← Back
New Organometallic Ruthenium(II) Compounds Synergistically Show Cytotoxic, Antimetastatic and Antiangiogenic Activities for the Treatment of Metastatic Cancer.
Accepted Article
Accepted Article
Title: New Organometallic Ruthenium(II) Compounds Synergistically
Show Cytotoxic, Antimetastatic and Antiangiogenic Activities for
the Treatment of Metastatic Cancer
Authors: Yuchen Wang, Jiahui Jin, Liwei Shu, Tongyu Li, Siming Lu,
Mohamed Kasim Mohamed Subarkhan, Chao Chen, and
Hangxiang Wang
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202002970
Link to VoR: https://doi.org/10.1002/chem.202002970
01/2020
10.1002/chem.202002970
Chemistry - A European Journal
New Organometallic Ruthenium(II) Compounds Synergistically
Show Cytotoxic, Antimetastatic and Antiangiogenic Activities
for the Treatment of Metastatic Cancer
Yuchen Wang,a,b Jiahui Jin,c Liwei Shu,a Tongyu Li,a Siming Lu,d Mohamed Kasim
a
The First Affiliated Hospital; Key Laboratory of Combined Multi-Organ Transplantation,
Ministry of Public Health, School of Medicine, Zhejiang University, Hangzhou, 310003, PR
China.
b
Department of Chemical Engineering, Zhejiang University, Hangzhou, 310027, PR China
c
Xingzhi College, Zhejiang Normal University, Jinhua, 321004, PR China
d
Department of Laboratory Medicine, the First Affiliated Hospital, School of Medicine,
Zhejiang University, Hangzhou, 310003, PR China.
e
College of Life Sciences, Huzhou University, Huzhou, 313000, PR China.
Conflicts of Interest Disclosure Statement: The authors declare no potential conflicts of
interest.
Corresponding Authors
*Hangxiang Wang (wanghx@zju.edu.cn)
*Chao Chen (chenc@zjhu.edu.cn)
Acknowledgements
The authors greatly acknowledge financial support from the Zhejiang Province
Preeminence Youth Fund (Grant No. LR19H160002), the National Natural Science
Foundation of China (Grant Nos. 81773193 and 81571799), and the National Science and
Technology Major Project (Grant No. 2017ZX10203205).
Running title: Ruthenium(II) compounds show efficacy against metastatic cancer
Word count: 4206 (main text) Figures: 7 Tables: 1
1
This article is protected by copyright. All rights reserved.
Accepted Manuscript
Mohamed Subarkhan,a Chao Chene,* Hangxiang Wanga,*
10.1002/chem.202002970
Chemistry - A European Journal
Abstract
In this study, we newly designed and synthesized a small library of ten structurally
related C,N-cyclometalated ruthenium(II) complexes containing various pyridinefunctionalized NHC ligand and chelating bipyridyl ligands (e.g., 2,2’-bipyridine, 5,5’dimethyl-2,2’-bipyridine, and 1,10-phenanthroline (phen)). The complexes were well
structure analyses. Among the new ruthenium(II) derivatives, we identified that the complex
Ru8 bearing bulky moieties (i.e., phen and pentamethyl benzene) had the most potent
cytotoxicity against all tested cancer cell lines, generating dose- and cell line-dependent
IC50 values at the range of 3.3-15.0 μM. More significantly, Ru8 not only efficiently inhibited
the antimetastasis process against invasion and migration of tumor cells but also exhibited
potent antivascular effects by suppressing HUVEC cells migration and tube formation in
vitro and blocking vessel generation in vivo (chicken chorioallantoic membrane model). In a
metastatic A2780 tumor xenograft-bearing mouse model, administration of Ru8
outperformed antimetastatic agent NAMI-A and clinically approved cisplatin in terms of
antitumor efficacy and inhibition of metastases to other organs. Overall, these data provided
compelling evidence that the new cyclometalated ruthenium complex Ru8 is an attractive
agent because of synergistically suppressing bulky tumors and metastasized tumor nudes.
Therefore, the complex Ru8 deserves further investigations.
KEYWORDS: ruthenium(II) complexes, cytotoxicity, antiangiogenesis, antimetastasis,
cancer chemotherapy
2
This article is protected by copyright. All rights reserved.
Accepted Manuscript
characterized by NMR, electrospray ionization-mass spectrometry, and single-crystal X-ray
10.1002/chem.202002970
Chemistry - A European Journal
Introduction
Chemotherapy is still regarded as the backbone for the management of patients with
cancer and has improved the life expectancy of countless patients. Platinum-based drugs,
including cisplatin, oxaliplatin, and carboplatin, have been extensively applied in the clinic
for cancer chemotherapy [1]. However, the clinical efficacy of these platinum agents in
(e.g., neurotoxicity and nephrotoxicity), inherent or acquired resistance and inability to
impede tumor metastasis [3]. Clinically, metastatic cancer accounts for the majority of
human deaths. Systemic administration of platinum drugs has produced local tumor control
but is not capable of suppressing treatment escape pathways. For example, emerging
evidence indicates that the repeated use of platinum chemotherapies is frequently
associated with the dissemination of cancer cells, which eventually facilitates metastases to
distant organs [4]. Towards the goal of addressing this medication challenge, new effective
therapies that simultaneously combine cytotoxic, antimetastatic, and antiangiogenic
activities are urgently needed and would be likely to provide long-term survival benefit to
patients when clinically used.
Of the alternatives to platinum-based drugs, significant advances have been made with
ruthenium-based agents that are appealing to exploration [5]. Compared to platinum drugs,
ruthenium complexes generally show fewer side effects, no cross-resistance with cisplatin,
and more diverse biological activities in vitro and in vivo [6]. Currently, several ruthenium(III)
complexes (Figure 1A), including NAMI-A (imidazolium trans[tetrachloro(dimethylsulfoxide)(1H-imidazole)ruthenate(III)]), which is effective against solid
metastatic tumors [4], and KP1019 (indazolium trans-[tetrachlorobis(1Hindazole)ruthenium(III)]), which is used to combat resistant tumors [7], have entered clinical
trials [8]. Unfortunately, these compounds have not yet shown the expected favorable
clinical outcomes in cancer patients [9], stressing the need for the development of new
metallodrugs. In addition to ruthenium(III) agents, ruthenium(II) arene complexes have
attracted particular interest as anticancer drug candidates for clinical investigation due to
certain advantages, including stability, aqueous solubility, structural diversity, and broad
modes of action in biological systems [10]. Recently, a novel ruthenium (II) complex
TLD1433, a photosensitizer, has entered the clinical trial for the treatment of non-muscle
3
This article is protected by copyright. All rights reserved.
Accepted Manuscript
patients has been greatly compromised by several drawbacks, including severe toxicities [2]
10.1002/chem.202002970
Chemistry - A European Journal
invasive bladder cancer through the combination of photochemotherapy (PCT) and
photodynamic therapy (PDT) [11]. Of the numerous ruthenium(II) compounds evaluated,
some complexes [12] exhibited clinically relevant antimetastatic and antiangiogenic potential.
It was reported that a series of ruthenium(II) complexes bearing the cinnamic acid moiety
displayed desirable antimetastatic activities against migration and invasion in cancer cells
. We recently also demonstrated that tetranuclear ruthenium(II) arene complexes are
promising anticancer agents that simultaneously span cytotoxic and antimetastatic
mechanisms, as well as alleviated systemic toxicity in animals [14]. More intriguingly,
ruthenium(II) complexes generally exhibit a lower systemic toxicity than other metallodrugs.
Cyclometallated ruthenium(II) complexes have recently demonstrated great potential
as a new class of metallodrug candidates [9]. In these cycloruthenated complexes, the
ligands are generally constructed with N- and C-donors. Compared to the ruthenium-N, the
metal-to-ligand bond distances of ruthenium-C are significantly shorter, which endows the
complexes with high stability in biological systems [15]. In addition, the cyclometalation
decreases the valence charge of ruthenium(II) complexes and contributes to an increase in
the lipophilicity and cellular uptake of the complexes [16]. As one of the well-documented
cycloruthenated complexes, [Ru(bpy)(phpy)(dppz)]+ (Figure 1A) showed superior activity
against a panel of 2D cancer cell lines and 3D multicellular tumor spheroids compared to
cisplatin [17]. Moreover, a cyclometallated ruthenium (II) complex RDC11 was recently
demonstrated to possess promising antiangiogenic activity through the inhibition of the
HIF1 pathway [18]. Furthermore, recent studies showed that cyclometallated ruthenium(II)
polypyridyl complexes could be activated by near-infrared light irradiation, thereby serving
candidates for PCT and PDT [19]. In these scaffolds, cyclometallated ligands were usually
based on phenylpyridine and its derivatives, in which the structure diversity was limited due
to the requirement of tedious synthesis for chemical derivatization of the phenylpridine ring.
Accordingly, N-Heterocyclic carbenes (NHCs), in which the N-substituents of imidazole ring
could be easily converted to various functional moieties , have attracted increasing interest.
