Difuran-substituted quinoxalines as a novel class of PI3Kα H1047R mutant inhibitors: Synthesis, biological evaluation and structure-activity relationship

Article abstract of DOI:10.1016/j.ejmech.2018.07.061

Phosphatidylinositol 3-kinase α (PI3Kα) is the most frequently mutated kinase in human cancers, making it an attractive therapeutic target for cancer treatment. We identified a structurally novel PI3Kα H1047R mutant inhibitor Hit-01 (EC₅₀ = 76.0 μM) through a high-throughput screening campaign. Chemical optimizations enabled us to discover compound 7b, which strongly inhibited PI3Kα H1047R mutant with an EC₅₀ value of 0.137 μM, over 500-fold more potent than Hit-01. Western blotting analysis suggested that 7b could decrease the phosphorylation level of p-AKT, another proof that 7b inhibited PI3Kα H1047R function. Cell viability assay revealed that 7b inhibited HCT116 cancer cell growth with an IC₅₀ value of 11.23 μM. In addition, 7b was found to arrest cell cycle at G1 phase and induce cell apoptosis via up-regulation of caspase-3, caspase-8 and caspase-9 protein expressions. Collectively, all these data demonstrated that 7b could be a promising lead for the development of structurally novel PI3Kα inhibitors.

Full text of DOI:10.1016/j.ejmech.2018.07.061

Accepted Manuscript
Difuran-substituted quinoxalines as a novel class of PI3Kα H1047R mutant inhibitors:
Synthesis, biological evaluation and structure-activity relationship
Ning Zhang, Zhimei Yu, Xiaohong Yang, Yan Zhou, Qing Tang, Ping Hu, Jia Wang,
Shao-Lin Zhang, Ming-Wei Wang, Yun He
PII:
S0223-5234(18)30626-3
10.1016/j.ejmech.2018.07.061
EJMECH 10596
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 13 May 2018
Revised Date: 25 July 2018
Accepted Date: 25 July 2018
Please cite this article as: N. Zhang, Z. Yu, X. Yang, Y. Zhou, Q. Tang, P. Hu, J. Wang, S.-L. Zhang, M.-
W. Wang, Y. He, Difuran-substituted quinoxalines as a novel class of PI3Kα H1047R mutant inhibitors:
Synthesis, biological evaluation and structure-activity relationship, European Journal of Medicinal
Chemistry (2018), doi: 10.1016/j.ejmech.2018.07.061.
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
Graphical Abstract
EC50 (µM)
IC50 (µM)
HCT116
Improved potency
(Hit-01 to 7b)
Comp.
PI3Kα H1047R
PI3Kα WT
mutant
76.0
NT
> 100
11.2
76.0 / 0.137 > 500 (enzymatic level)
100 / 11.2 > 9 (cellular level)
Hit-01
7b
0.137
0.093

ACCEPTED MANUSCRIPT
Title:
Difuran-substituted quinoxalines as a novel class of PI3Kα H1047R mutant inhibitors:
Synthesis, biological evaluation and structure-activity relationship
Author names and affiliations:
Ning Zhang1, Zhimei Yu1, Xiaohong Yang1, Yan Zhou2, Qing Tang1, Ping Hu1, Jia
Wang2, Shao-Lin Zhang1*, Ming-Wei Wang2*, and Yun He1*
1
School of Pharmaceutical Sciences, Chongqing Key Laboratory of Natural Product
Synthesis and Drug Research, Chongqing University, Chongqing, 401331, P. R.
China
2
The National Center for Drug Screening and the CAS Key Laboratory of Receptor
Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences (CAS),
189 Guo Shou Jing Road, Shanghai, 201203, China
* Corresponding address:
Tel.: +86-23-65678465; E-mail: zhangsl@cqu.edu.cn; mwwang@simm.ac.cn;
yun.he@cqu.edu.cn
Abstract:
Phosphatidylinositol 3-kinase α (PI3Kα) is the most frequently mutated kinase in
human cancers, making it an attractive therapeutic target for cancer treatment. We
identified a structurally novel PI3Kα H1047R mutant inhibitor Hit-01 (EC50 = 76.0
µM) through a high-throughput screening campaign. Chemical optimizations enabled
us to discover compound 7b, which strongly inhibited PI3Kα H1047R mutant with an
EC50 value of 0.137 µM, over 500-fold more potent than Hit-01. Western blotting
analysis suggested that 7b could decrease the phosphorylation level of p-AKT,
another proof that 7b inhibited PI3Kα H1047R function. Cell viability assay revealed
that 7b inhibited HCT116 cancer cell growth with an IC50 value of 11.23 µM. In
addition, 7b was found to arrest cell cycle at G1 phase and induce cell apoptosis via
up-regulation of caspase-3, caspase-8 and caspase-9 protein expressions. Collectively,
all these data demonstrated that 7b could be a promising lead for the development of
structurally novel PI3Kα inhibitors.
Keywords:
Proliferation; Anticancer drug; Phosphatidylinositol 3-kinase inhibitor; Apoptosis;
1

ACCEPTED MANUSCRIPT
Structure-activity relationship
2

ACCEPTED MANUSCRIPT
1. Introduction
The phosphatidylinositol-3-kinase (PI3K) signaling pathway plays key regulatory
roles in multiple cellular functions including cell growth, survival, proliferation,
differentiation and invasion [1-5]. The PI3Ks family could be divided into several
different subtypes according their structural characteristics and substrate specificity
[6-9]. Among which, the PI3Kα is frequently activated, amplified and mutated in
human cancers, which could be an attractive therapeutic target [8, 10-13]. PI3Kα
comprises a catalytic subunit, p110α, and a regulatory subunit, p85α. The p110α
subunit consists of five domains, namely adaptor-binding domain, RAS-binding
domain, and the C2, helical, and kinase domains [14]. The gene encoding p110α is
frequently mutated in a variety of human tumors, including breast, colon and
endometrial cancers, and glioblastomas [15-20]. Nearly 80% of the mutations were
clustered in two hotspots [21-23]. One is located in the helical domain of p110α
(E542K, E545K and Q546K), which activates PI3Kα by relieving the auto-inhibition
imposed by the p85α nSH2 domain on the p110α subunit [24-26]. Another hotspot
(H1047R) is located in the kinase domain, which induces conformational changes that
increase PI3Kα membrane association and lipid binding [27-29]. These two hotspot
mutations act synergistically, but work independently to induce aberrant PI3K activity
and malignant transformation in cancer cells [30]. Inhibition of PI3Kα is a promising
approach to kill or, at least, greatly inhibit the growth of cancer cells [31-36].
To date, several inhibitors targeting PI3K have been developed, some of them are
under preclinical evaluation or early clinical studies [37-40]. BKM120, a pan-PI3K
inhibitor, is one of the most advanced drug lead in clinical trials for solid tumors [41,
42], which is active against a number of tumor types [43], however, such broad
anticancer activities usually result in potential risk of adverse events, including
hyperglycemia and liver toxicity [41, 44]. BYL-719, a PI3Kα inhibitor, exhibited
strong antitumor activities in vitro and tumor xenograft models, and is particularly
effective against cancer cell lines with mutant or amplified PIK3CA (PI3K, catalytic,
alpha polypeptide) [45-47]. Therefore, targeting the PI3Kα mutant enzyme may be a
3

ACCEPTED MANUSCRIPT
promising approach to develop efficacious inhibitors [48]. Recently, many groups
have demonstrated that PIK3CA mutations occur at high frequency in all kinds of
solid tumors [40, 49]. Moreover, most of the mutations concentrate on the two
hotspots [40, 50, 51]. Specific inhibitors that target the mutant forms of PI3Kα could
enhance treatment efficiency while decreasing various side-effects [52].
Although mutant selective inhibitors have significant advantages, the discovery of
such inhibitors remains a great challenge [22]. Via a high-throughput screening
campaign, an in-house library with diverse chemical structures was screened against
the PI3Kα H1047R mutant protein, from which Hit-01 (Figure 1) was identified to
weakly inhibit the enzyme function [53]. Herein, we report the synthesis, biological
evaluation and structure−activity relationship analysis of a series of Hit-01 analogues
as novel PI3Kα H1047R mutant inhibitors with greatly improved potency.
Figure 1 Chemical structures of reported PI3Ks inhibitors and Hit-01
2. Results and discussion
2.1 Chemistry
Hit-01 contains a quinoxaline core with two appending furan rings. First, we
synthesized Hit-01 and its analogues with the reaction of furoin (2) using
3,5-dichlorobenzene-1,2-diamine catalyzed by amberlyst-15 (Scheme 1). To identify
the optimal substitutions on the quinoxaline core, analogues with various
halo-substitutions (3a-h) were synthesized with yields ranging from 46 to 78%. In
addition, to explore whether the two furan groups are essential for biological activities,
we prepared 4a-c, with thienyl (4a), phenyl (4b), and morpholinyl (4c) moieties
attaching to the quinoxaline core respectively.
4

