Synthesis and evaluation of 7-substituted-5,6-dihydrobenzo[c]acridine derivatives as new c-KIT promoter G-quadruplex binding ligands

Article abstract of DOI:10.1016/j.ejmech.2017.02.051

It has been shown that treatment of cancer cells with c-KIT G-quadruplex binding ligands can reduce their c-KIT expression levels thus inhibiting cell proliferation and inducing cell apoptosis. Herein, a series of new 7-substituted-5,6-dihydrobenzo[c]acri

Full text of DOI:10.1016/j.ejmech.2017.02.051

Accepted Manuscript
Synthesis and evaluation of 7-substituted-5,6-dihydrobenzo[c]acridine derivatives as
new c-KIT promoter G-quadruplex binding ligands
Qian-Liang Guo, Hua-Fei Su, Ning Wang, Yu-Ting Lu, Tian-Miao Ou, Jia-Heng Tan,
Ding Li, Zhi-Shu Huang
PII:
S0223-5234(17)30124-1
10.1016/j.ejmech.2017.02.051
EJMECH 9247
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 27 December 2016
Revised Date: 18 February 2017
Accepted Date: 20 February 2017
Please cite this article as: Q.-L. Guo, H.-F. Su, N. Wang, Y.-T. Lu, T.-M. Ou, J.-H. Tan, D. Li, Z.-S.
Huang, Synthesis and evaluation of 7-substituted-5,6-dihydrobenzo[c]acridine derivatives as new c-KIT
promoter G-quadruplex binding ligands, European Journal of Medicinal Chemistry (2017), doi: 10.1016/
j.ejmech.2017.02.051.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphic Abstract
Synthesis and evaluation of 7-substituted-5, 6-dihydrobenzo[c]acridine derivatives as new
c-KIT promoter G-quadruplex binding ligands
Qian-Liang Guo†, Hua-Fei Su†, Ning-Wang, Yu-Ting Lu, Tian-Miao Ou, Jia-Heng Tan, Ding Li, Zhi-Shu Huang*
A series of new 7-substituted-5, 6-dihydrobenzo[c]acridine derivatives were synthesized and evaluated as c-KIT
G-quadruplexes DNA ligands.

ACCEPTED MANUSCRIPT
Title page
Synthesis and evaluation of 7-substituted-5,6-dihydrobenzo[c]acridine derivatives as new c-KIT
promoter G-quadruplex binding ligands
Qian-Liang Guo†, Hua-Fei Su†, Ning Wang, Yu-Ting Lu, Tian-Miao Ou, Jia-Heng Tan, Ding Li, Zhi-Shu Huang*
School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou, 510006, People’s Republic of China
*
To whom correspondence should be addressed. Tel. and Fax: 8620-39943056, E-mail: ceshzs@mail.sysu.edu.cn.
†
These authors contributed equally.
Abstract
It has been shown that treatment of cancer cells with c-KIT G-quadruplex binding ligands can reduce their
c-KIT expression levels thus inhibiting cell proliferation and inducing cell apoptosis. Herein, a series of new
7
-substituted-5,6-dihydrobenzo[c]acridine derivatives were designed and synthesized. Subsequent biophysical
evaluation demonstrated that the derivatives could effectively bind to and stabilize c-KIT G-quadruplex with
good selectivity against duplex DNA. It was found that 12-N-methylated derivatives with a positive charge
introduced at 12-position of 5,6-dihydrobenzo[c]acridine ring had similar binding affinity but lower stabilizing
ability to c-KIT G-quadruplex DNA, compared with those of nonmethylated derivatives. Further molecular
modeling studies showed possible binding modes of G-quadruplex with the ligands. RT-PCR assay and
Western blot showed that compound 2b suppressed transcription and translation of c-KIT gene in K562 cells,
which was consistent with the property of an effective G-quadruplex binding ligand targeting c-KIT oncogene
promoter. Further biological evaluation showed that compound 2b could induce apoptosis through activation
of the caspase-3 cascade pathway.
Keywords: G-quadruplex; c-KIT gene; 5,6-dihydrobenzo[c]acridine derivatives; transcriptional control; cell
apoptosis.
Abbreviations
CD, circular dichroism;
FRET, fluorescence resonance energy transfer;
MST, microscale thermophoresis;
MTT, methyl thiazolyl tetrazolium;
RT-PCR, reverse transcription-polymerase chain reaction;
SAR, structure-activity relationship;
SPR, surface plasmon resonance;
Tm, melting temperature;
UV, ultraviolet.
1

ACCEPTED MANUSCRIPT
1
. Introduction
G-quadruplex is made up of stacked G-tetrad subunits, where four coplanar guanines are linked
together through Hoogsteen hydrogen bonds [1, 2]. G-quadruplex motif widely exists in eukaryotic
genome [3], including human telomeric DNA [4], immunoglobulin switch regions, mutational hot
spots, and regulatory elements in oncogene promoters, such as c-MYC [5], VEGF [6], BCL-2 [7],
KRAS [8]. Two G-quadruplex forming motifs have been identified in c-KIT (c-KIT1 and c-KIT2)
promoter [9−12]. These unique quadruplex sequences play important roles in regulating biological
events, not only in cell growth, proliferation, and apoptosis, but also in the development and
metastasis of tumor cells. Therefore G-quadruplexes are recognized as promising targets for the
design of anti-cancer drugs.
The proto-oncogene c-KIT, which encodes a transmembrane tyrosine receptor, belongs to class
III of the receptor tyrosine kinase family [10, 11]. The c-KIT oncogene plays key roles in relaying an
extracellular signal to control cell growth, proliferation, differentiation and migration [12]. The
excessive activation of c-KIT or specific mutation in the c-KIT has been found to be associated with
gastrointestinal stromal tumors (GISTs), and kinase domain mutation facilitating auto
phosphorylation of c-KIT protein has been found in myeloid leukemias [13−15]. Previous studies
have shown that a small molecule Gleevec (imatinib mesylate) can effectively inhibit the c-KIT
kinase activity in vitro and in vivo [16]. However, resistance is emerging as a major clinical problem
in the administration of Gleevec for patients because of secondary mutations, amplification of KIT,
and alternative signaling kinases driving cell growth [17]. Recently it has been shown that treatment
of cells with c-KIT promoter G-quadruplex binding ligands can reduce their c-KIT expression levels
2

ACCEPTED MANUSCRIPT
thus exhibiting an alternative pathway for the design of small molecules inhibiting cell growth
[
18−22]. Therefore, the development of small molecules as c-KIT promoter G-quadruplex binding
ligands has been considered as a promising strategy to overcome the c-KIT protein mutation related
resistance.
Acridine is a fused polycyclic aromatic molecule, which is a main scaffold in some natural
acridine alkaloids. In recent years, acridine derivatives have been modified as telomeric [23-25] or
c-MYC G-quadruplex binding ligands [26, 27] and some excellent ligands have been showed in Fig.
1A, such as BOQ1 [20], RHPS4 [28], and BRACO-19 [23]. Some studies have shown that the fused
crescent-shaped planar structure can offer effective π-orbital stacking on the G-quadruplex [29]. In
our previous studies, 8-substituted 12-N-methylated-5,6-dihydrobenzo[c]acridine derivatives (Fig.
1
B) have been studied as binding ligands for c-myc G-quadruplex DNA [26]. In order to develop new
c-KIT G-quadruplex binding ligands, we designed and synthesized a series of 7-substituted-
,6-dihydrobenzo[c]acridine derivatives (Fig. 1C). These small molecules have a slightly-twisted
5
planar ring scaffold which contains a saturated C-C bond at the 5,6-position of acridine scaffold and
an expanded aromatic system with a fused benzene ring. In order to improve the G-quadruplex DNA
binding potency and selectivity of the derivatives, the open-loop methoxy group and cationic amino
side chains were introduced to the dihydrobenzo[c]acridine scaffold [23, 26]. In addition, in order to
know whether the introduction of a positive charge through methylation at nitrogen atom of
12-position of 5,6-dihydrobenzo[c]acridine scaffold could similarly increase their interactions with
c-KIT G-quadruplex through electrostatic interactions as shown in our previous results [29], two
series of derivatives (Series I and Series II) without or with 12-N-methyl group respectively were
3

