Design, synthesis, antitumor evaluation, 3D-QSAR and molecular docking studies of novel 4-aminoacridone compounds

Article abstract of DOI:10.1007/s00044-017-1953-3

4-aminoacridone was efficiently synthesized using an optimized method and condensed with a variety of different aldehydes to give the corresponding Schiff bases, 1a–k. The antiproliferative activities of these compounds were measured against several human

Full text of DOI:10.1007/s00044-017-1953-3

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res
DOI 10.1007/s00044-017-1953-3
ORIGINAL RESEARCH
Design, synthesis, antitumor evaluation, 3D-QSAR and molecular
docking studies of novel 4-aminoacridone compounds
1
,2 ●
1
2 ●
3 ●
2 ●
4 ●
Lin Tian
Chang J. Feng Tong X. Li Zhao Li Wei H. Yang
Xiang E. Han
Received: 31 January 2016 / Accepted: 3 June 2017
Springer Science+Business Media, LLC 2017
©
Abstract 4-aminoacridone was efficiently synthesized
using an optimized method and condensed with a variety of
different aldehydes to give the corresponding Schiff bases,
were conducted to evaluate the multidrug resistance mod-
ulatory effects of these imine compounds in the adenosine
tri-phosphate binding site of P-glycoprotein and the trans-
membrane binding pocket. These docking experiments
revealed the occurrence of important interactions between
these molecules and the active site of the transmembrane
binding pocket, predicting multidrug resistance modulatory
behavior.
1
a–k. The antiproliferative activities of these compounds
were measured against several human cancer cell lines
in vitro, including A549, HeLa, SGC-7901, and Raji cells.
The results of these bioassays indicate that these com-
pounds possess antiproliferative activity for the HeLa and
Raji cell lines. In particular, compounds 1d and 1k con-
taining 4-(N,N-dimethyl)phenyl and 2,4-dichlorophenyl
groups, respectively, showed greater potency and selectivity
●
●
Keywords 4-aminoacridone derivative Antitumor activity
●
●
3D-QSAR Multidrug resistance (MDR) modulator
towards HeLa cells than any of the other cell lines (IC50
=
Docking analysis
7
.75 and 8.88 μM, respectively). Three-dimensional contour
maps based on highly predictive 3D-quantitative
2
structure–activity relationship studies (R = 0.674, R =
cv
0
.956) are used to explain the structure-activity relation-
ships of these compounds. Furthermore, docking studies
Introduction
Acridone derivatives not only possess excellent fluores-
cence properties (Swist and Soloduch 2013), but also
exhibit a wide range of biological activities (Sathish et al.
Electronic supplementary material The online version of this article
(
doi:10.1007/s00044-017-1953-3) contains supplementary material,
2010). For example, several studies have shown that acri-
which is available to authorized users.
done derivatives exhibit potent inhibitory activities against a
large number of biological targets, including cathepsins L
and V (Marques et al. 2012) and acetylcholinesterases
*
*
Lin Tian
xzittl@163.com
Xiang E. Han
xiangenh@163.com
(Mohammadi-Khanaposhtani et al. 2015), as well as bio-
logical activity in several cell lines, including A549 lung
carcinoma cells (Mouafo and Roy 2010) and CCRF-CEM
cells (Cao et al. 2008). To prepare novel acridone deriva-
tives with enhanced levels of activity and selectivity,
researchers have directed significant efforts towards the
design and synthesis of substituted acridone ring systems,
and several interesting studies have emerged from this work
(Tabarrini et al. 2006). In particular, acridone-4-carboxylic
acid derivatives, such as the corresponding amides, have
1
2
3
4
School of Chemical Engineering and Technology, China
University of Mining & Technology, 221116 Xuzhou, China
School of Chemical Engineering and Technology, Xuzhou
Institute of Technology, 221018 Xuzhou, China
School of Food Science and Technology, Xuzhou Institute of
Technology, 221018 Xuzhou, China
School of Chemical Engineering and Technology, Jiangsu Normal
University, 221116 Xuzhou, China

Med Chem Res
been reported to exhibit enhanced levels of cytotoxicity
towards a wide variety of cancer cells, compared with
acridone alone (Wang et al. 2015).
However, there have been very few reports pertaining to
the imine derivatives of acridone, where the amino group is
directly linked to the 4-position of the acridone ring. With
this in mind, we synthesized 11 novel Schiff base com-
pounds based on the condensation of 4-aminoacridone with
Bruker DPX 400 MHz spectrometer. The chemical shifts (δ)
have been reported in p.p.m. relative to tetramethylsilane,
which was used as an internal reference. HRMS analyses
were performed using a MicroTOF-Q spectrometer (Bruker).
X-Ray crystal data were collected using a Bruker Smart
APEX II CCD diffractometer. The chemical reactions were
monitored using thin layer chromatography on silica gel
plates and visualized with UV light.
1
1 different aldehydes. The 4-aminoacridone scaffold
required for this work was efficiently synthesized following
2-[(2-Aminophenyl)amino]benzoic acid
the optimization of conventional methods (Boyerl et al.