ACCEPTED MANUSCRIPT
Scheme 1 Reagents and conditions: (a) Et3N, THF/H2O = 1/4, 1,3-dimethylbenzimidazolium
iodide, reflux, overnight, 83%; (b) Amberlyst-15, H2O, reflux, 6 h, 46-78%; (c) Pd2(dba)3, BINAP,
K2CO3, dioxane, overnight.
With the improved enzymatic potencies for 3d, 3g and 3h (Table 1), we made
further efforts to enhance their activities. As shown in Scheme 2, target compounds
5a-d and 6a-c were successfully prepared by the Suzuki reaction of 3d with various
benzyl borates using Pd2(dba)3 as the catalyst and XPhos as a ligand, respectively.
However, we failed to obtain compounds 7a-n and 8a-e using the same reagents and
conditions. The reaction led only to bis-substitution of the two bromine atoms or
elimination of bromine in 3h. Significant efforts were made to optimize the reaction,
and we eventually found that Pd(OAc)2, s-phos and K3PO4 were the best combination
to accomplish the transformation smoothly, and 7a-n and 8a-e were thus obtained
with satisfactory yields (32-59%).
5

ACCEPTED MANUSCRIPT
O
O
N
N
N
O
2
a
R9
N
N
1
O
N
or
or
O
Y
O
R8
R6
Br
O
R9
R8
B
R7
O
3d
5
6
R6
R7
O
O
O
Br
N
N
Br
Br
N
N
Br
N
b
2
R9
1
O
O
O
Y
N
O
R8
R6
R9
R8
B
R7
3h
O
R6
7
8
R7
N
N
5a R6 = F
R7 = H R8 = H R9 = H
N
N
8a
8d
6a
6b
Y =
Y =
Y =
Y =
5b R6 = OMe R7 = H R8 = H R9 = H
5c R6 = OH R7 = H R8 = H R9 = H
5d R6 = CF3 R7 = H R8 = H R9 = H
O
O
8b
8c
Y =
Y =
8e
Y =
N
S
S
O
6c
Y =
O
R6 = H
7b R6 = F
R7 = H R8 = H R9 = H
R7 = H R8 = H R9 = H
R7 = F R8 = H R9 = H
R7 = H R8 = F R9 = H
R6 = Cl
7i R6 = Br
R7 = H R8 = H
R7 = H R8 = H
R9 = H
R9 = H
R9 = H
R9 = H
7a
7h
R6 = H
R6 = H
R6 = OH R7 = H R8 = H
R6 = OMe R7 = H R8 = H
7c
7j
7d
7k
R6 = F
R7 = F R8 = H R9 = H
R7 = H R8 = F R9 = H
R6 = H
R6 = H
R7 = H R8 = CF3 R9 = H
R7 = H R8 = OMe R9 = H
R7 = OMe R8 = H R9 = OMe
7e
7l
R6 = F
7f
7m
7g R6 = CF3 R7 = H R8 = H R9 = H
7n R6 = H
Scheme 2 Reagents and conditions: (a) Pd2(dba)3, XPhos, K2CO3, dioxane, overnight, yields:
38-57%. (b) Pd(OAc)2, s-Phos, K3PO4, t-BuOH/H2O = 20/1, 80 oC, 2.5-4 h, yields: 32-59%.
2.2 Inhibition of PI3Kα H1047R mutated kinase and cell proliferation
All synthesized compounds were evaluated for their antiproliferative activities on
PI3Kα H1047R mutated cell line HCT-116 and a PI3Kα H1047R non-mutation cancer
cell line MCF-7, as well as their inhibitory potency against PI3Kα H1047R mutated
kinase. The ZSTK474 and DOX were used as positive control compounds in the cell
viability assay, and results of the biological evaluation are summarized in Table 1.
For the first series of compounds 3a-h, the substitutions of R1-R4 played important
roles for exerting their enzymatic inhibitory activities. Compared with Hit-01,
compound 3a with a chlorine atom relocated from position 3 to 2 of the quinoxaline
core totally lost its biochemical activity. However, removal of the chlorine atom at
position 1 of the quinoxaline core afforded 3b with much improved activity. This
observation suggested that substitutions at position 2 might enhance compounds’
6

ACCEPTED MANUSCRIPT
potencies. Thus, different groups were introduced to position 2 on the quinoxaline
core. Compounds 3c, 3e and 3f were inactive, while 3d, 3g and 3h exhibited weak
activities. In addition, to explore whether the two furan rings appending to the
quinoxaline core are essential for biological activities, heterocyclic or aryl groups
were incorporated into Hit-01. To our disappointment, all these analogues (4a-c) lost
their inhibitory properties.
Table 1 Inhibitory activities of quinoxalines against two cancer cell lines and PI3K
α
H1047R
mutated kinase
EC50 (µM)
PI3Kα (H1047R)
76.0 ± 5.7
> 100
IC50 (µM)a
Compound
HCT-116
> 50
> 50
> 50
> 50
> 50
> 50
> 50
MCF-7
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
Hit-01
3a
3b
3c
3d
3e
3f
18.1 ± 3.9
> 100
33.5 ± 3.6
> 100
> 100
47.8 ± 4.5
39.3 ± 5.2
> 100
> 50
> 50
> 50
> 50
3g
3h
4a
4b
4c
5a
5b
5c
5d
6a
6b
6c
7a
7b
7c
7d
7e
7f
> 100
> 100
> 50
12.7 ± 2.7
87.3 ± 4.8
43.5 ± 4.2
11.8 ± 1.1
> 100
39.7 ± 3.2
> 50
> 50
45.3 ± 3.9
> 50
> 100
> 100
> 50
> 50
> 50
20.5 ± 2.3
0.137 ± 0.03
2.9 ± 0.5
1.9 ± 0.4
7.3 ± 0.7
3.2 ± 0.7
0.169 ± 0.06
> 100
11.23 ± 1.33
43.07 ± 2.35
38.05 ± 3.12
> 50
16.55 ± 1.02
13.46 ± 2.77
> 50
13.66 ± 2.18
19.56 ± 3.96
35.07 ± 2.76
> 50
27.94 ± 3.28
15.51 ± 2.54
> 50
7g
7h
7i
> 100
> 50
> 50
> 100
> 50
> 50
7j
> 100
0.516 ± 0.07
> 50
14.9 ± 1.69
> 50
25.18 ± 1.32
7k
7l
7

ACCEPTED MANUSCRIPT
> 100
> 100
> 50
> 50
> 50
> 50
7m
7n
10.8 ± 1.1
> 100
45.7 ± 2.21
> 50
> 50
> 50
8a
8b
1.2 ± 0.4
13.5 ± 0.9
18.6 ± 1.0
0.05 ± 0.01
ND
38.5 ± 4.67
> 50
> 50
1.31 ± 0.24
1.96 ± 0.09
40.9 ± 3.36
> 50
> 50
5.27 ± 0.73
3.28 ± 0.52
8c
8d
8e
ZSTK474
DOX
a Values are average of three independent experimental measurements and expressed as Mean ±
SD. ND: not determined.
Inspired by the above observations, we conducted further modifications on
compounds 3d and 3h. As shown in Table 1, compounds 5a-d displayed moderate
enzyme inhibitory activities, and 5a and 5d showed improved activities as compared
to Hit-01. Replacing the substituted-phenyl group at position 2 of the quinoxaline
core by polar heterocyclic moieties generated compounds 6a-c, which lost their ability
to inhibit the enzymatic function as well as cancer cell growth. In this round of SAR
campaign, the most potent compound (5d) showed an EC50 value of 11.8 µM against
PI3Kα H1047R protein, which is far from satisfaction. Thus, we continued our
structural modifications on Hit-01 with the aim to further improve its biochemical
potency.
As shown in Table 1, compounds 7a-g and 7l displayed not only improved EC50
value for inhibiting the mutated kinase function, but also showed enhanced
anti-proliferative activities against cancer cells. Among this series, compound 7b
demonstrated the best inhibitory activity for PI3Kα H1047R mutant (EC50 = 0.137
µM). Moreover, 7b exhibited potent antiproliferative activities for HCT116 (IC50
=
11.23 µM) and MCF-7 (IC50 = 13.66 µM) cells, respectively. However, compounds 7c
and 7d, with a fluorine atom at meta- and para- position of the phenyl ring
respectively showed reduced activities, suggesting that fluorine substitutions on the
phenyl ring had a detrimental effect on the potency. Besides, compound 7g (EC50 =
0.169 µM for PI3Kα H1047R mutant) also showed a similar inhibition effect as 7b,
which means trifluoromethyl group was another favorable substitution. To figure out
whether para-substitution on phenyl ring affect the activity, compounds 7l and 7m
8

