A new rhodium(I) NHC complex inhibits TrxR: In vitro cytotoxicity and in vivo hepatocellular carcinoma suppression

Article abstract of DOI:10.1016/j.ejmech.2019.111721

Thioredoxin reductase (TrxR) is often overexpressed in different types of cancer cells including hepatocellular carcinoma (HCC) cells and regarded as a target with great promise for anticancer drug research and development. Here, we have synthesized and characterized nine new designed rhodium(I) N-heterocyclic carbene (NHC) complexes. All of them were effective towards cancer cells, especially complex 1e was more active than cisplatin and manifested strong antiproliferative activity against HCC cells. In vivo anticancer studies showed that 1e significantly repressed tumor growth in an HCC nude mouse model and ameliorated liver lesions in a chronic HCC model caused by CCl₄. Notably, a mechanistic study revealed that 1e can strongly inhibit TrxR system both in vitro and in vivo. Furthermore, 1e promoted intracellular ROS accumulation, damaged mitochondrial membrane potential, promoted cancer cell apoptosis and blocked the cells in the G1 phase.

Full text of DOI:10.1016/j.ejmech.2019.111721

European Journal of Medicinal Chemistry 183 (2019) 111721
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
A new rhodium(I) NHC complex inhibits TrxR: In vitro cytotoxicity and
in vivo hepatocellular carcinoma suppression
Rong Fan 1, Mianli Bian a 1, Lihong Hu , Liu Wukuna
a
,
,
a
, b, c, *
a
School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing, 210023, China
State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing, 210009, China
State Key Laboratory of Coordination Chemistry, Nanjing University, Nanjing, 210023, China
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 June 2019
Received in revised form
Thioredoxin reductase (TrxR) is often overexpressed in different types of cancer cells including hepa-
tocellular carcinoma (HCC) cells and regarded as a target with great promise for anticancer drug research
and development. Here, we have synthesized and characterized nine new designed rhodium(I) N-het-
erocyclic carbene (NHC) complexes. All of them were effective towards cancer cells, especially complex
27 August 2019
Accepted 18 September 2019
Available online 21 September 2019
1
e was more active than cisplatin and manifested strong antiproliferative activity against HCC cells.
In vivo anticancer studies showed that 1e significantly repressed tumor growth in an HCC nude mouse
model and ameliorated liver lesions in a chronic HCC model caused by CCl . Notably, a mechanistic study
4
Keywords:
revealed that 1e can strongly inhibit TrxR system both in vitro and in vivo. Furthermore, 1e promoted
intracellular ROS accumulation, damaged mitochondrial membrane potential, promoted cancer cell
apoptosis and blocked the cells in the G1 phase.
Rhodium(I) NHC complexs
Thioredoxin reductase
Apoptosis
Reactive oxygen species
Hepatocellular carcinoma
© 2019 Elsevier Masson SAS. All rights reserved.
1. Introduction
mechanism. The thioredoxin (Trx) system is one of the most
important members in the redox control system, which comprises
Hepatocellular carcinoma (HCC) is a commonly malignant can-
cer with high mortality, and its death rate ranks third among all
cancer death rates in China [1,2]. More specifically, the number of
HCC cases in China occupied a large proportion of the HCC cases in
the world [3]. And the five-year survival rate of HCC is below 5%,
which is one of the worst prognosis cancers [4]. Oral sorafenib is
currently the only drug that significantly prolongs the survival of
patients with advanced HCC. However, none of the new molecu-
larly targeted drugs, including sorafenib, that has been approved
for marketing, have achieved very satisfactory results. Therefore,
improved HCC treatment methods still need to be developed [5,6].
Redox dysregulation stemming from metabolic alterations and
relying on mitogenic and survival signals through reactive oxygen
species (ROS) embodies an exceptional sensitivity to tumor cells
that can be selectively targeted by redox chemotherapeutic agents
NAPDH, Trx and thioredoxin reductase (TrxR) [8e11]. As a redox
regulatory protein, Trx can be oxidized by abundant ROS.
TrxR activates Trx by reduction and provides electrons in
numerous oxidation/reduction cycles. Many studies have shown
that TrxR involved in multifarious physiological and pathological
processes in bodies, including apoptosis, cancer and chronic
inflammation [12].
In previous years, several studies have reported that Trx and
TrxR are overexpressed in HCC [13,14]; TrxR contributes to tumor
growth through the hypoxia-inducible factor 1 pathway and shows
a crucial role in maintaining cancer phenotype and metastasis [14].
Targeted inhibition of TrxR leads to the production and accumu-
lation of ROS in HCC cells, inhibition of cell proliferation and
mitochondrial membrane damage [13]. Accumulating evidences
demonstrate that TrxR/Trx is an important modulator in tumor
development [10,15]. In the nude mouse xenograft model of HCC
cells, the progressions of tumor were slower and the sizes of tumor
were smaller in the TrxR knockdown tumor group than in the TrxR
normal tumor group [16e18]. Therefore, targeting and inhibiting
TrxR/Trx is a hopeful strategy for HCC treatment.
[
7]. Redox control has become an essential biological regulatory
*
Corresponding author. School of Pharmacy, Nanjing University of Chinese
Medicine, Nanjing, 210023, China ,
E-mail addresses: liuwukun0000@njucm.edu.cn, liuwukun0000@hotmail.com
W. Liu).
Rong Fan and Mianli Bian contributed equally to this work.
As Trx/TrxR is involved in the pathological process of cancer,
inhibition of TrxR is a significant clinical goal. Recently, some
(
1
https://doi.org/10.1016/j.ejmech.2019.111721
223-5234/© 2019 Elsevier Masson SAS. All rights reserved.
0

2
R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
important studies have reported TrxR inhibitors (e.g. metal-
containing inhibitors) [10,19e23], some of which exhibited strong
TrxR inhibition effects and good antiproliferative effects on many
tumor cells, such as gold(I) NHC complexes [24e31]. Interestingly,
rhodium(I) NHC compounds, which possess a square-planar ge-
ometry as cisplatin, also inhibited TrxR activity in Ott and co-
workers’ studies [32,33]. The ion rhodium(I) is isoelectronic with
platinum(II).
NMR spectra of ligands (1d-9d) and complexes (1e-9e) were
characteristic of complex formation, as shown in Table S1. The
carbene carbon chemical shift increased by approximately 45 ppm
after coordination to rhodium. From the positive ESI mass spec-
þ
trum, we can observe the peak of [M-X] (where X ¼ Br or Cl) for
þ
compounds 1d-9d and [M-Cl] peaks for complexes 1e-9e.
Crystals of 1e were grown by slowly diffused n-hexane into a
thick 1e dichloromethane solution. As shown in Fig. 1, the Rh-
Generally, TrxR inhibitors display anticancer effects by pro-
moting the apoptosis of tumor cells while HCC is closely related to
the overexpression of TrxR. To develop an inhibitor that can
effectively inhibit the overexpression of TrxR in HCC cells, here, we
present the synthesis, characterization, and cytotoxicity of nine
rhodium(I) NHC compounds possessed pharmacologically active
Ccarbene distance of 1e is 1.945 Å. Crystallographic of 1e has given in
Table S2.
2.2. Interaction with GSH and NAC
4
,5-diarylimidazoles (Schemes 1 and 2, Table 1) [28,34,35]. The
For further study, the stability of the rhodium(I) NHC complexes
in the presence of biologically important thiols were examined. We
tested the reactivity of 1e in the presence of glutathione (GSH), N-
most promising rhodium(I) complex can inhibit TrxR, increase
intracellular ROS accumulation and promote HCC cells apoptosis. In
addition, this complex has also been shown to significantly sup-
press tumor growth in an HCC nude mouse model and ameliorate
3
acetylcysteine (NAC) and deuteroxide in acetonitrile-d by exam-
1
ining the H NMR spectra (Fig. S1). When 1e was treated with
equivalent GSH, several changes in the spectra (Fig. S1A) could be
observed over time: at 5.53 ppm, 2.92 ppm, 3.21 ppm new peaks
appeared, in which the 5.53 ppm peak was enhanced over time, and
according to the literature that the peaks at 3.26 ppm were attrib-
4
liver lesions in a chronic HCC model caused by CCl .
2
. Results and discussion
uted to the
reactions between 1e and GSH. When 1e was treated with NAC
Fig. S1B), the peak at 5.53 ppm also appeared, and a series of peaks
b-CHcys groups of GSSG [37], indicating the existence of
2.1. Synthesis and characterization
(
The 4,5-diarylimidazole precursors were synthesized in terms of
literature methods [28,34,35], and the rhodium(I) NHC complexes
were prepared by the classical silver conversion method (Scheme
(8.68 ppm, 7.40 ppm, 7.22 ppm, 4.66, and 4.07 ppm) appeared and
were enhanced over time. The 5.53 ppm peak may be the RheS
complex signal, and the 8.68 ppm, 7.40 ppm, 7.22 ppm, 4.66 ppm,
and 4.07 ppm peaks may be attributed to the NHC ligand signals,
which indicated that 1e can react with NAC and lose its NHC li-
gands. As a control, the stability of complex 1e in deuteroxide was
2
) [32,33,36]. NHCs (1d-9d) were treated with a half equivalent of
Ag O in dichloromethane to generate a silver intermediate; then, a
2
half equivalent of bis[chloride(1,5-cyclooctadiene) rhodium(I)] was
added, and approximately 6 h later, the rhodium(I) NHC complexes
were obtained and purified by filtration over Celite, column chro-
matography and recrystallization. The structures were analyzed by
1
tested (Fig. S1C). The H NMR spectra of 1e in deuteroxide was not
observed significant changes after 96 h, indicating that 1e was
1
relatively stable in deuteroxide. The H NMR spectra changes of 1e
1
13
H, C NMR, and ESI-MS, and the purity of the complexes deter-
and GSH/NAC were caused by the reaction between 1e and GSH/
NAC, rather than by the reaction with deuteroxide. From these
experimental results, we speculated that 1e could interact with
mercapto biomacromolecules.
1
mined by elemental analysis. Compared to 1d-9d, the H NMR
spectra of 1e-9e appeared the proton signals of 1,5-cyclooctadiene.
13
The chemical shift changes of NHC-Ccarbene signals between the
C
Scheme 1. The structures of 1e-9e.

R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
3
Scheme 2. Synthesis routes for rhodium(I) NHC complexes (1e-9e).
Table 1
The substituent groups of compounds 1a-1e, 2a-2e, 3a-3e, 5d-5e, 6d-6e, 7d-7e, 8d-
8e.
Compounds
R
F
MeO
Br
1
R
2
X
1a-1e
2a-2e
3a-3e
5d-5e
Et
Et
Et
Br
Br
Br
Cl
F
6
7
d-6e
d-7e
F
F
Br
Cl
8d-8e
F
Cl
2.3. In vitro antiproliferative activity
Complexes 1e-9e were tested for antiproliferative activity in
HCC cells (HepG2), breast cancer cells (MCF-7), and human colon
cancer cells (HT-29) by the MTT assay. Cisplatin was used as the
positive control. As outlined in Table 2, 1e showed lower IC50 values
in all three carcinoma cell lines compared to cisplatin. When the
aromatic ring para location is substituted by fluorine, methoxy,
bromine or a benzene ring (1e-4e), a smaller substituent size leads
to stronger antiproliferative activity against MCF-7 cells and HepG2
cells (size of substituents: F<Br<OMe < Ph; antiproliferative activ-
ity: F(1e) > Br(3e) > OMe(2e) > Ph(4e)). The substituents at the N-
atoms seemed to play a large role in anticancer activities, because
Fig. 1. The crystal structure of 1e.
to the proper lipophilicity of the complexes being the key for the
antiproliferative activities. Complex 9e was less active than 1e-8e
against all three cell lines, possibly because the 4,5-diaryl moiety
expanded the conjugated region of the NHC and increased its
stability.
compounds with an N
or higher antiproliferative activity than those with an N
chain (5e, 7e-8e) against all three cell lines. This result may be due
1
-aliphatic chain (1e, 6e) exhibited equivalent
In addition, the antiproliferative activity of 1e (the most active
1
-aromatic
complex against HepG2 cells with an IC50 value of 1.33
mM) in two
other HCC cell lines and one normal liver cell line (LO
2
) were

