Superior anticarcinogenic activity of trans, trans-conjugated linoleic acid in N-methyl-N-nitrosourea-induced rat mammary tumorigenesis

Article abstract of DOI:10.1021/jf100117a

The anticarcinogenic activity of a mixture of trans,trans-conjugated linoleic acid (trans,trans-CLA) was investigated in rat mammary tumorigenesis induced by N-methyl-N-nitrosourea (MNU), with references to cis-9,trans-11-CLA and trans-10,cis-12-CLA isomers. Female, 7-week-old Sprague-Dawley rats were intraperitoneally injected with MNU (50 mg/kg of body weight) and then subjected to one of five diets (control, 1% trans,trans-CLA, 1% cis-9,trans-11-CLA, 1% trans-10,cis-12-CLA, and 1% linoleic acid; 8 rats/group) for 16 weeks. Food and water were made available ad libitum. trans,trans-CLA significantly (p < 0.05) reduced tumor incidence, number, multiplicity, and size and significantly (p < 0.05) increased apoptosis, relative to cis-9,trans-11-CLA and trans-10,cis-12-CLA. The molecular mechanism of trans,trans-CLA was elucidated by measuring apoptosis-related gene products and fatty acid composition in tumors. trans,trans-CLA led to increases in the p53 protein and Bax protein levels but suppressed the expression of Bcl-2 protein. The activation of caspase-3 led to the cleavage of poly(ADP-ribose) polymerase, which resulted in the execution of apoptosis. In addition, trans,trans-CLA reduced cytosolic phospholipase A2, cyclooxygenease-2, and peroxisome proliferator-activated receptor γ protein levels. These results suggest that the trans,trans-CLA inhibits MNU-induced rat mammary tumorigenesis through the induction of apoptosis in conjunction with the reduction of arachidonic acid metabolites and that the efficacy of trans,trans-CLA is superior to cis-9,trans-11-CLA and trans-10,cis-12-CLA.

Full text of DOI:10.1021/jf100117a

5670 J. Agric. Food Chem. 2010, 58, 5670–5678
DOI:10.1021/jf100117a
Superior Anticarcinogenic Activity of trans,trans-Conjugated
Linoleic Acid in N-Methyl-N-nitrosourea-Induced
Rat Mammary Tumorigenesis
‡
M
OHAMMAD A. ISLAM,† YOUNG S. KIM,† TAE W. O
H
,† GON S. KIM,‡ CHUNG K. WON
,
^
,†
H
OON G. KIM,§ MYUNG S. CHOI, JEONG O. KIM
,
AND
Y
EONG L. H *
A
†Division of Applied Life Science (BK21 Program), Graduate School, and Institute of Agriculture and
Life Science, ‡Laboratory of Biochemistry, School of Veterinary Medicine, and Divison of Environmental
Forest Science, Gyeongsang National University, Jinju 660-701, Republic of Korea, §Department of
Internal Medicine, Gyeongsang National University Hospital, Gyeongsang National University,
Jinju 660-751, Republic of Korea, and ^HK Biotech Company, Limited, Jinju 660-844, Republic of Korea
The anticarcinogenic activity of a mixture of trans,trans-conjugated linoleic acid (trans,trans-CLA)
was investigated in rat mammary tumorigenesis induced by N-methyl-N-nitrosourea (MNU), with
references to cis-9,trans-11-CLA and trans-10,cis-12-CLA isomers. Female, 7-week-old Sprague-
Dawley rats were intraperitoneally injected with MNU (50 mg/kg of body weight) and then subjected
to one of five diets (control, 1% trans,trans-CLA, 1% cis-9,trans-11-CLA, 1% trans-10,cis-12-CLA,
and 1% linoleic acid; 8 rats/group) for 16 weeks. Food and water were made available ad libitum.
trans,trans-CLA significantly (p < 0.05) reduced tumor incidence, number, multiplicity, and size and
significantly (p < 0.05) increased apoptosis, relative to cis-9,trans-11-CLA and trans-10,cis-12-CLA.
The molecular mechanism of trans,trans-CLA was elucidated by measuring apoptosis-related gene
products and fatty acid composition in tumors. trans,trans-CLA led to increases in the p53 protein
and Bax protein levels but suppressed the expression of Bcl-2 protein. The activation of caspase-3
led to the cleavage of poly(ADP-ribose) polymerase, which resulted in the execution of apoptosis. In
addition, trans,trans-CLA reduced cytosolic phospholipase A2, cyclooxygenease-2, and peroxisome
proliferator-activated receptor γ protein levels. These results suggest that the trans,trans-CLA
inhibits MNU-induced rat mammary tumorigenesis through the induction of apoptosis in conjunction
with the reduction of arachidonic acid metabolites and that the efficacy of trans,trans-CLA is superior
to cis-9,trans-11-CLA and trans-10,cis-12-CLA.
KEYWORDS: Conjugated linoleic acid (CLA); trans,trans-CLA; N-methyl-N-nitrosourea (MNU); rat
mammary tumorigenesis; apoptosis
INTRODUCTION
lines tested to date (4, 5). Because CLA contains several isomers,
individual CLA isomers (mainly cis-9,trans-11-CLA and trans-
10,cis-12-CLA) have been the ones most extensively studied in
carcinogen-induced carcinogenesis inexperimentalanimals (6-8)
and cancer cell lines (9). To date, it appears that both cis-9,trans-
11-CLA and trans-10,cis-12-CLA isomers are similarly active in
the inhibition of carcinogen-induced carcinogenesis. For exam-
ple, diets containing 1% cis-9,trans-11-CLA and trans-10,cis-
12-CLA are equally effective in reducing tumors and inducing
apoptosis in the colonic mucosa of rats treated with 1,2-dimethyl-
hydrazine (DMH) (7), and both cis-9,trans-11-CLA and trans-10,
cis-12-CLA isomers significantly decrease N-methyl-N-nitrosourea
(MNU)-induced rat mammary tumors (6). However, differential
effects between cis-9,trans-11-CLA and trans-10,cis-12-CLA on
tumorigenesis have been demonstrated in human cancer cell
studies, with the trans-10,cis-12-CLA isomer exhibiting higher
potency against HT-29 colorectal cancer cells (9), MG-63 osteo-
sarcoma cells (10), and MCF-7 breast cancer cells (11), relative to
the potency of cis-9,trans-11-CLA. The differential effects of
Conjugated linoleic acid (CLA) is a collective term that refers
to a mixture of positional (7,9; 8,10; 9,11; 10,12; 11,13; and 12,14)
and geometric (trans,cis; cis,trans; cis,cis; and trans,trans) isomers
of octadecadienoic acid (C18:2) with a conjugated double-bond
system (1). The predominant CLA isomers in synthetic CLA,
synthesized from linoleic acid (LA) by alkaline isomerization, are
cis-9,trans-11-CLA and trans-10,cis-12-CLA isomers; small
amounts of trans-10,trans-12-CLA, trans-9,trans-11-CLA, and
other isomers also occur (2). Many studies have investigated the
beneficial effect of CLA against some types of cancer, athero-
sclerosis, hypertension, diabetes, and inflammation (3).
