Identification of novel oxidized levuglandin D2in marine red alga and mouse tissue

Article abstract of DOI:10.1194/jlr.M017053

In animals, the product of cyclooxygenase reacting with arachidonic acid, prostaglandin(PG)H₂, can undergo spontaneous rearrangement and nonenzymatic ring cleavage to form levuglandin(LG)E₂ and LGD ₂. These LGs and their isomers are highly reactive γ-ketoaldehydes that form covalent adducts with proteins, DNA, and phosphatidylethanolamine in cells. Here, we isolated a novel oxidized LGD ₂ (ox-LGD₂) from the red alga Gracilaria edulis and determined its planar structure. Additionally, ox-LGD₂ was identified in some tissues of mice and in the lysate of phorbol-12-myristate-13-acetate (PMA)-treated THP-1 cells incubated with arachidonic acid using LC-MS/MS. These results suggest that ox-LGD₂ is a common oxidized metabolite of LGD₂. In the planar structure of ox-LGD₂, H8 and H12 of LGD₂ were dehydrogenated and the C9 aldehyde was oxidized to a carboxylic acid, which formed a lactone ring with the hydrated ketone at C11. These structural differences imply that ox-LGD₂ is less reactive with amines than LGs. Therefore, ox-LGD₂ might be considered a detoxification metabolite of LGD₂. Copyright

Full text of DOI:10.1194/jlr.M017053

Supplemental Material can be found at:
http://www.jlr.org/content/suppl/2011/09/05/jlr.M017053.DC1
.html
Identification of novel oxidized levuglandin D2 in marine
red alga and mouse tissue
Yoshikazu Kanai,* Sadahiko Hiroki,* Hiroyuki Koshino,† Keiichi Konoki,* Yuko Cho,*
Mirriam Cayme,§ Yasuo Fukuyo,** and Mari Yotsu-Yamashita*,1
Graduate School of Agricultural Science,* Tohoku University, Sendai, Miyagi 981-8555, Japan; RIKEN
Advanced Science Institute,† Wako, Saitama 351-0198, Japan; National Fisheries Research and Development
Institute,§ Quezon City 1103, The Philippines; and Asian Natural Environmental Science Center,**
The University of Tokyo, Bunkyo-ku, Tokyo 113-8657, Japan
Abstract In animals, the product of cyclooxygenase react-
ing with arachidonic acid, prostaglandin(PG)H2, can un-
dergo spontaneous rearrangement and nonenzymatic ring
cleavage to form levuglandin(LG)E2 and LGD2. These LGs
and their isomers are highly reactive ␥-ketoaldehydes that
form covalent adducts with proteins, DNA, and phosphati-
dylethanolamine in cells. Here, we isolated a novel oxidized
LGD2 (ox-LGD2) from the red alga Gracilaria edulis and de-
termined its planar structure. Additionally, ox-LGD2 was
identified in some tissues of mice and in the lysate of phor-
bol-12-myristate-13-acetate (PMA)-treated THP-1 cells incu-
bated with arachidonic acid using LC-MS/MS. These results
suggest that ox-LGD2 is a common oxidized metabolite of
LGD2. In the planar structure of ox-LGD2, H8 and H12 of
LGD2 were dehydrogenated and the C9 aldehyde was oxi-
dized to a carboxylic acid, which formed a lactone ring with
the hydrated ketone at C11. These structural differences
imply that ox-LGD2 is less reactive with amines than LGs.
Therefore, ox-LGD2 might be considered a detoxification
metabolite of LGD2.—Kanai, Y., S. Hiroki, H. Koshino, K.
Konoki, Y. Cho, M. Cayme, Y. Fukuyo, and M. Yotsu-Yamashita.
Identification of novel oxidized levuglandin D2 in marine
red alga and mouse tissue. J. Lipid Res. 2011. 52:
2245–2254.
donic acid, to generate a large array of lipid peroxida-
tion products, including a series of stereoisomers and
regioisomers of PGH2 (H2-isoprostanes) (Fig. 1) (2, 3). These
H2-isoprostanes undergo nonenzymatic rearrangement to
form D2- and E2-isoprostanes, as well as isolevuglandins as
shown in Fig. 1. Isolevuglandins (isoLG) are sometimes
referred to by the chemically inaccurate name of isoketals.
LGs and isoLGs are the most reactive aldehydes in the
mammalian biological system, reacting almost instanta-
neously to form covalent adducts with free amines in pro-
teins (4–6), aromatic amines in DNA (7, 8), and the
ethanolamine headgroup of phosphatidylethanolamine
(9, 10). The levels of LG and isoLG in these protein and
DNA adducts are known to increase in several pathologi-
cal conditions (11, 12).
Marine red algae of the genus Gracilaria are unique plants
containing prostaglandins. PGE2, PGA2, PGF2␣, 15-keto
PGE2, and PGE1 were found in G. lichenoides (13), G. ver-
rucosa (vermiculophylla), G. choda (14, 15), and G. asiatica
(16). PGs in G. verrucosa and G. chorda were suspected as
causative toxins for lethal food poisoning that occurred in
Japan (14, 17). Although the biosynthetic and metabolic
pathways and functions of PGs in red algae have not been
fully clarified, it has been suggested that, in G. vermiculo-
phylla, arachidonic acid is released from the membrane
lipids, such as monogalactosyldiacylglycerol and digalacto-
syldiacylglycerol, by the action of glycerolipid acyl-hydro-
lase (Fig. 1) (18). LG, isoLG, and isoprostane esters have
not yet been identified in red algae.
Supplementary key words prostaglandins • Gracilaria • LC-MS/MS •
NMR
In animals, the product of cyclooxygenase (COX) react-
ing with arachidonic acid, prostaglandin(PG)H2, can un-
dergo spontaneous rearrangement and nonenzymatic
ring cleavage to form highly reactive levuglandin (LG)E2
or LGD2 (1), if PGH2 is not utilized by cellular PG syn-
thases. However, reactive oxygen species (ROS) can also
interact with polyunsaturated fatty acids, such as arachi-
Abbreviations: COSY, correlation spectroscopy; HMBC, heteronu-
clear multiple-bond correlation; HSQC, heteronuclear single-quantum
coherence; LG, levuglandin; isoLG, isolevuglandin; ox-LGD2, oxidized
levuglandin D2; MRM, multiple reaction monitoring; PG, prostaglandin;
PMA, phorbol-12-myristate-13-acetate; ROESY, rotating-frame Overhauser
enhancement spectroscopy; TOCSY, total correlation spectroscopy.
1 To whom correspondence should be addressed.
This work was supported by the Cabinet Office of the Government of Japan
through its funding program for the Next-Generation World-Leading Research-
ers (LS012) and by the Japan Society for the Promotion of Science through Grant-
in-Aid for Scientific Research No. 21380070.
