Novel potent vasodilating agents: Evaluation of the activity and potency of LINS01005 and derivatives in rat aorta

Article abstract of DOI:10.1016/j.ejps.2019.105171

Cardiovascular diseases (CVDs) present high prevalence rates in the current world. It is estimated that approximately one-third of the global deaths are related to CVDs, and thus there is still a need for novel drugs to treat these disorders. We serendipitously discovered that LINS01005 (5a) is a potent vasodilating agent in rat aorta, and therefore a set of analogues were evaluated for the vasodilating potency in Wistar and SHR rat thoracic aorta precontracted with norepinephrine, with endothelium intact (E+) or denuded (E–) aortic rings. Compounds 5a and 5b were the most potent, showing submicromolar potency for endothelium intact vessels (EC₅₀ 853 and 941 nM, respectively) and micromolar values for E– vessels (EC₅₀ 2.4 and 7.1 μM, respectively). These compounds were indeed significantly more potent vasodilating agents in SHR-derived aortic rings (p < 0.001), showing nanomolar potency for 5a [EC₅₀ 2.4 nM (E+) 9.0 nM (E–)] and 5b [EC₅₀ 20 nM (E+) 2.1 μM (E–)]. SAR analysis though PCA and HCA were performed, suggesting that N-phenylpiperazine is essential to the activity, while increasing volume in the substituted aromatic moiety is detrimental to the potency. This is the first report of the vasodilating properties of such compounds, and studies regarding the mechanism of action are in progress in our group.

Full text of DOI:10.1016/j.ejps.2019.105171

EuropeanJournalofPharmaceuticalSciences143(2020)105171
Contents lists available at ScienceDirect
European Journal of Pharmaceutical Sciences
journal homepage: www.elsevier.com/locate/ejps
Novel potent vasodilating agents: Evaluation of the activity and potency of
LINS01005 and derivatives in rat aorta
Milton Ginoza†, Gustavo A.B. Fernandes†, Michelle F. Corrêa, João Paulo S. Fernandes⁎
Departamento de Ciências Farmacêuticas, Universidade Federal de São Paulo, Rua São Nicolau 210, Diadema, SP 09913-030, Brazil
A R T I C L E I N F O
A B S T R A C T
Keywords:
Cardiovascular diseases (CVDs) present high prevalence rates in the current world. It is estimated that ap-
proximately one-third of the global deaths are related to CVDs, and thus there is still a need for novel drugs to
treat these disorders. We serendipitously discovered that LINS01005 (5a) is a potent vasodilating agent in rat
aorta, and therefore a set of analogues were evaluated for the vasodilating potency in Wistar and SHR rat
thoracic aorta precontracted with norepinephrine, with endothelium intact (E+) or denuded (E–) aortic rings.
Compounds 5a and 5b were the most potent, showing submicromolar potency for endothelium intact vessels
(EC50 853 and 941 nM, respectively) and micromolar values for E– vessels (EC50 2.4 and 7.1 µM, respectively).
These compounds were indeed significantly more potent vasodilating agents in SHR-derived aortic rings (p <
0.001), showing nanomolar potency for 5a [EC50 2.4 nM (E+) 9.0 nM (E–)] and 5b [EC50 20 nM (E+) 2.1 µM
(E–)]. SAR analysis though PCA and HCA were performed, suggesting that N-phenylpiperazine is essential to the
activity, while increasing volume in the substituted aromatic moiety is detrimental to the potency. This is the
first report of the vasodilating properties of such compounds, and studies regarding the mechanism of action are
in progress in our group.
Vasodilating agents
SAR analysis
Thoracic aorta
Chemometric analysis
1. Introduction
Marona et al., 2009; Srour et al., 2016).
In order to reduce the blood pressure and therefore the risk of CVDs,
The cardiovascular diseases (CVDs) are a group of hearth and blood
vessels disorders which include high prevalence conditions such as
hypertension, chronic congestive heart failure and coronary diseases.
According to World Health Organization (WHO), CVDs together are
responsible for 31% of all global deaths per year, meaning about 17.9
million people worldwide (WHO, 2019). Hypertension is the most im-
portant cardiovascular disease presenting a very complex and multi-
factorial clinical condition with high prevalence rates in Brazil and
worldwide, being a common public health issue (Malachias et al.,
2016). Hypertension is characterized by abnormal and persistent ele-
vation of blood pressure (Franklin and Wong, 2013; Whelton et al.,
2018), and therefore it is one of the major risks for coronary artery
disease and heart failure (Malachias et al., 2016; Whelton et al., 2018;
WHO, 2019). People suffering from hypertension has functional and
structural alterations on the arteries such as the increase of vascular
resistance (Laurent and Boutouyrie, 2015). Considering the high pre-
valence of hypertension and CVDs, the need for new drugs or novel
treatments has always attracting the attention of several research
groups (Bouider et al., 2015; Canbakan, 2013; Kümmerle et al., 2009;
several vasodilating agents are already available, such as angiotensin-
converting enzyme inhibitors, angiotensin receptor blockers and cal-
cium antagonists. However, these agents are frequently used in asso-
ciation with other drugs, since several patients become refractory to the
therapy (Canbakan, 2013). Moreover, none of the available agents are
free of side effects. This illustrates the need of searching novel vaso-
dilating agents (McInnes, 2007; Singh et al., 2017). These agents can be
also used to treat several other diseases beyond hypertension, such as
cerebrovascular disease, renal failure and hearth diseases (Srour et al.,
2016).
