A ketone-functionalized Zn-MOF for solvent-free cyanosilylation of aldehyde and treatment activity against osteosarcoma trough increasing Mg63 cells autophagy

Article abstract of DOI:10.1080/00958972.2020.1819538

In this work, 5,5′-carbonyldiisophthalic acid (H₄cipa) was used as a ligand to react with Zn(NO₃)₂·6H₂O in a mixed solvent of DMF and water to prepare a new metal-organic framework (MOF) with the chemical formul

Full text of DOI:10.1080/00958972.2020.1819538

Journal of Coordination Chemistry
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/gcoo20
A ketone-functionalized Zn-MOF for solvent-free
cyanosilylation of aldehyde and treatment activity
against osteosarcoma trough increasing Mg63
cells autophagy
Tao Peng , Peng Jia , Rui Wang , Haoyu Feng & Xiao-Ming Han
To cite this article: Tao Peng , Peng Jia , Rui Wang , Haoyu Feng & Xiao-Ming Han (2020): A
ketone-functionalized Zn-MOF for solvent-free cyanosilylation of aldehyde and treatment activity
against osteosarcoma trough increasing Mg63 cells autophagy, Journal of Coordination Chemistry,
DOI: 10.1080/00958972.2020.1819538
To link to this article: https://doi.org/10.1080/00958972.2020.1819538
View supplementary material
Published online: 23 Sep 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/10.1080/00958972.2020.1819538
A ketone-functionalized Zn-MOF for solvent-free
cyanosilylation of aldehyde and treatment activity
against osteosarcoma trough increasing Mg63
cells autophagy
a
a
b
b
a
Tao Peng , Peng Jia , Rui Wang , Haoyu Feng and Xiao-Ming Han
a
Department of Orthopedics, Shandong Provincial ENT Hospital, Shandong Provincial ENT Hospital
b
Affiliated to Shandong University, Jinan, Shandong, China; Department of Medicine, Shanxi
Hospital, Taiyuan, Shanxi, China
ABSTRACT
ARTICLE HISTORY
Received 21 April 2020
Accepted 16 August 2020
0
In this work, 5,5 -carbonyldiisophthalic acid (H cipa) was used as a
4
ligand to react with Zn(NO
and water to prepare a new metal-organic framework (MOF) with
the chemical formula of [Zn (cipa)(H O) ](DMF) (1). Due to its O
3
)
2
ꢀ6H
2
O in a mixed solvent of DMF
KEYWORDS
2
2
2
Zn(II)-framework; porous
structure; nano-cages;
cyanosilylation reaction;
osteosarcoma
donor sites within the one-dimensional channels and high density
of open metal sites, the activated 1a was applied as a heteroge-
neous catalyst for the formation of C–C bonds of carbonyl com-
pounds in the absence of solvents, which shows high efficiency
and size-selectivity. Furthermore, the possible reaction mechanism
has been proposed based on the experimental results. The treat-
ment activity of the new compound on the Mg63 osteosarcoma
cell line was evaluated and the possible mechanism was dis-
cussed. First, flow cytometry was performed to detect the levels
of the Mg63 cancer cells’ autophagy levels after indicated com-
pound treatment. Next, real time RT-PCR assay was further con-
ducted to measure the relative expression levels of the
autophagic genes under treatment.
CONTACT Xiao-Ming Han
h_xiaoming11@126.com
Department of Orthopedics, Shandong Provincial ENT
Hospital, Shandong Provincial ENT Hospital Affiliated to Shandong University, Jinan, Shandong, China
Supplemental data for this article is available online at https://doi.org/10.1080/00958972.2020.1819538.
ß 2020 Informa UK Limited, trading as Taylor & Francis Group

2
T. PENG ET AL.
1. Introduction
Osteosarcoma, also known as osteogenic sarcoma or osteosarcoma, is one of the most
common primary malignant tumors of bone. Osteosarcoma is highly malignant and
develops rapidly, and often occurs in lung metastases [1]. The incidence of osteosar-
coma accounts for the highest percentage of all malignant bone tumors (44.6%),
accounting for 15.5% of primary bone tumors, about 3 times that of chondrosarcoma
and 7 times that of fibrosarcoma. There are more men than women with osteosar-
coma, and the ratio of men to women is 2.3: 1 [2]. The age of onset is 4 to 60 years
old, with 15 to 25 years old being the most common, accounting for more than 3/4. A
recent study found that multiple mechanisms can regulate the process of cellular
autophagy [3]. The autophagy signaling pathway has attracted wide attention from
researchers.
Cyanohydrins have a significant function within biology and chemistry. They are
extensively used as multi-functional components of fine chemicals, agricultural
chemicals and drugs, such as b-amino alcohols, a-hydroxy acids, etc. [4–6]. Generally
speaking, adding cyanide into carbonyl compounds is one of the basic methods to
prepare them, and it is often at the synthetic chemistry forefront. In the last few dec-
ades, multiple accelerators or activators have been reported for the conversion.
Under pressure to develop environment-friendly processes, organic catalysts have
developed rapidly within the cyanation of carbonyl compounds catalyzed by trime-
thylsilyl cyanide (TMSCN) [7–9]. Metal-organic frameworks (MOFs) are novel materials,
which were formed through the coordination of metal clusters and organic ligands.
MOFs have many unique properties, such as high porosity, large specific surface
area, and adjustable pore size as well as various functions, which make them hetero-
geneous catalysts with shape- and size-selectivity [10–13]. Recent literature has
revealed that MOFs with coordinatively unsaturated Zn(II) metal sites are promising

