Enhancement of glioblastoma multiforme therapy through a novel Quercetin-Losartan hybrid

Article abstract of DOI:10.1016/j.freeradbiomed.2020.08.007

Glioblastoma multiforme (GBM) is the most common and aggressive primary malignant brain tumor. Maximal surgical resection followed by radiotherapy and concomitant chemotherapy with temozolomide remains the first-line therapy, prolonging the survival of patients by an average of only 2.5 months. There is therefore an urgent need for novel therapeutic strategies to improve clinical outcomes. Reactive oxygen species (ROS) are an important contributor to GBM development. Here, we describe the rational design and synthesis of a stable hybrid molecule tethering two ROS regulating moieties, with the aim of constructing a chemopreventive and anticancer chemical entity that retains the properties of the parent compounds. We utilized the selective AT₁R antagonist losartan, leading to the inhibition of ROS levels, and the antioxidant flavonoid quercetin. In GBM cells, we show that this hybrid retains the binding potential of losartan to the AT₁R through competition-binding experiments and simultaneously exhibits ROS inhibition and antioxidant capacity similar to native quercetin. In addition, we demonstrate that the hybrid is able to alter the cell cycle distribution of GBM cells, leading to cell cycle arrest and to the induction of cytotoxic effects. Last, the hybrid significantly and selectively reduces cancer cell proliferation and angiogenesis in primary GBM cultures with respect to the isolated parent components or their simple combination, further emphasizing the potential utility of the current hybridization approach in GBM.

Full text of DOI:10.1016/j.freeradbiomed.2020.08.007

FreeRadicalBiologyandMedicine160(2020)391–402
Contents lists available at ScienceDirect
Free Radical Biology and Medicine
journal homepage: www.elsevier.com/locate/freeradbiomed
Original article
Enhancement of glioblastoma multiforme therapy through a novel
Quercetin-Losartan hybrid
Antonis D. Tsiailanisa,1, Alexander Renziehausenb,1, Sofia Kiriakidic, Eirinaios I. Vrettosa,
Georgios S. Markopoulosd,e, Nisar Sayyada, Baydaa Hirmizf, Marie-Isabel Aguilarf,
Mark P. Del Borgof, Evangelos Kolettasd,e, Robert E. Widdopg, Thomas Mavromoustakosc,
Tim Crookb,∗∗, Nelofer Syedb,∗∗∗, Andreas G. Tzakosa,h,∗
a University of Ioannina, Section of Organic Chemistry and Biochemistry, Department of Chemistry, Ioannina, Greece
b John Fulcher Neuro-Oncology Laboratory, Imperial College London, Hammersmith Hospital, London, UK
c National and Kapodistrian University of Athens, Department of Chemistry, Athens, Greece
d University of Ioannina, School of Medicine, Faculty of Health Sciences, Laboratory of Biology, University Campus, 45110, Ioannina, Greece
e Institute of Molecular Biology and Biotechnology, Foundation for Research and Technology, Biomedical Research Division, 45115, Ioannina, Greece
f Monash Biomedicine Discovery Institute and Department of Biochemistry and Molecular Biology, Monash University, Clayton, VIC, 3800, Australia
g Monash Biomedicine Discovery Institute and Department of Pharmacology, Monash University, Clayton, VIC, 3800, Australia
h University Research Center of Ioannina (URCI), Institute of Materials Science and Computing, Ioannina, Greece
A R T I C L E I N F O
A B S T R A C T
Keywords:
Glioblastoma multiforme (GBM) is the most common and aggressive primary malignant brain tumor. Maximal
surgical resection followed by radiotherapy and concomitant chemotherapy with temozolomide remains the
first-line therapy, prolonging the survival of patients by an average of only 2.5 months. There is therefore an
urgent need for novel therapeutic strategies to improve clinical outcomes. Reactive oxygen species (ROS) are an
important contributor to GBM development. Here, we describe the rational design and synthesis of a stable
hybrid molecule tethering two ROS regulating moieties, with the aim of constructing a chemopreventive and
anticancer chemical entity that retains the properties of the parent compounds. We utilized the selective AT1R
antagonist losartan, leading to the inhibition of ROS levels, and the antioxidant flavonoid quercetin. In GBM
cells, we show that this hybrid retains the binding potential of losartan to the AT1R through competition-binding
experiments and simultaneously exhibits ROS inhibition and antioxidant capacity similar to native quercetin. In
addition, we demonstrate that the hybrid is able to alter the cell cycle distribution of GBM cells, leading to cell
cycle arrest and to the induction of cytotoxic effects. Last, the hybrid significantly and selectively reduces cancer
cell proliferation and angiogenesis in primary GBM cultures with respect to the isolated parent components or
their simple combination, further emphasizing the potential utility of the current hybridization approach in
GBM.
Reactive oxygen species (ROS)
Glioblastoma multiforme (GBM)
Antioxidants
AT1R
AT2R
Quercetin
Losartan
1. Introduction
followed by radiotherapy and chemotherapy with temozolomide [2].
However, adjuvant chemotherapy adds only an average of 2.5 months
Glioblastoma multiforme (GBM) is the most common and lethal
primary malignant brain tumor among adults [1] with an increasing
incidence. Standard of care management for GBM is surgical resection
to the median overall survival of the patients, highlighting the need for
novel and more effective therapeutic approaches. The obvious in-
adequacies of the current agents used against GBM, along with the high
Abbreviations: AII, Angiotensin 2; AT1R/AT2R, Angiotensin subtype 1/2 Receptor; RAS, renin-angiotensin system; ARB, Angiotensin II Receptor Blocker; BBB,
blood-brain barrier; Bcl, B-cell lymphoma; DCFDA, 2′,7′–dichlorofluorescin diacetate; GBM, Glioblastoma multiforme; NADH, nicotinamide adenine dinucleotide;
ROS, reactive oxygen species; Q-L, Quercetin-Losartan
∗ Corresponding author.
∗∗ Corresponding author.
∗∗∗ Corresponding author.
E-mail addresses: tr.crook@gmail.com (T. Crook), n.syed@imperial.ac.uk (N. Syed), atzakos@uoi.gr (A.G. Tzakos).
1 Equal contribution.
https://doi.org/10.1016/j.freeradbiomed.2020.08.007
Received 22 June 2020; Received in revised form 31 July 2020; Accepted 8 August 2020
Availableonline18August2020
0891-5849/©2020ElsevierInc.Allrightsreserved.

