Enzymatic studies of isoflavonoids as selective and potent inhibitors of human leukocyte 5-lipo-oxygenase

Article abstract of DOI:10.1111/cbdd.12469

Continuing our search to find more potent and selective 5-LOX inhibitors, we present now the enzymatic evaluation of seventeen isoflavones (IR) and nine isoflavans (HIR), and their in vitro and in cellulo potency against human leukocyte 5-LOX. Of the 26 compounds tested, 10 isoflavones and 9 isoflavans possessed micromolar potency, but only three were selective against 5-LOX (IR-2, HIR-303, and HIR-309), with IC₅₀ values at least 10 times lower than those of 12-LOX, 15-LOX-1, and 15-LOX-2. Of these three, IR-2 (6,7-dihydroxy-4-methoxy-isoflavone, known as texasin) was the most selective 5-LOX inhibitor, with over 80-fold potency difference compared with other isozymes; Steered Molecular Dynamics (SMD) studies supported these findings. The presence of the catechol group on ring A (6,7-dihydroxy versus 7,8-dihydroxy) correlated with their biological activity, but the reduction of ring C, converting the isoflavones to isoflavans, and the substituent positions on ring B did not affect their potency against 5-LOX. Two of the most potent/selective inhibitors (HIR-303 and HIR-309) were reductive inhibitors and were potent against 5-LOX in human whole blood, indicating that isoflavans can be potent and selective inhibitors against human leukocyte 5-LOX in vitro and in cellulo. Of the 26 compounds tested, 10 isoflavones and 9 isoflavans possessed micromolar potency, but only three were selective against 5-LOX (IR-2, HIR-303, and HIR-309), with IC₅₀ values at least 10 times lower than those of 12-LOX, 15-LOX-1, and 15-LOX-2. Docking and steered molecular dynamics were performed to determinate the structure-activity relationship.

Full text of DOI:10.1111/cbdd.12469

Chem Biol Drug Des 2014
Research Article
Enzymatic Studies of Isoflavonoids as Selective
and Potent Inhibitors of Human Leukocyte
5-Lipo-Oxygenase
Carolina Mascayano1,*, Victoria Espinosa2, Silvia
In response to extracellular signals, arachidonic acid (AA)
2
3
3
is released from the cell membrane producing an intracel-
lular cascade of events, including the activation of lipo-ox-
ygenases (LOX) and cycloxygenases (COX) (1,2), whose
products generate pain, fever, and inflammation (3,4). Two
such products of LOX, leukotriene (LT) and lipoxin (LP),
participate in different physiologic and pathologic process
(5–7). 5-LOX catalyzes the synthesis of LTs, by sequential
reactions on AA to produce 5-hydroperoxyeicosatetraenoic
acid (5-HpETE) and then leukotriene A4 (LTA4), the latter
being converted to leukotriene B4 (LTB4) by LTA4 hydroxy-
lase (8). These lipid mediators are involved in inflammatory
and allergic diseases, such as asthma, rhinitis, ulcerative
colitis, cardiovascular disease, diabetes, metabolic syn-
drome and also in some types of cancers (9,10). For these
reasons, inhibition of 5-LOX is an important target for the
treatment of these pathologies.
ꢀ
Sepulveda-Boza , Eric K. Hoobler , Steve Perry ,
Giovanni Diaz3 and Theodore R. Holman3,*
1Departamento de Ciencias del Ambiente, Facultad de
ꢀ
Quımica y Biologıa, Universidad de Santiago, Chile, Casilla
442, Correo 2, Santiago, Chile
ꢀ
2
ꢀ
Laboratorio de Investigacion Cientıfica Emory Black,
Escuela de Medicina, Facultad de Ciencias Medicas,
ꢀ
ꢀ
Universidad de Santiago, Chile, Casilla 442, Correo 2,
Santiago, Chile
3Department of Chemistry and Biochemistry, University of
California, Santa Cruz, CA 95064
*Corresponding authors: Carolina Mascayano,
carolina.mascayano@usach.cl; Theodore R. Holman,
holman@ucsc.edu
Continuing our search to find more potent and selective
5-LOX inhibitors, we present now the enzymatic evalua-
tion of seventeen isoflavones (IR) and nine isoflavans
(HIR), and their in vitro and in cellulo potency against
human leukocyte 5-LOX. Of the 26 compounds tested,
10 isoflavones and 9 isoflavans possessed micromolar
potency, but only three were selective against 5-LOX
(IR-2, HIR-303, and HIR-309), with IC50 values at least
10 times lower than those of 12-LOX, 15-LOX-1, and
15-LOX-2. Of these three, IR-2 (6,7-dihydroxy-4-meth-
oxy-isoflavone, known as texasin) was the most selec-
tive 5-LOX inhibitor, with over 80-fold potency
difference compared with other isozymes; Steered
Molecular Dynamics (SMD) studies supported these
findings. The presence of the catechol group on ring A
(6,7-dihydroxy versus 7,8-dihydroxy) correlated with
their biological activity, but the reduction of ring C, con-
verting the isoflavones to isoflavans, and the substituent
positions on ring B did not affect their potency against
5-LOX. Two of the most potent/selective inhibitors (HIR-
303 and HIR-309) were reductive inhibitors and were
potent against 5-LOX in human whole blood, indicating
that isoflavans can be potent and selective inhibitors
against human leukocyte 5-LOX in vitro and in cellulo.
