Formation of cholesterol ozonolysis products in vitro and in vivo through a myeloperoxidase-dependent pathway

Article abstract of DOI:10.1194/jlr.M006775

1 3 β-Hydroxy-5-oxo-5,6-secocholestan-6-al (sec osterol- A) and its aldolization product 3 β-hydroxy-5 β- hydroxy-B-norcholestane-6 β-carboxaldehyde (secosterol-B) were recently detected in human atherosclerotic tissues and brain specimens, and they may play pivotal roles in the pathogenesis of atherosclerosis and neurodegenerative diseases. However, as their origin remains unidentifi ed, we examined the formation mechanism, the stability, and the fate of secosterols in vitro and in vivo. About 40% of secosterol-A remained unchanged after 3 h incubation in the FBS-free medium, whereas 20% and 40% were converted to its aldehyde-oxidation product, 3 β-hydroxy-5-oxo-secocholestan- 6-oic acid, and secosterol-B, respectively. In the presence of FBS, almost all secosterol-A was converted immediately to these compounds. Secosterol-B in the medium, with and without FBS, was relatively stable, but β30% was converted to its aldehyde-oxidation product, 3 β-hydroxy-5 β-hydroxy- B-norcholestane-6-oic acid (secoB-COOH). When neutrophillike differentiated human leukemia HL-60 (nHL-60) cells activated with PMA were cultured in the FBS-free medium containing cholesterol, significantly increased levels of secosterol-A and its aldehyde-oxidation product, but not secosterol-B, were formed. This secosterol-A formation was decreased in the culture of PMA-activated nHL-60 cells containing several reactive oxygen species (ROS) inhibitors and scavengers or in the culture of PMA-activated neutrophils isolated from myeloperoxidase (MPO)-defi cient mice. Our results demonstrate that secoterol-A is formed by an ozone-like oxidant generated with PMA-activated neutrophils through the MPO-dependent mechanism. - Tomono, S., N. Miyoshi, H. Shiokawa, T. Iwabuchi, Y. Aratani, T. Higashi, H. Nukaya, and H. Ohshima. Formation of cholesterol ozonolysis products in vitro and in vivo through amyeloperoxidase-dependent pathway.

Full text of DOI:10.1194/jlr.M006775

Formation of cholesterol ozonolysis products in vitro and
in vivo through a myeloperoxidase-dependent pathway
Susumu Tomono,*,§ Noriyuki Miyoshi,1,*,§ Hidemi Shiokawa,*,§ Tomoe Iwabuchi,*,§
Yasuaki Aratani,** Tatsuya Higashi,†,§ Haruo Nukaya,*,§ and Hiroshi Ohshima*,§
Laboratory of Biochemistry,* Graduate School of Nutritional and Environmental Sciences, Laboratory of
Analytical and Bio-Analytical Chemistry,† Graduate School of Pharmaceutical Sciences, and the Global
Center of Excellence (COE) Program,§ University of Shizuoka, Shizuoka, Japan; and Graduate School of
Nanobioscience,** Yokohama City University, Kanagawa, Japan
Abstract 3␤-Hydroxy-5-oxo-5,6-secocholestan-6-al (seco-
sterol-A) and its aldolization product 3␤-hydroxy-5␤-
hydroxy-B-norcholestane-6␤-carboxaldehyde (secosterol-B)
were recently detected in human atherosclerotic tissues and
brain specimens, and they may play pivotal roles in the
pathogenesis of atherosclerosis and neurodegenerative dis-
eases. However, as their origin remains unidentified, we ex-
amined the formation mechanism, the stability, and the fate
of secosterols in vitro and in vivo. About 40% of secosterol-A
remained unchanged after 3 h incubation in the FBS-free
medium, whereas 20% and 40% were converted to its
aldehyde-oxidation product, 3␤-hydroxy-5-oxo-secocholestan-
6-oic acid, and secosterol-B, respectively. In the presence of
FBS, almost all secosterol-A was converted immediately to
these compounds. Secosterol-B in the medium, with and
without FBS, was relatively stable, but ف30% was converted
to its aldehyde-oxidation product, 3␤-hydroxy-5␤-hydroxy-
B-norcholestane-6-oic acid (secoB-COOH). When neutrophil-
like differentiated human leukemia HL-60 (nHL-60) cells
activated with PMA were cultured in the FBS-free medium
containing cholesterol, significantly increased levels of
secosterol-A and its aldehyde-oxidation product, but not
secosterol-B, were formed. This secosterol-A formation was
decreased in the culture of PMA-activated nHL-60 cells con-
taining several reactive oxygen species (ROS) inhibitors and
scavengers or in the culture of PMA-activated neutrophils
isolated from myeloperoxidase (MPO)-deficient mice.
Our results demonstrate that secoterol-A is formed by an
ozone-like oxidant generated with PMA-activated neutro-
phils through the MPO-dependent mechanism.—Tomono,
S., N. Miyoshi, H. Shiokawa, T. Iwabuchi, Y. Aratani, T.
Higashi, H. Nukaya, and H. Ohshima. Formation of choles-
terol ozonolysis products in vitro and in vivo through a
myeloperoxidase-dependent pathway. J. Lipid Res. 2011. 52:
87–97.
Supplementary key words secosterol • neutrophils • myeloperoxi-
dase • inflammation
A reactive species with a chemical signature similar to
that of ozone has been proposed to be generated by the
antibody-catalyzed oxidation of water with singlet oxygen
during the oxidative burst of activated human neutrophils
and in inflamed tissues (1, 2). The formation of an ozone-
like oxidant from singlet oxygen by neutrophils can be cata-
lyzed not only by antibodies but also by amino acids such as
tryptophan, methionine, and cysteine (3). To prove the for-
mation of an ozone-like oxidant by neutrophils, previous
studies used chemical reactions, such as the conversion
of indigo carmine to isatin sulfonic acid or the oxidation
of vinylbenzoic acid to 4-carboxybenzaldehyde. However,
these reactions are not sufficiently specific to ozone to
conclude an ozone production by neutrophils (4, 5).
Further evidence for ozone formation in vivo was based
on the detection and formation of the cholesterol ozonoly-
sis products 3␤-hydroxy-5-oxo-5,6-secocholestan-6-al (seco-
sterol-A, also called atheronal-A) and its aldolization product
3␤-hydroxy-5␤-hydroxy-B-norcholestane-6␤-carboxaldehyde
(secosterol-B, also called atheronal-B) in human tissues.
These secosterols were previously reported to be formed
only by ozone among the various reactive oxygen species
(ROS) such as singlet oxygen, superoxide anion, hydroxyl
Abbreviations: ␣, ␤-unsaturated-secosterol-B, 3␤-hydroxy-B-norcho-
lest-5-ene; DH, dansyl hydrazine; HOCl, hypochlorous acid; LPS,
lipopolysaccharide; MPO, myeloperoxidase; nHL-60, neutrophil-like
differentiated human leukemia HL-60; PA, 2-picolylamine; PEG,
polyethylene glycol; ROS, reactive oxygen species; secoA-COOH,
3␤-hydroxy-5-oxo-secocholestan-6-oic acid; secoB-COOH, 3␤-hydroxy-5␤-
hydroxy-B-norcholestane-6-oic acid; secosterol-A, 3␤-hydroxy-5-oxo-5,6-
secocholestan-6-al; secosterol-B, 3␤-hydroxy-5␤-hydroxy-B-norcholestane-
6␤-carboxaldehyde; SOD, superoxide dismutase; WT, wild-type.