In addition, NHC ligands serve as excellent σ-donors, which may make the ruthenium
complexes potentially stable [20]. Prompted by these studies, we designed novel effective
NHC-coordinated cyclometallated ruthenium(II) complexes that spanned cytotoxic,
antimetastatic and antiangiogenic mechanisms to address the therapeutic challenges
4
This article is protected by copyright. All rights reserved.
Accepted Manuscript
[13]
10.1002/chem.202002970
Chemistry - A European Journal
against metastatic cancer.
Towards this goal, we synthesized a small library of structurally relevant ruthenium(II)
complexes, and their chemical structures were characterized by NMR spectra and X-ray
crystallography studies. The cytotoxic activity in five cancer cell lines (including one
cisplatin-resistant cancer cell) was assessed by the MTT assay, yielding dose-dependent
potency across distinct cancer cell lines. The unique feature of this agent is the ability to
simultaneously kill cancer cells and impede metastasis, as well as angiogenesis in cellbased assays and in vivo studies. Our results showed that NHC-coordinated
cyclometallated ruthenium(II) complexes possess multiple biological activities, thereby
making them attractive for further investigations and clinical development.
Results
Synthesis of ruthenium(II) complexes
Ruthenium-NHC complexes Ru1 and Ru5 were synthesized according to the protocol
[21]
shown in Figure 1B. The reaction of the in situ generated silver-NHC complexes, a
carbene transfer agent, with one equivalent of [Ru(p-cymene)Cl2]2 and NH4PF6 in
acetonitrile readily produced [RuL1(CH3CN)4](PF6)2 (Ru1) and [RuL2(CH3CN)4](PF6)2 (Ru5)
in high yields (56-66%). The NMR spectra of Ru1 and Ru5 showed disappearance of the
imidazolium C2-H protons. In addition, a substantial downfield shift of Ru-C carbon signal
was observed for Ru1 at δ 189.5 ppm and Ru5 at 188.2 ppm, evidencing Ru-C bond
formation. Furthermore, the peaks around 2.0 ppm ascribed to acetonitrile appeared in the
purified complexes, suggesting successful coordination of acetonitrile ligand.
Motivated by the ease of compound synthesis, we further attempted to expand the
library of ruthenium(II) derivatives. For this purpose, we reacted Ru1 or Ru5 with bidentate
ligands (e.g., 2,2’-bipyridine (bpy), 5,5’-dimethyl-2,2’-bipyridine (dmbpy), and 1,10phenanthroline (phen)) in reflux acetonitrile. This procedure efficiently yielded additional six
ruthenium-NHC complexes (Figure 1B). Following a similar protocol, we prepared two
chelate ruthenium-NHC complexes coordinating phen ligand [RuL3(phen)(CH3CN)2](PF6)2
(Ru9) and [RuL4(phen)(CH3CN)2](PF6)2 (Ru10), in which the R group was substituted to
pentafluorophenyl and methyl, respectively, compared to Ru4 and Ru8. The resulting
5
This article is protected by copyright. All rights reserved.
Accepted Manuscript
IC50 values. Of the ten ruthenium complexes studied, Ru8 showed the highest cytotoxic
10.1002/chem.202002970
Chemistry - A European Journal
complexes Ru1-10 were unambiguously characterized by NMR (1H and 13C) (Figure S120). The complexes were further dried and finally subjected to elemental analysis and
electrospray ionization-mass (ESI-MS) spectrometry.
The stability of organometallic complexes plays an important role for their biological
activities. Hence, choosing the complexes Ru6 and Ru8 as model complexes, we assessed
DMSO, PBS, and RPMI-1640 supplemented with 10% fetal bovine serum (FBS), almost
negligible variation was observed for both complexes, manifesting that they were stable in
solutions (Figure S28).
Single-crystal crystallography studies
Having these ruthenium(II) complexes in hand, we selected Ru1-4 and Ru8-10 for
structural validation in the solid state by a single-crystal X-ray structure analysis. Single
crystals suitable for X-ray crystallography studies were obtained by slowly diffusing diethyl
ether into a freshly prepared and concentrated solution of the complexes in acetonitrile. The
molecular structures created by the ORTEP (Oak Ridge thermal-ellipsoid plot) diagram are
presented Figure S21-27. In the complex Ru8, the ruthenium(II) metal is hexa-coordinated
by two ligands (i.e., L2 and phen) and two identical acetonitrile molecules in an octahedral
geometry, which is consistent with the results of NMR characterization. Moreover, the
carbene ligand, one acetonitrile ligand, and one nitrogen atom of phen are presented in the
same equatorial plane, while the remaining coordinated nitrogen atoms of acetonitrile and
chelating ligands lie on the axial positions (Figure 2). The planar aromatic structures are
usually considered to have high affinities to intercalate into DNA double helix [22]. In
addition, the cationic charge may readily cause Ru8 to associate with negatively charged
DNA [23]. The angles (N4-Ru-N5 and C6-Ru-N1) of the adjacent coordination atom are close
to the right angle of 79.4 to 78.7°. Interestingly, the distance of ruthenium-N4 (2.126 Å) at
the opposite position of the carbene ligand is slightly longer than the other four ruthenium-N
bonds (2.044-2.079 Å), which is caused by the trans-effect of the strongly coordinated
carbene ligand. In addition, single crystals of the complexes Ru1-4 and Ru9-10 were
resolved by the same X-ray diffraction analysis, which possess similar structures with Ru8
as shown in Figure S21-27.
6
This article is protected by copyright. All rights reserved.
Accepted Manuscript
the in vitro stability using UV-vis spectrometry. Upon incubation with different media (e.g.,
10.1002/chem.202002970
Chemistry - A European Journal
In vitro cytotoxic activity
The ruthenium(II) complexes Ru1-10 were tested for their cytotoxicity against the
following cell lines: A549 (human lung cancer cells), A549/cisR (cisplatin-resistant human
lung cancer cells), A2780 (human ovarian cancer cells), Huh-7 (human hepatocellular
carcinoma cells), B16-F10 (mice melanoma cancer cells) and HUVEC (human umbilical
the MTT assay. IC50 values (half-maximal inhibitory concentration) were extrapolated from
the dose-response curves to determine the antiproliferative activity of the complexes, and
these values are summarized in Table 1. Cisplatin, a widely used metallodrug in the clinic
to treat cancer [24], was included as the control. Compared to the parent complexes Ru1 or
Ru5, incorporation of bpy, dmbpy, and phen moieties to the ruthenium metal center (i.e.,
complexes Ru2-4 or Ru6-8) improved the cytotoxic activity in all tested cancerous cell lines.
Specifically, the substitution with the phen motif significantly yielded better cytotoxicity than
the others as evidenced by the reduced IC50 values. Interestingly, compared to complexes
Ru1-4 (containing phenyl substitution), Ru5-8 (containing pentamethyl benzene ring)
exhibited ~3- to 8-fold improvements in activity, manifesting the strong structure-activity
relationships.
To further investigate how NHC ligands affect the cytotoxic potency of the compounds,
we additionally synthesized Ru9 and Ru10, in which pentafluorophenyl- and methylsubstituted NHC ligands were coordinated, respectively. Unfortunately, we failed to obtain
more potent complexes according to the cytotoxicity assay data. Together, of the ten
compounds studied, we identified that Ru8 was the most potent against cancer cells
showing comparable cytotoxic activity with cisplatin, and preferentially killed cancer cells
over the non-cancerous cells due to the relatively high IC50 values in HUVEC. Very
interestingly, in cisplatin-resistant A549/cisR cells, the resistance fold of Ru8 was reduced
to 2.2-fold, indicating the ability of reversing drug resistance (Table 1).
UV-vis spectrometric titration using Ru8 was further exploited to examine whether the
interactions between the ruthenium(II) complex and DNA occurred. Upon addition of calf
thymus DNA (CT-DNA), the absorption peaks ascribed to Ru8 centered at 344 nm
gradually decreased (Figure S29). The result revealed the presence of interactions
between Ru8 and DNA base pairs exhibited superior cytotoxicity by interacting with DNA
7
This article is protected by copyright. All rights reserved.
Accepted Manuscript
vein endothelial cells). Following drug exposure for 72 h, cell viability was determined by
10.1002/chem.202002970
Chemistry - A European Journal
base pairs. Moreover, the complex Ru8 showed a favorable partition coefficient, which
indicates good lipophilicity and can contribute to compound penetration across cell
membranes (Table S3). Therefore, the potent cytotoxicity of Ru8 may be attributable to the
bulky volume of the chelating ligands, which favors high cellular uptake and strong
interaction with DNA [17]. Due to the observed high activity in cell-based assays, we
Mode-of-action to induce cell death
We next evaluated cell proliferation inhibited by Ru8 by detecting DNA synthesis using
a 5-ethynyl-2’-deoxyuridine (EdU) assay. This method enables the incorporation of EdU, a
thymidine analogue, into cellular DNA during replication [25]. Subsequently, labeling the EdU
moiety with green fluorescent dye using the click reaction can be used to detect
proliferating cells (Figure 3A). In untreated cells, a large fraction of proliferative cells
(52.2±0.8%) was observed with green fluorescence. However, treatment with complex Ru6
or Ru8 significantly reduced the DNA replication ratios, showing a dose-dependent manner
(Figure 3C). Notably, compared to Ru6, complex Ru8 was more effective in blocking DNA
replication over a wide range of drug concentrations. After treatment with 16 μM Ru8, DNA
synthesis was almost terminated, and the cell proliferation rate was dramatically reduced to
7.7% (Figure 3C).
