Benzofuranyl-2-imidazoles as imidazoline I2 receptor ligands for Alzheimer's disease

Article abstract of DOI:10.1016/j.ejmech.2021.113540

Recent findings unveil the pharmacological modulation of imidazoline I₂ receptors (I₂-IR) as a novel strategy to face unmet medical neurodegenerative diseases. In this work, we report the chemical characterization, three-dimensional quantitative structure-activity relationship (3D-QSAR) and ADMET in silico of a family of benzofuranyl-2-imidazoles that exhibit affinity against human brain I₂-IR and most of them have been predicted to be brain permeable. Acute treatment in mice with 2-(2-benzofuranyl)-2-imidazole, known as LSL60101 (garsevil), showed non-warning properties in the ADMET studies and an optimal pharmacokinetic profile. Moreover, LSL60101 induced hypothermia in mice while decreased pro-apoptotic FADD protein in the hippocampus. In vivo studies in the familial Alzheimer's disease 5xFAD murine model with the representative compound, revealed significant decreases in the protein expression levels of antioxidant enzymes superoxide dismutase and glutathione peroxidase in hippocampus. Overall, LSL60101 plays a neuroprotective role by reducing apoptosis and modulating oxidative stress.

Full text of DOI:10.1016/j.ejmech.2021.113540

European Journal of Medicinal Chemistry 222 (2021) 113540
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Benzofuranyl-2-imidazoles as imidazoline I2 receptor ligands for
Alzheimer's disease
a
,
1
a
, 1
ꢀ
ꢀ
~ꢀ
c
ꢀ b
Sergio Rodriguez-Arevalo , Andrea Bagan , Christian Grinan-Ferre
,
Foteini Vasilopoulou , Merce Pallas , Iria Brocos-Mosquera d, Luis F. Callado c
,
b
b
,
, d
ꢁ
ꢁ
e
e
e
f
g
ꢀ
ꢀ
ꢀ
ꢀ
M. Isabel Loza , Anton L. Martínez , Jose Brea , Belen Perez , Elies Molins ,
Steven De Jonghe h, Dirk Daelemans h, Milica Radan i, Teodora Djikic i, Katarina Nikolic i,
j
j
j
ꢀ
ꢀ
Elena Hernandez-Hernandez , M. Julia García-Fuster , Jesús A. García-Sevilla ,
,
*
Carmen Escolano a
a Laboratory of Medicinal Chemistry (Associated Unit to CSIC), Department of Pharmacology, Toxicology and Medicinal Chemistry, Faculty of Pharmacy and
Food Sciences, Institute of Biomedicine (IBUB), University of Barcelona, Av. Joan XXIII, 27-31, E-08028, Barcelona, Spain
b
ꢁ
Pharmacology Section, Toxicology and Medicinal Chemistry, Faculty of Pharmacy and Food Sciences, Institut de Neurociencies, University of Barcelona, Av.
Joan XXIII, 27-31, E-08028, Barcelona, Spain
c Department of Pharmacology, University of the Basque Country, UPV/EHU, E-48940, Leioa, Bizkaia, Spain
d
ꢀ
ꢀ
Centro de Investigacion Biomedica en Red de Salud Mental, CIBERSAM, Spain
e
ꢀ
ꢀ
Innopharma Screening Platform, BioFarma Research Group, Centro de Investigacion en Medicina Molecular y Enfermedades Cronica (CIMUS), Universidad
de Santiago de Compostela, E-15782, Santiago de Compostela, Spain
f Department of Pharmacology, Therapeutic and Toxicology, Autonomous University of Barcelona, E-08193, Cerdanyola, Spain
g
ꢁ
Institut de Ciencia de Materials de Barcelona (CSIC), Campus UAB, E-08193, Cerdanyola, Spain
h KU Leuven, Department of Microbiology, Immunology and Transplantation, Rega Institute, Laboratory of Virology and Chemotherapy, Herestraat 49, 3000,
Leuven, Belgium
i Department of Pharmaceutical Chemistry, Faculty of Pharmacy, University of Belgrade, 11000, Belgrade, Serbia
j IUNICS University of the Balearic Islands (UIB), Health Research Institute of the Balearic Islands (IdISBa), E-07122, Palma de Mallorca, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Recent findings unveil the pharmacological modulation of imidazoline I2 receptors (I2-IR) as a novel
strategy to face unmet medical neurodegenerative diseases. In this work, we report the chemical char-
acterization, three-dimensional quantitative structure-activity relationship (3D-QSAR) and ADMET in
silico of a family of benzofuranyl-2-imidazoles that exhibit affinity against human brain I2-IR and most of
them have been predicted to be brain permeable. Acute treatment in mice with 2-(2-benzofuranyl)-2-
imidazole, known as LSL60101 (garsevil), showed non-warning properties in the ADMET studies and
an optimal pharmacokinetic profile. Moreover, LSL60101 induced hypothermia in mice while decreased
pro-apoptotic FADD protein in the hippocampus. In vivo studies in the familial Alzheimer's disease 5xFAD
murine model with the representative compound, revealed significant decreases in the protein
expression levels of antioxidant enzymes superoxide dismutase and glutathione peroxidase in hippo-
campus. Overall, LSL60101 plays a neuroprotective role by reducing apoptosis and modulating oxidative
stress.
Received 23 February 2021
Received in revised form
26 April 2021
Accepted 10 May 2021
Available online 20 May 2021
Keywords:
Imidazoline I2 receptors
Imidazoline I2 receptor ligands
Neuroprotection
5xFAD
Benzofuranyl-2-imidazoles
Oxidative stress
© 2021 Elsevier Masson SAS. All rights reserved.
1. Introduction
bind with high affinity to [3H]idazoxan and with lower affinity to
[3H]p-aminoclonidine and [3H]clonidine [2,3]. I2-IR are present in
Imidazoline I2 receptors (I2-IR) are heterogeneous entities, often
described as nonadrenergic binding sites for imidazolines [1], that
many organs, tissues and cell types, including brain, kidney, liver,
astrocytes, platelets [4], pancreatic cells and, vascular smooth
muscle cells [5]. Modifications in the levels of I2-IR have been
associated with analgesia [6], inflammation [7] and with human
brain disorders [8] such as depression [9], Alzheimer's type
* Corresponding author.
E-mail address: cescolano@ub.edu (C. Escolano).
Both authors contribute equally to this work.
1
https://doi.org/10.1016/j.ejmech.2021.113540
0223-5234/© 2021 Elsevier Masson SAS. All rights reserved.

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
dementia [10], Parkinson's disease [11], and glial tumors [12]. The
fact that I2-IR are altered in many pathophysiological processes and
the availability of known I2-IR ligands have permitted to place I2-IR
in a privileged position as new promising therapeutic targets.
Representative I2-IR ligands [13] emerged from the literature as
useful tools to reveal the biological implications of these non-
structurally described receptors (Fig. 1). The compelling evidence
has demonstrated the neuroprotective role of I2-IR through the
pharmacological activities observed for their ligands. Agmatine,
identified as the endogenous I2-IR ligand, has modulated actions in
several neurotransmitters leading to neuroprotection in both
in vitro and in rodent models [14]. Idazoxan reduced neuron
damage in the rat brain hippocampus after global ischemia [15]. 2-
(2-Benzofuranyl)-2-imidazoline (2-BFI), a selective I2-IR ligand,
provided neuroprotective benefits against cerebral ischemia in
models of chronic opioid therapy [16]. Idazoxan and 2-BFI have
been proposed to exert neuroprotection by direct blocking of N-
The amyloid cascade hypothesis, that dominates the field of AD, has
been replaced by alternative explanations arising from the
connection of mitochondrial dysfunction and increased ROS. There
is evidence that indicates a pro-oxidant ability of Ab, mediating an
accelerated production of ROS by directly binding to the mito-
chondrial membranes. Consequently, mitochondrial dynamics and
function are altered, disrupting the energy metabolism, and leading
to the loss of synaptic function. An excess in ROS/RNS production
and a mitochondrial dysfunction could lead to oxidative stress (OS),
which is implicated in several neurodegenerative diseases, such as
AD [26e28]. In fact, there is growing evidence for the contribution
of OS and neuroinflammation to the pathogenesis of AD [29,30].
Recent studies have investigated the role of I2-IR in OS processes.
Selective 2-BFI [31] decreased OS and altered the levels of antiox-
idant enzymes in an AD rat model and protected against OS-
induced astrocytic cell death. Moreover, our group reported
decreased levels of hydrogen peroxide levels and OS markers
induced by two new I2-IR ligands in aged SAMP8 mice [32].
In this manuscript, we focused our attention on 2-(2-
benzofuranyl)-2-imidazole, named as LSL60101 (garsevil), first
described in 1995 by García-Sevilla's group as a I2-IR selective
ligand involved in astrocyte activation and neuronal regeneration
[33,34]. In the following years, some outstanding papers described
the biological relevance of LSL60101 in the attenuation of morphine
tolerance and hence proposing a neuroprotective role [35], such as
provoking morphological/biochemical changes in astroglia that
were neuroprotective after neonatal axotomy [36], and producing
discriminable stimulus [37], amongst others. Of note, astrocytes as
the main supportive cells in the CNS are significantly involved in
the redox homeostasis, and consequently, this could be an indica-
tive of a possible effect of LSL60101 on OS balance.
methyl-
D
-aspartate receptor (NMDA) mediated intracellular [Ca2þ
]
influx [17].
Two out of these seven ligands depicted in Fig. 1, CR4056 and
11C]BU99008, are in the process of validating their therapeutic
[
potential by progressing in clinical trials for osteoarthritis [18], and
for PET diagnosis for patients that suffer from Alzheimer's disease
(AD) [19,20], respectively. Due to the clinical implications of I2-IR,
the discovery of new I2-IR ligands that could modulate the phar-
macology involved is a challenging goal for a medicinal chemistry
program. In this framework, we recently provided two structurally
new families of I2-IR ligands and validated their properties
ameliorating the devastating cognitive decline in two murine
models of neurodegeneration [21e23].
At the subcellular level in the central nervous system (CNS), I2-IR
are mainly located on the outer membrane of mitochondria in as-
trocytes [24,25]. Mitochondria are one of the main sources of
reactive oxygen species (ROS) and reactive nitrogen species (RNS).
From the structural chemical point of view, the nature of known
I2-IR ligands (Fig. 1) is relatively restricted and the pharmacophore
moiety is generally related to 2-imidazoline-like structures.
Fig. 1. Representative I2-IR ligands.
2

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
Structural comparison of LSL60101 with other known I2-IR ligands,
and in particular with 2-BFI that shares a benzofuran moiety, sug-
gest a pharmacomodulation involving an unsaturation of the imi-
dazoline ring, a drug optimization strategy appealed in the design
of new drugs. The presence of an imidazole ring in the successful
CR4056 ligand encourages the proposal.
To access hydroxybenzofuran-2-imidazole derivatives 6b and
6c, hydrolysis of the methylether group of 5b and 5c was achieved
by treatment with hydrobromic acid. The alkylation of the N-
imidazole of compound 5a with methyl iodide gave compound 7a
and with ethyl iodide yielded 7aa. Analogously, the reaction of 5b
and 5c with ethyl bromide furnished the expected products 7b and
7c in excellent yields.
Herein we describe the synthesis and full characterization of ten
benzofuranyl-2-imidazole derivatives to shed light on the
structural-activity relationship of this family. Note, that the struc-
tural exploration involves substituents, either eletron donating
(methoxy) or electron attractor by inductive effect (bromide),
located in different positions of the phenyl ring, and alkylation of
the imidazole ring of the benzofuranyl-2-imidazole scaffold. We
assessed the pharmacological profile of the ten analogues and
selectivity through competition binding studies against the selec-
tive I2-IR radioligand [3H]2-BFI. Selectivity versus two related tar-
gets, the imidoline I1 receptor (I1-IR) and the a2-adrenergic
receptor (a2-AR) was evaluated through competition studies using
the selective radioligands [3H]clonidine and [3H]RX821002 (2-
methoxyidazoxan), respectively. Complementarily, we performed
three-dimensional quantitative structureꢀactivity relationship
(3D-QSAR) studies of this compound family and predicted in silico
the ADMET properties. LSL60101 endowed with the best I2-IR af-
finity and an excellent selectivity index regarding I1-IR and a2-AR
was selected for further studies. We carried out preliminary drug
metabolism and pharmacokinetics (DMPK) studies for LSL60101,
including chemical stability, PAMPA-BBB permeability assay, solu-
bility, cytotoxicity, microsomal stability, cytochromes inhibition,
and safety. The hypothermic properties and FADD multifunctional
protein (as a marker of neuroplasticity) regulation were also
studied following several LSL60101 treatments in mice. Pharma-
cokinetics was carried out prior to an in vivo treatment in a proper
murine model of AD. Thus, we further assessed the neuroprotective
effects of LSL60101 by evaluating specific OS markers under
oxidative damage and several transcription factors related to OS
machinery in 5xFAD mice (an early-onset mouse transgenic model
of AD).
All final products 5a-5d, 6b-6c and 7a, 7aa, 7b and 7c, were
completely characterized (see experimental section and supporting
information S19eS29) and all the tested compounds possess a
purity of at least 95% (see supporting information S30eS49).
2.2. Pharmacological evaluation
2.2.1. Radioligand I2-IR binding assays
The pharmacological profile of the ten compounds with the
structures 5, 6 and 7 (Scheme 1) was evaluated through competi-
tion binding studies against the selective I2-IR radioligand [3H]2-
BFI and the selective a2-AR radioligand [3H]RX821002. The
studies were performed in membranes from post-mortem human
frontal cortex, a brain area that shows an important density of I2-IR
and a2-AR. Idazoxan, a compound with well-established affinity for
I2-IR (pKi ¼ 7.41 0.63) and a2-AR (pKi ¼ 8.35 0.16) was used as
reference. The inhibition constant (Ki) for each compound was
obtained and is expressed as the corresponding pKi. The selectivity
for these two receptors was expressed by the I2/a2 index, calculated
as the antilogarithm of the ratio between pKi values for I2-IR and pKi
values for a2-AR. Competition experiments against [3H]2-BFI were
biphasic for most of the compounds (Table 1).
We have previously reported the affinity of 2-BFI in I2-IR human
brain (pKiH ¼ 9.08 and pKiL ¼ 7.15). The structural differences be-
tween 2-BFI and LSL60101 rely on the presence of an additional
double bond in the five membered ring, from a 2-imidazoline to an
imidazole conferring a planar structure to LSL60101 (see X-ray
crystallography discussion in the experimental section). The
mentioned structural difference rendered a decrease in the affinity
to pKiH ¼ 8.17 0.19 and pKiL ¼ 6.02 0.10, with 34% occupancy of
the high affinity site. The decreased affinity upon a2-AR plays in our
favor and selectivity had an excellent ratio of 3090.
2. Results and discussion
Next, electron-donating groups in the phenyl ring, such as a
methoxy group was considered. Thus, compounds 5b and 5c
bearing a methoxy group in the position ꢀ5 and ꢀ6, gave affinity
2.1. Chemistry
The preparation of the required final benzofuranyl-2-imidazoles
was accomplished, based on previous described procedure [33],
starting from the corresponding commercially available benzo-
furan-2-carboxylic acid derivatives (Scheme 1). Except for the
commercially available benzofuran-2-carbonitrile, 2a, the other
derivatives 2b, 2c, and 2d, were prepared in two steps. Treatment of
the corresponding benzofuran-2-carboxylic acid derivatives with
thionyl chloride and ammonium hydroxide furnished the carbox-
amides 1b, 1c, and 1d in excellent yields. Dehydration reaction with
phosphorus oxychloride gave benzofuran-2-carbonitriles, 2b, 2c,
and 2d, that were efficiently transformed in the corresponding
benzofuran-2-carbimididates hydrochlorides 3a, 3b, 3c and 3d af-
ter treatment with ethereal HCl 2 M. To construct the imidazole
moiety, the reaction with 2,2-dimethoxyethylamine was under-
taken to give 4a, 4b, 4c and 4d in quantitative yields, that were
treated with aqueous hydrochloric acid accomplish the attack of
the nitrogen atom to the ketal electrophilic carbon, affording ben-
zofuranyl-2-imidazoles 5a (named LSL60101), 5b, 5c and 5d.
Recrystallization of 5a as monocrystal confirmed its structure by X-
ray crystallographic analysis (see supporting information S51).
values
of
pKi ¼ 6.41 0.16
and
pKiH ¼ 6.77 0.29
and
pKiL ¼ 4.58 0.39, respectively. The introduction of a 5-bromine
furnished compound 5d with similar affinity values as LSL60101
(pKiH ¼ 8.63 0.51 and pKiL ¼ 5.85 0.18) but showing less selec-
tivity against a2-AR. The hydroxybenzofuran-2-imidazole de-
rivatives 6b and 6c showed affinities of pKiH ¼ 9.57 0.63 and
pKiL ¼ 4.6 0.26 and pKi ¼ 5.48 0.11, respectively, with a drop in
the I2/a2 selectivity in relation to their methoxy derivative partners
5b and 5c. Compounds 7a, 7aa and 7b bearing a N-alkylated
imidazole nucleus showed affinity that better fit to a two-site
binding model with pKiH ¼ 6.99 0.28 and pKiL ¼ 5.35 0.21,
pKiH ¼ 7.09 0.42 and pKiL ¼ 4.72 0.15, and pKiH ¼ 6.95 0.16 and
pKiL ¼ 4.16 0.12, with high affinity site occupancy of 38, 22 and
35%, respectively. Compared to the non-alkylated partners
LSL60101 and 5b, the additional N-alkyl group did not result in a
significant increase in the pKi value but in a drop in the I2/a2
selectivity. N-Ethyl substituted 7c gave an outstanding
pKiH ¼ 9.13 0.47 value of I2 affinity with a 12% high occupancy
affinity site, whereas the non-alkylated homologous 5c displayed a
and pKiL ¼ 5.11 0.06.
3