1 To whom correspondence should be addressed.
This work was supported by the Global Centers of Excellence (COE) Program of
Japan’s Ministry of Education, Science, Culture, and Sport; Grants-in-Aid
18509001 (HO), 21300280 (HO), 19700592 (NM), and 21680052 ( NM)
from Japan’s Ministry of Education, Science, Culture, and Sport; Grants-in-Aid
19-19 and 21-1-2 for Cancer Research from Japan’s Ministry of Health, Labor,
and Welfare; and Grant KHC1023 and a grant from the Japan Health Science
Foundation for Health Sciences Focusing on Drug Innovation.
Manuscript received 11 March 2010 and in revised form 9 August 2010.
Published, JLR Papers in Press, October 4, 2010
DOI 10.1194/jlr.M006775
e-mail: miyoshin@u-shizuoka-ken.ac.jp
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
87

metabolites of secosterol-A and -B. On the basis of these
findings, we have shown evidence suggesting that the for-
mation of secosterols is mediated by an MPO-dependent
system in vivo at least partly through an oxidant with the
chemical signature of ozone.
radicals, and ozone (6–9). Secosterols were recently detected
in human atherosclerotic tissues (10) and in human brain
specimens from patients with Alzheimer’s disease (11). They
were also detected in Lewy body dementia (12) by the analy-
sis of their hydrazine derivatives with LC-MS. Higher con-
centrations of secosterol-A were detected in the diseased
arteries of patients with atherosclerosis. This secosterol-A
concentration was further increased upon activation with
PMA, suggesting that human leukocytes within atheroscle-
rotic tissues are activated to produce an ozone-like oxidant
(10). However, it has been recently reported that secosterol-A
and -B were generated in an ozone-independent manner
via the Hock-cleavage of 5␣-hydroperoxy cholesterol, which
can arise from the singlet oxygen ene reaction with
cholesterol (13, 14). Secosterol-B is formed easily under
acidic conditions in organic solvents (13, 14), whereas seco-
sterol-A is either not formed at all or is a minor component
in the aqueous buffer (9). We reported recently that almost
equal amounts of secosterol-A and -B were formed by the
reaction of cholesterol with human myeloperoxidase (MPO)
in the presence of its substrates hydrogen peroxide (H2O2)
and ClϪ (15). On the other hand, five times more seco-
sterol-B was formed than secosterol-A when cholesterol was
incubated with hypochlorous acid (HOCl) and hydrogen
peroxide. However, in both the reactions, immunoglobulin
G (IgG) did not enhance the formation of secosterols,
suggesting that singlet oxygen (1O2) and possibly another
oxidant, but not an ozone-like oxidant, mediated the forma-
tion of secosterols (15).
MATERIALS AND METHODS
Materials
Allopurinol, apocynin, cholesterol, 3,4-13C-cholesterol, cata-
lase (bovine), cell permeable catalase, covalently linked to poly-
ethylene glycol (PEG-catalase), superoxide dismutase (SOD)
(bovine), PEG-SOD, PMA, immunoglobulin G (bovine serum),
p-toluenesulfonic acid, sodium azide, and LDL were purchased
from Sigma, St. Louis, MO. Dansyl hydrazine (DH) and RPMI
1640 medium were purchased from Invitrogen (Carlsbad, CA)
and Nissui Pharmaceutical Co., Ltd., Tokyo, Japan, respectively.
All other chemicals were obtained from Wako Pure Chemical In-
dustries, Osaka, Japan.
Synthesis of ozonolysis products of cholesterol
Secosterols and their metabolites were synthesized according
to the method reported by Wentworth et al. (10). The purity of
1
the product was verified by TLC and H-NMR. The stock solu-
tions (10 mM) of secosterols were prepared in ethanol and stored
at Ϫ20°C until use.
Fate of secosterols in culture media
One hundred pmol secosterol-A and -B were added into 1 ml
of RPMI 1640 medium in the absence or presence of 10% FBS.
After the incubation for 0-3 h, the media were spiked with stable
isotopes labeled synthesized 3,4-13C-secosterol-A or -B (5 nmol
each) and mixed vigorously with 2 ml chloroform-methanol (2:1)
for 1 min. After centrifugation at 3,000 rpm for 10 min, the or-
ganic phase was separated, washed with water twice, and evapo-
rated to dryness in vacuo. The secosterols were analyzed as
described below using a calibration curve prepared with several
concentrations of standard compounds.
Although the exact mechanism for the formation of
secosterols in human tissues remains unidentified, seco-
sterols have been reported to react with the amyloid ␤
protein in Alzheimer’s disease, resulting in the misfolding
and aggregation of this protein (11, 16, 17), which ac-
celerate amyloidogenesis. Secosterols can also accelerate
␣-synuclein fibrilization (12), which has been associated
with Parkinson’s disease and Lewy body dementia (18).
Secosterols exert cytotoxic effects on various culture cells
(10, 19, 20).
Formation of secosterols by HL-60 cells
Human leukemia HL-60 cells (RIKEN Cell Bank, Tsukuba,
Ibaraki, Japan) were maintained in RPMI 1640 supplemented with
10% heat-inactivated FBS, 2 mM glutamine, 50 U/ml penicillin
and 50 µg/ml streptomycin in an atmosphere of 95% air and 5%
CO2 at 37°C. HL-60 cells seeded at 5 × 105 cells/ml were differenti-
ated into neutrophils by the treatment with 1.25% dimethylsulfox-
ide for 5 days (21). Then the medium was substituted for a fresh
one containing either 10% FBS or no FBS. When the FBS-free
medium was used, 6 µg/ml IgG and/or 30 ␮g/ml cholesterol were
exogenously added into the medium, as FBS (10%) contained the
compounds at these concentrations. The cells were then activated
with 10 nM PMA for 0-3 h. When inhibitors of MPO (sodium
azide, aminobenzohydrazide and salicylhydroxamic acid) were ex-
amined, they were added 1 h before the addition of PMA. Other
inhibitors and chemicals were added just prior to the PMA activa-
tion. The harvested cells and the culture medium spiked with
a stable isotope labeled synthesized 3,4-13C-secosterol-A and -B
(5 nmol or 10 pmol each in Figs. 2 and 3, respectively) were mixed
with 25 ml chloroform-methanol (2:1). Secosterols were extracted
and quantified by external or internal standard methods as shown
in Figs. 2 and 3, respectively, as described below. The trace amounts
of secosterol-A and -B (about 0.7 and 15 nM, respectively) were
always detected in the cell-free culture media containing 10% FBS
We recently reported a highly sensitive method for the
quantification of secosterols based on derivatization with
fluorescent dansyl hydrazine (DH) and detection by
LC-MS/MS system, in which both secosterols were con-
firmed to be stable during the derivatization process (15).
This sensitive method has allowed us to perform quantita-
tive analysis of secosterols in cell culture and in animal tis-
sues. In this study, we examined the molecular mechanism
for the formation of secosterols in vivo. We demonstrate
here that the formation of secosterol-A and -B are en-
hanced by the inflamed stimuli not only in the cultures of
neutrophil-like differentiated HL-60 cells or mouse neu-
trophils, but also in plasma samples and the liver of mice
after lipopolysaccharide (LPS) administration. In ad-
dition, we have studied the fate of secosterols added to
the culture medium with and without FBS. Several prod-
ucts, including 3␤-hydroxy-5-oxo-secocholestan-6-oic acid
(secoA-COOH), 3␤,5␤-dihydroxy-B-norcholestane-6-oic
acid (secoB-COOH), and 3␤-hydroxy-B-norcholest-5-ene
(␣, ␤-unsaturated-secosterol-B), were newly identified as
88
Journal of Lipid Research Volume 52, 2011

(Fig. 2). Similarly, commercially available cholesterol also con-
tained detectable amounts of secosterol-A and -B at the levels of
ف7 and 200 µmol/mol cholesterol, respectively.