Previously, we had investigated the inhibitory activity of
isoflavonoids on 5-LOX using a suspension of porcine
blood leukocytes (11), and the relationship between struc-
ture and potency showed that isoflavans were more effec-
tive inhibitors than their corresponding isoflavones. The
data were rationalized on the basis of the conformational
change of ring C after hydrogenation and the interruption
of the fully conjugated system. Subsequently, we investi-
gated the effect of isoflavonoids on purified human lipo-ox-
ygenases (platelet 12-LOX, reticulocyte 15-LOX-1, and
epithelial 15-LOX-2) (12) and determined that aromaticity
and the oxidation state of the isoflavonoid ring C were
important for both inhibitory potency and selectivity. The
isoflavones and isoflavanones preferentially inhibited 12-
LOX, whereas the isoflavans preferentially inhibited 15-
LOX-1. Simple modifications of the basic isoflavonoid
structure produced a number of selective inhibitors of both
LOX, indicating that the isoflavonoid skeleton is a viable
scaffold for selective inhibitor development. Recently, we
reported results supporting our earlier work and indicating
that the catechol moiety and the flexibility of ring C are
important features for increased inhibitor potency, with this
new group of isoflavans being more potent and selective
against human 12-LOX and 15-LOX-1 (13).
Key words: human lipo-oxygenase, IC50 values, structure–
activity relationship, isoflavans derivative, steered molecular
dynamics
In an effort to discover potent and selective 5-LOX inhibi-
tors, we now present the in vitro and in cellulo potency
Received 8 July 2014, revised 2 October 2014 and accepted
for publication 17 October 2014
ª 2014 John Wiley & Sons A/S. doi: 10.1111/cbdd.12469
1

Mascayano et al.
evaluation of 26 related isoflavones (IR) and isoflavans
(HIR) against human leukocyte 5-LOX. In addition, pseud-
operoxidase assays, docking, and steered molecular
dynamics (SMD) studies were carried out for the most
selective inhibitors to elucidate the relationship between
the structural features of the isoflavonoids and their 5-LOX
inhibitory potency.
Preparation of isoflavones (general procedure)
To a solution of the benzylphenylketone (0.18 mol) in dry
DMF (200 mL), BF3.Et2O was added dropwise (0.88 mol)
(12,14). This solution was then warmed to 50 °C and a
solution of methanesulfonyl chloride (0.56 mol) in DMF
(100 mL) was slowly added. The resulting mixture was
then heated to 100 °C for 2 h. After cooling, it was poured
into water (4 L) and left overnight to give a precipitate,
which was stirred for 2 h in cold methanol (50 mL), fil-
tered, and crystallized in the appropriate solvent. The isof-
lavones were characterized by standard characterization in
supporting information (Data S1).
Methods and Materials
Synthesis of isoflavones and isoflavans
All starting materials were commercially available (Sigma-
Aldrich, St. Louis, MI, USA), with purity higher than 98%,
and were used without further purification. The isoflavo-
noids were obtained by classic electrophilic substitution
of appropriate phenols with benzyl cyanides (Houben-Ho-
esch reaction). The resulting hydroxybenzylketones were
cyclized to the isoflavones using DMF/MeSO2Cl as a
carbon atom donor in the presence of BF3Et2O. The
isoflavans were obtained by catalytic hydrogenation of
the corresponding isoflavones with Pd/C (10%) in acetic
acid containing 0.1% concentrated sulfuric acid as
shown in Schemes 1.
Expression and purification of human 5-lipo-
oxygenase
5-LOX was expressed as a non-tagged protein and used
as a crude ammonium sulfate protein fraction (17,18),
while 15-LOX-2 was expressed and purified as a his-
tagged protein, as published previously (19).
IC50 assay
The inhibition percentages were determined by following
the formation of the conjugated diene product at 234 nm
(e = 25 000 M/cm) with a Perkin Elmer (Santa Clara, CA,
USA) Lambda 40 UV/Vis spectrophotometer relative to
control rates of carrier solvent DMSO as previously pub-
lished. It is important to mention that all tested isoflavonoids
showed absorbance between 255 and 320 nm (18,19). The
reactions were done in a volume of 2 mL and constantly
stirred using a magnetic stir bar at room temperature
(23 °C). Reactions with the crude, ammonium sulfate pre-
cipitated 5-LOX were carried out in 25 mM HEPES (pH 7.3),
0.3 mM CaCl2, 0.1 mM EDTA, 0.2 mM ATP, 0.01% Triton-X-
100, and 10 lM AA. The concentrations of AA for 5-LOX
assays were quantitatively determined by allowing the enzy-
matic reaction to go to completion. IC50 values were
obtained by determining the enzymatic rate at various inhib-
itor concentrations and plotted against inhibitor concentra-
tion, followed by a hyperbolic saturation curve fit. The data
used for the saturation curves were obtained in duplicate or
triplicate, depending on their quality.
To verify the purity of each compound, HPLC analyses
were performed using a Merck-Hitachi Intelligent L-6200A
Pump, an L-4250 UV-Vis Detector, and a D-7000 HSM
System Manager Report, a C18 reverse phase column
(Hypersil ODS-5, 250 9 4 mm), and a flow rate of 1
mL/min. The isoflavones (IR) were detected at 260 nm,
and isoflavans (HIR) at 295 nm. Two different solvent sys-
tems were used as follows: system 1: (A) acetonitrile and
(B) 1% acetic acid and system 2: (A) acetonitrile and (B) a
1:1 mixture of 1% acetic acid/methanol. A gradient of
30 min of duration was used in both cases, beginning with
30% of (A), reaching 99% at 30 min, and (B) starting with
70% and ending with 1% in 30 min. Melting points were
recorded using a capillary Microthermal instrument and
were not corrected. The purity of all compounds evaluated
was higher than 95%. The isoflavones (IR) were prepared
by reported procedures (12,14,15) from the corresponding
intermediate benzylphenylketones.