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
cost of de novo drug discovery, imply an urgent need to explore alter-
native strategies to identify effective therapies. Repurposing of drugs
represents one such strategy that can both accelerate the drug discovery
process and minimize the overall development costs of new therapeutic
options.
important in cell signaling and thus its inhibition could confer anti-
oxidant effects [19]. AT1R blockers (ARBs) also have anti-angiogenic
properties [20,21]. Losartan, the prototype of sartan class, is one of the
most commonly used AT1R blockers. Chauhan et al. have reported its
anti-proliferative effects in several cancer cell lines [22] and Pender-
grass et al. showed that AII stimulated the ROS production, an effect
which was abolished by losartan [19]. Furthermore, it has recently been
shown that losartan enhances the delivery of chemotherapeutics by
altering tumor microenvironment leading to enhanced response to
chemotherapy [23,24]. Notably, a phase II trial has been conducted in
locally advanced pancreatic cancer with a combined treatment of
chemoradiation and losartan [25]. The improved drug response to lo-
sartan seems to result from decompression of blood vessels [22] and
improved vascular perfusion, reduction of the extracellular matrix,
depletion of collagen and associated physical barriers that normally
hamper drug delivery [23,24]. This is of great importance since the
abnormal vascular structure is a major factor limiting the efficacy of
radiotherapy and chemotherapy in GBM.
Gliomagenesis is a multistep process, involving several factors
leading to changes in cell physiology and homeostasis. An important
factor in GBM development is the levels of reactive oxygen species
(ROS). ROS are formed under physiological conditions due to the par-
tial reduction of molecular oxygen, but elevated levels induce cellular
stress, promote cell and macromolecule damage and contribute to tu-
morigenesis [3,4]. In normal cells, there is a balance between levels of
ROS and natural antioxidants [5]. In contrast, cancer cells are known to
exist under increased and persistent oxidative stress due to elevated
levels of ROS and a resulting cellular redox imbalance. Redox im-
balance has been described in various cancers including aggressive
types such as GBM. A number of factors drive redox imbalance in-
cluding oncogenic stimulation, increased metabolic activity, and mi-
tochondrial malfunction [6]. Free radicals, and particularly, hydrogen
peroxide, cause tissue damage, facilitating tumor growth and invasion
[7]. The central nervous system is particularly susceptible to the da-
maging effects of ROS [8,9] because of its increased metabolic rate and
reduced capacity for cellular regeneration. Furthermore, redox metals
are implicated in carcinogenesis, potentially attributable to their ability
to create ROS or to their capacity to be tethered with thiols [10]. A
number of studies have shown that ROS sensitize cancer cells to che-
motherapeutic agents. As a result, there is considerable current interest
into the use of ROS-producing molecules in GBM (called oxidative
therapy), although their use is associated with significant toxicity.
Conversely, antioxidant supplementation is less favoured because of the
potential reduction of chemotherapeutic efficacy and less favourable
time-dependent clinical outcomes [11]. Representative examples of
well-known flavonoids which can act as chemopreventive agents by
protecting glial cells from oxidative stress, excitotoxicity and neuroin-
flammation are quercetin, catechins and proanthocyanidins [12]. Op-
timal levels of ROS for higher survival rates are not yet known.
Therefore, the discovery of novel chemopreventive compounds that
could regulate ROS levels and also exert a chemotherapeutic effect is of
high clinical importance. Of note, antioxidants are a class of pharma-
ceutical compounds that not only decrease ROS levels but also possess
synergistic antitumor effects when used together with other che-
motherapeutic agents such as temozolomide [13].
Besides losartan, that has been identified to be a regulator of ROS,
natural products have served as a rich and profitable source of anti-
oxidants and drugs [26]. Quercetin, one of the most abundant natural
product, is considered as a robust antioxidant because it scavenges free
radicals and binds to transition metal ions [27]. Interestingly, quercetin
can modulate oxidative stress by regulating several signaling pathways
resulting in cell cycle arrest and cell apoptosis [28–31]. It has also been
shown to be an anticancer agent because of its ability to alter cell
proliferation, differentiation, and survival rate by targeting a variety of
key molecules responsible for tumor cell growth In particular, our
group recently reported that quercetin directly binds to the Bcl-2 family
proteins and triggers their pro-apoptotic activity [32]. We have also
developed quercetin prodrugs that could enhance its Bcl-2/Bcl-xL
binding profile [33,34]. Numerous studies have revealed that quercetin
may inhibit processes vital to the malignant growth of GBM such as cell
proliferation, survival and invasion [35,36], as also that it may have
therapeutic synergy with other agents by sensitizing GBM cells [37,38].
In addition, Choi et al., reported that flavonoids enhance the bioavail-
ability of losartan in rats [39]. Notably, quercetin can effectively pe-
netrate the blood–brain barrier (BBB) inducing apoptosis [40,41] and
consequently, it could be a potent agent against GBM.
Towards this end, we hypothesized that tethering the AT1R blocker
losartan with quercetin could lead to a novel compound with enhanced
pharmaceutical profile. This hybrid (designated Q-L), shown in Scheme
1, could prove to be an alternative strategy against GBM, mainly by
modulating the cellular ROS levels. Specifically, combining the anti-
oxidant ability of quercetin and the binding of losartan on the AT1R
(resulting in reduced levels of ROS), Q-L could regulate the overall le-
vels of ROS present in GBM tumors. This is of great significance to
maintain the cellular redox balance, since ROS imbalance is one of the
hallmarks of cancer [42]. Tight constrain of ROS is essential, as mod-
erate levels of ROS assist the control of cell proliferation and differ-
entiation [43]. Also, taking into consideration the ability of losartan to
The renin-angiotensin system (RAS) has been extensively studied as
a biochemical pathway involved in carcinogenesis and progression of
GBM [14]. Numerous studies have demonstrated the functions of the
RAS in cancer in general and in the development of glioma [15]. An-
giotensin II (AII) has two main receptors, the subtype 1 receptor (AT1R)
and the subtype 2 receptor (AT2R). Activation of the AT1R leads to
tumor cell proliferation and angiogenesis [16,17] and its inhibition is
considered a potential target for drug development. Activation of the
AT1R stimulates the production of endogenous ROS [18] which are
Scheme 1. Structures of A. Quercetin, B. Losartan and C. the Q-L hybrid.
392

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
alter the tumor microenvironment and enhance response to che-
motherapeutics, along with the anticancer activity of quercetin, we
hypothesized that Q-L could function as an effective therapy against
GBM.
we have shown that quercetin can complexate with Zinc (II) on the CO-
4 carbonyl oxygen and the deprotonated C5–OH, which is of im-
portance as Zinc (II) has antagonistic role with iron and copper and
prevents the deleterious free radical reactions (e.g., Fenton reaction)
[53].
Along these lines, we rationally designed the Q-L hybrid with the
aim to retain the characteristics of the isolated components. First, the
synthesized hybrid was validated for its binding potential to the AT1R
through radioligand competition-binding experiments in HEK-293 cells,
stably transfected with either the AT1R or the AT2R. To probe the in-
tracellular antioxidant capacity of the hybrid with respect to the parent
compounds, the intracellular ROS levels were quantified by flow cyto-
metry analysis following 2′,7′–dichlorofluorescin diacetate (DCFDA)
staining. Also, its efficacy against established GBM cell lines and pri-
mary explants derived from neurosurgically-resected GBM was ex-
plored. In addition, its ability to inhibit proliferation and angiogenesis
was elucidated and the cell cycle distribution of glioblastoma cells was
also evaluated using flow cytometric analysis and was compared with
the parent compounds.
Having determined that ring-B is the proper conjugation site, we
next considered the linkage moiety. Taking into account the superior
BBB passage of hydrophobic molecules relative to hydrophilic ones
[54,55], we aimed to construct a lypophilic hybrid. We have previously
shown that ester derivatization of rosmarinic acid increase its perme-
ability and potency [56]. We therefore hypothesized that building an
ester bond on the targeted bioconjugate would render the new hybrid
more efficient in crossing the BBB. We also calculated the LogP of
quercetin, losartan and the hybrid Q-L and as expected we found the
hybrid to be more lipophilic than the isolated components (Quercetin:
LogP = 0.35; Losartan: LogP = 4.95; Q-L: LogP = 5.49).
Having determined the conjugation site and the linkage type be-
tween the two components we then efforted to validate these first with
in silico studies. The designed molecular hybrid should be able to retain
the key interactions with the AT1R, in order not to abolish its affinity.
The in silico results initially confirm that additional hydrophobic space
existed in the AT1R ligand binding groove that is able to host a losartan
hybrid. By trial and error, a hybrid of losartan with quercetin was found
to fit into an extra hydrophobic cavity giving a higher docking score,
when compared with quercetin or losartan alone. Based on these ob-
servations, a Q-L hybrid was designed by tethering losartan and quer-
cetin with a succinate linker that could also stimulate the release of the
parent components after hydrolysis (Scheme 1) [57]. Τhe conforma-
tional analysis for each of the three ligands (i.e. losartan, quercetin and
Q-L) produced 270 structures from which the five lowest in energy for
each were chosen as input for induced fit docking. For the hybrid Q-L,
both possible isomers were evaluated, i.e. para or meta to losartan as
presented in Fig. 1. In total 20 (4 × 5) initial conformers were docked
to AT1R activated and AT1R inactivated. Induced fit docking (IFD) with
Glide generated 10 conformer structures which were sorted based on
decreasing lowest potential energy according to IFD score, which con-
stitutes a combination of GScore and Prime energy of the created pro-
tein-ligand complex. Selection of the best conformer was based on the
lowest energy score. The XP GScore which represents the binding en-
ergy for each ligand/receptor complex is presented in Table 1 for each
ligand docked in AT1R activated and AT1R inactivated forms.
2. Results and discussion
2.1. Rational design and in silico evaluation of losartan, quercetin and the
hybrid to bind the AT1R
Starting from the selection of the two components for the molecular
hybridization approach, we chose two compounds that are able to pe-
netrate the blood–brain barrier (BBB), and thus, could be effective
against GBM. First, we selected quercetin based on our experience of
developing relevant hybrids [33,44] and due to its ability to effectively
penetrate the BBB inducing apoptosis [41]. The second component,
losartan, also crosses the BBB acting as an AT1R blocker [45], which is
overexpressed in aggressive tumors, including GBM [46].
Next, the most crucial factor that should be examined is the con-
jugation site. The molecular hybridization approach requires proper
design, since the conjugation of two entities does not directly result in
additive activity of the isolated components. Taking into consideration
the metabolic pathways of both components, we ended up tethering the
ring-B of quercetin, which is susceptible to metabolism through O-
methylation via catechol-O-methyltransferase and glucuronidation via
uridine diphosphate glucuronosyltransferases, and the primary alcohol
of losartan which is converted to an aldehyde intermediate E-3179,
which has no AT1R blocking ability [47]. In order to further validate
this hypothesis, we utilized the TMIC biotransformation software [48],
indicating that the designed Q-L hybrid leads to reduced number of
metabolites with respect to the parent compounds (6 metabolites for Q-
L against 10 metabolites in total for quercetin and losartan). Although
the catechol group of quercetin is a key antioxidant moiety, it has been
proved that other groups of quercetin may contribute to its antioxidant
capacity. Specifically, the simultaneous presence of the C3–OH (ring-C)
and C5–OH (ring-A) groups and the 2,3-double bond (ring-C) in con-
jugation with the CO-4 carbonyl group contribute to the overall anti-
oxidant effects of quercetin [49]. The specific groups are also re-
sponsible for the effective DNA protection by quercetin [50]. In
addition to these, our group has exploited the antioxidant capacity of
each –OH group of quercetin [51]. We concluded that after the oxida-
tion of the catechol moiety in quercetin, the C3–OH is lost and cannot
be involved in the antioxidant effects of quercetin and its bio-
transformation. However, when the catechol unit is blocked through
conjugation, the C3–OH plays a pivotal role in the antioxidant prop-
erties of quercetin. Notably, we also observed oxidation of the re-
sorcinol moiety at ring A, which also pinpoints the oxidation potential
of the C5–OH of ring-A. In addition, transition metal chelators, and
especially iron, function as antioxidants by scavenging free ROS, re-
The docking results indicate that the Q-L hybrid (in both isomers)
binds strongly to the AT1R, both in its inactivated (Fig. 1 for the meta-
isomer and Fig. S1 for the para-isomer) and activated (Fig. 2 for the
meta-isomer and Fig. S2 for the para-isomer) crystal structures. In
particular, it exerts stronger binding than either of its parent molecules
losartan and quercetin. The stronger binding applies both to the acti-
vated and the inactivated versions of AT1R. However, the meta-sub-
stituted isomer seems to bind more strongly to the inactivated structure
of AT1R, whereas the para-substituted isomer binds slightly more
strongly to the activated one.
Regarding the prevalent ligand-receptor interactions, both hybrid
isomers retain the key losartan-AT1R interactions, as shown in Figs. 1
and 2 for the meta isomer, and Figs. S1 and S2 for the para isomer.
However, when docked to the inactive AT1R, the hydrogen bonding
with Arg167 is preserved but it is between the carbonyl group of the
linker and the residue, instead of the tetrazole group of losartan as in
native losartan. In addition, the pi-pi stacking with Trp84 from the
imidazole ring is similar to the native losartan [58], while forming
additional ones that arise from the quercetin part of the molecule, such
as hydrogen bonding and pi-cation interactions with several amino
acids, as depicted in Figs. 1 and 2. The same interactions are observed
in the case of the para isomer presented in Figs. S1 and S2.
•
ducing the amount of iron and therefore decrease the OH levels gen-
erated by Fenton reactions [52]. Quercetin is known to chelate transi-
tion metals, contributing to its antioxidant capability. Along these lines,
393