668.5/170.2, 236.1; that of ␣, ␤-unsaturated-secosterol-B was m/z
648.4/170.0, 365.1; and the PA-derivatives of secoA-COOH and
secoB-COOH were m/z 525.4/109.1, 381.4.
Animal experiments
Animals used in this study were kept according to guidelines for
the care and use of laboratory animals at the University of Shi-
zuoka, and all experiments were approved by the local animal ethi-
calcommittee.MpoϪ/Ϫ mice,backcrossedtoaC57BL/6background
10 times, were genotyped using polymerase chain reaction-ampli-
fied DNA from tail clippings (22). Wild-type (WT) male C57BL/6
control mice were obtained from the Japan SLC (Hamamatsu, Ja-
pan). Neutrophils were isolated from the bone marrow of the un-
treated WT and MpoϪ/Ϫ mice (11 weeks of age) using the Percoll
(Sigma) density gradient isolation method. Isolated neutrophils,
with over 95% purity and viability confirmed by the trypan blue
exclusion assay and the HE stain, were suspended in RPMI 1640 in
the presence or absence of 10% FBS. After the PMA stimuli for 3 h,
the cells and medium were harvested and then mixed with 4 ml
chloroform-methanol (2:1) containing stable isotope labeled syn-
thesized 3,4-13C-secosterol-A and -B (10 pmol each). Secosterols
extracted were derivatized and analyzed by LC-MS/MS as described
below. The levels of secosterols were determined in the plasma
samples of mice (6-7 weeks of age, n = 2-4 in each group), which
had been treated with or without an intraperitoneal injection of 30
mg/kg body weight of E. coli LPS. At 6 or 22 h after LPS administra-
tion, the mice were euthanized, and blood samples were obtained
with a heparinized syringe from the inferior vena cava. The plasma
samples were stored at Ϫ80°C until analysis. 3,4-13C-Secosterol-A
and -B (10 pmol each) were added to 20 ␮l of plasma before being
mixed with 100 ␮l chloroform-methanol (2:1). Then the secoste-
rols were analyzed by LC-MS/MS as described below. Under these
conditions, the recoveries of secosterol-A and -B (10 pmol each)
were 100.1 and 101.3%, respectively.
RESULTS
Fate of secosterol-A and -B in the culture medium with
and without 10% FBS
We determined the stability and fate of secosterol-A and
-B exogenously added to the culture medium in the pres-
ence and absence of 10% FBS. When we initially examined
the stability of secosterols in culture media, we employed
10 pmol of 3,4-13C-secosterol-A or -B as an internal stan-
dard, which resulted in a poor total recovery of compounds
(20-40%) (data not shown). However, when 500 times
more 3,4-13C-secosterol-A or -B (5 nmol each) were added
before lipid extraction with chloroform-methanol, their
total recovery was markedly improved and reached be-
tween 75% and 100%. This increased recovery could be
due to the replacement by excess 3,4-13C-secosterol-A or -B
of initially added secosterols, which were probably present
as reversibly bound forms with medium components. Us-
ing these improved conditions, we found that almost all
secosterol-A added to the FBS-containing medium dis-
appeared very rapidly, and 60-80% of secosterol-A was im-
mediately converted to secosterol-B (Fig. 1A). We also
observed that smaller parts of secosterol-A were converted
to the aldehyde-oxidation products of secosterols, secoA-
COOH and secoB-COOH (18% and 3%, respectively). In
contrast, secosterol-A added to the FBS-free medium dis-
appeared relatively slowly compared with the FBS-contain-
ing medium (Fig. 1B). About 40%, 38%, and 17% of added
secosterol-A were detected as an unchanged compound,
secosterol-B and secoA-COOH, respectively, after 3 h in-
cubation (Fig. 1B). Conversely, it was found that about
60% of the original amount of secosterol-B remained un-
changed in 10% FBS-containing medium, but 20% and
0.8% of secosterol-B were converted immediately to secoB-
COOH and ␣, ␤-unsaturated-secosterol-B, respectively
(Fig. 1C). No further increase or decrease of these com-
pounds was observed during a period of 3 h incubation
(Fig. 1C). The stability and the fate of secosterol-B added
to the FBS-free medium was similar to that observed in the
10% FBS-containing medium (Fig. 1D). No conversions of
secosterol-B to secosterol-A were detected during incuba-
tion in the media, both with and without FBS (Fig. 1C, D).
In addition to these metabolites, we also detected small
peaks with m/z values corresponding to other secosterol
metabolites, including dehydrated derivatives of secoA-
COOH and secoB-COOH (data not shown). Our results
indicate that secosterol-A is very rapidly converted mainly
to secosterol-B and secoA-COOH by some components in
FBS, and some parts of secosterol-B are converted to secoB-
COOH by some constituents in the culture medium.
Analysis of secosterols by LC-MS/MS
The secosterols -A, -B, and ␣, ␤-unsaturated-secosterol-B (struc-
tures, see Fig. 7) were analyzed after derivatization with DH
as described in our previous study (15). Briefly, aliquots of ex-
tracted samples from the culture media or animal tissues were
dissolved in acetonitrile containing 0.5 mg/ml DH and 0.1 mg/ml
p-toluensulfonic acid, and derivatized for 4 h at room temperature
in darkness. The derivatized mixture was evaporated to dryness
in vacuo, and the residue was finally dissolved in 1 ml acetonitrile.
SecoA-COOH and SecoB-COOH (structures, see Fig. 7) were
analyzed by LC-MS/MS after derivatization with 2-picolylamine
(PA) (23). Briefly, aliquots of extracted samples were dissolved in
acetonitrile (100 µl). Then, freshly prepared solutions of tri-
phenylphosphine (10 mM) in acetonitrile (10 µl), 2,2′-dipyridyl
disulfide (10 mM) in acetonitrile (10 µl) and PA (10 µg) in
acetonitrile (10 µl) were successively added, and the mixture was
incubated at 60°C for 10 min. After removal of the solvent in
vacuo, the products were dissolved in 1 ml acetonitrile. A 10 µl
aliquot from each individual sample was used for LC-MS/MS
analysis. LC-MS/MS analyses, using nanospace SI-1 (SHISEIDO
Co., Ltd., Tokyo, Japan) and API2000™ (Applied Biosystems,
Forester City, CA) were performed with an electrospray ioniza-
tion device running in a positive ionization mode. The deriva-
tives were separated using a TSK-GEL ODS-100V column (150 ×
2.0 mm, 3 µm TOSOH, Tokyo, Japan) with a linear gradient of
70% solvent A (0.1% formic acid in water) and 30% solvent B
(acetonitrile containing 0.1% formic acid) to 100% solvent B in
20 min, followed by 100% solvent B for 20 min at the flow rate of
0.2 ml/min, and analyzed in MRM mode. The ion transitions
monitored for DH derivatives of secosterol-A and -B were m/z
666.5/170.2, 236.1; those of 3,4-13C-secosterol-A and -B were m/z
Formation of secosterol-A and -B by neutrophil-like
differentiated HL-60 cells
On the basis of the above findings, we examined the
molecular mechanism for the formation of secosterols by
MPO-dependent formation of secosterols in vitro and in vivo
89

Fig. 1. Comparison of the stability and fate of secos-
terol-A and secosterol-B incubated in the culture
medium with 10% FBS or without FBS. 100 pmol of
synthesized secosterol-A (A and B) or secosterol-B (C
and D) were incubated in 1 ml of the medium. At
the indicated time, lipid fractions from harvested sam-
ples were extracted in the presence of excess 3,4-13C-
secosterol-A or -B (5 nmol), derivatized with DH or
PA, and then analyzed as described in the text. Values
are means SD (n = 3). DH, dansyl hydrazine; PA,
2-picolylamine; secosterol-A, 3␤-hydroxy-5-oxo-5,6-
secocholestan-6-al; secosterol-B, 3␤-hydroxy-5␤-hydroxy-
B-norcholestane-6␤-carboxaldehyde; secoA-COOH,
3␤-hydroxy-5-oxo-secocholestan-6-oic acid; SecoB-
COOH, 3␤-hydroxy-5␤-hydroxy-B-norcholestane-6-
oic acid; ␣, ␤-unsaturated-secosterol-B, 3␤-hydroxy-B-
norcholest-5-ene.