Preparation of intermediate benzylphenylketones
(general procedure)
Pseudoperoxidase activity assay
The reductive properties of the compounds were deter-
mined by monitoring the pseudo-peroxidase activity of 5-
LOX in the presence of the inhibitor and 13-HPODE (20).
Activity was characterized by direct measurement of the
product degradation following the decrease of absorbance
at 234 nm using a Perkin Elmer Lambda 45 UV/Vis spec-
trometer (50 mM sodium phosphate (pH 7.4), 0.3 mM
CaCl2, 0.1 mM EDTA, 0.01% Triton-X-100, and 10 lM 13-
HPODE). All reactions were performed in 2 mL of buffer
and constantly stirred with a rotating stir bar (22 °C).
Reaction was initiated by addition of 10 lM inhibitor (1:1
ratio to product), and a positive result for activity reflected
Dry HCl was passed into a cooled (0 °C), stirred mixture
of the substituted phenylacetonitrile (0.34 mol) and anhy-
drous zinc chloride (30 g, 0.22 mol) in dry diethyl ether
(200 mL) (12,14). The corresponding polyhydroxybenzene
(0.28 mol) was added portionwise with constant bubbling
of gaseous HCl. The reaction mixture was then stirred at
room temperature for a few hours. The ketiminium chloride
intermediate was separated as an oil, washed with diethyl
ether, and hydrolyzed by refluxing in 5% HCl (1 L) for 4–
5 h. The ketone that separated upon cooling was filtered
and recrystallized in the appropriate solvent.
2
Chem Biol Drug Des 2014

Enzymatic Studies of Isoflavonoids as Inhibitors of 5-hLOX
ꢁ
a loss of more than 40% of product absorption using zilu-
eton, a known reductive inhibitor. Individual controls were
conducted with inhibitor alone with product and enzyme
alone with product. These negative controls formed the
baseline for the assay, reflecting non-pseudoperoxidase
dependent hydroperoxide product decomposition. To rule
out the auto-inactivation of the enzyme from pseudo-per-
oxidase cycling, 15-LOX-1 residual activities were mea-
sured after the assay was complete, 20 lM AA was added
to the reaction mixture and the residual activity was deter-
mined by comparing the initial rates with inhibitor and 13-
(S)-HPODE versus inhibitor alone, as the inhibitor by itself
inherently lowers the rate of the oxygenation. Activity is
characterized by direct measurement of the product for-
mation with the increase of absorbance at 234 nm.
3O8Y, 2.39 A resolution) and porcine leukocyte 12-lipo-
ꢁ
oxygenase (PDB code: 3RDE, 1.89 A resolution) (24,25)
was performed with the AutoDock4 package (26), using a
Lamarckian algorithm and assuming total flexibility of the
inhibitors and partial flexibility of the His residues co-ordi-
nated to Fe3+ inside the binding site. The grid maps were
made up of 60 9 60 9 60 points, with a grid-point spac-
ꢁ
ing of 0.375 A. The AutoTors option was used to define
the ligand torsions, and the docking results were then
analyzed by a ranked cluster analysis, resulting in confor-
mations with the highest overall binding energy (most neg-
ative –DGbinding value).
The SMD simulations were performed using NAMD 2.6
(27) with the Charmm33b1 (28) force field. Each 12- and
5-lipo-oxygenase systems with IR-2 were put into a water
ꢁ3
box 100 9 100 9 100 and 90 9 120 9 110 A , respec-
ꢁ
COX-1 and COX-2 inhibition
tively, with a layer of water of at least 15 A in each
Ovine COX-1 (cat. no. 60100) and human COX-2 (cat. no.
60122) were purchased from Cayman Chemical. Approxi-
mately 2 lg of either COX-1 or COX-2 was added to buf-
fer containing 100 lM AA, 0.1 M Tris–HCl buffer (pH 8.0),
5 mM EDTA, 2 mM phenol, and 1 lM hematin at 37 °C.
Data were collected using a Hansatech DW1 oxygen elec-
trode chamber. Inhibitors were incubated with the respec-
tive COX isozyme for 20 min and added to the reaction
mixture, and the rate of oxygen consumption was
recorded. Ibuprofen, aspirin, and the carrier solvent,
DMSO, were used as positive and negative controls,
respectively (21).
dimension. Both systems were neutralized, and a cutoff
ꢁ
of 10 A for non-bonded interactions was applied. We
performed 250 ps of water equilibration, 10 000 steps of
minimization and 50 ps of heating from the 0 K up to
300 K before each main SMD simulation. For IR-2, the
1,2 -ns-long SMD simulations were carried out. The com-
pound was pulled out of the binding site and entered
the solvent region. The external force was attached to
the center of mass of the molecule, and the velocity of
ꢁ
the SMD reference point was 0.00005 A/ps in both
enzymes, and the spring constant was 4 kcal/(mol-A2),
the iron insight into the binding site was fixed to calculate
the direction of the vector. To obtain a better statistical
result, we repeated all simulations five times. The temper-
ature was kept constant using the Langevin method
(310 K). All graphical analysis was performed using the
VMD software (29).
