Investigation of HMG-CoA reductase inhibitory and antioxidant effects of various hydroxycoumarin derivatives

Article abstract of DOI:10.1002/ardp.201900378

Cardiovascular diseases are one of the primary causes of deaths worldwide, and the development of atherosclerosis is closely related to hypercholesterolemia. As the reduction of the low-density lipoprotein cholesterol level is critical for treating these

Full text of DOI:10.1002/ardp.201900378

Received: 3 January 2020
Revised: 16 June 2020
Accepted: 19 June 2020
|
|
DOI: 10.1002/ardp.201900378
F U L L P A P E R
Investigation of HMG‐CoA reductase inhibitory and
antioxidant effects of various hydroxycoumarin derivatives
Lalehan Ozalp1 | Özkan Danış1
Cihan Gündüz2 | Ayse Ogan1
| Basak Yuce‐Dursun1 | Serap Demir1
|
1Department of Chemistry, Faculty of Arts and
Sciences, Marmara University, Istanbul, Turkey
Abstract
2Department of Chemistry & Biochemistry,
Manhattan College, White Plains, NY, USA
Cardiovascular diseases are one of the primary causes of deaths worldwide, and the
development of atherosclerosis is closely related to hypercholesterolemia. As the
reduction of the low‐density lipoprotein cholesterol level is critical for treating these
diseases, the inhibition of 3‐hydroxy‐3‐methyl‐glutaryl coenzyme A (HMG‐CoA)
reductase, which is essentially responsible for cholesterol biosynthesis, stands out as
a key solution to lower plasma cholesterol levels. In this study, we synthesized
several dihydroxycoumarins and investigated their antioxidant and in vitro
HMG‐CoA reductase inhibitory effects. Furthermore, we carried out in silico studies
and examined the quantum‐chemical properties of the coumarin derivatives. We
also performed molecular docking experiments and analyzed the binding strength of
each coumarin derivative. Our results revealed that compound IV displayed the
highest HMG‐CoA reductase inhibitory activity (IC50 = 42.0 µM) in vitro. Cupric‐
reducing antioxidant capacity and ferric‐reducing antioxidant power assays de-
monstrated that coumarin derivatives exhibit potent antioxidant activities.
Additionally, a close relationship was found between the lowest unoccupied mole-
cular orbital energy levels and the antioxidant activities.
Correspondence
Özkan Danış, Department of Chemistry,
Faculty of Arts and Sciences, Marmara
University, 34722 Istanbul, Turkey.
Email: odanis@marmara.edu.tr
Funding information
Marmara Üniversitesi, Grant/Award Numbers:
FEN‐A‐110411‐0103, Fen‐D‐220513‐0222
K E Y W O R D S
antioxidant, coumarin, HMG‐CoA reductase, lipid lowering, radical scavenging
|
1
INTRODUCTION
may also be involved in the formation and development of AS. In the case
of oxidative stress, vascular wall cells produce excessive reactive oxygen
Cardiovascular diseases are the major cause of deaths worldwide, and
they are expected to remain a leading cause of mortality in the near
future.[1] The majority of coronary artery diseases leading to myocardial
infarction are caused due to atherosclerosis (AS), which is mainly char-
acterized by the deposition of lipids in the artery wall and the infiltration
of inflammatory cells.[2] Hypercholesterolemia and other dyslipidemias
are major factors for the development of AS, and the reduction of
low‐density lipoprotein (LDL) cholesterol level is the primary target of
therapy. For this reason, the inhibition of 3‐hydroxy‐3‐methyl‐glutaryl
coenzyme A (HMG‐CoA) reductase (HMGR), which is the key enzyme for
cholesterol biosynthesis, is crucial for lowering plasma cholesterol
levels.[3] Besides the deposition of lipids in the artery wall, oxidative stress
species (ROS), which can cause peroxidation of lipids and irreversible
damage of cell membranes, proteins, and DNA.[2]
Statins have been proved to be remarkably successful HMGR
inhibitors in clinical use. They bear HMG‐like moieties and bind
covalently and competitively to the catalytic site of HMGR.[4] Statins
are also known to possess anti‐inflammatory and antioxidant effects.
Besides, they reduce oxidized LDLs and inhibit their uptake by
macrophages.