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
Scheme. 1. Reagents and conditions: (i) SOCl2, toluene, 3 h, reflux; 25% aq. NH4OH, rt; (ii) POCl3, dichloroethane, 75 ꢁC, 2 h; (iii) Et2O.HCl 2 M, 4 ꢁC, 48 h; (iv) 2,2-
dimethoxyethylamine, methanol, 60 ꢁC, 16 h; (v) HCl 2 M, 60 ꢁC, 16 h; (vi) HBr 47%, 100 ꢁC, 7 h; (vii) NaH, methyl iodide or ethyl bromide, DMF, 0 ꢁC to rt, 75 min.
The study highlighted LSL60101 as the most promising candi-
date of the benzofuran-2-imidazole family to tackle further in vitro
and in vivo studies.
and pKiL ¼ 1.16, but a higher percentage fraction of occupancy (79%)
[34]. In mice brain cortical membranes, the competition curve
against [3H]2-BFI binding was significantly better with a two-site fit
than a one-site binding model, providing a pKiH 9.92 0.17 and
pKiL ¼ 6.00 0.14. These values were close to those found in human
tissues including a similar occupancy of the high site (41%) sup-
porting the in vivo experiments in mice.
2.2.2. Comparison of I2-IR binding affinities of LSL60101 across
species
The literature is non-uniform in the I2-IR affinity values across
species (human, rat, rabbit, mouse, and monkey), given the radio-
ligand considered as a reference (idazoxan, 2-BFI) and the tissues
used for analysis (from different anatomical parts: kidney, whole
brain, cortex, etc). Thus, we made an effort in the comparison of
affinities of standard I2-IR ligands, 2-BFI, tracizoline, and clinically
2.3. 3D-QSAR study
The 3D-QSAR method was used to analyse the most significant
descriptors that were interpreted in terms of identifying and
quantifying structural elements important for I2-IR and a2-AR ac-
tivity. The study was conducted on structurally diverse I2-IR ligands
that were divided into two clusters, based on their chemical
structures. Cluster 1 represents compounds synthesized in this
prominent
[
11C]BU99008 and CR4056 [23]. When evaluating
LSL60101 a two-site model of binding was observed in human post-
mortem brain cortical membranes with a pKiH 8.17 0.19 and
pKiL ¼ 6.02 0.10, and an occupancy of the high site of 34%. In rat
cerebral cortex using idazoxan as radioligand, LSL60101 was re-
ported a lower affinity than in human tissues, with a pKiH ¼ 6.45
manuscript, while cluster 2 contains bicyclic
I2-IR ligands previously described by our group (Fig. S1) [23]. To
a-iminophosphonate
4

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
Table 1
I2-IR and a2-AR Binding Affinities (pKi) of compound idazoxan and 2-BFI and new compounds 5a-5d, 6b, 6c, 7a, 7aa, 7b and 7c.
Compound
High-affinity site %
[3H]-RX821002
a2 pKi
Selectivity I2/a2
a[3H]2-BFI
b[3H]-2-BFI
R1/R2/R3
I2 pKi one site
I2 pKi two sites
General structure
Idazoxan
7.41 0.63
7.87 0.74
5.76 0.57
40
7
8.35 0.16
e
2-BFI
8.31 0.13
6.67 0.09
9.08 0.22
8.17 0.19
7.15 0.31
6.02 0.10
58
34
9
4
4.58 0.22
3.18 0.17
5370
3090
5a, LSL60101
R1 ¼ R2 ¼ R3 ¼ H
5b
6.41 0.16
5.88 0.16
6.28 0.18
4.87 0.23
5.48 0.11
5.98 0.08
5.05 0.10
4.96 0.15
5.26 0.08
3.94 0.07
3.01 0.45
4.92 0.25
3.84 0.16
3.76 0.12
3.75 0.12
3.77 0.08
4.21 0.19
3.17 0.25
295
741
23
e
e
e
R1 ¼ OMe, R2 ¼ R3 ¼ H
5c
6.77 0.29
4.58 0.39
46
9
5
5
R1 ¼ H, R2 ¼ OMe, R3 ¼ H
5d
8.63 0.51
9.57 0.63
5.85 0.18
4.6 0.25
20
29
R1 ¼ Br, R2 ¼ H, R3 ¼ H
6b
11
R1 ¼ OH, R2 ¼ H, R3 ¼ H
6c
52
e
e
e
R1 ¼ H, R2 ¼ OH, R3 ¼ H
7a
6.99 0.28
5.35 0.21
38 10
170
19
R1 ¼ R2 ¼ H, R3 ¼ Me
7aa
7.09 0.42
6.95 0.16
9.13 0.47
4.72 0.15
4.16 0.12
5.11 0.06
22
35
12
6
3
2
R1 ¼ R2 ¼ H, R3 ¼ Et
7b
6
R1 ¼ OMe, R2 ¼ H, R3 ¼ Et
7c
123
R1 ¼ H, R2 ¼ OMe, R3 ¼ Et
a
Selectivity I2-IR/a2-AR expressed as the antilog (pKi I2-IR ꢀ pKi a2-AR).
b
The best fit of the data for most of the compounds was to a two-site binding model with high pKi (pKiH) and low pKi (pKiL) affinities for both binding sites, respectively.
compare and validate our results, we have added three I2-IR stan-
dard ligands (tracizoline, idazoxan, and BU99008, Fig. 1), in both
data sets. Structural diversity of prepared data set enabled us not
only to deeper analyse the most important structural characteris-
tics, but also to suggest modifications to come up with novel
compounds with improved I2-IR activity and selectivity.
The Pentacle program [38] was used for calculation of GRID
independent descriptors (GRIND and GRIND2) and 3D-QSAR model
building. The reliability and predictive power of the created 3D-
QSAR models were assessed using different internal and external
validation parameters (Tables S1 and S2). Obtained results indi-
cated that both models possessed good predictive capability and
could be used for activity prediction of newly designed compounds.
PLS coefficient plots presented the most important variables with
positive and negative influence on I2-IR and a2-AR activity (Figs. S2
and S3).
activity. This variable implies the importance of hydrogen bond
donor group, eNH from imidazoline ring, located at a certain dis-
tance from hydrophobic region, benzofuranyl ring (Fig. 2a). Con-
trary, it did not possess such a significant influence in the
compounds from cluster 2, since they lack hydrogen bond donor
groups (Fig. S4). GRIND variable OeN1 (var394: 6.40e6.80 Å)
signified a positive influence of the imidazole ring on I2-IR activity,
describing the distance between hydrogen bond acceptor and
donor probes located around nitrogen atoms (Fig. 2a). On the other
hand, it did not show a significant effect on a2-AR activity.
Furthermore, the importance of hydrogen bonding interactions in
the binding site of I2 receptor was confirmed with variables var452
(O-TIP: 8.00e8.40 Å) and var503 (N1-TIP: 6.80e7.20 Å), which
underlined optimal distances between the steric region around the
benzofuranyl ring and eNH from the imidazoline ring as a
hydrogen bond donor or acceptor (Fig. 2a). Therefore, we can
conclude that the substitution of the eNH imidazole group (R3) and
losing of the hydrogen bond donating characteristics negatively
correlated with I2-IR binding activity and selectivity (7a, 7aa, 7c,
7b). Analysis of a2-AR 3D-QSAR model showed that the sole
introduction of a hydrogen bond acceptor-group, at a certain dis-
tance from hydrophobic or steric region around heterocyclic ring,
induced a positive impact on a2-AR activity.
Compounds from cluster 1 are compounds reported here
(Scheme 1) with pKi ranges between 4.87 and 6.67 for I2-IR, and
3.01e4.92 for a2-AR, while cluster 2 compounds present our bicy-
clic a-iminophosphonate derivatives (Fig. S1) with pKi ranging from
4.02 to 8.56 for I2-IR, and 3.38e6.77 for a2-AR. Comparing to
molecules from cluster 2, cluster 1 compounds possess lower pKi
values for I2-IR as well as a2-AR.
The most potent compound from cluster 1, LSL60101
(pKi ¼ 6.67), displayed all significant variables with the most posi-
The compound with the lowest activity within the cluster 1 is 6b
(pKi ¼ 4.87). The introduction of a hydroxy or methoxy group on
the carbon atom of the benzofuranyl nucleus resulted in reduced
affinity, which could be described with var197 (TIP-TIP:
14.00e14.40 Å) and var350 (DRY-TIP: 10.40e10.80 Å) (Fig. 2b).
Additionally, the unfavourable impact of these groups is also
defined with the negative variable var408 (OeN1), which explains
the distance of 12.00e12.40 Å between eNH of imidazoline, as a
hydrogen bond donor and the oxygen atom in hydroxy or methoxy
substituent as a hydrogen bond acceptor (Fig. 2b). Based on these
findings we can conclude that the introduction of a hydrogen bond
tive impact on the activity. Positive influence on both I2-IR and a2
-
AR activity showed variables var24 (DRY-DRY: 9.60e10.00 Å) and
var25 (DRY-DRY: 10.00e10.40 Å), respectively. They described the
optimal distance range between two hydrophobic regions, around
benzofuranyl and imidazole rings, which may be crucial for
establishing favourable van der Waals interactions with amino
acids in the active pocket of both receptors (Fig. 2a). Moreover, I2-IR
3D-QSAR model pointed out var232 (DRY-O: 6.40e6.80 Å) as the
variable with the strongest positive influence on I2-IR binding
5