1
991). The structures of all of the compounds synthesized
A mixture of o-chlorobenzoic acid (6.23 g, 10 mmol), o-
phenylenediamine (1.10 g, 10.1 mmol), anhydrous potas-
sium carbonate (2.72 g, 20 mmol), and copper powder (240
mg) in dry N,N-dimethylformamide (DMF) (40 mL) was
stirred for 8 h at 170 °C under an N2 atmosphere. The
mixture was then cooled to room temperature and filtered to
remove the copper powder. The filtrate was acidified to pH 1
with concentrated hydrochloric acid to give a precipitate,
which was collected by filtration and purified by recrys-
tallization from ethanol to give the desired product as a
in the current study were fully characterized using infrared
1
13
ray (IR), H-nuclear magnetic resonace (NMR), C-NMR,
and high-resolution mass spectrometry (HRMS). The X-ray
crystal structure of compound 1g was determined using X-
ray crystallography. All of the Schiff base compounds
prepared in the current study were screened in terms of their
antitumor activities against several human cancer cell lines
in vitro, including A549, HeLa, SGC-7901, and Raji cell
lines. The results show that compound 1d is potently
cytotoxic against HeLa cells (IC50 = 7.75 μM). We also
performed quantitative structure–activity relationship
1
yellow solid in 90% yield. M.p.: 208.2–209.9 °C. H NMR
(400 MHz, DMSO-d ): δ 9.04 (s, 1H), 7.86 (dd, 1H, J = 7.9,
6
(
QSAR) studies for compounds exhibiting inhibitory
1.6 Hz), 7.30 (dt, 1H, J = 7.2, 1.6 Hz), 7.04 (dd, 1H, J = 7.8,
1.6Hz), 6.99 (dt, 1H, J = 7.8, 1.4 Hz), 6.83 (dd, 1H, J = 7.9,
1.2 Hz), 6.67 (dt, 1H, J = 7.9 and 0.8 Hz), 6.63 (m, 2H).
activity against Raji cells, which resulted in a statistically
significant CoMFA model with high predictive abilities
(
2
2
R
= 0.674, R = 0.952).
cv
Multidrug resistance (MDR) can lead to a decrease in the
4-Aminoacridone
intracellular concentration of a drug, which can have an
adverse impact on the therapeutic use and effectiveness of
chemotherapeutic agents (Krishna and Mayer 2000). P-
glycoprotein (P-gp), which is a plasma membrane protein
and member of the ABC family of proteins, plays an
important role in MDR (Kumar et al. 2014). It is noteworthy
that acridone derivatives modulate MDR mechanisms that
occur through ATP binding and transmembrane domain of
P-gp (Zhang et al. 2015). Docking analysis was also used in
the current study to identify the binding pattern of 1d in the
active site of P-gp, as well as evaluating the MDR mod-
ulatory potential of the other compounds.
2-[(2-aminophenyl)amino]benzoic acid (2.28 g) was subjected
to a dehydration reaction in polyphosphoric acid (PPA; 60
mL) for 4 h at 120 °C. The reaction was then cooled to room
temperature and poured into water (1000 mL). The resulting
aqueous solution was neutralized (pH 7.0–8.0) by the addition
of an aqueous solution of NaOH, resulting in the formation of
a yellow precipitate. The precipitate was collected by filtration
and washed with water, followed by petroleum ether, to give
the crude product, which was recrystallized from EtOH to
give the desired product as a yellow solid in 73% yield. M.p.
1
341.0–342.5 °C. H NMR (400 MHz, DMSO-d ): δ 10.63
6
(
(
s, 1H), 8.22 (d, 1H, J = 9.2 Hz), 7.74–7.70 (m, 1H), 7.67
d, 1H, J = 7.6 Hz), 7.54 (t, 1H, J = 4.4 Hz), 7.03–7.02 (m,
Material and methods
Chemistry
2H), 5.56 (s, 2H).
General procedure for the synthesis of the 4-aminoacridone
Schiff bases
All of the chemical reagents used in the current study were
acquired from various commercial suppliers and purified
using standard methods before being used. Melting points
4-Aminoacridone (210 mg, 1.0 mmol) was added to abso-
lute ethanol (20 mL), and the resulting mixture was stirred
until it formed a solution. An aromatic aldehyde derivative
(1.5 mmol) was then added to the solution in a dropwise
manner, followed by two drops of acetic acid, and the
resulting mixture was heated at reflux for 2–4 h. The
(
(
M.p.) were determined using a digital M.p. apparatus
Shenke, Shanghai, China). IR spectra were recorded using a
Bruker Alfer FT-IR spectrometer (Bruker, Karlsruhe, Ger-
1
many). H NMR spectra were recorded in DMSO-d using a
6

Med Chem Res
reaction mixture was then cooled to room temperature and
evaporated at room temperature to produce a residue, which
was purified using column chromatography over silica gel
to give the crude product. The crude products were
recrystallized from ethanol.
128.35(C,C-7), 126.30(C,C-15), 125.84(C,C-11), 123.62(C,
C-13), 121.77(C,C-6), 121.70(C,C-8), 121.10(C,C-20),
120.55(C,C-1), 119.07(C,C-5), 115.75(C,C-2), 112.36,
+
56.24(CH –O). HRMS: m/z 345.1237 for [M + H] , calcd
3
+
for C H N O = 345.1234.
2
0 17 2 3
4
-(2-Hydroxylbenzylideneamino)acridin-9(10H)-one (1a)
4-(4-N,N-Dimethylbenzylideneamino)acridin-9(10H)-one
1d)
(
Isolated as a brown solid in 73% yield. M.p. 233.2–234.5 °C.
−
1
IR (cm ) 3394, 1621 (νC=N), 1624, 1335, 910, 744, 557.
Isolated as a brown solid in 67% yield. M.p. 254.5–257.2 °C.
