Aromatic Regions Govern the Recognition of NADPH Oxidase Inhibitors as Diapocynin and its Analogues

Article abstract of DOI:10.1002/ardp.201700041

Oxidative stress is related to the pathogenesis and progress of several human diseases. NADPH oxidase (NOX), and mainly the NOX2 isoform, produces superoxide anions (O₂ ^{? ?}). To date, it is known that NOX2 can be inhibited by preventing the assembly of its subunits, p47phox and p22phox. In this work, we analyzed the binding to NOX2 of the apocynin dimer, diapocynin (C1), a known NOX2 inhibitor, and of 18 designed compounds (C2–C19) which have chemical relationships to C1, by in silico methods employing a p47phox structure from the Protein Data Bank (PDB code: 1WLP). C1 and six of the designed compounds were recognized in the region where p22phox binds to p47phox and makes π–π interactions principally with W193, W263, and Y279, which form an aromatic-rich region. C8 was chosen as the best compound according to the in silico studies and was synthesized and evaluated in vitro. C8 was able to prevent the production of reactive oxygen species (ROS) similar to C1. In conclusion, targeting the aromatic region of p47phox through π-interactions is important for inhibiting NOX activity.

Full text of DOI:10.1002/ardp.201700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
Archiv der Pharmazie
Full Paper
Aromatic Regions Govern the Recognition of NADPH Oxidase
Inhibitors as Diapocynin and its Analogues
1
1,2
1
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Martha E. Macıas Perez , Maricarmen Hernandez Rodrıguez , Laura C. Cabrera Perez ,
3
1,2
4
ꢀ
ꢀ
ꢀ
M. Jonathan Fragoso-Vazquez , Jose Correa-Basurto , Itzia I. Padilla-Martınez ,
2
5
6
7
ꢀ
ꢀ
1
ꢀ
David Mendez Luna , Elvia Mera Jimenez , Cesar Flores Sandoval , Feliciano Tamay Cach ,
and Martha C. Rosales-Hernandez
ꢀ
1
ꢀ
ꢀ
ꢀ
ꢀ
Laboratorio de Biofısica y Biocatalisis, Seccion de Estudios de Posgrado e Investigacion de la Escuela
ꢀ
ꢀ
ꢀ
Superior de Medicina del Instituto Politecnico Nacional, Ciudad de Mexico, Mexico
Laboratorio de Modelado Molecular y Diseno de Farmacos, Seccion de Estudios de Posgrado e
Investigacion de la Escuela Superior de Medicina del Instituto Politecnico Nacional, Ciudad de Mexico,
2
~
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Mexico
3
4
5
6
7
ꢀ
ꢀ
ꢀ
ꢀ
Departamento de Quımica Organica, Escuela Nacional de Ciencias Biologicas del Instituto Politecnico
Nacional, Ciudad de Mexico, Mexico
Unidad Profesional Interdisciplinaria de Biotecnologıa, Instituto Politecnico Nacional, Ciudad de
Mexico, Mexico
Laboratorio de Cultivo Celular de la Seccion de Estudios de Posgrado e Investigacion de la Escuela
Superior de Medicina del Instituto Politecnico Nacional, Ciudad de Mexico, Mexico
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Gerencia de Desarrollo de Materiales y Productos Quımicos, Instituto Mexicano del Petroleo, Eje Central
ꢀ
ꢀ
ꢀ
ꢀ
(Lazaro Cardenas), Ciudad de Mexico, Mexico
ꢀ
ꢀ
ꢀ
ꢀ
Laboratorio de Investigacion de Bioquımica, Seccion de Estudios de Posgrado e Investigacion de la
Escuela Superior de Medicina del Instituto Politecnico Nacional, Ciudad de Mexico, Mexico
ꢀ
ꢀ
ꢀ
Oxidative stress is related to the pathogenesis and progress of several human diseases. NADPH oxidase
(NOX), and mainly the NOX2 isoform, produces superoxide anions (O2•ꢀ). To date, it is known that
NOX2 can be inhibited by preventing the assembly of its subunits, p47phox and p22phox. In this work,
we analyzed the binding to NOX2 of the apocynin dimer, diapocynin (C1), a known NOX2 inhibitor, and
of 18 designed compounds (C2–C19) which have chemical relationships to C1, by in silico methods
employing a p47phox structure from the Protein Data Bank (PDB code: 1WLP). C1 and six of the
designed compounds were recognized in the region where p22phox binds to p47phox and makes p–p
interactions principally with W193, W263, and Y279, which form an aromatic-rich region. C8 was chosen
as the best compound according to the in silico studies and was synthesized and evaluated in vitro. C8
was able to prevent the production of reactive oxygen species (ROS) similar to C1. In conclusion,
targeting the aromatic region of p47phox through p-interactions is important for inhibiting NOX
activity.
Keywords: Drug design / Reactive oxygen species / Virtual screening
Received: February 15, 2017; Revised: July 20, 2017; Accepted: July 21, 2017
DOI 10.1002/ardp.201700041
Additional supporting information may be found in the online version of this article at the publisher’s web-site.
:
Introduction
ꢀ
Correspondence: Dr. Martha C. Rosales-Hernandez, Plan de San
ꢀ
ꢀ
ꢀ
ꢀ
Luis y Dıaz Miron s/n Casco de Santo Tomas, Delegacion Miguel
Oxidative stress plays a central role in the development of
cardiovascular and microvascular complications in various
ꢀ
ꢀ
Hidalgo, CP.11340, Mexico, D.F., Mexico.
E-mail: marcrh2002@yahoo.com.mx
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(1 of 11) e1700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
M. E. Macıas Perez et al.
ꢀ
ꢀ
Archiv der Pharmazie
chronic pathologies such as hypertension and diabetes [1].
There is evidence that endothelium-dependent relaxation is
diminished in these patients and in animal models because of
the increased production of reactive oxygen species (ROS) or
damage to antioxidant defenses [2]. Increased ROS produc-
tion gives rise to endothelial dysfunction, decreasing the
catalytic activity of endothelial nitric oxide synthase (eNOS)
and increasing the activity of endothelin-1 (ET-1) as well as the
inducible nitric oxide synthase (iNOS), which is responsible for
producing excessive amounts of nitric oxide and is related to
hypertension and diabetes [3, 4]. It has also been reported
that there are several functional changes in the smooth
muscle cells of hypertensive and diabetic animals [2, 5].