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
Fig. 2. Representation of positive interactions (in red) of LSL60101 (a) and negative interactions (in blue) of 6b; (b) in I2-IR 3D-QSAR model. The steric hot spots (TIP) are presented
in green, hydrophobic regions (DRY) in yellow, H-bond acceptor regions (N1) in blue, and H-bond donor regions (N1) in red.
acceptor substituent on the benzofuranyl ring may not be consid-
ered as the most complemental with the binding site of I2-IR.
I2-IR ligands with potential therapeutic applications in the neuro-
protective field. Guaranteed by the in silico parameters the in vitro
permeability (Pe) of all the novel compounds was determined by
using the PAMPA-BBB permeability assay (Table 2). The new com-
pounds prepared were well above the threshold established for
high BBB permeation (Pe > 5.198 ꢂ 10ꢀ6 cm sꢀ1), except for com-
pounds 6b and 6c. The aforementioned two compounds bear a
hydroxyl group increasing their polarity and decreasing their
capability to permeate the artificial PAMPA-BBB membrane with Pe
2.4. In silico ADMET analysis of physicochemical and
pharmacokinetic parameters
The drug discovery pipeline has more and more relied on in
silico predictions to optimize lead compounds and reduce in-
vestments [39]. In silico ADMET prediction aims to evaluate indi-
values of 0.1 0.03 ꢂ 10ꢀ6 cm sꢀ1 and 0.32 0.1 ꢂ 10ꢀ6 cm sꢀ1
,
vidual
ADMET
behaviours
of
examined
compounds.
respectively. In particular, the most I2-IR affine compound
LSL60101 had a Pe value of 13.6 0.4 ꢂ 10ꢀ6 cm sꢀ1 and was
considered suitable to envisage further in vitro and in vivo studies
oriented to in-depth the pharmacological profile of the new family
of I2-IR ligands.
According to these data, compounds including a hydroxyl group
in its structure, such as 6b and 6c, were not suitable for considering
their potential in neurodegenerative diseases. In particular,
LSL60101 showed the best affinity and selectivity values and since
it had a good ability to cross the BBB it will be used to undertake
further studies.
Pharmacokinetic properties such as absorption, distribution,
metabolism, excretion, and toxicity (ADMET) profiling of com-
pounds were determined using ADMET Predictor software [40],
while physico-chemical parameters were assessed with Swis-
sADME online programme [41]. The obtained results are presented
in Tables S3 and S4. Based on the results obtained from the per-
formed calculations, we can conclude that all studied compounds
present good water solubility and lipophilicity. Compounds re-
ported here satisfy the Lipinski's Rule of 5, which supports their
drug-likeness properties and potential chance to be orally
bioavailable. Moreover, the polarity of compounds was evaluated
by the TPSA (topological polar surface area) descriptor and results
revealed that benzofuranyl-2-imidazoles possessed lower polarity,
similar to idazoxan, when compared to previously described bicy-
2.6. Selectivity I2-IR versus I1-IR in LSL60101
clic a-iminophosphonate derivatives (see supporting information,
LSL60101 showed a remarkable affinity for I2-IR and selectivity
I2/a2-AR with a ratio 3090. Then, we assessed the affinity/selectivity
Fig. S1) [23]. Regarding pharmacokinetic properties, we noted that
all molecules possessed high BBB permeation. Compared to stan-
dard molecules, such as idazoxan, all examined compounds
possessed lower percentage of unbound drug in plasma. Further-
more, this family showed lower metabolic CYP risk and TOX risk
Table 2
Permeability results (Pe 10ꢀ6 cm sꢀ1) from the PAMPA-BBB assay for new report
compounds and their prediction of BBB permeation.
comparing to idazoxan and bicyclic
a-iminophosphonate de-
rivatives. P-gp is believed to play an important role in drug distri-
bution and resistance to CNS drug treatment. The examined
compounds were not identified as potential substrates for P-gp
transporters. Since blocking hERG channels represents a major
therapeutic challenge in drug discovery, we note that the com-
pounds synthesized in this manuscript did not show affinity to
inhibit hERG channels.
The absence of warnings of this theoretical study gave us con-
fidence for undertaking further in vitro and eventually, in vivo ex-
periments to assess the benzofuranyl-2-imidazole family and
LSL60101 as a neuroprotective agent.
Compound
aPe 10¡6 cm s¡1
bPrediction
Idazoxan
2-BFI
5a, LSL60101
5b
5c
5d
6b
6c
7a
7aa
7b
7c
3.3 0.1
6.1 0.2
13.6 0.4
8.3 1.2
7.6 0.1
20.3 0.2
0.32 0.1
0.1 0.03
17.4 0.85
19.2 1.1
12.4 0.8
14.85 0.1
CNS
CNSþ
CNSþ
CNSþ
CNSþ
CNSþ
CNS-
CNS-
CNSþ
CNSþ
CNSþ
CNSþ
a
PBS/EtOH (70:30) was used as solvent. Values are expressed as mean SD of at
least three independent experiments.
2.5. Blood brain barrier permeation assay
b
The in vitro permeability (Pe) of fourteen commercial drugs through lipid extract
Considering the localization of I2-IR in the CNS, a good ability to
cross the BBB is an essential requirement for developing effective
of porcine brain membrane together with the test compounds were determined (for
the commercial drug values see Table S5).
6

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
for the very close receptors I1-IR. Specific binding of [3H]clonidine
(20 nM) to I1-IR of rat or human hypothalamic membranes was
accomplished after pre-incubation with benextramine (10 mM) to
affinity values of LSL60101 it is not expected that at therapeutic
doses the compound may interfere with the cytochrome P450-
mediated metabolism of other drugs.
alkylate population of a2-AR. Under these experimental conditions
[3H]clonidine only labelled I1-IR and in drug competition experi-
ments moxonidine, a known I1-IR selective compound, showed
The plasma stability of LSL60101 assessed in humans pooled
from healthy donors was measured up to 6 h (0 h, 1 h, 2 h and 6 h)
revealing 100, 83, 77 and 57% of the remaining percentage
(Table S9). In mouse plasma LSL60101 was inert under the condi-
tions studied, remaining as 100% of the initial compound after 6 h
(Table S10).
The extent of plasma protein binding was slightly superior in
human that in mouse plasma. The fraction unbound value is also
reported in Table S11. Whereas a 7.1% of LSL60101 can be found in
humans as free drug, a 16.0% was observed in mouse.
subnanomolar affinity for these I1-sites; KiH ¼ 0.2 nM; KiL ¼ 12
LSL60101 displayed a very low affinity for I1-IR, in rat samples
Ki ¼ 115 M and in human samples Ki ¼ 250 M.
mM.
m
m
2.7. Acute toxicity
Based on the excellent affinity/selectivity upon I2-IR of
LSL60101, the safety evaluation was undertaken to determine the
acute toxicity of the compound. LSL60101 was safe up to a dose
level of 100 mg/kg body weight after an intraperitoneal adminis-
tration, since the LD50 was considered to be greater than 100 mg/kg
(Probit analysis) in male albino mice (lack of toxicity in the range of
doses tested: 10e100 mg/kg; n ¼ 10 mice per dose tested).
The effect of LSL60101 over the activity of hERG, an important
safety issue in drug discovery, was assessed and showed an
inhibitory activity (%) of 4 1 at 10
any worries on this issue.
mM concentration, discarding
2.9. Cytotoxicity
2.8. ADME-DMPK profiling of LSL60101
In order to exclude compounds with adverse cellular effects, the
newly synthesized compounds were evaluated for cytotoxicity.
Human primary cells are physiologically more relevant. However,
because of the difficulties of working with primary cultures,
immortalized cell lines were used as an in vitro screening tool for
cytotoxicity. In a panel of cancer cell lines, including LN-229
(glioblastoma), Capan-1 (pancreatic adenocarcinoma), HCT-116
(colorectal carcinoma), NCIeH460 (lung carcinoma), DND-41
(acute lymphoblastic leukemia), HL-60 (acute myeloid leukemia),
K-562 (chronic myeloid leukemia) and Z-138 (non-Hodgkin lym-
phoma) cell lines, none of the compounds displayed any cytotox-
With a compound showing promising binding properties/
selectivity, devoid to a2-AR and I1-IR, and optimal safety, we un-
dertook in vitro assays to define its physicochemical properties and
chemical stability.
The solubility of LSL60101 in 1% DMSO and 99% PBS buffer was
excellent (52.5 mM). For a better definition, the solubility of
LSL60101 at different pH 1.13, 4.61 and 6.97, simulating the pH
range of the gastric and intestinal fluids, was studied (see sup-
porting information S10). Following the solubility terms stablished
in European Pharmacopoeia vs.10.0, LSL60101 is a soluble substance
at pH 1.13 and 4.61 and a very slightly soluble substance at pH 6.97
(Table S6).
icity at 100 mM (the highest concentration tested), as determined by
a real-time IncuCyte proliferation assay.
The chemical evaluation of LSL60101 implied forced degrada-
tion studies under different stress conditions for a period of nine
weeks, monitoring weekly the assays by HPLC and 1H NMR. In
particular, LSL60101 was subjected to the effect of daylight with
temperatures between 0 and 23 ꢁC and a relative humidity of
25e85%, to the effect of high temperature (thermal stability at
75 ꢁC), and to the continuous light of a 100W (230V) bulb. Analysis
by HPLC showed that the compound was completely stable under
all the aforementioned conditions. Calculated lipophilicity of
LSL60101 referring to the consensus log Po/w value calculated using
the SwissADME program for five predictive log Po/w models, iLOGP,
XLOGP3, WLOGP, MLOGP and SILICOS-IT, which gave values of 1.51,
2.16, 2.82, 1.15 and 3.02, respectively. Therefore, the calculated log
Po/w and the other parameters of Lipinski were within the limits
and the compounds are suitable for undertaking the characteriza-
tion of in vitro ADME properties.
Microsomal stability, which is widely used to determine the
likely degree of primary metabolic clearance in the liver, was
assessed in human and mouse recombinant microsomes. The data
shown in Table S7 reveals a 36% percentage of remaining com-
pound, after 60 min of incubation in mouse microsomes, and 58% in
human, indicating moderate differences in the metabolism
depending on the species. The t1/2 is 1.4-fold bigger in human than
in mice, therefore the difference should be taken into consideration
through additional preclinical studies.
2.10. Pharmacokinetics
The pharmacokinetic profile of LSL60101 was investigated prior
to the treatment of a murine model of AD (5xFAD). Following a
single oral administration of 10 mg/kg of LSL60101 in CD1 mice,
plasma concentrations of drug were found after 15 min of treat-
ment and were detected for 24 h. Absorption of drug from the
gastrointestinal tract was slowly reaching Cmax (3.24 mM) at 2 h
after dosing and t1/2 was around 7 h. The narrow differences in
b
AUCt0 and AUC∞0 showed complete exposure, good bioavailability
and appropriate elimination of LSL60101 to reach the therapeutic
potential of I2-IR ligands in the experimental conditions described
(Fig. 3).
Furthermore, the remarkable affinity of LSL60101 for I2-IR (pKi
in mouse brain cortical membranes ranged from 9.92 to 6) and the
excellent selectivity ratio I2/a2-AR (3090) could guarantee, from the
kinetic profile observed in plasma, that the concentrations reached
at the site of action are enough to demonstrate the efficacy of this
drug.
Overall, these studies confirmed that LSL60101 is orally
bioavailable and metabolically stable and can be used for further
in vivo experiments.
2.11. Hypothermic effects of LSL60101 in naïve mice
The inhibition potential of LSL60101 was evaluated using re-
combinant human cytochrome P450 enzymes [CYP1A2, CYP2C9,
CYP2C19, CYP2D6, CYP3A4 (7-BFC) and CYP3A4 (DBF)] and probe
substances with fluorescent detection. The results depicted in
Table S8 showed no inhibition of the cytochromes considered at
Several previous studies have proven the induction of acute
hypothermia by I2-IR ligands in rats [42,43] and mice [21e23]. In
this regard, the present study evaluated the hypothermic effects of
LSL60101 in a wide range of doses (1, 5, 10, 20, 30, 50 mg/kg i.p.) in
adult male and female CD1 mice at different times post-injection.
The results showed a sharp drop in core body temperature
10 mM concentration. Taking into account the range of nM in the
7

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
Fig. 3. Plasma concentration of LSL60101 at different times (15 mine24 h) after an oral administration of 10 mg/kg, determined by HPLC/UV-VS at 290 nm. Basic pharmacokinetic
parameters were calculated.
(ranging from ꢀ1.1 to ꢀ3.9 ꢁC) for the larger dose tested (50 mg/kg)
and observed 1 and up to 2 h post-injection (Fig. 4). Core body
temperature returned to basal levels 3 h after drug administration.
The present results, in line with prior data [42], suggest the need
for high doses of LSL60101 to induce the expected hypothermia
characteristic of I2-IR ligands [21e23,43]. As for the role of this
acute pharmacological effect, hypothermia is known to provide
neuroprotection in models of cerebral ischemia; even mild tem-
perature drops can cause significant neuroprotection [44]. In fact,
hypothermia has been used to improve the neurological outcome
under various pathological conditions, including stroke and trau-
matic brain injury [45,46]. In this line of thought, LSL60101 has
proven to partially prevent neuronal death in rats following
neonatal axotomy [36], and thus, its hypothermic effects might be a
relevant feature that could be mediating certain degree of
neuroprotection.
Although the agonist or antagonist nature of I2-IR ligands has
been a topic of debate [2], it has been suggested that I2-IR agonists
reliably decrease body temperature in a highly quantitative manner
in rodents [42], suggesting this assay could be used as a sensitive
in vivo assay for studying I2 receptor ligands, and in line with the
hypothermic effects observed in the present study in mice treated
with LSL60101.
controlling cell fate regulation, balancing pro-apoptotic and/or
neuroprotective actions in rodents [21,47,48]. The acute treatment
with
a high dose of LSL60101 (50 mg/kg, i.p.) significantly
decreased (by _34%) hippocampal FADD protein content as
compared to vehicle-treated mice (Fig. 5). The present results
replicated earlier studies in which other I2-IR ligands decreased
FADD hippocampal content [21,23], and suggested the induction of
non-apoptotic (e.g., neuroplastic and or neuroprotective) actions
initiated by acute LSL60101 treatment in mice brain and in parallel
to the hypothermic effects.
2.13. Effects of chronic LSL60101 at a low dose in a mice model of
Alzheimer's disease
Accumulating evidence support that I2-IR exert neuroprotective
roles in a plethora of neurodegenerative disorders, such as AD
[8,9,11,49,50]. Previous studies suggested that I2-IR in the CNS are
mainly located on the outer membrane of mitochondria in astro-
cytes [51]. Given that alterations in mitochondrial function pro-
mote an increased ROS production, which combined with an
altered antioxidant defence contribute to the early stages of AD
before the development of Ab pathology and cognitive dysfunction
[29,30,52], here we tested the effects of a low dose of LSL60101
administered following a chronic paradigm (1 mg/kg/day, for 4
weeks) on specific OS markers under oxidative damage in 5XFAD
mice, an early-onset mouse transgenic model of AD, and as
compared to a wild type strain. The results showed increased gene
expression of the so-called antioxidant response element (AREs),
such as heme oxygenase 1 (Hmox1), aldehyde dehydrogenase 2
(Aldh2) and iNOS [53] in 5xFAD vs. wild type mice (Fig. 6). By
contrast, in 5xFAD mice treated with LSL60101, the expression of
those decreased in a significant way.
2.12. Effects of acute LSL60101 on hippocampal FADD protein
content in naïve mice
In the context of neuroprotection, FADD adaptor emerges as a
key multifunctional protein involved in the mechanisms
On the other hand, an excess in ROS is removed by antioxidant
enzymes (e.g., superoxide dismutase, SOD1, and glutathione
peroxidase, GPX-1). 5XFAD mice showed a significant decrease in
both proteins levels in reference to wild type mice. Interestingly,
LSL60101 was able to significantly increase the levels of these key
protein, indicating that LSL60101 induced the activation of a
cellular signalling cascade that led to a reduced OS, and that in turn
could have a neuroprotective role in an oxidative environment
related with neurodegenerative processes such as AD (Fig. 7).
These results are in agreement with recent studies that
demonstrated a key role for I2-IR ligands in the OS process. In
particular, 2-BFI (a selective ligand to I2-IR), decreased OS and
altered the level of anti-oxidant enzymes in an AD rat model [54],
protecting against OS-induced astrocytic cell death [55]. Ultimately,
previous work from our group reported a decrease in hydrogen
peroxide levels and OS markers induced by the I2-IR ligands MCR5
Fig. 4. Acute effects of LSL60101 treatment on core body temperature in mice. Sym-
bols represent means SEM of the difference (
D
, 1, 2 or 3 h minus basal value) in body
temperature (ꢁC) for each treatment group. ***p
<
0.001 and *p < 0.05 when
comparing the dose of 50 mg/kg with the control group (repeated measures ANOVA
followed by Sidak's comparison test).
8

