Synthesis and anti-gastric cancer activity evaluation of novel triazole nucleobase analogues containing steroidal/coumarin/quinoline moieties

Article abstract of DOI:10.1016/j.ejmech.2019.07.023

A series of novel triazole nucleobase analogues containing steroidal/coumarin/quinoline moieties have been synthesized based on copper-catalyzed azide-alkyne cycloaddition (CuAAC). The anti-cancer activity of the new triazole nucleobase analogues was stud

Full text of DOI:10.1016/j.ejmech.2019.07.023

Accepted Manuscript
Synthesis and anti-gastric cancer activity evaluation of novel triazole nucleobase
analogues containing steroidal/coumarin/quinoline moieties
Jian-Wei Zhao, Zeng-Hui Wu, Jia-Wen Guo, Ming-Jie Huang, Ya-Zhen You, Hong-
Min Liu, Li-Hua Huang
PII:
S0223-5234(19)30644-0
DOI:
https://doi.org/10.1016/j.ejmech.2019.07.023
EJMECH 11520
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 26 February 2019
Revised Date: 4 June 2019
Accepted Date: 7 July 2019
Please cite this article as: J.-W. Zhao, Z.-H. Wu, J.-W. Guo, M.-J. Huang, Y.-Z. You, H.-M. Liu, L.-H.
Huang, Synthesis and anti-gastric cancer activity evaluation of novel triazole nucleobase analogues
containing steroidal/coumarin/quinoline moieties, European Journal of Medicinal Chemistry (2019), doi:
https://doi.org/10.1016/j.ejmech.2019.07.023.
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

ACCEPTED MANUSCRIPT
Synthesisandanti-gastriccanceractivityevaluationofnoveltriazolenucleobase
analoguescontainingsteroidal/coumarin/quinolinemoieties
a, #
c, #
a
a
a
Jian-Wei Zhao , Zeng-Hui Wu , Jia-Wen Guo , Ming-Jie Huang , Ya-Zhen You ,
b,
a,
Hong-Min Liu *, Li-Hua Huang *
a
College of Chemistry and Molecular Engineering, Zhengzhou University, 100 Kexue
Avenue, Zhengzhou, Henan 450001, PR China.
b
School of Pharmaceutical Sciences, Zhengzhou University, Zhengzhou 450001, PR
China; Key Laboratory of Technology of Drug Preparation (Zhengzhou University),
Ministry of Education of China, Zhengzhou University, 100 Kexue Avenue,
Zhengzhou, Henan 450001, PR China.
c
Luoyang Center for Disease Control and Prevention, 9 Zhenghe Road, Luoyang,
Henan 471023, PR China.
ABSTRACT:
A series of novel triazole nucleobase analogues containing steroidal/coumarin/quinoline
moieties have been synthesized based on copper-catalyzed azide-alkyne cycloaddition
(
CuAAC). The anti-cancer activity of the new triazole nucleobase analogues was studied
in gastric cancer cell lines (MGC-803, SGC-7901) and normal gastric epithelial cells
GES-1) in vitro. Some of the synthesized compounds could significantly inhibit the
(
proliferation of these tested cancer cells. Among the tested compounds, compound 20c
demonstrated good anti-proliferation activity against MGC-803 cells (IC = 1.48 µM)
5
0
and SGC-7901 (IC = 2.28 µM) cells as well as the best selectivity between the cancer
5
0
and normal cells. Further mechanistic studies indicated that compound 20c could
down-regulate the expression of TGF β1 both in the tested gastric cancer cell lines and
inhibit the cell migration and invasion. The results of the study indicate that compound

ACCEPTED MANUSCRIPT
2
0c could be used as a promising skeleton for anti-gastric cancer agents with improved
efficacy and less side effects.
Keywords: Hybrid; 1,2,3-Triazole; Nucleobase; Anti-cancer activity; Migration;
Invasion
1
. Introduction
Gastric cancer (GC), a very aggressive tumor, is one of the most common cancers
worldwide, with almost one million new cases diagnosed globally each year, and tolls the
third leading mortality related to cancer [1]. Nevertheless, undesirable drug-induced
toxicities and deleterious side effects owing to insufficient selectivity against healthy
mammalian cells are two main limitations for clinical uses [2]. Therefore, it is urgent to
develop less toxic agents for the treatment of gastric cancer (GC).
Nucleobases, a well-known class of nitrogen-containing heterocyclic compounds,
are basic units of nucleotides which form DNA and RNA [3,4]. Because of their
important roles in diverse fundamental biological processes, nucleobases have been
widely used as pharmacophores, particularly as anticancer pharmacophores in
medicinal chemical research [5-7]. Modified nucleobases are extensively involved in
current clinical treatment for a broad range of human cancers. Many nucleobase
analogues were also reported to exhibit potent cytotoxic activity in vitro against various
human cancer cell lines [7] through different mechanisms. 1,2,3-Triazole scaffold is
another important class of nitrogen containing heterocycle which has been widely
applied as a powerful linker in drug discovery for its convenient synthesis via click
reaction [8]. Moreover, introduction of 1,2,3-triazole moiety is favorable for improving the
solubility [9], enhancing binding affinity to biological targets as well as generating synergistic
biological activities [10]. Thus, various 1,2,3-triazole-nucleobase analogues obtained from
Copper (I) catalyzed azide-alkyne [3+2] cycloaddition (CuAAC) [11,12] have attracted
considerable attention in drug design. A nucleobase-methylene-1,2,3-triazole moiety is
shared in common among those nucleobase analogues to impart diverse biological
activities such as inhibiting diguanylate cyclases (DGCs) targeting the I-sit (compound

ACCEPTED MANUSCRIPT
I) [13] (Fig. 1), anticancer activity (compounds II and III) [14,15] (Fig. 1) and potent
hCA I inhibitors (compounds IV and V) [16] (Fig. 1).
Fig. 1. Structures of nucleobase analogues with a 1,2,3-triazole linker.
In addition, our previous study revealed that 4,7-dichloroisatin-triazole conjugates
VI and VII (Fig. 2) exerted selective inhibition against MGC-803 cells and were less
toxic to normal cells GES-1 [17,18]. Based on the above mentioned findings,
2
,6-dichloropuryl-triazole scaffold, derived from 4,7-dichloroisatin-triazole through
the scaffold hopping strategy, was chosen as a corefor designing new nucleobase analogues
with anti-gastric cancer properties.

ACCEPTED MANUSCRIPT
Fig. 2. Structures of non-nucleobase 1,2,3-triazole tethered molecules
containing privileged moieties.
Steroids are a fundamental class of biological molecules and have continuously attracted
much interest in drug design for their unique rigid frameworks and excellent ability of binding
to specific hormonal receptors [19]. Modifications on C-16 of steroids can lead to structural
variations, whereby cell viability regarding cytotoxic and anticancer compounds co-occurred
[20-24]. Based on these observations, we designed and synthesized a novel series of steroidal
nucleobase analogues 11-13 according to the hybrid approach (Fig. 3).
On the other hand, coumarin and quinoline are benzohetercyclic compounds with
versatile pharmacological activities, and play important roles in discovery of new anticancer
agents. Coumarin [25,26] and quinoline [27-29] derivatives have been reported to have potent
anti-proliferative effect against many types of cancer. Besides, the structure-activity
relationship (SAR) investigations of 4-(1,2,3-triazol-1-yl) coumarin and quinoline derivatives
VIII and IX (Fig. 2) suggested that 1,2,3-triazole nucleus at C-4 position of
coumarin/quinoline was indispensable for antitumoractivity[30,31]. Therefore, for the sake of
the structural diversity and optimizing for better efficacy and lower toxicity, quinoline and
coumarin scaffolds were incorporated into the designed hybrids to replace steroidal skeletons,

