Alkynyl Gold(I) complexes derived from 3-hydroxyflavones as multi-targeted drugs against colon cancer

Article abstract of DOI:10.1016/j.ejmech.2019.111661

The design of multi-targeted drugs has gained considerable interest in the last decade thanks to their advantages in the treatment of different diseases, including cancer. The simultaneous inhibition of selected targets from cancerous cells to induce their death represents an attractive objective for the medicinal chemist in order to enhance the efficiency of chemotherapy. In the present work, several alkynyl gold(I) phosphane complexes derived from 3-hydroxyflavones active against three human cancer cell lines, colorectal adenocarcinoma Caco-2/TC7, breast adenocarcinoma MCF-7 and hepatocellular carcinoma HepG2, have been synthesized and characterized. Moreover, these compounds display high selective index values towards differentiated Caco-2 cells, which are considered as a model of non-cancerous cells. The antiproliferative effect of the most active complexes [Au(L2b)PPh₃] (3b) and [Au(L2c)PTA] (4c) on Caco-2 cells, seems to be mediated by the inhibition of the enzyme cyclooxygenase-1/2 and alteration of the activities of the redox enzymes thioredoxin reductase and glutathione reductase. Both complexes triggered cell death by apoptosis, alterations in cell cycle progression and increased of ROS production. These results provide support for the suggestion that multi-targeting approach involving the interaction with cyclooxygenase-1/2 and the redox enzymes that increases ROS production, enhances cell death in vitro. All these results indicate that complexes [Au(L2b)PPh₃] and [Au(L2c)PTA] are promising antiproliferative agents for further anticancer drug development.

Full text of DOI:10.1016/j.ejmech.2019.111661

European Journal of Medicinal Chemistry 183 (2019) 111661
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Alkynyl Gold(I) complexes derived from 3-hydroxyflavones as
multi-targeted drugs against colon cancer
a
,
b
a
b
a
ꢀ
ꢀ
ꢀ
Ines Marmol , Pilar Castellnou , Raquel Alvarez , M. Concepcion Gimeno ,
M. Jesús Rodríguez-Yoldi b **, Elena Cerrada a
Departamento de Química Inorganica, Instituto de Síntesis Química y Catalisis Homogenea-ISQCH, Universidad de Zaragoza-C.S.I.C., 50009, Zaragoza,
,
, *
a
ꢀ
ꢀ
ꢀ
Spain
b
ꢀ
Departamento de Farmacología y Fisiología, Unidad de Fisiología, Universidad de Zaragoza, CIBERobn, IIS Aragon, IA2, 50013, Zaragoza, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The design of multi-targeted drugs has gained considerable interest in the last decade thanks to their
advantages in the treatment of different diseases, including cancer. The simultaneous inhibition of
selected targets from cancerous cells to induce their death represents an attractive objective for the
medicinal chemist in order to enhance the efficiency of chemotherapy. In the present work, several
alkynyl gold(I) phosphane complexes derived from 3-hydroxyflavones active against three human cancer
cell lines, colorectal adenocarcinoma Caco-2/TC7, breast adenocarcinoma MCF-7 and hepatocellular
carcinoma HepG2, have been synthesized and characterized. Moreover, these compounds display high
selective index values towards differentiated Caco-2 cells, which are considered as a model of non-
cancerous cells. The antiproliferative effect of the most active complexes [Au(L2b)PPh3] (3b) and
[Au(L2c)PTA] (4c) on Caco-2 cells, seems to be mediated by the inhibition of the enzyme cyclooxygenase-
1/2 and alteration of the activities of the redox enzymes thioredoxin reductase and glutathione reduc-
tase. Both complexes triggered cell death by apoptosis, alterations in cell cycle progression and increased
of ROS production. These results provide support for the suggestion that multi-targeting approach
involving the interaction with cyclooxygenase-1/2 and the redox enzymes that increases ROS production,
enhances cell death in vitro. All these results indicate that complexes [Au(L2b)PPh3] and [Au(L2c)PTA] are
promising antiproliferative agents for further anticancer drug development.
Received 14 June 2019
Received in revised form
20 August 2019
Accepted 29 August 2019
Available online 7 September 2019
Keywords:
Alkynyl
Gold complexes
Hidroxyflavones
Thioredoxin reductase
COX
ROS
Multi-target
© 2019 Elsevier Masson SAS. All rights reserved.
1. Introduction
growing up during the last decades.
On this regard, gold derivatives and particularly auranofin
Cancer chemotherapy based on metallodrugs has been used
worldwide, mainly in the case of platinum derivatives, which are
active against many tumor types [1]. The activity of platinum-based
derivatives (cisplatin, carboplatin, oxaliplatin, etc) is mediated by
their ability to form DNA adducts leading to DNA lesions that
interfere with transcription and results in cellular death [2].
Regardless of their success in chemotherapy, platinum derivatives
produce undesirable side effects and/or acquired resistance that
limits their effectiveness [3]. Consequently, the development of
novel compounds with metals other than platinum has been
(2,3,4,6-tetraacetyl-1-thio-b-D-glucopyranosato-S-triethylphos-
phane gold(I)), have been essayed as biological active products in
cancer, neurodegenerative disorders, acquired immunodeficiency
syndrome, parasitic and bacterial infection [4]. Moreover, auranofin
and related derivatives can reverse important aspects of immune
evasion [5]. Additionally, auranofin was approved by FDA (Food and
Drug Administration) for the treatment of arthritis rheumatoid in
1985. Consequently, gold compounds could be considered as an
alternative to platinum complexes.
Drugs targeting DNA, which is ubiquitously present in cancerous
and healthy cells, tend to be more toxic than derivatives with
cancer-specific targets. Accordingly, other pharmacological targets
of transition metal agents have been investigated, supporting
several proteins as important binding partners [6,7]. Most of the
lately approved organic drugs target specific proteins over-
* Corresponding author.
** Corresponding author.
E-mail addresses: mjrodyol@unizar.es (M.J. Rodríguez-Yoldi), ecerrada@unizar.
es (E. Cerrada).
https://doi.org/10.1016/j.ejmech.2019.111661
0223-5234/© 2019 Elsevier Masson SAS. All rights reserved.

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
2
expressed or unique to cancer cells [8]. Likewise, the mechanism of
gold derivatives is implicated with enzyme inhibition. Thus, aur-
anofin induces apoptosis in cisplatin-resistant cell lines [9],
showing a different mechanism of action than cisplatin. Its anti-
cancer activity has been mainly attributed to the inhibition of the
enzyme thioredoxin reductase (TrxR) [9e11], which is overex-
pressed in many cancer cells. Additionally, a second mechanism
based in the inhibition of the ubiquitin-proteasome pathway has
been also endorsed to auranofin [12,13]. Interestingly, this pathway
is upregulated in many cancers and several cancer cells have
dysfunctional ubiquitin-proteasome system with increased pro-
teasome activity [14]. Thus, gold complexes have been proposed to
act as multi-target drugs, which might enhance their clinical effect
in comparison to classic platinum-based chemotherapeutic drugs.
Multi-targeted compounds are emerging as a therapeutic approach
for a variety of cancers [15]. They are able to act on a diverse set of
regulatory pathways with a subsequent reduction of systemic
toxicities, even addressing resistance issues.
3-Hydroxyflavones (Fig. 1), also named flavonols are together
with flavanols the most common type of flavonoids, being also
widely distributed in fruits and vegetables [16,17]. Quercetin and
kaempferol are the most studied and most abundant [18e21].
Many beneficial properties have been endorsed to both that include
anti-oxidative activity, anti-inflammatory effects, ameliorative ef-
fects against different diseases such as atherosclerosis, thrombosis
and arrhythmia, in addition to anticancer properties [22e25]. Their
antioxidant power is one of the most known biological properties,
being described by several mechanism pathways that includes: ROS
inhibition, leukocyte immobilization, nitric oxide inhibition or in-
hibition of xanthine oxidase [26]. The anti-inflammatory and
anticancer properties of phytocompounds seems to be closely
related since both have in common the inhibition of the enzyme
cyclooxygenase (COX) [27]. COX enzymes are involved in prosta-
glandin synthesis from arachidonic acid and play a critical role in
inflammation and therefore might be related to tumorigenesis.
Researchers have traditionally focused on the inhibition of the
isoform COX-2, since it is overexpressed on many tumour cells.
However, inhibition of the basal isoform COX-1 has also success-
fully decreased cancer cells proliferation [28].
development of new metallodrugs.
Thus, we report herein the synthesis of new flavone-derived
ligands with a propargyl ether group, the corresponding phos-
phane gold(I) derivatives and their evaluation against a panel of
cancer cells and a model of the intestinal barrier (differentiated
Caco-2 cells). To support the hypothesis of multi-target compounds
we have studied the effect of some of the new derivatives towards
the activity of redox enzymes (thioredoxin reductase and gluta-
thione reductase) along with COX-1/2 enzyme. Measurement of
reactive oxygen species (ROS) levels and type of cell death triggered
by gold complexes were also investigated.
2. Results and discussion
2.1. Synthesis of ligands
The synthesis of 3-hydroxyflavones (HL-1a-d) was performed in
one step, similarly to the previously published method by Ozturk
et al. [37]. This easy method avoids the isolation of 2-
hydroxychalcones, obtained as intermediates, after a Claisen-
Schmidt condensation between 2-hydroxyacetophenone and the
corresponding benzaldehyde. Flavone-derived ligands (HL-2a-d)
were synthesized by the base-mediated reaction between 3-
hydroxyflavones and propargyl bromide (Scheme 1). The forma-
tion of the flavone with the propargyl ether group was evidenced
by the appearance of a doublet in the 1H NMR spectra characteristic
of the methylene unit at around 5 ppm and a triplet for the acet-
ylenic proton at 3.5 ppm (JH-H z 2.4 Hz). As occurred in the 3-
hydroxyflavones HL-1a-d used as precursors, there is a significant
difference in the displacements assigned for H3’,5’ (B ring) in their
1H NMR spectra (see experimental). Thus, electron donating
groups, such as eOMe, lead to upfield displaced signals while
electron withdrawing units drive to downfield displacements.
The crystalline structure of HL-2d has been established by X-ray
analysis (Fig. 2). The molecule of the flavone presents its typical
structure with the aromatic rings in the same plane. The single CeO
distances are very dissimilar, the oxygen inside the ring show bond
lengths of O3eC11 1.3676(16) and O3eC12 1.3737(16) Å, whereas
the oxygen bonded to the propargyl group shows bond lengths of
O1eC4 1.3776(17) and O1eC3 1.4524(17) Å, the latter being longer
than those to the sp2 carbon atoms. In a similar manner O4 display
distances of O4eC16 1.3645(17) and O4eC19 1.4365(17) Å. The C¼O
bond distance is O2eC5 1.2366(17) Å, which is typical of a double
bond. The propargyl group present CeC distances of C2eC3
1.470(2) and C1-C2 1.186(2) Å, which are typical for single and triple
CeC bonds respectively. In the crystal the formation of several
The presence of the hydroxyl group in C3 position and the keto
motif in C4 position enable easy coordination to metal complexes
leading to new complexes with biological properties [29e33]. Be-
sides, the eOH group can be functionalized leading to new flavone-
derived ligands with a propargyl ether group able to metal coor-
dination [34]. With this regard and in the particular case of gold,
alkynyl derivatives with anticancer properties are less represented
in comparison with other organometallic gold complexes, such as
N-heterocyclic carbene or cyclometalated compounds [35]. How-
ever, the remarkable stability exerted by alkynyl gold species even
under physiological conditions, due to the high bond dissociation
energy around the gold centre [36], makes them attractive for the
secondary bonds such as O/H hydrogen bonds or
observed, which give rise to a three dimensional network structure.
The shortest are O4/HeC of 2.718 Å or a slipped
3.665 Å.
p-p stacking is
p-p stacking of
2.2. Synthesis of gold complexes
Treatment of the alkyne ligands HL-2a-d with [AuCl(PR3)]
(PR3 ¼ PPh3 and PTA (PTA ¼ 1,3,5-triaza-7-phosphaadamantane))
in the presence of sodium methoxide afforded the corresponding
alkynyl phosphane gold(I) derivatives 3a-d and 4a-d (Scheme 1) as
air-stable solids. The disappearance of the acetylenic triplet and the
downfield shift of the methylene signal in their 1H NMR spectra are
indicative of gold coordination. The spectra of complexes 3a-
d display a more complicated pattern in comparison to their pre-
cursors, as a consequence of the presence of the PPh3 in the aro-
matic region. Similar spectroscopic data are observed for
complexes 4a-d to those of their precursors, in addition to the AB
system and a singlet at 4.2 ppm characteristics of the NCH2N and
Fig. 1. 3-Hydroxyflavone structure.