The LINS01 compounds were initially designed as ligands of hista-
mine H3 and H4 receptors due to the similarity with H3 and H4 ligands
such as ABT-239 and JNJ-7777120 (Corrêa et al., 2019, Corrêa et al.,
2017). Some compounds in the set showed nanomolar affinities for such
targets, with general selectivity towards the H3 receptors over H1, H2
and H4. During our investigations regarding the pharmacology of these
compounds in rat thoracic aorta, we serendipitously observed that the
compound LINS01005 (5a, Fig. 1) showed promising vasodilating effect
(personal data). Recently, we reported that the LINS01 compounds are
⁎ Corresponding author.
E-mail address: joao.fernandes@unifesp.br (J.P.S. Fernandes).
† These authors contributed equally.
https://doi.org/10.1016/j.ejps.2019.105171
Received 18 September 2019; Received in revised form 23 November 2019; Accepted 29 November 2019
Availableonline02December2019
0928-0987/©2019ElsevierB.V.Allrightsreserved.

M. Ginoza, et al.
EuropeanJournalofPharmaceuticalSciences143(2020)105171
Fig. 1. General synthetic procedure for the compounds. a. allyl bromide (2 eq.), K2CO3 (2 eq.), acetone, 60 °C, 8–12 h; b. 180 °C, DMF, MW 300 W, 1–2 h; c. I2 (1.1
eq.), water, 4–6 h; d. R’-piperazine K2CO3 (1 eq.), THF, MW 300 W, 1.5 h or reflux, 24–48 h.
considerably selective H3 receptor antagonists with very low affinity to
other histamine receptor subtypes, especially H1 and H2 receptors
(Corrêa et al., 2019). Accordingly, the vasodilating effect of LINS01005
is not related to these histamine receptors, and to date its molecular
mechanism of action is still unknown.
2017, Corrêa et al., 2017).
1-[(5‑methoxy‑2,3-dihydro-1-benzofuran-2-yl)methyl]−4-phenylpiper-
azine (5d). 65% yield. tR 5.194 min. 1H NMR (CDCl3, 300 MHz): δ 2.62
(dd, 1H, J = 13.3, 4.0 Hz); 2.74 (dt, 3H, J = 10.6, 5.2 Hz); 2.84 (dd,
1H, J = 13.3, 7.8 Hz); 2.97 (dd, 1H, J = 15.8, 8.0 Hz); 3.09–3.33 (m,
6H); 3.75 (s, 3H); 4.99 (dt, 1H, J = 12.8, 4.1 Hz); 6.60–6.79 (m, 3H);
6.86 (t, 1H, J = 7.3 Hz); 6.94 (d, 2H, J = 7.9 Hz); 7.20–7.33 (m, 2H).
13C NMR (CDCl3, 75 MHz): δ 34.68; 49.06; 53.88; 56.03; 63.21; 80.91;
109.48; 111.27; 112.74; 116.09; 119.73; 127.48; 129.14; 151.33;
153.62; 154.04. LRMS (EI): m/z (rel.%) 350 [M+] (20); 335 (5); 175
(100); 132 (31); 70 (56). HRMS (ESI+) calculated for C23H31N2O+
m/z 351.2430; experimental m/z 351.2431.
Considering the potential of LINS01005 as vasodilating agent, we
decided to further explore its potency as vasodilator in rat thoracic
aorta. Moreover, a set of derivatives containing different substituents in
the dihydrobenzofuran and piperazine moieties were prepared and
evaluated to assess how these substitutents can affect the vasodilating
potency of such compounds and therefore determine the initial struc-
ture-activity relationships.
1-[(5-t‑butyl‑2,3-dihydro-1-benzofuran-2-yl)methyl]−4-phenylpiper-
azine (5e). 50% yield. tR 4.681 min. 1H NMR (CDCl3, 300 MHz): δ 1.29
(s, 9H); 2.63 (dd, 1H, J = 13.3, 4.0 Hz); 2.71–2.79 (m, 4H); 2.84 (dd,
1H, J = 13.3, 7.9 Hz); 2.94 (dd, 1H, J = 15.7, 7.9 Hz); 3.21–3.34 (m,
5H); 4.99 (dq, 1H, J = 8.3, 4.0 Hz); 6.73 (d, 1H, J = 8.3 Hz); 6.85 (t,
1H, J = 7.2 Hz); 6.94 (d, 2H, J = 8.1 Hz); 7.13 (d, 1H, J = 8.4 Hz);
7.20–7.34 (m, 3H). 13C NMR (CDCl3, 75 MHz): δ 31.74; 34.27; 34.49;
49.07; 53.88; 63.36; 80.90; 108.81; 116.08; 119.68; 121.92; 124.76;
126.00; 129.12; 143.42; 151.36; 157.29. LRMS (EI): m/z (rel.%) 370
[M+] (18); 175 (100); 132 (27); 70 (56). HRMS (ESI+) calculated for
C23H31N2O+ m/z 371.2117; experimental m/z 371.2116.
2. Material and methods
The starting materials and chemicals used were purchased from
Sigma-Aldrich and LabSynth and used as provided. The NMR spectra
were recorded in a Bruker Advance 300 operating at 300 MHz for 1H
and 75 MHz for 13C analysis. Chemical shifts (δ) were measured from
tetramethylsilane (TMS) as internal reference and are reported in ppm.
Coupling constants (J) are reported in Hz, when applicable. The low
resolution mass spectra (LRMS) were obtained through electron ioni-
zation (70 eV) on a Shimadzu GCMS-QP2010 plus mass spectrometer.