Next, the acridine orange/ethidium bromide (AO/EB) staining assay was conducted to
validate cell apoptosis induced by Ru8. AO penetrates intact cell membranes and emits
green fluorescence inside cells, while EB only enters apoptotic cells with damaged
membranes and shows orange fluorescence [26]. As shown in Figure 3B, significant
apoptosis was induced in A549 cells after treatment with Ru8 for 48 h. Compared to Ru6,
the apoptotic ratio was more profound in Ru8-treated cells, manifesting the higher
cytotoxicity of complex Ru8. Upon treatment with Ru8, the number of total cells was
distinctly reduced, whereas the apoptosis rate was markedly increased accompanied with
morphological changes of tumor cells (Figure 3D). Thus, the AO/EB results were consistent
with the IC50 values derived from the MTT assay.
Western blot was additionally performed to verify the mode of action of Ru8 in cell
proliferation inhibition and apoptosis induction. Upregulation of apoptosis-related proteins
8
This article is protected by copyright. All rights reserved.
Accepted Manuscript
selected Ru8 for evaluation in ensuing experiments.
10.1002/chem.202002970
Chemistry - A European Journal
(e.g., c-PARP, c-Caspase 3 and c-Caspase 9) confirmed that the cell apoptosis could be
efficiently induced by Ru8 via the classical apoptotic pathway (Figure 3E) [27]. In addition,
G2/M phase transition-related proteins (e.g., cdc25c, cdc2 and cyclin B1) in Ru8-treated
A2780 cells were significantly downregulated compared with that of Ru6 treatment. These
results indicate that the complex Ru8 inhibited cell proliferation by arresting the cell cycle at
phosphorylated cdc25c (p-cdc25c) and cdc2 (p-cdc2) might be attributable to the total
reduction of cdc25c and cdc2 proteins. Thus, the western blot results further confirmed the
cellular mechanism of anticancer Ru8 compound and evidenced that Ru8 had higher
activities than its structurally similar complex Ru6 to inhibit cell proliferation and induce
apoptosis.
We anticipate that the high antiproliferation and apoptosis inducing ability of Ru8 may
be attributable to the coordination of the phen ligand which could benefit the interactions
with DNA. Furthermore, Ru8 was more active than Ru6, which possesses a similar
structure, further validating the importance of the coordination of the phen ligand.
In vitro antimetastatic activity
The migration and invasion of cancer cells are closely related to cancer metastasis to
distant organs [28]. Upon destroying the extracellular matrix, cancer cells invade into lymph
nodes and migrate into healthy tissues through circulation systems, thus promoting the
progression of tumors [12, 29]. Initially, a scratch wound-healing assay was included here to
evaluate the antimetastatic activity of Ru8 using the highly metastatic human HCC Huh-7
cell line (Figure 4A). For comparison, NAMI-A, a potent antimetastatic agent undergoing
clinical trials [30], was selected as a reference. Strikingly, exposure of cells to Ru8 reduced
cell migration, showing a dose-dependent suppression activity. The wound-healing ratio
after incubation with 32 μM Ru8 was reduced to 2.9% compared to untreated cells with
29.6% migrating cells (Figure 4B). Unexpectedly, NAMI-A yielded less antimetastatic
activity than Ru8, suppressing the gap closure with 19.0% at 32 μM, which was comparable
with the activity of Ru8 at 4 μM (Figure 4B). To eliminate the influence of cytotoxicity, the
wound-healing assay was further performed at low drug concentrations (e.g., IC10 and IC20,
Table S4) [31]. Encouragingly, Ru8 remained active against migration as compared to
9
This article is protected by copyright. All rights reserved.
Accepted Manuscript
G2/M phase through cdc25c/cdc2/cyclin B1 pathway [27]. Furthermore, downregulation of
10.1002/chem.202002970
Chemistry - A European Journal
untreated cells. However, at the concentrations of IC10 or IC20, NAMI-A failed to inhibit the
gap closure in the cells (Figure S30).
A transwell assay was further conducted to investigate the efficacy of Ru8 on Huh-7
cell invasion. The cell invasion ratio can be evaluated through counting the number of
invaded cells stained with crystal violet [32]. Following treatment with Ru8, Huh-7 cell
only showed limited activity in preventing cell invasion even at relatively high concentration
(i.e., 32 μM). Similarly, treatments of NAMI-A at low concentrations abolished the
antiinvasive effect, whereas the number of invaded cells in the Ru8-treated group was
remarkably diminished even at the IC10 and IC20 concentrations (Figure S31).
Inspired by these results, western blot was conducted to investigate the mechanisms of
Ru8 against Huh-7 cell metastasis. The enhancement of matrix metalloproteinases (MMPs)
level and the dysregulation of epidermal growth factor receptor (EGFR) were closely
associated with tumor metastasis [33]. Although the total expression of EGFR remained
unaffected after Ru8 treatment, the metastasis-relevant proteins, including MMP9 and pEGFR, were significantly downregulated (Figure 4E). Furthermore, Ru8 showed stronger
inhibitory activity than NAMI-A. Thus, all data demonstrated the potential of using Ru8 as a
promising candidate to combat metastatic cancer.
In vitro and in vivo antiangiogenic activity
Endothelial cell migration plays a critical role in the development of tumor vessels;
therefore, targeting this process is essential for antivascular therapy [34]. Thus, we selected
human umbilical vein endothelial cells (HUVECs) as a model to examine the antiangiogenic
activity of ruthenium(II) complexes. On the basis of the results described above, only Ru8
was used in the ensuing studies. First, we used the wound-healing assay to evaluate the
inhibition of migration of HUVECs in vitro after drug treatment. Cells were treated with
serum-free culture medium in the case of proliferation. Treatment with Ru8 suppressed the
migration of HUVECs in a dose-dependent manner (Figure 5A). For example, the wound
closure ratios were 24.4%, 28.3%, 16.0% and 7.7% after Ru8 treatment at concentrations
at 4, 8, 16, and 32 μM, respectively, whereas the wound closure ratio of untreated cells was
50.0% (p < 0.01, Figure 5B). More strikingly, the wound closure after Ru8 treatment at low
10
This article is protected by copyright. All rights reserved.
Accepted Manuscript
invasion was markedly reduced compared to untreated cells (Figure 4C and 4D). NAMI-A
10.1002/chem.202002970
Chemistry - A European Journal
concentrations was significantly inhibited, manifesting that Ru8 serves as a potent
antivascular agent (Figure S32). A tube formation assay was next performed to assess the
antiangiogenic capability of Ru8. For this purpose, HUVECs were seeded on Matrigel
containing proangiogenic factors. Robust capillary networks should form between single
HUVECs. Interestingly, addition of Ru8 at non-cytotoxic concentrations for 5 h resulted in
concentration (4 μM) for 2 h destroyed the tube-like structures, presenting only 16.5% of
tube formation relative to untreated cells (100%, p < 0.001, Figure 5C and 5D). Increasing
the drug concentration further yielded pronounced activity. Hence, Ru8 showed the
potential to impair the formation of tubes, suggesting potent antiangiogenic activity.
Based on these in vitro results, we next used the chicken embryo chorioallantoic
membrane (CAM) model to investigate if Ru8 is capable of inhibiting angiogenesis and
further blood vessel formation in vivo [35]. The CAM model is extensively used to analyze
the antiangiogenic activity of drug candidates due to its rapid capillary proliferation. At
embryo development day (EDD) 8, fertilized chicken egg tissues were topically treated with
Ru8 at concentrations ranging from 1.03 to 32 μM for 36 h. Figure 6A and Figure S34
show typical images for CAM. The saline-treated CAM developed vascular networks
densely and robustly. Following topical administration of Ru8, significant reduction of
neovascularization on the chick embryo was observed. Moreover, the vessel formation was
suppressed in a concentration-dependent manner. At the concentrations of 16 and 32 μM,
Ru8 substantially destroyed the capillary bed and reduced small vessels observed. The
photographs and quantification of the number of branching points are shown in Figure 6B.
Therefore, the in vivo CAM assays provided compelling evidence that Ru8 inhibits
angiogenesis, indicating an antivascular effect in vivo.
In vivo antitumor efficacy
Prior to the in vivo antitumor activity, we evaluated the maximum tolerated dose (MTD)
of using the complex Ru8. The healthy ICR mice were intraperitoneally administered with
Ru8 at the doses of 5, 10 and 12.5 μmol/kg every day for five injections. Saline was
included as a reference. As shown in Figure S35, the body weights of the mice receiving
Ru8 at 10 μmol/kg were stable, and no incidence of mouse death was observed during the
experiment. Thus, we estimated that the MTD of Ru8 was at least 10 μmol/kg (9.6 mg/kg).