4
R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
Table 2
imidazolium salt 1d against purified TrxR enzyme were measured
by the DTNB assay [38]. Compared to non-rhodium-containing
The antiproliferative activity of 1e-9e against MCF-7, HT-29 and HepG2 cell lines
IC50 values, Mean ± SD [ M]).
(
m
compound 1d (EC50 > 10
mM), 1e displayed a stronger inhibitory
Compound
MCF-7
HT-29
HepG2
effect on TrxR enzyme (EC50: 1.89 ± 0.44 mM), suggesting TrxR may
Cisplatin
4.95 ± 1.16
4.47 ± 1.96
6.21 ± 0.89
4.12 ± 0.98
7.72 ± 0.79
17.58 ± 3.12
3.83 ± 0.21
5.54 ± 0.92
3.72 ± 0.84
16.19 ± 2.13
9.51 ± 2.02
6.51 ± 1.24
11.33 ± 2.14
6.78 ± 2.12
5.17 ± 1.24
13.44 ± 1.68
6.04 ± 1.61
6.48 ± 1.61
7.36 ± 2.12
15.63 ± 2.45
2.18 ± 0.23
1.33 ± 0.21
5.84 ± 0.61
4.96 ± 1.86
8.87 ± 2.79
6.01 ± 2.48
4.04 ± 1.92
4.46 ± 1.45
3.86 ± 0.02
16.55 ± 2.89
indeed be a biological target for 1e. In addition, we also measured
the TrxR inhibitory activity of 1e in cells.
Briefly, different concentrations of 1e were used to dispose of
HepG2 cells for 48 h and 72 h. A TrxR Assay Kit was used to measure
the activity of TrxR and 1e showed TrxR inhibitory activity (with an
1e
2e
3e
4e
5e
6e
7e
8e
9e
IC50 z 2.5 mM) both dose- and time-dependently, indicating that 1e
may represent a fresh kind of TrxR inhibitor with potential for
further research. Besides, we investigated the effect of 1e on
intracellular mRNA levels of TrxR in HepG2 cells by real-time PCR
analysis. As shown in Fig. 2C, 1e effectively reduced the expression
of TrxR. These alterations were confirmed by immunofluorescence
analysis (Fig. 2D). Considering that the underlying mechanism of 1e
may inhibit the overexpressed TrxR in HCC cells, we subsequently
examined the relationship between TrxR expression levels and 1e
treatment concentrations in drug-sensitive HepG2 cells by western
blotting (Fig. 2B). TrxR expression was upregulated in HCC cells, and
the expression was decreased by 1e in a dose-dependent manner.
In summary, all these results confirm that TrxR could be a possible
biological target for complex 1e.
evaluated. As showed in Fig. S2A, 1e exhibited strong anti-
proliferative activity (IC50 values of 1.78 M in Hep3B, 4.01 M in
SMMC-7721) in HCC cell lines. However, no selectivity was
observed for HCC cells versus LO M). To
normal cells (IC50 ¼ 2.18
obtain a safe dose of 1e for further study, we evaluated the ALT and
LDH levels in a normal liver cell line (LO ) (Figs. S2B and S2C). The
doses of 1.5 M, 3 M, and 6 M of 1e caused minimal harm to
m
m
2
m
2
m
m
m
normal cells; in other words, these doses are safe. Thus, follow up
studies were mainly focused on complex 1e.
2.5. Fluorescence microscopy
2.4. TrxR inhibition
The cellular uptake and distribution of drugs in tumor cells is the
As mentioned above, TrxR was overexpressed in various cancer
key to the anticancer activity of drugs [39]. To investigate the
subcellular distribution of rhodium(I) NHC complexes, we chose
complex 5e, which possesses a large conjugated structure and can
emit fluorescence at a specific wavelength. The emission spectra
and excitation spectra of 5e are shown in Fig. S3. Due to the intrinsic
cells including HCC cells and studies have shown that rhodium(I)
NHC complexes can bind with TrxR [32,33,36]. Thus, we further
tested the reactivity of 1e with TrxR in purified TrxR enzyme and
HepG2 cells (Fig. 2A). The inhibitory activity of TrxR by 1e and
ꢀ
Fig. 2. Quantification of TrxR activity in HepG2 cells treated with 1e at 37 C for 48 h and 72 h (A), Western blot of HepG2 cells treated with 1e for 24 h (B), mRNA levels of TrxR in
HepG2 cells treated with 1e for 24 h (C), immunofluorescence analysis of HepG2 cells treated with 1e for 24 h (D, original magnification, 40 ꢁ ). Error bars: S.D., n ¼ 3. statistical
significance of differences in mean values: *p < 0.05 and ***p < 0.001.

R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
5
emission of 5e, the localization of the complex in HepG2 cells was
further investigated by fluorescence microscopy. After incubating
analyse the production of ROS in 1e-treated HepG2 cells. The level
of ROS generation was higher in cells treated with 1e (6 M) than in
m
HepG2 cells with 5e (10
m
M) for 24 h and Red-Mito-Tracker dye
untreated cells (Fig. 3C and D). Then, we tested the protection of
HepG2 cells by NAC when HepG2 cells were incubated with the
indicated concentrations of NAC and 1e for 48 h. NAC as an anti-
oxidant can remove excess ROS in cells [41]. Interestingly, treat-
ment with 1e prominently reduced the viability of HepG2 cells
(Fig. S4), which was partially or completely abrogated by treatment
with 1 mMe4 mM NAC, respectively. These findings suggested that
1e can activate ROS-induced apoptosis in HepG2 cells.
(
200 nM, a mitochondria-specific reactive dye) for 0.5 h, we
observed an obvious blue fluorescence in the cytoplasm. By the
merge of all the luminescent spots (caused by both 5e and Red-
Mito-Tracker dye), we can see that the blue fluorescence was spe-
cifically localized in mitochondria (Fig. 3A). Accordingly, we can
speculate that rhodium(I) NHC complexes were mainly distributed
in mitochondria and may interact with mitochondrial TrxR.
2.7. MMP and ATP
2.6. ROS analysis
Studies have shown that high levels of ROS can trigger mito-
Intracellular accumulation of ROS, which damages DNA through
chondrial membrane permeabilization and mitochondrial mem-
brane potential depolarization, leading to mitochondrial
dysfunction and activation of intrinsic apoptotic pathway [42]. We
used the JC-1 kit to investigate MMP upon treatment with 1e. After
treatment with 1e, we observed a dose-dependent decrease in
mitochondrial membrane polarization (orange fluorescence turned
oxidative stress, disrupts mitochondrial membrane potential,
causing mitochondrial dysfunction leading to apoptosis and cell
death [40]. Trx acts as an intracellular ROS scavenger whose activity
is affected by TrxR activity. When the activity of TrxR was inhibited,
the activity of Trx was also inhibited, resulting in accumulation of
ROS in cancer cells. Thus, HepG2 cells were inoculated with 1e
green fluorescence). In particular, 6
to the membrane compared with the control group, as expected
Fig. 3E).
MMP is vital to ATP generation and ATP is essential to the efflux
of the drug, while ATP is derived from ATP hydrolysis by ATPases
43]. To verify whether our compound 1e would affect ATPase, we
treated HepG2 cells with 1e at different concentrations (1.5 M,
M), and the results showed that samples treated with
e had significantly lower ATPase activity than that in the DMF
mM of 1e led to higher damage
(
1.5 mM, 3 mM, and 6 mM) for 24 h, fluorescence analysis was used to
(
[
m
3
1
mM, and 6 m
control (Fig. 3B). These data indicated that 1e inhibited ATPase
activity.
2.8. Release of cytochrome c and apoptosis
Study has demonstrated that most metal-based antitumor drugs
normally kill tumor cells by activating the apoptotic pathway [44].
In addition, ROS can promote a series of mitochondria-related
events such as apoptosis by affecting MMP. Thus, the apoptosis of
HepG2 cells induced by 1e was researched by propidium iodide (PI)
staining and FITC-Annexin V staining. For HepG2 cells treated with
1e, we measured the state of apoptosis by flow cytometry. HepG2
cells were treated with 1e for 72 h, separately. The results
demonstrated that 1e enhanced apoptosis in HepG2 cells in a dose-
dependent manner (Fig. 4A). The initial synthetic caspases in the
cells are inactive zymogen form and then activated by hydrolysis of
the N-terminal sequence. Activated caspase can hydrolyze the
corresponding substrate and induce apoptosis by cascade amplifi-
cation [45]. Thus, we also investigated the protein levels of the
cleaved active forms of caspases. The results demonstrated that 1e
dose-dependently upregulated the expression of apoptosis-related
factors Cleaved-Caspase-3 and Cleaved-Caspase-9 (Fig. 4B).
Hoechst 33258 staining was used to verify the ability of 1e-induced
apoptosis by analysing the morphology of the bare cell nucleus
(Fig. 4C). HepG2 cells were treated with 1e for 24 h and then
stained by Hoechst 33258 morphologically detecting apoptosis via
fluorescence microscopy. The control cells displayed a weak regular
blue fluorescence. While in 1e-treated group, bright blue fluores-
cence was observed in most HepG2 cells, and the nuclei appeared
hyper condensed (brighter blue stained). When HepG2 cells were
treated with complex 1e (6 mM) for 24 h, the obviously increased
Fig. 3. The uptake of 5e (10
m
M) in HepG2 cells (A, original magnification, 40 ꢁ ),
apoptotic nuclei containing condensed chromatin demonstrated
that apoptosis of HepG2 cells was occasioned by 1e in a dose-
dependent way. These results suggested that 1e induced
apoptosis of HepG2 cells along with caspase activation.
ATPase activity in HepG2 cells after treated with 1e for 24 h (B), ROS positive cells after
treated with 1e for 24 h (C), ROS generation after treated with 1e for 24 h (D, original
magnification, 20 ꢁ ), mitochondrial membrane damage imaged by JC-1 staining after
treated with 1e for 24 h (E, original magnification, 40 ꢁ ). Error bars: S.D., statistical
significance of differences in mean values: *p < 0.05, **p < 0.01 and ***p < 0.001.
Apoptosis leads to changes in mitochondrial membrane

6
R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
Fig. 4. Apoptosis of HepG2 cells treated with 1e at different concentrations for 72 h (A), Western blot analysis of HepG2 cells after treatment with 1e at different concentrations for
4 h (B), and Hoechst 33258 staining after treatment with 1e for 24 h (C, original magnification, 40 ꢁ ). Error bars: S.D., n ¼ 3. Statistical significance of differences in mean values:
p < 0.05, **p < 0.01 and ***p < 0.001.
2
*
permeability, AIF/Bak transfers from mitochondria to nucleus.
Increased Bax/Bcl-2 ratio promotes cytochrome c set free into the
cytosol and further triggers the cysteine protease cascades. PARP
cleavage involves the response of cells to apoptotic signal stimu-
lation and the repair of DNA damage, and serves as a marker of
apoptosis. Therefore, the effects of 1e on PARP cleavage was
examined by Western blot assay. As shown in Fig. 5A and B, the
results demonstrated that 1e enhanced the apoptotic family-
related proteins localized to the mitochondrial inner membrane
in HepG2 cells in a dose-dependent manner. These were consistent
with the findings from FITC Annexin V staining.
A dose-dependent expression level decrease was caused by the
treatment with 1e. Additionally, the upregulation of cell cycle
control inhibitors of Cyclin-CDK complexes (p15, p21 and p27)
causes to cell cycle arrest in the G0/G1 phase. Particularly, over-
expression of p21 and p27 activates Bax through different inhibi-
tion signals, attenuating cancer cell proliferation and inducing
apoptosis (Fig. 6C).
2.10. Interaction with DNA
It is widely accepted that DNA symbolizes a primary target of
anticancer metal compounds [46e52] and rhodium(I) NHC com-
plexes have also been proven to target DNA [53e55]. To more
clearly find out the anticancer activity of our rhodium(I) NHC
complexes, the binding ability of 1e to DNA was investigated by
UVeVis absorption, fluorescence emission and CD spectroscopy
(Fig. S5). Sodium dodecyl sulfate (SDS) was added as a probe to
confirm whether 1e interacts with DNA via electrostatic in-
teractions. When 1e was combined with SDS, the UV spectra
changed. No significant distinction was observed between the ab-
sorption spectra of 1e in the absence and presence of SDS (Fig. S5A).
2
.9. Cell cycle analysis
Next, the result of 1e on cell cycle progression was studied by PI
staining. Cells were treated with 1e for 72 h, showing that the
proportion of cells that accumulated in the G0/G1 phase increased
compared to the S and G2/M phases (Fig. 6A). Cycle-related pro-
teins’ levels were studied by Western blot analysis (Fig. 6B). The
expression levels of Cyclin D1 and Cyclin E1 were significantly
reduced, while the levels of CDK4 and CDK6 was almost unchanged.