Despite all the abovementioned benefits, statins are also known
to display a number of side effects. The most common adverse side
effects are elevated liver enzymes and myalgia, which is a muscle
ache and can easily be more severe with higher doses, even leading
|
Arch Pharm. 2020;e1900378.
wileyonlinelibrary.com/journal/ardp © 2020 Deutsche Pharmazeutische Gesellschaft
1 of 9
https://doi.org/10.1002/ardp.201900378

2 of 9
OZALP ET AL.
|
to rhabdomyolysis. Therefore, the design of HMGR inhibitors with
fewer side effects is receiving increasing attention. Coumarin and its
derivatives that were shown to have a lipid‐lowering potential[5] have
attracted intense interest due to their diverse pharmacological
properties with low toxicity.[6]
statins, exhibit intrinsic antioxidant activities with both antihydroxyl
and peroxyl radical activity.[10,11] The suggestion that some coumarin
derivatives with the antioxidant activity would be an inhibitor of
HMGR on the basis of structural similarities with statins attracted
our attention. These similarities depend on the presence of two
phenyl rings and the presence of hydroxyl groups in different posi-
tions on these rings. In our initial study, we reported that compound
IV [7,8‐dihydroxy‐3‐(4‐methylphenyl)coumarin] has a considerable
lipid‐lowering and antioxidant activity,[5] which suggests that syn-
thetic ortho‐6,7‐dihydroxy‐4‐methyl coumarins might be able to
scavenge ROS and modulate HMGR activity.
Human HMGR comprises 888 amino acids, making up to three
domains. The catalytic domain (residues 459–888) of HMGR is a
tetrameric structure and the active sites are located at the hydro-
phobic interface of the two monomers of a dimer (Figure 1).[7] HMGR
accommodates an N‐terminal transmembrane domain (residues
1–340) that is bound to the endoplasmic reticulum, and it is usually
not incorporated in the X‐ray crystal structures. Statins occupy the
area where the HMG portion of HMG‐CoA binds and do not spread
to the proximity of NADPH.[8] They interact with both monomers of a
dimer.[9]
|
2.1
Antioxidant activity assays
In this study, we have investigated antioxidant and in vitro
HMGR inhibitory effects of dihydroxycoumarins. Previously, our
group has reported the synthesis and in vivo lipid‐lowering activity of
7,8‐dihydroxy‐3‐(4‐methylphenyl)coumarin in rats.[5] Therefore, to
evaluate the possible role of HMGR inhibition in the lipid‐lowering
activity, we have synthesized structurally analogous dihydrox-
ycoumarin derivatives and investigated their antioxidant and HMGR
inhibitory effects in vitro. Later, we pursued in silico studies to gain
insight into the varying inhibition intensities. Toward this end, we
calculated several quantum mechanical (QM) properties of the di-
hydroxycoumarin derivatives and pravastatin, a known statin in-
hibitor of HMGR. Subsequently, molecular docking simulations were
performed. Selected hits were then analyzed and their binding en-
ergies were evaluated.
Table 1 shows the antioxidant activities of coumarins obtained from
DPPH (1,1‐diphenyl‐2‐picrylhydrazyl), CUPRAC (cupric‐reducing an-
tioxidant capacity), FRAP (ferric‐reducing antioxidant power), and
metal chelating assays. Among the five compounds examined, com-
pound II exhibited the strongest efficiency, followed by I, III, and IV,
whereas compound V rendered the weakest effect. As seen in
Table 1, TEACFRAP and TEACCUPRAC values of all compounds de-
creased in the following order: II > I > III > IV > V. Ferrous ion che-
lating activity of coumarins is expressed as EC50 (mM), and
compounds I, II, III, and IV displayed nearly the same metal chelating
activity when compared with standard antioxidants, but compound V
did not show any activity.
Coumarins recognized as possessing a potent antioxidant activity
are also strong scavengers of DPPH, a relatively stable nitrogen‐
centered free radical.[12] Our results demonstrated that the free ra-
dical scavenging activity of the coumarin compounds is concentration
dependent. As the concentration of the test compounds increases,
the radical scavenging activity increases, and a lower EC50 value
reflects a better protective action. The antioxidant activity of these
coumarin derivatives could be attributed to the electron‐donating
|
2
RESULTS AND DISCUSSION
Human HMGR, the rate‐limiting enzyme in cholesterol biosynthesis,
is tightly related to the ROS content. Inhibitors of HMGR, known as
FIGURE 1 (a) The homotetrameric
structure of human 3‐hydroxy‐3‐methyl‐
glutaryl coenzyme A (HMG‐CoA) reductase
(HMGR) enzyme. (b) Dimer of HMGR,
including chains A and B. One monomer is
colored for clarity. The binding region of
HMG‐CoA is mainly on one monomer,
neighboring the NADPH‐binding region of
another monomer. Image obtained by using
Discovery Studio 4.5

3 of 9
OZALP ET AL.
|
TABLE 1 Radical scavenging and reducing power of tested coumarin derivatives
Compound
DPPHa
FRAPb
CUPRACb
Metal chelatingc
0.782 0.011
0.728 0.016
0.734 0.027
0.746 0.009
2.702 0.101
0.905 0.089
0.984 0.057
0.963 0.007
1.191 0.039
I
74.70 0.057
64.27 0.109
92.64 0.841
94.85 0.917
6,604.92 2.157
7.24 0.089
2.28 0.091
3.61 0.107
2.06 0.058
2.00 0.101
1.22 0.055
1.16 0.017
1.22 0.056
0.39 0.008
1.00 0.051
2.44 0.016
2.82 0.108
2.25 0.139
2.24 0.121
1.18 0.182
1.04 0.089
1.43 .009
1.04 0.057
1.00 0.037
II
III
IV
V
Ascorbic acid
BHA
331.42 1.899
1,077.10 2.114
93.19 1.089
BHT
Trolox
Note: Results are given as the mean value standard deviation (n = 3); (Tukey's test, p < .05).