ACCEPTED MANUSCRIPT
were synthesized and the results suggested that an electron-drawing R8 (7l, EC50
=
0.516 µM) was beneficial for the potency, while introduction of an electron-donating
group (-OCH3) resulted in a complete loss of activity (7m). At last, polar heterocyclic
groups were introduced into 3h, the resulting compounds 8a-e showed decreased
activities. In all, we modified Hit-01 and identified two potent compounds 7b and 7g,
which strongly inhibited PI3Kα H1047R mutant and cancer cell proliferation. Since
these compounds displayed similar anti-proliferative activities on both PI3Kα
H1047R mutated cell line HCT-116 and PI3Kα non-mutation cancer cell line MCF-7,
these compounds may inhibit both PI3Kα H1047R mutant and PI3Kα wild-type
proteins.
Table 2. Inhibitory activity (EC50, µM) against PI3Kα wild type and H1047R mutant.
PI3Kα wild-type
EC50 (µM) a
54.6 ± 6.5
PI3Kα H1047R
EC50 (µM) a
76.0 ± 5.7
Compounds
SI b
1.39
1.21
3.83
0.92
1.47
1.74
Hit-01
3d
27.6 ± 3.5
33.5 ± 3.6
12.4 ± 2.1
47.8 ± 4.5
3g
12.8 ± 1.8
11.8 ± 1.1
5d
0.093 ± 0.05
0.137 ± 0.03
7b
0.097 ± 0.02
0.169 ± 0.06
7g
a Values are average of three independent experimental determinations and the EC50 values were
expressed as Mean ± SD. b SI stands for selectivity index, which was calculated by comparing the
EC50 values of compounds inhibiting PI3Kα H1047R mutant against the EC50 value of the same
compound inhibiting PI3Kα wild-type in a biochemical assay.
To verify our hypothesis, some of the synthesized compounds were selected and
measured for their inhibitory activity against PI3Kα wild-type and the mutant. As
expected, all selected compounds showed similar inhibitory activities against PI3Kα
wild-type and H1047R mutant (Table 2). Of note, compounds 7b and 7g displayed
the best potencies against these two proteins, with EC50 values of 0.093 and 0.097 µM
against PI3Kα wild type, and 0.137 and 0.167 µM against PI3Kα H1047R mutant,
9

ACCEPTED MANUSCRIPT
respectively. Although compounds 7b and 7g did not exhibit selectivity index on the
PI3Kα wild type and mutant proteins, they could be used as novel and potent PI3Kα
inhibitors. Then we compared the biochemical activities of 7b with other isoforms of
PI3K and mTORC1. As summarized in Table 3, 7b displayed relative weak inhibitory
potencies against other PI3K isoforms, with over 40-fold selectivity for α over β, and
over 100-fold for α over γ and δ. Meanwhile, 7b have little effect on mTORC1
function when the tested concentration up to 30 µM.
Table 3. Enzymatic activities of compound 7b against PI3Ks and mTORC1 (IC50, µM a)
Kinase
PI3Kα
PI3Kβ
PI3Kγ
Compound 7b
0.093 ± 0.05
3.34 ± 0.13
9.89 ± 0.41
15.67 ± 0.52
> 30
BEZ-235
ZSTK474
0.040 ± 0.01
0.015 ± 0.003
0.067 ± 0.021
0.093 ± 0.032
0.079 ± 0.024
0.050 ± 0.017
0.044 ± 0.014
0.051 ± 0.018
0.010 ± 0.004
ND
PI3Kδ
mTORC1
a Values are average of three independent experimental determinations. ND: not determined.
2.3 Cell cycle arrest
Since 7b showed the highest potency, we next conducted additional studies to
investigate its biological roles in HCT116 cancer cell. We examined the effect of
compound 7b on cell cycle distribution using the propidium iodide (PI) staining kit.
HCT116 cells were treated with compound 7b (5, 10, 20 µM) or control for 24 h,
stained with PI and then analyzed on a flow cytometer.
As shown in Figure 2, compound 7b led to a significant accumulation of cells at
G1 phase from 38.72% to 44.42% (5 µM), 57.67% (10 µM) and 70.72% (20 µM),
accordingly. At the same time, it reduced the cells at S phase from 29.81 to 18.28%,
and G2 phase from 31.47% to 11.00%, respectively. Meanwhile, we found that cell
cycle arrest at G1 phase induced by compound 7b was concentration-dependent,
which might be one of the possible mechanisms for its cytotoxicity.
10

ACCEPTED MANUSCRIPT
Figure 2 Compound 7b induced cell cycle arrest in the HCT-116 cell. Cells were treated with 5,
10 and 20 µM of 7b for 24 ꢀh, which were then harvested, washed with PBS, and fixed with 70%
ice-cold ethanol at 4ꢀ°C. After overnight incubation, the cell pellets were collected by
centrifugation, resuspended in 50ꢀ µg/mL of RNase A in PBS, and incubated at 37ꢀ°C for 1 h. Next,
PI dye (50 ꢀµg/mL) was added, and the mixture was incubated at 37ꢀ°C for 15ꢀ min. Cell cycle
analysis was performed via PI fluorescence and flow cytometer. Data were expressed as Mean ±
SD (n= 3). *p < 0.05 and **p < 0.01 vs. the control.
2.4 Cell apoptosis
To explore the mechanism of cancer cells death, 7b was used to induce HCT116
cell apoptosis, which were examined with Annexin V-FITC/PI FACS assay. As
shown in Figure 3, the percentages of apoptotic population in HCT116 cells treated
with 7b at 6.25, 12.5 and 25 µM for 24 h were 14.60, 18.50, and 37.74%, respectively,
suggesting a dose-dependent apoptosis induction by 7b.
To further explore the apoptosis mechanism by compound 7b in HCT116 cells, we
examined expression of caspase-3, caspase-8 and caspase-9 after 7b treatment.
Caspase-8 is the most upstream protease. Once activated, caspase-8 would cleave and
activate downstream effector caspases like caspase-9 and caspase-3, which are
recognized as bio-markers for cell apoptosis. As shown in Figure 4, expression of
caspase-3, caspase-8 and caspase-9 were increased in a dose-dependent manner after
7b treatment, suggesting that the caspase activation pathway was involved in the
induction of apoptosis in HCT116 cancer cells by this compound.
11

ACCEPTED MANUSCRIPT
Figure 3 Pro-apoptotic effect of compound 7b on HCT116 cell. (A) Apoptotic assay by flow
cytometry. HCT116 cell were treated with compound 7b at 0, 6.25, 12.5, and 25 µM for 24 h.
Then cells were stained with Annexin V-FITC / PI, then detected by a flow cytometer. Cells in the
upper right quadrant indicate PI positive / Annexin V positive, late apoptotic, or necrotic cells, and
cells in the lower right quadrant indicate early apoptotic cells. (B) Bar graph represents statistics
of total apoptotic cell percentages from three dependent experiments. *p < 0.05 and **p < 0.01 vs.
the control.
Figure 4 Effects of compound 7b on caspase-3, caspase-8 and caspase-9 expression. Cells were
treated with compound 7b for 72 h, then total protein was extracted and subjected to Western blot
analysis, β-actin was used as an internal control. Bar graphs represent the expression level of
Caspase proteins, which was quantified by an Image J software. Data were expressed as Means ±
SD (n= 3). *p < 0.05 and **p < 0.01, vs. the control.
12

ACCEPTED MANUSCRIPT
2.5 Inhibition of PI3Kα H1047R mutant enzymatic activity by 7b
Since 7b inhibited PI3Kα H1047R mutant, we docked 7b into the PI3Kα H1047R
protein structure (PDB code: 3hhm), and found that 7b formed hydrogen bond
interactions with two important amino acids Lys802 and Asp933 (Figure 5). As both
Lys802 and Asp933 play crucial roles in PI3Kα phosphorylation, these interactions
could potentially explain the inhibitory activities of 7b on PI3Kα function. To verify
this hypothesis, we assessed in vitro PI3Kα inhibitory activity of compound 7b in
cultured HCT116 cells. The cells were treated with 7b at 5, 10 and 20 µM,
respectively, and subjected to Western blotting analysis. As expected, 7b caused a
decrease in p-AKT phosphorylation level (Figure 6), demonstrating that it is capable
of inhibiting PI3Kα H1047R activities.
A
B
Figure 5 Molecular docking of compound 7b with the PI3Kα H1047R mutant protein (PDB
code: 3hhm). A, Docking model of 7b in the binding pocket. B, Stereo view of 7b in the binding
pocket. Amino acid residues and compound 7b are shown as stick models, H-bonds are shown as
yellow dashed lines. The 3D graphical presentations were drawn by PyMol.
13

ACCEPTED MANUSCRIPT
Figure 6 Effects of compound 7b on the AKT phosphorylation. Cells were treated with compound
7b for 24 h, total protein was extracted and subjected to Western blot analysis, β-actin was used as
an internal control. Bar graphs represent the expression level of AKT and p-AKT proteins, which
were quantified by an Image J software. Data were expressed as Mean ± SD (n= 3). *p < 0.05 and
**p < 0.01 vs. the control.
3. Conclusion
In conclusion, several series of novel compounds containing a quinoxaline core
were synthesized and evaluated for their inhibitory effects on PI3Kα H1047R mutant
protein. Heterocyclic or aryl groups were incorporated into Hit-01 to replace the two
furan rings appending to the quinoxaline core. However, the resulting analogues (4a-c)
lost their inhibitory properties, suggesting that the two furan groups are critical for its
potency. This observation is in line with our docking studies. While keeping the furan
rings and the quinoxaline core, we modified 3a. The resulting analogues 5a and 5d
bearing electron-withdrawing groups at the 2-position of benzene ring exhibited
stronger inhibitory activities than the related analogues 5b-c with electron-donating
groups at the 2-position of benzene ring. Further SARs analysis enabled us to discover
compound 7b, which strongly inhibited both PI3Kα H1047R mutant and wild-type
14