ACCEPTED MANUSCRIPT
respectively (Fig. 3).
Cl
Cl
N
N
N
N
H2N
N
N
Cl
N
N
N
H
N
N
O
N
O
N
6
-Chloropurinyl
2-Amino-6-chloropurinyl
N
F
6-Chloro-2-fluoropurinyl
Thymine
Cl
Cl
N
O
Cl
N
N
Cl
N
O
N
Scaffold hopping
Fragment A
N
Fragment B
N
R
N
N
Bioactive
Scaf
N
N
O
O
fold
Previous Work
O
Coumarin
21
HO
N
Dehydroepiandrosterone (DHEA)
O
O
Quinoline
11
20
HO
Hydroepiandrosterone (HEA)
O
Androstane-5α-3, 17-dione
13
12
Fig. 3. Designed nucleobase analogues tethered with a 1,2,3-triazole linker in this work.
The molecular hybridization is a widely used strategy in drug discovery, which forms
new molecular entities by incorporating two or more bioactive substructures through suitable
linkages. It is generally believed that these hybridized molecules possess improved or new
biological properties relative to their individual components [12]. In the present report, as
continuation of our studies on the development of anti-gastric cancer agents, we designed
through the molecular hybridization approach and synthesized a series of 1,4-disubstitued
1
,2,3-triazole-nucleobase analogues containing steroidal/coumarin/quinoline moieties. The
anti-proliferative of those compounds were examined against gastric cancer cells
MGC-803 and SGC-7901 cells). The preliminary structure activity and relationships
SARs) of those compounds were further discussed. Finally, the mechanistic study of
compound 20c was investigated.
(
(
2
. Chemistry

ACCEPTED MANUSCRIPT
The target compounds 11-13 were prepared via the standard CuAAC reactions as shown
in Scheme 1. According to previous literature [24], bromination of 17-oxo steroids 1-2 with
CuBr in methanol under reflux provided 16α-bromo 3-4 in 95-98% yield. Subsequently,
2
16α-bromo-3β-hydroxy-5α-androstan-17-one 4 was oxidized using Jones reagent to afford
16α-bromo-5α-androstan-3,17-dione 5 in 98% yield. Next, azidation of 16α-bromo-17-oxo-
androstanes 3-5 with NaN in DMF gave azido-steroids 6-8 according to the experimental
3
procedures described in the literature [17]. On the other hand, to obtain final products 11-13,
propargyl nucleobases 9a-9e were also prepared via nucleophilic addition of propargyl
bromide with commercially available nucleobases in the presence of potassium carbonates
following the procedure reported by Khunt et al [32]. Finally, Cu-catalyzed
Huisgen-Sharpless-Meldal [3+2] dipolar cycloaddition of 16β-azido steroidal ketones6-8 with
propargyl nucleobases 9a-9e provided the desired products 11-13 in 75-83% yield.

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of target nucleobase analogues 11-13 containing steroidal moiety. Reagents
and conditions: (i) CuBr , methanol, refluxing, 12h, 95-98%; (ii) Jones Reagent, acetone, ice bath,
2
9
8%; (iii) NaN , DMF, rt, 20min, 95%; ( vi ) DMF, K CO , 14h, rt, 62-72%; (v) sodium ascorbate,
3 2 3
CuSO , THF: H O (1:1), 5 h, rt, 75-83%.
4
2
The synthetic route of target nucleobase analogues 20-21 was shown in Scheme 2.
4
-Chloro-bicyclic heterocycle compounds 16 and 18 were prepared from chlorination of
commercially available 4-hydroxycoumarin 14 and 4-hydroxyquinoline 15 using POCl3,
respectively according to the procedure reported by Rui Li et al [30]. Treatment of 16 and 17
with NaN in DMF smoothly afforded the corresponding azides 18 and 19 in good yields.
3
Finally, 1,2,3-triazole tethered nucleobases 20 and 21 were obtained from copper catalyzed
‘‘click” chemistry between the azides (18 and 19) and alkynes 9, respectively, following the
procedure aforementioned.
Scheme 2. Synthesis of target nucleobase analogues 20-21 containing quinoline/coumarin substituents.
Reagents and conditions: (i) POCl , reflux, 5 h, 64–78%; (ii) NaN , DMF, 65℃, 5-8h, 95–98%; (iii)
3
3
NaN , DMF, rt, 8h, 98% (iv) sodium ascorbate, CuSO ·5H O, THF: H O (1:1), 5 h, rt, 64-80%.
3
4
2
2
1
13
Characterization of all the compounds was carried out by means of H, C NMR and
HREI-mass spectra. The detailed physical and analytical data are given in the experimental
part. The structure of compound11a was confirmed by 2D NMR (HSQC, COSY, HMBC and
NOSY) (see the spectra in the supporting information). The structure of compound 20c was
also confirmed by X-ray crystallography (CCDC number: 1888124, please see the X-ray
crystal structure in the supporting information).

ACCEPTED MANUSCRIPT
. Results and discussion
3
3
.1. Anti-proliferative activity and SAR discussion
The anticancer activity of synthesized compounds was evaluated against gastric cancer
cells (MGC-803 and SGC-7901) and non-cancer cell lines (GES-1) with aid of the MTT
method as described in a previous Letter [33]. Their inhibitory effects on cell viability were
evaluated through comparison with those of 5-fluorouracil (5-FU), a well-known anticancer
drug. The preliminary results are listed in Tables 1-5. Based on the obtained data,
structure-activity relationship was analyzed and summarized in Fig. 4.
For steroidal-nucleobase compounds 11-13 (Table 1), we mainly focused on two
aspects to study the structure-activity relationship: (1) For the sake of examining the impact
of the olefinic bond at C-5 position, the carbonylgroup and the hydroxyl group at C-3 position of
steroidal skeleton on the anti-proliferative activity of the hybrids, dehydroepiandrosterone
(DHEA), hydroepiandrosterone (HEA) and androstane-5α-3,17-dione were incorporated into
the designed hybrids, respectively; (2) In order to investigate the influence of different
substituents at the 2′- position of purinyl moiety on anti-proliferative activity, 2,6-dichloro,
6-chloro-2-fluoro, 2-amino-6-chloro and 6-chloropurinyl nucleobase analogues were taken
into account.
As seen in Table 1, 2,6-dichloropurine derivatives 11a and 13a and all of
-chloro-2-fluoropurine derivatives 11c-13c exhibited high activity against SGC-7901 cells
with IC from 1.86µM to 6.12µM. Among them, compounds11a, 13a and 13c displayed the
6
50
higher activity against SGC-7901 cells (IC = 1.86, 2.36 and 2.53 µM) compared to 5-FU.
50
The highest activity was displayed by DHEA nucleobase analogue 11a with IC of 1.86 µM.
50
In the case of MGC-803 cells, compound 13a with IC value of 2.65 µM was found to be the
50
most potent compound and showed lower IC value compared to 5-FU. Compounds 11a, 11c
50
and 13c had slightly lower activity compared to5-FU with IC values 5.99, 6.10 and 4.22µM,
50
respectively. However, compound 12a had a significant decrease in activity against MGC-807

ACCEPTED MANUSCRIPT
cells (IC = 26.53 µM) and was devoid of the activity against SGC-7901 cells (IC > 100
50
50
µM). In addition, compound 12c was less potent than the other corresponding
′-chloro-2′-fluoropurinyl nucleobase derivatives (11c and 13c) against the tested cancer cell
lines with IC value of 10.78 and 6.12 µM, respectively.
6
50
Table 1
In vitro anti-proliferative activity of steroidal nucleobase analogues 11-13 against human gastric
cancer cells.
Compd.
R1
R2
IC (µM)
50
MGC-803
SGC-7901
1
1a
1b
Cl
H
5.99 ± 0.78
> 100
1.86 ± 0.27
1
> 100
1
1c
F
6.10 ± 0.78
> 100
4.06 ± 0.61
> 100
11d
12a
12b
NH
Cl
H
2
2
2
26.53 ± 1.42
4.97 ± 0.69
10.78 ± 1.03
18.48 ± 1.27
2.65 ± 0.42
13.59 ± 1.13
4.22 ± 0.63
> 100
77.51 ± 1.89
6.12 ± 0.79
49.17 ± 1.69
2.36 ± 0.37
10.37 ± 1.02
2.53 ± 0.40
1
2c
F
12d
13a
13b
NH
Cl
H
O
O
1
3c
F
13d
NH
11.37 ± 1.06
3.38 ± 0.53
59.64 ± 1.78
2.96 ± 0.47
5-FU
These results clearly indicate that: (1) to conjoin the steroidal skeleton of
androstane-5α-3,17-dione into 2′,6′-dichloropurine derivative 13a or 6′-chloro-2′-fluoropure

