Novel ruthenium complexes ligated with 4-anilinoquinazoline derivatives: Synthesis, characterisation and preliminary evaluation of biological activity

Article abstract of DOI:10.1016/j.ejmech.2014.02.062

The ruthenium DMSO complexes cis-Ru^IIC1₂(DMSO) ₄ and [(DMSO)₂H][trans-Ru^IIICl ₄(DMSO)₂] reacted with 4-(3′-chloro-4′- fluoroanilino)-6-(2-(2-aminoethyl)aminoethoxy)-7-methoxyquinazoline (L1), 4-(3′-chloro-4′-fluoroanilino)-6-(2-(1H-imidazol-1-yl)ethoxy) -7-methoxy quinazoline (L2), N-(benzo[d]imidazol-4-yl)-6,7-dimethoxyquinazolin- 4-amine hydrochloride (L3), 5-(6,7-dimethoxyquinazolin-4-ylamino)quinolin-8-ol hydrochloride (L4), respectively, to afford [Ru^IICl ₂(DMSO)₂(L1)] (1), [Ru^IIICl₃(DMSO) (L1)] (2), [Ru^IIICl₄(DMSO)(H-L2)] (3), [Ru ^IIICl₄(DMSO)(H-L3)] (4), and [Ru^IIICl ₃(DMSO)(H-L4)] (5), which were characterised by mass spectrometry, NMR, elementary analysis and single crystal X-ray diffraction (complex 1). Experimental screening (ELISA) showed that complexes 1, 2 and 3 are remarkably inhibitory towards epidermal growth factor receptor (EGFR) with IC₅₀ values at submicromolar or nanomolar level. Docking studies indicated that complexation with ruthenium has little interference with the formation of the two essential H-bonds between the N3 of the quinazoline ring in L1 and L2 and O-H of Thr766 through a water molecule, and the N1 of the quinazoline ring and N-H of Met769 in EGFR. Moreover, complex 2 was shown to be more active against the EGF-stimulated proliferation of human breast cancer cell line MCF-7 than the better EGFR inhibitor 4-(3′-chloro-4′-fluoroanilino)-6,7- dimethoxyquinazoline, being more potential to induce early-stage apoptosis than gefitinib. These imply that apart from inhibiting EGFR, complex 2 may involve in regulating other biological events related to the proliferation of MCF-7, implicating a novel type of multi-targeting metal-based anticancer agents.

Full text of DOI:10.1016/j.ejmech.2014.02.062

European Journal of Medicinal Chemistry 77 (2014) 110e120
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Novel ruthenium complexes ligated with 4-anilinoquinazoline
derivatives: Synthesis, characterisation and preliminary evaluation of
biological activity
,
b
,
,
b
,
,
,
,b
Liyun Ji a 1, Wei Zheng a 1, Yu Lin a b, Xiuli Wang c, Shuang Lü a b, Xiang Hao a
,
Qun Luo a , Xianchan Li b, Ling Yang c, Fuyi Wang a
,
b
,
a
,
,b,
*
*
a Beijing National Laboratory for Molecular Sciences, PR China
b CAS Key Laboratory of Analytical Chemistry for Living Biosystems, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, PR China
c Laboratory of Pharmaceutical Resource Discovery, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, PR China
a r t i c l e i n f o
a b s t r a c t
The ruthenium DMSO complexes cis-RuIIC12(DMSO)4 and [(DMSO)2H][trans-RuIIICl4(DMSO)2] reacted
with 4-(30-chloro-40-fluoroanilino)-6-(2-(2-aminoethyl)aminoethoxy)-7-methoxyquinazoline (L1), 4-(30-
chloro-40-fluoroanilino)-6-(2-(1H-imidazol-1-yl)ethoxy)-7-methoxy quinazoline (L2), N-(benzo[d]imi-
dazol-4-yl)-6,7-dimethoxyquinazolin-4-amine hydrochloride (L3), 5-(6,7-dimethoxyquinazolin-4-
ylamino)quinolin-8-ol hydrochloride (L4), respectively, to afford [RuIICl2(DMSO)2(L1)] (1), [RuIIICl3(DM-
SO)(L1)] (2), [RuIIICl4(DMSO)(H-L2)] (3), [RuIIICl4(DMSO)(H-L3)] (4), and [RuIIICl3(DMSO)(H-L4)] (5),
which were characterised by mass spectrometry, NMR, elementary analysis and single crystal X-ray
diffraction (complex 1). Experimental screening (ELISA) showed that complexes 1, 2 and 3 are remark-
ably inhibitory towards epidermal growth factor receptor (EGFR) with IC50 values at submicromolar or
nanomolar level. Docking studies indicated that complexation with ruthenium has little interference
with the formation of the two essential H-bonds between the N3 of the quinazoline ring in L1 and L2 and
OeH of Thr766 through a water molecule, and the N1 of the quinazoline ring and NeH of Met769 in
EGFR. Moreover, complex 2 was shown to be more active against the EGF-stimulated proliferation of
human breast cancer cell line MCF-7 than the better EGFR inhibitor 4-(30-chloro-40-fluoroanilino)-6,7-
dimethoxyquinazoline, being more potential to induce early-stage apoptosis than gefitinib. These
imply that apart from inhibiting EGFR, complex 2 may involve in regulating other biological events
related to the proliferation of MCF-7, implicating a novel type of multi-targeting metal-based anticancer
agents.
Article history:
Received 28 August 2013
Received in revised form
24 February 2014
Accepted 28 February 2014
Available online 1 March 2014
Keywords:
Ruthenium complexes
EGFR inhibitors
Multi-targeting
Anticancer agents
MCF-7
Ó 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
effects such as neuro-, hepato- and nephrotoxicity, and inherent or
acquired resistance [2,3]. During the past three decades, therefore,
Since the serendipitous discovery of the biological activity of
cisplatin in 1965 [1] and its subsequent clinical use for the treat-
ment of various solid tumours including genitourinary, colorectal,
and non-small cell lung cancers [2], medicinal inorganic chemistry
has become a subject of intensive studies and continues to attract
much attention in the drug discovery field [3,4]. However, the
clinical use of cisplatin is largely restricted by dose-limiting side-
continuous effort has been devoted to the development of new
platinum drugs and other metal-based, in particular ruthenium-
based anticancer drugs to circumvent these limitations [3e8].
Current interest in ruthenium anticancer complexes was stim-
ulated by the discovery of (H2im)[trans-RuCl4(Him)(DMSO)]
(NAMI-A, Him ¼ imidazole) [9,10] and Indazolium trans-[tetra-
chlorobis(1H-indazole)ruthenate(III)] (KP1019) [11], of which both
are being in clinic trails. NAMI-A is in vivo active to prevent the
development and growth of pulmonary metastases in all the solid
tumours [6,9]. KP1019 can induce apoptosis at non-toxic levels via
the mitochondrial pathway and exhibits promising activity against
certain types of tumours which are not successfully treatable with
cisplatin [11,12].
* Corresponding authors. CAS Key Laboratory of Analytical Chemistry for Living
Biosystems, Institute of Chemistry, Chinese Academy of Sciences, No. 2 the First
North Street, Zhongguancun, Beijing 100190, PR China.
E-mail addresses: quoluo@iccas.ac.cn (Q. Luo), fuyi.wang@iccas.ac.cn (F. Wang).
These authors contributed equally to this work.
1
http://dx.doi.org/10.1016/j.ejmech.2014.02.062
0223-5234/Ó 2014 Elsevier Masson SAS. All rights reserved.