ACCEPTED MANUSCRIPT
proteins with EC50 values of 0.137, and 0.093 µM, respectively, while had little effect
on other PI3K isoforms. Western blotting assay indicated that 7b reduced the p-AKT
phosphorylation level, demonstrating its inhibitory activity against the PI3Kα
H1047R mutant. Moreover, 7b was docked well into the PI3Kα H1047R mutant
active site and formed strong hydrogen bonds with the protein. In addition, 7b
reduced HCT116 cancer cell proliferation with an IC50 of 11.23 µM. Furthermore,
studies on potential cytotoxic mechanisms indicated that compound 7b could arrest
cell cycle at G1 phase. Annexin V-FITC assay and Western blotting analysis revealed
that compound 7b induced HCT116 apoptosis by up-regulating caspase-3, caspase-8
and caspase-9 expressions. Wortmannin is a known and potent covalent pan-PI3K
inhibitor. Its non-specificity led to side / off-target effects. Our results suggest that 7b
is a selective PI3Kα inhibitor, which reversibly bind to the PI3Kα pocket. With these
superior properties, compound 7b could be a promising lead for developing potent
PI3Kα inhibitors for cancer treatment.
4. Experimental section
4.1 Materials and methods
All chemicals were purchased from commercial sources and used as received.
Reactions were monitored by thin-layer chromatography (TLC) and carried out on
Merck Kieselgel 60 F254 plates, which could be visualized under UV light at 254 nm.
1
Column chromatography was performed on silica gel (60−120 mesh). All H NMR
spectra were obtained on an Agilent 400 MR spectrometer at ambient temperature and
13
reported in ppm downfield from TMS (0 ppm). All C NMR spectra were obtained
with proton decoupling on an Agilent 400 MR DD2 (100 MHz) or 600 MR DD2
spectrometer and reported in ppm relative to CDCl3 (77.16 ppm). Coupling constants
were reported as Hz with multiplicity denoted as s (singlet), d (doublet), t (triplet), q
(quartet), m (multiplet), and br (broad). The NMR data was processed by software
Mest Re-Nova (Ver. 9.0.0.12821, Mestrelab Research S.L.). High resolution mass
spectra (HRMS) were obtained on a Bruker SolariX 7.0T spectrometer. Melting point
was recorded on a WRS-2A digital melting point apparatus. Compounds exhibited
15

ACCEPTED MANUSCRIPT
purities of more than 95% as determined by peak area integration in reverse phase
chromatography. The mobile phase is the mixtures of A and B, where A is 0.05%
TFA water and B is acetonitrile, with a flow rate of 7 mL/min.
4.2 Chemical synthesis
4.2.1
General
procedure
for
the
preparation
of
1,2-di(furan-2-yl)-2-hydroxyethan-1-one (2).
Furfural (10.0 mmol, 960 mg) and 1,3-dimethylbenzimidazolium iodide (5.0 mmol,
1370 mg) were dissolved in 25 mL THF/H2O (1/4). Then TEA (0.25 mL) was added,
the mixture was heated to reflux for 5 h and cooled to room temperature. Ethyl acetate
(50 mL) was added, the organic phase was washed with water (3 × 50 mL), dried over
MgSO4, and concentrated. The crude products were purified by column
chromatography, affording 2 (530 mg, 55% yield) as yellow solid. 1H NMR (400
MHz, CDCl3) δ 7.61 (dd, J = 1.7, 0.8 Hz, 1 H), 7.37 (dd, J = 1.8, 0.8 Hz, 1 H), 7.25
(dd, J = 3.7, 0.8 Hz, 1 H), 6.40 (d, J = 3.3 Hz, 1 H), 6.54 (dd, J = 3.7, 1.7 Hz, 1 H),
6.35 (dd, J = 3.3, 1.9 Hz, 1 H), 5.80 (s, 1 H), 4.19 (s, 1 H). The data is in line with the
literature reported previously [54].
4.2.2 General procedure for the preparation of compounds 3a-h and Hit-01
Furoin 2 (1.0 mmol, 192 mg) and o-phenylenediamines (1.0 mmol) were dissolved
in 3 mL H2O. Amberlyst-15 (100 mg) was added, the mixture was then heated to
reflux for 10 h and cooled to room temperature. Ethyl acetate (10 mL) was added, the
organic phase was washed with water (3 × 10 mL), dried over MgSO4, and
concentrated. The crude products were purified by column chromatography, affording
3a-h and Hit-01.
4.2.2.1 5,7-Dichloro-2,3-di(furan-2-yl)quinoxaline (Hit-01)
1
Hit-01 was obtained as brown solid. Yield 69%. H NMR (400 MHz, CDCl3) δ
8.04–7.84 (m, 1H), 7.61 (d, J = 15.8 Hz, 3H), 6.91–6.44 (m, 4H). 13C NMR (100
MHz, CDCl3) δ 150.55, 150.15, 144.92, 144.55, 143.61, 142.48, 141.27, 136.01,
135.29, 133.87, 130.58, 126.94, 114.40, 114.12, 112.16, 112.11. HRMS (m/z): calcd
for C16H9Cl2N2O2 330.9963 [M+H]+; found 330.9971.
16

ACCEPTED MANUSCRIPT
4.2.2.2 5,6-Dichloro-2,3-di(furan-2-yl)quinoxaline (3a)
The product was obtained as brown solid. Yield 75%. 1H NMR (400 MHz, CDCl3)
δ 7.98–7.88 (m, 1H), 7.66–7.52 (m, 3H), 6.82 (dd, J = 3.5, 0.8 Hz, 1H), 6.71 (dt, J =
3.6, 0.9 Hz, 1H), 6.64–6.49 (m, 2H). 13C NMR (100 MHz, CDCl3) δ 150.61, 150.24,
144.76, 144.68, 142.60, 142.60, 139.70, 138.74, 138.17, 134.50, 131.35, 127.97,
114.32, 114.00, 112.21, 112.07. HRMS (m/z): calcd for C16H9Cl2N2O2 330.9963
[M+H]+; found 330.9971.
4.2.2.3 6-Chloro-2,3-di(furan-2-yl)quinoxaline (3b)
The product was obtained as brown solid. Yield 71%. 1H NMR (400 M, DMSO-d6):
δ 8.19 (1H, J = 2.1 Hz, d), 8.16 (1H, J = 9.0 Hz, d), 7.94–7.87 (3H, m), 6.78–6.72 (4H,
m). The data is in line with the literature reported previously [55].
4.2.2.4 6-Fluoro-2,3-di(furan-2-yl)quinoxaline (3c)
The product was obtained as brown solid. Yield 78%. 1H NMR (400 MHz, CDCl3)
δ 8.11 (m, 1H), 7.74 (dd, J = 9.3, 3.1 Hz, 1H), 7.64–7.57 (m, 2H), 7.50 (td, J = 8.5,
2.8 Hz, 1H), 6.74–6.61 (m, 2H), 6.56 (m, 2H). 13C NMR (100 MHz, CDCl3) δ 163.11
(J = 252 Hz), 150.60, 150.50, 144.53, 144.14, 143.28, 141.93, 141.51, 141.38, 137.74,
131.19, 120.82, 113.60, 112.87, 112.68, 112.01. The data is in line with the literature
reported previously [56].
4.2.2.5 6-Bromo-2,3-di(furan-2-yl)quinoxaline (3d)
The product was obtained as brown solid. Yield 55%. 1H NMR (400 MHz, CDCl3)
δ 8.31 (d, J = 2.1 Hz, 1H), 7.98 (d, J = 8.9 Hz, 1H), 7.80 (dd, J = 8.9, 2.1 Hz, 1H),
13
7.62 (d, J = 1.6 Hz, 2H), 6.70 (t, J = 3.3 Hz, 2H), 6.57 (dd, J = 3.5, 1.8 Hz, 2H). C
NMR (100 MHz, CDCl3) δ 150.55, 150.49, 144.52, 144.41, 143.20, 142.71, 141.10,
139.30, 133.84, 131.32, 130.29, 124.31, 113.63, 113.38, 112.03, 112.00. The data is
in line with the literature reported previously [57].
4.2.2.6 2,3-Di(furan-2-yl)-6-methylquinoxaline (3e)
The product was obtained as brown solid. Yield 59%. 1H NMR (400 MHz, CDCl3)
δ 7.99 (d, J = 8.6 Hz, 1H), 7.92–7.82 (m, 1H), 7.59 (d, J = 0.9 Hz, 2H), 7.54 (dd, J =
8.6, 1.9 Hz, 1H), 6.61 (d, J = 3.5 Hz, 2H), 6.53 (m, 2H), 2.56 (s, 3H). 13C NMR (100
17