Manuscript received 10 May 2011 and in revised form 30 August 2011.
e-mail: myama@biochem.tohoku.ac.jp
Published, JLR Papers in Press, September 5, 2011
DOI 10.1194/jlr.M017053
The online version of this article (available at http://www.jlr.org)
contains supplementary data in the form of four figures.
Copyright © 2011 by the American Society for Biochemistry and Molecular Biology, Inc.
This article is available online at http://www.jlr.org
Journal of Lipid Research Volume 52, 2011
2245

Supplemental Material can be found at:
http://www.jlr.org/content/suppl/2011/09/05/jlr.M017053.DC1
.html
Fig. 1. Pathways leading to the formation of PGs, LGs, and isoLGs esters in the red algae genus Gracilaria
and animals. PL, phospholipid; GL, galactolipid.
then 200 ml). After the volatile substance in the ethyl acetate ex-
tract was evaporated under low pressure, ox-LGD2 was purified by
sequential reversed phase chromatography. First, a Cosmosil
5C18AR-II (i.d. 1.0 × 25 cm, Nacalai tesque, Tokyo, Japan) was
used with aqueous 20 mM ammonium formate and methanol
(1:1-3:7, v/v, gradient, flow rate of 1 ml/min), and then a Develo-
sil ODS-SR-5 (i.d. 0.8 × 25 cm, Nomura Chemical, Seto, Japan)
was used with aqueous 20 mM ammonium formate and methanol
(4:6, v/v, flow rate of 1 ml/min). Lastly, a Mightysil RP-18 GPII
(i.d. 0.46 × 25 cm, 5 ␮m, Kanto Chemical, Tokyo, Japan) with an
acetonitrile-water-acetic acid mixture (40:60:0.1, v/v/v, flow rate
of 0.5 ml/min) was used. After repeating the same purification
procedure four times, pure ox-LGD2 (ف0.1 mg, estimated by
1H NMR) was obtained from a total of 800 g of wet G. edulis.
In the present study, we found a novel oxidized deriva-
tive of LGD2, named ox-LGD2, in the red alga Gracilaria
edulis, and we determined its planar structure by spectro-
scopic methods. Furthermore, we identified ox-LGD2 in
some tissues of mice using LC-MS/MS and in a phorbol-
12-myristate-13-acetate (PMA)-stimulated THP-1 cell lysate
incubated with arachidonic acid.
MATERIALS AND METHODS
Materials
Authentic PGs were purchased from Cayman Chemical (Ann
Arbor, MI). All solvents were HPLC grade and purchased from
Wako Pure Chemical Industries (Osaka, Japan). 3,5-Di-tert-butyl-
4-hydroxytoluene (BHT) was obtained from Sigma Chemical Co.
(St. Louis, MO). Other reagents were special grade and pur-
chased from Wako Pure Chemical Industries. Gracilaria edulis was
collected in December 2004 at a beach in La Union, The Philip-
pines. The frozen alga was transported to Tohoku University, Ja-
pan, and kept at Ϫ20°C until use. G. vermiculophylla was collected
in May 2005 in Sendai Bay, Japan.
Structural determination of ox-LGD2
NMR spectra were obtained on a Varian 600 MHz NMR spec-
trometer (Agilent Technologies, Santa Clara, CA) with CD3CN as
1
the solvent. The signals of CHD2CN at 1.9 ppm in the H NMR
spectra and those of CD313CN at 118 ppm in the 13C NMR spectra
were used as the internal references. Signals were assigned based
on the analyses of correlation spectroscopy (COSY), total correla-
tion spectroscopy (TOCSY; mixing time 80 ms), heteronuclear
single-quantum coherence (HSQC), heteronuclear multiple-
bond correlation (HMBC), and rotating-frame Overhauser
enhancement spectroscopy (ROESY; mixing time 400 ms) spec-
tra. High-resolution fast atom bombardment (HRFAB)-MS (neg-
ative, matrix:m-nitrobenzyl alcohol) was recorded by a JEOL
JMS700 MS Station (Akishima, Japan). The UV spectrum of ox-
LGD2 was measured in methanol by a Shimadzu UV1800 spec-
trometer (Kyoto, Japan). LC-MS/MS was performed by an
API2000 mass spectrometer (AB SCIEX, Foster City, CA) with an
ESI ion source in the negative ion mode.
Purification of ox-LGD2 from the red alga G. edulis
The presence of a novel PG-related compound in G. edulis was
suggested by an unknown ion detected at m/z 365 in an ESI-MS
spectrum of the ethyl acetate extract from G. edulis, operating in
the negative ion mode and scanning between m/z 100 and 500.
The compound that gave this unknown ion at m/z 365 was puri-
fied from this alga as follows. PGs and other polar lipids were
extracted from G. edulis according to the method reported by
Fusetani and Hashimoto (14), with slight modifications. Frozen
G. edulis (200 g wet weight) was cut into approximately 1 cm pieces
and soaked in water (600 ml) at room temperature overnight.
The supernatant of the water extract, obtained by centrifugation
at 1,000 g at 4°C for 10 min, was adjusted to pH 3.5 with 1 N HCl,
and then the lipids were extracted with ethyl acetate (300 ml and
Chemical derivation of ox-LGD2 from PGH2
Ox-LGD2 was chemically derived from PGH2 by two-step reac-
tion. First, LGs (as a mixture with PGE2 and PGD2) were de-
rived from PGH2 (5 ␮g) by incubation in 1.5 ␮l of DMSO at 37°C
2246
Journal of Lipid Research Volume 52, 2011

Supplemental Material can be found at:
http://www.jlr.org/content/suppl/2011/09/05/jlr.M017053.DC1
.html
for 1 h (19). Second, we applied Pinnick oxidation (20) to oxi-
4 weeks if age, 20-21 g body weight) purchased from Japan SLC
(Hamamatsu, Japan) were euthanized with diethyl ether. The tis-
sues (testis, liver, kidney, lung, heart, brain, and muscle) from
five mice (labeled mouse A–E) and the liver from one mouse
(mouse F) were dissected on ice and immediately frozen in liquid
nitrogen. Eicosanoids were extracted according to the method
utilized by Yang et al. (23) with slight modifications. Briefly, the
frozen tissues were ground to fine powders using a mortar in liq-
uid nitrogen. Parts of the ground tissues (20-25 mg) were trans-
ferred into sealed microcentrifuge tubes, and three volumes of
ice-cold PBS buffer containing 0.1% (w/v) BHT and 1 mM EDTA
were added to the tissue samples. The samples were then homog-
enized by an ultrasonic processor (Vibra cell, Sonics and Materi-
als Inc., Danbury, CT) on ice for 2 min. The homogenate was
transferred into a glass tube (i.d. 1.0 × 10 cm), and 20 ␮l of 1 N
citric acid was added. The eicosanoids were then extracted with 1 ml
of hexane-ethyl acetate (1:1, v/v) and mixed by vortex for 2 min.