R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
7
Fig. 5. The expression of apoptosis-inducing factor (AIF) by immunofluorescence analysis (A), the expression of apoptosis-inducing factor (AIF) and Bax/Bcl-2 by Western blot (B).
Statistical significance of differences in mean values: *p < 0.05, **p < 0.01 and ***p < 0.001.
From the UVeVis spectrum, the ct-DNA titration experiment
showed a clear isosbestic point at 296 nm. (Fig. S5B). For the ab-
sorption band at approximately 258 nm, a slight hyperchromicity
and an ~2 nm blueshift were occurred upon addition of ct-DNA.
These results indicated that the changes in absorption spectra
observed after the addition of ct-DNA were not due to the elec-
trostatic binding of the 1e to the polyanion DNA phosphate back-
bone, but rather to the binding of 1e to the DNA grooves or its
be used to identify whether the compound and ct-DNA act in an
insertion or non-insertion mode. As shown in Fig. S5C, the EB-ct-
DNA system produced the strongest fluorescence emission at
613 nm without 1e; upon titration of 1e, the fluorescence intensity
was weakened, progressively, which revealed the presence of a
pattern of insertion of 1e with DNA. The fluorescence quenching
q
constant (K ) was analyzed through the Stern-Volmer equation
according to previous literature [56]. A circular dichroism spectral
assay of 1e (Fig. S5D) showed that 1e bound to ct-DNA, and when
the 1e/ct-DNA ratio reached 6/1, the intensity of positive and
negative optical rotation has changed a lot, further confirming that
the DNA binding mode of 1e occurs via the intercalation binding
involvement in the insertion of p- p stacking interactions.
In addition, the binding modes of 1e and ct-DNA were evaluated
by monitoring the fluorescence spectra in competitive binding ct-
DNA experiments between 1e and ethidium bromide (EB). After
the small molecule compound supersedes EB, the fluorescence
0
mode. Considering that 1e s analogue complex 5e was mainly
ꢂ6
emission intensity of the EB-ct-DNA system ([EB] ¼ 2.0 ꢁ 10 M,
localized in mitochondria, we speculate that this complex may
react with mitochondrial DNA [57e59], which requires further
ꢂ5
[
ct-DNA] ¼ 2.0 ꢁ 10 M) will be changed significantly, which can

8
R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
chronic liver injury, which led to pathological morphological
changes in liver, including inflammation, steatosis, and necrosis. To
assess the protective effect of 1e (10 mg/kg), we used mice to
explore its therapeutic potential in preventing or treating liver
injury. We therefore analyzed biochemical markers of liver injury
after CCl
4
treatment for 14 weeks. All mice were sacrificed after the
had cirrhotic features
experiment. All animal livers dealt with CCl
4
with micronodular pale surfaces, which indicated that the mouse
liver cirrhosis model was successfully modeled (Fig. 8A). In addi-
tion, H&E staining showed that treatment with 10 mg/kg of 1e
significantly ameliorated the morphological changes in liver tissue
(Fig. 8B). The accumulation of collagen accompanies fibrogenesis;
therefore, liver tissue sections were stained with Sirius red and
Masson’s reagent to examine collagen expression in the mouse
liver. This analysis showed that collagen was markedly deposited in
4
the CCl -injured liver but that the amount was reduced in the liver
of mice in the 1e treatment groups (Fig. 8C and Fig. S6A). During
chronic liver disease, inflammation is commonly accompanied by
hepatocellular injury. Severe inflammatory cell infiltration in livers
exposed to CCl
4
was suppressed by 1e, as characterized by a
ꢂ
decrease in the number of intrahepatic CD45 and F4/80-positive
cells (Fig. 8D and Fig. S6B). As demonstrated by the ELISA results,
1
(
e treatment also decreased the serum levels of IL-6 and IL-8
Fig. 9A). Thus, we proposed that 1e protected the liver against
CCl -induced injury by suppressing inflammation. The levels of AST,
ALT, ALP, and LDH were effectively increased in two groups (the
model group) and decreased by 1e in CCl -treated mice (Fig. 9B).
More importantly, the expression of TrxR in serum and liver tissue
was high in CCl -injured liver, and treatment with 1e significantly
4
4
4
reversed these changes (Fig. 10A and B). Overall, these results can
provide more in vivo evidence that treatment of 1e did reduce the
accumulation of collagen and inhibit the inflammation in HCC
model mice.
Fig. 6. Effect of 1e on cell cycle progression (A), Western blot analysis of HepG2 cells
after treatment with 1e at different concentrations for 24 h (B, C). Error bars: S.D.,
n ¼ 3. Statistical significance of differences in mean values: *p < 0.05, **p < 0.01 and
3. Conclusions
*
**p < 0.001.
In summary, our study demonstrates that a novel rhodium(I)
complex 1e triggers antiproliferative effects in vitro and exhibits
promising in vivo anticancer effects in HCC. To our knowledge, this
study showed for the first time of the in vivo biological activity of
rhodium(I) NHC complex. This complex inhibited the expression of
the TrxR system and induced the accumulation of ROS, the
apoptosis of cancer cells (Fig. 11). Moreover, 1e inhibited the
mitochondrial apoptotic pathway, causing high expression of the
apoptogenic factor Bax and low expression of the antiapoptotic
cytokine Bcl-2. Besides, 1e induced DNA damage and led to cell
cycle arrest. In light of this report we propose that 1e represents a
new TrxR inhibitor and a potential candidate for the treatment of
HCC.
research to prove.
2.11. In vivo antitumor efficacy
To assess the potency of complex 1e in inhibiting HCC tumor
growth in vivo, we constructed HepG2 tumor xenograft models in
nude mice by subcutaneously injecting HepG2 cells in the expo-
nential phase into the right armpit of the mice. When successful
modeling, the tumor volume of nude mice were 180e200 mm , and
these tumor model nude mice were randomly divided into two
3
groups (n ¼ 4): (1) the vehicle-treated group and (2) the complex
1
e-treated (10 mg/kg) group. During the 14-day treatment period,
4. Experimental section
there was significant weight loss in the 1e treatment group
Fig. 7A). Treatment with 10 mg/kg 1e remarkably inhibited the
(
4.1. Materials and methods
growth of HepG2 tumor volume (Fig. 7B). Compared with the
control, the tumor weight of mice treated with 1e 10 mg/kg was
significantly reduced, and the tumor growth inhibition rate reached
All the reagents for synthesis were purchased from Energy
Chemical (China) and used as received without further purification.
Cisplatin was purchased from Yuanye Bio. (Shanghai, China).
Deuterated solvents were purchased from Cambridge Isotope Lab
Inc. (CIL). Solutions for all spectroscopic and biological studies were
prepared with solvents of HPLC grade and doubly distilled water
4
5%. (Fig. 7C and D).
Many animal models have provided pertinent information on
the pathogenesis and molecular mechanisms of HCC, which is
critical for preclinical studies of drugs [60]. Chronic exposure of
mouse liver to hepatotoxin carbon tetrachloride (CCl
4
) is a mature
from Millipore (>18.2 MU). Antibodies, such as activated caspases-
method of inducing liver damage and promoting fibrosis and
cirrhosis [61]. Hence, to further evaluate the anticancer activities of
3, caspase-9, p15, p21, p27, NAPDH, for WB or IF, were purchased
from Abcam Co. (China). JC-1 assay kit was purchased from KeyGen
Co. (Jiangsu, China); ROS assay kit, ATP assay kit, mitochondria and
nuclei Isolation kit were purchased from Beyotime Biotechnology
1
e against HCC in vivo, we chose CCl
4
-induced chronic HCC mice.
Injection of CCl into the mice approximately for 14 weeks induced
4

R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
9
Fig. 7. Effect of treatments on the mean body weight of mice (A), tumor growth curves in mice receiving the 1e treatment schedule. Values are the mean ± SEM, n ¼ 4/group (B),
representative images of tumors explanted from control and treated mice at sacrifice (upper panel) and relative tumor weight (lower panel) after tumor challenge; values are the
mean ± SEM (C) and (D). Values are the mean ± SEM. The significance was determined by one-way ANOVA followed by Tukey’s multiple comparison post-test; each group was
compared to the control. (***p < 0.001).
(
Jiangsu, China). TrxR activity assay kit, purified TrxR (rat liver),
d
10.94 (s, 1H, NCHN), 7.58 (d, J ¼ 8.4 Hz, 4H, ArphH), 7.14 (d,
ATPase activity assay kit and Calf-thymus (ct-DNA) were purchased
from Sigma-Aldrich Co.; MitoTracker ®Red CMXRos (M7512) was
purchased from Yeasen (Shanghai, China).
J ¼ 8.4 Hz, 4H, ArphH), 4.33 (q, J ¼ 7.3 Hz, 4H, NCH
2
CH
3
), 1.51 (t,
137.44
13
ꢀ
J ¼ 7.3 Hz, 6H, NCH
2
CH
3
). C NMR (126 MHz, CDCl
3
, 20 C):
d
(NCHN), 132.88, 131.81, 123.53, 125.52 (ArphC), 130.90 (NC ¼ CN),
1
13
þ
H NMR and C NMR spectra were obtained on Bruker Avance
43.62 (NCH
2
CH
3
), 15.82 (NCH
2
CH
3
). ESI-MS(þ)[m/z]: 435.0 [M-Br] .
0
III HD, 500 M spectrometers. The tetramethylsilane signal as a
reference. The purities of 1e-9e (>95%) were performed by the
centre for instrumental analysis China Pharmaceutical University,
Nanjing, and obtained on PE 2400 Series II. UVeVis spectra were
recorded on a Shimadzu UV2400 spectrophotometer. Fluorescence
excitation and emission spectra were recorded on a Cary Eclipse
Varian fluorescence spectrophotometer (band pathways were
4,5-di([1,1 -biphenyl]-4-yl)-1,3-diethyl-1H-imidazol-3-ium bro-
1
ꢀ
3
, 20 C):
mide 4d. Yield 78%; white solid. H NMR (500 MHz, CDCl
10.96 (s, 1H, NCHN), 7.66 (d, J ¼ 8.2 Hz, 4H, ArphH), 7.58 (d,
d
J ¼ 7.3 Hz, 4H, ArphH), 7.46 (t, J ¼ 7.6 Hz, 4H, ArphH), 7.41e7.33 (m,
6H, ArphH), 4.41 (q, J ¼ 7.3 Hz, 4H, NCH
2
CH
, 20 C):
3
), 1.56 (t, J ¼ 7.3 Hz, 6H,
13
ꢀ
NCH
2
CH
3
). C NMR (126 MHz, CDCl
3
d
143.23, 139.31,
130.76, 129.02, 128.26, 127.95, 127.06, 123.64 (ArphC), 131.54
5
nm). CD spectra were obtained from vibrational CD spectrometer
(NC ¼ CN), 137.06 (NCHN), 45.52 (NCH
2 3 2 3
CH ), 15.99 (NCH CH ). ESI-
þ
Bruker PMA50. Fluorescent microscopy was performed using a
Leica DMi8 fluorescence microscope. ESI-MS spectra were got from
the Agilent 6125. X-Ray Single Crystal Diffractometer was BRUKER
Smart APEX II CCD, Flow Cytometer was BD C6 Plus. MTT and
protein assays were quantified by PerkinElmer Fusion Reader
MS(þ)[m/z]: 429.2 [M-Br] .
3-(anthracen-9-ylmethyl)-1-ethyl-4,5-bis(4-fluorophenyl)-1H-
1
imidazol-3-ium chloride 5d. Yield 67%; yellow solid. H NMR
ꢀ
(500 MHz, DMSO‑d
6
, 20 C):
d
8.87 (s, 1H, NCHN), 8.70 (s, 1H,
AranthH), 8.23 (d, J ¼ 8.2 Hz, 2H, AranthH), 8.19 (d, J ¼ 8.7 Hz, 2H,
AranthH), 7.72 (dd, J ¼ 8.6, 5.4 Hz, 2H, AranthH), 7.64 (m, 4H, ArphH),
7.54 (dd, J ¼ 8.6, 5.4 Hz, 2H, AranthH), 7.34 (dd, J ¼ 17.7, 2.2 Hz, 4H,
(
Packard BioScience Company). Analyses indicated by the symbols
of the elements or functions were within ±0.4% of the theoretical
values.
ArphH), 6.16 (s, 2H, NCH
2
Aranth), 3.97 (q, J ¼ 7.1 Hz, 2H, NCH
2
CH
, 20 C):
3
),
13
ꢀ
0
.99 (t, J ¼ 7.2 Hz, 3H, NCH
2
CH
3
). C NMR (126 MHz, CDCl
3
d
164.63, 164.51, 162.63, 162.50, 133.06, 132.99, 132.72, 132.65,
20.85, 120.82, 120.70, 120.67, 116.58, 116.40, 116.08, 115.91 (ArphC),
4.2. Synthesis and characterization
1
1
36.16 (NCHN), 132.13, 131.31 (NC ¼ CN), 131.19, 130.84, 130.32,
The synthesis of 1a-3a, 1b-3b, 1c-4c,1d-8d were seen in sup-
porting information.
129.29, 127.98, 125.41, 123.20, 122.83 (AranthC), 45.49 (NCH
2
Aranth),
þ
4
3.56 (NCH CH ), 15.39 (NCH CH
3
2
3
2
3
). ESI-MS(þ)[m/z]: 475.2 [M-Cl] .
-(cyclopropylmethyl)-1-ethyl-4,5-bis(4-fluorophenyl)-1H-imida-
1
4
.2.1. General procedure for the synthesis of 1d-8d
Bromoethane or halohydrocarbon (10.0 mmol) and imidazole
zol-3-ium bromide 6d. Yield 72%; white solid. H NMR (500 MHz,
CDCl
ꢀ
3
, 20 C):
d
10.85 (s, 1H, NCHN), 7.32e7.26 (m, 4H, ArphH), 7.15
CH
), 4.11 (d, J ¼ 7.4 Hz,
1
2
3
c-4c (2.0 mmol) were dissolved in 30 mL CH CN, and refluxed for
(
2
m, 4H, ArphH), 4.37 (q, J ¼ 7.3 Hz, 2H, NCH
H, NCH ), 2.19 (s, 1H, NCH CHCH CH
NCH CH
), 0.71 (q, J ¼ 5.7 Hz, 2H, NCH
J ¼ 5.1 Hz, 2H, NCH
2
3
e4 d. Then removed the solvent in vacuo, purified by silica gel
2
C
3
H
5
2
2
2
), 1.52 (t, J ¼ 7.3 Hz, 3H,
column chromatography.
,5-bis(4-bromophenyl)-1,3-diethyl-1H-imidazol-3-ium bromide
2
3
2
CHCH
2
CH
2
), 0.50 (q,
13
ꢀ
4
2
CHCH
2
CH
2
). C NMR (126 MHz, CDCl
3
, 20 C):
1
ꢀ
3
d. Yield 70%; pale yellow solid. H NMR (500 MHz, CDCl , 20 C):
3