ACCEPTED MANUSCRIPT
MHz, CDCl3) δ 150.92, 144.05, 143.93, 142.55, 141.81, 141.04, 140.69, 139.08,
132.71, 128.59, 127.93, 112.75, 112.50, 111.83, 111.80, 21.87. The data is in line with
the literature reported previously [58].
4.2.2.7 2,3-Di(furan-2-yl)-6-(trifluoromethyl)quinoxaline (3f)
The product was obtained as brown solid. Yield 46%.1H NMR (400 MHz, CDCl3) δ
8.42 (s, 1H), 8.21 (d, J = 8.8 Hz, 1H), 7.88 (dd, J = 8.9, 2.0 Hz, 1H), 7.63 (dd, J = 4.2,
13
1.7 Hz, 2H), 6.76 (dd, J = 7.5, 3.5 Hz, 2H), 6.58 (dt, J = 3.4, 1.6Hz, 2H). C NMR
(100 MHz, CDCl3) δ 150.34, 144.84, 144.65, 144.10, 143.60, 139.47, 131.92, 131.59,
130.17, 127.05, 127.01, 125.86, 125.83, 114.23, 113.87, 112.14, 112.09. HRMS (m/z):
calcd for C17H10F3N2O2 331.0616 [M+H]+; found 331.0610.
4.2.2.8 2,3-Di(furan-2-yl)-6-phenylquinoxaline (3g)
The product was obtained as brown solid. Yield 55%. 1H NMR (400 MHz, CDCl3):
δ 8.39–8.35 (d, 1H, J = 2.0 Hz), 8.22–8.17 (d, 1H, J = 8.7 Hz), 8.06–8.01 (m, 1H),
7.80–7.75 (d, 2H, J = 7.2 Hz), 7.66–7.63 (s, 2H), 7.55–7.40 (m, 3H), 6.70–6.54 (m,
4H); The data is in line with the literature reported previously [57].
4.2.2.9 6,7-Dibromo-2,3-di(furan-2-yl)quinoxaline (3h)
The product was obtained as brown solid. Yield 67%.1H NMR (400 MHz, CDCl3) δ
8.40 (s, 2H), 7.61 (d, J = 1.8 Hz, 2H), 6.72 (d, J = 3.5 Hz, 2H), 6.56 (dd, J = 3.6, 1.8
Hz, 2H). 13C NMR (100 MHz, CDCl3) δ 150.33, 144.68, 143.31, 139.69, 132.95,
126.83, 113.96, 112.11. HRMS (m/z): calcd for C16H9Br2N2O2 418.8953 [M+H]+;
found 418.8968.
4.2.3.1 General procedure for the preparation of compound 4a
Thenoin (1.0 mmol, 225 mg) and o-phenylenediamine (1.0 mmol) were dissolved
in 3 mL H2O. Amberlyst (100 mg) was added, the mixture was then heated to reflux
for 10 h and cooled to room temperature. Ethyl acetate (10 mL) was added, the
organic phase was washed with water (3 × 10 mL), dried over MgSO4, and
concentrated. The crude products were purified by column chromatography, affording
1
4a. Yield 49%. M.p. 129–130 °C; H NMR (400 MHz, CDCl3) δ 8.10–8.07 (m, 2H),
7.75–7.72 (m, 2H), 7.50 (d, J = 5.2 Hz, 2H), 7.27–7.24 (m, 2H), 7.03–7.06 (m, 2H).
18

ACCEPTED MANUSCRIPT
The data is in line with the literature reported previously [59].
4.2.2.9 2,3-Diphenylquinoxaline (4b)
1
The product was obtained as brown solid. Yield 50%. H NMR (400 MHz,
DMSO-d6): δ 8.19–8.14 (2H, m), 7.95–7.81 (2H, m), 7.50–7.47 (4H, m), 7.45–7.33
(6H, m). The data is in line with the literature reported previously [60].
4.2.3.2 General procedure for the preparation of compound 4c
2,3-Dibromoquinoxaline (1.0 mmol, 285 mg) and morpholine (2.0 mmol, 180 mg)
were dissolved into 5 mL dioxane, then Pd2(dba)3 (0.01 mmol), BINAP (0.02 mmol),
K2CO3 (2.0 mmol) were added to the mixture, respectively. The reaction was carried
o
out at 80 C in N2 atmosphere. After completion of the reaction, the mixture was
cooled to room temperature and evaporated the solvent under reduced pressure. The
affording crude product was purified by column chromatography to give 4c. Yield
49%. M.p. 200–201 °C. 1H NMR (400 MHz, DMSO) δ 7.77–7.52 (m, 2H), 7.49–7.28
(m, 2H), 3.75 (t, J = 4.5 Hz, 8H), 3.56–3.41 (m, 8H). The data matched the reported
literature [61].
4.2.4 General procedures for the preparation of compounds 5a–d and 6a–c.
Compound 3d (1.0 mmol) and its borate (1.0 mmol) were added to 5 mL dioxane.
Pd2(dba)3 (0.01 mmol), BINAP (0.02 mmol), K2CO3 (2 mmol) were added to the
o
mixture, respectively. The reaction was carried out at 80 C in N2 atmosphere. After
cooling to room temperature, the solvent of the mixture was evaporated under reduced
pressure to afford crude product. Purification was made through column
chromatography to give compounds 5a–d and 6a–c.
4.2.4.1 6-(2-Fluorophenyl)-2,3-di(furan-2-yl)quinoxaline (5a)
The product was obtained as brown solid. Yield 57%. 1H NMR (400 MHz, CDCl3)
δ 8.32 (s, 1H), 8.18 (d, J = 8.7 Hz, 1H), 7.97 (d, J = 8.9 Hz, 1H), 7.61 (d, J = 13.6 Hz,
3H), 7.39 (q, J = 6.9 Hz, 1H), 7.32–7.16 (m, 3H), 6.68 (t, J = 3.6 Hz, 2H), 6.58–6.48
(m, 2H). 13C NMR (100 MHz, CDCl3) δ 161.14, 158.66, 150.80, 144.27, 144.23,
142.97, 142.70, 140.53, 139.96, 138.03, 131.69, 131.66, 130.83, 130.80, 130.02,
129.94, 128.91, 128.87, 128.84, 127.68, 127.55, 124.69, 124.66, 116.52, 116.29,
19

ACCEPTED MANUSCRIPT
113.13, 113.09, 111.93. HRMS (m/z): calcd for C22H14FN2O2 357.0973 [M+H]+;
found 357.0979.
4.2.4.2 2,3-Di(furan-2-yl)-6-(2-methoxyphenyl)quinoxaline (5b)
The product was obtained as brown solid. Yield 57%. 1H NMR (400 MHz, CDCl3)
δ 8.30 (d, J = 2.0 Hz, 1H), 8.15 (d, J = 8.7 Hz, 1H), 7.99 (dd, J = 8.7, 2.0 Hz, 1H),
7.78–7.68 (m, 2H), 7.67–7.57 (m, 2H), 7.13–6.97 (m, 2H), 6.66 (dd, J = 6.4, 3.4 Hz,
2H), 6.56 (dt, J = 3.2, 1.5 Hz, 2H), 3.87 (s, 3H). 13C NMR (100 MHz, CDCl3) δ
159.96, 150.88, 150.82, 144.21, 144.10, 142.95, 142.66, 142.06, 140.96, 139.68,
131.86, 129.82, 129.27, 128.49, 125.41, 114.56, 113.00, 112.82, 111.90, 55.38.
HRMS (m/z): calcd for C23H17N2O3 369.1158 [M+H]+; found 369.1163.
4.2.4.3 2-(2,3-Di(furan-2-yl) quinoxaline-6-yl)phenol (5c)
1
The product was obtained as brown solid. Yield 42%. H NMR (400 MHz,
DMSO-d6) δ 9.89 (s, 1H), 8.22 (dt, J = 1.8, 0.9 Hz, 1H), 8.09 (q, J =1.1 Hz, 2H), 7.89
(q, J = 1.2 Hz, 2H), 7.48 (dd, J = 7.6, 1.6 Hz, 1H), 7.35–7.22 (m, 1H), 7.11–6.90 (m,
13
2H), 6.81–6.58 (m, 4H). C NMR (100 MHz, DMSO-d6) δ 155.15, 150.89, 150.87,
145.31, 145.29, 142.62, 142.16, 141.56, 140.33, 139.35, 133.16, 131.07, 130.08,
128.30, 128.25, 126.43, 120.24, 116.80, 113.21, 112.65, 112.62. HRMS (m/z): calcd
for C22H15N2O3 355.1004 [M+H]+; found 355.1005.
4.2.4.4 2,3-Di(furan-2-yl)-6-(2-(trifluoromethyl)phenyl)quinoxaline (5d)
The product was obtained as brown solid. Yield 55%. 1H NMR (400 MHz, CDCl3)
δ 8.15 (d, J = 8.7 Hz, 1H), 8.09 (d, J = 1.9 Hz, 1H), 7.81 (dd, J = 7.9, 1.3 Hz, 1H),
7.71 (m, 1H), 7.66–7.59 (m, 3H), 7.58–7.49 (m, 1H), 7.43 (dd, J = 7.6, 1.4 Hz, 1H),
13
6.69 (m, 2H), 6.57 (m, 2H). C NMR (100 MHz, CDCl3) δ 150.78, 144.30, 144.25,
143.12, 142.88, 142.07, 139.96, 139.89, 139.81, 139.79, 131.86, 131.83, 131.81,
131.54, 128.97, 128.96, 128.69, 128.39, 128.30, 128.05, 126.32, 126.27, 125.37,
122.65, 113.16, 113.13, 111.94. HRMS (m/z): calcd for C23H14F3N2O2 407.1001
[M+H]+; found 407.1004.
4.2.4.5 2,3-Di(furan-2-yl)-6-(pyridin-4-yl)quinoxaline (6a)
The product was obtained as brown solid. Yield 38%. 1H NMR (400 MHz, CDCl3)
20

