Molecular docking and synthesis of novel quinazoline analogues as inhibitors of transcription factors NF-κB activation and their anti-cancer activities

Article abstract of DOI:10.1016/j.bmc.2012.10.051

NF-kB is a transcription factor protein complex that can be found in almost all animal cell types and is a key player in some cancers and inflammatory responses. It can enhance the proliferation rate, reduce apoptosis, as well as creating more blood flow to ensure the survival of cancer, thus blocking the NF-kB pathway has potential therapeutic benefit. We designed a series of compounds based on a quinazoline scaffold pharmacophore model which may have high binding affinity with the p50 subunit of NF-kB. The compound series with phenyl substitution at the 2 position of the quinazoline proved to be more effective at inhibiting NF-kB function both theoretically and experimentally. These compounds also reduce the proliferation of numerous tumor cell lines and the mean GI₅₀ for compound 2a is 2.88 μM against the NCI-60 cell line. At the same time, compound 2a can induce significant apoptosis in EKVX cell line at the concentration of 1 μM.

Full text of DOI:10.1016/j.bmc.2012.10.051

Bioorganic & Medicinal Chemistry 21 (2013) 540–546
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Molecular docking and synthesis of novel quinazoline analogues as inhibitors
of transcription factors NF-jB activation and their anti-cancer activities
⇑
Lu Xu, Wade A. Russu
Department of Pharmaceutics and Medicinal Chemistry, Thomas J. Long School of Pharmacy and Health Sciences, University of the Pacific, Stockton, CA 95211, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
NF-kB is a transcription factor protein complex that can be found in almost all animal cell types and is a
key player in some cancers and inflammatory responses. It can enhance the proliferation rate, reduce
apoptosis, as well as creating more blood flow to ensure the survival of cancer, thus blocking the
NF-kB pathway has potential therapeutic benefit. We designed a series of compounds based on a quinaz-
oline scaffold pharmacophore model which may have high binding affinity with the p50 subunit of
NF-kB. The compound series with phenyl substitution at the 2 position of the quinazoline proved to be
more effective at inhibiting NF-kB function both theoretically and experimentally. These compounds also
Received 24 August 2012
Revised 20 October 2012
Accepted 30 October 2012
Available online 15 November 2012
Keywords:
Quinazoline
NF-jB cancer
Apoptosis
reduce the proliferation of numerous tumor cell lines and the mean GI50 for compound 2a is 2.88
against the NCI-60 cell line. At the same time, compound 2a can induce significant apoptosis in EKVX cell
line at the concentration of 1 M.
lM
l
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
subsequent proteosomal degradation of I
translocation of transcriptionally active NF-
What gene sets are actively transcribed depends on which NF-
dimer form is active. There are several strategies for inhibiting
NF- B activity.
The activity of NF-
rect strategies involve inhibitors that interact with one or more of
the NF- B family proteins to prevent their function. These inhibi-
tors may prevent dimerization of NF- B family members or pre-
vent DNA binding. Indirect strategies involve inhibitors of
molecules that affect NF- B function. Inhibitors that target mole-
cules upstream of NF- B such as IKK, cytokines and cytokine
receptors or prevent NF- B degradation, such as proteasome inhib-
itors have been investigated.8
The NF- B inhibitors that are predicted to interact directly with
p50 to prevent DNA binding include the triazine, coumarin, and
quinazoline chemical classes, among others.9–11 Extensive docking
studies have been performed on some of these in order to identify
important elements for binding and pharmacophore develop-
ment.12 Here we report our efforts toward the discovery of new
inhibitors of the NF-jB controlled transcriptional pathway by mak-
ing use of previously reported pharmacophore models and lead
molecules based on a quinazoline scaffold.
j
j
B, and the release and
B to the cell nucleus.6
The constitutive activation of nuclear kappa B (NF-
j
B) family of
jB
transcription factors has been implicated in various disease states
including inflammation and cancer.1 The relevance of this tran-
scriptional activity to cancer is thought to be linked to the tran-
scription control of key antiapoptotic genes that encode B-cell
lymphoma-2 (Bcl-2) and inhibitor of apoptosis (IAP) family pro-
teins.2 Over expression of these antiapoptotic genes can prevent
tumor cells from undergoing programmed cell death. These effects
in turn contribute to processes of tumorigenesis and resistance to
therapies.3
j
j
B may be inhibited directly or indirectly. Di-
j
j
j
j
The NF-
proteins, named Rel (c-Rel), RelA (p65), RelB, NF-
NF-
j
B proteins consist of five family members. These five
j
j
B1 (p50), and
j
B2 (p52), share a common domain called a Rel homology do-
j
main (RHD) and form active transcription factors through homo4
and hetero-dimerization5 capable of binding DNA at specific pro-
moter sequences. The p50 and p52 proteins come from larger pro-
teins, p105 and p100 respectively, through post translational
processes. Control of NF-
of signal transduction pathway activation. For example, one well
studied signaling pathway that activates NF- B controlled gene
transcription is the tumor necrosis factor (TNF) pathway.6 Inactive
jB transcriptional activation is the result
j
dimers of NF-jB are held in complex with inhibitors of jB (IjB) in
the cytosol of unstimulated cells.7 Upon stimulation with TNF
the TNF receptor (TNFR) initiates a phosphorylation cascade that
2. Results and discussion
results in inhibitor of
jB kinase (IKK) phosphorylation of and
2.1. Design and molecular docking
⇑
We designed a series of potential inhibitors of NF-
tional activation based upon a reported pharmacophore model for
jB transcrip-
Corresponding author.