cultured neutrophil-like differentiated HL-60 (nHL-60)
cells with or without PMA activation. When nHL-60 cells
were incubated with 10 nM PMA for 3 h in the FBS-free
medium into which cholesterol and IgG were added, sig-
nificantly elevated levels of secosterol-A but not secosterol-B
were formed (Fig. 2A, B). Moreover, increased levels
of the aldehyde-oxidation products of secosterols, secoA-
COOH and secoB-COOH, and ␣, ␤-unsaturated-secosterol-B
were detected in the cholesterol-containing FBS-free me-
dium (Fig. 2C-E). The formation of secosterol-A but not
secosterol-B by the activated nHL-60 cells in the FBS-free
medium occurred in a time-dependent manner as well as
in the concentration-dependent manner of cholesterol or
IgG added to the culture medium (Fig. 3A, B). No in-
creased formations of secosterol-A were detected when
BSA or heat-inactivated IgG were used in the place of IgG
(Fig. 3C). Furthermore, heat-inactivated nHL-60 cells (ac-
tivated with PMA) lost their ability to produce secosterol-
A, even in the presence of IgG and cholesterol in the
FBS-free medium (Fig. 3C). Additionally, levels of secos-
terol-A but not secosterol-B increased significantly (about
2-fold) when the FBS-free medium containing exogenous
LDL was incubated with PMA-activated nHL-60 cells (data
not shown). On the other hand, when the nHL-60 cells
were incubated in the culture medium containing 10%
FBS and upon activation with 10 nM PMA for 3 h, an in-
creased level of secosterol-B but not secosterol-A was ob-
served (Fig. 2A, B). When the nHL-60 cells were treated
with PMA in the culture medium containing 10% FBS, the
amounts of secosterol-B were also increased dose-depen-
dently with increasing concentrations of PMA (up to 10
nM) and in a time-dependent manner (up to 3 h of incu-
bation) (data not shown). Furthermore, the increased for-
mations of secoA-COOH and secoB-COOH are clearly
correlated with secosterol-A and secosterol-B formation,
respectively (Fig. 2C, D). These results indicate that a ma-
jor part of secosterol-A produced by the activated nHL-60
cells was immediately metabolized to its oxidation prod-
ucts and secosterol-B by components present in FBS.
Effects of inhibitors, scavengers, and antioxidants on
PMA-induced secosterol formation
To elucidate the underlying molecular mechanisms for
the formation of secosterols, we examined the effects of vari-
ous enzyme inhibitors, the scavengers of reactive species, and
antioxidant enzymes on secosterol-A formation by PMA-acti-
vated nHL-60 cells in the serum-free culture medium to
which cholesterol and IgG were exogenously added (Table
1). It was found that apocynin (a NADPH oxidase inhibitor),
allopurinol (an inhibitor for a xanthine oxidase and a free
radical signal generated by activated neutrophils), and MPO
inhibitors (sodium azide, aminobenzohydrazide, and salicyl-
hydroxamic acid) significantly inhibited the secosterol-A for-
mation. Similarly, methionineand␤-carotene(singletoxygen
scavengers) as well as vinylbenzoic acid (a scavenger for ozone
and possibly other reactive species) significantly inhibited
90
Journal of Lipid Research Volume 52, 2011

secosterol-A formation. Cell permeable catalase, covalently
linked to polyethylene glycol (PEG-catalase) alone or PEG-
catalase together with PEG-SOD added into the culture me-
dium significantly decreased the amounts of secosterol-A
formed by the activated nHL-60 cells, whereas neither PEG-
SOD, SOD, nor catalase showed the inhibitory effects on its
formation. Similar to the above findings, the formation of
secosterol-B by the PMA-activated nHL-60 cells in the 10%
FBS-containing medium was effectively inhibited by the same
compounds as described above, including inhibitors for NA-
DPH oxidase, xanthine oxidase, and MPO; the scavengers of
reactive species such as methionine, ␤-carotene, and vinyl-
benzoic acid; PEG-catalase alone, and PEG-catalase plus
PEG-SOD (Table 1). These results imply that several reactive
species are involved in the formation of secosterols in the
culture of PMA-activated nHL-60 cells.
Formation of secosterols by activated neutrophils
obtained from wild-type and MPO-deficient mice
As our previous study (15) and the experimental results
above suggested the involvement of MPO in the formation
of secosterols by activated neutrophils, we employed neutro-
phils obtained from WT and MPO-deficient mice. Similar to
the findings with the PMA-activated nHL-60 cells, cultured
WT neutrophils upon activation with PMA significantly pro-
duced secosterol-A and -B in the medium containing no FBS
(but containing exogenously added cholesterol and IgG) or
10% FBS, respectively (Fig. 4A, B). However, no increased
formation of secosterol-A or -B was found when PMA-acti-
vated neutrophils from MPO-deficient mice were incubated
under the same conditions as WT neutrophils (Fig. 4A, B).
Presence of secosterols in plasma and liver of WT and
MPO-deficient mice
We determined the levels of secosterols in the samples of
plasma and liver collected from WT and MPO-deficient mice.
Fig. 5 shows typical chromatograms obtained for the analyses
of secosterols in mice plasma. The background levels of
secosterol-A and -B were significantly lower in the plasma
samples of MPO-deficient mice than WT mice (Fig. 6A, B).
Similarly, the levels of secosterol-B detected in homogenates
of the liver were also significantly lower in MPO-deficient
mice than in WT mice (Fig. 6C). To examine the effect of
an acute inflammation stimulus on the formation of secoste-
rols, the mice were injected intraperitoneally with LPS. As
shown in Fig. 6A and B, the levels of secosterol-A and -B in the
plasma samples collected from WT mice were increased
time-dependently after LPS injection, whereas the plasma
levels of secosterols in MPO-deficient mice did not increase
even after LPS injection.