CLA is a potent anticancer agent in carcinogen-induced
carcinogenesis in most animal models and human cancer cell
*To whom correspondence should be addressed: Division of Applied
Life Science, Graduate School, Gyeongsang National University,
Jinju 660-701, Korea. Telephone: þ82-55-751-5471. Fax: þ82-55-757-0178.
E-mail: ylha@gnu.ac.kr.
pubs.acs.org/JAFC
Published on Web 04/07/2010
© 2010 American Chemical Society

Article
J. Agric. Food Chem., Vol. 58, No. 9, 2010 5671
these two CLA isomers on tumorigenesis in animal models and
human cancer cell lines have yet to be clarified.
randomized by body weight and divided into control, 1% trans,trans-
CLA, 1% cis-9,trans-11-CLA, 1% trans-10,cis-12-CLA, and 1% LA
treatment groups (8 rats for each group). Amounts added as ingredients
to each diet were adjusted accordingly by the purity of CLA isomers and
LA. Rats were given an intraperitoneal (i.p.) injection of MNU at a single
dose of 50 mg/kg of body weight, and 1 day later, they were subjected to
diets for 16 weeks. Diet and water were made available ad libitum. Starting
at the initiation of tumors and continuing thereafter, body weight and food
intake were measured weekly. All rats were sacrificed 17 weeks after
mammary tumor initiation. Rat care and experimental procedures were in
accordance with Gyeongsang National University Animal Ethics Guide-
lines (GNU-LA-18).
Mammary Tumor Analysis. Tumor weight, size, number, and
incidence were recorded from rats sacrificed at the termination of the
animal experiment. After all debris had been trimmed off, all of the
mammary tumors of each treatment group were longitudinally cut into
two pieces and saved for further analysis. One piece from each tumor was
fixed immediately in 10% phosphate-buffered formalin (pH 7.3) for
histological examination and terminal deoxynucleotidyl transferase-
mediated dUTP nick end-labeling (TUNEL) assay, and the rest of the
pieces were kept in a deep freezer maintained at -71 ꢀC until analyzed for
biochemical markers.
Histological Examination and TUNEL Assay. Mammary tumors
fixed in 10% buffered formalin were embedded in paraffin, followed by the
preparation of paraffin sections with a thickness of 4 μm and staining with
hematoxylin and eosin (16). The mammary tumors were diagnosed and
classified as fibroadenoma or adenocacinoma according to the criteria
for the classification of rat mammary tumors (17). TUNEL assay was
applied to detect apoptotic cells in mammary tumors using a Frag EL
DNA fragmentation detection kit (18). The sections (4 μm thick), adding
dioxigenin-labeled nucleotides to the 30-OH end of the DNA, were
mounted in Merckoglas (Merck, Whitehouse Station, NJ) and observed
using a model DM6000B microscope (Leica Wetzlar, Germany).
DNA Fragmentation Analysis. Mammary tumor was sliced into a
mortar, frozen by the slow addition of liquid nitrogen, disrupted, and
homogenized completely. DNA was isolated from the homogenized
tissues using the i-genomic CTB DNA Extraction Mini Kit (11). DNA
fragmentation patterns were analyzed by electrophoresis in a 1.2%
agarose gel, and the DNA was then treated with ethidium bromide for
visualization under UV light.
Western Blotting. Mammary tumors were homogenized in RIPA
buffer [150 mM NaCl, 0.1% NP40, 0.5% deoxycholic acid, 0.1% sodium
dodecyl sulfate (SDS), and 50 mM Tris at pH 7.4] containing protease
inhibitors, 21 μM leupeptin, and 15 μM pepstatin (10). The homogenates
were centrifuged at 13000 rpm for 5 min at 4 ꢀC, and the supernatant was
collected for protein samples. The protein content was determined by the
Bradford protein assay protocol. Western blotting for p53, Bax, Bcl-2,
caspase-3, PARP, PPARγ, cPLA2, COX-2, caspase-9, and caspase-8
proteins were performed according to previously described methods
(10, 11). Bound antibodies were detected under UV light with the
assistance of an ECL Western blotting detection kit. The relative protein
levels were determined using an Eastman Kodak model Gel Logic 100
Imaging System (Eastman Kodak, Rochester, NY).
In contrast, little attention has been paid to the inhibitory
effects of a mixture of trans,trans-CLA, containing trans-7,trans-9-,
trans-8,trans-10-, trans-9,trans-11-, trans-10,trans-12-, trans-11,
trans-13-, and trans-12,trans-14-CLA isomers (2) or its indivi-
dual trans,trans-CLA isomers on cancer cell lines and chemical-
induced animal carcinogenesis. In the current study, we found
that trans,trans-CLA exhibited the greatest potency in growth
inhibition of an osteosarcoma cell line, MG-63 (10), and a breast
cancer cell line, MCF-7 (11), relative to cis-9,trans-11-CLA and
trans-10,cis-12-CLA. This inhibition was dose-dependent. The
trans-9,trans-11-CLA isomer inhibits the growth of Caco-2 colon
cancer cells (12) and the development of azoxymethane (AOM)-
induced colonic aberrant crypt foci in rats by inducing apopto-
sis (13). Meanwhile, except for one previously published paper
by Xasui et al. (13) on colon cancer, no reports are available
concerning the growth inhibitory effect of trans,trans-CLA on
animal tumorigenesis. Hence, it would be of significance to
examine the anticarcinogenic activity of trans,trans-CLA in
carcinogen-induced animal carcinogenesis, especially rat mam-
mary carcinogenesis, as compared to that of cis-9,trans-11-CLA
and trans-10,cis-12-CLA.
