New Co(II)-organic framework for cyanosilylation reactions and treatment effect against esophageal cancer by inhibiting cancer cell viability, migration, and invasion

Article abstract of DOI:10.1080/00958972.2020.1788001

A new metal–organic framework with the chemical formula [Co₂(ptptc)·(H₂O)₃](MeOH) (1) was synthesized using Co(NO₃)₂· 6H₂O and terphenyl-3,3″,5,5″-tetracarboxylic acid (H₄ptptc)

Full text of DOI:10.1080/00958972.2020.1788001

Journal of Coordination Chemistry
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/gcoo20
New Co(II)-organic framework for cyanosilylation
reactions and treatment effect against esophageal
cancer by inhibiting cancer cell viability, migration,
and invasion
Hui Zan , Yongdong Huang , Changgao Wang , Zihao Liu , Nan Du & Xiaoman
Wang
To cite this article: Hui Zan , Yongdong Huang , Changgao Wang , Zihao Liu , Nan Du & Xiaoman
Wang (2020): New Co(II)-organic framework for cyanosilylation reactions and treatment effect
against esophageal cancer by inhibiting cancer cell viability, migration, and invasion, Journal of
Coordination Chemistry, DOI: 10.1080/00958972.2020.1788001
To link to this article: https://doi.org/10.1080/00958972.2020.1788001
View supplementary material
Published online: 06 Jul 2020.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=gcoo20

JOURNAL OF COORDINATION CHEMISTRY
https://doi.org/10.1080/00958972.2020.1788001
New Co(II)-organic framework for cyanosilylation
reactions and treatment effect against esophageal
cancer by inhibiting cancer cell viability, migration,
and invasion
a
a
a
b
b
Hui Zan , Yongdong Huang , Changgao Wang , Zihao Liu , Nan Du and
c
Xiaoman Wang
a
Department of Gastroenterology, Hainan General Hospital, Hainan Affiliated Hospital of Hainan
b
Medical University, Haikou, China; Department of Medicine, Chinese PLA General Hospital, Beijing,
China; Department of Gastroenterology, Union Hospital, Tongji Medical College, Huazhong
c
University of Science and Technology, Wuhan, China
ABSTRACT
ARTICLE HISTORY
A new metal–organic framework with the chemical formula
Received 12 February 2020
Accepted 26 May 2020
[
Co
2
(ptptc)ꢀ(H
2
O)
3
](MeOH) (1) was synthesized using Co(NO
3
)
2
ꢀ
0
0
00
6
H O and terphenyl-3,3 ,5,5 -tetracarboxylic acid (H ptptc) as the
2
4
KEYWORDS
starting materials under solvothermal conditions. Removal of the
lattice solvent molecules and the coordinated water molecules
resulted in porous framework 1a having a high density of open
metal sites. In the presence of trimethylsilyl cyanide, catalytic
activity for the cyanosilylation of the aromatic aldehydes was
observed, with different substituents in distinct positions, and
provided cyanohydrin trimethylsilyl ethers with shape- and size-
selective catalysis. Furthermore, the biological function and
detailed mechanism of anti-cancer action of the compound were
assessed. The CCK-8 assay was carried out to determine the com-
pound’s inhibitory function on cell viability. The influence of the
compound on cell invasion and migration was studied via the
transwell method. Finally, the reverse transcription polymerase
chain reaction was used to reveal the mechanism of the com-
pound by measuring the expression level of racl1 and racl2 in
CAES-17 esophageal cancer cells.
Co(II)-framework; porous
structure; cyanosilylation;
anti-cancer activity;
migration and invasion
CONTACT Xiaoman Wang
wangxm629@163.com
Department of Gastroenterology, Union Hospital, Tongji
Medical College, Huazhong University of Science and Technology, Wuhan 430022, China
ß 2020 Informa UK Limited, trading as Taylor & Francis Group