Although there are several treatment options for hyper-
tension and diabetes, they do not prevent or avoid the
development of complications from vascular diseases pro-
duced by ROS production. Hence, antioxidant molecules have
been used, though they have only been employed in
experimental and clinical trials [6, 7]. Antioxidant molecules
are able to scavenge ROS and some exhibit inhibitory activities
against specific targets related to ROS production, which
would be more beneficial than traditional treatments for
various pathologies that involve a vascular disease. NADPH
oxidase is the main enzyme that produces superoxide anion
(O2•ꢀ); it is widely distributed in endothelial cells, where the
predominant isoforms are NOX1, NOX2, and NOX5 [8]. Thus,
NADPH oxidase is an important therapeutic target to
counteract the damage caused by the overproduction of
O2•ꢀ. Through these two mediators, NO• and O2•ꢀ can react,
yielding peroxynitrite (ONOO•ꢀ), which is cytotoxic, dimin-
ishes the quantity of NO•, and affects muscular relaxation [8].
It has recently been reported that the inhibition of NADPH
oxidase improves cardiac function in diabetes [9, 10].
However, compounds such as diphenyleneiodonium (DPI)
and diapocynin, which can inhibit NADPH oxidase, have been
only used in experimental models [11]. It has been reported
that 80% of administered apocynin is recovered unchanged in
the urine when it was administered to rats [12]; only a small
amount (<20%) of apocynin can be converted to diapocynin
by myeloperoxidase (MPO) [11]. Diapocynin is one of the most
studied compounds and inhibits NOX2 through several
mechanisms, one of which is preventing the binding between
NOX2 subunits [11]. NOX2 is an oligomeric protein composed
of three cytosolic (p60phox, p47phox, and p40phox) subunits
and two transmembrane (p91phox and p22phox) subunits.
Accordingly, O2•ꢀ production by NOX2 requires that p22phox
form a complex with p47phox [13]. A proposed NADPH
oxidase inhibition mechanism by diapocynin is one in which
the compound binds to p47phox and prevents p22phox
assembly [11]. It is worthy to mention that binding between
p22phox and p47phox is mediated by p–p and hydrophobic
interactions; thus, the inhibition of binding occurs because
diapocynin interacts with p47phox through similar interac-
tions due to its aromatic nature.
interaction between the p47phox and p22phox NADPH
oxidase subunits, which could represent an important
strategy for inhibiting ROS production. Therefore, this work
was focused on studying the binding of diapocynin to
p47phox to explore whether this compound was recognized
in the same place as p22phox for the design of compounds
that are structurally related to diapocynin (Fig. 1; C1).
Subsequently, the target compounds were evaluated by in
silico methods to assess their binding to and affinity for
p47phox to select the best compound. This compound was
then synthesized and submitted for in vitro antioxidant
evaluations as well as testing in endothelial cells from the
aorta of non-diabetic and diabetic rats to measure the
inhibition of ROS production.
Results
In silico studies
Docking
Given that NADPH oxidase activity depends on p47phox–
p22phox complex formation, several compounds were
designed to prevent p47phox–p22phox binding and inhibit
NADPH oxidase activation. Based on information regarding
the protein surfaces involved in the recognition between
p47phox and p22phox, docking studies were performed. C1
(Fig. 1) was recognized in the 1WLP structure, examining the
short chain of p22phox binding to p47phox for: C1 (Fig. 1) was
recognized in the 1WLP structure without the short chain of
p22phox binding to p47phox. To show that the regions where
both subunits interact could be targeted by the designed
compounds, we demonstrated that amino acid residues I164,
A165, Y167, V186, E187, W193, F195, W204, P206, S208, and
F209 in the SH3A domain as well as Y237, D243, E244, W263,
Y274, Y279, and M278 in the p47phox SH3B domain were
important for the recognition of p22phox as has been
reported [14]. The results from a blind docking study showed
that C1 was recognized by p47phox in the same place where
p22phox binds (Fig. 2A and B), regions which are rich in
aromatic residues. This is important because C1 has been
reported as an NADPH oxidase inhibitor, suggesting that p–p
interactions are important for NADPH oxidase inhibition.
Because C1 is related to designed compounds (C2–C19), some
of them were recognized in the same place as C1 on p47phox.
The best compounds were selected by choosing those that
demonstrate interactions with amino acid residues (in
p47phox) similar to C1 and had a DG value of less than
ꢀ5 kcal/mol, which was also shown for C1 (Table 1). Only six
compounds, C8, C10, C11, C17, C18, and C19, satisfied the
criteria mentioned above (Fig. 2). Furthermore, C1 and listed
compounds showed p–p interactions with W263, Y167, F209,
W193, Y274, and W279 and hydrogen bonds with S208
(Table 1). Figure 2C–H shows the interactions between the
p47phox and each compound, where all the compounds are
located in the same region. It is also possible to observe that
the most important interaction is with the aromatic rings; due
Consequently, structurally designed compounds based on
an aromatic system could be capable of preventing the
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(2 of 11) e1700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
Aromatic Regions Recognize NADPH Oxidase Inhibitors
Archiv der Pharmazie
Figure 1. Chemical structures of designed diapocynin analogues and diapocynin (C1).
to the presence of a hydroxyl group, a p–H interaction can be
formed in addition to the p–p interactions with W193, W263,
F209, W204, and Y274. Thus, the compounds could prevent
the interaction between p47phox and p22phox. This effect
also explains why the apocynin trimer is a good NADPH
oxidase inhibitor.
As shown in Fig. 2 the compounds that bound to the p47phox
surface are akin to C1, showing comparable DG values
(Table 1), despite of having different substituents, specifically
where the keto group is located in C1. The in silico studies
demonstrated that six compounds could be promising NADPH
oxidase inhibitors because they have an analogous binding
Figure 2. Interactions obtained from the docking of diapocynin and the designed compounds with the 1WLP structure.
(A) Interactions between p47phox and p22phox in the 1WLP structure. (B) Interactions between the designed compounds and C1 in
the same place as p22phox. (C) General binding and the amino acid residues involved in the recognition between p47phox and
diapocynin (C1, yellow), (D) C8 (blue), (E) C10 (lime), (F) C11 (red), (G) C17 (pink), and (H) C19 (purple).