Fig. 2. Formation of secosterols and their oxidation products by
nHL-60 cells in the presence or absence of FBS. nHL-60 cells were
treated with 0.1% v/v ethanol (control) or 10 nM PMA in ethanol
for 1 or 3 h in a serum-free medium containing 30 ␮g/ml choles-
terol and 6 µg/ml IgG or in 10% FBS-medium. Lipid fractions ex-
tracted from the harvested cells and the medium were derivatized
with DH or PA. Secosterol-A (A), secosterol-B (B), secoA-COOH
(C), secoB-COOH (D), and ␣, ␤-unsaturated-secosterol-B (E) were
analyzed as described in the text. Values are means SD (n = 5).
Statistically significant: *P < 0.05, **P < 0.01 versus control. DH,
dansyl hydrazine; nHL-60, neutrophil-like differentiated human
leukemia HL-60; PA, 2-picolylamine; secoA-COOH, 3␤-hydroxy-
5-oxo-secocholestan-6-oic acid; secoB-COOH, 3␤-hydroxy-5␤-
hydroxy-B-norcholestane-6-oic acid; secosterol-A, 3␤-hydroxy-
5-oxo-5,6-secocholestan-6-al; secosterol-B, 3␤-hydroxy-5␤-hydroxy-
B-norcholestane-6␤-carboxaldehyde.
DISCUSSION
Currently there are two proposed mechanisms for the
secosterol formation in vivo: a) the oxidation of choles-
terol with an ozone-like oxidant to generate secosterol-A
and its conversion to secosterol-B by aldolization (2, 10);
and b) the formation of 5␣-hydroperoxy cholesterol by the
reaction of cholesterol with singlet oxygen and its Hock-
MPO-dependent formation of secosterols in vitro and in vivo
91

Fig. 3. Effects of incubation time and concentrations of IgG and cholesterol on the formation of seco-
sterol-A and secosterol-B by PMA-activated nHL-60 cells. A, B: nHL-60 cells were incubated with 10 nM PMA
in the presence of 30 ␮g/ml cholesterol or 6 µg/ml IgG for 3 h or indicated modified condition. Secosterols
were extracted from the harvested cells and medium, and then analyzed as described in the text. The gray
bar (M; medium) and dotted lines represent the levels of preformed secosterols detected in serum-free me-
dium containing 30 ␮g/ml cholesterol or 6 µg/ml IgG. C: Comparison of formations of secosterol-A and
secosterol-B by activated nHL-60 cells in the absence or presence of IgG (6 µg/ml), heat-inactivated IgG
(6 µg/ml heated at 100°C for 5 min), and BSA (6 µg/ml). Heat-inactivated nHL-60 cells (100°C, 5 min) were
also incubated in the FBS-free medium containing cholesterol and IgG in the presence of PMA. Values
are means SD (n = 3). Statistically significant: *P < 0.05 when compared with control (C). Dotted lines
represent the levels of preformed secosterols in the FBS-free medium containing cholesterol. nHL-60, neu-
trophil-like differentiated human leukemia HL-60; secosterol-A, 3␤-hydroxy-5-oxo-5,6-secocholestan-6-al;
secosterol-B, 3␤-hydroxy-5␤-hydroxy-B-norcholestane-6␤-carboxaldehyde.
cleavage to generate both secosterol-A and secosterol-B
(13, 14). However, recently Wentworth et al. (9) reported
that only ozone can react with cholesterol to form seco-
sterol-A and that other oxidants, such as singlet oxygen,
form secosterol-B but not -A as a major component. In the
present study, we investigated the underlying molecular
mechanism for the formation of secosterols by PMA-
activated human neutrophil-like HL-60 cells in culture
and in mice in vivo.
When we initially determined the stability and the fate
of secosterols added to the culture medium, we could
identify only 20-40% of the added compounds. However,
the total recovery was markedly improved to 75-100%
when 500 times more 3,4-13C-secosterol-A or -B were added
to the medium before lipid extraction with chloroform-
methanol. This might be due to excess 3,4-13C-secosterol-A
or -B replacing the previously added secosterols, which
could be present as reversibly bound forms with medium
components. Using this method, it was found that seco-
sterol-A added to the FBS-free medium decreased more
slowly than in the FBS-containing medium, and 40% re-
mained unchanged during 3 h incubation. Conversely,
secosterol-A added to the FBS-containing medium imme-
diately disappeared. In contrast, although ف30% of seco-
sterol-B added to the FBS-free medium was converted to
secoB-COOH, 60-70% of secosterol-B remained unchanged
92
Journal of Lipid Research Volume 52, 2011

TABLE 1. Effects of inhibitors for oxidant-generating enzymes, antioxidants, and scavengers of ROS
on the formation of secosterols by PMA-activated nHL-60 cells
Secosterol-A (nM) Formed
in FBS-free Medium
Secosterol-B (nM) Formed
in 10% FBS Medium
Treatment
Concentration
% Inhibitiona
% Inhibitiona
Medium alone
nHL-60 without PMA treatment
nHL-60 with PMA
0.5 0.4
0.8 0.4
3.2 1.1b
14.5 0.3
20.3 2.2
35.8 4.4b
+
+
+
+
+
+
+
+
+
+
+
+
+
+
apocynin
allopurinol
SOD
catalase
SOD + catalase
PEG-SOD
PEG-catalase
PEG-SOD + PEG-catalase
sodium azide
aminobenzohydrazide
salicylhydroxamic acid
methionine
100 µM
100 µM
1.0 0.4c
0.9 0.1c
3.1 0.3
2.2 0.3c
2.8 0.1
2.8 0.5
1.4 0.2c
1.8 0.7c
0.5 0.4c
0.2 0.1d
1.8 0.5c
0.7 0.3c
0.5 0.3c
0.6 0.4c
91.7
95.8
4.2
13.9 2.2d
15.2 3.5c
39.2 2.2
26.9 14.7
42.5 7.1
42.9 7.0
17.2 2.0c
21.9 0.8c
15.6 2.9c
18.1 2.9c
21.6 2.1c
7.1 0.1d
4.6 1.1d
24.9 2.5c
>
>
<
100
100
0
57.4
0
100 U/ml
1000 U/ml
100 U / 1000 U/ml
100 U/ml
200 U/ml
100 U / 200 U/ml
200 µM
100 µM
100 µM
100 µM
50 µM
41.7
16.7
16.7
75.0
58.3
100
100
58.3
100
100
100
<
<
>
0
100
89.7
100
100
91.6
100
100
70.3
>
>
>
>
>
>
>
>
>
␤-carotene
vinylbenzoic acid
100 µM
nHL-60 cells were activated with 10 nM PMA for 3 h in the indicated medium in the presence or absence of
test compounds. Formation of secosterol-A or secosterol-B was determined in the FBS-free medium containing 30
␮g /ml cholesterol and 6 µg/ml IgG, or in the 10% FBS medium, respectively. Values are means SD (n = 3). nHL-
60, neutrophil-like differentiated human leukemia HL-60; PA, 2-picolylamine; PEG, polyethylene glycol; ROS,
reactive oxygen species; SOD, superoxide dismutase.
a Percentage of inhibition representing relative inhibitory effect was calculated for the following equation:
A ꢀ C
B ꢀ C
§
©
·
%inhibition = 100 u 1ꢀ
¨
¸
¹
where A is the amount of secosterol formed by the PMA-activated cells in the presence of a test compound; B
is the amount of secosterol formed by the PMA-activated cells in the absence of a test compound; and C is the
amount of secosterol presented in the cell culture without PMA treatment.
b P < 0.05 comparing with and without PMA activation.
c P < 0.05 comparing the presence and absence of a test compound.
d P < 0.01 comparing the presence and absence of a test compound.
regardless of the presence or absence of FBS. Moreover,
we detected small peaks with m/z values corresponding to
dehydrated products of secoA-COOH and secoB-COOH
(5-oxo-secocholestan-6-oic acid and 5␤-hydroxy-B-nor-
cholestane-6-oic acid, respectively). About 20% of seco-
sterol-A added to the 10% FBS-containing medium remained
unidentified. Secosterol-A could be converted to the above
minor or unidentified metabolites, or it could be bound to
proteins present in the FBS. Further studies are needed to
identify the fate of secosterol-A in vivo.
