Virtual screening and synthesis of quinazolines as novel JAK2 inhibitors

Article abstract of DOI:10.1016/j.bmc.2010.11.057

JAK2 is an important target in multiple processes associated with tumor growth. In this study, virtual screening was employed for hit compound identification with chemical libraries using SurflexDock. Subsequently, hit optimization for potent and selective candidate JAK2 inhibitors was performed through synthesis of diverse C-1 substituted quinazoline derivatives. A novel compound 5p, (6,7-dimethoxyquinazolin-4-yl)naphthalen-1-ylamine, was thus obtained. JAK2 inhibitory activity of 5p was 43% at 20 μM and this was comparable to AG490, a representative JAK2 inhibitor. Moreover, 5p showed a positive correlation between JAK2 inhibition and cytotoxicity; 5p treatment in HT-29 cells strongly inhibited JAK2 activation and subsequent STAT3 phosphorylation, reduced anti-apoptotic protein levels, and finally induced apoptosis. This suggests that compound 5p is a candidate inhibitor of JAK2 and its downstream STAT3 signaling pathway for antitumor therapy. In the docking model, the quinazoline template of 5k, the lead compound, occupied a hydrophobic region such as Leu856, Leu855, Ala880, Leu932 and Gly935, and the highly conserved hydrogen bond was created by 6-OMe of the ring template, which binds to the NH of Arg980. Moreover, hydrophobic interactions were identified between morpholine moiety and the hydrophobic region formed by Leu855, Ala880, Tyr931, Val911 and Met929. Also, compound 5k more strongly inhibited JAK2 phosphorylation in mouse embryonic stem cells than AG490. Our study shows the successful application of virtual screening for lead discovery and we propose that the novel compound 5p can be an effective JAK2 inhibitor candidate for further antitumor agent research.

Full text of DOI:10.1016/j.bmc.2010.11.057

Bioorganic & Medicinal Chemistry 19 (2011) 968–977
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Virtual screening and synthesis of quinazolines as novel JAK2 inhibitors
Su Hui Yang a, Daulat Bikram Khadka a, Suk Hee Cho a, Hye-Kyung Ju b,c, Kwang Youl Lee a, Ho Jae Han d,
Kyung-Tae Lee b,c, Won-Jea Cho a,
⇑
a College of Pharmacy and Research Institute of Drug Development, Chonnam National University, Gwangju 500-757, Republic of Korea
b Department of Biomedical Science, College of Medical Science, Kyung-Hee University, Seoul 130-701, Republic of Korea
c Department of Pharmaceutical Biochemistry, College of Pharmacy, Kyung-Hee University, Seoul 130-701, Republic of Korea
d Department of Veterinary Physiology, College of Veterinary Medicine, Chonnam National University, Gwangju 500-757, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
JAK2 is an important target in multiple processes associated with tumor growth. In this study, virtual
screening was employed for hit compound identification with chemical libraries using SurflexDock. Sub-
sequently, hit optimization for potent and selective candidate JAK2 inhibitors was performed through
synthesis of diverse C-1 substituted quinazoline derivatives. A novel compound 5p, (6,7-dimethoxyqui-
Received 15 September 2010
Revised 23 November 2010
Accepted 23 November 2010
Available online 30 November 2010
nazolin-4-yl)naphthalen-1-ylamine, was thus obtained. JAK2 inhibitory activity of 5p was 43% at 20 lM
and this was comparable to AG490, a representative JAK2 inhibitor. Moreover, 5p showed a positive cor-
relation between JAK2 inhibition and cytotoxicity; 5p treatment in HT-29 cells strongly inhibited JAK2
activation and subsequent STAT3 phosphorylation, reduced anti-apoptotic protein levels, and finally
induced apoptosis. This suggests that compound 5p is a candidate inhibitor of JAK2 and its downstream
STAT3 signaling pathway for antitumor therapy. In the docking model, the quinazoline template of 5k,
the lead compound, occupied a hydrophobic region such as Leu856, Leu855, Ala880, Leu932 and
Gly935, and the highly conserved hydrogen bond was created by 6-OMe of the ring template, which binds
to the NH of Arg980. Moreover, hydrophobic interactions were identified between morpholine moiety
and the hydrophobic region formed by Leu855, Ala880, Tyr931, Val911 and Met929. Also, compound
5k more strongly inhibited JAK2 phosphorylation in mouse embryonic stem cells than AG490. Our study
shows the successful application of virtual screening for lead discovery and we propose that the novel
compound 5p can be an effective JAK2 inhibitor candidate for further antitumor agent research.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
JAK2
Quinazolines
Virtual screening
SurflexDock
Apoptosis
Stem cell
AG490
1. Introduction
the probable mechanism, cytokine binds to the corresponding
cell-surface receptor to which JAK2 tyrosine kinase attaches.
JAK2 is a member of the Janus kinases (JAKs), which are
intracellular non-receptor protein tyrosine kinases including Jak1,
Jak2, Jak3, and Tyk2. Each kinase has seven Janus homology
domains: a typical kinase domain (JH1) close to the C-terminus,
a pseudokinase domain (JH2) that is nearly identical to JH1 in
amino acid sequence, and the remaining homology regions
(JH3–JH7) containing the FERM domain, which regulates binding
to cytokine receptors.1 JAK2 associates with various cytokine
receptors; it is responsible for signal transduction by mediating
tyrosine phosphorylation.2 The cellular responses to cytokines
are essential for normal cell development and function. In other
words, if the kinase activity is not properly regulated, there can
be detrimental effects such as abnormal cell proliferation.
The JH1 domain of JAK2 contains typical tyrosines (e.g., Y1007/
Y1008) and autophosphorylates the dual tyrosines, leading to a
conformational change in the receptor for JAK2 activation.4 When
the phosphorylation occurs, the suppressive function of JH2 is
relieved.5 The activated JAK2 phosphorylates specific tyrosine
and serine residues of the receptor, creating docking sites for the
signal transducers and activators of transcription (STATs). Then,
STATs bind the receptor, allowing JAK2 in turn to phosphorylate
STATs. Finally, phosphorylated STATs are dissociated from the
receptor, form dimers, and translocate into the nucleus, where
gene transcription is regulated.
The role of the JH2 domain of JAK2 in controlling kinase activity
is worth noting due to the pathological consequences of mutations
in this domain. In 1995, the possibility of de-regulation of JAK2
activity was first suggested through a single mutation in the JH2
domain.6 Later, quite surprisingly, five independent research
groups simultaneously reported the presence of the JAK2 V617F
mutation in several myeloproliferative diseases; it was detected
in most polycythemia vera patients and in half of essential
Jak2 kinase activity is controlled mainly by the inhibitory effect
of the JH2 domain against the activation of the JH1 domain.2b,3 In
⇑
Corresponding author. Fax: +82 62 530 2911.
E-mail address: wjcho@jnu.ac.kr (W.-J. Cho).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.11.057

S. H. Yang et al. / Bioorg. Med. Chem. 19 (2011) 968–977
969
thrombocythemia patients.7 This mutation was somatically ac-
H3CO
H3CO
N
quired and was limited to the myeloid lineages, implying that
the mutation was present and maintained in a myeloid lineage
progenitor stem cell. In addition, chromosomal translocation of
JAK2 or other JAK2 somatic mutations such as amino acid substitu-
tions and deletions are characterized in various leukemias.8 There-
fore, the development of a potent, selective JAK2 inhibitor seems
quite promising for anticancer therapy.