The ion-radical (M+) and its fragments (in% relative abundance) are
reported in mass/charge ratio (m/z). High resolution mass spectra
(HRMS) were recorded in a MicroToF Bruker Daltonics through elec-
tron-spray ionization in positive mode (ESI+) and reported in m/z for
the monoprotonated form of the compounds. The compounds’ purity
was assessed by HPLC-DAD, using an isocratic acetonitrile:water mix-
ture as eluent and a C-18 column, and monitored in the 250 nm wa-
velength. The compounds were considered adequate for testing when
purity was over 95%. The retention times (tR) are reported in minutes
and presented along with spectral data.
2.2. Animals
Thirty adult (>12 weeks old) Wistar and eight adult spontaneously
hypertensive rats (SHR) were obtained from CEDEME (Center for
Development of Experimental Medicine and Biology, Federal University
of São Paulo, Brazil) and used to obtain the thoracic aortic rings. The
rats were housed by four animals/home cage and maintained during
acclimation and experimental time with free access to food and water
and under a 12:12 h light cycle (lights on at 6:00 a.m.), controlled
temperature (22 °C
2) and relative humidity (53% 2). Food and
water were provided ad libitum. All procedures were conducted in
agreement to the Brazilian law for the use of animals in research (Law
Number 11.794/2008) and were approved by the local ethics com-
mittee for animal usage of Federal University of São Paulo (CEUA
UNIFESP, approval no. 1012050417). The animals were sacrificed by
intraperitoneal administration using a mixture of ketamine (100 mg/
kg) and xylazine chloride (10 mg/kg) (Jodas et al., 2014) prior to the
surgeries.
2.1. Preparation of the compounds
The compounds were synthesized as previously by our group
(Fig. 1) (Corrêa et al., 2017, Corrêa et al., 2017, 2016). Compounds 5b-f
were prepared from the corresponding 4-substituted phenols, which
were allylated with allyl bromide, followed by Claisen rearrangement
under microwave conditions. The obtained orto-allyl phenols were
converted to the iodomethyl-dihydrobenzofurans 4a-e through iodine-
promoted cyclization, using an eco-friendly procedure (Corrêa et al.,
2016). The last step consisted in the reaction with R’-piperazines (1-
phenyl, 1-methyl or 1-allyl), using a recently reported method which
uses a microwave reactor, providing increased yield than the conven-
tional synthetic methodology (Corrêa et al., 2019). This method was
successfully employed to prepare the novel compounds 5d and 5e, with
comparable yields. However, the conventional methodology (reflux)
led to better yields for compounds 6a, 7a and 8a. The NMR data for the
compounds 5a, 5b, 5c, 5f, 6a, 7a and 8a are in line with the previous
literature reports (Corrêa et al., 2019c, Corrêa et al., 2019, Corrêa et al.,
2.3. Preparation of the aortic rings
The used procedure was following previous literature reports
(Rattmann et al., 2006; Vasconcelos et al., 2018a). The thoracic aorta
was dissected, put into a Petri dish, cut in four rings (2–4 mm) and
placed between two metallic hooks inside of a vertical chamber from
the isolated organ bath apparatus (model IOB, AVS Projetos Inc.) filled
with a nutritive Krebs-Henseleit solution composed by (in mmol/L)
NaCl 130; KCl 4.7; KH2PO4 1.18; MgSO4.7H2O 1.17; NaHCO3 14.9;
2

M. Ginoza, et al.
EuropeanJournalofPharmaceuticalSciences143(2020)105171
EDTA 0.026; CaCl2.2H2O 1.6; D-glucose 5.55 (pH 7.4). The rings were
kept at 37 °C and continuously oxygenated with carbogen mixture (5%
CO2 in O2). The hooks were attached in a force transducer (TIM 200)
coupled to a computational system (AQCAD version 2.3.8, AVS Projetos
Inc.). The system recorded the value of each contraction continuously.
All the rings were subjected to constant tension (~1 gf) during 60 min
and afterwards the solution was substituted every 10 min until re-sta-
bilization before each experiment. The responsiveness of the rings was
tested by the contraction with KCl solution (80 mM), followed by
washing and ring re-stabilization (60 min). The endothelium viability
was checked by the contraction with norepinephrine hydrochloride
(NE, 0.3 µM) followed by relaxation with acetylcholine chloride (1 µM).
The ring was considered viable when the contraction was reduced by
75% through endothelium dependent mechanism. The endothelium
was considered intact (E+) if at least 50% relaxation was achieved by
acetylcholine chloride addition (Rattmann et al., 2006). Acetylcholine
chloride was also used as positive control. The endothelium-denuded
rings (E–) were prepared through gentle rubbing the rings between the
fingers, and the absence of functional endothelium was checked
through the non-response after the application of 1 µM acetylcholine
chloride.
without (E–) endothelium, as well as the potencies in Wistar and SHR
thoracic aorta were compared through analysis of variance (ANOVA)
followed by Bonferroni's post-hoc test. The significance level considered
in all cases was 95% (p < 0.05).
2.6. Descriptors calculations
The molecular descriptors were calculated using Marvin Sketch
16.1.8.0 (ChemAxon Inc.) using the default configuration in the soft-
ware, which were preliminary selected through visual inspection from a
scatter plot between vasodilating activity vs. descriptor. The selected
descriptors included the acidic constant pKa of the nitrogen N4, the
distribution coefficient in pH 7.4 (logD7.4), polarizability (α), steric
effects at R and R’ (SER and SER’, respectively), topological descriptors
Platt, Balaban, Randič indexes, van der Waals volume (VvdW), solvent
accessible area (SAA) and total molar refractivity (A) and at R and R’
(AR and AR’, respectively). The values for these descriptors are pre-
sented in the Supplementary Table 1.