ACCEPTED MANUSCRIPT
derivative 13c can ameliorate or maintain the anti-proliferativeactivity against MGC-807 cells
and SGC-7901 cells. (2) Reduction of the olefinic bond at C-5 position and the retention of
hydroxyl group at C-3 position of steroidal skeleton (compounds 12a and 12c) concurrently
may impair anti-proliferative effects.
On the other hand, replacement of the halogen atom at the 2- position of purinyl moiety
with an amino group led to lower anticancer activity by a magnitude of 3-20 fold
(
compounds 12d and 13d), or even a substantial loss of anti-proliferative effects (compound
1d) against both of the tested cancer cell lines compared to corresponding 2′,6′-dichloro and
′-chloro-2′-fluoropurinyl nucleobase derivatives 11a, 13a, 11c and 13c. Similarly,
1
6
monosubstituted purinyl nucleobase analogues 11b and 13b also deteriorated or
lost the activity against MGC-803 and SGC-7901 cells, except 12b. These results suggest that
2′- halo groups of purinyl moiety such as 2′-chloro (11a and 13a) and 2′-fluoro (11c and 13c)
are favorable for anticancer activity, while introduction of an amino group or hydrogen atom
to C-2′ position of purinyl moiety would dramatically bring down or even counteract
anti-proliferative activity.
Given the better anti-proliferative effects againstthe tested cancer cell lines in comparison
with 5-FU, compound 13a was selected to examine for its toxicity against GES-1 (normal
human gastric epithelial cell line). As seen in Table 2, there was no obvious difference
between compound 13a and 5-FU in terms of selectivity, which means compound13a did not
have selectivity to the tested cancer cell lines over GES-1 cells.
Table 2
In vitro anti-proliferative activity of steroidal nucleobase analogues 13a against non-cancer gastric
cell lines GES-1.
Compd.
R1
R2
IC (µM)
Fold selectivity
50
MGC-803
SGC-7901
GES-1
A
B

ACCEPTED MANUSCRIPT
1
3a
Cl
2.65 ± 0.42
3.38 ± 0.53
2.36 ± 0.37
2.96 ± 0.47
3.11 ± 0.49
3.69 ± 0.89
1.2
1.1
1.3
1.2
5-FU
A= IC50 (GES-1) / IC50 ( MGC-803);
B= IC50 (GES-1) / IC50 (SGC-7901).
For quinoline and coumarin derivatives (Table 3), all of 2,6-dichloropurine and
6-chloro-2-fluoropurine derivatives (20a, 21a, 20c and 21c) exhibited high activity against
MGC-803 and SGC-7901 cells with IC from 1.28 µM to 4.37 µM. Compounds 21a,
5
0
and 20c and 21c displayed the higher activity against the both of tested cancer cells
compared to 5-FU. Among them, the highest activity against MGC-803 and SGC-7901
cells were displayed by quinoline nucleobase analogue 20c (IC = 1.48 µM) and
5
0
coumarin nucleobase derivative 21c (IC = 1.28 µM), respectively. However, 6-dichloropurinyl,
50
2
′-amino-6′-chloropurinyl nucleobase analogues 20b, 21b, 20d and 21d had weak to
moderate or even no activity against the tested cell lines, compared to corresponding
nucleobase analogues 20a, 21a, 20c and 21c. The above observations suggested that the
presence of 2,6-dichloropurine and 6-chloro-2-fluoropurine moiety in the quinoline
nucleobase analogues (20a and 20c) and coumarin nucleobase derivatives (21a and 21c)
could be beneficial for the anticancer activity. On the contrary, anti-proliferative activity
may sharply be weaken resulting from the introduction of 6-dichloropurinyl,
2-amino-6-chloropurinyl nucleobase into nucleobaseanalogues (20b, 21b, 20d and 21d).
Table 3
In vitro anti-proliferative activity of quinoline/coumarin nucleobase analogues 20-21 against human
gastric cancer.

ACCEPTED MANUSCRIPT
Compd.
R1
R2
IC (µM)
50
MGC-803
SGC-7901
2
0a
0b
Cl
H
3.75 ± 0.57
4.37 ± 0.64
2
51.95 ± 1.72
1.48 ± 0.17
20.74 ± 1.32
3.21 ± 0.51
24.69 ± 1.39
2.86 ± 0.46
> 100
21.87 ± 1.34
2.28 ± 0.36
40.22 ± 1.60
1.99 ± 0.30
21.80 ± 1.34
1.28 ± 0.08
> 100
2
0c
F
20d
21a
21b
NH
Cl
H
2
2
1c
1d
-FU
F
2
NH
2
5
3.38 ± 0.53
2.96 ± 0.47
Next, quinoline/ coumarin nucleobase analogues which IC value lower than 5-FU
50
against MGC-803 and SGC-7901 cells were selected to examine for their toxicity against
GES-1. As seen in Table 4, all of these tested compounds 20a, 20c and 21c, exhibit better
selectivity than 5-FU. The quinoline derivative 20c bearing 6′-chloro-2′-fluoropurinyl moiety
had the best selectivity. Based on these findings, it can be presumed that the conjugated
system composed of the 1,2,3-triazole and coumarin or quinoline motifs may enhance
selectivity to the tested cancer cell lines over GES-1 cells.
Notably, compound 20c not only had the best anti-proliferative activityagainst the tested
cancer cells, but also displayed the highest selectivity between the tested cancer cells and
normal cells (around 10.7 and 6.9 fold selectivityto MGC-803 cells and SGC-7901 cells over
GES-1 cells, respectively).
Table 4
In vitro anti-proliferative activity of quinoline/coumarin nucleobase analogues 20c, 21a and 21c
against non-cancer gastric cell lines GES-1.

ACCEPTED MANUSCRIPT
R2
N
N
N
N
N
N
Cl
N
R1
Compd.
R1
R2 IC (µM)
Fold selectivity
50
MGC-803
SGC-7901
GES-1
A
B
2
0c
F
1.48 ± 0.17
2.28 ± 0.36
15.85 ± 1.20
10.7
6.9
2
1a
1c
-FU
Cl
F
3.21 ± 0.51
2.86 ± 0.46
1.99 ± 0.30
1.28 ± 0.08
11.04 ± 1.04
5.86 ± 0.77
3.69 ± 0.89
3.4
2.0
1.1
5.5
4.6
1.2
2
5
3.38 ± 0.53 2.96 ± 0.47
A= IC50 (GES-1) / IC50 (MGC-803);
B= IC50 (GES-1) / IC50 (SGC-7901).
Additionally, previous investigation of our group [34,35] demonstrated that
1
,2,3-triazole-pyrimidine hybrids displayed high activities against the cancer cell proliferation.
Furthermore, with the aim of evaluating the contribution of purinyl group to the
anti-proliferative activity and increasing the structural diversity, thymine, a
pyrimidine nucleobase as a vital building unit of DNA and RNA, was incorporated into the
designed hybrids (11e-13e). Unfortunately, most of pyrimidine derivatives (11e and 13e)
exhibited weak or even no activity against the tested cell lines, except for compound 12e
(
IC = 4.37 µM). It is obvious that introduction of the thymine group in the designed
50
hybrids led to a significant reduction in the anti-proliferative effects or even entirely caused
loss of the activity against the tested cancer cellliners, which was similar to the introduction of
6-dichloro or 2′-amino-6′-chloropurinyl into the designed nucleobase analogues (11b-13b or
11d-13d).
Table 5
In vitro anti-proliferative activity of pyrimidine nucleobase analogues 11e-13e and 20e-21e against
human gastric cancer.