N
HN
N
O
(6,7-Dimethoxyquinazolin-4-yl)morpholin-4-yl-amine (5k)
As the first JAK2 inhibitor, AG490 was identified through high-
throughput screening in 1995.9 It is a member of the tyrphostin
family of tyrosine kinase inhibitors, and it efficiently blocks the
growth of leukemic cells by inducing cellular apoptosis. However,
AG490 lacks specificity for JAK2 so that the induced apoptosis in
cancer cells treated with AG490 may not result from JAK2 inhibi-
tion. Additionally, subsequent reports revealed that AG490 inhibits
JAK2/STAT3 signaling in vitro, but only at high concentrations
Figure 2. Structure of hit compound selected from virtual screening.
more potent molecules than lead compound 5k, we tried to syn-
thesize diverse modified compounds. The quinazoline derivatives
were synthesized as illustrated in Scheme 1. The commercially
available 2-amino-4,5-dimethoxybenzoic acid 1 was reacted with
formamidine acetate to give 6,7-dimethoxy-3H-quinazolin-4-one
2 in 89% yield. Treatment of the quinazolinone 2 with thionyl chlo-
ride afforded the corresponding imine chloride 3 in 92% yield. The
imine chloride 3 was then reacted with various amines in isopro-
panol or DMF to afford the desired quinazolines 5 in moderate
yields.
(IC50 = 50–100 lM), which translates into limited activity in
vivo.9,10 This has led to a search for a selective JAK2 inhibitor but
there is no effective inhibitor of JAK2 that is currently available
for clinical use.
Recently, several X-ray structures of JAK2 in complexes with a
number of small molecule inhibitors, including quinoxaline,11 aza-
indole,12 and 2-aminopyrazolo[1,5-a]pyrimidine13 as shown in
Figure 1, have been determined.
2.2. Identification of hit by virtual screening
Virtual screening has been considered an important in silico
technique for finding novel compounds and has led to many suc-
cesses in drug development. Structure-based virtual screening
methods are emerging as reliable and complementary approaches
through the determination of the X-ray crystal structures of
ligand–protein complexes.
For a virtual screening and a docking study, we used Surflex-
Dock in Sybyl version 8.1.1 (Tripos Associates) operating under
Red Hat Linux 4.0 on an IBM computer (Intel Pentium 4, 2.8 GHz
CPU, 1 GB memory). The structures of the inhibitors were drawn
in the Sybyl package with standard bond lengths and angles and
minimized using the conjugate gradient method. The Gasteiger–
Huckel charge, with a distance-dependent dielectric function,
was applied for the minimization process. We chose the 3E63
(PDB code #) structure from the Protein Data Bank, and the struc-
ture was polished as follows. The JAK2 protein was analyzed by
Structure-Preparation Tool in Sybyl. The angles of seven side
chains such as Glu846, Asn859, Gln872, Arg947, Gln1003,
Lys1053 and Asp1092 were corrected. After adding hydrogen, the
charges loaded on the protein and ligand with AMBER FF99 and
Gasteiger–Marsili, respectively. The side chain amides were also
fixed and two bumping amino acids such as Ser904 and Tyr972
were adjusted.
We performed flexible docking with SurflexDock on a relatively
small library containing 2000 drug-like molecules extracted from
commercial and in-house databases. SurflexDock performs the
docking through automatic way to the ligand binding pocket using
a protomol-based method. The protomol is a critical and essential
factor of the docking algorithm and is a computational demonstra-
tion of proposed ligands that interact with the binding site.
SurflexDock’s scoring function uses hydrophobic, polar, repulsive,
entropic, and solvation terms. After extracting the aminoindazole
ligand which was in active site of 2E63, the protomol was produced.
In this study, we describe the virtual screening of molecules to
recognize potential hits for JAK2 inhibitors using a SurflexDock
program. Novel potential leads were selected as identified mole-
cules using consensus docking score, and these compounds were
evaluated for JAK2 inhibitory activities. Through virtual screening,
the lead compound (6,7-dimethoxy-quinazolin-4-yl)-morpholin-
4-yl-amine (5k) was found as shown in Figure 2. Next, we tried
to optimize the lead compound (5k) through chemical modifica-
tion of quinazoline by introducing various amines on the C-1 posi-
tion. The synthesized quinazoline analogs were utilized to evaluate
their effects on JAK2 inhibition and cytotoxicity against three can-
cer cell lines. In addition, we investigated the effects of the potent
compound 5p on the downstream JAK2/STAT3 signaling pathway
and the mechanism of compound 5p to induce apoptosis by flow
cytometry.
2. Results and discussion
2.1. Chemistry
In order to improve biological activities or diminish toxicity, or
increase absorption lead optimization is operated. For finding out
N
H3CO
OH
N
N
N
O
N
HO
HO
N
H3CO
N
N
NEt2
H
Cl
OCH3
CN
H2N
N
N
O
N
H
N
O S
HN
O
CH3
2,8-diarylquinoxaline
2-Aminipyrazolo[1,5-a]pyrimidine
Cyclic azaindole
AG 490
Figure 1. Structures of the inhibitors interacting with ATP-binding site of JAK2.

970
S. H. Yang et al. / Bioorg. Med. Chem. 19 (2011) 968–977
NH2
H3CO
H3CO
N
O
H3CO
H3CO
NH2
OH
NH2
SOCl2, DMF
reflux
, MeOH
NH
reflux
O
1
2
H3CO
H3CO
N
H3CO
H3CO
N
i-PrOH, reflux
N.HCl
+
H2N
R
N
or DMF, 80°C
HN
Cl
R
3
5
Scheme 1. Synthesis of quinazolines 5.
The chemical structures were drawn in Sybyl and saved as mol2
file into database. The SurflexDock was carried out using the above
database as ligand source with maximum number of poses per
ligand with 20.
Docking calculations of the selected compounds led to the
identification of plausible inhibitors producing 12 hits with high
scores. The hit compounds were primarily used to test JAK2
inhibition activity. Among the 12 compounds, compound 5k, a
quinazoline derivative, displayed a good docking score (6.25)
2.3. JAK2 inhibitory activity and cytotoxicity of quinazoline
compounds
All synthesized quinazoline compounds were tested for JAK2
inhibition and in vitro cytotoxicity using HTScanÒ JAK2 Kinase Assay
Kit. AG490, a well-known JAK2 inhibitor, was used as a positive
control. The JAK2 inhibitory effect was assessed after treatment
with 20 lM of each compound. Most compounds did not have
potent JAK2 inhibitory activity, but compounds 5k and 5p showed
similar JAK2 inhibition (44.42% and 43.03%, respectively) to AG490
(Table 1). Also, the inhibition of JAK2 activity was detected at
and the most active potency (44.42% inhibition at 20
lM).
Thus, the virtual screening approach yielded novel and
a
potent hit compound JAK2 inhibitor from a limited selection of
compounds.
10, 20, and 50 lM of compound 5p in a dose-dependent manner
(Figure 4). In many cases, the enzyme activity does not correlate
well with cytotoxicity and this can be explained by different solu-
bility or poor membrane permeability.
Cytotoxicity was determined against three different human
tumor cell lines: SK-OV-3 (human ovarian tumor), HeLa (human
cervix tumor), and HT-29 (human colon tumor). The IC50 values
are displayed in Table 1. While no significant cytotoxic effect was
detected with AG490, compound 5p showed potent cytotoxicity at
The quinazoline template occupied a hydrophobic region such
as Leu856, Leu855, Ala880, Leu932, and Gly935. The highly con-
served hydrogen bond was created by 6-OMe of the ring template,
which binds to the NH of Arg980. Moreover, hydrophobic interac-
tions were identified between the morpholine moiety and the
hydrophobic region formed by Leu855, Ala880, Tyr931, Val911,
and Met929 as illustrated in Figure 3.