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
3
Scheme 1. Preparation of flavones and their gold complexes.
i) NaOH and H2O2, ii) K2CO3 and propargyl bromide, iii) KOH/MeOH and [AuClPR03].
Fig. 2. Crystal structure of compound HL-2d and part of the three-dimensional network through hydrogen bonding.
NCH2P moieties of the PTA molecule respectively. The 31P{1H} NMR
spectra show one resonance at around þ41 ppm in the case of the
triphenylphosphane derivatives and at about ꢀ49 ppm for PTA
complexes. In addition, their infrared spectra also display changes
triphenylphosphane, as similar absorption bands are also observed
in related Au(PPh3) complexes [38]. In all the complexes the tran-
sitions remain without any changes in shape or displacement in the
absorbance maximum (none apparent red- or blue shift), in addi-
tion to lacking of absorbance at around 500 nm (characteristic of
gold reduction) over 24 h, implying a substantial stability of the
chromophore under physiological conditions.
in the n(C≡C) vibration in comparison to the alkyne flavones. The
free ligands exhibit a stretching vibration of C≡C as a weak band at
2110 cmꢀ1, which is slightly displaced to around 2130 cmꢀ1 in the
gold complexes, and a narrow band around 3200 cmꢀ1 due to the
terminal alkyne (≡CeH) stretch, which disappears once coordi-
nated to the gold phosphane units.
2.4. Lipophilicity
On the basis of Lipinski's rule, lipophilicity and hydrophilicity
are essential on the performance of drug research. High lip-
ophilicity or high hydrophilicity can hinder biomedical applications
[39]. Consequently, a balanced relationship between lipophilicity
and hydrophilicity would be important in drug delivery process.
The balance between lipophilicity and hydrophilicity can be
measured by the partition coefficient water/n-octanol, logP, by
experimental procedures such as the shake flask method, which
studies the distribution of the drug between an equal amount of n-
octanol and water (or buffered aqueous solution) [40]. Our alkynyl
gold(I) derivatives display values of logP (at pH ¼ 7.4) in the range
0.23e1.2 (Table 1), being higher in the cases of the triphenylphos-
phane compounds that implies higher lipophilicity in comparison
with the PTA counterparts.
2.3. Solution stability
The stability of the complexes was analysed by UVevis ab-
sorption spectroscopy in PBS solution (PBS buffer at pH ¼ 7.4). So-
lutions suitable for spectrophotometric analysis were prepared by
diluting dimethylsulfoxide (DMSO) mother solutions of the com-
plexes with PBS buffer. The resulting solutions were monitored
over 24 h at 37 ꢁC. The spectra of the complexes show intense
absorption bands at ca. 210 nm and two lower energy absorption
bands with low intensity in the region of 250e320 nm, which could
be assigned as p/p* intraligand transitions of the alkynyl ligands
(Fig. S38). The higher energy absorption band around 210 nm is also
tentatively assigned to the intraligand transition characteristic of

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
4
Table 1
Volmer equation (see supplementary material) and the corre-
IC50 (m
M)a values of 3-hydroxyflavones and flavone-derived ligands on undifferen-
sponding plot F0/F vs [complex] (F and F0 are the corresponding
intensities in the presence and the absence of the quencher agent)
gives a linear plot (Fig. 4A and Fig. S40) in the cases of complexes 3a,
3b and 4b, characteristic of the presence of a single mechanism of
quenching, static or dynamic [43]. However, complexes 3c, 4a, 4c
and 4d give a plot with a positive deviation (concave towards the y
axis), which suggests the presence of a combination of static and
dynamic quenching by the fluorophore.
tiated Caco-2/TC7.
Compound
Assay method
MTT
SRB
HL-1a
HL-1b
HL-1c
HL-1d
HL-2a
HL-2b
HL-2c
HL-2d
89.1 4.98
22.0 0.25
16.5 0.69
22.3 2.51
63.56 17.87
>100
16.7 8.28
10.4 0.24
13.5 0.33
14.8 2.82
54.87 2.05
65.64 38.92
27.19 3.02
49.07 3.84
The quenching of the emission can be quantified only in the
cases of linear representation of the plot F0/F vs [complex]. Thus,
complexes 3a, 3b and 4b afford a quenching constant value (Ksv) in
the order of 1x104 M-1 (Table 2). The value of the bimolecular
quenching constant, kq, which represents the efficiency of
quenching or how accessible the fluorophores are to the quencher,
can afford an idea of the type of quenching mechanism. Thus, a kq
value higher than the diffusion-controlled rate constant of the
39.09 30.98
64.14 13.95
a
Mean SE of at least three determinations.
Flavonoids are highly lipophilic compounds, being therefore
difficult their oral bioavailability. Consequently, alternative routes
of administration have been explored for their delivery, such as the
use of nanoparticles, lipid nanocapsules, microparticles or micro-
sponges as delivery systems [41]. The functionalisation of the
flavone with the propargyl ether moiety and the subsequent co-
ordination of the gold-phosphane unit, has resulted in a decrease of
their lipophilic character and a balanced relationship between
lipophilicity and hydrophilicity.
biomolecule in water, 1010
M
ꢀ1sꢀ1, could indicate some type of
binding interaction implying a possible contribution of static
quenching, since a dynamic quenching depends on diffusion. kq
calculated (Ksv/
t0 by using the standard value of 10ꢀ8 s for t0) in the
case of complexes 3a, 3b and 4b afford a value of ca. 1012, pointing
to a static quenching rather than dynamic quenching.
Analysis of the absorption spectra of the fluorophores can also
be useful to characterize static and dynamic quenching. In view of
implication of complex formation in a static quenching, changes in
the absorption spectra of BSA could be expected; whereas these
spectra would remain unchanged in a dynamic quenching, since
only excited states of BSA are affected. BSA displays two absorption
2.5. Bovine serum albumin interaction
Considering that the majority of chemotherapeutic treatments
use bloodstream for drug transport and that blood plasma proteins
are able to interact with the drug, we have studied the interaction
of some of the new complexes with albumin.
peaks at around 210 nm, which represents the content of
a-helix in
the protein and at 280 nm, which arises from
aromatic amino acid residues [44].
p/p* transition of
Addition of increasing amounts of [Au(L2b)PTA] (4b) to a solu-
tion of BSA (10 M in PBS) (Fig. 5) resulted in a decrease of the
absorbance intensity at 280 nm (saturation taking place beyond
40 M) that suggest an interaction of the complex and the protein,
probably as a consequence of a conformational change. Similar
results have been described with other drugs [45].
Dynamic quenching does not affect the absorption of quenching
molecule as only affects its excited states, consequently the changes
observed in the spectrum could be indicative of static quenching.
The absorption spectra of the rest of the complexes in a 1:1 stoi-
chiometry gave also a decrease of the BSA absorption, in accordance
with the presence of a static quenching mechanism (see supple-
mentary material).
Bovine serum albumin (BSA) is frequently used as a model
protein for HSA (human serum albumin) due to their similarity and
commercial availability [42].
The interaction of the alkynyl gold derivatives with BSA has
been studied by the addition of increasing amounts of the com-
plexes to a solution of BSA followed by the measurement of the
quenching of BSA fluorescence emission, due to the tryptophan
residues, upon excitation at 295 nm. No emission was observed for
the gold compounds in the range 310e400 nm.
A concentration dependent quenching of the fluorescence is
observed, lacking changes in shape and a slight displacement in the
position of the maximum (ca. 6e8 nm) has been observed (see
Fig. 3 as example). Fluorescence data were analysed by the Stern-
m
m
Fluorescence intensity data has allowed us to calculate the
values of the binding constant for each experiment and the number
of binding sites by means of the representation of log (F0eF/F)
versus log[complex] (Fig. 4B and Fig. S41). The binding constant
values (summarized in Table 1) are in the range 2.4 x 104 to 2.3 x
106 Mꢀ1. These values are in agreement with those observed for
dithiocarbamate gold(III) derivatives [46] and for [Au(C≡CPh)(PTA)]
[47], although higher than other alkynyl or thiolate phosphane
gold(I) complexes previously reported by us [48e53].
The main molecule binding sites of BSA are located in the sub-
domain IIA and IIIA, commonly named as sites I and II [54]. Site I
shows affinity for warfarin, phenylbutazone, etc. meanwhile site II
is well suited for ibuprofen, diazepam, etc [55]. In order to localize
our derivatives, we have performed displacement experiments
with ibuprofen and phenylbutazone as site markers. Complex 4d,
with the highest binding constant value, was added in increasing
amounts to a mixture of BSA with ibuprofen or BSA with phenyl-
butazone and monitored the emission intensity quenching (see
Figs. S42eS46). The emission data were analysed similarly than the
above described experiments. As a result, the new binding constant
Fig. 3. Fluorescence emission spectra of BSA at 298 K in the presence of different
concentrations of [Au(L2a)PTA] (4a) Arrow indicates the increase of the quencher
concentration (0e30 mM).

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
5
Fig. 4. Stern-Volmer plots for the quenching of BSA with increasing amounts of PTA complexes (complex concentration from 0
m
M to 40
m
M) at 298 K
(
lexc ¼ 295 nm,
[BSA] ¼ 50
m
M). A) Stern-Volmer equation used: F0/F ¼ 1 þ Ksv[complex]. The slope of the best fit linear trend provides the Stern-Volmer quenching constant Ksv for PTA complexes.
B) Stern-Volmer equation used: log{(F0eF)/F} ¼ logKb þ nlog[complex]. The intercept of the best fit linear trend provides the Stern-Volmer quenching constant Kb.
Table 2
hand, polyphenols might trigger an increase in the activity of the
mitochondrial dehydrogenase enzyme, which would lead to an
enhanced reduction of MTT that not truly correlates with cell
proliferation. Additionally, polyphenols are able to reduce MTT to
formazan themselves, leading to false positive [56]. Therefore, al-
ternatives to MTT assay are recommended to the evaluation of the
antiproliferative effect of plant-derived molecules, including DNA-
based methods and the sulforhodamine B assay.
When comparing the IC50 values of the four 3-hydroxyflavones
HL1a-d obtained by SRB and MTT (Table 2), we found that SRB
method provided values 1.2 to 5 times lower than the MTT tech-
nique, which might be indicative of slight interference with the
second procedure. Furthermore, the obtained IC50 values of
flavone-derived ligands HL2a-d were significantly higher when
determined by MTT than with SRB; in addition, results obtained
with MTT assay were found to be less reproducible since higher
standard error values were found. We concluded that SRB was a
more reliable technique to evaluate the antiproliferative effect of 3-
hydroxyflavones and flavone-derived ligands. However, such in-
terferences with the MTT assay were not observed for the eight gold
Values of Stern-Volmer quenching constant (Ksv), the number of binding sites and
the apparent binding constant (Kb) for the interaction of complexes with BSA and
with BSA-Ibuprofen or BSA-Phenylbutazone for complex 4d.
System
Ksv (Mꢀ1
)
Kb (Mꢀ1
)
n
BSA-[Au(L2a)PPh3] (3a)
BSA-[Au(L2b)PPh3] (3b)
BSA-[Au(L2c)PPh3] (3c)
BSA-[Au(L2a)PTA] (4a)
BSA-[Au(L2b)PTA] (4b)
BSA-[Au(L2c)PTA] (4c)
BSA-[Au(L2d)PTA] (4d)
BSA-Ibuprofen-4d
3.05 x 104
3.74 x 105
1.41 x 105
1.22 x 105
5.19 x 105
2.45 x 104
4.95 x 105
2.35 x 106
4.47 x 104
2.04 x 106
1.24
1.08
1.14
1.25
1.03
1.23
1.34
1.02
1.37
5.67 x 104
e
e
1.77 x 104
e
e
e
e
BSA-Phenylbutazone-4d
derivatives at concentrations below 20 mM except for [Au(L2d)
PPh3] (3d) (Fig. S49). All complexes were tested by both assay
techniques on undifferentiated Caco-2/TC7 cells at a range of con-
centrations (from 1.25 to 20 mM). The resulting dose-response
curves and the IC50 values obtained with the SRB technique and
MTT assay (Tables S1 and S3 respectively) were comparable. We
hypothesized that the interaction between flavone-containing
molecules and the MTT assay -and the subsequent false positives-
were strongly influenced by concentration; thus, MTT was
considered as an adequate assay procedure when incubation with
gold derivatives was performed at concentration values below
Fig. 5. UVevisible absorption spectra of BSA with increasing amounts of [Au(L2b)PTA]
M] and [4b] (10e50 M). Equal concentration of 4b was also
added to the reference cell (in each measurement) in order to eliminate its absorbance.
(4b) at 298K. [BSA ¼ 10
m
m
20
mM.
Cytotoxicity of the eight novel gold derivatives was also evalu-
ated on breast adenocarcinoma MCF-7 and hepatocellular carci-
noma HepG2 in addition to colorectal adenocarcinoma Caco-2 (TC7
clone). To further determine the selectivity of the complexes,
differentiated Caco-2 cells were used as a model of the intestinal
barrier. As can be observed on Table 3, all complexes reduced cell
proliferation on each cancer cell line analysed and the obtained IC50
values were lower than the ones of 3-hydroxyflavones and flavone-
derived ligands. These results suggest that coordination of the gold
phosphane unit to flavones enhances their natural antiproliferative
effect and leads to more efficient chemotherapeutic drugs.
(Table 1) is affected in the presence of ibuprofen, since its value has
decreased remarkably, while the addition of phenylbutazone
maintained its value in the same order. This result indicates that
complex 4d competes with ibuprofen in subdomain IIA and its
binds in site I of BSA.
2.6. Biological studies
2.6.1. Analysis of the antiproliferative effect
Undifferentiated Caco-2/TC7 cells showed a better response to
treatment with all complexes than MCF-7 and HepG2 cells. The IC50
values obtained on undifferentiated Caco-2/TC7 and MCF-7 cells
were significantly lower when compared to cisplatin. Auranofin, on
the other hand, displayed a higher antiproliferative effect on the
three cancer cell lines than our compounds in most of the cases,
The toxicity of 3-hydroxyflavones and flavone-derived ligands
was firstly evaluated on undifferentiated Caco-2/TC7 cells by two
different methods: sulforhodamine B (SRB) and MTT (Table 2).
Some authors have reported that flavone-containing molecules
such as polyphenols might interfere with MTT assay. On the one