E-mail address: wrussu@pacific.edu (W.A. Russu).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.10.051

L. Xu, W. A. Russu / Bioorg. Med. Chem. 21 (2013) 540–546
541
Table 1
chosen for further study (Table 1). Figure 1 depicts the hypothesis
for binding to the p50 protein of 2a and 2b. Docking results are re-
ported in Table 1.
Calculated binding energies of compounds docked to
p50. Energies were calculated using AutoDock 4.2
Compound
DG (kcal/mol)
Examination of Table 1 reveals that for every selected pair the
2-phenyl analog is predicted to bind with a more negative free en-
ergy and thus a stronger association with the p50 protein. In each
case, the predicted binding geometry of the lowest energy docked
complex placed the 2-phenyl substituent in a hydrophobic pocket
where favorable intermolecular interactions with the aromatic side
chain of Phe53 and other hydrophobic amino acids contribute to
binding. Figure 1 depicts the predicted geometry of 2a compared
to 2b bound to the p50 protein. Other notable intermolecular inter-
actions include the 2a indole ring system with His64 through pi–pi
stacking and with the Val58 backbone amide through an H-bond.
1a
1b
2a
2b
3a
3b
4a
4b
5a
5b
6a
6b
À9.09
À7.63
À9.78
À7.81
À8.41
À6.27
À8.86
À6.89
À9.09
À8.22
À8.52
À6.98
2.2. Synthesis of quinazoline analogues
binding p50, derived from a series of 6-aminoquinazoline ana-
logs.13 Upon examining the proposed binding mode of one of these
analogs to p50 we discovered the possibility that substitution at
the 2-position of the quinazoline ring may be tolerated and result
in a gain of binding affinity. We also reasoned, based on the bind-
ing hypothesis, that the 6-amino group may not be necessary for
binding to the target. A series of 2-phenylquinazolines linked from
the 4-position via aminoethyl or piperazinyl group to an aromatic
substituent were designed. A second series without a 2-phenyl
substituent were also designed for comparison. A total of 22 mol-
ecules were designed based upon readily available starting materi-
als, a previously report pharmacophore model and were docked to
the NF-kB p50 protein (Autodock 4.2). The molecules were synthe-
sized and evaluated for their inhibitory effect against the HepG2
The designed molecules were synthesized in few steps starting
from either 4-chloroquinazoline or 2-phenyl-4-chloroquinazoline.
The quinazoline starting materials were either allowed to react
with the appropriate amine to give the final molecule directly or
with mono-Boc piperazine followed by removal of the Boc group
and subsequent reductive amination with an aldehyde in order
to provide the final molecule. Molecules 1–6 a and b series analogs
were synthesized according to Scheme 1. Yields of final recrystal-
lized products for the one or three step synthetic procedure were
moderate to good.
2.3. TNFa enhances anti-proliferation effects of designed
cell line at 1 lM. A two-tailed Pearson correlation was performed
molecules
which revealed a relationship between inhibition and docking free
energy (see Supplementary data Fig. 1). Representative examples
displaying low, medium and high binding free energies were
The HepG2 cell line was used to assess whether the paired
molecular series antitumor cell effects correlate with NF-
jB
Figure 1. Representations of lowest energy docking poses of compounds 2a (panel A) and 2b (panel B) bound to the NF-
p50 and 2a (panel C) and 2b (panel D) are highlighted by 2-D interaction maps.
jB p50 protein. Intermolecular interactions between

542
L. Xu, W. A. Russu / Bioorg. Med. Chem. 21 (2013) 540–546
Scheme 1. Chemical synthesis of molecules 1–6.
signaling. When stimulated with tumor necrosis factor (TNF
HepG2 cell line activates both pro and anti-apoptosis genes.14
These anti-apoptosis genes are expressed in an NF- B dependent
manner. Therefore, HepG2 cell growth will be inhibited to a greater
extent when treated with an NF- B pathway inhibitor in the pres-
ence of TNF , than in the absence of TNF , by increased apoptosis
a) the
j
Table 2
Percent cell growth of HepG2 cells treated with compounds 1–6a in the
absence and presence of TNF
a
j
a
a
Compound
% Cell growth ( SD)b
without TNF
due to shifting the balance between pro and anti-apoptosis pro-
a
With TNFa
teins toward pro-apoptosis. Table 2 reports the percent growth of
1a
1b
2a
2b
3a
3b
4a
4b
5a
5b
6a
6b
À30.7 (2.7)
À20.1 (1. 7)
À35.3 (2.3)
À22.3 (2.9)
À37.6 (2.6)
À5.3 (1.4)
15.5 (3.1)
À46.8 (7.0)
À28.1 (2.3)
À46.7 (6.4)
À30.0 (3.0)
À49.1 (6.1)
À13.3 (6.6)
13.0 (2.4)
HepG2 cells treated with 50
or absence of TNF . For molecules 1–3 it is evident that they are
more effective in the presence of TNF (p <0.05), and that the
lM of molecules 1–6 in the presence
a
a
growth inhibitory effect is of a greater magnitude for the 2-phenyl
substituted analogs. This result supports the hypothesis that the
2-phenyl substituted analogs are better NF-
For molecules 4–6 the results varied. For example, 4b was more
effective in the presence of TNF whereas 4a was not. Neither of
molecule 5a or 5b showed a significant difference in the presence
of TNF (p >0.4). Molecule 6a was better at inhibiting the growth of
HepG2 cells in the presence of TNF whereas 6b did not show a
jB pathway inhibitors.