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
Fig. 1. Ligand Interaction Diagrams (LID) and 3D structure of the Q-L hybrid meta-isomer docked to the inactivated AT1R (pdb ID: 4YAY).
the AT1R binding site with good affinity, we then synthesized the hy-
brid as depicted in Scheme 2.
Table 1
Binding Energy (kcal/mol) for hybrid, losartan and quercetin to AT1R (in both
activated and inactivated form).
Ligand
The designed hybrid Q-L was achieved through coupling of the
primary hydroxyl group of losartan and the phenolic group of quercetin
(3′ resulting in the major product or 4′ resulting in the minor product)
via ester bond, using hemisuccinate as a linker. To achieve the synthesis
of the relevant prodrug, the tetrazole ring of losartan was first protected
with the trityl group [59]. Trityl-protected losartan 2 was converted to
its hemisuccinate derivative 3 at the primary –OH using DIPEA and
succinic anhydride [60]. The esterification reaction of 3 with quercetin
(3′-OH, major) was conducted using EDCI. HCl and HOBt in DMF [61].
The reaction resulted in 4 as a mixture of regioisomers of quercetin at 3′
and 4′ positions at a ratio of 3:1 (Fig. S3), which is common for the
catechol group of quercetin [62]. Prior to the final acidic (CH2Cl2/TFA/
TIS, 8.5/1/0.5 v/v) deprotection of the trityl group, the compound was
purified with silica gel column chromatography.
XP GScore (kcal/mol)
AT1R activated
AT1R inactivated
Hybrid meta-substituted
Hybrid para-substituted
Losartan
−15.445
−15.946
−9.601
−15.225
−14.147
−10.218
−8.051
Quercetin
−10.637
2.2. Chemical synthesis of the molecular hybrid Q-L
Having determined from the rational design studies, the docking
calculations and the potential of the Q-L hybrid to beneficially occupy
Fig. 2. Ligand Interaction Diagrams (LID) and 3D structure of the Q-L hybrid meta-isomer docked to the activated AT1R (pdb ID: 6DO1).
394

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
Scheme 2. Chemical synthesis of the Q-L hybrid. Reagents and conditions: (a) Chlorotriphenylmethane, triethylamine, CH2Cl2, rt, 2 h; (b) Succinic anhydride,
DIPEA, acetonitrile, rt, 12 h; (c) Quercetin, EDCI. HCl, HOBt, DMF, 60 °C, 1 h; (d) CH2Cl2/TFA/TIS (8.5/1/0.5 v/v), 0 °C to rt, 4 h.
Molecular hybrids should present adequate chemical stability to
prevent an unwanted premature drug release. In order to evaluate the
chemical stability of the Q-L hybrid, we first established an HPLC
method that quercetin, losartan and the Q-L hybrid are identified in
different retention times (Fig. S4). Then, to monitor the stability profile
of Q-L and the potential degradation and consequent formation of the
parent components over time, we performed chemical stability ex-
periments in buffer pH 7.4. The hybrid proved to be stable at 37 °C even
after 8 h of incubation (Fig. S5).
2.3. Probing the binding to the receptor site by radioligand binding studies
In order to validate the in silico findings, we explored the binding
affinity of the synthesized hybrid to the AT1R and AT2R in radioligand
competition-binding experiments in HEK-293 cells, stably transfected
with plasmids encoding either AT1R or AT2R, as previously described
[63]. [125I]-[Sar1, Ile8]AII was used as the radioligand in the competi-
tion assays at increasing concentrations of ligand. The results shown in
Table 2 and in Fig. 3 indicate that losartan and the Q-L hybrid bind
avidly to the AT1R but not to the AT2R, highlighting that we achieved to
retain, through this bioconjugation the binding and selectivity for the
AT1R.
2.4. Q-L hybrid retains the antioxidant properties of quercetin in GBM cell
lines
Fig. 3. Radioligand competition-binding experiments performed in HEK-
293 cells stably expressing (a) AT1R or (b) AT2R where increasing concentra-
tions of modified non-peptide analogues were tested against [125I]-[Sar1,
Ile8]AII binding. Each point represents the mean for three separate experiments,
each performed in triplicate. IC50 values are listed in Table 2.
Quercetin is an effective antioxidant [64], a property that in part
contributes to its anticancer effects [65]. To determine the potency of
Q-L hybrid to mimic the antioxidant effects of quercetin, U87 and T98G
GBM cell lines were treated for 24 h with or without 200 μΜ Η2Ο2 and
with either 30 μΜ of losartan, quercetin, their combination and Q-L
hybrid. ROS levels were quantified by flow cytometry analysis fol-
lowing 2′,7′–dichlorofluorescin diacetate (DCFDA) staining (Fig. 4).
First, we compared ROS production of all compounds in the absence of
H2O2 and found that they had no significant effect compared to control
cells. Next, we analyzed ROS levels in the presence of H2O2. Losartan
had no significant effect on ROS production. However, we confirmed
that quercetin alone and in combination with losartan resulted in a
significant inhibition in ROS production (p < 0.05), 27.5% and 24% or
Table 2
Binding affinity values presented in IC50 values (M) of losartan and Q-L hybrid in comparison with the native ligand AII.
Compound
AT1R–IC50 (M)
Standard Error of the mean (SEM)
AT2R–IC50 (M)
Standard Error of the mean (SEM)
0,09114 × 10−10
AII
1,66 × 10−9
1,03 × 10−8
1,4 × 10−7
0,2452 × 10−9
0,1122 × 10−8
0,1058 × 10−7
1,21 × 10−10
> 10 μM
Losartan
Q-L Hybrid
> 10 μM
*Results are the competition binding from three separate experiments (each in triplicate).
395

