide groups (k1–10).
3
Az2k1–20, 40 in total) were synthesized through attachment
of aromatic groups to the C6 position of flavone through the
triazole moiety (Figure 2C).
MAO inhibitory potency of compounds of this class appears
to be mainly dependent on the alkyl chain length. By com-
parison of molecules with different chain lengths (n = 1, 2,
4
) (Figure 2C), some trends become apparent. With increas-
ing chain length (n = 1, 2–4), the inhibitory activities against
MAO-A and MAO-B gradually decrease. This is obvious
in compounds Az1k1 (n = 1) (MAO-A, 183.7 μm; MAO-
B, 110.9 μm), Az1k4 (n = 2) (MAO-A, 265.7 μm; MAO-B,
123.8 μm), and Az1k8 (n = 4) (MAO-A, 326.6 μm; MAO-B,
138.6 μm). Other trends can be observed in the selectivity
2
.2
|
In vitro MAO inhibitory activity
All molecules in the library were initially evaluated for inhi-
bition of both MAO-A and MAO-B. The IC of the test com-
5
0
pounds was monitored by fluorescence quenching after the
addition of flavones.[
43,50]
The IC and selectivity index (SI
50

Jia et al.
7
|
of inhibitors. Az1k5 is a strong inhibitor of both MAO-A
RTE cells and did not cause the cell death even at very high
concentrations.
(
25.3 μm) and MAO-B (28.5 μm), while Az1k3 (15.6 μm)
is a better MAO-B inhibitor and Az2k1 (32.91 μm) a better
MAO-A inhibitor. Surprisingly, all Az2k series compounds
with chlorine substituted at the C4′ position exhibit weak
inhibition of MAO-A, but stronger inhibition of MAO-B.
Comparing different functional groups with the same chain
length, it was observed that Az1k3 (MAO-A, 84.6 μm; MAO-
B, 15.6 μm) and Az2k3 (MAO-A, 372 μm; MAO-B, 47 μm)
containing p-chlorophenyl substituent have better potency
when the chain length n is equal to 1, while Az2k6 (MAO-
A, 678 μm; MAO-B, 121 μm) and Az2k10 (MAO-A, ND;
MAO-B, 377 μm) containing p-fluorophenyl substituent are
more potent when n is 2 or 4.
2
.4
|
Molecular docking studies
To further investigate the interactions between the most
potent inhibitors and MAOs, we performed docking stud-
ies to explore possible binding modes. Using AUTODOCK
4.2.6, the docking of three compounds to the active center
of MAOs was examined. Az2k19 (the best MAO inhibitor
in this study), Az2k17 (which has only slightly lower po-
tency), and Az1k15 (the most selective MAO-B inhibitor)
were selected based on their favorable inhibitory properties.
The X-ray crystallographic structures of MAO-A (PDB ID
code: 2Z5X) and MAO-B (PDB ID code: 2KB3) were se-
lected as the docking models. The docking geometries of
Az2k19/Az1k15 with MAO-A and MAO-B and of Az2k17
with MAO-B are shown in Figures 3 and 4, respectively.
There are three functional domains (the substrate cavity, the
entrance cavity, and the “aromatic cage”) in the active site of
Class II of inhibitors (Az1k11–20, Az2k11–20) was
synthesized through a click reaction of azide Az1 (or azide
Az2) with alkynes (k11–20). The values of SI
and
MAO-A/B
IC
for Az1k11 (SI = 4.1, IC = 84.7 μm), Az1k13
50
5
0 MAO-B
(
SI > 3.1,IC = 322 μm),Az2k13(SI > 3.4,IC = 291 μm)
50 50
show that introduction of an ether bond decrease inhibition
of MAO-A (Table 2). Meanwhile, compounds having m-
phenol groups (Az1k15, SI > 13.2; Az1k16, SI > 10.3) and
p-cyclopropyl aminophenyl groups (Az1k16, SI > 10.7) ex-
hibit a high selectivity for inhibition of MAO-B. The results
in Table 2 demonstrate that the introduction of a secondary
amine substituent at the C6 position of flavone is beneficial
in this case. It can be easily seen that compounds substituted
with secondary amine show strong inhibition of MAO, espe-
cially Az1k19 (IC MAO-A = 56.5 μm, MAO-B = 38.4 μm,
[
4,31]
MAO.