2
H. ZAN ET AL.
1. Introduction
As one of the fatal malignancies worldwide, esophageal cancer has a poor prognosis
and high mortality rate [1]. In addition, in the last few decades, the incidence of
esophageal cancer has increased dramatically within the Western world [2]. As
reported, in 2016, the United States had 16,910 new esophageal cancer cases, and
15,910 people died because of esophageal cancer [3]. Although the treatment and
management of esophageal cancer have been improved to a certain extent, the 5 year
survival rate is only about 10%, and the 5 year survival rate after esophagectomy is
15–40% [4, 5]. Therefore, a new treatment plan for esophageal cancer is
urgently needed.
Metal–organic frameworks (MOFs) are a novel kind of porous material, which have
an adjustable pore size, high specific surface area, hybrid components, and functional
modification. MOFs have a wide application prospect in the fields of gas separation/
adsorption, drug release, and heterogeneous catalysis [6–10]. Compared with a homo-
geneous catalyst, a heterogeneous catalyst has the advantages of easy recycling, less
waste, and product cleaning. In addition, porosity can improve the selectivity and
reactivity. When a Lewis acid/basic functional ligand and Lewis acid metal center are
used as the organic transformation catalytic site, MOFs can be applied as a system of
efficient heterogeneous catalysis [11–14]. In general, the Lewis acid metal centers are
available because of readily accessible defects or metal locations within the frame-
work. In recent years, various catalytic systems of organic transformation based on
MOFs have received several excellent reviews [15–17]. One of the valuable catalytic
systems based on MOFs may be the cyanosilylation reaction of trimethylsilyl cyanide
with aldehydes because the cyanohydrin is an important fine chemical and pharma-
ceutical intermediate [18]. In this research, a novel MOF with the chemical compos-
ition [Co (ptptc)ꢀ(H O) ](MeOH) (1) was synthesized using Co(NO ) ꢀ6H O and
2
2
3
3 2
2
0
0
00
terphenyl-3,3 ,5,5 -tetracarboxylic acid (H ptptc) as the starting materials via the
4

JOURNAL OF COORDINATION CHEMISTRY
3
simple solvothermal method and characterized via thermogravimetric analysis (TGA),
powder X-ray diffraction (PXRD), elemental analysis, and single-crystal X-ray diffraction.
Removal of the lattice solvent molecules and the coordinated water molecules
resulted in porous framework 1a having a high density of open metal sites. In the
presence of the trimethylsilyl cyanide, the catalytic activity for the cyanation of the
aromatic aldehydes was observed, including different substituents in distinct positions,
and provided the cyanohydrin trimethylsilyl ethers with shape- and size-selective
catalysis. In the biological study, the CCK-8 method was carried out, and the results
indicated that the compound has a significant inhibitory effect on the CAES-17
esophageal cancer cell viability. The transwell assay showed that the compound could
inhibit cancer cell invasion and migration in vitro. Finally, in the mechanism study,
racl1 and racl2 expression levels in CAES-17 esophageal cancer cells revealed that the
anti-tumor property of the compound was due to the promotion of reactive oxygen
species (ROS) production.
2. Experimental
2.1. Chemicals and measurements
All solvents and metal salts were obtained from the Tianjin Guangfu Reagent
Company. Benzaldehyde and its derivatives and the trimethylsilyl cyanide (TMSCN)
used in the catalytic studies were supplied by Beijing Bailingwei Technology Co., Ltd.
The H ptptc ligand was purchased from Jinan Henghua Chemical Reagent Factory.
4
Fourier transform infrared (FT-IR) spectra were collected using a Nicolet Avatar 360 FT-
IR spectrophotometer. The Q50 TGA was utilized for TGA under nitrogen flow, and the
ꢁ
ꢂ1
heating rate was 5 Cꢀmin . PXRD patterns were recorded on a Rigaku Dmax 2000 X-
ray diffractometer with graphite-monochromated Cu Ka radiation (k ¼ 0.154 nm) at a
ꢁ
ꢂ1
step size of 0.05 2-theta and counting time of 7 s step . In the PXRD measurements,
the voltage was set to 40 kV, and the current was set to 25 mA. Gas chromatography
analysis was performed with an Agilent 7890B gas chromatograph. Gas adsorption iso-
therms (N ) under low pressure (reach 1 bar) were determined using a porosity ana-
2
lyzer and Micromeritics ASAP 2020 surface area analyzer.
2
.2. Preparation and characterization of [Co (ptptc)ꢀ(H O) ](MeOH) (1) and
2
2
3
[Co (ptptc)] (1a)
2
A mixture of H ptptc (0.0105 g, 0.025 mmol) and Co(NO ) ꢀ6H O (0.0291 g, 0.1 mmol)
4
3 2
2
was added into 8 mL of H O-methanol-DMF (1:2:5 by volume) solution; the suspension
2
was ultrasonically treated for half an hour. The solution was then sealed in a pressure-
ꢁ
resistant Teflon reactor (20 mL) and heated in an oven for 1 day at 85 C. Purple crys-
tals of 1 with a block shape (8 mg) were obtained by centrifugation and washed with
DMF and MeOH three times. From these crystals, a suitable crystal without crack and
impurity was chosen for the single-crystal X-ray diffraction study (the yield was 52.3%,
on the basis of H ptptc ligand). Anal. Calcd for 1 (C H Co O ): C, 45.57; H, 3.33%.
4
23 20
2 12
ꢂ1
Found for 1: C, 45.76; H, 3.23%. IR (KBr, cm ): 3442 (br), 3130 (w), 2928 (w), 1635 (s),
601 (m), 1542 (w), 1509 (m), 1458 (m), 1417 (m), 747 (m), 587 (m).
1