Samples were centrifuged at 1,800 g at 4°C for 10 min, and then
the upper organic layer was collected. Next, the solvent was evap-
orated to dryness under a stream of nitrogen gas at room tem-
perature, and then the residue was dissolved in 200 ␮l of
methanol. An aliquot (10 ␮l) was injected into an LC-MS/MS in
MRM mode for quantitation of PGs and ox-LGD2 (see below).
For the samples in which ox-LGD2 was not detected, the remain-
ing sample solutions were concentrated to 70 ␮l by a stream of
nitrogen gas, and an aliquot (20 ␮l) was applied to LC-MS/MS
(MRM) again. The final limit of detection for PGE2, PGD2, and
ox-LGD2 was 0.004, 0.008, and 0.004 ␮g/g tissue, respectively. Af-
ter MRM analysis, the remaining sample solution (19/20, v/v)
prepared from the liver of mouse B was concentrated to 40 ␮l,
and 20 ␮l of this solution was injected on column for product ion
scan analysis for ox-LGD2 ([M-H]Ϫ m/z 365).
dize aldehyde in LGs to carboxylic acid. To the reaction mixture
of the first reaction, 1 ␮M KH2PO4 (10 ␮l, aqueous), water (10 ␮l),
t-butanol (30 ␮l), 2-methyl-2-butene (10 ␮l), and 56 mM NaClO2
(5 ␮l, aqueous) were added, and the mixture was kept at 10°C for
1 h. After the reactions, the mixture was acidified with 1 N HCl, and
lipophilic compounds were extracted with ethyl acetate (500 ␮l).
The ethyl acetate layer was washed with 500 ␮l of water and dried
by a stream of nitrogen gas. The resultant residue was dissolved
in 200 ␮l of methanol, and an aliquot (10 ␮l) of the methanolic
solution was applied to LC-MS/MS in the multiple reaction mode
(MRM) as described below.
Examination of reactivity of ox-LGD2 with the thiol
in cysteine
As ox-LGD2 has conjugated double bonds with carbonyl group,
purified ox-LGD2 was examined for its reactivity with the thiol in
cysteine as a nucleophile under the similar reaction condition
reported for the Michael addition of glutathione to ⌬7-PGA1
methyl ester (21). Purified ox-LGD2 (12 ng, 0.033 nmol) dis-
solved in 0.05 ml of methanol was mixed with a solution of
L-cysteine (4 ng, 0.033 nmol) in 0.1 M phosphate buffer (pH 7.4,
0.05 ml) and incubated at 30°C for 4 h. For reference, purified
ox-LGD2 was similarly incubated without L-cysteine. Next, metha-
nol was removed from the reaction mixture by a stream of nitro-
gen gas, and the remaining ox-LGD2 was extracted with ethyl
acetate from the aqueous layer acidified with 1 N HCl. After re-
moval of ethyl acetate by stream of nitrogen gas, the remaining
residue was dissolved in methanol (0.1 ml), and an aliquot (10 ␮l)
of this solution was applied to LC-MS/MS (MRM) to quantify
unreacted ox-LGD2. This experiment was performed three times.
Preparation of crude PG mixtures from red algae
To examine the effect of BHT on the level of ox-LGD2, the
frozen and ground liver from mouse F was divided into two por-
tions, and a part (20 mg) of each was homogenized in the same
buffer as described above with or without BHT. The sample solu-
tions for LC-MS/MS (MRM) analysis were prepared as described
above.
PGs were extracted from G. edulis and G. vermiculophylla by two
different methods for comparison: water extraction and ethanol
extraction. The frozen algae (10 g) were finely (2-3 mm) and
quickly cut on ice and mixed completely. Next, 1.0 g was used for
each extraction method. The water extraction was prepared ac-
cording to Fusetani and Hashimoto (14) as follows. The algae
(1.0 g) were homogenized with 8 ml of water containing 0.002%
(w/v) BHT at 20°C for 60 s. The homogenate was centrifuged at
2,000 g for 15 min. The pH of the supernatant was adjusted to 3.0
with 1N HCl, and the PGs were extracted with 8 ml of ethyl ace-
tate. After washing the ethyl acetate layer with 0.8 ml of water, the
solvent was evaporated under nitrogen gas, and the residue was
dissolved in 2 ml of methanol. The ethanol extraction was pre-
pared according to a method by Powell (22). Briefly, the algae
(1.0 g) were homogenized in 10 ml of ethanol-water 4:1 (v/v)
containing 0.002% (w/v) BHT at 20°C for 60 s. The homogenate
was shaken at 5°C for 1 h and then centrifuged at 1,000 g for
10 min. A 3 ml volume of the ethanol soluble fat was diluted with
the same volume of water, and the pH was adjusted to 3.0 with 1N
HCl. A 2 ml volume of the acidic solution was applied to an ODS-
silica Sep-Pak cartridge (Waters, Milford, MA), and polar and
nonpolar lipids were removed from the column with 10 ml of
ethanol-water 15:85 (v/v) and 10 ml of petroleum ether. Next,
the PGs were eluted with 6 ml of ethyl acetate. The solvent was
evaporated under a stream of nitrogen gas, and the residue was
dissolved in 2 ml of methanol. Aliquots of the water (1/200) and
ethanol (1/1000) extracts, each from 1.0 g (wet mass) of the al-
gae, were injected into an LC-MS/MS in multiple reaction mode
(MRM) as described below.
Formation of ox-LGD2 in the lysate of PMA-treated
THP-1 cells
The human monocytic leukemia cell line THP-1 was obtained
from the Cell Resource Center for Biomedical Research, Tohoku
University, and RIKEN BioResource Center (Tsukuba, Japan). The
cells were maintained in a RPMI 1640 culture medium contain-
ing 10% (v/v) fetal bovine serum and penicillin/streptomycin
Animal experiments
Animal experiments were approved by the Animal Ethical
Committee of Tohoku University. Six normal mice (ddY, male,
Fig. 2. COSY spectrum of ox-LGD2 (0.1 mg, CD3CN, 600 MHz).
The signal corresponding to CHD2CN was set at 1.9 ppm.
Identification of novel oxidized levuglandin D2
2247

Supplemental Material can be found at:
http://www.jlr.org/content/suppl/2011/09/05/jlr.M017053.DC1
.html
Fig. 3. HMBC spectrum of ox-LGD2 (0.1 mg, CD3CN, 600 MHz, 2,3JCH=8 Hz, 55 h). The cross peak of H13/
C8 was shown more clearly in the spectrum of the crude ox-LGD2 (data not shown), supporting the assignment
of C8. The signal of CHD2CN was set at 1.9 ppm and that of CD313CN was set at 118 ppm as references.