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
4. Experimental section
4.1. Chemistry
Reagents, solvents and starting products were acquired from
commercial sources. The term “concentration” refers to the vacuum
evaporation using a Büchi rotavapor. When indicated, the reaction
products were purified by “flash” chromatography on silica gel
(35e70 mm) with the indicated solvent system. IR spectra were
performed in a Spectrum Two FT-IR Spectrometer, and only note-
worthy IR absorptions (cmꢀ1) are listed. NMR spectra were recor-
ded in DMSO‑d6 at 400 MHz (1H) and 100.6 MHz (13C), and
chemical shifts are reported in
d values downfield from TMS or
relative to residual DMSO‑d6 (2.50 ppm, 39.5 ppm) as an internal
standard. Data are reported in the following manner: chemical
shift, multiplicity, coupling constant (J) in hertz (Hz) and integrated
intensity. Multiplicities are reported using the following abbrevia-
tions: s, singlet; d, doublet; dd, double doublet; q, quadrupet; t,
triplet; m, multiplet; br s, broad signal. The accurate mass analyses
were carried out using a LC/MSD-TOF spectrophotometer. HPLC-MS
(Agilent 1260 Infinity II) analysis was conducted on a Poroshell
120 EC-C15 (4.6 mm ꢂ 50 mm, 2.7
m
m) at 40 ꢁC with mobile phase A
(H2O þ 0.05% formic acid) and B (ACN þ 0.05% formic acid) using a
gradient elution and flow rate 0.6 mL/min. The DAD detector was
set at 254 nm, the injection volume was 5 mL, and oven temperature
Fig. 5. Acute effects of LSL60101 (30 or 50 mg/kg, i.p.) on hippocampal FADD protein.
Columns are means SEM of FADD content (% C) for each treatment group. Individual
symbols are shown for each animal. One-way ANOVA followed by Sidak's multiple
comparisons tests: **p < 0.01 vs. C (control-treated mice). Representative immuno-
was 40 ꢁC. All tested compounds possess a purity of at least 95%.
4.1.1. General procedures for the synthesis of benzofuran-2-
carboxamides 1b, 1c and 1d
blots depicting the labelling of FADD and
treatment group are shown below.
b-actin (as a loading control) for each
Thionyl chloride (1.65 equiv) was added to a suspension of
benzofuran-2-carboxylic acid derivatives (1 equiv) in anhydrous
toluene (0.4 mmol/mL). After stirring the mixture for 3 h, at reflux,
the reaction was cooled and concentrated. Then, the resulting
benzofuran-2-carbonyl chloride derivative (1 equiv) was added in
small portions to an ice-cold solution of 25% aq. NH4OH (0.5 mmol/
mL). Upon completion of the addition the reaction mixture was
allowed to reach rt and a precipitate was formed. The solid was
collected by filtration, washed with cold water and dried under
vacuum.
and MCR9 in aged SAMP8 mice [22]. Recently, we demonstrated
that LSL60101 can ameliorate the AD hallmarks and neuro-
inflammation [56]. Moreover, the chronic treatment with LSL60101
led to the induction of reactive astrocytes and the up-regulation of
the expression of the astrocyte marker glial fibrillary acidic protein
(GFAP) [34]. As mentioned, astrocytes are the main supportive cells
in the CNS and are significantly involved in the redox homeostasis
[57], and thus, and as a consequence, this could be proposed as a
possible effect of LSL60101 on OS balance.
4.1.1.1. 5-Methoxybenzofuran-2-carboxamide (1b). Following the
general procedure, thionyl chloride (17.2 mmol, 1.25 mL), 5-
bromobenzofuran-2-carboxylic acid (10.4 mmol, 2.0 g), anhydrous
toluene (25 mL) and 25% aq. NH4OH (20 mL) gave 1b (1.87 g, 98%) as
a beige solid.
3. Conclusions
We have evaluated the binding and selective properties upon I2-
IR in human brain tissues of a series of benzofuranyl-2-imidazoles
diversely substituted in both benzofuranyl and imidazole rings. Due
to the lack of structural description of these receptors, 3D-QSAR
and in silico physicochemical properties were performed in order to
determine the relevant elements that may allow the further
structural optimization of new molecules. The secure theoretical
DMPK of the family led us to undertake in vitro studies including
PAMPA to confirm their penetration into the CNS to address
neurodegenerative issues. The safe in vitro DMPK and cytotoxicity
assays of the selected LSL60101 opened the door to in vivo studies.
After the determination of its pharmacokinetic profile, the treat-
ment of animals with LSL60101 confirmed a decrease in the con-
tent of hippocampal FADD protein, a key signalling mediator of
neuroprotective actions. The I2-IR ligand LSL60101 also fostered a
diminution in oxidative stress biomarkers in an AD murine model
(5xFAD). Thus, the modulation of I2-IR by LSL60101 is proposed as a
promising opportunity for addressing AD therapeutics and invites
for the further design of new promising benzofuranyl-2-imidazole-
base structures to be added to the scarce arsenal of I2-IR ligands.
4.1.1.2. 6-Methoxybenzofuran-2-carboxamide (1c). Following the
general procedure, thionyl chloride (8.60 mmol, 0.63 mL), 6-
methoxybenzofuran-2-carboxylic acid (5.20 mmol, 1.0 g), anhy-
drous toluene (13 mL) and 25% aq. NH4OH (10 mL) gave 1c (930 mg,
97%) as a beige solid.
4.1.1.3. 5-Bromobenzofuran-2-carboxamide (1d). Following the
general procedure, thionyl chloride (13.7 mmol, 1.0 mL), 5-
bromobenzofuran-2-carboxylic acid (8.30 mmol, 2.0 g), anhydrous
toluene (21 mL) and 25% aq. NH4OH (17 mL) gave 1d (1.98 g, 99%) as
a white solid.
4.1.2. General procedure for the synthesis of benzofuran-2-
carbonitrile 2b, 2c and 2d
Phosphoryl chloride (3 equiv) was added to a solution of
benzofuran-2-carboxamide derivative (1 equiv) in dichloroethane
(0.48 mmol/mL). The reaction was stirred at 75 ꢁC for 2 h. Then, the
reaction mixture was evaporated and neutralized with saturated
NaHCO3 solution. The aqueous phase was extracted with AcOEt, the
9

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
Fig. 6. Chronic effects of LSL60101 (1 mg/kg/day, per os.) on Hmox1, Aldh2 and iNOS gene expression in hippocampus. Columns are the mean SEM for each treatment group.
Individual symbols are shown for each animal. Means were compared with two-way ANOVAs, followed by Tukey-Kramer multiple comparison post-hoc analysis: ****p < 0.0001;
***p < 0.001; **p < 0.01; *p < 0.05 vs. Wt Controls.
Fig. 7. Chronic effects of LSL60101 (1 mg/kg/day, per os.) on the expression of SOD1 and GPX-1 protein levels in hippocampus. Columns are the mean SEM for each treatment
group. Individual symbols are shown for each animal. Means were compared with two-way ANOVAs, followed by Tukey-Kramer multiple comparisons post-hoc analysis. *p < 0.01
vs. Wt control mice).
combined organic phases were washed with brine, dried over
Na2SO4, and evaporated to give a residue, which was purified by
flash column chromatography.
phosphoryl chloride (23.8 mmol, 2.21 mL) gave 2d (1.55 g, 88%) as a
white solid, after column chromatography (dichloromethane
100%).
4.1.2.1. 5-Methoxybenzofuran-2-carbonitrile (2b). Following the
general procedure, 1b (9.62 mmol, 1.84 g), dichloroethane (20 mL)
and phosphoryl chloride (28.9 mmol, 2.70 mL) gave 2b (1.42 g, 85%)
as a beige solid, after column chromatography (dichloromethane
100%).
4.1.3. General procedure for the synthesis of methyl benzofuran-2-
carbimididate hydrochlorides 3a, 3b, 3c and 3d
The 2-cyanobenzofuran derivative (1 equiv) was dissolved in
ethereal HCl 2 M (0.25 mmol/mL) and methanol (5 mmol/mL). The
resulting mixture was kept at 4 ꢁC for 48 h. The resulting solid was
filtered, washed with cold ether and dried in order to obtain the
desired carbimidate hydrochloride.
4.1.2.2. 6-Methoxybenzofuran-2-carbonitrile (2c). Following the
general procedure, 1c (4.87 mmol, 930 mg), dichloroethane (10 mL)
and phosphoryl chloride (14.6 mmol, 1.36 mL) gave 2c (740 mg,
88%) as a beige solid, after column chromatography (dichloro-
methane 100%).
4.1.3.1. Methyl benzofuran-2-carbimididate hydrochloride (3a).
Following the general procedure, benzofuran-2-carbonitrile
(6.99 mmol, 1.0 g), ethereal HCl 2 M (28 mL) and methanol
(1.4 mL) gave 3a (1.29 g, 87%) as a white solid.
4.1.2.3. 5-Bromobenzofuran-2-carbonitrile (2d). Following the gen-
eral procedure, 1d (7.92 mmol, 1.90 g), dichloroethane (17 mL) and
10