ACCEPTED MANUSCRIPT
Compd.
R
IC (µM)
5
0
MGC-803
SGC-7901
1
1
1
2
2
1e
2e
3e
0e
1e
> 100
> 100
33.78 ± 1.53
67.08 ± 1.83
> 100
4.37 ± 0.64
85.58 ± 1.93
> 100
> 100
31.99 ± 1.5
2.96 ± 0.47
5-FU
3.38 ± 0.53
Based on above analysis of the structure-activity relationship, we can conclude that the R2
group attached to the 2-position of the purine scaffold are important for the activity. The chlorine
and fluorine atoms are beneficial for the anticancer activity, while the hydrogen and amino groups
havebeen found to bedetrimental to theactivity.Replacement ofthepurinescaffold with thymine
ring also leads to the decrease of the anticancer activity. For the R group, we have found that the
1
heterocyclic coumarin and quinoline skeletons are preferred for the activity over the steroidal
nucleus in loweringthetoxicityagainst normal cells (Fig. 4).
In view of an effective growth inhibitory effect on both of the tested cancer cells and
obvious difference in viability between the tested cancer cells and the normal cells, compound
20c was chosen to further evaluate its possible anticancer mechanism of action against
MGC-803 and SGC-7901 cells in all the subsequent experiments.

ACCEPTED MANUSCRIPT
Fig. 4. Summary of the structure-activity relationship.
3.2. Migration and invasion inhibition of compound 20c against MGC-803 and SGC-7901
We explored the effect of compound 20c on the migration ability of MGC-803 and
SGC-7901. Initially, the cell migration ability was evaluated with wound healing assay.
As shown in Fig. 5A, compound 20c significantly inhibited migration of MGC-803 cells
in a concentration-dependent manner, even at 0.1 µM. The cell migration can be
completely blocked by compound 20c at 0.2 µM or 0.4 µM. Besides, SGC-7901 cells
treated with compound 20c also performed weak migration ability dose dependently as
shown in Fig. 5B. With the treatment of compound 20c as low as 0.4 µM, the cell
migration can be blocked fully.

ACCEPTED MANUSCRIPT
Fig. 5. Inhibitory effect of compound 20c on MGC-803 (A) and SGC-7901 (B) cell migration with
wound healing assay for 24h treatment.
To further quantify the cell migration inhibitory effect of compound 20c, transwell
experiment coupled with high content analysis was applied. For transwell experiment,
different concentrations of compound 20c were applied for 24 h in MGC-803 and SGC-7901
cells. As shown in Fig. 6A & B, such lower concentrations of compound 20c can inhibit the
migration of both MGC-803 (Fig. 6A) and SGC-7901 (Fig. 6B) cells does-dependently. As
cell invasion is a key step in cell migration from primary carcinoma to metastasis. Transwell
coated with matrigel was applied to evaluate the cell invasion inhibitory effect of compound
20c by incorporating with high content analysis. For this experiment, cells were treated with
indicated concentrations of compound 20c for 24 h either (Fig. 6C & D), the results indicated
that compound 20c may suppress MGC-803 (Fig. 6C) and SGC-7901 (Fig. 6D) cells invasion
dose-dependently. All these results indicate that compound 20c may inhibit gastric cancer cell
MGC-803 and SGC-7901 migration and invasion.

ACCEPTED MANUSCRIPT
Fig. 6. Inhibitory effect of compound 20c on MGC-803 (A & C) and SGC-7901 (B & D) cell
migration and invasion. (A & B) Migration ability of MGC-803 (A) and SGC-7901 (B) cells with
2
4 h treatment of compound 20c at indicated concentration with the aid of transwell experiment; (C
D) Invasion ability of MGC-803 (C) and SGC-7901 (D) cells with 24 h treatment of compound
0c at indicated concentration with the aid of matrigel coated transwell experiment. All experiments
were performed three times at least. **: p < 0.01.
&
2
3
.3. Mechanism study for the inhibitor effect of compound 20c against cell migration
and invasion.
TGF β1 (Transforming growth factor β1) is a polypeptide member of the transforming
growth factor β superfamily of cytokines. It was first identified in human platelets as a protein
with a molecular mass of 25 kilodaltons with a potential role in wound healing [36]. As
reported, TGF β1 is a secreted protein that performs many cellular functions, including the
control of cell growth, cell proliferation, cell differentiation, and apoptosis. In gastric cancer,
TGF β1 was overexpressed and closely related to the invasion and metastasis of gastric cancer
[37, 38]. Hence, amount of TGFβ1 was evaluated in the cell culture medium using Elisa when

ACCEPTED MANUSCRIPT
cells were treated with compound 20c for 24 h. As can be seen in Fig. 7A & D, when
compound 20c was applied to MGC-803 and SGC-7901 cells, respectively, the amount of
TGF β1 in cell culture medium was sharply decreased in both MGC-803 (Fig. 7A) and SGC-7901
(Fig.7D)cells.AsTGFβ1isthewell-knowninducerofEMT(epithelial-mesenchymal transition),
which may contribute to cell migration and invasion [39], several biomarkers of EMT were
evaluated by RT-qPCR for the cells treated with compound 20c. As indicated in Fig. 7, we
found that 24h treatment of compound 20c led to the accumulation of epithelial cell marker
E-Cadherin mRNA (Fig. 7B for MGC-803, Fig. 7E for SGC-7901) and decreasing amount of
mesenchymal cell markerN-Cadherin mRNA (Fig. 7C for MGC-803, Fig. 7F for SGC-7901)
dose-dependently in both MGC-803 and SGC-7901 cells. All these results indicate that
compound 20c may reverse EMT by decreasing the expression of TGF β1 and inhibit cell
migration and invasion in cellular level.
Fig. 7. Relative amount of TGF β1 protein, E-Cadherin mRNA and N-Cadherin mRNA in MGC-803
and SGC-7901 with indicated treatment. (A-C) Relative amount of TGF β1 in MGC-803 cell culture
medium (A), and the amount of E-Cadherin mRNA (B) and (C) N-Cadherin mRNA in MGC-803
with indicated treatment; (D-F) Relative amount of TGF β1 in SGC-7901 cell culture medium (D),
and the amount of E-Cadherin mRNA (E) and (F) N-Cadherin mRNA in SGC-7901 with indicated
treatment. All experiments were performed three times at least. **: p < 0.01.