LTB4 inhibition assay
Whole human blood was purchased through Innovative
Research. Blood was dispensed in 150 lL samples fol-
lowed by addition of inhibitor or control (the DMSO vehi-
cle), and then incubation for 15 min at 37 °C. Blood
coagulation was then stimulated by introduction of cal-
cium ionophore A23817 (freshly diluted from a 50 mM
stock solution in DMSO to 1.5 mM in Hanks’ balanced
salt solution (HBSS)) along with incubation for 30 min at
37 °C. Samples were then centrifuged at 7500 9 g
(300 9 g) for 10 min at 4 °C. Plasma was then sepa-
rated and diluted (1:100) with HBSS for LTB4 detection
using an ELISA detection kit (Cayman). Inhibitors were
added at 10 or 15 lM concentrations (0.5 lM for control
compound setileuton), and IC50 values were generated
using a one point IC50 estimation equation. Drug efficacy
Results and Discussion
Twenty-six isoflavone (IR) and isoflavan (HIR) derivatives
were analyzed as potential inhibitors of human leukocyte
5-LOX (Table 1), of which nineteen proved to have IC50
values lower than 10 lM. For comparison, baicalein (the
flavone 5,6,7 trihydroxy-2-phenyl chromen-4-one) showed
an IC50 of 0.85 lM (30). The interpretation of the structure–
function relationship was focused on three structural
aspects, the catechol group on ring A (6,7-dihydroxy or
7,8-dihydroxy), the oxidation state of ring C (i.e. isoflav-
ones versus isoflavans), and finally the position and chemi-
cal features of substituents on ring B.
was determined using
donors (22).
a minimum of two different
The role of the catechol moiety in the LOX inhibitors was
previously investigated (11–13,31), but the impact of its
position for IR (isoflavones) and HIR (isoflavans) derivatives
on the biological activity against human 5-LOX has not
been fully discussed. Of the current 26 derivatives, only
Docking and steered molecular dynamics studies
The 5-LOX structure was built with the GaussianView soft-
ware (23). ChelpG charges were obtained at the B3LYP/6-
31G** level of theory, employing the Gaussian 03 package
(23). Docking of all inhibitors into the active site of the
crystal structures of human 5-lipo-oxygenase (PDB code:
the compounds that possessed
a catechol moiety
manifested submicromolar potency, indicative of either a
Chem Biol Drug Des 2014
3

Mascayano et al.
Table 1: Structures of isoflavonoids and baicalein and their inhibitory activity (IC50) against human leukocyte 5-LOX and human 15-LOX
type 2
12-LOX (12,13)
[lM ꢀ SD]
15-LOX-1 (12,13)
15-LOX-2
5-LOX
Name
R6
R7
R8
R30
R40
[lM ꢀ SD]
[lM ꢀ SD]
[lM ꢀ SD]
IR-1
IR-2
IR-3
IR-4
IR-5
IR-6
IR-8
IR-10
IR-201
IR-203
IR-206
IR-213
IR-301
IR-303
IR-308
IR-309
IR-406
HIR-1
HIR-3
HIR-7
HIR-9
HIR-11
HIR-205
HIR-301
HIR-303
HIR-309
Baicalein
OH
OH
OH
OH
OH
OH
OH
OH
H
H
H
H
H
H
H
H
H
OH
OH
OH
OH
H
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
H
H
H
H
H
H
H
H
CH3
CH3
CH3
CH3
OH
OH
OH
OH
H
H
H
H
H
H
H
H
H
H
H
Cl
F
H
H
H
CF3
H
H
Cl
CH3
OH
OCH3
CH3
H
Cl
NO2
H
8.7 ꢀ 1
>100
49 ꢀ 8
>100
>100
>100
–
–
>40
1.2 ꢀ 0.2
0.36 ꢀ 0.3
>40
2.3 ꢀ 0.3
18 ꢀ 7
>40
>40
>100
>100
>100
0.59 ꢀ 0.07
12 ꢀ 3
>100
>100
>100
>100
50 ꢀ 20
7.8 ꢀ 0.8
6.2 ꢀ 0.7
11 ꢀ 0.7
>100
0.5 ꢀ 0.1
2.3 ꢀ 0.7
16 ꢀ 2
0.21 ꢀ 0.02
11 ꢀ 3
29 ꢀ 5
>40
–
31
5.8 ꢀ 0.5
0.28 ꢀ 0.3
4.2 ꢀ 0.5
>100
>100
>100
>30
>30
1.5 ꢀ 0.5
0.39 ꢀ 0.05
0.49 ꢀ 0.2
12 ꢀ 2
H
–
–
–
–
–
–
OH
CH3
NO2
H
OH
CH3
H
1.5 ꢀ 0.3
300
>100
–>100
600
0.9 ꢀ 0.1
0.25 ꢀ 0.05
0.3 ꢀ 0.03
0.3 ꢀ 0.03
5 ꢀ 0.5
1.6 ꢀ 0.3
0.78 ꢀ 0.08
3.6 ꢀ 0.3
>100
17 ꢀ 1.7
5.5 ꢀ 0.6
0.35 ꢀ 0.06
15 ꢀ 1.4
17 ꢀ 4
>40
>100
>100
>100
–
H
H
H
H
NO2
OH
CH3
11 ꢀ 1.2
H
OCH3
Cl
OH
CH3
H
71 ꢀ 30
0.3 ꢀ 0.05
0.28 ꢀ 0.09
–
O-CH2-O
CH3
OCH3
H
H
H
0.4 ꢀ 0.08
8.3 ꢀ 0.9
0.5 ꢀ 0.1
0.69 ꢀ 0.2
7.1 ꢀ 0.9
OH
CH3
OH
OH
OH
–
–
–
H
H
H
H
>40
0.31 ꢀ 0.06
0.18 ꢀ 0.2
0.15 ꢀ 0.01
0.85 ꢀ 0.2
7.7 ꢀ 1.3
6.4 ꢀ 1
0.86 ꢀ 0.3
34 ꢀ 0.5
5.8 ꢀ 0.9
9.1 ꢀ 0.8
34 ꢀ 7
33 ꢀ 7
>100
CH3
chelative or reductive inhibitory mechanism. Of the IRs, the
7,8-catechols (IR-301 and IR-303) were more active than
the related 6,7-catechols (IR-1 and IR-3, respectively).