ꢀ
6
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
Table 3
Distribution coefficients and IC50
(
m
M)a values of the complexes on undifferentiated Caco-2/TC7, MCF-7 and HepG2 cancer cell lines and on differentiated Caco-2 cells (non-
cancerous model) compared with auranofin and cisplatin.
Compound
logD7.4
Caco-2/TC7
MCF-7
HepG2
Dif Caco-2
[Au(L2a)PPh3] (3a)
[Au(L2b)PPh3] (3b)
[Au(L2c)PPh3] (3c)
[Au(L2d)PPh3] (3d)
[Au(L2a)PTA] (4a)
[Au(L2b)PTA] (4b)
[Au(L2c)PTA] (4c)
[Au(L2d)PTA] (4d)
Cisplatin
0.99
0.92
1.17
1.2
0.78
0.80
0.23
0.44
ꢀ0.53
ꢀ2.53
3.81 1.18
1.52 0.91
5.34 0.05
4.78 0.62
7.68 1.74
6.42 0.35
2.33 1.26
5.22 0.43
37.24 5.15
1.80 0.1
2.08 0.17
13.87 0.78
8.99 3.89
3.40 0.85
18.49 0.90
8.80 3.14
7.57 0.08
9.19 2.89
41.82 0.07
0.77 0.05
32.34 4.27
3.38 0.07
34.14 5.88
47.97 5.32
10.84 0.67
11.25 0.63
5.88 0.04
10.70 1.35
49.85 6.66
0.92 0.08
131.30 38.54
43.89 0.02
114.38 9.24
75.45 11.28
18.25 1.34
24.39 0.86
25.46 0.66
25.29 0.60
151.13 58.12
6.21 0.44
Auranofin
a
Mean SE of at least three determinations by using MTT method.
although the IC50 values of complexes [Au(L2b)PPh3] (3b)
a remarkably higher SI value than cisplatin (4.3-up fold) and aur-
anofin (1.92-up fold). To further analyse the toxicity of our com-
plexes on a non-cancerous model, differentiated Caco-2 cells were
incubated 72h with the complexes at the IC50 values previously
obtained on undifferentiated Caco-2 cells (Fig. S51). No significant
changes in cell viability were observed, with suggests that these
concentration values resulted toxic on cancerous cells rather than
on the non-cancerous ones.
According to the results obtained on this section in terms of
antiproliferative effect and selectivity on cancer cells, [Au(L2b)
PPh3] (3b) and [Au(L2c)PTA] (4c) were selected as the most
promising complexes and their mechanism of action on undiffer-
entiated Caco-2/TC7 cells was further analysed.
(1.52 0.91
comparable to the obtained upon treatment with auranofin
(1.80 0.10 M) on undifferentiated Caco-2 cells. Whereas the IC50
mM) and [Au(L2c)PTA] (4c) (2.33 1.26 mM) were
m
values of the PTA gold derivatives (complexes 4a, 4b, 4c and 4d) on
HepG2 cells were comparable to the measured on Caco-2 and MCF-
7, the cytotoxicity of the related PPh3 derivatives against HepG2
cells was 3e4 times lower, except for [Au(L2b)PPh3] (3b), which
indeed displayed an IC50 value 3.33 times higher than the PTA
counterpart 4b. Curiously, the opposite effect was observed on
MCF-7 cell line, in which complex 3b displayed the highest IC50
value compared with the other three PPh3 complexes. In general,
similar IC50 values were observed in related alkynyl gold(I) phos-
phane complexes on MCF-7 or Caco-2 cells [35,50,51,57e60].
Caco-2 cell line is considered a model of the intestinal barrier
due to the spontaneous differentiation acquired after reaching
confluence. Therefore, we tested the toxicity of the series of gold
derivatives on differentiated Caco-2 cells in order to obtain a first
estimation in regard to their behaviour on non-cancerous tissue
(Table 3), and the Selectivity Index (SI) was in addition calculated
(Table S1) as previously indicated by Badisa et al. [61]. Whereas the
SI values of the PTA complexes were in the same range than the
obtained for cisplatin and auranofin on the three cancer models,
PPh3 derivatives highlighted due to their high selectivity to undif-
ferentiated Caco-2 cells, with SI values 5 to 10 times higher than
those obtained for cisplatin and auranofin. Complexes 3a, 3c and 3d
showed in addition higher SI values than cisplatin and auranofin on
MCF-7. Finally, [Au(L2b)PPh3] (3b) was the only complex to display
2.6.2. Measurement of COX-1/2 activity
Given the affinity of flavones for the enzyme cyclooxygenase
(COX) [62,63], we considered both isoforms 1 and 2 as likely targets
of our flavone gold derivatives [Au(L2b)PPh3] (3b) and [Au(L2c)
PTA] (4c). Therefore, we measured the activity of COX-1 (Fig. 6A)
and COX-2 (Fig. 6B) from undifferentiated Caco-2/TC7 cells after 2
and 24h of incubation. A reduction of the activity of the enzyme
COX-1 is observed along the first 2h of incubation with both com-
plexes. However, non-inhibitory effect was detected after overnight
incubation with complex 3b. Nevertheless, in the case of complex
4c, inhibition of COX-1 was noticed at both incubation periods,
being 16.3 times lower after 24h of incubation. Our data suggest
that cellular compensation mechanisms might be occurring,
although further assays are needed to validate this hypothesis.
Fig. 6. Measurement of COX activity. A) Determination of COX-1 activity from undifferentiated Caco-2/TC7 cells after 2 and 24h incubation with [Au(L2b)PPh3] (3b) and [Au(L2c)
PTA] (4c) at their IC50 values. B) Determination of COX-2 activity from undifferentiated Caco-2/TC7 cells after 2 and 24h incubation with [Au(L2b)PPh3] (3b) and [Au(L2c)PTA (4c) at
their IC50 values. *p < 0.05 vs negative control. #p < 0.05 vs gold complex.

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
7
Regarding to the enzyme COX-2, [Au(L2b)PPh3] (3b) significantly
decreased its activity at 2h that was maintained after 24h of incu-
bation, whereas treatment with [Au(L2c)PTA] (4c) did not display
any effect on COX-2 activity. Therefore, these results suggest that 3b
and 4c might interact with different targets.
Few research data are available in regards to the inhibitory effect
of COX triggered by gold complexes. Auranofin was recently re-
ported to be able to inhibit COX-2 [64]. In addition, Ott et al. re-
ported its ability for COX-1 inhibition in human platelets [65],
gold derivatives and glutathione reductase (GR), which has also a
key role in controlling redox balance [74]. The function of GR and
TrxR are closely related and the inhibition of any of them leads to
cancer cell death due to an aberrant increase in ROS levels.
Misfunction of one of these systems might lead to over-
compensation of the other in order to avoid cell failure [75,76].
Furthermore, authors have reported that certain flavanols are able
to directly inhibit TrxR [77] and GR [78]. Therefore, we investigated
the influence of [Au(L2b)PPh3] (3b) and [Au(L2c)PTA] (4c) on the
activity of TrxR (Fig. 7A) and GR (Fig. 7B) from undifferentiated
Caco-2/TC7 cells after 2 and 24h of incubation with the complexes.
As shown in Fig. 7A, both complexes [Au(L2b)PPh3] (3b) and
[Au(L2c)PTA] (4c) were able to inhibit TrxR activity after short in-
cubation time (2h), which was maintained upon overnight treat-
ment. Indeed, after 24h incubation we observed a higher decrease
in TrxR activity induced by complex 4c, whereas the effect caused
by complex 3b remained unchanged. Regarding to interaction with
GR (Fig. 7B), complex 3b did not induce any changes after 2h in-
cubation, whereas a great increase in GR activity was noticed after
24h of incubation. According to previous results from Jortzik et al.
[79], inhibition of TrxR induced by gold complexes might lead to
upregulation of the expression of those enzymes involved in both
thioredoxin and glutathione systems in order to restore redox
balance. Our present data suggest that inhibition of TrxR triggered
by complex 3b leads to a compensatory overexpression of GR. On
the other hand, complex 4c effectively inhibited GR activity
although initially the activity of the enzyme was increased in
comparison to negative control. In conclusion, herein we have
demonstrated that complex 3b inhibits TrxR, thus leading to an
increase in GR activity, whereas complex 4c was found able to
inhibit both TrxR and GR after 24h incubation. Taken together,
these results and our previous findings regarding to the effect on
COX-1/2 activity, our gold complexes seem to act through a multi-
target mechanism of action.
although authors evaluated high concentrations (10 and 100 mM)
that according to our data on differentiated Caco-2 cells (Table 2)
would be considerably cytotoxic for non-cancerous tissue. How-
ever, gold coordination to 3-hydroxyflavones might increase its
affinity for this enzyme, leading to stronger anticancer and anti-
inflammatory properties. Similar results have been obtained with
drugs containing well-known COX inhibitors including valdecoxib
[66] and acetylsalicylic acid [67] among others.
Although inhibition of both isoforms 1 and 2 has been related to
antitumor effect, inhibition of COX-2 is preferred due to the minor
incidence of side effects. COX-1 is involved on gastric mucosa
maintenance and its pharmacological inhibition is related to gastric
injury [68]. Therefore, researchers have focused on the develop-
ment of novel selective COX-2 inhibitors. In line with this, the
absence of inhibitory effect of COX-1 activity after 24h incubation
with complex 3b is of great interest since it might be indicative of
COX-2 selectivity and consequently might induce less side effects
than complex 4c.
The different effect induced on COX-1/2 activity by complexes
3b and 4c might explain the results obtained on MCF-7 cell line. As
shown in Table 3 [Au(L2b)PPh3] (3b) displayed less toxicity on this
cell model in comparison with undifferentiated Caco-2 and HepG2,
whereas the IC50 values of [Au(L2c)PTA] (4c) were comparable
between the three cell lines. Contrary to HepG2 [69] and Caco-2
[70], MCF-7 human breast tumor cell line lacks endogenous COX-
2 expression [71]. Consequently, the reduced antitumor effect of
complex 3b on MCF-7 cells is not surprising since at least one of its
targets is absent. Furthermore, these results suggest that complex
3b might be a multi-target compound, given that it displays toxicity
on MCF-7 despite the lack of the enzyme COX-2.
Villegas et al. [80] found that COX-1 inhibition with piroxicam
resulted in glutathione (GSH) depletion along with a decrease in GR
activity on a rat gastric mucosa. Although COX-2 selective inhibi-
tion eventually resulted in the same situation, treatment with
piroxicam triggered GSH depletion in a faster manner. We have also
observed an interplay between COX-1 and GR inhibition on un-
differentiated Caco-2 cells treated with [Au(L2c)PTA] (4c). Since
inhibition of COX-1 was noticed previously to the decrease in GR
activity, it is feasible to suggest that the inhibition of GR induced
upon treatment with complex 4c might be a consequence of direct
inhibition of the enzyme, a downstream event of COX-1 inhibition
2.6.3. Analysis of redox enzymes TrxR and GR inhibition
Most of gold complexes are able to interact with Thioredoxin
reductase due to the affinity between the gold atom and the sele-
nocysteine residue present on the active site of the enzyme [72,73].
Less data has been collected in regards to the interaction between
Fig. 7. Analysis of the activity of redox enzymes. A) Measurement of the activity of TrxR from undifferentiated Caco-2/TC7 cells after 2 and 24h of incubation with [Au(L2b)PPh3]
(3b) and [Au(L2c)PTA] (4c) at their IC50 values. B) Measurement of the activity of GR from undifferentiated Caco-2/TC7 cells after 2 and 24h incubation with [Au(L2b)PPh3] (3b) and
[Au(L2c)PTA] (4c) at their IC50 values. *p < 0.05 vs negative control. #p < 0.05 vs gold complex.