ACCEPTED MANUSCRIPT
designed and synthesized. Moreover, the types of linkage (‒NH‒ and ‒CONH‒) between scaffold
and side chain with terminal amino group were also studied. Subsequent biophysical methods
including fluorescence resonance energy transfer (FRET), surface plasmon resonance (SPR),
microscale thermophoresis (MST), circular dichroism (CD) spectroscopy, ultraviolet (UV)
spectroscopy, and fluorescence titration experiments were used to evaluate the interactions of
acridine derivatives with c-KIT G-quadruplex DNA. Moreover, molecular modeling was also
employed for investigating the binding mode of the compounds with G-quadruplex DNA.
Furthermore, the inhibitory activities of compounds on transcription and expression of c-KIT gene in
K562 cells (a human erythroleukemic cell line) were studied by using reverse
transcription-polymerase chain reaction (RT-PCR) and Western blot. The effects of compounds on
cell proliferation and cell apoptosis were also evaluated by using MTT assay, flow cytometry, and
Western blot.
2
. Results and discussion
2.1. Chemistry
The synthetic routes for 7-substituted-5,6-dihydrobenzo[c]acridine derivatives were shown in
Scheme 1. The synthesis of key intermediate 11 [26],
-chloro-2,3-dimethoxy-5,6-dihydrobenzo[c]acridine was carried out with veratrole and succinic
7
anhydride as starting materials, and 4-(3,4-dimethoxyphenyl)-4-oxobutanoic acid (7) was obtained
o
by using Friedel-Crafts acylation in the presence of AlCl in CH Cl at 0-60 C. Then, reduction of
3
2
2
compound 7 with Et SiH in CF COOH, followed with intra-cyclization with PPA afforded
3
3
4

ACCEPTED MANUSCRIPT
intermediate 9. Condensation of 9 with 2-aminobenzoic acid by using Friedländer reaction gave
intermediate 10, which was then heated under reflux in POCl to give key intermediate 11 (Scheme
3
1A). Next, compounds 1a-1e and 2a-2e were synthesized through the introduction of side chains
(Scheme 1B). The introduction of alkylamino side chain into acridine scaffold was carried out by
initially refluxing in p-toluenesulfonic acid prior to treatment with diamines to give 1a-1e and 2a-2e.
In addition, methylation of intermediate 11 with methyl trifluoromethanesulfonate (MeOTf) in
toluene at room temperature afforded the salt 12. Similarly, the salt 12 could be transformed to 3a-3c
and 4a-4c with amines in THF (Scheme 1B). Finally, with the catalysis of KOH, compound 9 was
condensed with indoline-2,3-dione to afford acid 13, which could be successfully transformed to 5c,
6b, and 6d with amines by using BOP reagent (Scheme 1C).
2.2. Stabilizing ability and selectivity of the derivatives on c-KIT G-quadruplex
To determine the ligand induced increase in the melting temperature (∆T ) of a folded c-KIT
m
G-quadruplex [30], FRET-melting experiment was employed with c-KIT1 and c-KIT2 gene promoter
G-quadruplex sequence containing fluorophores at both 5′-end and 3′-end (F-c-KIT1-T,
5
5
′-FAM-d[AG AG CGCTG AGGAG ]-TAMRA-3′;
F-c-KIT2-T,
3
3
3
3
′-FAM-d[G CG CGCGAG AG )]-TAMRA-3′).
The
oligomer
F10T
3
3
3
4
(5′-FAM-d[TATAGCTATA-HEG-TATAGCT-ATA]-TAMRA-3′) that could self-associate into a hairpin
structure was used as a duplex DNA control. 8-substituted 12-N-methyl-5,6-dihydrobenzo[c]acridine
derivative 21c [26] previously reported by us was used as reference compound. The results shown in
Table 1 and Table S1 revealed that the new synthetic ligands had a wide range of stabilizing ability
5

ACCEPTED MANUSCRIPT
on the G-quadruplex and quite weak effects on the stability of duplex DNA. It was found that the
o
∆
T values were ranged from 3.4 to 22.9 C when F-c-KIT1-T was incubated with 10-fold excess
m
ligands in 60 mM KCl (Table 1). Most of nonmethylated compounds (1a-1c, 1e, and 2a-2e) showed
better stabilizing ability on c-KIT1 G-quadruplex than the reference compound 21c. To our surprise,
12-N methylated compounds 3a-3c and 4a-4c showed lower stabilizing ability for c-KIT1
G-quadruplex compared to corresponding nonmethylated compounds 1a-1c and 2a-2c. In addition,
o
because of a high self-stability of F-c-KIT2-T (T = 75.5±0.6 C, 60 mM KCl), the stabilizing
m
ability of compounds in a 5-fold excess was also studied for c-KIT1 and c-KIT2 G-quadruplex (Table
S1), As a result, the stabilizing abilities for c-KIT1 and c-KIT2 of the novel compounds were not
similar. Although most of nonmethylated compounds (1a-1c, 1e, and 2a-2e) similarly showed better
stabilizing ability on c-KIT2 G-quadruplex than methylated compounds (3a-3c and 4a-4c) or the
reference compound 21c, but the increase on melting temperatures on c-KIT2 G-quadruplex were
clearly lower than c-KIT1 G-quadruplex.
Analysis of the △T values of 7-substituted-5,6-dihydrobenzo[c]acridine derivatives gave the
m
following structure-activity relationship (SAR): (1) The introduction of an alkylamino side chain
dramatically increased the stabilization ability of compounds to G-quadruplex DNA. It was found
that intermediate 11 without the side chain hardly showed stabilizing effect on DNA. (2) The good
electron-donating effect of nitrogen atom of the linkage ‒NH‒ between scaffold and side chain
(1a-1e, 2a-2e) could increase the electron density of 5,6-dihydrobenzo[c]acridine, thereby achieving
effective stabilization and selective recognition of compounds on G-quadruplex [31, 32]. Thus, these
compounds had obviously higher stabilizing ability for G-quadruplex than the derivatives with a ‒
6

ACCEPTED MANUSCRIPT
CONH‒ linkage (5c, 6b, and 6d). (3) The distances apart of positive charges in the end of side chain
could influence the stabilizing ability of ligands to G-quadruplex. With the same terminal bases,
compounds 2a-2e with three methylene groups (n = 3) between the linkage ‒NH‒ and basic
N-terminus showed stronger stabilizing ability to the G-quadruplex than compounds 1a-1e with two
methylene groups (n = 2), and compound 2e showed the highest thermodynamic stability to
o
G-quadruplex (△T value of 22.9 C for F-c-KIT1-T, Table 1). Similar results could be found in
m
methylated compounds 3a-3c (n = 2) and 4a-4c (n = 3). The results indicated that the compounds
with the appropriate length of side chains were easy to locate in and interact with the groove of
G-quadruplex DNA [33]. (4) Among the compounds with the same length of side chain and different
basic terminal, the FRET-melting results showed that the compounds with 2-methylpiperazino
terminal (1e, 2e) had greater effect on stabilization of G-quadruplex than those with other types of
terminal, such as dimethylamino, diethylamino, pyrrolidino, and piperidino terminal.
To further evaluate the selectivity of the derivatives for G-quadruplex DNA over duplex DNA, a
FRET-based competition assay was carried with non-fluorescent complementary DNA ds26
(5′-GTTAGCCTAGCTTAAGCTAGGCTAAC-3′). As shown in Fig. 2, in the presence of 15- or
5
0-folds excess of competitor ds26, the △T values of F-c-KIT1-T raised by the ligands were only
m
slightly affected. The results combined with the data from FRET-melting assay indicated that all
compounds had good selectivity to G-quadruplex DNA over duplex DNA.
2.3. Binding affinity and selectivity of the derivatives on c-KIT G-quadruplex DNA
SPR experiments were carried out quantitatively by using biotinylated c-KIT1 G-quadruplex
7