Figure 3. (A) Ribbon diagram illustrating the crystal structure of JAK2 with compound 5k bound to the active site; (B) graphical representation of the binding mode of
compound 5k to the ATP-binding site of JAK2. For clarity, only the Arg980 residue is displayed, and hydrogen bond interactions are represented by yellow dashed lines.

S. H. Yang et al. / Bioorg. Med. Chem. 19 (2011) 968–977
971
Table 1
JAK2 inhibitory effect and cytotoxicity of quinazolines 5a–r
No.
Compound
R
JAK2 inhibitory effect (%)
IC50 (lM)
SK-OV-3
37.476
HeLa
HT-29
OH
1
2
5a
5b
23.799
43.946
59.718
92.994
33.126
74.595
>100
51.472
>100
OH
OCH3
3
4
5
6
7
5c
5d
5e
5f
29.418
>100
à
à
à
—
>100
—
—
OCH3
CH3
10.589
18.426
>100
>100
>100
91.901
99.547
57.865
CH3
Cl
Cl
§
§
§
§
5g
—
—
—
—
§
§
§
§
8
5h
5i
—
—
—
—
CN
CN
§
§
§
§
9
—
—
—
—
CH3
à
10
5j
29.345
44.415
>100
>100
—
>100
>100
Br
11
12
5k
5l
O
>100
N
N
§
§
§
§
N
—
—
—
—
13
14
5m
5n
29.457
38.036
>100
>100
>100
>100
>100
>100
N
O
OCH3
OCH3
OCH3
15
16
5o
5p
25.297
43.025
>100
>100
>100
16.729
8.019
7.650
Cl
N
S
17
5q
14.444
>100
>100
>100
>100
N
18
5r
14.475
55.081
>100
>100
>100
à
Reference
AG490
—
78.523
Not HCl salt.
No significant effect.
Not tested due to solubility problem of the compound.
à
§

972
S. H. Yang et al. / Bioorg. Med. Chem. 19 (2011) 968–977
Figure 4. The JAK2 inhibitory effect (%) of 5p and AG490 at varying concentrations;
JAK2 kinase activities were measured using Assay Kit (Cell Signaling Technology)
system. Data are the means SD of three independent experiments.
low micromolar concentration. Interestingly, a positive correlation
between JAK2 inhibition and cytotoxicity was detected in 5p treat-
ment but not in 5k treatment. So, further in vitro screening was
carried out using 5p.
As compound 5k was identified as the lead compound through
the initial screening for JAK2 inhibition by JAK2 Kit (44.42% of JAK2
inhibition), we investigated to see whether 5k efficiently inhibits
phosphorylation of Jak2 protein in mouse embryonic stem cells
(ESC) (Figure 5). As a result, the level of phosphorylated JAK2
(p-JAK2) was increased with IL-6 pretreatment, and compound
5k markedly inhibited the level of p-JAK2 while AG490 reduced
the level of p-JAK2 to the basal level.
Figure 6. (A) Expression level of JAK2 and p-JAK2 after treatment with 5p and
AG490; HT-29 cells were treated with IL-6 for 0.5 h, then with 5p (10, 20 M) and
AG490 (50, 100 M). The expressions were examined using Western blotting.
l
l
GAPDH was used as an internal control. A representative immunoblot of three
separate experiments is shown. (B) The inhibitory effect on luciferase activity of
STAT3 after treatment with 5p and AG490 at varying concentrations; HT-29 cells
were transfected with the STAT3-dependent luciferase reporter construct pSTAT3-
Luc. After treatment with 5p or AG490 for 1 h in transfected HT-29 cells, the
luciferase assay was performed. Data are the means SD of three independent
experiments. **p <0.01, ***p <0.001 vs control group; significance was determined
using Student’s t-test.
2.4. Effects of quinazoline 5p on inhibition of phosphorylation
of JAK2 and STAT3
To examine whether JAK2 activity is down-regulated by com-
pound 5p, the phosphorylation of JAK2 was examined by Western
blot analysis in HT-29 cells. As shown in Figure 6A, 5p treatment
significantly attenuated p-JAK2 level dose-dependently (10,
change with 5p treatment, indicating that 5p inhibits JAK2
activation.
20 lM). On the other hand, AG490 treatment at the same concen-
JAKs have been reported to be important for mediating the con-
stitutive STAT activation that has been detected in many human
cancer cells.14 STAT3 is one of the signaling pathways induced by
activated JAK2, and it plays an essential role in mitogenic and cell
tration did not inhibit IL-6 induced phosphorylation of JAK2
protein (data not shown). With fivefold more AG490 (50,
100
lM), a slight decrease in p-JAK2 protein expression was
detected. Moreover, the total protein expression of JAK2 did not
Figure 5. Expression level of p-JAK2 after treatment with 5k and AG490 in mouse ESC; mouse ESC was treated with IL-6 for 1 h, then with 5k and AG490 (20
expressions were examined using Western blotting. A representative immunoblot is shown. Relative optical density (% of control) is displayed next. Data are the means SD
**
lM). The
*
of three independent experiments. p <0.1, p <0.01 vs control group; significance was determined using a Student’s t-test.

S. H. Yang et al. / Bioorg. Med. Chem. 19 (2011) 968–977
973
Figure 7. The expression level of c-myc and Bcl-XL with 5p (20
lM) in HT-29 cells; cells were treated with 5p (20
l
M) for the indicated time, then total RNA and lysates were
prepared for RT-PCR (A) and Western blotting (B) of c-myc and Bcl-XL. GAPDH was used as an internal control.
survival.15 To verify STAT3 regulation by compound 5p, we mea-
sured the transcriptional activity of STAT3 by using pSTAT3-Luc
plasmid, which was generated by inserting multidimerized
STAT3-specific binding sites upstream of a luciferase reporter gene.
Cells were transiently transfected with this plasmid and then stim-
ulated with IL-6 (20 ng/ml) either in the presence or absence of
compound 5p. Pretreatment with 5p significantly reduced
STAT3-dependent luciferase activity, but 5p and AG490 have not
shown dose-dependency (Figure 6B).
2.6. Cytotoxicity of quinazoline 5p and AG490 in HT-29 cell
The cytotoxicity of compound 5p and AG490 was determined
by increasing their concentrations in HT-29 cells (Figure 8).
AG490 had no cytotoxicity until it was increased to 25
about 75 M was required to inhibit cell growth by 50%. However,
the cytotoxicity of 5p increased in a dose-dependent manner, and
about 7 M 5p was required for 50% inhibition of cell growth.
lM, and
l
l
Additionally, our results agreed with the previous report that
AG490 inhibited JAK2/STAT3 signaling in vitro only at high
concentrations.
2.7. Cell cycle dynamics of quinazoline 5p by flow cytometry
As compound 5p had a significant cytotoxic effect on human
colon adenocarcinoma HT-29 cells, we further examined the pre-
cise effect of 5p on the cell cycle of HT-29 cells by flow cytometry
2.5. Effects of quinazoline 5p on the expression of c-myc and
Bcl-XL
(Figure 9A). When cells were treated with 5p (5, 10 and 20 lM)
for 24 h, the cell cycle arrest at G2/M phase was significantly
To determine whether this inhibitory effect of 5p on c-myc and
Bcl-XL expression is associated with the induction of apoptosis in
HT-29 cells, we quantified the mRNA and protein levels of c-myc
and Bcl-XL by RT-PCR and Western blotting, respectively.
increased, accompanying a decrease in G1 phase when compared
The increase in activated STAT3 induces target proteins such as
c-myc or Bcl-XL, which are anti-apoptotic.16 Since these target pro-
teins are involved in the control of apoptosis and cell cycle progres-
sion, STAT3 is pathogenetically important by regulating the
expression of anti-apoptotic proteins.16,17 Based on the data pre-
sented here (Figure 7A and B), the expressions of c-myc and
Bcl-XL were affected by treatment with 20 lM 5p, and both levels
were decreased in a time-dependent manner. These results suggest
that 5p-mediated suppression of the JAK2/STAT3 pathway was
accompanied by down-regulation of c-myc and Bcl-XL, finally
resulting in apoptosis. The inhibitory effect of 5p on the anti-apop-
totic proteins caused cytotoxic effect on cells that corresponds to
the result of cytotoxicity of 5p as explained below.