ACCEPTED MANUSCRIPT
δ 8.85–8.69 (m, 2H), 8.42 (d, J = 2.0 Hz, 1H), 8.23 (d, J = 8.7 Hz, 1H), 8.02 (dd, J =
8.8, 2.1 Hz, 1H), 7.71–7.66 (m, 2H), 7.64 (d, J = 1.7 Hz, 2H), 6.73 (dd, J = 5.3, 3.5
Hz, 2H), 6.58 (dd, J = 3.5, 1.8 Hz, 2H). 13C NMR (100 MHz, CDCl3) δ 150.64,
150.44, 149.27, 148.55, 146.87, 144.49, 143.31, 143.09, 140.66, 140.59, 139.73,
129.95, 128.98, 127.27, 121.73, 113.55, 113.46, 112.05, 112.03. HRMS (m/z): calcd
for C21H14N3O2 340.1012 [M+H]+; found 340.1006.
4.2.4.6 6-(Benzofuran-2-yl)-2,3-di(furan-2-yl)quinoxaline (6b)
The product was obtained as brown solid. Yield 45%. 1H NMR (400 MHz, CDCl3)
δ 8.56 (s, 1H), 8.13 (q, J = 8.8 Hz, 2H), 7.68–7.45 (m, 4H), 7.38–7.04 (m, 4H), 6.69
13
(dd, J = 10.2, 3.4 Hz, 2H), 6.57 (d, J = 4.0 Hz, 2H); C NMR (100 MHz, CDCl3) δ
155.28, 154.36, 150.79, 144.27, 143.16, 142.35, 140.79, 140.54, 132.02, 129.49,
128.93, 127.17, 125.13, 124.20, 123.20, 121.29, 113.20, 111.96, 111.93, 111.34,
103.80. HRMS (m/z): calcd for C24H15N2O3 379.1004 [M+H]+; found 379.1002.
4.2.4.7 6-(Benzo[b]thiophen-2-yl)-2,3-di(furan-2-yl)quinoxaline (6c)
The product was obtained as brown solid. Yield 45%. 1H NMR (400 MHz, CDCl3)
δ 8.39 (s, 1H), 8.09 (q, J = 8.8 Hz, 2H), 7.87–7.58 (m, 5H), 7.34 (q, J = 6.0, 4.9 Hz,
13
2H), 6.68 (t, J = 3.9 Hz, 2H), 6.57 (s, 2H). C NMR (100 MHz, CDCl3) δ 150.77,
150.69, 144.34, 144.27, 143.20, 142.54, 142.36, 140.82, 140.48, 140.38, 139.97,
136.10, 129.48, 128.64, 125.60, 125.03, 124.73, 123.99, 122.34, 121.41, 113.26,
113.19, 111.97, 111.95. HRMS (m/z): calcd for C24H15N2O2S 395.0783 [M+H]+;
found 395.0779.
4.2.5 General procedures for the preparation of compounds 7a–n and 8a–e
Compound 3h (1.0 mmol) and its relative borates (1.0 mmol) were added to 5 mL
dioxane. Pd(OAc)2 (0.01 mmol), s-phos (0.02 mmol), K3PO4 (2.0 mmol) were added
o
to the mixture, which were then heated to 80 C in N2 atmosphere. After completion
of the reaction, the mixture was cooled to room temperature and evaporated the
solvent under reduced pressure. The affording crude products were purified by
column chromatography to give compounds 7a–n and 8a–e.
4.2.5.1 6-Bromo-2,3-di(furan-2-yl)-7-phenylquinoxaline (7a)
21

ACCEPTED MANUSCRIPT
The product was obtained as brown solid. Yield 43%. 1H NMR (400 MHz, CDCl3)
δ 8.49 (d, J = 1.0 Hz, 1H), 8.09 (d, J = 1.0 Hz, 1H), 7.63 (m, 2H), 7.55–7.39 (m, 6H),
13
6.72 (dd, J = 3.5, 0.9 Hz, 2H), 6.58 (m, 2H). C NMR (100 MHz, CDCl3) δ 150.64,
150.56, 144.73, 144.52, 144.42, 143.18, 140.05, 139.84, 139.46, 132.72, 130.34,
129.50, 128.24, 128.09, 125.49, 113.61, 113.42, 112.04, 112.01, 109.99. HRMS (m/z):
calcd for C22H14BrN2O2 417.0173 [M+H]+; found 417.0168.
4.2.5.2 6-Bromo-7-(2-fluorophenyl)-2,3-di(furan-2-yl)quinoxaline (7b)
The product was obtained as brown solid. Yield 45%. 1H NMR (400 MHz, CDCl3)
δ 8.49 (s, 1H), 8.08 (s, 1H), 7.63 (dd, J = 7.5, 1.7 Hz, 2H), 7.53–7.39 (m, 1H), 7.42–
7.34 (m, 1H), 7.32–7.14 (m, 2H), 6.73 (t, J = 3.2 Hz, 2H), 6.58 (td, J = 3.9, 1.8 Hz,
2H). 13C NMR (100 MHz, CDCl3) δ 160.72, 158.25, 150.56, 144.59, 144.44, 143.41,
143.10, 140.42, 139.37, 139.24, 132.37, 131.26, 131.13, 130.55, 130.47, 126.16,
123.99, 115.89, 115.67, 113.76, 113.47, 112.04. HRMS (m/z): calcd for
C23H13BrFN2O2 435.0138 [M+H]+; found 435.0141
4.2.5.3 6-Bromo-7-(3-fluorophenyl)-2,3-di(furan-2-yl)quinoxaline (7c)
The product was obtained as brown solid. Yield 39%. 1H NMR (400 MHz, CDCl3)
δ 8.49 (s, 1H), 8.07 (s, 1H), 7.63 (d, J = 5.7 Hz, 2H), 7.51–7.41 (m, 1H), 7.31–7.22 (m,
3H), 7.15 (td, J = 8.5, 2.5 Hz, 1H), 6.73 (d, J = 3.4 Hz, 2H), 6.59 (d, J = 4.3 Hz, 2H).
13C NMR (100 MHz, CDCl3) δ 163.50, 161.05, 150.56, 150.49, 144.61, 144.50,
143.37, 143.30, 143.22, 141.81, 141.73, 140.19, 139.35, 132.89, 130.41, 129.76,
129.68, 125.33, 125.30, 124.87, 116.81, 116.59, 115.34, 115.13, 113.80, 113.57,
112.08, 112.05. HRMS (m/z): calcd for C23H13BrFN2O2 435.0138 [M+H]+; found
435.0141
4.2.5.4 6-Bromo-7-(4-fluorophenyl)-2,3-di(furan-2-yl)quinoxaline (7d)
The product was obtained as brown solid. Yield 41%. 1H NMR (400 MHz, CDCl3)
δ 8.47 (s, 1H), 8.05 (s, 1H), 7.63 (d, J = 4.8 Hz, 2H), 7.49 (dd, J = 8.3, 5.3 Hz, 2H),
13
7.17 (t, J = 8.5 Hz, 2H), 6.72 (t, J = 4.1 Hz, 2H), 6.65–6.50 (m, 2H). C NMR (100
MHz, CDCl3) δ 163.96, 161.50, 150.58, 150.52, 144.56, 144.48, 143.61, 143.25,
143.18, 140.07, 139.41, 135.80, 135.77, 132.79, 131.33, 131.25, 130.36, 125.34,
22