Abbreviations: BHA, butylated hydroxyanisole; BHT, butylated hydroxytoluene; CUPRAC, cupric‐reducing antioxidant capacity; DPPH, 1,1‐diphenyl‐2‐
picrylhydrazyl; FRAP, ferric‐reducing antioxidant power.
aAntioxidant activity was measured by using the DPPH radical assay and data are expressed as EC50 (μM).
bFRAP and CUPRAC results were expressed as trolox equivalent antioxidant capacity (mM).
cThe metal chelating activity was determined by inhibiting ferrous–ferrozine complex formation and data are expressed as EC50 (mM).
nature of substituents like –OH, –CH3, and also C═C on a coumarin
scaffold, as they play a key role in reducing free radicals such as
DPPH, which can consequently prevent oxidative damage to cells.
Although compound V has a similar structure with other coumarins,
it displayed the lowest DPPH activity. This low activity may be at-
tributed to the para position of 5‐OH and 4‐phenyl groups instead of
ortho position when compared with other compounds. As shown in
Figure 2, the results of the time‐dependent free radical scavenging
activities of the studied compounds and the standard antioxidant
trolox have showed that all coumarin compounds have a very fast
initial reaction with DPPH when compared with Trolox.
Coumarins possess antioxidant activities probably due to their
structural analogy with flavonoids, and it has been reported that they
could bind to transition metal ions.[13] The abilities of the coumarin
FIGURE 2 DPPH (1,1‐diphenyl‐2‐picrylhydrazyl) scavenging capacity (% inhibition) of coumarin derivatives as a function of the reaction time (s)

4 of 9
OZALP ET AL.
|
compounds to reduce ferric ions (FRAP), cupric ions (CUPRAC), and
their ferrous ion chelating activities were compared with that of
Trolox, butylated hydroxyanisole (BHA), butylated hydroxytoluene
(BHT), and ascorbic acid, and it was found that coumarin derivatives,
except compound V, have more potent antioxidant activities than the
standard antioxidants that were used in the study. The FRAP and
CUPRAC assays offer accepted methods for the evaluation of the
antioxidant activity. It is clear from the results that the CUPRAC and
FRAP activities of compound II are the highest among all the species
studied. The main strategy to avoid ROS generation that is associated
with redox active metal catalysis involves chelating of the metal ions,
thus inhibiting hydroxyl radical and hydrogen peroxide formation
produced by Fenton's reactions. Our results revealed that compound
II also exhibits a mild chelating activity in vitro; therefore, it could
have therapeutic potential in the treatment of oxidative stress‐
related diseases.
According to our results, compound IV displayed the highest
HMGR inhibitory activity (IC50 = 42.0 µM) when compared with
other derivatives, whereas compound V shows the lowest activity
(Figure 3). However, the inhibitory activities of the studied coumarin
derivatives on HMGR are lower than that of standard inhibitor,
pravastatin (IC50 = 6.9 µM). In our previous study, we investigated in
vivo antioxidative and lipid‐lowering effects of 7,8‐dihydroxy‐3‐(4‐
methylphenyl)coumarin (compound IV) in hyperlipidemic rats and
reported that this compound displayed potent antioxidant and lipid‐
lowering effects.[5] Our results suggest that the apparent in vivo lipid‐
lowering activities of coumarin derivatives could be caused by the
inhibition of the HMGR activity.
|
2.3
QM descriptors
QM studies were carried out to derive theoretical data. Electron
densities of frontier molecular orbitals provide insights into
donor–acceptor interactions that play a key role in several pharma-
cological mechanisms.[15] In addition to this, the frontier orbitals have
a strong correlation with the antioxidative activity.[16] The energy of
the highest occupied molecular orbitals (HOMO) is directly propor-
tional to the electron‐donating ability of a compound.[17,18] Figure S1
shows that electron‐donating groups increase the HOMO energy,
whereas electron‐withdrawing groups (EWGs) such as –NO2, as in
compound II, reduces the HOMO energy, as well as the lowest un-
occupied molecular orbital (LUMO) energy. It was also observed that
HOMO orbitals are localized in the coumarin ring, whereas LUMO
orbitals are delocalized in the whole molecule.