4
H. ZAN ET AL.
The solvent-free samples of 1 (denoted as 1a) could be prepared using the follow-
ing manipulation: about 100 mg of freshly synthesized 1 in the above experiment was
placed into an adsorption sample tube (weighed before activation) and heated in vac-
uum (18 m Torr) at 373 K for 15 h to obtain 1a. The absence of the coordinated and
lattice solvent molecules in 1a was confirmed via the TGA measurements.
Compound 1’s X-ray data were obtained through utilizing a Rigaku SuperNova Dual
Source diffractometer with an Eos detector using a micro-focus sealed X-ray tube,
SuperNova (Cu) X-ray Source (k ¼ 1.54184 Å). The CrysAlisPro software was used to ana-
lyze strength data and then convert the strength data into HKL files [19]. Empirical
absorption correction was performed using spherical harmonics, implemented in
SCALE3 ABSPACK scaling algorithm [20]. On the basis of olex2.solve embedded in the
OLEX 2 software [21, 22], 1’s original structure was established and then modified by the
SHELXL-2014 program on the basis of the least square approach [23]. In 1 non-H atoms
were refined with anisotropic parameters. All H atoms were geometrically fixed to their
connected C atoms through applying AFIX commands. The center phenyl ring of the
4
ꢂ
ptptc ligand was disordered in two positions, and the FLAT, SIMU, and ISOR com-
mands were used to fix the disordered phenyl ring and the thermal vibration of the C
atoms. As a result of the highly disordered nature of the lattice MeOH molecules, they
could not be well evaluated via the structural refinement; thus, their electronic contribu-
tions were removed via the SQUEEZE operation embedded in the software PLATON [24].
The crystal data for 1, which were used in all calculations, were as follows (C H Co O ,
22
16
2 11
M ¼ 574.21 g/mol): monoclinic, space group P2 (no. 4), a ¼ 9.9742(2) Å, b ¼ 17.9136(4) Å,
1
ꢁ
3
c ¼ 14.4262(3) Å, b ¼ 101.125(2) , V ¼ 2529.15(9) Å , Z ¼ 2, T ¼ 293.15 K, l(CuKa) ¼
ꢂ1
3
ꢁ
ꢁ
5
8
.388 mm , Dcalc ¼ 0.754 g/cm , 15,860 reflections measured (7.96 ꢃ 2H ꢃ 129.99 ),
023 unique (R ¼ 0.0504, R ¼ 0.0632). The final R was 0.0625 (I > 2r(I)) and wR
int
sigma
1
2
was 0.1714 (all data). The CCDC number for 1 is 1999639.
2.3. Heterogeneous catalytic cyanosilylation tests
The catalyst 1a (5 mg, 0.1 mol%), aldehyde (1 mmol) and TMSCN (2.0 mmol) were
placed into a Pyrex glass screw cap vial (volume ca. 5 mL), which was vigorously
stirred at room temperature under a nitrogen atmosphere. The reaction progress was
monitored by gas chromatography (GC) analysis, which was performed using an
Agilent 7820 GC with a cross-linked (95%)-dimethyl-(5%)-diphenylpolysiloxane column
ꢁ
(
HP-5, 30 m ꢄ 0.32 mm ꢄ 0.25 lm), injector temperature of 250 C, flame ionization
ꢁ
ꢁ
detector (FID) temperature of 300 C, and oven temperature program of 45 C
3 min)–20 Cꢀmin –280 C (2 min). All products (cyanohydrin trimethylsilyl ethers)
ꢁ
ꢂ1
ꢁ
(
were identified by comparing their GC retention times.
2.4. Cell viability assay
Aiming to explore the synthesized compound inhibitory function against human esopha-
geal cancer cell line CAES-17, the cell counting kit-8 (CCK-8) was performed on the basis
of instructions with a few modifications. In brief, the CAES-17 esophageal cancer cells in
the logarithmic growth phase were collected and inoculated into 96-well plates and

