Synthesis of 13R,20-dihydroxy-docosahexaenoic acid by site-directed mutagenesis of lipoxygenase derived from Oscillatoria nigro-viridis PCC 7112

Article abstract of DOI:10.1016/j.bbrc.2020.09.079

Lipoxygenases (LOXs) are implicated in the biosynthesis of pro- and anti-inflammatory lipid mediators involved in immune cell signaling, most of which catalyze peroxidation of polyunsaturated fatty acids by distinct regio- and stereoselectivity. Current reports suggested that conserved amino acid, Gly in R-LOXs and Ala in S-LOXs, in the catalytic domain play an important role in determining the position as well as the stereochemistry of the functional group. Recently, we have confirmed that the catalytic specificity of cyanobacterial lipoxygenase, named Osc-LOX, with alanine at 296 was 13S-type toward linoleic acid, and producing a 17S- hydroxy-docosahexaenoic acid from docosahexaenoic acid (DHA). Here, we aimed to change the catalytic property of LOX from13S-LOX to 9R-LOX by replacing Ala with Gly and to produce a lipid mediators different from the wild-type using DHA. Finally, we succeeded in generating human endogenous a 13R-hydroxy-docosahexaenoic acid and a 13R,20-dihydroxy-docosahexaenoic acid from DHA through an enzymatic reaction using the Osc-LOX-A296G. Our study could enable physiological studies and pharmaceutical research for the 13R,20-dihydroxy-docosahexaenoic acid.

Full text of DOI:10.1016/j.bbrc.2020.09.079

Biochemical and Biophysical Research Communications 533 (2020) 893e898
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications
journal homepage: www.elsevier.com/locate/ybbrc
Synthesis of 13R,20-dihydroxy-docosahexaenoic acid by site-directed
mutagenesis of lipoxygenase derived from Oscillatoria nigro-viridis
PCC 7112
,
, **
Jong-Jae Yi a b, Sun-Yeon Heo a, Jung-Hyun Ju a, Baek-Rock Oh a, Woo Sung Son b
,
Jeong-Woo Seo a
,
*
a Microbial Biotechnology Research Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB), Jeongeup-Si, 56212, Republic of Korea
b Department of Pharmacy, College of Pharmacy and Institute of Pharmaceutical Sciences, CHA University, Pocheon-Si, Gyeonggi-do, 11160, Republic of
Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Lipoxygenases (LOXs) are implicated in the biosynthesis of pro- and anti-inflammatory lipid mediators
involved in immune cell signaling, most of which catalyze peroxidation of polyunsaturated fatty acids by
distinct regio- and stereoselectivity. Current reports suggested that conserved amino acid, Gly in R-LOXs
and Ala in S-LOXs, in the catalytic domain play an important role in determining the position as well as
the stereochemistry of the functional group. Recently, we have confirmed that the catalytic specificity of
cyanobacterial lipoxygenase, named Osc-LOX, with alanine at 296 was 13S-type toward linoleic acid, and
producing a 17S- hydroxy-docosahexaenoic acid from docosahexaenoic acid (DHA). Here, we aimed to
change the catalytic property of LOX from13S-LOX to 9R-LOX by replacing Ala with Gly and to produce a
lipid mediators different from the wild-type using DHA. Finally, we succeeded in generating human
endogenous a 13R-hydroxy-docosahexaenoic acid and a 13R,20-dihydroxy-docosahexaenoic acid from
DHA through an enzymatic reaction using the Osc-LOX-A296G. Our study could enable physiological
studies and pharmaceutical research for the 13R,20-dihydroxy-docosahexaenoic acid.
Received 7 September 2020
Accepted 20 September 2020
Available online 29 September 2020
Keywords:
Lipoxygenase
Coffa-brash principle
Site-directed mutagenesis
13R-HDHA
13R,20-diHDHA
© 2020 Elsevier Inc. All rights reserved.
1. Introduction
C17, and/or C-20), theoretically.
