Organoantimony(III)halide complexes with azastibocine framework as potential antitumor agents: Correlation between cytotoxic activity and N→Sb inter-coordination

Article abstract of DOI:10.1016/j.ejmech.2019.05.054

The relationship between chemical structure and in vitro cytotoxic activities of a series of azastibocine-framework organoantimony(III)halide complexes against cancerous (HepG2, MDA-MB-231, MCF-7 and HeLa)and nonmalignant (HEK-293)cell lines was studied f

Full text of DOI:10.1016/j.ejmech.2019.05.054

Accepted Manuscript
Organoantimony(III) halide complexes with azastibocine framework as potential
antitumor agents: Correlation between cytotoxic activity and N→Sb inter-coordination
Jian Lei, Yongping Liu, Yingcan Ou, Chak-Tong Au, Yi Chen, Shuang-Feng Yin
PII:
S0223-5234(19)30464-7
DOI:
https://doi.org/10.1016/j.ejmech.2019.05.054
EJMECH 11364
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 11 March 2019
Revised Date: 2 May 2019
Accepted Date: 19 May 2019
Please cite this article as: J. Lei, Y. Liu, Y. Ou, C.-T. Au, Y. Chen, S.-F. Yin, Organoantimony(III) halide
complexes with azastibocine framework as potential antitumor agents: Correlation between cytotoxic
activity and N→Sb inter-coordination, European Journal of Medicinal Chemistry (2019), doi: https://
doi.org/10.1016/j.ejmech.2019.05.054.
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
Graphic abstract

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
Organoantimony(III) Halide Complexes with Azastibocine
Framework as Potential Antitumor Agents: Correlation between
Cytotoxic Activity and N→Sb Inter-Coordination
a ,b, 1
a, 1
a
c
a, b, *
Jian Lei
, Yongping Liu , Yingcan Ou , Chak-Tong Au , Yi Chen
,
b,**
Shuang-Feng Yin
a
School of Medicine, Hunan University of Chinese Medicine, Changsha 410208, PR
China
b
State Key Laboratory of Chemo/Biosensing and Chemometrics, College of
Chemistry and Chemical Engineering, Hunan University, Changsha 410082, PR
China
1
1
1
1
1
1
c
College of Chemistry and Chemical Engineering, Hunan Institute of Engineering,
Xiangtan, 411104, PR China
* Corresponding author. E-mail address: cscy1975@163.com (Y. Chen).
** Corresponding author. E-mail address: sf_yin@hnu.edu.cn (S.-F. Yin).
1
Lei and Liu contributed equally to this work.
1

ACCEPTED MANUSCRIPT
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Abstract
The relationship between chemical structure and in vitro cytotoxic activities of a
series of azastibocine-framework organoantimony(III) halide complexes against
cancerous (HepG2, MDA-MB-231, MCF-7 and HeLa) and nonmalignant (HEK-293)
cell lines was studied for the first time. A positive correlation between cytotoxic
activity and the length of N→Sb coordinate bond on azastibocine framework of same
nitrogen substituent was observed. By comparison, the organoantimony(III) complex
6-cyclohexyl-12-fluoro-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine
(C4)
exhibited the highest selectivity index, giving a IC (nonmalignant)/IC (cancerous)
5
0
50
ratio of up to 8.33. The results of cell cycle analysis indicated that the inhibitory effect
of C4 on the cellular viability was caused by cell cycle arrest mainly at the S phase.
The necrosis induced by C4 was confirmed by the Trypan blue dye exclusion test and
the increase of lactic dehydrogenase (LDH) released in the culture medium.
Furthermore, evaluation of the levels of intracellular reactive oxygen species (ROS) in
MDA-MB-231 cells, by quantifying the relative fluorescence units (RFU) using
spectrofluorometer, indicated that cytotoxic activity of C4 is dependent on the
production of ROS. This work established the correlation between cytotoxic activity
and N→Sb inter-coordination, a finding that provided theoretical and experimental
basis for in-depth design of antimony-based organometallic complexes as potential
anticancer agents.
2

ACCEPTED MANUSCRIPT
36
37
38
Keywords
Organoantimony(III) complex; Cytotoxic Activity; Structure-activity relationship;
Necrosis; Reactive oxygen species
39
40
41
42
43
Abbreviations
CCK-8, Cell Counting Kit-8; IC , half-maximal inhibitory concentration; SI,
5
0
selectivity index; PI, propidium iodide; PS, phosphatidylserine; LDH, lactic
dehydrogenase; ROS, reactive oxygen species; DCFH-DA,
5-(and-6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate; NAC, N-acetylcysteine
3

ACCEPTED MANUSCRIPT
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
1. Introduction
The rich diversity of coordination chemistry provides exciting prospects for the
design of novel metallodrugs with unique biological activity [1]. Since the landmark
report of cisplatin compounds in cancer therapy, continuous efforts have been made in
the investigation of new metallic and organometallic complexes for the purpose of
developing metal-based chemotherapeutics of low toxicity that are endowed with
anticancer activity [2–6].
In the past century, antimony-based drugs have been extensively investigated
because of their clinically therapeutic efficacy against leishmaniosis [7,8]. Like
cancer, the parasite divides rapidly and may share some common pathways that could
be targets for chemotherapy [7,8]. Therefore, there is an increasing interest in
antitumor application of antimony compounds, and significant progress has been
made [9,10]. For instance, inorganic trivalent antimonials such as Sb O were found
2
3
to show biological effects on acute promyelocytic leukemia (APL) that closely
resemble those of As O [11]. A number of antimony(III) compounds complexed with
2
3
polydentate ligands such as carboxylic acids [12–14], hydrazones [15], thioamides
[16], and dithiocarbamates [17], were synthesized and assessed as biomedical agents
for anticancer purpose. Currently, the most studied antimony(III) compounds in the
context of antitumor activity are the ones with at least one antimony–carbon covalent
bond, the so-called organoantimony complexes [18–24]. The first example showing
the antitumor property of organoantimony(III) species was reported by Silvestru and
coworkers, wherein they found that diphenylantimony(III) derivatives of
4