1
−1
H NMR (400 MHz, DMSO-d ): δ 11.60 (s, 1H), 10.93
s, 1H), 9.03 (s, 1H), 8.27 (d, 1H, J = 7.2 Hz), 8.19 (d, 1H,
IR (cm ): 3417, 3322, 1639 (νC=N), 1608, 1508, 1474,
6
1
(
1395, 1250, 756, 656. H NMR (400 MHz, DMSO-d ): δ
6
J = 8.0 Hz),7.98–7.95 (m, 2H), 7.76–7.73 (m, 1H), 7.60
10.77 (s, 1H), 8.72 (s, 1H), 8.27 (d, 1H, J = 6.8 Hz),
8.09–8.04 (m, 4H), 7.76 (t, 1H, J = 8.4 Hz), 7.63 (d, 1H, J
= 6.4 Hz), 7.30–7.25 (m, 2H), 6.88 (d, 2H, J = 9.2 Hz),
(
(
dd, 1H, J = 7.2 and 7.6 Hz), 7.51–7.46 (m, 1H), 7.32–7.27
1
3
m, 2H), 7.06 (t, 1H, J = 7.6 Hz). C NMR (125 MHz,
177.11(C=O), 163.63(C–OH), 160.00
C=N), 141.24(C,C-10), 139.83(C,C-4), 135.74(C,C-2),
1
3
DMSO-d6):
(
δ
3.07 (s, 6H). C NMR (125 MHz, DMSO-d ): δ 177.29
6
(C=O), 161.02(C=N), 153.23(C,C-18), 141.05(C,C-10),
140.59(C,C-4), 136.509(C,C-2), 133.589(C,C-7), 132.029
(C,C-11), 124.25(C,C-13), 122.96(C,C-16=C-20), 121.69
(C,C-6), 121.55(C,C-15), 121.07(C,C-8), 120.21(C,C-1),
1
34.21(C,C-7), 133.76(C,C-18), 132.35(C,C-20), 126.28(C,
C-11), 124.37(C,C-13), 122.53(C,C-6), 121.92(C,C-8),
21.72(C,C-1), 121.58(C,C-19), 121.18(C,C-15), 121.15(C,
1
C-5), 119.77(C,C-12), 118.94(C,C-3), 117.06(C,C-17).
119.05(C,C-5), 111.77(C,C-17=C-19), 79.64(C,CH –N),
3
+
+
HRMS: m/z 315.1134 for [M + H] , calcd for
79.30(C,CH –N). HRMS: m/z 342.1608 for [M + H] ,
3
+
+
C H N O = 315.1128.
calcd for C H N O = 342.1601.
20 20 3
2
0
15
2
2
4
-(Benzylideneamino)acridin-9(10H)-one (1b)
4-(4-Bromobenzylideneamino)acridin-9(10H)-one (1e)
Isolated as a brown solid in 82% yield. M.p. 241.5–244.2 °C,
Isolated as a brown solid in 87% yield. M.p. 256.5–258.2 °C.
−
1
−1
IR (cm ): 3449, 1625 (νC=N), 1623, 1477, 1449, 755, 688.
IR (cm ): 3412, 3224, 1620 (ν
), 1598, 1569, 1523,
C=N
1
1
H NMR (400 MHz, DMSO-d ): δ 10.82 (s, 1H), 8.94 (s,
1483, 1449, 1341, 1066, 1008. H NMR (400 MHz,
6
1
1
7
1
4
H), 8.28–8.23 (m, 3H), 8.19 (d, 1H, J = 8.0 Hz), 8.06 (d,
DMSO-d ): δ 10.83 (s, 1H), 8.94 (s, 1H), 8.27 (d, 1H, J =
6
H, J = 8.0 Hz), 7.77–7.70 (m, 2H), 7.62–7.60 (m, 3H),
8.4 Hz), 8.21–8.17 (m, 3H), 8.05 (d, 1H, J = 8.4 Hz), 7.84
(d, 2H, J = 8.8 Hz), 7.78–7.72 (m, 2H), 7.33–7.28 (m, 2H).
1
3
.33–7.28 (m, 2H). C NMR (125 MHz, DMSO-d ): δ
6
1
3
77.18(C=O), 161.96(C=N), 141.12(C,C-10), 139.72(C,C-
), 136.49(C,C-15), 136.43(C,C-2), 133.72(C,C-7), 132.40
C NMR (125 MHz, DMSO-d ): δ 177.17(C=O), 160.79
6
(C=N), 141.12(C,C-10), 139.35(C,C-4), 136.55(C,C-15),
135.63(C,C-2), 133.79(C,C-7), 132.26(C,C-11), 132.03(C,
C-17=C-19), 126.31(C,C-13), 126.06(C,C-16=C-20),
124.74(C,C-6), 121.91(C,C-18), 121.81(C,C-8), 121.46(C,
C-1), 121.14(C,C-5), 121.03(C,C-12), 119.00(C,C-3).
(
(
(
C,C-18), 130.24(C,C-11), 129.17(C,C-16=C-20), 126.29
C,C-13), 124.43(C,C-17=C-19), 121.85(C,C-6), 121.75
C,C-8), 121.47(C,C-1), 121.13(C,C-5), 120.98(C,C-12),
+
1
19.02(C,C-3). HRMS: m/z 299.1182 for [M + H] , calcd
+
+
for C H N O = 299.1179.
HRMS: m/z 377.0287 for [M + H] , calcd for
2
0 15 2
+
C H BrN O = 377.0284.
2
0
14
2
4
-(3-Methoxyl-4-hydroxylbenzylideneamino)acridin-9
(
10H)-one (1c)
4-(4-Fluorobenzylideneamino)acridin-9(10H)-one (1f)
Isolated as a brown solid in 87% yield. M.p. 216.5–219.6 °C.
IR (cm ): 3422, 1624 (νC=N), 1523, 1288, 755. H NMR
Isolated as a yellow solid in 76% yield. M.p. 205.2–208.2 °C.