It has been known that the regio- and stereospecificity toward
Lipoxygenases (LOXs) are essential enzymes for biosynthesis of
lipid mediators that are bioactive molecules involved in immune
cell signaling. LOXs catalyze regio- and stereospecific peroxidation
by the initiation of hydrogen abstraction in polyunsaturated fatty
acids (PUFAs) containing one or more 1-cis,4-cis-pentadiene
structure to yield hydroperoxides [1]. Hydrogen abstraction arises
at a bis-allylic methylene group of PUFAs, and the peroxidation
occurs either at the [þ2] or [ꢀ2] carbon position by radical rear-
rangement [2]. linoleic acid (LA) containing one methylene group
(C-11) can be peroxygenated at only two carbon positions (C-9 and/
or C-13), whereas, docosahexaenoic acid (DHA) containing five
methylene groups (C-6, C-9, C-12, C-15, and C-18) can be peroxy-
genated at ten positions (C-4, C-7, C-8, C-10, C-11, C-13, C-14, C-16,
substrates of LOXs are controlled by three different hypothetical
mechanisms as follows: (i) The orientation theory proposed that
positional specificity is determined by the orientation, such as
head-to-tail or tail-to-head, of substrates to entry into the catalytic
pocket [3]. For instance, the difference in regiospecificty between
5S-LOX and 15S-LOX is due to the entry direction of the substrates.
In the case of 5-LOX, the substrates enter into the substrate-binding
pocket with its carboxylate group ahead. In contrast, for 15-LOX,
the substrates slide into the catalytic site with its methyl-end first.
(ii) The space theory suggested that regiospecificity depends on the
volume of the catalytic pocket [4]. The volume of the active site of
human 12-LOX is about 6% bigger than that of human 15-LOX
resulting in a frameshift of the substrate is occurred. Thus, the
position of peroxidation is determined by how deeply the substrate
enters the active site. (iii) According to Coffa-Brash principle, Gly
and Ala are key amino acids to determine of regio- and stereo-
specific (R/S) oxidation at C-1 and C-5 of the1-cis,4-cis-pentadiene.
* Corresponding author.
** Corresponding author.
E-mail address: jwseo@kribb.re.kr (J.-W. Seo).
https://doi.org/10.1016/j.bbrc.2020.09.079
0006-291X/© 2020 Elsevier Inc. All rights reserved.

J.-J. Yi, S.-Y. Heo, J.-H. Ju et al.
Biochemical and Biophysical Research Communications 533 (2020) 893e898
Ala is highly conserved in S-type LOXs, whereas relatively the
smaller amino acid Gly is conserved in R-type LOXs [5]. If these
three mechanisms could be applied to LOXs, they could be used as
tools to produce the lipid mediators that exist in the human im-
mune system but cannot be generated via in vitro enzyme reaction.
In our previous study, it has been identified that cyanobacterial
lipoxygenase (Osc-LOX) derived from Oscillatoria nigro-viridis PCC
7112 has alanine at 296 position and the catalytic specificity was
13S-type toward LA. Also, it has been confirmed that wild-type Osc-
LOX generates a 17S-hydroperoxy-DHA as intermediate product
from DHA by hydrogen abstraction for the C-15 methylene group
(under revision). The aim of this study was to investigate the
applicability of the Coffa-Brash principle for Osc-LOX and to
observe the changes in the regio- and stereospecificity of the
enzyme. Hence, we carried out site-directed mutagenesis to sub-
stitute 296 alanine with glycine.
2.3. Effects of pH and temperature in enzyme reactions
Purified Osc-LOX-A296G was used to confirm the optimum
catalytic conditions of pH and temperature toward linoleic acid (LA;
C18:2 n-6D9,12, TCI chemicals, Tokyo, Japan). All reactions for assay
were carried out using 3 m mM LA for 10 min.
g mLꢀ1 enzyme and 50
The optimum pH was monitored over a range of pH values from 5.0
to 10.0 at room temperature. The buffers were 2-(N-morpholino)
ethanesulfonic acid (MES) buffer (50 mM, pH 5.0), sodium phos-
phate buffer (50 mM, pH 6.0 and 7.0), Tris-HCl buffer (50 mM, pH
8.0), and sodium tetraborate buffer (50 mM, pH 9.0 and 10.0). To
investigate optimum temperature, the reaction mixtures (enzyme
with 50 mM Tris-HCl buffer, pH 8.0) without substrate were pre-
heated for 30 min at different temperatures ranging from 20 to
60 ꢁC. And then, the activity was measured by adding substrates for
10 min.
2. Materials and methods
2.4. Specific activity and kinetics toward PUFAs
2.1. Mutagenesis and expression of osc-LOX-A296G
The specific activity and kinetic parameters of Osc-LOX-A296G
toward arachidonic acid (AA; 20:4 n-6D5,8,11,14, TCI chemicals,
Tokyo, Japan), eicosapentaenoic acid (EPA; 20:5 n-3D5,8,11,14,17, TCI
chemicals, Tokyo, Japan), and docosahexaenoic acid (DHA; 22:6 n-
3D4,7,10,13,16,19, Glentham Life Science, Corsham, UK) were deter-
mined by monitoring the increase in the formation rate of the
conjugated diene (monohydroperoxide) at absorbance at 234 nm.