À22.0 (2.5)
À29.9 (2.4)
14.6 (1.2)
13.0 (2.4)
À21.7 (5.9)
30.7 (0.4)
À16.7 (4.5)
À26.0 (5.7)
13.7 (1.2)
À17.6 (4.8)
a
a
a
50
l
M.
a
b
SD: standard deviation for 3 independent experiments.
significant difference (p >0.4). Molecules 1–3 and 6 behaved as ex-
pected in that the 2-phenyl analogs were better at inhibiting cell
growth in the presence of TNFa than in the absence of TNFa. Only
molecule 4 showed the opposite tendency.
shown in Figure 2 is the effect of 2a on NF-jB dependent Luciferase
expression at different doses compared to the commercially avail-
2.4. Designed molecules downregulate NF-
transcription
jB dependent gene
able NF- B pathway inhibitor QNZ. The molecules 2a and QNZ
have similar activity.
j
We next turned our attention to assessing whether the new
compounds could inhibit NF- B dependent gene transcription.
For this inquiry we used a Luciferase reporter assay. HepG2 cells
2.5. Designed molecule induce apoptosis in tumor cells
Encouraged by the growth inhibition of human tumor cell lines
j
were transfected with the Luciferase gene under control of an
by the new designed molecules and their ability to inhibit NF-jB
transcriptional activity, we next assessed whether these activities
were accompanied by apoptosis. We compared the relative
NF-
2-H series were assayed in addition to a commercially available
quinazoline-based NF-
B inhibitor called QNZ.11,16 The results of
j
B promoter.15 The molecules from both the 2-phenyl and
j
amount of apoptosis of EKVX cells treated with TNF
sence of and presence of 2a using terminal deoxynucleotidyltrans-
ferase (TdT) nick-end labeling (TUNEL) assay. The TNF stimulated
cells did not show apoptosis when stained with TUNEL reagent in
the absence of 2a. However, TNF treated cells in the presence of
a in the ab-
the Luciferase reporter assay are summarized in Figure 2. The
results of the Luciferase reporter assay mirror the HepG2 growth
inhibition experiment in that 1a–3a and 6a significantly decrease
a
(p <0.05) NF-
of TNF , whereas 1b–3b and 6b did so to a lesser extent. There
does not appear to be a significant difference in the ability of 4a
and 5a, to inhibit the NF- B Luciferase expression, compared to
j
B dependent Luciferase expression in the presence
a
a
2a showed an increase in the number of TNNEL positive cells when
examined under fluorescence microscopy. This is a positive indica-
j
tion that 2a can initiate apoptosis in the presence of TNF
together with the cell growth inhibition experiments and NF-
dependent Luciferase expression strongly suggests that molecule
a, which
their 2-H analogs 4b and 5b (p = 0.8 and 0.1 respectively) which
is also consistent with the aforementioned growth inhibition
experiment, and suggests that these molecules may exert their
antitumor activity by a different or multiple mechanisms. Also
jB
2a exerts its antitumor cell activity by inhibition of NF-jB.
Figure 3 depicts the TUNEL assay results.

L. Xu, W. A. Russu / Bioorg. Med. Chem. 21 (2013) 540–546
543
Figure 2. Luciferase reporter assay performed with transfected HepG2 cells pretreated with 1
lM of indicated molecules 1–6 and stimulated with TNF
a. (A) Bars represent
relative NF- B dependent luciferase activity compared to TNFa stimulated cells in the absence of molecules 1–6. Control represents luciferase activity in non-TNFa
j
stimulated HepG2 cells in the absence of experimental molecules; (B) Comparison of NF-jB dependent luciferase activity in TNFa stimulated HepG2 cells pre-treated with
either QNZ or 2a.
Table 3
NCI dose response report of 2a and 2b
Tissue
type
Cell line
GI50
(
l
M) Tissue
Cell line
GI50
(
l
M)
type
2a
2b
2a
2b
Colon
cancer
COLO
205
HCC-
2998
2.2 28.2 Ovarian
cancer
2.1 26.3
IGROV1
3.2 14.1
2.9 19.1
OVCAR-3
HCT-116 2.8 6.3
OVCAR-4
OVCAR-8
NCI/ADR-
RES
3.6
2.2
2.6
1.5
5.5
5.6
HCT-15
KM12
2.5 24.0
2.8 19.1
SW-620
SF-268
3.0 41.7
SK-OC-3
3.7 58.9
2.5 33.9
CNS
cancer
4.6 34.7 Renal
cancer
786-0
SF-295
SF-539
SNB-19
SNB-75
U251
2.0 3.5
A498
2.0
3.1
1.9 30.9
3.9 30.2
2.8 20.4
2.8
2.2
2.0 39.8
5.1 50.1
5.9 22.9
3.2 30.9
CAKl-1
RXF 393
SN12C
TK-10
Melanoma LOX
IMVI
2.2 15.5 Breast
cancer
MCF-7
4.2 11.2
MALME-
3 M
M14
2.8 29.5
MDA-MB-
231/ATCC
HS 578T
BT-549
3.1 21.9
2.6 24.6
2.3 10.2
3.4 60.3
4.4 12.3
MDA-
MB-435
SK-MEL-
2
SK-MEL-
28
UACC-
357
UACC-62 2.0 14.1
3.0 30.9
6.2 25.1
T-47D
UO-31
PC-3
2.4 12.0
3.0 14.5
2.4 19.5
4.7 27.5
3.2 9.8
Prostate
cancer
DU-145
(NCI-60) within the context of the Developmental Therapeutics
Program.17 Molecules 2a and 2b were selected for dose response
testing and fifty percent growth inhibition concentrations (GI50
)
were determined according to the SRB protocol.18 Table 3 gives a
comparison of GI50 values for both 2a and 2b on all cell lines tested.
In all cases but two, molecule 2a, the 2-phenyl substituted analog,
is more effective at inhibiting the growth of tumor cells.