11
This article is protected by copyright. All rights reserved.
Accepted Manuscript
dramatic suppression of tube formation (Figure S33). Treatment with Ru8 at higher
10.1002/chem.202002970
Chemistry - A European Journal
However, the half lethal dose (LD50) of cisplatin was reported to be less than 10 μmol/kg
(3.0 mg/kg) with intraperitoneal injection every other day for 5 times [14]. Hence, we chose
the tolerated dose at 2 μmol/kg of Ru8 for further in vivo therapeutic assessment.
Ovarian cancer is the second most common cause of malignancy-associated deaths
among women due to its high capacity of metastasis and invasion in patients [36]. We
model. This model was established in Balb/c nude mice by intraperitoneal implantation of
human ovarian cancer A2780 cells [37]. When the cells were injected, they disseminated
throughout the abdomen, and the tumor cells metastasized to the ovary. Following cell
implantation, we intraperitoneally injected cisplatin, NAMI-A and Ru8 at a dose of 2 μmol/kg
to initiate therapy, and the daily injections lasted for one week. On day 21
postadministration, mice were sacrificed to examine tumors that localized in the abdomen
and metastasized to ovaries. Obviously, mice treated with cisplatin, NAMI-A and Ru8
showed local tumor control (Figure 7A and 7B), as well as reduced metastatic tumor
burdens in ovaries (Figure 7C and 7D), compared to the DMSO-treated group. Noticeably,
Ru8 demonstrated significantly higher tumor inhibitory activity than the other treatments,
showing an inhibition rate of ~91.4% for abdominal tumors (p < 0.01 versus cisplatin or
NAMI-A treatments, Figure 7B). Further considering the metastatic potential of A2780 cells
to ovaries, we weighed ovaries in each group to assess the antimetastatic capacity.
Encouragingly, in Ru8-treated mice, the ovaries remained in the normal range comparable
to healthy mice (Figure 7C and 7D). In addition, the stable body weights of the mice
supported the low toxicity and high tolerability of Ru8 in animals (Figure 7E). Taken
together, these results demonstrated that Ru8 efficaciously inhibits tumor growth and
suppresses metastases to other organs, making it promising for further clinical investigation
as a multifunctional anticancer agent.
Discussion
The most extensively used metallodrugs such as cisplatin, oxaliplatin, and carboplatin
are known to yield serious side effects, including peripheral neurotoxicity, nephrotoxicity,
and hair loss [38]. In addition, although systemic administration of platinum agents could gain
the local tumor control, this modality generally is not capable of addressing the issues of
tumor recurrence and metastasis. Unfortunately, the majority (∼90%) of cancer patients
12
This article is protected by copyright. All rights reserved.
Accepted Manuscript
therefore assessed the therapeutic efficacy of using Ru8 in a metastatic ovarian tumor
10.1002/chem.202002970
Chemistry - A European Journal
died from metastasis in the clinic [39]. Therefore, it can be rationally envisioned that single
agents synergistically combining anti-metastatic and cytotoxic activities could be promising
to efficiently treat metastatic cancer. To address these unmet medical needs, numerous
transition metal (e.g., Ru, Ag, Au, Cu, Ir, Ni, Pd, and Os) complexes have been investigated
as alternative anticancer therapeutics [4, 40]. Among them, ruthenium complexes have
at the different stages [7, 41]. As a well-documented example, NAMI-A was demonstrated to
suppress tumor metastasis but this agent was not sufficient to kill cancer cells
simultaneously [42]. Consequently, there is a considerable incentive for the development of
new generation metallotherapeutics that synergistically inhibit tumor metastasis and growth
with reduced in vivo toxicity.
For this purpose, in the present study, we designed and synthesized a series of novel
cyclometalated ruthenium(II) complexes (Ru1-10) coordinating with diverse NHC moieties
and bipyridyl ligands (e.g., bpy, dmpy and phen). The geometrical structures of these
complexes were validated via single-crystal X-ray analysis. Compared with many previously
reported ruthenium complexes bearing polypyridyl ligands [16], the cyclometallated ligands
endowed the overall molecules with higher stability and lipophilicity, which could facilitate
cellular uptake and, thereby enhancing the cytotoxic activity [9, 43]. In contrast to the
phenylpyridine (a widely used pharmacophore as cyclometallated ligands), the NHC
moieties provided more chances in the development of structure diversity by converting the
N-substituents of imidazole ring to various functional moieties [44].
The differences in IC50 values clearly revealed the structure-activity relationship for
these compounds. For example, compared with the complexes Ru1 and Ru5, their
derivatives showed increased cytotoxicity due to the substitution of acetonitrile moieties
with bpy, dmbpy and phen, which indicates that the incorporation of the planar aromatic
chelating ligands with bulky volume may be contributable to their high activity [22a, 45].
Specifically, the substitution of the phen group rendered the complex Ru8 with the strong
ability to interact with DNA. Furthermore, we investigated the effect of NHC ligands and
showed that the substitution of pentamethyl benzene moiety was beneficial to its cytotoxic
activity. Of the ten complexes evaluated, Ru8 was proven to be the most potent in inducing
13
This article is protected by copyright. All rights reserved.
Accepted Manuscript
attracted a surge of interest. Several ruthenium compounds have entered the clinical trials
10.1002/chem.202002970
Chemistry - A European Journal
cell apoptosis via the classical apoptotic pathway and inhibiting cell proliferation by
arresting the cell cycle at G2/M phase against tested cancer cells [46].
In addition to the cytotoxic activity, Ru8 was further validated to effectively impede
metastasis of highly invasive Huh-7 cells through the inhibition of cell migration and
invasion through downregulation of some metastasis-associated proteins (e.g., MMP9 and
a potent antimetastatic agent confirmed in many in vitro and in vivo studies [47].
Furthermore, Ru8 suppressed angiogenesis by inhibiting HUVEC migration and tube
formation in vitro, as well as by reducing neovascularization in the CAM model. Thus, we
hypothesized that the multifunctional Ru8 could suppress the tumor progression by
preventing the cancer cells metastasis to distant organs and interrupting the development
of tumor vessels. To test this, we finally employed a metastatic ovarian tumor mouse model
to evaluate the anticancer efficiency of Ru8. Animal studies showed that Ru8 more potently
inhibited the proliferation of tumor cells and suppressed the growth of metastases in ovaries
compared with cisplatin and NAMI-A.
In summary, we newly synthesized ten C,N-cyclometalated ruthenium(II) complexes
Ru1-10 and the optimized complex Ru8 not only showed potent cytotoxic activity but also
effectively impeded tumor metastasis and angiogenesis, thereby synergistically enhancing
the efficacy against metastatic cancer. Collectively, these results showed the effectiveness
of novel cyclometalated ruthenium-NHC complexes, representing a promising scaffold for
further clinical investigations.
Experimental Section
Materials and methods
[Ru(p-cymene)Cl2]2 and calf thymus DNA (CT-DNA) was purchased from SigmaAldrich (Shanghai, China). 2,2’-bipyridine (bpy), 5,5’-dimethyl-2,2’-bipyridine, and 1,10phenanthroline were purchased from Tokyo Chemical Industry (Shanghai, China). Other
reagents were purchased from J&K Chemical (Shanghai, China). All the solvents were
purchased from Tianjin Yongda Chemical Reagent (Tianjin, China). All reactions were
performed in dry solvents. 1H and 13C NMR spectra were recorded on a Bruker Avance-400
(400 MHz) spectrometer at 400 MHz for 1H and 100 MHz for 13C. Chemical shifts () are
14
This article is protected by copyright. All rights reserved.
Accepted Manuscript
p-EGFR). Surprisingly, Ru8 showed much stronger antimetastatic capability than NAMI-A,
10.1002/chem.202002970
Chemistry - A European Journal
expressed in ppm downfield to TMS at = 0 ppm and coupling constants (J) are expressed
in Hz. Elemental analyses were performed with a Flash EA 1112 from ThermoFinnigan.
Mass spectrometry ESI-MS was recorded on an AB Triple TOF 5600 + System (AB SCIEX,
Framingham, USA).