10
R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
Fig. 8. The liver morphological changes after treatment with vehicle, CCl
CCl , and both CCl
4
, and both CCl
4
and 1e (A), liver issue conventional slice with H&E staining after treatment with vehicle,
, and both CCl and 1e (C, original
4
4
and 1e (B, original magnification, 20 ꢁ ), liver issue conventional slice with Sirius red staining after treatment with vehicle, CCl
4
4
magnification, 20 ꢁ ), liver issue conventional slice with CD45 staining (D, original magnification, 20 ꢁ ). (For interpretation of the references to colour in this figure legend, the
reader is referred to the Web version of this article.)
d
164.71, 162.70, 132.83, 132.76, 132.68, 132.61, 121.01, 120.98,
4.2.2. General procedure for the synthesis of 1e-9e
120.89,120.86,116.87,116.85,116.69,116.67 (ArphC),136.80 (NCHN),
One equivalent of the imidazolium cholrine/bromine 1d-9d
1
31.07, 130.98 (NC ¼ CN), 52.77 (NCH
2
C
3
H
5
), 43.55 (NCH
2
CH
3
),
(0.20 mmol) and 0.6 equivalent of Ag
added, and then 15 mL of anhydrous dichloromethane was added,
and reacted overnight under N atmosphere at ambient tempera-
ture. After that, 0.5 equivalent of bis[chloride(1,5-cyclooctadiene)
rhodium(I)] (49.3 mg, 0.10 mmol) was added in the mixture, and
continued to stir for 6 h in darkness. Then the solution was filtered
over a Celite bed (281 nm) and the filtrate concentrated in vacuo.
Next, the residue was purified by silica gel column chromatography,
2
O (27.8 mg, 0.12 mmol) were
15.83 (NCH
2
CH
3
), 11.10 (NCH
2
CH), 4.89 (NCH CHCH CH ). ESI-
2
2
2
þ
MS(þ)[m/z]: 339.1 [M-Br] .
2
1-ethyl-4,5-bis(4-fluorophenyl)-3-(naphthalen-1-ylmethyl)-1H-
1
imidazol-3-ium chloride 8d. Yield 38%; white solid. H NMR
ꢀ
(
500 MHz, DMSO‑d
6
, 20 C):
d
9.82 (s, 1H, NCHN), 8.02e7.98 (m, 1H,
ArnaphH), 7.95 (d, J ¼ 8.5 Hz, 2H, ArnaphH), 7.62e7.55 (m, 4H, ArphH),
7
.47 (t, J ¼ 7.7 Hz, 1H, ArnaphH), 7.42 (dd, J ¼ 8.5, 5.4 Hz, 2H, ArphH),
7.34 (d, J ¼ 8.8 Hz, 2H, ArphH), 7.19 (dd, J ¼ 16.9, 8.0 Hz, 3H, ArnaphH),
2 2
and recrystallized from CH Cl /n-hexane to get the yellow crys-
5
.93 (s, 2H, NCH
2
Arnaph), 4.16 (q, J ¼ 7.2 Hz, 2H, NCH
2
CH
3
), 1.31 (t,
talline solid.
13
ꢀ
(h h
2, 2-cycloocta-1,5-diene)(4,5-bis(4-fluorophenyl)-1,3-
J ¼ 7.2 Hz, 3H, NCH
2
CH
3
). C NMR (126 MHz, DMSO‑d
6
, 20 C):
Chlorido
diethylimidazol-2-ylidene) rhodium(I) 1e. Yield 71%; yellow solid.
d
1
(
1
164.41, 164.30, 162.44, 162.33, 133.88, 133.81, 133.74, 133.67,
22.01, 121.98, 121.85, 121.82, 116.74, 116.57, 116.40 (ArphC), 136.76
NCHN), 133.60, 130.36, 130.15, 129.59, 129.22, 127.42, 126.79,
26.58, 125.89, 123.15 (ArnaphC), 48.76 (NCH Arnaph), 131.55, 131.48
CH ), 15.22 (NCH CH
). ESI-MS(þ)[m/z]:
1
ꢀ
H NMR (500 MHz, CDCl
ArphH), 7.01 (dd, J ¼ 8.3 Hz, 5.4 Hz, 4H, ArphH), 5.05 (s, 2H, CH ¼),
4.77 (m, 2H, NCH CH ), 4.54 (m, 2H, NCH CH
2.50e2.36 (m, 4H, CH ), 2.02e1.93 (m, 4H, CH
3
, 20 C):
d
7.12 (dd, J ¼ 8.3, 5.4 Hz, 4H,
2
2
3
2
3
), 3.43 (s, 2H, CH ¼),
(
4
NC ¼ CN), 43.37 (NCH
2
3
2
3
2
2
), 1.25 (t, J ¼ 7.2 Hz,
þ
13
ꢀ
25.1 [M-Cl] .
6H, NCH
2
CH
3
).
C
NMR (126 MHz, CDCl
3
,
20 C):
d
182.46

R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
11
Fig. 9. ELISA measurement of IL-6, IL-8 levels in serum (A), determination of serum ALT, AST, ALP and LDH levels (B) after treatment with vehicle, CCl
4
, and both CCl
4
and 1e. For the
###
statistics of each panel in this figure, data are expressed as mean ± SD (n ¼ 5);
4
p < 0.001 compared with Vehicle, *p < 0.05, **p < 0.01 and ***p < 0.001 compared with group CCl .
Fig. 10. ELISA and Western blot measurement of TrxR levels in serum (A) and in liver
tissue (B) after treatment with Vehicle, CCl , and both CCl and 1e. For the statistics of
each panel in this figure, data are expressed as mean ± SD (n ¼ 5);
compared with Vehicle, ***p < 0.001 compared with group CCl
4
4
#
##
p < 0.001
4
.
(
NCNeRh), 163.71, 161.73, 132.03, 124.77, 115.97, 115.80 (ArphC),
1
30.78 (NC ¼ CN), 98.35,98.41 (CH ¼), 68.33, 68.44 (CH ¼), 44.21
Fig. 11. Proposed anticancer mechanisms of 1e.
(
[
NCH
2
CH
3
), 32.94 (CH
2
), 28.21 (CH
2
), 16.12 (NCH CH
2
3
). ESI-MS(þ)
þ
m/z]: 523.1 [M-Cl] . Anal. Calcd for C27
H
30ClF N Rh: C, 58.02; H,
2 2
5
.41; N, 5.01; Found: C, 57.90; H, 5.43; N, 5.03.
). 13C NMR (126 MHz, CDCl
ꢀ
3 3
, 20 C): d 180.92 (NCNeRh),
NCH
1
2
CH
59.49, 131.51, 121.27, 113.95 (ArphC), 131.15 (NC ¼ CN), 97.99, 97.94
CH ¼), 68.33, 68.21 (CH ¼), 55.19 (OCH ), 44.08 (NCH CH ), 32.95
CH ), 28.91 (CH ), 16.24 (NCH CH
). ESI-MS(þ)[m/z]: 547.2 [M-
36ClN Rh: C, 59.75; H, 6.22; N, 4.81;
2
2
Chlorido(
,3-diethylimidazol-2-ylidene) rhodium(I) 2e. Yield 91.8%; yellow
h ,h -cycloocta-1,5-diene)(4,5-bis(4-methoxyphenyl)-
1
(
(
3
2
3
1
ꢀ
solid. H NMR (500 MHz, CDCl
3
, 20 C):
d
7.05 (d, J ¼ 8.7 Hz, 4H,
2
2
2
3
ArphH), 6.80 (d, J ¼ 8.7 Hz, 4H, ArphH), 5.02 (s, 2H, CH ¼), 4.73 (m,
þ
Cl] . Anal. Calcd for C29
H
2 2
O
2
H, NCH
2
CH
3
), 4.50 (m, 2H, NCH
2
CH
3
), 3.76 (s, 6H, OCH
3
), 3.42 (s,
Found: C, 59.70; H, 6.15; N, 5.03.
2
H, CH ¼), 2.41 (m, 4H, CH
2
), 1.95 (m, 4H, CH
2
),1.22 (t, J ¼ 7.1 Hz, 6H,
2
2-cycloocta-1,5-diene)(4,5-bis(4-bromophenyl)-1,3-
Chlorido (h ,h