ACCEPTED MANUSCRIPT
4
. Conclusion
In summary, a series of novel nucleobase analogues containing steroidal/coumarin/quinoline
moieties via 1,2,3-triazole linker were prepared according to the molecular hybridization
design approach. Among these compounds, in comparison of 5-FU, compounds 20c, 21a
and 21c showed better activity against gastric cancer cell lines (MGC-803 and SGC-7901)
and lower cytotoxicity against non-cancer cell lines (GES-1). Moreover, compound 20c
not only had the best anti-proliferative activity against MGC-803 cells and SGC-7901
cells with IC of 1.48 and 2.28 µM, respectively, but also it displayed the highest
50
selectivity between the tested cancer cells and normal cells (10.7-fold selectivity to the
MGC-803 cancer cells over the GES-1 normal cells and 6.9-fold selectivity to the
SGC-7901 cells over the GES-1 normal cells, respectively). The further mechanism
study demonstrated that compound 20c can down regulate the expression of TGF β1 in
both MGC-803 and SGC-7901, and inhibit the cell migration and invasion.
In addition, preliminary structure-activity relationship research revealed that
incorporation of quinoline to the hybrids was the best optimal for bioactivity;
coumarin/quinoline skeletons at N-1 position of 1,2,3-triazole core can make positive
contribution to an enhancement of activity and a decrease in toxicity; chlorine atom and
fluorine atom at C-2 position of purinyl moiety played a vital role to strengthen
anticancer activity. All of these findings suggest that the quinoline nucleobase analogue
20c bearing 6′-chloro-2′-fluoropurinyl moiety is a promising skeleton for the
development of anti-gastric cancer agents with improved efficacy and fewer side effects.
Relevant optimization study is currently underway in our laboratory.
5. Experimental
5
.1. General methods
1
Proton nuclear magnetic resonance spectra ( H NMR) and carbon nuclear
1
3
magnetic resonance spectra ( C NMR) were recorded in DMSO-d on a spectrometers
6
1
13
working at 600 MHz or 400 MHz for H, 150 MHz or 100 MHz for C.
1
13
Tetramethylsilane (TMS) for H (600 MHz or 400 MHz) and C NMR (150 MHz or
00 MHz) was used as an internal standard. Chemical shift values are reported in ppm
1

ACCEPTED MANUSCRIPT
relative to TMS as an internal reference or residual solvent peak. Data are represented
as follows: chemical shift, multiplicity (s, singlet; d, doublet; m, multiplet; q, quartet),
coupling constant in hertz (Hz). High resolution mass spectra (HRMS) data were
TM
measured on Micromass Q-TOF Micro
(Waters, USA) by means of the ESI
+
technique, with accurate masses reported for the molecular ion [M+H] . The melting
points of these compounds were determined by an X-4 micro-melting point apparatus
(Beijing, China). All chemicals and anhydrous solvents were purchased from
commercial sources and used without further purification. All reactions were carried
out in an open air, monitored by thin layer chromatography (TLC). All products were
purified by flash chromatography. The data of 20c was collected on a Bruker APEX-III
CCD diffractometer at room temperature using Mo Kα radiation (λ = 0.71073 Å). The
structure was resolved and refined using the Olex2 software [33,34]. All non-H atoms were
refined with anisotropic thermal parameters and the H atoms of carbon or nitrogen
molecules were located in geometrically constrained positions and refined by riding.
5
.2. Synthesis of 16α-bromo-androstan-3β-ol-5-en-17-one (3)
A mixture of dehydroepiandrosterone 1 (5.00 g, 17.40 mmol) and CuBr (11.50 g,
2
5
1.60 mmol, 3.00 eq.) in methanol (100 mL) was stirred at 70 ℃for 12 h and monitored
by TLC. After cooling, the inorganic salt was filtered off. The filtrate was evaporated in
vacuo. The residue was dissolved in DCM (100 mL), then washed with brine (20 mL × 3).
The organic phase was separated and dried over anhydrous Na SO . Then the solvent was
2
4
evaporated in vacuo and the crude product was purified by flash column chromatography to
give the title compound 3 as a white solid (6.20 g, 97%).
Compound 4 was obtained from epiandrosterone 2 by using the same method that
described above.
5
.3. Synthesis of 16α-bromo-5α-androstan-3,17-dione (5)
To a solution of 5α-androstan-16α-bromo-3β-ol-17-one 4 (6.20 g, 16.80 mmol) in
acetone (120 mL), Jones reagent (9.4 mL, 18.50 mmol) was added dropwise. The
mixture was stirred at 0 ℃for 1 h and monitored by TLC. The reaction was quenched
with ethanol (2 mL) and concentrated in vacuo. The residue was dissolved in DCM (150

ACCEPTED MANUSCRIPT
mL), and then washed with brine (20 mL × 3). The organic phase was separated and dried
over anhydrous Na SO . Then the solvent was evaporated in vacuo and the crude product was
2
4
purified by flash column chromatography to give thetitle compound 5 as a white solid (6.10 g,
98%).
5.3. Synthesis of 16β-azido-androstan-3β-ol-5-en-17-one (6)
Sodium azide (3.25 g, 50.00 mmol) was added to a solution of compound 3 (6.20 g,
1
6.60 mmol) in DMF (100 mL), and the mixture was stirred at room temperature for 2 h
and monitored by TLC. The mixture was quenched with H O (100 mL) and extracted
2
with EtOAc (150 mL × 3). The combined organic layers were washed with brine (40
mL × 3). The organic phase was separated and dried over anhydrous Na SO . Then the
2
4
solvent was evaporated in vacuo and the crude product was purified by flash column
chromatography to give the title compound 6 as a white solid (5.28 g, 95%).
Compounds 7 and 8 were obtained from compounds 4 and 5, respectively, by
using the same method that described above.
5
.5. Synthesis of 2,6-dichloro-9-(prop-2-yn-1-yl)-9H-purine (9a)
To a stirred solution of 2,6-dichloro-9H-purine (5.00 g, 26.40 mmol) in DMF (100
mL) were added 3-bromopropyne 10 (3.77 g, 31.70 mmol, 1.20 eq.) and potassium
carbonate (5.52 g, 40.00 mmol, 1.50 eq.) successively. The mixture was stirred
overnight at room temperature and monitored by TLC, then was quenched with H O
2
(200 mL) and extracted with EtOAc (100 mL × 3). The combined organic layers were
washed with brine (30 mL × 3). The organic phase was separated and dried over anhydrous
Na SO . Then the solvent was evaporatedin vacuo and the crude product was purified byflash
2
4
column chromatography to give the title compound 9a as a white solid (4.25 g, 72%).
Compounds 9b-9e were obtained from 6-Chloropurine, 6-Chloro-2-fluoropurine,
2-amino-6-chloropurine, 5-methyl-2,4-dioxypyrimidine, respectively, by using the same
method that described above.
5
.6. General procedure for preparation of target compounds (11a-13e)

ACCEPTED MANUSCRIPT
CuSO ·5H O (37.50 mg, 0.15 mmol, 0.05 eq.) in 20 mL of water and sodium
4
2
ascorbate (120.00 mg, 0.60 mmol, 0.20 eq.) were added to a solution of corresponding
azide steroidal compounds (6, 7 and 8) (3.00 mmol) and alkyne compounds 9a-9e (3.00
mmol) in tetrahydrofuran (20 mL). The mixture was stirred at room temperature for 5 h
and monitored by TLC. Then the mixture was concentrated in vacuo to remove
tetrahydrofuran and then extracted with DCM (50 mL × 3). The combined organic
layers were washed with brine (10 mL × 3). The organic phase was separated and dried
over anhydrous Na SO . Then the solvent was evaporated in vacuo and the crude product was
2
4
purified by flash column chromatography to give the target compounds.
5
.6.1. 16β-(4-((2,6-dichloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-androstan-
3
β-ol-5-en-17-one (11a)
1
Yield 73%; a white solid; m.p. 206-208 ºC; H NMR (400 MHz, CDCl ) δ 8.37 (s,
3
1
H, 8′′-H), 7.85 (s, 1H, 5′-H), 5.54 (dd, 2H, J = 17.2, 15.4 Hz, -CH ), 5.39-5.38 (m, 1H,
2
6
1
1
3
-H), 4.71 (dd, 1H, J = 10.3, 8.4 Hz, 16-H), 3.58-3.50 (m, 1H, 3-H), 1.10 (s, 3H, 19-CH3),
13
.06 (s, 3H, 18-CH ); C NMR (100 MHz, CDCl ) δ 211.0, 153.0, 152.8, 151.9, 145.9,
3
3
41.2, 140.6, 130.7, 124.1, 120.4, 71.5, 64.9, 50.2, 47.5, 47.5, 42.1, 39.0, 37.1, 36.7,
1.7, 31.5, 30.8, 30.7, 29.4, 20.1, 19.4, 14.3; HRMS (ESI) m/z calcd for C H Cl N O
2
2
7
31
2
7
+
[
M+H] 556.1995, found 556.1993.
5
.6.2. 16β-(4-((6-dichloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-androstan-
3
β-ol-5-en-17-one (11b)
1
Yield 81%; a white solid; m.p. 155-157 ºC; H NMR (400 MHz, DMSO-d ) δ 8.81
6
(
2H, 8′′-H, 2′′-H), 8.23 (s, 1H, 5′-H), 5.64 (s, 2H, -CH ), 5.30 (s, 1H, 6-H), 5.24 (t, 1H, J = 9.0
2
13
Hz, 16-H), 1.02 (s, 3H, 19-CH ), 0.99 (s, 3H, 18-CH ); C NMR (100 MHz, DMSO-d )
3
3
6
δ 212.0, 151.7, 149.1, 147.4, 141.5, 141.5, 130.7, 124.7, 119.7, 69.9, 64.3, 49.7, 46.6,
4
6.0, 42.1, 38.8, 36.8, 36.3, 31.3, 31.2, 30.3, 30.1, 29.7, 19.7, 19.1, 14.0; HRMS (ESI)
+
m/z calcd for C H ClN O [M+H] 522.2384; found 522.2385.
2
7
32
7
2
5
.6.3. 16β-(4-((6-chloro-2-fluoro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-androsta-
β-ol-5-en-17-one (11c)
3