On the other hand, the HIRs exhibited little difference
when the catechol position was changed (6,7-dihydroxy
versus 7,8-dihydroxy), as seen with the comparable
potency of HIR-1, -3 and -9 and HIR-301, -303 and 309.
These results suggested that changes in the catechol
position had larger effects on potency for the IR derivatives
than the HIR derivatives, due to the tight binding con-
straints for isoflavone (vide infra). Nevertheless, this differ-
ence is minor considering that the most potent IR and HIR
derivatives have comparable IC50 values, regardless of the
position of the catechol.
Among the non-catechol derivatives, only IR-203 (IC50
1.5 lM) and IR-406 (IC50 5 lM) inhibited 5-LOX, with nei-
ther having submicromolar potency, indicating the need of
4
Chem Biol Drug Des 2014

Enzymatic Studies of Isoflavonoids as Inhibitors of 5-hLOX
Table 2: Qualitative results of pseudoperoxidase assays for two
A
standard compounds and three top inhibitors of human 5-LOX
Compound
Redox
Zileuton
Setileuton
HIR-303
HIR-309
IR-8
Yes
No
Yes
Yes
No
The oxidation state of ring C was not correlated with the
potency of the inhibitors studied, as seen with the isoflav-
one/isoflavan pairs: IR-3/HIR-3, IR-301/HIR-301, IR-303/
HIR-303, and IR-309/HIR-309, all having similar poten-
cies. The IR-1/HIR-1 pair was an exception, with an over
100-fold difference in potency which may indicate that a
different docking mode for this pair. Nevertheless, consid-
ering that the majority of the derivatives showed no
potency difference between the IR and HIR derivatives, it
appears that for 5-LOX, the oxidation state of ring C is
not a critical factor for inhibitor potency. This differs from
our previous results, where we observed that the oxida-
tive state of the isoflavonoid had a large impact on 12-
LOX and 15-LOX-1 potency and selectivity (12,13). In
comparison, the structure–activity relationship of the isof-
lavones and isoflavans against porcine 5-LOX (11)
showed similar behavior with human 5-LOX. This low
sensitivity to the conformational change between isoflav-
ones and isoflavans for 5-LOX may be due to its inher-
ently larger active site, compared to the other LOX
isozymes. One of the above pairs, IR-303 (gray) and
HIR-303 (pink), were docked to the human 5-LOX active
site and little difference was found between them. The
catechol group was positioned at a mean distance of
B
Figure 1: Docking results of IR-8 (A) and HIR-309 (B) into
binding site of crystal structure of human 5-LOX (PDB ID 3O8Y).
ꢁ
4.9 A from iron for both inhibitors (Figure S1), explaining
the similar experimental results.
Docking studies also show that in the case of the 6,7-di-
hydroxy-isoflavones, the 7-hydroxyl of the catechol is at a
ꢁ
distance of 4 A from the active site iron. When the 6,7-di-
hydroxyisoflavone is reduced to obtain the 6,7-dihydroxyi-
soflavan, which changes the conformation of ring C, the
7-hydroxyl group of the ring B catechol group is still
positioned the same as for the 6,7-dihydroxyisoflavone
(Figure S2). These poses support the inhibitor data where
similar potencies between the isoflavones and isoflavans
were seen (i.e. IR-3, IC50 = 0.36 lM (gray) and HIR-3,
IC50 = 0.28 lM (pink)).
Figure 2: Variation of the pulling force (in pN) exerted on IR-2
during its unbinding from the 5-hLOX (black) and 12-pLOX (red)
pockets.
a catechol group for potent inhibitory activity (12,13). To
understand the inhibitory behavior of non-catechol com-
pounds more thoroughly, additional evidence would be
required; however, we can highlight that in our previous
studies, IR-203 and IR-406 were unable to inhibit other
human lipo-oxygenases (13), indicating they are selective
for 5-LOX. An explanation could be based on the larger
size of the 5-LOX active site, allowing for additional inter-
actions with our tested compounds.
The effect on 5-LOX inhibition by para or meta substitu-
ents on ring B is harder to interpret than the impact of the
catechol position, previously described for ring A. The ste-
ric effects are not sufficient to explain the results. For
example, IR-2 (IC50 1.2 lM) had a 40-methoxy group which
is bigger than the 40-hydroxyl on IR-1 (IC50 > 40 lM), but
its potency is greater. The electronegativity of the substitu-
ents on ring B also does not explain the results obtained,
Chem Biol Drug Des 2014
5

Mascayano et al.
Table 3: Evaluation of inhibitory activity (IC50) against COX1 and
COX2 and selectivity of some of the better inhibitors of human
5-LOX
because we found a compound with an electron with-
drawing nitro group (IR-6, IC50 1.5 lM) and a compound
with an electron donating methoxy group (IR-2, (IC50
1.2 lM) both being good inhibitors. This observation was
previously reported in our studies of the isoforms 12-LOX
and 15-LOX-1 (12,13). The lack of importance of the ring
B substituents for the inhibitory activity is reinforced with
the results found for the following compounds: 7,8-IR (IR-
301,-303,-308,-309), 6,7-HIR (HIR-1,-3,-7,-9), and 7,8-
HIR (HIR-301,-303,-309). For all the above compounds,
changes in size, electronegativity and position on ring B
produced only minimal differences in their ability to inhibit
5-LOX.