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
Fig. 4. Q-L hybrid demonstrates similar antioxidant capacity and ROS inhibition, in comparison with quercetin in GBM cell lines. GBM cell lines U87 and
T98G were treated with 30 μM of either losartan (L), quercetin (Q), losartan together with quercetin (L + Q) or the Q-L hybrid, with or without co-treatment with
200 μΜ Η2Ο2. 24 h post-treatment oxidized DCFDA was analyzed using flow cytometry. The mean fluorescence of three independent experiments (n = 3) is
presented. The results of significant differences by one-way ANOVA with Tukey's multiple comparisons test are presented (*P ≤ 0.05).
26.3% and 25.1% in U87 or T98G cells, respectively. Q-L hybrid ex-
alone. Importantly, Q-L hybrid had a significant effect on both cell
lines, leading to early S-phase cell cycle arrest and increases in cells in
sub-G1 in both cell lines, while the effects on T98G were more marked
(Fig. 6A). The aforementioned effects were also confirmed by light
microscopy, as growth inhibition and/or cytotoxic effects were ob-
served in each treatment condition (Fig. 6B). In conclusion, Q-L hybrid
effectively alters the cell cycle distribution of GBM cells, leading to cell
cycle arrest and to the induction of cytotoxic effects.
hibited
a slightly enhanced effect in inhibiting ROS production
(p < 0.05), 28.1% or 27.2% in U87 or T98G cells, respectively, in
comparison both with quercetin and the co-administration of quercetin
and losartan.
2.5. Q-L hybrid inhibits GBM proliferation compared with the parent
molecules
Having determined that the Q-L hybrid has a similar binding effi-
ciency to AT1R as losartan and retains the antioxidant capacity of
quercetin, we next aimed to elucidate its inhibitory effects on GBM cells
in vitro. We tested two established GBM cell lines (LN229 and U87). We
first determined the expression levels of AGTR1 in these cell systems.
AGTR1 expression was readily detected in the established GBM cell line
U87 and was similar to the primary brain microvascular endothelial cell
line hCMEC/D3 (Fig. 5A) but was expressed at low levels in LN229.
LN229 and U87 cells were treated with losartan, quercetin, quercetin
together with losartan and with the Q-L hybrid (Fig. 5B). Losartan alone
had no effect on the growth of GBM cells, whereas quercetin alone
caused a reduction in cell growth of 25% for LN229 (p = 0.0002) and
27% for U87 (p = 0.0010). The combination of losartan and quercetin
further increased the anti-proliferative effect to 33% for LN229
(P < 0.0001) and 32% for U87 (p = 0.0002). Of note, the most potent
inhibition of GBM cell growth was achieved when cells were treated
with the Q-L hybrid which caused a 39% decrease in LN229 cell growth
(p < 0.0001) and a 38% decrease in U87 (p < 0.0001).
2.7. Q-L hybrid significantly inhibits GBM-induced angiogenic processes
Signaling via AT1R is involved in regulating angiogenesis and nu-
merous studies have shown that quercetin is anti-angiogenic [35]. To
determine the effect of the Q-L hybrid and its components on angio-
genesis, we employed an in vitro assay using the brain microvascular
endothelial cell line hCMEC/D3 and medium conditioned by the GBM
cell line U87. The respective results are summarized in Fig. 7. Losartan
alone had no significant effect on the induction of angiogenesis by
U87 cells but quercetin monotherapy (non-significantly) reduced seg-
ment formation and more strikingly reduced mesh formation (a mea-
sure of the integrity of the vascular network) by 25% (p = 0.0355). The
Q-L hybrid significantly reduced the number of segments formed by
25% (p = 0.017) and caused a highly significant inhibition in the
number of meshes between these segments (p = 0.0034). The ability to
significantly reduce both segment and mesh formation implies that the
Q-L hybrid will potently inhibit neovascularization induced by GBM
cells.
2.6. Q-L hybrid induces cell cycle arrest and cytotoxic effects in GBM cell
lines
3. Conclusions
Despite promising recent advances in the treatment of GBM, there is
a clear and urgent need for new therapeutic strategies to improve the
clinical outcomes associated with this disease. A viable and practicable
approach is to repurpose already existing marketed agents as GMB
therapeutics. The main goal of the specific work was to establish a new
chemical entity through the compound hybridization approach which
retains the characteristics of the isolated components. The molecular
hybridization approach is a complex technique that mandates correct
design, since an improper conjugation could lead to the total abolition
of activity. Herein, we described the design and synthesis of a novel
hybrid molecule, consisting of losartan and quercetin (designated Q-L),
which is able to modulate the ROS levels through a dual mode of action.
We succeeded in constructing Q-L in such a manner that is chemically
The antiproliferative and/or cytotoxic effects of chemotherapeutic
agents impact the cell cycle distribution of cancer cells, ultimately
leading to cell cycle arrest and to the induction of cell death. Quercetin
has been previously associated with cell cycle arrest in U87 and T98G
GBM cell lines [66]. To evaluate the impact of the Q-L hybrid on the cell
cycle phase of the GBM cell lines U87 and T98G, we performed cell
cycle phase distribution analysis by flow cytometry. We found that in
most cases, cell cycle phase distributions in both cell lines were affected
by losartan, quercetin or their combination. In detail, losartan induced
a G1 cell cycle phase arrest in U87 cells and quercetin induced an early
S-phase cell cycle arrest in both cell lines, while the combinatorial effect
of both quercetin and losartan was similar to the one of quercetin acting
396

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
Fig. 5. Q-L hybrid is a potent inhibitor of
GBM cell growth irrespective of AGTR1
expression. GBM cell lines LN229 and U87
were analyzed for AGTR1 expression by
hydrolysis probe qPCR (A). These cells were
used in SRB proliferation and cytotoxicity
assays to determine the effect of the Q-L
hybrid on GBM cell growth (B). LN229 and
U87 were treated with 30 μM of losartan,
30 μM quercetin, 30 μM quercetin together
with 30 μM losartan and with the Q-L hy-
brid also at 30 μM. Cells were analyzed 5
days post-treatment. Data presented are
mean absorbance (490 nm) values normal-
ized to the untreated control (UC) +/− SD.
One-way ANOVA with Tukey's multiple
comparisons test was performed to de-
termine significant differences between the
treatment groups (***Pp
****p ≤ 0.0001).
≤
0.001,
stable in physiological pH, while it functions as a chemopreventive
agent and retains the desired properties of its two isolated components
at similar levels e.g. binding affinity/selectivity for AT1R, anti-
angiogenic and antioxidant properties. Q-L hybrid showed similar ROS
inhibition capacity in comparison with quercetin in GBM cell lines and
also induced changes in cell cycle distribution quantified by flow cy-
tometry. In addition, the Q-L hybrid inhibited cancer cell proliferation
and angiogenesis in primary GBM cell cultures at similar levels in
comparison to its two components. Although there is currently interest
in the utilization of ROS-producing therapeutics for GBM, the Q-L hy-
brid that possesses ROS-scavenging properties is a novel class of mo-
lecules and may represent a new therapeutic route against GBM. Such
molecules could be further stabilized by encapsulation in supramole-
cular vectors or be efficiently delivered into the brain using alternative
techniques, including intranasal administration.
purchased from Carlo Erba Reagents; Methanol (MeOH), di-
chloromethane (CH2Cl2) and acetonitrile (MeCN) from Fisher Scientific;
Diethyl ether from Sigma Aldrich; Silica gel (60) was used for column
chromatography. TLC was carried out on silica gel plates (UV254) and
was of the highest available purity from Alfa Aesar. DMSO-d6 (99.8%)
was purchased from Euriso-top. Thin-layer chromatography (TLC) was
performed on TLC plates, silica gel coated with fluorescent indicator
F254. Column chromatography was performed using SiO2 and the ap-
propriate elution solvent mixture as indicated in each experimental
procedure. The 1H NMR and 13C NMR spectra were recorded on a
Bruker AV500, using DMSO-d6 and 298 K. Chemical shifts for 1H NMR
spectra are reported as d units of parts per million (ppm) and relative to
the signal of DMSO-d6 (d 2.50 multiplet). Multiplicities are given as
follows: s (singlet); d (doublet); dd (doublet of doublets); t (triplet); bs
(broad singlet); m (multiplet). Coupling constants are reported as J
values in Hz. For the MS characterization of the synthesized com-
pounds, the samples were injected directly to an EVOQ™ Elite ER LC-TQ
system (Bruker, Germany).
4. Materials and methods
4.1. Chemicals
4.2. Chemical synthesis
All materials and reagents were used without further purification.
Chlorotriphenylmethane,
succinic
anhydride,
1-(3-
4.2.1. Synthesis of (2-butyl-4-chloro-1-((2'-(1-trityl-1H-tetrazol-5-yl)-
[1,1′-biphenyl]-4-yl)methyl)-1H-imidazole-5-yl)methanol (2)
Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDCI.HCl)
and Triisopropylsilane (TIS) were purchased from Alfa Aesar;
Triethylamine (TEA), N,N-Diisoproylethylamine (DIPEA), losartan and
quercetin from Sigma Aldrich; Sodium sulfate from Carlo Erba Reactifs
sds; N-Hydroxybenzotriazole (HOBt) from GL Biochem; Trifluoroacetic
acid (TFA) from Fischer Scientific; N,N-Dimethylformamide (DMF) was
Triethylamine (165.0 μL, 2.5 mmol) was added to the solution of
losartan (200 mg, 0.4729 mmol) in 10 mL anhydrous CH2Cl2 under an
inert atmosphere. After 5 min, a solution of chlorotriphenylmethane
(184.56 mg, 0.6620 mmol) in 5 mL anhydrous CH2Cl2 was added
dropwise and the reaction was continued at room temperature for 2 h.
397