ACCEPTED MANUSCRIPT
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
dithiophosphorus ligands exhibited inhibitory effects on the growth of Ehrlich ascites
tumor [18]. Meanwhile, the pharmacological applications of antimony(V)-based
organometallic complexes in cancerous cells proliferation were also reported [25–29].
The potential inhibitory effects of triphenylantimony(V) polyamines on cancer-related
cell lines were described by Carraher et al. [25]. Demicheli and coworkers disclosed
the in vitro cytotoxic activity of a series of pentavalent organoantimony complexes
against malignant carcinoma cells, such as human chronic myelogenous leukemia
(K562) and murine metastatic melanoma (B16F10) [26,27]. Arylantimony(V)
derivatives of exo-7-oxa-bicyclo[2,2,1]heptane (ene)-3-arylamide-2-acid and
arylhydroxamic acid were introduced by Li and coworkers as potential antitumor
agents against a diversity of human neoplastic cell lines [27,28]. Among those
antimony-based metallic and organometallic complexes, the presence of coordination
between antimony center and ligand is relevant to their cellular activity. In most cases,
the coordination of antimony by ligands resulted in compounds with antitumor
potency greater than that of the uncoordinated ligands [13,15,27–29] as well as their
antimony-containing precursors [15,28,29]. The strong development in this field
motivated us to elucidate the relationship between coordination effect and antitumor
potency for the purpose of developing organoantimony-based chemotherapeutics in
cancer therapy. To our knowledge, relatively few information is available in the
literature about the related chemistry between coordination effect and antitumor
activity of antimony-based organometallic complexes.
Previously our group reported the anti-proliferative activity of heterocyclic
5

ACCEPTED MANUSCRIPT
8
8
9
9
9
9
9
9
9
9
9
9
8
9
0
1
2
3
4
5
6
7
8
9
hypervalent organoantimony(III) complexes on human alveolar adenocarcinoma cell
line (A549) [30]. Although full details are yet to be revealed, preliminary data have
indicated that anti-proliferative activity detected over these complexes could be
attributed to the coordination between the antimony and nitrogen atoms. Very
recently, Hadjikakou and coworkers developed a series of antimony(III) halide
complexes with dithiocarbamate ligands, and they found that the halogen type causes
no influence on the antitumor activity [17]. With this in mind, we herein revealed for
the first time the correlation between cytotoxicity and coordination effect by studying
the in vitro structure-activity relationship (SAR) of a series of azastibocine-framework
organoantimony(III) halide complexes against different cell lines, including solid
cancerous (human liver hepatocellular carcinoma cell line, HepG2; human breast
cancer cell line, MDA-MB-231; human breast adenocarcinoma cell line, MCF-7; and
human cervical carcinoma cell line HeLa) and nonmalignant (human embryonic
kidney cell line, HEK-293) cells. Moreover, considering the selectivity index that is
derived from IC (nonmalignant)/IC (cancerous) ratio, the most potent compound C4
1
1
1
1
1
1
00
01
02
03
04
05
5
0
50
was selected for further mechanistic investigation. Finally, we describe
organoantimony complex C4 driven induction of necrosis in MDA-MB-231 cells
through intracellular ROS production.
6

ACCEPTED MANUSCRIPT
06 2 Results and Discussion
1
107
2.1 Chemistry
108
109
110
111
Fig.1. a) Molecular structures of organoantimony(III) halide complexes with azastibocine
framework. b) ORTEP diagram of the X-ray molecular structure of organoantimony(III)
complexes C1–C12.
7

ACCEPTED MANUSCRIPT
Table 1
Selected bond lengths (Å) and angles (deg) of organoantimony(III) halide complexes C1–
C12.
112
113
114
115
116
117
118
119
120
121
The synthesis of organoantimony chlorides, which were labeled C1, C5 and C9,
was carried out as previously reported [30]. Organoantimony bromides, iodides, and
fluorides could be prepared by mixing the organoantimony chlorides with the
corresponding inorganic salts (i.e. potassium bromide, potassium iodide or silver
fluoride) in 1:10 or 1:1 molar ratio as specified in the Experimental Section. Fig. 1a
depicts the general molecular structures of the complexes of formula
t
RN(CH C H )SbX, where R = Cy (cyclohexyl), Ph (phenyl), or Bu (tertiary butyl)
2
6
4
and X = Cl, Br, I, F. These complexes readily recrystallized from the reaction
mixtures as colorless crystals in moderate to excellent yields and could be kept in
1
open air for a long-term period without showing any detectable change in H NMR
8