|
2.2
In vitro HMGR inhibitory activity
Human HMGR, the rate‐limiting enzyme in cholesterol biosynthesis,
is transmembrane glycoprotein that catalyzes the NADPH‐
a
dependent four‐electron reduction of HMG‐CoA to CoA and meva-
lonate. Inhibitors of HMGR, such as statins, are effective agents used
to reduce serum cholesterol levels and prevent coronary artery dis-
eases.[5] For known pitavastatin analogs, a phenyl group is connected
at position 4 in the quinoline, and it is located in the ortho position to
the pharmacophore group at position 3. This substitution pattern is
also found in other artificial HMGR inhibitors such as fluvastatin,
atorvastatin, and rosuvastatin, which belong to the indole, pyrrole,
and pyrimidine derivatives, respectively.[14]
The energy of the HOMO orbitals is proportional to the ten-
dency of the molecule toward an electrophile attack. In contrast, the
FIGURE 3 HMG‐CoA (3‐hydroxy‐3‐methyl‐glutaryl coenzyme A) reductase (HMGR) inhibitory activities (%) and IC50 values of the
coumarins investigated

5 of 9
OZALP ET AL.
|
TABLE 2 Energies of HOMO and LUMO orbitals calculated by the
M062X method
This shows that EWG renders the compound an easier target for an
electron transfer, due to their lowering effect on LUMO orbital en-
ergies (Table 2). In contrast, compounds III and IV bear the same
electron‐donating group at adjacent positions. However, due to this
difference in position, the molecule has a slight change in HOMO
energy in both gas and water phase. There is an increase in the
HOMO energy of the methyl group at 4‐position (IV), compared with
one that bears the same group at 3‐position (III). Compound V, with
the lowest antioxidant and inhibitory activity, has OH– substituents
at the 5‐ and 7‐positions, unlike others that have substituents at the
7‐ and 8‐positions and also bear the phenyl ring at the 4‐position
instead of the 3‐position. This provided the molecule with more lo-
calized HOMO orbitals, located in the coumarin scaffold. Figure S1
displays that LUMO orbitals are more delocalized in compound II.
This is in accordance with the order of LUMO energies (Table 2). This
arises from the fact that EWGs lower the energy of LUMO orbitals,
rendering them more susceptible to a nucleophilic attack.
Frontier
orbital
Energy in gas
phase (eV)
Energy in water
phase (eV)
Coumarin
I
HOMO
LUMO
−7.073
−1.004
−6.813
−0.744
II
HOMO
LUMO
−7.481
−1.726
−6.978
−1.463
III
IV
V
HOMO
LUMO
−7.036
−0.966
−6.784
−0.713
HOMO
LUMO
−6.973
−0.955
−6.733
−0.705
HOMO
LUMO
−7.365
−0.885
−7.157
−0.713
Abbreviations: HOMO, highest occupied molecular orbital; LUMO, lowest
unoccupied molecular orbital.
|
Docking scores
energy of the LUMO levels reflects the susceptibility of the molecule
to a nucleophilic attack.[15] We analyzed HOMO and LUMO energies
and evaluated correlations between inhibition capacity and anti-
oxidant activity of the coumarin derivatives. According to Table 2,
HOMO energy values of coumarin derivatives that are obtained from
M062X calculations in the water phase are strongly correlated with
their inhibitory activities. The HOMO energies are ordered as
IV > III > I > II > V, which is in accordance with the experimental data.
However, further QM research is required to illuminate the inhibition
mechanism that involves possibly multiple reactions in the active site
of the HMGR enzyme.[19]
2.4
Molecular docking studies were carried out to see the affinities of
coumarin derivatives toward HMGR. Scores are given in Table 3.
According to the docking scores, compound II has the highest
affinity to the HMGR enzyme. This may be attributed to more de-
localized LUMO orbitals on the coumarin ring of compound II (Figure
S1), which enables a nucleophilic attack, particularly on the coumarin
ring. Furthermore, being the only coumarin that bears an EWG on its
phenyl ring, compound II has the highest electron density on its
phenyl ring. Additional electrostatic interactions that are formed
with II contribute to strong affinity of II toward HMGR, as the cat-
alytic site of HMGR has a positive electrostatic potential.[15] Pra-
vastatin, a known good inhibitor of HMGR, follows compound II in
binding affinity. Compound IV, of which lipid‐lowering activity on rats
has been validated earlier, comes after compound II.[5]
Furthermore, according to the DPPH assay, there is a strong
correlation between LUMO energy levels and the antioxidant activity
of studied compounds. Compound II, which was found to have the
highest antioxidant activity, also has the lowest LUMO energy level,
both in gas and water phase. The low LUMO energy level increases
the susceptibility of the molecule to a nucleophilic attack, therefore
its potential to reduce DPPH, FRAP, and CUPRAC.
Among 3‐phenyl‐substituted coumarins, compound II stands out
as the only compound that bears an EWG on its phenyl ring. Com-
pound I does not possess any substituent on the phenyl ring.