Figure 3. EKVX cells in the absence (control) of or pretreated with 2a (1
lM) were
stimulated with TNF (20 ng/mL) for 18 h and then subjected to TUNEL assay.
a
2.6. NCI-60 cell line activity
3. Conclusion
In order to test the designed molecules ability to inhibit the
growth of a variety of tumor cells, these molecules were screened
against the National Cancer Institute’s 60 tumor cell line panel
We have described the design, synthesis and evaluation of qui-
nazoline based NF-jB pathway inhibitors. Based on a previously

544
L. Xu, W. A. Russu / Bioorg. Med. Chem. 21 (2013) 540–546
reported pharmacophore model, we incorporated a 2-phenyl sub-
stitution of the quinazoline ring that enhanced the inhibition of
J = 0.48 Hz, J = 8.22 Hz, 1H), 7.72–7.66 (m, 2H), 7.52–7.44 (m, 4H),
7.38 (d, J = 8.1 Hz, 1H), 7.32 (ddd, J = 1.20 Hz, J = 7.08 Hz,
J = 8.28 Hz 1H), 7.23 (ddd, J = 1.02 Hz, J = 7.08 Hz, J = 8.10 Hz, 1H),
7.14 (ddd, J = 0.90 Hz, J = 7.08 Hz, J = 7.92 Hz, 1H), 7.06 (d,
J = 2.20 Hz, 1H), 5.84 (t, J = 5.46 Hz, 1H), 4.12 (q, J = 6.72 Hz 2H),
3.25 (t, J = 7.08 Hz, 2H); 13C NMR (CDCl3) d 160.8, 159.6, 150.6,
139.2, 136.5, 132.5, 130.2, 128.8, 127.5, 125.4, 122.4, 120.6,
119.7, 118.9, 113.8, 113.3, 111.5, 41.6, 25.1; MS Calcd for
NF-
sessed by Luciferase and TUNEL assays respectively. While these
compounds were designed upon and our results support a NF-
jB dependent gene transcription and initiated apoptosis as as-
jB
p50 binding hypothesis, other possible mechanisms such as inhibi-
tion of upstream kinases cannot be ruled out at this time. We are
currently engaged in a more detailed mechanistic investigation of
2a and will report our findings in due course.
C17H15N3O2 364.1688. Found MS (DART-TOF) m/z 364.1789
(M+H)+.
4. Experimental
4.1. General
4.2.4. N-(2-(1H-indol-3-yl)ethyl)quinazolin-4-amine (2b)
The title compound was prepared starting from 4-chloroquinaz-
oline and 2-(1H-indol-3-yl)ethanamine using the same procedure
described for 1a to afford a pale yellow solid (52.7% yield): mp
162 °C; 1H NMR (CDCl3) d 8.69 (s, 1H), 8.19 (s, 1H), 7.81 (d,
J = 7.74 Hz, 1H), 7.69 (ddd, J = 1.38 Hz, J = 7.08 Hz, 8.46 Hz, 1H),
7.66 (dd, J = 1.02 Hz, J = 7.92 Hz, 1H), 7.43 (dd, J = 1.20 Hz,
J = 8.22 Hz, 1H), 7.40 (dt, J = 0.90 Hz, J = 8.28 Hz, 1H), 7.36 (ddd,
J = 1.20 Hz, J = 7.02 Hz, J = 8.22 Hz, 1H), 7.23 (ddd, J = 1.02 Hz,
J = 7.26 Hz, J = 8.28 Hz,1H), 7.14 (ddd, J = 1.02 Hz, J = 7.14 Hz,
J = 8.10 Hz, 1H), 7.09 (d, J = 2.4 Hz, 1H), 5.84 (s, 1H), 4.00 (q,
J = 6.54 Hz 2H), 3.20 (td, J = 0.54 Hz, J = 6.54 Hz, 2H); 13C NMR
(CDCl3) d 159.3, 155.5, 149.4, 136.6, 132.6, 128.6, 127.5, 125.9,
122.5, 122.3, 120.5, 119.8, 118.8, 115.1, 113.1, 111.5, 41.5, 24.9;
MS Calcd for C18H16N4 288.1375. Found MS (MALDI-TOF) m/z
289.1457 (M+H)+.
All the chemicals were obtained from Sigma–Aldrich and
ARCOS, and were used for synthesis without further purification.
Proton and 13C NMR were measured by JEOL ECA 600 MHz instru-
ment. Mass spectra were analyzed by KRATOS/ SHIMADZU AXIMA-
CFR MALDI-TOF. The luciferase activities were measured using
GloMax 96 Microplate Luminometer with Dual injectors (Promega,
Medison, WI).
4.2. Synthetic procedures
4.2.1. Synthesis of N-(2-(benzo[d][1,3]dioxol-5-yl)ethyl)-2-
phenylquinazolin-4-amine (1a)
Triethylamine (0.28 ml, 2 mmol) was added to a suspension of
4-chloro-2-phenylquinazoline (0.36 g, 1.5 mmol) and 3,4-methy-
lenedioxyphenethylamine hydrochloride (0.30 g, 1.5 mmol) in
10 ml 2-propanol. The reaction was heated at reflux and monitored
using TLC. Once the reaction was finished, solvent was removed
under reduced pressure and the residue was purified using column
chromatography (3:1 Hexane/Ethyl acetate) to give an off-white
solid. (210.5 mg, 38.3% yield): mp 105 °C; 1H NMR (CDCl3) d
8.60–8.57 (m, 2H), 7.92 (d, J = 7.92 Hz, 1H), 7.71 (ddd, J = 1.38 Hz,
J = 7.08 Hz, 1H), 7.54–7.58 (m, 1H), 7.52–7.345 (m, 3H), 7.38
(ddd, J = 1.20 Hz, J = 7.08 Hz, J = 8.28 Hz, 1H), 6.78 (t, J = 1.68 Hz,
1H), 6.77 (s, 1H), 6.72 (dd, J = 1.56 Hz, 1H), 5.94 (s, 2H), 5.75 (t,
J = 4.68 Hz 1H), 4.00 (q, J = 6.90 Hz, 2H), 3.00 (t, J = 6.90 Hz 2H);
13C NMR (CDCl3) d 160.5, 159.4, 150.5, 147.9, 146.3, 138.9, 132.8,
132.5, 130.0 128.8, 128.4, 128.3, 125.4, 121.7, 120.3, 113.7, 109.1,
108.4, 100.9, 42.4, 35.1; MS Calcd for C23H19N3O2 369.1477. Found
MS (DART-TOF) m/z 369.1527 (M+H)+.