The enzymatic reaction was carried out using 50 mM Tris-HCl
The DNA sequence of wild-type osc-lox (accession number
AFZ09286.1, 1713 bp) was found in the GeneBank database. Site-
directed mutagenesis of Osc-LOX derived from Oscillatoria nigro-
viridis PCC 7112 was designed by multiple sequence alignment
analysis with conserved sequence of heterologous. The gene
encoding Osc-LOX-A296G was synthesized by Bioneer Inc. and it
was inserted between NdeI and XhoI (New England Biolabs, Beverly,
MA, USA) sites of the pET-28a plasmid (Novagen, Madison, WI, USA)
buffer (pH 8.0), 3
m
g mLꢀ1 enzyme and varying amounts of the
M at 30 ꢁC for 5 min in a quartz cuvette
substrate from 25 to 125
m
with the N-terminal His-tag. E. coli DH5
a
(RBC, Banqiao, Taiwan) was
(10 mm), and all reactions were repeated three times. The absor-
bance was recorded every 2 s by an Ultrospec 3100 Pro spectro-
photometer (Amersham Biosciences, Cambridge, UK). The
extinction coefficient, ε ¼ 2.3 ꢂ 104 Mꢀ1 cmꢀ1 for the formation of
conjugated diene was used to calculate enzymatic activity [7]. One
Unit of the enzyme was defined by the amount required for the
used to amplify of plasmids number. E. coli BL21 (DE3) (Novagen,
Madison, WI, USA) was transformed with constructed plasmids and
used for enzyme expression. Host cells harboring pET-28a/osc-lox-
A296G were cultivated in 1 L of LB medium containing 50
m
g mLꢀ1
kanamycin at 37 ꢁC. Expression of Osc-LOX-A296G was induced by
adding 0.01 mM isopropyl-
b
-thiogalactopyranoside when reaching
transformation of 1 mmol into hydroperoxy fatty acids (HpFAs) per
minute. For kinetic parameters, reaction rates were calculated from
the initial linear portion of the curve. The Michaelis constant (Km)
an optical density of 0.6 (600 nm), and incubating at 20 ꢁC for 24 h on
shaking incubator at 180 rpm.
and turnover number (kcat
) were calculated by nonlinear-
2.2. Enzyme purification and measurement of molecular weight
regression fits to the Michaelis-Menten equation using the Sig-
maPlot 10.0 software.
After expression, cultured cells containing Osc-LOX-A296G were
harvested by centrifugation at 3500 ꢂ g for 15 min at 4 ꢁC. Cell
pellet was resuspended in 50 mL lysis buffer (50 mM Tris-HCl pH
7.5, 500 mM NaCl,10 mM imidazole, 0.1 mM PMSF, and 5% glycerol).
The resuspended cells were lysed by a sonicator. Cell debris was
eliminated by centrifugation at 13,000 ꢂ g for 30 min at 4 ꢁC and
2.5. Enzyme reactions for conversion of LA and DHA into hydroxy
fatty acids (HFAs)
Enzymatic reactions were performed to convert LA and DHA
into HFAs. The reaction mixture was prepared so as to contain
50 mM substrate in 50 mM Tris-HCl buffer (pH 8.0). For DHA con-
the supernatant was filtered by a syringe filter (0.45
mm). The sol-
uble fraction was loaded onto a cobalt affinity column (HiTrap Talon
crude 5 mL, GE Healthcare, WI, USA) equilibrated with lysis buffer.
Cobalt column was washed by 5 CV wash buffer (50 mM imidazole
added to lysis buffer) to remove nonspecific binding proteins. The
crude proteins were fractionated by increasing concentrations of
imidazole up to 700 mM at flow rate of 5 mL minꢀ1. Fractions
harboring target protein were concentrated by Amicon Ultra-15
(10 kDa cut-off). Concentrated proteins were loaded onto a size-
exclusion chromatography (SEC) column (HiLoad 16/600 Super-
dex 200 prep grade, GE Healthcare, WI, USA) that had been previ-
ously equilibrated with SEC buffer (50 mM Tris-HCl pH 7.5, 150 mM
NaCl) and was separated at a flow rate of 1 mL minꢀ1. Also, SEC was
utilized to confirm the molecular weight and oligomeric state. The
version, enzyme reaction was initiated by adding Osc-LOX-A296G
of 50, 100, and 200 Units per mL of reaction volume, respectively,
while 100 Units were used for LA conversion. The reaction mixture
was incubated at 30 ꢁC with magnetic stirring for 30 min. Hydro-
peroxide products were reduced by the addition of sodium boro-
hydride to a final concentration of 25 mM for 10 min, after which
enzyme reactions were finished by adding glacial acetic acid
(5
m
L mLꢀ1). The converted products were refined using HP20
(Mitsubishi Chemical, Tokyo, Japan) resin one-twentieth of the
reaction volume. The HFAs were identified by HPLC and LC-MS/MS
analysis.