−
1
1
−1
IR (cm ): 3144, 3034, 1618 (νC=N), 1522, 1219, 1002,
1
(
400 MHz, DMSO-d ): δ 10.83 (s, 1H), 9.89 (s, 1H), 8.77
836, 755. H NMR (400 MHz, DMSO-d ): δ 10.83 (s, 1H),
6
6
(
s, 1H), 8.27 (d, 1H, J = 7.6 Hz), 8.12 (d, 1H, J = 7.2 Hz),
8.95 (s, 1H), 8.28 (d, 1H, J = 8.4 Hz), 8.19 (dd, 1H, J = 8.8
Hz), 8.05 (d, 1H, J = 8.4 Hz), 7.78–7.67 (m, 4H), 7.33–7.28
8
1
8
.05 (d, 1H, J = 7.6 Hz), 7.92 (d, 1H, J = 1.6 Hz), 7.77 (t,
H, J = 8.4 Hz), 7.65 (d, 1H, J = 7.2 Hz), 7.58 (dd, 1H, J =
.4 Hz), 7.31–7.27 (m, 2H), 6.98 (d, 1H, J = 7.6 Hz), 3.96
1
3
(m, 2H). C NMR (125 MHz, DMSO-d ): δ 177.18(C=O),
6
160.51(C=N), 141.10(C,C-10), 139.49(C,C-4), 136.51(C,
C-2), 133.71(C,C-7), 132.67(C,C-15), 126.30(d, 1C, J =
56 Hz), 124.46(C,C-16=C-20), 121.83(C,C-11), 121.41(C,
C-13), 121.12(C,C-6), 120.89(C,C-8), 118.97(C,C-5),
1
3
(
s, 3H). C NMR (125 MHz, DMSO-d ): δ 177.24(C=O),
6
1
61.32(C=N), 151.28(C,C-18), 148.43(C,C-17), 141.04(C,
C-10), 140.12(C,C-4), 136.45(C,C-10), 133.64(C,C-2),

Med Chem Res
1
16.36(C,C-12), 116.14(C,C-3). HRMS: m/z 317.1092 for
C-6), 121.49(C,C-8), 121.14(C,C-5), 120.76(C,C-12),
+
+
[
M + H] , calcd for C H FN O = 317.1085.
119.08(C,C-3), 107.64(C, C-16=C-20), 60.72(C, C-21=C-
2
0
14
2
+
2
3), 56.58(C,C-22). HRMS: m/z 389.1499 [M + H] , calcd
+
4
-(2-Bromobenzylideneamino)acridin-9(10H)-one (1g)
for C H N O = 389.1496.
23 21 2 4
Isolated as a brown solid in 72% yield. M.p. 211.2–213.2 °C.
4-(3,4-Dimethoxylbenzylideneamino)acridin-9(10H)-one
(1j)
−1
IR (cm ): 3064, 1619 (νC=N), 1561, 1523, 1477, 1340,
1
1
1
7
261, 1183, 754, 675. H NMR (400 MHz, DMSO-d ): δ
6
0.88 (s, 1H), 8.92 (s, 1H), 8.52 (s, 1H), 8.62 (dd, 1H, J =
.6 Hz), 8.28 (dd, 1H, J = 8.4 Hz, 8.0 Hz), 8.21 (dd, 1H, J
Isolated as a yellow solid in 83% yield. M.p. 187.3–189.2 °C.
−1
IR (cm ): 3415, 3330, 1619 (νC=N), 1599, 1508, 1342,
1
=
=
8.0 and 8.4 Hz), 8.02 (d, 1H, J = 9.2 Hz), 7.83 (d, 1H, J
8.0 Hz), 7.77–7.73 (m, 1H), 7.70 (dd, 1H, J = 8.0 Hz),
1275, 1240, 1139, 1018, 860, 757, 636. H NMR (400
MHz, DMSO-d ): δ 10.86 (s, 1H), 8.85 (s, 1H), 8.28 (d, 1H,
6
7
2
1
.66 (t, 1H, J = 7.6 Hz), 7.58–7.54 (m, 1H), 7.34–7.28 (m,
J = 8.0 Hz), 8.05 (d, 1H, J = 8.4 Hz), 7.96 (d, 1H, J = 1.6
Hz), 7.77–7.72 (m, 1H), 7.67–7.65 (m, 2H), 7.31 (t, 2H, J
= 8.0 Hz), 7.17 (d, 1H, J = 8.4 Hz), 3.95 (s, 3H), 3.89 (s,
1
3
H). C NMR (125 MHz, DMSO-d ): δ 177.11(C=O),
6
59.84(C=N), 141.14(C,C-10), 139.64(C,C-4), 136.44(C,
1
3
C-15), 134.42(C,C-2), 134.07(C,C-17), 133.80(C,C-7),
33.70(C,C-11), 128.53(C,C-18), 126.30(C,C-13), 126.19
C,C-19), 125.06(C,C-20), 121.95(C,C-6), 121.79(C,C-8),
21.57(C,C-16), 121.31(C,C-1), 121.31(C,C-5), 121.17(C,
3H). C NMR (125 MHz, DMSO-d ): δ 177.22(C=O),
6
1
(
1
161.24(C=N), 152.78(C,C-18), 149.43(C,C-17), 141.07(C,
C-10), 139.95(C,C-4), 133.65(C,C-2), 129.48(C,C-7), 126.
30(C,C-15), 125.73(C,C-11), 123.87(C,C-13), 121.78(C,C-
6), 121.74(C,C-20), 121.48(C,C-8), 121.12(C,C-5), 120.66
+
C-12), 118.99(C,C-3). HRMS: m/z 377.0287 for [M + H] ,
calcd for C H BrN O = 377.0284.
+
(C,C-12), 119.10(C,C-3), 111.64(C,C-16), 111.32(C,C-19),
2
0
14
2
+
5
6.13(C,C-21=C-22). HRMS: m/z 359.1396 for [M + H] ,
+
4
-(3-Fluorobenzylideneamino)acridin-9(10H)-one (1h)
calcd for C H N O = 359.1390.
22 19 2 3
Isolated as a yellow solid in 70% yield. M.p. 219.5–222.0 °C.
IR (cm ): 3022, 1629 (νC=N), 1525, 1348, 747, 551. H
4-(2,4-Dichlorobenzylideneamino)acridin-9(10H)-one (1k)
Isolated as a yellow solid in 69% yield. M.p. 161.3–163.2 °C.
−
1
1
NMR (400 MHz, DMSO-d ): δ 10.79 (s, 1H), 8.98 (s, 1H),
6
−
1
1
8
.28 (dd, 1H, J = 8.0 Hz), 8.23–8.18 (m, 2H), 8.06 (d, 1H,
IR (cm ): 3417, 1647 (νC=N), 1523, 1473, 1295, 755. H
J = 8.0 Hz), 8.00 (d, 1H, J = 7.6 Hz), 7.79–7.75 (m, 2H),
NMR (400 MHz, DMSO-d ): δ 10.85 (s, 1H), 9.10 (s, 1H),
6
7
.69–7.64 (m, 1H), 7.50–7.45 (m, 1H), 7.34–7.29 (m, 1H).