ACCEPTED MANUSCRIPT
DNA and biotinylated duplex DNA attached to a streptavidin-coated sensor chip [34], which was
useful to investigate the binding affinity and selectivity of the synthesized acridine derivatives. The
binding constants (K ) were determined through equilibrium analysis as shown in Table 2.
D
It was found that most of the derivatives exhibited strong binding affinity to c-KIT1
G-quadruplex with their K values lower than 8.0 µM, while these compounds had weak or no
D
c-KIT1
obvious binding to duplex DNA. A comparison of the reference compound 21c (KD
= 2.8 µM,
duplex
KD
>20 µM) with some of 7-substituted-5,6-dihydrobenzo[c]acridine derivatives, such as 2c
c-KIT1
duplex
c-KIT1
duplex
c-KIT1
(KD
= 2.9 µM, KD
>20 µM), 2e (KD
= 2.8 µM, KD
>20 µM), 3c (KD
= 2.2 µM,
duplex
c-KIT1
duplex
c-KIT1
duplex
KD
>20 µM), 4c (KD
= 3.3 µM, KD
>20 µM), and 6b (KD
= 2.8 µM, KD
>20
µM), indicated that these compounds all had the similar binding affinity results, which revealed that
acridine scaffold with the introduction of alkylamino side chain at 7-position had efficient binding
affinity and selectivity for c-KIT1 G-quadruplex. Besides, the length of side chain and the types of
basic terminals showed no obvious effect on the binding affinity to c-KIT1 G-quadruplex.
Microscale thermophoresis (MST) is a highly sensitive probe for many kinds of
binding-induced interactions related with such as molecular size, charge, hydration shell or
conformation [35, 36]. In our studies, MST was also employed to study the binding affinity of
compound 2b on c-KIT1 G-quadruplex DNA dissolved in Tris-HCl buffer. Data shown in Fig. S2
demonstrated that compound 2b had a higher binding affinity for c-KIT1 G-quadruplex DNA with a
K of 1.5 µM (Fig. S2A) and a lower binding affinity with duplex DNA with a K of 10.5 µM (Fig.
D
D
S2B). These results were consistent with our SPR experimental data, and implied that compound 2b
could selectively recognize c-KIT1 G-quadruplex DNA over duplex DNA.
8

ACCEPTED MANUSCRIPT
2.4 Binding property of the derivatives on G-quadruplex
In general, the introduction of steady positive charges via the alkylation of nitrogen atoms
within nitrogen-containing compounds can significantly improve their G-quadruplex binding and
stabilization abilities [37]. However, in this study, combined with the results of FRET-melting and
SPR experiments, 12-N-methylated compounds (3a-3c and 4a-4c) showed a similar binding affinity
and lower stabilizing ability for c-KIT1 G-quadruplex compared with corresponding nonmethylated
compounds (1a-1c and 2a-2c). To clarify this unusual finding, CD, UV and fluorescence titration
were performed to further determine the binding properties of representative ligands (2b and its
methylated analog 4b) to G-quadruplex DNA. Molecular modeling approaches were also employed
to illustrate G-quadruplex DNA- ligand interactions.
The CD spectroscopy [34] was performed with c-KIT promoter G-rich sequence for c-KIT1
(
d[AG AG CGCTG AGGAG ]) and c-KIT2 (d[G CG CGCGAG AG ]) treated with compound 2b
3 3 3 3 3 3 3 4
or 4b in the presence or absence of 100 mM KCl. As shown in Fig. 3A, the CD spectra of c-KIT1
without any metal cations at room temperature didn’t exhibit the features of G-quadruplex (black).
After the treatment with 2b or 4b, the CD spectra of c-KIT1 presented a small negative peak at 240
nm and a large positive peak at 262 nm, which was the feature of parallel G-quadruplex. In the
presence of 100 mM KCl, the CD spectrum of c-KIT1 without ligands also displayed the feature of
parallel G-quadruplex (Fig. 3B). Correspondingly, the ligands 2b and 4b had no obvious effect on
+
the formation of parallel G-quadruplex with K . The CD spectroscopy of c-KIT2 presented similar
results (Fig. S3). Overall, the ligands 2b and 4b both had the ability to promote the c-KIT G-rich
sequences to form the parallel G-quadruplex in the absence of metal cations.
9

ACCEPTED MANUSCRIPT
To further investigate the mode of ligands binding to G-quadruplex DNA, UV titration
experiments were performed. As shown in Fig. S4, binding of compound 2bto c-KIT1 G-quadruplex
DNA exhibited marked hypochromicity and red shifts, showing a maximum at 357 nm and a clear
isosbestic point at 372 nm. The remarkable changes upon binding of 2b regarding hypochromicity,
the red shifts and the isosbestic point revealed π–π stacking interactions between the chromophore of
2b and the G-quartet [38]. Similarly, binding of compound 4b to c-KIT1 G-quadruplex DNA
exhibited remarkable hypochromicity and red shifts with a maximum at 393 nm and a clear
isosbestic point at 405 nm, indicating that 4b bound to target DNA with a similar binding mode to
that of 2b. In addition, as shown in Fig. S4B and S4D, the fitting of the experimental data
demonstrated that 2b (K = 3.3 µM) had higher affinity to c-KIT1 G-quadruplex DNA than 5b (K =
D
D
9.1 µM), which was fully in line with the results obtained from the SPR measurements.
Fluorescence titration experiments were also performed to investigate the mode of ligands
binding to c-KIT1 DNA, as shown in Fig. S5. Similar results revealed that ligands 2b and 4b bound
to c-KIT1 G-quadruplex DNA exhibiting marked hypochromicity, which showed a maximum
emission at 490 nm for 2b and 530 nm for 4b. In addition, as shown in Fig. S6, the fitting curve of
the Job’s plot experiments indicated the formation of a 1:1 complex with one ligand for one
G-quadruplex. Furthermore, The ESI-MS experiments were also used to intuitively assess the
stoichiometries of ligand-quadruplex interactions [39]. c-KIT1 G-quadruplex DNA was studied upon
+
addition of a buffer containing co-existing NH ion with the ligand 2b or 4b respectively. Upon
4
addition of 5-fold equivalents of ligands, the mass spectra signals of ligand-DNA adducts appeared
in the spectra corresponding to 1:1 ligand-quadruplex complexes (Fig. S7).
1
0

ACCEPTED MANUSCRIPT
To elucidate the mode of binding and the major factors governing the G-quadruplex recognition
and help rationalize the experimental results, molecular docking was carried out using ligands (2b,
4b) and c-KIT1(PDB code: 4WO2, a dimeric parallel propeller-type G-quadruplex based on X-ray
structure[40]) or c-KIT2 (PDB code: 2KYP, a monomeric all-parallel-stranded G-quadruplex based
on NMR structure[41]). The 4WO2 crystal structure contains two GQ molecules (A and B) in the
asymmetric unit. They are closely similar except for differences in the conformations of residues A1,
C11 and T12 at their stacking interfaces. We have taken the coordinates of 4WO2 quadruplex A as
the starting structure for all our simulations. For the G-quadruplex structure, there were two common
binding sites for small molecules: the external G-quartets and the groove regions [42]. The docking
results of ligands with c-KIT1 model (PDB code: 4WO2) revealed that the ligand 2b bound with its
A and D rings of acridine scaffold being stacked on guanine residue by π-π stacking interaction at
dG10 site and dG6 site in 5’-end respectively, and the protonated nitrogen atom on the diethylamino
side chain of the ligand involved in electrostatic interactions with the negative electrostatic potential
phosphodiester backbone (Fig. 4A). For compound 4b, the ligand-DNA interaction similarly
included π-π stacking interaction to the G-quartets and electrostatic interactions to the phosphate
backbone, but the methyl group at 12-position had an effect on the planarity of 5,6-dihydrobenzo[c]
acridine scaffold, which was an obstacle to π–π stacking interactions of D rings of acridine scaffold
as shown in Fig. 4B. Moreover, the docking results of ligands with c-KIT2 model (PDB code: 2KYP)
revealed π-π stacking interaction at dG4 and dG8 site in 5’-end, and electrostatic interactions with
the negative electrostatic potential phosphate backbone (Fig. S8). The effect of the 12-methyl
substituent of acridine scaffold on ligand-DNA interaction was mainly responsible for its weaker
11

ACCEPTED MANUSCRIPT
binding efficacy than the nonmethylated derivatives.
2.5. Antitumor activities of the derivatives
MTT assay was employed to evaluate the cytotoxicity of 7-substituted-5,6-dihydrobenzo[c]
acridine derivatives against human leukemia cell K562, lung adenocarcinoma cell line A549, cervical
cancer cell line Hela, and human BJ cell. As shown in Table 3, those nonmethylated compounds
(1a-1e, 2a-2e) that exhibited stronger stabilizing ability to c-KIT1 G-quadruplex DNA showed higher
cytotoxicity for different cell lines. In comparison, the methylated compounds (3a-3c, 4a-4c) and the
compounds 5c, 6b, and 6d with amide bond connections had weaker cytotoxicity, which might be
due to their weaker interacting ability to target DNA and/or their lower cellular uptake. Moreover,
most of nonmethylated derivatives showed stronger cytotoxicity for A549 cell lines than the other
two cancer cell lines and relatively weaker inhibitory effect on normal human BJ cells, indicating
their selectivity for cancer cells.
Based on the above results, we found that compound 2b exhibited the strongest inhibitory effect
on Hela cells (IC = 3.2 µM) and good cytotoxicity for K562 and A549 cells (IC = 9.2 and 5.7 µM,
5
0
50
respectively), especially had a weak cytotoxicity to BJ cells (IC = 29.4 µM). Thus, 2b was selected
5
0
for our further investigation. The potential cytotoxic effect of compound 2b on K562 cells, a chronic
myelogenous leukemia cell line with c-KIT expression, was performed by using the short-term
proliferation assay with imatinib as a positive control (Fig. 5). As expected, compound 2b treatment
clearly inhibited the cell proliferation of K562 cells in a dose-dependent manner in comparison with
the vehicle control.
1
2