ACCEPTED MANUSCRIPT
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
spectra (see Supplementary Materials for details).
The molecular structures of C1–C12 are unambiguously characterized by NMR
spectroscopy, elemental analysis as well as single crystal X-ray diffraction analysis.
The ORTEP diagrams of these complexes revealed that the central
antimony-containing part shows analogous pseudo-trigonal bipyramidal structures
with a butterfly-shaped ligand where the nitrogen and halogen atoms are located at the
apical positions and the two carbon atoms adjacent to the antimony atom are at the
equatorial positions along with a lone electron pair of antimony (Fig. 1b). The lengths
of N→Sb coordinate bond in these complexes are within the 2.387(2)–2.522(2) Å
range (Table 1), which are slightly longer than the sum of the covalent radii (2.11 Å)
but much shorter than the sum of the van der Waals radii (3.74 Å) [31,32]. The results
evidence the inter-coordination between the antimony and the nitrogen atoms, and the
strength of N→Sb coordination on the azastibocine framework could be adjusted by
synergistic modulating the property of nitrogen substituents and halogen atoms
adjacent to the central antimony atom. The lengths of N→Sb coordinate bond of
phenyl derivatives (C5–C8) are obviously longer than those of cyclohexyl derivatives
(C1–C4). Given the inverse electronic properties of phenyl and cyclohexyl group, it is
apparent that the strength of N→Sb inter-coordination is weakened with the increase
of electron-withdrawing ability of the nitrogen substituents. The steric effect of the
nitrogen substituents also significantly affect the intramolecular interaction between
the antimony and nitrogen atoms, as seen in the lengths of N→Sb coordinate bond of
tert-butyl derivatives C9–C12. In addition, the coordinate bond lengths of
9

ACCEPTED MANUSCRIPT
144
145
146
organoantimony(III) fluorides [2.450(2) Å, C4; 2.522(2) Å, C8; 2.495(5) Å, C12] are
distinctly longer than those of chlorides, bromides and iodides, owing to the strong
electron-withdrawing ability of fluorine atom.
Table 2
Crystallographic data for organoantimony(III) halide complexes C1–C12.
147
2.2 Biological activity
1
0

ACCEPTED MANUSCRIPT
148
2.2.1 Anticancer/cytotoxic activity and structure-activity relationship (SAR) study
Table 3
Cytotoxic effect of organoantimony(III) halide complexes C1–C12, nitrogen-containing
precursors P1–P3 and cisplatin on various cancerous and nonmalignant cell lines after 24 h.
Cancerous Cell Lines: HepG2 (human liver hepatocellular carcinoma cell line),
MDA-MB-231 (human breast cancer cell line), MCF-7 (human breast adenocarcinoma cell
line) and HeLa (human cervical carcinoma cell line). Nonmalignant Cell Line: HEK-293
(
human embryonic kidney cell line). IC : concentration that is cytotoxic against 50% of cell
50
a
lines. SI: selectivity index. The IC values were determined through non-linear regression
5
0
analysis; Each well was triplicated and each experiment was repeated at least three times. IC50
b
values quoted are mean ± SD (µM). Calculated as the ratio between IC in HEK-293 cells
5
0
c
and IC50 in HepG2. Calculated as the ratio between IC in HEK-293 cells and IC50 in
50
d
e
MDA-MB-231. Calculated as the ratio between IC in HEK-293 cells and IC in MCF-7.
5
0
50
Calculated as the ratio between IC in HEK-293 cells and IC in HeLa.
5
0
50
149
150
151
152
153
154
155
The anticancer activity of the synthesized organoantimoy(III) halide complexes was
evaluated in a panel of four human solid tumor cell lines derived from hepatocellular
carcinoma (HepG2), breast cancer (MDA-MB-231 and MCF-7) and cervical
carcinoma (HeLa), together with cisplatin as positive control. Cell viability was
assessed using Cell Counting Kit-8 (CCK-8) as described and the results of
half-maximal inhibitory concentration (IC ) that represents the drug concentration
5
0
required to inhibit cell growth by 50% are summarized in Table 3. The results
1
1

ACCEPTED MANUSCRIPT
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
demonstrated that all of the organoantimony(III) halide complexes exhibited much
higher cytotoxic activity against the selected cancerous cell lines than cisplatin and
the nitrogen-containing precursors P1–P3. It was also observed that the modulation of
the type of the nitrogen substituent and halogen atom on the
tetrahydrodibenzo[c,f][1,5]azastibocine framework would result in distinct
antineoplastic performance towards the same cancerous cell line.
To further elucidate the related chemistry between coordination effect and
cytotoxic activity of organoantimony complexes, a systematic structure-activity
relationship study was explored. Given the different electronic and steric properties of
the nitrogen substituents as well as based on the results of our preliminary study on
antimony-based organometallic antitumor agents [30], a ranking of cytotoxic activity
across the organoantimony complexes with the same nitrogen substituent (e.g.
cyclohexyl, phenyl and tertiary butyl) and N→Sb coordinate bond length was
performed (Fig. 2a–c). In each case, the nonlinear correlations between the lengths of
N→Sb coordinate bond, which were modulated simply by the halogen type, and the
cytotoxic activities of the corresponding organoantimony halide complexes were
found, respectively. When the complexes were ranked by decreasing coordination
effect, i.e., with increasing N→Sb coordinate bond lengths, the complexes having the
same nitrogen substituent showed increasing inhibitory activities on all four cancerous
cell lines. On the other hand, the most potent cytotoxic activities were observed over
the organoantimony fluorides (C4, C8 and C12) in each set, albeit there is no trend
for the complexes bearing the other halogens (e.g. chloride, bromide and iodide).
1
2