The purpose of this study was to evaluate the anticarcinogenic
activity of trans,trans-CLA in MNU-induced rat mammary tumo-
rigenesis, relative to cis-9,trans-11-CLA, trans-10,cis-12-CLA,
and LA. We also examined whether apoptosis mediated by
mitochondrial dysfunction is one of the mechanistic anticarcino-
genic actions of trans,trans-CLA.
MATERIALS AND METHODS
Materials. LA (99.0%), MNU, leupeptin, pepstatin, and Mayer’s
hematoxylin were obtained from Sigma-Aldrich (St. Louis, MO). The
i-genomic CTB DNA Extraction Mini Kit was obtained from iNtRON
Biotechnology (Seongnam, Republic of Korea). Rabbit polyclonal p53,
Bcl-2, caspase-3, and Bax antibodies were purchased from Delta Biolabs
(Campbell, CA). Rabbit polyclonal caspase-9, cytosolic phospholipase A2
(cPLA2), cyclooxygenase-2 (COX-2), peroxisome proliferator-activated
receptor γ (PPARγ), and poly(ADP-ribose) polymerase (PARP) anti-
bodies and anti-rabbit, anti-goat and anti-mouse IgG-horseradish peroxi-
dases were obtained from Santa Cruz Biotechnology (Santa Cruz, CA).
Monoclonal caspase-8 antibody was obtained from Oncogene Research
Products (San Diego, CA). Monoclonal anti-β-actin was purchased from
Sigma-Aldrich. The enhanced chemiluminescence (ECL) Western blotting
kit was purchased from Amersham Biosciences (Buckingham, U.K.). Frag
EL DNA fragmentation detection kit was purchased from Calbiochem
(Burlingame, CA). A mixture of fatty acid methyl ester standards was
obtained from Supelco (Bellefonte, PA). All other reagents used were of
analytical grade.
Preparation of CLA Isomers. CLA methylester was prepared from
LA by alkaline isomerization in propylene glycol, followed by methylation
with 1.0 N H2SO4/methanol (14). The methyl esters of cis-9,trans-11-CLA
and trans-10,cis-12-CLA were separated from the CLA methyl esters by
low-temperature crystallization (15). The trans,trans-CLA methyl ester was
prepared from CLA methyl ester according to the method described
previously (2). The purity of CLA isomer samples was found to be 94.5%
for cis-9,trans-11-CLA, 99.1% for trans-10,cis-12-CLA, and 99.0% for
trans,trans-CLA by gas chromatography (GC) (14). The isomeric composi-
tion of trans,trans-CLA was found to be 0.7% trans-7,trans-9-CLA, 8.3%
trans-8,trans-10-CLA, 41.0% trans-9,trans-11-CLA, 41.6% trans-10,trans-
12-CLA, 7.8% trans-11,trans-13-CLA, and 0.6% trans-12,trans-14-CLA,
analyzed by silver high-performance liquid chromatography (1, 2).
Animal Experiment. Specific pathogen-free (SPF), female, 7-week-
old Sprague-Dawley (SD) rats (Samtako Bio Korea, Osan, Republic of
Korea) were housed in polycarbonate cages (1 rat/cage) in a temperature
(22 ( 2 ꢀC)- and humidity (55 ( 5%)-controlled SPF facility with a 12 h
light/dark cycle system and were fed a basal AIN93G diet (pellet)
(Samtako Bio Korea) as a control diet. After 1 week, animals were
Determination of Phospholipids. Total mammary tumor lipid was
extracted, and the phospholipid fraction from total lipids was separated
using a silica Maxi-Clean cartridge as previously described (19). The
acquired phospholipid was methylated with 20% tetramethylguanidine
(TMG) in methanol (100 ꢀC, 10 min), followed by an additional reaction
with 1.0 N H2SO4/methanol (55 ꢀC, 5 min) (2). Fatty acid methyl esters
were analyzed by GC (Hewlett-Packard 5890) equipped with a flame ioniza-
tion detector (FID) and a fused silica capillary column Supelcowax-10
(60 m ꢀ 0.32 mm inner diameter, 25 μm film thickness). The oven temp-
erature was increased from 180 to 200 ꢀC at a rate of 2 ꢀC/min and then
held for 30 min. Injection port and detector temperatures were 240 and
260 ꢀC, respectively. Nitrogen (99.9%) was usedas a carrier gas with a flow
rate of 2 mL/min. The fatty acids of the samples were identified through
the comparison to the relative retention time of the standards.
Statistical Analysis. The results are presented as mean ( standard
deviation (SD). Statistical analysis was carried out using analysis of
variance (ANOVA), followed by Duncan’s multiple range test. Mean
differences at p < 0.05 were considered to indicate statistical significance.

5672 J. Agric. Food Chem., Vol. 58, No. 9, 2010
Islam et al.
RESULTS
tumor incidence. Tumor incidence in trans,trans-CLA-treated rats
was significantly (p < 0.05) reduced, as compared to that of trans-
10,cis-12-CLA-, cis-9,trans-11-CLA-, and LA-treated rats. Similar
results were obtained from the total tumor number, tumor multi-
plicity, and tumor size. Tumor incidence of rats treated with LA was
significantly (p < 0.05) reduced, relative to control rats. This could
be, in part, due to LA metabolite, such as (()-13-hydroxy-10-oxo-
trans-11-octadecenoid acid (13-HOA) acting as a anti-promoter in
mice (20), although LA acts as a tumor promoter in rats (21).
Further research should be performed to clarify this event.
Inhibition of MNU-Induced SD Rat Mammary Tumorigenesis
by trans,trans-CLA. Table 1 shows the effect of dietary trans,trans-
CLA on the tumor incidence, number, multiplicity, and average
tumor size of SD rats treated with MNU, with references to cis-9,
trans-11-CLA, trans-10,cis-12-CLA, and LA. The efficacy of trans,
trans-CLA on the inhibition of tumorigenesis was superior to the
other CLA isomers. Control rats given MNU alone had 100%
As seen in Figure 1, trans,trans-CLA-treated rats maintained a
higher body weight than cis-9,trans-11-CLA-, trans-10,cis-12-CLA-,
and LA-treated rats from week 6 after MNU treatment, which might
be associated with the enhanced food intake levels. The body weight
of trans,trans-CLA-treated rats at week 16 was maintained signifi-
cantly (p < 0.05) higher than other CLA-treated rats. The ability of
trans,trans-CLA to maintain the appetite and the body weight may
be related to a possible anti-cachectic action of this compound (22).