2.7. Principal component and hierarchical cluster analyses
The exploratory principal component (PCA) and hierarchical cluster
(HCA) analyses were carried out using the descriptors presented in
Table S1 using the PAST 3.17 software (Hammer et al., 2001). The
descriptor data were pretreated by autoscaling procedure, where each
value was subtracted by the column's mean and divided by the standard
deviation (van den Berg et al., 2006), to avoid the spans of the variables
in different ranges that could influence in the weight of each in-
dependent variable on the result of the analysis. The PCA eigenvectors
(principal components or PCs) were obtained by using the variance-
covariance matrix. The obtained component loadings are presented in
Table 2, and the scatter plot of the PCA data is presented in Fig. 4. The
HCA were performed through Ward's Method, which uses the Euclidean
similarity index to separate the compounds. The obtained dendrogram
plot is presented in Fig. 5.
2.4. Determination of the vasodilating potency of the compounds
The aortic rings were pre-contracted with NE to achieve an increase
>1 gf in the basal tension. Afterwards the compounds were added (10
µL) in different concentrations after NE contraction effect. The dihy-
drobenzofurans 5a-f, 6a, 7a and 8a were dissolved in dimetilsufoxide
(DMSO) to generate a stock solution of 1 mM and maximal con-
centration of 0.01% in the test medium. DMSO was also tested alone
(without compounds) to confirm no activity and no interference in the
rat's aorta rings assay. The tested concentrations of the compounds
ranged from 100 µM to 0.1 nM. Vasodilating effect of each compound
was determined as percentage of the reduction in the tension caused by
NE (100%) as the following formula:% tension
sion + compound/∆ NE tension) x 100.
= (∆ NE ten-
3. Results and discussion
2.5. Statistics
3.1. Effects of the compounds on isolated aortic rings
Potency data is expressed as means standard error of mean (SEM)
of a given number of observations (n). Potency was determined as
concentration to reduce the contraction induced by NE in 50% (EC50),
using non-linear regression in the GraphPad Prism 5.0 software. The
data was converted to the corresponding pEC50 [log(1/EC50), Table 1].
The difference between the potencies in the aortic rings with (E+) and
The search for novel vasodilating agents have been the focus of
several research reports. For instance, Silva et al. (2002) reported the
effects of LaSSBio-294 on rat aorta, showing IC50 value of 74 µM on
endothelium-intact rings. This effect seems to be dependent of GMPc,
since the activity was abolished when the endothelium was removed.
Table 1
Potency of the vasodilating activity of the compounds in thoracic aorta from Wistar and SHR rats (n = 6–10).
Compound
R
R’
pEC50
(
SEM)
Wistar E+
Wistar E–
SHR E+
SHR E–
5a
5b
5c
5d
5e
5f
H
Ph
Ph
Ph
Ph
Ph
Ph
Me
Allyl
H
6.07
0.16
5.61
0.08
8.62
7.69
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
0.22
0.13
8.05
5.67
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
0.25
0.13
Cl
6.02
0.31
5.14
0.21
0.34
Me
MeO
t-Bu
Ph
H
5.45
0.26
0.18
5.83
6.18
<4.00
<4.00
<4.00
<4.00
<4.00
<4.00
<4.00
<4.00
<4.00
<4.00
<4.00
6a
7a
8a
H
H
3

M. Ginoza, et al.
EuropeanJournalofPharmaceuticalSciences143(2020)105171
Table 2
each compound (Fig. 2), it can be noted that although compound 5d
showed higher EC50 value, the contraction achieved at maximal tested
concentration on E+ was only 36.3%, while compounds 5a and 5b
could reduce the contraction induced by NE to 28.0% and 22.2%, re-
spectively. When considering the effect in the Wistar's endothelium-
denuded vessels (E–), compound 5d did not reached 50% contraction,
and thus it showed EC50 values > 100 µM (pEC50 < 4.00). On the other
hand, compounds 5a and 5b showed EC50 2.4 and 7.1 µM, respectively.
Compounds 5e and 5f showed poor efficacy as vasodilators, with re-
laxing effects that not reached 50% reduction of the NE-induced con-
traction in both E+ or E– vessels. No important effects were observed
for compounds 6a, 7a or 8a for both E+ or E– vessels.
Loadings of the descriptors in each PC.
Descriptors
PC1
PC2
pKa N4
logD7.4
α
−0.23908
0.28646
0.29495
0.26555
0.24336
0.29589
−0.27609
0.29457
0.29473
0.29253
0.29616
0.25213
0.26414
0.43991
−0.15589
0.07133
SER
−0.35424
0.4946
SER’
Platt index
Balaban index
Randič index
VvdW
SAA
0.033006
−0.19742
0.018816
0.024644
0.02675
Considering these data, compounds 5a and 5b can be considered the
most promising in the set. The comparison between the effect on E+
and E– vessels through ANOVA showed that there is not significant
differences (p > 0.05), suggesting that the vasodilating effects of
LINS01 compounds on Wistar rats’ aorta are independent of endothelial
factors, with exception made to compound 5d, which showed sig-
nificant different effects (p < 0.01) between E+ and E– vessels, and
Bonferroni's post-hoc test showed that these differences occurred in all
concentrations tested (p < 0.05). This suggests that the presence of the
methoxy group may exert important influence on the mechanisms be-
hind the vasodilating effect. Although the mechanism of action for all
these compounds remains unknown, studies regarding the possible
pathways involved in the vasodilating effect are in progress by our
group.