4.2.5. Synthesis of 2-phenyl-4-(piperazin-1-yl)quinazoline
Potassium carbonate (1.38 g, 10 mmol) was added to
a
suspension of 4-chloro-2-phenylquinazoline (1.2 g, 5 mmol) and
1-Boc-piperazine (0.93 g, 5 mmol) in 7 ml DMF. The reaction was
heated while refluxing and monitored using TLC. Once finished,
the reaction was acidified with acetic acid and a white precipitate
formed. The product was filtered and partitioned (CH2Cl2: water
1:1). The organic layer was concentrated under pressure and white
solid was formed. The white solid was added to 50 ml methanolic
HCl and stirred for 1 h followed by the neutralization with NaHCO3.
The precipitate (white solid, 1.18 g, 81.4% yield) was filtered and
used in the next step without further purification. 1H NMR (CDCl3)
d 8.56–8.53 (m, 2H), 7.96 (dd, J = 0.66 Hz, J = 8.40 Hz, 1H), 7.88 (dd,
J = 0.90 Hz, J = 8.28 Hz, 1H), 7.71 (ddd, J = 1.38 Hz, J = 6.90 Hz,
J = 8.40 Hz, 1H), 7.50–7.43 (m, 3H), 7.40 (ddd, J = 1.20 Hz,
J = 6.90 Hz, J = 8.28 Hz, 1H), 3.83 (t, J = 4.80 Hz, 4H), 3.13 (t,
J = 4.80 Hz, 4H); MS Calcd for C18H18N4 290.15. Found MS (MAL-
DI-TOF) m/z 290.8 (M+H)+.
4.2.2. Synthesis of N-(2-(benzo[d][1,3]dioxol-5-
yl)ethyl)quinazolin-4-amine (1b)
The title compound was prepared starting from 4-chloroquinaz-
oline and 3,4-methylenedioxyphenethylamine hydrochloride
using the same procedure described for 1a to afford a white solid
(31.2% yield): mp 187 °C; 1H NMR (CDCl3) d 8.66 (s, 1H), 7.80 (d,
J = 7.74 Hz, 1H), 7.70 (ddd, J = 1.38 Hz, J = 7.02 Hz, J = 8.40 Hz, 1H),
7.55 (d, J = 1.56 Hz, 1H), 7.41 (ddd, J = 1.20 Hz, J = 7.02 Hz,
J = 8.10 Hz, 1H), 6.75 (d, J = 7.86 Hz, 1H), 6.73 (d, J = 1.74 Hz, 1H),
6.67 (dd, J = 1.68 Hz, J = 7.92 Hz, 1H), 5.93 (s, 2H), 5.84 (s, 1H),
3.87 (q, J = 6.84 Hz, 2H), 2.94 (t, J = 6.90 Hz, 2H); 13C NMR (CDCl3)
d 159.4, 155.5, 149.5, 148.1, 146.4, 132.6, 128.7, 126.1, 121.8,
120.4, 115.0, 109.2, 108.6, 101.1, 42.5, 35.0; MS Calcd for
4.2.6. Synthesis of 4-((4-(2-phenylquinazolin-4-yl)piperazin-1-
yl)methyl)phenol (3a)
2-phenyl-4-(piperazin-1-yl)quinazoline (0.29 g, 1 mmol) and 4-
hydroxybenzaldehyde (0.12 g, 1 mmol) dissolved in 10 ml 1,2-
dichloroethene and NaBH(OAc)3 (0.21 g, 1.3 mmol) was added.
The reaction was stirred at room temperature overnight and then
partitioned between saturated aqueous NaHCO3 and ethyl acetate.
The organic layer was concentrated under reduced pressure and
then purified using column chromatography (6:1 Hexane/Ethyl
acetate) to give pale yellow solid. (0.132 g, 33.3% yield): 1H NMR
(DMSO-d6) d 9.27 (s, 1H), 8.46–8.43 (m, 2H), 7.96 (dd, J = 0.66 Hz,
J = 8.22 Hz, 1H), 7.84 (dd, J = 0.84 Hz, J = 8.22 Hz, 1H), 7.77 (ddd,
J = 1.38 Hz, J = 6.90 Hz, J = 8.28 Hz, 1H), 7.47 (m, 4H), 7.10 (dt,
J = 2.58 Hz, J = 8.58 Hz, 2H), 6.69 (dt J = 2.76 Hz, J = 9.30 Hz, 2H),
3.78 (t, J = 4.14 Hz, 4H), 3.41 (s, 2H), 2.55 (t, J = 4.62 Hz, 4H); MS
Calcd for C25H24N4O 396.1950. Found MS (MALDI-TOF) m/z 396.7
(M+H)+.
C
17H15N3O2 293.1164. Found MS (DART-TOF) m/z 293.6 (M+H)+.