JOURNAL OF COORDINATION CHEMISTRY
5
ꢁ
cultured in an incubator at 5% CO and 37 C for half a day. Then the cells were treated
2
with serial dilution of the compound for 48 h with serial concentrations (1, 2, 4, 8, 10, 20,
40, 80 lM). Cisplatin was used as the positive control, and the ligand as well as the
Co(NO ) ꢀ6H O were also used as controls. After indicated treatment, the cell culture
3
2
2
medium was discarded, and the CCK-8 solution in 100 lL fresh medium was supple-
mented into each well. After another 4 h incubation, the absorbance of each well was
determined at 450 nm using a spectrophotometer. These measurements were performed
three or more times, and the results were presented as mean ± SD.
2.5. Transwell assay
The cell migration and invasion ability of the CAES-17 esophageal cancer cells after
compound treatment were determined by transwell measurement according to the
protocol with some modifications. Cell migration experiments were carried out by
placing a 24-well chamber on a 24-well plate. The CAES-17 esophageal cancer cells
4
were implanted into the upper cavity with a density of 3 ꢄ 10 cells/well. The comple-
ment medium was added to the inferior chamber. The compound, Cisplatin, ligand as
well as the Co(NO ) ꢀ6H O were added into the wells for treatment at the concentra-
3
2
2
ꢁ
tion of 4 lM. At 37 C and 5% CO , the cells were allowed to migrate for 1 day, and
2
then the upper cells were vibrated. Afterward, the lower cells were labeled with 0.5%
crystal violet, and the cells were counted under a light microscope. To perform the
cell invasion test, the 24-well perforation chambers were pre-coated by a gel matrix
(BD Biosciences, NJ, USA), followed by cell plantation and invasion. The following steps
were performed according to the aforementioned protocols.
2.6. Ros gene measurement
After compound treatment for the indicated time, the relative expression level of ROS-
related genes produced in the cells was detected by real-time reverse transcription
polymerase chain reaction (RT-PCR) in accordance with the protocols. In brief, the cells
6
were inoculated into 6-well plates with a quantity of 1 ꢄ 10 cells per well and then
treated with the compound, Cisplatin, and ligand as well as Co(NO ) ꢀ6H O were
3
2
2
added into the wells for treatment at the concentration of 4 lM. Then the cells were
washed with phosphate-buffered saline (PBS) three times and harvested, and the total
RNA in the cells was extracted using the RNA extraction kit. After measuring total RNA
concentration, RNA was transcribed reversely into the cDNA using the manufacturing
kit on the basis of the manufacturer’s recommendations. SYBR Green Master Mix
(Roche) was used in performing RT-PCR; the relative expression level of racl1 and racl2
in CAES-17 esophageal cancer cells were detected with gapdh, which was used as the
ꢂDDCt
internal control. The results were calculated from three tests using the 2
method.
3. Results and discussion
3.1. Crystal structure of 1
0
0
00
The target complex 1 was obtained via the reaction of terphenyl-3,3 ,5,5 -tetracarbox-
ylic acid (H ptptc) with Co(NO ) ꢀ6H O in a mild reaction in the mixed solvent DMF-
4
3 2
2

6
H. ZAN ET AL.
4ꢂ
Figure 1. (a) Complex 1’s asymmetric unit view. (b) The connection mode for the ptptc ligand.
1’s one-dimensional channels view running along the a-axis (c) and c-axis (d).
ꢁ
methanol-H O at 85 C for 24 h and was formulated as [Co (Hptptc) ꢀ(H O) ](MeOH) on
2
2
2
2
3
the basis of the results from the TGA data, elemental analysis, and single-crystal X-ray
diffraction. The crystal structure solution and the results of refinement based on room
temperature data showed that 1 featured a three-dimensional framework structure
and crystallized in the chiral monoclinic space group P2 . As shown in Figure 1(a), 1’s
1
4
ꢂ
basic repeating unit was composed of a fully deprotonated ptptc ligand, two crys-
tallographically unique Co(II) ions, a disordered lattice MeOH molecule, and three coor-
dinated water molecules in agreement with the TGA and elemental analyses (Figure
2
þ
S1). The Co ions had two distinct coordination environments. Co1 was coordinated
to six oxygen atoms; two of them were from the monodentate carboxylate group, and
the other four were from the chelated carboxylate group. Co2 was also coordinated
via the six oxygen atoms; three of which were from the coordination water molecule,
and the other three O atoms were from the monodentate coordination carboxylate
group. The Co(II)-O bond distances were in the region of 1.992(3) Å to 2.198(3) Å,
which were comparable with those observed in other Co(II)-MOFs constructed from
0
tetracarboxylic acid linkers such as f[Co (tpta) (4,4 -bpy)(H O) ]ꢀ3H Og (H tpta ¼
4
2
2
3
2
n
4
0
0
00
00
00
0
0
[
1,1 :3 ,1 -terphenyl]-3,3 ,5,5 -tetracarboxylic acid, 4,4 -bpy ¼ 4,4 -bipyridine, Co–O dis-
0
tances fall between 1.958(2) and 2.361(2) Å) and [Co (L)(H O) ]ꢀ5H O (H L ¼ 5,5 -(eth-
2
2
3
2
4
ane-1,2-diyl)-bis(oxy)diisophthalic acid, Co–O distances fall between 1.981(4) to
.313(4) Å) [25, 26]. The resulting clusters of fCo O g were connected via tetrapod
2
2
6
4
ꢂ
organic ligands. The ptptc ligand connected seven Co(II) ions using its four carboxyl-
1
1
2
1
1
1
1
1
ate groups in a (m -g : g )-(m -g : g )-(m -g : g )-(m -g : g ) mode (Figure 1(b)). The
2
2
2
1