Figure 9. Effects of 5p and AG490 on cell cycle arrest in HT-29 cells; cells were
treated with 5p (5, 10 and 20 lM) for 24 h (A) or AG490 (10, 20 and 30 lM) for 48 h
(B), and then harvested, and fixed in 70% ethanol. After staining with propidium
iodide, DNA contents were analyzed by flow cytometry. Data are the means SD of
**
p
***
p
Figure 8. Cytotoxicity (%) of 5p and AG490 in the HT-29 cell line at varying
three independent experiments.
<0.01,
<0.001 vs non-treated control
concentrations.
group; significance was determined by the Student’s t-test.

974
S. H. Yang et al. / Bioorg. Med. Chem. 19 (2011) 968–977
with the untreated control cells. Moreover, treatment with AG490
(10, 20 and 30 M) for 48 h also induced G2/M phase arrest (Figure
(18 mL). The resulting residue was filtered, washed with water,
and dried to give compound 2 as a dark brown solid (1.83 g,
89%). 1H NMR (300 MHz, DMSO-d6) d: 12.11 (1H, s, NH), 8.0 (1H,
s, –N@CH), 7.44 (1H, s, aromatic-H), 7.14 (1H, s, aromatic-H),
3.90 (3H, s, –OCH3), 3.87 (3H, s, –OCH3).
l
9B). These results explained the positive correlation between JAK2
inhibition and cytotoxicity of compound 5p in HT-29 cells in that
5p inhibited JAK2/STAT3 signaling, reduced consequent anti-apop-
totic protein levels, and finally caused apoptosis by inducing G2/M
arrest.
4.1.2. General procedure for the synthesis of quinazolines 5
A mixture of compound 2 (1.82 g, 8.8 mmol) and DMF (30
drops) in SOCl2 (50 mL) was refluxed for 7 h. The excess SOCl2
was removed by vacuum distillation. The residue was stirred with
diethyl ether, filtered, washed, and dried to give compound 3 as a
beige solid (1.83 g, 92%). A mixture of compound 3 and various
amines in 2-propanol or DMF was refluxed. Upon completion of
the reaction, it was cooled in an ice bath. The residue was filtered,
washed, and dried to isolate the product.
3. Conclusion
A structure-based virtual screening approach was used to iden-
tify the compounds that inhibit JAK2 activity. A docking study was
applied to find a lead compound in a time- and cost-efficient man-
ner. This approach employed the SurflexDock program in the Sybyl
package with an in-house and commercial chemical database. We
identified a C-1 morpholine-substituted quinazoline compound as
a novel strong inhibitor of JAK2 through virtual screening of 2000
compounds, and 12 compounds were selected for actual testing in
a biochemical assay. Among these molecules, one compound sup-
pressed the enzymatic activity of JAK2 with 44% inhibition at
4.1.3. 3-(6,7-Dimethoxyquinazolin-4-ylamino)phenol hydro-
chloride (5a)
A mixture of compound 3 (120 mg, 0.53 mmol) and 3-amino-
phenol (87 mg, 0.80 mmol) in 2-propanol afforded product 5a as
a light green solid (109 mg, 62%). 1H NMR (300 MHz, DMSO-d6)
d: 11.14 (1H, s, OH), 9.76 (1H, s, NH), 8.83 (1H, s, –N@CH), 8.21
(1H, s, aromatic-H), 7.32 (1H, s, aromatic-H), 7.26 (1H, d,
J = 8.1 Hz, aromatic-H), 7.11–7.07 (2H, m, aromatic-H), 6.75 (1H,
d, J = 8.1 Hz, aromatic-H), 4.02 (6H, s, –2OCH3). ESI-MS: m/z (332,
MHꢀ).
20 lM, and was used the starting compound as a JAK2 inhibitor.
We selected the hit compound 5k as the lead for the design of
new JAK2 inhibitors for further synthetic processes. Substituents
at the C-1 position were synthesized for optimization of JAK2
inhibitory activity. In the docking model, the quinazoline template
of 5k occupied hydrophobic regions such as Leu856, Leu855,
Ala880, Leu932, and Gly935, and the highly conserved hydrogen
bond was created by 6-OMe of the ring template, which binds to
the NH of Arg980. Moreover, hydrophobic interactions were iden-
tified between the morpholine moiety and the hydrophobic region
formed by Leu855, Ala880, Tyr931, Val911, and Met929.
Among the newly synthesized derivatives, compound 5p
showed a positive correlation between JAK2 inhibition and cyto-
toxicity. 5p treatment in HT-29 cells strongly inhibited JAK2 acti-
vation and the subsequent STAT3 phosphorylation, reduced the
anti-apoptotic protein levels, and finally induced apoptosis, proba-
bly due to G2/M arrest. We have therefore demonstrated that com-
pound 5p is a potent and selective inhibitor of the JAK2/STAT3
pathway. Compound 5p was identified as a candidate for further
exploration of the hypothesis that modulation of JAK2/STAT3 sig-
naling pathway could be beneficial in cancer treatment.
4.1.4. 4-(6,7-Dimethoxyquinazolin-4-ylamino)phenol hydro-
chloride (5b)
Compound 3 (200 mg, 0.89 mmol) and 4-aminophenol (129 mg,
1.16 mmol) in 2-propanol gave product 5b as a green solid
(184 mg, 62%). 1H NMR (300 MHz, DMSO-d6) d: 11.02 (1H, s, NH),
9.70 (1H, s, OH), 8.75 (1H, s, –N@CH), 8.13 (1H, s, aromatic-H),
7.40 (2H, d, J = 8.7 Hz, aromatic-H), 7.26 (1H, s, aromatic-H), 6.87
(2H, d, J = 6.6 Hz, aromatic-H), 3.99 (6H, s, –2OCH3). ESI-MS: m/z
(298, MH+).
4.1.5. (6,7-Dimethoxyquinazolin-4-yl)-(2-methoxyphenyl)-
amine hydrochloride (5c)
Compound 3 (80 mg, 0.35 mmol) and 2-methoxyaniline (88 mg,
0.71 mmol) in 2-propanol gave product 5c as a dark brown solid
(64 mg, 53%). 1H NMR (300 MHz, DMSO-d6) d: 11.06 (1H, s, NH),
8.71 (1H, s, –N@CH), 8.16 (1H, s, aromatic-H), 7.44–7.38 (2H, m,
aromatic-H), 7.31 (1H, s, aromatic-H), 7.22 (1H, d, J = 7.8 Hz,
aromatic-H), 7.07 (1H, t, J = 7.5 Hz, aromatic-H), 4.00 (3H, s,
–OCH3 of quinazoline), 3.98 (3H, s, –OCH3 of quinazoline), 3.79
(3H, s, –OCH3 of anisidine). ESI-MS: m/z (348, MH+).
4. Experimental
4.1. General remarks
Chemicals were provided by Aldrich Chemical Co. or Tokyo
Chemical Industry Co. and utilized without further purification.
Solvents were distilled prior to use. Nuclear magnetic resonance
(NMR) spectra were taken on Bruker AMX-R300 for 1H NMR and
were calibrated with tetramethylsilane. The NMR data are dis-
played as follows: chemical shifts (d) are recorded in ppm, coupling
constants (J) in hertz (Hz), integrity in the number of protons, and
multiplicity in s (singlet), d (doublet), t (triplet), and m (multiplet).