L. Ji et al. / European Journal of Medicinal Chemistry 77 (2014) 110e120
111
Another reason that ruthenium attracts increasing attention in
drug discovery field is that the well known octahedral ruthenium
centre provides many synthetic opportunities for controlling and
tuning the biological activities of inorganic pharmaceuticals by
organising a wide range of multi-functional ligands in the three-
dimensional space [3,13e19]. On the one hand, subject to extra-/
intracellular hydrolysis or reduction, either RuII or RuIII centre itself
in these ruthenium complexes is the main reactive site towards
biological molecules as has been reported for RuII in the ruthenium
organometallic ruthenium fragments, affording novel dual-
functional ruthenium complexes [34]. The complexation with
organoruthenium fragments confers the 4-anilinoquinazoline
pharmacophores higher potential inducing cellular apoptosis
while the highly inhibitory activity of 4-anilinoquinazolines against
EGFR and the reactivity of the ruthenium centre to 9-ethylguanine
well preserve [34]. In the present work, we have designed and
synthesised a series of RuII/RuIII DMSO complexes bearing 4-
anilinoquinazoline derivatives, which are analogues of gefitinib, a
specific inhibitor of EGFR for clinic treatment of locally advanced or
metastatic non-small cell lung cancer (NSCLC) [35]. Our goal is
conferring the ruthenium complexes inhibitory potency against
EGFR while maintaining the original activity, for instance cytotox-
icity, of the ruthenium pharmacophores as much as possible.
arene complexes [(h
6-arene)Ru(YZ)(X)][PF6] [20e22], where X is a
halide, YZ a chelating diamine such as ethylenediamine (en), and
for RuIII in KP1019 [12,23]. On the other hand, ruthenium has been
used as a scaffold to organise various well-established bioactive
organic pharmacophores around the metal centre [14,16]. In these
cases, the Ru centre is either active [24e28] or solely a building
block but not involved in any direct interactions with biological
targets [29,30]. For example, coordinating with a potential leaving
group (chloride), the Ru centre in ruthenium arene complexes
bearing protein kinase inhibiting tyrphostin or topoisomerase
inhibiting flavonoid moieties are active towards DNA while the
bioactive ligands show inhibitory potential towards epidermal
growth factor receptor (EGFR) and topoisomerase, respectively
[24,27], which conferring the target complexes with dual- or multi-
functional potential. However, in a series of organometallic ruthe-
nium complexes mimicing the three dimensional structure of
staurosporine [14,29e31], a highly potent protein kinase inhibitor
(PKI), the ruthenium centre is inert while the entire molecules
show highly inhibitory potency towards protein kinases such as
2. Results and discussion
2.1. Chemistry
To develop a suitable ligand containing the 4-anilinoquinazoline
pharmacophore for coordination to either RuII or RuIII, very
recently, we have synthesised a series of 4-anilinoquinazoline de-
rivatives by modifying the pharmacophore using various sub-
stituents at either the 6- or 4-position of quinazoline core (Scheme
1) [33]. The precursor compound 4-(30-chloro-40-fluoro)-6-
hydroxy-7-methoxyquinazoline reacted with 1, 2-dibromoethane
in the presence of potassium carbonate at 353 K, giving rise to
the modified 4-anilinoquinazoline intermediate 4-(30-chloro-40-
fluoroanilino)-6-(2-bromoethoxy)-7-methoxyquinazoline (L0), and
then L0 reacted with ethylenediamine (en) and imidazole (Im) in
acetonitrile at 353 K to form 4-anilinoquinazoline derivatives L1
Pim-1 [32] and glycogen synthase kinase 3b (GSK3b) [30].
Recently, we have synthesised a series of EGFR inhibiting 4-
anilinoquinazoline derivatives [33] which can be linked to
Scheme 1. Synthesis and Chemical structures of 4-anilinoquinazoline derivatives and Ru(II) (1) and Ru(III) (2, 3, 4 and 5) complexes containing 4-anilinoquinazoline derivatives and
DMSO ligands.

112
L. Ji et al. / European Journal of Medicinal Chemistry 77 (2014) 110e120
Table 1
and L2, respectively (Scheme 1). The reactions of the precursor
compound 4-chloro-6,7-dimethoxyquinazoline (L00) with 5-
aminobenzoimidazole and 5-aminoquinolin-8-ol gave rise to 4-
anilinoquinazoline derivatives L3 and L4, respectively [33], as
shown in Scheme 1.
Crystallographic data of complex 1.
Formula
C23H33C13FN5O4RuS2
734.08
Triclinic
P-1
12.5570(17)
14.3849(17)
18.7823(19)
93.090(7)
3091.7(6)
4
Molecular weight
Crystal system
Space group
The reactions between the 4-anilinoquinazoline derivative L1
ꢀ
a (A)
with the ruthenium-DMSO complexes cis-RuIIC12(DMSO)4 and
ꢀ
b (A)
[(DMSO)2H][trans-RuIIICl4(DMSO)2]
afforded
complexes
ꢀ
c (A)
(ꢁ)
[RuIICl2(DMSO)2(L1)] (1) and [RuIIICl3(DMSO)(L1)] (2), respectively.
Slow diffusion of diethyl ether into the DMSO/acetone (1:5) solu-
tion of 1 gave rise to yellow plate crystals suitable for X-ray
diffraction analysis. The X-ray structure and atom numbering
scheme are shown in Fig. 1, crystallographic data and the selected
bond lengths and angles are listed in Tables 1 and 2. It can be seen
that the coordination sphere around the ruthenium centre consti-
tutes a distorted octahedron with coordination of the bidentate
(NeN) ligand L1, and two chloride ions in a trans-configuration and
two DMSO through the S atom in a cis-configuration. The RueS
bond distance for the two DMSO ligand are 2.2474(18) and
b
3
ꢀ
V (A )
Z
Crystal size (mm)
Crystal description
Crystal colour
0.20 ꢂ 0.13 ꢂ 0.07
plate
yellow
1.577
0.945
173(2)
Dx (Mg mꢀ3
)
m
(mmꢀ1
T (K)
)
ꢀ
Wavelength (A)
Data-collection mode
qmax (ꢁ)
No. of integrated refl.
Rint
Final R and Rw
No. of parameters
Drmax
0.71073 (Mo-K
scans
a)
u
27.48
34,682
0.0507
0.0821, 0.2301
723
ꢀ
2.2557(18) A, and the SeO bond lengths are 1.483(6) and 1.499(5)
ꢀ
A, respectively. The SeO bond lengths are shorter than that in free
ꢀꢀ3
ꢀ
,
Drmin (e A
)
3.084, ꢀ1.109
DMSO molecule (1.531(5) A) [36], indicating a considerable in-
crease of double-bond character of the SeO bond upon coordina-
tion to ruthenium via the sulphur atom. The bonds between Ru(II)
ion and the NeN bidentate ligand (2.120(6) and 2.182(5)A,
containing 4-anilinoquinazoline ligands slightly shift to visible light
region and the absorption signals increase in intensity, especially at
ꢀ
350 nm, compared to those of the RuII arene complex [(h6
-
respectively) are similar to those found in Ru(II) complexes con-
taining polypyridine ligand [37], but slightly longer than those in
the ruthenium(II) phosphine/diimine/picolinate complexes [38].
The reactions of Ru(III) complex [(DMSO)2H][trans-
RuIIICl4(DMSO)2] with 4-anilinoquinazoline derivatives L2, L3 and
L4 afforded complexes 3, 4 and 5, respectively (Scheme 1). In the 4-
anilinoquinazoline derivative L3, the hydrogen atom can be located
on either of the two nitrogen atoms in the benzoimidazole moiety.
Thus L3 exists in two non-equivalent tautomeric forms, of which
both can coordinate to ruthenium via the non-protonated N.
Indeed, our LC-MS analysis shows that complex 4 presents in a pair
of tautomeric forms as evidenced by identical mass spectra (data
not shown) and different HPLC retention time (Fig. 2a).
Fig. 2 shows the HPLC chromatograms (Fig. 2a) of the five
ruthenium complexes in DMSO as well as the UVeVis spectra
recorded during the HPLC analysis (Fig. 2b). It can be seen that (i)
there is little hydrolysis reaction occurring during the HPLC sepa-
ration; and (ii) apart from the absorption bands corresponding to
the aromatic rings of the 4-anilinoquinazoline ligands, these
complexes show strong ligand-to-metal charge transfer (LMCT)
absorption centred at 342 nm with a shoulder band at ca. 400 nm
upon the coordination of ruthenium (II or III) with the ligands.
Complex 5 bearing the bidentate (Ne, Oe) ligand L4 exhibits the
weakest LMCT absorption at 342 nm but the strongest absorption at
400 nm. The LMCT absorption bands of these ruthenium complexes
biphenyl)Ru(ethylenediamine)Cl]þ [39] and the RuIII complex Na
[trans-RuCl4(indazole)2] [40].
