← Back
Development of novel ruthenium(II)-arene complexes displaying potent anticancer effects in glioblastoma cells.
View Article Online
Dalton
View Journal
Transactions
An international journal of inorganic chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: P. Kumar, I.
Mondal, R. Kulshreshtha and A. K. Patra, Dalton Trans., 2020, DOI: 10.1039/D0DT02167A.
This is an Accepted Manuscript, which has been through the
Volume 46
Number 10
Dalton 1 P 4 a g M e a s r c 3 h 0 7 2 3 0 - 1 3 7 412 Royal Society of Chemistry peer review process and has been
accepted for publication.
Transactions
Accepted Manuscripts are published online shortly after acceptance,
An international journal of inorganic chemistry
rsc.li/dalton before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
ISSN 0306-0012
shall the Royal Society of Chemistry be held responsible for any errors
CDNan o- O dHuM eg tetlM ea llrs u U o r Wc N eyn .I cy CSllit A ce c Tp act hI ai O o erbn n N e s en ( teX a sH l.t +a, bXil i=z eSd, Spea,r Teen)t sulfenyl, selenenyl, or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
rsc.li/dalton
Page 1 of 37 Dalton Transactions
Development of Novel Ruthenium(II)-Arene Complexes Displaying Potent
Anticancer Effects in Glioblastoma
Priyaranjan Kumar,a Indranil Mondal,b Ritu Kulshreshtha,b and Ashis K. Patra*,a
t
p
i
r
c
s
u
n
a
M
d
e
t
p
e
c
c
A
s
n
o
i
t
c
a
s
n
a
r
T
n
o
t
l
a
D
Author address: aDepartment of Chemistry, Indian Institute of Technology Kanpur, Kanpur
208016, Uttar Pradesh, India. E-mail: akpatra@iitk.ac.in
bDepartment of Biochemical Engineering and Biotechnology, Indian Institute of Technology
Delhi, Hauz Khas, New Delhi 110 016, India
1
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 2 of 37
Abstract
Glioblastoma (GB) is a highly aggressive and malignant brain tumor that is highly resistant to
conventional multimodal treatments leading to its abysmal prognosis. We designed two
organometallic half-sandwich Ru(II)-6-p-cymene complexes containing Schiff-base derived
from 3-aminoquinoline and 2-hydroxy-benzaldehyde (L1) and 2-hydroxy-naphthaldehyde
(L2), namely [Ru(6-p-cymene)(L1)Cl] (1) and [Ru(6-p-cymene)(L2)Cl] (2) and studied their t
p
activity on GB cells. Both the complexes were structurally characterized using single-crystal i
r
c X-ray diffraction showing half-sandwich three-legged piano-stool geometry. We studied the
s
physicochemical behavior, solution speciation, aquation kinetics, and photosubstitution u
n
reactions using various spectroscopic methods. The complexes exhibit moderate binding
a
affinity with calf-thymus (CT)-DNA (K b ~105 M-1). The complexes effectively interact with M
human serum albumin (HSA) (K ~105 M-1) with preferential tryptophan binding as determined
d
from synchronous fluorescence studies. The in vitro studies showed significant e
t
antiproliferative activity against an aggressive human GB cell line LN-229 (IC 50 = 22.8 μM) p
e
with a moderate selectivity relative to the normal mouse fibroblast L929 cells. Notably, [Ru(6-
c
p-cymene)(L1)Cl] (1) have higher selectivity index (S.I.) than [Ru(6-p-cymene)(L2)Cl] (2) or c
A
cisplatin. We evaluated the clonogenic potential of the GB cells using the colony formation
s
assay in the presence of the complex 1. Excitingly, it shows ~75% inhibition of clonogenic
n
potential at IC concentration. Complex 1 also effectively lowered the migratory potential of 50 o
GB cells as accessed from wound healing assay. The studied compound lead to apoptosis of i
t
c
GB cells as evidenced by nuclear condensation, blebbing, and enhanced caspase 3/7 activity
a
s
with anticipated utility in GB treatment using photochemotherapy.
n
a
r
Introduction
T
Glioblastoma (GB) is a highly aggressive, malignant, and deadliest form of brain tumor that
n
shows dismal prognosis, with a very high mortality rate [1]. Less than 5% of the patients o
t
survive more than 5 years upon a treatment, and average survival is only 3 months without any l
a
treatment after diagnosis. GB is probably the most challenging and unique tumor to treat D
clinically due to its very invasive and somewhat illusive multifocal nature of growth throughout
the white matter in the brain [2]. Because of its stealthy and invasive growth through central
nervous system (CNS), the surgery remains quite challenging, very often inoperable, and
associated with severe cognitive dysfunctions. Over the last 3-4 decades, therapeutic
intervention failed to have a noteworthy impact on GB survival. Gamma knife radiosurgery is
2
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 3 of 37 Dalton Transactions
one of the technique neurosurgeons uses to remove as well as block the anticipated proliferation
of cancerous tissue in white matter tracks [3]. The surgical removal of the tumor is augmented
by radiation and cytotoxic chemotherapy. To date, only a few drugs are approved by the US
FDA, such as temozolomide (TMZ), bevacizumab, carmustine, and lomustine (Fig. 1).
Treatment with temozolomide (TemodarTM), a promiscuous DNA methylating agent, notably
demonstrated notable efficacy in combination with radiotherapy. The acquired resistance
t
p
towards radiation and chemotherapy of GB makes it even more challenging to treat [4]. Despite
i
r
significant progress in GB genomics, clinical medicine and diagnosis, the combined treatment c
s
regimen from surgery and subsequent concomitant radiotherapy and TMZ chemotherapy, its
u
prognosis is still inferior with a median survival of terrifying 14.6 months [5]. Clinically n
a
approved DNA damaging drugs (e.g., doxorubicin, irinotecan, etoposide, carmustine and
M
TMZ) remained disappointing for GB targeted therapy [1]. Therefore, combating the
d
aggressive GB tumor chemotherapeutically at the molecular level requires a new drug
e
discovery approach and seems an uphill task for medicinal chemists and clinicians. The t
p
molecularly targeted drugs under clinical trial are mostly based on oncogenic receptor tyrosine e
c
kinase (RTK) pathways. The PI3K-AKT and RAS oncogenic pathways, along with various
c
growth factors, and their receptors are often targeted with organic inhibitors [1, 5]. A
s
n
o
i
t
c
a
s
n
a
r
T
n
o
t
l
a
D
Fig. 1 Top: Schematic representation of glioblastoma (GB) in cerebral hemisphere and FDA approved
molecular drugs for glioblastoma after surgical resection. Temozolomide is the current standard of
treatment; bevacizumab is a monoclonal antibody (mAb) targeted VEGF-inhibitor. Bottom: Structure
of the most successful antimetastatic (NAMI-A) and anticancer ruthenium compounds.
The successful chemotherapy from platins (cisplatin, carboplatin, oxaliplatin) to numerous
cancer types envisioned that metallodrugs serendipitously could serve startling efficiency than
3
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 4 of 37
organic anticancer drugs [6]. However, the dose-dependent side effects and the acquired drug
resistance of platins limits its applicability and simultaneously opens up the opportunity for
discovery of new-generation targeted metallodrugs [7−9]. Most of the drugs show side effects
because of the poor selectivity index of its target, and the enhanced selectivity index (S.I.) is
desirable and directly increases the efficacy of a drug.