Mass spectra were obtained on Shimadzu LCMS-2010EV utilizing
the electron-spray ionization (ESI) method. Thin-layer chromatog-
raphy (TLC) was carried out using plates coated with Silica Gel 60
F254 purchased from Merck. Column chromatography was per-
formed with Merck Silica Gel 60 (70–230 mesh).
4.1.6. (6,7-Dimethoxyquinazolin-4-yl)-(4-methoxyphenyl)-
amine hydrochloride (5d)
Compound
3 (120 mg, 0.53 mmol) and 4-methoxyaniline
(131 mg, 1.06 mmol) in 2-propanol gave 5d as a dark green solid
(115 mg, 63%). 1H NMR (300 MHz, DMSO-d6) d: 11.22 (1H, s, NH),
8.78 (1H, s, –N@CH), 8.21 (1H, s, aromatic-H), 7.55 (2H, d,
J = 6.9 Hz, aromatic-H), 7.30 (1H, s, aromatic-H), 7.06 (2H, d,
J = 6.9 Hz, aromatic-H), 4.00 (6H, s, –2OCH3 of quinazoline), 3.81
(3H, s, –OCH3 of anisidine). ESI-MS: m/z (346, MHꢀ).
4.1.7. (6,7-Dimethoxyquinazolin-4-yl)-o-tolylamine hydro-
chloride (5e)
4.1.1. 6,7-Dimethoxy-3H-quinazolin-4-one (2)
Compound 3 (90 mg, 0.40 mmol) and 2-methylaniline (52 mg,
0.48 mmol) in 2-propanol gave product 5e as a white solid
(70 mg, 53%). 1H NMR (300 MHz, DMSO-d6) d: 11.28 (1H, s, NH),
8.71 (1H, s, –N@CH), 8.21 (1H, s, aromatic-H), 7.42–7.39 (1H, m,
aromatic-H), 7.36–7.33 (3H, m, aromatic-H), 7.31 (1H, s,
A solution of 2-amino-4,5-dimethoxybenzoic acid 1 (2.02 g,
10 mmol) and formamidine acetate (2.10 g, 20 mmol) in 2-
methoxyethanol (50 mL) was refluxed overnight. After evaporation
of the solvent, the residue was stirred after adding 10% NH4OH

S. H. Yang et al. / Bioorg. Med. Chem. 19 (2011) 968–977
975
aromatic-H), 4.00 (3H, s, –OCH3), 3.99 (3H, s, –OCH3), 2.21 (3H, s, –
low solid (55 mg, 41%). 1H NMR (300 MHz, DMSO-d6) d: 10.95
(1H, s, NH), 8.55 (1H, s, –N@CH), 8.04 (1H, s, aromatic-H), 7.32
(1H, s, aromatic-H), 3.97 (6H, s, –2OCH3), 3.96–3.87 (6H, m, N–
CH2–CH2–), 2.79 (3H, s, –CH3), 2.78–2.73 (2H, m, N–CH2–CH2–).
ESI-MS: m/z (304, MH+).
CH3). ESI-MS: m/z (330, MHꢀ).
4.1.8. (6,7-Dimethoxyquinazolin-4-yl)-p-tolylamine
hydrochloride (5f)
Compound 3 (90 mg, 0.40 mmol) and 4-methylaniline (52 mg,
0.48 mmol) in 2-propanol gave product 5f as a light green solid
(80 mg, 60%). 1H NMR (300 MHz, DMSO-d6) d: 11.23 (1H, s, NH),
8.79 (1H, s, –N@CH), 8.22 (1H, s, aromatic-H), 7.54 (2H, d,
J = 8.4 Hz, aromatic-H), 7.31 (1H, s, aromatic-H), 7.30 (2H, d,
J = 8.1 Hz, aromatic-H), 4.00 (3H, s, –OCH3), 3.99 (3H, s, –OCH3),
2.36 (3H, s, –CH3). ESI-MS: m/z (330, MHꢀ).
4.1.15. (6,7-Dimethoxyquinazolin-4-yl)-(2-morpholin-4-
ylethyl)amine hydrochloride (5m)
Compound 3 (100 mg, 0.44 mmol) and 4-(2-aminoethyl)-mor-
pholine (88 mg, 0.67 mmol) in 2-propanol gave product 5m as a
beige solid (114 mg, 73%). 1H NMR (300 MHz, CDCl3) d: 8.56 (1H,
s, –N@CH), 7.21 (1H, s, aromatic-H), 6.92 (1H, s, aromatic-H),
4.02 (3H, s, –OCH3), 4.01 (3H, s, –OCH3), 3.76 (4H, t, J = 4.5 Hz,
O–CH2–CH2–), 3.71 (2H, t, J = 4.7 Hz, aliphatic-H), 2.76 (2H, t,
J = 5.9 Hz, aliphatic-H), 2.59 (4H, t, J = 4.4 Hz, N–CH2–CH2–). ESI-
MS: m/z (319, MH+).
4.1.9. (2,3-Dichlorophenyl)-(6,7-dimethoxyquinazolin-4-
yl)amine hydrochloride (5g)
Compound
3 (80 mg, 0.35 mmol) and 2,3-dichloroaniline
(115 mg, 0.71 mmol) in 2-propanol gave product 5g as a white so-
lid (77 mg, 57%). 1H NMR (300 MHz, DMSO-d6) d: 11.61 (1H, s, NH),
8.82 (1H, s, aromatic-H), 8.24 (1H, s, aromatic-H), 7.75 (1H, dd,
J = 6.6, 2.4 Hz, aromatic-H), 7.71 (1H, s, aromatic-H), 7.56 (1H, dd,
J = 8.4, 2.4 Hz, aromatic-H), 7.35 (1H, s, aromatic-H), 4.01 (3H, s,
–OCH3), 4.00 (3H, s, –OCH3). ESI-MS: m/z (386, MH+).
4.1.16. (6,7-Dimethoxyquinazolin-4-yl)-(2-methoxyethyl)amine
hydrochloride (5n)
Compound 3 (80 mg, 0.35 mmol) and 2-methoxyethylamine
(54 mg, 0.71 mmol) in 2-propanol gave product 5n as a brown so-
lid (29 mg, 28%) after re-crystallization with 2-propanol. 1H NMR
(300 MHz, DMSO-d6) d: 9.73 (1H, s, NH), 8.75 (1H, s, –N@CH),
7.95 (1H, s, aromatic-H), 7.21 (1H, s, aromatic-H), 3.95 (3H, s, –
OCH3), 3.94 (3H, s, –OCH3), 3.85 (2H, dd, J = 10.9, 5.2 Hz, ali-
phatic-H), 3.62 (2H, t, J = 5.4 Hz, aliphatic-H), 3.29 (3H, s, –CH3).
ESI-MS: m/z (264, MH+).
4.1.10. 4-(6,7-Dimethoxyquinazolin-4-ylamino)benzonitrile
hydrochloride (5h)
Compound 3 (80 mg, 0.35 mmol) and 4-cyanoaniline (86 mg,
0.71 mmol) in 2-propanol gave product 5h as a white solid
(79 mg, 66%). 1H NMR (300 MHz, DMSO-d6) d: 11.21 (1H, s, NH),
8.90 (1H, s, –N@CH), 8.23 (1H, s, aromatic-H), 8.02 (2H, d,
J = 8.4 Hz, aromatic-H), 7.95 (2H, d, J = 8.7 Hz, aromatic-H), 7.32
(1H, s, aromatic-H), 4.03 (3H, s, –OCH3), 4.01 (3H, s, –OCH3). ESI-
MS: m/z (341, MHꢀ).