In view of the potential importance of aquation in the biological
mechanism of action of the ruthenium complexes 1e5 as reported
to the ruthenium(II) arene complexes [17,39,41e43] and ruth-
enium(III) complexes such as NAMI-A and KP1019 [40,44e47],
complex 2 was chosen for a preliminary hydrolysis study. As shown
in Fig. 3, apart from the HPLC peak corresponding to the intact
complex 2, there are two new HPLC peaks observed for the solution
of complex 2 in the mixture of DMSO/H2O (1:9, v/v) incubated at
310 K for 30 min. The intact species decreased and the new species
b and c, which were identified by LCeMS to be L1 (m/z 406.2 for the
most abundant isotopomer of [M þ H]þ) and the hydrolytic product
[RuIIICl2(DMSO)(L1)(H2O)]þ (2a, m/z 655.6 for [M
ꢀ
H2O]þ),
increased in content with increase in incubation time. Although
there was no HPLC peak detected and corresponding to the hy-
drolytic product [RuIIICl3(DMSO)(H2O)]þ (2b) formed by dissocia-
tion of complex 2 because this highly polar species may has no
Table 2
ꢁ
ꢀ
Selected bond lengths (A), angles and torsion ( ) for complex 1.
Ru(1)-N(1)
Ru(1)-S(1)
Ru(1)-Cl(1)
S(1)-O(3)
O(1)-C(4)
C(1)-C(2)
N(1)-C(1)
2.120(6)
2.2474(18)
2.4670(14)
1.483(6)
1.454(7)
1.499(10)
1.501(9)
Ru(1)-N(2)
Ru(1)-S(2)
Ru(1)-Cl(2)
S(2)-O(4)
O(2)-C(13)
C(3)-C(4)
2.182(5)
2.2557(18)
2.4497(14)
1.499(5)
1.432(8)
1.511(9)
N(2)-C(2)
1.497(8)
S(1)-Ru(1)-Cl(1)
S(1)-Ru(1)-Cl(2)
S(2)-Ru(1)-Cl(1)
S(2)-Ru(1)-Cl(2)
N(1)-Ru(1)-Cl(1)
N(1)-Ru(1)-Cl(2)
N(2)-Ru(1)-Cl(1)
N(2)-Ru(1)-Cl(2)
Ru(1)-N(1)-C(1)
Ru(1)-N(2)-C(2)
90.20(6)
94.24(6)
96.53(6)
88.68(6)
87.91(16)
86.82.(16)
87.64(14)
86.88(14)
109.5(4)
107.1(4)
S(1)-Ru(1)-N(1)
S(1)-Ru(1)-N(2)
S(2)-Ru(1)-N(1)
S(2)-Ru(1)-N(2)
N(1)-Ru(1)-N(2)
Cl(1)-Ru(1)-Cl(2)
S(1)-Ru(1)-S(2)
N(1)-C(1)-C(2)
N(2)-C(2)-C(1)
N(2)-C(3)-C(4)
87.23(16)
167.92(15)
175.47(16)
98.40(14)
80.8(2)
172.94(5)
93.64(7)
106.5(6)
110.0(6)
113.0(5)
S(1)-Ru(1)-N(2)-C(2)
S(2)-Ru(1)-N(2)-C(2)
ꢀ3.1(10)
Cl(1)-Ru(1)-N(2)-C(2)
Cl(2)-Ru(1)-N(2)-C(2)
76.8(4)
ꢀ98.8(4)
Fig. 1. X-ray crystal structure and atom numbering scheme for [RuIICl2(DMSO)2(L1)]
(1) at 50% probability thermal ellipsoids.
173.0(4)

L. Ji et al. / European Journal of Medicinal Chemistry 77 (2014) 110e120
113
2.2. Enzyme assays
We have previously determined the IC50 values, which are the
concentration of the enzyme inhibitors requested to gain 50% in-
hibition against the targeted enzyme, of the 4-anilinoquinazoline
derivatives L0eL4 (Scheme 1) against EGFR by enzyme-linked
immunosorbent assay (ELISA) to be 3.8, 12.1, 60.2 nM, >5
mM and
>5 M, respectively [33]. These indicate that the substitution by N-
m
containing groups ethylenediamine (en) and imidazole (Im) via the
C2 alkyl linker at the 6-methoxyl position of the 4-
anilinoquinazoline pharmacophore has no significant effect on its
inhibition capacity against EGFR, but the modification on the ani-
line moiety remarkably reduces the inhibitory potency of 4-
anilinoquinazoline pharmacophore as happened to L3 and L4
(Scheme 1).
Here, we applied the same method to characterise the inhib-
itory activity of the five ruthenium complexes bearing a 4-
anilinoquinazoline ligand (Scheme 1) against EGFR. With the
well-known EGFR inhibitor L0 [48] as a reference, at the dose of
10
mM, the relative inhibition efficiency (%) of the 4-
anilinoquinazoline derivatives (L0eL4) and their ruthenium
complexes 1e5 were measured to be 100 (L0), 100 (L1), 100 (L2),
77 (L3), 58 (L4), 99.95 (1), 98.56 (2), 98.47 (3), 23.8 (4) and 51 (5),
respectively. These indicate that Ru(II) or Ru(III) coordination to
the en in L1, Im in L2 or the quinoline group in L4 has no pro-
nounced effect on the inhibitory activity of the respective 4-
anilinoquinazoline derivatives towards EGFR. However, the RuIII
DMSO unit ligated with L3 through coordination with the nitro-
gen atom of benzoimidazole in L3 significantly reduces the
inhibitory potency of L3.
Next, the IC50 values to EGFR of the active complexes 1, 2 and 3
were determined, the concentration-dependent inhibitory curves
are shown in Fig. 4 and IC50 values are listed in Table 3. The results
indicate that Ru-coordination to en in L1 pronouncedly reduces the
inhibitory activity of L1 towards EGFR, but the IC50 values of the
resulting complexes 1 and 2 are still at sub-micromolar level.
Notably, although the Im modification at the 6-position leads to a
pronounced reduction of the inhibitory capacity of the 4-
anilinoquinazoline pharmacophore, the activity of the 4-
anilinoquinazoline derivative L2 against EGFR remain intact sub-
ject to the Ru(III) coordination to the Im group.
Fig. 2. (a) HPLC Chromatograms with UV detection at 350 nm of complexes 1e5
(1 mM) in DMSO; (b) UVeVis spectra of complexes 1e5 recorded during the HPLC
analysis, of which the chromatograms are shown in (a).
retention in the C18 reverse phase HPLC column, these results
unambiguously indicate that complex 2 can be hydrolysed via the
substitution of either the 4-anilinoquinazoline ligand (L1) or one of
the chloride ligands. The HPLC time course (Fig. 3) indicate that the
hydrolysis of complex 2 did not reach equilibrium status until 12 h,
suggesting that the hydrolysis rate of complex 2 is much lower than
those of [(h
6-biphenyl)Ru(ethylenediamine)Cl]þ [39] and indazo-
lium [trans-tetrachlorobis(1H-indazole)ruthenate(III)] (KP1019)
[44], of which the half lives of hydrolysis under physiological
relevant conditions are 9.02 and 17.1 min, respectively.
Fig. 3. HPLC time course with UV detection at 260 nm for the aquation of complex 2
(1 mM) in mixture of DMSO and H2O (v/v: 1/9) incubated at 310 K. Peak assignment: a,
Fig. 4. Concentrationeresponse inhibition curves of the 4-anilinoquinazoline de-
intact
complex
2
([RuIIICl3(DMSO)(L1)]);
b,
hydrolytic
product
rivatives L1 and L2 and the ruthenium complexes 1,
mean ꢃ SD of triplicate determinations.
2 and 3 on EGFR. Points:
[RuIIICl2(DMSO)(L1)(H2O)]þ; c, L1 dissociated from complex 2.

114
L. Ji et al. / European Journal of Medicinal Chemistry 77 (2014) 110e120
Table 3
IC50 for inhibition of EGFR and of the growth of MCF-7 cancer cells of ruthenium
complexes 1e5.
Compound
IC50 to EGFR/nMa
IC50 to MCF-7/m
Mb
ꢀ EGF
þ EGF
1
2
3
4
283 ꢃ 11.0
168 ꢃ 7.4
60.8 ꢃ 3.5
>100
>100
92.1 ꢃ 2.2
18.3 ꢃ 1.9
>100
>100
>10
>5
m
M
>100
>100
>100
>100
5
m
M
L0
3.8 ꢃ 2.5
>100
13.0 ꢃ 3.4
55.3 ꢃ 3.1
RM116c
N. A.d
21.8 ꢃ 1.7
a
The IC50 values were determined in the presence of 100 mM ATP.
The cells were exposed with each tested compound for 48 h.
RM116 ¼ [(
b
c
h
6-p-cym)Ru(en)Cl]PF6.
d
Not applicable.