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
8
or a combination of both. Future assays are needed.
the most relevant steps of intrinsic apoptosis is the loss of mito-
chondrial membrane potential (MMP) that leads to the eventual
collapse of the whole organelle. The analysis of mitochondrial
integrity upon 24h incubation with complex 3b was performed by
measuring the incorporation of the cationic dye DiIC1 (1, 10,3,3,30-
haxamethylindodicarbo-cyanine iodide). As can be observed in
Fig. 9C, complex 3b increased the percentage of cells with altered
MMP when compared with untreated cells, which might be
indicative of the induction of mitochondrial apoptosis.
2.6.4. Determination of ROS levels
As aforementioned, inhibition of TrxR and GR has been related
to a dramatic increase in ROS levels that triggers cell death [81]. In
light of this, we measured ROS production on undifferentiated
Caco-2/TC7 cells incubated with [Au(L2b)PPh3] (3b) and [Au(L2c)
PTA] (4c) for 1, 3 and 24h (Fig. 8).
Treatment with both complexes resulted in an increased ROS
generation upon overnight incubation. According to results ob-
tained after 1 and 3h incubation respectively, the pro-oxidant
effect of [Au(L2b)PPh3] (3b) was slightly faster than the induced
by [Au(L2c)PTA] (4c). Similar increase in ROS and reduction of
thioredoxin reductase activity have been found in gold(I) and
gold(III) derivatives [47,82]. As previously shown on Fig. 7,
although both complexes 3b and 4c are able to inhibit TrxR after
2h incubation, treatment with 4c induced initially an increase in
GR activity. Therefore, despite complex 4c triggered a greater
decrease in TrxR than complex 3b at short incubation time, the
increased activity of GR maintained redox balance and no signif-
icant changes in ROS levels were observed at 1h of incubation.
Moreover, some authors have reported that inhibition of COX-2
lead to a overproduction of ROS [83,84]. Thus, the observed ef-
fect on ROS production induced by [Au(L2b)PPh3] (3b) might be a
consequence of the inhibition of both TrxR and COX-2, which
could explain the significant (p < 0.05) increase in ROS levels after
1h incubation.
The pro-apoptotic effect of complex 3b lead to alterations in cell
cycle progression, which produced a reduction in G2 population
concomitant to an increase in number of cells undergoing G1 phase,
similarly to auranofin [85] (Fig. 9D).
On the other hand, [Au(L2c)PTA] (4c) did not display any sig-
nificant changes in apoptotic population after 24h incubation,
although a 8.5-up fold increase in late apoptotic cells population
was detected after 48h of incubation (Fig. 10A). The delay in
apoptosis induction compared to [Au(L2b)PPh3] (3b) might be
related to the previously discussed slower pro-oxidant effect
(Fig. 8), given the key role of redox homeostasis on cancer cell
survival. Nevertheless, the antiproliferative effect of [Au(L2c)PTA]
(4c) after overnight incubation was enhanced by pre-incubation
with the autophagy inhibitor chloroquine (CQ) (Fig. 10B). Due to
the aberrant proliferation of cancer cells, increased basal rates of
autophagy have been found in comparison to non-cancerous cells,
and in the concrete case of colorectal cancer (CRC) upregulation of
autophagy is related to tumour mass growth and spread.
Moreover, autophagy is closely related to prognosis, since it can
lead to drug resistance and chemotherapy failure [86]. Conse-
quently, the therapeutic modulation of autophagy has been pro-
posed as a novel approach to CRC treatment in order to increase
chemosensitivity. Shi et al. [87] found that pre-treatment of Caco-
2 cells with the autophagy inhibitor 3-methyladenine resulted in an
enhancement of cell sensitivity to oxaliplatin. Our present findings
demonstrate that the chemotherapeutic potential of autophagy
inhibition is not limited to platinum-based drugs since gold com-
plexes can also be benefited from this strategy.
2.6.5. Cell death studies
Once it has been evaluated the mechanism of action of the
multi-target complexes [Au(L2b)PPh3] (3b) and [Au(L2c)PTA (4c),
we analysed the type of cell death triggered by both of them.
We found that [Au(L2b)PPh3] (3b) induced apoptosis on Caco-2/
TC7 after 24h incubation at its IC50 concentration (Fig. 9). As shown
in Fig. 9A, total apoptotic events increased a 5.9-up fold after
treatment with complex 3b, with a significant increase in late
apoptotic population in comparison with negative control. More-
over, an increase in the number of cells undergoing early apoptosis
were also noticed, although it was statistically non-significant.
Contrarily, no significant changes in the percentage of cells un-
dergoing necrosis were noticed (Fig. S52). Induction of apoptosis
was further confirmed by the observed increase in the percentage
of cells with activated caspase 3 (Fig. 9B).
3. Conclusions
In this work, the synthesis of new flavone-derived ligands with a
propargyl ether group, the corresponding alkynyl gold(I) de-
rivatives and the evaluation of their anticancer activity is described.
The complexes are stable at 37 ꢁC under physiological conditions,
display balanced relationship between lipophilicity and hydrophi-
licity and good binding propensity to BSA, being the hydrophobic
cavity of site I in the subdomain IIA of BSA the most probable
location of the gold complex. Although free flavones are active
against Caco-2/TC7 human colon cancer cells, a large enhancement
of the cytotoxicity has been obtained after gold(I) coordination. In
addition, such coordination has allowed the interaction of the gold
complexes with COX-1/2 enzyme as well as with the redox en-
zymes TrxR and GR, which leads to an increased ROS production on
colorectal adenocarcinoma cell line Caco-2. Through a multi-
targeted mechanism of action, these series of gold complexes
have triggered apoptotic cell death via the intrinsic pathway and
have altered cell cycle progression. Moreover, toxicity evaluation on
differentiated Caco-2 cells as a model of non-cancerous tissue
suggests that these gold complexes might act preferentially on
cancer cells rather than on the differentiated ones. Finally, pre-
liminary studies showed that the concomitant administration of
[Au(L2c)PTA] with autophagy inhibitors such as chloroquine might
enhance its antiproliferative effect due to the blockage of cyto-
protective autophagy. Taken together, our results suggest a prom-
ising future for these complexes on cancer chemotherapy.
Gold complexes are known to preferentially trigger mitochon-
drial or intrinsic apoptosis rather than by extrinsic pathway. One of
Fig. 8. Time-course (1, 3 and 24h) measurement of ROS generation on undifferentiated
Caco-2/TC7 cells upon incubation with [Au(L2b)PPh3] (3b) and [Au(L2c)PTA (4c) at
their IC50 values. *p < 0.05 vs negative control.

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
9
Fig. 9. Analysis of the type of cell death induced on undifferentiated Caco-2/TC7 cells after 24h incubation with [Au(L2b)PPh3] (3b) at its IC50 value. A) Percentages of alive, necrotic,
early apoptotic and late apoptotic cells. B) Percentages of cells with activated caspase 3. C) Analysis of the integrity of mitochondrial membrane potential in terms of loss of
fluorescence. Percentages of each cell population are included. D) Cell cycle analysis. Percentages of each cell population are included. *p < 0.05 vs negative control.
4. Experimental
H2O2 (33% w/v). After 3 h of stirring, the mixture was poured into
an ice-water solution affording yellow solids, which were filtered
off and dried in air.
4.1. General procedure for the preparation of the 3-hydroxyflavones
[37] (HL-1a-d)
NaOH (0.4 g) was added to a methanol (20 mL) solution of 2-
hydroxyacetophenone (3.32 mmol) and benzaldehyde or the
different para-substituted benzaldehydes (3.32 mmol). The reac-
tion was refluxed for 3 h until it acquires an intense orange color.
The reaction was cooled to room temperature and followed by the
addition of NaOH (0.4 g) dissolved in 20 mL of water and 1.45 mL of