4.1.17. [2-(3,4-Dimethoxyphenyl)ethyl]-(6,7-
dimethoxyquinazolin-4-yl)-amine Hydrochloride (5o)
Compound 3 (100 mg, 0.44 mmol) and 3,4-dimethoxypheneth-
ylamine (125 mg, 0.67 mmol) in 2-propanol gave product 5o as a
brown solid (115 mg, 64%) after re-crystallization with 2-propanol.
1H NMR (300 MHz, DMSO-d6) d: 8.36 (1H, s, –N@CH), 7.59 (1H, s,
aromatic-H), 7.08 (1H, s, aromatic-H), 6.76 (3H, t, J = 7.9 Hz, aro-
matic-H), 3.89 (3H, s, –OCH3), 3.87 (3H, s, –OCH3), 3.74 (3H, s, –
OCH3), 3.72 (3H, s, –OCH3), 2.94 (2H, t, J = 7.5 Hz, aliphatic-H),
2.74 (2H, t, J = 7.1 Hz, aliphatic-H). ESI-MS: m/z (406, MH+).
4.1.11. 5-(6,7-Dimethoxyquinazolin-4-ylamino)-2-
methylbenzonitrile hydrochloride (5i)
Compound 3 (80 mg, 0.35 mmol) and 3-cyano-4-methylaniline
(97 mg, 0.71 mmol) in 2-propanol gave product 5i as a white solid
(86 mg, 69%). 1H NMR (300 MHz, DMSO-d6) d: 11.28 (1H, s, NH),
8.88 (1H, s, –N@CH), 8.21 (1H, s, aromatic-H), 8.17 (1H, s, aro-
matic-H), 7.92 (1H, d, J = 8.4 Hz, aromatic-H), 7.59 (1H, d,
J = 8.4 Hz, aromatic-H), 7.30 (1H, s, aromatic-H), 4.02 (3H, s, –
OCH3), 4.01 (3H, s, –OCH3), 2.53 (3H, s, –CH3). ESI-MS: m/z (357,
MH+).
4.1.18. (6,7-Dimethoxyquinazolin-4-yl)naphthalen-1-ylamine
hydrochloride (5p)
Compound 3 (80 mg, 0.35 mmol) and 1-aminonaphthalene
(102 mg, 0.71 mmol) in 2-propanol gave product 5p as a purple so-
lid (92 mg, 71%). 1H NMR (300 MHz, DMSO-d6) d: 11.62 (1H, s, NH),
8.66 (1H, s, –N@CH), 8.30 (1H, s, aromatic-H), 8.06 (2H, t, J = 6.9 Hz,
aromatic-H), 7.89 (1H, d, J = 7.8 Hz, aromatic-H), 7.69–7.64 (2H, m,
aromatic-H), 7.62–7.53 (2H, m, aromatic-H), 7.31 (1H, s, –OCH3),
4.06 (6H, s, –2OCH3). ESI-MS: m/z (368, MH+).
4.1.12. (4-Bromophenyl)-(6,7-dimethoxyquinazolin-4-yl)amine
hydrochloride (5j)
Compound 3 (80 mg, 0.35 mmol) and 4-bromoaniline (48 mg,
0.43 mmol) in DMF afforded product 5j as a white solid (70 mg,
57%). 1H NMR (300 MHz, DMSO-d6) d: 11.19 (1H, s, NH), 8.83
(1H, s, –N@CH), 8.21 (1H, s, aromatic-H), 7.69 (4H, s, aromatic-
H), 7.30 (1H, s, aromatic-H), 4.01 (3H, s, –OCH3), 4.00 (3H, s, –
OCH3). ESI-MS: m/z (394, MHꢀ).
4.1.19. (4-Chlorobenzothiazol-2-yl)-(6,7-dimethoxyquinazolin-
4-yl)amine hydrochloride (5q)
Compound 3 (80 mg, 0.35 mmol) and 2-amino-4-chlorobenzo-
thiazole (102 mg, 0.71 mmol) in 2-propanol gave product 5q as a
green solid (16 mg, 9%) after re-crystallization with 2-propanol.
1H NMR (300 MHz, DMSO-d6) d: 9.08 (1H, s, –N@CH), 8.35 (1H, s,
aromatic-H), 8.03–7.98 (1H, m, aromatic-H), 7.57 (1H, d,
J = 7.8 Hz, aromatic-H), 7.40 (1H, s, aromatic-H), 7.31–7.28 (1H,
m, aromatic-H), 4.00 (6H, s, –2OCH3). ESI-MS: m/z (409, MH+).
4.1.13. (6,7-Dimethoxyquinazolin-4-yl)morpholin-4-ylamine
hydrochloride (5k)
Compound 3 (100 mg, 0.44 mmol) and 1-aminomorpholine
(92 mg, 0.89 mmol) in 2-propanol gave product 5k as a yellow so-
lid (42 mg, 29%). 1H NMR (300 MHz, DMSO-d6) d: 8.41 (1H, s, –
N@CH), 7.93 (1H, s, aromatic-H), 7.20 (1H, s, aromatic-H), 3.93
(6H, s, –2OCH3), 3.79–3.76 (2H, m, N–CH2–CH2–), 3.70–3.66 (2H,
m, N–CH–CH2–), 2.99 (2H, m, O–CH2–CH2–), 2.90 (2H, m, O–
CH2–CH2–). ESI-MS: m/z (327, MH+).
4.1.20. (6,7-Dimethoxyquinazolin-4-yl)pyridin-2-
ylmethylamine (5r)
Compound 3 (100 mg, 0.44 mmol) and 2-aminomethylpyridine
(243 mg, 2.22 mmol) in dioxane gave product 5r as a beige solid
(55 mg, 42%) through column purification. 1H NMR (300 MHz,
DMSO-d6) d: 8.63 (1H, d, J = 4.6 Hz, –N@CH–CH2), 8.59 (1H, s, –
N@CH–N), 7.72 (1H, t, J = 7.7 Hz, aromatic-H), 7.40 (1H, d,
4.1.14. (6,7-Dimethoxyquinazolin-4-yl)-(4-methylpiperazin-1-
yl)amine (5l)
Compound 3 (100 mg, 0.44 mmol) and 1-amino-4-mehtylpiper-
azine (79 mg, 0.67 mmol) in 2-propanol gave product 5l as a yel-

976
S. H. Yang et al. / Bioorg. Med. Chem. 19 (2011) 968–977
J = 7.8 Hz, aromatic-H), 7.22 (1H, s, aromatic-H), 7.18 (1H, m, aro-
matic-H), 7.07 (1H, s, aromatic-H), 5.30 (3H, s, aliphatic-H and
NH), 4.02 (6H, s, –2OCH3). ESI-MS: m/z (297, MH+).
then sequentially incubated with the primary antibody overnight
at 4 °C and secondary antibodies for 1 h at room temperature. Blots
were processed for visualization using an enhanced chemilumines-
cence system (Pierce Biotechnology Inc., Rockford, IL, USA) and ex-
posed to Kodak XAR-5 film (Rochester, NY, USA) to obtain the
fluorographic images.