4.2.3. N-(2-(benzo[d][1,3]dioxol-5-yl)ethyl)quinazolin-4-amine
(2a)
The title compound was prepared starting from 4-chloroquinaz-
oline and 2-(1H-indol-3-yl)ethanamine using the same procedure
described for 1a to afford a white solid (26.6% yield): mp 166 °C;
1H NMR (CDCl3) d 8.64–8.58 (m, 2H), 8.19 (s, 1H), 7.91 (dd,

L. Xu, W. A. Russu / Bioorg. Med. Chem. 21 (2013) 540–546
545
4.2.7. Synthesis of 4-((4-(quinazolin-4-yl)piperazin-1-yl)methyl)
phenol (3b)
7.73 (ddd, J = 1.56 Hz, J = 7.08 Hz, J = 8.46 Hz, 1H), 7.45 (ddd,
J = 1.20 Hz, J = 6.84 Hz, J = 8.22 Hz, 1H), 7.19 (td, J = 1.68
Hz, J = 8.04 Hz, 1H), 6.99 (dd, J = 1.38 Hz, J = 7.38 Hz, 1H), 6.85
(dd, J = 1.02 Hz, J = 8.04 Hz, 1H), 6.80 (td, J = 1.02 Hz, J = 7.38 Hz,
1H), 3.84 (s, 4H), 3.78 (s, 2H), 3.76 (s, 4H); 13C NMR (CDCl3) d
164.6, 157.6, 154.1, 151.9, 132.7, 129.2, 128.9, 125.7, 124.8,
120.8, 119.5, 116.7, 116.3, 61.7, 52.5, 49.6; MS Calcd for
The title compound was prepared starting from 4-(piperazin-1-
yl)quinazoline and 4-hydroxybenzaldehyde using the same proce-
dure described for 3a to afford white solid (44.7% yield). Mp
188–189 °C; 1H NMR (CDCl3) d 8.69 (s, 1H), 7.89 (d, J = 8.46 Hz,
1H), 7.86 (dd, J = 0.90 Hz, J = 8.46 Hz, 1H), 7.71 (ddd, J = 1.20
Hz, J = 6.84 Hz, J = 8.22 Hz, 1H), 7.42 (ddd, J = 1.20 Hz, J = 7.08 Hz,
J = 8.28 Hz, 1H), 7.15 (d, J = 8.46 Hz, 2H), 6.79 (d, J = 8.46 Hz, 2H),
3.83 (t, J = 4.86 Hz, 4H), 3.52 (s, 2H), 2.64 (t, J = 4.80 Hz, 4H); MS
C
19H20N4O 320.1637. Found MS (DART-TOF) m/z 321.1730 (M+H)+.
4.2.12. Synthesis of 4-(4-benzylpiperazin-1-yl)-2-
phenylquinazoline (6a)
Calcd for
C19H20N4O 320.1637. Found MS (DART-TOF) m/z
321.1715 (M+H)+.
The title compound was prepared starting from2-phenyl-4-
(piperazin-1-yl)quinazoline and benzaldehyde using the same
procedure described for 3a to afford white solid (40.2% yield):
mp 118 °C; 1H NMR (CDCl3) d 8.58–8.55 (m, 2H), 7.97 (dd,
J = 0.72 Hz, J = 8.46 Hz, 1H), 7.88 (dd, J = 1.02 Hz, J = 8.46 Hz, 1H),
7.71 (ddd, J = 1.38 Hz, J = 6.90 Hz, J = 8.46 Hz, 1H), 7.51–7.45 (m,
3H), 7.41–7.33 (m,5H), 7.29 (tt, J = 1.56 Hz, J = 6.54 Hz, 1H), 3.89
(t, J = 4.68 Hz, 4H), 3.61 (s, 2H), 2.69 (t, J = 4.80 Hz, 4H); 13C NMR
(CDCl3) d 164.9, 159.5, 152.9, 132.4, 130.2, 129.3, 129.0, 128.5,
128.4, 127.3, 125.0, 124.9, 115.5, 63.2, 53.1, 49.9; MS Calcd for
4.2.8. Synthesis of 4-bromo-2-((4-(2-phenylquinazolin-4-yl)
piperazin-1-yl)methyl)phenol (4a)
The title compound was prepared starting from 2-phenyl-4-
(piperazin-1-yl)quinazoline and 5-bromo-2-hydroxybenzaldehyde
using the same procedure described for 3a to afford white solid
(36.8% yield). Mp 189–190 °C; 1H NMR (CDCl3)
d 8.54 (d,
J = 1.56 Hz, 1H), 8.53 (s, 1H), 7.99 (d, J = 8.40 Hz, 1H), 7.87 (d,
J = 8.22 Hz, 1H), 7.74 (t, J = 7.86 Hz, 1H), 7.51–7.41 (m, 4H), 7.29
(dd, J = 2.40 Hz, 1H), 7.13 (d, J = 2.22 Hz, 1H), 6.75 (d, J = 8.58 Hz,
1H), 5.29 (s, 1H), 3.93 (s, 4H), 3.77 (s, 2H), 2.81 (s, 4H); MS Calcd
for C25H23BrN4O 474.1055. Found MS (DART-TOF) m/z 475.1095
(M+H)+.
C
25H24N4 380.2001. Found MS (DART-TOF) m/z 381.2075 (M+H)+.
4.2.13. Synthesis of 4-(4-benzylpiperazin-1-yl)quinazoline (6b)
The title compound was prepared starting from 4-
(piperazin-1-yl)quinazoline and benzaldehyde using the same
procedure described for 3a to afford pale yellow solid (35.1%
yield): mp 100–102 °C; 1H NMR (CDCl3) d 8.71 (s, 1H), 7.88
(dd, J = 0.54 Hz, J = 8.42 Hz, 1H), 7.87 (dd, J = 0.90 Hz, J = 8.52 Hz,
1H), 7.71 (ddd, J = 1.38 Hz, J = 7.02 Hz, 1H), 7.42 (ddd,
J = 1.20 Hz, J = 7.08 Hz, J = 8.28 Hz, 1H), 7.37–7.31 (m, 4H), 7.27
(tt, J = 1.74 Hz, J = 6.84 Hz, 1H), 3.80 (t, J = 4.68 Hz, 4H), 3.59 (s,
2H), 3.65 (t, J = 4.80 Hz, 4H); 13C NMR (CDCl3) d 164.6, 154.1,
151.8, 137.6, 132.5, 129.3, 128.6, 128.4, 128.3, 127.4, 125.4,
125.1, 63.1, 53.0, 49.8; MS Calcd for C19H20N4 304.1688. Found
MS (DART-TOF) m/z 305.1777 (M+H)+.