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
Fig. 6. Q-L hybrid leads to cell cycle arrest and cytotoxicity in GBM cell lines. GBM cell lines U87 and T98G were treated with 30 μM of either losartan (L),
quercetin (Q), losartan together with quercetin (L + Q) or the Q-L hybrid; (A) 24 h post treatment cell cycle distribution was analyzed following staining with PI.
Representative results are presented, while the percentage of cells in each cell cycle phase is the mean of three independent experiments; (B). 24 h post-treatment,
cells were photographed under a microscope in 40X magnification for observation of cytotoxic effects.
TLC (CH2Cl2/MeOH, 9.5/0.5 v/v) shows the formation of a new non-
polar spot and consumption of almost all losartan. Reaction solvent is
evaporated on a rotary evaporator and the residue was dried well to
give 2 (310 mg, 98.5%) as a white solid, which was used for the next
reaction without further purification.
anhydride (29.37 mg, 0.2935 mmol) in anhydrous acetonitrile (5 mL)
under inert atmosphere and the reaction was continued at RT for 12 h.
TLC (CH2Cl2/MeOH, 10/1 v/v) shows the complete consumption of 2.
The solvent is evaporated on rotary evaporator and the residue is dis-
solved in 5 mL water and extracted with ethyl acetate (3 × 5 mL). The
combined organic phases are washed with brine, dried over sodium
sulfate and concentrated to give 3 (86 mg, 95.86%) as a brown solid,
which was used without further purification. 1H-NMR of (3)
(500 MHz, DMSO-d6, 25 οC): δ = 12.23 (bs, 1H, COOH), 7.84–7.81
(m, 1H, 21-H), 7.66–7.63 (m, 1H, 19-H), 7.59–7.56 (m,1H, 20-H),
7.47–746 (m, 1H, 18-H), 7.43–7.27 (m, 9H, 28 and 29-H), 7.11 (d,
4.2.2. Synthesis of 4-((2-butyl-4-chloro-1-((2'-(1-trityl-1H-tetrazol-5-yl)-
[1,1′-biphenyl]-4-yl)methyl)-1H-imidazole-5-yl)methoxy)-4-oxobutanoic
acid (3)
Diisopropylethylamine (DIPEA) (818.15 μL, 4.6968 mmol) was
added to the solution of 2 (78 mg, 0.1174 mmol) and succinic
398

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
Fig. 7. Q-L hybrid significantly inhibits
GBM-induced
angiogenic
processes.
Medium was conditioned by the GBM cell
line U87 for 24 h in the presence of 30 μM of
losartan, 30 μM quercetin, 30 μM quercetin
together with 30 μM losartan and the Q-L
hybrid also at 30 μM. Conditioned medium
was used to plate hCMEC/D3 brain micro-
vascular endothelial cells onto matrigel
coated angiogenesis slides and tube struc-
tures formed were imaged 6 h post-plating.
Images were analyzed using the ImageJ
angiogenesis analyzer plugin. Data pre-
sented are mean values normalized to the
conditioned media control (CM) +/− SD.
One-way ANOVA with Tukey's multiple
comparisons test were performed to de-
termine significant differences (*p ≤ 0.05,
**p ≤ 0.01).
J = 8.2 Hz, 2H, 15-H), 6.91–6.88 (m, 6H, 27-H), 6.88 (d, J = 8.1 Hz,
2H, 14-H), 5.20 (s, 2H, 12-H), 4.92 (s, 2H, 6-H), 2.46 (t, J = 7.7 Hz, 2H,
30-H), 2.36 (t, J = 6.3 Hz, 2H, 3-H), 2.27 (t, J = 6.3 Hz, 2H, 2-H),
1.53–1.46 (m, 2H, 31-H), 1.23–1.16 (m, 2H, 32-H), 0.78 (t, J = 7.4 Hz,
3H, 33-H) ppm. 13C-NMR of (3) (500 MHz, DMSO-d6, 25 οC):
δ = 174.06 (C1), 172.44 (C4), 164.28 (C23), 149.48 (C10), 141.83
(C17), 141.79 (C22), 141.66 (C26), 140.34 (C16), 136.07 (C13), 131.29
(C18), 131.11 (C19), 130.92 (C21), 130.23 (C27), 139.94 (C15), 129.05
(C7), 128.96 (C29), 128.61 (C28), 128.49 (C20), 126.27 (C14), 121.42
(C8), 83.09 (C25), 54.48 (C6), 47.12 (C12), 29.42 (C31), 29.08 (C3),
28.98 (C2), 26.41 (C30), 22.18 (C32), 14.13 (C33) ppm. Mass: MS
(ESI−) m/z for C45H41ClN6O4: calcd: 764.29; found: 763.27 [M-H]-.
C60H49ClN6O10: calcd: 1048.32; found: 1047.30 [M-H]-.
4.2.4. Synthesis of (1-((2'-(1H-tetrazol-5-yl)-[1,1′-biphenyl]-4-yl)methyl)-2-
butyl-4-chloro-1H-imidazole-5-yl)methyl (2-hydroxy-5-(3,5,7-trihydroxy-4-
oxo-4H-chromen-2-yl)phenyl)succinate (Q-L)
Compound 4 (48 mg) is dissolved in a 5 mL (CH2Cl2/TFA/TIS, 8.5/
1/0.5 v/v) cleavage cocktail at 0 °C to rt for 4 h. TLC (CH2Cl2/MeOH,
9.5/0.5 v/v) shows the formation of a new polar spot and almost
complete The reaction solvent is removed on rotary evaporator and the
residue is washed with hexane (1 mL × 2) and diethyl ether (1 mL × 2)
to give pure Q-L (35 mg, 94.82%) as a yellow solid. 1H-NMR of (5)
(500 MHz, DMSO-d6, 25 οC): δ = 16.25 (bs, 1H, 24-NH), 12.45 (s, 1H,
Q5-OH), 10.85 (s, 1H, Q7-OH), 10.42 (s, 1H, Q4′-OH), 9.60 (s, 1H, Q3-
OH), 7.97 (dd, J = 8.5, 2.1 Hz, 1H, Q6′-H), 7.86 (d, J = 2.2 Hz, 1H,
Q2′-H), 7.67 (m, 1H, 21-H), 7.59 (m, 1H, 19-H), 7.58 (m, 1H, 20-H),
7.67 (m, 1H, 18-H), 7.11 (m, 1H, Q5′-H), 7.11 (d, J = 8.1 Hz, 2H, 15-
H), 6.96 (d, J = 8.1 Hz, 2H, 14-H), 6.47 (d, J = 2.1 Hz, 1H, Q8-H), 6.23
(d, J = 2.1 Hz, Q6-H), 5.26 (s, 2H, 12-H), 5.06 (s, 2H, 6-H), 2.81 (m,
2H, 3-H), 2.54 (m, 2H, 25-H), 2.46 (t, J = 6.5 Hz, 2H, 2-H), 1.51 (m,
2H, 26-H), 1.28 (m, 2H, 27-H), 0.83 (t, J = 7.30 Hz, 3H, 28-H) ppm.
13C-NMR of (5) (500 MHz, DMSO-d6, 25 οC): δ = 172.22 (C4), 171.18
(C1), 164.98 (C Q7), 161.71 (C Q5), 157.01 (C Q9), 151.81 (C Q4′),
149.66 (C10), 146.44 (C Q2), 141.87 (C17), 139.30 (C16), 138.91 (C
Q3′) 136.82 (C13), 131.30 (C21), 131.28 (C19), 131.26 (C18), 129.94
(C15), 129.13 (C7), 128.53 (C20), 127.21 (C Q6′), 126.56 (C14),
123.11 (C Q2′), 121.46 (C8), 117.60 (C Q5′), 103.94 (C Q10), 98.91 (C
Q6), 94.17 (C Q8), 54.73 (C6), 47.18 (C12), 29.52 (C26), 29.02 (C3),
28.91 (C2), 26.47 (C25), 22.27 (C27), 14.22 (C28) ppm. Mass: MS
(ESI−) m/z for C41H35ClN6O10: calcd: 806.21; found: 805.2 [M-H]-.
4.2.3. Synthesis of (2-butyl-4-chloro-1-((2'-(1-trityl-1H-tetrazol-5-yl)-
[1,1′-biphenyl]-4-yl)methyl)-1H-imidazole-5-yl)methyl
(2-hydroxy-5-
(3,5,7-trihydroxy-4-oxo-4H-chromen-2-yl)phenyl) succinate (4)
The solution of 3 (50 mg, 0.065 mmol), quercetin (29.63 mg,
0.098 mmol), N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide hy-
drochloride (EDCI.HCl), (31.35 mg, 0.163 mmol) and 1-
Hydroxybenzotriazole hydrate (HOBt) (17.67 mg, 0.1308 mmol) in
anhydrous DMF is heated at 60 °C for 2 h under inert atmosphere. TLC
(CH2Cl2/MeOH, 9.5/0.5 v/v) shows the formation of a new spot and
almost complete consumption of 4. The reaction solvent is removed
under high vacuum and the residue is purified using silica gel column
chromatography (CH2Cl2/methanol) to give 4 (25 mg, 36.45%) as a
yellow sticky solid. 1H-NMR of (4) (500 MHz, DMSO-d6, 25 οC):
δ = 12.44 (s, 1H, Q5-OH), 10.85 (s, 1H, Q7-OH), 10.41 (s, 1H, Q4′-
OH), 9.60 (s, 1H, Q3-OH), 7.97 (dd, J = 8.76, 2.26 Hz, 1H, Q6′-H),
7.86 (d, J = 2.26 Hz, 1H, Q2′-H), 7.81 (m, 1H, 21-H), 7.60 (m, 1H, 19-
H), 7.56 (m, 1H, 20-H), 7.44 (m, 1H, 18-H), 7.39 (m, 3H, 29-H), 7.35
(m, 6H, 28-H), 7.11 (d, J = 8.76 Hz, 1H, Q5′-H), 7.08 (d, J = 8.20 Hz,
2H, 15-H), 6.90 (m, 6H, 27-H), 6.86 (d, J = 8.2 Hz, 2H, 14-H), 6.45 (d,
J = 2.0 Hz, 1H, Q8-H), 6.23 (d, J = 2.0 Hz, Q6-H), 5.19 (s, 2H, 12-H),
4.97 (s, 2H, 6-H), 2.77 (m, 2H, 2-H), 2.46 (m, 2H, 3-H), 2.43 (m, 2H,
30-H), 1.48 (m, 2H, 31-H), 1.17 (m, 2H, 32-H), 0.77 (t, J = 7.35 Hz,
3H, 33-H) ppm. 13C-NMR of (4) (500 MHz, DMSO-d6, 25 οC):
δ = 176.69 (C Q4), 172.15 (C4), 171.16 (C1), 164.91 (C Q7), 164.22
(C23), 161.73 (C Q5), 157.08 (C Q9), 151.67 (C Q4′), 149.46 (C10),
146.37 (C Q2), 141.85 (C17), 141.70 (C26), 140.30 (C16), 138.87 (C
Q3′) 136.94 (C Q3), 136.20 (C13), 131.38 (C18), 131.22 (C19), 130.99
(C21), 130.24 (C27), 130.09 (C15), 128.96 (C29), 128.61 (C28), 128.58
(C20), 127.19 (C Q6′), 126.34 (C14), 123.19 (C Q2′), 121.34 (C8),
117.66 (C Q5′), 103.92 (C Q10), 99.00 (C Q6), 94.22 (C Q8), 83.09
(C25), 54.74 (C6), 47.23 (C12), 29.52 (C31), 29.05 (C3 and C2), 26.48
(C30), 22.18 (C32), 14.20 (C33) ppm. Mass: MS (ESI−) m/z for
4.3. Q-L docking to AT1R and AT2R
Docking calculations have been performed in order to predict the
binding of the designed hybrid molecule to AT1R and define the binding
mode with which the hybrid exerts its inhibitory action. A hybrid 2D
structure was sketched in Schrodinger's 2017-1 Maestro molecular
modeling platform [67] and LigPrep was used to prepare a 3D model.
The geometries were optimized with MacroModel in order to relax the
structure [68–70]. The force field used for the minimization was
OPLS3. The hybrid compound was subjected to proper treatment of
ligand protonation states for the physiological pH of 7.4. The three-
dimensional hybrid structure was further minimized, with MacroModel
in vacuum and with OPLS3, using a conjugate gradient (CG) method
with a threshold of 0.01 kcal/mol. This minimized structure was further
used as input to a mixed-torsional/low-sampling conformational search
399