Ruthenium is a suitable choice of metal after platinum for metallodrug design towards targeted
t
p
chemotherapy due to physiologically accessible multiple oxidation states (+II/III), the
i
r
formation of rigid bonds with hard and soft donor ligands, biocatalytic potentials, transferrin c
s
binding as iron mimicking and tunable ligands substitution kinetics similar to Pt(II) drugs [10].
u
The photochemistry and favorable optical properties of Ru-drugs is fascinating and assists in n
a
photodynamic therapy (PDT) or photoactivated chemotherapy (PACT). This is apt for
M
advanced non-invasive spatiotemporally controlled and innovative targeted photo-activated
d
chemotherapy (PACT) modality [11, 12]. The TLD1433 is the first Ru(II)-polypyridyl based
e
PDT agent entered for human clinical trial for treating non-muscle invasive bladder cancer t
p
[12b]. The antimetastatic NAMI-A, anticancer NKP-1339 and organometallic Ru(II)-arene e
c
complexes namely RM175 [(η6‑biphenyl)Ru(en)Cl]+ (en = ethylenediamine), RAPTA series
c
of complexes [η6arene)Ru(PTA)Cl ]+ (PTA = 1,3,5‑triaza‑7‑phosphaadamentane) (Fig. 1) and A
2
RDC11 [Ru(phenanthroline)(κ‑C,N‑(2‑phenyl‑pyridine)(NCMe) ]PF have been reported for
2 6 s
excellent anticancer activity and significantly reduced side effects and also non-cross resistant n
o
with platin drugs [13−15]. Interestingly, the primary molecular targets of these Ru-complexes
i
t
are not limited to DNA, but also enzyme, genes, and proteins crucial in a postgenomic era, and c
a
not entirely driven by cytotoxicity. In the recent years, the screening of a wide spectrum of Ru-
s
arene Schiff-base complexes conjugated with pyridine/quinoline amines with substituted n
a
aromatic aldehydes has been reported for overcoming multi-drug resistance (MDR) with
r
T
apoptosis independent pathways [16, 17].
n
o
(a) (b) t
l
a
N N N
D
Ru Cl Ru Cl Ru Cl
N N N
O Aquation O O
Photoactive
GBInhibition
(1) (2)
4
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 5 of 37 Dalton Transactions
Fig. 2 (a) Design scheme for Ru(II)-arene complexes. (b) [Ru(6-p-cymene)(L1)Cl] (1) and [Ru(6-p-
cymene)(L2)Cl] (2) under study.
We designed two Ru(II)-6-p-cymene half-sandwich complexes using Schiff bases obtained
from the conjugation of quinoline amine and 2-hydroxy-benzaldehyde (L1) (1) and 2-hydroxy-
1-naphthaldehyde (L2) (2) (Fig. 2). To our curiosity, we have tested in vitro cytotoxicity of
t
p
these complexes with highly aggressive glioblastoma (GB) cells, LN229, considering the
i
r
absence of any systematic investigation of cytotoxic metal complexes for GB in literature. We
c
s
also envision the potential application of PACT or PDT in combination with chemotherapy
u
using effective Ru-photosensitizers for glioblastoma.
n
We report here synthesis, characterization and crystal structures, solvation kinetics, and a
M
binding studies with CT-DNA and HSA. The photoactivity in the low energy UV-A light (λ
exc
d
= 365 nm) was investigated for possible phototherapeutic applications. The in vitro MTT-based
e
cytotoxicity studies with glioblastoma (GB) cell line LN229 and normal fibroblast cell line
t
p
L929 showed both the complexes are significantly potent with a better selectivity index (S. I.)
e
for compound 1. The therapeutic potency of complex 1 further shows ~75% inhibition of c
c
clonogenic potential and lower the migratory potential as revealed from the wound healing
A
assay. The apoptotic cell death pathway is evident from the nuclear localization, condensation,
s
and blebbing and increased caspase 3/7 activity. n
o
Result and Discussion
i
t
Physiochemical characterizations. c
a
The Schiff base ligands were synthesized in high yields from 3-aminoquinoline with 2-hydroxy
s
benzaldehye (L1) or 2-hydroxy-1-naphthaldehye (L2) via refluxing in EtOH. [Ru(6-p- n
a
cymene)(L1)Cl] (1) and [Ru(6-p-cymene)(L2)Cl] (2) complexes were synthesized in high r
T
yields via stoichiometric reaction of respective deprotonated ligand with dimeric [Ru(6–p-
n
cym)(μ-Cl)Cl] in CHCl at 50 0C for 4 h. Complexes were non-hygroscopic crystalline solids
2 3 o
and soluble in CH Cl , CHCl , MeOH, CH CN, DMF and DMSO and insoluble in H O, Et O t
2 2 3 3 2 2 l
a
and pentane/hexane. Both the complexes were well characterized using ESI-MS, FTIR, NMR,
D
and UV-vis studies. The solid-state structures of both the complexes was determined using
single-crystal X-ray diffraction studies showing the typical three-legged piano-stool structure.
Selected spectral data are given in Table 1.
5
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 6 of 37
t
p
i
r
c
s
u
Fig. 3 (a) Electronic absorption spectra of complexes 1 and 2 in DMF (C = 39 M) at 298 K. (b) An
n
overlay of the excitation and emission spectra of the complexes 1 and 2 in DMF (C = 20 M) at 298 K. a
M
Excitation parameters: = 550 nm and exc.and em. slit width = 10 nm. Emission parameters: =
em ex
355 nm and exc.and em. slit width = 5 nm. d
e
t
p
The (+)-ve mode ESI-MS analysis of the complexes in CHCl showed molecular ion peaks
3
e
corresponding to [MCl-]+ at m/z for 483.10 (1) and 533.12 (2) arising from the dissociation of c
c
the chloride with characteristic isotopic patterns of Ru (Fig. S1, ESI†). The FT-IR spectra of
A
the complexes displayed a series of strong stretching peaks ranging 16061443 cm-1 (1) and
s
16161429 cm-1 (2) corresponding to and functionalities of the ligands [18]. The n
C=C C=N
o
absence of any broad band at ~3300 cm-1 typical for which confirms the deprotonated
O-H,
i
t
Schiff-base metal coordination (RuII-OPh) (Fig. S2, ESI†). [19]. Further, the structural integrity c
a of the complexes in solution was determined by 1H-NMR and 13C-NMR (Fig. S3−S6, ESI†).
s
and chemical shift values are assigned in experimental sections. The absence of 1H NMR n
a
signals corresponding to –OH group also confirms RuII-OPh coordination. The –CH (s, 3H)
3
r
of p-cymene resonated at 2.14 (1) and 2.17 (2) ppm. A septate signal corresponding to -CH- T
(CH ) of p-cymene was observed at 2.63 (1) and 2.64 ppm (2). The electronic absorption n
3 2
o
spectra showed three maxima (λ ) bands for 1 at 274 nm, 325 nm and 447 nm, and for
max t
l
complex 2 at 275 nm, 323 nm and 461 nm (Fig. 3 (a), Table 1). The highest energy band a
D
corresponds to π−π* of ligands (L1 and L2), while lowest energy visible region bands
corresponds MLCT transitions, whereas the bands ~ 325 nm for both the complexes are
probably due to n−π* transitions [20]. The complexes were fluorescent and showed identical
emission maximum at 417 nm when excited at 355 nm in DMF solution (Fig. 3b). The
excitation from lowest energy band (λ = 447 nm (1) and λ = 461 nm (2)) showed very weak
ex ex
emission at 560 nm (1) and 532 nm (2). The excitation spectra displayed bands at 275 nm, 288
6
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 7 of 37 Dalton Transactions
nm, and 357 nm (λ = 550 nm) in DMF and show features from absorption spectral pattern
em
(Fig. 3(b)).
Table 1. Physiochemical parameters of the complexes 1 and 2.
Complex IRa/cm-1, λ b/nm, (ε/M−1cm−1)
C=C, C=N max
t
1 1606, 1527, 1492, 1459, 1443 274 (16300), 325 (7800), 447 (2780) p
i
2 1616, 1603, 1581, 1535, 1453, 1429 275 (23750), 323 (11220), 461 (3810)
r
c
ain KBr phase. bUV-visible spectra in DMF.
s
u
n
Single crystal X-ray structure
a
Both the complexes Ru(6-p-cymene)(L1)Cl] (1) and [Ru(6-p-cymene)(L2)Cl] (2) and ligand M
2-[(E)-(3-quinolinylimino)methyl)naphthol (L2) were crystallized at ambient conditions and
d
structures were determined by single-crystal X-ray diffraction studies. The ORTEP view of 1, e
t
2, and L2 was shown in Fig. 4 and Fig. S7(a), respectively. The crystallographic parameters
p
are noted in Table S1, and selected bond parameters are given in Table S2 in ESI†. The unit e
c
cell of L2 and complexes 1 and 2 contain 8, 8 and 4 molecules respectively (Fig. S7(b) and Fig.
c
A
S8 in ESI†). Complex 1 and L2 crystallized in monoclinic C2/c and complex 2 in P2 /c space
1
groups. The free ligand L2 shows minor deviation from the coplanarity between quinoline and s
n
naphthol rings, and displayed ideal cisodial orientation of imine-N and phenolic-OH group for
o
coordination with RuII centre (Fig. S7, ESI†). The coordination environment around the Ru(II) i
t
c
centre found identical (distorted octahedral) with a tetrahedral like geometry from (6- p-
a
cymene) capping one vertex, the chelating N,O bidentate Schiff-bases and monodentate s
n
chloride, making an overall “piano stool” geometry typical for Ru(II)-arene complexes
a
[21,22]. The RuII-6-cymene (centroid) distances are ~1.67 Å (1, 2). The Ru-N bond distances r
T
are 2.097(4) Å (1) and 2.096(3) Å (2). The shortest and largest bonds are Ru-O (2.078(3) Å for
n
1 and 2.059(2) Å for 2) and Ru-Cl (2.4285(10) Å for 1 and 2.4243(9) Å for 2) matching with o
t
the reported RuII-arene complexes [23−25]. The shorter bond lengths in complex 2 than 1 for
l
a
the Ru-N1/O1/Cl1 bonds suggest more acceptor ability of L2 through extended conjugation D
in complex 2. The dihedral angle between the quinoline ring and phenyl/naphthyl groups are
20.06 (L2), 48.04 (1), and 42.99 (2) indicates significant twisting of ligand planarity upon
binding to RuII-(6-cym) moiety (Fig. S9 in ESI†). Both the complexes 1 and 2 show weak
intermolecular stacking interactions among planar aromatic rings at 3.575 Å and 3.504 Å,
respectively (Fig. S10 in ESI†).