(100 U/ml, 100 ␮g/ml). The cells were differentiated into mac-
rophage-like phenotypes by incubation with 40 nM of phorbol-
12-myristate-13-acetate (PMA) at 37°C for 72 h in 5% CO2 (24).
After washing with PBS, the cells were harvested and frozen at
Ϫ80°C. Next, the cells (7.5 × 105 cells) were thawed in 1 ml of 40 mM
Tris-HCl (pH 7.8) buffer containing 2 mM EGTA, 5 mM trypto-
phan, and 2 ␮M hematin (24). An ethanolic solution of arachi-
donic acid (2 ␮g/2 ␮l) was added to this cell suspension, and
the suspension was vigorously vortex-mixed. After incubation at
25°C for 5 min, the cell suspension was acidified by addition of
1 N HCl (50 ␮l), and the lipids were extracted by 4 ml of ethyl
acetate. Ethyl acetate was evaporated by a stream of nitrogen
gas, and the obtained residue was dissolved in 200 ␮l of metha-
nol. An aliquot (10 ␮l) of this methanolic solution was injected
into an LC-MS/MS in MRM mode for PGE2 and ox-LGD2 (see,
below).
RESULTS
Planar structure of ox-LGD2
Pure ox-LGD2, obtained from G. edulis, showed a UVmax
at 267 nm (␧ 25,000) in methanol, suggesting the presence
of a conjugated triene- or dienone-like structure. Ox-LGD2
has a molecular formula C20H30O6, as determined by HR-
FAB-MS (found at m/z 365.1964 [M-H]Ϫ; calculated for
C20H29O6 at 365.1964), which is indicative of six degrees of
1
unsaturation. The H NMR (supplementary Fig. I), COSY
(Fig. 2), and TOCSY (supplementary Fig. II) spectra were
measured to elucidate the structure of ox-LGD2. Two car-
bon chains of C2-C7 and C13-C20, which contain double
bonds at C5-C6 and C13-C14, respectively, were deduced by
analyses of the COSY and TOCSY. The stereochemistry of
the latter double bond was determined to have an E geom-
etry due to its large vicinal 3JHH value of 16.2 Hz between the
olefinic proton signals of H13 and H14. Additionally, H14
had a correlation with an oxymethine proton signal at 4.19
ppm (H15) in the COSY. These partial structures suggest
that ox-LGD2 is an analog of PG, although a singlet methyl
proton signal (Me10) at 1.58 ppm was also observed. In the
HMBC spectrum (Fig. 3), obvious long-range correlations
LC-MS/MS analysis of ox-LGD2 and PGs
The ion transitions m/z 365/177 and 365/195 were monitored
for quantitation of ox-LGD2 by LC-MS/MS analysis in MRM
mode. For further identification of ox-LGD2, ox-LGD2 in the liver
of mouse B was analyzed in product ion scan mode, setting
[M-H]Ϫ m/z 365 as the precursor ion and scanning between m/z
100 and 400. The collision energy was set at Ϫ20 eV.
PGs were quantified by LC-MS/MS in MRM mode by follow-
ing the method of Cao (25) with slight modification. The PGs
and ox-LGD2 were separated using a Cosmosil 5C18AR-II
(i.d. 0.46 × 15 cm) column with acetonitrile-water-acetic acid
(35:65:0.1, v/v/v) with a flow rate of 0.3 ml/min at 25°C for the
algal samples to separate PGA2 and PGB2 (Fig. 8). For the analy-
sis of the chemical reaction products derived from PGH2
(Fig. 6), mouse tissues (Fig. 10) and the lysate of the THP-1 cells
(Fig. 11), an alternative LC solvent, acetonitrile-water-acetic acid
(40:60:0.1, v/v/v), was used for faster elution. The ion transi-
tions monitored for PGE2 and PGD2 were m/z 351/333 and
351/271, and those for PGA2, PGB2, and PGJ2 were m/z 333/315
and 333/271. The collision energy was set at Ϫ15 eV for the PGs
and at Ϫ20 eV for ox-LGD2. The calibration curves were con-
structed from the different amounts of each authentic PG and
purified ox-LGD2 injected into the LC-MS/MS instrument and
their peak areas in the MRM chromatograms. The calibration
curve obtained for each PG and ox-LGD2 showed good linearity
(r2 = 0.99).
Fig. 4. Planar structure of ox-LGD2. A: Ox-LGD2 with key HMBC
and ROESY correlations. B: PGB2. C: Predicted structure for
ox-LGE2.
2248
Journal of Lipid Research Volume 52, 2011

Supplemental Material can be found at:
http://www.jlr.org/content/suppl/2011/09/05/jlr.M017053.DC1
.html
TABLE 1. NMR spectral data of ox-LGD2 and PGB2 in CD3CN
ox-LGD2
PGB2
Multiplicity (coupling
Multiplicity (coupling
Position
␦
H
constant J)
␦
C
␦
constant J)
␦
C
H
1
—
2.25
1.60
2.14
5.40
5.39
3.04, 2.95
—
ND
33.0
24.7
26.8
131.1
125.6
21.5
127.2
ND
25.2
105.6
155.6
117.5
145.5
71.7
37.5
25.5
32.2
23.1
14.0
—
175.0
33.4
25.2
2
m
m
m
m
m
m
2.25
1.58
2.17
5.31
5.26
2.94
—
m
m
m
m
m
m
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
26.9
130.3
127.8
21.9
139.7
209.1
26.2
—
1.58
—
—
s
2.60
2.28
—
m
dd (10.9, 6.8)
34.1
—
164.3
123.4
143.9
72.1
37.7
26.2
32.4
23.1
14.3
6.46
6.39
4.19
1.48
1.36
1.27
1.27
0.85
d (16.2)
dd (16.2, 5.4)
6.79
6.29
4.20
1.48
1.37
1.27
1.27
0.85
d (15.8)
dd (15.8, 5.9)
brt
m
m
m
m
q (5.9)
m
m
m
m
t (7.0)
t (7.0)
Chemical shifts are in ␦ values. Multiplicities and coupling constant J (in parentheses) are in Hertz. CD313CN
at 118.0 ppm and CHD2CN at 1.90 ppm. 13C at 150 MHz and 1H at 600 MHz. 13C chemical shift was roughly
determined by HSQC and HMBC correlations. ND, not determined; m, multiple; d, doublet; dd, double doublet;
t, triplet; brt, broad triplet; s, singlet; q, quartet.
Based on these data (Table 1, see also HSQC spectrum in
supplementary Fig. III), the planar structure of ox-LGD2
was elucidated as shown in Fig. 4A. From the molecular ion
of ox-LGD 2 ([M-H] m/z 365), the major product ions were
shown at m/z 177 and 195 (Fig. 5A), which were inter-
preted as shown in Fig. 5B.
The structure of ox-LGD2 was also confirmed by chemi-
cal derivation of ox-LGD2 from PGH2 as shown in Fig. 6A.