Imidazolium salt HL1PF6 (381 mg, 1 mmol) was dissolved in 25 mL of acetonitrile
(CH3CN) containing dispersive silver oxide (140 mg, 0.6 mmol) and then stirred at 50 °C for
24 h. After cooling the solution to room temperature, [Ru(p-cymene)Cl2]2 (306 mg, 0.5
mmol) was added to the mixture, which was then stirred for 2 h. The reaction mixture was
then heated to reflux for 24 h after adding NH4PF6 (815 mg, 5 mmol). The mixture was
filtered through vacuum filtration, concentrated under reduced pressure and precipitated by
addition of diethyl ether (Et2O) to obtain Ru1. Yield: 521 mg, 66%, light yellow solid. 1H
NMR (400 MHz, CD3CN): δ 8.86-8.88 (m, 1H), 8.09-8.13 (m, 1H), 8.06-8.07 (d, J = 2.0,
1H), 7.83-7.85 (d, J = 4.0, 1H), 7.36-7.46 (m, 5H), 7.28-7.29 (d, J = 2.0, 2H), 5.63 (s, 2H),
2.25 (s, 3H), 2.10 (s, 6H), 1.96 (s, 3H). 13C NMR (100 MHz, CD3CN) δ 189.49 (Ru-C),
155.44, 153.45, 141.08, 137.41, 129.49, 128.62, 127.22, 125.28, 122.84, 118.46, 112.19,
53.41, 3.85, 3.63. Anal. Calcd for C23H25F12N7P2Ru:C, 34.95; H, 3.19; N, 12.40. Found: C,
34.86; H, 3.16; N, 12.31. ESI-MS: calcd C23H25F12N7P2Ru for [M-CH3CN-PF6]+ 605.0597,
found 605.0598.
Synthesis of [RuL1(bpy)(CH3CN)2](PF6)2, (Ru2)
A mixture of Ru1 (158 mg, 0.2 mmol) and the chelating ligand, 2,2’-bipyridine (byp) (32
mg, 0.2 mmol) in CH3CN was refluxed and stirred for 24 h. The solution was evaporated
under vacuum to obtain the crude product, which was further purified by flash column
chromatography via dichloromethane-methanol (15:1). Yield: 97 mg, 56%, light yellow solid.
1
H NMR (400 MHz, CD3CN) δ 8.90-8.92 (m, 1H), 8.07-8.11 (m, 1H), 7.90-7.91 (d, J = 2.0,
1H), 7.78-7.80 (d, J = 4.0, 1H), 7.43-7.46 (m, 1H), 6.70-6.71 (d, J = 2.0, 1H), 5.65 (s, 2H),
2.50 (s, 3H), 2.31 (s, 6H), 2.30 (s, 3H), 2.28 (s, 6H), 2.15 (s, 6H), 1.96 (s, 3H). 13C NMR
(100 MHz, CD3CN) δ 188.17, 155.46, 153.44, 141.03, 136.85, 134.37, 134.02, 128.87,
127.14, 125.32, 124.25, 122.76, 122.72, 112.09, 49.58, 16.94, 16.65, 16.41, 4.14, 3.62.
15
This article is protected by copyright. All rights reserved.
Accepted Manuscript
Synthesis of [RuL1(CH3CN)4](PF6)2, (Ru1)
10.1002/chem.202002970
Chemistry - A European Journal
Anal. Calcd for C28H35F12N7P2Ru: C, 40.29; H, 3.15; N, 11.34. Found: C, 40.32; H, 3.14; N,
11.29. ESI-MS: calcd C28H35F12N7P2Ru for [M-CH3CN-PF6]+ 675.1379, found 675.1362.
Synthesis of [RuL1(dmbpy)(CH3CN)2](PF6)2, (Ru3)
The synthetic protocol of Ru3 was similar with Ru2. 5,5’-dimethyl-2,2’-bipyridine
solid. 1H NMR (400 MHz, CD3CN) δ 9.04 (s, 1H), 8.34-8.36 (d, J = 4.0, 1H), 8.13-8.15 (m,
2H), 8.10-8.12 (m, 1H), 7.89-7.93 (m, 1H), 7.79-7.81 (d, J = 4.0, 1H), 7.67-7.69 (m, 1H),
7.60 (s, 1H), 7.57-7.58 (d, J = 2.0, 1H), 7.38-7.45 (m, 6H), 7.01-7.05 (m, 1H), 5.69-5.83 (m,
2H), 2.62 (s, 3H), 2.24 (s, 3H), 2.13 (s, 3H), 1.97 (s, 3H). 13C NMR (100 MHz, DMSO-d6) δ
193.06, 155.71, 155.34, 155.04, 153.15, 151.19, 150.92, 140.41, 140.13, 138.99, 138.62,
137.79, 137.35, 129.26, 128.25, 127.71, 127.01, 126.42, 126.08, 123.58, 123.39, 123.02,
118.83, 112.49, 53.04, 18.73, 18.30, 4.44, 3.95. Anal. Calcd for C31H31F12N7P2Ru: C, 41.71;
H, 3.50; N, 10.98. Found: C, 41.52; H, 3.39; N, 10.79. ESI-MS: calcd C31H31F12N7P2Ru for
[M-PF6]+ 748.1332, found 748.1342.
Synthesis of [RuL1(phen)(CH3CN)2](PF6)2, (Ru4)
The synthetic protocol of Ru4 was similar with Ru2, and 1,10-phenanthroline (phen)
(36 mg, 0.2 mmol) was used as the chelating ligand. Yield: 256 mg, 69%, orange solid. 1H
NMR (400 MHz, CD3CN) δ 9.64-9.66 (m, 1H), 8.88-8.90 (m, 1H), 8.45-8.48 (m, 1H), 8.238.29 (m, 3H), 8.20 (s, 1H), 8.13-8.16 (d, J = 6.0, 1H), 7.80-7.88 (m, 2H), 7.62-7.63 (d, J =
2.0, 1H), 7.53-7.56 (m, 1H), 7.37-7.48 (m, 5H), 7.25-7.27 (m, 1H), 6.86-6.90 (m, 1H), 5.715.92 (m, 2H), 2.32 (s, 3H), 1.80 (s, 3H). 13C NMR (100 MHz, CD3CN) δ 194.32, 157.38,
155.55, 152.49, 152.09, 148.98, 147.06, 140.35, 138.60, 137.60, 137.47, 131.50, 131.17,
129.56, 128.72, 128.37, 128.32, 127.53, 126.89, 125.84, 125.63, 122.64, 118.56, 112.27,
53.81, 4.22, 3.82. Anal. Calcd for C31H27F12N7P2Ru: C, 41.90; H, 3.06; N, 11.03. Found: C,
41.72; H, 3.17; N, 11.12. ESI-MS: calcd C31H27F12N7P2Ru for [M-PF6]+ 744.1019, found
744.1024.
Synthesis of [RuL2(CH3CN)4](PF6)2, (Ru5)
The synthetic protocol of Ru5 was similar with Ru1, and HL2PF6 (451 mg, 1 mmol)
was used as imidazolium salt. Yield: 482 mg, 56%, light yellow solid. 1H NMR (400 MHz,
CD3CN) δ 8.90-8.92 (m, 1H), 8.07-8.11 (m, 1H), 7.90-7.91 (d, J = 2.0, 1H), 7.78-7.80 (d, J =
16
This article is protected by copyright. All rights reserved.
Accepted Manuscript
(dmbpy) (37 mg, 0.2 mmol) was used as the chelating ligand. Yield: 105 mg, 59%, orange
10.1002/chem.202002970
Chemistry - A European Journal
4.0, 1H), 7.43-7.46 (m, 1H), 6.70-6.71 (d, J = 2.0, 1H), 5.65 (s, 2H), 2.50 (s, 3H), 2.31 (s,
6H), 2.30 (s, 3H), 2.28 (s, 6H), 2.15 (s, 6H), 1.96 (s, 3H). 13C NMR (100 MHz, CD3CN) δ
188.17, 155.46, 153.44, 141.03, 136.85, 134.37, 134.02, 128.87, 127.14, 125.32, 124.25,
122.76, 122.72, 112.09, 49.58, 16.94, 16.65, 16.41, 4.14, 3.62. Anal. Calcd for
C28H35F12N7P2Ru: C, 39.08; H, 4.10; N, 11.39. Found: C, 39.03; H, 4.14; N, 11.41. ESI-MS:
Synthesis of [RuL2(bpy)(CH3CN)2](PF6)2, (Ru6)
The synthetic protocol of Ru3 was similar with Ru2. A mixture of Ru5 (172 mg, 0.2
mmol) and the chelating ligand, 2,2’-bipyridine (byp) (32 mg, 0.2 mmol) in CH3CN was
refluxed and stirred for 24 h. Then, the solution was evaporated under vacuum to obtain the
crude product, and further purified by flash column chromatography. Yield: 123 mg, 66%,
light orange solid. 1H NMR (400 MHz, DMSO-d6) δ 9.46-9.47 (m, 1H), 8.89-8.91 (d, J = 4.0,
1H), 8.69-8.71 (d, J = 6.0, 1H), 8.57 (s, 1H), 8.46-8.50 (m, 1H), 8.18-8.20 (d, J = 4.0, 1H),
8.00-8.09 (m, 4H), 7.39-7.43 (m, 2H), 7.16-7.19 (m, 1H), 7.07-7.08 (d, J = 2.0, 1H), 5.705.80 (m, 2H), 2.53 (s, 3H), 2.39 (s, 6H), 2.37 (s, 3H), 2.28 (s, 3H), 2.27 (s, 6H). 13C NMR
(100 MHz, DMSO-d6) δ 191.74, 157.79, 156.68, 155.71, 154.95, 151.54, 151.18, 140.58,
139.81, 138.43, 136.01, 134.03, 133.29, 128.95, 128.59, 128.34, 127.78, 127.08, 124.58,
124.46, 123.43, 123.09, 118.61, 112.43, 49.07, 17.39, 17.18, 16.91, 4.26, 3.80. Anal. Calcd
for C34H37F12N7P2Ru: C, 43.69; H, 3.99; N, 10.49. Found: C, 43.61; H, 4.07; N, 10.43. ESIMS: calcd C34H37F12N7P2Ru for [M-PF6]+ 790.1801, found 790.1799.