2.6. HPLC analysis
SEC standards, thyroglobulin (670 kDa),
g-globulin (158 kDa),
ovalbumin (44 kDa), myoglobulin (17 kDa), and vitamin B12
(1.35 kDa), were used for calibration. The molecular weight was
calculated by comparison with the elution time of standards.
Agilent 1200 series (Agilent Technologies, Waldbronn, Ger-
many) was utilized for analysis of Osc-LOX-A296G reaction prod-
ucts. Chiral-phase HPLC (CP-HPLC) was performed isocratically on a
894

J.-J. Yi, S.-Y. Heo, J.-H. Ju et al.
Biochemical and Biophysical Research Communications 533 (2020) 893e898
CHIRALCEL OD-H column (25 cm ꢂ 4.6 mm, 5
mm particle size,
DAICEL, Tokyo, Japan) using the solvent consisting of n-hexane/2-
propanol/acetic acid (95:5:0.1, by volume). The flow rate was
1.0 mL minꢀ1, and the column temperature was maintained at
30 ꢁC. Normal-phase HPLC (NP-HPLC) was equipped with a
SUPELCOSIL LC-DIOL column (25 cm ꢂ 3 mm, 5
mm particle size,
SUPELCO, Sigma-Aldrich). A mobile phase consisting of heptane/2-
propanol/acetic acid (95:5:0.1, by volume) was used at a flow rate of
0.5 mL minꢀ1 and a column temperature of 10 ꢁC for NP-HPLC
analysis. All samples were prepared in mobile phase and were
injected 10 mL. The reaction products were detected by monitoring
absorbance at 237, 242, and 270 nm, and identified by comparing
retention times using HFA standards, 9S-, 9R-, 13S-, 13R-hydrox-
yoctadecadienoic acid (HODE), and ( )13-hydroxy docosahexae-
noic acid (13-HDHA) (Cayman Chemical, MI, USA). The amounts of
conversion products were measured by correlating peak areas with
concentrations of reference material in linear calibration curves.
Converted products were purified by a SUPELCOSIL LC-DIOL col-
umn with a fraction collector and they were used for NMR analysis.
2.7. LC-MS/MS analysis
LC-MS/MS analysis of HFAs converted from DHA was performed
by an Agilent TOF spectrometer G6550A (Agilent Technologies,
Santa Clara, CA, U.S.) equipped with an electrospray ionization (ESI)
interface. The conversion products were separated on a ZORBAX
Fig. 1. Amino acid sequence alignment of various S- and R-LOXs with Osc-LOX-A296G.
The highlighted amino acid indicate that Ala is conserved in S-type LOX and Gly is
conserved in R-type LOX.
Eclipse Plus C18 column (10 cm ꢂ 2.1 mm, 1.8
mm particle size,
Agilent) with a isocratic solvent of water/acetonitrile/formic acid
(90:10:0.1, by volume) at a column temperature of 40 ꢁC and a flow
rate 0.3 mL minꢀ1. All analyzed samples were prepared in absolute
A296G was purified by following two-step chromatography: first,
Osc-LOX-A296G was isolated over an imidazole concentration
range of 170e190 mM by cobalt affinity column. (Fig. 2A, middle).
Next, SEC was performed for obtaining further purified protein and
for measurement of protein molecular weight. Osc-LOX-A296G was
eluted at 82 min. The molecular weight was confirmed as
approximately 67.55 kDa compared to the retention times of the
standars, indicating that it exists as a monomer (Fig. 2A, right and
2B). We have obtained 12.5 mg of the purified functional enzyme
with purity of 93% from a 1-L culture, which was 2.5-fold higher
than soybean LOX obtained by E. coli BL21 (DE3) [8]. Moreover,
murine 12R-LOX was expressed in E. coli system at low yield that
could not be used in enzyme reaction [9]. This result suggests that
bacterial LOXs could be efficiently expressed with higher yield
more than eukaryotic LOXs.
ethanol and were injected 2 mL. Sample ionization was performed
in negative mode using the following conditions: nitrogen
(14 L minꢀ1, 200 ꢁC) drying gas; 35 psi nebulizer gas; 3.5 kV
capillary voltage; 350 ꢁC capillary temperature; and 20V frag-
mentor voltage in negative mode. Spectra were recorded over an m/
z range of 0e500 with an accumulation rate of 2 spectra/s. Data
were processed using Mass Hunter Workstation Acquisition soft-
ware and the molecular masses of its fragments were confirmed by
ChemDraw Professional v 15.1.