4.2.9. Synthesis of 4-bromo-2-((4-(quinazolin-4-yl)piperazin-1-
yl)methyl)phenol (4b)
The title compound was prepared starting from 4-(piperazin-1-
yl)quinazoline and 5-bromo-2-hydroxybenzaldehyde using the
same procedure described for 3a to afford white solid (37.3%
yield): mp 176–178 °C; 1H NMR (CDCl3) d 8.74 (s, 1H), 7.91 (dd,
J = 0.46 Hz, J = 8.40 Hz, 1H), 7.84 (dd, J = 0.84 Hz, J = 8.40 Hz, 1H),
7.74 (ddd, J = 1.38 Hz, J = 6.90 Hz, J = 8.28 Hz, 1H), 7.45 (ddd,
J = 1.38 Hz, J = 7.02 Hz, J = 8.28 Hz, 1H), 7.27 (dd, J = 2.58 Hz,
J = 8.58 Hz, 1H), 7.11 (d, J = 2.40 Hz, 1H), 6.73 (d, J = 8.64 Hz, 1H),
3.82 (s, 4H), 3.74 (s, 2H), 2.75 (s, 4H); 13C NMR (CDCl3) d 164.6,
156.8, 154.1, 151.9, 132.8, 131.9, 131.4, 128.9, 125.8, 118.2,
4.3. Molecular docking
116.7, 111.2, 61.1, 52.5, 49.6; MS Calcd for
C19H19BrN4O
398.0742. Found MS (DART-TOF) m/z 399.0788 (M+H)+.
Model building and binding energy calculation of designed
compounds were carried out by Autodock 4.2 and the structural
information was illustrated by Discovery Studio 3.1. In brief, the
4.2.10. Synthesis of 2-((4-(2-phenylquinazolin-4-yl)piperazin-1-
yl)methyl)phenol (5a)
crystal structure of NF-jB & DNA complex (PDB code: 1LE5) was
The title compound was prepared starting from 2-phenyl-4-
(piperazin-1-yl)quinazoline and 2-hydroxybenzaldehyde using
the same procedure described for 3a to afford white solid (33.2%
yield): mp 171–172 °C; 1H NMR (CDCl3) d 8.56–8.53 (m, 2H),
7.99 (dd, J = 0.66, J = 8.46 Hz, 1H), 7.86 (dd, J = 0.90 Hz,
J = 8.28 Hz, 1H), 7.73 (ddd, J = 1.38 Hz, J = 6.90 Hz, J = 8.46 Hz, 1H),
7.51–7.45 (m, 3H), 7.42 (ddd, J = 1.20 Hz, J = 6.90 Hz, J = 8.28 Hz,
1H), 7.21 (td, J = 1.74 Hz, J = 7.92 Hz, 1H), 7.01 (dd, J = 1.20 Hz,
J = 7.38 Hz, 1H), 6.87 (dd, J = 1.20 Hz, J = 8.10 Hz, 1H), 6.81 (td,
J = 1.20 Hz, J = 7.38 Hz, 1H), 3.92 (s, 4H), 3.80 (s, 2H), 2.80 (s, 4H);
13C NMR (CDCl3) d 164.8, 159.4, 157.5, 152.8, 138.4, 132.6, 130.3,
129.2, 129.1, 128.8, 128.4, 125.1, 124.6, 120.7, 119.4, 116.2,
115.4, 61.6, 52.4, 49.7; MS Calcd for C25H24N4O 396.1950. Found
MS (DART-TOF) m/z 396.2042 (M+H)+.
obtained from Protein Data Bank. The DNA molecule was removed
and hydrogen atoms were added. One hundred different conforma-
tions of each designed molecule were generated by Lamarkian
algorithm and docked to the putative drug binding region.13 The
lowest binding energy of each molecule conformer was calculated
using the default parameters of Autodock 4.
4.4. Cell growth inhibition assay
Cells were plated in 96-well plates and allowed to adhere for
24 h and the T0 plate was fixed and treated according to the SRB
protocol. The compounds at appropriate concentration were added
into the sample plates and incubate for 1 h at 37 °C followed by
stimulation with 20 ng/ml TNF-
then incubated for an additional 48 h before fixation. The plates
were washed and air dried. Into each well was added 50 l 0.4%
w/v sulphorhodamine B (SRB) (Sigma–Aldrich) at room tempera-
ture for 30 minutes, followed by washing with 1% acetic acid.
The bound stain was solubilized with 10 mM Tris base and the
OD was read at 515 nm. One-way ANOVA was employed to detect
significant differences between mean values.
a of some wells. The cells were
4.2.11. Synthesis of 2-((4-(quinazolin-4-yl)piperazin-1-yl)
methyl)phenol (5b)
l
The title compound was prepared starting from 4-(piperazin-1-
yl)quinazoline and 2-hydroxybenzaldehyde using the same
procedure described for 3a to afford white solid (50.4% yield):
mp 118–120 °C; 1H NMR (CDCl3)
d
8.74 (s, 1H), 7.90 (dd,
J = 0.66 Hz, J = 8.40 Hz, 1H), 7.84 (dd, J = 0.90 Hz, J = 8.46 Hz, 1H),

546
L. Xu, W. A. Russu / Bioorg. Med. Chem. 21 (2013) 540–546
4.5. Luciferase assay
Research Fellowship (WAR) from UOP FRC. NMR instrumentation
was facilitated by National Science Foundation grant MRI-
0722654. The authors thank Dr. William Chan, Dr. Andreas Franz,
Patrick Batoon and Liang Zhao for technical help and useful
discussion. The authors also appreciate Dr. Kathy M. Giacomini
and Dr. Ligong Chen for their gracious support of LX during
internship at UCSF.