TABLE 3 Docking scores of coumarin derivatives and pravastatin
Lowest binding
Mean binding
Compounds
energy (kcal/mol)
energy (kcal/mol)
I
−7.36
−9.52
−7.4
−7.31
−8.34
−7.28
−7.36
−5.09
−7.81
II
III
FIGURE 4 Coumarin II bound to positively charged pocket in
HMGR (3‐hydroxy‐3‐methyl‐glutaryl coenzyme A reductase).
Surface visualization was performed using the APBS (Adaptive
Poisson–Boltzmann Solver) Electrostatics plugin for PyMOL
(http://www.pymol.org)
IV
−7.39
−6.45
−8.15
V
Pravastatin

6 of 9
OZALP ET AL.
|
FIGURE 5 The binding mode of compound
II in the active site of HMGR (3‐hydroxy‐3‐
methyl‐glutaryl coenzyme A reductase). Image
obtained by Discovery Studio 4.5
TABLE 4 ADME properties concerning
ADME properties
Pravastatin
I
II
III
IV
V
Lipinski's rule of five
Molecular weight (Da)
The number of H bond acceptors
The number of H bond donors
Log P
424.53
254.24 299.24 268
268.26 254.24
7
4
6
4
2
4
2
4
4
2
2
2
2.34
2.5
1.85
2.82 2.82
2.43
Abbreviation: ADME, absorption, distribution, metabolism, and excretion.
|
2.5
Binding site of coumarins and electrostatic
observed with Arg590 (3.75 Å). Alkyl–π interactions were observed
with Leu853 and NADPH (4.44, 5.27 Å). His866 contributes to π–π
stacked interactions, with distances of 3, 3.81, and 4.87 Å.
potential
The coumarin scaffolds of the studied compounds are expected to be rich
in electron density, as they bear electron‐donating groups such as –OH.
This is supported by the localized HOMO orbitals (Figure S1). Con-
sidering the positive electrostatic potential in the catalytic site in HMGR
(Figure 4), it can be understood that these electron‐rich compounds'
binding is favored in the pocket.
|
2.6
ADME (absorption, distribution, metabolism,
and excretion) properties
For all coumarin derivatives and pravastatin, Lipinski's rule of five para-
meters, which are important to evaluate the druglikeness of drug lead
compounds, was applied using the web server (http://www.swissadme.
ch/).[20] Briefly, the rule states that the molecular weight of an oral drug
should be lower than 500 Da, the number of hydrogen bond acceptors
should be at most 10, the number of hydrogen bond donors should be at
most 5, and log P should be lower than 5. As seen in Table 4, none of the
coumarin derivatives violate the “Rule of Five” parameters.[21]
Figure 5 shows the binding site of II in the active site. Hydrogen
bonds and π electron‐originated interactions are found to be domi-
nant in the stabilization of II. Hydrogen bonds are formed with
Asn686, Cys688, Asp690, Lys691, and Lys735 with distances of 2.72,
1.76, 2.23, 1.61, and 2.47 Å, respectively. Cation–π interactions were
TABLE 5 Toxicity prediction of synthesized compounds and
pravastatin
Pravastatin
I
II
III
IV
V
|
2.7
Toxicity
LD50 (mg/kg)
8,939
83
350 600 350 250 3,200
The C‐3 and C‐4 epoxidation reaction that unsubstituted coumarins
undergo in the liver has been shown to cause toxicity. CYP2A6 cy-
tochrome P‐450 oxidases generate this reaction, followed by a series
Hepatotoxicity (%)
Cytotoxicity (%)
71
96
61
71
71
95
71
95
68
86
61

7 of 9
OZALP ET AL.
|
SCHEME 1 The scheme of synthesis and chemical structures of the coumarin derivatives
|
EXPERIMENTAL
of reactions that finally give stable complexes with heavy metals,
which are thus toxic to the liver.[22] It was shown that 3,4‐double
bond has a strong correlation with coumarin toxicity.[23] Ergo,
coumarins with cyclic substituents at position 3 are promising, as
they are likely to be resistant to C‐3 and C‐4 epoxidation. Table 5
shows toxicity percentages obtained from web server, http://tox.
charite.de/.[24]
4
|
4.1
Materials
HMG‐CoA Reductase Assay Kit (#CS1090), Trolox, BHT, and BHA
were purchased from Sigma‐Aldrich Co., St. Louis, MO. Neocuproine
was purchased from Merck KGaA, Darmstadt, Germany. All other
chemicals used were of analytical grade. For quantum‐chemical
analysis, Gaussian 09[25] software was used. AutoDock4[26] was used
for docking experiments. Discovery Studio 4.5[27] was used for
visualization.
|
3
CONCLUSION
We investigated the inhibitory activities of several coumarin deriva-
tives against the HMGR enzyme. HMGR is known to be tightly related
to the ROS content. Therefore, we also analyzed antioxidant activities
of these compounds. The results of in vitro experiments showed that
compound IV showed the highest inhibitory activity, whereas com-
pound II showed the highest antioxidant activity. Docking studies re-
vealed that compound II has also the strongest affinity toward HMGR.