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(3 of 11) e1700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
M. E. Macıas Perez et al.
ꢀ
ꢀ
Archiv der Pharmazie
Table 1. Free energy values (DG ꢀ kcal/mol) obtained by docking with p47phox from 1WLP structure and the designed
compounds and diapocynin (C1).
Compound
DG (ꢀkcal/mol)
Amino acids of interaction
Y167, P206, S208, F209, D261, W263, Y274
C1
C8
C10
C11
C17
C18
C19
ꢀ5.70
ꢀ5.19
ꢀ5.07
ꢀ5.42
ꢀ5.29
ꢀ5.40
ꢀ5.92
W193, G192, P206, S208, F209, G262, W263
Y167,S191, G192, W193, P206, Y237, G262, W263, P276, M278, Y279
G192, W193, P206, S208 K235, Y237, G262, W263, M278, Y279
G192, W193, F206, S208, F209, D261, G262, W263, Y274
Y167, P206, S208, F209, E244, D261, G262, W263, Y274
W193, P206, S208, F209, E244, D261, W263, Y274
mode to C1 and similar DG values, demonstrating that
p-interactions could be the most important feature during
the recognition between the p47phox and its ligands.
recognized in the same form as diapocynin. It is important to
mention that these compounds have the highest GAP values
(Fig.3).GAPisameasureofthereactivity,wherelowGAPvalues
mean high reactivity. Figure 3 shows that C2, C6–C9, C15–C19
have low GAP values similar to C1.
Frontier molecular orbitals
The molecular orbitals were determined to analyze the
chemical descriptors of compounds C8, C10, C11, C17–C19
and identify what makes them good NADPH oxidase inhibitors
according to docking calculations. First, the optimized molecu-
lar structures with a minimum energy were obtained as shown
in Fig. 3. The docking analyses showed that all the compounds
have p–p interactions, but the presence of different substitu-
ents in the aromatic ring could influence ligand recognition.
Optimized structures of the compounds that have a chemical
group with a negativeor positive charge in their chain were not
Because compounds C8, C17–C19 are recognized in the same
place as C1 by docking studies and have low GAP values, it was
possible to distinguish that compounds C8 and C17 have a
hydrophobic chain, whereas C18 and C19 have either a
hydroxyl, ether or ester group. The compounds with negatively
charged groups allow for hydrogen bonding and show that the
compound has low LUMO energy, allowing for highest
occupied molecular orbital (HOMO)–lowest unoccupied molec-
ular orbital (LUMO) interactions between the ligand and the
protein. Even though the protein can recognize these ligands
Figure 3. Optimized molecular structures of compounds C1, C8, C10, C11, C17, and C19 and the graphic of the HOMO (highest
occupied molecular orbital), LUMO (lowest unoccupied molecular orbital) and GAP energies.
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(4 of 11) e1700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
Aromatic Regions Recognize NADPH Oxidase Inhibitors
Archiv der Pharmazie
with chemical cores similar to C1 and polar groups in the
aromatic ring, the polar groups could weaken the electronic
effects of the aromatic ring and prevent effective p–p
interaction with the protein aromatic region.
Thus, compound C8 was selected to be synthesized and
assessed because this compound has hydrocarbon chains and
aromatics rings.
with superoxide dismutase (SOD) to corroborate superoxide
anion production. The diminished absorbance from this
reaction indicated that SOD catalyzes the conversion of
superoxide anion to hydrogen peroxide (H2O2) (data not
shown because these values are subtracted from the samples
to obtain the total production of superoxide anion in each
sample). When diapocynin was added to the reaction, the
absorbance also decreased, indicating that diapocynin can
scavenge the superoxide anion; however, this fact was more
evident with compound C8 (Fig. 5A).
Chemical synthesis
Compounds C1 and C8 were obtained as depicted in Fig. 4. To
form these derivatives, two monomers were reacted to obtain
the corresponding dimer. Diapocynin synthesis was performed
as previously reported, with slight modifications in the
purification step, where the pH was maintained low by using
a6 Naqueoussolutionofhydrochloricacid[15]. Thesynthesisof
C8 was conducted to maintain the groups in the aromatic ring,
such as apocynin, and with a flexible hydrocarbon chain to
increase its affinity toward p47phox; this effect of the
hydrocarbon chain over the affinity has been previously
reported [16]. C8 synthesis was achieved by mixing 2-
methoxy-4-propylphenol in a solution of H2O/MeOH (3:1 v/v).
In both cases, the reaction was performed in the presence of
ferrous sulfate (FeSO4•7H2O). To obtain the corresponding
dimers, the compounds were purified and characterized by
nuclear magnetic resonance (1H and 13C NMR), infrared (FT-IR)
and mass spectrometry (MS). The results were consistent with
previous reports for the mp [17, 18], NMR signals and MS
measurements of diapocynin [15].
DPPH reduction determination
Figure 5B shows the results obtained from the DPPH reduction
measurements; we observed that 5-aminosalicylic acid
(5-ASA) and C8 can reduce DPPH at concentrations as low
as 0.013 and 0.026 mM, respectively. Diapocynin was able to
reduce DPPH nearly 45% but only at a concentration of
0.408 mM, while 5-ASA and compound C8 reached this
percentage at lower concentrations (0.026 mM).
Fenton reaction
The electron paramagnetic resonance (EPR) spectra from the
reaction of the compounds as free radical scavengers of the
hydroxyl radicals produced by the Fenton reaction are shown
in Fig. 5C. In Fig. 5C, the control corresponds to the Fenton
reaction without compound. The EPR spectrum has hyper-
fine splitting constants of aN ¼ 15.14 G and aH ¼ 3.3 because
of the PBN-CH3 adduct. However, considering the reactions
reported by Polyakov [19], it is necessary for hydroxyl radical
formation to generate PBN-CH3. The same EPR spectra shape
was observed when 5-ASA was evaluated (Fig. 5C; 5-ASA). In
the presence of 5-ASA, the signal from the adduct dimin-
Antioxidant activity
Superoxide anion determination
The activities of C1 and C8 as superoxide anion scavengers was
evaluated by the reduction of cytochrome c. Superoxide anion
was produced by the reaction between hypoxanthine and
xanthine, which was shown by the high cytochrome c
absorbance values (Fig. 5A). The same reaction was performed
ished in intensity (Fig. 5C). Furthermore,
difference in the area under the curve was obtained
between the control and 5-ASA
(79289 ꢁ 2022)
(35305 ꢁ 2601). This difference in EPR intensity was obtained
a significant
Figure 4. ReactionpathwayforcompoundsC1andC8.