The produced ox-LGD2 was shown as the peak at the same
retention time (24.5 min) as that of purified ox-LGD2
(Fig. 6B) by LC-MS/MS (MSM) monitoring the ion transi-
tion of m/z 365/195 and 365/177. The yield of ox-LGD2
from PGH2 was roughly estimated as 1% (mol/mol) with-
out optimization.
to four quaternary carbons at ␦ 105.6 (C11), 127.2 (C8),
155.6 (C12), and 174.9 (C1) ppm were observed. The
chemical shifts of C13 and C14 (117.5 and 145.5 ppm) with
H13 and H14 (6.46 and 6.39 ppm) and C8 and C12 (127.2
and 155.6) suggests that C13, C14, C8, and C12 are conju-
gated olefin carbons. HMBC correlations clarified these
connectivities around the quaternary and carbonyl carbons
by giving cross peaks between C8/H13, C11/Me10, C11/
H13, C12/Me10, C12/H13, and C12/H14 (Fig. 4A). These
HMBC correlations suggested that ox-LGD2 has a fully sub-
stituted double bond between C8 and C12, and it is conju-
gated with another double bond between C13 and C14 and
a methyl-substituted hemiacetal carbon at C11 (Fig. 4A).
The remaining substitution group at C8 was speculated to
be a carbonyl group on the basis of the chemical shift values
of C8 (127.2 ppm), C12 (155.6 ppm), C13 (117.5 ppm),
and C14 (145.5 ppm), in which the ␤ and ␦ position car-
bons (C12 and C14) were shifted to low field. Although the
presence of a carbonyl group at C9 was not directly indi-
cated by the C9/CH2-7 HMBC correlation, the molecular
formula of ox-LGD2 (C20H30O6) and the resemblance of
the UV and NMR spectral data of ox-LGD2 with that of
PGB2 (Fig. 4B; PGB2: UVmax at 269 nm, ␧ 28500 in methanol;
NMR data of PGB2, see Table 1) strongly suggested the
presence of a conjugated carbonyl group at C9. Chemical
shift differences for C8 and C12 between ox-LGD2 and
PGB2 are due to the structural differences between the ␣,␤-
unsaturated-␥-lactone and the cyclopentenone. The LGE2-
type structure (ox-LGE2, Fig. 4C) was excluded as a possible
structure for ox-LGD2 because the HMBC correlation be-
tween the hemiacetal carbon at 105.6 ppm and H13 should
Fig. 5. MS/MS analysis of ox-LGD2. A: Product ion spectrum of
the [M-H]Ϫ ion at m/z 365 of ox-LGD2. Purified ox-LGD2 (18 ng)
was injected directly into an MS/MS by an LC pump with methanol
(0.2 ml/min). Collision energy: Ϫ20 eV. B: Interpretation of the
major product ions.
4
be explained by unlike JCH correlation in this structure.
The Z geometry of the C5-C6 double bond was indicated by
ROE observed between H4/CH2-7 (Fig. 4A) in the ROESY.
Identification of novel oxidized levuglandin D2
2249

Supplemental Material can be found at:
http://www.jlr.org/content/suppl/2011/09/05/jlr.M017053.DC1
.html
Fig. 6. Chemical derivation of ox-LGD2 from PGH2. A: Scheme of chemical derivation of ox-LGD2 from
PGH2. B: LC-MS/MS (MRM) chromatograms of purified ox-LGD2 (2 ng) and chemically derived ox-LGD2
obtained by monitoring the ion transition of m/z 365/195. LC was performed using a Cosmosil 5C18ARII
(i.d. 0.46 × 15 cm) column with acetonitrile-water-acetic acid (40:60:0.1, v/v/v) with a flow rate of 0.3 ml/
min at 25°C.
Examination of reactivity of ox-LGD2 with the
chromatograms for the PGs and ox-LGD2 in the ethanol
extracts from the two species of red algae are shown in Fig.
8. Additionally, the contents of the PGs and ox-LGD2 in
their respective water and ethanol extracts are summarized
in Table 2. Although PGE2 and PGA2 were detected in both
of these algae, ox-LGD2 and PGB2 were identified only in
G. edulis, not in G. vermiculophylla, even though the level of
PGE2 was much higher in G. vermiculophylla than in G. edu-
lis. PGD2 and PGJ2 were less than their limits of detection,
0.056 and 0.13 ␮g/g, respectively, in both G. edulis and G.
vermiculophylla. Except for PGA2, PGs and ox-LGD2 were
higher in the ethanol extract than in the water extract.
thiol in cysteine
Typical MRM mass chromatograms of purified ox-LGD2
incubated with or without cysteine are shown in Fig. 7. The
mean and SD values of recovered ox-LGD2 after incuba-
tion with or without cysteine in three repeated experi-
ments were 8.5 0.5 and 9.2 0.7 ng, respectively, which
were not significantly different. These data suggest that ox-
LGD2 is not reactive with the thiol in cysteine under this
condition,eventhoughox-LGD2 possessesan␣,␤-unsaturated
carbonyl group. Probably the presence of two constituents
at C8 and C12, composed of a long carbon chain, decreases
its reactivity with nucleophiles.
Identification and quantitation of ox-LGD2 in
mouse tissues
Quantitation of PGs and ox-LGD2 in red algae
PGs were extracted by two different methods from two
species of red algae, Gracilaria edulis and G. vermiculophylla,
and quantified using LC-MS/MS in MRM mode. MRM ion
We examined the presence of ox-LGD2, PGE2, and PGD2
in seven tissues (testis, liver, kidney, lung, heart, brain, and
muscle) of five normal ddY strain male mice (labeled
Fig. 7. Reactivity of ox-LGD2 with the thiol in cysteine. Purified ox-LGD2 (12 ng, 0.033 nmol) in 0.05 ml of
methanol was mixed with the solution of L-cysteine (4 ng, 0.033 nmol) in 0.1 M phosphate buffer (pH 7.4,
0.05 ml) and incubated at 30°C for 4 h. For reference, purified ox-LGD2 was similarly incubated without
L-cysteine. After incubation, remaining ox-LGD2 was quantified using LC-MS/MS (MRM). The mean and SD
values (n = 3) of recovered ox-LGD2 after incubation with or without cysteine were 8.5 0.5 and 9.2 0.7 ng,
respectively, which were not significantly different between them. Typical MRM mass chromatograms of ox-
LGD2 incubated with (B) and without (A) cysteine are shown. LC was performed using a Cosmosil 5C18ARII
(i.d. 0.46 × 15 cm) column with acetonitrile-water-acetic acid (40:60:0.1, v/v/v) with a flow rate of 0.3 ml/
min at 25°C. MA, measured peak area; MH, measured peak height; RT, retention time.