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
10
Fig. 10. Analysis of cell death induced on undifferentiated Caco-2/TC7 cells after incubation with [Au(L2c)PTA] (4c) at its IC50 value. A) Time-dependent effect of apoptosis induction
after 24 and 48h incubation of undifferentiated Caco-2/TC7 cells with [Au(L2c)PTA] (4c) at its IC50 value. B) Effect of pre-incubation with chloroquine (10
undifferentiated Caco-2/TC7 cells viability upon 24h treatment with [Au(L2c)PTA]. *p < 0.05 vs negative control. #p < 0.05 vs gold complex.
mM, 1h) on changes in
2-phenyl-3-hydroxy-4H-chromen-4-one (HL-1a, R ¼ H). Yellow
solid in 58% yield. 1H NMR (400 MHz, dmso-d6, 25 ꢁC)
d
(ppm) ¼ 8.69 (m, 2H, H2’,6’), 8.02 (dd, J ¼ 8.1 Hz; 1.2 Hz 1H, H6),
7.56 (m, 2H, H9þH8), 7.37 (dd, J ¼ 10.7; 5.0 Hz, 2H, H5’,3’), 7.24 (ddd,
8.0; 6.6; 1.4 Hz, 1H, H7), 7.16 (dd, J ¼ 10.7; 4.2 Hz, 1H, H4’). 13C{1H}
NMR (100 MHz, CDCl3):
d
(ppm) ¼ 177.3 (C¼O), 154.5 (C3), 144.4
(C2), 134.0, 132.6 (C7,9), 128.5 (C3’,5’), 128.0 (C8), 126.6 (C2’,6’), 125.2
(C6), 123.7 (C4’), 121.7, 118.7 (C7,9). IR:
1519 cmꢀ1
n
(-OH) 3385 cmꢀ1
,
n
(C¼O)
.
2-phenyl-3-(prop-2-in-1-yloxy)-4H-chromen-4-one (HL-2a, R ¼
2-(4-bromophenyl)-3-hydroxy-4H-chromen-4-one
(HL-1b,
1
H). White solid in 50% yield. H NMR (400 MHz, dmso-d6, 25 ꢁC)
d
R ¼ Br). Yellow solid in 67% yield. 1H NMR (400 MHz, dmso-d6,
(ppm) ¼ 8.12 (m, 3H, H6 þ H2’,6’), 7.86 (ddd, J ¼ 8.5; 6.9; 1.6 Hz,
25 ꢁC)
d
(ppm) ¼ 8.65 (d, J ¼ 8.0 Hz; 2H, H2’,6’), 8.00 (m, 1H, H6), 7.56
1H,H8), 7.79 (m, 1H, H9), 7.59 (m, 3H, H3’,4’,5’), 7.53 (m, 1H, H7), 4.98
(m, 2H, H9þH8), 7.51 (d, J ¼ 8.9 Hz, 2H, H5’,3’), 7.22 (ddd, J ¼ 6.8; 5.8;
(d, J ¼ 2.4 Hz, 2H, -CH2-), 3.48 (t, J ¼ 2,4 Hz, 1H, -C≡C-H). 13C{1H}
2.2 Hz, 1H, H7). 13C{1H} NMR (100 MHz, CDCl3):
d
(ppm) ¼ 180.3
NMR (100 MHz, CDCl3):
d
(ppm) ¼ 176.6 (C]O), 156.4, 155.2, 138.4,
(CO), 156.2 (C4’), 143.0, 133.6, 132.6 (C8,9), 131.4 (C3’,5’), 128.1 (C2’,6’),
134.4 (C8), 131.4, 130.8, 129.1, 129.1, 125.7 (C2’,3’,4’,5’,6’), 125.4 (C7),
125.3 (C6), 123.6 (C7), 121.5, 120.4, 118.6 (C8,9). IR: (-OH) 3335 cmꢀ1
n ,
123.7 (C6), 118.9 (C9), 79.7 y 79.1 (C≡C), 59.1 (-CH2). IR:
n(≡CeH)
n
(C¼O) 1624 cmꢀ1
.
3253 cmꢀ1
,
n
(C≡C) 2113 cmꢀ1
,
n
(C¼O) 1611 cmꢀ1. Elemental anal-
2-(4-clorophenyl)-3-hydroxy-4H-chromen-4-one (HL-1c, R ¼ Cl).
ysis calcd. (%) for C18H12O3 (276.28): C 78.25, H 4.38; found: C 77.88,
H 4.26.
1
Yellow solid in 62% yield. H NMR (400 MHz, dmso-d6, 25 ꢁC)
d
(ppm) ¼ 8.73 (d, J ¼ 8.6 Hz, 2H, H2’ 6’), 8.00 (m, 1H, H6), 7.56 (dt,
2-(4-bromophenyl)-3-(prop-2-in-1-yloxy)-4H-chromen-4-ona
J ¼ 8.5; 5.0 Hz, 2H, H9,8), 7.39 (d, J ¼ 8.9 Hz, 2H, H5’,3’), 7.23 (ddd,
(HL-2b, R ¼ Br). White solid in 52% yield. 1H NMR (400 MHz, dmso-
J ¼ 8.0; 6.2, 1.8 Hz, 1H, H7). 13C{1H} NMR (100 MHz, CDCl3):
d(ppm)
d6, 25 ꢁC)
d
(ppm) ¼ 8.11 (m, 1H, H6), 8.07 (m, 2H, H2’,6’), 7.88 (d,
177.9 (C¼O), 154.4 (C4’), 154.1, 142.3 (C3, C2), 134.8 (C6), 131.6 (C9),
J ¼ 8.0 Hz, 1H, H8), 7.81 (d, J ¼ 8.0 Hz, 3H, H9þH3’,5’), 7.53 (ddd,
J ¼ 8.0; 7.0; 1.1 Hz, 1H, H7), 4.98 (d, J ¼ 2.5 Hz, 2H, -CH2-), 3.48 (t,
J ¼ 2,5 Hz,1H, -C≡C-H). 13C{1H} NMR (100 MHz, CDCl3):
129.9,128.1 (C3’,5’), 126.7 (C2’,6’),125.3,122.7 (C7), 121.7,118.4 (C8). IR:
n
(-OH) 3351 cmꢀ1
,
n
(C¼O) 1594 cmꢀ1
.
2-(4-metoxiphenyl)-3-hydroxy-4H-chromen-4-one
(HL-1d,
d
(ppm) ¼ 174.1 (C]O), 154.9 and 155.3 (C3,5), 138.3 (C4’), 134.7 (C7),
R ¼ OCH3). Yellow solid in 59% yield. 1H NMR (400 MHz, dmso-d6,
131.2 and 131.3 (C2’,6’ and C3’,5’), 130 (C1’), 125.8 and 125.3 (C6,8),
25 ꢁC)
d
(ppm) ¼ 8.66 (d, J ¼ 9.1 Hz, 2H, H2’,6’), 8.01 (dd, J ¼ 8.0;
124.9 and 123.9 (C10,2), 118.8 (C9), 79.9 and 78.9 (C≡C), 59.2 (-CH2-).
1.2 Hz, 1H, H6), 7.54 (ddd, J ¼ 12.2; 10.1; 4.7 Hz, 2H, H9,8), 7.24 (m,
IR:
n
(≡CeH) 3228 cmꢀ1
,
n
(C≡C) 2117 cmꢀ1
,
n
(C¼O) 1605 cmꢀ1
.
2H, H5’,3’), 6.95 (d, J ¼ 9.1 Hz, 1H, H7). 13C{1H} NMR (100 MHz,
Elemental analysis calcd. (%) for C18H11BrO3 (355.18): C 60.87, H
3.12; found: C 60.52, H 3.13.
CDCl3):
d
(ppm) 176.6 (C¼O), 159.3(C4’), 154.3, 145 (C3), 132.3 (C8,9),
128.4 (C2’,6’), 126.5, 125.1 (C6), 123.7 (C7), 121.7, 118.52 (C8,9), 114.9
2-(4-clorophenyl)-3-(prop-2-in-1-yloxy)-4H-chromen-4-one (HL-
(C3’,5’), 114.0, 55.6 (-OCH3). IR:
n
(-OH) 3195 cmꢀ1
,
n
(C¼O)
2c, R ¼ Cl). White solid in 50% yield. 1H NMR (400 MHz, dmso-d6,
1595 cmꢀ1
25 ꢁC)
d
(ppm) ¼ 8.14 (t, J ¼ 6.4 Hz, 2H, H2’,6’), 8.10 (dd, J ¼ 8.0;
1.4 Hz, 1H, H6), 7.87 (m, 1H, H8), 7.79 (d, J ¼ 8.0 Hz, 3H, H9 þH3’5’),
4.2. General procedure for the preparation of the propargyl flavones
HL-2a-d
7.65 (d, J ¼ 8.7 Hz,1H, H7), 4.98 (d, J ¼ 2.4 Hz, 2H, -CH2-), 3.46 (m,1H,
-C≡C-H). 13C{1H} NMR (100 MHz, CDCl3):
d
(ppm) ¼ 174.2 (C]O),
155.1 (C3,5), 138.2, 134.8 (C8,9), 130.9 (C2’,6’), 129.1 (C3’,5’), 125.8, 125.4,
K2CO3 in excess and propargylbromide (1.6 equivalents) were
added to a solution of flavones HL-1a-d. The reaction was heated
under reflux for 6 h, and after this time, the excess of base was
filtered off and the residue was evaporated to dryness.
123.7 (C7), 118.8 (C9,8), 79.9 and 79.0 (C≡C), 59.25 (-CH2-). IR:
n
(≡CeH) 3238 cmꢀ1
,
n
(C≡C) 2112 cmꢀ1
,
n
(C¼O) 1628 cmꢀ1
.
Elemental analysis calcd. (%) for C18H11ClO3 (310.73): C 69.58, H
3.57; found: C 69.24, H 3.75.

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
11
2-(4-metoxiphenyl)-3-(prop-2-in-1-yloxy)-4H-chromen-4-one
dmso-d6)
d
(ppm) ¼ 8.20 (d, J ¼ 8.6 Hz, 2H, H2’,6’), 8.09 (d, J ¼ 7.63 Hz,
(HL-2d, R ¼ OMe). White solid in 52% yield. 1H NMR (400 MHz,
1H, H6), 7.79 (m, 2H, H9,7), 7.52 (m, 16H, PPh3þH8), 7.10 (d, J ¼ 8.6 Hz,
dmso-d6, 25 ꢁC)
d
(ppm) ¼ 8.17 (m, 2H, H2’,6’), 8.11 (m, 1H, H6), 8.09
2H, H3’,5’), 4.97 (s, 2H, -CH2-), 3.81 (s, 3H, -CH3) 31P{1H} NMR
(dd, J ¼ 8.0; 1.4 Hz, 1H, H8), 7.84 (m, 1H, H9), 7.77 (d, J ¼ 7.8 Hz, 3H,
H3’,5’), 7.50 (m, 1H, H7), 4.96 (d, J ¼ 2.4 Hz, 2H, -CH2-), 3.86 (s, 3H,
OMe), 3.46 (s, 1H, -C≡C-H). 13C{1H} NMR (100 MHz, CDCl3):
(162 MHz, dmso-d6)
CDCl3):
(ppm) ¼ 174.8 (C¼O), 161.3, 155.2, 134.2 (d, J ¼ 13.9 Hz,
orto, PPh3), 132.5, 131.5 (d, J ¼ 2.2 Hz, Cpara, PPh3), 130.9,130.1, 129.4,
129.1 (d, J ¼ 11.3 Hz, Cmeta, PPh3), 125.9, 124.3, 123.9, 117.8, 113.7,
d
(ppm): 41.2 (s). 13C{1H} NMR (100 MHz,
d
C
d
(ppm) ¼ 174.7(C¼O), 161.6 (C4’), 156.7 and 155.2 (C3,5), 138.0 (C1’),
133.3 (C8),130.7(C2’,6’), 125.7 and 124.7 (C6,8), 124.0 and 123.7 (C10,2),
98.3 and 98.7 (C≡C), 60.6 (-CH2-), 55.3 (OMe). IR: n ,
(C≡C) 2130 cmꢀ1
117.9 (C9), 113.8 (C3’,5’), 78.7 and 75.9 (C≡C), 59.25 (-CH2-), 55.4
n
(C¼O) 1627 cmꢀ1. Elemental analysis calcd. (%) for C37H28AuO4P
(OCH3). IR:
n
(≡CeH) 3286 cmꢀ1
,
n
(C≡C) 2161 cmꢀ1
,
n
(C¼O)
(764.56): C 58.12, H 3.69; found: C 57.85, H 3.54.
1634 cmꢀ1. Elemental analysis calcd. (%) for C19H14O4 (306.31): C
74.50, H 4.61; found: C 74.15, H 4.85.
4.3. Synthesis of the [Au(L2)PR03] complexes
To a solution of KOH (0.225 mmol) in MeOH (ca. 10 mL) con-
taining L2a-d (0.15 mmol) was added [AuCl(PR03)] (PR03 ¼ PPh3,
PTA) (0.15 mmol). The gold derivatives [AuCl(PPh3)] and
[AuCl(PTA)] were prepared as published elsewhere [88,89]. The
corresponding solids precipitated in the methanolic solution, and
were isolated by filtration after 20 h of stirring at room tempera-
ture, washed with methanol and diethyl ether, and dried in vacuo.
Using this method, the following complexes were prepared:
[Au(L2a)PTA] (4a). Yield: 55%. Light yellow solid. 1H NMR
(400 MHz, dmso-d6)
d
(ppm) ¼ 8.17 (m, 2H, H2’,6’), 8.10 (d, J ¼ 6.9 Hz,
1H, H6), 7.84 (d, J ¼ 7.5 Hz, 1H, H7), 7.78 (d, J ¼ 8.5 Hz, 1H, H9), 7.54
(m, 4H, H3’,5’þH4þH8), 4.99 (s, 2H, -CH2-), 4.44 and 4.31 (AB system,
J ¼ 12.0 Hz, 6H, NCH2N), 4.18 (s, 6H, NCH2P). 31P{1H} NMR (162 MHz,
dmso-d6)
d
(ppm): ꢀ49.2 (s). 13C{1H} NMR (100 MHz, CDCl3):
d
(ppm) ¼ 174.8 (C¼O), 156.4, 155.3, 139.1, 133.2, 131.3, 130.4 (C9),
129.0 and 128.3 (C3’,5’ and C2’,6’), 125.9 (C6), 124.5 (C8), 124.2, 117.9
(C7), 98.6 (C≡C), 73.2 (d, J ¼ 6.9 Hz, NCH2N), 60.6 (-CH2-), 52.3 (d,
J ¼ 20 Hz, NCH2P). IR:
n
(C≡C) 2136 cmꢀ1
,
n
(C¼O) 1630 cmꢀ1
.
Elemental analysis calcd. (%) for C24H23AuN3O3P (629.39): C 45.80,
H 3.68, N 6.68; found: C 45.50, H 3.71, N 6.50.
[Au(L2b)PTA] (4b). Yield: 55%. Light yellow solid. 1H NMR
(400 MHz, dmso-d6)
d
(ppm) ¼ 8.11 (m, 3H, H2’,6’þH6), 7.84 (d,
[Au(L2a)PPh3] (3a). Yield: 65%. White solid. 1H NMR (400 MHz,
J ¼ 7.1 Hz, 1H, H7), 7.78 (m, 3H, H3’,5’þH6), 7.52 (t, J ¼ 7.3 Hz, 1H, H8),
4.98 (s, 2H, -CH2-), 4.44 and 4.31(AB system, J ¼ 12.0 Hz, 6H,
NCH2N), 4.18 (s, 6H, NCH2P). 31P{1H} NMR (162 MHz, dmso-d6)
dmso-d6)
d
(ppm) ¼ 8.27 (m, 2H, H2’,6’), 8.20 (dd, J ¼ 8.1, 1.4 Hz, 1H,
H6), 7.92 (ddd, J ¼ 8.4; 7; 1.6 Hz, 1H, H7), 7.85 (d, J ¼ 8.0 Hz, 1H, H9),
7.62 (m, 19H, PPh3 þH3’,4’,5’þH8), 5.08 (s, 2H, -CH2-) 31P{1H} NMR
d
(ppm): ꢀ49.4 (s). 13C{1H} NMR (100 MHz, dmso-d6):
(162 MHz, dmso-d6)
CDCl3):
d
(ppm): 41.4 (s). 13C{1H} NMR (100 MHz,
d
(ppm) ¼ 174.4 (C¼O), 155.1, 138.7, 134.7, 131.9 and 131.1 (C3’,5’ and
d
(ppm) ¼ 175.2 (C¼O), 156.7, 155.3, 139.3, 134.2 (d,
C2’,6’), 130.4, 125.7, 125.4, 124.9, 123.7, 118.9, 87.2 (C≡C), 72.2 (d,
J ¼ 13.9 Hz, Corto, PPh3), 133.1, 131.5 (d, J ¼ 2.2 Hz, Cpara, PPh3), 130.3,
J ¼ 7.7 Hz, NCH2N), 60.6 (-CH2-), 51.1 (d, J ¼ 20 Hz, NCH2P). IR:
n(C≡C)
130.1, 129.4, 129.2, 129.1 (d, J ¼ 11.3 Hz, Cmeta, PPh3), 128.2, 125.9,
2120 cmꢀ1
,
n
(C¼O) 1627 cmꢀ1. Elemental analysis calcd. (%) for
124.4, 124.3, 117.9, 98.2 and 98.7 (C≡C), 60.7 (-CH2-). IR:
n
(C≡C)
C24H22AuBrN3O3P (708.29): C 40.69, H 3.13, N 5.93; found: C 40.38,
2136 cmꢀ1
,
n
(C¼O) 1626 cmꢀ1. Elemental analysis calcd. (%) for
H 2.83, N 5.88.
C
36H26AuO3P (734.53): C 58.87, H 3.57; found: C 58.52, H 3.4.
[Au(L2c)PTA] (4c). Yield: 60%. Light yellow solid. 1H NMR
[Au(L2b)PPh3] (3b). Yield: 65%. White solid. 1H NMR (400 MHz,
(400 MHz, dmso-d6)
d
(ppm) ¼ 8.20 (d, J ¼ 8.7 Hz, 2H, H2’,6’), 8.10 (d,
dmso-d6)
d
(ppm) ¼ 8.13 (m, 3H, H2’,6’þH6), 7.85 (ddd, J ¼ 8.4, 7.2,
J ¼ 6.7 Hz, 1H, H6), 7.84 (d, J ¼ 6.9 Hz, 1H, H7), 7.78 (d, J ¼ 8.4 Hz, 1H,
H9),7.65 (d, J ¼ 8.7 Hz, 2H, H3’,5’), 7.52 (t, J ¼ 7.4 Hz, 1H, H8), 4.98 (s,
2H, -CH2-), 4.44 and 4.31 (AB system, J ¼ 12.0 Hz, 6H, NCH2N), 4.17
1.6 Hz, 1H, H7), 7.75 (m, 3H, H3’,5’þH9), 7.54 (m, 16H, PPh3þH8), 4.98
(s, 2H, -CH2-). 31P{1H} NMR (162 MHz, dmso-d6)
d(ppm): 41.3 (s).
13C{1H} NMR (100 MHz, CDCl3):
139.3, 134.2 (d, J ¼ 13.9 Hz, Corto, PPh3), 133.3, 131.5 (d, J ¼ 2.4 Hz,
para, PPh3), 131.4, 130.8, 130.5, 130.0, 129.8, 129.3, 129.1 (d,
d
(ppm) ¼ 175.0 (C¼O), 155.7, 155.2,
(s, 6H, NCH2P). 31P{1H} NMR (162 MHz, dmso-d6)
d
(ppm): ꢀ49.2 (s).
13C{1H} NMR (100 MHz, dmso-d6):
d
(ppm) ¼ 174.5 (C¼O), 155.1,
C
138.7, 135.9, 134.7, 130.9 and 128.9 (C3’,5’ and C2’,6’), 130.0, 125.7,
125.4, 123.7, 118.9, 99.7 (C≡C), 72.2 (d, J ¼ 8.2 Hz, NCH2N), 60.5
J ¼ 11.3 Hz, Cmeta, PPh3), 125.9, 124.9, 124.6, 124.2, 117.9, 98.3 and
97.9 (C≡C), 60.8 (-CH2-). IR:
n
(C≡C) 2130 cmꢀ1
,
n
(C¼O) 1631 cmꢀ1
.
(-CH2-), 51.3 (d, J ¼ 21.2 Hz, NCH2P). IR:
n
(C≡C) 2122 cmꢀ1
,
n
(C¼O)
Elemental analysis calcd. (%) for C36H26AuClO3P (813.43): C 53.16, H
3.10; found: C 52.85, H 2.9.
1628 cmꢀ1. Elemental analysis calcd. (%) for C24H22AuClN3O3P
(663.84): C 43.42, H 3.32, N 6.33; found: C 43.07, H 2.98, N 6.72.
[Au(L2d)PTA] (4d). Yield: 65%. Light yellow solid. 1H NMR
[Au(L2b)PPh3] (3c). Yield: 65%. White solid. 1H NMR (400 MHz,
dmso-d6)
d
(ppm) ¼ 8.22 (m, 2H, H2’,6’), 8.11 (dd, J ¼ 7.8, 1.6 Hz, 1H,
(400 MHz, dmso-d6)
d
(ppm) ¼ 8.18 (d, J ¼ 8,7 Hz, 2H, H2’,6’), 8.08 (d,
H6), 7.85 (ddd, J ¼ 8.4, 6.8, 1.6 Hz, 1H, H7), 7.77 (d, J ¼ 8 Hz, 1H, H9),
7.54 (m,18H, PPh3þH3’,5’þH8), 4.99 (s, 2H, -CH2-). 31P{1H} NMR
J ¼ 8.5 Hz, 1H, H6), 7.82 (d, J ¼ 7.7 Hz, 1H, H7), 7.76 (d, J ¼ 8.5 Hz, 1H,
H9), 7.51 (t, J ¼ 7.0 Hz, 1H, H8), 7.12 (d, J ¼ 8.7 Hz, 2H, H3’,5’), 4.97 (s,
2H, -CH2-), 4.44 and 4.31 (AB system, J ¼ 12.0 Hz, 6H, NCH2N), 4.18
(162 MHz, dmso-d6)
CDCl3):
PPh3), 133.6, 132.1, 130.7, 129.3 (d, J ¼ 11.3 Hz, Cmeta, PPh3), 128.5,
125.6, 118.5, 97.4 and 96.9 (C≡C), 60.5 (-CH2-). IR: (C^C)
2130 cmꢀ1 (C≡C) 1625 cmꢀ1. Elemental analysis calcd. (%) for
d
(ppm): 41.3 (s). 13C{1H} NMR (100 MHz,
d
(ppm) ¼ 173.7 (C¼O), 152.2, 134.2 (d, J ¼ 13.6 Hz, Corto
,
(s, 6H, NCH2P). 31P{1H} NMR (162 MHz, dmso-d6)
d
(ppm): ꢀ49.1 (s).
13C{1H} NMR (100 MHz, dmso-d6):
d
(ppm) ¼ 173.7 (C¼O), 161.2,
n
155.6, 155.1, 134.3, 130.9 and 114.4 (C3’,5’ and C2’,6’), 125.5, 125.3,
123.7, 118.8, 93.3 (C≡C), 72.2 (d, J ¼ 7.7 Hz, NCH2N), 60.3 (-CH2-),
,
n
C
36H26AuClO3P (768.97): C 56.23, H 3.28; found: C 55.90, H 3.04.
55.9 (OMe), 51.5 (d, J ¼ 20 Hz, NCH2P). IR:
n
(C≡C) 2123 cmꢀ1
,
n
(C¼O)
[Au(L2b)PPh3] (3d). Yield: 71%. White solid. 1H NMR (400 MHz,
1600 cmꢀ1
. Elemental analysis calcd. (%) for C25H25AuN3O4P