4.2. JAK2 activity assay
For determination of in vitro Jak2 kinase activity, we used
HTScanÒ Jak2 Kinase Assay Kit (Cell Signaling Technology), which
contains recombinant wild-type JAK2. Analysis was performed
according to the manufacturer’s protocol. Briefly, the kinase reac-
tion was set up in the total volume as follows: 60 mmol/L Hepes
4.6. Luciferase assay
HT-29 cells were transfected with STAT3-dependent luciferase
reporter construct pSTAT3-Luc plasmid together with phRL-TK
using Lipofectamine LTX reagent (Invitrogen, Carlsbad, CA, USA)
according to the manufacturer’s protocols. Luciferase activity in
cell lysates was determined by a dual luciferase reporter assay sys-
tem (Promega, Madison, USA) and normalized to the Renilla lucif-
erase activity of co-transfected phRL-TK. After 24 h, cells were
serum-starved overnight and then treated with the test sample
for 1 h. Cell extracts were prepared and assayed according to
the manufacturer’s protocols (Dual Luciferase Assay System;
Promega).
(pH 7.5), 5 mmol/L MgCl2, 5 mmol/L MnCl2, 3
1.25 mmol/L DTT, 20 mol/L ATP (200 mol/L ATP for competition
experiments), 1.5 mol/L substrate peptide (0.3 and 0.15 mol/L
lmol/L Na3VO3,
l
l
l
l
for substrate competition experiments), and 10 U kinase. Increas-
ing doses of test sample were added to the reaction and incubated
at 25 °C for 1 h. The reaction was stopped by adding 50
50 mmol/L EDTA (pH 8). Each reaction (25 L) was transferred in
triplicate to a 96-well streptavidin-coated plate (Perkin-Elmer Life
Sciences); 75 L distilled water/well was added to each well and
lL/well of
l
l
the reaction was incubated at room temperature for 1 h. After
washing with PBS/T, monoclonal anti-phosphotyrosine antibody
(p-Tyr100; Cell Signaling Technology) was added (1:1000 in PBS/T
with 1% bovine serum albumin) and incubated at room tempera-
ture for 1 h. After repeated washing with PBS/T, anti-mouse IgG,
horseradish peroxidase-linked antibody (Cell Signaling Technol-
ogy) was added (1:500 in PBS/T with 1% bovine serum albumin)
and incubated for 30 min at 37 °C. After subsequent washing,
TMB substrate was added, incubated for 5 min at room tempera-
ture, and stopped with an equal volume of 1 mol/L HCl. Absorbance
was measured using a standard ELISA reader at 450 nm.
4.7. RT-PCR
Total cellular RNA was isolated using Easy BlueÒ Kits (Intron
Biotechnology, Seoul, Korea) according to the manufacturer’s pro-
tocols. From each sample, 1
(RT) using MuLV reverse transcriptase, 1 mM dNTP, and oligo(dT
12–18) 0.5 g/ L. Then PCR analyses were performed on the aliquots
lg of RNA was reverse-transcribed
l
l
of the cDNA preparations to detect c-myc, Bcl-xL and GAPDH (as an
internal standard) gene expression using a thermal cycler (Perkin-
Elmer Cetus, Foster City, CA, USA). The reactions were carried out
in a volume of 25 lL containing 1 U of Taq DNA polymerase,
4.3. Cell culture
0.2 mM dNTP, 10ꢁ reaction buffer, and 100 pmol of 50 and 30 prim-
ers. After initial denaturation for 2 min at 94 °C, 30 amplification
cycles were performed for c-myc (20 s of 94 °C denaturation, 10 s
of 61 °C annealing, and 20 s of 72 °C extension), Bcl-xL (20 s of
94 °C denaturation, 30 s of 53 °C annealing, and 30 s of 72 °C exten-
sion) and GAPDH (60 s of 94 °C denaturation, 60 s of 61 °C anneal-
ing, and 120 s of 72 °C extension). The PCR primers used in this
study are listed below and were purchased from Bioneer (Seoul,
Korea): sense strand c-myc, 50-GAT TCT CTG CTC TCCTCG ACG
GAG-30, anti-sense strand c-myc, 50-GCG CTG CGT AGT TGT GCT
GAT GTG-30, sense strand Bcl-XL, 50-CCC AGA AAG GAT ACA GCT
GG-30, anti-sense strand Bcl-XL, 50-GCG ATC CGA CTC ACC AAT
AC-30, sense strand GAPDH, 50-CCT GTT CGA CAG TCA CCG-30,
anti-sense strand GAPDH, 50-CGA CCA AAT CCG TTG ACT CC-30.
After amplification, portions of the PCRs were electrophoresed on
2% agarose gel and visualized by ethidium bromide staining and
UV irradiation.
SC-OV-3, HeLa, and HT-29 cells were purchased from Korean
Cell Line Bank. These cells were cultured in RPMI-1640 medium
with 2.0 g/L sodium bicarbonate plus 10% fetal bovine serum,
100 U/mL penicillin, and 100 lg/mL of streptomycin. All cell lines
were maintained in a humidified incubator with an atmosphere
of 95% air and 5% CO2 at 37 °C.
4.4. Cytotoxicity assay
Cytotoxicity was assessed by the MTT assay. Briefly, cells were
seeded at 1 ꢁ 105 cells/mL in each well containing 100
lL of
RPMI-1640 medium supplemented with 10% FBS in a 96-well
plate. After 24 h, various concentrations of tested samples were
added. After 48 h, 50 lL of MTT (5 mg/mL stock solution, in PBS)
were added per well and the plates were incubated for an addi-
tional 4 h. The medium was discarded and the formazan blue
formed in the cells was dissolved with 100 lL of DMSO. The optical
density was measured using a standard ELISA reader at 540 nm.
4.8. Flow cytometric detection
HT-29 cells were incubated until reaching 70–80% confluence.
After treatment with or without treated samples, cells were har-
vested and fixed in 70% ethanol for 1 h on ice. After washing with
4.5. Protein extraction and Western blotting
After treatment with tested samples, cells growing on 10-cm
PBS, cells were labeled with propidium iodide (50
lg/mL) in the
dishes were lysed in 300
at pH 7.4, 1 M DTT, 20
phosphate, 10 M NaF, 1% Triton X-100, 1 mM PMSF, 1 mM
Na3VO4, 2 M aprotinin, 100 M leupeptin, 2 M pepstatin, and
0.5 M okadaic acid) on ice. The cell lysate was centrifuged at
14,000g for 15 min, and the supernatant fraction was collected
for immunoblot. Equal amounts of protein (30–40 g/well) were
lL of ice-cold lysis buffer (20 mM Hepes
presence of RNase A (100 g/mL) and incubated at room tempera-
l
l
l
M EGTA, 10% glycerol, 50 M glycero-
l
ture in the dark for 30 min and analyzed using the fluorescence-
activated cell sorting (FACS) cater-plus flow cytometry (Becton–
Dickinson Co., Heidelberg, Germany).
l
l
l
l
l
Acknowledgment
l
loaded and separated by SDS–PAGE (8–12%) and then electroblot-
ted onto hybond-PVDF membranes (GE Healthcare Bio-Sciences
Corp., NJ, USA). Membranes were blocked with 5% skim milk and
This work was supported by Korea Research Foundation Grant
(KRF-2007-314-H00006).

S. H. Yang et al. / Bioorg. Med. Chem. 19 (2011) 968–977
977
D.; Barker, H. F.; Taylor, P. C.; O’Driscoll, A.; Benedetti, F.; Rudolph, C.; Kolb, H.
J.; Hochhaus, A.; Hehlmann, R.; Chase, A.; Cross, N. C. Cancer Res. 2005, 65,
2662; (c) Mercher, T.; Wernig, G.; Moore, S. A.; Levine, R. L.; Gu, T. L.; Frohling,
S.; Cullen, D.; Polakiewicz, R. D.; Bernard, O. A.; Boggon, T. J.; Lee, B. H.;
Gilliland, D. G. Blood 2006, 108, 2770.
References and notes
1. (a) Zhou, Y. J.; Chen, M.; Cusack, N. A.; Kimmel, L. H.; Magnuson, K. S.; Boyd, J.