7
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 8 of 37
t
p
i
r
c
s
u
n
a
M
Fig. 4 ORTEP views of 1 and 2 with 50% probability of thermal ellipsoid and labelling of metal and
d
heteroatoms. (a) Ru(6-p-cymene)(L1)Cl] (1) (CCDC 2006007): Ru1-N1, 2.097(4) Å; Ru1-O1, e
t
2.078(3) Å; Ru1-Cl1, 2.4285(10) Å. (b) [Ru(6-p-cymene)(L2)Cl] (2) (CCDC 2006008): Ru1-N1, p
e
2.096(3) Å; Ru1-O1, 2.059(2) Å; Ru1-Cl1, 2.4243(9) Å. The hydrogen atoms were not shown for
c
clarity.
c
A
Solvation of Complexes s
n
The hydrolysis or aquation is the key step of the activation of several metal-based
o
chemotherapeutics. A majority of the metallodrugs (e.g., platins) are truly based on tuneable i
t
c
and variable ligand exchange rates with H O [26, 27]. All the notable Ru-drugs (NAMI-A, 2
a
NKP1339, RM175, RAPTA-C) contain Ru-Cl bonds with anticipation of ligand exchange s
n
reaction from a potential donor from biological targets. The labile RuCl bonds undergoes
a
hydrolysis and forms RuOH bond, and the later can bind to DNA nucleobases (guanine N7 r
2
T
and adenine) by substitution of labile water. In general, the rate of hydrolysis of
n
monofunctional Ru(II)-arene complexes (containing labile ligands such as halides) correlated
o
with their cytotoxicity and pharmacokinetics profile. The inert Ru-complexes remain inactive t
l
a
while moderate hydrolysis exhibits pronounced targeted cytotoxicity. Further, the rate of
D
hydrolysis is tuneable and depends on the substituent present on the arene ring, and other
secondary ligands attached to Ru(II) [28−30].
The solvation or aquation of the complexes 1 and 2 was studied in 5 mM Tris-HCl/NaCl buffer
(pH = 7.2), as well in DMF from the time-dependent UV-visible and fluorescence spectral
measurements at 298 K due to their well-defined spectroscopic signatures. The UV-vis spectra
8
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 9 of 37 Dalton Transactions
of complex 1 and 2 dissolved in aqueous buffer showed concomitant small increase in
absorbance at 395 nm (1) and 379 nm (2) (Fig. 5(a) and Fig. S11(a)). The insignificant
absorption spectral changes indicate possible aquation reaction (Ru-ClRu-OH ) without any
2
major electronic or structural perturbation affecting the chromophoric ligands bound to Ru(II).
The rate constants (k ) for these hydrolysis reactions were determined using these bands (Fig.
H2O
5(b) and Fig. S11(b)). The rate constants of hydrolysis (k H2O ) and half-life time (t 1/2 ) of the t
p
complexes were obtained from the first-order exponential curve fitting by following equation
i
r
(1) (Table 2). The concomitant changes in absorption spectra in DMF and the rate constants of c
s
solvation in DMF (k ) and half-life (t ) were obtained by monitoring maximum change in
DMF 1/2 u
absorbance at 325 nm (1) and 324 nm (2) in DMF (Table 2, Fig. S12, ESI†) n
a
M
Table 2. Solvation kinetics parameters of the complexes 1 and 2 at 298 K.
d
e
Complex Aquation (H O) Solvation (DMF)
2
t
p
ak (s-1) bt (min) ck (s-1) dt (min)
H2O 1/2 DMF 1/2 e
c
1 5.58 10-4 20.7 4.13 10-4 28.0 c
A
2 8.47 10-5 136.4 3.67 10-4 31.5
s
n
aHydrolysis rate constant in 5 mM Tris-HCl/NaCl buffer (pH = 7.2), bHalf-life time of hydrolysis,
o
cSolvation rate constant in DMF, dHalf-life of solvation. i
t
c
The differences in solvation rate constants for the complex 1 than 2 can be accounted from a
s
better -acceptor ability of complex 2 from extended aromatic conjugations in L2, which
n
decreases the electron density and lability around Ru(II) 4d6 centre [31]. The rate of hydrolysis a
r
of complex 1 is faster than cisplatin, which exhibits hydrolysis rates of 7.56 10-5 and 6.32 T
10-5 s-1 for the first and second chloride ligands [32]. n
o
t
l
a
D
9
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 10 of 37
t
p
i
r
c
s
u
n
a
M
d
e
t
p
e
c
c
Fig. 5 (a) The changes in absorption spectral profile for the complex 1 (58 M) in 5 mM Tris-HCl/NaCl A
buffer (pH = 7.2)-DMF mixture (97:3). Inset: Evolution of UV-visible difference spectra upon aquation
s
of complex 1. (b) Absorption traces at 395 nm and monoexponential fit upon hydrolysis for complex 1. n
o
(c) The changes in fluorescence spectral profile for the complex 1 (10 M) in 5 mM Tris-HCl/NaCl
i
buffer (pH = 7.2)-DMF mixture (200:1) (d) Increase of emission intensity at 421 nm (1) and 424 nm t
c
(2) for the complexes upon hydrolysis in 5 mM Tris-HCl/NaCl buffer (pH 7.2). = 275 nm, Ex. and a exc
s
Em. slit width = 10 nm at 298 K.
n
a
The hydrolysis of the complexes further evaluated from fluorescence measurement in buffer at
r
T
298 K, which showed concomitant increase of emission intensity at 421 nm (1) and 424 nm (2)
n
(Fig. 5(c), (d) and Fig. S13, ESI†). The enhancement in emission intensity for complex 1 was
o
more pronounced than complex 2 in agreement with a faster rate of hydrolysis of complex 1 as
t
l
observed from absorption spectral measurements. The aquated products apparently exhibit a
D
higher emission intensity than the original complexes. This can be probably from the heavy
atom displacement (Cl) by H O; the former facilitates the triplet-state population and thereby
2
decrease in fluorescence intensity [33].
Photoreactivity
10
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 11 of 37 Dalton Transactions
The Ru(II)-polypyridyl complexes having excellent photoreactivity due to rich photochemistry
mediated by energy transfer from photogenerated triplet 3MLCT excited to low-lying triplet
metal-centered (3MC) excited states [34]. The Ru(II)-arene complexes containing secondary
aromatic chromophores can be potentially photoactivated to undergo ligand dissociation or de-
arenation reactions. These photo-substituted Ru-products sometime are avid binder of DNA,
or capable of generating ROS for PDT [35, 36]. In addition, light activation of Ru(II)
t
p
complexes selectively trigger the release of Ru-bound bioactive ligands/drugs for feasible
i
r
PACT [37, 38]. c
s
u
n
a
M
d
e
t
p
e
c
c
A
s
n
o
i
t
c
a
s
n
a
r
T
n
o
Fig. 6 The photo-irradiation of UV-A light of 365 nm (6 W) to the complexes at 298 K. (a) 1 (11 M) t
l
a
for 062 min and (c) 2 (8 M) for 042 min resulted in alteration in electronic absorption spectra. (b)
D
The evolution of A absorption band for complex 1 as a function of photo-exposure time with
348 nm
monoexponential fitting. (d) The variation of A absorption band for complex 2 with photo-exposure
365 nm
time with monoexponential fitting. Inset (a) and (c) show the evolution of a new absorption band in the
range of 500800 nm (1) and 520800 nm (2), respectively.