ACCEPTED MANUSCRIPT
1
Yield 72%; a white solid; m.p. 176-178 ºC; H NMR (400 MHz, CDCl ) δ 8.36 (s,
3
1
H, 8′′-H), 7.86 (s, 1H, 5′-H), 5.52 (dd, 2H, J = 16.0, 5.6 Hz, -CH ), 5.39-5.38 (m, 1H, 6-H),
2
4
.71 (dd, 1H, J = 10.2, 8.5 Hz, 16-H), 3.57-3.50 (m, 1H, 3-H), 1.10 (s, 3H, 19-CH ), 1.06
3
1
3
(
s, 3H, 18-CH ); C NMR (100 MHz, CDCl ) δ 210.9, 157.3 (d, J = 219 Hz), 153.3
3 3 FC
(
d, J = 16 Hz), 152.8 (d, J = 17 Hz), 145.9 (d, J = 4 Hz), 141.2, 140.6, 130.2 (d, J
FC FC FC FC
=
5 Hz), 124.1, 120.4, 71.5, 64.9, 50.2, 47.6, 47.5, 42.1, 39.0, 37.1, 36.7, 31.7, 31.5,
+
3
0.9, 30.7, 29.4, 20.1, 19.4, 14.4; HRMS (ESI) m/z calcd for C H ClFN O [M+H]
27 31 7 2
5
40.2290; found 540.2291.
5
1
.6.4.
6β-(4-((2-amino-6-chloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-androstan-
3
β-ol-5-en-17-one (11d)
1
Yield 79%; a white solid; m.p. 191-193 ºC; H NMR (400 MHz, DMSO-d ) δ 8.20
6
(
s, 1H, 8′′-H), 8.12 (s, 1H, 5′-H), 6.96 (s, 2H, N-H), 5.37 (s, 2H, -CH ), 5.30 (s, 1-H, 6-H),
2
5
.23 (dd, 1H, J = 10.1, 8.6 Hz, 16-H), 3.30-3.24 (m, 1-H, 3-H), 1.03 (s, 3H, 19-CH ), 0.99
3
1
3
(
s, 3H, 18-CH ); C NMR (100 MHz, DMSO-d ) δ 212.0, 159.9, 153.9, 149.4, 143.0,
3
6
1
3
42.1, 141.5, 124.4, 123.1, 119.8, 69.9, 64.2, 49.7, 46.7, 46.0, 42.1, 38.1, 36.8, 36.3,
1.4, 31.2, 30.3, 30.1, 29.7, 19.8, 19.1, 14.0; HRMS (ESI) m/z calcd for C H ClN O
2
2
7
33
8
+
[
M+H] 537.2493, found 537.2492.
5.6.5. 16β-(4-((5-methyl-pyrimidine-2,4-dione-1-yl)methyl)-1H-1,2,3-triazol-1-yl)-androstan-
3
β-ol-5-en-17-one (11e)
1
Yield 78%; a white solid; m.p. 179-181 ºC; H NMR (400 MHz, DMSO-d ) δ
6
1
1.30 (s, 1H, N-H), 8.11 (s, 1H, 5′-H), 7.64 (s, 1H, 6′′-H), 5.31 (s, 1H, 6-H), 5.25 (t, 1H, J =
1
3
1
3
8.6 Hz, 16-H), 4.90 (s, 2H, -CH ), 3.28-3.24 (m, 1H, 3-H), 1.76 (s, 3H, 5′′-CH ), 1.04 (s,
2
3
1
3
H, 19-CH ), 1.00 (s, 3H, 18-CH ); C NMR (100 MHz, DMSO-d ) δ 212.1, 164.2,
3
3
6
50.7, 142.3, 141.5 141.2, 124.5, 119.8, 108.8, 69.9, 64.2, 49.7, 46.7, 46.0, 42.1, 36.8,
6.3, 31.4, 31.2, 30.3, 30.1, 29.8, 19.8, 19.1, 14.0, 13.4, 11.9; HRMS (ESI) m/z calcd
+
for C H N O [M+H] 494.2767, found 494.2767.
2
7
35
5
4
5
.6.6. 16β-(4-((2,6-dichloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-5α-androstan-
β-ol-17-one (12a)
3

ACCEPTED MANUSCRIPT
1
Yield 76%; a white solid; m.p. 248-250 ºC; H NMR (400 MHz, CDCl ) δ 8.35 (s,
3
1
1
H, 8′′-H), 7.83 (s, 1H, 5′-H), 5.54 (dd, 2H, J = 17.5, 15.5 Hz, -CH ), 4.67 (dd, 1H, J =
2
1
3
0.2, 8.4 Hz, 16-H), 1.07 (s, 3H, 19-CH ), 0.86 (s, 3H, 18-CH ); C NMR (100 MHz,
3
3
CDCl ) δ 211.1, 153.0, 152.8, 152.0, 145.8, 140.6, 130.7, 124.0, 71.1, 65.0, 54.5, 47.9,
3
4
7.2, 44.8, 39.0, 38.0, 36.8, 35.7, 34.2, 31.8, 31.4, 31.0, 29.3, 28.2, 20.3, 14.6, 12.3;
+
HRMS (ESI) m/z calcd for C H Cl N O [M+H] 558.2151, found 558.2025.
2
7
33
2
7
2
5
.6.7. 16β-(4-((6-dichloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-5α-androstan-
3
β-ol-17-one (12b)
1
Yield 72%; a white solid; m.p. 166-168 ºC; H NMR (400 MHz, CDCl ) δ 8.76 (s,
3
1
H, 2′′-H), 8.37 (s, 1H, 8′′-H), 7.84 (s, 1H, 5′-H), 5.59 (dd, 2H, J = 17.7, 15.5 Hz, -CH ), 4.67
2
(
dd, 1H, J = 10.2, 8.4 Hz, 16-H), 3.65-3.58 (m, 1H, 3-H), 1.10 (s, 3H, 19-CH ), 1.07 (s,
3
1
3
3
1
3
5
H, 18-CH ); C NMR (100 MHz, CDCl ) δ 211.2, 152.0, 151.5, 151.2, 145.3, 141.1,
3
3
31.5, 123.9, 71.0, 65.0, 54.5, 47.9, 47.2, 44.8, 39.0, 38.0, 36.9, 35.7, 34.2, 31.8, 31.4,
+
1.0, 29.3, 28.2, 20.3, 14.6, 12.3; HRMS (ESI) m/z calcd for C H ClN O [M+H]
27
34
7
2
24.2541; found 524.2543.
5
.6.8. 16β-(4-((6-chloro-2-fluoro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-5α-androstan-
3
β-ol-17-one (12c)
1
Yield 79%; a white solid; m.p. 201-203 ºC; H NMR (400 MHz, DMSO-d ) δ 8.81
6
(
s, 1H, 8′′-H), 8.22 (s, 1H, 5′-H), 5.58 (s, 2H, -CH ), 5.24 (dd, 1H, J = 10.1, 8.5 Hz, 16-H),
2
1
3
1
1
1
3
.00 (s, 3H, 19-CH ), 0.79 (s, 3H, 18-CH ); C NMR (100 MHz, DMSO-d ) δ 212.1,
3 3 6
56.3 (d, J = 213 Hz), 153.7 (d, J = 16 Hz), 150.4 (d, J = 18 Hz), 148.3, 141.2,
FC
FC
FC
29.9 (d, J = 5 Hz), 124.8, 69.2, 64.3, 53.9, 47.0, 45.7, 44.3, 38.0, 36.5, 35.3, 33.6,
FC
1.3, 31.3, 30.6, 29.6, 28.0, 19.9, 14.2, 12.0; HRMS (ESI) m/z calcd for
+
C H ClFN O [M+H] 542.2447; found 542.2446.
27
33
7
2
5
.6.9. 16β-(4-((2-amino-6-chloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-5α-androstan
3β-ol-17-one (12d)
Yield 77%; a white solid; m.p. 218-220 ºC; H NMR (400 MHz, DMSO-d ) δ 8.21
-
1
6
(
s, 1H, 8′′-H), 8.11 (s, 1H, 5′-H), 6.98 (s, 2H, N-H), 5.37 (s, 2H, -CH ), 5.22 (dd, 1H, J = 10.2,
2