JOURNAL OF COORDINATION CHEMISTRY
3
candidates for the cyanosilylation reactions. For instance, Bharadwaj and coworkers
reported a Zn(II)-MOF with water bound metal centers, which could act as an effect-
ive catalyst for cyanosilylation reactions and Knoevenagel condensations with size
selective performance [14]. Recycling experiments showed that this Zn(II)-MOF could
be reused at least five times without any loss of activity. Gu et al. obtained a
DABCO-functionalized Zn-MOF with one-dimensional channels, which could behave
as a size selective catalyst for the cyanosilylation reaction [8]. Yuan’s group con-
structed a Zn(II)-MOF with coordinatively unsaturated metal sites using a pyridyl–tri-
carboxylate ligand, which can be used as a highly efficient size selective
heterogeneous catalyst for the solvent-free cyanosilylation of acetaldehydes under
mild conditions [15]. Based on the above results from the literature, both the open
metal sites and the open O donor sites could act as the binding sites for the cyanosi-
0
lylation reaction. In this work, 5,5 -carbonyldiisophthalic acid (H cipa) was used as a
4
ligand to react with Zn(NO ) ꢀ6H O in a mixed solvent of DMF and water to prepare
3
2
2
a
new metal-organic framework (MOF) with the chemical formula of
[
Zn (cipa)(H O) ](DMF) (1). Due to its O donor sites within the one-dimensional chan-
2 2 2
nels and high density of open metal sites, the activated 1a was applied as a hetero-
geneous catalyst for the formation of C–C bonds of carbonyl compounds in the
absence of solvents, which shows high efficiency and size-selectivity. Furthermore,
the possible reaction mechanism has been proposed based on the experimental
results. Through biological functional study, the treatment activity of the compound
on the Mg63 osteosarcoma cell line was evaluated in vitro, and the related mechan-
ism was explored at the same time. The results of the flow cytometry indicated that
the compound could significantly reduce the autophagy levels in the Mg63 osteosar-
coma cells. In addition to this, the data of the real time RT-PCR also suggested that
the relative expression levels of the autophagic genes were promoted by the com-
pound exposure.
2. Experimental
2.1. Chemicals and measurements
0
Zn(NO ) ꢀ6H O (AR) was purchased from Beijing Beihua Chemicals Co., Ltd. 5,5 -
3
2
2
Carbonyldiisophthalic acid (97%) was obtained from Shanghai Guoyao Chemicals
Co., Ltd. TMSCN and various aldehydes used in the catalytic reaction were bought
from Beijing Bailingwei Technology Co., Ltd. The solvents used in this research
were purchased from Tianjin Guangfu Chemicals Co., Ltd. With the Nicolet Avatar
3
60 FT-IR spectrophotometer we determined FT-IR spectra. The heat analysis equip-
ment of Q50 TGA (TA) was utilized for the analysis of thermogravimetry under
ꢁ
ꢂ1
nitrogen flow; the heating rate was 5 Cꢀmin . On a MiniFlex (Cu Ka, k ¼ 1.5418 Å)
we determined bulk samples’ powder X-ray diffraction (PXRD) at room temperature.
The gas chromatography analysis was performed with an Agilent 7890B gas chro-
matograph. The isotherms of gas adsorption (N ) at low pressure (up to 1 bar) were
2
determined using a Micrometrics ASAP 2020 porosity analyzer along with specific
surface area.