Synthesis of [RuL2(dmbpy)(CH3CN)2](PF6)2, (Ru7)
The synthetic protocol of Ru7 was similar with Ru6. 5,5’-dimethyl-2,2’-bipyridine
(dmbpy) (37 mg, 0.2 mmol) was used as the chelating ligand. Yield: 110 mg, 57%, orange
solid. 1H NMR (400 MHz, DMSO-d6) δ 9.18-9.19 (d, J = 2.0, 1H), 8.73-8.75 (d, J = 4.0, 1H),
8.54-8.57 (m, 2H), 8.28-8.30 (m, 1H), 8.19-8.20 (d, J = 2.0, 1H), 8.00-8.04 (m, 1H), 7.867.88 (m, 1H), 7.66 (s, 1H), 7.44-7.46 (m, 1H), 7.16-7.20 (m, 1H), 7.02-7.03 (d, J = 2.0, 1H),
5.69-5.78 (m, 2H), 2.68 (s, 3H), 2.54 (s, 3H), 2.38 (s, 6H), 2.37 (s, 3H), 2.29 (s, 3H), 2.28
(s, 6H), 2.22 (s, 3H). 13C NMR (100 MHz, DMSO-d6) δ 189.92, 153.70, 153.34, 152.91,
151.16, 149.12, 148.83, 138.33, 138.07, 136.95, 136.60, 135.42, 133.99, 131.90, 131.23,
126.56, 126.08, 124.91, 121.55, 121.36, 121.01, 120.92, 116.59, 110.41, 47.03, 16.69,
16.34, 15.29, 15.08, 14.76, 2.25, 1.75. Anal. Calcd for C36H41F12N7P2Ru: C, 44.91; H, 4.29;
17
This article is protected by copyright. All rights reserved.
Accepted Manuscript
calcd C28H35F12N7P2Ru for [M-CH3CN-PF6]+ 675.1379, found 675.1362.
10.1002/chem.202002970
Chemistry - A European Journal
N, 10.18. Found: C, 44.69; H, 4.33; N, 10.07. ESI-MS: calcd C36H41F12N7P2Ru for [M-PF6]+
818.2114, found 818.2082.
Synthesis of [RuL2(phen)(CH3CN)2](PF6)2, (Ru8)
The synthetic protocol of Ru3 was similar with Ru6. 1,10-phenanthroline (phen) (36
(400 MHz, DMSO-d6) δ 9.86-9.87 (m, 1H), 9.11-9.13 (m, 1H), 8.67-8.69 (d, J = 4.0, 1H),
8.62 (s, 1H), 8.42-8.47 (m, 3H), 8.30-8.33 (d, J = 6.0, 1H), 8.18-8.20 (d, J = 4.0, 1H), 7.947.98 (m, 1H), 7.75-7.79 (m, 1H), 7.29-7.30 (d, J = 2.0, 1H), 7.14-7.15 (d, J = 2.0, 1H), 6.997.02 (t, J = 6.0, 1H), 5.77-5.89 (m, 2H), 2.60 (s, 3H), 2.44 (s, 6H), 2.29 (s, 9H), 2.25 (s, 3H).
13
C NMR (100 MHz, DMSO-d6) δ 191.55, 157.46, 154.97, 152.51, 151.44, 149.47, 148.14,
146.31, 141.17, 140.47, 138.79, 137.51, 136.01, 134.09, 133.32, 130.93, 130.61, 129.08,
128.42, 128.28, 128.08, 127.24, 126.66, 126.42, 123.53, 122.97, 118.69, 112.35, 49.12,
17.41, 17.21, 16.96, 4.30, 3.78. Anal. Calcd for C36H37F12N7P2Ru: C, 45.10; H, 3.89; N,
10.23. Found: C, 45.44; H, 3.93; N, 10.32. ESI-MS: calcd C36H37F12N7P2Ru for [M-PF6]+
814.1801, found 814.1811.
Synthesis of [RuL3(phen)(CH3CN)2](PF6)2, (Ru9)
Ru9 was synthesized by using HL3PF6 as imidazolium salt. A mixture of HL3PF6 (235
mg, 0.5 mmol) and Ag2O (65 mg, 0.3 mmol) in acetonitrile was stirred at 50 °C for 24 h.
After cooling to room temperature, [Ru(p-cymene)Cl2]2 (153 mg, 0.25 mmol) was added and
stirred for 2 h. After that, NH4PF6 (408 mg, 2.5 mmol) was added and the mixture was
heated to reflux for 24 h. The resulting mixture was directly added by the chelating ligand
1,10-phenanthroline (phen) (90 mg, 0.5 mmol) without purification. The solution was further
refluxed and stirred for another 24 h. After cooling to room temperature, the solution was
evaporated under vacuum to obtain the crude product of Ru9, which was further purified by
flash column chromatography. Yield: 298 mg, 61%, orange solid. 1H NMR (400 MHz,
DMSO-d6) δ 9.82-9.84 (d, J = 4.0, 1H), 9.10-9.12 (d, J = 4.0, 1H), 8.76 (s, 1H), 8.66-8.68 (d,
J = 4.0, 1H), 8.42-8.47 (m, 2H), 8.29-8.34 (m, 2H), 8.17-8.19 (d, J = 4.0, 1H), 7.96-8.00 (t, J
= 8.0, 1H), 7.81 (s, 1H), 7.69-7.73 (m, 1H), 7.27-7.29 (d, J = 4.0, 1H), 7.00-7.03 (t, J = 6.0,
1H), 5.96 (s, 2H), 2.54 (s, 3H), 2.23 (s, 3H). 13C NMR (100 MHz, DMSO-d6) δ 193.30,
157.34, 154.83, 152.53, 151.51, 148.07, 146.22, 140.52, 138.85, 137.60, 130.94, 130.60,
128.40, 128.27, 127.88, 127.25, 126.18, 124.45, 123.13, 119.50, 112.46, 41.65, 4.23, 3.87.
18
This article is protected by copyright. All rights reserved.
Accepted Manuscript
mg, 0.2 mmol) was used as the chelating ligand. Yield: 125 mg, 65%, orange solid. 1H NMR
10.1002/chem.202002970
Chemistry - A European Journal
Anal. Calcd for C31H22F17N7P2Ru: C, 38.05; H, 2.27; N, 10.02. Found: C, 38.25; H, 2.22; N,
9.98. ESI-MS: calcd C31H22F17N7P2Ru for [M-PF6]+ 834.0548, found 834.0525.
Synthesis of [RuL4(phen)(CH3CN)2](PF6)2, (Ru10)
The synthetic protocol of Ru7 was similar with Ru9, and HL4PF6 (160 mg, 0.5 mmol)
DMSO-d6) δ 9.80-9.81 (m, 1H), 9.08-9.10 (m, 1H), 8.67-8.68 (d, J = 2.0, 1H), 8.63-8.65 (m,
1H), 8.43-8.45 (m, 2H), 8.27-8.29 (d, J = 4.0, 1H), 8.16-8.19 (m, 2H), 8.00-8.01 (d, J = 2.0,
1H), 7.94-7.99 (m, 1H), 7.69-7.73 (m, 1H), 7.26-7.27 (d, J = 2.0, 1H), 6.97-7.00 (m, 1H),
4.53-4.58 (m, 2H), 2.55 (s, 3H), 2.29 (s, 3H), 1.57-1.61 (t, J = 8.0, 3H). 13C NMR (100 MHz,
DMSO-d6) δ 189.38, 154.70, 152.84, 150.40, 149.40, 146.08, 144.14, 138.31, 136.60,
135.36, 128.79, 128.50, 126.32, 126.11, 125.95, 125.08, 124.54, 124.23, 123.35, 120.78,
116.16, 110.11, 43.41, 14.92, 2.20, 1.57. Anal. Calcd for C26H25F12N7P2Ru: C, 37.78; H,
3.05; N, 11.86. Found: C, 37.70; H, 3.11; N, 11.84. ESI-MS: calcd C26H25F12N7P2Ru for [MPF6]+ 682.0862, found 682.0849.