11
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 12 of 37
The photo-irradiation of both the complexes in DMF solution with low power UV-A light of
365 nm (6 W) at RT resulted in concomitant changes of absorption bands (Fig. 6). The complex
1 showed the formation of a new band at 348 nm, along with and a broad band at ~580 nm
(Fig. 5(a) and inset). The band of 447 nm disappears over the photo-exposure time with the
evolution of isosbestic points at 330, 432 and 505 nm, indicating the formation of a single
photoproduct. Similarly, the complex 2 showed a decrease in intensity at 323 nm, and the
t
p
disappearance of 461 nm band with simultaneous evolution of a new band at 365 nm. The
i
r
photoirradiation also resulted in the appearance of a new low-energy band in the range of 500- c
s
800 nm with ~ 675 nm (Fig. 6(c) and inset). The photo illumination resulted in clear
max u
isosbestic points at 310, 343, 448, and 525 nm suggesting unique photochemical product n
a
formation. The changes in A (1) and A (2) were used to calculate the kinetic parameters.
348 365 M
The calculation suggested the half lifetime (t ) 14.5 min (1) and 17.0 min (2), and the first-
1/2
d
order rate constants (7.96±0.46) × 10-4 and (6.79±0.59) × 10-4 s-1 for complex 1 and 2
e
respectively (Fig. 6(b), (d) and Table S3, ESI†). These significant changes in absorption t
p
possibly arising from the arene-loss and/or formation of solvated species with potential PACT e
c
application, as reported for several Ru(II)-arene complexes [39].
c
A
CT-DNA Binding
s
n
DNA is the most explored biological target for the anticancer agents, especially platins and
o
many Ru(II) compounds [40]. The monofunctional organoruthenium complexes are known for
i
t
binding of double helical DNA with guanine N-7 site selectivity [41]. The double helical c
a
architecture of DNA consisting of nucleobases linked with phosphate-deoxyribose chains
s
n
exhibit a variety of interactions: non-covalent (H-bonding and interactions), covalent
a
(cross-linking and adduct formation) and electrostatic interactions. These interactions lead to
r
T
structural perturbations of DNA and ultimately resulted in inhibition of the replication
n
processes, thus binding efficacy of metallodrugs with DNA correlated with the potential of
o
complexes to damage cancer cells. t
l
a
D
12
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 13 of 37 Dalton Transactions
t
p
i
r
c
s
u
n
Fig. 7 The binding propensity of the complexes 1 and 2 with CT-DNA. (a) The continuous decrease of a
M
absorbance of complex 1 (39 M) upon gradual addition of CT-DNA (025 M) in 5 mM Tris-
HCl/NaCl buffer (pH = 7.2)-DMF mixture (50:1). Inset: Determination of binding constant (K ) of
b d
complex 1 with CT-DNA from the slope to intercept ratio of the linear fitted plot of [DNA] vs. DNA/ - e
a
t
from equation (2). (b) The decrease of emission intensity of EthB (12.5 M) bound CT-DNA upon p
f
e
addition of complex 1 (0483 M), = 546 nm, Ex and Em slit width = 10 nm. Inset: An overlay of
ex c
the variation of I/I (604 nm) vs. [Complex]. c 0
A
The absorption spectral studies provided information about binding affinity and showed
s
significant hypochromic shifts due to interactions and bathochromic shifts upon n
o
stabilization of duplex DNA [42]. The gradual addition of CT-DNA at constant complex
i
t
concentration lead to a significant decrease in absorbance of * and MLCT bands of the c
a
complexes (Fig. 7a and Fig. S14, ESI†) The results indicate possible non-covalent interactions
s
between the complex and DNA. The intrinsic DNA binding constants (K ) calculated from n
b
a
equation (2) (Table 3).
r
T
Table 3. The binding parameters for the complexes 1 and 2 with CT-DNA and HSA
n
o
Complex aK bK cC dK eK fK gn
b app 50 HSA q t
l
a
1 1.91 105 3.30 105 378 3.75 105 3.75 1013 6.12 105 1.04 D
2 5.32 105 6.07 105 206 3.01 105 3.01 1013 8.74 104 0.91
aK ,Intrinsic binding constant with CT-DNA. bK ,apparent binding constant with CT-DNA. cC , the
b app 50
complex concentration (M) corresponding to the 50% decrease of emission intensity of EthB-DNA
adduct. dK ,Stern-Volmer quenching constant of fluorescence for HSA. eK , Quenching rate constant.
HSA q
fK, HSA binding constant of complexes. gn, number of binding sites.
13
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 14 of 37
Further, to get more insight of the possible intercalation of complexes to DNA, ethidium
bromide (EthB) displacment assay was performed. The efficient quenching by aqueous buffer
makes EthB weakly emissive, and in the presence of DNA showed enhanced fluorescence
intensity [43]. The fluorescence emission of EthB-DNA adduct upon addition of complexes 1
and 2 showed a significant decrease of intensity at 604 nm (Fig. 7b and Fig. S15 ESI†). The
spectral changes correspond to the displacement of EthB from its DNA adduct. The
t
p
competitive binding of complexes to DNA determined from their apparent binding constants
i
r
(K ) (Table 3). c app
s
The complexes were further studied using circular dichroism (CD) spectral titration for
u
their ability to perturb helicity of chiral ds-DNA structure from. The results showed that the n
a
complexes results in minimal structural perturbation of DNA double-helix (Fig. S16, ESI†).
M
Human Serum Albumin (HSA) binding
d
e
Human serum albumin (HSA), being the most abundant protein in blood plasma, plays a crucial
t
p
role in numerous physiological processes-regulation of osmotic pressure and pH, and
e
transportation of endogenous metabolites. Serum proteins expresses affinity towards drugs and c
c other toxicants, and thereofore capable of modulating their action or delivery [44, 45]. The
A
optimum drug binding affinity of HSA is of significant concern for therapeutic effect and drug
s
development. The single tryptophan residue in domain IIA (Trp-214) is responsible for the
n
fluorescence of the HSA and highly sensitive to the microenvironment and thus serves as a o
i
spectral probe for the ligand binding [46]. t
c
a
s
n
a
r
T
n
o
t
l
a
D
Fig. 8 (a) The changes in emission intensity of HSA with increasing concentration complex 1 (03.7
M) in 0.8% DMF-5 mM Tris-HCl/NaCl buffer (pH 7.2) mixture, = 295 nm, = 346 nm, slit
exc em
10/5 nm. Inset: An overlay of Stern-Volmer quenching plot for the complexes 1 and 2. (b) An overlay
of log [Complex] vs. log [(I -I)/I] for the complexes 1 and 2.
0
14
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 15 of 37 Dalton Transactions
The lead Ru(III) antimetastatic agent NAMI-A and anticancer agent KP1019 show a higher
binding affinity to HSA and structurally well characterized [47−49]. The half sandwich
ruthenium-arene complexes can covalently bind to the amino acid residues of HSA and could
even induce oxidation of cysteine thiolate (Cys-34) to sulfinate [50].
The interaction of complexes 1 and 2 with HSA was studied from fluorescence quenching
measurements. The addition of complexes 1 and 2 to the HSA resulted in a significant decrease
t
p
in protein fluorescence with no shift of indicating the retention of the overall polarity of
em i
r
HSA (Fig. 8a and Fig. S17 ESI†). The second emission peak at ~ 413 nm originated from c em
s
intrinsic emission of complex 1 upon binding to a hydrophobic pocket of HSA. The quenching
u
of emission may originate from the numerous supramolecular interactions; excited-state and/or n
a
ground-state complex formation, collisional quenching, and ligand-induced conformational
M
changes [51].
d
The Stern-Volmer quenching constant (K ) and Stern-Volmer quenching rate constants were
HSA e
determined from equation (3) and equation (4) respectively (Table 3). The HSA binding t
p
constant for the complexes (K) and the number of sites (n) available was measured from the e
c
modified Stern-Volmer equation (5), suggesting a moderate binding affinity to HSA at single
c
binding site. A
Conformational analysis from synchronous fluorescence
s
n
In a multichromophoric system, the synchronous fluorescence spectra can provide information
o
about the microenvironment of specific residue. The spectral advantages over fluorescence are
i
t
the spectral simplifications, reduced bandwidth, and retained sensitivity of fluorescence. The c
a
technique utilizes a simultaneous scanning of excitation and emission monochromators by
s
n
keeping fixed wavelength differences () between them. The = 15 nm and = 60 nm
a
signifies the molecular microenvironment of tyrosine and tryptophan residue of HSA
r
T
respectively [52].
n
The synchronous fluorescence spectral measurement with = 15 nm and = 60 nm
o
for HSA upon addition of complexes 1 and 2 resulted in the significant quenching of emission t
l
a
at 284 nm [(% I/I = 67 (1) and 66 (2)) 279 nm (% I/I = 43 (1) and 44 (2)] respectively (Fig. 9
0 0 D
and Fig. S18). These results indicate the tryptophan as preferential binding site for the from
both the complexes.