12
R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
diethylimidazol-2-ylidene) rhodium(I) 3e. Yield 33.5%; yellow solid.
C
29
H
32ClF
N, 4.98.
Chlorido(
2
N
2
Rh: C, 59.55; H, 5.51; N, 4.79; Found: C, 59.49; H, 5.56;
2, 2-cycloocta-1,5-diene)(3-benzyl-4,5-bis(4-
1
ꢀ
H NMR (500 MHz, CDCl
.01 (d, J ¼ 8.2 Hz, 4H, ArphH), 5.06 (s, 2H, CH ¼), 4.78 (dq, J ¼ 14.2,
.1 Hz, 2H, NCH CH CH ), 3.43
s, 2H, CH ¼), 2.44 (m, 4H, CH
3
, 20 C):
d
7.46 (d, J ¼ 8.2 Hz, 4H, ArphH),
7
7
(
h
h
2
3
), 4.56 (dq, J ¼ 14.0, 7.0 Hz, 2H, NCH
2
3
fluorophenyl)-1-ethylimidazol-2-ylidene) rhodium(I) 7e. Yield: 62%;
1
ꢀ
3
, 20 C): d 7.19e7.12 (m, 5H,
2
), 1.98 (m, 4H, CH
NMR (126 MHz, CDCl
2
), 1.25 (t, J ¼ 7.1 Hz,
yellow solid. H NMR (500 MHz, CDCl
ArbenzylH), 7.07e6.99 (m, 4H, ArphH), 6.86e6.78 (m, 4H, ArphH), 6.11
(d, J ¼ 15.3 Hz, 1H, NCH Ph), 5.73 (d, J ¼ 15.3 Hz, 1H, NCH Ph), 5.09
(p, J ¼ 8.1 Hz, 2H, CH ¼ ), 4.81 (dq, J ¼ 14.3, 7.2 Hz, 1H, NCH CH ),
4.70 (dq, J ¼ 14.3, 7.2 Hz, 1H, NCH CH ), 3.54 (m,1H, CH ¼), 3.33 (m,
1H, CH ¼),2.56e2.46 (m, 1H, CH ), 2.43e2.33 (m, 2H, CH ), 2.17 (m,
1H, CH ), 2.06e1.90 (m, 3H, CH ), 1.83 (m, 1H, CH
J ¼ 7.2 Hz, 3H, NCH C NMR (126 MHz, CDCl
13
ꢀ
6
H, NCH
2
CH
3
).
C
3
,
20 C):
d
182.96
(
9
NCNeRh), 132.05, 131.60, 127.57, 123.22 (ArphC), 130.70 (NC ¼ CN),
2
2
8.55, 98.50 (CH ¼), 68.48, 68.37(CH ¼), 44.29 (NCH
2
CH ), 32.93
3
2
3
(CH
2
), 28.87 (CH
2
), 16.22 (NCH
2
CH
3
). ESI-MS(þ)[m/z]: 701.7 [M-
2
3
þ
þ
þ
HþNa] , 645.1 [M-Cl] . MALDI-TOF/TOF MS: 644.9907 [M-Cl] .
2
2
Anal. Calcd for C27
H30Br
2
ClN
2
Rh: C, 47.64; H, 4.44; N, 4.12; Found: C,
2
2
2
), 1.31 (t,
13
ꢀ
4
7.69; H, 4.42; N, 4.17.
2
CH
3
).
3
, 20 C):
2
2
0
Chlorido(
h
,h
-cycloocta-1,5-diene)(4,5-di([1,1 -biphenyl]-4-yl)-
d 183.62, 183.21 (NCNeRh), 163.74, 163.58, 161.75, 161.60, 132.32,
1
,3-diethylimidazol-2-ylidene) rhodium(I) 4e. Yield 31%; yellow
132.26, 132.04, 131.97, 124.63, 124.60, 124.55, 124.53, 115.96, 115.78,
115.45, 115.28 (ArC), 136.94, 128.27, 127.47, 127.28 (ArbenzylC), 131.47,
131.21 (NC ¼ CN), 98.94, 98.88, 98.39, 98.33 (CH ¼), 68.92, 68.81,
1
ꢀ
solid. H NMR (500 MHz, CDCl
8
3
, 20 C):
d
7.56 (dd, J ¼ 11.7, 8.1 Hz,
H, ArphH), 7.42 (t, J ¼ 7.6 Hz, 4H, ArphH), 7.35 (d, J ¼ 7.4 Hz, 2H,
ArphH), 7.26 (d, J ¼ 7.7 Hz, 4H, ArphH), 5.07 (s, 2H, CH ¼), 4.86 (m,
68.61, 68.49 (CH ¼), 53.25 (NCH
2
Ph), 44.31 (NCH CH
), 28.46 (CH ), 16.19 (NCH CH
2 2
32ClF N Rh: C, 61.89; H,
2
3
), 33.34 (CH
2
),
2
2
NCH
H, NCH
.54e2.36 (m, 4H, CH
2
CH
3
), 4.64 (m, 2H, NCH
2
CH
3
), 3.49 (s, 2H, CH ¼),
), 1.31 (t, J ¼ 7.2 Hz, 6H,
32.40 (CH
2
), 29.15 (CH
2
2
2
3
). ESI-MS(þ)
þ
2
), 1.99 (m, 4H, CH
2
ꢀ
[m/z]: 585.1 [M-Cl] . Anal. Calcd for C32
H
13
2
CH
3
). C NMR (126 MHz, CDCl
3
, 20 C):
d
181.85 (NCNeRh),
5.19; N, 4.51; Found: C,61.78; H, 5.28; N, 4.67.
2
2
141.10, 140.00, 130.57, 128.84, 127.85, 127.69, 127.19, 126.97 (ArphC),
Chlorido(
h ,h -cycloocta-1,5-diene)(4,5-bis(4-fluorophenyl)-3-
1
31.39 (NC ¼ CN), 98.39, 98.24 (CH ¼), 68.47, 68.36 (CH ¼), 44.30
(naphthalen-1-ylmethyl)-1-ethylimidazol-2-ylidene) rhodium(I) 8e.
1
ꢀ
(
[
NCH
2
CH
3
), 32.97 (CH
2
2
), 28.92 (CH ), 16.33 (NCH
2
CH
3
). ESI-MS(þ)
Yield: 23%; yellow solid. H NMR (500 MHz, CDCl , 20 C): d 8.12 (d,
3
þ
þ
m/z]: 639.42 [M-Cl] . MALDI-TOF/TOF MS: 639.2273 [M-Cl] . Anal.
J ¼ 8.3 Hz, 1H, ArnaphH), 7.85 (d, J ¼ 7.9 Hz, 1H, ArnaphH), 7.73 (d,
J ¼ 8.2 Hz, 1H, ArnaphH), 7.58 (t, J ¼ 7.2 Hz, 1H, ArnaphH), 7.53 (t,
J ¼ 7.2 Hz, 1H, ArnaphH), 7.32 (t, J ¼ 7.7 Hz, 1H, ArnaphH), 7.22 (dd,
J ¼ 8.6, 5.3 Hz, 2H, ArphH), 7.05 (t, J ¼ 8.6 Hz, 2H, ArphH), 6.97 (dd,
J ¼ 5.3 Hz, 2H, ArphH), 6.93 (d, 1H, ArnaphH), 6.86 (d, J ¼ 16.5 Hz, 1H,
Calcd for C39
H, 5.97; N, 4.22.
2
H40ClN Rh: C, 69.38; H, 5.97; N, 4.15; Found: C, 69.15;
2
h ,h
2-cycloocta-1,5-diene)([3-(anthracen-9-ylmethyl)]-
Chlorido(
,5-bis(4-fluorophenyl)-1-ethylimidazol-2-ylidene) rhodium(I) 5e.
4
1
ꢀ
Yield 63%; yellow solid. H NMR (500 MHz, CDCl
3
, 20 C):
d
8.36 (d,
NCH
NCH
1H, NCH
2
Arnaph), 6.71 (t, J ¼ 8.6 Hz, 2H, ArphH), 6.00 (d, J ¼ 16.5 Hz, 1H,
Arnaph), 4.99 (p, J ¼ 8.1 Hz, 2H, CH ¼), 4.90 (dq, J ¼ 14.3, 7.2 Hz,
J ¼ 8.9 Hz, 2H, AranthH), 8.24 (s, 1H, AranthH), 7.86 (d, J ¼ 8.4 Hz, 2H,
2
AranthH), 7.52e7.46 (m, 2H, AranthH), 7.43e7.37 (m, 2H, AranthH),
2
CH
3
), 4.65 (dq, J ¼ 14.1, 7.1 Hz, 1H, NCH
2
CH
3
), 3.54 (t,
), 2.29
), 1.51e1.39
). C NMR (126 MHz,
184.44, 184.03 (NCNeRh), 163.76, 163.45, 161.77,
7
.32 (d, J ¼ 15.2 Hz, 1H, NCH
ArphH), 6.85 (t, J ¼ 8.6 Hz, 2H, ArphH), 6.53 (d, J ¼ 15.1 Hz, 1H,
NCH
Aranth), 6.03e5.95 (m, 2H, ArphH), 5.93 (t, J ¼ 8.7 Hz, 2H,
ArphH), 5.21e5.10 (m, 2H, CH ¼), 4.86e4.74 (m, 2H, NCH CH ), 3.69
), 2.44e2.31 (m, 3H, CH ),
), 2.04e1.90 (m, 3H, CH
), 1.31 (t, J ¼ 7.2 Hz,
2
Aranth), 6.94 (dd, J ¼ 8.6, 5.4 Hz, 2H,
J ¼ 6.9 Hz,1H, CH ¼), 3.12 (s,1H, CH ¼), 2.54e2.44 (m,1H, CH
(m, 1H, CH ), 2.00e1.87 (m, 2H, CH ), 1.78 (m, 2H, CH
), 1.35 (t, J ¼ 7.2 Hz, 3H, NCH
2
2
2
2
13
2
(m, 2H, CH
2
2 3
CH
ꢀ
2
3
CDCl
3
, 20 C):
d
(
2
3
s, 2H, CH ¼), 2.63e2.54 (m, 1H, CH
2
2
161.47, 132.10, 132.04, 124.74, 124.71, 124.19, 124.16, 116.04, 115.87,
115.57, 115.39 (ArphC), 133.45, 133.25, 128.67, 127.82, 126.49, 125.95,
125.01, 124.05, 122.60 (ArnaphC), 131.11, 130.28 (NC ¼ CN), 99.08,
99.03, 97.78, 97.72 (CH ¼), 69.13, 69.01, 68.59, 68.47 (CH ¼), 49.93
.14e2.07 (m, 1H, CH
2
2
ꢀ
13
H, NCH CH ). C NMR (126 MHz, CDCl , 20 C): d 183.11, 182.70
2 3 3
(
1
(
1
(
(
1
NCNeRh), 163.54, 162.45, 161.56, 160.48, 131.97, 131.90, 131.24,
31.17, 124.50, 124.47, 123.68, 123.66, 115.61, 115.44, 113.67, 113.50
ArC), 131.84, 131.21 (NC ¼ CN), 131.07, 130.82, 128.77, 128.65,
26.53, 126.03, 124.89, 124.60 (AranthC), 99.09, 99.03, 98.51, 98.46
CH ¼), 69.55, 69.44, 67.95, 67.88 (CH ¼), 48.46 (NCH Arnath), 44.53
NCH CH ), 33.59 (CH ), 32.44 (CH ), 29.32 (CH ), 28.31 (CH ),
6.25 (NCH CH
). ESI-MS(þ)[m/z]:743.14 [MþNa] . Anal.Calcd for
36ClF Rh: C, 66.63; H, 5.03; N, 3.88; Found: C, 66.56; H, 4.99;
N, 4.15.
Chlorido( 2 2-cycloocta-1,5-diene)([3-(cyclopropylmethyl)]-4,5-
bis(4-fluorophenyl)-1-ethylimidazol-2-ylidene) rhodium(I) 6e. Yield
(NCH
2
Arnaph), 44.40 (NCH
), 16.32 (NCH
Cl] . Anal. Calcd for C36 34ClF
Found: C, 63.94; H, 5.07; N, 4.37.
2
CH
3
), 33.55 (CH
CH
). ESI-MS(þ)[m/z]: 635.2 [M-
Rh: C, 64.44; H, 5.11; N, 4.17;
2 2
), 31.84 (CH ), 28.85
(CH ), 28.16 (CH
2
2
2
3
þ
H
2 2
N
2
2
2
2
3
2
2
2
2
Chlorido(h ,h -cycloocta-1,5-diene)(1,3-diethylimidazol-2-
ylidene) rhodium(I) 9e. Yield 54%; yellow solid. H NMR (500 MHz,
þ
1
2
3
ꢀ
C
40
H
2
N
2
CDCl
NCH
4H, CH
3
, 20 C):
d
6.87 (s, 2H, CH ¼), 5.02 (s, 2H, CH ¼), 4.69 (m, 2H,
CH
), 1.54 (t, J ¼ 7.3 Hz, 6H, CH
181.60 (NCNeRh), 119.75 (NCH ]CHN),
CH ), 32.95
). ESI-MS(þ)[m/z]: 335.09 [M-
Rh: C, 48.60; H, 6.53; N, 7.56; Found:
2
CH
3
), 4.52 (m, 2H, NCH
), 1.94 (m, 4H, CH
2
3
), 3.31 (s, 2H, CH ¼), 2.46e2.37 (m,
13
h
,h
2
2
d
2 3
CH ). C NMR
ꢀ
(126 MHz, CDCl
3
, 20 C):
1
ꢀ
4
4
5
4
0%; yellow solid. H NMR (500 MHz, CDCl
H, ArphH), 7.08e6.97 (m, 4H, ArphH), 5.10 (d, J ¼ 6.9 Hz, 1H, CH ¼),
.05 (dd, J ¼ 12.6, 7.8 Hz, 1H, CH ¼), 4.79e4.70 (m, 2H, NCH CH ),
.66 (td, J ¼ 14.2, 7.1 Hz, 1H, NCH CH(cyclopropane)), 4.32 (dd, J ¼ 14.0,
.1 Hz, 1H, NCH
CH(cyclopropane)), 3.48 (t, J ¼ 7.2 Hz, 1H, CH ¼ ), 3.41
s, 1H, CH ¼), 2.53e2.36 (m, 4H, CH ), 2.06e1.92 (m, 4H, CH ), 1.27
t, J ¼ 7.2 Hz, 3H, NCH CH ), 1.14 (m, 1H, NCH
m, 2H, CH2(cyclopropane)), 0.25e0.18 (m, 1H, CH2(cyclopropane)), ꢂ0.16
3
, 20 C):
d
7.19e7.11 (m,
98.23, 99.17 (CH ]CH), 67.91, 67.80 (CH ¼), 45.59 (NCH
(CH ), 28.87 (CH ), 16.31 (NCH CH
Cl] , Anal. Calcd for C15 24ClN
C, 48.65; H, 6.61; N, 7.53.
2
3
2
2
2
3
þ
2
3
H
2
2
7
2
(
(
(
(
1
1
1
6
(
1
2
2
2
3
2
CH (cyclopropane)), 0.38
4.3. X-ray crystallographic analysis
13
ꢀ
m, 1H, CH2(cyclopropane)). C NMR (126 MHz, CDCl
82.67 (NCNeRh), 163.72, 161.73, 132.31, 132.25, 132.14, 132.07,
25.34, 125.31, 124.86, 124.83, 115.95, 115.88, 115.77, 115.70 (ArphC),
31.05, 130.79 (NC ¼ CN), 98.49, 98.44, 98.28, 98.22 (CH ¼), 68.78,
8.66, 68.25, 68.14 (CH ¼), 53.86 (NCH
NCH CH ), 33.23, 32.61 (CH ), 29.24, 28.58 (CH
1.61 (NCH
3
, 20 C):
d
182.68,
Monocrystal of 1e was obtained by slowly diffusing n-hexane
into concentrated solution of 1e of dichloromethane. The data were
collected at 173(2) K on a BRUKER Smart APEX II CCD area-detector
diffractometer with graphite-monochromated Mo Ka radiation
2
CH(cyclopropane)), 44.29
(
l
¼ 0.71073 Å). Single crystal structure analysis and correction
2
3
2
2
), 16.15 (NCH
2
CH
3
),
were performed using the SHELX-2014 kit program. The Cambridge
Crystallographic Data Centre: CCDC 1916369. DOI: 10.5517/
ccdc.csd.cc22b4cy.
2
CH (cyclopropane)), 5.01 (CH
2
(cyclopropane)), 4.33 (CH
2
þ
(
cyclopropane)). ESI-MS(þ)[m/z]: 549.2 [M-Cl] . Anal. Calcd for