8.68 (d, 1H, J = 8.4 Hz), 8.27 (d, 1H, J = 8.0 Hz), 8.21 (d,
1H, J = 6.0 Hz), 8.01 (d, 1H, J = 8.8 Hz), 7.84 (d, 1H, J =
6.0 Hz), 7.76–7.67 (m, 3H), 7.34–7.28 (m, 2H). C NMR
1
3
C NMR (125 MHz, DMSO-d ): δ 177.20(C=O), 164.12
6
1
3
(
1
C–F), 161.70(C–N), 141.11(C,C-10), 139.03(C,C-4),
38.91(C,C-15), 136.63(C,C-2), 133.77(C,C-7), 131.23(C,
C-19), 127.16(C,C-11), 126.31(C,C-13), 124.94(C,C-6),
21.83(C,C-20), 121.43(C,C-9), 121.13(C,C-5), 121.01(C,
(125 MHz, DMSO-d ): δ 177.09(C=O), 156.25(C=N),
6
141.11(C,C-10), 139.26(C,C-4), 137.64(C,C-2), 136.58(C,
C-7), 136.54(C,C-15), 133.82(C,C-16), 132.12(C,C-11),
131.75(C,C-20), 129.93(C,C-13), 128.33(C,C-18), 126.31
(C,C-6), 125.30(C,C-8), 121.96(C,C-5), 121.82(C,C-12),
1
C-12), 119.29(C,C-3), 119.08(C,C-18), 115.57(C,C-16).
HRMS: m/z 317.1092 for [M + H] , calcd for
C H FN O = 317.1085.
+
+
+
118.95(C,C-3). HRMS: m/z 367.0400 for [M + H] , calcd
2
0
14
2
+
for C H Cl N O = 367.0399.
2
0
13
2 2
4
-(3,4,5-Trimethoxylbenzylideneamino)acridin-9(10H)-one
(
1i)
Single crystal cultivation and structure identification
Isolated as a yellow solid in 85% yield. M.p. 195.7–198.2 °C.
Crystal growth and conditions: yellow crystals of 1g sui-
table for X-ray crystallography were obtained at room
temperature from an EtOH solution by slow evaporation of
the solvent. The resulting crystals were then mounted on the
goniometer of a single crystal diffractometer. Crystal data
were collected at 296 K using Mo Kα radiation (λ =
0.710713 Å) in the w/v scan mode and analyzed for Lorentz
and polarization effects siemens area detector absorption.
The structure was solved using the direct method and
refined by full-matrix least-squares fitting on F2 using
SHELX-97.
−1
IR (cm ): 3431, 3385, 1685 (νC=N), 1587, 1505, 1463,
1
1
392, 1331, 1234, 1128, 992, 846, 628. H NMR (400
MHz, DMSO-d ): δ 10.86 (s, 1H), 8.85 (s, 1H), 8.28 (d, 1H,
6
J = 9.6 Hz), 8.17 (d, 1H, J = 9.2 Hz), 8.04 (d, 1H, J = 8.4
Hz), 7.72–7.30 (m, 1H), 7.68 (dd, 1H, J = 7.6 Hz), 7.58 (s,
1
3
2
H), 7.33–7.27 (m, 2H), 3.95 (s, 6H), 3.79 (s, 3H).
C
NMR (125 MHz, DMSO-d ): δ 177.20(C=O), 161.40
6
(
C=N), 153.54(C,C-17=C-19), 141.31(C,C-10), 141.04(C,
C-18), 139.65(C,C-4), 136.46(C,C-2), 133.71(C,C-11),
31.85(C,C-15), 126.31(C,C-7), 124.27(C,C-13), 121.85(C,
1

Med Chem Res
Cytotoxicity assay against A549, Hela, SGC-7901, and
Raji cell lines
validation and the SAMPLS program were used to obtain
the optimal number of components (Noc) and cross-
validated coefficient (R cv). After the optimal Noc was
2
Cells were cultured in RPMI-1640 medium with 10% fetal
determined, anon-cross-validated analysis was performed
bovine serum, 100 U/mL of penicillin and 100 μg/mL of
without column filtering to obtain the regression coefficients
(R ), which determine the external predictive ability.
2
streptomycin in a 5% CO humidified atmosphere at 37 °C.
2
Cytotoxicity was evaluated using an MTT assay. Briefly,
cells were seeded in 96-well tissue culture plates and incu-
bated with compounds (0–100 μmol/L) for 48 h at 37 °C and
Molecular docking
5
% CO . Following compound treatment, MTT (0.5 mg/
2
The 3D structures of P-gp (PDB ID: 1MV5 and 3G60) were
downloaded from the Protein Data Bank (http://www.rcsb.
org/pdb/home/home.do). The 3D structures of 1d were
drawn using the ChemBioDraw Ultra 12.0 and ChemBio3D
Ultra 12.0 software packages (CambridgeSoft, MA, USA).
The AutoDockTools 1.5.6 package (http://mgltools.scripps.
edu) was used to generate the docking input files. The
search grid for the transmembrane binding site of P-gp was
identified as center_x: 19.133, center_y: 52.68 and center_z:
mL) was added, and the resulting mixtures were incubated
for 4 h. The reaction was then terminated by the removal of
the supernatant and the remaining formazan crystals in each
well were dissolved in 150 μL of DMSO. The optical
density of each well was measured at 570 nm using a
PowerWaveX Microplate Scanning Spectrophotometer
(
Bio-tek Instruments, Inc., Vermont, America). The per-
centage inhibition of cell proliferation of each compound
was then calculated at various concentrations.