15
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 16 of 37
t
p
i
r
c
s
u
Fig. 9 The synchronous fluorescence measurement of HSA (2.4 M) upon addition of complexes (03.7
n
M) 1 and 2 in 0.8% DMF-5 mM Tris-HCl/NaCl buffer (pH 7.2) mixture at 298 K. (a) Overlay of I/I a
0
M
at 284 nm vs. [complex] for = 15 nm, inset: Spectral traces of the HSA upon titration with complex
1. (b) Overlay of I/I at 279 nm vs [complex] for = 60 nm, inset: Spectral traces of the HSA upon d
0
e
titration with complex 1. Slit width for exc./em. = 10/5 nm.
t
p
Cellular Studies: In vitro cytotoxicity e
c
c The cytotoxic effects of both the complexes 1 and 2 along with cisplatin, a well-known
A
metallodrug, have been studied by MTT assay in the human brain GB cell line LN229. The
s
cytotoxicity was also compared with normal mouse fibroblast subcutaneous cell line L929.
n
Each compound was tested at a minimum of six concentrations, and the IC value was o
50
i
determined from the dose-response curves (Fig. 10). While both the Ru-complexes showed t
c
similar cytotoxicity in the GB LN229 cell line (Fig. 10a); however, the complex 2 exhibited a
s
much higher cytotoxicity towards normal fibroblast cells than 1 (Fig. 10b).
n
a
Table 4. IC values and selectivity index (S.I.) of complexes 1 and 2 with GB LN229 and
50 r
T
normal mouse fibroblast L929 cells.
n
Complexes 1 2 o
t
Cell lines IC (μM)a S.I. b IC (μM)a S.I.b l
50 50 a
D
LN229 22.8 ± 1.5 3.0 21.7 ± 1.5 2.0
L929 68.5 ± 0.7 - 44.0 ± 0.3 -
aConcentrations in μM determined for complexes after 24 h incubation.
bSelectivity index (S.I.) for complexes measured after 24 h incubation,
see the experimental section for detail.
16
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 17 of 37 Dalton Transactions
t
p
i
r
c
s
u
n
a
M
d
e
t
p
e
c
c
A
s
n
o
Fig. 10 Dose-dependent cytotoxicity of the complexes 1 and 2 and cisplatin: (a) LN229 GB cell line, i
t
c
(b) L929 fibroblast cell line, (c) cisplatin in both LN229 and L929. Data are expressed as the mean (±
a
SD) for three independent experiments. s
n
The % cell viability was used to calculate the selectivity index (S.I.), the parameter indicating a
r
selectivity of a drug towards cancer cells compared to normal cells, was determined according T
to equation (7). [Ru(6-p-cymene)(L1)Cl] (1) was found more selective towards GB cells (S. n
o
I. = 3), while [Ru(6-p-cymene)(L2)Cl] (2) displayed moderate selectivity (S. I. = 2) (Table
t
l
4). Both the complexes exhibit higher cytotoxicity than cisplatin for the GB cells (Fig. 10c).
a
D
From the dose-survival curve, it is evident that cisplatin has a poor selectivity towards cancer
cells, and indiscriminately kills normal cells equally well; probably indicating its well-known
side effects (Fig. 10c). Due to better efficacy and selectivity, further detailed anticancer studies
were carried out with complex 1.
Effect on cellular proliferation
17
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 18 of 37
The effect on cellular proliferation over a prolonged time was investigated by incubating GB
LN229 cells with [Ru(6-p-cymene)(L1)Cl] (1) at its IC concentration (22.8 µM) and
50
monitoring cell viability at various time periods by MTT cell viability assay. It was observed
that complex 1 significantly inhibits the proliferation of LN229 GB cells (Fig. 11).
t
p
i
r
c
s
u
n
a
M
d
e
t
p
e
c
c
A
s
Fig. 11 The complex 1 at its IC concentration inhibits the proliferation of LN229 cells as detected by n
50
o
MTT cell proliferation assay. The graphical data points represent mean ± SD of at least three
i
independent experiments (*p > 0.01 and <0.05, **p < 0.01, ***p < 0.001). Error bars denote ± SD. t
c
a
The clonogenic potential is the ability to proliferate into multiple cells from a single cell; it also
s
n
gives an effective measure of cellular proliferation. The ability to form a colony of cells from
a
a single cell was measured for both untreated and complex 1 treated cells using the colony
r
T
formation assay. Interestingly, colony formation was significantly inhibited in complex 1
n
treated cells. Both the number of colonies and their sizes were diminished significantly (~75%)
o
in the latter group of cells (Fig. 12). t
l
a
D
18
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 19 of 37 Dalton Transactions
t
p
i
r
c
s
Fig. 12 LN229 cells treated with complex 1 (22.8 μM) show poor clonogenic potential as determined
u
by the colony formation assay: (A) A six well plate displaying colonies formed in untretaed and n
a
Complex 1 treated cells, (B) microscopic images of formed colonies, (C) quantitative estimation of
M
number of colonies formed. The graphical data points represent mean ± SD of three independent
experiments. Error bars denote ± SD. d
e
t
Effect on cellular migration
p
e
The antimetastatic property of Ru-complexes (e.g., NAMI-A, RAPTA-C) are uniquely c
c
advantageous along with their cytotoxicity [53]. Enhanced cellular migration is a crucial
A
property of cancer cells that contributes to rapid tumor progression and metastasis. To evaluate
s
whether complex [Ru(6-p-cymene)(L1)Cl] (1) has any effect on migration of GB cells, the in n
o
vitro scratch assay or wound healing assay was performed. A scratch wound was created in the
i
monolayer of LN229 cells, and the cells were incubated with serum-free media with/without t
c
complex 1. It was observed that the cells treated with complex 1, had significantly lower a
s
migratory potential as compared to the untreated control group of cells (Fig. 13). The results
n
show that complex 1 effectively restrain the migratory potential of GB cells. a
r
T
n
o
t
l
a
D
19
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 20 of 37
t
p
i
r
c
s
u
n
a
Fig. 13 Migration inhibition of glioblastoma LN229 cells in presence of complex 1 (22.8 μM) as M
detected by in vitro scratch assay. (A) and (B) represent the migration of untreated cells on day 0 and
d
day 2 respectively; while (C) and (D) represent the same for complex 1 treated cells; E is a quantitative e
t
representation of relative migration between untreated and treated cells. The graphical data points
p
represent mean ± SD of two independent experiments. Error bars denote ± SD. e
c
c Apoptosis assessment
A
To check whether complex 1 imparts anti-GB activity by inducing apoptosis, we incubated the
s
n
GB LN229 cells with complex 1 for 48 hours and performed DAPI nuclear staining to detect
o
any morphological changes using fluorescence microscopy. Furthermore, the highly sensitive
i
t
Caspase-Glo 3/7 luminescent assay was used to measure caspase-3/7 activities as a genuine c
a
marker for apoptotic cell death [54]. As demonstrated in Fig. 14B and D, the cells treated with
s
n
complex 1 showed the characteristics of apoptotic nuclei morphology, i.e. fragmented and
a
condensed DNA with nuclear blebbing, while the untreated cells (Fig. 14A and C) maintained
r
T
smooth and round contours, which are characteristic of healthy nuclei.
n
The results from the DAPI staining experiment was further validated with the caspase 3/7 o
t
assay. We observed ~64% increase in the caspase 3/7 activity in cells treated with complex 1 l
a
when compared to the untreated control (Fig. 14D). Caspase activation indicates the onset of D
the cellular apoptotic pathways in LN229 cells in the presence of the complex 1.