We next examined whether secosterol-A and -B are
formed by nHL-60 cells in culture (Fig. 2). In the case of the
experiments with the FBS-free medium, we added both
cholesterol and IgG in the same concentrations as those
present in 10% FBS. It was found that significantly increased
amounts of secosterol-A and its oxidation product secoA-
COOH, but not secosterol-B, were formed by the incuba-
tion of activated nHL-60 cells. Similarly, the formation of
secosterol-A was also observed with activated neutrophils
isolated from WT mice (Fig. 4). Under these conditions,
the formation of secosterol-A was dependent on the con-
centrations of cholesterol and IgG added exogenously (Fig.
3). Several enzyme inhibitors and scavengers for different
reactive species also inhibited the nHL-60 cell-mediated for-
mation of secosterol-A in the FBS-free medium, implying
that the activated cells generate superoxide, hydrogen perox-
ide, hypochlorous acid, and singlet oxygen, all of which are
required to form secosterol-A (Table 1). On the basis of
these findings, we can speculate that secosterol-A was
formed in the FBS-free medium by the reaction of exoge-
nously added cholesterol with an ozone-like oxidant(s) gen-
erated from other reactive species (superoxide, hydrogen
peroxide, hypochlorous acid, and singlet oxygen) in the
presence of IgG. In addition, the greater amount of seco-
sterol-B formation observed in PMA-activated nHL-60 cells
or WT mice neutrophil cultures containing 10% FBS could
be mainly derived from a Hock-cleavage of 5␣-hydroperoxy
cholesterol, as it cannot be explained merely by the aldoliza-
tion of generated secosterol-A.
In addition to our findings showing that secosterol-A or
-B added to the culture medium were immediately oxi-
dized to form secoA-COOH or secoB-COOH (Fig. 1), we
found for the first time that secoA-COOH and secoB-
COOH were formed in the PMA-activated nHL-60 cells
culture (Fig. 2). SecoA-COOH has been reported to in-
duce the fibrilization of apolipoprotein C-II (24) and
␣-synuclein (12). As no data have been available concern-
ing the presence of these oxidized secosterols in vivo, fur-
ther studies are needed to establish the levels of metabolites
of secosterols in in vivo samples.
We previously reported that almost equal amounts of
secosterol-A and -B were formed by the reaction of choles-
MPO-dependent formation of secosterols in vitro and in vivo
93

Fig. 4. Formation of secosterols by neutrophils isolated from WT
or MPO-deficient mice. Isolated neutrophils were incubated with
(closed bars) or without (open bars) 10 nM PMA in the medium
without FBS but containing 30 ␮g /ml cholesterol and 6 µg/ml
IgG (A) or containing 10% FBS (B). Values are means SD (n = 3).
Statistically significant: *P < 0.05 compared the amounts formed
by the cells without PMA activation. Dotted lines represent the lev-
els of preformed secosterols in the medium used in each assay. M,
medium; MPO, myeloperoxidase; WT, wild-type.
terol with human MPO in the presence of its substrates
hydrogen peroxide and ClϪ (15). The results in this study
also indicate that inhibitors of MPO significantly attenu-
ated the formation of secosterols by the PMA-activated
nHL-60 cells. These data strongly suggest that MPO is in-
volved in the formation of secosterols. To elucidate the
role of MPO, we examined the formation of secosterols by
neutrophils isolated from WT and MPO-deficient mice
(Fig. 4). It was found that similar to the results obtained
with the PMA-activated nHL-60 cells, increased amounts
of secosterol-A and -B were formed when PMA-activated
neutrophils isolated from WT mice were cultured in the
medium without and with 10% FBS, respectively. Interest-
ingly, these increased formations of secosterols were not
observed when neutrophils isolated from MPO-deficient
mice were cultured under the same conditions. Further-
more, low levels of secosterol-A and -B were detected in the
plasma samples and liver obtained from WT mice, whereas
their levels were much lower in those obtained from MPO-
deficient mice (Fig. 6). Also interestingly, upon the induc-
tion of acute inflammation by LPS injection, the plasma
levels of both secosterol-A and -B were increased signifi-
cantly in the WT mice, whereas no increase was found in
the MPO-deficient mice. These results clearly indicate that
MPO is required for the formation of secosterols in vivo.
We have also shown that the formation of secosterol-A
but not -B due to the reaction of cholesterol with PMA-
Fig. 5. LC-MS/MS analyses of secosterol-A and secosterol-B in
mice plasma. Typical chromatograms obtained by monitoring the ion
transition of m/z 666.5/170.2 for the DH derivatives of secosterol-A
and secosterol-B extracted from a plasma sample of WT mouse
(A) and MPO-deficient mouse (C) and 668.5/170.2 for those of
internal standards (IS) (3,4-13C-secosterol-A and -B) (B, D). No in-
terfering peaks were detected when the derivatized extract of WT
mouse plasma without addition of IS was analyzed for 668.5/170.2
(E). DH, dansyl hydrazine; MPO, myeloperoxidase; secosterol-A,
3␤-hydroxy-5-oxo-5,6-secocholestan-6-al; secosterol-B, 3␤-hydroxy-
5␤-hydroxy-B-norcholestane-6␤-carboxaldehyde; WT, wild-type.
activated nHL-60 cells in the FBS-free medium is enhanced
by IgG, but it is decreased by the various enzyme inhibitors
and scavengers for different reactive species and by neu-
trophils from MPO-deficient mice. These results support
the notion proposed by Babior et al. (1) and Nieve and
Wentworth (2) that a reactive species with a chemical sig-
nature similar to that of ozone is generated by activated
human neutrophils. On the basis of our results and those
of Babior et al. (1) and Nieve and Wentworth (2), we
can speculate the following reaction mechanism for the
94
Journal of Lipid Research Volume 52, 2011

hypochlorous acid; iv) hypochlorous acid in the presence
of hydrogen peroxide generates singlet oxygen; v) IgG
catalyzes the formation of an ozone-like oxidant in the
presence of singlet oxygen and hydrogen peroxide; and
vi) an ozone-like oxidant reacts with cholesterol to form
secosterol-A (Fig. 7).
On the other hand, increased amounts of secosterol-B
but not -A were formed when the PMA-activated nHL-60
cells and neutrophils isolated from WT mice were cultured
in the medium containing 10% FBS (Figs. 2 and 4). Fur-
thermore, higher levels of secosterol-B than -A were de-
tected in the samples of plasma and liver collected from WT
mice compared to those from MPO-deficient mice (Fig. 6).
The secosterol-B levels were also significantly increased in
the plasma samples of LPS-treated WT mice (Fig. 6). The
origin of secosterol-B formed under these conditions is
currently unknown, although there are two possibilities as
described above: a) the conversion of secosterol-A to seco-
sterol-B by aldolization (2, 10), and b) the direct formation
of secosterol-B by the Hock-cleavage of 5␣-hydroperoxy
cholesterol (13, 14). On the basis of our results, the former
pathway seems to be more important than any other. How-
ever, it is clear that under inflammatory conditions, in-
creased amounts of secosterol-B could be formed and may
exert adverse effects as discussed below.