ACCEPTED MANUSCRIPT
178
179
180
181
182
183
184
185
186
187
According to this relationship and experimental results, we deduce that the cytotoxic
activity of these complexes was determined by the level of exposure of the
antimony(III) center. With decreasing extent of N→Sb coordination, the cytotoxic
activities of organometal moiety towards cancerous cells were increased. In other
words, the nitrogen-containing ligand may serve as an auxiliary moiety to decrease
the cytotoxicity of the antimony(III) center. With this in mind, we speculate that there
are two potential functions of N→Sb inter-coordination in these organoantimony(III)
complexes: (1) maintaining the skeleton structure to improve the air-stability and
water-tolerance; (2) reducing the level of exposure of the antimony(III) center to
decrease the cytotoxicity of trivalent antimonial.
1
1
88
89
Fig. 2. Correlation between N→Sb coordinate bond length in organoantimony complexes
with the same nitrogen substituent (a: cyclohexyl; b: phenyl; c: tert-butyl) and corresponding
anticancer activity against human tumor cells: HepG2 (red line), MDA-MB-231 (purple line),
MCF-7 (blue line) and HeLa (green line).
190
191
192
193
194
195
196
197
To test the effect of these organoantimony(III) halide complexes on nonmalignant
cells, we employed human embryonic kidney cells (HEK-293). The study of the
effects of these twelve complexes on nonmalignant cells is of particular interest in
terms of their potential application in clinical practice. After 24 h incubation, the IC50
values of complexes C1–C12 on HEK-293 are within the 2.49–10.23 µM range. By
1
3

ACCEPTED MANUSCRIPT
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
comparison with the commercial cisplatin, the majority of the synthesized
organoantimony(III) complexes exhibited relatively higher selectivity index in the
biological
tests,
especially
so
with
the
complex
C4
6-cyclohexyl-12-fluoro-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine, which gives
a IC (nonmalignant)/IC (cancerous) ratio of up to 8.33. With such a performance,
5
0
50
C4 opens a potential therapeutic window for antimony-based organometallic
antitumor drugs, and we focused our investigation on the mechanistic study of C4 in
MDA-MB-231 cells.
2.2.2 Cell Cycle Analysis
Generally, metal-based chemotherapeutic agents exhibit perturbation effect on cell
cycle that is composed of G , S, G , M, and G phases in proliferating cells [33,34].
1
2
0
At different stages of the cell cycle, cell nuclei contain different amounts of DNA.
Therefore, according to the DNA content detected by propidium iodide (PI) staining,
cell cycle is commonly described on the basis of G /G , S, and G /M phases. To
0
1
2
investigate whether the cytotoxic effect of C4 was caused by cell cycle arrest, PI
staining and flow cytometry analysis of cells was performed in MDA-MB-231 cells.
As shown in Fig. 3, after independently incubated with C4 at 0.25, 0.5, 1, and 2
equipotent concentrations of IC50 (0.91 µM) for 24 h, the MDA-MB-231 cells in
G /G phase showed a significant population decrease from 50.4% (untreated cells) to
0
1
3.7% (1.82 µM of C4), accompanied by a dose-dependent increase in the fraction of
cells in the S (68.9% compare to 39.8% in untreated cells) and G /M phases (26.8%
2
versus 9.8%). The results confirmed the retardation effect of organoantimony fluoride
1
4

ACCEPTED MANUSCRIPT
220
221
222
C4 on the cell cycle progression, and it is reasonable to deduce that the observed
inhibitory effect on the cellular viability of MDA-MB-231cells was caused by cell
cycle arrest mainly at the S phase, which represents a period of DNA replication.
223
224
225
226
227
Fig. 3. Effect of C4 on the cell cycle distribution of MDA-MB-231 cells. a) Flow cytometry
analysis was performed on MDA-MB-231 cells treated with C4 for 24 h. b) With the
percentages of cells in different phases quantified. Data were shown as mean ± S.D. (n=3). *P
< 0.05, **P < 0.01 compared with the population of cells in the control group.
228
229
230
231
2.2.3 Apoptosis assay
Because prolonged cell cycle arrest may influence cellular viability through
processes including anti-proliferation and cell death [35,36], cell apoptosis has been
adopted as a vehicle for cancer treatment [37]. We next evaluated the apoptotic
1
5

ACCEPTED MANUSCRIPT
232
233
234
235
236
237
238
239
240
241
activity of C4-treated MDA-MB-231 cells by means of Annexin-V and PI double
staining using a FACSCalibur flow cytometer. During apoptosis, phosphatidylserine
(PS) is translocated from the cytoplasmic face of the plasma membrane to the cell
surface. Thus, apoptotic cells can be identified by the presence of PS on the cell
2
+
surface. Annexin V has a strong Ca -dependent affinity for PS, and therefore can be
used as a probe for detecting apoptosis. PI can only pass through the membranes of
later apoptotic and necrotic cells, and consequently intercalate into their nucleic acids,
since the integrity of their plasma and nuclear membranes was decreased. This allows
+
-
+
+
apoptotic cells (annexin-V /PI and annexin-V /PI ) to be distinguished from living
-
-
-
+
(annexin-V /PI ) and necrotic (annexin-V /PI ) cells in flow cytometry analysis.
2
42
43
2
Fig. 4. a) Apoptosis of MDA-MB-231 cells induced by the C4 (0, 0.25 × IC , 0.5 × IC , 1 ×
5
0
50
1
6