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
4.1.3.2. Methyl 5-methoxybenzofuran-2-carbimididate hydrochloride
(3b). Following the general procedure, 2b (8.08 mmol, 1.40 g),
ethereal HCl 2 M (32 mL), methanol (2 mL) gave 3b (1.62 g, 83%) as a
white solid.
2H), 7.88 (d, J ¼ 7.5 Hz, 1H), 8.12 (s, 1H). 13C NMR (100.6 MHz)
d
109.8, 111.6, 120.8 (2C), 122.9, 124.4, 127.1, 127.3, 135.1, 140.6, 154.6.
HRMS C11H9N2O [MþH]þ 185.0709; found, 185.0706. Purity 99.6%
(tR ¼ 3.08 min).
4.1.3.3. Methyl 6-methoxybenzofuran-2-carbimididate hydrochloride
(3c). Following the general procedure, 2c (4.04 mmol, 700 mg),
ethereal HCl 2 M (16 mL) and methanol (1 mL) gave 3c (797 mg,
82%) as a white solid.
4.1.5.2. 2-(5-Methoxybenzofuran-2-yl)-1H-imidazole hydrochloride
(5b). Following the general procedure, 4b (4.10 mmol, 1.14 g), HCl
2 M (41 mL) and ethereal HCl 2 M (2.7 mL) gave 5b (950 mg, 93%) as
a white solid. IR (ATR) 3449, 3076, 2587,1647,1457,1434,1206,1157,
1018, 812, 776, 712 cmꢀ1 1H NMR (400 MHz, DMSO‑d6)
. d 3.81 (s,
4.1.3.4. Methyl 5-bromobenzofuran-2-carbimididate hydrochloride
(3d). Following the general procedure, 2d (4.82 mmol, 1.10 g),
ethereal HCl 2 M (19 mL) and methanol (1 mL) gave 3d (1.20 g, 86%)
as a white solid.
3H), 7.08 (dd, J ¼ 9.0, 2.5 Hz, 1H), 7.38 (d, J ¼ 2.5 Hz, 1H), 7.61 (d,
J ¼ 9.0 Hz, 1H), 7.82 (s, 2H), 8.10 (s, 1H). 13C NMR (100.6 MHz)
d 55.7,
104.4, 110.2, 112.2, 116.5, 120.7 (2C), 127.8, 135.0, 141.1, 149.5, 156.5.
HRMS C12H11N2O2 [MþH]þ 215.0815; found, 215.0816. Purity 98.5%
(tR ¼ 3.19 min).
4.1.4. General procedure for the synthesis of N-(2,2-
dimethoxyethyl)benzofuran-2-carboximidamide 4a, 4b, 4c and 4d
A solution of 2,2-dimethoxyehtylamine (1.1 equiv) and methyl
benzofuran-2-carbimidate hydrochloride derivative (1 equiv) in
methanol (0.47 mmol/mL) was stirred at 60 ꢁC for 16 h. The mixture
was evaporated to dryness, which was used directly in the next step
without further purification.
4.1.5.3. 2-(6-Methoxybenzofuran-2-yl)-1H-imidazole hydrochloride
(5c). Following the general procedure, 4c (2.93 mmol, 815 mg), HCl
2 M (29 mL) and ethereal HCl 2 M (2.0 mL) gave 5c (702 mg, 96%) as
a white solid. IR (ATR) 3349, 3089, 2731, 1614,1492, 1269, 1151, 1110,
1023, 842, 773, 709 cmꢀ1. 1H NMR (400 MHz, DMSO‑d6)
d
3.86 (s,
3H), 7.02 (dd, J ¼ 8.5, 2.0 Hz, 1H), 7.23 (s, 1H), 7.74 (d, J ¼ 8.5 Hz, 1H),
7.77 (s, 2H), 8.06 (s, 1H). 13C NMR (100.6 MHz)
55.9, 95.8, 110.2,
d
4.1.4.1. N-(2,2-dimethoxyethyl)benzofuran-2-carboximidamide (4a).
Following the general procedure, 3a (5.20 mmol, 1.10 g), 2,2-
dimethoxyethylamine (5.72 mmol, 0.62 ml) and methanol (11 mL)
gave 4a (1.29 g, quantitative) as a beige solid.
113.8, 120.2, 120.4 (2C), 123.2, 135.2, 139.6, 156.0, 159.8. HRMS
C
(tR ¼ 3.20 min).
12H11N2O2 [MþH]þ 215.0815; found, 215.0814. Purity 99.6%
4.1.5.4. 2-(5-Bromobenzofuran-2-yl)-1H-imidazole
hydrochloride
4.1.4.2. N-(2,2-Dimethoxyethyl)-5-methoxybenzofuran-2-
carboximidamide (4b). Following the general procedure, 3b
(5d). Following the general procedure, 4d (1.99 mmol, 650 mg),
HCl 2 M (20 mL) and ethereal HCl 2 M (1.3 mL) gave 5d (536 mg,
82%) as a white solid. IR (ATR) 3379, 3171, 2703, 1650, 1566, 1447,
(4.34 mmol, 1.05 g),
2,2-dimethoxyethylamine
(4.78 mmol,
0.52 mL) and methanol (9 mL) gave 4b (1.29 g, quantitative) as a
1143, 1047, 879, 749, 718 cmꢀ1. 1H NMR (400 MHz, DMSO‑d6)
d 7.62
beige solid.
(dd, J ¼ 9.0, 2.0 Hz, 1H), 7.70 (d, J ¼ 9.0 Hz, 1H), 7.83 (s, 2H), 8.07 (s,
1H), 8.12 (d, J ¼ 2.0 Hz, 1H). 13C NMR (100.6 MHz)
d 108.9, 113.7,
4.1.4.3. N-(2,2-Dimethoxyethyl)-6-methoxybenzofuran-2-
carboximidamide (4c). Following the general procedure, 3c
(2.93 mmol, 708 mg), 2,2-dimethoxyethylamine (3.22 mmol,
0.35 mL) and methanol (6 mL) gave 4c (815 mg, quantitative) as a
beige solid.
116.6, 121.1 (2C), 125.3, 129.4, 129.8, 134.6, 142.0, 153.3. HRMS
C
11H8BrN2O [MþH]þ 262.9815; found, 262.9813. Purity 98.6%
(tR ¼ 3.47 min).
4.1.6. General procedure for the synthesis of 1-alkyl-2-(benzofuran-
2-yl)-1H-imidazole hydrochlorides 6b and 6c
4.1.4.4. N-(2,2-Dimethoxyethyl)-5-bromobenzofuran-2-
carboximidamide (4d). Following the general procedure, 3d
(2.01 mmol, 584 mg), 2,2-dimethoxyethylamine (2.21 mmol,
0.24 mL) and methanol (4 mL) gave 4d (657 mg, quantitative) as a
beige solid.
2-(Benzofuran-2-yl)-1H-imidazole hydrochlorides derivatives
were neutralized with NaOH 2 N and the free base was extracted
with AcOEt. The combined organic phases were washed with brine,
dried over Na2SO4, and evaporated to give the desired amine. To a
solution of the free base (1 equiv) generated from the corre-
sponding 2-(benzofuran-2-yl)-1H-imidazole hydrochloride was
treated with HBr 47% acid solution (0.5 mL) and the mixture was
stirred at 100 ꢁC for 7 h. After cooling the resulting solid was
filtered, dissolved in water and basified with saturated NaHCO3
solution. The free base was extracted with AcOEt. The combined
organic phases were washed with brine, dried over Na2SO4, and
evaporated to give a residue which was dissolved in diethyl ether/
ethanol. Ethereal HCl 2 M (1.5 mmol/mL) was added and the
precipitated salt was collected by filtration and was crystallized
with acetonitrile.
4.1.5. General procedure for the synthesis of 2-(benzofuran-2-yl)-
1H-imidazole hydrochlorides LSL60101, 5b, 5c and 5d
The
corresponding
N-(2,2-dimethoxyethyl)benzofuran-2-
carboximidamide was treated with HCl 2 M (0.1 mmol/mL) and
the resulting mixture was stirred at 60 ꢁC for 16 h. After cooling, the
solution was washed with dichloromethane. The aqueous layer was
basified with NaOH 5 M and the free base extracted with AcOEt. The
combined organic phases were washed with brine, dried over
Na2SO4, and evaporated to give a residue which was dissolved in
diethyl ether/ethanol (5:1). Ethereal HCl 2 M (1.5 mmmol/mL) was
added and the precipitated salt was collected by filtration and was
crystallized with acetonitrile.
4.1.6.1. 2-(5-Hydroxybenzofuran-2-yl)-1H-imidazole hydrochloride
(6b). Following the general procedure, 5b (0.93 mmol, 200 mg),
HBr 47% acid solution (2 mL) and ethereal HCl 2 M (0.6 mL) gave 6b
(191 mg, 81%) as a yellowish solid. IR (ATR) 3266, 3009, 2819, 2731,
4.1.5.1. 2-(Benzofuran-2-yl)-1H-imidazole
hydrochloride
(5a).
Following the general procedure, 4a (5.20 mmol, 1.29 g), HCl 2 M
(52 mL) and ethereal HCl 2 M (3.5 mL) gave LSL60101 (960 mg, 84%)
as a white solid. IR (ATR) 3477, 3168, 2535, 1651, 1457, 1310, 1140,
1593, 1443, 1370, 1246, 1195, 1157, 1090, 850, 802, 700 cmꢀ1
.
1H
6.99 (dd, J ¼ 9.0, 2.0 Hz, 1H), 7.14 (d,
J ¼ 2.0 Hz, 1H), 7.51 (d, J ¼ 9.0 Hz, 1H), 7.80 (s, 2H), 7.97 (d, J ¼ 1.0 Hz,
NMR (400 MHz, DMSO‑d6)
d
1009, 879, 737, 706 cmꢀ1
.
1H NMR (400 MHz, DMSO‑d6)
d
7.40 (t,
1H). 13C NMR (100.6 MHz)
127.9, 135.2, 140.7, 148.9, 154.5. HRMS
d 106.4, 109.9, 111.9, 116.7, 120.6 (2C),
J ¼ 7.5 Hz, 1H), 7.52 (t, J ¼ 8.5 Hz, 1H), 7.74 (d, J ¼ 8.5 Hz, 1H), 7.83 (s,
C
11H9N2O2 [MþH]þ
11

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
201.0659; found, 201.0656. Purity 98.1% (tR ¼ 2.92 min).
2532,1610, 1490, 1420,1255, 1209, 1021, 928, 810, 758, 709 cmꢀ1. 1H
NMR (400 MHz, DMSO‑d6)
d
1.49 (t, J ¼ 7.0 Hz, 3H), 3.82 (s, 3H), 4.54
4.1.6.2. 2-(6-Hydroxybenzofuran-2-yl)-1H-imidazole hydrochloride
(6c). Following the general procedure, 5c (0.93 mmol, 200 mg),
HBr 47% acid solution (1.9 mL) and ethereal HCl 2 M (0.6 mL) gave
6c (183 mg, 77%) as a yellowish solid. IR (ATR) 3388, 3092, 2813,
(q, J ¼ 7.0 Hz, 2H), 7.10 (dd, J ¼ 9.0, 2.5 Hz, 1H), 7.35 (d, J ¼ 2.5 Hz,
1H), 7.66 (d, J ¼ 9.0 Hz, 1H), 7.85 (d, J ¼ 2.0 Hz, 1H), 7.98 (d,
J ¼ 2.0 Hz, 1H), 8.06 (s, 1H). 13C NMR (100.6 MHz)
d 15.2, 44.3, 55.7,
104.1, 111.7, 112.5, 116.8, 120.7, 123.8, 127.5, 134.1, 140.3, 149.6, 156.5.
HRMS C14H15N2O2 [MþH]þ 243.1128; found, 243.1128. Purity 98.4%
(tR ¼ 3.44 min).
2755, 1624, 1430, 1377, 1283, 1150, 1122, 1096, 839, 766, 708 cmꢀ1
.
1H NMR (400 MHz, DMSO‑d6)
1H), 7.63 (d, J ¼ 8.5 Hz, 1H), 7.77 (s, 2H), 8.01 (s, 1H), 10.30 (br s, 1H).
13C NMR (100.6 MHz)
97.5, 110.6, 114.4, 118.9, 120.2 (2C), 123.2,
d
6.92 (dd, J ¼ 8.5, 2.0 Hz, 1H), 7.10 (s,
d
4.1.7.4. 1-Ethyl-2-(6-methoxybenzofuran-2-yl)-1H-imidazole hydro-
chloride (7c). Following the general procedure, 5c (0.93 mmol,
200 mg), ethyl bromide (0.98 mmol, 0.07 mL), sodium hydride 60%
in mineral oil (1.40 mmol, 56 mg), DMF (2.7 mL) and ethereal HCl
2 M (0.6 mL) gave 7c (212 mg, 87%) as a white solid. IR (ATR) 3474,
3151, 2728, 1614, 1495, 1417, 1276, 1150, 1108, 1021, 928, 853, 813,
135.4, 138.7, 156.2, 158.4. HRMS C11H9N2O2 [MþH]þ 201.0659;
found, 201.0657. Purity 97.6% (tR ¼ 2.91 min).
4.1.7. General procedure for the synthesis of 1-alkyl-2-(benzofuran-
2-yl)-1H-imidazole hydrochlorides 7a, 7aa, 7b and 7c
2-(Benzofuran-2-yl)-1H-imidazole hydrochlorides derivatives
were neutralized with NaOH 2 N and the free base was extracted
with AcOEt. The combined organic phases were washed with brine,
dried over Na2SO4, and evaporated to give the desired amine. To a
solution of the free base (1 equiv) generated from the corre-
sponding 2-(benzofuran-2-yl)-1H-imidazole hydrochloride in DMF
(0.34 mmol/mL) at 0 ꢁC was added sodium hydride (1.5 equiv, 60%
in mineral oil). After 30 min at rt, methyl iodide/ethyl bromide (1.1
equiv) was added dropwise over 15 min at 0 ꢁC. Then, the mixture
was stirred for 30 min at rt, poured into water and extracted with
AcOEt. The organic layer was washed with and the product was
extracted with HCl 1 M. The aqueous layer was basified with NaOH
2 M and the free base was extracted with AcOEt. The combined
organic phases were washed with brine, dried over Na2SO4, and
evaporated to give a residue which was dissolved in diethyl ether/
ethanol. Ethereal HCl 2 M (1.5 mmol/mL) was added and the
precipitated salt was collected by filtration and was crystallized
with acetonitrile.
778, 701 cmꢀ1
.
1H NMR (400 MHz, DMSO‑d6)
d
1.49 (t, J ¼ 7.0 Hz,
3H), 3.86 (s, 3H), 4.52 (q, J ¼ 7.0 Hz, 2H), 7.03 (dd, J ¼ 8.5, 2.0 Hz,1H),
7.33 (s, 1H), 7.73 (d, J ¼ 8.5 Hz, 1H), 7.82 (d, J ¼ 2.0 Hz, 1H), 7.95 (d,
J ¼ 2.0 Hz, 1H), 8.05 (s, 1H). 13C NMR (100.6 MHz)
d 15.2, 44.2, 55.9,
95.8, 112.0, 114.2, 119.9, 120.3, 123.1, 123.6, 134.3, 138.6, 156.1, 159.9.
HRMS C14H15N2O2 [MþH]þ 243.1128; found, 243.1128. Purity 99.2%
(tR ¼ 3.37 min).
4.2. X-ray crystallographic analysis
Crystals of LSL60101 were obtained from slow evaporation of
methanol solutions. The single crystal X-Ray diffraction data set
was collected at 294 K up to a max 2q
of ca. 57ꢁ on a Bruker Smart
APEX II diffractometer, using monochromatic MoK
l
a radiation
¼ 0.71073 Å and 0.3ꢁ separation between frames. Data integration
was performed using SAINT V6.45A and SORTAV (Blessing, 1995) in
the diffractometer package. The crystal and collection data and
structural refinement parameters are given in Table S14. The
structure was solved by direct methods using SHELXT-2014 (Shel-
drick, 2014) and Fourier's difference methods, and refined by least
squares on F2 using SHELXL-2014/7 (Sheldrick, 2014) inside the
WinGX program environment (Farrugia, 2012). Atom coordinates
are given in Table S15 and bond distances and angles in Table S16.
The crystal structure shows the chlorohydrate form of LSL60101,
which is as almost planar, as well as a methanol solvent molecule.
Anisotropic displacement parameters were used for non-H atoms
(Table S17) and the H-atoms were positioned in calculated posi-
tions (except H2) and refined riding on their parent atoms. Fig. 8
exhibits an ORTEP view of the molecule with the atom labelling,
as well as its closest intermolecular bonds (Table S18) with the
chloride anion (N12eH12/Cl1) and the methanol molecule
(N9eH9/O2). These two intermolecular bonds are slightly out of
the plane of the LSL60101 molecule, producing a small torsion of
the imidazolium group respect to the molecular mean plane
4.1.7.1. 1-Methyl-2-(benzofuran-2-yl)-1H-imidazole hydrochloride
(7a). Following the general procedure, LSL60101 (0.76 mmol,
140 mg), methyl iodide (0.84 mmol, 0.05 mL), sodium hydride 60%
in mineral oil (1.14 mmol, 45.8 mg), DMF (2.3 mL) and ethereal HCl
2 M (0.5 mL) gave 7a (150 mg, 84%) as a white solid. IR (ATR) 3306,
3095, 2569, 1634, 1445, 1270, 1188, 1131, 1033, 883, 756, 708 cmꢀ1
.
1H NMR (400 MHz, DMSO‑d6)
7.55 (t, J ¼ 8.5 Hz, 1H), 7.77 (d, J ¼ 8.5 Hz, 1H), 7.80 (d, J ¼ 2.0 Hz, 1H),
7.86e7.91 (m, 2H), 8.01 (s, 1H). 13C NMR (100.6 MHz)
36.8, 111.5,
d
4.13 (s, 3H), 7.43 (t, J ¼ 7.5 Hz, 1H),
d
112.2, 121.0, 123.2, 124.9, 126.0, 127.4, 127.9, 135.6, 140.5, 154.9.
HRMS C12H11N2O [MþH]þ 199.0866; found, 199.0866. Purity 99.7%
(tR ¼ 3.15 min).
4.1.7.2. 1-Ethyl-2-(benzofuran-2-yl)-1H-imidazole
hydrochloride
(7aa). Following the general procedure, LSL60101 (1.36 mmol,
250 mg), ethyl bromide (1.42 mmol, 0.10 mL), sodium hydride 60%
in mineral oil (2.04 mmol, 82 mg), DMF (4.0 mL) and ethereal HCl
2 M (0.9 mL) gave 7aa (310 mg, 92%) as a white solid. IR (ATR) 3394,
3104, 2618, 1628, 1428, 1266, 1177, 1111, 929, 836, 761, 708 cmꢀ1. 1H
(t(O1eC2eC8eN9) ¼ ꢀ4.3(11),
t(C3eC2eC8eN12) ¼ ꢀ1.8(14)º).
Methanol is also bonded to the chloride anion through the contact
O2eH2$$$Cl1i (i ¼ x-1,y,z), giving rise to corrugated chains along a,
assembled by parallel stacking along c. Crystallographic data
(excluding structure factors) for the reported structure has been
deposited in the Cambridge Crystallographic Data Centre as sup-
plementary publication, CCDC No. 2063533. Copies of this infor-
mation may be obtained free of charge from The Director, CCDC, 12
Union Road, Cambridge CB2 1EZ, UK. Fax: þ44 1223 336 033. E-
mail: data_request@ccdc.cam.ac.uk. Web page: http://www.ccdc.
cam.ac.uk.
NMR (400 MHz, DMSO‑d6)
d
1.50 (t, J ¼ 7.0 Hz, 3H), 4.55 (q,
J ¼ 7.0 Hz, 2H), 7.41 (d, J ¼ 7.5 Hz, 1H), 7.53 (t, J ¼ 8.0 Hz, 1H), 7.76 (d,
J ¼ 8.5 Hz, 1H), 7.86e7.88 (m, 2H), 8.00 (d, J ¼ 1.5 Hz, 1H), 8.12 (s,
1H). 13C NMR (100.6 MHz)
d 15.2, 44.3, 111.6, 111.8, 120.8, 122.8,
123.9, 124.5, 126.8, 127.5, 134.1, 139.7, 154.6. HRMS C13H13N2O
[MþH]þ 213.1022; found, 213.1021. Purity 100% (tR ¼ 3.35 min).
4.1.7.3. 1-Ethyl-2-(5-methoxybenzofuran-2-yl)-1H-imidazole hydro-
chloride (7b). Following the general procedure, 5b (0.93 mmol,
200 mg), ethyl bromide (0.98 mmol, 0.07 mL), sodium hydride 60%
in mineral oil (1.40 mmol, 56 mg), DMF (3 mL) and ethereal HCl 2 M
(0.6 mL) gave 7b (220 mg, 90%) as a white solid. IR (ATR) 3410, 3143,
4.3. Binding studies
4.3.1. Preparation of cellular membranes
Male Swiss mice (final age 8e10 weeks) were killed, and the
12