Compound
COX1 IC50
COX2 IC50
(10 lM in assay)
[lM ꢀ SD]
[lM ꢀ SD]
Setileuton
Zileuton
Ibuprofen
Aspirin
Baicalein
IR-2
IR-8
IR-301
HIR-303
HIR-309
>150
>150
4.0 ꢀ 0.4
100%
>150
>150
>50
>150
>150
>150
>100
–
8 ꢀ 2
19 ꢀ 3
–
>150
>150
>150
>150
>150
Given the importance of the catechol moiety, it is logical to
assume that these isoflavonoids are chelative and/or
reductive inhibitors. To determine whether a particular
inhibitor is chelative in nature, the EPR technique is
required to observe a direct change in the iron ligation.
This is a difficult experiment to do with 5-LOX, given the
unstable nature of the enzyme (32,33). However, testing
the reductive nature of an inhibitor is markedly easier with
the pseudoperoxidase assay, which measures the reduc-
tion of the hydroperoxide product by 5-LOX, with the con-
comitant oxidation of the inhibitor. This test was performed
with three potent 5-LOX inhibitors (IR-8, HIR-303, and
HIR-309) and two known inhibitors, the chelative and
reductive zileuton and the non-reductive setileuton (34).
The results showed that both, HIR-303 and HIR-309, with
7,8-dihydroxy groups, tested positive in the pseudoperoxi-
dase assay, which indicates that they are reductive inhibi-
tors, most likely through an inner sphere mechanism that
requires iron chelation. The radical scavenger properties
presented by phenolic compounds with one or two hydro-
xyl groups were previously studied, especially for the cate-
chol flavonoids taxifolin, luteolin, and quercetin, which can
react with the metal ion producing ortho-benzoquinones
by oxidation (35). However, IR-8, an isoflavone with a 6,7
dihydroxy group, was not active in the pseudoperoxidase
group was located near the iron atom. Nevertheless,
these two metal chelating properties correlate well with a
possible hydrogen transfer and change in oxidation state
from ferric to ferrous iron supporting the pseudoperoxi-
dase activity. (Figure 1B shows the docking of HIR-309,
as an example.)
Twelve of the 26 isoflavonoid derivatives were screened
against 15-LOX-2, and none showed potent activity
(Table 1). Only HIR-9 had an IC50 value below 10 lM,
which is over 10-fold less potent than its IC50 value
against 5-LOX (IC50 = 0.5 lM). This lack of potency
against 15-LOX-2 is a common feature for isoflavones and
has been described previously (13). Nevertheless, HIR-9 is
as potent (IC50 8.3 lM) as NDGA (IC50 = 11 lM) (12), one
of the few compound which inhibit 15-LOX-2 at all (36).
An additional aspect of this study was to evaluate whether
some of the potent and selective 5-LOX inhibitors were
also active against mammalian cyclooxygenase (COX), an
enzyme also involved in arachidonic acid metabolism. The
COX isozymes results showed that isoflavans HIR-303 and
HIR-309 and isoflavone IR-2 do not inhibit COX-1 or COX-
2 at IC50 concentrations higher than 150 lM (Table 3).
Ibuprofen and aspirin were used as positive controls and
shown to inhibit COX-2 activity with IC50 < 20 lM.
assay, indicating that it is not
a reductive inhibitor
(Table 2). An explanation for this behavior might rely on
the interruption of a fully conjugated system in isoflavans,
and the important role of the acidity of the -OH group
located at C-7 on ring A, as can be observed by analyz-
ing the pKa values for isoflavones and isoflavans, which
are 6.7 and 9, respectively. This change in pKa could
be responsible for the different reductive properties of
the inhibitors, allowing HIR derivatives to act as reduct-
ive inhibitors in the pseudoperoxidase assay, but not IR
derivatives (12).
Among all the compounds tested, IR-2 was the most
selective against 5-LOX (Table 1 and 2). Simulations by
SMD studies showed difference of affinity when we ana-
lyzed their trajectory in both sites of 5 and 12-LOX (IC50
1.2 and 100 lM, respectively). The SMD results in 5-LOX
showed two significant maximum forces at 2500 and 5500
pN, the first peak (A) corresponds to the break of the inter-
action between catechol and iron inside to the binding site
and the second peak corresponding to the rupture of the
interaction between catechol and Pro152 and Leu153 resi-
dues (B). Finally, the SMD study for 12-LOX did not pro-
vide relevant results in affinity, and the only important
signal was caused by the break of interaction between the
catechol and the iron at 1300 pN. (Figure 2).
To display the position of these three inhibitors inside the
binding site of 5-LOX, docking studies were performed.
The data reveal that the catechol group of IR-8
was located symmetrically in front of the metal (see
Figure 1A). On the other hand, the isoflavans, HIR-303
and HIR-309, were unable to orient their catechol groups
symmetrically opposite to the metal, and only the 7-OH
6
Chem Biol Drug Des 2014

Enzymatic Studies of Isoflavonoids as Inhibitors of 5-hLOX
References
Table 4: Ex vivo inhibition of LTB4 production (IC50) in whole
blood
1. Yamamoto S. (1992) Lipoxygenases: molecular
Compound
IC50 [lM ꢀ SD]
structures
and
functions.
Biochim
Biophys
IR-8
19.4 ꢀ 4.0
1.4 ꢀ 1.2
1.2 ꢀ 0.1
0.2 ꢀ 0.1
Acta;1128:117–131.
HIR-303
HIR-309
Setileuton
2. Malhotra S., Deshmukh S.S., Dastidar S.G. (2012)
COX inhibitors for airway inflammation. Expert Opin
Ther Targets;16:195–207.