A
0.016298
0.4604
AR
AR’
−0.39115
Other derivatives were prepared showing improved potency, such as
LaSSBio-1456 [IC50 3.1 µM (E+) and 17.8 µM (E–)] (Kümmerle et al.,
2009). Recently, Magalhães’ group (Teófilo et al., 2017; Vasconcelos
et al., 2018a) have been investigating the effects of nitro-derivatives
with potencies between 1 mM and 61 µM, with no difference between E
+ and E– aortic rings. Further, Egyptian researchers prepared and
evaluated benzofuran-containing compounds in rat thoracic aorta
(Srour et al., 2016) derived from previously reported benzimidazoles
(Nofal et al., 2013) with similar potencies, ranging from IC50 145 to
488 nM.
The obtained EC50 values suggest that LINS01 compounds are very
promising vasodilating agents. Compounds 5a, 5b and 5d showed
submicromolar potency in Wistar-derived aorta, suggesting to be more
potent than prazosin (IC50 487 µM) (Nofal et al., 2013) and amiodarone
(IC50 312 µM) (Srour et al., 2016), which are drugs with clinical use to
treat CVDs, and comparable to other potent compounds such as N-
acylhydrazones (Kümmerle et al., 2009). Moreover, these compounds
are more potent than other reported in literature such as pyridazines
Table 1 summarizes the obtained results for the vasodilating activity
of LINS01 compounds on Wistar and SHR rats-derived thoracic aorta
rings. The most potent compound in Wistar's endothelium-intact aorta
rings was 5d (E+ EC50 649 nM), followed by compounds 5a and 5b (E
+ EC50 853 and 941 nM, respectively). However, ANOVA showed no
statistical differences between the potencies of these compounds (p >
0.05). Considering the maximal effect at the higher concentration of
Fig. 2. Effects of the compounds 5a (A), 5b (B), 5c (C) and 5d (D) on Wistar rat aorta rings (n = 6–10) with (E+) and without (E–) endothelium. Vertical bars
represent the standard error of mean ( SEM).
4

M. Ginoza, et al.
EuropeanJournalofPharmaceuticalSciences143(2020)105171
Fig. 3. Effects of the compounds 5a (A) and 5b (B) on SHR rat aorta rings (n = 6–10) with (E+) and without (E–) endothelium. Vertical bars represent the standard
error of mean ( SEM).
(George and Saleh, 2016) or flavonoids (Herrera et al., 1996), and also
higher potency than other benzofuran-containing compounds such as
benzofuran-pyridines (Srour et al., 2016).
as t‑butyl or phenyl. The steric and topological descriptors in Table S1
support this hypothesis.
To evaluate the SAR by a more systematic manner, a set of mole-
cular descriptors were calculated and used to perform a statistical
analysis through PCA and HCA. The PCA and HCA are exploratory
statistical methods to find what molecular features are playing the role
in a given set of compounds (Varela et al., 2018; Vendrame et al.,
2002). PCA uses the selected descriptors (Table 2 and S1) to build new
orthogonal variables which explain the variance of the descriptor's
values, known as PCs. These PCs load the information from the de-
scriptors, used to identify the SAR data for a set of compounds. HCA
uses the same descriptors to generate a separation tree (known as
dendogram) according to the similarity of the molecules, clustering the
compounds by a similarity index (Euclidean distances). When the
compounds are grouped in the same cluster mean that they present high
similarity, which can be correlated to the activity. It must be stressed
that PCA and HCA are different but complementary analysis
(Ferreira et al., 2014), and thus if both present similar results, the ob-
tained SAR can be considered reliable and not a product from chance
(Varela et al., 2018).
In order to compare the effects of selected compounds in a more
reactive blood vessel, the compounds 5a and 5b were also evaluated in
thoracic aorta rings from SHR (Table 1). Similarly to the observed on
Wistar aorta, these compounds showed to be more potent in the en-
dothelium-intact than in the endothelium-denuded aorta, but with in-
creased potency. Compound 5a showed nanomolar potency (EC50
2.4 nM) in E+ and also in E–, with comparable potency (EC50 9.0 nM).
In both cases, this compound practically reverted the contraction in-
duced by NE, leading the tension to the baseline at 1 µM, and leading to
negative NE-induced tension values when the concentration was 10 µM
(−16% and −27% to E+ and E–, respectively; Fig. 3A). Similarly,
compound 5b was also more potent in SHR vessels, however with lower
potency [EC50 20 nM (E+) and 2.1 µM (E–)] than 5a. The maximal
effect for compound 5b was 14% to E+ and 29% E– at the higher
concentration tested (~100 µM, Fig. 3B).
Two-way ANOVA showed that there are significant differences be-
tween the effects of 5b in E+ and E– vessels (p < 0.01), but not to the
effects of compound 5a, suggesting that similarly to the observed for
Wistar-derived vessels, the substituent in R’ may play a role in the
mechanism of vasodilating effect. Moreover, the effects in SHR vessels
were also compared to those obtained to Wistar vessels through two-
way ANOVA, and significant differences were identified (p < 0.001),
suggesting that LINS01 compounds are more effective in SHR than in
Wistar rats. This also suggests that the effects of LINS01 compounds are
dependent on possible misregulated pathways in the SHR rat cells.
Further studies are in progress in our group to unveil the mechanisms
involved in the vasodilating effects of LINS01 compounds.