Histology of Tumors from SD Rats Treated with trans,trans-
CLA. Figure 2 shows the results of microscopic analysis of tumors
stained with hematoxylin and eosin (17). The tumors in control
rats were all adenocarcinomas. The lesions were rather cellular
and consisted of a proliferation of epithelial and stromal compo-
nents; the epithelial elements ranged from reactive to malignant
in nature. The majority of the tumors was adenomatous and
exhibited features of florid sclerosing adenosis mixed with low-
grade ductal carcinoma in situ. Focal areas exhibited features
Table 1. Effect of trans,trans-CLA on the Incidence, Total Number, Multi-
plicity, and Size of Tumors in SD Rats Induced by MNU
incidence
(%)b
total
tumorc
multiplicity
(tumor/rat)d
tumor size
(cm)e
treatmenta
control
1% LA
100 a
83 b
67 c
50 d
33 e
30
27
18
16
12
5.0 ( 0.8f a
4.5 ( 0.8 a
3.0 ( 0.6 b
2.7 ( 0.4 b
2.0 ( 0.3 c
4.5 ( 0.8 a
3.9 ( 0.5 a
2.8 ( 0.4 b
2.4 ( 0.6 b
1.8 ( 0.3 c
1% cis-9,trans-11-CLA
1% trans-10,cis-12-CLA
1% trans,trans-CLA
a Rats were injected, i.p., with a single dose of MNU (50 mg/kg of body weight)
and, 1 day later, subjected to one of the following diets for 16 weeks: control, 1% LA,
1% cis-9,trans-11-CLA, 1% trans-10,cis-12-CLA, and 1% trans,trans-CLA. b Per-
centages of tumor-bearing rats against eight rats in a given treatment group. Tumor
incidence with different letters is significantly different at p < 0.05 by the χ2 test.
c Total number of tumors induced in a given treatment group with eight rats.
d Number of tumors per tumor-bearing rat. e Average tumor size g0.3 mm in
diameter. f Mean ( SD (n = 8). Means with different letters are significantly different
at p < 0.05 by Duncan’s multiple range tests.
Figure 1. Effect of dietary trans,trans-CLA on body weight and food intake of rats treated with MNU. Data are presented as mean ( SD of eight independent
rats. Means with different letters are significantly different at p < 0.05 by Duncan’s multiple range tests.

Article
J. Agric. Food Chem., Vol. 58, No. 9, 2010 5673
Figure 2. Histochemical analysis of MNU-induced mammary tumors from rats treated with trans,trans-CLA. Tumor tissues were stained with hematoxylin
and eosin.
Figure 3. Induction of apoptosis in MNU-induced rat mammary tumors by trans,trans-CLA. Apoptosis of tumors was assayed by TUNEL (A-E) and DNA
fragmentation (F) assays. (A-D) Control, cis-9,trans-11-CLA, trans-10,cis-12-CLA, and trans,trans-CLA treatments, respectively. The apoptotic cell index is
shown in E. The size markers used are indicated in F. Means with different letters are significantly different at p < 0.05 by Duncan’s multiple range tests.
characteristic of adenocarcinoma, including increased mitotic
indices, atypical mitotic figures, moderate-to-severe cytologic
atypia, coagulative tumor cell necrosis, and jagged infiltrating
margins. The lesions were well-circumscribed in some areas and
were associated with a brisk host response composed of reac-
tive stromal cells and a mixed inflammatory infiltrate. Angio-
genesis was much more prominent in these lesions than in the
adenomas of the supplemented rats. In contrast, tumors from
trans,trans-CLA-treated rats contained epithelial cells that appeared
mildly to moderately cytologically atypical and had low mitotic
indices. Lesions were well-circumscribed and exhibited prominent
papillary architecture. In the tumors from rats treated with cis-9,
trans-11-CLA, trans-10,cis-12-CLA, and LA, lesions were mode-
rately cellular and consisted of epithelial and stromal components.
The overall cytoarchitectural features were indicative of a fibro-
epithelial lesion, such as a fibroadenoma. The stromal and vascular
proliferation was much less than that seen in adenocarcinoma
tumors of control rats. These results suggest that the anticarcino-
genic potential of trans,trans-CLA was superior to other CLA
compounds in rat mammary turmorigenesis induced by MNU.

5674 J. Agric. Food Chem., Vol. 58, No. 9, 2010
Islam et al.
Figure 4. Gene expressions andenzyme activations relatedtoapoptosis inMNU-induced rat mammary tumors bytrans,trans-CLA. The band intensities were
quantified relative to the control. Bars indicate means ( SD (n = 3). (A-D) p53, Bax and Bcl-2, caspase-9, and caspase-3, respectively. Means with different
letters are significantly different at p < 0.05 by Duncan’s multiple range tests.
Induction of Apoptosis in Tumors from SD Rats Treated with
trans,trans-CLA. Apoptosis was determined by TUNEL (panels
A-E of Figure 3) and DNA fragmentation (Figure 3F) assays,
which are considered hallmarks of apoptosis, for mammary
tumor tissues from rats treated with trans,trans-CLA, with
references to cis-9,trans-11-CLA, trans-10,cis-12-CLA, and LA.
TUNEL analysis showed more frequent occurrence of darkly
stained TUNEL-positive nuclei in the mammary epithelium of
trans,trans-CLA-treated rats than in the mammary epithelia of
the control and cis-9,trans-11-CLA-, and trans-10,cis-12-CLA-
treated rats. The apoptotic cell indices were significantly higher
(p < 0.05) in the tumors from trans,trans-CLA-treated rats than
in the tumors from cis-9,trans-11-CLA- and trans-10,cis-12-CLA-
treated rats, although cis-9,trans-11-CLA- and trans-10,cis-12-
CLA-treated rats maintained significantly higher (p < 0.05) indices
in the tumors than LA-treated or control rats (Figure 3E).
Mammary tumors from rats treated with trans,trans-CLA
produced a distinct smear of DNA fragments (Figure 3F), a
typical characteristic of cells undergoing apoptosis. The trans-10,
cis-12-CLA also produced a characteristic smearing but at a lesser
extent than trans,trans-CLA. No or little smearing was seen in
tumors from the control, cis-9,trans-11-CLA-, and LA-treated
rats. Given the results from TUNEL and DNA fragmentation
assays, the induction of apoptosis by trans,trans-CLA was greater
than that of cis-9,trans-11-CLA, trans-10,cis-12-CLA, or LA.