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(5 of 11) e1700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
M. E. Macıas Perez et al.
ꢀ
ꢀ
Archiv der Pharmazie
Figure 5. Free radical scavenging activity determination of C1 and compound C8 on superoxide anions, DPPH radicals and hydroxyl
radicals. (A) Scavenging of superoxide anions by compound C8 and C1 shown by the reduction of cytochrome c. (B) Reduction of
DPPH radicals by 5-ASA, compound C8 and C1. (C) Scavenging of hydroxyl radicals by 5-ASA, compounds C8 and C1.
when compounds C1 and C8 were used (Fig. 5C), which could
be due to the interaction between the compound and the
hydroxyl radical. The signal of the PBN-CH3 adduct was
diminished, possibly because the compound reacted with the
hydroxyl radical and prevented the PBN-CH3 adduct from
forming. Moreover, EPR measurements showed that dia-
pocynin and C8 are good hydroxyl radical scavengers. These
results are consistent with the findings obtained with DPPH,
and we can therefore confirm that diapocynin and com-
pound C8 are free radical scavengers.
yellow, indicating increased ROS production, and it
increases with respect to time. The b value is higher at
15 min in the ECDRs, while in ECDRs incubated with C1 or
C8, ROS production was not observed at 5 or 15 min. The
color intensities in Fig. 6D correspond to the graphics in
Fig. 6C.
Therefore, ROS production is increased in ECRDs at 15 min,
while ECRDs incubated with diapocynin to C8 showed no
ROS production, which supports the theory that these
compounds inhibit ROS production. Furthermore, the effects
of compounds C1 and C8 on the superoxide anion produc-
tion (O2•ꢀ) was evaluated in neutrophils activated with
ROS detection in endothelial cells by DCFH-DA
The CIELAB parameters determined by confocal microscopy
in ECs and ECDRs with and without preincubation with
diapocynin and C8 are shown in Fig. 6. Figure 6A shows that
the luminosity (L) increased with respect to time in the
samples from ECs and ECDRs and that L is higher in ECDRs at
15 min, indicating an increase in ROS production. However,
the ECDRs incubated with C1 or C8 showed no ROS
production at either 5 or 15 min. Figure 6B shows the a
value obtained from the EC and ECDR samples was more
negative at 15 min, indicating the prevalence of the green
color that identifies the production of DCF and conse-
quently indicates ROS production. The a value is more
negative in ECDRs at 15 min, whereas when ECDRs are
incubated with C1 or C8, ROS production was not observed
at 5 or 15 min. In Fig. 6C, the positive value of b tends to be
lipopolysaccharide. The results showed that compounds C1
•ꢀ
and C8 were able to effectively inhibit O2
production
evaluated by EPR using CMH as the spin probe (Supporting
Information).
Discussion
Various pathologies are associated with the generation of
oxidative stress, which is one of the principal events related
to their genesis and progression. One example is diabetes,
where endothelial dysfunction is produced as a complica-
tion [3, 4]. Oxidative stress is produced as a consequence
of the imbalance between oxidative and antioxidative
molecules.
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(6 of 11) e1700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
Aromatic Regions Recognize NADPH Oxidase Inhibitors
Archiv der Pharmazie
Figure 6. CIELAB parameters determined by confocal microscopy in ECs and ECDRs with and without preincubation of diapocynin
and C8, stimulated with PMA using DCFH-DA, and measured at 5 and 15 min. (A) CIELAB parameter L. (B) CIELAB parameter a.
(C) CIELAB parameter b. (D) Photograph of cells observed by confocal microscopy.
NADPH oxidase, mainly the NOX2 isoform, is implicated in the
excessive production of superoxide anion (O2•ꢀ) during
pathological conditions and, consequently, the reduction of
nitric oxide (NO) concentrations. This is because O2•ꢀ and NO
can react to produce ONOO•ꢀ, which is more toxic toward
The in silico studies showed that compounds without polar
groups in the aromatic ring had similar affinities to C1 and
displayed similar binding forms. Furthermore, the GAP value
showed that these compounds had higher GAP energies
diminishing their reactivities. This could be due to the
presence of the electron-donating or electron-withdrawing
group, which affects the p–p interaction between the ligand
and the site formed by aromatic residues W193, W263, F209,
W204, and Y274 in p47phox, the site where p22phox binds.
This interaction is important in the SH3A domain for p22phox
recognition because of its interactions with W204 and F195.
Therefore, it is important to maintain the aromatic rings in the
ligand to allow for the p–p, p–H, and p–CH interactions
between the ligand and protein. These results are significant
because the in silico studies showed that it is important to
maintain the p-interaction to increase the affinity between
p47phox and its ligands [16, 22].
cells. Great efforts have been expended to inhibit the
•ꢀ
excessive production of O2
by NOX2 using different
mechanisms [20]. It is most feasible to inhibit NOX2 by
preventing the assembly of the p22phox and p47phox
subunits. Several compounds have been reported as NOX2
inhibitors, though the most studied inhibitor that prevents
the interaction between these subunits is C1 [21].
The design of compounds with chemical similarities to C1
could provide selective NOX2 inhibitors by targeting several
protein cavities. Therefore, 18 compounds were designed in
this work to maintain the two aromatic rings in C1. In some
cases, hydroxyl and methoxyl substituents were included on
the keto group of C1 in the aromatic ring to allow
additional non-covalent interactions at the binding sites
identified by in silico studies describing several C1 analog
ligands.
In this sense, it is well known that hydrophobicity is an
important chemical characteristic that allows phenolic acids,
such as the ones obtained from caffeic and ferulic acids, to have
better biological activities in various diseases [16]. Considering
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(7 of 11) e1700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
M. E. Macıas Perez et al.
ꢀ
ꢀ
Archiv der Pharmazie
these factors, protocatechuic acid alkyl esters were reported as
NADPH oxidase inhibitors with high activity, including the
heptyl protocatechuate named P7, with the conclusion that
increasing the hydrophobicity of the compound provoked a
strong inhibitory effect on NADPH oxidase. However, the study
does not show the possible mechanisms and interactions of the
compounds at the molecular level. We decided to synthesize
compound C8 based on the results obtained form the in silico
studiesinthiswork;C8hasa hydrophobicchainonthe aromatic
ring, and its p–p interactions are maintained, which is an
important characteristic of apocynin analogues for inhibiting
NADPH oxidase [22].
had similar DG values, principally interacted with the SH3A
domain, and showed that the p-interactions were the most
important chemical characteristics for inhibiting NOX2. C8
displayed antioxidant activity and was able to inhibit ROS
production in ECs and ECDRs similar to diapocynin. Therefore,
compound C8 should be evaluated using in vivo models where
ROS production is increased, such as with hypertension or
diabetes.