The PCA scatter plot (Fig. 4) showed a clear separation of the
compounds into three clusters, comprised by the active N-phenylpi-
perazine derivatives 5a-d (cluster A), the poorly active N-phenylpi-
perazine derivatives 5e and 5f (cluster B) and the completely inactive
compounds 6a, 7a and 8a (cluster C). PCA showed that the two main
PCs (PC1 and PC2) explained 97.0% of the total variance of the model,
being 87.2% for PC1 and 9.8% for PC2. Table 2 shows the weight
(loadings) of each descriptor in each PC. Data presented in Table 2
indicate that properties related to volume and ionization have high
impact on separation of the compounds by PC1 and PC2, suggesting
that volume and basicity of the nitrogen plays the role in the activity.
Positive values for PC1 and negative values for PC2 separate the more
active compounds from cluster A. As can be noted in Fig. 4, compounds
in cluster A are bigger and less basic than compounds in cluster C,
mainly given by the N-phenylpiperazine motif, and accordingly this
group seems extremely important to keep the vasodilating activity. On
the other hand, PC2 also suggests that the increasing volume in R leads
to positive value for PC2 and is detrimental to the vasodilating activity.
Compounds 5e and 5f in cluster B present bulky groups such as t‑butyl
and phenyl, respectively, in R’ and showed to be poorly active.
Similar results were obtained in the HCA, as can be seen in Fig. 5.
The dendrogram shows that compounds that contain the N-phenylpi-
perazine motif in clusters A and B are sharply different from the in-
active compounds 6a, 7a and 8a, which are separated in the cluster C.
Additionally, the more active N-phenylpiperazine compounds 5a-d are
grouped in the same cluster A, while compounds 5e and 5f are in a
separated cluster B, corroborating to the observation that the higher the
volume in R, the lower the activity. Since the results from both
3.2. SAR analysis
The results provided some structure-activity relationship (SAR) data
regarding these compounds. As can be seen in Table 1, the presence of
methyl or allyl group (compounds 6a and 7a, respectively) in the pi-
perazine moiety (R’) completely abolished the vasodilating activity,
suggesting that the N-phenylpiperazine moiety (5a) is essential to the
activity. Additionally, the presence of a substituent in the dihy-
drobenzofuran nucleus (R) plays the role in the potency. When small
substituents are attached in the aromatic ring, the potency is main-
tained. For instance, compounds 5b and 5c present similar EC50 values
to those obtained for 5a, and compound 5d was slightly more potent
than 5c. However, the activity of compounds with bulkier groups such
as t‑butyl (5e) or phenyl (5f) is strongly decreased. This suggests that
there is a steric limit to bind to the pharmacological target, which al-
lows the binding of the compound 5d presenting the methoxy group,
but does not allow the interaction of molecules with bigger groups such
5

M. Ginoza, et al.
EuropeanJournalofPharmaceuticalSciences143(2020)105171
Fig. 4. Scatter plot for the PCA results. Cluster A is comprised by the more active compounds, cluster B by the poorly active N-phenylpiperazine compounds and
cluster C by the inactive compounds.
methodologies were almost identical, the proposed SAR can be con-
sidered feasible and must be taken into account for designing com-
pounds with improved activity .
agents in rat aorta, that can be considered promising agents for treating
cardiovascular and other vascular-related diseases. Prototype com-
pound 5a (LINS01005) should be highlighted due to its high potency,
showing submicromolar EC50 values for Wistar rats and nanomolar
EC50 values for SHR. The significant differences between the effects on
Wistar and SHR rats suggests that the effect is possibly linked to the
misregulated pathways involved in the SHR aorta reactivity. The SAR
4. Conclusion
In summary, the present paper reports novel potent vasodilating
Fig. 5. Dendrogram obtained by the HCA. Cluster A is comprised by the more active compounds, cluster B by the poorly active N-phenylpiperazine compounds and
cluster C by the inactive compounds.
6

M. Ginoza, et al.
EuropeanJournalofPharmaceuticalSciences143(2020)105171
studies indicated that activity is highly dependent of the presence of the
N-phenylpiperazine motif, and potency depends on volume of the
substituents. Further modifications should consider these features to
improve the efficacy of future analogues. Although the studies to unveil
the exact mechanism of action of such compounds are in progress in our
group, this is the first report of the activity of such compounds, and
brings important contribution to the rational design of novel vasodila-
tors.
on blood pressure, glucose metabolism and albuminuria of spontaneously hy-
pertensive rats. J. Bras. Nefrol. 36. https://doi.org/10.5935/0101-2800.20140040.
Kümmerle, A.E., Raimundo, J.M., Leal, C.M., da Silva, G.S., Balliano, T.L., Pereira, M.A.,
de Simone, C.A., Sudo, R.T., Zapata-Sudo, G., Fraga, C.A.M., 2009. Studies towards
the identification of putative bioactive conformation of potent vasodilator arylidene
N-acylhydrazone derivatives. Eur. J. Med. Chem. 44, 4004–4009. https://doi.org/10.
1016/j.ejmech.2009.04.044.
Laurent, S., Boutouyrie, P., 2015. The structural factor of hypertension. Circ. Res. 116,
1007–1021. https://doi.org/10.1161/CIRCRESAHA.116.303596.