20
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 21 of 37 Dalton Transactions
t
p
i
r
c
s
u
n
a
M
Fig. 14 The effect of complex 1 (22.8 μM) on apoptosis. Representative photomicrographs showing d
e
nuclear morphology of LN229 cells after 48 h as observed by nuclear DAPI staining, (A) untreated
t
cells; (B) cells treated with complex 1; (C) Untreated and complex 1 treated (D) cells from a biological p
e
repeat of the same experiment (at 40X magnification),showing nuclear condensation and blebbing
c
indicating apoptotic cells (D). (E) Complex 1 treated cells show an increased caspase 3/7 activity as c
A
compared to that of untreated cells using Caspase-Glo 3/7 assay kit. The graphical data points represent
mean ± SD of two independent experiments. Error bars denote ± SD. s
n
Conclusion o
i
t
c
We designed two half-sandwich Ru(II)-p-cymene complexes with two Schiff base ligands
a
derived from 3-aminoquinoline with 2-hydroxy benzaldehye or 2-hydroxy-1-naphthaldehye. s
n
We utilized these complexes to target human glioblastoma LN299 cancer line as possible
a
repurposed chemotherapeutic intervention for GB. The complexes were studied from several r
T
spectral and analytical techniques, and the ‘piano stool’ structure of the complexes was
n
elucidated from X-ray diffraction studies. The complexes undergo solvation in the presence of
o
Tris-buffer and DMF via replacement of labile chloride ligand by the solvent molecules. The t
l
a
complexes were avid binder to CT-DNA, and HSA binding measurement showed preferences
D
for tryptophan-214 (Trp-214) residue. The complexes are photoactive upon illumination of low
energy UV-A light (λ = 365 nm), which is correlated with photo-induced ligand dissociation.
exc
These photoactive complexes may find its application towards photodynamic therapy (PDT)
and photoactivated chemotherapy (PACT) considering rich Ru-photochemistry as an
alternative modality of GB treatment. The complexes show encouraging inhibition of
21
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 22 of 37
glioblastoma (GB) cell lines with moderate selectivity. The higher selectivity for aggressive
brain tumor cells may minimize toxic side effects associated with conventional
chemoradiotherapy. We observed the desirable outcome of complex 1 displaying significant
antiproliferative activity and lowered cellular migratory potential. The mechanistic studies of
complex 1 showed the apoptosis of GB cells with nuclear condensation, blebbing and increased
caspase 3/7 activity. We are exploring PACT activity and in vivo applications for glioblastoma
t
p
from these compounds.
i
r
c
Experimental Section
s
u
Materials and Methods n
a
M
All chemicals and regents purchased were analytical grade and used directly without any
further purifications. Analytical grade solvents were used for spectral studies while a standard d
e
method was followed for solvent purification used for synthesis [55]. 2-hydorxy-benzaldehyde
t
p
and 2-hydorxy-1-naphthaldehyde were received from Merck (Mumbai, India) and Alfa Aesar,
e
respectively. 3-Aminoquinoline purchased from Avra Synthesis Pvt. Ltd., saccharin sodium
c
c salt dihydrate from SDFCL (India), RuCl .xH O from Arora Matthey Pvt. Ltd. (Kolkata, India)
3 2
A
and tris-(hydroxymethyl)-aminomethane (Tris) from Merck life science Pvt. Ltd. (Mumbai)
s
India. Tris-HCl-NaCl buffer solution was prepared in Milli-Q water (18.2 M) and stored at
n
o
4C. -Phellandrene and bioanalytical grade calf thymus (CT) DNA, ethidium bromide (EthB)
i
and human serum albumin (HSA, lyophilized powder, Fatty acid free, Globulin free, ≥99%) t
c
were obtained from Sigma Chemicals (U. S. A.). The metal precursor [(6–p-cymene)Ru(μ- a
s
Cl)Cl] was prepared from following standard literature procedures from RuCl .xH O and - n
2 3 2
a
Phellandrene [56]. Elemental analyses were measured from Perkin-Elmer 2400 series-II
r
T
analyzer. Fourier transform infrared (FTIR) spectral data recorded in KBr pellets from Perkin-
Elmer 1320 instrument. Molecular masses were determined from Waters Q-TOF Premier n
o
electron spray ionization (ESI-MS) spectrometer. 1H and 13C NMR recorded in JEOL 400 MHz
t
l
spectrometer using deuterated solvents as a secondary reference. Absorption spectra and a
D
circular dichroism (CD) spectra measured in JASCO V-670 and JASCO J-815
spectrophotometers, respectively. Agilent Cary Eclipse spectrophotometer used for
fluorescence measurement.
Synthesis of ligands
22
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 23 of 37 Dalton Transactions
The ligands 2-[(E)-(3-quinolinylimino)methyl]phenol (L1) and 2-[(E)-(3-
quinolinylimino)methyl)naphthol were synthesized from modified literature and characterized
from reported methods [57, 58]. To the solution of 3-aminoquinline in dry ethanol, 2-hydorxy-
benzaldehyde (L1) and 2-hydorxy-1-naphthaldehyde (L2) were added, and the mixture was
refluxed for 6 h. The L1 appeared as orange-red and L2 as a yellow color precipitate, which
was separated and washed thoroughly with 3 Et 2 O. The single crystal of L2 was obtained t
p
from the slow evaporation of a mixture of ethanol and Et O at room temperature.
2 i
r
c
Synthesis of complexes s
u
Complexes [Ru(6-p-cymene)(L1)Cl] (1) and [Ru(6-p-cymene)(L2)Cl] were synthesized by n
a
following a general synthetic procedure. To the [(6–p-cymene)Ru(μ-Cl)Cl] (61.24 mg, 0.10
2 M
mM) dissolved in 4 mL CHCl , the L1 (2 equiv., 49.66 mg, 0.20 mM for complex 1) or L2 (2
3
d
equiv., 59.67 mg, 0.20 mM for complex 2) ligand dissolved also in CHCl was added dropwise
3 e
and neutralized with single equivalent of methanolic KOH solution. The mixture was warmed t
p
at 50 C for 4 h. The resulting brick red solution was dried in vacuum and washed with 3 e
c
Et O. Single-crystals of suitable dimensions of both the complexes were grown in CHCl , by
2 3 c
A
slow diffusion with n-pentane for complex 1, while complex 2 crystals obtained with slow
evaporation mixed with Et O at RT. s
2
n
o
[Ru(6-p-cymene)(L1)Cl] (1). Yield: 105 mg (95%). Brick red powder. Anal. Calcd for
i
t
C H ClN ORu: C, 60.28; H, 4.86; N, 5.41. Found: C, 60.19; H, 4.75; N, 5.34. ESI-MS
26 25 2 c
a (CHCl ): m/z 483.10 [M-Cl-]+ (calc. for C H N ORu: m/z 483.10). FT-IR (KBr matrix, /cm-
3 26 25 2
s
1): 3432 (s, br), 3053 (w), 2962 (m), 2925 (w), 2870 (w), 1936 (w), 1606 (vs, sh), 1527 (s, sh), n
a
1492 (m), 1459 (m), 1443 (vs, sh), 1414 (w), 1387 (w), 1360 (m), 1325 (m), 1276 (w), 1262
r
(w), 1191 ( m), 1164 (w), 1147 (s, sh), 1135 (m), 1055 (w), 1028 (m), 989 (w), 930 (w), 898 T
(m), 863 (w), 801 (w), 785 (m), 752 (vs, sh), 662 (w), 633 (w), 597 (w), 494 (m) [vs, very n
o
strong; s, strong; m, medium; w, weak; br, broad; sh; sharp]. 1H NMR (400 MHz, Chloroform-
t
l
d) δ 9.28 (s, 1H, imine-H), 8.43 (d, J = 2.1 Hz, 1H, quinolinyl), 8.19 (d, J = 8.4 Hz, 1H), 7.92 a
D
(d, J = 7.9 Hz, 1H), 7.84 (s, 1H), 7.78 (t, J = 7.9 Hz, 1H), 7.65 (t, J = 7.2 Hz, 1H), 7.31 – 7.22
(m, 1H), 7.05 – 6.97 (m, 2H, phenylenyl), 6.46 (t, J = 7.3 Hz, 1H, phenylenyl), 5.38 (d, J = 6.1
Hz, 1H, p-cym-H), 5.29 (d, J = 6.1 Hz, 1H, p-cym-H), 5.01 (d, J = 5.7 Hz, 1H, p-cym-H), 4.24
(d, J = 5.7 Hz, 1H, p-cym-H), 2.63 (hept, J = 6.6, 6.1 Hz, 1H, iPr-CH), 2.14 (s, 3H, p-cym-
CH ), 1.17 (d, J = 6.9 Hz, 3H, iPr-CH ), 1.13 (d, J = 6.8 Hz, 3H, iPr-CH ). 13C NMR (100
3 3 3
MHz, Chloroform-d) δ 165.80 (imine-H), 165.47 (quinolinyl), 151.41, 147.31, 146.93, 136.33,
23
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 24 of 37
135.65, 129.89, 129.61, 128.75, 128.20, 128.03, 127.49, 123.10, 118.30 (phenylenyl), 114.76
(phenylenyl), 101.93 (p-cym-C-CH ), 98.56 (p-cym-C-iPr), 86.85 (p-cym-CH), 83.52 (p-cym-
3
CH), 82.95 (p-cym-CH), 80.29 (p-cym-CH), 30.63 (iPr-CH), 22.89 (iPr-CH ), 21.80 (iPr-CH ),
3 3
18.71 (p-cym-CH ). UV- visible in DMF [ , nm (, M-1cm-1)]: 274 (16300), 325 (7800),
3 max
447 (2780).