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
12
(659.42): C 45.54, H 3.82, N 6.37; found: C 45.18, H 3.71, N 5.98.
in cell proliferation were analysed by MTT or SRB and the obtained
absorbance values were converted into percentage of growth in-
hibition. Absorbance was measured with SPECTROstar Nano (BMG
Labtech).
4.4. Crystal structure determination
Crystals were mounted in inert oil on glass fibres and trans-
ferred to the cold gas stream of a Smart APEX CCD diffractometers
equipped with a low-temperature attachment. Data were collected
Conflicts of interest
using monochromated MoK
a
radiation (
l
¼ 0.71073 Å). Scan type
Authors declare no conflict of interest.
Acknowledgments
6. Absorption corrections based on multiple scans were applied
using SADABS. The structures were solved by direct methods and
refined on F2 using the program SHELXL-2016 [90]. All non-
hydrogen atoms were refined anisotropically. CCDC 1922088 con-
tains the supplementary crystallographic data for this paper. These
data can be obtained free of charge via www.ccdc.cam.ac.uk/data_
request/cif, or by emailing data_request@ccdc.cam.ac.uk, or by
contacting The Cambridge Crystallographic Data Centre, 12 Union
Road, Cambridge CB2 1EZ, UK; fax: þ44 1223 336033.
ꢀ
ꢀ
ꢀ
Authors thank to Centro de Investigacion Biomedica de Aragon
~
(CIBA), Espana for technical assistance: http://www.iacs.aragon.es,
ꢀ
use of Servicio General de Apoyo a la Investigacion-SAI, Universidad
de Zaragoza. We thank The Ministerio de Economía y Com-
petitividad (CTQ2016-75816-C2-1-P and SAF2016-75441-R),
CIBEROBN under Grant (CB06/03/1012) of the Instituto Carlos III,
European Grant Interreg/SUDOE (Redvalue, SOE1/PI/E0123) and
Aragon Regional Government (B16-R17 and E07-17R) for financial
support.
4.5. Cell culture
Human colorectal adenocarcinoma Caco-2 cells were kindly
ꢀ
provided by Dr. Edith Brot-Laroche (Universite Pierre et Marie
Appendix A. Supplementary data
Curie-Paris 6 UMR S872, Les Cordeliers, France). Human breast
adenocarcinoma MCF-7 cells were kindly provided by Carlos J.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.ejmech.2019.111661.
ꢁ
ꢀ
Ciudad and Dr. Veronica Noe (Departamento de Bioquímica y
Fisiología, Facultad de Farmacia, Universidad de Barcelona, Spain).
Human hepatocellular carcinoma HepG2 cells were kindly pro-
vided by Dr. María Angeles Alava (Departamento de Bioquímica y
Biología Molecular, Universidad de Zaragoza, Spain). All cell lines
were maintained in a humidified atmosphere of 5% CO2 at 37 ꢁC.
Cells (passages 20e40) were grown in Dulbecco's Modified Eagles
medium (DMEM) (Gibco Invitrogen, Paisley, UK) supplemented
with 20% fetal bovine serum, 1% non-essential amino acids, 1%
Abbreviations
1,10,3,3,30-hexamethylindodicarbo-cyanine iodide: DiIC1
1,3,5-triaza-7-phosphaadamantane PTA
2,3,4,6-tetraacetyl-1-thio-b-D-glucopyranosato-S-
triethylphosphane gold(I) auranofin
3-(4,5-dimethyl-2-thiazoyl)-2,5-diphenyltetrazolium
bromide: MTT
penicillin (1000 U/mL), 1% streptomycin (1000 mg/mL) and 1%
amphotericin (250 U/mL). Culture medium was replaced every two
days and cells were passaged enzymatically with 0.25% trypsin-1
mM EDTA and sub-cultured on 25 cm2 flasks at a density of
2$104 cells/cm2.
Bovine serum albumin BSA
Chloroquine: CQ
Colorectal cancer CRC
Cyclooxygenase COX
Experiments in undifferentiated Caco-2 cells as well as on MCF-
7 and HepG2 cells were performed 24h post-seeding. For assays on
differentiated Caco-2 cells, cells were cultured on 96-wells plates
under standard culture conditions for 7e9 days, until reaching 80%
confluence as confirmed by optic microscopy observance.
Dimethylsulfoxide: DMSO
Ethylenediaminetetraacetic acid EDTA
Food and Drug Administration FDA
Glutathione GSH
Glutathione reductase GR
Human serum albumin HSA
Mitochondrial membrane potential MMP
Reactive oxygen species ROS
Selectivity Index SI
4.6. Cell treatment
Complexes were initially solved on DMSO to a concentration of
20 mM and then diluted on cell culture without fetal bovine serum
to the required work concentrations. For treatment, cell culture
medium was replaced with medium containing complexes and
cells were incubated at 37 ꢁC for a variable time depending on the
assay.
Sulforhodamine B SRB
Thioredoxin reductase TrxR
References
[1] E. Wong, C.M. Giandomenico, Current status of platinum-based antitumor
drugs, Chem. Rev. 99 (1999) 2451e2466.
[2] Y. Jung, S.J. Lippard, Direct cellular responses to platinum-induced DNA
damage, Chem. Rev. 107 (2007) 1387e1407.
4.7. Cell proliferation assay and IC50 calculation
[3] D. Wang, S.J. Lippard, Cellular processing of platinum anticancer drugs, Nat.
Rev. Drug Discov. 4 (2005) 307e320.
[4] T. Onodera, I. Momose, M. Kawada, Potential anticancer activity of auranofin,
Chem. Pharm. Bull. 67 (2019) 186e191.
Two different methods were used to determine IC50: 3-(4,5-
dimethyl-2-thiazoyl)-2,5-diphenyltetrazolium bromide (MTT)
assay and sulforhodamine B (SRB) assay. MTT assay was performed
ꢀ
[5] B. Englinger, C. Pirker, P. Heffeter, A. Terenzi, C.R. Kowol, B.K. Keppler,
W. Berger, Metal drugs and the anticancer immune response, Chem. Rev. 119
(2019) 1519e1624.
[6] A. Bergamo, P.J. Dyson, G. Sava, The mechanism of tumour cell death by metal-
based anticancer drugs is not only a matter of DNA interactions, Coord. Chem.
Rev. 360 (2018) 17e33.
[7] G. Palermo, A. Magistrato, T. Riedel, T. von Erlach, C.A. Davey, P.J. Dyson,
U. Rothlisberger, Fighting cancer with transition metal complexes: from naked
DNA to protein and chromatin targeting strategies, ChemMedChem 11 (2016)
as previously described by Marmol et al. [91]; SRB assay was per-
ꢀ
formed as previously described by Jimenez et al. [92].
For IC50 calculation, cells were grown in 96-wells plates at a
density of 4000 cells per well and incubated overnight at standard
culture conditions. Then, cells were exposed to a range of concen-
trations of complexes (0e20
mM for complexes 3a-d and 4a-d;
20e100 M for complexes HL-1a-d and HL-2a-d) for 72h. Changes
m