ACCEPTED MANUSCRIPT
115.27, 115.05, 113.70, 113.52, 112.06, 112.04. HRMS (m/z): calcd for
C23H13BrFN2O2 435.0138 [M+H]+; found 435.0141
4.2.5.5 6-Bromo-7-(2,3-difluorophenyl)-2,3-di(furan-2-yl)quinoxaline (7e)
The product was obtained as brown solid. Yield 39%. 1H NMR (400 MHz, CDCl3)
δ 8.49 (s, 1H), 8.07 (s, 1H), 7.63 (dd, J = 7.0, 1.7 Hz, 2H), 7.34–7.24 (m, 1H), 7.20
13
(m, 1H), 7.13 (m, 1H), 6.74 (d, J = 3.4 Hz, 2H), 6.58 (td, J = 3.8, 1.8 Hz, 2H). C
NMR (100 MHz, CDCl3) δ 150.55 (J = 287.8 Hz), 150.54, 150.47, 147.87 (J = 247.4
Hz), 144.67, 144.50, 143.60, 143.21, 140.57, 139.17, 137.99, 132.55, 131.20, 129.84,
126.15, 123.99, 117.73, 117.56, 113.91, 113.60, 112.09, 112.05. HRMS (m/z): calcd
for C22H12BrF2N2O2 453.0044 [M+H]+; found 453.0047.
4.2.5.6 6-Bromo-7-(2,4-difluorophenyl)-2,3-di(furan-2-yl)quinoxaline (7f)
The product was obtained as brown solid. Yield 40%. 1H NMR (400 MHz, CDCl3)
δ 8.48 (s, 1H), 8.05 (s, 1H), 7.63 (d, J = 6.6 Hz, 2H), 7.35 (q, J = 7.7 Hz, 1H), 6.99 (dt,
13
J = 18.0, 8.1 Hz, 2H), 6.73 (d, J = 3.5 Hz, 2H), 6.58 (d, J = 4.2 Hz, 2H). C NMR
(100 MHz, CDCl3) δ 150.55, 150.47, 144.64, 144.50, 143.51, 143.17, 140.47, 139.20,
138.35, 132.47, 132.27, 132.22, 132.17, 132.13, 131.31, 126.07, 113.86, 113.57,
112.09, 112.05, 111.47, 111.44, 111.26, 111.22, 104.53, 104.28, 104.02. HRMS (m/z):
calcd for C23H13BrFN2O2 435.0138 [M+H]+; found 435.0141
4.2.5.7 6-Bromo-2,3-di(furan-2-yl)-7-(2-(trifluoromethyl)phenyl)quinoxaline (7g)
The product was obtained as brown solid. Yield 49%. 1H NMR (400 MHz, CDCl3)
δ 8.47 (s, 1H), 8.02 (s, 1H), 7.82 (d, J = 7.7 Hz, 1H), 7.69–7.51 (m, 4H), 7.35 (d, J =
7.6 Hz, 1H), 7.26 (d, J = 1.9 Hz, 1H), 6.73 (dd, J = 6.2, 3.5 Hz, 2H), 6.58 (m, 2H). 13C
NMR (150 MHz, CDCl3) δ 150.58, 150.49, 144.61, 144.44, 143.51, 143.21, 142.06,
140.40, 138.83, 138.42, 131.94, 131.63, 131.39, 130.38, 128.83, 128.63, 128.58,
126.18, 126.15, 126.11, 126.08, 113.78, 113.50, 112.06, 112.03. HRMS (m/z): calcd
for C23H13BrF3N2O2 485.0107 [M+H]+; found 485.0102.
4.2.5.8 6-Bromo-7-(2-chlorophenyl)-2,3-di(furan-2-yl)quinoxaline (7h)
The product was obtained as brown solid. Yield 49%. 1H NMR (400 MHz, CDCl3)
δ 8.49 (s, 1H), 8.02 (s, 1H), 7.63 (dd, J = 9.1, 1.7 Hz, 2H), 7.53 (dd, J = 7.6, 1.8 Hz,
23

ACCEPTED MANUSCRIPT
1H), 7.45–7.29 (m, 4H), 7.26 (s, 2H), 6.77–6.67 (m, 2H), 6.58 (m, 2H). 13C NMR
(100 MHz, CDCl3) δ 150.64, 150.54, 144.59, 144.43, 143.41, 143.09, 142.56, 140.44,
139.27, 138.95, 133.43, 132.21, 130.94, 130.59, 129.80, 129.54, 126.65, 126.21,
113.73, 113.46, 112.06, 112.04. HRMS (m/z): calcd for C22H13BrClN2O2 450.9843
[M+H]+; found 450.9851.
4.2.5.9 6-Bromo-7-(2-bromophenyl)-2,3-di(furan-2-yl)quinoxaline (7i)
The product was obtained as brown solid. Yield 43%. 1H NMR (400 MHz, CDCl3)
δ 8.49 (s, 1H), 8.00 (s, 1H), 7.75–7.68 (m, 1H), 7.63 (m, 2H), 7.43 (m, 1H), 7.33 (m,
13
2H), 6.73 (dd, J = 4.7, 3.6 Hz, 2H), 6.58 (m, 2H). C NMR (100 MHz, CDCl3) δ
150.65, 150.55, 144.59, 144.43, 144.10, 143.40, 143.10, 140.98, 140.44, 139.28,
132.67, 132.22, 130.83, 130.46, 129.90, 127.25, 126.15, 123.45, 113.73, 113.47,
112.06, 112.04. HRMS (m/z): calcd for C22H13Br2N2O2 494.9338 [M+H]+; found
494.9342.
4.2.5.10 2-(7-Bromo-2,3-di(furan-2-yl)quinoxaline-6-yl)phenol (7j)
The product was obtained as brown solid. Yield 40%. 1H NMR (400 MHz, CDCl3)
δ 8.41 (s, 1H), 8.00 (s, 1H), 7.60 (m, 2H), 7.36 (td, J = 7.6, 1.8 Hz, 1H), 7.21 (dd, J =
13
7.8, 1.8 Hz, 1H), 7.08–6.98 (m, 3H), 6.71 (m, 2H), 6.56 (m, 2H), 5.89 (s, 1H). C
NMR (100 MHz, CDCl3) δ 152.89, 150.31, 144.63, 144.59, 143.20, 142.89, 140.85,
140.10, 139.04, 132.56, 131.27, 130.73, 130.22, 127.04, 126.93, 120.50, 116.45,
113.90, 113.78, 112.09. HRMS (m/z): calcd for C22H14BrN2O3 433.0182 [M+H]+;
found 433.0177.
4.2.5.11 6-Bromo-2,3-di(furan-2-yl)-7-(2-methoxyphenyl)quinoxaline (7k)
The product was obtained as brown solid. Yield 40%. 1H NMR (400 MHz, CDCl3)
δ 8.46 (s, 1H), 8.04 (s, 1H), 7.62 (d, J = 8.0 Hz, 2H), 7.45 (t, J = 7.9 Hz, 1H), 7.25 (d,
J = 4.0 Hz, 2H), 7.14–6.93 (m, 2H), 6.70 (dd, J = 5.5, 3.4 Hz, 2H), 6.57 (d, J = 4.4 Hz,
2H), 3.80 (s, 4H). 13C NMR (100 MHz, CDCl3) δ 156.68, 150.77, 150.66, 144.41,
144.26, 143.04, 142.84, 142.54, 140.20, 139.48, 131.94, 130.76, 130.73, 129.98,
129.20, 127.38, 120.44, 113.40, 113.16, 111.97, 111.95, 110.95, 55.53. HRMS (m/z):
calcd for C23H16BrN2O3 447.0339 [M+H]+; found 447.0335.
24

ACCEPTED MANUSCRIPT
4.2.5.12 6-Bromo-2,3-di(furan-2-yl)-7-(4-(trifluoromethyl)phenyl)quinoxaline (7l)
The product was obtained as brown solid. Yield 59%. 1H NMR (400 MHz, CDCl3)
δ 8.51 (s, 1H), 8.08 (s, 1H), 7.75 (d, J = 8.0 Hz, 2H), 7.71–7.60 (m, 4H), 6.75 (t, J =
13
3.6 Hz, 2H), 6.59 (m, 2H). C NMR (150 MHz, CDCl3) δ 150.48, 150.43, 144.66,
144.55, 143.48, 143.28, 143.25, 143.09, 140.28, 139.32, 132.99, 130.46, 130.26,
129.92, 125.17, 125.15, 125.12, 125.09, 124.54, 113.91, 113.67, 112.10, 112.07.
HRMS (m/z): calcd for C23H13BrF3N2O2 485.0107 [M+H]+; found 485.0102.
4.2.5.13 6-Bromo-2,3-di(furan-2-yl)-7-(4-methoxyphenyl)quinoxaline (7m)
The product was obtained as brown solid. Yield 44%. 1H NMR (400 MHz, CDCl3)
δ 8.47 (s, 1H), 8.06 (s, 1H), 7.62 (m, 2H), 7.53–7.43 (m, 2H), 7.08–6.94 (m, 2H), 6.71
(t, J = 3.2 Hz, 2H), 6.57 (m, 2H), 3.88 (s, 3H). 13C NMR (100 MHz, CDCl3) δ 159.63,
150.65, 150.59, 144.46, 144.40, 144.39, 143.07, 143.01, 139.90, 139.53, 132.70,
132.22, 130.80, 130.17, 125.89, 113.52, 113.37, 112.02, 112.00, 55.30. HRMS (m/z):
calcd for C23H16BrN2O3 447.0339 [M+H]+; found 447.0335.
4.2.5.14 6-Bromo-7-(3,5-dimethoxyphenyl)-2,3-di(furan-2-yl)quinoxaline (7n)
The product was obtained as brown solid. Yield 34%. 1H NMR (400 MHz, CDCl3)
δ 8.48 (s, 1H), 8.10 (s, 1H), 7.69–7.55 (m, 2H), 7.26 (s, 1H), 6.72 (t, J = 3.2 Hz, 2H),
6.65 (d, J = 2.2 Hz, 2H), 6.63–6.50 (m, 3H), 3.84 (s, 7H). 13C NMR (100 MHz,
CDCl3) δ 160.36, 150.62, 150.56, 144.62, 144.52, 144.44, 143.21, 143.14, 141.61,
140.06, 139.36, 132.76, 130.20, 125.23, 113.64, 113.45, 112.04, 112.03, 107.75,
100.56, 55.46. HRMS (m/z): calcd for C24H18BrN2O4 477.0518 [M+H]+; found
477.0512.
4.2.5.15 6-Bromo-2,3-di(furan-2-yl)-7-(pyridin-4-yl)quinoxaline (8a)
The product was obtained as brown solid. Yield 30%. 1H NMR (400 MHz, CDCl3)
δ 8.80–8.69 (m, 2H), 8.50 (d, J = 2.1 Hz, 1H), 8.06 (d, J = 2.1 Hz, 1H), 7.64 (dd, J =
5.4, 1.8 Hz, 2H), 7.51–7.40 (m, 2H), 6.76 (dd, J = 5.4, 3.5 Hz, 2H), 6.59 (m, 2H). 13C
NMR (100 MHz, CDCl3) δ 150.44, 150.40, 149.35, 147.83, 144.77, 144.64, 143.72,
143.40, 141.56, 140.50, 139.30, 133.25, 130.45, 124.38, 123.65, 114.09, 113.81,
112.15, 112.11. HRMS (m/z): calcd for C21H13BrN3O2 418.0185 [M+H]+; found
25