2.3. Docking analysis
We have previously established an indirect docking analysis
method using the Surflex-Dock module of Sybyl X 1.1 program with
introduction of the orders “-fmatch” and “-cpen” [33]. Herein, we
applied this similar method to dock complexes 1, 2 and 3 into the
ATP-binding pocket in the EGFR kinase domain. It is worthy of
pointing out that because the hydrolysis of these ruthenium com-
plexes are slow (supra vide), the complexes were docked in their
intact forms as shown in Scheme 1. Interestingly, although Ru(II)
coordination to the bidentate en group leads to formation of an
additional H-bond between the O at 7-postion of quinazoline ring
and the amide N of Asp 776 (Fig. 5a), and Ru(III) coordination to the
en group of L1 leads to formation of two additional H-bonds be-
tween the NeH of en and the COOꢀ of Asp 776, and the NeH of en
and the backbone CeO of Leu694 (Fig. 5b), the ruthenium coordi-
nation reduces the inhibition capacity of L1 (Table 3). In contrast,
without inducing formation of new H-bond between L2 and EGFR
kinase domain (Fig. 5c), Ru(III) binding to the imidazole group of L2
has little impact on the inhibition activity of L2 against EGFR. One of
possible explanation is that compared to the non-coordinated L1,
the poses of ligand L1 of complexes 1 and 2 at the ATP binding cleft
of EGFR are changed significantly due to Ru coordination (Fig. 5a, b).
However, the Ru coordination to the Im group does not change the
conformation of the ligand L2 of complex 3 at the ATP-binding cleft
of EGFR (Fig. 5c).
2.4. Antiproliferation potency
To evaluate the anticancer potency of the EGFR inhibiting
ruthenium complexes 1e3, we employed sulforhodamine B (SRB)
colourimetric assay to screen the in vitro antiproliferation activity
of ruthenium complexes 1, 2 and 3 towards human breast cancer
cell line MCF-7, which has been reported to overexpress EGF re-
ceptor [49,50], in the absence and the presence of EGF. The well-
established EGFR inhibitor L0 [48] and the cytotoxic ruthenium
arene complex [(h
6-p-cym)RuII(en)Cl]PF6 (RM116) [51] were used
Fig. 5. The docked conformers/poses of complexes (a) 1, (b) 2 and (c) 5 (coloured by
atoms) and their 4-anilinoquinazoline ligands L1 (red, for 1 and 2) and L2 (red, for 3) at
the ATP binding cleft of EGFR kinase as generated via Surflex docking-scoring com-
binations. The dotted yellow lines illustrate the positions of probable H-bonding in-
teractions as calculated by the H-Bond calculator imbedded in the Surflex-Dock
module of Sybyl X 1.0 program. (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.)
as positive references during the in vitro screening. The dose-
dependent inhibition curves of the tested compounds on the
growth of MCF-7 are shown in Fig. 6, and the resulting IC50 values
are listed in Table 3.
The results indicate that in the presence of 10 nM EGF, the EGFR
inhibiting L0 exhibited dose-dependent inhibition on the prolifer-
ation of MCF-7 cancer cells, the average IC50 value from three ex-
[35]. In contrast, the cytotoxic organometallic ruthenium complex
RM116 exhibited EGF-independent inhibition on the proliferation
periments was 55.3 mM. While even at 100 mM, L0 showed little
inhibition on the proliferation of MCF-7 in the absence of EGF. This
implies that the 4-anilinoquinazoline derivative (L0) indeed exert
its inhibition on proliferation of MCF-7 via blocking EGFR signalling
of MCF-7, the IC50 values range from 13.0 to 21.8
mM, similar to that
of RM119 against human ovarian cancer line A2780 [51].

L. Ji et al. / European Journal of Medicinal Chemistry 77 (2014) 110e120
115
MCF-7 cells. The inhibitory potency of complex 2 on the EGF-
induced growth of MCF-7 is similar to that of the non-specific
and cytotoxic organoruthenium complex RM116 [51] and pro-
nouncedly higher than that of L0, though the inhibitory capacity
of complex 2 on EGFR is lower than that of L0. These imply that
other than blocking EGFR signalling, complex 2 may inhibit the
proliferation of MCF-7 via other ways, for example inducing
apoptosis via the mitochondrial pathway through hydrolytic
activation as does KP1019 [11], or via blocking DNA synthesis and
replication as supposed to cytotoxic ruthenium complex RM116
[52].
The cytotoxicity of complex 2 on the growth of the normal
human embryonic kidney 293 (HEK293) cells was also tested. The
dose-dependent inhibition curves are showed in Fig. 6c, corre-
sponding to IC50 values being 195 ꢃ 10 and 188 ꢃ 18
mM in the
absence and in the presence of EGF, respectively. These results
further verify the highly selective inhibition of complex 2 on the
EGF-induced growth of MCF-7 cancer cells.
2.5. Induction of apoptosis
In order to understand the mechanism of action of complex 2,
the capacity of complex 2 to induce apoptotic cell death was
evaluated by fluorescence microscopy imaging of MCF-7 stained
with 4,6-diamidino-2-phenylindole (DAPI). The monofunctional
EGFR inhibitor gefitinib is thought to exert its effect by blocking the
mitogen-activated protein kinase (MAPK) signalling pathway and
to have less capacity to induce apoptotic cell death [53]. The
ruthenium arene compound RM116 containing the cytotoxic
pharmacophore (arene)Ru(en) is anticipated to be more active to
induce apoptosis [51]. Indeed, the microscopic image (Fig. 7c) gives
clear evidence of the formation of more apoptotic bodies, charac-
terised by the fragmentation of nuclei with condensed chromatin,
upon the treatment of MCF-7 with the cytotoxic RM116 than those
resulting from the treatment with EGFR inhibiting gefitinib and
complex 2 (Fig. 7b and d). Furthermore, the flow cytometric
quantification indicated that complex 2 (Fig. 7h) induced more
early-stage apoptosis and necrosis than gefitinib did (Fig. 7f),
though the inhibitory potency of gefitinib (IC50 ¼ 33 nM [54]) is
higher than that of complex 2 (IC50 ¼ 168 nM). These results sug-
gest that (i) ligation with the EGFR inhibiting anilinoquinazoline
derivative L1 makes the NAMI-like complex 2 in vitro cytotoxic
towards primary cancer cell line MCF-7, and (ii) complexation with
the RuIII(DMSO) chlorido fragment confers the effective EGFR in-
hibitor L1 with additional capacity inducing cellular apoptosis, in
particular early-stage apoptosis, to the blockade of MAPK
signalling.
3. Conclusions
In conclusion, in this work, we demonstrate that the modification
by ethylenediamine or imidazole group at 6-postion of the 4-
anilinoquinazoline derivatives L0 allows this well-established phar-
macophore to serve as a chelating or monodentate ligands of which
the complexation with RuII and RuIII DMSO fragments affords com-
plexes to give rise to a new type of Ru complexes with interesting
inhibitory potency against EGFR. Moreover, the RuIII DMSO complex
[RuIIICl3(DMSO)(L1)] (2, L1 ¼ 4-(30-chloro-40-fluoroanilino)-6-(2-(2-
Fig. 6. (a, b) Dose-dependent inhibition curves of complex 2 and 4-anilinoquinazoline
derivative L0 on the growth of human breast cancer cell line MCF-7 in the absence (a)
and in the presence (b) of EGF (10 nM). (c) Dose-dependent inhibition curves of
complex 2 on the growth of human embryonic kidney cell line HEK293 in the absence
or in the presence of EGF (10 nM).
aminoethyl)-aminoethoxy)-7-methoxyqui
nazoline)
exhibits
Interestingly,
bearing
the
EGFR
inhibiting
4-
higher in vitro inhibitory potency specifically on the EGF-induced
proliferation of cancer cell line MCF-7 than the non-coordinated 4-
anilinoquinazoline derivative L0 (4-(30-chloro-40-fluoroanilino)-6,7-
dimethoxyquinazoline) which is more active against EGFR than
complex 2. The apoptosis assays indicate that complex 2 is more
active to induce early-stage apoptosis than gefitinib, a clinically used
anilinoquinazoline L1 or L2 [33], ruthenium complexes 1, 2 and
3 showed different antiproliferation activity on MCF-7. The RuIII
complex 2 ([RuIIICl3(DMSO)(L1)]) exhibited a remarkably EGF-
stimulation-dependent inhibition on proliferation of MCF-7,
while complexes 1 and 3 are inactive towards the growth of

116
L. Ji et al. / European Journal of Medicinal Chemistry 77 (2014) 110e120
Fig. 7. Confocal fluorescent images (left) with emission at 461 nm and flow cytometric quantification (right) of viability (bottom left quadrant), early-stage apoptosis (bottom right
quadrant), late-stage apoptosis (top right quadrant) and necrosis (top left quadrant) of MCF-7 cells treated with 50 M of different compounds in the presence of 10 nM EGF at 310 K
for 24 h. The number in each quadrant indicates the respective percentages of total cell populations. (a, e) untreated control; (b, f) gefitinib; (c, g) RM116; (d, h) complex 2.
m
anticancer drug. These imply that complex 2 may exert inhibition on
the proliferation of MCF-7 not only via blocking the EGFR signalling,
but alsovia other pathways, forexample, blocking DNA synthesis and
replication. In other words, complex 2 may represent a novel class of
multi-targeting anticancer agents involving both blockage of EGFR
signalling and induction of early-stage apoptosis cascades. To this
regard, it is worthy of further studies in details for elucidation of the
molecular mechanism of action of complex 2.