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
brain cortex dissected and stored at ꢀ70 ꢁC until assays were per-
formed. All animal experimental protocols were performed in
agreement with European Union regulations (O.J. of E.C. L 358/118/
12/1986).
Human brain samples were obtained at autopsy in the Basque
Institute of Legal Medicine, Bilbao, Spain. Samples from the pre-
frontal cortex (Brodmann's area 9) were dissected at the time of
autopsy and immediately stored at ꢀ70 ꢁC until assay. The study
was developed in compliance with policies of research and ethical
review boards for postmortem brain studies.
To obtain cellular membranes (P2 fraction) the different samples
were homogenized using an ultraturrax in 10 vol of homogeniza-
tion buffer (0.25 M sucrose, 5 mM TriseHCl, pH 7.4). The crude
homogenate was centrifuged for 5 min at 1000 g (4 ꢁC) and the
supernatant was centrifuged again for 10 min at 40,000 g (4 ꢁC).
The resultant pellet was washed twice in 5 vol of homogenization
buffer and recentrifuged in similar conditions. Protein content was
measured according to the method of Bradford using BSA as
standard.
bound ligand by rapid filtration under vacuum (1450 Filter Mate
Harvester, PerkinElmer) through GF/C glass fiber filters. The filters
were then rinsed three times with 300 mL of binding buffer, air-
dried (120 min), and counted for radioactivity by liquid scintilla-
tion spectrometry using a MicroBeta TriLux counter (PerkinElmer).
Specific binding was determined and plotted as a function of the
compound concentration. Nonspecific binding was determined in
the presence of idazoxan (10ꢀ5 M), a compound with well stab-
lished affinity for I2-IR and a2-adrenergic receptors, in [3H]2-BFI
and [3H]RX821002 assays. To obtain the inhibition constant (Ki)
analyses of competition experiments were performed by nonlinear
regression using the GraphPad Prism program. Ki values were
normalized to pKi values. I2-IR/a2 selectivity index was calculated as
the antilogarithm of the difference between pKi values for I2-IR and
pKi values for a2-AR.
4.4. Acute toxicity
Lethal dose (LD50) is a statistical derived amount of a compound
that can be expected to cause death in 50% of the animals, rodents
in general and is used as a measure of its acute toxicity. In this
particular study, LD50 was calculated using the Logarithmic-Probit
method, as initially described in Miller and Tainter (1944) [58],
and the calculations further elaborated by Randhawa (2009) [59].
Male albino mice were administered i.p. with LSL60101 at a range
of doses (10e100 mg/kg; n ¼ 10 mice per dose tested), percentage
mortalities were transformed to probits for each dose, and log-
doses vs. probits were plotted to calculate LD50 (see further de-
tails in Ref. [59]).
4.3.2. Competition binding assays
The pharmacological activity of the compounds was evaluated
through competition binding studies against the I2-IR selective
radioligand [3H]2-BFI or the a2-adrenergic receptor selective radi-
oligand [3H]RX821002. Specific binding was measured in 0.25 mL
aliquots (50 mM Tris-HCl, pH 7.5) containing 100 mg of membranes,
which were incubated in 96-well plates either with [3H]2-BFI
(2 nM) for 45 min at 25 ꢁC or [3H]RX821002 (1 nM) for 30 min at
25 ꢁC, in the absence or presence of the competing compounds
(10ꢀ12 to 10ꢀ3 M, 10 concentrations).
4.5. 3D-QSAR study. Data set preparation
Specific binding of [3H]clonidine (20 nM) to rat or human hy-
pothalamic membranes pre-incubated with benextramine
The original data set was divided on training set, that was used
for model building and test set, used for model evaluation. The I2-IR
3D-QSAR model contains 24 compounds (Fig. S1), with 16 com-
pounds in the training set and 8 compounds in the test set, while
data set for the a2-AR 3D-QSAR model contains 22 compounds,
with 14 compounds in the training set and 8 compounds in the test
set (Tables S1 and S2). In order to compare and validate our results,
we added three I2-IR standard ligands (tracizoline, idazoxan, and
BU99008, Fig. 1), in both data sets. Test set compounds were chosen
based on PCA (Principal Component Analysis) plot, considering that
pKi values were homogeneously distributed in the whole range.
After dividing all data sets into training and test sets, variable
selection was performed using fractional factorial design (FFD), and
Partial Least Square (PLS) regression was applied for building 3D-
QSAR models (Figs. S2 and S3).
(100 mM) to alkylate the population of a2-adrenoceptors. Under
these experimental conditions [3H]clonidine only labelled I1-sites.
In drug competition experiments, moxonidine (the reference
compound for I1-IR) showed sub-nanomolar affinity for these I1-
sites (KiH ¼ 0.2 nM; KiL ¼ 32 mM).
Incubations were terminated by separating free ligand from
Dominant forms of ligands at physiological pH 7.4 were ob-
tained with the Marvin Sketch 5.5.1.0 program [60], while geom-
etry was optimized with semiempirical/PM3 (Parameterized Model
revision 3) method [61,62] followed by ab initio Hartree-Fock/3-
21G method [63] using Gaussian 09 software [64] included in
ChemBio3D Ultra 13 program [65].
The Pentacle program [38] was used for calculation of GRID
independent descriptors (GRIND and GRIND2) and 3D-QSAR model
building. Computation of descriptors is based on Molecular Inter-
action Fields (MIF), by using four different probes: O probe
(hydrogen bond acceptor groups), N1 probe (hydrogen bond donor
groups), TIP probe (the shape of molecule), and DRY probe (hy-
drophobic interactions). After MIFs calculation, ALMOND algorithm
was used for the extraction of the most relevant regions, which
represented the positions of favourable interactions between the
ligand and probe. In the final step, Consistently Large Auto and
Cross Correlation (CLACC) algorithm was used to plot node-node
energies, between the same or a different probe, into auto- and
Fig. 8. X-ray crystal structure of LSL60101.
13

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
cross-correlograms, with the smoothing window set to 0.8 Å [66].
4.8. In vivo studies in mice
4.6. Cytotoxicity assays
Studies and procedures involving mouse brain dissection and
extractions were approved by the respective Local Bioethical
Committees (Universitat de les Illes Balears-CAIB and University of
Barcelona-GenCat) and followed the ARRIVE [67] and standard
ethical guidelines (European Communities Council Directive 2010/
63/EU and Guidelines for the Care and Use of Mammals in Neuro-
science and Behavioural Research, National Research Council 2003).
All efforts were made to minimize the number of animals used and
their suffering.
Cancer cell lines LN-229, Capan-1, HCT-116, NCIeH460, HL-60,
K-562 and Z-138 were acquired from the American Type Culture
Collection (ATCC, Manassas, VA, USA) and the DND-41 cell line was
purchased from the Deutsche Sammlung von Mikroorganismen
und Zellkulturen (DSMZ Leibniz-Institut, Germany). All cell lines
were cultured as recommended by the suppliers. Adherent cell
lines LN-229, Capan-1, HCT-116 and NCIeH460 were seeded at a
density between 500 and 1500 cells per well, in 384-well tissue
culture plates (Greiner). After overnight incubation, cells were
treated with different concentrations of the test compounds. Sus-
pension cell lines HL-60, K-562, Z-138 and DND-41 were seeded at
densities ranging from 2500 to 5500 cells per well in 384-well
culture plates containing the test compounds at the same con-
centration points. The plates were incubated and monitored at
37 ꢁC for 72 h in an IncuCyte® (Essen BioScience Inc., Sartorius) for
real-time imaging of cell proliferation. Brightfield images were
taken every 3 h, with one field imaged per well under 10x magni-
fication. Cell growth was then quantified based on the percent
cellular confluence as analysed by the IncuCyte® image analysis
software, and used to calculate CC50 values by linear interpolation.
4.9. Hypothermia in naïve mice
This study was performed in male and female adult CD-1 mice
bred and housed in the animal facility at the University of the
Balearic Islands in standard cages in set environmental conditions
(22 ꢁC, 70% humidity, and 12 h light/dark cycle, lights on at 8:00
a.m.) with free access to a standard diet and tap water. Prior to any
experimental procedures, mice were habituated to the experi-
menter by being handled and weighed for 2 days. Mice were
treated with a single dose of LSL60101 (1, 5, 10, 20, 30 or 50 mg/kg,
i.p., n ¼ 5 for treatment group) or vehicle (2 ml/kg of saline, i.p.,
n ¼ 5), and changes in rectal temperature were measured by
comparing basal values (before drug treatment) with that obtained
1, 2 and 3 h post-treatment by a rectal probe connected to a digital
thermometer (Compact LCD display thermometer, SA880-1 M, RS,
Corby, UK).
4.7. Pharmacokinetic analysis and analysis conditions
The pharmacokinetic study was carried on in male CD1 mice
(Envigo Laboratories) with a body weight between 40 and 50 g
(n ¼ 3e4 per group). Animals were randomized to be included in
the treated or control groups. A single intraperitoneal dose of
LSL60101 (10 mg/kg, 10 ml/kg) was administered early in the
morning (between 8 and 11 a.m.) without anaesthesia. Compound
4.10. Western blot analysis for FADD protein in naïve mice
For evaluating FADD protein regulation, a subgroup of mice
received a single dose of LSL60101 (30 or 50 mg/kg, i.p., n ¼ 7 for
treatment group) or vehicle (2 ml/kg of saline, i.p., n ¼ 6) and were
sacrificed 1 h after treatment. The hippocampus was freshly
dissected and kept at ꢀ80 ꢁC until hippocampal sample proteins
was dissolved in 10% of 2-hydroxypropyl)-b-cyclodextrin in phys-
iological saline). Mice were monitored for signs of pain or distress
during the time between injection and euthanasia. Mice were
sacrificed by cervical dislocation and blood (0.6 mL) was collected
at different time points (15 min, 30 min, 45 min, 60 min, 2 h, 3 h,
4 h, 5 h, 8 h and 24 h after injection) in tubes with serum gel and
clotting activator (Sarstedt Micro tube 1.1 mL Z-Gel). Samples were
centrifugated at 10.000 rpm for 10 min to obtain plasma and stored
at ꢀ80 ꢁC up to analysis of compound concentration by HPLC-UV.
Experimental procedures were in line with the Directive 2010/63/
EU and approved by the Institutional Animal Care and Generalitat
de Catalunya (#10291, 1/28/2018).
LSL60101 plasma concentrations versus time curves for the
mean of animals were analysed by a non-compartmental model
based on statistical moment theory using the “PK Solutions”
computer program. The pharmacokinetic parameters calculated
were the area under the concentration vs time curve (AUC),
calculated using the trapezoidal rule in the interval 0e24 h; the
(40 mg) were separated by SDS-PAGE on 10% polyacrylamide min-
igels (Bio-Rad) and transferred onto nitrocellulose membranes by
standard Western blot (Wb) procedures as described previously
[21]. Membranes were incubated overnight with anti-FADD (H-
181; sc-5559; Santa Cruz Biotechnology, Santa Cruz, CA), and
following secondary antibody (anti-rabbit) incubation and ECL
detection system (Amersham, Buckinghamshire, UK), proteins
were visualized on autoradiographic films (Amersham ECL Hyper-
film). The amount of FADD protein in mice brain samples from
different treatment groups was compared with that of vehicle-
treated controls (100%) in the same gel by densitometric scanning
(GS-800 Imaging Densitometer, Bio-Rad) of immunoreactive bands
(integrated optical density, IOD). Each brain sample was quantified
in 2e4 gels, and the mean value was used as a final estimate.
Quantification of
Sigma) was used as a house-keeping control (no differences be-
tween treatment groups, 15 g per sample).
b-actin contents (clone AC-15; Ab, no. A1978;
half-life (t1/2b), determined as ln2/b, being b, calculated from the
slope of the linear, least-squares regression line; the Cmax and Tmax
were read directly from the mean concentration curves.
m
The HPLC system was a PerkinElmer LC (Perkin Elmer INC,
Massachussets, U.S.) consisting of a Flexar LC pump, a chromatog-
raphy interface (NCI 900 network), a Flexar LC autosampler PE, and
4.11. 5XFAD as an animal model of AD: pharmacological treatments
The 5XFAD is a double transgenic APP/PS1 that co-expresses five
mutations of AD and presents robust oxidative levels [68]. 6-
month-old female 5XFAD mice (n ¼ 20) and wild type (Wt) mice
(n ¼ 20) were used to carry out the molecular analyses. The animals
were randomly allocated into four experimental groups: Control
a Waters 2487 dual
l absorbance detector. The chromatographic
column was a kromasil 100-5-C18 (4.0 ꢂ 200 mm-Teknokroma
Analítica S.A. Sant Cugat, Spain). Flux was 0.8 ml/mn and the mo-
bile phase consisted in 0.05 M KH2PO4 (30%):acetonitrile (70%) in
isocratic conditions. The elution time of LSL60101 was 4.4 min, and
5XFAD and Wt, administered with vehicle (2-hydroxypropyl)-b-
it was detected at 290 nm. The assay had a range of 0,025-5
The calibration curves were constructed by plotting the peak area
ratio of the analysed peak against the known concentrations.
m
g/mL.
cyclodextrin 1.8%), and treated 5XFAD and Wt administered with
LSL60101 diluted in vehicle (1 mg/kg/day). Treatment length was 4
weeks. Water consumption was controlled each week and the I2-IR
14