JOURNAL OF COORDINATION CHEMISTRY
7
Figure 2. (a) 1’s and 1a’s PXRD pattern. (b) 1’s and 1a’s TGA curves. (c) 1a’s sorption data of N2
under (d) 1a’s distribution of pore size.
4
ꢂ
connection of the fCo O g clusters with the ptptc ligand resulted in a three-dimen-
2
6
sional skeleton, which had one-dimensional helical channels running along the c-axis.
2
The channel measurement was approximately 7.6 ꢄ 9.2 Å without regard to coordi-
nated water molecules (Figure 1(c)). Meanwhile, the framework of 1 contained a solv-
3
ent accessible free volume of 1719 Å as calculated via the PLATON software,
corresponding to 68% of the crystal volume. Apart from the channels along the c-axis,
fan-shaped channels were also found along the b-axis, and the aperture was 6.2 ꢄ 8.6
2
Å , thereby creating a 2 D pore system in 1 (Figure 1(d)).
3.2. Pxrd, TGA analysis, and gas sorption properties
Complex 1’s bulk phase was determined via PXRD analysis. The diffraction pattern was
consistent with the simulation results, revealing that 1 was a pure phase (Figure 2(a)).
Considering the following bioactivity tests, the stability of 1 in the PBS solution was
studied. The freshly as-prepared compound 1 was soaked in PBS for 1 day, and then
its PXRD patterns were collected, which revealed a similar profile with that of the
newly prepared samples, confirming its good framework stability in PBS. Compound
ꢁ
1
’s thermal stability was assessed via TGA in a N atmosphere at 25–800 C (Figure
2
2
(b)). The TGA data of 1 demonstrated that the first weight loss of 12.7% occurred
ꢁ
ꢁ
between ca. 81 C and 247 C, which was equivalent to the removal of one lattice
MeOH and three coordination water molecules in the pores (calcd: 13.2%). After a

8
H. ZAN ET AL.
ꢁ
ꢁ
short plateau from 247 C to 295 C, a second weight loss could be observed at
2
ꢁ
94–507 C, which corresponded to the entire skeleton collapse and organic ligand
decomposition. The remaining weight corresponded to the formation of Co O , whose
2
3
PXRD patterns were collected and indexed using the JADE 6.5 software with the PDF
004 database (calcd: 27.2%, observed: 28.1%, JCPDS 00-002-0770). The IR spectrum of
2
1
are reported in the Experimental section. The strong and broad stretching absorp-
ꢂ1
tion band of 3442 cm was due to the vibratory OH group from the hydrogen bond
between the lattice water and MeOH molecules. No evident band was observed
ꢂ1
around 1700 cm , demonstrating that the H ptptc ligands in 1 were all completely
4
ꢂ1
deprotonated. Strong characteristic bands at 1635, 1509, 1458 and 1417 cm for 1
can be assigned to the asymmetric and symmetric vibrations of the carboxyl groups.
The values of Dm[m (COO)ꢂm (COO)] bands showed that the carboxyl groups exhibited
as
s
ꢂ1
ꢂ1
monodentate (218 cm ) and chelate (Dm ¼ 51 cm ) coordination modes. As previ-
ously discussed, 1 was microporous and had one-dimensional channels along the dir-
ection of the b-axis. To measure the sample’s permanent porosity, the measurement
of nitrogen adsorption and desorption was carried out. Before measuring adsorption,
the synthesized 1 (100 mg) was placed into an adsorption sample tube and heated in
vacuum (18 m Torr) at 373 K for 15 h to obtain 1a without solvent. PXRD analysis dem-
onstrated that the activated MOF retained the original skeleton structure. In addition,
1
a’s TGA curve revealed that no significant weight loss was found from room tem-
ꢁ
perature to 272 C, confirming the complete activation of the as-prepared samples.
1
a’s permanent porosity was studied via the experiment of N adsorption at 77 K. The
2
results revealed that 1a had type IV characteristics, and the absorption capacity was
3
approximately 354 cm /g (Figure 2#). The accompanying two-step hysteresis might
be due to the rotation implied by the disordered distribution of the dynamic organic
connector or the possible crystal defects implied via the mesoporous characteristics of
the adsorption isotherm. The Langmuir surface areas and Brunauer–Emmett–Teller
2
ꢂ1
(
BET) surface can be calculated as 1447 and 1087 m g , respectively. Moreover, the
pore size distribution measured at the center of 8.6 Å was similar to 1’s crystal struc-
ture (Figure 2(d)).
3.3. Catalytic properties of 1a
Considering that 1 had unsaturated metal centers (UMCs) and a large channel that
might serve as a latent Lewis acid catalyst, the cyanation of carbonyl compounds was
preformed to detect its catalytic performance. In the absence of solvent, benzaldehyde
was used as a standard molecule, and the experimental model was carried out by
changing the amount and temperature of catalyst. As shown in table 1, 1a had out-
standing catalytic activity; the conversion of benzaldehyde was 98.5% in half a day at
room temperature, and the catalyst loading was 0.1 mol% Co, which was only one fif-
teenth of the Cr-MIL-101 (1.5 mol% Cr-active site). In addition, Co-MOF, as a catalyst,
did not require dichloromethane as an organic solvent in contrast to Cr-MIL-101 [14].
1
a can be removed only after filtration for 2 h, and the reaction was stopped com-
pletely. The conversion rate was only 45% after only staying for 15 h, indicating that