4. Experimental section
4.1. Materials
RuCl3$3H2O (Ru > 36.7%) was purchased from Shenyang Jingke
Reagent Co. (China), the Catalyst Pd/C from Beijing Ouhe Technol-
ogy Co. (China), 4-chloro-6,7-dimethoxylquinazoline (AR grade)
from Shanghai FWD Chemicals Co. (China), 5-nitro-8-

L. Ji et al. / European Journal of Medicinal Chemistry 77 (2014) 110e120
117
hydroxylquinoline from Wuhan Kaitong Fine Chemicals Co. (China),
and Imidazole from Tianjin Jinke Fine Chemical Research Institute
(China). Organic solvents including absolute methanol, absolute
ethanol, absolute ether, acetonitrile, dichloromethane and DMSO
were all analytical grade and used directly without further
purification.
Column chromatography silica gel was purchased from Qingdao
Jiyida Silica Reagent Manufacture (China), and thin layer chroma-
tography silica gel from Yantai Institute of Chemical Industry
Research (China). Trifluoroacetic acid (TFA) was purchased from
Sigma, chromatographic grade acetonitrile from Tedia Company
(China).
ligand. Anal. Calcd for C21H27Cl4FN5O3RuS$3H2O (F.W.: 745.9): C,
33.83; H, 4.46; N, 9.39. Found: C, 33.80; H, 4.13; N, 9.18.
4.2.3. [RuIIICl4(DMSO)(H-L2)] (3)
Complex 3 was synthesised following a similar procedure re-
ported in the literature [57,58]. The precursor complex [(DMSO)2H]
[trans-RuCl4(DMSO)2] (0.036 mmol) was dissolved in 4 mL of EtOH/
HCl (0.1 M), the resulting solution was stirred for 5 min at room
temperature, and 0.072 mmol of 4-anilinoquinazoline derivative L2
was then added. The orange solution was stirred for 24 h, and the
yellow solid appeared in the solution was filtered off, washed with
ethanol and diethyl ether, and dried in vacuum to give rise to
complex 3 (10.6 mg, yield: 40%). ESI-MS (negative): m/z 735.2 for
The protein tyrosine kinase epidermal growth factor receptor
(EGFR), and the epidermal growth factor (EGF) were purchased
from Sigma, and other biological agents including the ELISA kits for
EGFR inhibitor screening from Cell Signaling Technology Inc. (USA).
[M ꢀ H]ꢀ (theoretical m/z 734.9). 1H NMR (DMF-d7)
d (ppm):
ꢀ13.09 (very broad) for S-DMSO ligand. Anal. Calcd for
C22H24Cl5FN5O3RuS (F.W.: 735.9): C, 35.91; H, 3.29; N, 9.52. Foud: C,
Sodium
silicate
nonahydrate
(Na2SiO3$9H2O),
cetyl-
35.88; H, 3.70; N, 8.93.
trimethylammonium bromide (CTAB), iron nitrate nonahydrate
(Fe(NO3)3$9H2O), ethylene glycol, ammonium hexafluorotitanate
((NH4)2TiF6), boric acid (H3BO3) were purchased from Shanghai
General Chemical Reagent Manufacture (China). The deionised
water used in the experiments was prepared by a Milli-Q system
(Millipore, Milford, MA).
4.2.4. [RuIIICl4(DMSO)(H-3)] (4)
This complex was prepared following a procedure similar to that
for synthesis of complex 3. The precursor complex [(DMSO)2H]
[trans-RuCl4(DMSO)2] (0.036 mmol) was dissolved in 4 mL of EtOH/
HCl (0.1 M). The resulting solution was stirred for 5 min at room
temperature and 0.072 mmol of 4-anilinoquinazoline derivative L3
was then added. The orange solution was stirred for 24 h, and the
yellow solid appeared in the solution was filtered off, washed with
ethanol and diethyl ether, and dried in vacuum to give rise to
complex 4 (10 mg, yield: 43%). ESI-MS (negative): m/z 643.1 for
4.2. Synthesis of ruthenium complexes 1e5
The ruthenium DMSO precursor complexes cis-RuIIC12(DMSO)4
and
[(DMSO)2H][trans-RuIIICl4(DMSO)2]
were
synthesised
[M ꢀ H]ꢀ (theoretical m/z 642.9). 1H NMR (DMF-d7)
(ppm): ꢀ13.17
d
following the procedures reported in the literature [55,56]. The 4-
anilinoquinazoline derivatives L0eL4 (Scheme 1) were syn-
thesised following the procedures described in our previous work
[33].
(very broad) for S-DMSO ligand. Anal. Calcd for C19H22Cl4N5O3R-
uS$H2O (F.W.: 661.9): C, 34.50; H, 3.66; N, 10.59. Found: C, 34.11; H,
3.81; N, 10.13. IR (cmꢀ1): 3452w, br, 1635vs, 1573m, 1512vs, 1437m,
1409m, 1280m, 1221m, 1065s, 1022s, 852m.
4.2.1. [RuIICl2(DMSO)2(L1)] (1)
4.2.5. [RuIIICl3(DMSO)(H-L4)] (5)
The 4-anilinoquinazoline derivative L1 (40.58 mg, 0.1 mmol)
was added to cis-RuIIC12(DMSO)4 (48.5 mg, 0.1 mmol) in dry
ethanol (10 mL), and the mixture was held at 385 K for 6 h. A good
quantity of precipitate formed, and the precipitate was filtered off,
washed with ethanol and ether, and dried in vacuum to give rise to
[RuIICl2(DMSO)2(L1)] (1) (40 mg, yield: 55%). ESI-MS (positive): m/z
736.2 for [M þ H]þ (theoretical m/z 736.0). 1H NMR (DMF-d7)
The precursor complex Na[trans-RuCl4(DMSO)2] (0.036 mmol)
was dissolved in absolute methanol (4 mL). The resulting solution
was stirred for
5 min at room temperature, and the 4-
anilinoquinazoline derivative L4 (0.036 mmol) in methanol was
then added. The resulting solution was stirred for 6 h, and the
precipitate appeared in the solution was filtered off, washed with
methanol and diethyl ether, and dried in vacuum to afford complex
5 (49% yield). ESI-MS (negative): m/z 634.1 for [M ꢀ H]ꢀ (theoretical
d(ppm): 8.58 (s, 1H), 8.33 (d, 1H), 8.31 (d, 1H), 7.95e7.92 (m, 1H),
m/z 633.9). 1H NMR (DMF-d7)
d
(ppm): ꢀ6.38 (broad), ꢀ8.36 (very
7.48e7.46 (m, 1H), 7.45e7.43 (m, 1H), 7.29 (m, 1H), 4.49 (2H, br s),
4.39 (1H, br s), 4.00 (s, 3H), 3.28e3.26 (m, 8H), 2.58 (s, 12H). 13C
broad) for S-DMSO ligand. Anal. Calcd for C21H22Cl3N4O4RuS$2H2O
(F.W.: 670.9): C, 37.65; H, 3.91; N, 8.36. Found: C, 38.27; H, 4.43; N,
7.70. IR (cmꢀ1): 3446w, br, 3066m, 3005m, 2941m, 2844m, 2634s,
1583s, 1512s, 1463m, 1436m, 1409m, 1381m, 1365m, 1309m,
1296m, 1279m, 1228s, 1071m, 1022s.