4
T. PENG ET AL.
Table 1. Crystallographic parameters along with refinement details for 1.
Empirical formula
17 10 2
C H O11Zn
Formula weight
Temperature/K
Crystal system
Space group
a/Å
1562.97
150.15
tetragonal
I4/mmm
30.461(2)
30.461(2)
b/Å
c/Å
23.0136(13)
90
ꢁ
a/
ꢁ
b/
90
90
21354(3)
8
0.982
1.381
71114
ꢁ
c/
3
Volume/Å
Z
3
q
calcg/cm
-
1
l/mm
Reflections collected
Independent reflections
Data / restraints / parameters
Goodness-of-fit on F
5161 [Rint ¼ 0.1100, Rsigma ¼ 0.0832]
5161 / 3 / 226
1.053
2
Final R indexes [I> ¼2r (I)]
R
R
1
¼ 0.0688, xR
2
¼ 0.1852
Final R indexes [all data]
1
¼ 0.0882, xR
2
¼ 0.1977
-3
Largest diff. peak / hole /e Å
CCDC
1.31 / ꢂ0.64
1968357
2
1
.2. Preparation and characterization for catena-[(l-5,5’-carbonylbis(benzene-
,3-dicarboxylato))-bis-aqua-bis-zinc(II) dimethylformamide] (1) and catena-
[(l-5,5’-carbonylbis(benzene-1,3-dicarboxylato))-bis-zinc(II)] (1a)
A mixture of H cipa of 10 mg and 0.028 mmol along with Zn(NO ) ꢀ6H O, which was
4
3 2
2
2
0 mg and 0.083 mmol, were mixed with HBF (50% in H O, 0.6 mL) and DMF (2.0 mL).
4 2
The mixture was transferred to a vial with a screw cap, which was sealed and heated
ꢁ
in the oven for three days at 90 C. Colorless single crystals with block-shape were col-
lected and then washed using DMF several times to obtain 1 and the yield was 8 mg
(
52% on the basis of ligand H cipa). Anal. Calcd for 1 (C H NO Zn ): C, 40.43; H,
4 20 17 12 2
2
.88; N, 2.36%. Found for 1: C, 40.23; H, 2.77; N, 2.54%. Main FT-IR absorptions (KBr
ꢂ1
pellets, cm ): 3382, 2973, 1646, 1447, 1378, 1286, 1044, 734.
The solvent-free samples of 1 (denoted as 1a) were prepared using the following
procedure: 200 mg of 1 was soaked in MeOH (20 mL) for three days, during which
fresh MeOH was replaced every 12 h. Then the solvent-exchanged samples were
ꢁ
added into an adsorption test tube and heated at 110 C for one day under a
dynamic vacuum.
Applying an Oxford Xcalibur E diffractometer we acquired compound 1’s X-ray data
using the software CrysAlisPro to analyze strength data and then convert it into HKL
files [16]. Empirical absorption correction was applied using spherical harmonics,
implemented in SCALE3 ABSPACK scaling algorithm [17]. Complex 1’s starting struc-
ture pattern was established by the program of Olex2.solve embedded in the OLEX2
software and modified via the program SHELXL-2014 according to the least square
method [18,19]. Complex 1’s non-H atoms were refined with anisotropic parameters;
hydrogens were fixed to their linked C atoms using AFIX commands. Due to the highly
disordered nature of the lattice DMF molecule, it could not be refined, so its electronic

JOURNAL OF COORDINATION CHEMISTRY
5
contribution was removed via the SQUEEZE manipulation embedded in the PLATON
software. The detailed analysis of the SQUEEZE results is discussed in the Supporting
Information. The final chemical formula of 1 was determined via a combination of the
results from the single crystal X-ray diffraction, elemental analysis and the TGA data
[20]. Table 1 describes compound 1’s crystal.
2.3. The heterogeneous catalytic cyanosilylation with 1a
Dried Zn(II)-MOF catalyst (0.005 mmol) was added to a mixture of the corresponding
carbonyl compound (0.5 mmol) and trimethylsilyl cyanide (0.133 mL, 1.0 mmol) under
ꢁ
nitrogen atmosphere. Then, the mixture was stirred at 40 C. The reaction was moni-
tored by GC (FID from AGILENT 7820) using a cross-linked (95%)-dimethyl-(5%)-diphe-
nylpolysiloxane column (HP-5, 30 m ꢃ 0.32 mm
ꢃ
0.25 lm), helium, injector
temperature 250 C, detector temperature 300 C, and oven temperature program
ꢁ
ꢁ
ꢁ
ꢁ
ꢂ1
ꢁ
4
5 C (3 min)–20 Cꢀmin –280 C (2 min). The reaction conversions were determined
by gas chromatography (GC) analysis. For filtration tests, the catalyst was separated
after a reaction time of 2 h with 50% conversion. The reaction was allowed to proceed
in the filtrate, with no additional conversion after 17 h.
2.4. Cell counting kit-8
To determine the inhibitory effect of the synthetic compound on the viability of the
Mg63 osteosarcoma cell, the CCK-8 assay was performed. All operations were carried
out under the guidance of the instructions with some modifications. In brief, the
Mg63 osteosarcoma cells were used for the indicated evaluations. Mg63 cells used in
this research were purchased from ATCC and cultured in DMEM culture medium with
the penicillin-streptomycin solution. After growth into the logical growth phage, the
4
cells were collected and seeded into the 96 well plates at the final density of 10 cells
ꢁ
per well. The plates were then placed in an incubator at 37 C, 5%CO . Then the cells
2
were exposed to the compound at a series concentration (1, 2, 4, 8, 10, 20, 40, 80 lM).
The cells were incubated for a further 48 h, and the cell supernatant was discarded.
Fresh medium was added containing 10% CCK-8 solution. Finally, the absorbance of
each well was measured with a spectrophotometer at 490 nm. This experiment was
repeated at least three times, and the results were presented with mean ± SD.
2.5. Flow cytometry
The flow cytometry was performed in this research to detect the influence of the com-
pound on the Mg63 osteosarcoma cell autophagy. Before the experiment, the Mg63
osteosarcoma cells were first transfected with LC3-GFP plasmid, and then the follow-
ing protocols were performed according to the instructions as previously described.
Mg63 cells harboring the LC3-GFP plasmid in the logarithmic growth phase were col-
5
lected and inoculated into 6 well plates and the density of 1 ꢃ 10 cells per well. After
incubation in an incubator for 12 h, the compound was added into the wells at con-
centrations of 5 lM and 10 lM. The ligand and metal ion were used for the control