ACCEPTED MANUSCRIPT
To investigate whether compound 2b induces apoptosis of K562 cells, flow cytometry assay
was employed. Firstly, K562 cells were stained with Annexin V-FITC/PI in apoptotic cells analyses.
The dose-dependent apoptosis ratios were 0.1%, 1.8%, 3.6%, 8.4% and 64.8% when treated with
compound 2b at concentrations of 0, 1.0, 3.0, 6.0 and 9.0 µM, respectively. As shown in Fig. 6,
treatment with 9.0 µM of 2b produced a more obvious apoptotic response (apoptosis ratio of 64.8%)
than that with 0.5 µM of imatinib (apoptosis ratio of 51.2%). These results clearly indicated that
compound 2b efficiently induced apoptosis of K562 cells. Therefore, compound 2b may be a good
candidate for further cellular studies.
2.6. Down-regulation of c-KIT transcription and expression by 2b in K562 cells
To determine the effect of compound 2b on the transcription and translation of c-KIT gene, the
c-KIT mRNA level of K562 cells was analyzed by using real-time polymerase chain reaction
RT-PCR). K562 cells were treated with varying concentrations of 2b for 48 h, and the level of c-KIT
mRNA was measured relative to the expression of β-actin, a housekeeping gene. As shown in Fig.
A, the treatment of K562 cells with 2b caused a reduction of c-KIT mRNA in a dose-dependent
(
7
manner. Level of c-KIT mRNA was reduced to about 52% relative to a β-actin control with treatment
of 9.0 µM compound 2b. No significant change in c-KIT transcription level was found with imatinib
treatment.
To confirm that the observed reduction in c-KIT mRNA level leads to a reduction of c-KIT
protein, Western blotting was employed by using anti-KIT antibodies. This experiment showed that
treatment of K562 cells with 2b resulted in a dose-dependent reduction in the level of c-KIT protein,
1
3

ACCEPTED MANUSCRIPT
with an obvious loss in total c-KIT after 48 h treatment with 9.0 µM 2b (Fig. 7B).
2.7. Possible mechanism for 2b-induced apoptosis in K562 cells
Compound 2b had a relatively strong inhibitory activity on proliferation of K562 cells and
induced apoptosis of K562 cells. Next, we explored the possible mechanism of 2b-induced apoptosis
of K562 cells. Caspase, a kinase protein plays an important role in the regulation of cell apoptosis
[
43]. It has been well-known that caspase-3 mediates cleavage of many substrates, including Poly
ADP)-Ribose Polymerase (PARP), and acts as a major executor of apoptosis [44]. Therefore, in the
(
present study, caspase-3 and PARP proteins were monitored by using Western blotting method. As
shown in Fig. 8, PARP was cleaved to an 89 kDa fragment after cells were treated with compound 2b
for 48 h. As the derivative concentration increased, the level of cleaved-PARP was up-regulated.
Besides, compound 2b activated caspase-3, reflected by down-regulation of pro-caspase-3 and
up-regulation of cleaved-caspase-3. Taken together, these results indicated that compound 2b could
activate the caspase-dependent pathway related with apoptosis.
It has been reported that gene members of the Bcl-2 family regulate the mitochondrial cell death
pathway. In particular, the stoichiometries of Bax (pro-apoptotic gene) and Bcl-2 (anti-apoptotic gene)
are important for the following downstream activation of caspase protein [45, 46]. In this study,
several mitochondrial related apoptotic proteins, including Bcl-2, Bad and Bax were investigated by
using Western blotting with specific antibodies as indicated. The expression levels of Bax and Bad
were significantly increased in a dose-dependent manner in 2b-treated K562 cells. In contrast, the
expression level of Bcl-2 was decreased upon compound 2b treatment.
1
4

ACCEPTED MANUSCRIPT
3
. Conclusion
A series of new 7-substituted- 5,6-dihydrobenzo[c]acridine derivatives were designed,
synthesized, and evaluated. The studies on the interactions of the derivatives with c-KIT
G-quadruplexes showed that the ligands presented promising abilities on stabilizing and binding to
G-quadruplex DNA in c-KIT oncogene promoter instead of duplex DNA. The CD, UV, and
fluorescence titration results indicated that the two types of ligands (Series I for 12-N-nonmethylated
and Series II for 12-N-methylated) were capable of inducing or maintaining the formation of the
G-quadruplex structure in the presence or absence of metal cations, and had similar binding property
for the interaction of ligand with G-quadruplex (π–π stacking interactions between the ligands and
the G-quartets, and the formation of a 1:1 complex). Molecular docking results interpreted that the
introduction of positive charges via the methylation of nitrogen atoms in acridine scaffolds didn’t
improve their quadruplex-stabilization abilities for the reason of an obstacle to π–π stacking
interactions.
Subsequent bioassays verified that the representative compound 2b exhibited an obvious
anti-proliferative effect on human cancer cells. One such ligand was shown to inhibit the expression
of c-KIT mRNA and presented a concomitant reduction of c-KIT protein in K562 cells. Besides,
compound 2b induced apoptosis through suppression of Bcl-2, up-regulation of Bax as well as
activation of caspase-3. However, our current study still lacks sufficient data to demonstrate a direct
action of 2b on the G-quadruplex structure of the c-KIT gene promoter in vivo, and the detailed
molecular mechanism of total c-KIT loss whether caused by the suppression of gene transcription or
the inhibition of the RNA polymerase need further confirmation [47]. Moreover, whether the
1
5

ACCEPTED MANUSCRIPT
apoptotic effect of c-KIT suppression was related to the down-regulation of antiapoptotic proteins
and how c-KIT was involved in apoptosis remain obscure. Detailed investigations are required to
answer these questions.
To sum up, the induction/stabilization of G-quadruplex DNA by the acridine derivatives
influenced the biological events in which down-regulation of c-KIT gene was involved. And this
study also suggested that the novel acridine compounds might be potential candidates for the
development of new anti-cancer agents targeting c-KIT G-quadruplex DNA.
4
. Experimental section
4.1. Synthesis and characterization
All commercial chemicals used as starting materials were analytical grade and utilized without
1
13
further purification. H and C NMR spectra were recorded using TMS as the internal standard in
DMSO-d , D O or CDCl with a Bruker BioSpin GmbH spectrometer at 400 MHz and 100 MHz,
6
2
3
respectively; Mass spectra (MS) were recorded on a Shimadzu LCMS-2010A instrument with an ESI
or ACPI mass selective detector, and high resolution mass spectra (HRMS) were recorded on
Shimadzu LCMS-IT-TOF. All the compounds were purified by using flash column chromatography
with silica gel (200-300 mesh) or Al O (200-300 mesh). The purities of synthesized compounds
2
3
were confirmed to be higher than 95% by using analytical HPLC equipped with a dual pump
Shimadzu LC-20AB system with an C18 column (4.6×250 mm, 5 mm), which was eluted with
-1
methanol-water (35:65-50:50) containing 0.1% TFA at a flow rate of 0.5 mL·min . Melting points
(m.p.) were determined by using capillary tubes with a MSRSOptiMelt automated melting point
1
6