NMR (DMF-d7)
d (ppm): 157.1 (Ar-C), 155.7 (Ar-C), 153.6 (Ar-C),
149.2 (Ar-C), 148.4 (Ar-C), 138.1 (Ar-C), 123.9 (Ar-C), 122.6 (Ar-C),
119.9 (Ar-C), 117.1 (Ar-C), 117.0 (Ar-C), 109.9 (Ar-C), 108.2 (Ar-C),
103.8 (Ar-C), 68.6 (OCH2), 57.3 (OCH3), 56.4 (NHCH2), 50.6 (NHCH2),
46.1 (NH2CH2), 45.4 (SOCH3), 44.8 (SOCH3), 44.3 (SOCH3), 43.6
(SOCH3).
It is notable that the NMR characterisation of complexes 2e5
was not informative due to the paramagnetic effect of the RuIII ions.
However, one or two broadened 1signals of the two eCH3 groups in
the DMSO ligand of these complexes are distinguished, signifi-
cantly moving towards high field, and can be indicative of the
presence of asymmetric methyl groups in complexes 2 and 5 and
symmetric methyl groups in complexes 3 and 4.
X-ray diffraction-quality crystals were grown by slow diffusion
of diethyl ether into the DMSO/acetone (1:5) solutions of 1.
4.2.2. [RuIIICl3(DMSO)(L1)] (2)
The 4-anilinoquinazoline derivative L1 (40.58 mg, 0.1 mmol)
dissolved in 6 mL EtOH/MeCN, was added to [(DMSO)2H][trans-
RuCl4(DMSO)2] (55.6 mg, 0.1 mmol) in dry ethanol (4 mL), and the
mixture was stirred for 30 min at room temperature. Then 4 mL
H2O was added and the mixture was stirred for another 30 min. The
yellow solid, appeared in the solution, was filtered off, washed with
ethanol and diethyl ether, and dried in vacuum to afford
[RuIIICl3(DMSO)(L1)] (2) (43 mg, yield: 62%). ESI-MS (positive): m/z
693.1 for [M þ H]þ (theoretical m/z 693.0). 1H NMR (DMF-d7)
4.3. X-ray crystallography
Single-crystal X-ray diffraction study for [RuIICl2(DMSO)2(L1)]
(1) was carried out using graphite monochromated MoKa radiation
ꢀ
(l
¼ 0.71073 A) on a Rigaku Saturn 724 CCD area detector. All data
were collected at 173 K, and structure solution and refinement were
performed using the SHELXL-97. Standard data relating to the X-ray
crystal structure of complex 1 have been deposited in the
d
(ppm): ꢀ5.30 (very broad), ꢀ14.18 (very broad) for S-DMSO

118
L. Ji et al. / European Journal of Medicinal Chemistry 77 (2014) 110e120
Cambridge Crystallographic Data Centre with the CCDC deposition
number CCDC866471.
incubated at room temperature for 5 min, followed by addition of
25 L of ATP/substrate mixture, and then the resulting mixture was
incubated at 310 K for 1 h. The phosphorylation reaction was
m
4.4. High performance liquid chromatography (HPLC)
terminated by the addition of 50
pH 8).
mL/well stop buffer (50 mM EDTA,
An Agilent 1200 series quaternary pump and a Rheodyne
Each well of a microtitre plate was coated with 100
10 m
mL of
sample injector with a 20-
m
L loop, an Agilent 1200 series UVeVis
g mLꢀ1 streptavidin in carbonateebicarbonate buffer and
DAD detector and Chemstation data processing system were used
to separate the hydrolytic mixtures of complex 2. The HPLC analysis
was carried out on an Agilent Eclipse XDB-C18 reversed-phase
incubated overnight at 277 K, and then blocked with 1.5% bovine
serum albumin (BSA) in PBS/T (PBS solution contain 0.05% Tween-
20) at 310 K for 2 h, followed by three times of washing with PBS/T
column (150 ꢂ 4.6 mm, 5
m
M, USA). The mobile phases were water
prior to use. Then, aliquot (25
mixture and 75 L/well of D2O were added to the wells (in tripli-
cate) for incubation at 310 K for 1 h. Following three times of
washing with PBS/T, 100 L of primary antibody (Phospho-Tyrosine
Mouse mAb, 1:1000 in PBS/T with 1.5% BSA) was added to each well
and the plate was incubated at 310 K for another 1 h. The plate was
mL/well) of each enzymatic reaction
containing 0.1% TFA (solvent A), and acetonitrile containing 0.1%
TFA (solvent B). The gradient at 1.0 mL minꢀ1 was as follows: 10%
solvent B to 80% from 0 to 20 min, 80% from 20 to 22 min, then
resetting to 10% at 23 min.
m
m
4.5. Electrospray ionization mass spectrometry (ESI-MS)
again washed three times with PBS/T, and then 100 mL of secondary
antibody (HRP-labelled Goat Anti-Mouse lgG, 1:1000 in PBS/T with
1.5% BSA) was added to each well for 1 h of incubation at 310 K,
followed by three times of washing with PBS/T. Finally, 100 mL of
TMB substrate system was added to each well and the plate was
incubated at 310 K for 15 min, and the reaction was stopped by
addition of 100 mL of 2 M H2SO4, and the plate was read on the ELISA
Negative- or positive-ion ESI mass spectra were obtained on a
Micromass Q-TOF spectrometer (Waters) equipped with a Mas-
slynx (ver. 4.0) data processing system for analysis and post pro-
cessing. For the online LC-ESI-MS assays, an Agilent 1200 system
was interfaced with the mass spectrometer, using the same column
and gradients as described above for the HPLC assays with a flow
rate of 1 mL/min and a splitting ratio of 1/10 into mass spectrom-
eter. The spray voltage and the cone voltage were 2.8e3.8 kV and
55e70 V, respectively. The desolvation temperature was 393 K and
the source temperature 373 K. Nitrogen was used as both cone gas
plate reader (SpectraMax M5 Molecular Devices Corporation).
4.9. Docking analysis
The docking analysis was performed, a fully automatic docking
tool available, running on Dual-core Intel(R) E5300 CPU 2.60 GHz,
RAM Memory 2 GB under the Windows XP system. The crystal
structure of the EGFR-erlotinib complex was collected from PDB
under code 1M17 [59]. All the hydrogen atoms were added to
define the correct configuration and tautomeric states. Then the
modelled structure was energy-minimised using AMBER7FF99
force field with distance dependent dielectric function and current
charges. The Powell energy minimisation algorithm was used for
the energy minimisation. After extracting the binding ligand erlo-
tinib, the structure corresponding to the constringent energy
gradient (0.05 kcal molꢀ1) was used for re-docking and scoring
calculations of erlotinib to check the accuracy of the Surflex-Dock
program. The 4-anilinoquinazoline-type analogues were then
separately docked into the binding pocket for docking-scoring
analysis.
and desolvation gas with a flow rate of 50 L hL1 and 500 L hL1
,
respectively. The collision energy was set up to 10 V. The spectra
were acquired in the range of 200e1200 m/z. The mass accuracy of
all measurements was within 0.1 m/z unit, and all m/z values are the
mass-to-charge ratios of the most abundant isotopomer for
observed ions.
4.6. Elemental analysis and NMR spectroscopy
Elemental analysis was carried out on a Flash EA 1112 element
analysis instrument (ThermoQuest). 1H NMR spectra were obtained
on a Bruker Avance 400 spectrometer (Germany).
4.7. Hydrolysis
The stock solution of complex 2 (8 mM) was prepared by dis-
solving the complex in DMSO. To initial the hydrolysis of complex 2,
4.10. In vitro antiproliferation assays
10 mL DMSO solution of complex 2 was mixed with 90 mL, and the
resulting mixture was incubated at 310 K. Then HPLC data were
recorded at four time intervals, 0.5, 1, 2 and 12 h. And the products
generated from the hydrolysis were identified by LC-ESI-MS.
The human breast cancer cell line MCF-7 was obtained from the
Centre for Cell Resource of Shanghai Institute for Biological Sci-
ences, Chinese Academy of Science. MCF-7 cells were maintained in
RPMI 1640 (Invitrogen, USA) media supplemented with 10% foetal
calf serum (HyClone, USA). On requested, an aliquot of 100 ng mLꢀ1
epidermal growth factor (Sigma, USA) was added into the media.