The results are in accordance with common pharmacophore features
of the coumarin derivatives. The presence of an EWG, –NO2, on the
phenyl ring has rendered the compound II to be a good/valuable an-
tioxidant. Compounds III and IV show higher inhibition activities than
I, supported by relatively higher binding affinities, which is related to
the presence of –CH3 group instead of –H, providing more interactions
with the cavity. The position of phenyl ring at position 4 of compound
V was found to be responsible for poor inhibition activity, as it yielded
an arrangement that is not favorable for binding to the catalytic site.
Previously, our group has shown that compound IV might be a drug
lead for its in vivo lipid‐lowering effects.[5] The findings of this study
suggest that compound II also might have a potential to be a drug lead
for oxidative stress‐related and atherosclerotic diseases; thus, further
in vivo studies are recommended to evaluate its lipid‐lowering effects.
The InChI keys of the investigated compounds, together with
some biological activity data, are provided as Supporting Informa-
tion Data.
|
4.2
Methods
|
4.2.1
Chemistry
3‐Phenylcoumarin derivatives (I, II, III, and IV) were prepared by the
reaction of substituted hydroxybenzaldehydes with the correspond-
ing arylacetic acids under the traditional Perkin conditions.[28–30]
4‐Phenylcoumarin derivative (V) was obtained by condensation of
selected phenol with ethyl‐3,4‐dimethoxybenzoylacetate according
to the Pechmann reaction.[13] Scheme 1 shows the scheme of
synthesis and chemical structures of the coumarin derivatives. To the
best of the authors' knowledge, compound III has only been designed
under a patent (WO 9424119, 1994) without proper characteriza-
tion. Therefore, compound III have been characterized with 1H and
13C nuclear magnetic resonance, and the spectra have been given in
Figure S2.

8 of 9
OZALP ET AL.
|
|
4.2.2
DPPH radical scavenging activity
in 500 μl methanol were added to a solution of 2 mM FeCl2 (500 μl).
The reaction was initiated by the addition of 5 mM ferrozine (200 μl)
in methanol. Then, the mixture was shaken vigorously at room
temperature for 10 min. Absorbance of the solution was then mea-
sured spectrophotometrically at 562 nm. The percentage inhibition
of ferrozine–Fe2+ complex formation was calculated as [(A0 – As)/
As] × 100, where A0 is the absorbance of the control and As is the
absorbance of the compound/standard. The ferrous ion chelating
effects of the compounds were expressed in terms of EC50 (con-
centration required for chelating 50% of ferrous ions).
The free radical scavenging activities were measured by DPPH using
the method of Blois.[31] Briefly, 0.1‐mM solution of DPPH in metha-
nol was prepared, and 1 ml of this solution was added to 250 μl of
coumarins in methanol containing 5 μl dimethyl sulfoxide (DMSO) at
different concentrations (10–100 μg/ml). The mixture was shaken
vigorously and allowed to be kept at room temperature for 30 min.
Later, the absorbance was spectrophotometrically measured at
517 nm. The radical scavenging activities of the samples were ex-
pressed in terms of EC50 (concentration required for a 50% decrease
in absorbance of DPPH radical). To determine the rate of the reac-
tion, assay was repeated for coumarin derivatives and Trolox at
50 μM final concentration, and the decrease of absorbance was
monitored every 2 s for 90 s at 517 nm using the Helios Zeta UV–Vis
spectrophotometer (Thermo Fisher Scientific, Waltham, MA).
|
4.2.6
HMGR inhibition assay
Coumarin derivatives were screened for their ability to inhibit the
catalytic activity of human recombinant HMGR in vitro. The assay is
based on the spectrophotometric measurement of the decrease in
absorbance at 340 nm, which represents the oxidation of NADPH by
the catalytic subunit of HMGR in the presence of the substrate,
HMG‐CoA. The HMGR assay was performed using the manu-
facturer's protocol. Briefly, about 6 µg of the enzyme was incubated
at 37°C with 400 µM NADPH, 0.3 mg/ml HMG‐CoA, and different
concentrations (10–25–50 μM) of coumarin derivatives in a 96‐well
plate, and their absorbance change was monitored at 37°C using
Epoch microplate spectrophotometer (BioTek, Winooski, VT). The
concentration of an inhibitor required to inhibit 50% of the HMGR
under the assay conditions was defined as IC50. Pravastatin was
utilized as a control for positive inhibition.
|
4.2.3
FRAP assay
Ferric ion (Fe3+) reducing power was evaluated according to the
protocol of Oyaizu[32] with adjustments. BHT, BHA, and Trolox were
used as positive controls. 100 μl of samples (dissolved in phosphate
buffer 50 μM, pH 6.6, containing 5 μl DMSO) were mixed with
phosphate buffer (200 μl, 50 μM, pH 6.6) and K3Fe(CN)6 (200 μl, 1%).