ACCEPTED MANUSCRIPT
instrument without correction.
4
.1.1. Synthesis of intermediates
The intermediates 7-13 were prepared following a process as shown in Schemes 1A according
to our previous literature [26], and the detail of synthesis was described in Supporting Information.
4.1.2. General procedure A: preparation of 1a-1e, 2a-2e
Compound 11 (1 eq.) and p-toluenesulfonic acid (3 eq.) were mixed in pressure tubing and
o
heated under 80 C for 30 min. Then, the mixture was cooled down to room temperature and
o
alkylamine (2 eq.) was slowly added. The reaction mixture was heated at 120 C for 10 h, and then
cooled down to room temperature, which was washed with 1 N NaOH. It was extracted with ethyl
acetate, purified by using flash column chromatography to give compound 1a-1e and 2a-2e as
yellow oil or solid.
1
2
2
4
.1.2.1. N -(2,3-dimethoxy-5,6-dihydrobenzo[c]acridin-7-yl)-N ,N -dimethylethane-1,2-diamine (1a).
Compound 11 was treated with N,N-dimethylethane-1,2-diamine according to general procedure A to
afford 1a. After silica gel column separation, the desired product was obtained as a brown solid in 29%
1
yield. H NMR (400 MHz, CD OD): δ 8.11 (dd, J = 8.5, 0.8 Hz, 1H), 7.99 (d, J = 8.4 Hz, 1H), 7.90
3
(s, 1H), 7.62 (ddd, J = 8.3, 6.9, 1.3 Hz, 1H), 7.44 (ddd, J = 8.2, 6.9, 1.2 Hz, 1H), 6.90 (s, 1H), 3.96 (s,
3H), 3.90 (s, 3H), 3.47 (t, J = 6.5 Hz, 2H), 3.06– 2.97 (m, 2H), 2.91– 2.81 (m, 2H), 2.55 (t, J = 6.5
13
Hz, 2H), 2.29 (s, 6H). C NMR (100 MHz, CD OD): δ 154.9, 152.0, 151.9, 149.5, 148.6, 134.2,
3
1
7

ACCEPTED MANUSCRIPT
1
2
29.8, 129.3, 128.7, 125.6, 123.4, 123.0, 117.8, 111.9, 110.9, 60.3, 56.6, 56.5, 47.2, 45.5 (2C), 28.7,
+
5.3. HRMS (ESI): Calcd for [M+H] (C H N O ) requires m/z 378.2176, found 378.2175.
23
27
3
2
1
2
2
4
.1.2.2. N -(2,3-dimethoxy-5,6-dihydrobenzo[c]acridin-7-yl)-N ,N -diethylethane-1,2-diamine (1b).
Compound 11 was treated with N,N-diethylethane-1,2-diamine according to general procedure A to
afford 1b. After silica gel column separation, the desired product was obtained as a pale yellow oil in
1
1
7% yield. H NMR (400 MHz, DMSO-d ): δ 8.11 (d, J = 8.4 Hz, 1H), 7.92 (s, 1H), 7.89 (d, J = 8.4
6
Hz, 1H), 7.58 (t, J = 7.5 Hz, 1H), 7.41 (t, J = 7.5 Hz, 1H), 6.92 (s, 1H), 3.85 (d, J = 9.3 Hz, 6H), 3.40
(d, J = 5.8 Hz, 2H), 2.99 (t, J = 6.5 Hz, 2H), 2.83 (t, J = 6.4 Hz, 2H), 2.56 (s, 2H), 2.49 – 2.43 (m,
1
3
4
1
2
4
H), 0.93 (t, J = 7.0 Hz, 6H); C NMR (100 MHz, DMSO-d ): δ 152.4, 150.0, 149.9, 147.7, 147.2,
6
31.8, 129.1, 128.2, 127.5, 123.8, 122.5, 121.27, 114.9, 110.9, 109.0, 55.6, 55.5, 52.6, 46.1(2C), 45.8,
+
7.2, 23.7, 11.7(2C). HRMS (ESI): Calcd for [M+H] (C H N O ) requires m/z 406.2489, found
2
5
31
3
2
06.2498.
4.1.2.3.
2,3-dimethoxy-N-(2-(pyrrolidin-1-yl)ethyl)-5,6-dihydrobenzo[c]acridin-7-amine
(1c).
Compound 11 was treated with 2-(pyrrolidin-1-yl) ethan-1-amine according to general procedure A
to afford 1c. After silica gel column separation, the desired product was obtained as a pale yellow
o
1
solid in 32% yield. m.p. 151.4 – 152.4 C; H NMR (400 MHz, CD OD): δ 8.07 (t, J = 4.0 Hz, 2H),
3
8
3
2
.00 (d, J = 8.3 Hz, 1H), 7.62 – 7.55 (m, 1H), 7.43 – 7.35 (m, 1H), 6.75 (s, 1H), 5.26 (s, 1H), 4.06 (s,
H), 3.94 (s, 3H), 3.42 (d, J = 5.1 Hz, 2H), 3.05 (dd, J = 8.3, 5.8 Hz, 2H), 2.89 (t, J = 7.0 Hz, 2H),
1
3
.73 – 2.66 (m, 2H), 2.60 – 2.48 (m, 4H), 1.91 – 1.77 (m, 4H); C NMR (100 MHz, CD OD): δ
3
1
8

ACCEPTED MANUSCRIPT
1
1
53.1, 150.3, 149.8, 148.3, 147.8, 131.9, 129.6, 128.3, 128.2, 124.2, 122.0, 121.9, 116.4, 110.0,
+
08.9, 56.1, 55.9, 55.4, 53.6, 47.7(2C), 27.9, 24.1, 23.7(2C). HRMS(ESI): Calcd for [M+H]
(
C H N O ) requires m/z 404.2333, found: 404.2331.
25 29 3 2
4.1.2.4.
2,3-dimethoxy-N-(2-(piperidin-1-yl)ethyl)-5,6-dihydrobenzo[c]acridin-7-amine
(1d).
Compound 11 was treated with 2-(piperidin-1-yl) ethan-1-amine according to general procedure A to
afford 1d. After silica gel column separation, the desired product was obtained as a colorless oil in 22%
1
yield. H NMR (400 MHz, CDCl ): δ 8.11 – 8.04 (m, 2H), 8.02 (d, J = 8.4 Hz, 1H), 7.57 (t, J = 7.6
3
Hz, 1H), 7.38 (t, J = 7.6 Hz, 1H), 6.73 (s, 1H), 5.31 (s, 1H), 4.04 (s, 3H), 3.92 (s, 3H), 3.38 (t, J = 5.5
Hz, 2H), 3.08 – 3.01 (m, 2H), 2.92–2.84 (m, 2H), 2.52 – 2.46 (m, 2H), 2.41 (s, 4H), 1.63 (dt, J =
1
3
1
1
2
4
0.8, 5.5 Hz, 4H), 1.48 (d, J = 4.7 Hz, 2H); C NMR (100 MHz, CDCl ): δ 153.3, 150.4, 150.1,
3
48.3, 148.0, 132.0, 129.8, 128.3, 124.2, 122.1, 122.0, 116.3, 110.3, 109.2, 58.2, 56.0, 54.2, 45.6,
+
7.9, 26.3, 24.5, 24.2. HRMS (ESI): Calcd for [M+H] (C H N O ) requires m/z 418.2495, found
2
6
31
3
2
18.2492.
4.1.2.5. 2,3-dimethoxy-N-(2-(4-methylpiperazin-1-yl)ethyl)-5,6-dihydrobenzo[c] acridin-7-amine
(1e).
Compound 11 was treated with 2-(4-methylpiperazin-1-yl) ethan-1-amine according to
general procedure A to afford 1e. After silica gel column separation, the desired product was obtained
1
as a brown yellow oil in 39% yield. H NMR (400 MHz, CD OD): δ 8.10 (d, J = 8.4 Hz, 1H), 7.98 (d,
3
J = 8.4 Hz, 1H), 7.92 (s, 1H), 7.61 (t, J = 7.5 Hz, 1H), 7.43 (t, J = 7.5 Hz, 1H), 6.86 (s, 1H), 5.51 (s,
1H), 3.97 (s, 3H), 3.90 (s, 3H), 3.41 (t, J = 5.5 Hz, 2H), 2.98 (t, J = 6.5 Hz, 2H), 2.83 (t, J = 6.5 Hz,
1
9