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
13
1199e1210.
Linking flavonoids to gold - a new family of gold compounds for potential
therapeutic applications, Eur. J. Inorg. Chem. (2015) 4275e4279.
[8] G. Chopra, R. Samudrala, Exploring polypharmacology in drug discovery and
repurposing using the CANDO platform, Curr. Pharmaceut. Des. 22 (2016)
3109e3123.
ꢀ
[35] E. Cerrada, V. Fernandez-Moreira, M.C. Gimeno, Gold and platinum alkynyl
ꢀ
complexes for biomedical applications, in: P. Perez (Ed.), Advances in
[9] C. Marzano, V. Gandin, A. Folda, G. Scutari, A. Bindoli, M.P. Rigobello, Inhibition
of thioredoxin reductase by auranofin induces apoptosis in cisplatin-resistant
human ovarian cancer cells, Free Radical Biol. Med. 42 (2007) 872e881.
[10] T. Onodera, I. Momose, M. Kawada, Potential anticancer activity of auranofin,
Chem. Pharm. Bull. 67 (2019) 186e191.
[11] X.N. Zhang, K. Selvaraju, A.A. Saei, P. D'Arcy, R.A. Zubarev, E.S.J. Arner,
S. Linder, Repurposing of auranofin: thioredoxin reductase remains a primary
target of the drug, Biochimie 162 (2019) 46e54.
Organometallic Chemistry, Elsevier Academic Press, 2019, pp. 227e258.
[36] H.T. Liu, X.G. Xiong, P.D. Dau, Y.L. Wang, D.L. Huang, J. Li, L.S. Wang, Probing
the nature of gold-carbon bonding in gold-alkynyl complexes, Nat. Commun.
4 (2013) 2223, https://doi.org/10.1038/ncomms3223.
[37] S. Gunduz, A.C. Goren, T. Ozturk, Facile syntheses of 3-hydroxyflavones, Org.
Lett. 14 (2012) 1576e1579.
[38] D. Li, X. Hong, C.-M. Che, W.-C. Lo, S.-M. Peng, Luminescent gold(I) acetylide
complexes. Photophysical and photoredox properties and crystal structure of
[12] N. Liu, X. Li, H. Huang, C. Zhao, S. Liao, C. Yang, S. Liu, W. Song, X. Lu, X. Lan,
X. Chen, S. Yi, L. Xu, L. Jiang, C. Zhao, X. Dong, P. Zhou, S. Li, S. Wang, X. Shi,
P.Q. Dou, X. Wang, J. Liu, Clinically used antirheumatic agent auranofin is a
proteasomal deubiquitinase inhibitor and inhibits tumor growth, Oncotarget
5 (2014) 5453e5471.
[13] K. Patel, Z.S.O. Ahmed, X.M. Huang, Q.Q. Yang, E. Ekinci, C.M. Neslund-Dudas,
B. Mitra, F. Elnady, Y.H. Ahn, H.J. Yang, J.B. Liu, Q.P. Dou, Discovering protea-
somal deubiquitinating enzyme inhibitors for cancer therapy: lessons from
rational design, nature and old drug reposition, Future Med. Chem. 10 (2018)
2087e2108.
[14] A. Arlt, I. Bauer, C. Schafmayer, J. Tepel, S.S. Muerkoster, M. Brosch, C. Roder,
H. Kalthoff, J. Hampe, M.P. Moyer, U.R. Folsch, H. Schafer, Increased protea-
some subunit protein expression and proteasome activity in colon cancer
relate to an enhanced activation of nuclear factor E2-related factor 2 (Nrf2),
Oncogene 28 (2009) 3983e3996.
[15] R.G. Kenny, C.J. Marmion, Toward multi-targeted platinum and ruthenium
drugsda new paradigm in cancer drug treatment regimens? Chem. Rev. 119
(2019) 1058e1137.
[{Au(CCPh)}2(m-Ph2PCH2CH2PPh2)], J. Chem. Soc. Dalton Trans. (1993)
2929e2932.
[39] C.K. Lee, T. Uchida, K. Kitagawa, A. Yagi, N.S. Kim, S. Goto, Relationship be-
tween lipophilicity and skin permeability of various drugs from an ethanol/
water/lauric acid system, Biol. Pharm. Bull. 17 (1994) 1421e1424.
[40] E. García-Moreno, E. Cerrada, M.J. Bolsa, A. Luquin, M. Laguna, Water-soluble
phosphanes derived from 1,3,5-Triaza-7-phosphaadamantane and their
reactivity towards gold(I) complexes, Eur. J. Inorg. Chem. (2013) 2020e2030.
[41] F. Menaa, A. Menaa, B. Menaa, Chapter 65 - polyphenols nano-formulations
for topical delivery and skin tissue engineering, in: R.R. Watson,
V.R. Preedy, S. Zibadi (Eds.), Polyphenols in Human Health and Disease, Aca-
demic Press, San Diego, 2014, pp. 839e848.
[42] D. Carter, J.X. Ho, Advances in Protein Chemistry, Academic Press, New York,
1994.
[43] J.R. Lakowicz, In 'Principles of Fluorescence Spectroscopy', Springer, Baltimore,
Maryland, USA, 2006.
[44] H. Polet, J. Steinhardt, Binding-induced alterations in ultraviolet abssorption
of native serum albumin, Biochemistry 7 (1968) 1348e1356.
[16] E.B. Mojzer, M.K. Hrncic, M. Skerget, Z. Knez, U. Bren, Polyphenols: extraction
methods, antioxidative action, bioavailability and anticarcinogenic effects,
Molecules 21 (2016) 901e939.
[45] J.A. Molina-Bolivar, F. Galisteo-Gonzalez, C.C. Ruiz, M.M.O. Donnell, A. Parra,
Spectroscopic investigation on the interaction of maslinic acid with bovine
serum albumin, J. Lumin. 156 (2014) 141e149.
[17] C. Manach, A. Scalbert, C. Morand, C. Remesy, L. Jimenez, Polyphenols: food
sources and bioavailability, Am. J. Clin. Nutr. 79 (2004) 727e747.
[18] R.J. Nijveldt, E. van Nood, D.E.C. van Hoorn, P.G. Boelens, K. van Norren,
P.A.M. van Leeuwen, Flavonoids: a review of probable mechanisms of action
and potential applications, Am. J. Clin. Nutr. 74 (2001) 418e425.
[19] A.F. Almeida, G.I.A. Borge, M. Piskula, A. Tudose, L. Tudoreanu, K. Valentova,
G. Williamson, C.N. Santos, Bioavailability of quercetin in humans with a focus
on interindividual variation, Compr. Rev. Food Sci. F. 17 (2018) 714e731.
[20] R.V. Patel, B.M. Mistry, S.K. Shinde, R. Syed, V. Singh, H.S. Shin, Therapeutic
potential of quercetin as a cardiovascular agent, Eur. J. Med. Chem. 155 (2018)
889e904.
[21] R.L. Nagula, S. Wairkar, Recent advances in topical delivery of flavonoids: a
review, J. Control. Release 296 (2019) 190e201.
[22] J. Wang, X. Fang, L. Ge, F. Cao, L. Zhao, Z. Wang, W. Xiao, Antitumor, antiox-
idant and anti-inflammatory activities of kaempferol and its corresponding
glycosides and the enzymatic preparation of kaempferol, PLoS One 13 (2018),
e0197563.
[23] S.G. Darband, M. Kaviani, B. Yousefi, S. Sadighparvar, F.G. Pakdel, J.A. Attari,
I. Mohebbi, S. Naderi, M. Majidinia, Quercetin: a functional dietary flavonoid
with potential chemo-preventive properties in colorectal cancer, J. Cell.
Physicol. 233 (2018) 6544e6560.
[24] B. Butun, G. Topcu, T. Ozturk, Recent advances on 3-hydroxyflavone de-
rivatives: structures and properties, mini-review, Med. Chem. 18 (2018)
98e103.
[25] M. Asensi, A. Ortega, S. Mena, F. Feddi, J.M. Estrela, Natural polyphenols in
cancer therapy, Crit. Rev. Clin. Lab. Sci. 48 (2011) 197e216.
[26] F. Perez-Vizcaino, C.G. Fraga, Research trends in flavonoids and health, Arch.
Biochem. Biophys. 646 (2018) 107e112.
[46] J. Quero, S. Cabello, T. Fuertes, I. Marmol, R. Laplaza, V. Polo, M.C. Gimeno,
M.J. Rodriguez-Yoldi, E. Cerrada, Proteasome versus thioredoxin reductase
competition as possible biological targets in antitumor mixed thiolate-
dithiocarbamate gold(III) complexes, Inorg. Chem. 57 (2018) 10832e10845.
[47] I.M.C. Sanchez-de-Diego, R. Perez, Sonia Gascon, Ma J. Rodriguez-Yoldi,
E. Cerrada, The anticancer effect related to disturbances in redox balance on
Caco-2 cells caused by an alkynyl gold(I) complex, J. Inorg. Biochem. 166
(2017) 108e121.
[48] E. Atrian-Blasco, S. Gascon, M.J. Rodriguez-Yoldi, M. Laguna, E. Cerrada, Novel
gold(I) thiolate derivatives synergistic with 5-fluorouracil as potential selec-
tive anticancer agents in colon cancer, Inorg. Chem. 56 (2017) 8562e8579.
[49] E. Garcia-Moreno, A. Tomas, E. Atrian-Blasco, S. Gascon, E. Romanos,
M.J. Rodriguez-Yoldi, E. Cerrada, M. Laguna, In vitro and in vivo evaluation of
organometallic gold(I) derivatives as anticancer agents, Dalton Trans. 45
(2016) 2462e2475.
[50] I. Marmol, M. Virumbrales-Munoz, J. Quero, C. Sanchez-De-Diego,
L. Fernandez, I. Ochoa, E. Cerrada, M.J.R. Yoldi, Alkynyl gold(I) complex trig-
gers necroptosis via ROS generation in colorectal carcinoma cells, J. Inorg.
Biochem. 176 (2017) 123e133.
ꢀ
ꢀ
ꢀ
[51] E. Garcia-Moreno, S. Gascon, M.J. Rodriguez-Yoldi, E. Cerrada, M. Laguna, S-
propargylthiopyridine phosphane derivatives as anticancer agents: charac-
terization and antitumor activity, Organometallics 32 (2013) 3710e3720.
[52] E. Garcia-Moreno, S. Gascon, E. Atrian-Blasco, M.J. Rodriguez-Yoldi, E. Cerrada,
M. Laguna, Gold(I) complexes with alkylated PTA (1,3,5-triaza-7-
phosphaadamantane) phosphanes as anticancer metallodrugs, Eur. J. Med.
Chem. 79 (2014) 164e172.
ꢀ
ꢀ
[53] E. García-Moreno, S. Gascon, J.A.G.d. Jalon, E. Romanos, M.J. Rodriguez-Yoldi,
E. Cerrada, M. Laguna, In vivo anticancer activity, toxicology and histopath-
ological studies of the thiolate gold(I) complex [Au(Spyrimidine)(PTA-CH2Ph)]
Br, Anti-Cancer Agents, Med. Chem. 15 (2015) 773e782.
[27] S.C. Gupta, A.B. Kunnumakkara, S. Aggarwal, B.B. Aggarwal, Inflammation, a
double-edge sword for cancer and other age-related diseases, Front. Immunol.
9 (2018) 2160.
[28] Y. Wang, W. Wang, K.Z. Sanidad, P.A. Shih, X. Zhao, G. Zhang, Eicosanoid
signaling in carcinogenesis of colorectal cancer, Cancer Metastasis Rev. 37
(2e3) (2018) 257e267. Cancer Metastasis Rev., 37 (2018) 257-267.
[29] M. Grazul, E. Budzisz, Biological activity of metal ions complexes of chro-
mones, coumarins and flavones, Coord. Chem. Rev. 253 (2009) 2588e2598.
[54] G. Sudlow, D.J. Birkett, D.N. Wade, Further characterization of specific drug
binding sites on human serum albumin, Mol. Pharmacol. 12 (1976)
1052e1061.
[55] G.W. Zhang, Y.D. Ma, Mechanistic and conformational studies on the inter-
action of food dye amaranth with human serum albumin by multi-
spectroscopic methods, Food Chem. 136 (2013) 442e449.
[56] P. Wang, S.M. Henning, D. Heber, Limitations of MTT and MTS-based assays
for measurement of antiproliferative activity of green tea polyphenols, PLoS
One 5 (2010), e10202.
ꢂ
ꢀ
ꢀ
ꢀ
ꢀꢂ
[30] P. Mladenka, K. Macakova, T. Filipský, L. Zatloukalova, L. Jahodar, P. Bovicelli,
I.P. Silvestri, R. Hrdina, L. Saso, In vitro analysis of iron chelating activity of
flavonoids, J. Inorg. Biochem. 105 (2011) 693e701.
[31] A. Kurzwernhart, S. Mokesch, E. Klapproth, M.S. Adib-Ravazi, M.A. Jakupec,
C.G. Hartinger, W. Kandioller, B.K. Keppler, Flavonoid-based organometallics
with different metal centers - investigations of the effects on reactivity and
cytotoxicity, Eur. J. Inorg. Chem. (2016) 240e246.
[57] R.G. Balasingham, C.F. Williams, H.J. Mottram, M.P. Coogan, S.J.A. Pope, Gold(I)
complexes derived from alkynyloxy-substituted anthraquinones: syntheses,
luminescence, preliminary cytotoxicity, and cell imaging studies, Organome-
tallics 31 (2012) 5835e5843.
[32] R. Gaur, L. Mishra, Bi-nuclear Ru(II) complexes of bis-chalcone and bis-
flavonol: synthesis, characterization, photo cleavage of DNA and Topoisom-
erase I inhibition, RSC Adv. 3 (2013) 12210e12219.
[33] A. Kurzwernhart, W. Kandioller, S. Bachler, C. Bartel, S. Martic, M. Buczkowska,
G. Muhlgassner, M.A. Jakupec, H.B. Kraatz, P.J. Bednarski, V.B. Arion, D. Marko,
B.K. Keppler, C.G. Hartinger, Structure-activity relationships of targeted Ru-
II(eta(6)-p-Cymene) anticancer complexes with flavonol-derived ligands,
J. Med. Chem. 55 (2012) 10512e10522.
[58] A. Meyer, C.P. Bagowski, M. Kokoschka, M. Stefanopoulou, H. Alborzinia,
S. Can, D.H. Vlecken, W.S. Sheldrick, S. Wolfl, I. Ott, On the biological prop-
erties of alkynyl phosphine gold(I) complexes, Angew. Chem. Int. Ed. 51
(2012) 8895e8899.
[59] B. Bertrand, A. Casini, A golden future in medicinal inorganic chemistry: the
promise of anticancer gold organometallic compounds, Dalton Trans. 43
(2014) 4209e4219.
[60] C. Sanchez-de-Diego, I. Marmol, R. Perez, S. Gascon, M.J. Rodriguez-Yoldi,
E. Cerrada, The anticancer effect related to disturbances in redox balance on
[34] N. Mirzadeh, S.H. Priver, A. Abraham, R. Shukla, V. Bansal, S.K. Bhargava,