ACCEPTED MANUSCRIPT
2
44
45
IC , and 1 × IC ) for 48 h. b) All data were obtained from three independent experiments
5
0
50
2
and presented as the mean ± S.D. *P < 0.05，**P < 0.01 compared with control.
246
247
248
249
250
251
252
253
254
255
In MDA-MB-231 cells incubated with compound C4 at a 0.25 equipotent
concentration of IC50 for 48 h, the proportion of apoptotic (early + late) cells
increased to 2.07%, relatively slight compared to that of DMSO alone (1.51%). In
contrast, even at a low C4 dosage of 0.23 µM, the treatment resulted in 10.43% of the
MDA-MB-231 cells becoming necrotic, obviously higher than the 2.81% value of
necrotic cells in untreated control (Fig. 4). With the increase of C4 dosage to 0.5, 1,
and 2 equipotent concentrations of IC , there was dose-dependent enhancement of
5
0
both apoptotic (early + late) and necrotic cells to 8.76% and 38.88%, respectively. It is
apparent that the main cause of cell death in human breast adenocarcinoma was the
necrosis induced by compound C4.
256
257
258
259
260
261
262
263
264
265
2.2.4 Lactate dehydrogenase release assay
Lactate dehydrogenase (LDH) is a stable cytoplasmic enzyme that is found in all
cells. The leaking of LDH enzyme from the cytosol into the surrounding culture
medium is generally regarded as an indicator of necrosis, and reflects a loss of
membrane integrity in dead cells [38]. Because compound C4 failed to significantly
induce apoptosis, while the Annexin V-FITC/propidium iodide staining revealed that
C4 induced necrosis robustly, an LDH release assay was performed to further
investigate the effects of C4 on the LDH activity of treated MDA-MB-231 cells as
specified in the experimental section. After treatment of MDA-MB-231 cells with
compound C4 (0.23, 0.46, 0.91, and 1.82 µM for 48 h), there was somewhat elevation
1
7

ACCEPTED MANUSCRIPT
266
267
268
269
270
of LDH release levels in the culture medium (Fig. 5a). Especially in the cases where
the cancerous cells were exposed to C4 at concentration of 0.91 and 1.82 µM, the
levels of LDH release were almost 3.21 and 3.05 fold (P < 0.01) higher than that in
non-treated supernatant, respectively. With the increased levels of LDH release, we
confirmed that the necrosis of MDA-MB-231 cells was induced by C4.
271
272
273
274
275
Fig. 5. a) LDH release from MDA-MB-231 cells treated with different concentrations of C4
for 48ꢀh. b) MDA-MB-231 cells treated with various concentrations of C4 for 48ꢀh, and the
necrotic cells were counted by Trypan blue exclusion assay. The values are presented as mean
± S.D. of three independent experiments. *P < 0.05, **P < 0.01.
276
277
278
279
280
281
282
2.2.5 Trypan blue exclusion assay
The Trypan blue dye exclusion test was used to determine the number of viable
cells present in a cell suspension. It is based on the principle that live cells will
exclude membrane-impermeable dyes such as trypan blue, whereas trypan blue can
penetrate inside necrotic cells and stain them [39,40]. In the present study, after
independent treatment with C4 at concentrations of 0.23, 0.46, 0.91, and 1.82 µM for
48 h, the MDA-MB-231 cells were trypsinized and stained with 0.2% Trypan blue
1
8

ACCEPTED MANUSCRIPT
283
284
285
286
287
288
289
solution at 1:1 dilution for 5 min, and, thereafter, loaded into a hemocytometer for
separate counting of stained (necrotic) and unstained (viable) cells. As shown in Fig.
5b, complex C4 dramatically induced necrosis in a dose-dependent manner, and
almost 50% of MDA-MB-231 cells was identified with necrosis after C4 (1.82 µM)
incubation of 48 h, which is in a agreement with the results presented in Fig. 4 and
Fig. 5a. With such notation, the necrosis induced by antimonial C4 has been
repeatedly proven.
290
291
292
293
294
295
296
297
298
299
300
301
302
303
2.2.6 Reactive oxygen species (ROS) assay
Reactive oxygen species (ROS) are radicals, ions or molecules that have a single
unpaired electron in their outermost shell of electrons. Recent evidence has shown
that ROS play a key role as a messenger for signal transduction and cell cycling in
normal cells. However, unnecessary increases of ROS could result in cell death [41].
Previous works revealed that metal-based chemotherapeutic agents (e.g. cisplatin)
induced apoptosis and/or necrosis through ROS generation in several cancer cell lines
[34,42]. It is widely accepted that the anticancer effect of these chemotherapeutics is
due to the induction of oxidative stress and ROS-mediated cell injury in cancer. To
explore whether cell necrosis induced by C4 was dependent on the level of ROS, we
assessed the production of intracellular ROS in C4-treated MDA-MB-231 cells by
using 5-(and-6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) as
fluorescence probe and H O as a positive control. Intracellular ROS levels were
2
2
measured and quantified by change in relative fluorescence units (RFU). As shown in
1
9

ACCEPTED MANUSCRIPT
304
305
306
307
308
309
310
311
312
313
314
315
316
Fig. 6a, compared to the negative untreated control, treatment with C4 at graded
concentrations induced a significant elevation of ROS production in a dose-dependent
manner. When the exposure was 1.82 µM for 48 h, C4 promoted an increase of 4.61
fold (P < 0.01) of ROS production compared to that of untreated control. The
excessive increase of ROS production in cells might be attributed to the cytotoxic
activity of compound C4 through the activation of mitochondria-initiated events.
Furthermore, we examined the effect of N-acetylcysteine (NAC), a well-known
antioxidant, on the necrotic activity of C4-treated MDA-MB-231 cells by means of
Trypan blue dye exclusion test, as shown in Fig 6b. After pretreatment with NAC (5
µM) for 2 h, the population of necrotic MDA-MB-231 cells induced by C4 was
significantly decreased from 30.16% to 19.79%, evidencing the retarding effect of
NAC. It is consistent with the notion that the C4-driven induction of necrosis in
MDA-MB-231 cells is dependent on the production of intracellular ROS.
3
3
17
18
Fig. 6. a) ROS generation in C4 treated MDA-MB-231 cells. Relative fluorescence units
(RFU) of DCF was measured using a spectrofluorometer with excitation at 485ꢀnm and
emission at 530ꢀnm. b) Cells were pretreated with NAC (5 µM) for 2 h, followed by treatment
3
19
20
3
2
0