2.8. Nuclear magnetic resonance (NMR) spectroscopy
NMR analysis was performed on Bruker Avance III HD spec-
trometer (800-MHz) equipped with a 5 mm triple resonance in-
verse (TCI) Cryo Probe. 5e10 mg HFAs were dissolved in 700 mL
methanol-D4 and measured at 298 K. Besides 1D 1H and 13C NMR,
the following 2D-NMR were performed: 1He1H COSY, 1He1H
TOCSY, 1He13C HSQC, and 1He13C HMBC. NMR data were assigned
using NMRViewJ software.
The effects of pH and temperature on the enzyme activity were
investigated by monitoring the conversion yield of HpFA from LA
using the spectroscopic method with 234 nm UV. The maximum
activity on the various pH and temperatures was defined as 100%,
and the relative activity was plotted as a percentage. Osc-LOX-
A296G showed maximum activity at pH 8.0 (Fig. 2C). Some plant-
derived acidic LOXs have been found, but most LOXs prefer
neutral or alkaline pH conditions [10,11]. The optimum tempera-
ture was confirmed to be 30 ꢁC (Fig. 2D), and the enzyme was stable
from 20 to 35 ꢁC without aggregation. However, its activity rapidly
declined above 40 ꢁC and was totally lost at temperatures above
55 ꢁC with aggregation. Although a slight quantitative change
compared to the wild-type, these results showed the altered amino
acids did not affect the optimum activity by pH and temperature.
3. Results and discussion
Site-directed mutagenesis was performed to investigate
whether substitution of amino acid could control catalytic prop-
erties. The amino acid sequence-alignment analysis was carried out
with conserved sequences of previously identified LOXs of animals,
plants, and bacteria. Alanine (S-type) and glycine (R-type) in the
catalytic pocket of LOXs are conserved as stereo- and regiospeci-
ficity determinant known as the Coffa-Brash sequence [5] (Fig. 1).
Wild-type Osc-LOX contains alanine at amino acid 296 and it has
been identified as linoleate 13S-LOX in our previous study. The gene
(1713 bp) encoding mutant Osc-LOX that alanine 296 was replaced
by glycine was cloned into pET-28a. Osc-LOX-A296G was expressed
by E. coli system. Molecular weight of A296G including a His-tag
was theoretically estimated of 67.2 KDa (Fig. 2A, left). Osc-LOX-
The specific activity (
m
mol minꢀ1 mgꢀ1) and catalytic efficiency
(minꢀ1 ꢀ1) toward AA, EPA, and DHA were calculated as follows:
mM
4768.9 and 9.7; 3148.9 and 6.2; 3084.1 and 6.0, respectively
(Table S1). The Michaelis-Menten plots of each substrates are pro-
vided in Fig. S1. These results showed that enzymatic activity was
significantly decreased compared with wild-type. In particular, the
895

J.-J. Yi, S.-Y. Heo, J.-H. Ju et al.
Biochemical and Biophysical Research Communications 533 (2020) 893e898
Fig. 2. Purification and enzymatic characterization of Osc-LOX-A296G. (A) Left, expression and solubility were confirmed by SDS-PAGE. The molecular weight of Osc-LOX-A296G
(67.2 kDa) containing a His-tag was confirmed by comparison with protein marker (lane-1, after induction; lane-2, insoluble friction; lane-3, soluble fraction). Middle, purified
enzyme by cobalt affinity chromatography. Right, highly purified enzyme by size-exclusion chromatography (SEC). (B) The molecular weight and oligomeric state of purified enzyme
were determined using SEC on a HiLoad 16/600 Superdex 200 pg. Target protein was eluted at a position corresponding to 67.5 kDa (filled circle) as a monomer in aqueous solution.
Effect of pH (C) and temperature (D) on the enzymatic activity.
specific activity and catalytic efficiency of the variant toward AA
were reduced by 37.1% and 43.3%, respectively. It has been previ-
ously reported that site-directed mutagenesis, which switches Ala
to Gly, bring about reducing enzymatic activity with changing
regio- and stereochemistry [9].