Malachias, M., Souza, W., Plavnik, F., Rodrigues, C., Brandão, A., Neves, M., Bortolotto,
L., Franco, R., Figueiredo, C., Jardim, P., Amodeo, C., Barbosa, E., Koch, V., Gomes,
M., Paula, R., Póvoa, R., Colombo, F., Ferreira Filho, S., Miranda, R., Machado, C.,
Nobre, F., Nogueira, A., Mion Júnior, D., Kaiser, S., Forjaz, C., Almeida, F., Martim,
J., Sass, N., Drager, L., Muxfeldt, E., Bodanese, L., Feitosa, A., Malta, D., Fuchs, S.,
Magalhães, M., Oigman, W., Gomes, O., Pierin, A., Feitosa, G., Bortolotto, M.,
Magalhães, L., Silva, A., Ribeiro, J., Borelli, F., Gus, M., Passarelli Júnior, O., Toledo,
J., Salles, G., Martins, L., Jardim, T., Guimarães, I., Antonello, I., Lima Júnior, E.,
Matsudo, V., Silva, G., Costa, L., Alessi, A., Scala, L., Coelho, E., Souza, D., Lopes, H.,
Gowdak, M., Cordeiro Júnior, A., Torloni, M., Klein, M., Nogueira, P., Lotaif, L.,
Rosito, G., Moreno Júnior, H., 2016. Capítulo 1 - Conceituação, Epidemiologia e
Prevenção Primária. Arq. Bras. Cardiol. 107. https://doi.org/10.5935/abc.
20160151.
Acknowledgements
The authors are grateful to São Paulo Research Foundation –
FAPESP [grant numbers 2016/25028-3, 2016/23139-2 and 2017/
05441-6] for the research funding to JPSF's working group and for the
scholarships to MFC and GABF. JPSF is also thankful to National
Council for Scientific and Technological Development – CNPq [grant
number 306355/2018-2] for the scientific fellowship award.
Marona, H., Szkaradek, N., Rapacz, A., Filipek, B., Dybała, M., Siwek, A., Cegła, M.,
Szneler, E., 2009. Preliminary evaluation of pharmacological properties of some
xanthone derivatives. Bioorg. Med. Chem. 17, 1345–1352. https://doi.org/10.1016/
j.bmc.2008.12.031.
Supplementary materials
McInnes, G.T., 2007. Vasodilators, in: Comprehensive Hypertension. Elsevier, pp.
1037–1048. https://doi.org/10.1016/B978-0-323-03961-1.50087-8.
Nofal, Z.M., Srour, A.M., El-Eraky, W.I., Saleh, D.O., Girgis, A.S., 2013. Rational design,
synthesis and QSAR study of vasorelaxant active 3-pyridinecarbonitriles in-
corporating 1H-benzimidazol-2-yl function. Eur. J. Med. Chem. 63, 14–21. https://
doi.org/10.1016/j.ejmech.2013.01.042.
Supplementary material associated with this article can be found, in
the online version, at doi:10.1016/j.ejps.2019.105171.
References
Rattmann, Y.D., Cipriani, T.R., Sassaki, G.L., Iacomini, M., Rieck, L., Marques, M.C.A., da
Silva-Santos, J.E., 2006. Nitric oxide-dependent vasorelaxation induced by extractive
solutions and fractions of Maytenus ilicifolia Mart ex Reissek (Celastraceae) leaves. J.
Ethnopharmacol. 104, 328–335. https://doi.org/10.1016/j.jep.2005.09.026.
Silva, C.L.M., Noël, F., Barreiro, E.J., 2002. Cyclic GMP-dependent vasodilatory proper-
ties of LASSBio 294 in rat aorta. Br. J. Pharmacol. 135, 293–298. https://doi.org/10.
1038/sj.bjp.0704473.
Bouider, N., Fhayli, W., Ghandour, Z., Boyer, M., Harrouche, K., Florence, X., Pirotte, B.,
Lebrun, P., Faury, G., Khelili, S., 2015. Design and synthesis of new potassium
channel activators derived from the ring opening of diazoxide: study of their vaso-
dilatory effect, stimulation of elastin synthesis and inhibitory effect on insulin re-
lease. Bioorg. Med. Chem. 23, 1735–1746. https://doi.org/10.1016/j.bmc.2015.02.
043.
Canbakan, B., 2013. Rational approaches to the treatment of hypertension: drug ther-
apy—monotherapy, combination, or fixed-dose combination? Kidney Int. Suppl. 3,
349–351. https://doi.org/10.1038/kisup.2013.75.
Singh, A., Laribi, S., Teerlink, J.R., Mebazaa, A., 2017. Agents with vasodilator properties
in acute heart failure. Eur. Heart J. 38, 317–325. https://doi.org/10.1093/eurheartj/
ehv755.
Corrêa, M.F., Barbosa, Á., Fernandes, G.A.B., Baker, J.G., Fernandes, J.P.S., 2019.
Pharmacological and SAR analysis of the LINS01 compounds at the human histamine
H1, H2, and H3 receptors. Chem. Biol. Drug Des. 93, 89–95. https://doi.org/10.
1111/cbdd.13387.
Srour, A.M., Abd El-Karim, S.S., Saleh, D.O., El-Eraky, W.I., Nofal, Z.M., 2016. Rational
design, synthesis and 2D-QSAR study of novel vasorelaxant active benzofuran-pyr-
idine hybrids. Bioorg. Med. Chem. Lett. 26, 2557–2561. https://doi.org/10.1016/j.
bmcl.2016.03.054.
Corrêa, M.F., Barbosa, Á.J.R., Sato, R., Junqueira, L.O., Politi, M.J., Rando, D.G.,
Fernandes, J.P.S., 2016. Factorial design study to access the “green” iodocyclization
reaction of 2-allylphenols. Green Process. Synth. 5, 145–151. https://doi.org/10.
1515/gps-2015-0101.