ACCEPTED MANUSCRIPT
321
322
323
with an IC concentration of C4 (0.91 µM) for 48 h before determination of cell death by
5
0
Trypan blue dye exclusion assay. The results are presented as mean ± S.D. (*P < 0.05, **P <
0.01).
2
1

ACCEPTED MANUSCRIPT
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
24 3. Conclusion
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
In the present study, a series of organoantimony(III) halide complexes with
azastibocine framework were synthesized, characterized and evaluated for in vitro
cytotoxic activity on human liver hepatocellular carcinoma cell line (HepG2), human
breast cancer cell line (MDA-MB-231), human breast adenocarcinoma cell line
(MCF-7), human cervical carcinoma cell line (HeLa) and human embryonic kidney
cell line (HEK-293). The results reveal a positive correlation between cytotoxic
activity and the N→Sb coordinate bond lengths in the complexes with same nitrogen
substituent. Among all tested trivalent organoantimony complexes, the complex
6-cyclohexyl-12-fluoro-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine
(C4)
exhibited the highest selectivity index. Further mechanistic investigation indicated
that C4 causes cell cycle arrest mainly at the S phase, and consequently results in
necrosis of MDA-MB-231 cells through the production of intracellular reactive
oxygen species. The results indicate that the azastibocine-framework organoantimony
halide complexes could be promising candidates for the development of new drugs in
cancer therapy. Specifically, the elucidation of correlation between cytotoxicity and
intermolecular interaction has provided theoretical and experimental basis for
in-depth design of antimony-based antineoplastic metallodrugs.
2
2

ACCEPTED MANUSCRIPT
342 4. Experimental
343
344
345
346
347
348
349
350
351
352
353
354
4.1. Chemistry
The commercially available starting materials were purchased from Adamas-beta,
and were used as received unless otherwise noted. The preparation of N-containing
precursors (P1–P3) and organoantimony(III) chlorides (C1, C4 and C9) were
prepared according to literature procedures [30,43]. Melting points were determined
over a XT-4 micro melting point apparatus (Beijing Tech Instrument Co., Ltd.).
Nuclear magnetic resonance (NMR) data were obtained on a Bruker-400 spectrometer
1
13
19
(400 MHz for H, 100 MHz for C and 376 MHz for F spectroscopy). Chemicals
shifts are reported in ppm (δ) relative to internal tetramethylsilane (TMS). In this
paper, data are reported as follows: Chemical shift, multiplicity (s = singlet, d =
doublet, t = triplet, q= quartet, m = multiplet). Coupling constants (J) are reported in
hertz. Elemental analyses were performed using a VARIO EL III instrument.
355
356
357
358
359
360
n
Scheme 1. Synthesis of organoantimony(III) halide complexes C1–C12. (a) BuLi (2.0
equiv.), Et O, -60 ℃ to room temperature. (b) SbCl (1.1 equiv.), Et O, -78 ℃ to room
2
3
2
temperature; C1, 72%; C5, 78%; C9, 80%. (c) KBr or KI (10.0 equiv.), CH Cl /H O, room
2
2
2
temperature; C2, 92%; C3, 86%; C6, 87%; C7, 85%; C10, 91%; C11, 81%. (d) AgF (1.0
equiv.), CH Cl /H O, room temperature, dark; C4, 93%; C8, 95%; C12, 76%.
2
2
2
361
4.1.1. General procedure for the synthesis of organoantimony(III) chlorides
2
3

ACCEPTED MANUSCRIPT
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
To a solution of N-containing precursors (25 mmol) in anhydrous ether (100 mL),
n-BuLi in hexane (2.5 M, 20 mL) was added dropwise under nitrogen with vigorous
stirring at -30 °C, and the mixture thus obtained was warmed to room temperature for
3 h. Then a solution of SbCl (5.8 g, 25.5 mmol) in anhydrous ether (60 mL) was
3
added slowly at -78 °C within 30 min. After subject to stirring at room temperature
overnight, the solution was subject to evaporation under reduced pressure. The
residue was extracted with dichloromethane (100 mL) and washed with a solution of
NH Cl (1M) in H O. The organic layer was washed with deionized water (3 × 50
4
2
mL), dried over anhydrous Na SO , subject to filtration, and concentrated in vacuo.
2
4
The residue was purified by recrystallization in CH Cl /n-hexane mixture to afford
2
2
the corresponding organoantimony chloride as colorless crystals.
4.1.2. General procedure for the synthesis of organoantimony(III) bromides
To a solution of organoantimony chloride (5.0 mmol) in CH Cl (30 mL), a solution
2
2
of KBr (6.0 g, 50 mmol) in deionized water (25 mL) was added under open-flask
conditions. After being stirred at room temperature overnight, the solution was
extracted with CH Cl (3 × 30 mL). The combined organic layer was dried over
2
2
anhydrous Na SO , and subject to filtration. The filtrate was mixed with n-hexane and
2
4
kept at room temperature for 24 h to afford the corresponding organoantimony
bromide as colorless crystals.
4.1.3. General procedure for the synthesis of organoantimony(III) iodides
To a solution of organoantimony chloride (5.0 mmol) in CH Cl (30 mL), a solution
2
2
of KI (8.3 g, 50 mmol) in deionized water (25 mL) was added under open-flask
2
4