The substrate cavity is directly in front of the fla-
vin adenine dinucleotide (FAD) cofactor containing residues
178–221. “Aromatic cage” is consisted by FAD, Tyr407, and
Tyr444 in MAO-A and FAD, Tyr435, and Tyr398 in MAO-
B. The entrance cavity is situated nearer to the surface of the
protein and is composed of residues Phe108-Pro118 (MAO-
A) or Phe99-Tyr112 (MAO-B).[
4,32,51]
Figure 3A shows that Az2k19 almost completely occupies
all entrances into the substrate cavity of MAO-B, situating the
p-chlorophenyl across the entrance and substrate cavities sep-
arated by residues Leu171, Ile 199, Phe168, and Tyr326. The
triazole and p-fluorophenyl moiety are positioned in the “ar-
omatic cage”, forming hydrogen bonds with the FAD coen-
zyme (O/N distance: 2.6 Å) and with Tyr435 (O/O distance:
3.1 Å). Az2k19 also occupies the substrate cavity, stabilized
by hydrogen bond interactions with residue Ile198 (O/O dis-
tance: 2.8 Å) and hydrophobic interactions with Tyr 326,
Phe 343, Leu171, Cys172, Gln206, Ile199, Tyr188, Phe168,
and Trp119 (in the entrance cavity). Figure 3B clearly shows
that Az2k19 occupies both cavities of MAO-A, situating the
triazole and p-fluorophenyl moiety in the “aromatic cage”,
forming hydrogen bond interactions with the FAD coen-
zyme (N/N distance: 2.7 Å) and with Asn181 (O/H distance:
1.9 Å), and hydrophobic interactions with Tyr407, Tyr444,
Ile180, Ile207, Ile335, Phe352, Phe208, Phe350, Gln215,
Cys323, Leu337, and Val210. The p-chlorophenyl moiety oc-
cupies the entrance cavity interacting with lipophilic residues
Ala111, Phe108, Leu97, Ile325, and Val210. This docking
data also explain why Az2k19 shows strong inhibition of
both isoenzymes compared with compound Az2k19.
5
0
SI = 1.5) and Az2k19 (MAO-A = 1.6 μm, MAO-B = 2.1 μm,
SI = 0.7) which also possess fluorine leading to even stron-
ger inhibition.
Overall, compounds containing amide bonds and amine
functional groups have stronger inhibitory effects than those
containing an ether group. However, compounds containing
ether group exhibit good selectivity. Despite the limited num-
ber of secondary amine derivatives, it is worth noting that
compounds containing this functional group display excep-
tionally high MAO inhibition potency. Furthermore, fluorine
substitution significantly improves MAO inhibition.
2
.3
|
Cytotoxicity tests
1
1 compounds (Az1k3, Az1k5, Az1k6, Az1k10, Az1k11,
Az1k15, Az1k19, Az2k1, Az2k3, Az2k15, and Az2k19)
were investigated for cytotoxicity against RTE (rat tracheal
epithelial) cells employing the MTT-based assay. The re-
sult showed that all inhibitors displayed low cytotoxicity
(
IC > 400 μm) against the cell line. It is also worth noting
50
that cell viability is 87.1% when treated with Az1k19 which
is the best MAO inhibitor in our library, and 77.5% when
treated with Az1k15 which is the most selective MAO-B
inhibitor. In other words, all inhibitors were non-toxic to
To investigate the binding models of Az1k15 with MAO-A
and MAO-B, the docking studies were also performed on this
molecule and the results are displayed in Figure 4. As shown

8
Jia et al.
|
A
A
B
B
FIGURE 4 3D docking models of target compounds and MAOs. (A)
Az1k15 with MAO-A; (B) Az1k15 with MAO-B. The enzyme is shown
as carbon; the target compounds and FAD are reported as sticks. The
interacting residues are displayed as green or yellow lines. Yellow dotted
line indicates the intermolecular hydrogen bond
C
Tyr60, Tyr398, Ile199, Ile198, Cys172, Phe343, Leu171, and
Gln206. All of the docking results in Figures 3A-C and 4B
are in accordance with the experimental data further explain-
ing why Az1k15 (IC = 75.8 μm) has better selectivity to-
5
0
ward MAO-B than Az2k17 (IC = 350 μm), but weaker than
5
0
Az2k19 (IC = 2.1 μm).
5
0
2
.5
|
CoMFA analysis
FIGURE 3 3D docking models of target compounds and MAOs.
A) Az2k19 with MAO-B; (B) Az2k19 with MAO-A; (C) Az2k17 with
(
To further design better inhibitors towards MAO-A and
MAO-B, the CoMFA method was employed for rapid predic-
tion of quantitative structure–activity relationship (QSAR).