ACCEPTED MANUSCRIPT
418.0181.
4.2.5.15 6-(Benzofuran-2-yl)-7-bromo-2,3-di(furan-2-yl)quinoxaline (8b)
The product was obtained as brown solid. Yield 34%. 1H NMR (400 MHz, CDCl3)
δ 8.75 (s, 1H), 8.50 (s, 1H), 7.76 (s, 1H), 7.72–7.61 (m, 3H), 7.55 (d, J = 8.2 Hz, 1H),
7.42–7.34 (m, 1H), 7.31–7.20 (m, 1H), 6.75 (dd, J = 9.1, 3.5 Hz, 2H), 6.58 (dt, J = 3.7,
13
1.8 Hz, 2H). C NMR (100 MHz, CDCl3) δ 154.63, 152.01, 150.61, 150.54, 144.66,
144.50, 143.49, 143.37, 140.04, 139.28, 134.03, 132.37, 129.26, 128.70, 125.59,
123.19, 122.00, 121.75, 113.93, 113.62, 112.10, 112.04, 111.23, 108.64. HRMS (m/z):
calcd for C24H14BrN2O3 457.0117 [M+H]+; found 457.0113.
4.2.5.16 6-(Benzo[b]thiophen-2-yl)-7-bromo-2,3-di(furan-2-yl)quinoxaline (8c)
The product was obtained as brown solid. Yield 32%. 1H NMR (400 MHz, CDCl3)
δ 8.52 (d, J = 1.8 Hz, 1H), 8.32 (d, J = 1.7 Hz, 1H), 7.88 (m, 2H), 7.74–7.57 (m, 3H),
7.53–7.32 (m, 2H), 6.75 (d, J = 3.5 Hz, 2H), 6.59 (dt, J = 4.8, 2.2 Hz, 2H). 13C NMR
(100 MHz, CDCl3) δ 150.51, 150.48, 144.69, 144.56, 143.49, 143.38, 140.50, 140.45,
140.26, 139.71, 139.19, 137.27, 133.40, 131.59, 125.64, 124.94, 124.93, 124.63,
124.17, 122.13, 113.96, 113.67, 112.12, 112.07. HRMS (m/z): calcd for
C24H14BrN2O2S 472.9953 [M+H]+; found 472.9939.
4.2.5.17 6-Bromo-2,3-di(furan-2-yl)-7-(pyrimidin-5-yl)quinoxaline (8d)
The product was obtained as brown solid. Yield 32%. 1H NMR (400 MHz, CDCl3)
δ 9.31 (s, 1H), 8.95 (s, 2H), 8.52 (s, 1H), 8.09 (s, 1H), 7.64 (dd, J = 5.3, 1.7 Hz, 2H),
6.78 (t, J = 3.7 Hz, 2H), 6.59 (td, J = 3.9, 1.8 Hz, 2H). 13C NMR (100 MHz, CDCl3) δ
158.18, 156.82, 150.36, 150.33, 144.86, 144.73, 143.87, 143.50, 140.72, 139.37,
137.28, 133.75, 133.37, 130.87, 124.15, 114.27, 113.97, 112.19, 112.14. HRMS (m/z):
calcd for C20H12BrN4O2 419.0096 [M+H]+; found 419.0102.
4.2.5.18
Tert-butyl-2-(7-bromo-2,3-di(furan-2-yl)quinoxaline-6-yl)-1H-pyrrole-1-carboxylate
(8e)
The product was obtained as brown solid. Yield 32%. 1H NMR (400 MHz, CDCl3)
δ 8.38 (s, 1H), 8.08 (s, 1H), 7.63 (dd, J = 4.4, 1.8 Hz, 2H), 7.48 (dd, J = 3.4, 1.8 Hz,
26

ACCEPTED MANUSCRIPT
1H), 6.73 (dd, J = 12.2, 3.4 Hz, 2H), 6.58 (dd, J = 3.5, 1.8 Hz, 2H), 6.39–6.16 (m, 2H),
1.28 (s, 10H). 13C NMR (100MHz, CDCl3) δ 150.63, 150.54, 148.75, 144.55, 144.39,
143.10, 142.93, 140.35, 139.26, 138.81, 131.46, 131.36, 130.25, 128.37, 122.35,
115.58, 113.67, 113.38, 112.06, 112.03, 110.75, 84.12, 27.49. HRMS (m/z): calcd for
C25H21BrN3O4 506.0697 [M+H]+; found 506.0700.
4.3 Cell culture
HCT116 and MCF7 cancer cell lines were purchased from Shanghai Cell Bank of
Chinese Academy of Sciences (Shanghai, China). The cells were cultured in DMEM
supplemented with 10% FBS, 1% penicillin and streptomycin at 37 °C in a 5% CO2
humidified atmosphere. Cells grown at exponential stage were used for all the
biological experiments.
4.4 Cell viability
Cell viability was assessed by Cell Counting Kit-8 (Beyotime, Jiangsu, China)
following manufacturer’s instructions. Compounds were dissolved in DMSO and
diluted with culture medium. Cells at a density of 5 × 103 per well were seeded in
96-well plates and cultured overnight, which were then treated with either vehicle (1%
DMSO PBS buffer) or desired concentrations of compounds (50, 25, 12.5, 6.25, 3.12,
1.56, 0.78 and 0.39 µM) for 72 h at 37 °C. Cell viability was measured by the CCK-8
kit and the absorbance was measured using a microplate reader (Bio-Rad Laboratories)
at 450 nm. At last, resultant OD450 nm values were expressed as IC50 values, which
were the mean values derived from three independent experiments.
4.5 Cell cycle
Cells were seeded in 6-well plates at density of 1 × 105 cells/well and treated with
different concentrations of compound 7b (5, 10 and 20 µM) for 24 h at 37 °C. Then,
cells were harvested and fixed in 70% precooled ethanol. The cells were then treated
with RNase A, and stained by PI (Product #: C1052, Beyotime, Jiangsu, China).
Finally, the suspended cells were analyzed with a flow cytometer (Accuri C6, BD
Biosciences), a minimum total of 10,000 events were recorded. The data were
27

ACCEPTED MANUSCRIPT
analyzed with the FlowJo.
4.6 Cell apoptosis
HCT116 cells were seeded at a density of 1 × 105 cells/well on each well of 6-well
plates and allowed to grow overnight. Then the cells were treated with 7b (5, 10 and
20 µM) for 24 h, cells without treatment were used as control group. Then the cells
were trypsinized, washed with cold PBS for three times, centrifuged at 1200 rpm for 5
min, and the supernatants were discarded. Then cells were stained by an
Annexin-V-fluorescein isothiocyanate (FITC) kit in the binding buffer for 15 min at
room temperature. Subsequently, the cells were labeled by PI and the apoptotic cells
were measured by a flow cytometer (Accuri C6), data were analyzed with BD Accuri
C6 Plus.
4.7 Western blotting
After treating HCT116 cells with 7b (5, 10 and 20 µM) for 24 h, the total proteins
were extracted and their concentrations were balanced to the same level using BCA
Protein Assay Reagent (Beyotime, Jiangsu, China), followed by 8 min protein
denaturation with SDS loading buffer at 100 °C. Proteins (30 µg) were separated by
sodium dodecyl sulfate-polyacrylamide (SDS-PAGE) electrophoresis and transferred
onto polyvinylidene fluoride (PVDF) membranes, which were blocked with 5%
fat-free dry milk in 1 × Tris-buffered saline (TBS) containing 0.05% Tween 20 for 2 h
at room temperature. Then the membranes were incubated with AKT (Beyotime, no:
AA326), caspase-3 (Beyotime, no: AC030), caspase-8 (Beyotime, no: AC056),
caspase-9 (Beyotime, no: AC062) and β-actin (Beyotime, no: AF0003) antibodies at
4
°C
for
overnight,
followed
by
treatment
with
secondary
horseradish-peroxidase-conjugated anti-rabbitIgG (Beyotime, no: A0208) for 2 h.
Membranes were finally scanned in a ChemiDoc MP Imaging System (Bio-Rad) after
2 min incubation in Clarity Western ECL Substrate (Bio-Rad).
4.8 EC50 of PI3Kα H1047R mutant and wide-type kinases activity
EC50 values of the selected compounds against PI3Kα H1047R mutant were
determined using the HTRF assay as described in the literature [53]. Compounds were
28