Experimental
In silico evaluations
The in vitro results showed that C8 exhibits antioxidant
activity similar to those previously reported for C1 [11].
In endothelial cells from non-diabetic and diabetic rats,
compound C8 was able to inhibit ROS production; similar
results were shown for C1. Several compounds have been
synthesized to prevent ROS production in ECs by inhibiting
the NOX2 enzyme; for example, compounds 10 and 12 were
the most potent in protecting the cells against cytotoxicity
induced by lipopolysaccharide (LPS) [23]. It is important to
note that compound 10 has a methylamine group between
the two apocynin aromatic rings. Compound 12 was obtained
by coupling apocynin to lipoic acid and displayed increased
hydrophobicity due to the presence of the hydrocarbon chain,
which allows a p–CH interaction with the rich aromatic region
in p47phox. Thus, as previously mentioned, the effect of
compound C8 could be due to the aromatic rings and the
presence of the hydrophobic chains. However, the in silico
results showed that the presence of this substituent did not
improve the affinity for p47phox due to its similarity to C1.
However, the molecular hydrophobicity (partitioning coeffi-
cient, log P) of compound C8 (log P ¼ 5.33) was better than
that of apocynin (log P ¼ 1.17) and diapocynin (log P ¼ 2.52),
as demonstrated by log P values obtained by using the
Molinspiration web server [24]. Clearly, these data should be
corroborated via further in vitro studies.
Furthermore, several studies have shown how the
intensity of confocal microscopy analysis is diminished
because of the presence of compounds evaluated as ROS
inhibitors; this was observed in this study with compounds
C1 and C8. However, the intensity abolished by compound
C8 could be due to the concentration employed, because
fluorescence intensity was observed with all the reported
compounds and diapocynin at 10 mM [23]. More studies
should be performed to assess compound C8 at different
concentrations and in other in vitro and in vivo models.
Additional studies using C8 as a lead molecule to design
better compounds to inhibit the assembly of p22phox and
p47phox should also be conducted.
To perform the in silico studies, the protein selection
considered only the p47phox domain because of its
importance during NADPH oxidase activation. Several
p47phox segments have been solved by X-ray crystallogra-
phy or NMR studies to yield three-dimensional (3-D)
structures, which are stored in the Protein Data Bank
(PDB) [25]. We used the 3-D p47phox structure, selecting
the most complete structure without mutations that
corresponds to the p47phox–p22phox complex (PDB code:
1WLP) [26]; this structure includes the SH3A and SH3B
domains. These regions are important for p47phox and
p22phox binding [14]. The 3-D ligand minimum energies of
the designed compounds, which are structurally related to
C1 (Fig. 1), were determined using the means from the
Gaussian 98 software at the AM1 level [27].
Docking procedure
Docking studies were performed to identify and analyze
where the target ligands interact with p47phox (1WLP). Prior
to docking, the water molecules and p22phox segment
present in 1WLP 3-D structure were removed. Then,
hydrogen atoms were added using the PSFGEN program,
which is included in the Visual Molecular Dynamics (VMD) 1.8
program [28]. The protein was minimized for 2000 steps
using the CHARMM27 force field [29] implemented in the
Nanoscale Molecular Dynamics (NAMD) 2.6 program [30]. All
the flexible bonds in the ligand were identified; their partial
atomic charges (Gasteiger–Marsili formalism) were calcu-
lated using AutoDock Tools 1.5.2 (21). For the proteins, the
Kollman charges for all the atoms were computed consider-
ing the polar hydrogens. All the other parameters were
maintained at their default values. The protein exploration
and binding site definitions were prepared using a GRID-
based procedure [28].
In the 1WLP structure, blind docking was achieved in a
̊
̊
60 ꢂ 60 ꢂ 60 A grid with 0.375 A spacing. The hybrid Lamarck-
ian Genetic Algorithm was used in all the simulations with an
initial population of 100 randomly placed individuals and
1 ꢂ 107 evaluations. Docked orientations within a root-mean
̊
square deviation (RMSD) of 0.5 A were clustered together.
Conclusion
The lowest binding free energy cluster returned for each
compound docked to the 1WLP structure was used for further
analysis. The interactions between the ligands and p47phox
Compounds C8, C10, C11, C17–C19 were recognized to have
the same interactions at C1 in the 1WLP structure. They also
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(8 of 11) e1700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
Aromatic Regions Recognize NADPH Oxidase Inhibitors
Archiv der Pharmazie
were visualized using AutoDock Tools 1.5.2 [31], and figures
were created using VMD 1.9.1 [28].
All the computational work was performed using CUDA on
an Intel Core i7–980 ꢂ 3.33 GHz Linux workstation with 12 GB
of RAM, (2x) NVIDIA Geforce GTX580 video cards, and a (1x)
NVIDIA Geforce GTX560 video card.
Brown solid: Rf ¼ 0.18 (EtOH/AcOEt 9:1); mp ¼ 306–308°C.
Solubility: AcOEt, DME, DMSO, and EtOH. FT IR (ATR) lmax
3311, 3013, 2948, 2853, 1663, 1586, 1506, 1285, and 842 cmꢀ1
.
1H NMR (300 MHz, DMSO-d6) d 9.47 (s, 2H, 2OH), 7.49 (d,
J ¼ 2 Hz, 2H, H-4, H-40), 7.48 (d, J ¼ 2 Hz, 2H, H-6, H-60), 3.90 (s,
6H, H-7, H-70), 2.50 (s, 6H, H-9, H-90). 13C NMR (75.4 MHz,
DMSO-d6) d 196.9 (C-8), 149.8 (C-3), 148.1 (C-4), 128.5 (C-1),
126.0 (C-6), 125.1 (C-5), 110.3 (C-2), 56.6 (C-7), 27.0 (C-9). ESI-
MS calcd. for C18H18O6: (MþNa)þ 353.1001. Found: 353.0976.