0
1
.174 with dimensions of size_x: 15, size_y: 15 and size_z:
5, whereas the search grid for the ATP binding site of P-gp
was identified as center_x: 3.091, center_y: 52.697 and
center_z: 107.494, with dimensions of size_x: 15, size_y: 15
and size_z: 15. The value of exhaustiveness was set at 20.
The default parameters were used for the Vina docking
experiments. The best-scoring pose as judged by the Vina
docking score was chosen and visually analyzed using
version 1.7.6 of the PyMOL 1.7.6 (http://www.pymol.org/).
3
D-QSAR studies
The 11 compounds generated in the current study were
divided into training and testing sets including compounds
1
randomly selected. The inhibition ratio values were con-
verted into logarithmic values (inhibition ratio × 100) for the
QSAR study. The three-dimensional (3D) structures of these
compounds were built using the SYBYLx2.0 software
c and 1i, respectively. The testing set compounds were
(
Tripos, SaintLouis, America). Partial atomic charges were
Results and discussion
Chemistry
calculated using the Gasteiger–Huckel method, and energy
minimizations were performed using the Tripos force field
and the Powell conjugate gradient algorithm with a con-
vergence criterion of 0.001 kcal/mol. The steric and elec-
trostatic interaction fields for CoMFA were calculated using
the default parameters from SYBYL: 2.0 kcal/mol grid point
The target compounds were synthesized as outlined in
Scheme 1. The properties of acridone derivatives have tra-
ditionally been difficult to evaluate because of the lack of an
efficient synthetic method for the construction of these
compounds. Many scientists have obtained satisfactory
results through the development of novel synthetic methods
(Shi et al. 2013). Using a conventional method, compound
1 was synthesized in DMF in the absence of a nitrogen
protecting group in 83% yield, and 4-aminoacridone was
synthesized in the strong oxidizing/dehydrating agent
3
spacing, sp carbon probe atom with +1 charge and a van
der Waals radius of 1.52 Å, and column filtering of 2.0 kcal/
mol. The descriptors calculated from the CoMFA analysis
were used as independent variables and the experimental
logarithmic 100 × inhibition ratio (lgIR) values were used as
dependent variables in a partial least-squares (PLS) analysis
to derive the 3D-QSAR model. Leave-one-out cross-
Scheme 1 Synthesis of 4-
aminoacridone and its
derivatives (yields: 67–87%)

Med Chem Res
H SO with a low yield of 31% (Boyerl et al. 1991). To
Biological activity
2
4
provide reliable access to large quantities of 4-aminoacri-
done, we investigated the optimization of the experimental
conditions. The results of a series of screening experiments
revealed that the initial reaction of 2-chlorobenzoic acid
with benzene-1,2-diamine in the presence of copper powder
All of the Schiff base compounds (1a–k) synthesized in the
current study and 4-aminoacridone were evaluated to
determine their cytotoxicity against several human cancer
cell lines, including A549, HeLa, SGC-7901, and Raji cells.
Cisplatin was used as a positive control. These experiments
revealed that most of these Schiff bases exhibit potent
cytotoxicity towards HeLa and Raji cells. However, only a
few compounds exhibited cytotoxicity towards A549 cells,
with most compounds also showing low activity against
SGC-7901 cells. The results of these experiments are
summarized in Table 1, where the activities of the different
compounds are presented as inhibition ratios and IC50
values.
As shown in Table 1, compound 1d exhibited the highest
activity of all of the compounds tested against the HeLa
cells, with an IC50 value of 7.75 µM. In contrast, com-
pounds 1a and 1k showed lower levels of activity, with IC50
values greater than 30 μM. Compounds 1c, 1e, 1f, and 1i
also showed weak activity against HeLa cells. Analysis of
the structure–activity relationships for these compounds
revealed that the nature of the substituent groups on the
phenyl ring has a pronounced effect on their activities
against HeLa cells. For example, compounds such as 1d,
and potassium carbonate in DMF gave the S Ar product 2-
N
[
(2-aminophenyl)amino]benzoic acid with an isolated yield
of up to 90%. The subsequent dehydration of 2-[(2-ami-
nophenyl)amino]benzoic acid in the presence of the mild
dehydrating agent PPA, resulted in the formation of 4-
aminoacridone in a much higher isolated yield of 73%
whilst avoiding the need for a strongly oxidizing dehydra-
tion reagent. The Schiff bases were synthesized using a
conventional condensation method. The X-ray crystal
structure of 1g was solved and it clearly shows the desired
Schiff base functional group (Fig. 1).
1
k, 1a, and 1c, bearing a Cl, OH, or N(Me) group at R or
2 1
R displayed greater cytotoxicity. Compound 1k, bearing a
3
2,4-dichlorosubstituted phenyl ring or compound 1d, bear-
ing a larger –N(Me) substituent at the 4-position of its
2
phenyl ring, exhibited potent activity against HeLa cells.
However, these two compounds showed much weaker
activities against the other three tested cell lines, high-
lighting their selectivity for HeLa cells.
Fig. 1 X-Ray crystal structure of compound 1g
Table 1 Antitumor activities
Compd A549
HeLa
SGC-7901
Raji
(μmol/L) of the Schiff base
derivatives of 4-aminoacridone
Inhibition
ratio (%)
IC50
Inhibition
(μM) ratio (%)
IC50
(μM)
Inhibition
ratio (%)
IC50 (μM) Inhibition
IC50
(μM)
ratio (%)
Cisplatin
5.34
4.62
2.27
5.20
a
4
1
1
1
1
1
1
1
1
1
1
1
a
-AA
22.41
21.27
42.43
34.14
39.55
65.11
44.50
76.70
45.18
26.99
33.30
77.28
41.02
25.23
28.42
34.02
47.60
56.73
46.00
62.65
82.13
a
b
c
d
e
f
71.33
24.0
63.21
13.74
66.02
36.50 0.67
7.75 0.74
64.77
62.21
63.96
95.61
48.12 74.36
78.24
56.62 44.26
104.25
53.13
62.65
30.63
34.00
33.00
67.49
60.27
61.65
52.5
8.43
g
h
i
35.64 32.37
54.41
75.58
82.62 22.97
66.69
59.5
/
/
60.86
73.41
87.03
j
43.12
k
45.48 76.26
8.88 63.39
72.63
Expressed 4-aminoacridone

Med Chem Res
Table 2 lgIR (logarithmic 100 × inhibition ratio) of the Schiff bases
of 4-aminoacridone against Raji cells
3
D-QSAR analysis
The majority of the Schiff base compounds synthesized in
the current study exhibit better antitumor activity towards
the HeLa and Raji cells than other cells. Consequently, a
QSAR study on the biological testing data for the HeLa and
Raji cells was performed. However, a very low regression
compd. lgIR
exp. cal.
compd. lgIR
exp. cal.
err.
err.