ACCEPTED MANUSCRIPT
1
3
2
H), 2.75 – 2.28 (m, 10H), 2.26 (s, 3H); C NMR (100 MHz, CD OD): δ 155.0, 152.2, 151.9, 149.5,
3
148.7, 134.2, 129.9, 129.3, 128.7, 125.5, 123.5, 123.1, 117.9, 111.9, 110.9, 58.8, 56.6, 56.5(2C), 56.0,
+
5
3.4(2C), 46.4, 46.0, 28.6, 25.4. HRMS(ESI): Calcd for [M+H] (C H N O ) requires m/z 433.2598,
2
6
32
4
2
found: 433.2598.
1
3
3
4
.1.2.6. N -(2,3-dimethoxy-5,6-dihydrobenzo[c]acridin-7-yl)-N ,N -dimethylpropane-1,3-diamine (2a).
Compound 11 was treated with N,N-dimethylpropane-1,3-diamine according to general procedure A
to afford 2a. After silica gel column separation, the desired product was obtained as a brown yellow
1
oil in 36% yield. H NMR (400 MHz, CD OD): δ 8.16 (d, J = 8.4 Hz, 1H), 8.00 (d, J = 8.3 Hz, 1H),
3
7.91 (s, 1H), 7.65 (t, J = 7.5 Hz, 1H), 7.47 (t, J = 7.6 Hz, 1H), 6.93 (s, 1H), 3.99 (s, 3H), 3.93 (s, 3H),
3.50 (t, J = 6.6 Hz, 2H), 3.04 (t, J = 6.7 Hz, 2H), 2.89 (t, J = 6.7 Hz, 2H), 2.46 (t, J = 7.3 Hz, 2H),
1
3
2
.27 (s, 6H), 1.90 – 1.77 (m, 2H)； C NMR (100 MHz, CD OD): δ 154.5, 152.1, 151.9, 149.5, 148.2,
3
134.2, 129.9, 128.9, 128.3, 125.6, 123.5, 122.7, 117.1, 111.9, 110.8, 58.4, 56.7, 56.5, 48.3, 45.4(2C),
+
2
9.1, 28.7, 25.5. HRMS(ESI): Calcd for [M+H] (C H N O ) requires m/z 392.2333,
24
29
3
2
found:392.2336.
1
3
3
4
.1.2.7. N -(2,3-dimethoxy-5,6-dihydrobenzo[c]acridin-7-yl)-N ,N -diethylpropane-1,3-diamine (2b).
Compound 11 was treated with N,N-diethylpropane-1,3-diamine according to general procedure A to
afford 2b. After silica gel column separation, the desired product was obtained as a colorless oil in
1
4
2% yield. H NMR (400 MHz, DMSO-d ): δ 8.71 (d, J = 8.4 Hz, 1H), 8.59 (d, J = 8.5 Hz, 1H), 8.14
6
(s, 2H), 7.88 (t, J = 7.6 Hz, 1H), 7.63 (t, J = 7.6 Hz, 1H), 7.11 (s, 1H), 3.99 (s, 3H), 3.88 (s, 5H), 3.16
2
0

ACCEPTED MANUSCRIPT
13
–
2.99 (m, 8H), 2.88 (t, J = 7.0 Hz, 2H), 2.25–2.08 (m, 2H), 1.20 (t, J = 7.1 Hz, 6H)； C NMR (100
MHz, DMSO-d ): δ 155.4, 152.0, 147.8, 146.7, 138.1, 134.2, 132.1, 125.7, 123.8, 120.3, 119.0,
6
117.2, 111.2, 110.7, 110.5, 56.5, 55.8, 47.8, 46.0, 44.4, 26.8, 24.1, 23.7, 8.3(2C). HRMS (ESI):
+
Calcd for [M + H] (C H N O ) requires m/z 420.2651, found 420.2644.
26
33
3
2
4.1.2.8.
2,3-dimethoxy-N-(3-(pyrrolidin-1-yl)propyl)-5,6-dihydrobenzo[c]acridin-7-amine
(2c).
Compound 11 was treated with 3-(pyrrolidin-1-yl) propan-1-amine according to general procedure A
to afford 2c. After silica gel column separation, the desired product was obtained as a yellow green
1
oil in 27% yield. H NMR (400 MHz, CD OD): δ 8.16 (d, 1H), 8.00 (d, J = 8.5 Hz, 1H), 7.93 (s, 1H),
3
7
3
2
1
2
4
.64 (t, J = 7.5 Hz, 1H), 7.46 (t, J = 7.6 Hz, 1H), 6.93 (s, 1H), 3.96 (s, 6H), 3.50 (t, J = 6.7 Hz, 2H),
.05 (t, J = 6.8 Hz, 2H), 2.87 (t, 2H), 2.58 (t, J = 7.4 Hz, 2H), 2.54 (t, J = 7.2 Hz, 4H), 1.91 – 1.84 (m,
1
3
H), 1.84 – 1.77 (m, 4H)； C NMR (100 MHz, CD OD): δ 155.0, 152.0, 151.8, 149.5, 148.7, 134.2,
3
29.8, 129.3, 128.8, 125.5, 123.4, 123.0, 117.4, 111.8, 110.8, 56.6, 56.5, 55.3, 55.0(2C), 48.5, 30.6,
+
8.8, 25.7, 24.2(2C). HRMS (ESI): Calcd for [M+H] (C H N O ) requires m/z 418.2489, found:
2
6
31
3
2
18.2481.
4.1.2.9.
2,3-dimethoxy-N-(3-(piperidin-1-yl)propyl)-5,6-dihydrobenzo[c]acridin-7-amine
(2d).
Compound 11 was treated with 3-(piperidin-1-yl) propan-1-amine according to general procedure A
to afford 2d. After silica gel column separation, the desired product was obtained as a brown yellow
1
oil in 69% yield. H NMR (400 MHz, CD OD): δ 8.17 (d, J = 8.4 Hz, 1H), 7.99 (d, J = 8.4 Hz, 1H),
3
7.91 (s, 1H), 7.63 (t, J = 7.5 Hz, 1H), 7.45 (t, J = 7.6 Hz, 1H), 6.91 (s, 1H), 3.98 (s, 3H), 3.92 (s, 3H),
2
1

ACCEPTED MANUSCRIPT
3
1
1
5
.45 (t, J = 6.5 Hz, 2H), 3.03 (t, J = 6.6 Hz, 2H), 2.87 (t, J = 6.6 Hz, 2H), 2.50 – 2.35 (m, 6H), 1.88 –
1
3
.77 (m, 2H), 1.64 – 1.55 (m, 4H), 1.49 – 1.43 (m, 2H)； C NMR (100 MHz, CD OD): δ 154.8,
3
52.1, 151.9, 149.5, 148.5, 134.2, 129.8, 129.2, 128.6, 125.5, 123.5, 123.0, 117.4, 111.8, 110.9, 58.1,
+
6.6, 56.5, 55.5(2C), 48.4, 28.7, 28.3, 26.4(2C), 25.7, 25.0. HRMS (ESI): Calcd for [M+H]
(
C H N O ) requires m/z 432.2646, found: 432.2650.
27 33 3 2
4.1.2.10. 2,3-dimethoxy-N-(3-(4-methylpiperazin-1-yl)propyl)-5,6-dihydrobenzo[c]acridin-7-amine
(2e). Compound 11 was treated with 3-(4-methylpiperazin-1-yl) propan-1-amine according to
general procedure A to afford 2e. After silica gel column separation, the desired product was obtained
1
as a brown yellow oil in 42% yield. H NMR (400 MHz, CD OD): δ 8.01 (d, J = 8.3 Hz, 1H), 7.91 (d,
3
J = 8.3 Hz, 1H), 7.83 (s, 1H), 7.51 (t, J = 7.4 Hz, 1H), 7.31 (t, J = 7.5 Hz, 1H), 6.72 (s, 1H), 3.88 (s,
3
2
1
5
4
H), 3.79 (s, 3H), 3.22 (t, J = 6.2 Hz, 2H), 2.81 (t, 2H), 2.68 (t, J = 6.9 Hz, 2H), 2.52 – 2.13 (m, 10H),
1
3
.13 (s, 3H), 1.66 – 1.56 (m, 2H)； C NMR (100 MHz, CD OD): δ 154.9, 152.0, 151.8, 149.4, 148.7,
3
34.0, 129.7, 129.4, 128.8, 125.5, 123.4, 123.0, 117.3, 111.6, 110.7, 57.1, 56.6, 56.4, 55.6(2C),
+
3.6(2C), 48.2, 46.0, 28.7, 28.3, 25.7. HRMS (ESI): Calcd for [M+H] (C H N O ) requires m/z
27
34
4
2
47.2755, found: 447.2757.
4.1.3. General procedure B: preparation of 3a-3c and 4a-4c
The mixture of salt compound 12 (1 eq.) and alkylamine (2 eq.) was heated under reflux in
anhydrous THF for 2 h. Then, the mixture was cooled down to room temperature and purified by
using flash column chromatography to give compound 3a-3c and 4a-4c as a yellow oil or solid.
2
2