It is widely accepted that oxysterols resulting from the
chemical or enzymatic oxidation of cholesterol play pivotal
roles in the development of cardiovascular diseases and
atherosclerotic lesions (25–28). They might also be in-
volved in the development of important degenerative
diseases, such as Alzheimer’s disease (29), osteoporosis
(30–32), and age-related macular degeneration (32).
In addition, it has been reported that secosterols induced
apoptotic cell death at ␮M concentrations in several cultured
cells, such as T- and B-cells, macrophage, cardiomyoblasts,
and abdominal aortic endothelial cells (10, 19, 20). We
also observed that secosterols are much more strongly cyto-
toxic (more than 10-fold) compared with five known toxic
oxysterols, including 7␤-hydroxycholesterol, 7-ketocho-
lesterol, 5␤,6␤-epoxycholesterol, and 25-hydroxycholesterol
for 48 h incubation in HL-60 cells (Tomono et al., unpub-
lished observations). Further studies are warranted to
elucidate the roles of cholesterol ozonolysis products, in-
cluding their metabolites, in the development of athero-
sclerosis, neurodegenerative diseases, and other oxidative
stress-related disorders, especially in the context of their
reactions with tissue components, such as proteins, nucleic
acids, and lipids, and their resulting effects on the biologi-
cal functions of these components.
In conclusion, we demonstrated that secosterols are
formed by PMA-activated human neutrophil-like cells as well
as activated neutrophils obtained from WT mice. These acti-
vated neutrophils produce increased amounts of secosterol-A
in the medium without 10% FBS, to which cholesterol was
added exogenously. This secosterol-A formation was en-
hanced significantly by IgG, and the experiments with some
enzyme inhibitors and scavengers of ROS imply that the reac-
tion requires the generation of superoxide, hydrogen perox-
ide, hypochlorous acid, and singlet oxygen. On the basis of
Fig. 6. Levels of secosterols detected in plasma samples and in
the liver of WT or MPO-deficient mice. A, B: The levels of seco-
sterol-A and secosterol-B in plasma samples of mice collected at 0, 6
and 22 h after the intraperitoneal injection of LPS (30 mg/kg BW).
Values are means SD (n = 2-4 in each time point). Statistically
significant: *P < 0.05, **P < 0.01 compared with the levels detected
in untreated mice; #P < 0.01 compared with the levels between WT
and MPO-deficient mice at each time point. C: The levels of seco-
sterol-B in the liver samples collected from untreated WT (n = 8)
and MPO-deficient (n = 7) mice. Median (m) values are presented.
LPS, lipopolysaccharide; MPO, myeloperoxidase; secosterol-A, 3␤-
hydroxy-5-oxo-5,6-secocholestan-6-al; secosterol-B, 3␤-hydroxy-5␤-
hydroxy-B-norcholestane-6␤-carboxaldehyde; WT, wild-type; BW,
body weight.
formation of secosterol-A by PMA-activated nHL-60 cells
in the FBS-free medium: i) the activated cells produce su-
peroxide anion by NADPH oxidase and/or xanthine oxi-
dase; ii) superoxide is converted to hydrogen peroxide by
SOD; iii) hydrogen peroxide is used by MPO to generate
MPO-dependent formation of secosterols in vitro and in vivo
95

Fig. 7. Proposed mechanisms for the formation of
secosterol-A, secosterol-B, and their oxidation products
by activated neutrophils. ␣, ␤-unsaturated-secosterol-B,
3␤-hydroxy-B-norcholest-5-ene; HOCl, hypochlorous
acid; MPO, myeloperoxidase; secoA-COOH, 3␤-
hydroxy-5-oxo-secocholestan-6-oic acid; secoB-COOH,
3␤-hydroxy-5␤-hydroxy-B-norcholestane-6-oic acid;
secosterol-A, 3␤-hydroxy-5-oxo-5,6-secocholestan-6-al;
secosterol-B, 3␤-hydroxy-5␤-hydroxy-B-norcholestane-
6␤-carboxaldehyde.
structure of the principal product. J. Chem. Soc., Perkin Trans. 1:
1222–1223.
9. Wentworth, A. D., B. D. Song, J. Nieva, A. Shafton, S. Tripurenani,
and P. Wentworth, Jr. 2009. The ratio of cholesterol 5,6-seco-
sterols formed from ozone and singlet oxygen offers insight into
the oxidation of cholesterol in vivo. Chem. Commun. (Camb.) (21):
3098–3100.
10. Wentworth, P., Jr., J. Nieva, C. Takeuchi, R. Galve, A. D. Wentworth,
R. B. Dilley, G. A. DeLaria, A. Saven, B. M. Babior, K. D. Janda,
et al. 2003. Evidence for ozone formation in human atherosclerotic
arteries. Science. 302: 1053–1056.
11. Zhang, Q., E. T. Powers, J. Nieva, M. E. Huff, M. A. Dendle, J.
Bieschke, C. G. Glabe, A. Eschenmoser, P. J. Wentworth, R. A.
Lerner, et al. 2004. Metabolite-initiated protein misfolding may trig-
ger Alzheimer’s disease. Proc. Natl. Acad. Sci. USA. 101: 4752–4757.
12. Bosco, D. A., D. M. Fowler, Q. Zhang, J. Nieva, E. T. Powers, P. J.
Wentworth, R. A. Lerner, and J. W. Kelly. 2006. Elevated levels of
oxidized cholesterol metabolites in Lewy body disease brains accel-
erate alpha-synuclein fibrilization. Nat. Chem. Biol. 2: 249–253.
13. Brinkhorst, J., S. J. Nara, and D. A. Pratt. 2008. Hock cleavage of
cholesterol 5alpha-hydroperoxide: an ozone-free pathway to the
cholesterol ozonolysis products identified in arterial plaque and
brain tissue. J. Am. Chem. Soc. 130: 12224–12225.
14. Uemi, M., G. E. Ronsein, S. Miyamoto, M. H. G. Medeiros, and P.
Di Mascio. 2009. Generation of cholesterol carboxyaldehyde by the
reaction of singlet molecular oxygen [O2 (1Delta(g))] as well as
ozone with cholesterol. Chem. Res. Toxicol. 22: 875–884.
15. Tomono, S., N. Miyoshi, K. Sato, Y. Ohba, and H. Ohshima. 2009.
Formation of cholesterol ozonolysis products through an ozone-
free mechanism mediated by the myeloperoxidase-H2O2-chloride
system. Biochem. Biophys. Res. Commun. 383: 222–227.
16. Bieschke, J., Q. Zhang, E. T. Powers, R. A. Lerner, and J. W. Kelly.
2005. Oxidative metabolites accelerate Alzheimer’s amyloido-
genesis by a two-step mechanism, eliminating the requirement
for nucleation. Biochemistry. 44: 4977–4983.
17. Bieschke, J., Q. Zhang, D. A. Bosco, R. A. Lerner, E. T. Powers,
P. J. Wentworth, and J. W. Kelly. 2006. Small molecule oxidation
products trigger disease-associated protein misfolding. Acc. Chem.
Res. 39: 611–619.
18. Spillantini, M. G., and M. Goedert. 2000. The alpha-synucleino-
pathies: Parkinson’s disease, dementia with Lewy bodies, and mul-
tiple system atrophy. Ann. N. Y. Acad. Sci. 920: 16–27.