Teófilo, T.M., Arruda-Barbosa, L., Rodrigues-Silva, J.M., Vale, J.K.L., Borges, R.S., Duarte,
G.P., Magalhães, P.J.C., Lahlou, S., 2017. Mechanism of the vasorelaxant effect in-
duced by trans-4-methyl-β-nitrostyrene, a synthetic nitroderivative, in rat thoracic
aorta. Clin. Exp. Pharmacol. Physiol. 44, 787–794. https://doi.org/10.1111/1440-
1681.12771.
Corrêa, M.F., Barbosa, Á., Teixeira, L.B., Duarte, D.A., Simões, S.C., Parreiras-e-Silva, L.T.,
Balbino, A.M., Landgraf, R.G., Bouvier, M., Costa-Neto, C.M., Fernandes, J.P.S., 2017.
Pharmacological characterization of 5-Substituted 1-[(2,3-dihydro-1-benzofuran-2-
yl)methyl]piperazines: novel antagonists for the histamine H3 and H4 receptors with
anti-inflammatory potential. Front. Pharmacol. 8, 825. https://doi.org/10.3389/
fphar.2017.00825.
van den Berg, R.A., Hoefsloot, H.C.J., Westerhuis, J.A., Smilde, A.K., van der Werf, M.J.,
2006. Centering, scaling, and transformations: improving the biological information
content of metabolomics data. BMC Genomics 7, 142. https://doi.org/10.1186/1471-
2164-7-142.
Varela, M.T., Romaneli, M.M., Lima, M.L., Borborema, S.E.T., Tempone, A.G., Fernandes,
J.P.S., 2018. Antiparasitic activity of new gibbilimbol analogues and SAR analysis
through efficiency and statistical methods. Eur. J. Pharm. Sci. 122, 31–41. https://
doi.org/10.1016/j.ejps.2018.06.023.
Corrêa, M.F., Reiner, D., Fernandes, G.A.B., Varela, M.T., Aranha, C.M.S.Q., Stark, H.,
Fernandes, J.P.S., 2019c. Profiling of LINS01 compounds at human dopamine D2 and
D3 receptors. J. Chem. Sci accepted.
Corrêa, M.F., Varela, M.T., Balbino, A.M., Torrecilhas, A.C., Landgraf, R.G., Troncone,
L.R.P., Fernandes, J.P.S., dos, S., 2017. 1-[(2,3-Dihydro-1-benzofuran-2-yl) methyl]
piperazines as novel anti-inflammatory compounds: synthesis and evaluation on H 3
R/H 4 R. Chem. Biol. Drug Des. 90, 317–322. https://doi.org/10.1111/cbdd.12947.
Ferreira, J.E.V., da Costa, C.H.S., de Miranda, R.M., Figueiredo, A.F., Ginarte, Y.M.A.,
2014. An interdisciplinary study on monosubstituted benzene involving computation,
statistics and chemistry. Educ. Química 25, 418–424. https://doi.org/10.1016/
S0187-893X(14)70061-8.
Vasconcelos, T.B., de, Ribeiro-Filho, H.V., Lahlou, S., Pereira, J.G., de, C., Oliveira, P.S.L.,
de, Magalhães, P.J.C., 2018a. Vasodilator effects and putative guanylyl cyclase sti-
mulation by 2-nitro-1-phenylethanone and 2-nitro-2-phenyl-propane-1,3-diol on rat
aorta. Eur. J. Pharmacol. 830, 105–114. https://doi.org/10.1016/j.ejphar.2018.04.
029.
Vendrame, R., Ferreira, M.M.C., Collins, C.H., Takahata, Y., 2002. Structure-activity re-
lationships (SAR) of contraceptive progestogens studied with four different methods
using calculated physicochemical parameters. J. Mol. Graph. Model. 20, 345–358.
Whelton, P.K., Carey, R.M., Aronow, W.S., Casey, D.E., Collins, K.J., Dennison
Himmelfarb, C., DePalma, S.M., Gidding, S., Jamerson, K.A., Jones, D.W.,
MacLaughlin, E.J., Muntner, P., Ovbiagele, B., Smith, S.C., Spencer, C.C., Stafford,
R.S., Taler, S.J., Thomas, R.J., Williams, K.A., Williamson, J.D., Wright, J.T., 2018.
2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline
for the prevention, detection, evaluation, and management of high blood pressure in
adults: a report of the American college of cardiology/American heart association
task force on clinical Pr. Circulation 138. https://doi.org/10.1161/CIR.
0000000000000596.
Franklin, S.S., Wong, N.D., 2013. Hypertension and cardiovascular disease: contributions
of the framingham heart study. Glob. Heart 8, 49–57. https://doi.org/10.1016/j.
gheart.2012.12.004.
George, R.F., Saleh, D.O., 2016. Synthesis, vasorelaxant activity and 2D-QSAR study of
some novel pyridazine derivatives. Eur. J. Med. Chem. 108, 663–673. https://doi.
org/10.1016/j.ejmech.2015.12.015.
Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: paleontological statistics software
package for education and data analysis. Paleontol. Electron. 4, 1–9.
Herrera, M.D., Zarzuelo, A., Jiménez, J., Marhuenda, E., Duarte, J., 1996. Effects of fla-
vonoids on rat aortic smooth muscle contractility: structure-activity relationships.
Gen. Pharmacol. 27, 273–277.
WHO, W.H.O., 2019. Cardiovascular diseases [WWW document]. URLhttps://www.who.
int/health-topics/cardiovascular-diseases/.
Jodas, E.M.M.G., Voltera, A.F., Ginoza, M., Junior, Kohlmann, O., Santos, dos, N.B.,
Cesaretti, M.L.R., 2014. Effects of physical training and potassium supplementation
7