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
ligand concentration was adjusted accordingly to reach the precise
dose. Animals had free access to food and water, and were kept
under standard temperature conditions (22 2 ꢁC) and 12 h: 12 h
light-dark cycles (300 lux/0 lux).
normalize differences in sample loading and preparation. The
primers sequences used in this study are presented in
Supplementary Table S10b. Normalization of expression levels was
performed with b
-actin for SYBR® Green-based real-time PCR. Each
sample was analysed in duplicate, and the results represent the n-
fold difference of the transcript levels among different groups.
4.12. Western blot analysis in 5xFAD and Wt mice
For subcellular fractionation, 150 mL of buffer A (10 mM HEPES
4.14. Statistical analysis for 5xFAD and Wt mice comparisons
pH 7.9, 10 mM KCl, 0.1 mM EDTA pH 8, 0.1 mM EGTA pH 8, 1 mM
DTT, 1 mM PMSF, protease inhibitors) were added to each sample
and incubated on ice for 15 min. After this time, the samples were
The statistical analysis was conducted using the statistical
software GraphPad Prism version 8. Data were expressed as the
mean Standard Error of the Mean (SEM). Means were compared
with two-way ANOVAs, followed by Tukey-Kramer multiple com-
parisons post-hoc analysis. Statistical significance was considered
when p-values were <0.05. The statistical outliers were carried out
with Grubss’ test and subsequently removed from the analysis.
homogenized with a tissue homogenizer, 12.5 mL Igepal 1% were
added, and the each eppendorf was vortexed for 15 s. Following
30 s of full-speed centrifugation at 4 ꢁC, supernatants were
collected (cytoplasmic fraction); 80 mL of buffer C (20 mM HEPES
pH 7.9, 0,4 M NaCl, 1 mM EDTA pH 8, 0.1 mM EGTA pH 8, 20%
Glycerol 1 mM DTT, 1 mM PMSF, protease inhibitors) was added to
each pellet and incubated under agitation at 4 ꢁC for 15 min. Sub-
sequently, samples were centrifuged for 10 min at full speed at 4 ꢁC.
Supernatants were collected (nuclear fraction).
Declaration of competing interest
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
For Western blot analysis, aliquots of 20 mg of total hippocampal
protein were used. Protein samples from mice (n ¼ 3e5 per group)
were separated by Sodium dodecyl sulphate-polyacrylamide gel
electrophoresis (SDS-PAGE) (8e12%) and transferred onto (Poly-
vinylidene difluoride) PVDF membranes (Millipore). Afterward,
membranes were blocked in 5% non-fat milk in 0.1% Tris-buffered
saline - Tween20 (TBS-T) for 1 h at room temperature, followed
by overnight incubation at 4 ꢁC with the primary antibodies listed
in Table S12.
Membranes were washed and incubated with the correspond-
ing secondary antibodies for 1 h at room temperature. Immuno-
reactive proteins were viewed with a chemiluminescence-based
detection kit, following the manufacturer's protocol (ECL Kit; Mil-
lipore) and digital images were acquired using a ChemiDoc
XRS þ System (Bio-Rad). Semi-quantitative analyses were carried
out using ImageLab software (Bio-Rad), and results were expressed
in Arbitrary Units (AU), considering control protein levels as 100%.
Protein loading was routinely monitored by immunodetection of
Glyceraldehyde-3-phosphate dehydrogenase (GADPH) (Table S12).
Acknowledgment
ꢀ
This work was supported by Ministerio de Ciencia, Innovacion y
ꢀ
Universidades, Agencia Estatal de Investigacion (Spain, PID2019-
107991RB-I00, PID2019-106285RB), the Basque Government
(IT1211/19) and Generalitat de Catalunya (GC) (2017SGR106). The
project leading to these results has received funding from “la Caixa”
Foundation (ID 100010434) under agreement CI18-00002. This
activity has received funding from the European Institute of Inno-
vation and Technology (EIT). This body of the European Union re-
ceives support from the European Union's Horizon 2020 research
and innovation programme. C.G.-F, F.V., C.E., S.R.-A., A.B., and M.P.
Financial support was provided for F.V. (University of Barcelona,
APIF_2017), S.R.-A. (Generalitat de Catalunya, 2018FI_B_00227),
A.B. (Institute of Biomedicine UB_2018), J.A.G.-S. is a member
emeritus of the Institut d’Estudis Catalans (Barcelona, Catalonia).
E.H.-H. is supported by a predoctoral scholarship (FPI/2102/2018;
4.13. RNA extraction and gene expression determination in 5xFAD
and Wt mice
ꢀ
ꢀ
Consejería de Innovacion, Investigacion y Turismo del Gobierno de
las Islas Baleares y del Fondo Social Europeo). M.R., T.D., and K.N.
kindly acknowledge the project funded by the Ministry of Science
and Technological Development of the Republic of Serbia, Contract
No. 451-03-68/2020-14/200161, and HORISON 2020-COST-Action
CA18133 ERNEST: European Research Network on Signal Trans-
duction. E. M. acknowledges funding to Severo Ochoa: CEX2019-
917S. M.I.L, A.L.M. and J.B. gratefully acknowledge support from
Xunta de Galicia (ED431C 2018/21 and ED431G 2019/02) and Eu-
ropean Regional Development Fund (ERDF).
Total RNA isolation was carried out using TRIzol® reagent ac-
cording to manufacturer's instructions. The yield, purity, and
quality of RNA were determined spectrophotometrically with a
NanoDrop™ ND-1000 (Thermo Scientific) apparatus and an Agilent
2100B Bioanalyzer (Agilent Technologies). RNAs with 260/280 ra-
tios and RIN higher than 1.9 and 7.5, respectively, were selected.
Reverse Transcription-Polymerase Chain Reaction (RT-PCR) was
performed as follows: 2 mg of messenger RNA (mRNA) was reverse-
ꢀ
The authors thanks Dr. Rossend Obach and Dr. Angel Menargues
transcribed using the High Capacity cDNA Reverse Transcription Kit
(Applied Biosystems). Real-time quantitative PCR (qPCR) from 24
mice of both strains (n ¼ 4e6 per group) was used to quantify
mRNA expression of OS and inflammatory genes.
(LASA Laboratorios and later Ipsen Pharma SA, Barcelona, Spain) for
initially providing the compound LSL60101.
Appendix A. Supplementary data
SYBR® Green real-time PCR was performed on a Step One Plus
Detection System (Applied-Biosystems) employing SYBR® Green
PCR Master Mix (Applied-Biosystems). Each reaction mixture
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.ejmech.2021.113540.
contained 6.75
tration was 2 g), 0.75
100 nM), and 6.75
L of SYBR® Green PCR Master Mix (2X). Data
mL of complementary DNA (cDNA) (which concen-
m
mL of each primer (which concentration was
References
m
were analysed utilizing the comparative Cycle threshold (Ct)
method (DDCt), where the housekeeping gene level was used to
[1] P. Bousquet, J. Feldman, J. Schwartz, Central cardiovascular effects of alpha
adrenergic drugs: differences between catecholamines and imidazolines,
15