2250
Journal of Lipid Research Volume 52, 2011

Supplemental Material can be found at:
http://www.jlr.org/content/suppl/2011/09/05/jlr.M017053.DC1
.html
Fig. 8. LC-MS/MS (MRM) analysis of PGs and ox-LGD2 in red algae. MRM chromatograms obtained by monitoring the ion transition of
m/z 351/271 for PGE2 and PGD2; m/z 333/271 for PGA2, PGJ2, and PGB2; and m/z 365/195 for ox-LGD2. A: Authentic mixture of PGs (each
10 ng) and ox-LGD2 (10 ng). B: Ethanol extract of G. edulis. C: Ethanol extract of G. vermiculophylla. LC was performed using a Cosmosil
5C18ARII (i.d. 0.46 × 15 cm) column with acetonitrile-water-acetic acid (35:65:0.1, v/v/v) with a flow rate of 0.3 ml/min at 25°C.
mouse A–E) using LC-MS/MS in MRM mode. As shown in
Fig. 9, ox-LGD2 was detected (у0.004 ␮g/g) in all tissues
of three mice (A, B, C) but only in liver and lung of mouse
E (0.004 ␮g/g), whereas ox-LGD2 was less than limit of
detection (<0.004 ␮g/g) in the tissues of mouse D and the
tissues of mouse E (except liver and lung). The content of
ox-LGD2 in the mice tissues was not related to the content
of PGE2 and PGD2, and it was different in individual mice.
By comparison among the tissues, ox-LGD2 tended to be
high in liver and lung, whereas PGE2 and PGD2 were sig-
nificantly higher in lung (and brain for PGD2) than other
tissues. Representative MRM ion chromatograms of PGE2
and ox-LGD2 in the liver of mouse B are shown in Fig. 10C,
displaying the peak of ox-LGD2 at 24.6 min, almost same
retention time (24.5 min) of purified ox-LGD2 from the red
alga (Fig. 10A). Ox-LGD2 in the liver of mouse B was further
identified by the product ion scan spectrum for the peak
of ox-LGD2 (24.0-25.0 min) (Fig. 10D), showing the
same fragmentation pattern as that of purified ox-LGD2
(Fig. 10B).
␮g/g, respectively. When the same sample was homoge-
nized in the buffer without BHT, the values were 0.017
␮g/g (PGE2) and 0.13 ␮g/g (ox-LGD2), almost same as
those with BHT.
Formation of ox-LGD2 in the lysate of PMA-treated
THP-1 cells
In the lipophilic extract from the lysate of PMA-treated
THP-1 cells (7.5 × 105 cells), after incubation with arachi-
donic acid (2.0 ␮g) at 25°C for 5 min, the peaks corre-
sponding to PGE2 and ox-LGD2 were found at 15.3 and
24.7 min, respectively, in the MRM ion chromatograms, as
detected by the ion transitions m/z 351/271 and m/z
365/195 (Fig. 11). The amounts of PGE2 and ox-LGD2
were quantified as 13.6 ng and 3.8 ng, respectively. PGE2
and ox-LGD2 were less than the limits of detection (PGE2,
0.1 ng; ox-LGD2, 0.3 ng) when arachidonic acid was not
TABLE 2. Contents of PGs and ox-LGD2 in Gracilaria edulis
and G. vermiculophylla
Effect of BHT in the homogenization buffer on the level
of ox-LGD2
Concentrations (␮g/g wet mass)
G. edulis
G. vermiculophylla
Prostag
landin
The level of ox-LGD2 in mouse liver extracted with the
buffer in the presence or absence of BHT was determined
using LC-MS/MS in MRM mode to examine whether ox-
LGD2 is produced by autoxidation. In the lipophilic ex-
tract from the liver of mouse F homogenized in PBS buffer
containing 1 mM EDTA and 0.1% (w/v) BHT, the con-
tents of PGE2 and ox-LGD2 were 0.017 ␮g/g and 0.12
Water extract Ethanol extract Water extract Ethanol extract
PGE2
7.7
<0.056
16
3.0
<0.13
0.72
29
<0.056
11
9.4
<0.13
1.9
140
<0.056
42
<0.10
<0.13
<0.15
260
<0.056
11
<0.10
<0.13
<0.15
PGD2
PGA2
PGB2
PGJ2
ox-LGD2
Identification of novel oxidized levuglandin D2
2251

Supplemental Material can be found at:
http://www.jlr.org/content/suppl/2011/09/05/jlr.M017053.DC1
.html
Fig. 9. Contents of PGE2, PGD2, and ox-LGD2 in the tissues of five normal ddY strain male mice (Mouse
A–E), 4 weeks of age, 20-21 g body weight. Eicosanoids were extracted from frozen and powdered tissues of
mice in liquid nitrogen immediately after dissection. PGE2, PGD2 (A), and ox-LGD2 (B) were quantified
using LC-MS/MS (MRM) as described in the text.
added to the lysate (data not shown). We also confirmed
that the yield of ox-LGD2 was not significantly changed by
exclusion of 5 mM tryptophan from the incubation buffer
(40 mM Tris-HCl buffer, pH 7.8, containing 2 mM EGTA,
5 mM tryptophan, and 2 ␮M hematin) (24) used for this
experiment (see supplementary Fig. IV).
under the similar condition reported for the Michael addi-
tion of glutathione to ⌬7-PGA1 methyl ester (21). There-
fore, we predict ox-LGD2 to be a detoxification metabolite
of LGD2.
The reactions that produce ox-LGD2 from LGD2 (a two-
step oxidation, dehydrogenation of H8 and H12, and oxi-
dation of the C9 aldehyde to a carboxylic acid) can be
predicted. We presume that these oxidations are enzy-
matic oxidations, because the possibility of autoxidation
from LGD2 to ox-LGD2 in mice and red algae can be ex-
cluded by the following observations: i) ox-LGD2 was de-
tected in the extracts from mouse tissues, which were
immediately frozen in liquid nitrogen just after dissection
and homogenized in buffer containing BHT; ii) the level
of ox-LGD2 in the liver of mouse did not increase even
when the sample was homogenized in buffer without BHT;
and iii) ox-LGD2 was found only in G. edulis, not in G.
vermiculophylla, even though G. vermiculophylla contained
higher levels of PGE2 than did G. edulis (Table 2).
DISCUSSION
LGs, reactive byproducts of PG synthesis, have been re-
ported to form covalent adducts with proteins (4–6), DNA
(7, 8), and phosphatidylethanolamine (9, 10). However,
the oxidized forms of LGs, isoLGs, and isoprostanes have
not been reported. In the present study, we purified an
oxidized LGD2 (ox-LGD2) from the red alga Gracilaria edu-
lis, and we determined its planar structure by spectroscopic
methods and by chemical derivation from PGH2. Further-
more, ox-LGD2 was identified in mouse tissues and the
lysate of PMA-treated THP-1 cells incubated with arachi-
donic acid. These results suggest that ox-LGD2 is a com-
mon oxidized metabolite of LGD2. Because ox-LGD2 does
not possess a reactive aldehyde in its molecule, it must be
less reactive with amines than are LGs. We also confirmed
that the ox-LGD2 was not reactive to the thiol in cysteine
The stereochemistry of C15 in ox-LGD2 has not been
fully determined by some chemical methods (e.g., Mosher’s
method (26)) due to the limited sample amount (0.1 mg).