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
which yielded several conformers of the hybrid compound that were
energetically ranked and used for the following docking calculations
[69,71].
qPCR Mastermix (PrimerDesign) and run on a C1000 Thermal Cycler
combined with a CFX96 detection module (Bio-Rad) following the
manufacturers protocol. 50 ng of template cDNA was used per reaction.
Expression data was normalized to the mean Ct value of both reference
genes and reported as 2^-deltaCt.
Using the PDB ID: 4YAY [72] and 6DO1 [73] for inactivated and
activated AT1R respectively, initial docking calculations were per-
formed with Induced Fit Docking (IFD) method (Schrödinger, LLC, New
York, USA). The ligand was docked in 10 of the possible conformations
generated by Macromodel. Protein preparation constrained refinement
was applied in the Glide docking stage. Trimming side chains auto-
matically (based on B – factor) and Prime refinement of the protein side
chains were applied and the docking process was accomplished by
Glide/XP. Finally, the binding energy for each ligand was calculated.
Additional docking simulations were performed for losartan/AT1R ac-
tivated, losartan/AT1R inactivated, and quercetin/AT1R activated,
quercetin/AT1R inactivated complexes for comparison reasons.
4.5.3. Sulphorhodamine
proliferation assays
B (SRB) and cell counting Kit-8 (CCK8)
GBM cells were seeded onto 96-well plates (Corning) at
2 × 103 cells per well in DMEM supplemented with 2% FBS. Twenty-
four hours post-plating cells were treated with Quercetin, Losartan,
Quercetin and Losartan together and the Q-L hybrid alone for 24 h, 3
days, 6 days and 9 days. 30 μM concentration of drugs were used based
on previous dose response data obtained in the lab (unpublished). Cells
were analyzed using the CCK8 kit (Sigma) according to manufacturer's
instructions prior to being fixed with 10% tri-chloroacetate (TCA) for at
least 1 h at 4 °C and stained with SRB (Sigma) as described previously
[63]. Absorbance values were normalized to untreated controls.
4.4. Radioligand binding assays to the AT1R and AT2R
Experiments on whole cells were performed in both AT1R- and
AT2R- HEK-293 stable cell lines, as previously described in detail with
minor modifications [63]. Q-L hybrid and reference compounds were
4.5.4. Endothelial cell tube formation assays
examined for their ability to inhibit specific binding of
[
125I]-
Assays were performed as described previously by Renziehausen
et al. [36], using the method of Carpentier et al. [74]. To prepare
conditioned medium (CM), GBM cells were seeded onto 6-well plates
(Corning) at 1.5 × 105 cells/well in their respective culture media.
Cells were allowed to adhere for 24 h then washed twice with warm
PBS prior to replacing the medium with 2 mL of EBM-2 basal medium
without any supplements per well. The EBM-2 basal medium was
conditioned for 24 h, then carefully collected and centrifuged at 223
RCF for 5 min at RT to pellet cells and debris. The conditioned media
(CM) was used immediately in tube formation assays. To assess tube
formation growth factor reduced matrigel (10 mg/mL, Corning) was
thawed at 4 °C on ice for 48 h. U-angiogenesis slides (Ibidi) were coated
with 10 μL matrigel per well and incubated at 37 °C for 30 min. During
the incubation, hCMEC/D3 cells were washed twice with PBS, trypsi-
nized and counted. hCMEC/D3 cells were aliquoted into sterile 1.5 mL
microcentrifuge tubes (Eppendorf) at 4 × 104 cells and pelleted by
5 min centrifugation at 223 RCF. Cell pellets were resuspended in
200 μL GBM CM and plated onto matrigel coated angiogenesis slides in
triplicate (1 × 104 cells in 50 μL per well). Images were acquired using
a TMS inverted phase contrast microscope (Nikon) with a DinoEye
AM7023 eyepiece camera (Dino-Lite) at 2X magnification and analyzed
using the ImageJ angiogenesis analyzer plugin.1.
Sar1Ile8AII, and nonspecific binding was determined using 10 μM un-
labeled Ang II. Following non-linear regression, IC50 values, re-
presenting the concentration at which each ligand displaced 50%
binding of [125I]-Sar1Ile8AII in either AT1R- or AT2R- HEK-293 cell
lines, were determined.
4.5. Biological evaluation of the Q-L hybrid
4.5.1. Cell culture
Cells were routinely cultured in a humidified 5% CO2 incubator at
37 °C and passaged using trypsin/EDTA (Gibco). Established GBM cell
lines U87, T98G and LN229 were purchased from ATCC and cultured in
DMEM (Gibco) supplemented with 10% FBS. Human brain micro-
vascular endothelial cells hCMEC/D3 were purchased from Cellutions
Biosystems and cultured in EBM-2 media (Lonza) supplemented with
5% FBS, 1.4 μM Hydrocortisone (Sigma), 5 μg/mL Ascorbic Acid
(Sigma), 10 mM HEPES (Invitrogen), 1:100 chemically defined lipid
concentrate (Invitrogen) and 1 ng/mL hFGF-B (Peprotech).
4.5.2. Quantitative PCR to evaluate expression of AGTR1 in GBM cells and
hCMEC/D3
Total RNA was extracted from cultured cells using the RNeasy Mini
kit (Qiagen) according to the manufacturer's instructions including the
optional on-column DNase digestion step. Total RNA concentration was
determined using a NanoVue Plus spectrophotometer (GE Healthcare).
RNA degradation and genomic contamination were checked for by gel-
electrophoresis. Total RNA was reverse transcribed using random pri-
mers (Invitrogen) following the M-MLV Reverse Transcriptase protocol
(Promega). Custom designed hydrolysis probes were used in an effort to
increase detection efficiency. Probes for AGTR1 and the reference genes
TATA-binding protein (TBP) and hypoxanthine phosphoribosyl-
transferase 1 (HPRT1) were designed by PrimerDesign (Listed in
Table 3). Triplicate qPCR reactions were set up using the Precision 2X
4.5.5. Flow cytometry analyses
Flow cytometric analysis was performed to evaluate the effect of
quercetin, losartan and Q-L hybrid on the cell cycle distribution of
glioblastoma cells, based on a protocol by Vartholomatos et al. [75].
Briefly, cells were trypsinized, centrifuged, washed once with PBS and
incubated for 20 min at 37 °C in the dark with propidium iodide (PI)
solution (50 μg/mL PI, 20 mg/mL RNase A and 0.1% Triton X-100).
Analysis of DCFDA fluorescence for reactive oxygen species quantifi-
cation was performed according to protocol by Eruslanov and Kus-
martsev [76]. For both the aforementioned analyses, the fluorescence of
10,000 cells was measured in a flow cytometer (FACScalibur, BD
Biosciences). Fluorescence distributions were analyzed using Cell Quest
V3.1 software (BD Biosciences).
Table 3
Sequence of primers used in qPCR.
Gene
Primer
Sequence (5′-3′)
Product Size (bp)
5. Statistical analysis
AGTR1
TBP
Forward
Reverse
Forward
Reverse
Forward
Reverse
CAAAAGCCAAATCCCACTCAA
CACAAAATTACTTTATGGACAGGTTT
GTTTGCCAAGAAGAAAGTGAAC
GGGTCAGTCCAGTGCCAT
150
143
111
One-way ANOVA with Tukey's multiple comparisons test was per-
formed using Graphpad Prism to determine significant differences be-
tween treatment groups (the respective p-values are presented in each
assay).
HPRT1
TTAGTAGTGTTTCAGTAATGTTGACT
ATTTTTGGGAATTTATTGATTTGCAT
400