ꢀ
I. Marmol et al. / European Journal of Medicinal Chemistry 183 (2019) 111661
14
Caco-2 cells caused by an alkynyl gold(I) complex, J. Inorg. Biochem. 166
(2017) 108e121.
cell death by aurothioglucose, J. Biol. Chem. 287 (2012) 38210e38219.
[77] J. Lu, L.V. Papp, J. Fang, S. Rodriguez-Nieto, B. Zhivotovsky, A. Holmgren, In-
hibition of Mammalian thioredoxin reductase by some flavonoids: implica-
tions for myricetin and quercetin anticancer activity, Cancer Res. 66 (2006)
4410e4418.
[78] K. Zhang, E.B. Yang, W.Y. Tang, K.P. Wong, P. Mack, Inhibition of glutathione
reductase by plant polyphenols, Biochem. Pharmacol. 54 (1997) 1047e1053.
[79] E. Jortzik, M. Farhadi, R. Ahmadi, K. Toth, J. Lohr, B.M. Helmke, S. Kehr,
A. Unterberg, I. Ott, R. Gust, V. Deborde, E. Davioud-Charvet, R. Reau, K. Becker,
C. Herold-Mende, Antiglioma activity of GoPI-sugar, a novel gold(I)-phosphole
inhibitor: chemical synthesis, mechanistic studies, and effectiveness in vivo,
Biochim. Biophys. Acta 1844 (2014) 1415e1426.
[80] I. Villegas, M.J. Martin, C. La Casa, V. Motilva, C.A. De La Lastra, Effects of
oxicam inhibitors of cyclooxygenase on oxidative stress generation in rat
gastric mucosa. A comparative study, Free Radic. Res. 36 (2002) 769e777.
[81] C. Nicco, F. Batteux, ROS modulator molecules with therapeutic potential in
cancers treatments, Molecules 23 (2018) 84e100.
[82] M. Altaf, N. Casagrande, E. Mariotto, N. Baig, A.N. Kawde, G. Corona, R. Larcher,
C. Borghese, C. Pavan, A.A. Seliman, D. Aldinucci, A.A. Isab, Potent in vitro and
in vivo anticancer activity of new bipyridine and bipyrimidine gold (III)
dithiocarbamate derivatives, Cancers 11 (2019) 14.
[61] R.B. Badisa, S.F. Darling-Reed, P. Joseph, J.S. Cooperwood, L.M. Latinwo,
C.B. Goodman, Selective cytotoxic activities of two novel synthetic drugs on
human breast carcinoma MCF-7 cells, Anticancer Res. 29 (2009) 2993e2996.
[62] K. Vidyalakshmi, P. Kamalakannan, S. Viswanathan, S. Ramaswamy, Anti-in-
flammatory effect of certain dihydroxy flavones and the mechanisms
involved, Anti-Inflammatory Anti-Allergy Agents Med. Chem. 11 (2012)
253e261.
[63] R.J. Meshram, K.T. Bagul, S.P. Pawnikar, S.H. Barage, B.S. Kolte, R.N. Gacche,
Known compounds and new lessons: structural and electronic basis of
flavonoid-based bioactivities, J. Biomol. Struct. Dyn. (2019) 1e17.
[64] H.S. Youn, J.Y. Lee, S.I. Saitoh, K. Miyake, D.H. Hwang, Auranofin, as an anti-
rheumatic gold compound, suppresses LPS-induced homodimerization of
TLR4, Biochem. Biophys. Res. Commun. 350 (2006) 866e871.
[65] I. Ott, T. Koch, H. Shorafa, Z. Bai, D. Poeckel, D. Steinhilber, R. Gust, Synthesis,
cytotoxicity, cellular uptake and influence on eicosanoid metabolism of
cobalt-alkyne modified fructoses in comparison to auranofin and the cyto-
toxic COX inhibitor Co-ASS, Org. Biomol. Chem. 3 (2005) 2282e2286.
[66] S. Roscales, N. Bechmann, D.H. Weiss, M. Kockerling, J. Pietzsch, T. Kniess,
Novel valdecoxib derivatives by ruthenium(ii)-promoted 1,3-dipolar cyclo-
addition of nitrile oxides with alkynes - synthesis and COX-2 inhibition ac-
tivity, Medchemcomm 9 (2018) 534e544.
[67] G. Rubner, K. Bensdorf, A. Wellner, B. Kircher, S. Bergemann, I. Ott, R. Gust,
Synthesis and biological activities of transition metal complexes based on
acetylsalicylic acid as neo-anticancer agents, J. Med. Chem. 53 (2010)
6889e6898.
[83] M.W. Sung, D.Y. Lee, S.W. Park, S.M. Oh, J.J. Choi, E.S. Shin, S.K. Kwon, S.H. Ahn,
Y.H. Kim, Celecoxib enhances the inhibitory effect of 5-FU on human squa-
mous cell carcinoma proliferation by ROS production, The Laryngoscope 127
(2017) E117eE123.
[84] J.N. Boodram, I.J. McGregor, P.M. Bruno, P.B. Cressey, M.T. Hemann,
K. Suntharalingam, Breast cancer stem cell potent copper(II)-Non-Steroidal
anti-inflammatory drug complexes, Angew Chem. Int. Ed. Engl. 55 (2016)
2845e2850.
[68] U.S. Akarca, Gastrointestinal effects of selective and non-selective non-ste-
roidal anti-inflammatory drugs, Curr. Pharmaceut. Des. 11 (2005) 1779e1793.
[69] S.H. Bae, E.S. Jung, Y.M. Park, B.S. Kim, B.K. Kim, D.G. Kim, W.S. Ryu, Expression
of cyclooxygenase-2 (COX-2) in hepatocellular carcinoma and growth inhi-
bition of hepatoma cell lines by a COX-2 inhibitor, NS-398, Clin. Cancer Res. 7
(2001) 1410e1418.
ꢀ
[85] B.T. Elie, J. Fernandez-Gallardo, N. Curado, M.A. Cornejo, J.W. Ramos,
M. Contel, Bimetallic titanocene-gold phosphane complexes inhibit invasion,
metastasis, and angiogenesis-associated signaling molecules in renal cancer,
Eur. J. Med. Chem. 161 (2019) 310e322.
[70] H. Kamitani, M. Geller, T. Eling, Expression of 15-lipoxygenase by human
colorectal carcinoma Caco-2 cells during apoptosis and cell differentiation,
J. Biol. Chem. 273 (1998) 21569e21577.
[71] E. Half, X.M. Tang, K. Gwyn, A. Sahin, K. Wathen, F.A. Sinicrope, Cyclo-
oxygenase-2 expression in human breast cancers and adjacent ductal carci-
noma in situ, Cancer Res. 62 (2002) 1676e1681.
[72] Y. Cheng, Y. Qi, Current progresses in metal-based anticancer complexes as
mammalian TrxR inhibitors, Anti Cancer Agents Med. Chem. 17 (2017)
1046e1069.
[86] H. Zhou, M. Yuan, Q. Yu, X. Zhou, W. Min, D. Gao, Autophagy regulation and its
role in gastric cancer and colorectal cancer, Cancer Biomark. 17 (2016) 1e10.
[87] Y. Shi, B. Tang, P.W. Yu, B. Tang, Y.X. Hao, X. Lei, H.X. Luo, D.Z. Zeng, Autophagy
protects against oxaliplatin-induced cell death via ER stress and ROS in Caco-2
cells, PLoS One 7 (2012), e51076.
[88] Z. Assefa, B.G. McBurnett, R.J. Staples, J.P. Fackler, Structures and spectroscopic
properties of gold(I) complexes of 1,3,5-triaza-7-phosphaadamantane (tpa) .2.
Multiple-state emission from (tpa)AuX (X¼Cl, Br, I) complexes,, Inorg. Chem.
34 (1995) 4965e4972.
[73] A. Bindoli, M.P. Rigobello, G. Scutari, C. Gabbiani, A. Casini, L. Messori, Thio-
redoxin reductase: a target for gold compounds acting as potential anticancer
drugs, Coord. Chem. Rev. 253 (2009) 1692e1707.
[89] C.A. McAuliffe, R.V.D. Parish, P.D. Randall, Gold(I) complexes of unidentate and
bidentate phosphorus-, arsenic-, antimony-, and sulphur-donor ligands,
J. Chem. Soc., Dalton Trans. (1979) 1730e1735.
[74] N. Couto, J. Wood, J. Barber, The role of glutathione reductase and related
enzymes on cellular redox homoeostasis network, Free Radic. Biol. Med. 95
(2016) 27e42.
[75] X. Yan, X. Zhang, L. Wang, R. Zhang, X. Pu, S. Wu, L. Li, P. Tong, J. Wang,
Q.H. Meng, V.B. Jensen, L. Girard, J.D. Minna, J.A. Roth, S.G. Swisher,
J.V. Heymach, B. Fang, Inhibition of thioredoxin/thioredoxin reductase induces
synthetic lethality in lung cancers with compromised glutathione homeo-
stasis, Cancer Res. 79 (2019) 125e132.
[90] G. Sheldrick, Crystal structure refinement with SHELXL, Acta Crystallogr. C 71
(2015) 3e8.
[91] I. Marmol, M. Virumbrales-Munoz, J. Quero, C. Sanchez-de-Diego,
L. Fernandez, I. Ochoa, E. Cerrada, M.J.R. Yoldi, Alkynyl gold(I) complex trig-
gers necroptosis via ROS generation in colorectal carcinoma cells, J. Inorg.
Biochem. 176 (2017) 123e133.
[92] S. Jimenez, S. Gascon, A. Luquin, M. Laguna, C. Ancin-Azpilicueta,
M.J. Rodriguez-Yoldi, Rosa canina extracts have antiproliferative and antiox-
idant effects on caco-2 human colon cancer, PLoS One 11 (2016), e0159136.
[76] Y. Du, H. Zhang, J. Lu, A. Holmgren, Glutathione and glutaredoxin act as a
backup of human thioredoxin reductase 1 to reduce thioredoxin 1 preventing