6
T. PENG ET AL.
Figure 1. (a) View for the asymmetric unit for 1. (b) A view of the organic ligand coordination
mode. (c) The rhombicuboctahedral cage in 1. (d) A perspective view for the channels along with
cages formed within the three-dimensional skeleton.
treatments. PBS and cell apoptosis inducer were used as negative and positive control.
In the end, the cells in the different groups were harvested and then washed with PBS
three times. The fluorescence absorption value was measured with flow cytometry.
The experiment was repeated for three or more times.
2.6. Real time RT-PCR
Real time RT-PCR was further conducted to demonstrate the induction effect of the
compound on the Mg63 cell line autophagy levels by measuring the relative expres-
sion of the autophagic atg5 and atg7 genes. This experiment was finished in accord-
ance with the instructions with some modifications. Briefly, the Mg63 cancer cells
within the phase of logarithmic growth were collected and inoculated into 96 well
5
ꢁ
plates at the ultimate density 1 ꢃ 10 cells/well. After 12 h culturation at 37 C, 5%CO
2
environment, the cells were then treated with the compound at the concentration of
lM and 10 lM using the ligand and metal ion as the control treatment. After com-
5
pound treatment, the Mg63 cancer cells were collected and washed with PBS 3 times,
then the TRIzol Reagent (Sigma, St. Louis, MO, USA) was used to extract total RNA in
the cells. The quality and quantity of the extracted RNA was measured and then
reversely transcribed into cDNA using an RNA reverse transcription kit according to
protocols. In the end, the relative expression level of the atg5 and atg7 gene in the

JOURNAL OF COORDINATION CHEMISTRY
7
Figure 2. (a) 1a’s and 1’s pattern of PXRD. (b) 1a’s and 1’s curves of TGA. (c) 1a’s N adsorption
2
data at 77 K. (d) Calculating 1a’s distribution of pore size by Horvath–Kawazoe approach.
Mg63 cancer cells was measured with real-time RT-PCR. This experiment was per-
ꢂDDCt
formed in triplicate and analyzed utilizing the 2
approach.
3. Results and discussion
3.1. Crystal structure of 1
According to the results of elemental analysis and single-crystal X-ray diffraction as
well as thermogravimetry, in water along with DMF mixed solvent, Zn(NO ) ꢀ6H O
3
2
2
reacts with H cipa after 48 h to form colorless crystals of [Zn (cipa)(H O) ](DMF) .
4
2
2
2
2
According to the crystal data, which were collected at ambient temperature, 1 crystal-
lizes within the tetragonal space group I4/mmm; the unit cell parameters are a ¼
3
b ¼ 30.461(2) Å and c ¼ 18.6322(14) Å. The unit cell volume is 21354(2) Å . 1’s basic
molecular repeating unit contains four crystallography independently Zn(II) ions, one
2-
and a quarter fully deprotonated cipa ligands and four coordinated water molecules
Figure 1(a)). The crystallographic analysis shows that 1 has a three-dimensional struc-
(
4
–
ture composed of organic cdip connectors along with binuclear propeller Zn (COO)4
2
secondary building units (SBU). Within each binuclear SBU, every Zn(II) ion coordin-
ation geometry is regarded as a five-coordinate pyramid and the s value is 0, which is
2-
formed via four oxygen carboxylic acid oxygen atoms of four distinct cipa ligands
along with a coordinated water molecule. The length of the zinc-oxygen bonds at the
apex [2.114(9)–2.146(8) Å] were obviously longer than the 4 zinc-oxygen bond lengths