ACCEPTED MANUSCRIPT
4
1
.1.3.1. 7-((2-(dimethylamino) ethyl) amino)-2,3-dimethoxy-12-methyl-5,6-dihydrobenzo[c]acridin-
2-ium(3a). Compound 12 was treated with N,N-dimethylethane-1,2-diamine according to general
procedure B to afford 3a. After silica gel column separation, the desired product was obtained as a
1
yellow solid in 40% yield. m.p. 217 – 219 °C；H NMR (400 MHz, D O): δ 8.27 (d, J = 8.4 Hz, 1H),
2
8
.12 – 8.02 (m, 2H), 7.77 (t, J = 6.9 Hz, 1H), 7.09 (s, 1H), 7.00 (s, 1H), 4.18 (s, 3H), 3.98 (s, 3H),
.92 (t, J = 6.8 Hz, 2H), 3.85 (s, 3H), 2.81 (dd, J = 11.6, 5.0 Hz, 4H), 2.74 (d, J = 7.1 Hz, 2H), 2.37
3
1
3
(
s, 6H)； C NMR (100 MHz, D O): δ 154.6, 152.2, 151.2, 146.4, 140.8, 136.6, 133.5, 126.0, 124.6,
2
120.0, 118.5, 117.5, 115.7, 112.4, 110.8, 57.8, 56.1, 55.9, 45.2, 44.0 (2C), 42.9, 27.1, 23.4. HRMS
+
+
(
ESI): Calcd for [M – OTf] (C H N O ) requires m/z 392.2338, found 392.2329.
24 30 3 2
4
1
.1.3.2. 7-((2-(diethylamino) ethyl) amino)-2, 3-dimethoxy-12-methyl-5,6-dihydrobenzo[c]acridin-
2-ium trifluoromethanesulfonate (3b). Compound 12 was treated with N,N-diethylpropane-1,3-
diamine according to general procedure B to afford 3b. After silica gel column separation, the
1
desired product was obtained as a yellow oil in 46% yield. H NMR (400 MHz, CD OD): δ 8.46 (d, J
3
=
8.5 Hz, 1H), 8.22 (d, J = 8.7 Hz, 1H), 8.06 (t, J = 7.8 Hz, 1H), 7.78 (t, J = 7.7 Hz, 1H), 7.35 (s, 1H),
7
2
1
5
.17 (s, 1H), 4.36 (s, 3H), 4.00 (m, 5H), 3.94 (s, 3H), 2.99 (d, J = 7.3 Hz, 4H), 2.93 (t, J = 5.8 Hz,
1
3
H), 2.73 (dd, J = 13.3, 6.4 Hz, 4H), 1.12 (t, J = 7.0 Hz, 6H)； C NMR (100 MHz, CD OD): δ 156.2,
3
54.4, 153.8, 149.2, 142.5, 137.7, 134.5, 127.2, 125.7, 121.5, 120.1, 119.4, 116.9, 114.6, 112.1, 57.2,
-
+
6.7, 53.2, 47.6(2C), 46.0, 43.9, 28.9, 25.6, 11.8(2C). HRMS (ESI): Calcd for [M – OTf ]
+
2
(
C H N O ) requires m/z 420.2646, found: 420.2651.
2
6
34
3
2
3

ACCEPTED MANUSCRIPT
4
1
.1.3.3. 2,3-dimethoxy-12-methyl-7-((2-(pyrrolidin-1-yl)ethyl)amino)-5,6-dihydrobenzo[c]acridin-
2-ium trifluoromethanesulfonate (3c). Compound 12 was treated with 3-(pyrrolidin-1-yl)
propan-1-amine according to general procedure B to afford 3c. After silica gel column separation, the
1
desired product was obtained as a yellow oil in 59% yield. H NMR (400 MHz, CD OD): δ 8.33 (d, J
3
=
8.5 Hz, 1H), 8.10 (d, J = 8.7 Hz, 1H), 7.93 (t, J = 7.8 Hz, 1H), 7.66 (t, J = 7.5 Hz, 1H), 7.22 (s, 1H),
7.03 (s, 1H), 4.24 (s, 3H), 3.92 (t, J = 6.2 Hz, 2H), 3.87 (s, 3H), 3.80 (s, 3H), 2.97 (s, 2H), 2.88 (d, J
1
3
=
5.5 Hz, 2H), 2.83 (d, J = 6.3 Hz, 2H), 2.73 (s, 4H), 1.79 (s, 4H); C NMR (100 MHz, CD OD): δ
3
1
5
56.3, 154.9, 154.0, 149.2, 142.3, 137.9, 134.5, 127.5, 125.5, 121.5, 120.2, 119.7, 117.4, 114.6, 112.1,
+
7.2, 56.7, 55.9, 55.0(2C), 47.0, 44.0, 29.0, 25.9, 24.3(2C). HRMS (ESI): Calcd for [M – OTf]
(
C H N O ) requires m/z 418.2489, found: 418.2495.
26 32 3 2
4.1.3.4. 7-((3-(dimethylamino) propyl) amino)-2,3-dimethoxy-12-methyl-5,6-dihydrobenzo[c]acridin
-12-ium trifluoromethanesulfonate (4a). Compound 12 was treated with N,N-dimethylpropane-1,3-
diamine according to general procedure B to afford 4a. After silica gel column separation, the desired
1
product was obtained as a yellow solid in 20% yield. m. p. 162 – 165 °C； H NMR (400 MHz,
CDCl ): δ 8.55 (d, J = 8.4 Hz, 1H), 7.97 – 7.92 (m, 2H), 7.77 (t, J = 6.8 Hz, 1H), 7.64 (s, 1H), 6.94 (s,
3
1
2
1
2
H), 4.31 (s, 3H), 4.02 (s, 3H), 3.97 (s, 2H), 3.95 (s, 3H), 3.32 (s, 2H), 2.95 (s, 2H), 2.91 (s, 2H),
13
.84 (s, 6H), 2.26 (s, 3H); C NMR (100 MHz, CDCl ): δ 155.4, 153.3, 152.6, 147.9, 140.6, 136.5,
3
33.6, 127.0, 125.5, 119.9, 119.0, 118.6, 118.0, 115.8, 113.1, 110.9, 56.8, 56.3, 55.1, 44.6, 42.9, 29.7,
+
+
2
8.3, 25.4, 25.3(2C). HRMS (ESI): Calcd for [M – OTf] (C H N O ) requires m/z 406.2495,
25
31
3
2
4

ACCEPTED MANUSCRIPT
found 406.2481.
4.1.3.5. 7-((3-(diethylamino) propyl) amino)-2,3-dimethoxy-12-methyl-5,6-dihydrobenzo[c]acridin-
12-ium trifluoromethanesulfonate (4b). Compound 12 was treated with N,N-diethylpropane-1,3-
diamine according to general procedure B to afford 4b. After silica gel column separation, the
1
desired product was obtained as a yellow solid in 45% yield. m. p. 184 – 185 °C；H NMR (400 MHz,
CD OD): δ 8.19 (d, J = 8.5 Hz, 1H), 8.01 (d, J = 8.9 Hz, 1H), 7.95 – 7.88 (m, 1H), 7.65 (s, 1H), 7.00
3
(s, 2H), 4.13 (s, 3H), 3.85 (s, 3H), 3.79 (t, J = 7.0 Hz, 2H), 3.72 (s, 3H), 3.16 – 3.02 (m, 6H), 2.71 (s,
1
3
4
1
4
H), 2.12 – 1.96 (m, 2H), 1.13 (t, J = 7.3 Hz, 6H); C NMR (100 MHz, CDCl ): δ 155.5, 153.4,
3
52.5, 147.9, 140.6, 136.6, 133.6, 127.0, 125.5, 120.0, 118.8, 118.1, 116.0, 113.2, 110.9, 56.8, 56.4,
+
9.4, 46.7(2C), 44.5, 43.0, 28.4, 25.5, 24.8, 8.6(2C). HRMS (ESI): Calcd for [M – OTf]
+
2
(
C H N O ) requires m/z 434.2808, found 434.2797.
2
7
36
3
4.1.3.6. 2,3-dimethoxy-12-methyl-7-((3-(pyrrolidin-1-yl)propyl)amino)-5,6-dihydrobenzo[c]acridin-
12-ium trifluoromethanesulfonate (4c). Compound 12 was treated with 3-(pyrrolidin-1-yl)
propan-1-amine according to general procedure B to afford 4c. After silica gel column separation, the
1
desired product was obtained as a yellow solid in 35% yield. m. p. 261 – 264 °C；H NMR (400 MHz,
CDCl ): δ 9.25 (s, 1H), 9.03 (d, J = 8.4 Hz, 1H), 7.90 – 7.83 (m, 2H), 7.70 (t, J = 8.0 Hz, 1H), 7.04 (s,
3
1
H), 6.91 (s, 1H), 4.24 (s, 3H), 4.05 (t, J = 6.0 Hz, 2H), 4.00 (s, 3H), 3.93 (s, 3H), 3.43 (t, J = 7.6 Hz,
H), 3.09 (t, J = 6.8 Hz, 2H), 2.86 (t, J = 6.8 Hz, 2H), 2.58–2.51 (m, 2H), 2.09 (s, 4H), 1.88 (s, 4H);
2
1
3
C NMR (100 MHz, D O): δ 154.7, 153.1, 151.3, 146.6, 140.8, 137.0, 133.5, 126.4, 124.2, 120.5,
2
2
5