The mixture was incubated at 50°C for 20 min. A portion (250 μl) of
trichloroacetic acid (10%) was added to the mixture, which was then
centrifuged at 10,000g for 10 min. The upper layer of the solution
(700 μl) was mixed with FeCl3 (150 μl, 0.1%) and the absorbance was
measured at 700 nm. BHT, BHA, and Trolox were used as positive
controls. Results were expressed as the Trolox equivalent antioxidant
capacity (TEAC), which is defined as the millimolar concentration of
a Trolox solution having the antioxidant capacity equivalent to a
1.0‐mM solution of the substance under investigation.
|
4.2.7
Statistical analysis
All statistical analyses were performed using GraphPad Prism 6.0
(GraphPad Software, San Diego, CA) software in triplicate, and the
difference of parameters was statistically tested with unpaired Stu-
dent's t‐test. The level of significance was defined as p < .05.
|
4.2.4
CUPRAC assay
CUPRAC was determined as described by Apak et al.[33] with slight
changes. Briefly, 800 μl of coumarin derivatives were dissolved in
NH4Ac buffer, pH 7, containing 5 μl DMSO, and mixed with 160 μl
NH4Ac buffer, pH 7, 160 μl CuCl2 (10 mM), and 160 μl neocuproine
(7.5 mM). After 30 min, absorbance was measured at 450 nm. Results
were expressed as the TEAC, which is defined as the millimolar con-
centration of a Trolox solution having the antioxidant capacity
equivalent to a 1.0‐mM solution of the compound under investigation.
4.2.8
QM Calculations
|
Pravastatin and coumarin derivatives (compounds I, II, III, IV, and V)
were optimized with M062X functional and 6–31 G(d,p) basis set by
using Gaussian 09.[25] Temperature was set to 310 K. HOMO–LUMO
orbitals were produced and the energies of HOMO and LUMO levels
were calculated in gas and water phase.
|
4.2.9
Minimization of HMGR enzyme and docking
|
4.2.5
Metal chelation
calculations
Ferrous ion (Fe2+) chelating activity was determined by inhibiting
ferrous–ferrozine complex formation after treatment of test material
with ferrous ion (Fe2+). 20–100 μg/ml concentrations of compounds
The crystal structure of HMGR was obtained from the Brookhaven
Protein Data Bank (PDB ID: 1DQA). To eliminate any steric clashes in
the crystal structure, the dimer was first subjected to a minimization

9 of 9
OZALP ET AL.
|
procedure by using the steepest descent algorithm with GROMACS
5.1.2 package.[34] Minimization was complete as the maximum force
of 10 kJ/mol was reached. Water, CoA, and HMG were removed from
the minimized structure for molecular docking simulations. NADPH
was kept intact in the protein, as statins are not known to be in-
terfering with NADPH binding.[35] Before docking, HMG was re-
moved and redocked into the enzyme by using AutoDock4.[26] A root
mean square deviation of 0.941 Å confirmed the reliability of the
docking protocol. Optimized structures of compounds I, II, III, IV, V,
and pravastatin were docked into the dimeric structure. The calcu-
[11] R. Kleemann, T. Kooistra, Cardiovasc. Hematol. Disord.: Drug Targets
2005, 5, 441.
[12] M. Patel, J. Patel, J. Adv. Pharm. Educ. Res. 2011, 1, 52.
[13] O. Danis, S. Demir, C. Gunduz, Res. Chem. Intermed. 2016, 42, 6061.
[14] Z. Cai, W. Zhou, L. Sun, Bioorg. Med. Chem. 2007, 15, 7809.
[15] F. M. Atlam, M. K. Awad, E. A. El‐Bastawissy, J. Mol. Struct. 2014,
1075, 311.
[16] Y.‐Z. Zheng, D.‐F. Chen, G. Deng, R. Guo, Molecules 2018, 23, 1989.
[17] Y. Xue, Y. Zheng, L. An, Y. Dou, Y. Liu, Food Chem. 2014, 151, 198.
[18] G. Wang, Y. Xue, L. An, Y. Zheng, Y. Dou, L. Zhang, Y. Liu, Food Chem.
2015, 171, 89.
[19] B. E. Haines, O. Wiest, C. V. Stauffacher, Acc. Chem. Res. 2013, 46,
2416.
lations were confined within
a grid box with dimensions of
[20] A. Daina, O. Michielin, V. Zoete, Sci. Rep. 2017, 7, 42717.
[21] C. A. Lipinski, F. Lombardo, B. W. Dominiy, P. J. Feeney, Adv. Drug
Deliv. Rev. 1997, 23, 3.