X-ray Crystallography
Single crystal diffraction analysis was performed for Ru1-4 and Ru8-10 using a
Siemens Smart-CCD area-detector diffractometer with Mo Kα radiation (λ = 0.710 73 Å) in
ω scan mode. Absorption was corrected by multiscan. Oxford Diffraction CrysAlisPro
software was employed to collect data. Structures were clearly solved by direct methods
and refined by full-matrix least-squares on F2 using the SHELXTXL package. In addition,
nonhydrogen atoms were defined by full-matrix least-squares on F2 with anisotropic
temperature factors. Hydrogen atoms were regularly positioned based on various distances
of C-H as follows: 0.95 Å for aromatic CH; 0.99 Å for CH2; and 0.98 Å for CH3 on a riding
model with Uiso(H) = −1.2–1.5Ueq(C). All data were calculated and shown in Table S1 and
Table S2.
Supporting Information
The supporting information contains experimental sections, 1H and 13C NMR spectra
for the complexes Ru1-10, ORTEP diagrams for the complexes, and in vitro and in vivo
characterizations for the complexes. The tables show crystallographic data, lipophilicity and
cytotoxicity values for the complexes.
19
This article is protected by copyright. All rights reserved.
Accepted Manuscript
was used as imidazolium salt. Yield: 289 mg, 70%, orange solid. 1H NMR (400 MHz,
10.1002/chem.202002970
Chemistry - A European Journal
References
[1] T. C. Johnstone, K. Suntharalingam and S. J. Lippard, Chem. Rev. 2016, 116, 34363486.
[3] M. Ohmichi, J. Hayakawa, K. Tasaka, H. Kurachi and Y. Murata, Trends Pharmacol. Sci.
2005, 26, 113-116.
[4] A. Bergamo and G. Sava, Chem. Soc. Rev. 2015, 44, 8818-8835.
[5] a) R. G. Kenny and C. J. Marmion, Chem. Rev. 2019, 119, 1058-1137; b) W. Ma, L.
Guo, Z. Tian, S. Zhang, X. He, J. Li, Y. Yang and Z. Liu, Dalton Trans. 2019, 48, 47884793.
[6] S. Thota, D. A. Rodrigues, D. C. Crans and E. J. Barreiro, J. Med. Chem. 2018, 61,
5805-5821.
[7] C. G. Hartinger, M. A. Jakupec, S. Zorbas-Seifried, M. Groessl, A. Egger, W. Berger, H.
Zorbas, P. J. Dyson and B. K. Keppler, Chem. Biodiversity 2008, 5, 2140-2155.
[8] J. Li, Z. Tian, X. Ge, Z. Xu, Y. Feng and Z. Liu, Eur. J. Med. Chem. 2019, 163, 830-839.
[9] L. Zeng, P. Gupta, Y. Chen, E. Wang, L. Ji, H. Chao and Z. S. Chen, Chem. Soc. Rev.
2017, 46, 5771-5804.
[10] Q. Du, L. Guo, M. Tian, X. Ge, Y. Yang, X. Jian, Z. Xu, Z. Tian and Z. Liu,
Organometallics 2018, 37, 2880-2889.
[11] a) G. Golbaghi and A. Castonguay, Molecules 2020, 25, 265; b) S. Monro, K. L. Colon,
H. Yin, J. Roque, 3rd, P. Konda, S. Gujar, R. P. Thummel, L. Lilge, C. G. Cameron and S.
A. McFarland, Chem. Rev. 2019, 119, 797-828.
[12] W. L. Kwong, K. Y. Lam, C. N. Lok, Y. T. Lai, P. Y. Lee and C. M. Che, Angew. Chem.,
Int. Ed. 2016, 55, 13524-13528.
[13] A. E. Graminha, J. Honorato, L. L. Dulcey, L. R. Godoy, M. F. Barbosa, M. R.
Cominetti, A. C. Menezes and A. A. Batista, J. Inorg. Biochem. 2020, 206, 111021.
[14] M. K. M. Subarkhan, L. L. Ren, B. B. Xie, C. Chen, Y. C. Wang and H. X. Wang, Eur. J.
Med. Chem. 2019, 179, 246-256.
[15] F. Cisnetti and A. Gautier, Angew. Chem., Int. Ed. 2013, 52, 11976-11978.
20
This article is protected by copyright. All rights reserved.
Accepted Manuscript
[2] a) M. R. Trendowski, O. El Charif, P. C. Dinh, L. B. Travis and M. E. Dolan, Clin. Cancer
Res. 2019, 25, 1147-1155; b) Y. Yang, L. Guo, Z. Tian, X. Ge, Y. Gong, H. Zheng, S. Shi
and Z. Liu, Organometallics 2019, 38, 1761-1769.
10.1002/chem.202002970
Chemistry - A European Journal
[16] F. E. Poynton, S. A. Bright, S. Blasco, D. C. Williams, J. M. Kelly and T. Gunnlaugsson,
Chem. Soc. Rev. 2017, 46, 7706-7756.
[18] V. Vidimar, C. Licona, R. Ceron-Camacho, E. Guerin, P. Coliat, A. Venkatasamy, M.
Ali, D. Guenot, R. Le Lagadec, A. C. Jung, J. N. Freund, M. Pfeffer, G. Mellitzer, G. Sava
and C. Gaiddon, Cancer Lett. 2019, 440-441, 145-155.
[19] a) B. Peña, A. David, C. Pavani, M. S. Baptista, J.-P. Pellois, C. Turro and K. R.
Dunbar, Organometallics 2014, 33, 1100-1103; b) J. A. Solis-Ruiz, A. Barthe, G. Riegel, R.
O. Saavedra-Diaz, C. Gaiddon and R. Le Lagadec, J. Inorg. Biochem. 2020, 208, 111080.
[20] C. Yang, F. Mehmood, T. L. Lam, S. L. Chan, Y. Wu, C. S. Yeung, X. Guan, K. Li, C. Y.
Chung, C. Y. Zhou, T. Zou and C. M. Che, Chem. Sci. 2016, 7, 3123-3136.
[21] C. Chen, C. Lu, Q. Zheng, S. Ni, M. Zhang and W. Chen, Beilstein J. Org. Chem. 2015,
11, 1786-1795.
[22] a) S. J. Lippard, P. J. Bond, K. C. Wu and W. R. Bauer, Science 1976, 194, 726-728;
b) S. Arnott, P. J. Bond and R. Chandrasekaran, Nature 1980, 287, 561-563.
[23] M. R. Gill and J. A. Thomas, Chem. Soc. Rev. 2012, 41, 3179-3192.
[24] X. Wang, X. Wang and Z. Guo, Acc. Chem. Res. 2015, 48, 2622-2631.
[25] H. X. Wang, L. Q. Zhou, K. Xie, J. P. Wu, P. H. Song, H. Y. Xie, L. Zhou, J. L. Liu, X.
Xu, Y. Q. Shen and S. S. Zheng, Theranostics 2018, 8, 3949-3963.
[26] K. Xie, S. S. Song, L. Q. Zhou, J. Q. Wan, Y. T. Qiao, M. Wang, H. Y. Xie, L. Zhou, S.
S. Zheng and H. X. Wang, Int. J. Pharm. 2019, 556, 159-171.
[27] B. Xie, J. Wan, X. Chen, W. Han and H. Wang, Mol. Cancer Ther. 2020, 19, 822-834.
[28] H. Hamidi and J. Ivaska, Nat. Rev. Cancer 2018, 18, 532-547.
[29] X. Ge, S. Chen, X. Liu, Q. Wang, L. Gao, C. Zhao, L. Zhang, M. Shao, X. A. Yuan, L.
Tian and Z. Liu, Inorg. Chem. 2019, 58, 14175-14184.
[30] D. Cirri, M. G. Fabbrini, A. Pratesi, L. Ciofi, L. Massai, T. Marzo and L. Messori,
Biometals 2019, 32, 813-817.
[31] B. T. Elie, J. Fernandez-Gallardo, N. Curado, M. A. Cornejo, J. W. Ramos and M.
Contel, Eur. J. Med. Chem. 2019, 161, 310-322.
[32] F. U. Rahman, M. Z. Bhatti, A. Ali, H. Q. Duong, Y. Zhang, X. J. Ji, Y. J. Lin, H. Wang,
Z. T. Li and D. W. Zhang, Eur. J. Med. Chem. 2018, 157, 1480-1490.
21
This article is protected by copyright. All rights reserved.
Accepted Manuscript
[17] H. Y. Huang, P. Y. Zhang, B. L. Yu, Y. Chen, J. Q. Wang, L. N. Ji and H. Chao, J. Med.
Chem. 2014, 57, 8971-8983.
10.1002/chem.202002970
Chemistry - A European Journal
[33] a) M. Egeblad and Z. Werb, Nat Rev Cancer 2002, 2, 161-174; b) S. Kalyankrishna
and J. R. Grandis, J. Clin. Oncol. 2006, 24, 2666-2672.
[34] R. S. Kerbel, Science 2006, 312, 1171-1175.
[35] D. D. Sun, Y. N. Liu, Q. Q. Yu, Y. H. Zhou, R. Zhang, X. J. Chen, A. Hong and J. Liu,
Biomaterials 2013, 34, 171-180.