Four predictive CoMFA models were established for QSAR
analysis of the compounds in Table 1 towards MAO-A and
MAO-B (Model 1 and Model 2), and compounds in Table 2
towards MAO-A and MAO-B (Model 3 and Model 4). The
MAO-B. The enzyme is shown as carbon, and the target compounds and
FAD are reported as sticks. The interacting residues are displayed as green
or yellow lines. Yellow dash line indicates the intermolecular hydrogen
bond
in Figure 4A, Az1k15 did not enter the active site and instead
binds to the surface of MAO-A, which likely explains its rela-
tively weak inhibition of MAO-A. Meanwhile, Az1k15 is able
to fit into the substrate cavity forming hydrogen bond inter-
actions with Tyr435 (O/N distance: 2.7 Å, belongs to the “ar-
omatic cage”) and Tyr326 (O/O distance: 2.4 Å; Figure 4B).
It is also stabilized by hydrophobic interactions with residue
2
conventional correlation coefficients (r ) of models 1–4 are
0.69, 0.66, 0.79, and 0.73, respectively. The isocontour dia-
grams of the steric and electrostatic field contributions were
displayed in Figure 5 (compound Az1k1 as template) and
Figure 6 (compound Az1k11 as template). The steric field

Jia et al.
9
|
A
B
C
D
FIGURE 5 Steric contribution contour
maps of Az1k1 towards MAO-A (A) and
MAO-B (C). Electrostatic contribution contour
maps of Az1k1 towards MAO-A (B) and
MAO-B (D)
FIGURE 6 Steric contribution contour
maps of Az1k11 towards MAO-A (A) and
MAO-B (C). Electrostatic contribution contour
maps of Az1k11 towards MAO-A (B) and
MAO-B (D)
contours are represented with different colors: The green color
means a bulky group here would be favorable for higher in-
secticidal activity, while the yellow color means the opposite.
As shown in Figures 5A, C, 6A, C, there is a green region
located around the para-position of the benzene ring, indi-
cating that the bulky groups at this position will increase the
inhibition activity towards MAO-A and MAO-B. This is in
a good agreement with the experimental data of compounds
Az1k5 and Az1k19 with a methyl group at this position, and
Az1k3 and Az2k3 with a chlorine group at this position (the
IC values of Az1k5, Az1k19, Az1k3, and Az2k3 towards
5
0
MAO-A and MAO-B are all less than 85 μm). Similarly, the
electrostatic contours are displayed in two different colors:
Blue means that an increase in the positive charge is desir-
able for improving the activity, while the red contour defines
the opposite. Figures 5B, D, 6B, D show that the designed
compounds bearing an electron-withdrawing group at 2-,
3-, 4-positions of the benzene ring and an electron-donating

1
0
Jia et al.
|
[
10] J. Sterling, Y. Herzig, T. Goren, N. Finkelstein, D. Lerner, W. Goldenberg, I.
Miskolczi, S. Molnar, F. Rantal, T. Tamas, G. Toth, A. Zagyva, A. Zekany,
J. Finberg, G. Lavian, A. Gross, R. Friedman, M. Razin, W. Huang, B.
Krais, M. Chorev, M. B. Youdim, M. Weinstock, J. Med. Chem. 2002, 45,
group at the 5-positions would display higher inhibition ac-
tivity. These results provide useful information for further
optimization of the compounds.
5
260.
[
11] M. S. Song, D. Matveychuk, E. M. MacKenzie, M. Duchcherer, D. D.
Mousseau, G. B. Baker, Prog. Neuropsychopharmacol. Biol. Psychiatry
3
|
CONCLUSIONS
2
013, 44, 118.
[
[
12] P. Riederer, W. Danielczyk, E. Grünblatt, Neurotoxicology 2004, 25, 271.
13] C. J. Fowler, L. Oreland, B. A. Callingham, J. Pharm. Pharmacol. 1981, 33,
341.
In conclusion, we have successfully developed a click
chemistry-based strategy for the preparation of flavone
MAO inhibitors modified at C6 position which exhibit low
cytotoxicity. Among the molecules examined, compounds
containing fluorine, amide bonds, or amino bonds have im-
proved inhibitory activities. The most potent flavone MAO
inhibitor studied is Az2k19 (1.6 μm for MAO-A, 2.1 μm
for MAO-B), while Az1k15 and Az2k15 displayed better
selectivity toward MAO-B (SI > 10). Docking studies not
only demonstrate Az2k19 is a stronger inhibitor of MAO
and Az2k15 has better selectivity towards MAO-B, but also
are in accordance with our hypothesis that this series inhibi-
tors are most likely located at the substrate cavity and the
[
[
[
[
14] J. P. Johnston, Biochem. Pharmacol. 1968, 17, 1285.