JOURNAL OF COORDINATION CHEMISTRY
9
Table 1. Optimization of reaction condition.
ꢁ
a
Entry
Cat. mol%
TMSCN
Temp. ( C)
Conv. (%)
1
2
3
4
5
6
0
1
1
2 eq
2 eq
2 eq
2 eq
2 eq
2 eq
40
40
r.t.
r.t.
r.t.
r.t.
36.6
65.2
76.6
96.5
98.6
38.2
0.5
0.1
0.1(1)
aMeasured via GC on the basis of carbonyl substrate.
Table 2. Cyanation of carbonyl compounds catalyzed by TMSCN.
a
b
Entry
R
1
Conv. (%)
1
2
3
4
5
6
7
8
9
Ph
96
94
88
99s
96
95
82
99
74
0
2-CH
3
C
6
H
H
4
4-CH
4-NO
3
C
6
4
2 6 4
C H
n-Propyl
2-Furyl
(CH
2
)
4
n-Bu
1-Naphthyl
9-Anthryl
1
0
a
ꢁ
Reaction conditions: Me
atmosphere.
Measured via GC on the basis of carbonyl substrate.
3
SiCN of 1 mmol, aldehyde/ketone of 0.5 mmol, 1a (0.1% mmol), 40 C, under
N
2
b
no homogeneous catalyst type was found within the reaction solution; 1a was a genu-
ine heterogeneous catalyst.
Under an optimized condition (2 Eq. TMSCN, catalyst of 0.1 mol% and at room tem-
perature), nine kinds of carbonyl substrates containing cycloketones, aliphatic alde-
hydes, and aromatic aldehydes were cyanated. The results are summarized in table 2,
which revealed that the reaction had wide tolerance to different substrates. Under
standard conditions, when the aromatic aldehydes, which had electronic effect of sub-
stituents (electron-withdrawing NO and electron-donating CH ) or distinct substitu-
2
3
ents, were used, the corresponding products were obtained in excellent yield (over
0%) after 9–12 h. However, the catalytic efficiency of aliphatic aldehydes and ketones
9
was generally lower than that of aromatic aldehydes, which was due to their
decreased inherent reactivity. After a long reaction time of 2 days, cyclobutanone con-
version was only 83%. Aiming to investigate and explain the outstanding catalytic
2
þ
activity of open Co sites, 1 was employed under the same reaction conditions. The
conversion of the reaction was 38.2%, which was close to the reaction with no cata-
2
þ
lyst. This result might be due to the saturated octahedral configuration of the Co
2
þ
ions within 1. The results also revealed that the unsaturated Co site within 1a had a
significant function in catalysis. The channel of cobalt MOFs was a “microreactor,”
which offered a good environment for cyanation. The unsaturated Co Lewis acid
2
þ

1
0
H. ZAN ET AL.
Figure 3. Yields of the first to third runs for cyanosilylation of benzaldehyde using 1a as
the catalyst.
was important to aromatic substrate adsorption, thereby improving catalyst
1a’s activity.
The stability of the heterogeneous catalysts was important for recyclability. The
PXRD patterns of 1a before and after catalytic reaction remained almost identical, indi-
cating that the catalyst was stable in the reaction process (Figure 2(a)). The catalytic
activity was stable at least for three reaction cycles for the substrate benzaldehyde
with yields of 96%, 94% and 93% for three runs (Figure 3).
Based on the experimental results and former reports, a reasonable reaction mech-
anism was suggested to explain the cyanation of 1a catalyst. The unstable water mol-
ecules within compound 1a channel were removed via heating, thereby exposing the
previously saturated metal sites. Aldehydes were activated via Co(II) centers and subse-
quently reacted with TMSCN (scheme 1). Meanwhile, the O atom on the carboxyl
group can accept the tetrel bond donor site of the Si atom and conduct cyanosilyla-
tion of (CH ) SiCN and aldehydes. Therefore, the suggested transition state contained
3
3
(
CH ) SiCN and a catalyst based on MOF, in which carboxyl oxygen interacted with
3 3
high-valence silicon atoms to form the tetrel bond, which was conducive to cyanation.
Aldehydes replaced products, and the catalysts activated aldehydes continuously
within the next cycle of catalysis.
3
1
.4. Compound reduced the viability, migration, and invasion ability of CAES-
7 esophageal cancer cells
After synthesizing the compound, the related anti-cancer activity was explored
in vitro. First, the inhibitory function of the compound on the viability of cancer cells
was measured with the CCK-8 method. Based on the results shown in Figure 4(A), the
cell viability in the compound treatment group was evidently reduced compared with
that in the control group, indicating an evident difference between these two groups.
In addition, the results shown in Figure 4(B) showed that the compound had no cyto-
toxicity on the human normal esophageal cell line, suggesting that the compound
showed high selectivity during anti-cancer treatment. As previously reported, the