[Ru(6-p-cymene)(L2)Cl] (2). Yield: 112 mg (93%). Brick red powder. Anal. Calcd for t
p
C 30 H 27 ClN 2 ORu: C, 63.43; H, 4.79; N, 4.93. Found: C, 63.36; H, 4.73; N, 4.85. ESI-MS i
r
(CHCl ): m/z 533.12 [M-Cl-]+ (calc. for C H N ORu: m/z 533.12). FT-IR (KBr matrix, /cm- c
3 30 27 2
s
1): 3430 (vs, br), 2960 (m), 2926 (m), 1935 (w), 1616 (vs, sh), 1603 (vs), 1581 (s), 1535 (s, sh), u
n
1501 (w), 1453 (m), 1429 (s), 1399 (s), 1369 (s), 1188 (m), 1160 (m), 1136 (m), 1031 (w),
a
999 (w), 970 (w), 859 (w), 827 (m), 784 (w), 750 (s), 645 (w), 618 (m), 480 (w). 1H NMR M
(400 MHz, Chloroform-d) δ 9.33 (s, 1H, imine-H), 8.65 (s, 1H, quinolinyl), 8.43 (s, 1H), 8.22
d
(d, J = 8.3 Hz, 1H), 7.94 (d, J = 8.1 Hz, 1H), 7.79 (t, J = 7.5 Hz, 1H), 7.71– 7.63 (m, 3H), 7.57
e
t
(d, J = 7.8 Hz, 1H), 7.33 (t, J = 7.6 Hz, 1H, naphthalenyl), 7.17 (dd, J = 14.7, 8.2 Hz, 2H,
p
naphthalenyl), 5.43 (d, J = 6.0 Hz, 1H, p-cym-H), 5.34 (d, J = 6.1 Hz, 1H, p-cym-H), 5.01 (d, e
c
J = 5.5 Hz, 1H, p-cym-H), 4.28 (d, J = 5.5 Hz, 1H, p-cym-H), 2.64 (hept, J = 6.9, 6.4 Hz, 1H,
c
iPr-CH), 2.17 (s, 3H, p-cym-CH ), 1.18 (d, J = 6.9 Hz, 3H, iPr-CH ), 1.12 (d, J = 6.9 Hz, 3H, A
3 3
iPr-CH ). 13C NMR (100 MHz, Chloroform-d) δ 166.95 (imine-H), 158.82 (quinolinyl), s
3
n
152.47, 148.11, 146.79, 136.75, 134.82, 129.75, 129.61, 129.11, 128.93, 128.19, 127.98,
o
127.75, 127.63, 126.84, 125.51, 122.42, 118.66 (naphthalenyl), 108.74 (naphthalenyl), 101.84 i
t
c
(p-cym-C-CH ), 98.58 (p-cym-C-iPr), 86.88 (p-cym-CH), 84.32 (p-cym-CH), 83.03 (p-cym-
3
a
CH), 80.54 (p-cym-CH), 30.69 (iPr-CH), 22.89 (iPr-CH ), 21.78 (iPr-CH ), 18.77 (p-cym- s
3 3
n
CH ). UV- visible in DMF [ , nm (, M-1 cm-1)]: 275 (23730), 323 (11220), 461 (3810).
3 max
a
Solubility and Stability r
T
The complexes were stable in solid-state for several months and non-hygroscopic in nature. n
o
The complexes found soluble in MeOH, CH CN, DCM, CHCl , DMF and DMSO and were
3 3
t
l
insoluble in water and aliphatic hydrocarbons (n-pentane and n-hexane). The solution-state a
D
stability of the complexes was determined from time-dependent absorption spectral
measurement of DMF solution of the complexes; time-dependent 1H NMR scans of DMSO-d
6
solution and ESI-MS spectra in CHCl solution. All these time-dependent analyses confirmed
3
the structural integrity of the complexes in the solution.
Single crystal X-ray structure
24
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 25 of 37 Dalton Transactions
Red colour and block-shaped crystal of both the complexes were obtained with suitable
dimensions at ambient conditions. Data collection was performed on Bruker SMART II
diffractometer instrument. The X-ray radiation source used was Mo K and data integration
was processed from SAINT program [59]. The structure solution was achieved from the direct
method using SHELXS program [60, 61], and absorption correction was made from SADABS
program [62]. Full matrix least-squares refinement methods were performed combinedly from t
p
both the SHELX-2017 [61] and the Olex2 software [63]. The non-hydrogen atoms refinement
i
r
processed anisotropically for full convergence and hydrogen atoms refinement performed from c
s
riding model added for geometry requirement. An ORTEP view of the structures was shown
u
in Fig. 4, the detailed crystallographic parameters given in supporting information. The CCDC n
a
deposition identity number for the structures are 2006007 (1) and 2006008 (2), respectively.
M
Hydrolysis experiments d
e
The room temperature absorption spectra of the complexes were recorded in Tris-HCl/NaCl t
p
buffer (pH = 7.2) for 5 h. The time-dependent absorbance (at 395 nm for 1, at 379 nm for 2) e
c
was plotted versus time and fitted from origin 8.5 to give monoexponential fit and first-order
c
rate constant k and half-life t from equation (1). A
1/2
s
A = C + C e-kt ……………………(1)
0 1 n
o
where, the absorbance at time t is A and C and C are constant generated from computer fit
0 1 i
t
[64]. c
a
s
Photo-irradiation experiments
n
a
The effect of light irradiation (6 W) on solution-state of the complexes was measured from the
r
T
UV-vis measurement in DMF solution at 298 K. A continuous measurement of absorbance
n
with stipulated experimental time resulted in significant alteration of the electronic spectrum
o
of the samples with the formation of multiple isosbestic point formation. The kinetics of the
t
l
photochemical changes absorbance at 348 nm (1) and 365 nm (2) with a function of time. The a
D
samples were kept in the dark to avoid any unwanted photochemical reactions.
DNA binding measurements
The binding studies of complexes with CT-DNA were studied from both the absorption and
emission techniques. The concentration of DNA was determined from its known molar
extinction coefficient ( = 6600 M-1cm-1) [65] and the purity from RNA and protein was
25
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 26 of 37
confirmed from the absorbance ratio at 260 nm and 280 nm ~ 1.9 [66]. To the 5mM Tris-
HCl/NaCl buffer (pH = 7.2), the solution of the complex (39 M) in DMF was added to the
sample cuvette, and the same volume of DMF to the reference cuvette and absorbance was
recorded. The CT-DNA (025 M for (1) and 015 M for (2)) solution was added to the
reference and sample cuvette to eliminate the absorption of DNA itself. The absorption spectra
recorded after each addition of DNA and absorbance at 275 nm (1) and 273 nm (2) used for t
p
calculation of binding constant from the following i
r
c
[DNA]/(-) = [DNA]/(-) + 1/K ( -) ...................... (2) s
a f b f b b f
u
where, [DNA] is the concentration of CT-DNA, ε and ε are molar extinction coefficient for
a b n
the apparent and fully bound species while ε representsmolar extinction coefficient in absence a
f,
M
DNA [67]. The ratio of the slope to intercept of linear fitted [DNA]/(ε − ε) vs. [DNA] plots a f
gives the intrinsic equilibrium binding constant (K ). d
b
e
The ethidium bromide displacement assay and circular dichroism DNA binding studies were
t
p
performed by following our recent publication [18].
e
c
c
Protein binding experiments
A
The binding of the complexes with human serum albumin (HSA) was determined from the
s
fluorescence quenching experiments of tryptophan of the protein and synchronous fluorescence
n
measurements in 0.8% DMF-5mM Tris-HCl/NaCl buffer (pH 7.2) mixture. To the HSA (2.4 o
i
M), the addition of complexes 1 and 2 (04 M) lead to quenching of emission intensity at t
c
346 nm. The Stern-Volmer quenching constant (K ) was calculated from the equation (3) by a
HSA
s
slope of linear fit between I /I and [Q] [68]
0 n
I /I = 1 + K [Q]……………………….. (3) a
0 HSA
r
I /I = 1 + K [Q]………………………... (4) T
0 q 0
Where, the quencher (complex) concentration [Q] lead to intensity change from I of HSA to n
0
o
I. The symbols, K and represents the quenching rate constant and average life time of the
q 0 t
l
HSA (~ 10-8 s) in the absence of quencher. Further, the number of binding sites (n) and the a
D
binding constant (K) of the complexes 1 and 2 with HSA were measured from the linear fit of
the modified Stern-Volmer equation (5) [69].
log(I -I)/I = logK + nlog[Q]……………….(5)
0
The synchronous fluorescence measurement was also recorded after each addition of the
complexes to HSA with parameters = 15 nm and = 60 nm.