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
J. Pharmacol. Exp. Therapeut. 230 (1984) 232e236.
[2] P. Bousquet, A. Hudson, J.A. García-Sevilla, J.X. Li, Imidazoline receptor system:
the past, the present, and the future, Pharmacol. Rev. 72 (2020) 50e79.
[3] J.X. Li, Imidazoline I2 receptors: an update, Pharmacol. Ther. 178 (2017)
48e56.
[28] Z. Liu, T. Zhou, A.C. Ziegler, P. Dimitrion, L. Zuo, Oxidative stress in neurode-
generative diseases: from molecular mechanisms to clinical applications,
Oxid. Med. Cell Longev. 2017 (2017) 1e11.
[29] M. Johnstone, A.J. Geraing, K.M. Miller, A central role for astrocytes in the
inflammatory response to beta-amyloid: chemokines, cytokines and reactive
oxygen species are produced, J. Neuroinmunol. 93 (1999) 182e193.
[30] D.J. Bonda, X. Wang, G. Perry, A. Nunomura, M. Tabaton, X. Zhu, M.A. Smith,
Oxidative stress in Alzheimer disease: a possibility for prevention, Neuro-
pharmacology 59 (2010) 290e294.
ꢀ
[4] J. García-Sevilla, P.V. Escriba, M. Sastre, C. Walzer, X. Busquets, G. Jaquet,
ꢀ
D.J. Reis, J. Guimon, Immunodetection and quantitation of imidazoline re-
ceptor proteins in platelets of patients with major depression and in brains of
suicide victims, Arch. Gen. Psychiatr. 53 (1996) 803e810.
[5] J.A. Lowry, J.T. Brown, Significance of the imidazoline receptors in toxicology,
Clin. Toxicol. 52 (2014) 454e469.
[6] J.X. Li, Y. Zhang, Imidazoline I2 receptors: target for new analgesics? Eur. J.
Pharmacol. 658 (2011) 49e56.
[7] S. Regunathan, D.L. Feinstein, D.J. Reis, Anti-proliferative and anti-
inflammatory actions of imidazoline agents. Are imidazoline receptors
involved? Ann. N. Y. Acad. Sci. 881 (1999) 410e419.
[31] L.A. Lione, D.J. Nutt, A.L. Hudson, Characterisation and localisation of [3H]-2-
(2-benzofuranyl)-2-imidazoline binding in rat brain: a selective ligand for
imidazoline I2 receptors, Eur. J. Pharmacol. 353 (1998) 123e135.
~ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
[32] F. Vasilopoulou, C. Grinan-Ferre, S. Rodríguez-Arevalo, A. Bagan, S. Abas,
ꢁ
C. Escolano, M. Pallas, I2 imidazoline receptor modulation protects aged
SAMP8 mice against cognitive decline by suppressing the calcineurin
pathway, GeroScience 43 (2021) 965e983.
[8] K.L. Smith, D.S. Jessop, D.P. Finn, Modulation of stress by imidazoline binding
sites: implications for psychiatric disorders, Stress 12 (2009) 97e114.
[9] J.J. Meana, F. Barturen, I. Martín, J.A. García-Sevilla, Evidence of increased non-
adrenoreceptor [3H]idazoxan binding sites in the frontal cortex of depressed
suicide victims, Biol. Psychiatr. 34 (1993) 498e501.
[10] J. Ruíz, I. Martín, L.F. Callado, J.J. Meana, F. Barturen, J.A. García-Sevilla, Non-
adrenoreceptor [3H]idazoxan binding sites (I2-imidazoline sites) are increased
in postmortem brain from patients with Alzheimer's disease, Neurosci. Lett.
160 (1993) 109e112.
[33] J.A. García-Sevilla, J.J. Meana, F. Barturen, F.A. Geijo, A. Menargues, R. Obach,
F. Pla, Benzofuranylimidazole Derivatives and Therapeutical Compositions
Containing the Same, 1994. US5354769(A), October 11.
ꢀ
[34] R. Alemany, G. Olmos, P.V. Escriba, A. Menargues, R. Obach, J.A. García-Sevilla,
LSL60101, a selective ligand for imidazoline I2 receptors, on glial fibrillary
acidic protein concentration, Eur. J. Pharmacol. 280 (1995) 205e210.
[35] M.A. Boronat, G. Olmos, J.A. García-Sevilla, Attenuation of tolerance to opioid-
induced antinociception and protection against morphine-induced decrease
of neurofilament proteins by idazoxan and other I2-imidazoline ligands, Br. J.
Pharmacol. 125 (1998) 175e185.
[36] A. Casanova, G. Olmos, J. Ribera, M.A. Boronat, J.E. Esquerda, J.A. García-Sevilla,
Induction of reactive astrocytosis and prevention of motoneuron cell death by
the I2-imidazoline receptor ligand LSL60101, Br. J. Pharmacol. 130 (2000)
1767e1776.
[11] C. Gargalidis-Moudanos, N. Pizzinat, F. Javoy-Agid, A. Remaury, A. Parini, I2-
imidazoline binding sites and monoamine oxidase activity in human post-
mortem brain from patients with Parkinson's disease, Neurochem. Int. 30
(1997) 31e36.
ꢀ
[12] L.F. Callado, J.I. Martín-Gomez, J. Ruiz, J.M. Garibi, J.J. Meana, Imidazoline I2
receptors density increases with the malignancy of human gliomas, J. Neurol.
Neurosurg. Psychiatry 75 (2004) 785e787.
[37] N. MacInnes, S.L. Handley, Characterization of the discriminable stimulus
produced by 2-BFI: effects of imidazoline I2-site ligands, MAOIs,
b-carbolines,
[13] C. Dardonville, I. Rozas, Imidazoline binding sites and their ligands: an over-
view of the different chemical structures, Med. Res. Rev. 24 (2004) 639e661.
[14] G.M. Gilad, V.H. Gilad, Accelerated functional recovery and neuroprotection by
agmatine after spinal cord ischemia in rats, Neurosci. Lett. 296 (2000) 97e100.
[15] K. Maiese, L. Pek, S.B. Berger, D.J. Reis, Reduction in focal cerebral ischemia by
agents acting at imidazole receptors, J. Cerebr. Blood Flow Metabol. 12 (1992)
53e63.
agmatine and ibogaine, Br. J. Pharmacol. 135 (2002) 1227e1234.
[38] Pentacle, Version1.0.6, MolecularDiscoveryLtd., Perugia, Italy, 2009.
[39] A. Krüger, V. Gonçalves, C. Wrenger, T. Kronenberger, ADME Profiling in Drug
Discovery and a New Path Paved on Silica, Drug Discovery and Development -
New Advances, 2019.
[40] ADMET Predictor, V. 9.5, Simulations Plus Inc., Lancaster, CA, USA. https://
www.simulations-plus.com.
[16] Z. Han, M.J. Xiao, B. Shao, R.Y. Zheng, G.Y. Yang, K. Jin, Attenuation of ischemia
induced rat brain injury by 2-(2-benzofuranyl)-2-imidazoline, a high selec-
tivity ligand for imidazoline I(2)receptors, Neurol. Res. 31 (2009) 390e395.
[17] S.X. Jiang, R.Y. Zheng, J.Q. Zheng, X.L. Li, Z. Han, S.T. Hou, Reversible inhibition
of intracellular calcium influx through NMDA receptors by imidazoline (I)2
receptor antagonists, Eur. J. Pharmacol. 629 (2010) 12e19.
[18] L.C. Rovati, N. Brambilla, T. Blicharski, J. Connell, C. Vitalini, A. Bonazzi,
G. Giacovelli, F. Girolami, M. D'Amato, Efficacy and safety of the firts-in-classs
imidazoline-2 receptor ligand CR4056 in pain from knee osteoarthritis and
disease phenotypes: a randomized, double-blind, placebo-controlled phase 2
trial, Osteoarthritis Cartilage 28 (2020) 22e30.
[41] A. Daina, O. Michielin, V. Zoete, SwissADME: a free web tool to evaluate
pharmacokinetics, drug-likeness and medicinal chemistry friendliness of
small mol, Sci. Rep. 7 (2017) 1e13.
[42] D.A. Thorn, X. An, Y. Zhang, M. Pigini, J. Li, Characterization of the hypothermic
effects of imidazoline I2 receptor agonists in rats, Br. J. Pharmacol. 166 (2012)
1936e1945.
ꢀ
ꢀ
[43] E. Hernandez-Hernandez, M.J. García-Fuster, Evaluating the effects of 2-BFI
and tracizoline, two potent I2-imidazoline receptor agonists, on cognitive
performance and affect in middle-aged rats, Naunyn-Schmiedeberg's Arch.
Pharmacol. Pharmacology 394 (2021) 989e996.
[44] J.A. Craven, E.L. Conway, Effects of alpha 2-adrenoceptor antagonists and
imidazoline 2-receptor ligands on neuronal damage in global ischemia in the
rat, Clin. Exp. Pharmacol. Physiol. 24 (1997) 204e207.
[45] D.W. Marion, L.E. Penrod, S.F. Kelsey, W.D. Obrist, P.M. Kochanek, A.M. Palmer,
S.R. Wisniewski, S.T. DeKosky, Treatment of traumatic brain injury with
moderate hypothermia, N. Engl. J. Med. 336 (1997) 540e546.
[19] R.J. Tyacke, J.F.M. Myers, A.V. Venkataraman, I. Mick, S. Turton, J. Passchier,
S.M. Husband, E.A. Rabiner, R.N. Gunn, P.S. Murphy, C.A. Parker, D.J. Nutt,
Evaluation of 11C-BU99008, a PET ligand for the imidazoline2 binding site in
human brain, J. Nucl. Med. 59 (2018) 1597e1602.
[20] H. Wilson, G. Dervenoulas, G. Pagano, R.J. Tyacke, S. Polychronis, J. Myers,
R.N. Gunn, E.A. Rabiner, D. Nutt, M. Politis, Imidazoline
2
binding sites
[46] C.M. Maier, K.V. Ahern, M.L. Cheng, J.E. Lee, M.A. Yenari, G.K. Steinberg,
J.R. Kirsch, Optimal depth and duration of mild hypothermia in a focal model
of transient cerebral ischemia: effects on neurologic outcome, infarct size,
apoptosis, and inflammation, Stroke 29 (10) (1998) 2171e2180.
[47] E. Hernandez-Hernandez, A. Miralles, S. Esteban, M.J. García-Fuster, Improved
age-related deficits in cognitive performance and affective-like behavior
following acute, but not repeated, 8-OH-DPAT treatments in rats: regulation
of hippocampal FADD, Neurobiol. Aging 71 (2018) 115e126.
reflecting astroglia pathology in Parkinson's disease: an in vivo 11C-BU99008
PET study, Brain 10 (2019) 3116e3128.
ꢀ
ꢀ
[21] S. Abas, A.M. Erdozain, B. Keller, S. Rodríguez-Arevalo, L.F. Callado, J.A. García-
Sevilla, C. Escolano, Neuroprotective effects of a structurally new family of
high affinity imidazoline I2 receptors ligands, ACS Chem. Neurosci. 8 (2017)
737e742.
ꢀ
ꢀ
~ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
[22] C. Grinan-Ferre, F. Vasilopoulou, S. Abas, S. Rodríguez-Arevalo, A. Bagan,
ꢀ
ꢀ
ꢀ
F.X. Sureda, B. Perez, L.F. Callado, J.A. García-Sevilla, M.J. García-Fuster,
[48] E. Hernandez-Hernandez, A. Miralles, S. Esteban, M.J. García-Fuster, Repeated
treatment with the 2-adrenoceptor agonist UK-14304 improves cognitive
ꢁ
C. Escolano, M. Pallas, Behavioral and cognitive improvement induced by
a
novel imidazoline I2 receptor ligands in female SAMP8 mice, Neuro-
performance in middle-age rats: role of hippocampal Fas-associated death
domain, J. Psychopharmacol. 32 (2) (2018) 248e255.
therapeutics 16 (2019) 416e431.
[23] S. Abas, S. Rodríguez-Arevalo, A. Bagan, C. Grinan-Ferre, F. Vasilopoulou,
ꢀ
ꢀ
ꢀ
ꢀ
~ꢀ
ꢀ
[49] E. Comi, M. Lanza, F. Ferrari, V. Mauri, G. Caselli, L.C. Rovati, Efficacy of CR4056,
a first-in-class imidazoline-2 analgesic drug, in comparison with naproxen in
two rat models of osteoarthritis, J. Pain Res. 10 (2017) 1033e1043.
[50] N. Mirzaei, B.C. Mota, A.M. Birch, N. Davis, C. Romero-Molina, L. Katsouri,
E.O.C. Palmer, A. Golbano, L.J. Riggall, I. Nagy, R. Tyacke, D.J. Nutt, M. Sastre,
Imidazoline ligand BU224 reverses cognitive deficits, reduces microgliosis and
enhances synaptic connectivity in a mouse model of Alzheimer's disease, Br. J.
Pharmacol. 178 (3) (2021) 654e671.
[51] J.A. Garcia-Sevilla, P.V. Escriba, C. Walzer, C. Bouras, J. Guimon, Imidazoline
receptor proteins in brains of patients with Alzheimer's disease, Neurosci.
Lett. 247 (1998) 95e98.
[52] B. Uttara, A.V. Singh, P. Zamboni, R.T. Mahajan, Oxidative stress and neuro-
degenerative diseases: a review of upstream and downstream antioxidant
therapeutic options, Curr. Neuropharmacol. 7 (1) (2009) 65e74.
ꢀ
I. Brocos-Mosquera, C. Muguruza, B. Perez, E. Molins, F.J. Luque, P. Perez-
Lozano, S. de Jonghe, D. Daelemans, L. Naesens, J. Brea, M.I. Loza,
ꢀ
ꢀ
E. Hernandez-Hernandez, J.A. García-Sevilla, M.J. García-Fuster, M. Radan,
ꢁ
T. Djikic, K. Nikolic, M. Pallas, L.F. Callado, C. Escolano, Bicyclic
a-imino-
phosphonates as high affinity imidazoline I2 receptor ligands for alzheimer's
disease, J. Med. Chem. 7 (2020) 3610e3633.
[24] D.A. Ruggiero, S. Regunathan, H. Wang, T.A. Milner, D.J. Reis, Immunocyto-
chemical localization of an imidazoline receptor protein in the central nervous
system, Brain Res. 780 (1998) 270e293.
ꢀ
ꢀ
[25] J.A. García-Sevilla, P.V. Escriba, C. Walzer, C. Bouras, J. Guimon, Imidazoline
receptor proteins in brains of patients with Alzheimer's disease, Neurosci.
Lett. 247 (1998) 95e98.
[26] D.S. Albers, M.F. Beal, Mitochondrial dysfunction and oxidative stress in aging
and neurodegenerative diseases, J. Neural. Transm. Suppl. 5 (2000) 133e154.
[53] E. Birben, U.M. Sahiner, C. Sackesen, S. Erzurum, O. Kalayci, Oxidative stress
and antioxidant defense, World Allergy Organ. J. 5 (1) (2012) 9e19.
[54] J.S. Tian, Q.J. Zhai, Y. Zhao, R. Chen, L.D. Zhao, 2-(2-benzofuranyl)-2-
imidazoline (2-BFI) improved the impairments in AD rat models by
~
[27] J.P. Bolanos, M.A. Moro, I. Lizasoain, A. Almeida, Mitochondrial and reactive
oxygen and nitrogen species in neurological disorders and stroke: therapeutic
implications, Adv. Drug Deliv. Rev. 61 (2009) 1299e1315.
16

ꢀ
ꢀ
~ꢀ
ꢀ
S. Rodriguez-Arevalo, A. Bagan, C. Grinan-Ferre et al.
European Journal of Medicinal Chemistry 222 (2021) 113540
Applications, J. Comput. Chem. 10 (1989) 221e264.
[63] W.J. Hehre, L. Radom, PvR. Schleyer, J. Pople, AB Initio Molecular Orbital
Theory, Wiley, 1986.
[64] M.J. Frisch, Gaussian 98 (RevisionA.7), 1998.
[65] CambridgeSoft Corporation, ChemBio3D Ultra, Version 13.0, 2013. Cambridge,
MA, USA.
inhibiting oxidative stress, inflammation and apoptosis, J. Integr. Neurosci. 16
(4) (2017) 385e400.
[55] D.H. Choi, J.H. Yun, J. Lee, Protective effect of the imidazoline I2 receptor
agonist 2-BFI on oxidative cytotoxicity in astrocytes, Biochem. Biophys. Res.
Commun. 503 (4) (2018) 3011e3016.
ꢀ
ꢀ
~ꢀ
ꢀ
[56] F. Vasilopoulou, S. Rodríguez-Arevalo, A. Bagan, C. Escolano, C. Grinan-Ferre,
ꢁ
M. Pallas, Disease-modifying treatment with I2 imidazoline receptor ligand
LSL60101 in an Alzheimer's disease mouse model: a Comparative study with
donepezil, Br. J. Pharmacol. (2021) 1e17.
[66] A. Duran, M. Pastor, An advanced tool for computing and handling grid-
independent descriptors, User Manual (2011). Version 1.06.
[67] N. Percie du Sert, A. Ahluwalia, S. Alam, M.T. Avey, M. Baker, W.J. Browne,
A. Clark, I.C. Cuthill, U. Dirnagl, M. Emerson, P. Garner, S.T. Holgate,
D.W. Howells, V. Hurst, N.A. Karp, S.E. Lazic, K. Lidster, C.J. MacCallum,
M. Macleod, E.J. Pearl, O.H. Petersen, F. Rawle, P. Reynolds, K. Rooney,
E.S. Sena, S.D. Silberberg, T. Steckler, H. Würbel, Reporting animal research:
explanation and elaboration for the ARRIVE guidelines 2.0, PLoS Biol. 18 (7)
(2020), https://doi.org/10.1371/journal.pbio.3000411.
[57] Y. Chen, C. Qin, J. Huang, X. Tang, C. Liu, K. Huang, J. Xu, G. Guo, A. Tong,
L. Zhou, The role of astrocytes in oxidative stress of central nervous system: a
mixed blessing, Cell Prolif 53 (3) (2020) e12781.
[58] L.C. Miller, M.L. Tainter, Estimation of the ED50 and its error by means of
logarithmic-probit graph paper, Proc. Soc. Exp. Bio. Med. 57 (1944) 261e264.
[59] M.A. Randhawa, Calculation of LD50 values from the method of miller and
tainter, 1944, J. Ayub Med. Coll. Abbottabad 21 (3) (2009) 184e185.
[60] MarvinSketch 5.5.1.0, ChemAxon, Budapest, Hungary, 2011 software available
at, https://www.chemaxon.com.
[68] H. Oakley, S.L. Cole, S. Logan, E. Maus, P. Shao, J. Craft, A. Guillozet-Bongaarts,
M. Ohno, J. Disterhoft, L. Van Eldik, R. Berry, R. Vassar, Intraneuronal b-amy-
loid aggregates, neurodegeneration, and neuron loss in transgenic mice with
five familial Alzheimer's disease mutations: potential factors in amyloid pla-
que formation, J. Neurosci. 26 (2006) 10129e10140.
[61] J.J.P. Stewart, Optimization of parameters for semiempirical methods I.
Method, J. Comput. Chem. 10 (1989) 209e220.
[62] J.J.P. Stewart, Optimization of parameters for semiempirical methods II.
17