The cells were grown at 310 K in a humidified atmosphere con-
taining 5% CO2 for 2e3 days prior to screening experiments.
The IC50 values, this is the concentration of tested compounds
that inhibit 50% of cell growth of MCF-7 cell line were determined
using the sulforhodamine B (SRB) technique. Cells were plated at a
4.8. Kinase inhibition assay
Enzyme-linked immunosorbent assay (ELISA) was applied to
characterise the inhibition potency of ruthenium complexes 1e5
against EGFR. An aliquot (10 mL) of the enzyme solution was added
to 415 mL DTT kinase buffer which is consist of 5 mL DTT (1.25 M) and
1.25 mL 4 ꢂ HTScanÒ Tyrosine Kinase Buffer (240 mM HEPES pH
7.5, 20 mM MgCl2, 20 mM MnCl2, 12
mM Na3VO4). Each tested
density of 6500 cells/well in 150 mL media in 96-well plates and
complex was dissolved in dimethylsulphoxide (DMSO) to give a
4 mM solution which was diluted with 0.05% Tween-20 in deion-
grew in the absence or the presence of EGF for 24 h. The stock
solutions (2 or 4 mM) of all tested complexes were made up fresh in
10% DMSO and saline before diluted down in media to give the
required concentration for addition to the cells. The final concen-
tration of DMSO in media was 0.5%. Cells were then exposed to each
tested complex at various concentrations for 48 h and cell growth
measured using SRB assay following the procedure reported by
ised water to give a 40
The ATP/peptide mixture was prepared by addition of 10
10 mM ATP to 125 L of 6 M substrate peptide, and then diluted
with D2O to 250 L. An aliquot (12.5 L) of solution of a screened
complex was mixed with as-prepared EGFR solution (12.5 L) and
mM solution.
mL of
m
m
m
m
m

L. Ji et al. / European Journal of Medicinal Chemistry 77 (2014) 110e120
119
Skehan et al. [60] Briefly, after 48 h exposure to the tested com-
pounds, cells were fixed with 50 L of cold trichloroacetic acid
Appendix A. Supplementary data
m
(50%) per well (96-well plate) for 60 min at 310 K. After washed five
times with tap water, the cells were stained for 30 min at 298 K
with 0.5% acetic acid containing 0.4% SRB (Sigma). Then, each plate
was rinsed five times with 1% acetic acid and allowed to air dry. The
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.02.062.
References
resulting coloured residue was dissolved in 200
mL of Tris base
[1] B. Rosenber, L. Vancamp, T. Krigas, Nature 205 (1965) 698e699.
[2] Y.W. Jung, S.J. Lippard, Chemical Reviews 107 (2007) 1387e1407.
[3] P.C.A. Bruijnincx, P.J. Sadler, Current Opinion in Chemical Biology 12 (2008)
197e206.
(10 mM) and optical density (OD) value for each well was measured
using a microplate reader (SpectraMax M5 Molecular Devices
Corporation) at the wavelength of 570 nm. The inhibition rate (IR)
was calculated based on the equation as follow:
[4] L. Kelland, Nature Reviews Cancer 7 (2007) 573e584.
[5] P.C.A. Bruijnincx, P.J. Sadler, Advances in Inorganic Chemistry 61 (2009) 1e62.
[6] P.J. Dyson, G. Sava, Dalton Transactions (2006) 1929e1933.
[7] T.W. Hambley, Dalton Transactions (2007) 4929e4937.
[8] A. Levina, A. Mitra, P.A. Lay, Metallomics 1 (2009) 458e470.
[9] E. Alessio, G. Mestroni, A. Bergamo, G. Sava, Current Topics in Medicinal
Chemistry 4 (2004) 1525e1535.
h
ꢀ
ꢁ.
i
IRð%Þ ¼ 1 ꢀ ODcompound ꢀ ODblank ðODcontrol ꢀ ODblank
ꢂ 100%
Þ
[10] A. Bergamo, B. Gava, E. Alessio, G. Mestroni, B. Serli, M. Cocchietto, S. Zorzet,
G. Sava, International Journal of Oncology 21 (2002) 1331e1338.
[11] C.G. Hartinger, S. Zorbas-Seifried, M.A. Jakupec, B. Kynast, H. Zorbas,
B.K. Keppler, Journal of Inorganic Biochemistry 100 (2006) 891e904.
[12] M.A. Jakupec, M. Galanski, V.B. Arion, C.G. Hartinger, B.K. Keppler, Dalton
Transactions (2008) 183e194.
[13] C.G. Hartinger, P.J. Dyson, Chemical Society Reviews 38 (2009) 391e401.
[14] E. Meggers, Current Opinion in Chemical Biology 11 (2007) 287e292.
[15] E. Meggers, G.E. Atilla-Gokcumen, H. Bregman, J. Maksimoska, S.P. Mulcahy,
N. Pagano, D.S. Williams, Synlett (2007) 1177e1189.
All reported values were averages of three independent exper-
iments and expressed as mean ꢃ SD (standard deviation).
The human embryonic kidney 293 cells (HEK293) was gifted by
Professor Dihua Shangguan at the Institute of Chemistry, Chinese
Academy of Science. The IC50 values of complex 2 on the growth of
HEK293 in the absence and in the presence of EGF were measured
by the analogical method described above.
[16] T. Storr, K.H. Thompson, C. Orvig, Chemical Society Reviews 35 (2006) 534e
544.
4.11. Fluorescence microscopy
[17] F.Y. Wang, A. Habtemariam, E.P.L. van der Geer, R. Fernandez, M. Melchart,
R.J. Deeth, R. Aird, S. Guichard, F.P.A. Fabbiani, P. Lozano-Casal, I.D.H. Oswald,
D.I. Jodrell, S. Parsons, P.J. Sadler, Proceedings of the National Academy of
Sciences of the United States of America 102 (2005) 18269e18274.
[18] P. Sathyadevi, P. Krishnamoorthy, N.S.P. Bhuvanesh, P. Kalaiselvi, V.V. Padma,
N. Dharmaraj, European Journal of Medicinal Chemistry 55 (2012) 420e431.
[19] H.J. Yu, Y. Chen, L. Yu, Z.F. Hao, L.H. Zhou, European Journal of Medicinal
Chemistry 55 (2012) 146e154.
[20] H.M. Chen, J.A. Parkinson, R.E. Morris, P.J. Sadler, Journal of the American
Chemical Society 125 (2003) 173e186.
[21] H.K. Liu, S.J. Berners-Price, F.Y. Wang, J.A. Parkinson, J.J. Xu, J. Bella, P.J. Sadler,
Angewandte Chemie International Edition 45 (2006) 8153e8156.
[22] F.Y. Wang, J.J. Xu, A. Habtemariam, J. Bella, P.J. Sadler, Journal of the American
Chemical Society 127 (2005) 17734e17743.
Induction of apoptosis in human breast cancer cell line MCF-7
was evaluated by confocal microscope after DAPI (40, 6-
diamidino-2-phenylindole) staining. Briefly, 1 ꢂ 105 cells per well
were seeded on cover glasses in a 24-well plate and allowed to
attach for 16 h at 310 K. Cells were then treated with the tested
compounds at 310 K for 24 h. After removing the supernatant and
washing the cells with PBS three times, the cells were fixed with the
pre-cool methanol for 15 min. After that, the cells were treated with
1 m
g mLꢀ1 DAPI (Sigma) in deionised water in dark. Washing the
cells with PBS, the cover glasses were then mounted on microscopy
slides with glycerol. Fluorescence images were obtained on a Carl
Zeiss LSM 510 META confocal microscope (Oberkochen, Germany)
at excitation wavelength of 790 nm and emission wavelength of
461 nm.
[23] F. Kratz, M. Hartmann, B. Keppler, L. Messori, Journal of Biological Chemistry
269 (1994) 2581e2588.
[24] B. Biersack, M. Zoldakova, K. Effenberger, R. Schobert, European Journal of
Medicinal Chemistry 45 (2010) 1972e1975.
[25] A. Castonguay, C. Doucet, M. Juhas, D. Maysinger, Journal of Medicinal
Chemistry 55 (2012) 8799e8806.
[26] W. Ginzinger, G. Muhlgassner, V.B. Arion, M.A. Jakupec, A. Roller, M. Galanski,
M. Reithofer, W. Berger, B.K. Keppler, Journal of Medicinal Chemistry 55
(2012) 3398e3413.