ACCEPTED MANUSCRIPT
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
conditions. After subject to stirring at room temperature overnight, the solution was
extracted with CH Cl (3 × 30 mL). The combined organic layer was dried over
2
2
anhydrous Na SO , and subject to filtration. The filtrate was mixed with n-hexane and
2
4
kept at room temperature for 24 h to afford the corresponding organoantimony iodide
as colorless crystals.
4.1.4. General procedure for the synthesis of organoantimony(III) fluorides
To a solution of organoantimony chloride (5.0 mmol) in CH Cl (30 mL), a
2
2
solution of AgF (635 mg, 5.0 mmol) in deionized water (25 mL) was added
under open-flask conditions. After being stirred in the dark at room temperature
for 4 h, the mixture was subject to filtration. The filtrate was mixed with
n
-hexane and kept at room temperature for 24 h to afford the corresponding
organoantimony fluoride as colorless crystals.
4.1.5. Analytical data for the synthesized organoantimony halide complexes
12-chloro-6-cyclohexyl-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine
(C1).
1
Yield: 72% (7.8 g). Melting point: 254–256 °C. H NMR (400 MHz, CDCl , TMS): δ
3
8.30 (2H, d, J = 7.4 Hz), 7.39–7.27 (4H, m), 7.14 (2H, d, J = 7.4 Hz), 4.15 (4H, dd, J
= 61.9, 15.1 Hz), 3.09 (1H, t, J = 11.3 Hz), 2.05 (2H, d, J = 11.7 Hz), 1.86 (2H, d, J =
13
12.5 Hz), 1.69 (d, J = 13.0 Hz), 1.46–1.27 (m, 4H), 1.18–1.08 (m, 1H); C NMR
(100 MHz,CDCl , TMS): δ 144.0, 140.1, 134.9, 128.8, 128.7, 124.7, 65.4, 57.8, 29.5,
3
25.6, 25.4. Anal. Calc. for C H ClNSb: C, 55.27; H, 5.33; N, 3.22. Found: C, 55.39;
20
23
H, 5.25; N, 3.14.
12-bromo-6-cyclohexyl-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine
(C2).
2
5

ACCEPTED MANUSCRIPT
1
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
Yield: 92% (2.2 g). Melting point: 245–247 °C. H NMR (400 MHz, CDCl , TMS): δ
3
8.40–8.38 (2H, m), 7.37–7.27 (4H, m), 7.13 (2H, d, J = 6.6 Hz), 4.15 (4H, dd, J =
58.0, 15.1 Hz), 3.13–3.08 (1H, m), 2.04 (2H, d, J = 10.9 Hz), 1.85 (2H, d, J = 11.8
13
Hz), 1.68 (1H, d, J = 11.8 Hz), 1.45–1.27 (4H, m), 1.17–1.09 (1H, m); C NMR (100
MHz,CDCl , TMS): δ 144.0, 138.0, 136.3, 128.9, 128.7, 124.7, 65.5, 57.8, 29.6, 25.6,
3
25.4. Anal. Calc. for C H BrNSb: C, 50.14; H, 4.84; N, 2.92. Found: C, 50.29; H,
2
0
23
4.95; N, 2.99.
6-cyclohexyl-12-iodo-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine (C3). Yield:
1
86% (2.3 g). Melting point: 253–255 °C. H NMR (400 MHz, CDCl , TMS): δ 8.52
3
(2H, d, J = 7.0 Hz), 7.39–7.30 (4H, m), 7.12 (2H, d, J = 6.9 Hz), 4.14 (4H, dd, J =
52.4, 15.0 Hz), 3.12 (1H, t, J = 11.4 Hz), 2.07 (2H, d, J = 11.2 Hz), 1.88 (2H, d, J =
1
3
12.4 Hz), 1.71 (1H, d, J = 13.2 Hz), 1.47–1.31 (4H, m), 1.19–1.13 (1H, m); C NMR
(100 MHz,CDCl , TMS): δ 143.8, 139.4, 134.5, 129.2, 129.0, 124.7, 65.6, 57.6, 29.7,
3
25.7, 25.4. Anal. Calc. for C H INSb: C, 45.66; H, 4.41; N, 2.66. Found: C, 45.74;
2
0
23
H, 4.54; N, 2.79.
6-cyclohexyl-12-fluoro-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine
(C4).
1
Yield: 93% (1.9 g). Melting point: 237–239 °C. H NMR (400 MHz, CDCl ,
3
TMS):
4.09 (4H, dd,
Hz), 1.86 (2H, d,
1.18–1.09 (1H, m); C NMR (100 MHz,CDCl , TMS):
δ
7.97 (2H, d,
= 71.4, 15.2 Hz), 3.03 (1H, t,
= 12.5 Hz), 1.69 (2H, d,
J
= 7.3 Hz), 7.40–7.25 (4H, m), 7.14 (2H, t,
= 10.8 Hz), 2.00 (2H, d,
= 12.7 Hz), 1.45–1.26 (4H, m),
144.1, 144.0 (d,
= 6.0 Hz), 128.5, 128.3, 124.7, 65.2, 57.7 (d, = 2.1 Hz),
J
= 7.4 Hz),
J
J
J
= 11.4
J
J
13
δ
J =
3
6.4 Hz), 133.4 (d,
J
J
2
6