HepG2 cells were cultured in Dulbecco’s Modified Eagle
Medium (DMEM) supplemented with 10% fetal bovine serum,
100 U/ml of penicillin and 100
were plated in 96-well plates and transfected using the lipofect-
amine 2000 (Invitrogen) with 10 ng/well pGL 4.32 NF- B-Luc (Pro-
lg/ml of streptomycin. The cells
j
mega) and Renilla control in Opti-MEM reduced serum medium
(Invitrogen) following the protocol provided by manufacturers.
Test compounds were dissolved in DMSO and added at the appro-
priate concentration to the 96-well plate with fresh medium and
the plates were then incubated at 37 °C for 1 h. For induction of
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.10.051.
These data include MOL files and InChiKeys of the most important
compounds described in this article.
transcription, TNF-a was added (final concentration: 20 ng/ml)
and the cells were incubated for an additional 6 h before lysis with
passive lysis buffer. The luciferase and Renilla activity were mea-
sured in a 96-well plate using the Dual-Luciferase Reporter (DLR)
Assay Systems (Promega). The luciferase activity was determined
as the relative light units of firefly luciferase activity per unit of
Renilla reniformis luciferase. One-way ANOVA was employed to
detect significant differences beween mean values.
References and notes
1. Karin, M.; Yamamoto, Y.; Wang, Q. M. Nat. Rev. Drug Disc. 2004, 3, 17.
2. Karin, M.; Lin, A. Nat. Immunol. 2002, 3, 221.
3. Pommier, Y.; Sordet, O.; Antony, S.; Hayward, R. L.; Kohn, K. W. Oncogene 2004,
23, 2934.
4. Kieran, M.; Blank, V.; Logeat, F.; Vandekerckhove, J.; Lottspeich, F.; Le Bail, O.;
Urban, M.; Kourilsky, P.; Baeuerle, P.; Israel, A. Cell 1990, 1007, 62.
5. Baeuerle, P.; Baltimore, D. Genes Dev. 1989, 3, 1689.
6. Van Antwerp, D. J.; Martin, S. J.; Kafri, T.; Green, D. R.; Verma, I. M. Science 1996,
274, 787.
7. Baeuerle, P.; Baltimore, D. Science 1988, 242, 540.
8. Yamamoto, Y.; Gaynor, R. B. J. Clin. Invest. 2001, 107, 135.
9. Kobayashi, Y.; Yoshimori, A.; Kino, K.; Miyazawa, H.; Tanuma, S.-i. Bioorg. Med.
Chem. 2009, 17, 5293.
10. Pande, V.; Sharma, R. K.; Inoue, J.-I.; Otsuka, M.; Ramos, M. J. J. Comput. Aided
Mol. Des. 2003, 17, 825.
11. Tobe, M.; Isobe, Y.; Tomizawa, H.; Nagasaki, T.; Takahashi, H.; Fukazawa, T.;
Hayashi, H. Bioorg. Med. Chem. 2003, 11, 383.
12. Qian, L.; Shen, Y.; Chen, J.-C.; Wang, Y.-X.; Wu, X.-T.; Chen, T.-J.; Zheng, K.-C.
QSAR Comb. Sci. 2008, 27, 984.
13. Tsai, K.-C.; Teng, L.-W.; Shao, Y.-M.; Chen, Y.-C.; Lee, Y.-C.; Li, M.; Hsiao, N.-W.
Bioorg. Med. Chem. Lett. 2009, 19, 5665.
14. Van Horssen, R.; Ten Hagen, T. L. M.; Eggermont, A. M. M. Oncologist 2006, 11,
397.
4.6. TUNEL assay
Terminal deoxynucleotidyltransferase (TdT) nick-end labeling
(TUNEL) assay was performed on slides with a Trevigen TACS 2
TdT (TBL) kit (Trevigen) following the protocol provided by the
manufacturer. In brief, EKVX cells were cultured RPMI 1640 with
10% fetal bovine serum, 100 U/ml of penicillin and 100 lg/ml of
streptomycin and seeded on sterile glass coverslips placed in
12-well plates. The cells were treated with designed compound
at the appropriate concentration for 1 h at 37 °C and then stimu-
lated with 20 ng/ml TNF-
a for another 18 h. After fixing the cells
with 3.7% paraformaldehyde, the cells were covered with protein-
ase K solution for 30 min, followed by the treatment of labeling
reaction mix, stop buffer, Strep_Fluorescein Solution and DAPI
stain.
15. Matsumoto, N.; Ariga, A.; To-e, S.; Nakamura, H.; Agata, N.; Hirano, S.-i.; Inoue,
J.-i.; Umezawa, K. Bioorg. Med. Chem. Lett. 2000, 10, 865.
16. Obtained from Cayman Chemical Company, Ann Arbor, MI, USA.
17. Monga, M.; Sausville, E. A. Leukemia 2002, 16, 520.
18. Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J.; Vistica, D.;
Warren, J. T.; Bokesch, H.; Kenney, S.; Boyd, M. R. J. Natl. Cancer Inst. 1990, 82,
1107.
Acknowledgments
The authors appreciate funding from a Graduate Student
Research Grant (LX) from UOP PCSP committee and Eberhardt