Expression of Apoptotic-Related Proteins from SD Rats Treated
with trans,trans-CLA. The expression of p53, Bax, and Bcl-2 genes
was assessed by Western blotting to establish the role of apoptosis
in MNU-induced mammary tumorigenesis (Figure 4). The p53
tumor suppressor protein was significantly increased (p < 0.05)
in tumors from trans,trans-CLA-treated rats, relative to the
protein levels in tumors from cis-9,trans-11-CLA- and trans-10,
cis-12-CLA-treated rats. trans,trans-CLA treatment induced a
significantly higher level of expression of Bax in the tumors (p <
0.05), as compared to the treatment with cis-9,trans-11-CLA,
trans-10,cis-12-CLA, and LA. In contrast to p53 and Bax pro-
teins, the expression of anti-apoptotic Bcl-2 was significantly reduced
in trans,trans-CLA-treated mammary tumors (p < 0.05), relative to
the expression levels in the cis-9,trans-11-CLA and trans-10,cis-12-
CLA groups. Consequently, trans,trans-CLA-induced apoptosis in
MNU-induced mammary tumorigenesis was associated with a
reduced Bcl-2 protein level and an increased Bax protein level.

Article
J. Agric. Food Chem., Vol. 58, No. 9, 2010 5675
Figure 5. Gene expressions and enzyme activations related to apoptosis and/or AA metabolism in MNU-induced rat mammary tumors by trans,trans-CLA.
The band intensities were quantified relative to the control. Bars indicate means ( SD (n = 3). (A-D) Caspase-8, PARP, cPLA2, and PPARγ, respectively.
Means with different letters are significantly different at p < 0.05 by Duncan’s multiple range tests.
To determine whether caspase activation is a required signaling
event for apoptosis and whether increases in caspase activity
coincided with apoptotic cell death, the protein levels of caspase-3,
caspase-9, and caspase-8 were measured in tumors from rats
treated with trans,trans-CLA. trans,trans-CLA treatment signifi-
cantly elevated caspase-9 and caspase-3 protein levels (p < 0.05),
relative to the levels observed following cis-9,trans-11-CLA,
trans-10,cis-12-CLA, and LA treatments (Figure 4). The protein
levels of caspase-9 and caspase-3 in the tumors of rats treated with
trans,trans-CLA were 1.5- and 1.8-fold higher than those seen in
control rats, respectively. A similar result with the caspase-8
protein level was seen in tumors treated with trans,trans-CLA
(Figure 5). The degradation of PARP, an intrinsic substrate for
caspase-3 that is normally cleaved during apoptosis, in tumors
from rats treated with the trans,trans-CLA isomer was evident
(Figure 5). A 85 kDa product was found in these tumors, together
with the concomitant loss of the 116 kDa full-length PARP. The
relative breakdown ratio of the 85 kDa product to 116 kDa
was 0.2 in tumors from control rats but was elevated to 0.7 in
tumors from rats treated with trans,trans-CLA, 1.6 in tumors
from rats treated with cis-9,trans-11-CLA, and 1.7 in tumorsfrom
rats treated with trans-10,cis-12-CLA. Overall, the results indi-
cated that trans,trans-CLA-induced apoptosis in MNU-induced
mammary tumorigenesis was associated with an increased level of
p53 and Bax proteins, reduced Bcl-2 protein level, and activation
of caspase-8 and caspase-3, thus leading to PARP cleavage.
Reduction of cPLA2, COX-2, and PPARγ Expressions and
Composition of Major Fatty Acids in Tumors from SD Rats by
trans,trans-CLA. Arachidonic acid (AA) metabolites are closely
related to carcinogenesis in animal models (23). Appropriately,
we measured the expression of the cPLA2, COX-2, and PPARγ
genes, which are related to AA metabolism, in MNU-induced rat
mammary tumors treated with trans,trans-CLA (Figure 5). The
expression of cPLA2 protein was significantly (p < 0.05) lowered
in rat mammary tumors treated with trans,trans-CLA, relative to
that of control, cis-9,trans-11-CLA, and trans-10,cis-12-CLA treat-
ments. Similar results were seen for COX-2 protein expression.
The reduction efficacy of trans,trans-CLA on PPARγ pro-
tein expression was superior to that of cis-9,trans-11-CLA and
trans-10,cis-12-CLA but was not significant. LA treatment was
not effective on the reduction of cPLA2, COX-2, and PPARγ
protein expressions.

5676 J. Agric. Food Chem., Vol. 58, No. 9, 2010
Islam et al.
Table 2. Composition of Major Fatty Acids in the Phospholipid Fractions of Mammary Tumors from SD Rats Treated with MNU
treatmenta
fatty acidb
control
1% LA
1% cis-9,trans-11-CLA
1% trans-10,cis-12-CLA
1% trans,trans-CLA
palmitic (C16:0)
stearic (C18:0)
oleic (C18:1)
24.6c b
14.8 a
32.1 b
13.8 b
3.7 a
11.0 b
d
24.4 b
12.1 b
34.0 a
17.3 a
2.7 a
8.6 c
d
28.7 a
11.2 b
34.5 a
14.6 b
1.4 a
8.2 c
1.1 a
0.4 b
d
30.4 a
15.2 a
26.2 c
9.3 c
2.7 a
13.6 b
0.5 b
2.1 a
d
28.0 a
15.4 a
27.7 c
9.4 c
1.6 b
16.3 a
d
linoleic (C18:2)
R-linolenic (C18:3)
arachidonic (C20:4)
cis-9,trans-11-CLA
trans-10,cis-12-CLA
trans,trans-CLA
d
d
d
1.6
d
d
a Rats were injected ip with a single dose of MNU (50 mg/kg of body weight) and, 1 day later, subjected to one of the following diets for 16 weeks: control, 1% LA, 1% cis-9,trans-
11-CLA, 1% trans-10,cis-12-CLA, and 1% trans,trans-CLA. b The composition of fatty acids was calculated by the ratio of the chromatogram peak area of a given fatty acid to total
peak areas of interesting fatty acids obtained from GC. c Mean ( SD (n = 3). SD is less than 5% of the mean value. Means with different letters in the same row are significantly
different at p < 0.05 by Duncan’s multiple range test. d Not detectable.