3. Pontiki E. (2008) Lipoxygenase inhibitors: a compara-
tive QSAR study review and evaluation of new QSARs.
Med Res Rev;28:39–117.
The next step was to apply these inhibitors to a whole
blood, 5-LOX cellular model, which would demonstrate
not only cell permeability but also cellular stability. As seen
in Table 4, HIR-303, HIR-309, and IR-8 (in vitro IC50
values of 0.18, 0.15 and 0.39 lM, respectively) all inhibit
LTB4 production in cellulo, indicating 5-LOX inhibition.
However, the in cellulo IC50 values for HIR-303 (IC50
1.4 lM) and HIR-309 (IC50 1.2 lM) are approximately 10-
fold higher than their in vitro IC50 values, with IR-8 being
almost 50-fold less potent (IC50 19.4 lM). These data indi-
cate that while these compounds are active in the cell
against 5-LOX, they have lowered potency, possibly due
to either lowered cell permeability or structural modification
of the compounds in the cell. It should be noted that the
known 5-LOX inhibitor, setileuton, was used as a positive
control and showed high activity.
4. Ji R.R., Xu Z.Z., Strichartz G., Serhan C.N. (2011)
Emerging roles of resolvins in the resolution of inflam-
mation and pain. Trends Neurosci;34:599–609.
5. Kelavkar U.P., Cohen C., Kamitani H., Eling T.E., Badr
K.F. (2000) Concordant induction of 15-lipoxygenase-1
and mutant p53 expression in human prostate adeno-
carcinoma: correlation with Gleason staging. Carcino-
genesis;21:1777–1787.
6. Ding X.Z., Iversen P., Cluck M.W., Knezetic J.A.,
Adrian T.E. (1999) Lipoxygenase inhibitors abolish pro-
liferation of human pancreatic cancer cells. Biochem
Biophys Res Commun;261:218–223.
7. Bozza P.T., Bakker-Abreu I., Navarro-Xavier R.A., Ban-
deira-Melo C. (2011) Lipid body function in eicosanoid
synthesis: an update. PLEFA;85:205–213.
€
8. Riccioni G., Back M., Capra V. (2010) Leukotrienes
and atherosclerosis. Curr Drug Targets;11:882–887.
9. Moreno J.J. (2005) Arachidonic acid cascade enzyme
inhibition and cancer. Curr Enzyme Inhib;1:131–145.
10. Mitjavilaa M., Moreno J.J. (2012) The effects of po-
lyphenols on oxidative stress and the arachidonic acid
cascade. Implications for the prevention/treatment of
high prevalence diseases. Biochem Pharma-
col;84:1113–1122.
Conclusion
A series of isoflavones and isoflavans were synthesized
and evaluated against human leukocyte 5-LOX, with most
of the compounds that possess a catechol moiety mani-
festing submicromolar potency, regardless of its position
(i.e. 6,7- versus 7,8-). We also observe that substituents
on ring B and ring C oxidation are not relevant for
potency, but ring C oxidation is relevant for selectivity.
None of these molecules are potent against the COX iso-
zymes, and only one, HIR-9, shows modest potency
against 15-LOX-2. The pseudoperoxidase assay shows
that HIR-303 and HIR-309 are reductive inhibitors, but
IR-8 is not, possibly due to an interruption in the fully con-
jugated system and its effect of -OH group acidity in HIR-
303 and HIR-309, also both compounds are active against
5-LOX in whole blood, indicating effective cellular transport
and cellular potency. In summary, we have demonstrated
that both isoflavones and isoflavans can be potent and
selective inhibitors against human leukocyte 5-LOX in vitro
and in cellulo.
11. Voss C., Sepulveda-Boza S., Zilliken F.W. (1992) New
isoflavonoids as inhibitors of porcine 5-lipoxygenase.
Biochem Pharmacol;44:157–162.
12. Vasquez-Martinez Y., Ohri R.V., Kenyon V., Holman
ꢀ
T.R., Sepulveda-Boza S. (2007) Structure–activity rela-
tionship studies of flavonoids as potent inhibitors of
human platelet 12-hLO, reticulocyte 15-hLO-1, and
prostate
epithelial
15-hLO-2.
Bioorg
Med
Chem;15:7408–7425.
ꢀ
13. Mascayano C., Espinosa V., Sepulveda-Boza S., Hoo-
bler E.K., Perry S. (2013) In vitro study of isoflavones
and isoflavans as potent inhibitors of human 12- and
15-lipoxygenases. Chem Biol Drug Des;82:317–325.
ꢀ
14. Sepulveda-Boza S., Walizei G.H., Caroli-Rezende M.,
ꢀ
ꢀ
Vasquez Y., Mascayano C., Mejıas L. (2001) The prep-
aration of new isoflavones. Synthetic Comm;31:1933–
1940.
Acknowledgments
15. Albert A.I., Zilliken F.W. (1989) U.S. Patent 4 814, 346.
16. Zilliken F.W. (1984) EP0027796B1.
17. Tang L.J., Chen X., Sun Y.N., Ye J., Lu J., Han Y., Ji-
ang X., Cheng C.C., He C.C., Qiu P.H., Li. X.K. (2011)
Synthesis and biological studies of 40, 7, 8-trihydroxy-
Financial support from FONDECYT (project # 1120379)
and NIH (GM56062 to TRH) is gratefully acknowledged.
We also acknowledge a generous gift of setileuton from
Dr. David Maloney.
Chem Biol Drug Des 2014
7

Mascayano et al.
isoflavone
chem;105:1623–1629.
metal
complexes.