Ox-LGD2 isolated from G. edulis was shown as a single
1
isomer in the H NMR spectrum (supplementary Fig. I)
2252
Journal of Lipid Research Volume 52, 2011

Supplemental Material can be found at:
http://www.jlr.org/content/suppl/2011/09/05/jlr.M017053.DC1
.html
Fig. 10. Representative quantitation and identification of ox-LGD2 in the liver of mouse B. A: MRM ion
chromatograms of the mixture of authentic PGE2 (0.7 ng) and purified ox-LGD2 (2.5 ng). B: Product
ion spectrum of the [M-H]Ϫ ion at m/z 365 of ox-LGD2 (purified, 1.0 ng) eluted at 24.0-25.0 min. C: MRM
ion chromatograms of PGE2 (0.01 ng) and ox-LGD2 (0.06 ng) in an aliquot (1/20) of the lipophilic extract
from the liver of mouse B (20 mg). D: Product ion spectrum of ox-LGD2 (0.6 ng) eluted at 24.0-25.0 min in
the liver of mouse B. LC was performed using a Cosmosil 5C18ARII (i.d. 0.46 × 15 cm) column with ace-
tonitrile-water-acetic acid (40:60:0.1, v/v/v) with a flow rate of 0.3 ml/min at 25°C. Collision energy for
MRM and the product ion scan: Ϫ15 eV for PGE2, Ϫ20 eV for ox-LGD2.
and the MRM ion chromatograms (Fig. 8). Similarly, the
peaks corresponding to ox-LGD2 detected by LC-MS/MS
(MRM) were single peaks in both the lipophilic extracts
from mouse tissues (Fig. 10 for liver) and the PMA-treated
THP-1 cell lysate (Fig. 11). We also confirmed that the op-
tical rotation and 1H NMR of PGE2 isolated from red algae
were identical to those of authentic PGE2, which was de-
rived from PGH2 (Fig. 1). Therefore, the stereochemistry
at C15 of ox-LGD2 is presumed to be in an S configuration
like PGE2 (Fig. 1). However, the stereochemistry at C11 in
ox-LGD2 could be racemic due to the nature of the he-
miketal, which could mean that the two diastereomers of
bearing a ␥-hydroxybutenolide, like ox-LGD2, has been re-
ported (27), although we have not tested such activity for
ox-LGD2 due to a limited amount of purified ox-LGD2.
The level of ox-LGD2 should be examined in the tissues of
several animals, including humans, under various physio-
logical conditions.
In summary, our first identification of ox-LGD2 as a
probable detoxification metabolite of LGD2 provides
methods to examine its formation in animal tissues and
cells.
1
ox-LGD2 were not separated by H NMR and MRM. We
admit that this point should be further confirmed.
Although it is reported that approximately 20% of PGH2
nonenzymatically converts to both of LGE2 and LGD2 (1),
ox-LGE2 (Fig. 3C), the regioisomer of ox-LGD2, was not
found in this study in the red alga G. edulis by screening
using LC-diode array detection and LC-MS in single ion
monitoring (SIM) mode. The presence of ox-LGE2 in
mouse tissues and the THP-1 cell lysate has not been fully
examined by LC-MS/MS due to lack of authentic ox-LGE2.
In this study, we measured the contents of PGE2, PGD2,
and ox-LGD2 in liver of six mice (mouse A–F). The con-
tent of ox-LGD2 in the mice was 15, 60, 14, <4, 5, and 130
ng/g, respectively, and the content of PGE2 was 300, 10,
14, 6, 8, and 17 ng/g, respectively, suggesting that the con-
tent of ox-LGD2 in liver differed in individual mice and
was unrelated to the content of PGE2.
Fig. 11. Formation of PGE2 and ox-LGD2 by PMA-treated THP-1
cell lysate incubated with arachidonic acid. The lysate of PMA-
treated THP-1 cells (7.5 × 105 cells) was incubated with arachidonic
acid (2.0 ␮g) at 25°C for 5 min. The lipophilic extract from this
lysate was subjected to LC-MS/MS analysis for PGE2 and ox-LGD2
in MRM mode. The amounts of formed PGE2 and ox-LGD2 were
quantified as 13.6 ng and 3.8 ng, respectively. LC was performed
using a Cosmosil 5C18ARII (i.d. 0.46 × 15 cm) column with
acetonitrile-water-acetic acid (40:60:0.1, v/v/v) with a flow rate of
0.3 ml/min at 25°C.
Further study will be needed to examine the physiologi-
cal functions of ox-LGD2 and to identify the predicted en-
zymes that catalyze the oxidation of LGD2 to ox-LGD2.
Interestingly, a synthetic inhibitor of prostanoid production
Identification of novel oxidized levuglandin D2
2253

Supplemental Material can be found at:
http://www.jlr.org/content/suppl/2011/09/05/jlr.M017053.DC1
.html
15. Imbs, A. B., A. V. Vologodskaya, N. V. Nevshupova, S. V.
REFERENCES
Khotimchenko, and E. A. Titlyanov. 2001. Response of prostaglan-
din content in the red alga Gracilaria verrucosa to season and solar
irradiance. Phytochemistry. 58: 1067–1072.
1. Salomon, R. G., D. B. Miller, M. G. Zagorski, and D. J. Coughlin.
1984. Solvent-induced fragmentation of prostaglandin endoper-
oxides. New aldehyde products from PGH2 and a novel intramo-
lecular 1,2-hydride shift during endoperoxide fragmentation in
aqueous solution. J. Am. Chem. Soc. 106: 6049–6060.
2. Davies, S. S., V. Amarnath, and L. J. Roberts II. 2004. Isoketals:
highly reactive ␥-ketoaldehydes formed from the H2-isoprostane
pathway. Chem. Phys. Lipids. 128: 85–99.
3. Salomon, R. G., and D. B. Miller. 1985. Levuglandins: isola-
tion, characterization, and total synthesis of new secoprostanoid
products from prostaglandin endoperoxides. Adv. Prostaglandin
Thromboxane Leukot. Res. 15: 323–326.
4. Salomon, R. G., W. Sha, C. Brame, K. Kaur, G. Subbanagounder,
J. O’Neil, H. F. Hoff, and L. J. Roberts II. 1999. Protein adducts of
isolevuglandin E2, a product of the isoprostane pathway, inoxidized
low density lipoprotein. J. Biol. Chem. 274: 20271–20280.