Table 2 shows major fatty acid composition in the membrane
phospholipid fraction from tumors of rats treated with trans,
trans-CLA. trans,trans-CLA treatment significantly (p < 0.05)
increased the composition of AA in membrane phospholipids, as
compared to control and other treatments. Meanwhile, trans-10,
cis-12-CLA also significantly (p < 0.05) increased the composi-
tion of AA, but cis-9,trans-11-CLA and LA treatments rather
decreased relative to the control treatment. The increasing
efficacy of trans,trans-CLA on AA composition was signifi-
cantly (p < 0.05) higher than that of other CLA isomers. The
composition of palmitic acid by trans,trans-CLA, cis-9,trans-1
1-CLA, and trans-10,cis-12-CLA treatments was significantly
(p < 0.05) elevated, as compared to that of control and LA
treatments, but no significance was observed among that of
CLA isomer treatments.
mammary tumorigenesis (8) and various types of cancer cell
lines (9-12), whereas LA acts as a promoter (22). Likewise, in this
study, the potency of anticarcinogenic activities were directly
related to the configuration at the double bonds of CLA isomers;
trans,trans-CLA is a straight chain that resembles stearic acid
(C18:0) and shows the strongest anticarcinogenic activity,
whereas cis-9,trans-11 CLA resembles LA and shows the least
activity. This hypothesis of the structural configuration of trans,
trans-CLA may also suggest that the anticarcinogenic action of
the trans,trans-CLA in MNU-induced tumorigenesis occurs
through the modulation of the expressions of apoptosis-related
genes.
The molecular mechanism by which trans,trans-CLA inhibits
MNU-induced tumorigenesis in rats is not yet well-understood.
However, the inhibition by trans,trans-CLA can be explained, in
part, by the alteration in the caspase-dependent apoptosis caused
by Bax, Bcl-2, caspase-9, and caspase-3 (Figures 4 and 5). These
results are in agreement with the results shown in some human
cancer cell lines (24). In addition, trans,trans-CLA induced the
expression of the caspase-8 protein (Figure 5), which indicates
that caspase-8 is involved in the induction of apoptosis in tumor
cells from rats treated with trans,trans-CLA through the cleavage
of cytosolic, cytoskeletal, and nuclear proteins, as well as DNA (25),
and the release of cytochrome c from mitochondria (26). Such
extrinsic pathway events at an early stage should be further
examined at time intervals.
Although upregulation of p53, reciprocal regulation of Bax
and Bcl-2, and activation of caspase-8 and caspase-3 might be
sufficient to shift the balance toward apoptosis in MNU-induced
rat mammary cacrcinogenesis, AA and its metabolites are also
associated with apoptosis. trans,trans-CLA treatment increased
the AA composition in the membrane phospholipids (Table 2)
and suppressed the expression of cPLA2 and COX-2 proteins
(Figure 5), relative to those of trans-10,cis-12-CLA and cis-9,
trans-11-CLA. These results may indicate the reduced biosynth-
esis of prostaglandin E2 (PGE2) from AA, which is reduced in
concentration in cytosol, via the COX-2 and prostaglandin H
synthase pathways, because the cis-9,trans-11-CLA and trans-10,
cis-12-CLA isomers inhibit the formation of AA and PGE2 in the
mucosal cells of 1,2-dimethylhydrazine-induced rats (8) and in
cancer cells (27). These results of increased AA content in the
membrane and lowered cPLA2 and COX2 by trans,trans-CLA,
which might reduce the AA concentration and PGE2 in cytosol,
are in agreement with results that the positive association of
AA with apoptosis was proven only for free AA in cytosol or
exogenously added AA in the culture media by Monjazeb
et al. (23). There is growing evidence that AA and PGE2 exert
DISCUSSION
The present study clearly revealed that trans,trans-CLA in-
hibits MNU-induced mammary carcinogenesis in rats. The
inhibitory efficacy of trans,trans-CLA was superior to that of
the equally effective cis-9,trans-11-CLA and trans-10,cis-12-CLA
and could be, in part, associated with the induction of apoptosis
of tumor cells through antitumor promoter action. No direct
comparison data for the anticarcinogenicity of trans,trans-CLA
with cis-9,trans-11-CLA and trans-10,cis-12-CLA on chemical-
induced mammary carcinogenesis in rats is available in the
published literature, but this finding is in agreement with the
inhibitory efficacy of trans,trans-CLA and trans-9,trans-11-CLA
on human cancer cell lines investigated thus far. trans,trans-CLA
displays a stronger anti-proliferative activity in both MG-63
osteosarcoma cells (10) and MCF-7 breast cancer cells (11) by
inducing apoptosis, relative to that of cis-9,trans-11-CLA and
trans-10,cis-12-CLA, which exhibit a similar inhibitory effect.
trans-9,trans-11-CLA also has been demonstrated to exhibit an
even higher efficiency than both cis-9,trans-11-CLA and trans-10,
cis-12-CLA in inducing apoptosis in human colon cancer
cells (12). The collective data strongly suggests that the minor
trans,trans-CLA isomer should be considered, because it could
exert even more powerful antitumor effects than those observed
with other CLA isomers.
The anticarcinogenic action of trans,trans-CLA isomers is not
well-defined, but it may be related to its chemical structure,
because the anticarcinogenic efficacy of trans,trans-CLA is
superior to trans-10,cis-12-CLA and cis-9,trans-11-CLA in var-
ious cancer cell lines (10, 11). Furthermore, it is evident that
the anticarcinogenic action of trans-10,cis-12-CLA is even
greater than that of cis-9,trans-11-CLA in chemical-induced rat

Article
J. Agric. Food Chem., Vol. 58, No. 9, 2010 5677
anti-apoptotic effects in various cancer cells and chemical-
induced carcinogenesis in animal modes (28), confirming the data
obtained in the present study. Because cPLA2 is cleaved by
caspase-3 and/or a related caspase in HeLa cells undergoing
apoptosis (29), the activated caspase-3 in the tumors treated with
trans,trans-CLA might be involved in the cleavage of cPLA2,
leading to a lower AA content and COX-2 protein level in trans,
trans-CLA-treated tumors. These suggest that trans,trans-CLA
induced apoptosis of tumor cells through the reduction of cPLA2
and COX-2 activities, which act as tumor promoters.