G.; Lin, W.; Roberts, J. L.; Lengi, A.; Buckley, R. H.; Geahlen, R. L.; Candotti, F.;
Gadina, M.; Changelian, P. S.; O’Shea, J. J. Mol. Cell 2001, 8, 959; (b) Kisseleva, T.;
Bhattacharya, S.; Braunstein, J.; Schindler, C. W. Gene 2002, 285, 1; (c)
Funakoshi-Tago, M.; Pelletier, S.; Matsuda, T.; Parganas, E.; Ihle, J. N. EMBO J.
2006, 25, 4763.
2. (a) Yamaoka, K.; Saharinen, P.; Pesu, M.; Holt, V. E., 3rd; Silvennoinen, O.;
O’Shea, J. J. Genome Biol. 2004, 5, 253; (b) Sayyah, J.; Sayeski, P. P. Curr. Oncol.
Rep. 2009, 11, 117.
9. Meydan, N.; Grunberger, T.; Dadi, H.; Shahar, M.; Arpaia, E.; Lapidot, Z.; Leeder,
J. S.; Freedman, M.; Cohen, A.; Gazit, A.; Levitzki, A.; Roifman, C. M. Nature 1996,
379, 645.
10. (a) Ugo, V.; Marzac, C.; Teyssandier, I.; Larbret, F.; Lecluse, Y.; Debili, N.;
Vainchenker, W.; Casadevall, N. Exp. Hematol. 2004, 32, 179; (b) Faderl, S.;
Harris, D.; Van, Q.; Kantarjian, H. M.; Talpaz, M.; Estrov, Z. Blood 2003, 102, 630.
11. Pissot-Soldermann, C.; Gerspacher, M.; Furet, P.; Gaul, C.; Holzer, P.; McCarthy,
C.; Radimerski, T.; Regnier, C. H.; Baffert, F.; Drueckes, P.; Tavares, G. A.;
Vangrevelinghe, E.; Blasco, F.; Ottaviani, G.; Ossola, F.; Scesa, J.; Reetz, J. Bioorg.
Med. Chem. Lett. 2010, 20, 2609.
3. Ihle, J. N.; Gilliland, D. G. Curr. Opin. Genet. Dev. 2007, 17, 8.
4. Feng, J.; Witthuhn, B. A.; Matsuda, T.; Kohlhuber, F.; Kerr, I. M.; Ihle, J. N. Mol.
Cell. Biol. 1997, 17, 2497.
5. Saharinen, P.; Vihinen, M.; Silvennoinen, O. Mol. Biol. Cell 2003, 14, 1448.
6. Luo, H.; Hanratty, W. P.; Dearolf, C. R. EMBO J. 1995, 14, 1412.
7. (a) Levine, R. L.; Wadleigh, M.; Cools, J.; Ebert, B. L.; Wernig, G.; Huntly, B. J.;
Boggon, T. J.; Wlodarska, I.; Clark, J. J.; Moore, S.; Adelsperger, J.; Koo, S.; Lee, J.
C.; Gabriel, S.; Mercher, T.; D’Andrea, A.; Frohling, S.; Dohner, K.; Marynen, P.;
Vandenberghe, P.; Mesa, R. A.; Tefferi, A.; Griffin, J. D.; Eck, M. J.; Sellers, W. R.;
Meyerson, M.; Golub, T. R.; Lee, S. J.; Gilliland, D. G. Cancer Cell 2005, 7, 387; (b)
James, C.; Ugo, V.; Le Couedic, J. P.; Staerk, J.; Delhommeau, F.; Lacout, C.;
Garcon, L.; Raslova, H.; Berger, R.; Bennaceur-Griscelli, A.; Villeval, J. L.;
Constantinescu, S. N.; Casadevall, N.; Vainchenker, W. Nature 2005, 434, 1144;
(c) Baxter, E. J.; Scott, L. M.; Campbell, P. J.; East, C.; Fourouclas, N.; Swanton, S.;
Vassiliou, G. S.; Bench, A. J.; Boyd, E. M.; Curtin, N.; Scott, M. A.; Erber, W. N.;
Green, A. R. Lancet 2005, 365, 1054; (d) Kralovics, R.; Passamonti, F.; Buser, A.
S.; Teo, S. S.; Tiedt, R.; Passweg, J. R.; Tichelli, A.; Cazzola, M.; Skoda, R. C. N. Engl.
J. Med. 2005, 352, 1779; (e) Zhao, R.; Xing, S.; Li, Z.; Fu, X.; Li, Q.; Krantz, S. B.;
Zhao, Z. J. J. Biol. Chem. 2005, 280, 22788; (f) Levine, R. L.; Belisle, C.; Wadleigh,
M.; Zahrieh, D.; Lee, S.; Chagnon, P.; Gilliland, D. G.; Busque, L. Blood 2006, 107,
4139.
12. (a) Wang, T.; Ledeboer, M. W.; Duffy, J. P.; Salituro, F. G.; Pierce, A. C.; Zuccola,
H. J.; Block, E.; Shlyakter, D.; Hogan, J. K.; Bennani, Y. L. Bioorg. Med. Chem. Lett.
2010, 20, 153; (b) Wang, T.; Duffy, J. P.; Wang, J.; Halas, S.; Salituro, F. G.; Pierce,
A. C.; Zuccola, H. J.; Black, J. R.; Hogan, J. K.; Jepson, S.; Shlyakter, D.; Mahajan,
S.; Gu, Y.; Hoock, T.; Wood, M.; Furey, B. F.; Frantz, J. D.; Dauffenbach, L. M.;
Germann, U. A.; Fan, B.; Namchuk, M.; Bennani, Y. L.; Ledeboer, M. W. J. Med.
Chem. 2009, 52, 7938.
13. Ledeboer, M. W.; Pierce, A. C.; Duffy, J. P.; Gao, H.; Messersmith, D.; Salituro, F.
G.; Nanthakumar, S.; Come, J.; Zuccola, H. J.; Swenson, L.; Shlyakter, D.;
Mahajan, S.; Hoock, T.; Fan, B.; Tsai, W. J.; Kolaczkowski, E.; Carrier, S.; Hogan, J.
K.; Zessis, R.; Pazhanisamy, S.; Bennani, Y. L. Bioorg. Med. Chem. Lett. 2009, 19,
6529.
14. (a) Battle, T. E.; Frank, D. A. Curr. Mol. Med. 2002, 2, 381; (b) Yu, H.; Jove, R. Nat.
Rev. Cancer 2004, 4, 97.
15. Rane, S. G.; Reddy, E. P. Oncogene 2000, 19, 5662.
16. (a) Darnell, J. E., Jr. Science 1997, 277, 1630; (b) Zamo, A.; Chiarle, R.; Piva, R.;
Howes, J.; Fan, Y.; Chilosi, M.; Levy, D. E.; Inghirami, G. Oncogene 2002, 21,
1038.
17. Amin, H. M.; McDonnell, T. J.; Ma, Y.; Lin, Q.; Fujio, Y.; Kunisada, K.; Leventaki,
V.; Das, P.; Rassidakis, G. Z.; Cutler, C.; Medeiros, L. J.; Lai, R. Oncogene 2004, 23,
5426.
8. (a) Lacronique, V.; Boureux, A.; Valle, V. D.; Poirel, H.; Quang, C. T.; Mauchauffe,
M.; Berthou, C.; Lessard, M.; Berger, R.; Ghysdael, J.; Bernard, O. A. Science 1997,
278, 1309; (b) Reiter, A.; Walz, C.; Watmore, A.; Schoch, C.; Blau, I.;
Schlegelberger, B.; Berger, U.; Telford, N.; Aruliah, S.; Yin, J. A.; Vanstraelen,