15] B. Blackwell, Lancet 1963, 282, 849.
16] M. B. Youdim, Y. S. Bakhle, Br. J. Pharmacol. 2006, 147, 287.
17] R. A. Remick, C. Froese, F. D. Keller, Prog. Neuropsychopharmacol. Biol.
Psychiatry 1989, 13, 97.
[
18] S. Carradori, M. D’Ascenzio, P. Chimenti, D. Secci, A. Bolasco, Mol.
Divers. 2014, 18, 219.
[
[
19] S. Carradori, J. P. Petzer, Expert Opin. Ther. Pat. 2015, 25, 91.
20] D. Vina, S. Serra, M. Lamela, G. Delogu, Curr. Top. Med. Chem. 2012, 12,
2131.
[
[
21] A. L. Simons, M. Renouf, S. Hendrich, P. A. Murphy, J. Agric. Food Chem.
2
005, 53, 8519.
22] D. Kim, J. Park, J. Kim, C. Han, J. Yoon, N. Kim, J. Seo, C. Lee, J. Agric.
Food Chem. 2006, 54, 935.
[23] H. I. El-Subbagh, S. M. Abu-Zaid, M. A. Mahran, F. A. Badria, A. M. Al-
“
aromatic cage”. Taken together, these results show that it is
Obaid, J. Med. Chem. 2000, 43, 2915.
possible to develop new MAO inhibitors from C6 substitu-
tion of flavonoid derivatives and that these compounds may
have potential for the treatment of diseases associated with
MAOs.
[
[
[
[
24] A. Gaspar, T. Silva, M. Yánez, D. Vina, F. Orallo, F. Ortuso, F. Borges, J.
Med. Chem. 2011, 54, 5165.
25] C. Kanadaswami, L. T. Lee, P. P. H. Lee, J. J. Hwang, F. C. Ke, Y. T. Huang,
M. T. Lee, In Vivo 2005, 19, 895.
26] O. L. Woodman, W. F. Meeker, M. Boujaoude, J. Cardiovasc. Pharmacol.
2
005, 46, 302.
27] F. Chimenti, F. Cottiglia, L. Bonsignore, L. Casu, M. Casu, C. Floris, D.
ACKNOWLEDGMENTS
Secci, A. Bolasco, P. Chimenti, I. Guarino, J. Nat. Prod. 2006, 69, 945.
This study was supported by the National Nature Science
Foundation of China (Nos. 21272212, 21472172), Project
of Science Technology Department of Zhejiang Province
[28] X. H. Han, S. S. Hong, J. S. Hwang, M. K. Lee, B. Y. Hwang, J. S. Ro, Arch.
Pharmacal Res. 2007, 30, 13.
[
[
[
[
29] S. J. Lee, H. Y. Chung, I. K. Lee, S. U. Oh, I. D. Yoo, Food Sci. Biotechnol.
2
000, 9, 179.
(
2014C33141), Project of Science Technology Department
30] F. Chimenti, R. Fioravanti, A. Bolasco, P. Chimenti, D. Secci, F. Rossi, M.
Yanez, F. Orallo, F. Ortuso, S. Alcaro, J. Med. Chem. 2009, 52, 2818.
31] D. E. Edmondson, A. Mattevi, C. Binda, M. Li, F. Hubalek, Burger’s Med.
Chem. Drug Discov. 2003, 11, 1983.
32] G. La Regina, R. Silvestri, M. Artico, A. Lavecchia, E. Novellino, O.
Befani, E. Agostinelli, J. Med. Chem. 2007, 50, 922.
[33] F. Reck, F. Zhou, M. Girardot, G. Kern, C. J. Eyermann, N. J. Hales, R. R.
of Jinhua City (2013-3-003).
CONFLICT OF INTEREST
The authors declare no conflict of interest.
Rona, M. B. Gravestock, J. Med. Chem. 2005, 48, 499.
[
34] T. Z. Jones, P. Fleming, C. J. Eyermann, M. B. Gravestock, R. R. Ramsay,
Biochem. Pharmacol. 2005, 70, 407.
REFERENCES
[35] Z. Jia, Q. Zhu, Bioorg. Med. Chem. Lett. 2010, 20, 6222.
[
[
[
1] J. C. Shih, K. Chen, M. J. Ridd, Annu. Rev. Neurosci. 1999, 22, 197.