The activation of PPARγ by ligands leads to either inhibition
of cell proliferation or induction of apoptosis in human breast
tumors and other cancers and human breast cancer cell lines as
well (30). Of the PPARγ ligands, CLA isomers display high
affinity for and are an activator of PPARγ (31). Hence, the
anticarcinogenic activity of CLA is partially mediated by PPARγ
activation in susceptible tumors. However, in the present study,
PPARγ expression was significantly (p < 0.05) reduced by trans,
trans-CLA and other CLA isomers, relative to that of control and
LA treatment, with no significant difference between CLA
isomers (Figure 5). This inverse relation of PPARγ data shown
in the present study and in the literature might be partly due to the
fact that endogenous ligands for PPARγ are limited in the tumors
treated with trans,trans-CLA, because COX-2 products, such as
15-deoxy-D12,14-PGJ2, are probably the most potent endogenous
PPARγ ligands (32). The suppressed PPARγ and COX-2 pro-
teins by trans,trans-CLA might also be involved in the molecular
mechanism of anticarcinogenicity of trans,trans-CLA, including
cis-9,trans-11-CLA and trans-10,cis-12-CLA isomers. Further
studies should clarify this issue.
Because trans,trans-CLA acts as an anti-promoter in the
present study, it suppressed tumor promotion by reducing cPLA2
and COX-2 (Figure 5C), which act as tumor promoters, and by
probably blocking endogenous hormones from acting as tumor
promoters. It is also another mechanism that trans,trans-CLA
either prevented tumor promotion from inhibiting gap junction
function or increased gap junction function in tumor cells (33,34).
Further studies should be performed to clarify such events.
In conclusion, dietary trans,trans-CLA inhibits mammary
carcinogenesis in MNU-induced SD rats by inducing apoptosis
through the reciprocal expressions of Bcl-2 and Bax and the
caspase pathway and by reducing eicosanoids through the cPLA2
pathway. The inhibitory action of trans,trans-CLA is superior to
those ofthe cis-9,trans-11-CLA and trans-10,cis-12-CLA isomers.
The precise mechanistic action and signaling event involved in
trans,trans-CLA-induced apoptosis in mammary tumors remains
to be determined. We strongly suggest that the trans,trans-CLA
isomer should be considered in tumor therapy because of its
superior antitumor effects, as compared to the cis-9,trans-
11-CLA and trans-10,cis-12-CLA isomers.
(4) Ha, Y. L.; Storkson, J. M.; Pariza, M. W. Inhibition of benzo[a]-
pyrene induced mouse forestomach neoplasia by conjugated dienoic
derivatives of linoleic acid. Cancer Res. 1990, 50, 1097–1101.
(5) Miglietta, A.; Bozzo, F.; Bocca, C.; Gabriel, L.; Trombetta, A.;
Belotti, S.; Canuto, R. A. Conjugated linoleic acid induces apoptosis
in MDA-MB-231 breast cancer cells through ERK/MAPK signal-
ling and mitochondrial pathway. Cancer Lett. 2006, 234, 149–157.
(6) Hubbard, N. E.; Lim, D.; Erickson, K. L. Effect of separate
conjugated linoleic acid isomers on murine mammary tumorigenesis.
Cancer Lett. 2003, 190, 13–19.
(7) Park, H. S.; Chun, C. S.; Kim, S.; Ha, Y. L.; Park, J. H. Dietary
trans-10,cis-12 and cis-9,trans-11 conjugated linoleic acids induce
apoptosis in the colonic mucosa of rats treated with 1,2-dimethyl-
hydrazine. J. Med. Food 2006, 9, 22–27.
(8) Meng, X.; Shoemaker, S. F.; McGee, S. O.; Ip, M. M. t10,c12-
Conjugated linoleic acid stimulates mammary tumor progression in
Her2/ErbB2 mice through activation of both proliferative and
survival pathways. Carcinogenesis 2008, 29, 1013–1021.
(9) Cho, H. J.; Kim, W. K.; Jung, J. I.; Kim, E. J.; Lim, S. S.; Kwon,
D. Y.; Park, J. H. Y. trans-10,cis-12, not cis-9,trans-11, conjugated
linoleic acid decreases ErbB3 expression in HT-29 human colon
cancer cells. World J. Gastroenterol. 2005, 11, 5142–5150.
(10) Kim, Y. S.; Cerbo, R. M.; Hah, C. K.; Bahn, K. N.; Kim, J. O.; Ha,
Y. L. Growth inhibition of osteosarcoma cell MG-63 by a mixture of
trans,trans conjugated linoleic acid isomers: Possible mechanistic
actions. J. Food Sci. 2008, 73, 247–255.
(11) Islam, M. A.; Kim, Y. S.; Jang, W. J.; Lee, S. M.; Kim, H. G.; Kim,
S. Y.; Kim, J. O.; Ha, Y. L. A mixture of trans,trans conjugated
linoleic acid induces apoptosis in MCF-7 human breast cancer cells
with reciprocal expression of Bax and Bcl-2. J. Agric. Food Chem.
2008, 56, 5970–5976.
(12) Beppu, F.; Hosokawa, M.; Tanaka, L.; Kohno, H.; Tanaka, T.;
Miyashita, K. Potent inhibitory effect of trans-9,trans-11 isomer of
conjugated linoleic acid on the growth of human colon cancer cells.
J. Nutr. Biochem. 2006, 17, 830–836.
(13) Yasui, Y.; Suzuki, R.; Kohno, H.; Miyamoto, S.; Beppu, F.;
Hosokawa, M.; Miyashita, K.; Tanaka, T. 9-trans,11-trans con-
jugated linoleic acid inhibits the development of azoxymethane-
induced colonic aberrant crypt foci in rats. Nutr. Cancer 2007, 59,
82–91.
(14) Park, S. J.; Park, C. W.; Kim, S. J.; Kim, J. K.; Kim, Y. R.;
Park, K. A.; Kim, J. O.; Ha, Y. L. Methylation methods for
the quantitative analysis of conjugated linoleic acid (CLA) isomers
in various lipid samples. J. Agric. Food Chem. 2002, 50, 989–996.
(15) Kim, S. J.; Park, K. A.; Park, J. H. Y.; Kim, J. O.; Ha, Y. L.
Preparation of cis-9,trans-11 and trans-10,cis-12 conjugated linoleic
acid (CLA) isomers from synthetic CLA. J. Food Sci. Nutr. 2000, 5,
70–74.
(16) Wong, P.; Coulombe, P. A. Loss of keratin 6 (K6) proteins reveals a
function for intermediate filaments during wound repair. J. Cell Biol.