1a
1b
1c
1.754 1.781 −0.027 1g
1.531 1.550 −0.019
1.663 1.617
1.797 1.718
0.046 1h
0.079 1i
1.519 1.489
1.829 1.805
1.780 1.780
0.030
0.024
0.000
2
coefficient (R ) was obtained for the HeLa cells, although a
better result was obtained for the Raji cells. A fragment of
the structure of compound 1d was used as a template to
align the other compounds. The statistical results of the 3D-
QSAR model show that the cross-validated correlation
1d
1e
1f
1.915 1.918 −0.003 1J
1.797 1.767 −0.030 1k
1.486 1.539 −0.053
1.790 1.794 −0.004
2
coefficient R = 0.719 > 0.5, corresponding to N = 3,
cv
oc
whereas the 3D-QSAR of the non-cross-validated model
gave R = 0.952, S = 0.081, and F = 46.497. These results
E
indicate that the model provided good correlation and
accuracy characteristics for prediction of the compound
activities. The contributions of the steric and electrostatic
fields in the 3D-QSAR model were 58.8 and 41.2%,
respectively.
All of the synthesized compounds were randomly divi-
ded into two groups: 1c and 1i were used as a testing set,
whereas the remaining nine compounds were used as a
training set. PLS analysis was also conducted using com-
Fig. 2 CoMFA contour maps: a steric contour map b electrostatic
contour map
position of its phenyl ring. The two yellow contours around
pound 1b from the training set as a template for all of the
the R and R positions indicate that the introduction of
1
2
2
other compounds. The results of this analysis were R cv
=
small groups at these positions would lead to an improve-
ment in their activity. The CoMFA electrostatic fields in
blue show the regions where the occurrence of a positive
charge would lead to an increase in activity, whereas the red
regions indicate the areas where the introduction of nega-
tively charged groups would be favorable.
0
.674 > 0.5, N = 3. The 3D-QSAR of the non-cross-
oc
validated model gave R = 0.956, S = 0.040 and F =
E
3
6.588. It is clear from these results that the estimated
values are very similar to the test values.
Therefore, these results indicate that the CoMFA model
has a good predictive capability. The predicted lgIR values
are generally in good agreement with the experimental data
In the electrostatic map (Fig. 2b), the two red contours
around R and R indicate that the introduction of electro-
1
4
(
Table 2). The results of the CoMFA indicate that the
negative groups at these positions would lead to an increase
contributions of the steric and electrostatic fields in the 3D-
QSAR model are 64.2 and 35.8%, respectively. The steric
and electrostatic contribution contour maps of the model are
depicted in Fig. 2. The 3D contour maps show that changes
in the molecular fields are associated with differences in
biological activity. The green and yellow areas show the
steric fields, whereas the green contour regions suggest that
the introduction of larger substituents at these positions
would lead to an improvement in the biological activity.
Finally, the yellow region indicates that the introduction of a
bulky group would have an adverse impact on the inhibitory
activity.
in biological activity, especially at the R position. The
1
small blue contours surrounding the R substituent suggest
3
that the introduction of an electropositive group would
increase activity at the 3-position, whereas the red contour
at the 1-position indicates that the introduction of electro-
negative groups at this position would also be favorable for
increased activity.
Analysis of compound 1d, which exhibits the greatest
activity of all the compounds tested against Raji cells,
revealed that it has a large electropositive –N(CH ) group
3
2
at the 4-position of its phenyl ring and small groups (–H) at
the 2, 3, 5, and 6 positions of the same ring, which conforms
to the steric and electrostatic contribution contour maps. In
contrast, compound 1F, with a small electronegative group
(–F) at the 4-position of its phenyl ring, did not match the
model, and has the weakest activity. Based on the CoMFA
model, it is envisaged that the introduction of a large
electropositive group at the 4-position, together with a small
electronegative group at the 2-position would improve
As shown in the steric contour map (Fig. 2a), the green
contour around R indicates that the introduction of a bulky
3
group at this position would lead to an increase in biological
activity. For example, compounds 1d, 1e, and 1f bearing a
–
N(CH ) , –Br or –F substituent at the 4-, 5-, or 6-position
3 2
of their phenyl ring, respectively, show better inhibitory
activity than compound X, bearing a hydrogen atom at the 2

Med Chem Res
biological activity. The synthesis of such an optimized
compound will be the focus of future work.
Docking analysis
Acridone derivatives have been reported to modulate MDR
by binding to the ATP and transmembrane domains of P-gp
(
Zhang et al. 2015). To reveal the MDR modulatory activity
and binding behavior of compound 1d, we conducted
molecular docking studies to investigate the binding mode
of 1d to P-gp using Autodock vina 1.1.2. The docking of
compound 1d into the transmembrane binding pocket led to
the formation of much better interactions compared with the
ATP binding pocket.