JOURNAL OF COORDINATION CHEMISTRY
11
Scheme 1. Suggested mechanism of 1a catalytic reaction.
Zn(bbb)Cl ]ꢀDMF also showed treatment activity on esophageal cancer treatment with
[
2
IC of 8.9 ± 1.1 lM. The anti-cancer activity of 1 was stronger (4.05 ± 0.15 lM) than
5
0
that reported in literature [27, 28]. Moreover, the ligand and metal ions showed no
inhibitory activity on cancer cell viability. All the IC50 values were presented in table 3.
For the migration and invasion ability assay (Figure 4(C,D)), the results indicated that
the control group had a large number of transitional and infiltrating cells. The number
of transplanted and invasive cells in the combined treatment group was evidently
lower than that in the control group (p < 0.001). Furthermore, the ligand and metal
ions were unable to reduce the number of migrated and invaded cells after the indi-
cated treatment.
3.5. Compound inhibited the relative expression of ROS-related genes
In the previous study, we have demonstrated that the compound has a good anti-can-
cer activity in vitro, which could decrease the invasion ability, migration, and survival
ability of cancer cells. Thus, in this experiment, we want to further explore the mech-
anism of the anti-cancer activity of the compound. As reported, the ROS played an
important role in cancer cell viability; therefore, after compound treatment, the genes
related with ROS production were measured with RT-PCR. In Figure 5, a relatively high
level of racl1 and racl2 was observed in CAES-17 esophageal cancer cells, and its
expression level was significantly increased by the compound treatment, indicating

1
2
H. ZAN ET AL.
Figure 4. Cell invasion and migration and viability were decreased after the combined treatment.
The CAES-17 esophageal cancer cells were inoculated into plates and treated by the compound.
(A) The viability of CAES-17 esophageal cancer cells was measured by CCK-8 assay. (B) The viability
of NEC cells was measured by CCK-8 assay. (C, D) The transwell method was used to detect the
invasion and migration of tumor cells.
Table 3. IC50 Values of the compounds on the cancer cells or NEC.
Names
CAES-17
NEC
Compound
Cisplatin
Solution
Ligand
4.05 ± 015 lM
>80 lM
>80 lM
>80 lM
>80 lM
>80 lM
>80 lM
>80 lM
>80 lM
>80 lM
Ion
the outstanding promotion effect of the compound on ROS production. Furthermore,
the ligand and metal ions had no activity on the regulation of the relative expression
level of racl1 and racl2 in CAES-17 esophageal cancer cells after compound treatment.
4. Conclusion
We have prepared a novel metal–organic skeleton with high porosity under solvother-
mal conditions. Single-crystal X-ray analysis indicated that 1 showed a 3 D framework
based on the fCo (CO ) (H O) g cluster-based secondary building unit, which had 1 D
2
2 4
2
3
open channels running along the c-axis. Removal of the lattice solvents and the coor-
dinated water molecules resulted in the porous framework 1a having a high density
of open metal sites. In the presence of trimethylsilyl cyanide, the catalytic activity for