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
Declaration of competing interest
Sci. U. S. A. 116 (6) (2019) 2210–2219.
[24] Y. Tang, et al., Depletion of collagen by losartan to improve tumor accumulation
and therapeutic efficacy of photodynamic nanoplatforms, Drug. Deliv. Transl. Res.
9 (3) (2019) 615–624.
There are no conflicts of interest to declare.
[25] J.E. Murphy, et al., Potentially curative combination of TGF-b1 inhibitor losartan
and FOLFIRINOX (FFX) for locally advanced pancreatic cancer (LAPC): R0 resection
rates and preliminary survival data from a prospective phase II study, J. Clin. Oncol.
36 (15_suppl) (2018) 4116-4116.
Acknowledgements
[26] L.G. Milroy, et al., Modulators of protein-protein interactions, Chem. Rev. 114 (9)
(2014) 4695–4748.
The research work was supported by the Hellenic Foundation for
Research and Innovation (H.F.R.I.) under the “First Call for H.F.R.I.
Research Projects to support Faculty members and Researchers and the
procurement of high-cost research equipment grant” (Project Number:
991, PROTECT, to AGT). In addition, this research has been financially
supported by the Hellenic Foundation for Research and Innovation
(HFRI) under the HFRI PhD Fellowship grant (Fellowship Number:
14551). In addition, it has been financially supported by the Barrow
Neurological Foundation, Brain Tumour Research (BTR) and Brain
Tumour Research Campaign (BTRC).
[27] A. Primikyri, et al., Understanding zinc(II) chelation with quercetin and luteolin: a
combined NMR and theoretical study, J. Phys. Chem. B 119 (1) (2015) 83–95.
[28] A.W. Boots, et al., Quercetin reduces markers of oxidative stress and inflammation
in sarcoidosis, Clin. Nutr. 30 (4) (2011) 506–512.
[29] L. Gibellini, et al., Interfering with ROS metabolism in cancer cells: the potential
role of quercetin, Cancers 2 (2) (2010) 1288–1311.
[30] D. Xu, et al., Antioxidant activities of quercetin and its complexes for medicinal
application, Molecules 24 (6) (2019).
[31] D. Bao, et al., Protective effect of quercetin against oxidative stress-induced cyto-
toxicity in rat pheochromocytoma (PC-12) cells, Molecules 22 (7) (2017).
[32] A. Primikyri, et al., Direct binding of Bcl-2 family proteins by quercetin triggers its
pro-apoptotic activity, ACS Chem. Biol. 9 (12) (2014) 2737–2741.
[33] T.F. Kellici, et al., Rational design and structure-activity relationship studies of
quercetin-amino acid hybrids targeting the anti-apoptotic protein Bcl-xL, Org.
Biomol. Chem. 15 (37) (2017) 7956–7976.
Appendix A. Supplementary data
[34] A. Primikyri, et al., Probing the interaction of a quercetin bioconjugate with Bcl-2 in
living human cancer cells with in-cell NMR spectroscopy, FEBS Lett. 592 (20)
(2018) 3367–3379.
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.freeradbiomed.2020.08.007.
[35] Y. Liu, et al., Low concentration of quercetin antagonizes the invasion and angio-
genesis of human glioblastoma U251 cells, OncoTargets Ther. 10 (2017)
4023–4028.
References
[36] R. Vengoji, et al., Natural products: a hope for glioblastoma patients, Oncotarget 9
(31) (2018) 22194–22219.
[1] J. Mann, et al., Advances in radiotherapy for glioblastoma, Front. Neurol. 8 (2017)
748.
[37] M.H. Park, S. Min do, Quercetin-induced downregulation of phospholipase D1 in-
hibits proliferation and invasion in U87 glioma cells, Biochem. Biophys. Res.
Commun. 412 (4) (2011) 710–715.
[2] R. Stupp, et al., Radiotherapy plus concomitant and adjuvant temozolomide for
glioblastoma, N. Engl. J. Med. 352 (10) (2005) 987–996.
[3] M. Valko, et al., Free radicals, metals and antioxidants in oxidative stress-induced
cancer, Chem. Biol. Interact. 160 (1) (2006) 1–40.
[38] E. Pozsgai, et al., Quercetin increases the efficacy of glioblastoma treatment com-
pared to standard chemoradiotherapy by the suppression of PI-3-kinase-Akt
pathway, Nutr. Canc. 65 (7) (2013) 1059–1066.
[4] W.T. Chiu, et al., Contribution of reactive oxygen species to migration/invasion of
human glioblastoma cells U87 via ERK-dependent COX-2/PGE(2) activation,
Neurobiol. Dis. 37 (1) (2010) 118–129.
[39] D.H. Choi, C. Li, J.S. Choi, Effects of myricetin, an antioxidant, on the pharmaco-
kinetics of losartan and its active metabolite, EXP-3174, in rats: possible role of
cytochrome P450 3A4, cytochrome P450 2C9 and P-glycoprotein inhibition by
myricetin, J. Pharm. Pharmacol. 62 (7) (2010) 908–914.
[5] G.S. Devi, et al., Free radicals antioxidant enzymes and lipid peroxidation in dif-
ferent types of leukemias, Clin. Chim. Acta 293 (1–2) (2000) 53–62.
[6] H. Pelicano, D. Carney, P. Huang, ROS stress in cancer cells and therapeutic im-
plications, Drug Resist. Updates 7 (2) (2004) 97–110.
[40] L.G. Costa, et al., Mechanisms of neuroprotection by quercetin: counteracting oxi-
dative stress and more, Oxidative medicine and cellular longevity (2016) 2016: pp.
2986796–2986796.
[7] G.Y. Liou, P. Storz, Reactive oxygen species in cancer, Free Radic. Res. 44 (5)
(2010) 479–496.
[41] A. Ishisaka, et al., Accumulation of orally administered quercetin in brain tissue and
its antioxidative effects in rats, Free Radic. Biol. Med. 51 (7) (2011) 1329–1336.
[42] A. Glasauer, N.S. Chandel, Targeting antioxidants for cancer therapy, Biochem.
Pharmacol. 92 (1) (2014) 90–101.
[8] G. Waris, H. Ahsan, Reactive oxygen species: role in the development of cancer and
various chronic conditions, J. Carcinog. 5 (2006) 14.
[9] L. Minutoli, et al., ROS-mediated NLRP3 inflammasome activation in brain, heart,
kidney, and testis ischemia/reperfusion injury, Oxid. Med. Cell Longev. 2016
(2016) 2183026.
[43] B. Perillo, et al., ROS in cancer therapy: the bright side of the moon, Exp. Mol. Med.
52 (2) (2020) 192–203.
[10] M. Caffo, et al., Heavy metals and epigenetic alterations in brain tumors, Curr.
Genom. 15 (6) (2014) 457–463.
[44] M.V. Chatziathanasiadou, et al., Amplifying and broadening the cytotoxic profile of
quercetin in cancer cell lines through bioconjugation, Amino Acids 50 (2) (2018)
279–291.
[11] B. Poljsak, I. Milisav, The role of antioxidants in cancer, friends or foes? Curr.
Pharmaceut. Des. 24 (44) (2018) 5234–5244.
[45] Z. Li, J.S. Bains, A.V. Ferguson, Functional evidence that the angiotensin antagonist
losartan crosses the blood-brain barrier in the rat, Brain Res. Bull. 30 (1–2) (1993)
33–39.
[12] M. Vidak, D. Rozman, R. Komel, Effects of flavonoids from food and dietary sup-
plements on glial and glioblastoma multiforme cells, Molecules 20 (10) (2015)
19406–19432.
[46] M. Pinter, R.K. Jain, Targeting the renin-angiotensin system to improve cancer
treatment: implications for immunotherapy, Sci. Transl. Med. 9 (410) (2017)
eaan5616.
[13] D.D. McConnell, et al., Do anti-oxidants vitamin D3, melatonin, and alpha-lipoic
acid have synergistic effects with temozolomide on cultured glioblastoma cells?
Medicines (Basel) 5 (2) (2018).
[47] T. Watanabe, et al., Losartan metabolite EXP3179 activates Akt and endothelial
nitric oxide synthase via vascular endothelial growth factor receptor-2 in en-
dothelial cells: angiotensin II type 1 receptor-independent effects of EXP3179,
Circulation 112 (12) (2005) 1798–1805.
[14] T. Wegman-Ostrosky, et al., The renin-angiotensin system meets the hallmarks of
cancer, J. Renin-Angiotensin-Aldosterone Syst. JRAAS 16 (2) (2015) 227–233.
[15] A.R. Bradshaw, et al., Glioblastoma multiforme cancer stem cells express compo-
nents of the renin-angiotensin system, Front Surg 3 (2016) 51.
[16] M. de Gasparo, et al., International union of pharmacology. XXIII. The angiotensin
II receptors, Pharmacol. Rev. 52 (3) (2000) 415–472.
[48] Yannick Djoumbou Feunang, et al., BioTransformer: a comprehensive computa-
tional tool for small molecule metabolism prediction and metabolite identification,
J. Cheminf. 11 (2019), https://doi.org/10.1186/s13321-018-0324-5.
[49] P.C.H. Hollman, M.B. Katan, Absorption, metabolism and health effects of dietary
flavonoids in man, Biomed. Pharmacother. 51 (8) (1997) 305–310.
[50] M. Melidou, K. Riganakos, D. Galaris, Protection against nuclear DNA damage of-
fered by flavonoids in cells exposed to hydrogen peroxide: the role of iron chelation,
Free Radic. Biol. Med. 39 (12) (2005) 1591–1600.
[17] E. Januel, et al., Impact of renin-angiotensin system blockade on clinical outcome in
glioblastoma, Eur. J. Neurol. 22 (9) (2015) 1304–1309.
[18] S. Rajagopalan, et al., Angiotensin II-mediated hypertension in the rat increases
vascular superoxide production via membrane NADH/NADPH oxidase activation.
Contribution to alterations of vasomotor tone, J. Clin. Invest. 97 (8) (1996)
1916–1923.
[51] E. Heřmánková, et al., Redox properties of individual quercetin moieties, Free
Radic. Biol. Med. 143 (2019) 240–251.
[19] K.D. Pendergrass, et al., The angiotensin II-AT1 receptor stimulates reactive oxygen
species within the cell nucleus, Biochem. Biophys. Res. Commun. 384 (2) (2009)
149–154.
[52] J.P. Adjimani, P. Asare, Antioxidant and free radical scavenging activity of iron
chelators, Toxicology reports 2 (2015) 721–728.
[20] L.M. Willis, et al., Angiotensin receptor blockers and angiogenesis: clinical and
experimental evidence, Clin. Sci. (Lond.) 120 (8) (2011) 307–319.
[21] C. Xu, X. Wu, J. Zhu, VEGF promotes proliferation of human glioblastoma multi-
forme stem-like cells through VEGF receptor 2, Sci.World J. 2013 (2013) 417413.
[22] V.P. Chauhan, et al., Angiotensin inhibition enhances drug delivery and potentiates
chemotherapy by decompressing tumour blood vessels, Nat. Commun. 4 (2013)
2516.
[53] A. Primikyri, et al., Understanding zinc(II) chelation with quercetin and luteolin: a
combined NMR and theoretical study, J. Phys. Chem. B 119 (1) (2015) 83–95.
[54] K.E. Warren, Beyond the blood:brain barrier: the importance of central nervous
system (CNS) pharmacokinetics for the treatment of CNS tumors, including diffuse
intrinsic pontine glioma, Frontiers in Oncology 8 (239) (2018).
[55] W.A. Banks, Characteristics of compounds that cross the blood-brain barrier, BMC
Neurol. 9 (Suppl 1) (2009) (Suppl 1): p. S3-S3.
[23] Y. Zhao, et al., Losartan treatment enhances chemotherapy efficacy and reduces
ascites in ovarian cancer models by normalizing the tumor stroma, Proc. Natl. Acad.
[56] P.S. Gerogianni, et al., Lipophilic ester and amide derivatives of rosmarinic acid
protect cells against H2O2-induced DNA damage and apoptosis: the potential role
401