The theoretical binding mode of 1d to the transmem-
brane binding site of P-gp is shown in Fig. 3. The acridone
scaffold of 1d binds at the bottom of the pocket, where it
forms a series of high density van der Waals contacts,
whereas the N,N-dimethylphenyl group of 1d is located in a
hydrophobic pocket, surrounded by residues Ala-981, Ile-
Fig. 4 PyMoL image for the docking of compound 1k into the ATP
binding site of P-gp
Compound 1d docks into the ATP binding site of P-gp
and the theoretical binding mode between 1d and P-gp is
shown Fig. 4. The acridone scaffold of 1d binds at the
entrance to the pocket, where it forms a high density of van
der Waals contacts, whereas the N,N-dimethylphenyl moi-
ety of 1d is located at the bottom of the pocket, surrounded
by residues B/Glu-1356, B/Ile-1358, B/Gly-1379, and C/
Ser-2355. A detailed analysis showed that one of the phenyl
rings of the acridone core displays π–π stacking interactions
with the B/Tyr-1352 and C/Tyr-2352 amino acid residues.
Importantly, the carbonyl oxygen atom of the acridone ring
in 1d forms a series of essential coordination interactions
3
36, Met-68, and Tyr-949. Detailed analysis shows that two
π–π stacking interactions are present between the acridone
scaffold and the side chains of residues Phe-728 and Phe-
9
74. In addition, the phenyl group at the 10-position of the
acridone scaffold forms a π–π interaction with the side chain
of the Phe-332 residue. Importantly, the carbonyl oxygen
atom of the acridone core in 1d forms an essential hydrogen
bond with the Ser-975 residue, which is the main interaction
between 1d and P-gp.
2
+
with Mg , which is chelated by residue C/Asp-2353.
Conclusion
We have developed an optimized route for the synthesis of
4-aminoacridone based on conventional methods. This
material was subsequently used to prepare a series of Schiff
bases (1a–k), which were screened against several human
cancer cell lines to evaluate their antitumor activity. Nota-
bly, compounds 1d and 1k displayed potent antitumor
activity against HeLa cells. These data were used to gen-
2
erate a CoMFA model with R = 0.674 and R = 0.956.
cv
The low standard error of estimation and the high F values
confirm that this CoMFA model possesses a significant
prediction ability. The model predicts that the introduction
of a large electropositive group at the 4-position of 4-ami-
noacridone, as well as a small electronegative group added
to the 1,4-position, would lead to an improvement in the
activity against Raji cells. Docking analysis indicates that
1d forms stronger interactions with the transmembrane
Fig. 3 PyMoL image for the docking of compound 1k into the
transmembrane binding site of P-gp

Med Chem Res
Kumar R, Bahia MS, Silakari O (2014) Synthesis, cytotoxic activity,
binding pocket of P-gp than it does with the P-gp ATP
binding site, indicating that it could act as an MDR inhi-
bitor. Based on this analysis, it can be concluded that
compound 1d exhibits significant anti-cancer activity in
addition to its predicted effects against MDR.
1
0
and computational analysis of N -substituted acridone analogs.
Med Chem Res 24:1–13
Marques EF, Bueno MA, Duarte PD, Vieira PC, Corrêa AG (2012)
Evaluation of synthetic acridones and 4-quinolinones as potent
inhibitors of cathepsins L and V. Eur J Med Chem 54:10–21
Mohammadi-Khanaposhtani M, Saeedi M, Shafiee A, Akbarzadeh T
(
2015) Potent acetylcholinesterase inhibitors: design, synthesis,
Acknowledgements Lin Tian thanks Gui F. Dai, Zhengzhou Uni-
versity, Zhengzhou, for testing the activity of the compounds. The
authors thank the Chinese National Natural Science Foundation
biological evaluation, and docking study of acridone linked to
1,2,3-triazole derivatives. Eur J Med Chem 92:799–806
Mouafo RM, Roy R (2010) In vitro cytotoxic activity of isolated
acridones alkaloids from Zanthoxylum leprieurii Guill. et Perr.
Bioorg Med Chem 18:3605
(
31270577).
Sathish NK, GopKumar P, Prasad VVSR, Kumar SMS, Mayur YC
Compliance with ethical standards
(2010) Synthesis, chemical characterization of novel 1,3-dimethyl
acridones as cytotoxic agents, and their DNA-binding studies.
Med Chem Res 19:674–689
Shi F, Zhu RY, Liang X, Tu SJ (2013) Catalytic asymmetric 1,3-
dipolar cycloadditions of alkynes with isatin-derived azomethine
ylides: enantioselective synthesis of spiro[indoline-3,2′-pyrrole]
derivatives. Adv Synth Catal 355:2447–2458
Conflict of interest The authors declare that they have no competing
interests.
Swist A, Soloduch J (2013) Novel acridone-based branched blocks as
highly fluorescent materials. Syn Met 18:1–8
References
Tabarrini O, Dutartre H, Paeshuyse J, Neyts J (2006) Synthesis and
anti-BVDV activity of acridones as new potential antiviral agents.
J Med Chem 49:2621–2627
Boyerl G, Galyl J, Barbe J (1991) [Synthèse et caractérisation de bis
acridines pontées en positions 2, 3 ou 4]. J Heterocycl Chem
2
8:913–918
Wang XQ, Xia CL, Huang ZS (2015) Design, synthesis, and biolo-
gical evaluation of 2-arylethenylquinoline derivatives as multi-
functional agents for the treatment of Alzheimer’s disease. Eur J
Med Chem 89:349–361
Cao DR, Gao CM, Jiang YY, Tan CY (2008) Synthesis and potent
antileukemic activities of 10-benzyl-9(10H)-acridinones. Bioorg
Med Chem 16:8670–8675
Krishna R, Mayer LD (2000) Multidrug resistance (MDR) in cancer:
mechanisms, reversal using modulators of MDR and the role of
MDR modulators in influencing the pharmacokinetics of antic-
ancer drugs. Eur J Pharm Sci 11:265–283
Zhang B, Chen K, Wang N, Liu HX, Jiang YY (2015) Molecular
design, synthesis and biological research of novel pyridyl acri-
dones as potent DNA-binding and apoptosis-inducing agents. Eur
J Med Chem 93:214–226