J
Inorg
Bio-
30. Deschamps J.D., Kenyon V.A., Holman T.R. (2006)
Baicalein is a potent in vitro inhibitor against both retic-
ulocyte 15-human and platelet 12-human lipoxygenas-
es. Bioorg Med Chem;14:4295–4301.
31. Mascayano C., Caroli M., Rivera Y., Espinosa V.
(2011) Isoflavans derivatives as inhibitors of soybean
lipoxygenase: in-vitro and docking studies. J Child
Chem Soc;56:935–937.
32. Solomon E.I., Zhou J., Neese F., Pavel E.G. (1997)
New insights from spectroscopy into the structure/
function relationships of lipoxygenases. Chem
Biol;4:795–808.
33. Dunham W.R., Carroll R.T., Thompson J.F., Sands
R.H., Funk M.O. Jr (1990) The initial characterization of
18. Robinson S.J., Hoobler E.K., Riener M., Loveridge
S.T., Tenney K., Valeriote F.A., Holman T.R., Crews P.
(2009) Using enzyme assays to evaluate the structure
and bioactivity of sponge-derived meroterpenes. J Nat
Prod;72:1857–1863.
19. Ohri R.V., Radosevich A.T., Hrovat K.J., Musich C.,
Huang D., Holman T.R., Toste F.D. (2005) A Re(V)-cat-
alyzed C-N bond-forming route to human lipoxygenase
inhibitors. Org Lett;9:2501–2504.
20. Riendeau D., Falgueyret J.P., Guay J., Ueda N., Ya-
mamoto S. (1991) Pseudoperoxidase activity of 5-lip-
oxygenase stimulated by potent benzofuranol and N-
hydroxyurea inhibitors of the lipoxygenase reaction.
Biochem J;274:287–292.
€
the iron environment in lipoxygenase by Mossbauer
spectroscopy. Eur J Biochem;190:611–617.
21. Rai G., Kenyon V., Jadhav A., Schultz L., Armstrong
M., Jameson J.B., Hoobler E., Leister W., Simeono A.,
Holman T.R., Maloney D.J. (2010) Discovery of potent
and selective inhibitors of human reticulocyte 15-lipox-
ygenase-1. J Med Chem;53:7392–7404.
22. Spaethe S.M., Snyder D.W., Pechous P.A., Clarke T.,
Van Alstyne E.L. (1992) Guinea pig whole blood 5-lip-
oxygenase assay: utility in the assessment of potential
5-lipoxygenase inhibitors. Biochem Pharmacol;43:377–
382.
34. Hoobler E.K., Holz C., Holman T.R. (2013) Pseudoper-
oxidase investigations of hydroperoxides and inhibitors
with
human
lipoxygenases.
Bioorg
Med
Chem;21:3894–3899.
35. Baumann J., Bruchhausen F., Wurm G. (1980) Flavo-
noids and related compounds as inhibition of arachi-
donic acid peroxidation. Prostaglandins;20:627–639.
36. Tang S., Bhatia B., Maldonado C.J., Yang P., New-
man R.A., Liu J., Chandra D., Traag J., Klein R.D., Fi-
scher S.M., Chopra D., Shen J., Zhau H.E., Chung
L.W., Tang D.G. (2002) Evidence that arachidonate
15-lipoxygenase 2 is a negative cell cycle regulator in
23. Frisch M.J., Trucks G.W., Schlegel H.B., Scuseria
G.E., Robb M.A., Cheeseman J.R., Scalmani G. et al.
(2009) Gaussian 09, Revision A.1. Wallingford CT:
Gaussian, Inc.
normal
prostate
epithelial
cells.
J
Biol
Chem;277:16189–16201.
24. Gilbert N.C., Bartlett S.G., Waight M.T., Neau D.B.,
Boeglin W.E., Brash A.R., Newcomer M.E. (2011) The
structure of human 5-lipoxygenase. Science;331:217–
219.
Supporting Information
25. Xu S., Mueser T.C., Marnett L.J., Funk M.O. (2012)
Crystal structure of 12-lipoxygenase catalytic-domain-
inhibitor complex identifies a substrate-binding channel
for catalysis. Structure;20:1490–1497.
Additional Supporting Information may be found in the
online version of this article:
Scheme S1. Reagents and general conditions of classic
26. Morris G.M., Goodsell D.S., Halliday R.S., Huey R.,
Hart W.E., Belew R.K., Olson A.J.J. (1998) Automated
docking using a lamarckian genetic algorithm and
empirical binding free energy function. Comput
Chem;19:1639–1662.
synthesis of isoflavones and isoflavans.
Figure S1. Docking of 7,8-dihydroxyisoflavone and isofla-
van derivatives inside the 5-LOX binding site (IR-303 and
HIR-303, respectively).
27. Kale L., Skee L.R., Bhandarkar M., Brunner R., Gursoy
A., Krawetz N., Phillips J., Shinozaki A., Varadarajan K.,
Schulten K. (1999) NAMD2: greater scalability for parallel
molecular dynamics. J Comput Phys;151:283–312.
28. Brooks B.R., Bruccoleri R.E., Olafson B.D., States
D.J., Swaminathan S., Karplus M. (1983) CHARMM: a
program for macromolecular energy, minimization, and
dynamics calculations. J Comp Chem;4:187–217.
29. Humphrey W., Dalke A., Schulten K. (1996) VMD –
visual molecular dynamics. J Mol Graphics;14:33–38.
Figure S2. Docking results of 6,7-dihydroxyisoflavone and
6,7-dihydroxyisoflavan derivatives IR-3 and HIR-3, respec-
tively, inside the 5-LOX binding site.
Data S1. The isoflavones were characterized by standard
characterization.
8
Chem Biol Drug Des 2014