A.D. Tsiailanis, et al.
FreeRadicalBiologyandMedicine160(2020)391–402
of intracellular accumulation and labile iron chelation, Redox Biol. 15 (2018)
Res. 318 (8) (2012) 925–935.
548–556.
[67] D.R. Tomchick, et al., Structural and functional characterization of second-co-
ordination sphere mutants of soybean lipoxygenase-1, Biochemistry 40 (25) (2001)
7509–7517.
[57] G. Garcia, et al., New losartan-hydrocaffeic acid hybrids as antihypertensive-anti-
oxidant dual drugs: ester, amide and amine linkers, Eur. J. Med. Chem. 50 (2012)
90–101.
[68] T.A. Halgren, et al., Glide: a new approach for rapid, accurate docking and scoring.
2. Enrichment factors in database screening, J. Med. Chem. 47 (7) (2004)
1750–1759.
[58] H. Zhang, et al., Structure of the Angiotensin receptor revealed by serial femtose-
cond crystallography, Cell 161 (4) (2015) 833–844.
[59] D.J. Carini, et al., Nonpeptide angiotensin II receptor antagonists: the discovery of a
series of N-(biphenylylmethyl)imidazoles as potent, orally active antihypertensives,
J. Med. Chem. 34 (8) (1991) 2525–2547.
[69] W. Sherman, et al., Novel procedure for modeling ligand/receptor induced fit ef-
fects, J. Med. Chem. 49 (2) (2006) 534–553.
[70] R. Farid, et al., New insights about HERG blockade obtained from protein modeling,
potential energy mapping, and docking studies, Bioorg. Med. Chem. 14 (9) (2006)
3160–3173.
[60] T. Karampelas, et al., GnRH-Gemcitabine conjugates for the treatment of androgen-
independent prostate cancer: pharmacokinetic enhancements combined with tar-
geted drug delivery, Bioconjugate Chem. 25 (4) (2014) 813–823.
[61] R. Tanaka, et al., Novel 3alpha-methoxyserrat-14-en-21beta-ol (PJ-1) and 3beta-
methoxyserrat-14-en-21beta-ol (PJ-2)-curcumin, kojic acid, quercetin, and baica-
lein conjugates as HIV agents, Bioorg. Med. Chem. 17 (14) (2009) 5238–5246.
[62] M.K. Kim, et al., In vitro solubility, stability and permeability of novel quercetin-
amino acid conjugates, Bioorg. Med. Chem. 17 (3) (2009) 1164–1171.
[63] F. Magnani, et al., Electronic sculpting of ligand-GPCR subtype selectivity: the case
of angiotensin II, ACS Chem. Biol. 9 (7) (2014) 1420–1425.
[71] R.A. Friesner, et al., Extra precision Glide: docking and scoring incorporating a
model of hydrophobic enclosure for Protein−Ligand complexes, J. Med. Chem. 49
(21) (2006) 6177–6196.
[72] H. Zhang, et al., Structure of the angiotensin receptor revealed by serial femtose-
cond crystallography, Cell 161 (4) (2015) 833–844.
[73] L.M. Wingler, et al., Distinctive activation mechanism for angiotensin receptor re-
vealed by a synthetic nanobody, Cell 176 (3) (2019) 479–490 e12.
[74] A.F. Carpentier, et al., Steroid-sparing effects of angiotensin-II inhibitors in glio-
blastoma patients, Eur. J. Neurol. 19 (10) (2012) 1337–1342.
[75] G. Vartholomatos, et al., Intraoperative cell-cycle analysis to guide brain tumor
removal, Proc. Natl. Acad. Sci. U. S. A. 111 (36) (2014) E3755.
[76] E. Eruslanov, S. Kusmartsev, Identification of ROS using oxidized DCFDA and flow-
cytometry, Methods Mol. Biol. 594 (2010) 57–72.
[64] D.W. Lamson, M.S. Brignall, Antioxidants and cancer, part 3: quercetin, Alternative
Med. Rev. 5 (3) (2000) 196–208.
[65] A. Murakami, H. Ashida, J. Terao, Multitargeted cancer prevention by quercetin,
Canc. Lett. 269 (2) (2008) 315–325.
[66] J. Michaud-Levesque, N. Bousquet-Gagnon, R. Beliveau, Quercetin abrogates IL-6/
STAT3 signaling and inhibits glioblastoma cell line growth and migration, Exp. Cell
402