JOURNAL OF COORDINATION CHEMISTRY
13
Figure 5. Increased expression level of racl1 and racl2 in CAES-17 esophageal cancer cells after
compound treatment. The CAES-17 esophageal cancer cells were inoculated into plates and treated
by compound. The relative expression level of racl1 and racl2 in cells was measured with RT-PCR.
the cyanation of aromatic aldehydes was observed, with varying substituents in dis-
tinct positions, and provided the cyanohydrin trimethylsilyl ethers. In the biological
research, the anti-cancer activity of the compound action was evaluated, and the
related mechanism was explored. The CCK-8 assay indicated compound gas inhibitory
function on cancer cell viability. Moreover, the transwell assay revealed that the com-
pound could inhibit the invasion and migration ability of cancer cells. Furthermore,
the expression level of racl1 and racl2 in CAES-17 esophageal cancer cells revealed
that the anti-tumor ability of the compound was due to the increase of
ROS production.
Disclosure statement
No potential conflict of interest was reported by the author(s).
References
[
[
[
1] L. Zheng, S. Kocic, D. Radovanovic, I. Zivanovic Macuzic, I. Ilic. J. Buon., 24, 1666 (2019).
2] Q. Song, J.Z. Wu, S. Wang, W.H. Chen. Sci. Rep., 9, 15234 (2019).
3] Y. Xu, L. Wang, B. He, W. Li, Q. Wen, S. Wang, X. Sun, X. Meng, J. Yu. Radiat. Oncol., 14,
181 (2019).
[4] A. Rosemurgy, C. Wilfong, D. Craigg, F. Co, I. Sucandy, S. Ross. Am. Surg., 85, 944 (2019).
[5] K.J. Anderson, R.T. Cormier, P.M. Scott. World J. Gastroenterol., 25, 5732 (2019).
[6] Y. Xiong, Y.Z. Fan, D. Damasceno-Borges, C.X. Chen, Z.W. Wei, H.P. Wang, M. Pan, J.J.
Jiang, G. Maurin, C.Y. Su. Chemistry, 22, 16147 (2016).
[7] Y. Xiong, Y.Z. Fan, R. Yang, S. Chen, M. Pan, J.J. Jiang, C.Y. Su. Chem. Commun. (Camb.),
5
0, 14631 (2014).
[8] D.M. Chen, X.J. Zhang. J. Solid State Chem., 278, 120906 (2019).
[9] Y.H. Han, Y.C. Liu, X.S. Xing, C.B. Tian, P. Lin, S.W. Du. Chem. Commun. (Camb.), 51, 14481
(2015).
[10] D.M. Chen, X.J. Zhang. CrystEngComm., 21, 4696 (2019).

1
4
H. ZAN ET AL.
[11] A. Bhunia, S. Dey, J.M. Moreno, U. Diaz, P. Concepcion, K. Van Hecke, C. Janiak, P. Van
Der Voort. Chem. Commun. (Camb.), 52, 1401 (2016).
[
[
[
[
[
[
[
[
[
12] X. Cui, M.C. Xu, L.J. Zhang, R.X. Yao, X.M. Zhang. Dalton Trans., 44, 12711 (2015).
13] L.J. Zhang, C.Y. Han, Q.Q. Dang, Y.H. Wang, X.M. Zhang. RSC Adv., 5, 24293 (2015).
14] Z. Zhang, J. Chen, Z. Bao, G. Chang, H. Xing, Q. Ren. RSC Adv., 5, 79355 (2015).
15] S. Neogi, M.K. Sharma, P.K. Bharadwaj. J. Mol. Catal. A: Chem, 299, 1 (2009).
16] J.M. Gu, W.S. Kim, S. Huh. Dalton Trans., 40, 10826 (2011).
17] D. Wang, Z. Li. J. Catal., 342, 151 (2016).
18] M.Y. Sun, D.M. Chen. Polyhedron., 147, 80–85 (2018).
19] CrysAlisPro Version 1.171. 37.35 [J]. Agilent Technologies (2014).
20] ABSPACK, SCALE. CrysAlisPro, Agilent Technologies UK Ltd, Oxford, UK, 2014, 2015
(2011).
[
[
21] O.V. Dolomanov, L.J. Bourhis, R.J. Gildea, J.A.K. Howard, H. Puschmann. J. Appl.
Crystallogr., 42, 339 (2009).
22] L.J. Bourhis, O.V. Dolomanov, R.J. Gildea, J.A.K. Howard, H. Puschmann. Acta Crystallogr. A
Found Adv., 71, 59 (2015).
[23] G.M. Sheldrick. Acta Crystallogr. C Struc.t Chem., 71, 3 (2015).
[24] A.L. Spek. Acta Crystallogr. C Struct. Chem., 71, 9 (2015).
[25] L. Zhang, L. Liu, C. Huang, X. Han, L. Guo, H. Xu, H. Hou, Y. Fan. Cryst. Growth Des., 15,
3426 (2015).
[26] W.Q. Kan, Y.Y. Liu, J. Yang, Y.Y. Liu, J.F. Ma. CrystEngComm., 13, 4256 (2011).
[
27] Z. Human, A. Munyaneza, B. Omondi, N.M. Sanabria, R. Meijboom, M.J. Cronj ꢀe . Biometals,
2
8, 219 (2015).
[28] Z. Che, S. Wang, S. Liu, G. Li, Q. Wu, C. Lin, L. Kong, S. Wang. Spectrochim. Acta A Mol.
Biomol. Spectrosc., 135, 878 (2015).