82%). The final product formed by concentrations of 100e200 Unit
mLꢀ1 was detected at 16.3 min at a lmax of 270 nm. The conversion
of the final product from 13-HDHA was initiated at a concentration
of 100 Unit mLꢀ1, but the full conversion was completed using a
concentration of 200 Unit mLꢀ1, which was yielded 30
mM corre-
To confirm the catalytic effect of substituted amino acid on the
regio- and stereospecificity, we performed enzymatic reactions
with LA. HpFA was generated and its hydroperoxide group was
converted into alcohol groups by sodium borohydride. The HFA
deoxygenated was identified by comparing the retention times
with those of standards using CP-HPLC. The 13S-HODE formed by
wild-type was used as a reference to confirm the changed prop-
erties by A296G. The retention times of four standards, 9S-, 9R-,
13S-, and 13R-HODE, were 12.2, 11.28, 10.8, and 8.7 min, respec-
tively. The HFA produced by Osc-LOX-A296G was eluted at accu-
rately the same time as the 9R-HODE standard (Fig. 3A). The
sponding to a conversion rate of 73% from the first product. The
concentration of two products was determined by reference to a
calibration curve generated using purified products by preparative
HPLC. We have estimated that the final product has a molecular
structure of 13,20-dihydroxy- docosahexaenoic acid. Our assump-
tion could be supported by the following two pieces of evidence.
First, the peak corresponding to 13-HDHA decreased and the latter
peak was newly generated by the increasing concentration of the
enzyme, thus the final product was generated from the first prod-
uct. LOX is known to be an irreversible enzyme, so higher con-
centrations of the enzyme are required to catalyze dihydroperoxy
fatty acid [12]. Finally, the final product was detected at a lmax of
270 nm UV with a triplet spectrum (260, 270, and 280 nm); this
spectrum represents that it has a conjugated triene. The second
hydroxyl group must be introduced into carbon-20 to form a con-
jugated triene, because HFAs have unique spectra reflecting their
conjugated double bonds [13]. Two products were further analyzed
by CP-HPLC comparing with standards, 13S- and 13R-HDHA, to
verify the regio- and stereochemistry. The retention times of the
first and final products were 11 and 52.8 min, respectively (Fig. 3C).
These results indicated that the first product is 13R-HDHA however
the final product did not match any standards.
concentration of 9R-HODE converted from 50 mM LA was 45.8 mM
(conversion rate 91.6%). Osc-LOX-A296G completely converted LA
into 9R-HODE, not 13S-HODE. Thus, these results indicate that the
property of wild-type as linoleate 13S-LOX was totally changed to
linoleate 9R-LOX by substitution Ala to Gly. According to reported
studies, the changes of regio- and stereospecificity, such as that the
13(S or R)-HODE to the 9(R or S)-HODE or vice versa, by the Ala and
Gly exchange were related to the size of amino acid adjacent to the
carbon where oxygen could be inserted [5].
Two products were generated from DHA depending on the
concentration of enzyme ranging from 50 to 200 Unit mLꢀ1
(Fig. 3B). In NP-HPLC, the first product generated at concentration
of 50 Unit mLꢀ1 was eluted at 5.78 min (lmax of 237 nm), which
matched the retention time of the ( )13-HDHA standard. DHA
appears to be totally consumed by enzyme reaction using 50 Unit
mLꢀ1 because the peak of remaining substrate was no detected
(data not shown), however, this result does not mean that DHA was
converted into first product as 100%. The concentration of the first
To identify total molecular mass of two products, we carried out
LC-MS analysis in negative ion mode [M-H-]. First product
confirmed as 13R-HDHA was detected at m/z 343.2 [M-H-] corre-
sponding to molecular mass of a monohydroxy-DHA (344.24 Da;
C
22H32O3), and the final product was detected at m/z 359.2 [M-H-]
corresponding to a dihydroxy-DHA (360.23 Da; C22H32O4) (Fig. S2).
Also, the final product was further analyzed by LC-MS/MS (Fig. 4).
Two fragment peaks at m/z 167 and 193 were formed by cleavage
product converted from 50 mM DHA was 41 mM (conversion rate
896

J.-J. Yi, S.-Y. Heo, J.-H. Ju et al.