ACCEPTED MANUSCRIPT
8
.5 Hz, 16-H), 4.45 (br, 1H, O-H), 3.38-3.32 (m, 1H, 3-H), 1.00 (s, 3H, 19-CH ), 0.79 (s,
3
1
3
3
1
3
H, 18-CH ); C NMR (100 MHz, DMSO-d ) δ 212.2, 159.9, 153.9, 149.4, 143.0,
3 6
42.1, 124.4, 123.1, 69.2, 64.3, 53.8, 47.0, 45.8, 44.3, 38.1, 38.1, 36.5, 35.3, 33.6, 31.3,
1.3, 30.6, 29.6, 28.0, 19.9, 14.3, 12.0; HRMS (ESI) m/z calcd for C H ClN O
2
2
7
35
8
+
[
M+H] 539.2650, found 539.2651.
5
.6.10. 16β-(4-((5-methyl-pyrimidine-2,4-dione-1-yl)methyl)-1H-1,2,3-triazol-1-yl)-5α-androstan-
3
β-ol-17-one (12e)
1
Yield 84%; a white solid; m.p. 163-165 ºC; H NMR (400 MHz, CDCl ) δ 9.66 (s,
3
1H, N-H), 7.84 (s, 1H, 5′-H), 7.36 (d, 1H, J= 1.2 Hz, 6′′-H), 4.96 (dd, 2H, J = 39.2, 15.0 Hz,
-
CH ), 4.74 (dd, 1H, J = 10.3, 8.4 Hz, 16-H), 3.65-3.57 (m, 1H, 3-H), 1.89 (d, 3H, J = 0.9
2
1
3
Hz, 5′′-CH ), 1.10 (s, 3H, 19-CH ), 0.86 (s, 3H, 18-CH ); C NMR (100 MHz, CDCl )
3
3
3
3
δ 211.4, 164.4, 151.2, 141.8, 140.3, 124.8, 111.3, 71.0, 65.0, 54.2, 50.7, 47.8, 47.2, 44.8,
4
2.9, 37.9, 36.9, 35.7, 34.2, 31.9, 31.3, 31.1, 29.7, 28.2, 20.3, 14.6, 12.3; HRMS (ESI)
+
m/z calcd for C H N O [M+H] 496.2924, found 496.2923.
2
7
37
5
4
5
3
.6.11. 16β-(4-((2,6-dichloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-5α-androstan-
,17-dione (13a)
1
Yield 75%; a white solid; m.p. 242-244 ºC; H NMR (400 MHz, CDCl ) δ 8.36 (s,
3
1
1
H, 8′′-H), 7.85 (s, 1H, 5′-H), 5.54 (dd, 2H, J = 17.3, 15.5 Hz, -CH ), 4.69 (dd, 1H, J =
2
1
3
0.3, 8.4 Hz, 16-H), 1.10 (s, 3H, 19-CH ), 1.07 (s, 3H, 18-CH ); C NMR (100 MHz,
3
3
CDCl ) δ 211.2, 210.8, 153.0, 152.8, 151.9, 145.9, 140.6, 130.7, 124.1, 64.9, 53.9, 47.8,
3
4
7.0, 46.5, 44.5, 39.0, 38.4, 38.0, 35.9, 34.2, 31.8, 30.7, 29.3, 28.4, 20.5,14.6, 11.5;
+
HRMS (ESI) m/z calcd for C H Cl N O [M+H] 556.1995, found 556.1993.
2
7
31
2
7
2
5
.6.12. 16β-(4-((6-chloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-5α-androstan-
3
,17-dione (13b)
1
Yield 75%; a white solid; m.p. 134-135 ºC; H NMR (400 MHz, CDCl ) δ 8.77 (s,
3
1
H, 2′′-H), 8.36 (s, 1H, 8′′-H), 7.84 (s, 1H, 5′-H), 5.59 (dd, 2H, J = 18.0, 15.4 Hz, -CH ), 4.67
2
1
3
(
dd, 1H, J = 10.2, 8.4 Hz, 16-H), 1.10 (s, 3H, 19-CH ), 1.07 (s, 3H, 18-CH ); C NMR
3 3
(
100 MHz, CDCl ) δ 211.2, 210.8, 152.0, 151.5, 151.2, 145.2, 141.1, 131.6, 124.0, 64.9,
3