Cell culture
26
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 27 of 37 Dalton Transactions
The mouse fibroblast cell line (L929) was procured from the cell repository of the National
Center for Cell Sciences (NCCS, Pune), India. The human glioblastoma (GB) cell line (LN229)
was a kind gift from Dr. Kunzang Chosdol (AIIMS, New Delhi, India). Both the cell lines were
maintained in DMEM (GIBCO) media, supplemented with 10% fetal bovine serum (FBS),
along with 100U/ml penicillin, and 100 mg/ml streptomycin. The cells were cultured at 37ºC
and 5% CO in a humidified atmosphere in an incubator.
2 t
p
i
Determination of IC 50 value and antiproliferative activity r
c
s
The in-vitro cytotoxicity of the ruthenium-based compounds and cisplatin was determined by
u
the MTT cell viability assay. The ruthenium-based drugs were dissolved in DMSO, while n
a
cisplatin was dissolved in an aqueous solution of 0.9% NaCl due to controversial reports of its
M
chemical structure alteration in DMSO (Hall et al., 2014). Briefly, about 1×105 cells of the GB
d
cell line (LN229) or the normal fibroblast cells (L929) were seeded in each well of a 96 well
e
plate, 24 hours prior to treatment. Cells were then treated with various concentrations of drugs t
p
and incubated for 24 hours (maximum final concentration of DMSO [70] used was 0.6%) at
e
c
37 ºC and 5% CO in an incubator. The following day, media was removed and 100 μl fresh
2
c
media containing 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyl tetrazolium bromide) MTT A
reagent (0.5 mg/ml) was introduced in each well, and the plate was incubated for an additional
s
n
two hours at 37 ºC in the dark. Tetrazolium crystals formed were dissolved by adding 100 μl
o
of DMSO per well. Absorbance at 595 nm was measured using a microplate reader. The 50%
i
t
inhibitory concentration (IC ) was determined by plotting the % cell viability against the drug c
50
a
concentration. Data points were collected in triplicate, and each experiment was repeated three
s
n
times. The % cell viability was calculated using the following formulae:
a
r
𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒 (𝑡𝑟𝑒𝑎𝑡𝑒𝑑)―𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒 (𝑏𝑙𝑎𝑛𝑘)
% 𝐶𝑒𝑙𝑙 𝑣𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑦= ……… (6) T
𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒 (𝑢𝑛𝑡𝑟𝑒𝑎𝑡𝑒𝑑)―𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒 (𝑏𝑙𝑎𝑛𝑘)
n
Selectivity index (S.I.) of the drug was calculated as follows: o
t
l
𝐼𝐶 50 value for normal cell line (L929) a
𝑆.𝐼. = …………(7)
𝐼𝐶 50 value for GB cell line (LN229) D
A drug displaying S.I. ≥ 3 is considered to be highly selective towards cancer [71].
To measure the effect of the drug on cellular proliferation, briefly, about 5 × 103 LN229 cells
were seeded in each well of a 96 well plate and drug treatment was given at the determined
IC value. The treated and untreated cells were scored at various time points using the above-
50
27
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Dalton Transactions Page 28 of 37
mentioned method. Relative proliferation was plotted by comparing the fold change in
absorbance values of both the groups as compared to day 0.
Colony formation assay
About 1000 LN229 cells were seeded in each well of a 6-well plate in the presence or absence
of the drug and allowed to grow up to two weeks until they form colonies. The colonies were
t
p
then fixed with 3.7% formaldehyde and stained with crystal violet. The experiment was
i
r
performed in triplicate and repeated three times. Images of the microscopic colonies, as well c
s
as the plate, were captured for comparison purposes.
u
n
Wound healing assay
a
M
LN229 Cells were seeded in a 12 well plate such that they become nearly 90-100% confluent
d
after 24 hours. The monolayer of cells was scraped with a sterile 200μl pipette tip followed by
e
removal of debris by washing with PBS. Serum-free media was introduced with or without t
p
drug and the wound closure process was monitored over a period of 24 hours by capturing e
c
image at various time points. The migration was quantified by using an arbitrary unit scale.
c
The experiment was performed three times in triplicate. A
s
Apoptosis assessment by DAPI staining
n
o
LN229 cells were seeded in 24-well plates at a seeding density of 1 × 105 cells/well and allowed
i
t
to grow overnight. The cells were then treated with 22.8 μM of the drug and incubated for 48 c
a
hours. After the treatment period, the cells were washed with PBS, and fixed with 3.7%
s
formaldehyde, followed by PBS wash. Next, the cells were permeabilized with 0.2% Triton X- n
a
100 and washed with PBS. Finally, the cells were incubated in DAPI labeling solution for 5
r
T
minutes at room temperature and imaged under a fluorescent microscope (λ ~359 nm, λ ~
ex em
n
461 nm). The experiment was repeated three times.
o
t
Apoptosis analysis by Caspase 3/7 Glo assay l
a
D
LN229 cells were subjected to 22.8 μM complex 1 treatment for 48 hours. After drug treatment,
5×103 cells from both treated and untreated control group of cells were treated with caspase 3/7
Glo reagent according to manufacturer’s instructions (Promega, USA). After an incubation of
1.5 hours at room temperature, luminescence (caspase activity), was recorded using a
luminometer (Berthold, Germany). The experiment was repeated twice.
28
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A
Page 29 of 37 Dalton Transactions
Notes
The authors declare no competing financial interest.
Acknowledgments
The Science and Engineering Research Board (Project number: EMR/2016/000521) is greatly
acknowledged for the financial support of this work. P. K. was supported by University Grant
t
p
Commission (U. G. C.) fellowship. R. K. was supported by internal grant from IIT Delhi for
i
r
this study. I. M. is thankful to the Ministry of Human Resource and Development (MHRD), c
s
Govt. of India for senior research fellowship.
u
n
Electronic Supplementary Information (ESI) available:
a
M
Characterization data, crystallographic details, additional spectral results for kinetics,
DNA/HSA binding. See DOI: xx.xxx/xxxxxxx. d
e
References
t
p
1. Bai, R.-Y.; Staedtke, V.; Riggins, G. J. Molecular targeting of glioblastoma:Drug
e
discovery and therapies. Trends. Mol. Med. 2011, 17, 301−312. c
c
2. (a) Chen, J.; McKay, R. M.; Parada, L. F.; Malignant glioma: lessons from genomics,
A
mouse models, and stem cells. Cell. 2012, 149, 36−47. (b) Li, S. C.; Vu, L. T.; Ho, H.
s
W.; Yin, H. Z.; Keschrumrus, V.; Lu, Q.; Wang, J.; Zhang, H.; Ma, Z.; Stover, A.; Weiss, n
o
J. H.; Schwartz, P. H. Loudon, W. G. Cancer stem cells from a rare form of glioblastoma
i
multiforme involving the neurogenic ventricular wall. Cancer Cell Int. 2012, 12:41 t
c
3. Crowley, R.W.; Pouratian, N.; Sheehan, J. P.; Gamma knife surgery for glioblastoma a
s
multiforme. Neurosurgical focus 2006, 20(4):E17.
n
a
4. Polivka Jr., J.; Polivka, J.; Holubec, L.; Kubikova, T.; Priban, V.; Hes, O.; r
T
Pivovarcikova, K.; Treskova, I. Advances in experimental targeted therapy and
n
immunotherapy for patients with glioblastoma multiforme. Anticancer Res. 2017, 37,
o
21−34. t
l
a
5. Wang, M.; Jiang, X.; Wu, F.; Xu, H.; Lin, Z.; Qi, B.; Xia, H. The advances in targeted
D
therapy and immunotherapy for glioblastoma: Basic research and clinical trials. Glioma
2018, 1, 79−88.
6. Kelland, L. The resurgence of platinum-based cancer chemotherapy, Nat. Rev. 2007, 7
573−584.
29
.MP
40:72:21
0202/82/8
no
yrarbiL
ytisrevinU
llenroC
yb
dedaolnwoD
.0202
tsuguA
72
no
dehsilbuP
View Article Online
DOI: 10.1039/D0DT02167A