Biochemical and Biophysical Research Communications 533 (2020) 893e898
Fig. 3. HPLC analysis of HFAs generated from LA and DHA by Osc-LOX-A296G. (A) CP-HPLC analysis for regio- and stereospecificity determination for linoleic acid. The product
generated by Osc-LOX A296G was identified 9R-HODE. (B) NP-HPLC profile of two products obtained from DHA. First product (13-HDHA) was generated by enzyme concentration of
50 Unit mLꢀ1. Final product was converted from 13-HDHA by enzyme concentration of 200 Unit mLꢀ1. (C) CP-HPLC profile to confirm the stereochemistry of the first product (13R-
HDHA).
between C-12 and C-13, respectively. These peaks indicate that two
hydroxyl groups may exist between C-13 and C-22. Another frag-
ment peak formed by cleavage between C-13 and C-14 was detec-
ted at m/z 137. This fragment including a methyl group contains one
hydroxyl group, and also means that another hydroxyl group was
formed at the C-13. A small fragment generated by cleavage be-
tween C-19 and C-20 was detected at m/z 59, consequently indi-
cating that it was hydroxylated at carbon-20. Thus, the final
product has two hydroxyl groups at C-13 and C-20.
The structural configurations of two products were accurately
determined by 800-MHz NMR spectroscopy. The spectra of 1D-
NMR and 2D-NMR for the first product identified as 13R-HDHA
were shown in Fig. S3 and Fig. S4, respectively, and assigned
chemical shifts were summarized in Table S2. The chemical shifts of
the C13eOH was recorded at 4.13 ppm (1H) and 73.94 ppm (13C) in
2D 1He13C HSQC spectrum (Fig. S7A). The 1H NMR spectrum
showed that a conjugated diene fragment as 14E,16Z was deter-
mined to be the trans-geometry between C-14 and C-15 based on
the coupling constants of J14 and J15 (15.92 Hz). Cis (Z) and trans (E)
coupling appear differently on 1H NMR spectrum [14]. The coupling
constant of the olefinic protons shows approximately ranging of
12e19 Hz in trans-geometry and 7e11 in cis-geometry. Thus, these
results indicated that the chemical structure of the first product
was as 13R-hydroxy-4Z,7Z,10Z,14E,16Z,19Z-docosahexaenoic acid.
The NMR spectra of final product were shown in Fig. S5 and Fig. S6,
and assigned chemical shifts were summarized in Table S2A. The
chemical shifts corresponding to C13eOH and C20eOH were
detected as follows: C13eOH, 4.17 ppm (1H) and 73.81 ppm (13C);
C20eOH, 4.04 ppm (1H) and 75.40 ppm (13C) in 2D 1He13C HSQC
spectrum (Fig. S7B). The 1H NMR spectrum exhibited that the ge-
ometry of conjugated triene fragment was 14E,16Z,18E based on the
coupling constants of J14-J15 (15.92 Hz) and J18-J19 (12.72 Hz).
Based on these results, the chemical structure of the final product
was determined to be 13R,20-dihydroxy-4Z,7Z,10Z,14E,16Z,18E-do-
cosahexaenoic acid (13R,20-diHDHA). However, we have could not
define the stereochemistry of C20eOH due to insufficient data, thus
it would be required additional NMR experiments such as NOESY or
ROESY.
In conclusion, we obtained two HFAs identified as a 13R-HDHA
and a 13R,20-diHDHA from DHA by Osc-LOX-A296G. It was re-
ported that the 13-HDHA inhibited IL-1b, which is a potent pro-
inflammatory cytokine, generating by microglial cells, and the
13,20-diHDHA converted from 13-HDHA was discovered in the
resolution phase of inflammation [15]. It was reported that the 13-
HDHA was synthesized by recombinant human cyclooxygenase-2
expressed in Sf9 insect cells. The 13,20-diHDHA was discovered
as a metabolite converted from 13-HDHA by unknown enzymatic
reaction in vivo [15]. To the best of our knowledge, this is the first
897

J.-J. Yi, S.-Y. Heo, J.-H. Ju et al.
Biochemical and Biophysical Research Communications 533 (2020) 893e898
Fig. 4. LC-MS/MS analysis of 13R,20-dihydroxy-docosahexaenoic acid (13R,20-diHDHA). This fianl product was generated from DHA via 13R-HDHA by Osc-LOX-A296G. The blue
rhombus square indicates total molecular mass of 13R,20-diHDHA and the empty red squares indicate molecular masses of fragments.
503 (2010) 161e174.
[2] J. Groot, G.A. Veldink, J. Vliegenthart, J. Boldingh, R. Wever, B. Van Gelder,
Demonstration by EPR spectroscopy of the functional role of iron in soybean
lipoxygenase-1, Biochim. Biophys. Acta 377 (1975) 71e79.