R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
13
4.4. Growth inhibitory assay
4.10. Immunofluorescence staining
The antiproliferation effect of rhodium(I) compounds and
The immunofluorescence analysis was carried out according to
cisplatin were measured using the MTT cytotoxicity assay. The
processes were described as: adding 0.5% MTT (0.5 mg/mL, DMSO)
previously reported method [64]. The nuclei of HepG2 cells were
dyed by 4 ,6-Diamidino-2-phenylindole (DAPI). HepG2 cells were
0
reagent to the cells treated with the rhodium(I) complexes, incu-
seeded in 24-well plates and cultured for 24 h, then treated with 1e
ꢀ
bating for 4 h at 37 C, 5% CO
carefully discarding the culture medium, adding 200
2
in a humidified environment, then
L DMSO per
(1.5
m
M, 3
mM, and 6
m
M) for 24 h. Incubating HepG2 cells with
ꢀ
m
antibodies overnight at 4 C, followed by treating with anti-mouse
or anti-rabbit IgG for 2 h at ambient temperature and then the
HepG2 cells were washed with PBS for three times. Immunofluo-
rescence was detected with a fluorescence microscope (Leica DMi8,
original magnification 40 ꢁ ).
well. The 96-well plate was shaken on a shaker at a low speed until
fully dissolved the formazan (about 10 min). Detected the absor-
bance of each well at 490 nm.
4.5. Stability analysis
4.11. Western blotting analysis
1
e was incubated with 10% D
2
O þ 90% acetonitrile-d
3
over a
6
1
HepG2 cells (1 ꢁ 10 ) were inoculated into a 10 cm diameter
period of 96 h and H NMR spectra were taken in different times.
ꢀ
tissue culture dish (10 mL/well), and cultured at 37 C, 5% CO
2
in a
humidified environment for 24 h, and then treated 1e for 24 h,
carefully discarded the medium, and washed with cold PBS for
three times. Lysed with RIPA buffer in an ice bath, then centrifuged,
and the supernatants were collected and quantified by DC Protein
4
.6. Interaction with GSH and NAC
The solution with 50% acetonitrile-d
3
þ 50% D
2
O containing the
complex 1e (2 mM) and GSH (2 mM) or NAC (2 mM) was analyzed
by H NMR at different times.
Assay (Bio-Rad). Samples (15e45 mg/lane) were separated by 12.5%
1
SDS-PAGE gel, Tris-Glycine as the surging solution. After the sepa-
ration process, the proteins were imprinted onto the PVDF mem-
branes and blocked overnight in TBST (20 mM Tris-HCl, pH ¼ 7.6,
4
.7. Purified TrxR enzyme assay [38,62]
0.1% V/V Tween-20) containing 5% BSA at room temperature. The
blots were incubated with TBST for 1 h at ambient temperature,
which contained primary antibody and 5% BSA. Then washed with
TBST for three times, next, incubated with the suitable secondary
antibody for 2 h, The bands of protein detected by chem-
iluminescence procedure (ECL, Amersham).
DMF or with 1e (1.5
dissolved in 50 L of reaction buffer and then incubated in 96-well
plates for 75 min. Added 200 L of reaction mixture to each well
and 25 L 20 mM NADPH solution. And the reaction was started by
adding 25 L of a solution of 20 mM DTNB. After appropriate mix-
mM, 3 mM, 6 mM) and rat liver TrxR were
m
m
m
m
ing, record the absorption data at 405 nm with a microplate reader.
4
.12. Fluorescence spectroscopy
The increase in DTNB concentration over time follows a linear trend
2
(
r ꢃ 0.99), and the enzyme activity is calculated as its slope
Complex 5e was dissolved in DMF to get a clear solution
(
increased absorbance per second). For each test compound, it was
(
100 mM), which was then diluted to a suitable concentration in
.1 mM Tris-HCl buffer (pH 7.8). The spectra of 5e were followed
from 400 nm to 480 nm (excitation spectra) and from 470 nm to
confirmed by non-interference test components by using a nega-
tive control experiment without an enzyme solution.
0
560 nm (emission spectra).
4.8. Quantitative real-time polymerase chain reaction
4.13. Cellular uptake experiment
Trizol reagent was used to extract HepG2 RNA according to the
manufacturer’s instructions. Real-time PCR was executed by 7500
RT-PCR System as described previously [63]. The mRNA levels of the
target genes were calculated and results are from triplicate exper-
iments. The following primers (GenScript, Nanjing, China) were
used:
HepG2 cells were seeded on glass slides and cultured in DMEM
until the cell density reached 60%. Then, they were incubated with
ꢀ
1
0
m
M of 5e at 37 C, 5% CO
2
atmosphere for 24 h. The medium was
carefully aspirated, washed three times with PBS (1 ꢁ ) and incu-
bated with 200 nM MitoTracker® Red CMX Ros (M7512, Yeasen) for
0
0
b
-actin: (forward) 5 -TGTGGATCAGCAAGCAGGAGTA-3 ,
ꢀ
ꢀ
3
0 min at 37 C. PBS (1 ꢁ ) was washed three more times at 37 C
0
0
(
reverse) 5 -TGCGCAAGTTAGGTTTTGTCA-3 .
and then fixed with 4% PFA for 15 min at room temperature. The
cells were washed with PBS (1 ꢁ ) three times again, the cells were
stored in PBS (1 ꢁ ) and fluorescence was captured by a Leica DMi8
fluorescence microscope to visualize 5e and MitoTracker® Red CMX
Ros respectively (original magnification 40 ꢁ ).
0
0
TrxR:(forward) 5 -GCCCTGCAAGACTCTCGAAATTA-3 ,
0
0
(
reverse) 5 -GCCCATAAGCATTCTCATAGACGA-3 .
4.9. Cellular activities of TrxR assay
When HepG2 cells grew to abundance of 70e80%, treated with
4.14. Intracellular ROS measurement
different concentrations of 1e and incubated for 48 h and 72 h.
Then, the culture was carefully discarded and washed twice with
phosphate buffer (PBS). Total cell protein was extracted by treat-
ment with RIPA buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 0.5%
deoxycholate, 2 mM EDTA, 0.1% SDS, 1% Triton X-100, 1 mM PMSF
ROS generation in HepG2 cells was assessed by the fluorescent
probe CM-DCFH2-DA (Molecular Probes, Invitrogen). HepG2 cells
4
(2 ꢁ 10 ) were inoculated in 24-well plates, incubated for 24 h,
washed with PBS buffer and treated with 10
20 min at 37 C in darkness. Then, the cells were washed with PBS
m
M CM-DCFH2-DA for
ꢀ
and 1 mM Na
3
VO
4
) in an ice bath and quantified using the Bradford
program. The activity of TrxR in cells was determined according to
buffer and incubated with different concentrations of 1e (1.5
M, 6 M). The fluorescence increase was visualized by a Leica
DMi8 fluorescent microscopy (original magnification 20 ꢁ ).
mM,
the TrxR activity detection kit (Beijing Solarbio Science
Technology).
&
3
m
m