8
T. PENG ET AL.
Table 2. Optimization of conditions for the cyanosilylation reaction.
ꢁ
a
Entry
Cat. mol %
TMSCN
Temp. ( C)
Conv. (%)
1
2
3
4
5
6
7
8
9
a
0
2
2
1
0.5
1
3 eq
3 eq
3 eq
3 eq
3 eq
2 eq
2 eq
2 eq
2 eq
rt
rt
40
76
98
98
84
96
94
66
75
40
40
40
40
40
40
40
1 (reused 1a)
1 (Zn(NO
2 (Zn(NO
3
)
)
2
ꢀ6H
2
O)
O)
3
2
ꢀ6H
2
Determined via GC on the basis of carbonyl substrate.
on the base [1.947(3)–1.938(4) Å]. The Zn(II)-O bond distances are comparable with
those observed in other MOFs based on the Zn (COO)4 cluster such as
2
[
2
Zn (TMBHB) ꢀ2NO ꢀ5DMFꢀ4CH CH OHꢀ6H O] (Zn(II)-O bond distance: 1.984(10)-
7
2
3
3
2
2
n
.146(18) Å) [21], [Zn (dbda)(H O) ] (Zn(II)-O bond distance: 2.004(11)-2.149(10) Å) [22].
2 2 2
The Zn-Zn separations are between 2.6461(17) and 2.6514(11) Å, which are similar to
the reported within other MOFs based on Zn(II) with SBUs of Zn (COO) [21–23]. It is
2
4
4
–
found that the organic cdip unit has two configurations, and the two benzene rings
ꢁ
have distinct dihedral angles [62.3(5) and 75.8(5) ] along with ketone-related C ꢂ C–C
ꢁ
bond angles [114.3(5) and 101.1(5) ] (Figure 1(b)). In general, 1’s structural porosity is
chiefly because of peripheral channels between the cages and the formation of rhom-
bic octahedral cage (Figure 1(c)). In each rhombic octahedral cage, there are eight tri-
angular windows along with six quadrilateral windows. The former has a radius of
2.0 Å and the latter is 3.8 Å, neglecting the atoms’ van der Waals radii of. The channels
on the periphery are modified via non-coordinative polar carbonyl functions with
ꢄ2.5 Å radius. This shows that the porosity could provide favorable conditions for gas
entry. Using Platon’s SOLV function, 1’s total accessible volume of solvent was calcu-
3
lated as 14081 Å , accounting for about 65.9% of the total structure volume (Figure
1(d)). The structure of 1 is isostructural to the Cu(II)-MOF based on the same lig-
and [24].
3.2. Pxrd, TGA analysis and gas sorption properties
By measuring the newly prepared samples at room temperature, 1’s phase purity was
determined via powder X-ray diffraction (PXRD). The solvated samples PXRD diagram
was consistent with that of the simulated samples in crystal data, revealing that the
prepared samples had high phase purity (Figure 2(a)). The newly prepared 1’s ther-
ꢁ
mogravimetric analysis shows that the weight loss was about 17.83% between 75 C
ꢁ
and 253 C, equivalent to the removal of one lattice DMF and two coordinated water
ꢁ
molecules (calcd: 18.35%) (Figure 2(b)). Above 450 C, the second weight loss could be
observed, indicating decomposition of the organic ligand and the collapse of the
whole framework. The final product is ZnO (JCPDS card no. 36-1451). Furthermore, the
samples were immersed in methanol for solvent exchange and then heated under a
ꢁ
dynamic vacuum at 110 C for one day to carry out 1’s desolvation. PXRD analysis of
the completely desolvated skeleton reveals that the crystallinity of the porous material
still remained. The absence of coordinated and lattice solvents has been confirmed via