ACCEPTED MANUSCRIPT
5
1
5
3.9, 47.8, 47.0, 46.5, 44.5, 39.0, 38.3, 38.0, 35.9, 34.1, 31.8, 30.7, 29.2, 28.4, 20.5,
+
4.6, 11.5; HRMS (ESI) m/z calcd for C H ClN O [M+H] 522.2384; found
2
7
32
7
2
22.2383.
5
3
.6.13. 16β-(4-((6-chloro-2-fluoro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-5α-androstan-
,17-dione (13c)
1
Yield 76%; a white solid; m.p. 196-198 ºC; H NMR (400 MHz, CDCl ) δ 8.35 (s,
3
1
H, 8′′-H), 7.86 (s, 1H, 5′-H), 5.52 (dd, 2H, J = 16.0, 5.3 Hz, -CH ), 4.70 (dd, 1H, J = 10.2,
2
1
3
8
.4 Hz, 16-H), 1.10 (s, 3H, 19-CH ), 1.07 (s, 3H, 18-CH ); C NMR (100 MHz, CDCl )
3 3 3
δ 211.3, 210.8, 157.3 (d, J = 219 Hz), 153.1 (d, J = 55 Hz), 153.0 (d, J = 55 Hz),
FC
FC
FC
1
4
45.9 (d, J = 4 Hz), 140.6, 130.2 (d, J = 5 Hz), 124.1, 64.9, 53.9, 47.8, 47.0, 46.5,
FC FC
4.5, 39.0, 38.3, 38.0, 35.9, 34.1, 31.8, 30.7, 29.3, 28.4, 20.5, 14.6, 11.5; HRMS (ESI)
+
m/z calcd for C H ClFN O [M+H] 540.2290; found 540.2289.
2
7
31
7
2
5
.6.14. 16β-(4-((2-amino-6-chloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)-5α-androstan-
3
,17-dione (13d)
1
Yield 74%; a white solid; m.p. 254-256 ºC; H NMR (400 MHz, DMSO-d ) δ
6
8
.22 (s, 1H, 8′′-H), 8.13 (s, 1H, 5′-H), 6.98 (s, 2H, N-H), 5.38 (s, 2H, -CH ), 5.23 (dd, 1H, J
2
1
3
=
10.2, 8.5 Hz, 16-H), 1.03 (s, 3H, 19-CH ), 1.01 (s, 3H, 18-CH ); C NMR (100 MHz,
3 3
DMSO-d ) δ 212.1, 210.3, 159.9, 153.9, 149.4, 143.1, 142.1, 124.4, 123.2, 64.3, 53.1,
6
4
1
6.9, 45.8, 45.6, 44.1, 38.1, 37.7, 37.6, 35.4, 33.4, 31.3, 30.2, 29.6, 28.0, 20.1, 14.3,
+
1.0; HRMS (ESI) m/z calcd for C H ClN O [M+H] 537.2493, found 537.2492.
2
7
33
8
2
5
.6.15. 16β-(4-((5-methyl-pyrimidine-2,4-dione-1-yl)methyl)-1H-1,2,3-triazol-1-yl)-5α-androstan-
3
,17-dione (13e)
1
Yield 83%; a white solid; m.p. 186-188 ºC; H NMR (400 MHz, CDCl ) δ 9.67 (s,
3
1H, N-H), 7.84 (s, 1H, 5′-H), 7.36 (d, 1H, J= 1.2 Hz, 6′′-H), 4.96 (dd, 2H, J = 39.2, 15.0 Hz ,
-
CH ), 4.74 (dd, 1H, J = 10.3, 8.4 Hz, 16-H), 1.89 (d, 3H, J = 0.9 Hz, 5′′-CH ), 1.10 (s, 3H,
2
3
1
3
1
1
9-CH ), 0.86 (s, 3H, 18-CH ); C NMR (100 MHz, CDCl ) δ 211.3, 211.0, 164.2,
3 3 3
51.1, 141.8, 140.2, 124.8, 111.3, 64.9, 53.9, 47.7, 47.0, 46.5, 44.5, 42.9, 38.4, 38.0,

ACCEPTED MANUSCRIPT
3
5.9, 34.1, 31.8, 30.7, 29.6, 28.4, 20.5, 14.6, 12.3, 11.5; HRMS (ESI) m/z calcd for
+
C H N O [M+H] 494.2767, found 494.2766.
27
35
5
4
5
.7. Synthesis of 4-chloroquinoline (16)
POCl (7 mL) was added to 4-hydroxyquinoline (2.00 g, 13.79 mmol) in a dry
3
round bottom flask. After refluxed for 5 h, the reaction mixture was cooled to room
temperature and then slowly added into ice. The solution was neutralized with solid
sodium bicarbonate in ice bath, and then extracted with DCM (30 mL × 3). The
combined organic layers were washed with brine (10 mL × 3). The organic phase was
separated and dried over anhydrous Na SO . Then the solvent was evaporated in vacuo and
2
4
the crude product was purified by flash column chromatography to give the title compound
1
6 as yellow liquid (1.75 g, 78%).
Compounds 17 was obtained from 4-hydroxycoumarin by using the same method
that described above.
5
.8. Synthesis of 4-azidoquinoline (18)
Sodium azide (1.40 g, 21.50 mmol) was added to a solution of 16 (1.75 g, 10.70
mmol) in DMF (20 mL), and the mixture was stirred at 65 ℃for 5 h. The mixture was
quenched with H O (50 mL) and extracted with EtOAc (50 mL × 3). The combined
2
organic layers were washed with brine (10 mL × 3). The organic phase was separated and
dried over anhydrous Na SO . Then the solvent was evaporated in vacuo and the crude
2
4
product was purified by flash column chromatographyto give the title compound 18 as white
solid (1.73 g, 95%).
5
.9. Synthesis of 4-azido-2H-chromen-2-one (19)
Sodium azide (1.54 g, 24.00 mmol) was added to a solution of 17 (2.84 g, 16.00
mmol) in DMF (30 mL), and the mixture was stirred at room temperature for 8 h. The
mixture was quenched with H O (70 mL) and extracted with EtOAc (50 mL × 3). The
2
combined organic layers were washed with brine (10 mL × 3). The organic phase was
separated and dried over anhydrous Na SO . Then the solvent was evaporated in vacuo and
2
4

ACCEPTED MANUSCRIPT
the crude product was purified by flash column chromatography to give the title compound 19
as white solid (2.90 g, 98%).
5
.10. General procedure for the synthesis of target compounds (20a-21e)
CuSO ·5H O (37.50 mg, 0.15 mmol, 0.05 eq.) in 20mL of water and sodium
4
2
ascorbate (120.00 mg, 0.60 mmol, 0.20 eq.) were added to the corresponding solution
of azide compounds 18 and 19 (3.00 mmol) and alkyne compounds 9a-9e (3.00 mmol)
in tetrahydrofuran (20 mL). The mixture was stirred at room temperature for 5 h and
monitored by TLC. Then the mixture was concentrated in vacuo to remove
tetrahydrofuran and then extracted with DCM (50 mL × 3). The combined organic
layers were washed with brine (10 mL × 3). The organic phase was separated and dried
over anhydrous Na SO . Then the solvent was evaporated in vacuo and the crude product was
2
4
purified by flash column chromatography to give the target compounds.
5
.10.1. 4 -(4-((2,6-dichloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)quinolone (20a)
1
Yield 83%; white solid; m.p. 208-210 ºC; H NMR (400 MHz, DMSO-d ) δ 9.13
6
(d, 1H, J = 8.0 Hz, 2-H), 8.94 (s, 1H, 8′′-H), 8.90 (s, 1H, 5′-H), 8.22 (d, 1H, J = 8.0 Hz,
8
6
1
-H), 7.95-7.90 (m, 2H, 5-H and 7-H ), 7.81 (d, 1H, J = 8.0 Hz, 3-H), 7.77-7.73 (m, 1H,
1
3
-H), 5.82 (s, 2H, -CH ); C NMR (100 MHz, DMSO-d ) δ 153.4, 151.2, 150.8, 149.7,
2
6
48.9, 148.4, 142.2, 140.1, 130.7, 130.5, 129.5, 128.4, 126.4, 122.8, 121.6, 116.9, 39.1;
+
HRMS (ESI) m/z calcd for C H Cl N [M+H] 397.0484, found 397.0485.
1
7
10
2
8
5
.10.2. 4-(4-((6-chloro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)quinoline (20b)
1
Yield 84%; white solid; m.p. 195-197 ºC; H NMR (400 MHz, CDCl ) δ 9.07 (d,
3
1
8
1
H, J = 8.0 Hz, 2-H), 8.78 (s, 1H, 2′′-H), 8.48 (s, 1H, 8′′-H), 8.27- 8.25 (m, 1H, 8-H),
.24 (s, 1H, 5′-H), 7.89-7.87 (m, 1H, 7-H), 7.84 (d, 1H, J = 8.0 Hz, 5-H), 7.67-7.63 (m,
13
H, 6-H), 7. 50 (d, 1H, J = 4.0 Hz, 3-H), 5.77 (s, 2H, -CH ); C NMR (100 MHz,
2
CDCl ) δ 152.2, 151.6, 151.4, 150.2, 149.8, 145.1, 142.0, 140.4, 131.7, 130.9, 130.2,
3
+
1
3
28.7, 125.2, 122.4, 122.0, 116.3, 39.0; HRMS (ESI) m/z calcdfor C H ClN [M+H]
17 11 8
63.0873, found 363.0872.
5
.10.3. 4-(4-((6-chloro-2-fluoro-9H-purin-9-yl)methyl)-1H-1,2,3-triazol-1-yl)quinoline(20c)