14
R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
4
.15. Measurement of MMP
size of tumor every 3 days and calculate the tumor volume (TV)
3
2
using the following formula: TV (mm ) ¼ width (length/2). The
4
HepG2 cells (2 ꢁ 10 cells/mL) were seeded on cover slips in a
inhibition rates of tumor growth (IRT) were calculated as follows:
2
4-well plate and incubated for 24 h in supplemental medium and
IRT ¼ 100% ꢁ (M represents the mean weight of the
1
e M
2
)/M
1
; M
1
ꢀ
then treated with 1e or DMF in 5% CO
for 2 h. After that, the cells were treated with JC-1 (5
0 min, washed with PBS buffer for twice, the cell staining was
observed with a fluorescence microscope (original magnification
2
culture conditions at 37 C
tumor in the blank group and M represents the mean weight of the
2
m
M) for
tumor in the drug treatment group.
3
4.19.1. Chronic liver cancer model caused by tetrachloromethane
All experiments involving mice were approved by the institu-
tional and local ethics committee (Beijing Weitong Lihua Labora-
tory Animal Technology Co., Ltd., China). According to National
Institutes of Health guidelines, all mice were gained humane
sympathy. Four-week-old male BALB/c mice (18e22 g) were
randomly divided into 3 groups of 5 mice each, and all animals
were fed under standardized conditions with a light/dark cycle of
appropriate humidity at room temperature, and provided water
and standard pathogen-free feed. Group one was the vehicle con-
4
0 ꢁ ).
4
.16. Cell cycle arrest
4
HepG2 cells (2 ꢁ 10 cells/well) were seeded in 6-well plates
and cultured for 24 h followed by treatment with DMF or 1e
1.5 M, 3 M or 6 M) for another 72 h. Cellular DNA flow cyto-
(
m
m
m
metric kits (Nanjing Key Gen Biotech) were used according to the
manufacturer’s protocol to determine the stage of the cell cycle,
specifically G0/G1, S, or G2/M, by flow cytometry (FACS Calibur; BD,
Franklin Lakes, NJ, USA).
trol, in which the mice were treated without CCl
Group two was the CCl model group. Group three was drug
treatment groups, in which the mice received CCl and 1e at the
same time. Groups two and three were injected with CCl (5 mg/kg)
4
or 1e treatment.
4
4
4
.17. Apoptosis analysis
4
three times a week and treated for 14 weeks to induce hepatic
carcinoma. Groups three was intraperitoneally injected with 1e
(10 mg/kg) daily for two weeks. After 14 weeks treatment, collected
the liver and blood and the livers were fixed in 4% para-
formaldehyde buffer for subsequent Western blot analysis and
histological analysis.
An annexin V-FITC apoptosis assay kit (Nanjing Key Gen Biotech
Co., Ltd.) was used according to the manufacturer’s protocol. The
percentages of annexin-positive HepG2 cells without PI staining
were determined by flow cytometry (FACS Calibur; BD). The data
were analyzed using Cell Quest software.
4.18. DNA binding test
4.20. Enzyme-linked immunosorbent assay
All spectral analyses for DNA binding experiments were tested
ELISA kits were used to determine the serum levels of tumor
necrosis factor, IL-6, IL-8 and TrxR following the manufacturer’s
protocols (Nanjing Sen Bei Jia Biological Technology Co., Ltd.,
Nanjing, China).
in Tris-HCl buffer (5 mM, KCl 50 mM, pH ¼ 7.8). The ct-DNA
ꢀ
(
4.439 mM) stock solution was stored in a refrigerator at 4 C and
stored for less than 5 d before use. 1e was dissolved in DMF and
mixed into a 2 mM stock solution, which was diluted to the cor-
responding concentration with Tris-HCl buffer for subsequent
spectral experiments. Similarly, 2 mM SDS stock solution was pre-
pared by dissolving SDS in Tris-HCl buffer.
4.21. Biochemical analysis
Serum was separated from whole blood and assayed for lactate
dehydrogenase (LDH) and alanine aminotransferase (ALT), alkaline
phosphatase (ALP) using a commercial assay kit (Nanjing Institute
of Bioengineering, China), and levels of aspartate aminotransferase
(AST) were tested according to the manufacturer’s agreement. The
plate reader PerkinElmer Fusion Reader was used to determine the
absorbance values.
4
.18.1. UVeVis spectral analysis
1
e stock solution was diluted 100 times with Tris-HCl buffer to
5
obtain 2.0 ꢁ 10 M solution (3 mL), then gradually add ct-DNA stock
solution or SDS stock solution, When the mixture solution had
reacted sufficiently, the UVeVis absorption spectra were recorded.
4.18.2. Fluorescence spectral analysis
Notes
Fluorescence emission spectroscopy was used to study the
competitive binding experiments between EB or complex 1e and
DNA by varying the concentration of 1e and maintaining the sta-
tionary concentration of EB-ct-DNA system.
The authors declare that they have no conflict of interest.
Acknowledgments
4
.18.3. Circular dichroism spectral analysis
We thank the financial supports of the National Natural Science
Foundation of China (No. 81703337), the Jiangsu Specially-
Appointed Professors program and the Open Project of State Key
Laboratory of Natural Medicines (No. SKLNMKF201808,
SKLNMKF201712), the State Key Laboratory of Coordination
Chemistry, Nanjing University, the Six Talent Peaks Project in
Jiangsu Province of China (No. SWYY-069).
ꢂ4
Complex 1e was added to ct-DNA (1.0 ꢁ 10 M) solution and
ꢀ
ꢂ4
incubated at 37 C for 5 min. The final concentration of 1e were 0,
2
range of 200e320 nm by CD spectrometer.
.0, 4.0, 6.0 ꢁ 10 M and each sample was recorded the data in the
4.19. Tumor nude mice model
7
HepG2 cells (1.5 ꢁ 10 /mouse) were inoculated subcutaneously
Appendix A. Supplementary data
into male BALB/c nude mice (18e22 g) to create HepG2 liver cancer
mouse models. After the tumor reached 180e200 mm , the mice
were randomly divided into two groups (n ¼ 4), and 1e (10 mg/kg)
or the same volume of normal saline was intraperitoneally injected
every day, sustained 15 days, using a vernier caliper to measure the
3
Supplementary data associated with this article can be found in
the online version, at https://doi.org/10.1016/j.ejmech.2019.111721.
These data include MOL file and InChiKeys of the most important
compounds described in this article.

R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
15
References
anticancer activity in 2D and 3D spheroidal models of HeLa cancer cells, Eur. J.
Med. Chem. 145 (2018) 291e301.
[
[
23] W. Cai, L. Zhang, Y. Song, B. Wang, B. Zhang, X. Cui, G. Hu, Y. Liu, J. Wu, J. Fang,
Small molecule inhibitors of mammalian thioredoxin reductase, Free Radical
Biol. Med. 52 (2012) 257e265.
24] J.L. Hickey, R.A. Ruhayel, P.J. Barnard, M.V. Baker, S.J. Berners-Price,
A. Filipovska, Mitochondria-targeted chemotherapeutics: the rational design
of gold(I) N-heterocyclic carbene complexes that are selectively toxic to
cancer cells and target protein selenols in preference to thiols, J. Am. Chem.
Soc. 130 (2008) 12570e12571.
[
1] W.Q. Chen, R.S. Zheng, P.D. Baade, S.W. Zhang, H.M. Zeng, F. Bray, A. Jemal,
X.Q. Yu, J. He, Cancer statistics in China, 2015, Ca-cancer, J. Clin. 66 (2016)
115e132.
[
2] Y. Jiang, A. Sun, Y. Zhao, W. Ying, H. Sun, X. Yang, B. Xing, W. Sun, L. Ren, B. Hu,
C. Li, L. Zhang, G. Qin, M. Zhang, N. Chen, M. Zhang, Y. Huang, J. Zhou, Y. Zhao,
M. Liu, X. Zhu, Y. Qiu, Y. Sun, C. Huang, M. Yan, M. Wang, W. Liu, F. Tian, H. Xu,
J. Zhou, Z. Wu, T. Shi, W. Zhu, J. Qin, L. Xie, J. Fan, X. Qian, F. He, F. He, X. Qian,
J. Qin, Y. Jiang, W. Ying, W. Sun, Y. Zhu, W. Zhu, Y. Wang, D. Yang, W. Liu,
Q. Liu, X. Yang, B. Zhen, Z. Wu, J. Fan, H. Sun, J. Qian, T. Hong, L. Shen, B. Xing,
P. Yang, H. Shen, L. Zhang, S. Cheng, J. Cai, X. Zhao, Y. Sun, T. Xiao, Y. Mao,
X. Chen, D. Wu, L. Chen, J. Dong, H. Deng, M. Tan, Z. Wu, Q. Zhao, Z. Shen,
X. Chen, Y. Gao, W. Sun, T. Wang, S. Liu, L. Lin, J. Zi, X. Lou, R. Zeng, Y. Wu,
S. Cai, B. Jiang, A. Chen, Z. Li, F. Yang, X. Chen, Y. Sun, Q. Wang, Y. Zhang,
G. Wang, Z. Chen, W. Qin, Z. Li, C. Chinese, Human Proteome Project, Prote-
omics identifies new therapeutic targets of early-stage hepatocellular carci-
noma, Nature 567 (2019) 257e261.
[
25] C. Angela, M. Luigi, Molecular mechanisms and proposed targets for selected
anticancer gold compounds, Curr. Top. Med. Chem. 11 (2011) 2647e2660.
26] A. Meyer, C.P. Bagowski, M. Kokoschka, M. Stefanopoulou, H. Alborzinia,
S. Can, D.H. Vlecken, W.S. Sheldrick, S. Wolfl, I. Ott, On the biological prop-
erties of alkynyl phosphine gold(I) complexes, Angew Chem. Int. Ed. Engl. 51
[
(2012) 8895e8899.
[
27] O. Rackham, A.M.J. Shearwood, R. Thyer, E. McNamara, S.M.K. Davies,
B.A. Callus, A. Miranda-Vizuete, S.J. Berners-Price, Q. Cheng, E.S.J. Arner,
A. Filipovska, Substrate and inhibitor specificities differ between human
cytosolic and mitochondrial thioredoxin reductases: implications for devel-
opment of specific inhibitors, Free Radical Biol. Med. 50 (2011) 689e699.
28] W. Liu, K. Bensdorf, M. Proetto, A. Hagenbach, U. Abram, R. Gust, Synthesis,
characterization, and in vitro studies of bis[1,3-diethyl-4,5-diarylimidazol-2-
ylidene]gold(I/III) complexes, J. Med. Chem. 55 (2012) 3713e3724.
[
[
[
3] F.M. Lu, H. Zhuang, Management of hepatitis B in China, Chin. Med. J. 122
(
2009) 3e4.
4] A. Jemal, R. Siegel, E. Ward, Y. Hao, J. Xu, M.J. Thun, Cancer statistics, 2009, CA
A Cancer J. Clin. 59 (2009) 225e249.
5] R.S. Finn, A.X. Zhu, W. Farah, J. Almasri, F. Zaiem, L.J. Prokop, M.H. Murad,
K. Mohammed, Therapies for advanced stage hepatocellular carcinoma with
macrovascular invasion or metastatic disease: a systematic review and meta-
analysis, Hepatology 67 (2018) 422e435.
[
[
29] T. Zou, C.T. Lum, S.S. Chui, C.M. Che, Gold(III) complexes containing N-het-
erocyclic carbene ligands: thiol “switch-on” fluorescent probes and anti-
cancer agents, Angew Chem. Int. Ed. Engl. 52 (2013) 2930e2933.
[
6] J.C. Nault, O. Sutter, P. Nahon, N. Ganne-Carri eꢀ , O. S eꢀ ror, Percutaneous treat-
[
[
[
30] J. G eꢀ rard, V. Anne, T. Siden, Ferrocifen type anti cancer drugs, Chem. Soc. Rev.
ment of hepatocellular carcinoma: state of the art and innovations, J. Hepatol.
44 (2015) 8802e8817.
68 (2017) 783e797.
31] W. Liu, R. Gust, Update on metal N-heterocyclic carbene complexes as po-
tential anti-tumor metallodrugs, Coord. Chem. Rev. 329 (2016) 191e213.
32] J.J. Zhang, J.K. Muenzner, M.A. Abu el Maaty, B. Karge, R. Schobert, S. Wolfl,
I. Ott, A multi-target caffeine derived rhodium(I) N-heterocyclic carbene
complex: evaluation of the mechanism of action, Dalton Trans. 45 (2016)
[
7] J.S. Dawane, V.A. Pandit, Understanding redox homeostasis and its role in
cancer, J. Clin. Diagn. Res. 6 (2012) 1796e1802.
[
8] E.S.J. Arn eꢀ r, Focus on mammalian thioredoxin reductases–Important seleno-
proteins with versatile functions, Biochim. Biophys. Acta Bioenerg. 1790
(
2009) 495e526.
9] S.E. Eriksson, S. Prast-Nielsen, E. Flaberg, L. Szekely, E.S.J. Arn eꢀ r, High levels of
thioredoxin reductase modulate drug-specific cytotoxic efficacy, Free
Radical Biol. Med. 47 (2009) 1661e1671.
13161e13168.
[
[
33] L. Oehninger, L.N. Kuster, C. Schmidt, A. Munoz-Castro, A. Prokop, I. Ott,
A chemical-biological evaluation of rhodium(I) N-heterocyclic carbene com-
plexes as prospective anticancer drugs, Chem. Eur J. 19 (2013) 17871e17880.
34] W. Liu, K. Bensdorf, M. Proetto, U. Abram, A. Hagenbach, R. Gust, NHC gold
halide complexes derived from 4,5-diarylimidazoles: synthesis, structural
analysis, and pharmacological investigations as potential antitumor agents,
J. Med. Chem. 54 (2011) 8605e8615.
1
[
[
10] J. Zhang, X. Li, X. Han, R. Liu, J. Fang, Targeting the thioredoxin system for
cancer therapy, Trends Pharmacol. Sci. 38 (2017) 794e808.
11] M. Bian, R. Fan, S. Zhao, W. Liu, Targeting the thioredoxin system as a strategy
for cancer therapy, J. Med. Chem. 62 (2019) 7309e7321. https://doi.org/10.
[
1021/acs.jmedchem.8b01595.
[
35] W. Liu, K. Bensdorf, A. Hagenbach, U. Abram, B. Niu, A. Mariappan, R. Gust,
Synthesis and biological studies of silver N-heterocyclic carbene complexes
derived from 4,5-diarylimidazole, Eur. J. Med. Chem. 46 (2011) 5927e5934.
36] L. Oehninger, S. Spreckelmeyer, P. Holenya, S.M. Meier, S. Can, H. Alborzinia,
J. Schur, B.K. Keppler, S. Wolfl, I. Ott, Rhodium(I) N-heterocyclic carbene
bioorganometallics as in vitro antiproliferative agents with distinct effects on
cellular signaling, J. Med. Chem. 58 (2015) 9591e9600.
[
12] I.S. Harris, A.E. Treloar, S. Inoue, M. Sasaki, C. Gorrini, K.C. Lee, K.Y. Yung,
D. Brenner, C.B. Knobbe-Thomsen, M.A. Cox, A. Elia, T. Berger, D.W. Cescon,
A. Adeoye, A. Brustle, S.D. Molyneux, J.M. Mason, W.Y. Li, K. Yamamoto,
A. Wakeham, H.K. Berman, R. Khokha, S.J. Done, T.J. Kavanagh, C.W. Lam,
T.W. Mak, Glutathione and thioredoxin antioxidant pathways synergize to
drive cancer initiation and progression, Cancer Cell 27 (2015) 211e222.
13] H. Lei, G. Wang, J. Zhang, Q.J. Han, Inhibiting TrxR suppresses liver cancer by
inducing apoptosis and eliciting potent antitumor immunity, Oncol. Rep. 40
[
[
[
[
[37] H. Petzold, P.J. Sadler, Oxidation induced by the antioxidantglutathione (GSH),
Chem. Commun. 50 (2008) 4413e4415.
(
2018) 3447e3457.
[38] I. Ott, X. Qian, Y. Xu, D.H.W. Vlecken, I.J. Marques, D. Kubutat, J. Will,
W.S. Sheldrick, P. Jesse, A. Prokop, C.P. Bagowski, A gold(I) phosphine complex
14] X. Zheng, W. Ma, R. Sun, H. Yin, F. Lin, Y. Liu, W. Xu, H. Zeng, Butaselen pre-
vents hepatocarcinogenesis and progression through inhibiting thioredoxin
reductase activity, Redox Biol. 14 (2018) 237e249.
containing
proliferative agent and angiogenesis inhibitor, J. Med. Chem. 52 (2009)
63e770.
a naphthalimide ligand functions as a TrxR inhibiting anti-
15] O. Rackham, A.M. Shearwood, R. Thyer, E. McNamara, S.M. Davies, B.A. Callus,
A. Miranda-Vizuete, S.J. Berners-Price, Q. Cheng, E.S. Arner, A. Filipovska,
Substrate and inhibitor specificities differ between human cytosolic and
mitochondrial thioredoxin reductases: implications for development of spe-
cific inhibitors, Free Radical Biol. Med. 50 (2011) 689e699.
7
[
39] J.L. Tsai, A.O. Chan, C.M. Che, A luminescent cyclometalated gold(iii)-avidin
conjugate with a long-lived emissive excited state that binds to proteins
and DNA and possesses anti-proliferation capacity, Chem. Commun. 51 (2015)
8547e8550.
[
[
[
16] L. Bj o€ rkhem, H. Teclebrhan, E. Kesen, J.M. Olsson, L.C. Eriksson, M. Bj o€ rnstedt,
[
[
40] H. Pelicano, D. Carney, P. Huang, ROS stress in cancer cells and therapeutic
implications, Drug Resist. Updates 7 (2004) 97e110.
41] S. Keisuke, H. Hitomi, T. Eriko, M. Reiko, I. Masako, I. Yuji, S. Kazuhiko,
S. Makoto, M. Kunitoshi, Antioxidant N-acetyl-L-cysteine (NAC) supplemen-
tation reduces reactive oxygen species (ROS)-mediated hepatocellular tumor
promotion of indole-3-carbinol (I3C) in rats, J. Toxicol. Sci. 36 (2011)
Increased levels of cytosolic thioredoxin reductase activity and mRNA in rat
liver nodules, J. Hepatol. 35 (2001) 259e264.
17] Y. Min-Hyuk, X. Xue-Ming, B.A. Carlson, V.N. Gladyshev, D.L. Hatfield, Thio-
redoxin reductase 1 deficiency reverses tumor phenotype and tumorigenicity
of lung carcinoma cells, J. Biol. Chem. 281 (2006) 13005e13008.
18] C. Skogastierna, M. Johansson, P. Parini, M. Eriksson, L.C. Eriksson, L. Ekstr o€ m,
775e786.
L. Bj o€ rkhem-Bergman, Statins inhibit expression of Thioredoxin reductase 1 in
rat and human liver and reduce tumor development, Biochem. Biophys. Res.
Commun. 417 (2012) 1046e1051.
[42] J.D. Ly, D.R. Grubb, A. Lawen, The mitochondrial membrane potential(Djm)in
apoptosis; an update, Apoptosis 8 (2003) 115e128.
[
43] Z. Chen, T. Shi, L. Zhang, P. Zhu, M. Deng, C. Huang, T. Hu, L. Jiang, J. Li,
Mammalian drug efflux transporters of the ATP binding cassette (ABC) family
in multidrug resistance: a review of the past decade, Cancer Lett. 370 (2016)
[
[
19] C. Anna, F. Alessandra, B. Alberto, P. Pascal, T. Siden, V. Anne, S. Mich eꢁ le,
J. G eꢀ rard, R. Maria Pia, Evidence for targeting thioredoxin reductases with
ferrocenyl quinone methides.
proliferative effect of hydroxyferrocifens on cancer cells, J. Med. Chem. 57
2015) 8849e8859.
A possible molecular basis for the anti-
153e164.
[
[
[
[
44] E.S.J. Arner, A. Holmgren, The thioredoxin system in cancer, Semin. Cancer
Biol. 16 (2006) 420e426.
45] L.U. Xiao ye, X.Y. Zhong, Caspases and apoptosis, Essays Biochem. 35 (2000)
(
20] O. Karaca, V. Scalcon, S.M. Meier-Menches, R. Bonsignore, J.M.J.L. Brouwer,
F. Tonolo, A. Folda, M.P. Rigobello, F.E. Kuhn, A. Casini, Characterization of
hydrophilic gold(I) N-heterocyclic carbene (NHC) complexes as potent TrxR
inhibitors using biochemical and mass spectrometric approaches, Inorg.
Chem. 56 (2017) 14237e14250.
21] J. Zhang, B. Zhang, X. Li, X. Han, R. Liu, J. Fang, Small molecule inhibitors of
mammalian thioredoxin reductase as potential anticancer agents: an update,
Med. Res. Rev. 39 (2018) 5e39.
9
e19.
46] C.G. Hartinger, P.J. Dyson, Bioorganometallic chemistry–from teaching para-
digms to medicinal applications, Chem. Soc. Rev. 38 (2009) 391e401.
47] A. Wee Han, M. Myatnoezin, S.J. Lippard, Transcription inhibition by
platinum-DNA cross-links in live mammalian cells, J. Am. Chem. Soc. 132
[
[
(2010) 7429e7435.
[
[
48] G. Gilles, O. Ingo, M.N. Nils, Organometallic anticancer compounds, J. Med.
Chem. 54 (2011) 3e25.
49] J.J. Wilson, S.J. Lippard, Synthetic methods for the preparation of platinum
22] T.S. Reddy, S.H. Priver, N. Mirzadeh, S.K. Bhargava, Synthesis of gold(I) phos-
phine complexes containing the 2-BrC6F4PPh2 ligand: Evaluation of