[37] L. Shu, L. Ren, Y. Wang, T. Fang, Z. Ye, W. Han, C. Chen and H. Wang, Chem.
Commun. 2020, 56, 3069-3072.
[38] L. B. Travis, S. D. Fossa, H. D. Sesso, R. D. Frisina, D. N. Herrmann, C. J. Beard, D.
R. Feldman, L. C. Pagliaro, R. C. Miller, D. J. Vaughn, L. H. Einhorn, N. J. Cox, M. E. Dolan
and G. Platinum Study, J. Natl. Cancer Inst. 2014, 106, 44-54.
[39] K. Pantel and C. Alix-Panabieres, Trends. Mol. Med. 2010, 16, 398-406.
[40] T. Lazarevic, A. Rilak and Z. D. Bugarcic, Eur. J. Med. Chem. 2017, 142, 8-31.
[41] C. G. Hartinger, S. Zorbas-Seifried, M. A. Jakupec, B. Kynast, H. Zorbas and B. K.
Keppler, J. Inorg. Biochem. 2006, 100, 891-904.
[42] S. Yuan, S. Chen, H. Wu, H. Jiang, S. Zheng, Q. Zhang and Y. Liu, Chem. Commun.
2020, 56, 1397-1400.
[43] L. I. Rylands, A. Welsh, K. Maepa, T. Stringer, D. Taylor, K. Chibale and G. S. Smith,
Eur. J. Med. Chem. 2019, 161, 11-21.
[44] a) D. Hu, C. Yang, C. N. Lok, F. R. Xing, P. Y. Lee, Y. M. E. Fung, H. B. Jiang and C.
M. Che, Angew. Chem., Int. Ed. 2019, 58, 10914-10918; b) M. Mora, M. C. Gimeno and R.
Visbal, Chem. Soc. Rev. 2019, 48, 447-462.
[45] S. Fruhauf and W. J. Zeller, Cancer Res. 1991, 51, 2943-2948.
[46] L. Shi, Y. Wang, Q. Wang, Z. Jiang, L. Ren, Y. Yan, Z. Liu, J. Wan, L. Huang, B. Cen,
W. Han and H. Wang, J. Controlled Release 2020, 324, 289-302.
[47] G. Sava, S. Zorzet, C. Turrin, F. Vita, M. Soranzo, G. Zabucchi, M. Cocchietto, A.
Bergamo, S. DiGiovine, G. Pezzoni, L. Sartor and S. Garbisa, Clin. Cancer Res. 2003, 9,
1898-1905.
22
This article is protected by copyright. All rights reserved.
Accepted Manuscript
[36] S. Lheureux, M. Braunstein and A. M. Oza, Ca-Cancer J. Clin. 2019, 69, 280-304.
10.1002/chem.202002970
Chemistry - A European Journal
Table 1. In vitro cytotoxicity of ruthenium(II) complexes in comparison with cisplatin. Cells
were treated for 72 h, and cell viability was determined by the MTT assay (expressed as
IC50 ± SD in μM). The ruthenium complexes were dissolved in DMSO and the DMSO
contents in culture media were less than 0.2% (v/v) to avoid the solvent impact to cell
Compounds
Ru1
Ru2
Ru3
Ru4
Ru5
Ru6
Ru7
Ru8
Ru9
Ru10
Cisplatinb
NAMI-A
IC50 (μM)a
A549
>100
94.7±7.7
80.2±4.3
35.0±2.4
28.5±2.2
14.3±1.2
10.3±0.6
4.7±0.4
12.5±1.0
11.5±0.9
3.2±0.2
>100
A549/cisR
>100
>100
90.7±3.4
40.4±2.1
36.4±1.9
30.9±1.7
19.9±1.5
10.5±0.5
24.9±2.2
23.3±2.2
12.5±0.7
>100
A2780
>100
>100
62.2±5.1
32.7±1.9
15.5±1.2
11.8±0.7
9.5±0.9
5.1±0.5
22.2±1.6
17.5±1.9
2.1±0.06
>100
Huh-7
>100
>100
84.5±9.1
97.9±9.3
81.6±17.6
33.7±4.3
33.1±9.7
15.0±2.5
18.2±1.3
15.7±1.3
3.7±0.3
>100
B16-F10
>100
29.9±4.6
19.15±5.6
10.9±1.4
9.0±1.2
7.3±0.7
5.7±0.8
3.3±0.2
6.9±1.9
4.7±2.0
1.3±0.3
>100
HUVEC
>100
>100
>100
>100
55.7±12.3
38.9±4.9
25.3±3.0
16.2±1.3
28.4±3.3
20.2±2.5
8.2±1.0
>100
a
IC50 = compound concentration required to inhibit tumor cell proliferation by 50%.
b
A clinical aqueous solution of cisplatin was used as control.
23
This article is protected by copyright. All rights reserved.
Accepted Manuscript
viability.
Figure 1. (A) Structures of biologically active ruthenium(II) complexes (NAMI-A, KP1019
and [Ru(bpy)(phpy)(dppz)]+. bpy = 2,2’-bipyridine; phpy = 2-phenylpyridine; dppz =
dipyrido[3,2-a:2’,3’-c]phenazine). (B) Chemical structures and synthesis of new
ruthenium(II) compounds (Ru1-10). Reagents and conditions: (i) Ag2O, CH3CN, 50 °C, 24
h; (ii) [Ru(p-cymene)Cl2]2, NH4PF6, CH3CN, reflux, 24 h; (iii) N-N ligand (bpy, dmbpy or
phen), CH3CN, reflux, 24 h.
24
This article is protected by copyright. All rights reserved.
Accepted Manuscript
10.1002/chem.202002970
Chemistry - A European Journal
Figure 2. X-ray crystal structure of the complex Ru8. Thermal ellipsoid is shown at the 50%
probability level.
25
This article is protected by copyright. All rights reserved.
Accepted Manuscript
10.1002/chem.202002970
Chemistry - A European Journal
Figure 3. (A) A Click-iT® EdU assay for quantifying the proliferation of A2780 cells. All cells
were treated with ruthenium(II) complexes Ru6 and Ru8 for 48 h at various concentrations.
(B) AO/EB assay quantified the apoptosis of A2780 cells. All cells were treated with Ru6
and Ru8 for 48 h at various concentrations. (C-D) Quantification of cell proliferation ratio
and apoptosis ratio. Data are presented as the means ± SD for n > 3 regions; ***p < 0.001.
(E) Western blot analysis of apoptosis- and cell cycle-associated proteins. A2780 cells were
treated with the complexes for 48 h.
26
This article is protected by copyright. All rights reserved.
Accepted Manuscript
10.1002/chem.202002970
Chemistry - A European Journal
Figure 4. Complex Ru8 shows antimetastatic activity on cancer cells. (A-B) Wound-healing
assay performed on Huh-7 cells. Cells were treated with different concentrations of Ru8 for
24 h. (C-D) Inhibition of transwell invasion in Huh-7 cells after treatment with different
concentrations of Ru8 for 24 h. Data are presented as the means ± SD for n > 3 regions; *p
< 0.05, ***p < 0.001. (E) Western blot analysis of metastasis-associated proteins. The cells
were treated with NAMI-A and Ru8 for 24 h.
27
This article is protected by copyright. All rights reserved.
Accepted Manuscript
10.1002/chem.202002970
Chemistry - A European Journal
Figure 5. Ru8 shows antiangiogenic activity in cell-based assays. (A-B) Wound-healing
assay performed on HUVECs. Cells were treated with different concentrations of Ru8 for 24
h. (C-D) Inhibition of tube formation in HUVECs after treatment with different concentrations
of Ru8 for 2 h. Data are presented as the means ± SD for n > 3 regions; **p < 0.01, ***p <
0.001.
28
This article is protected by copyright. All rights reserved.
Accepted Manuscript
10.1002/chem.202002970
Chemistry - A European Journal
Figure 6. (A) Representative images of the developmental CAM treated with ruthenium(II)
Ru8. Untreated CAM exhibited small and high dense vessels and capillary network.
Treatment with Ru8 hampered angiogenesis and vessel development in the CAM model.
(B) Quantitative analysis of the number of branching points. Data are presented as the
means ± SD for n > 3 regions; ***p < 0.001.
29
This article is protected by copyright. All rights reserved.
Accepted Manuscript
10.1002/chem.202002970
Chemistry - A European Journal
Figure 7. In vivo therapeutic efficacy of Ru8 against metastatic A2780 tumor model. (A)
Photographs of tumor-bearing nude mice after different treatments. Healthy mice without
tumor cell implantation were used as control. (B) Total abdominal tumor weight excised
from the mice after different treatments. Representative images of tumors and ovaries (C),
as well as ovary weight (D), in each group at the end of the study. (E) Body weights of
mice. Compounds were injected into mice at a dose of 2 μmol/kg every day for one week.
Data are presented as the means ± SD; *p < 0.05, **p < 0.01, ***p < 0.001.
30
This article is protected by copyright. All rights reserved.
Accepted Manuscript
10.1002/chem.202002970
Chemistry - A European Journal