2] M. D. Berry, A. V. Juorio, I. A. Paterson, Prog. Neurobiol. 1994, 42, 375.
3] D. E. Edmondson, A. Mattevi, C. Binda, M. Li, F. Hubalek, Curr. Med.
Chem. 2003, 11, 1983.
[36] F. Chimenti, R. Fioravanti, A. Bolasco, P. Chimenti, D. Secci, F. Rossi, M.
Yáñezb, F. Orallob, F. Ortusoc, S. Alcaroc, R. Cirillid, R. Ferrettid, M. L.
Sanna, Bioorg. Med. Chem. 2010, 18, 1273.
[37] Y. Loh, H. Shi, M. Hu, S. Q. Yao, Chem. Commun. 2010, 46, 8407.
[38] D. Van den Berg, K. R. Zoellner, M. O. Ogunrombi, S. F. Malan, G.
Terre’Blanche, N. Castagnoli, J. J. Bergh, J. P. Petzer, Bioorg. Med. Chem.
2007, 15, 3692.
[
[
[
[
[
[
4] C. Binda, P. Newton-Vinson, F. Hubálek, D. E. Edmondson, A. Mattevi,
Nat. Struct. Biol. 2002, 9, 22.
5] A. S. Kalgutkar, D. K. Dalvie, N. Castagnoli, T. J. Taylor, Chem. Res.
Toxicol. 2001, 14, 1139.
[39] C. L. Paradise, P. R. Sarkar, M. Razzak, J. K. De Brabander, Org. Biomol.
Chem. 2011, 9, 4017.
6] D. E. Edmondson, C. Binda, J. Wang, A. K. Upadhyay, A. Mattevi,
Biochemistry 2009, 48, 4220.
[40] K. E. Schulte, J. Reisch, Archiv der Pharmazie und Berichte der Deutschen
Pharmazeutischen Gesellschaft 1959, 292, 51.
7] K. F. Tipton, S. Boyce, J. O’Sullivan, G. P. Davey, J. Healy, Curr. Med.
Chem. 2004, 11, 1965.
[41] J. F. S. Carvalho, J. Louvel, M. L. J. Doornbos, E. Klaasse, Z. Y. Yu, J.
Brussee, A. P. IJzerman, J. Med. Chem. 2013, 56, 2820.
8] K. N. Westlund, R. M. Denney, R. M. Rose, C. W. Abell, Prog. Neurobiol.
1
994, 42, 375.
[42] J. Dinges, D. H. Albert, L. D. Arnold, K. L. Ashworth, I. Akritopoulou-Zanze, P.
F. Bousquet, J. J. Bouska, G. A. Cunha, S. K. Davidsen, G. J. Diaz, S. W. Djuric,
9] P. Riederer, M. B. Youdim, J. Neurochem. 1986, 46, 1359.

Jia et al.
11
|
A. F. Gasiecki, G. A. Gintant, V. J. Gracias, C. M. Harris, K. A. Houseman, P. H.
[48] K. Z. Myint, X. Q. Xie, Int. J. Mol. Sci. 2010, 11, 3846.
[49] J. K. Buolamwini, H. Assefa, J. Med. Chem. 2002, 45, 841.
Tapang, Z. R. Xia, H. Q. Zhang, J. Med. Chem. 2007, 50, 2011.
43] L. Li, C. Zhang, G. Y. J. Chen, B. Zhu, C. Chai, Q. H. Xu, E. K. Tan, Q. Zhu,
K. L. Lim, S. Q. Yao, Nat. Commun. 2014, 5, p. 3276.
[
[
[
[
[
[50] Q. Zhu, M. Uttamchandani, D. Li, M. L. Lesaicherre, S. Q. Yao, Org. Lett.
2003, 5, 1257.
44] B. Y. He, Y. J. Wang, Y. G. Zheng, W. Chen, Q. Zhu, Chem. Biol. Drug Des.
[51] C. Binda, A. Mattevi, D. E. Edmondson, J. Biol. Chem. 2002, 277,
2
013, 82, 307.
23973.
45] S. Y. Son, J. Ma, Y. Kondou, M. Yoshimura, E. Yamashita, T. Tsukihara,
Proc. Natl Acad. Sci. 2008, 105, 5739.
46] P. Barthe, C. Roumestand, M. J. Canova, L. Kremer, C. Hurard, V. Molle,
M. Cohen-Gonsaud, Structure 2009, 17, 568.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in
the supporting information tab for this article.
47] K. T. Gundertofte, P. O. Liljefors, I. P. Norrby, J. Comput. Chem. 1996, 17,
4
29.