2003, 163, 327–337.
(17) Hartman, W. H.; Uzello, L.; Sobin, L. H.; Stalsberg, H. Histological
Typing of Breast Tumors, 2nd ed.; World Health Organization (WHO):
Geneva, Switzerland, 1981; pp 15-25.
(18) Negoescu, A.; Lorimier, P.; Labat-Moleur, F.; Drouet, C.;
Robert, C.; Guillermet, C.; Brambilla, C.; Brambilla, E. In situ
apoptotic cell labeling by the tunnel method: Improvement and
evaluation on cell preparations. J. Histochem. Cytochem. 1996, 44,
959–968.
LITERATURE CITED
(1) Sehat, N; Kramer, J. K. G.; Mossoba, M. M.; Yurawecz, M. P.;
Roach, J. A. G.; Eulitz, K.; Morehouse, K. M.; Ku, Y. Identification
of conjugated linoleic acid isomers in cheese by gas chromatography,
silver ion high performance liquid chromatography and mass
spectral reconstructed ion profiles-Comparison of chromatographic
elution sequences. Lipids 1998, 33, 963–971.
(19) Hamilton, J. G.; Comai, K. Rapid separation of neutral lipids, free
fatty acids and polar lipids using prepacked silica sep-pak columns.
Lipids 1999, 23, 1146–1149.
(20) Yasuda, M.; Nishizawa, T.; Ohigashi, H.; Tanaka, T.; Hou, D.;
Colburn, N. H.; Murakami, A. Linoleic acid metabolite suppresses
skin inflammation and tumor promotion in mice: Possible roles of
programmed cell death 4 induction. Carcinogenesis 2009, 30, 1209–
1216.
(21) Fischer, S. M.; Leyton, J.; Lee, M. L.; Locniskar, M.; Belury, M. A.;
Maldve, R. E.; Slaga, T. J.; Bechtel, D. H. Differential effects of
dietary linoleic acid on mouse skin-tumor promotion and mammary
carcinogenesis. Cancer Res. 1992, 52, 2049s–2054s.
(2) Kim, S. J.; Park, S. J.; Kim, J. K.; Kim, J. H.; Shim, K. H.; Park, C. G.;
Kim, J. O.; Ha, Y. L. Preparation of a t,t conjugated linoleic acid
methylester (CLA-Me) isomers mixture from synthetic CLA by methyl-
ation with BF3/methanol. J. Agric. Food Chem. 2003, 51, 3208–3214.
(3) Serini, S.; Piccioni, E.; Merendino, N.; Calviello, G. Dietary poly-
unsaturated fatty acids as inducers of apoptosis: Implication for
cancer. Apoptosis 2009, 14, 135–152.

5678 J. Agric. Food Chem., Vol. 58, No. 9, 2010
Islam et al.
€
(22) Tisdale, M. J. Biology of cachexia. J. Natl. Cancer Inst. 1997, 89,
1763–1773.
(23) Monjazeb, A. M.; High, K. P.; Connoy, A.; Hart, L. S.; Koumenis,
C.; Chilton, F. H. Arachidonic acid-induced gene expression in colon
cancer cells. Carcinogenesis 2006, 27, 1950–1960.
(24) Majumder, B.; Wahle, K. W.; Moir, S.; Schofield, A.; Choe, S. N.;
Farquharson, A.; Grant, I.; Heys, S. D. Conjugated linoleic acids
(CLAs) regulate the expression of key apoptotic genes in human
breast cancer cells. FASEB J. 2002, 16, 1447–1449.
(30) Elstner, E.; Muller, C.; Koshizuka, K.; Williamson, E. A.; Park, D.;
Asou, H.; Shintaku, P.; Said, J. W.; Heber, D.; Koeffler, H. P.
Ligands for peroxisome proliferator-activated receptor γ and reti-
noic acid receptor inhibit growth and induce apoptosis of human
breast cancer cells in vitro and in BNX mice. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 8806–8811.
(31) McCarty, M. F. Activation of PPARγ may mediate a portion of the
anticancer activity of conjugated linoleic acid. Med. Hypotheses
2000, 55, 187–188.
(25) Kuwana, T.; Smith, J. J.; Muzio, M.; Dixit, V.; Newmeyer, D. D.;
Kornbluth, S. Apoptosis induction by caspase-8 is amplified through
the mitochondrial release of cytochrome c. J. Biol. Chem. 1998, 273,
16589–16594.
(26) Li, H.; Zhu, H.; Xu, C. J.; Yuan, J. Cleavage of Bid by caspase 8
mediates the mitochondrial damage in the Fas pathway of apoptosis.
Cell 1998, 94, 491–501.
(27) Liu, K. L.; Belury, M. A. Conjugated linoleic acid reduces arachi-
donic acid content and PGE2 synthesis in murine keratinocytes.
Cancer Lett. 1998, 127, 15–22.
(32) Schwab, M.; Reynders, V.; Loitsch, S.; Shastri, Y. M.; Steinhilber,
€
D.; Schroder, O.; Stein, J. PPARγ is involved in mesalazine-
mediated induction of apoptosis and inhibition of cell growth in
colon cancer cells. Carcinogenesis 2008, 29, 1407–1414.
(33) Trosko, J. E.; Ruch, R. J. Cell-cell communication in carcinogen-
esis. Front. Biosci. 1998, 3, 208–236.
(34) Trosko, J. E.; Ruch, R. J. Gap junctions as targets for cancer chemo-
prevention and chemotherapy. Curr. Drug Targets 2002, 3, 465–
482.
(28) Papadimitriou, A.; King, A. J. F.; Jones, P. M.; Persaud, S. J. Anti-
apoptotic effects of arachidonic acid and prostaglandin E2 in
pancreatic β-cells. Cell. Physiol. Biochem. 2007, 20, 607–616.
Received for review January 11, 2010. Revised manuscript received
March 15, 2010. Accepted March 25, 2010. This work was partly
supported by a scholarship from the BK21 Program and a National
Research Foundation of Korea (NRF) Grant funded by the Korean
Government (2009-0075483).
€
(29) Adam-Klages, S.; Schwandner, R.; Luschen, S.; Ussat, S.; Kreder,
€
D.; Kronke, M. Caspase-mediated inhibition of human cytosolic
phospholipase A2 during apoptosis. J. Immunol. 1998, 161, 5687–5694.