19. Sathishkumar, K., S. N. Murthy, and R. M. Uppu. 2007. Cytotoxic
effects of oxysterols produced during ozonolysis of cholesterol
in murine GT1–7 hypothalamic neurons. Free Radic. Res. 41:
82–88.
20. Sathishkumar, K., M. Haque, T. E. Perumal, J. Francis, and R. M.
Uppu. 2005. A major ozonation product of cholesterol, 3beta-
hydroxy-5-oxo-5,6-secocholestan-6-al, induces apoptosis in H9c2
cardiomyoblasts. FEBS Lett. 579: 6444–6450.
21. Collins, S. J., F. W. Ruscetti, R. E. Gallagher, and R. C. Gallo. 1978.
Terminal differentiation of human promyelocytic leukemia cells
these findings, we conclude that secosterol-A, at least in the
medium without 10% FBS, is formed by an ozone-like oxi-
dant, which is generated by the catalysis of IgG in the pres-
ence of singlet oxygen and hydrogen peroxide. In addition,
we clearly demonstrated that MPO plays an important role in
the formation of secosterols. Possibly, hypochlorous acid gen-
erated by MPO is needed to generate singlet oxygen and,
therefore, an ozone-like oxidant in the presence of hydrogen
peroxide. As MPO has been implicated in the pathogenesis
of atherosclerosis and neurodegenerative diseases, such as
Alzheimer’s disease (33, 34), further studies are needed to
elucidate the biological roles of the MPO-mediated forma-
tion of secosterols and their further oxidation products in
the development of these diseases.
The authors thank Yohei Wakao, Azusa Suzuki, Mina Yoshida,
Kei Yamamoto, and Noriko Matsumoto for their technical
support in animal experiments.
REFERENCES
1. Babior, B. M., C. Takeuchi, J. Ruedi, A. Gutierrez, and P. J.
Wentworth. 2003. Investigating antibody-catalyzed ozone gen-
eration by human neutrophils. Proc. Natl. Acad. Sci. USA. 100:
3031–3034.
2. Nieva, J., and P. J. Wentworth. 2004. The antibody-catalyzed water
oxidation pathway–a new chemical arm to immune defense? Trends
Biochem. Sci. 29: 274–278.
3. Yamashita, K., T. Miyoshi, T. Arai, N. Endo, H. Itoh, K. Makino, K.
Mizugishi, T. Uchiyama, and M. Sasada. 2008. Ozone production
by amino acids contributes to killing of bacteria. Proc. Natl. Acad.
Sci. USA. 105: 16912–16917.
4. Kettle, A. J., B. M. Clark, and C. C. Winterbourn. 2004. Superoxide
converts indigo carmine to isatin sulfonic acid: implications for
the hypothesis that neutrophils produce ozone. J. Biol. Chem. 279:
18521–18525.
5. Kettle, A. J., and C. C. Winterbourn. 2005. Do neutrophils produce
ozone? An appraisal of current evidence. Biofactors. 24: 41–45.
6. Gumulka, J., and L. L. Smith. 1983. Ozonization of cholesterol.
J. Am. Chem. Soc. 105: 1972–1979.
7. Jaworski, K., and L. L. Smith. 1988. Ozonization of cholesterol in
nonparticipating solvents. J. Org. Chem. 53: 545–554.
8. Paryzek, Z., J. Martynow, and W. Swoboda. 1990. The reaction of
cholesterol with ozone and alcohols: a revised mechanism and
96
Journal of Lipid Research Volume 52, 2011

induced by dimethyl sulfoxide and other polar compounds. Proc.
Natl. Acad. Sci. USA. 75: 2458–2462.
28. Larsson, D. A., S. Baird, J. D. Nyhalah, X-M. Yuan, and W. Li.
2006. Oxysterol mixtures, in atheroma-relevant proportions, dis-
play synergistic and proapoptotic effects. Free Radic. Biol. Med. 41:
902–910.
29. Vaya, J., and H. M. Schipper. 2007. Oxysterols, cholesterol homeo-
stasis, and Alzheimer disease. J. Neurochem. 102: 1727–1737.
30. Kha, H. T., B. Basseri, D. Shouhed, J. Richardson, S. Tetradis, T. J.
Hahn, and F. Parhami. 2004. Oxysterols regulate differentiation of
mesenchymal stem cells: pro-bone and anti-fat. J. Bone Miner. Res.
19: 830–840.
22. Aratani, Y., H. Koyama, S. Nyui, K. Suzuki, F. Kura, and N. Maeda. 1999.
Severe impairment in early host defense against Candida albicans in
mice deficient in myeloperoxidase. Infect. Immun. 67: 1828–1836.
23. Higashi, T., T. Ichikawa, S. Inagaki, J. Z. Min, T. Fukushima, and
T. Toyo’oka. 2010. Simple and practical derivatization procedure
for enhanced detection of carboxylic acids in liquid chromatog-
raphy-electrospray ionization-tandem mass spectrometry. J. Pharm.
Biomed. Anal. 52: 809–818.
24. Stewart, C. R., L. M. Wilson, Q. Zhang, C. L. Pham, L. J. Waddington,
M. K. Staples, D. Stapleton, J. W. Kelly, and G. J. Howlett. 2007.
Oxidized cholesterol metabolites found in human atherosclerotic
lesions promote apolipoprotein C–II amyloid fibril formation.
Biochemistry. 46: 5552–5561.
25. Bjorkhem, I., O. Andersson, U. Diczfalusy, B. Sevastik, R. J. Xiu, C.
Duan, and E. Lund. 1994. Atherosclerosis and sterol 27-hydroxylase:
evidence for a role of this enzyme in elimination of cholesterol from
human macrophages. Proc. Natl. Acad. Sci. USA. 91: 8592–8596.
26. Salonen, J. T., K. Nyyssonen, R. Salonen, E. Porkkala-Sarataho,
T. P. Tuomainen, U. Diczfalusy, and I. Bjorkhem. 1997. Lipoprotein
oxidation and progression of carotid atherosclerosis. Circulation.
95: 840–845.
31. Shouhed, D., H. T. Kha, J. A. Richardson, C. M. Amantea, T. J.
Hahn, and F. Parhami. 2005. Osteogenic oxysterols inhibit the
adverse effects of oxidative stress on osteogenic differentiation of
marrow stromal cells. J. Cell. Biochem. 95: 1276–1283.
32. Liu, H., L. Yuan, S. Xu, K. Wang, and T. Zhang. 2005. Cholestane-
3beta,5alpha,6beta-triol inhibits osteoblastic differentiation and
promotes apoptosis of rat bone marrow stromal cells. J. Cell.
Biochem. 96: 198–208.
33. Nicholls, S. J., and S. L. Hazen. 2009. Myeloperoxidase, modified
lipoproteins, and atherogenesis. J. Lipid Res. 50: S346–S351.
34. Maki, R. A., V. A. Tyurin, R. C. Lyon, R. L. Hamilton, S. T. DeKosky,
V. E. Kagan, and W. F. Reynolds. 2009. Aberrant expression of
myeloperoxidase in astrocytes promotes phospholipid oxidation
and memory deficits in a mouse model of Alzheimer disease.
J. Biol. Chem. 284: 3158–3169.
27. Brown, A. J., and W. Jessup. 1999. Oxysterols and atherosclerosis.
Atherosclerosis. 142: 1–28.
MPO-dependent formation of secosterols in vitro and in vivo
97