5. Salomon, R. G., G. Subbanagounder, U. Singh, J. O’Neil, and H. F.
Hoff. 1997. Oxidation of low-density lipoproteins produces levug-
landin-protein adducts. Chem. Res. Toxicol. 10: 750–759.
6. Brame, C. J., R. G. Salomon, J. D. Morrow, and L. J. Roberts II.
1999. Identification of extremely reactive gamma-ketoaldehydes
(isolevuglandins) as products of the isoprostane pathway and
characterization of their lysyl protein adducts. J. Biol. Chem. 274:
13139–13146.
7. Murthi, K. K., L. R. Friedman, N. L. Oleinick, and R. G. Salomon.
1993. Formation of DNA-protein cross-links in mammalian cells by
levuglandin E2. Biochemistry. 32: 4090–4097.
8. Carrier, E. J., V. Amarnath, J. A. Oates, and O. Boutaud. 2009.
Characterization of covalent adducts of nucleosides and DNA formed
by reaction with levuglandin. Biochemistry. 48: 10775–10781.
9. Bernoud-Hubac, N., L. B. Fay, V. Armarnath, M. Guichardant, S.
Bacot, S. S. Davies, L. J. Roberts II, and M. Lagarde. 2004. Covalent
binding of isoketals to ethanolamine phospholipids. Free Radic.
Biol. Med. 37: 1604–1611.
16. Sajiki, J., and H. Kakimi. 1998. Identification of eicosanoids in
the red algae, Gracilaria asiatica, using high-performance liquid
chromatography and electrospray ionization mass spectrometry. J.
Chromatogr. A. 795: 227–237.
17. Noguchi, T., T. Matsui, K. Miyazawa, M. Asakawa, N. Iijima, Y. Shida,
M. Fuse, Y. Hosaka, C. Kirigaya, and K. Watabe. 1994. Poisoning by
the red alga ‘ogonori’ (Gracilaria verrucosa) on the Nojima Coast,
Yokohama, Kanagawa Prefecture, Japan. Toxicon. 32: 1533–1538.
18. Illijas, M. I., M. Terasaki, R. Nakamura, N. Iijima, A. Hara, N.
Fusetani, and Y. Itabashi. 2008. Purification and characterization
of glycerolipid acyl-hydrolase from the red alga Gracilaria vermiculo-
phylla. Fish. Sci. 74: 670–676.
19. Davies, S. S., V. Amarnath, C. J. Brame, O. Boutaud, and L. J.
Roberts II. 2007. Measurement of chronic oxidative and inflamma-
tory stress by quantification of isoketal/levuglandin ␥-ketoaldehyde
protein adducts using liquid chromatography tandem mass spec-
trometry. Nat. Protoc. 2: 2079–2091.
20. Balkrishna, S. B., W. E. Childers, Jr., and H. W. Pinnick. 1981.
Oxidation of ␣,␤-unsaturated aldehydes. Tetrahedron. 37: 2091–2096.
21. Suzuki, M., M. Mori, T. Niwa, R. Hirata, K. Furuta, T. Ishikawa, and
R. Noyori. 1997. Chemical implications for antitumor and antiviral
prostaglandins: reaction of ⌬7-prostaglandin A1 and prostaglandin
A1 methyl esters with thiols. J. Am. Chem. Soc. 119: 2376–2385.
22. Powell, W. S. 1980. Rapid extraction of oxygenated metabolites of
arachidonic acid from biological samples using octadecylsilyl silica.
Prostaglandins. 20: 947–957.
23. Yang, P., D. Chan, E. Felix, T. Madden, R. D. Klein, I. Shureiqi, X.
Chen, A. J. Dannenberg, and R. A. Newman. 2006. Determination
of endogenous tissue inflammation profiles by LC/MS/MS: COX-
and LOX-derived bioactive lipids. Prostaglandins Leukot. Essent. Fatty
Acids. 75: 385–395.
24. Zhou, L., L. H. Shen, L. H. Hu, H. Ge, J. Pu, D. J. Chai, Q. Shao,
L. Wang, J. Z. Zeng, and B. He. 2010. Retinoid X receptor agonists
inhibit phorbol-12-myristate-13-acetate (PMA)-induced differentia-
tion of monocytic THP-1 cells into macrophages. Mol. Cell. Biochem.
335: 283–289.
25. Cao, H., L. Xiao, G. Y. Park, X. Wang, A. C. Azim, J. W. Christman,
and R. B. Breemen. 2008. An improved LC-MS/MS method for the
quantification of prostaglandins E2 and D2 production in biological
fluids. Anal. Biochem. 372: 41–51.
10. Sullivan, C. B., E. Matafonova, L. J. Roberts II, V. Amarnath, and
S. S. Davies. 2010. Isoketals forms cytotoxic phosphatidyletha-
nolamine adducts in cells. J. Lipid Res. 51: 999–1009.
11. Fukuda, K., S. S. Davies, T. Nakajima, B. H. Ong, S. Kupershmidt,
J. Fessel, V. Amarnath, M. E. Anderson, P. A. Boyden, P. C.
Viswanathan, et al. 2005. Oxidative mediated lipid peroxidation re-
capitulates proarrhythmic effects on cardiac sodium channels. Circ.
Res. 97: 1262–1269.
26. Ohtani, I., T. Kusumi, Y. Kashman, and H. Kakisawa. 1991. High-
field FT NMR application of Mosher’s method. The absolute config-
urations of marine terpenoids. J. Am. Chem. Soc. 113: 4092–4096.
27. Guerrero, M. D., M. Aquino, I. Bruno, M. C. Terencio, M. Paya, R.
Riccio, and L. Gomez-Paloma. 2007. Synthesis and pharmacologi-
cal evaluation of a selected library of new potential anti-inflamma-
tory agents bearing the ␥-hydroxybutenolide scaffold: a new class of
inhibitors of prostanoid production through the selective modula-
tion of microsomal prostaglandin E synthase-1 expression. J. Med.
Chem. 50: 2176–2184.
12. Zagol-Ikapitte, I., T. S. Masterson, V. Amarnath, T. J. Montine, K.
I. Andreasson, O. Boutaud, and J. A. Oates. 2005. Prostaglandin
H(2)-derived adducts of proteins correlate with Alzheimer’s dis-
ease severity. J. Neurochem. 94: 1140–1145.
13. Gregson, R. P., J. F. Marwood, and R. J. Quinn. 1979. The occur-
rence of prostaglandins PGE2 and PGF2␣ in a plant - the red alga
Gracilaria lichenoides. Tetrahedron Lett. 46: 4504–4506.
14. Fusetani, N., and K. Hashimoto. 1984. Prostaglandin E2: a candi-
date for causative agent of “Ogonori” poisoning. Bull. Japan. Soc.
Sci. Fish. 50: 465–469.
2254
Journal of Lipid Research Volume 52, 2011