4.12. Annexin v/propidium iodide (PI) double staining assay
[27] A. Kurzwernhart, W. Kandioller, C. Bartel, S. Bachler, R. Trondl,
G. Muhlgassner, M.A. Jakupec, V.B. Arion, D. Marko, B.K. Keppler,
C.G. Hartinger, Chemical Communications 48 (2012) 4839e4841.
[28] X.Y. Wei, L. Qi, G.L. Yang, F.Y. Wang, Talanta 79 (2009) 739e745.
[29] H. Bregman, P.J. Carroll, E. Meggers, Journal of the American Chemical Society
128 (2006) 877e884.
To further verify the induction of apoptosis by the synthesised
ruthenium complexes, MCF-7 cells were seeded in a density of
2 ꢂ 105 per well in a 6-well plate and allowed to attach for 16 h,
then the cells were exposed to each tested compounds, including
complex 2 and references, at 310 K for 24 h. The supernatant was
removed, and cells were detached by trypsinization after washing
by PBS. The cells were transferred to FACS tubes after washing by
PBS and centrifuged at 1000 rpm for 3 min. After re-suspension in
[30] H. Bregman, D.S. Williams, G.E. Atilla, P.J. Carroll, E. Meggers, Journal of the
American Chemical Society 126 (2004) 13594e13595.
[31] E. Meggers, Angewandte Chemie International Edition 50 (2011) 2442e2448.
[32] J.E. Debreczeni, A.N. Bullock, G.E. Atilla, D.S. Williams, H. Bregman, S. Knapp,
E. Meggers, Angewandte Chemie International Edition 45 (2006) 1580e1585.
[33] S. Lü, W. Zheng, L.Y. Ji, Q. Luo, X. Hao, X.C. Li, F.Y. Wang, European Journal of
Medicinal Chemistry 61 (2013) 84e94.
0.5 mL binding buffer, the cells were incubated with 5
mL Annexin-V
[34] W. Zheng, Q. Luo, Y. Lin, Y. Zhao, X.L. Wang, Z.F. Du, X. Hao, Y. Yang, S. Lü,
L.Y. Ji, X.C. Li, L. Yang, F.Y. Wang, Chemical Communications (2013), http://
dx.doi.org/10.1039/c1033cc43000f.
conjugate for 5 min, followed by addition of 5 L PI prior to the FACS
m
analysis. The FACS assays were preformed on a Calibur flow cy-
tometer (BD, Franklin Lakes, New Jersey, US), of which the FL1
channel was used to record the intensity of annexin V-FITC staining
and FL2 channel to record the intensity of PI staining. The data were
quantified by Sell Quest software (BD, Franklin Lakes, New Jersey,
US).
[35] J. Baselga, Science 312 (2006) 1175e1178.
[36] M. Calligaris, O. Carugo, Coordination Chemistry Reviews 153 (1996) 83e154.
[37] J.G. Liu, Q.L. Zhang, X.F. Shi, L.N. Ji, Inorganic Chemistry 40 (2001) 5045e5050.
[38] F.R. Pavan, G.V. Poelhsitz, M.I.F. Barbosa, S.R.A. Leite, A.A. Batista, J. Ellena,
L.S. Sato, S.G. Franzblau, V. Moreno, D. Gambino, C.Q.F. Leite, European Journal
of Medicinal Chemistry 46 (2011) 5099e5107.
[39] F. Wang, H.M. Chen, S. Parsons, L.D.H. Oswald, J.E. Davidson, P.J. Sadler,
Chemistry e A European Journal 9 (2003) 5810e5820.
[40] B. Cebrian-Losantos, E. Reisner, C.R. Kowol, A. Roller, S. Shova, V.B. Arion,
B.K. Keppler, Inorganic Chemistry 47 (2008) 6513e6523.
Acknowledgement
[41] I. Berger, M. Hanif, A.A. Nazarov, C.G. Hartinger, R.O. John, M.L. Kuznetsov,
M. Groessl, F. Schmitt, O. Zava, F. Biba, V.B. Arion, M. Galanski, M.A. Jakupec,
L. Juillerat-Jeanneret, P.J. Dyson, B.K. Keppler, Chemistry e A European Journal
14 (2008) 9046e9057.
We thank NSFC (Grant Nos: 21020102039, 21135006, 21127901,
21275148 (Q. L.) and 21321003), the 973 Program of MOST
(2013CB531805) for financial support.

120
L. Ji et al. / European Journal of Medicinal Chemistry 77 (2014) 110e120
[42] C. Gossens, A. Dorcier, P.J. Dyson, U. Rothlisberger, Organometallics 26 (2007)
3969e3975.
[43] A.F.A. Peacock, A. Habtemariam, R. Fernandez, V. Walland, F.P.A. Fabbiani,
S. Parsons, R.E. Aird, D.I. Jodrell, P.J. Sadler, Journal of the American Chemical
Society 128 (2006) 1739e1748.
[44] M. Groessl, C.G. Hartinger, P.J. Dyson, B.K. Keppler, Journal of Inorganic
Biochemistry 102 (2008) 1060e1065.
[45] A.V. Vargiu, A. Robertazzi, A. Magistrato, P. Ruggerone, P. Carloni, The Journal
of Physical Chemistry B 112 (2008) 4401e4409.
[51] R.E. Aird, J. Cummings, A.A. Ritchie, M. Muir, R.E. Morris, H. Chen, P.J. Sadler,
D.I. Jodrell, British Journal of Cancer 86 (2002) 1652e1657.
[52] O. Novakova, J. Kasparkova, V. Bursova, C. Hofr, M. Vojtiskova, H.M. Chen,
P.J. Sadler, V. Brabec, Chemistry & Biology 12 (2005) 121e129.
[53] M. Muhsin, J. Graham, P. Kirkpatrick, Nature Reviews Drug Discovery 2 (2003)
515e516.
[54] A.E. Wakeling, S.P. Guy, J.R. Woodburn, S.E. Ashton, B.J. Curry, A.J. Barker,
K.H. Gibson, Cancer Research 62 (2002) 5749e5754.
[55] E. Alessio, G. Balducci, M. Calligaris, G. Costa, W.M. Attia, G. Mestroni, Inor-
ganic Chemistry 30 (1991) 609e618.
[46] A.H. Velders, A. Bergamo, E. Alessio, E. Zangrando, J.G. Haasnoot, C. Casarsa,
M. Cocchietto, S. Zorzet, G. Sava, Journal of Medicinal Chemistry 47 (2004)
1110e1121.
[56] I.P. Evans, A. Spencer, G. Wilkinson, Journal of the Chemical Society Dalton
Transactions (1973) 204e209.
[47] M. Bacac, A.C.G. Hotze, K. van der Schilden, J.G. Haasnoot, S. Pacor, E. Alessio,
G. Sava, J. Reedijk, Journal of Inorganic Biochemistry 98 (2004) 402e412.
[48] K.H. Gibson, W. Grundy, A.A. Godfrey, J.R. Woodburn, S.E. Ashton, B.J. Curry,
L. Scarlett, A.J. Barker, D.S. Brown, Bioorganic & Medicinal Chemistry Letters 7
(1997) 2723e2728.
[49] M. Ono, A. Hirata, T. Kometani, M. Miyagawa, S. Ueda, H. Kinoshita, T. Fujii,
M. Kuwano, Molecular Cancer Therapeutics 3 (2004) 465e472.
[50] K.B. Reddy, G.L. Mangold, A.K. Tandon, T. Yoneda, G.R. Mundy, A. Zilberstein,
C.K. Osborne, Cancer Research 52 (1992) 3636e3641.
[57] J.J. Fiol, A. Garcia-Raso, F.M. Alberti, A. Tasada, M. Barcelo-Oliver, A. Terron,
M.J. Prieto, V. Moreno, E. Molins, Polyhedron 27 (2008) 2851e2858.
[58] A.Garcia-Raso,J.J.Fiol, A. Tasada, M.J.Prieto,V. Moreno, I.Mata, E.Molins,T.Bunic,
A. Golobic, I. Turel, Inorganic Chemistry Communications 8 (2005) 800e804.
[59] J. Stamos, M.X. Sliwkowski, C. Eigenbrot, Journal of Biological Chemistry 277
(2002) 46265e46272.
[60] P. Skehan, R. Storeng, D. Scudiero, A. Monks, J. Mcmahon, D. Vistica,
J.T. Warren, H. Bokesch, S. Kenney, M.R. Boyd, Journal of the National Cancer
Institute 82 (1990) 1107e1112.