JOURNAL OF COORDINATION CHEMISTRY
9
Table 3. The cyanosilylation reaction of different aldehydes in the existence of 1a.
a
b
Entry
R
1
R
2
Conv.(%)
1
2
3
4
5
6
7
8
9
a
Ph
H
96
94
88
99
78
66
0
4-CH
3
C
C
6
H
H
4
H
H
H
H
H
H
H
H
2-CH
3
6
4
6 4
4-ClC H
1-Nathphyl
9-Anthryl
2-Furyl
n-Bu
2 5
(CH )
99
99
The reaction conditions are Me
3
SiCN (1 mmol), 0.5 mmol of aldehyde/ketone, and 1a (0.013 mg, 0.005 mmol) at
ꢁ
4
0 C and under N
2
.
b
Determined via GC on the basis of carbonyl substrate.
the TGA curve of 1a as shown in Figure 2(b). The activated 1a’s permanent porosity
was demonstrated via the measurements of gas adsorption. According to Figure 2(c),
1
a shows a typical type I adsorption behavior. At 1 bar and 77 K, the saturated absorp-
3
ꢂ1
tion of N is 522.1 cm ꢀg . According to the adsorption of N at 77 K, the Brunauer-
2
2
2
ꢂ1
Emmett-Teller and Langmuir surface areas are 1539 and 2259 m ꢀg , respectively.
3
ꢂ1
1
a’s pore volume is 0.81 cm ꢀg . N isotherms’ distribution of pore size at 77 K was
2
analyzed via the Horvath-Kawazoe approach. It was found that Cu-MOF 1’s distribution
of pore size is about 7.6 Å, which is close to the distribution seen in the single-crystal
structure. This characteristic clearly shows 1’s microporous nature (Figure 2(d)).
3.3. Catalytic properties of 1a
In order to compare 1a’s catalytic activity with other MOFs, we examined carbonyl
compound’s cyanation. After removing the coordination water molecules, the strong
acid, unsaturated metal centers of 1a’s pores volume increases. First, the model experi-
ment was carried out using benzaldehyde as a standard molecule and changing the
amount and temperature of catalyst without solvent. We are pleased that, according
ꢁ
to Table 2, the conversion of benzaldehyde in a nitrogen atmosphere at 40 C was
96% with a 1a loading of 1 mol%, which is only one-eleventh of the conversion
observed in the relevant Mn-BTT reported [25]. In contrast, we found that our cata-
lyst’s activity was higher than the activity of some formerly reported catalysts (Ln-
MOFs along with Sc-MOF), in which the catalyst loading was 5 mol% [26,27]. 1a is
removed via filtration after 2 h reaction stops, providing only 51% of the total conver-
sion after 15 h. This shows that there is no homogeneous catalyst species within the
reaction solution, and 1a is indeed heterogeneous in nature. To study the important
roles of the open metal sites as well as the formation of a porous framework for 1a in
the catalytic reaction, a comparison experiment was carried out using the equivalent
amount of Zn(NO ) ꢀ6H O as the catalyst, which gave a yield of only 66% under the
3
2
2
same reaction conditions after 12 h (Table 2 entry 8). Further increase of the concen-
tration of Zn(NO ) ꢀ6H O showed only a slight increase of the conversion (from 66%
3
2
2
to 75%, Table 2 entry 9), and such an experimental phenomenon might be due to the
enhanced Lewis acidity of the open Zn(II) sites in 1a than that of the Zn(II) in
Zn(NO ) ꢀ6H O. More importantly, it should be pointed out that 1a could be
3
2
2

1
0
T. PENG ET AL.
Scheme 1. The suggested mechanism of 1a catalytic reaction.
conveniently separated from the reaction system via filtration and reused for a second
ꢁ
time after the activation at 80 C for 12 h without losing catalytic efficiency, but the
separation of Zn(NO ) ꢀ6H O from the reaction system needs the extraction or even
3
2
2
the silica gel column chromatography operations, without considering reusing experi-
ments. The results from the comparison experiments highlight the important role of
open metal sites as well as the open framework for 1a in the cyanosilylation reaction.
Under the high conversion rate and high efficiency conditions, we turned our atten-
tion to the investigation of the scope of 1a as a catalyst for general use; the results
are summarized in Table 3. The results showed that the reaction had a wide tolerance
to different substrates. Aryl- (Table 3, items 1-4), heteroaryl- (Table 3, item 7) and ali-
phatic aldehydes (Table 3, item 8) or a cyclic ketone (Table 3, item 9) were appropriate
substrates and the relevant cyanohydrin trimethylsilyl ethers were produced in high to
outstanding conversions. According to Table 3, neither substituent position on the aro-
matic ring or electronic effect revealed any obvious variety in the total conversion.
Furthermore, an obvious effect of size-selectivity was discovered in 1a; when the

JOURNAL OF COORDINATION CHEMISTRY
11
Figure 3. Inhibited viability of the Mg63 cancer cells after compound treatment. Mg63 cancer cells
in the logical growth phage were harvested and inoculated into 96 well plates and then treated
by a series of compounds. The CCK-8 determination was performed to detect the inhibitory activity
of the compound on the Mg63 cancer cells (A). The cytotoxicity of the compound on human
osteoblast hFOB 1.19 (B).
substrate was 1-naphthaldehyde, only 78% product was acquired (Table 3, item 5),
and 9-anthracene formaldehyde had a much lower conversion efficiency (Table 3, item
6). It also shows that cyanation takes place in MOF-1a pores.
According to the results of the experiments and prior reports, a reasonable reaction
mechanism was put forward to explain the cyanation process of 1a catalyst [27–33].
The unstable water molecules within 1a channels were removed via heating at the
aim of exposing the unsaturated metal centers. Activating the aldehydes occurred via
coordination to unsaturated centers of zinc followed by reaction with TMSCN.
Meanwhile, the TMSCN could also be stabilized by the O donor groups in the pores of
1
a. Fresh aldehydes replaced products, and the catalysts activate aldehydes continu-
ously within the next cycle of catalysis (Scheme 1).
3.4. Compound inhibited the viability of the Mg63 cancer cells
Cancer cells are the root of tumors and the result of normal cell mutation. Different
from normal cells, cancer cells have three characteristics of infinite growth, transform-
ation and metastasis, and can proliferate infinitely and destroy normal cell tissues.
Thus, in the biological functional study of this experiment, the CCK-8 preformation
was first conducted to determine the inhibitory effect of the compound on the Mg63
cancer cells using the ligand and metal ion as the control treatment. As the results
show in Figure 3, we can see that different from the control group, the compound
could reduce the viability of the Mg63 cancer cells, and there was a significant differ-
ence between these two groups. However, both the ligand and metal ion have no
effect on the cancer cell viability (A). In addition to this, the compound, as well as the
ligand and ion showed no cytotoxicity against human hFOB 1.19 osteoblast (B). This
result suggested that the compound has potentional inhibitory activity on the cancer
cells growth with low cytotoxicity.