[3] S.T. Prigge, B.J. Gaffney, L.M. Amzel, Relation between positional specificity
report of efficient production of human endogenous 13R-HDHA and
13R,20-diHDHA through in vitro enzymatic reactions. Thus, our
finding may enable physiological studies and pharmaceutical
research for the 13R,20-diHDHA.
Note: While submitting this manuscript, a paper describing
wild-type LOX was first submitted and is under revision for
publishing.
and chirality in mammalian lipoxygenases, Nat. Struct. Biol.
178e179.
5 (1998)
~
[4] S.A. Gillmor, A. Villasenor, R. Fletterick, E. Sigal, M.F. Browner, The structure of
mammalian 15-lipoxygenase reveals similarity to the lipases and the de-
terminants of substrate specificity, Nat. Struct. Biol. 4 (1997) 1003e1009.
[5] G. Coffa, A.R. Brash, A single active site residue directs oxygenation stereo-
specificity in lipoxygenases: stereocontrol is linked to the position of
oxygenation, Proc. Natl. Acad. Sci. Unit. States Am. 101 (2004) 15579e15584.
[7] C. Cossette, P. Patel, J.R. Anumolu, S. Sivendran, G.J. Lee, S. Gravel, F.D. Graham,
A. Lesimple, O.A. Mamer, J. Rokach, Human neutrophils convert the sebum-
derived polyunsaturated fatty acid sebaleic acid to a potent granulocyte
chemoattractant, J. Biol. Chem. 283 (2008) 11234e11243.
Declaration of competing interest
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
[8] D. Tomchick, P. Phan, M. Cymborowski, W. Minor, T. Holman, Structural and
functional characterization of second-coordination sphere mutants of soybean
lipoxygenase-1 y, Biochemistry 40 (2001) 7509e7517.
Acknowledgments
€
[9] S. Meruvu, M. Walther, I. Ivanov, S. Hammarstrom, G. Fürstenberger, P. Krieg,
R. Pallu, H. Kühn, Sequence determinants for the reaction specificity of murine
This work was supported by grants from the Korea Research
(12R)-lipoxygenase
- targeted substrate modification and site-directed
Foundation
(KRF)
for
Basic
Research
Promotion
mutagenesis, J. Biol. Chem. 280 (2005) 36633e36641.
(2017R1A2B4006016) and from the KRIBB Research Initiative Pro-
gram (KGM5482012). This study also partly supported by the GRRC
program of Gyeonggi province (GRRC-CHA2017-A01, Discovery and
Analysis of Regional Specialized Resources and Operation of
Regional Research Service Center).
[10] S. Hayward, T. Cilliers, P. Swart, Lipoxygenases: from isolation to application,
Compr. Rev. Food Sci. Food Saf. 16 (2017) 199e211.
[11] S. Liu, B. Han, Differential expression pattern of an acidic 9/13-lipoxygenase in
flower opening and senescence and in leaf response to phloem feeders in the
tea plant, BMC Plant Biol. 10 (2010) 228.
[12] O. Rådmark, O. Werz, D. Steinhilber, B. Samuelsson, 5-Lipoxygenase: regula-
tion of expression and enzyme activity, Trends Biochem. Sci. 32 (2007)
332e341.
[13] C. Serhan, J. Dalli, R. Colas, J. Winkler, N. Chiang, Protectins and maresins: new
pro-resolving families of mediators in acute inflammation and resolution
bioactive metabolome, Biochim. Biophys. Acta 1851 (2014).
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.bbrc.2020.09.079.
[14] E. Alexandri, R. Ahmed, H. Siddiqui, M. Choudhary, C. Tsiafoulis,
I. Gerothanassis, High resolution NMR spectroscopy as
a structural and
analytical tool for unsaturated lipids in solution, Molecules 22 (2017) 1663.
[15] C.N. Serhan, S. Hong, K. Gronert, S.P. Colgan, P.R. Devchand, G. Mirick,
R.L. Moussignac, Resolvins, A family of bioactive products of omega-3 fatty
acid transformation circuits initiated by aspirin treatment that counter
proinflammation signals, J. Exp. Med. 196 (2002) 1025e1037.
References
[1] I. Ivanov, D. Heydeck, K. Hofheinz, J. Roffeis, V.B. O’Donnell, H. Kuhn,
M. Walther, Molecular enzymology of lipoxygenases, Arch. Biochem. Biophys.
898