ACCEPTED MANUSCRIPT
118.8, 116.6, 116.5, 113.0, 111.0, 56.3, 56.1, 54.3, 52.2, 44.9(2C), 43.0, 27.4, 26.7, 24.2, 22.7.
+
HRMS (ESI): Calcd for [M + H] (C H N O ) requires m/z 432.2651, found 432.2645.
2
6
34
3
2
4.1.4. General procedure C: preparation of 5c, 6b and 6d
Compound 13 (1 eq.) was treated with alkylamine (1.2 – 1.5 eq.) and Et N (2 eq.) with the
3
o
catalysis of BOP reagent, and the mixture was continuously stirred at 60 C for about 10 h. Then, the
mixture was cooled down to room temperature and purified by using flash column chromatography
to give compound 5c, 6b and 6d as brown yellow solid.
4.1.4.1. 2,3-dimethoxy-N-(2-(pyrrolidin-1-yl)ethyl)-5,6-dihydrobenzo[c]acridine-7-carboxamide (5c).
Compound 13 was treated with 2-(pyrrolidin-1-yl) ethan-1-amine according to general procedure C
to afford 5c. After silica gel column separation, the desired product was obtained as a dark yellow
1
solid in 60% yield. m. p. 67 – 70 °C; H NMR (400 MHz, CD OD): δ 8.81 (s, 1H), 8.04 (d, J = 8.0
3
Hz, 1H), 7.99 (s, 1H), 7.81 (d, J = 8.0 Hz, 1H), 7.71 (t, J = 8.0 Hz, 1H), 7.53 (t, J = 8.0 Hz, 1H), 6.97
(s, 1H), 3.89 (s, 3H), 3.85 (s, 3H), 3.57 (s, 2H), 3.01 (s, 2H), 2.90 (t, J = 6.4 Hz, 4H), 2.80 (s, 4H),
1
3
1
1
2
.81 (s, 4H)； C NMR (100 MHz, DMSO-d ): δ 167.1, 152.9, 151.3, 148.5, 146.8, 141.5, 133.3,
6
29.6, 129.3, 126.7, 126.7, 125.8, 125.4, 124.0, 111.7, 109.0, 56.2, 56.1, 54.3, 54.0, 37.7, 27.1, 26.1,
+
3.5. HRMS (ESI): Calcd for [M + H] (C H N O ) requires m/z 432.2287, found 432.2281.
2
6
29
3
3
4.1.4.2 N-(3-(diethylamino)propyl)-2,3-dimethoxy-5,6-dihydrobenzo[c]acridine-7-carboxamide (6b).
Compound 13 was treated with N,N-diethylpropane-1,3-diamine according to general procedure C to
2
6

ACCEPTED MANUSCRIPT
afford 6b. After silica gel column separation, the desired product was obtained as a pale yellow solid
1
in 25% yield. m. p. 78 – 80 °C；H NMR (400 MHz, CDCl ): δ 8.25 (s, 1H), 8.12 (s, 1H), 8.10 (d, J =
3
8
4
2
.0 Hz, 1H), 7.78 (d, J = 8.0 Hz, 1H), 7.63 (t, J = 8.0 Hz, 1H), 7.45 (t, J = 8.0 Hz, 1H), 6.74 (s, 1H),
.06 (s, 3H), 3.94 (s, 3H), 3.71 (s, 2H), 3.10 (s, 2H), 2.93 (t, J = 8.0 Hz, 2H), 2.62 (t, J = 8.0 Hz, 2H),
1
3
.38 – 2.35 (dd, J = 8.0 Hz, J = 8.0 Hz, 4H), 1.81 (t, J = 6.0 Hz, 2H), 0.74 (t, J = 8.0 Hz, 6H); C
NMR (100 MHz, CDCl ): δ 167.2, 153.0, 150.8, 148.5, 147.2, 140.8, 132.7, 129.4, 128.9, 127.2,
3
126.3, 125.5, 124.5, 123.8, 110.5, 108.7, 56.1, 56.0, 52.8, 46.6, 40.3, 27.6, 26.2, 25.3, 11.4. HRMS
+
(
ESI): Calcd for [M + H] (C H N O ) requires m/z 448.2600, found 448.2594.
27
33
3
3
4.1.4.3 2,3-dimethoxy-N-(3-(piperidin-1-yl) propyl)-5,6-dihydrobenzo[c]acridine-7-carboxamide
(6d). Compound 13 was treated with 3-(piperidin-1-yl) propan-1-amine according to general
procedure C to afford 6d. After silica gel column separation, the desired product was obtained as a
1
pale yellow solid in 25% yield. m. p. 62 – 65 °C; H NMR (400 MHz, DMSO-d ): δ 8.84 (s, 1H),
6
8
3
1
1
2
4
.05(d, J = 8.4 Hz, 1H), 7.99 (s, 1H), 7.72 (t, J = 7.6 Hz, 2H), 7.55 (d, J = 7.6 Hz, 1H), 6.97 (s, 1H),
.89 (s, 3H), 3.85 (s, 3H), 3.42 (t, J = 5.6 Hz, 2H), 2.98 – 2.82 (m, 10H), 1.90 (s, 2H), 1.64 (s, 4H),
1
3
.49 (s, 2H); C NMR (100 MHz, DMSO-d ): δ 166.4, 152.4, 150.8, 148.1, 146.3, 141.1, 132.8,
6
29.1, 128.9, 126.3, 126.2, 125.2, 124.7, 123.5, 11.3, 108.5, 55.7, 55.6, 54.9, 52.9, 36.7, 26.6, 25.7,
+
4.8, 23.7, 22.4. HRMS (ESI): Calcd for [M + H] (C H N O ) requires m/z 460.2600, found
28
33
3
3
60.2608.
4.2. Materials
2
7

ACCEPTED MANUSCRIPT
All oligomers/primers used in this study were purchased from Invitrogen (China). Stock
solutions of all the derivatives (10 mM) were made using DMSO. Further dilutions to working
concentrations were made with double-distilled deionized water.
4.3. FRET-melting assay
FRET assay was carried out on a real-time PCR apparatus following previously published
procedures. The fluorescently labeled oligonucleotides included F-c-KIT1-T (5′-FAM-AGGG-
AGGGCGCTGGAGGAGGG-TAMARA-3′), F-c-KIT2-T (5′-FAM-GGGCGGGCGCGAGGGAGG-
GG-TAMARA-3′) and F10T (5′-FAM-TATAGCTATA-HEG-TATAGCTATA-TAMARA-3′). The
donor fluorophore FAM is 6-carboxyfluorescein; the acceptor fluorophore TAMRA is
6
-carboxytetramethylrhodamine; HEG linker [(-CH -CH -O-) ]) was used for the FRET probes.
2 2 6
Fluorescence melting curves were determined with a Roche Light Cycler 2 real-time PCR machine,
using a total reaction volume of 20 µL, with 0.4 µM of labeled oligonucleotide in Tris-HCl buffer (10
mM, pH 7.4) containing 60 mM KCl. Fluorescence readings with excitation at 470 nm and detection
o
o
at 530 nm were taken at intervals of 1 C over the range of 37-99 C with a constant temperature
being maintained for 30 s prior to each reading to ensure a stable value. The melting of the
G-quadruplex was monitored alone or in the presence of various concentrations of compounds and/or
double-stranded competitor ds26 DNA. Final analysis of the data was carried out using Origin 9.0
(Origin Lab Corp.).
4.4. Surface plasmon resonance
2
8