60 × 60 × 60 Å and centered at a sulfur atom of CoA, which was large
enough to include the binding site of HMG‐CoA. The Lamarckian
genetic algorithm[36] was employed, which is implemented in Auto-
Dock4. The maximum number of energy evaluations was set to
2,500,000 and 100 runs were performed for each ligand. All calcu-
lations were performed on Linux platform. Hits with the lowest Gibbs
binding energy (ΔG) of each compound were analyzed.
[22] V. D. Kancheva, L. Saso, P. V. Boranova, A. Khan, M. K. Saroj, M. K.
Pandey, S. Malhotra, J. Z. Nechev, S. K. Sharma, A. K. Prasad, M. B.
Georgieva, C. Joseph, A. L. DePass, R. C. Rastogi, V. S. Parmar, Bio-
chimie 2010, 92, 1089.
[23] J. H. Fentem, A. H. Hammond, M. J. Garle, J. R. Fry, Toxicol. In Vitro
1992, 6, 21.
[24] P. Banerjee, R. Preissner, Pharmazeutische Medizin 2019, 21, 02.
[25] M. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A.
Petersson, Gaussian 09, Revision A., Gaussian Inc., Wallingford,
CT 2009.
ACKNOWLEDGMENTS
This study was supported by Marmara University, Commission of
Scientific Research Project FEN‐A‐110411‐0103 and FEN‐D‐
220513‐0222. L. O. acknowledges the Council of Higher Education
(YÖK) of Turkey for the Ph.D. scholarship of 100/2000 program.
[26] G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner, R. K. Belew, D. S.
Goodsell, J. Comput. Chem. 2009, 30, 2785.
[27] D. Biovia, Discovery Studio Modeling Environment 2015, Dassault
Systems, version: 4.5 San Diego.
CONFLICT OF INTERESTS
[28] M. Bulut, Ç. Erk, J. Heterocycl. Chem. 2001, 38, 1291.
[29] Ü. Salan, M. Bulut, Heterocycles 2006, 68, 237.
[30] O. Danis, B. Yuce‐Dursun, C. Gunduz, A. Ogan, G. Sener, M. Bulut, A.
Yarat, Arzneimittelforschung 2010, 60, 617.
The authors declare that there are no conflicts of interests.
ORCID
[31] M. S. Blois, Nature 1958, 181, 1199.
Özkan Danış
http://orcid.org/0000-0003-1781-0520
[32] M. Oyaizu, Jpn. J. Nutr. Diet. 1986, 44, 307.
[33] R. Apak, K. Guclu, M. Ozyurek, S. E. Karademir, J. Agric. Food Chem.
2004, 52, 7970.
REFERENCES
[34] M. J. Abraham, T. Murtola, R. Schulz, S. Páll, J. C. Smith, B. Hess, E.
Lindahl, SoftwareX 2015, 1–2, 19.
[1] T. Inoue, M. Hayashi, K. Takayanagi, S. Morooka, Atherosclerosis 2002,
160, 369.
[35] E. S. Istvan, J. Deisenhofer, Science 2001, 292, 1160.
[36] 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.
[2] U. Indumathi, K. Kanchana, P. Sachdanandam, Biomed. Prev. Nutr.
2013, 3, 313.
[3] M. Graaf, D. Richel, C. van Noorden, H. Guchelaar, Cancer Treat. Rev.
2004, 30, 609.
[4] M. Son, A. Baek, S. Sakkiah, C. Park, S. John, K. W. Lee, J. Neurol. Sci.
2013, 8, 1.
SUPPORTING INFORMATION
Additional supporting information may be found online in the
Supporting Information section.
[5] B. Yuce, O. Danis, A. Ogan, G. Sener, M. Bulut, A. Yarat, Drug Res.
2009, 59, 129.
[6] F. Belluti, G. Fontana, L. Dal Bo, N. Carenini, C. Giommarelli, F.
Zunino, Bioorg. Med. Chem. 2010, 18, 3543.
[7] E. S. Istvan, M. Palnitkar, S. K. Buchanan, J. Deisenhofer, EMBO J.
2000, 19, 819.
How to cite this article: Ozalp L, Danış Ö, Yuce‐Dursun B,
Demir S, Gündüz C, Ogan A. Investigation of HMG‐CoA
reductase inhibitory and antioxidant effects of various
hydroxycoumarin derivatives. Arch Pharm. 2020;e1900378.
https://doi.org/10.1002/ardp.201900378
[8] Textbook of Drug Design and Discovery, 4th ed. (Eds: K. Stromgaard, P.
Krogsgaard‐Larsen, U. Madsen), CRC Press, Boca Raton 2009.
[9] G. A. Holdgate, W. H. J. Ward, F. McTaggart, Biochem. Soc. Trans.
2003, 31, 528.
[10] S. Colle, C. Taillefumier, Y. Chapleur, R. Liebl, A. Schmidt, Bioorg. Med.
Chem. 1999, 7, 1049.