1
2
T. PENG ET AL.
Figure 4. Increased Mg63 cells autophagy level after compound treatment. Mg63 cells were trans-
fected with LC3-GFP plasmid, seeded into 6 well plates and then exposed to the indicated concen-
tration of the compound. Flow cytometry was used to measure the fluorescence absorption value
of different groups after compound treatment, which reflected the cancer cell autophagy levels.

JOURNAL OF COORDINATION CHEMISTRY
13
Figure 5. Induced the atg5 and atg7 relative expression in the Mg63 cells after compound treat-
ment. Mg63 cancer cells in the logical growth phage were harvested and inoculated into 96 well
plates and then treated by a series of compounds. RT-PCR was performed to determine the relative
expression of atg5 and atg7 genes.
3.5. Compound increased the autophagy level in the Mg63 cell line
In the above experiment, we have confirmed the excellent anticancer activity of the com-
pound on Mg63 cancer cells with low cytotoxicity. However, the specific mechanism of
the compound against the Mg63 cancer cells was still unclear. Recent studies found that
multiple mechanisms can regulate the process of cellular autophagy. The autophagy sig-
naling pathway has attracted wide attention from researchers. Thus, in this present experi-
ment, flow cytometry was performed to detect the autophagy levels in the Mg63 cancer
cells after each indicated treatment. From the results in Figure 4, we found that the fluor-
escence absorption value of the Mg63 cells harboring LC3-GFP plasmid was significantly
increased in the treatment compound group, indicating the compound treatment could
evidently increase the Mg63 cells autophagy. The induction activity of the compound also
showed a dose relationship, while the ligand and the metal ion still had no influence on
the cell autophagy. This result revealed that the compound inhibits the osteosarcoma pro-
cess by increasing the autophagy level in the Mg63 cancer cells.
3.6. Compound induced the atg5 and atg7 relative expression in the Mg63 cells
Through the flow cytometry assay, we revealed that the compound significantly
increased the cancer cell autophagy levels. Whether the compound could regulate the
relative expression levels of the autophagic genes in the Mg63 cancer cells was next
examined. Real time RT-PCR was conducted for the atg5 and atg7 genes determin-
ation. The results in Figure 5 indicate that after compound treatment, the relative

1
4
T. PENG ET AL.
expression levels of the atg5 and atg7 in the Mg63 cancer cells were significantly
increased and showed dose relationship. This result was consistent with the previous
study, suggesting the induction of the compound on Mg63 cancer cells autophagy.
4. Conclusion
We prepared a new MOF based on Zn(II) ions as nodes via utilizing the carbonylated
0
tetracarboxylic acid ligand 5,5 -carbonyldiisophthalic acid (H cipa). Single crystal X-ray
4
diffraction studies showed that 1 has open metal sites running along one-dimensional
channels of the c-axis. Due to its O donor sites within the one-dimensional channels
and high density open metal sites, activated 1a was applied as a solvent-free hetero-
geneous catalyst for the formation of C–C bonds of carbonyl compounds, which
showed high efficiency and size-selectivity. Furthermore, according to the experimen-
tal results, we proposed a possible reaction mechanism. In biological research, the
anticancer activity, as well as the specific mechanism of the compound on the Mg63
osteosarcoma cells, was evaluated in vitro. The results of the flow cytometry indicated
that the compound could significantly reduce the autophagy levels in the Mg63 osteo-
sarcoma cells. In addition, the data of the real time RT-PCR also suggested that the
relative expression levels of the autophagic genes was promoted by the compound
exposure. In conclusion, the compound showed an excellent inhibitory effect on the
Mg63 osteosarcoma cells by stimulating the autophagy in the cells.
Acknowledgements
The research did not receive any specific funding.
Disclosure statement
The authors declare there is no conflict of interest regarding the publication of this paper.
Data availability
The data used to support the findings of this study are included within the article.
References
[1] S. Qiu, L. Tao, Y. Zhu. Med. Sci. Monit., 25, 8190 (2019).
[2] Y. Feng, Y. Liao, J. Zhang, J. Shen, Z. Shao, F. Hornicek, Z. Duan. Cell Commun. Signal., 17,
138 (2019).
[
[
[
3] T. Liu, L. Ling, Q. Zhang, Y. Liu, X. Guo. Orthop. Surg., 11, 826 (2019).
4] X. Feng, N. Guo, R. Li, H. Chen, L. Ma, Z. Li, L. Wang. J. Solid State Chem., 268, 22 (2018).
5] X. Feng, H. Chen, R. Li, M. Yang, S. Guo, L. Wang, Q. Liang, Z. Li. Polyhedron, 157, 420
(
2019).
6] X. Feng, Y.P. Shang, H. Zhang, R.F. Li, W.Z. Wang, D.M. Zhang, L.Y. Wang, Z.J. Li. RSC Adv.,
, 16328 (2019).
[
9
[
[
[
7] X. Cui, M.C. Xu, L.J. Zhang, R.X. Yao, X.M. Zhang. Dalton Trans., 44, 12711 (2015).
8] J.M. Gu, W.S. Kim, S. Huh. Dalton Trans., 40, 10826 (2011).
9] P. Phuengphai, S. Youngme, P. Gamez, J. Reedijk. Dalton Trans., 39, 7936 (2010).

JOURNAL OF COORDINATION CHEMISTRY
15
[
[
[
[
10] X. Feng, R. Li, N. Guo, Y.S. Seik, W. Ng, X. Liu, L. Wang. Inorg. Chim. Acta, 459, 87 (2017).
11] W. Liu, H. Li, H. Zhu, P. Xu. Materials, 13, 1180 (2020).
12] Z. Wang, X. Zhang, S. Jiang, Y. Qu, D. Ou, J. Wang. J. Alloys Compd., 828, 154412 (2020).
13] C. Duan, Y. Yu, J. Xiao, X. Zhang, L. Li, P. Yang, J. Wu, H. Xi. Sci. China Mater., 63, 667
2020).
(
[
[
[
14] S. Neogi, M.K. Sharma, P.K. Bharadwaj. J. Mol. Catal. A. Chem., 299, 1 (2009).
15] Y. Zhang, K. Su, M. Hao, L. Liu, Z.-B. Han, D. Yuan. CrystEngComm., 20, 6070 (2018).
16] CrysAlisPro.
CrysAlis171.NET).
17] SCALE3 ABSPACK C A P. Agilent Technologies UK Ltd. 2013.
Agilent
Technologies,
Version
1.171.35.21
(release
20-01-2012
[
[
18] O.V. Dolomanov, L.J. Bourhis, R.J. Gildea, J.A.K. Howard, H. Puschmann. J. Appl.
Crystallogr., 42, 339 (2009).
[19] G.M. Sheldrick. Acta Crystallogr. C. Struct. Chem., 71, 3 (2015).
[20] A.L. Spek. Acta Crystallogr. C. Struct. Chem., 71, 9 (2015).
[21] X. Zhao, X. Wang, S. Wang, J. Dou, P. Cui, Z. Chen, D. Sun, X. Wang, D. Sun. Cryst. Growth
Des., 12, 2736 (2012).
[22] X. Zhang, Z. Zhang, J. Boissonnault, S.M. Cohen. Chem. Commun. (Camb.), 52, 8585
(2016).
[
[
[
[
23] D.M. Chen, X.J. Zhang. CrystEngComm., 21, 4696 (2019).
24] G. Feng, Y. Peng, W. Liu, F. Chang, Y. Dai, W. Huang. Inorg. Chem., 56, 2363 (2017).
25] S. Horike, M. Dinca, K. Tamaki, J.R. Long. J. Am. Chem. Soc., 130, 5854 (2008).
26] R.F. D’Vries, V.A. de la Pe n~ a-O’Shea, N. Snejko, M. Iglesias, E. Guti ꢀe rrez-Puebla, M.A.
Monge. J. Am. Chem. Soc., 135, 5782 (2013).
[
27] R.F. D’Vries, M. Iglesias, N. Snejko, E. Guti ꢀe rrez-Puebla, M.A. Monge. Inorg. Chem., 51,
11349 (2012).
[
[
[
28] Z. Zhang, J. Chen, Z. Bao, G. Chang, H. Xing, Q. Ren. RSC Adv., 5, 79355 (2015).
29] C. Zhu, Q. Xia, X. Chen, Y. Liu, X. Du, Y. Cui. ACS Catal., 6, 7590 (2016).
30] A. Karmakar, A. Paul, G.M.D.M. R uꢀ bio, M.F.C. Guedes da Silva, A.J.L. Pombeiro. Eur. J. Inorg.
Chem., 2016, 5557 (2016).
ꢀ
[
[
[
31] L.M. Aguirre-D ꢀı az, M. Iglesias, N. Snejko, E. Guti ꢀe rrez-Puebla, M.A. Monge. CrystEngComm.,
1
5, 9562 (2013).
32] A. de Oliveira, A. Mavrandonakis, G.F. de Lima, H.A. De Abreu. ChemistrySelect., 2, 7813
2017).
33] J. Tao, Y. Guo, S. Li. J. Mol. Struct. Theochem., 899, 61(2009).
(