ACCEPTED MANUSCRIPT
29.5, 25.7, 25.5; F NMR (376 MHz, CDCl3):
C H FNSb: C, 57.45; H, 5.54; N, 3.35. Found: C, 57.53; H, 5.59; N, 3.44. FT-IR
19
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
δ -185.45. Anal. Calc. for
20
23
-
1
(KBr, cm ):
υ 2935, 2857, 1455, 1440, 1268, 1204, 1100, 975, 950, 934, 896, 758.
12-chloro-6-phenyl-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine (C5). Yield:
1
76% (8.1 g). Melting point: 222–224 °C. H NMR (400 MHz, CDCl , TMS): δ 8.40–
3
8.28 (2H, m), 7.47–7.42 (2H, m), 7.38–7.34 (4H, m), 7.30–7.19 (5H, m), 4.66 (4H, dd,
13
J = 70.8, 14.9 Hz); C NMR (100 MHz,CDCl , TMS): δ 147.9, 143.1, 140.5, 135.2,
3
129.5, 129.2, 129.2, 125.4, 125.3, 119.7, 61.2. Anal. Calc. for C H ClNSb: C, 56.05;
2
0
17
H, 4.00; N, 3.27. Found: C, 55.73; H, 4.12; N, 3.35.
12-bromo-6-phenyl-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine (C6). Yield:
1
87% (2.0 g). Melting point: 233–235 °C. H NMR (400 MHz, CDCl , TMS): δ 8.41–
3
8.29 (2H, m), 7.47–7.43 (2H, m), 7.40–7.35 (4H, m), 7.31–7.29 (2H, m), 7.25–7.21
1
3
(3H, m), 4.67 (4H, dd, J = 65.7 ,14.9 Hz); C NMR (100 MHz,CDCl , TMS) δ 147.8,
3
143.0, 138.2, 136.8, 129.6, 129.4, 129.3, 125.5, 125.3, 119.8, 61.2. Anal. Calc. for
C H BrNSb: C, 50.78; H, 3.62; N, 2.96. Found: C, 50.89; H, 3.69; N, 3.07.
20
17
12-iodo-6-phenyl-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine (C7). Yield:
1
85% (2.2 g). Melting point: 249–251
°
C. H NMR (400 MHz, CDCl , TMS):
δ
3
8.47–8.45 (2H, m), 7.39–7.30 (6H, m), 7.26–7.23 (2H, m), 7.19–7.16 (3H, m),
1
3
4.60 (4H, dd,
142.9, 140.0, 134.1, 129.6, 129.5, 129.4, 125.5, 125.3, 119.8, 60.9. Anal. Calc.
for C H INSb: C, 46.19; H, 3.30; N, 2.69. Found: C, 46.27; H, 3.33; N, 2.76
J = 66.6, 14.9 Hz); C NMR (100 MHz,CDCl , TMS): δ 147.6,
3
.
2
0
17
2
7

ACCEPTED MANUSCRIPT
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
12-fluoro-6-phenyl-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine (C8). Yield:
1
95% (2.0 g). Melting point: 216–218 °C. H NMR (400 MHz, CDCl , TMS):
δ
3
8.00 (2H, d,
J
= 7.3 Hz), 7.46 (2H, t,
J
= 7.3 Hz), 7.35–7.34 (4H, m), 7.20–7.25
13
(4H,m), 7.20 (1H, t,
J
= 6.8 Hz), 4.63 (4H, dd, J = 85.1, 15.0 Hz); C NMR
(100 MHz,CDCl , TMS):
δ
148.5, 144.7 (d,
6.2 Hz), 129.5, 128.9, 128.8, 125.2, 125.0, 119.4, 61.2 (d,
NMR (376 MHz, CDCl3):
4.16; N, 3.40. Found: C, 58.39; H, 4.24; N, 3.58
J
= 6.6 Hz), 143.1, 133.4 (d,
J
=
3
1
9
J
= 1.8 Hz); F
δ
-198.97. Anal. Calc. for C H FNSb: C, 58.29; H,
20 17
.
6-(tert-butyl)-12-chloro-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine
(C9).
1
Yield: 80% (8.1 g). Melting point: 213–215 °C. H NMR (400 MHz, CDCl , TMS): δ
3
8.32 (2H, d, J = 7.3 Hz), 7.37 (2H, t, J = 7.2 Hz), 7.31–7.28 (2H, m), 7.15 (2H, d, J =
13
7.5 Hz), 4.23 (4H, dd, J= 161.2, 15.4 Hz), 1.38 (9H, s); C NMR (100 MHz,CDCl ,
3
TMS): δ 145.1, 139.9, 134.9, 128.9, 128.4, 124.5, 60.4, 57.2, 27.0. Anal. Calc. for
C H ClNSb: C, 52.91; H, 5.18; N, 3.43. Found: C, 52.78; H, 5.26; N, 3.52.
18
21
12-bromo-6-(tert-butyl)-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine (C10).
1
Yield: 91% (2.0 g). Melting point: 244–246 °C
.
H NMR (400 MHz, CDCl , TMS):
3
δ 8.41 (2H, d, J= 7.2 Hz), 7.36–7.27 (4H, m), 7.15 (2H, d, J= 6.8 Hz), 4.22 (4H, dd,
1
3
J= 152.9, 15.4 Hz), 1.37 (9H, s); C NMR (100 MHz,CDCl , TMS): δ 145.0, 137.8,
3
136.4, 129.0, 128.5, 124.5, 60.6, 57.2, 27.1. Anal. Calc. for C H BrNSb: C, 47.72; H,
1
8
21
4.67; N, 3.09. Found: C, 47.82; H, 4.75; N, 3.16.
6-(tert-butyl)-12-iodo-5,6,7,12-tetrahydrodibenzo[c,f][1,5]azastibocine
(C11).
1
Yield: 81% (2.0 g). Melting point: 228–230
°
C. H NMR (400 MHz, CDCl ,
3
2
8