Frontier orbitals
The molecular structures were first optimized at the AM1
semiempirical level and subsequently optimized using DFT
calculations at the B3YLP6-31Gþ level. The optimized
structures were visualized with the Gauss View program.
The calculations were performed with the Gaussian 03
program package.
Synthesis of 3,30-dimethoxy-5,50-dipropyl-(1,10-biphenyl)-
2,20-diol (C8)
Seventy-five milligrams (0.3 mmol) of heptahydrate ferrous
sulfate (FeSO4 ꢃ 7H2O) and 810 mg (3 mmol) of potassium
persulfate (K2S2O7) were added to a solution of 1 g (6.0 mmol)
of 2-methoxy-4-propylphenol in 200 mL of hot water/MeOH
(3:1 v/v). The mixture was boiled and stirred for 1 h.
After cooling to room temperature, the precipitate was
collected and dissolved in 100 mL of a 4 N aqueous solution of
NaOH. Subsequently, crudeC8 was precipitated bytheaddition
of 100 mL of a 6 N aqueous solution of HCl. The resulting
precipitate was filtered and washed with hot deionized water
at 90°C (3 ꢂ 100 mL) and hot methanol (1 ꢂ 100mL). The
resultant solid was purified by flash column chromatography
with hexane/AcOEt (1:1) to obtain C8 at 66% yield.
Chemistry
All commercial grade reagents were obtained from Sigma–Aldrich
ꢀ
(Mexico) and used without further purification. Melting points
were determined in open capillary tubes with an ELECTROTHER-
MAL melting point apparatus and are uncorrected. 1H and
13C NMR spectra were recorded on a Bruker Advance III
spectrometer (1H 400 and 13C 125 MHz, respectively) for: 1H and
13C NMR spectra were recorded on a Bruker Advance III
spectrometer (1H 400 and 13C 125 MHz, respectively), on a Varian
Mercury spectrometer at (1H300and13C 75.4 MHz, respectively) or
Varian System (1H 500 and 13C 125 MHz, respectively), using
deuterated chloroform (CDCl3) or deuterated dimethyl sulfoxide
(DMSO-d6) as the solvent. Chemical shifts (d) are reported in ppm
downfield from the internal (CH3)4Si standard, and coupling
constants are reported in Hertz (Hz). IR spectra were run on a
Perkin-ElmerFT-IRspectrometer(SpectrumGX). Absorptionvalues
are expressed as wave numbers (cmꢀ1); only significant absorption
bands are given. Electrospray ionization (ESI) high-resolution mass
spectrometry was performed using a Bruker micrOTOF-Q II
instrument. Reactions were monitored by thin layer chromatog-
raphy (TLC) on aluminum-backed sheets with silica gel 60 GF254
(HX805651) and a fluorescent indicator and visualized with a UV
light lamp (254 nm). Flash chromatography was performed using
silica gel 60 (230–400 mesh).
Red solid: Rf ¼ 0.72 (hexane/AcOEt 1:1); mp ¼ 132–134°C.
Solubility: AcOEt, DME, DMSO, and EtOH. IR (ATR) lmax 3296,
3018, 2973, 2933, 2856, 1603, 1494, and 849 cmꢀ1 1H NMR
.
(500 MHz, CDCl3) d 6.74 (s, 2H, H-6, H-60), 6.72 (s, 2H, H-4, H-40),
6.02 (s, 2H, 2OH), 3.90 (s, 6H, H-7, H-70), 2.56 (t, 4H, J ¼ 7.5 Hz,
H-8, H-80), 1.65 (m, 4H, H-9, H-90), 0.96 (t, 6H, J ¼ 7.5 Hz, H-10,
H-100). 13C NMR (125 MHz, CDCl3) d 147.1 (C-2), 140.5 (C-3),
134.7 (C-5), 124.4 (C-1), 123.0 (C-6), 110.7 (C-4), 56.1 (C-7), 37.9
(C-8), 24.8 (C-9), 13.9 (C-10). ESI-MS calcd. for C20H26O4:
(MþNa)þ 353.1729. Found: 353.1625.
Antioxidant activity evaluation
Scavenging of superoxide anions by compounds C1 and C8
The method is based on cytochrome c reduction. In this assay,
The InChI codes of the investigated compounds are
provided as Supporting Information.
O2•ꢀ donates its unpaired electron to ferricytochrome c (Fe3þ
)
to form ferrocytochrome c (Fe2þ), resulting in an increase in
the absorbance at 550 nm. The reaction mixture contained
cytochrome c (92 mM in PBS), xanthine oxidase (167.5 mU/mL)
and hypoxanthine (200 mM in NaOH 30 mM) [32]. The
reduction of cytochrome c was determined using an ELISA
reader (Thermo Scientific Original Multiskan EX with an
AscentTM software) at a wavelength of 550 nm for 5 min at
room temperature. To confirm the production of O2•ꢀ, a
sample with superoxide dismutase (SOD) (10200 U/mL) and
catalase (120 U/mL) was used. To determine the ability of
compounds C8 and C1 to scavenge O2•ꢀ, these compounds
were employed at 100 mM [11] dissolved in DMSO (0.45%).
Synthesis of 1,10-(6,60-dihydroxy-5,50-dimethoxy-(1,10-biphenyl)-
3,30-diyl)-bis-(ethan-1-one), 5,50-dehydrodiacevanillone
(diapocynin; C1)
Seventy-five milligrams (0.3 mmol) of ferrous sulfate hepta-
hydrate (FeSO4 ꢃ 7H2O) and 810 mg (3 mmol) of potassium
persulfate (K2S2O7) were added to a solution of 1 g (6.0 mmol)
apocynin in 200 mL of water (90°C). The reaction mixture was
stirred in a boiling water bath for 0.5 h. After cooling to room
temperature, the precipitate was collected and dissolved in
100 mL of a 4 N aqueous solution of NaOH. Subsequently, the
product was precipitated by adding 100 mL of a 6 N aqueous
solution of HCl. The final precipitate was filtered and washed
with deionized water at 90°C (3 ꢂ 100 mL) and hot methanol
(1 ꢂ 100 mL) to obtain diapocynin at 63% yield.
DPPH reduction determination
A 3 ꢂ 10ꢀ5 M solution of DPPH dissolved in DMSO was
prepared. 5-ASA was used at different concentrations
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(9 of 11) e1700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
M. E. Macıas Perez et al.
ꢀ
ꢀ
Archiv der Pharmazie
(0.013, 0.026, 0.051, 0.102, 0.204, and 0.408 mM) as a positive
control because of its antioxidant activity [33]. The following
three different assays were performed: (i) mixing of 1 mL
DPPH solution with 1 mL DMSO; (ii) mixing of 5-ASA in DMSO
to different concentrations with 1 mL of DPPH solution; and
(iii) mixing different concentrations of 5-ASA with 1 mL of
DMSO. Finally, the samples were maintained in the dark for
60 min before the absorbance was measured at 517 nm using
DMSO as the blank on a Perkin-Elmer (Lambda 25) spectro-
photometer. The percentage of DPPH reduction was calcu-
lated as previously reported [34]. C1 and C8 were evaluated at
a mol to mol relation with 5-ASA. All the compounds were
dissolved in DMSO.
ROS production based on the fluorescence emitted by
dichlorofluorescein (DCF). Subsequently, 10 mM phorbol
12-myristate 13-acetate (PMA) was added to stimulate ROS
production. The samples were incubated at 37°C for 5 and
15 min [39]. Afterwards, three washes were performed and
the cells were observed by confocal microscopy (Confocal
Laser Scanning Microscope LSM 710 NLO; objective lens Plan-
Apochromat 40x/1.301LDICM27; acquisition filter Ex/Em 493/
540 nm; laser excitation 488 with 2% transmittance. Gain
master: 530–540; digital offset: ꢀ46; and digital gain: 1%).
Image analysis
The images obtained by confocal microscopy from the EC and
ECDR samples in the presence or absence of C1 or C8 were
cropped to remove the dark borders and analyzed using
image analysis. RGB images from the z-stack were converted
to three-dimensional chromatic coordinates in the CIELAB
space Lꢄ, aꢄ, and bꢄ using the ImageJ 1.43u software (National
Institutes of Health, Bethesda, MD, USA) [40].
Fenton reaction
The Fenton reaction was performed as has been reported
elsewhere [19]. All the reagent solutions were prepared
immediately before starting the experiment and deaerated
by N2 bubbling. Initially, a solution of FeCl2 (0.32 mM) in
dichloromethane was mixed with a 3 mM spin trap DMSO
solution containing N-tert-butyl-a-phenylnitrone (PBN), and
the reaction was initiated by the addition of 0.5 M H2O2. The
reaction solution (control) was transferred to an EPR capillary
tube, and an EPR spectrum was recorded. The same reaction
was performed using C1 (0.026 mM) dissolved in DMSO and
added to the reaction solution before the addition of H2O2.
Compound C8 was evaluated at a similar molar ratio as C1. All
the EPR measurements were conducted at room temperature
using a Bruker BioSpin’S E-SCAN spectrometer operating at
86 kHz field modulation. EPR spectra were recorded with the
following conditions: an X-band frequency of 9.737 GHz,
3474.556 ꢁ 72 G field center and sweep, 0.872 mW microwave
power, 0.04 s time constant, 1.10 G modulation amplitude and
2 ꢂ 102 receiver gain [35]. EPR spectra were recorded in digital
form (on average, 20 scans were used as a working spectrum).
The number of paramagnetic species contained in the samples
was obtained by the double integration of the EPR signals
using the WINEPR program; the g values were also calculated
using this program.
Statistical analysis
The results are presented as the mean ꢁ SE. For comparison,
several groups were analyzed by one-way analysis of variance
(ANOVA) followed by the Holm–Sidak test. P < 0.05 was
considered statistically significant. All the analyses were
performed using the statistical program Sigma Stat for
Windows software Version 2.03 (SPSS, Inc.), and the graphs
were constructed using the GraphPad Prism software Version
5.00.
This work was supported by grants from Consejo Nacional de
ꢀ
Ciencia y Tecnologıa 254600 and 782 JCB, SIPCOFAA-IPN,
SIP201660204, SIP20161374 and scholarships from Consejo
ꢀ
Nacional de Ciencia y Tecnologıa to study PhD to MEMP.
The authors have declared no conflicts of interest.
References
[1] D. Pitocco, F. Zaccardi, E. Di Stasio, F. Romitelli,
S. A. Santini, C. Zuppi, G. Ghirlanda, Rev. Diabet. Stud.
2010, 7, 15–25.
[2] A. Phaniendra, D. B. Jestadi, L. Periyasamy, Indian J. Clin.
Biochem. 2015, 30, 11–26.
[3] M. Roghani, M. R. N. Jalali, T. Baluchnejadmojarad,
M. R. M. Vaez, G. Naderi, F. D. Roghani, M. T. Joghataei,
Iran J. Pharm. Res. 2013, 12, 377–385.
ROS scavenging by compounds C1 and C8 in
endothelial cells (EC)
Endothelial cells isolated from the aorta of non-diabetic (EC)
and diabetic (ECDR) rats pretreated with streptozotocin
[36, 37] were seeded on plastic cover slips (Nunc 174950
Thermanox CoverslipTM Cell Culture). Both EC types were
isolated as reported elsewhere [38].
The cells were cultured in DMEM supplemented with 5%
FBS and incubated in a 5% CO2 atmosphere at 37°C until
reaching 70% confluence. Then, the cells were incubated for
18 h with C1 or C8 at 100 mM, which is one of the most
frequently evaluated concentrations in various ROS produc-
tion studies. Next, the cells were washed three times with
culture medium free of phenol red, and 10 mM dichloro-
fluorescein diacetate (DCFH-DA) was added to measure the
[4] M. Saboor, Moinuddin, M. Ajmal, S. Ilyas, J. Ayub. Med.
Coll. Abbottabad 2014, 26, 239–243.
ꢀ
[5] A. B. Salmon, A. Richardson, V. I. Perez, Free Radic. Biol.
Med. 2010, 48, 642–655.
[6] S. A. Bukhari, S. A. Naqvi, S. A. Nagra, F. Anjum, S. Javed,
M. Farooq, Pak. J. Pharm. Sci. 2015, 28, 483–491.
[7] C. Karaaslan, S. Suzen, Curr. Top. Med. Chem. 2015, 20,
894–903.
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(10 of 11) e1700041

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2017, 350, e1700041
Aromatic Regions Recognize NADPH Oxidase Inhibitors
Archiv der Pharmazie
[8] G. R. Drummond, C. G. Sobey, Endocrinol. Metab. 2014,
25, 452–463.
[9] H. Mollnau, M. Oelze, E. Zinßius, M. Hausding, Z. Wu,
S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin,
M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi,
R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford,
J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui,
K. Morokuma, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. Cioslowski,
J. V. Ortiz, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, C. Gonzalez, M. Challacombe,
P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong,
J. L. Andres, M. Head-Gordon, E. S. Replogle, J. A. Pople,
Gaussian 98, 1998, Revision A.7, Gaussian Inc.
€
€
M. Knorr, J. K. Ghaemi, S. Kroller-Schon, T. Jansen,
C. Teutsch, C. Foster, H. Li, P. Wenzel, E. Schulz,
€
T. Munzel, A. Daiber, Naunyn Schmiedebergs Arch.
Pharmacol. 2013, 386, 405–419.
€
[10] T. Munzel, T. Gori, R. M. Bruno, S. Taddei, Eur. Heart J.
2010, 131, 2741–2748.
[11] H. M. Ismail, L. Scapozza, U. T. Ruegg, O. M. Dorchies,
PLoS ONE 2014, 9, 1–8.
[12] B. A. Hart, S. Copray, I. Philippens, Biomed. Res. Int. 2014,
2014, 1–6.
[13] D. C. Ellinsworth, J. Pharmacol. Exp. Ther. 2015, 353,
458–464.
[28] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graph. 1996,
14, 33–38.
[14] F. Autore, B. Pagano, A. Fornili, K. Rittinger, F. Fraternali,
Biophys. J. 2010, 99, 3716–3725.
[29] X. Zhu, P. E. Lopes, A. D. MacKerell, Jr., Wiley Interdiscip.
Rev. Comput. Mol. Sci. 2012, 2, 167–185.
ꢀ
[15] K. Elbs, H. Lerch, J. prakt. Chem. 1916, 93, 1–9.
[30] L. Kale, R. Skeel, M. Bhandarkar, R. Brunner, A. Gursoy,
ꢀ
^
[16] C. M. De Faria, A. C. Nazare, M. S. Petronio, L. C. Paracatu,
M. L. Zeraik, L. O. Regasini, D. H. Silva, L. M. da Fonseca,
V. F. Ximenes, Curr. Med. Chem. 2012, 19, 4885–4893.
[17] J. C. Pew, J. Org. Chem. 1963, 28, 1048–1054.
N. Krawetz, J. Phillips, A. Shinozaki, K. Varadarajan,
K. Schulten, J. Comput. Phys. 1999, 151, 283–312.
[31] G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey,
W. E. Hart, R. K. Belew, A. J. Olson, J. Comput. Chem.
1998, 19, 1639–1662.
[32] D. K. Johnson, K. J. Schillinger, D. M. Kwait, C. V. Hughes,
E. J. McNamara, F. Ishmael, R. W. O’Donnell,
M. M. Chang, M. G. Hogg, J. S. Dordick, L. Santhanam,
L. M. Ziegler, J. A. Holland, Endothelium 2002, 9,
191–203.
€
€
ꢀ
[18] J. Hyotylainen, J. Knuutinen, E. Vilen, Chemosphere
1995, 30, 891–906.
[19] N. E. Polyakov, T. V. Leshina, T. A. Konovalova,
L. D. Kispert, Free Radic. Biol. Med. 2001, 31, 398–404.
[20] J. Streeter, W. Thiel, K. Brieger, F. J. Miller, Jr.,
Cardiovasc. Ther. 2013, 31, 125–137.
ꢀ
ꢀ
ꢀ
[21] M. E. Macıas-Perez, F. Martınez-Ramos, I. I. Padilla-
[33] R. Joshi, S. Kumar, M. Unnikrishan, T. Mukherjee, Free
Radic. Res. 2005, 39 1163–1172.
ꢀ
Martınez, J. Correa-Basurto, L. Kispert, J. E. Mendieta-
ꢀ
Wejebe, M. C. Rosales-Hernandez, Biosci. Rep. 2013, 33,
605–616.
[34] H. R. Balaji Raghavendran, A. Sathivel, T. Devaki,
Chemosphere 2005, 61, 276–281.
ꢀ
[22] J. Jiang, H. Kang, X. Song, S. Huang, S. Li, J. Xu, Int. J. Mol.
Sci. 2013, 14, 807–817.
[23] X. Lu, S. Wan, J. Jiang, X. Jiang, W. Yang, P. Yu, L. Xu,
Z. Zhang, G. Zhang, L. Shan, Y. Wang, Eur. J. Med. Chem.
2011, 46, 2691–2698.
[35] I. Spasojevic, Crit. Rev. Clin. Lab. Sci. 2011, 48, 114–142.
[36] A. Tawfik, T. Sanders, K. Kahook, S. Akeel, A. Elmarakby,
M. Al-Shabrawey, Invest. Ophthalmol. Vis. Sci. 2009, 50,
878–884.
[37] K. Serizawa, K. Yogo, K. Aizawa, Y. Tashiro, N. Ishizuka,
Cardiovasc. Diabetol. 2011, 10, 1–10.
[24] http://www.molinspiration.com/
[25] H. Berman, K. Henrick, H. Nakamura, Nat. Struct. Biol.
2003, 10, 1203–980.
[38] E. A. Jaffe, R. L. Nachman, C. G. Becker, C. R. Minick, J.
Clin. Invest. 1973, 52, 2745–2756.
[26] K. Ogura, I. Nobuhisa, S. Yuzawa, R. Takeya, S. Torikai,
K. Saikawa, H. Sumimoto, F. Inagaki, J. Biol. Chem. 2006,
281, 3660–3668.
[39] T. C. Wu, C. Y. Chao, S. J. Lin, J. W. Chen, PLoS ONE 2012,
7, e46067.
ꢀ
ꢀ
[40] M. X. Quintanilla-Carvajal, H. Hernandez-Sanchez,
ꢀ
[27] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski,
J. A. Montgomery, R. E. Stratmann, J. C. Burant,
L. Alamilla-Beltran, G. Zepeda-Vallejo, M. E. Jaramillo-
Flores, M. J. Perea-Flores, A. Jimenez-Aparicio,
ꢀ
ꢀ
G. F. Gutierrez-Lopez, Food Res. Int. 2014, 63, 2–8.
ß 2017 Deutsche Pharmazeutische Gesellschaft
www.archpharm.com
(11 of 11) e1700041