16
R. Fan et al. / European Journal of Medicinal Chemistry 183 (2019) 111721
anticancer complexes, Chem. Rev. 45 (2014) 4470e4495.
50] S.M. Meier-Menches, C. Gerner, W. Berger, C.G. Hartinger, B.K. Keppler,
Structure-activity relationships for ruthenium and osmium anticancer agents
towards clinical development, Chem. Soc. Rev. 47 (2017) 909e928.
51] O. Karaca, S.M. Meier-Menches, A. Casini, F.E. Kühn, On the binding modes of
metal NHC complexes with DNA secondary structures: implications for
therapy and imaging, Chem. Commun. 53 (2017) 8249e8260.
52] L. Ma, W. Na, R. Ma, C. Li, M.K. Tse, G. Zhu, Monochalcoplatin: an actively
transported, quickly reducible, and highly potent Pt(IV) anticancer prodrug,
Angew Chem. Int. Ed. Engl. 57 (2018) 9098e9102.
53] W. Streciwilk, A. Terenzi, F.L. Nardo, P. Prochnow, J.E. Bandow, B.K. Keppler,
I. Ott, Synthesis and biological evaluation of organometallic complexes
bearing bis-1,8-naphthalimide ligands, Eur. J. Inorg. Chem. 26 (2018)
consequences for anti-cancer activity, Metallomics 4 (2012) 628e632.
[
[
[
[
[58] J. Cao, Y. Zheng, X.W. Wu, C.P. Tan, M.H. Chen, N. Wu, L.N. Ji, Z. Mao, Anti-
cancer cyclometalated iridium(III) complexes with planar ligands: mito-
chondrial DNA damage and metabolism disturbance, J. Med. Chem. 62 (2019)
3311e3322.
[59] Z. Zhu, Z. Wang, C. Zhang, Y. Wang, H. Zhang, Z. Gan, Z. Guo, X. Wang,
Mitochondrion-targeted platinum complexes suppressing lung cancer
through multiple pathways involving energy metabolism, Chem. Sci. 10
(2019) 3089e3095.
-
€
[60] P. Newell, A. Villanueva, S.L. Friedman, K. Koike, J.M. Llovet, Experimental
models of hepatocellular carcinoma, J. Hepatol. 48 (2008) 858e879.
[61] M.T. Guimar a~ es, C. Eleazar, F.J.H. Da, L.O.A.C. Rodrigues, S.F. Duarte,
R.J.M.A. Fontenelle, G.O.F.H. Ferreira, D.A.L.A. Carneiro, Review of experi-
mental models for inducing hepatic cirrhosis by bile duct ligation and carbon
tetrachloride injection, Acta Cir. Bras. 27 (2012) 589e594.
3104e3112.
[
[
54] W. Streciwilk, A. Terenzi, X. Cheng, L. Hager, Y. Dabiri, P. Prochnow,
J.E. Bandow, S. Wolfl, B.K. Keppler, I. Ott, Fluorescent organometallic rho-
dium(I) and ruthenium(II) metallodrugs with 4-ethylthio-1,8-naphthalimide
ligands: antiproliferative effects, cellular uptake and DNA-interaction, Eur. J.
Med. Chem. 156 (2018) 148e161.
55] J.R. McConnell, D.P. Rananaware, D.M. Ramsey, K.N. Buys, M.L. Cole,
S.R. McAlpine, A potential rhodium cancer therapy: studies of a cytotoxic
organorhodium(I) complex that binds DNA, Bioorg. Med. Chem. Lett 23 (2013)
[62] K. Wang, C. Zhu, Y. He, Z. Zhang, W. Zhou, N. Muhammad, Y. Guo, X. Wang,
Z. Guo, Restraining cancer cells by dual metabolic inhibition with
mitochondrion-targeted platinum(II) complex, Angew. Chem. Int. Ed. 58
(2019) 4638e4643.
a
[63] H.J. Kim, H.S. Yoo, J.C. Kim, C.S. Park, M.S. Choi, M. Kim, H. Choi, J.S. Min,
Y.S. Kim, S.W. Yoon, J.K. Ahn, Antiviral effect of Curcuma longa Linn extract
against hepatitis B virus replication, J. Ethnopharmacol. 124 (2009) 189e196.
[64] S. Li, Q. Wang, Y. Tao, C. Liu, Swertiamarin attenuates experimental rat hepatic
2527e2531.
[
56] D.L. Ma, C.M. Che, A bifunctional platinum(II) complex capable of intercalation
and hydrogen-bonding interactions with DNA: binding studies and cytotox-
icity, Chem. Eur J. 9 (2003) 6133e6144.
fibrosis by suppressing angiotensin II-angiotensin type
extracellular signal-regulated kinase signaling, J. Pharmacol. Exp. Ther. 359
(2016) 247e255.
1
receptor-
[57] J. Reedijk, Fast and slow versus strong and weak metal-DNA binding: