European Journal of Clinical Nutrition (2006) 60, 386–392
& 2006 Nature Publishing Group All rights reserved 0954-3007/06 $30.00
www.nature.com/ejcn
ORIGINAL ARTICLE
Effects of docosahexaenoic acid supplementation on
blood lipids, estrogen metabolism, and in vivo
oxidative stress in postmenopausal vegetarian
women
WH Wu1, SC Lu2, TF Wang1, HJ Jou3 and TA Wang4
1
Graduate Program of Nutrition, Department of Human Development and Family Studies, National Taiwan Normal University, Taipei,
Taiwan; 2Department of Biochemistry and Molecular Biology, College of Medicine, National Taiwan University, Taipei, Taiwan;
3
Department of Obstetrics and Gynecology, Taiwan Adventist Hospital, Taipei, Taiwan and 4Department of Obstetrics and Gynecology,
Taipei City Hospital Yangming Branch, Taipei, Taiwan
Background: Vegetarians are generally deficient in long-chain n-3 fatty acids. Long-chain n-3 fatty acids have a beneficial effect
on plasma lipid levels, and some studies showed that they had breast cancer suppression effect. One of the biomarkers of breast
cancer risk is the ratio of urinary 2-hydroxyestrone (2-OHE1) to 16a-hydroxyestrone (16a-OHE1).
Objective: To investigate the effect of docosahexaenoic acid (DHA, 22:6n-3) supplementation on blood lipids, estrogen
metabolism and oxidative stress in vegetarians.
Design: Single-blind, randomized, placebo-controlled trial.
Interventions: Twenty-seven postmenopausal vegetarian women were recruited. After a 2-week run-in period with 6 g placebo
corn oil, the subjects were subsequently randomized to receive either 6 g corn oil (n ¼ 13) or 6 g DHA-rich algae oil (2.14 g of
DHA/day) (n ¼ 14) for 6 weeks. Two subjects in corn oil group withdrew before completion.
Main outcome measures: Plasma lipids, urinary 2-OHE1 and 16a-OHE1, urinary F2-isoprostanes and plasma a-tocopherol.
Results: Plasma LDL-DHA and EPA level increased significantly by DHA supplementation. DHA decreased plasma cholesterol (C)
levels (P ¼ 0.04), but did not influence the levels of plasma TG, LDL-C and HDL-C, a-tocopherol, urinary F2-isoprostanes,
2-OHE1, 16a-OHE1 and ratio of 2-OHE1 to 16a-OHE1 as compared to corn oil.
Conclusion: DHA supplementation at a dose of 2.14 g/day for 42 days decreases plasma cholesterol but neither does it show
beneficial effects on estrogen metabolism, nor does it induce deleterious effects on the observed in vivo antioxidant or oxidative
stress marker in postmenopausal vegetarian women.
Sponsorship: A grant (# DOH89-TD-1062) from Department of Health, Executive Yuan, Taiwan.
European Journal of Clinical Nutrition (2006) 60, 386–392. doi:10.1038/sj.ejcn.1602328; published online 9 November 2005
Keywords: DHA; algae oil; hydroxyestrone; blood lipids; oxidative stress; postmenopausal vegetarian women
Correspondence: Professor WH Wu, Graduate Program of Nutrition,
Department of Human Development and Family Studies, National Taiwan
Normal University, Taipei 106, Taiwan.
E-mail: [email protected]
Guarantor: WH Wu.
Contributors: WHW designed the study, contributed to method development,
data interpretation and preparation of the manuscript. SCL initiated the study,
obtained a research grant for the study and was responsible for all stages of
the study. TFW was responsible for biochemical and statistical analyses and
taking care of the study subjects. HJJ and TAW were the gynecologists involved
in subject recruitment, sample processing and advised on protocol design. All
authors contributed to the writing of the manuscript.
Received 27 October 2004; revised 16 August 2005; accepted 14 September
2005; published online 9 November 2005
Introduction
Vegetarians have a lower status of long-chain n-3 fatty acid
(Lee et al., 2000), principally eicosapentaenoic acid (EPA,
20:5n-3) and docosahexaenoic acid (DHA, 22:6n-3), because
of the exclusion of fish from their diet. In addition, the high
intake of linoleic acid (18:2n-6) from vegetable oils (Lu et al.,
2000b) competes with the conversion of a-linoleic acid
(18:3n-3) to EPA and DHA. Long-chain n-3 fatty acids have
been found to have beneficial effects on the regulation of
plasma lipid levels and cardiovascular function (Mori and
DHA and postmenopausal vegetarians
WH Wu et al
387
Beilin, 2001). Higher levels of platelet aggregation were
found in vegetarians when compared to their omnivorous
counterparts (Li et al., 1999), and long-chain n-3 fatty acid
supplementation has been shown to reduce platelet aggregation (Mezzano et al., 2000). DHA supplementation also
induced a moderate reduction in the total or LDL-cholesterol:HDL-cholesterol ratio and a slight reduction in triglyceride concentration in vegetarians (Conquer and Holub, 1996).
These observations suggest the deficit in long-chain n-3 fatty
acid may cause adverse effect in vegetarians.
Long-chain n-3 fatty acids have been consistently shown
to inhibit the proliferation of breast cancer cells in vitro
(Bernard-Gallon et al., 2002) and in animal models (Cave,
1997). However, results of epidemiological studies examining the association of breast cancer risk and fish or marine
n-3 long-chain fatty acids consumption were not entirely
consistent (Terry et al., 2003). One of the biomarkers of
breast cancer risk is the ratio of urinary 2-hydroxyestrone
(2-OHE1) to 16a-hydroxyestrone (16a-OHE1). The two major
estrogen metabolites are 2-OHE1 and 16a-OHE1; the former is
not estrogenic, whereas the latter is estrogenic and genotoxic
(Service, 1998). The proposal of using the ratio of urinary
2-OHE1 to 16a-OHE1 as a breast cancer risk marker is
suggested by some (Kabat et al., 1997; Muti et al., 2000) but
not all studies (Ursin et al., 2001). Although estrogen is a
major risk factor for breast cancer, its level is hardly altered
by diet intervention. On the other hand, the pathways of
estrogen metabolism can be manipulated by several dietary
factors (Michnovicz et al., 1997; Brooks et al. 2004) and
perhaps could be better used for fine-tuning breast cancer
risk within an ethnic group. The oxidative metabolism of
estrogen is catalyzed predominantly by hepatic cytochrome
P450 (CYP450) (Huang et al., 1998; Yamazaki et al., 1998).
Dietary fish oil increased the concentration of hepatic
overall CYP450 and some subtypes of CYP450 in rats (Valdes
et al., 1995; Chen et al., 2003). A preliminary study of
Osborne et al. (1998) indicated a decreased extent of 16ahydroxylation of estradiol in women after supplemented
with n-3 fatty acids. Vegetarians have been found to have
significantly lower risk of some cancers (Fraser, 1999) but not
breast cancer (Fraser, 1999; Dos Santos Silva et al., 2002).
Therefore, in this study, an attempt was made to determine if
estrogen metabolism was modulated to a less carcinogenic
pathway after the improvement of long-chain n-3 fatty acid
status by algae oil in postmenopausal vegetarian women.
Higher number of double bonds present in n-3 long-chain
fatty acids has been postulated to enhance lipid peroxidation. Nevertheless, the influence of fish oil supplementation
on some indicators of oxidative stress was not consistent
(Meydani et al., 1991; Higdon et al., 2000), partly because n-3
fatty acids also enhanced the activities and expression of
antioxidant enzymes, such as catalase, glutathione peroxidase and superoxide dismutase (Takahashi et al., 2002; Wang
et al., 2004). Therefore, to examine whether DHA supplementation actually induces oxidative stress in vivo in
vegetarians, levels of urinary F2-isoprostanes and plasma
a-tocopherol were measured. In addition, LDL oxidation
in vitro was also assessed in this study.
Materials and methods
Supplements
DHA-rich oil extracted from algae was used because fish oil
was not acceptable to vegetarians. The DHA oil was kindly
provided by Westar Nutrition Corp, CA, USA. It was free of
EPA and with the addition of 1 IU dl-a-tocopheryl acetate per
gram. Corn oil was used as a placebo oil because it was
devoid of n-3 fatty acids and was one of the popular cooking
oils used by the subjects recruited. Antioxidants were not
added to the corn oil and a- and g-tocopherol contents were
natural amounts found in corn oil. Fatty acid composition
analyzed in our laboratory and tocopherol contents of DHA
oil and placebo corn oil based on manufacturer’s specification are shown in Table 1.
Subjects
A total of 27 postmenopausal vegans and lacto-ovovegetarians were recruited from the local vegetarian societies and
religious groups. The inclusion criteria were as follows: (1)
having followed the vegetarian diet at least for 1 year; (2)
amenorrheic at least for 1 year; (3) age under 60 years; (4)
body mass index in the range of 18–26 kg/m2; (5) no history
of cardiovascular, metabolic or endocrinologic disease, as
evidenced by blood chemistry screening and personal interview; (6) no use of HRT at least for 6 months. The
characteristics of the subjects studied are described in Table 2.
None of them were nuns. All subjects received single-blinded
corn oil placebo during a 2-week run-in period and were
subsequently randomized by random digit table to receive
either 6 g placebo or 6 g DHA-rich algae oil (2.14 g of DHA/
day) per day for another 6 weeks. The amount of n-3 fatty
acids ingested per day was approximately equal to that
Table 1
Fatty acid composition and tocopherol contents of oils
Fatty acid (wt %)
14:0
16:0
18:1
18:2n-6
22:6n-3
a-TEb (mg/g)
a-TE (mg/g PUFA)
a-TE (mg/double bond)
Corn oil
DHA oil
—a
10.0
30.0
60.0
—
0.38c
1.27
62
23.5
19.2
21.7
—
35.6
0.45d
0.63
71
a
Trace.
a-TE, Tocopherol equivalent; 1 mg a-TE ¼ 1 mg a-tocopherol ¼ 10 mg gtocopherol ¼ 2.2 IU dl-a-tocopheryl acetate.
c
Calculated from 1 IU dl-a-tocopheryl acetate/g of DHA oil.
d
Calculated from (0.28 mg a-tocopherol þ 1 mg g-tocopherol)/g of corn oil.
b
European Journal of Clinical Nutrition
DHA and postmenopausal vegetarians
WH Wu et al
388
Table 2 Basic characteristics of the subjects
n
Age (year)
Height (cm)
BMI (kg/m2)
Years vegetarian
Years since menopause
Corn oil
DHA oil
11
52.375.1
154.775.3
22.973.2
12.879.0
4.474.4
14
52.674.4
153.475.2
23.373.0
17.4710.3
5.073.9
Values are means7s.d.
BMI, body mass index.
contained in 80 g of salmon (16 g fat/100 g salmon, 16 g n-3
fatty acids/100 g total fatty acids). The odor and taste of algae
oil were very mild. The subjects had never known or eaten
algae oil before entering this trial, and thus were not
expected to distinguish corn oil from algae oil. The subjects
were advised to maintain their usual life styles, levels of
physical activity, diet habits, and to keep their body weights
unvaried. They met technicians to pick up the oils and had
body weight measured each week. They were instructed to
keep the oil in the refrigerator. Fasting blood and first
morning urine samples were collected on the first morning
of intervention and the morning just following the end of
intervention. Written informed consent was obtained from
each participant before inclusion in the study. The protocol
was approved by the Human Experimentation Committee of
Taiwan Adventist Hospital, Taipei, Taiwan. Two subjects
assigned to control group withdrew from the study before
completion: one for family problem and the other for an
unreported reason.
Collection and analysis of blood and urine samples
After a 12-h fast, blood was collected from subjects into tubes
containing EDTA (2.8 mg/ml of blood). Plasma was separated
from whole blood by centrifugation at 2000 g for 15 min and
stored at 701C until the end of the study where aliquots
from each subjects were analyzed at the same run. Lowdensity lipoprotein (LDL) and high-density lipoprotein
(HDL) were isolated from plasma by sequential ultracentrifugation in NaBr density solution containing 10 mmol EDTA/
l at densities of 1.019–1.063, and 1.063–1.210, respectively.
The isolated LDL were dialyzed overnight and then diluted
to 200 mg protein/l. LDL oxidation was initiated by the
addition of CuSO4 to a final concentration of 5 mmol/l for the
measurements of thiobarbituric acid reactive substances
(TBARS) produced after 3 h oxidation as previously described
(Lu et al., 2000b). Fatty acid compositions of corn oil, algae
oil and LDL were analyzed according to the method of
Lepage and Roy (1986). Plasma cholesterol (C), LDL-C, HDLC and plasma triglycerides (TG) were measured by using
enzymatic kits (Randox Lab., Antrim, UK). Plasma atocopherol was determined by HPLC according to the
method described by Kaplan et al. (1987).
European Journal of Clinical Nutrition
Early morning spot urine samples were collected into a
tube containing vitamin C (1 mg/ml of urine) and immediately cooled to 41C. After centrifugation at 3000 g for 15 min,
the resulting clear supernatants were stored at 701C until
the end of the study where aliquots from each subjects were
analyzed at the same time. Urinary creatinine was determined with the help of a commercial kit (Randox Lab.
Antrim, UK) after heating at 1001C for 5 min to destroy
residual vitamin C. Urinary 2-OHE1 and 16a-OHE1 were
measured in triplicate using a competitive solid-phase
enzyme immunoassay (EIA) kit (Immuna Care Corporation,
Bethlehem, PA, USA) (Bradlow et al., 1998). The monoclonal
antibody to 2-OHE1 had a 100% crossreactivity with 2hydroxyestradiol (2-OHE2), and, thus, the results shown are
actually the sum of 2-hydroxyestrogens. Since the level of 2OHE2 is much less than 2-OHE1, the values were predominantly contributed by 2-OHE1. The samples from the same
person were analyzed sequentially in random order in the
same plate and each plate included equal number of samples
from each group to decrease the artificial variations arising
from sample location and plate to plate difference. The
kinetics of the immunoreaction was monitored at 405 nm at
2-min intervals for 20 min, using a MR5000 plate reader
(Dynatech Lab., VI, USA). The within-plate coefficients of
variation were 3.2–10 and 0.9–5.2% and between-plate
coefficients of variation were 7.7 and 3.1% for 2-OHE1 and
16a-OHE1, respectively. The results are expressed as ng/mg
creatinine to account for differences arising from variations
in urine concentration.
Urinary concentration of an F2-isoprostane, 8-iso-prostaglandin F2a was measured by competitive solid-phase EIA kits
(Assay Designs, Ann Arbor, MI, USA) and the results are
expressed as ng/mg creatinine.
Statistical analysis
Results were expressed in terms of unadjusted means and
standard deviations. The normality of data was checked by
Kolmogorov–Smirnov test. Normally distributed data, either
original or after transformation, were tested by analysis of
covariance (ANCOVA) with adjustment for baseline values.
Because the unstandardized residuals of log-transformed
data for LDL-EPA were not normally distributed, data of
LDL-EPA were analyzed by nonparametric test (Mann–
Whitney test) to compare changes from baseline between
two groups. Results were considered statistically significant
at Po0.05. All statistical analyses were conducted by using
SPSS 11.5.
Results
Compliance of the subjects was monitored by means of a
self-reported daily consumption sheet and confirmed by the
changes in LDL fatty acid composition following DHA
supplementation. DHA oil supplementation significantly
DHA and postmenopausal vegetarians
WH Wu et al
389
Discussion
increased the mean levels of LDL-DHA and eicosapentaenoic
acid (EPA) (Table 3), but did not change that of arachidonic
acid (AA) (Table 3). None of the LDL-fatty acid levels
changed significantly after placebo corn oil supplementation
(Table 3).
Plasma cholesterol level had a small but significant
decrease after DHA supplementation as compared to corn
oil supplementation. Levels of plasma TG, LDL-C, HDL-C,
urinary 2-OHE1, 16a-OHE1 and the ratio of 2-OHE1 to 16aOHE1 did not show a significant difference between two
groups after oil supplementations (Table 4). a-Tocopherol
status expressed as plasma levels or levels normalized by the
sum of plasma TG and cholesterol, and urinary F2-isoprostane did not have significant difference between two groups
after interventions (Table 4). TBARS production in in vitro
oxidized LDL increased significantly after DHA supplementation (Table 4).
Fatty acid status
The n-3 fatty acids supplemented in our study was DHA-rich
algae oil, which was devoid of EPA. Although a small portion
of DHA could be retroconverted to EPA (Nelson et al., 1997),
the conversion is limited. Our study showed that DHA
supplementation increased LDL-DHA by 175% and LDL-EPA
by 39% (Table 3). LDL-fatty acid compositions might not
closely reflect fatty acid status as plasma or erythrocyte
phospholipids did (Dougherty et al., 1987). But serum
cholesteryl ester fatty acid composition has also been widely
used as a biomarker for fatty acid intake (Zock et al., 1997). The
fatty acids in LDL are predominantly esterified to cholesterol
and they contribute to the majority of plasma cholesteryl ester
fatty acids. As a consequence, LDL-fatty acid composition was
used as a surrogate index of fatty acid status in this study.
Table 3 Fatty acid composition of LDLa
Corn oil
C14:0
C16:0
C16:1
C18:0
C18:1
C18:2
C20:4
C20:5
C22:6
(n-6)
(n-6)
(n-3)
(n-3)
Pb
DHA oil
Before
After
Change
Before
After
Change
0.5070.18
17.3571.59
1.8270.71
6.5571.42
14.7072.22
49.8373.51
5.2672.41
1.5771.60
1.6170.34
0.4870.16
18.2073.39
1.5770.44
7.4571.21
15.6572.33
49.6072.54
3.6572.18
1.0070.31
1.7170.64
0.0270.22
0.8572.77
0.2570.80
0.8971.46
0.9472.49
0.2373.21
1.6173.34
0.5771.53
0.1070.76
0.5770.28
19.4772.87
1.9370.55
6.8071.18
16.2772.79
46.7074.51
5.0072.18
1.2370.45
1.3570.54
0.6070.16
18.5971.77
1.6770.47
6.5770.98
14.8073.13
47.2374.17
4.2871.75
1.7271.03
3.7171.03
0.0370.35
0.8872.10
0.2670.64
0.2371.13
1.4873.47
0.5373.90
0.7272.23
0.4871.05
2.3670.85
0.088
0.423c
0.650
0.034c
0.224
0.522
0.333c
0.040d
0.000
a
Values are unadjusted means7s.d., n ¼ 11 and 14, for corn oil and DHA oil, respectively.
Comparing values after corn oil and DHA oil treatments by analysis of covariance adjusted for baseline value except for C20:5.
c
Data were log transformed before statistical analysis.
d
Comparing changes from baseline between two groups by nonparametric test (Mann–Whitney test).
b
Table 4 Levels of plasma lipids, urinary estrogen metabolites, oxidative stress and antioxidant status of subjectsa
Corn oil
Body wt (kg)
TG (mmol/l)
TC (mmol/l)
LDL-C (mmol/l)
HDL-C (mmol/l)
U-2-OHE1 (ng/mg creatinine)
U-16a-OHE1 (ng/mg creatinine)
U-2/16a-OHE1
a-Tocopherol (mmol/l)
a-Tocopherol (mmol/mol of lipids)
U-isoprostane (ng/mg creatinine)
LDL-TBARS (nmol/mg protein)
Pb
DHA oil
Before
After
Change
Before
After
Change
54.978.8
1.5170.31
4.1570.52
2.4970.32
1.0670.15
6.7873.54
5.4371.94
1.2870.51
25.3674.96
4.4870.68
4.2173.43
48.2577.90
55.178.8
1.5570.61
4.4270.38
2.6070.30
1.1070.26
7.1472.26
6.0372.30
1.2570.32
25.5475.92
4.2570.61
3.4773.00
49.16714.63
0.270.5
0.0470.57
0.2770.49
0.1070.37
0.0470.23
0.3772.99
0.6072.95
0.0470.64
0.1973.50
0.2370.34
0.7474.28
0.91710.19
54.977.8
1.40 7 0.62
4.0970.68
2.3270.53
1.0970.24
5.7373.64
5.0072.49
1.2070.56
23.2575.33
4.3270.91
3.9273.40
49.74715.42
55.077.8
1.1670.46
3.9670.77
2.2370.64
1.1770.27
6.5776.69
5.7873.30
1.0870.53
21.5375.31
4.2170.54
3.8772.88
61.86715.13
0.170.7
0.2570.59
0.1370.46
0.0870.50
0.0770.17
0.8477.38
0.7974.12
0.1270.63
1.7272.78
0.1170.69
0.0573.06
12.12715.91
0.597
0.105
0.040
0.181
0.607
0.300c
0.845
0.403
0.126
0.798
0.764c
0.039
a
Values are unadjusted means7s.d., n ¼ 11 and 14, for corn oil and DHA oil, respectively. U, urinary; 2-OHE1, 2-hydroxyestrone; 16a-OHE1, 16a-hydroxyestrone;
2/16aOHE1, ratio of 2-OHE1 to 16a-OHE1; TBARS, thiobarbituric acid reactive substances.
b
Comparing values after corn oil and DHA oil treatments by analysis of covariance adjusted for baseline value.
c
Data were log transformed before statistical analysis.
European Journal of Clinical Nutrition
DHA and postmenopausal vegetarians
WH Wu et al
390
Estrogen metabolites
Fish oil influenced the activity of hepatic CYP450 (Valdes
et al., 1995; Chen et al., 2003), which might consequently
influence the metabolism of estrogen. The lower status of n3 long-chain fatty acids in vegetarians seems to be a good
model to investigate the influence of DHA supplementation
on estrogen metabolism. However, our study did not find
significant changes in 2- and 16a-OHE1 after DHA or placebo
supplementation (Table 4). In a preliminary study, Osborne
et al. (1988) and Karmali (1989) described that women
supplemented with fish oil (1.53 g EPA þ 1.44 g DHA per day)
decreased the extent of 16a-hydroxylation of estradiol and
women with higher baseline values of 16a-hydroxylation
had a more profound decrease. Most of the women in the
study were already with an increased risk for breast cancer,
whereas those in our study were apparently healthy
postmenopausal vegetarians who were not considered to
have high breast cancer risk. It is probably hard to further
modulate the oxidative metabolism of estrogen in these
subjects. The effect of EPA and DHA on estrogen metabolism
might be different and needs further investigation. The 6week duration of intervention was short in this study;
however, the metabolic pathway of estrogens was found to
be altered in 7 days by indole-3-carbinol (Michnovicz et al.,
1997), 12 days by broccoli diet (Kall et al., 1996), 4 weeks or 6
weeks by soy (Lu et al., 2000a; Nettleton et al., 2005).
Therefore, 6-week intervention in this study might be
adequate. Nevertheless, the small number of subjects and
large standard deviations for the estrogen metabolites could
have limited the power to examine this effect.
Blood lipids
We only observed a small but significant decrease in plasma
cholesterol level (P ¼ 0.040), but not in plasma LDL- and
HDL-C after DHA supplementation (Table 4). Plasma TG level
decreased by 18%, which did not reach statistical significance and was not as pronounced as the findings reported in
other studies (25–30%) (Harris, 1997). The mild effect might
be due to the low daily dose of n-3 fatty acids and original
favorable lipid profiles in these postmenopausal vegetarians.
Lipid peroxides
The influence of n-3 fatty acids on lipid peroxidation in vivo
depends on the balance of oxidative stress and induced
antioxidative enzymes, but the net effect remains contradictory. This is probably due to the different doses of n-3
fatty acids used, different amounts of antioxidants supplied
and different methods used to quantify the oxidative stress
in previous studies. In this DHA oil, the amounts of atocopherol equivalent (a-TE) adjusted by numbers of double
bonds were slightly lower than those in corn oil (Table 1),
but its supplementation neither decreased a-tocopherol
status nor increased urinary F2 isoprostane level (Table 4).
F2 isoprostanes, a promising index of oxidative stress in vivo,
European Journal of Clinical Nutrition
are produced solely by free radical-induced peroxidation of AA
(Morrow and Roberts, 1996). The unchanged AA status after
DHA supplementation in this study was accompanied with an
unchanged urinary F2-isoprostane level (Table 4), which
represented no increase in AA peroxidation. Two research
groups found a decrease in F2-isoprostane with higher doses of
n-3 fatty acids than ours. One of these studies involved
postmenopausal women supplemented with fish oil at a dose
of 3.4 g n-3 fatty acids per day but significant differences of
plasma F2-isoprostane were eliminated when the values were
normalized to plasma AA concentrations (Higdon et al., 2000).
Subjects in the other study were hypertensive type II diabetic
patients (Mori et al., 2003). After having been supplemented
with purified EPA or DHA at a dose of 4 g per day, their AA
status and urinary F2-isoprostane was lower than the olive oil
control group. Therefore, when the level of F2-isoprostane was
used as a criterion, all the studies demonstrated no enhancement of in vivo oxidant stress after n-3 fatty acid supplementation. However, instead of F2-isoprostane, the peroxidation of
EPA and DHA produced F3- and F4-isoprostanes, respectively
(Nourooz-Zadeh et al., 1997, 1998). The influence of DHA
supplementation on the production of F4-isoprostanes needs
further evaluation. Some studies showing increased peroxidation and oxidative stress by n-3 fatty acid supplementation
were suggested by an increase of LDL oxidation in vitro
(Oostenbrug et al., 1994; Suzukawa et al., 1995). However,
when similar assay was used to examine whether the observed
unchanged in vivo oxidative stress status may be corroborated
in vitro, results showed that the level of LDL-TBARS increased
by DHA supplementation (Table 4). This would have been
expected from the 1.75-fold enrichment of DHA in LDL
(Table 4), as LDL-TBARS is produced by in vitro extensively
oxidized LDL. The higher antioxidant status (Lu et al., 2000b;
Manjari et al., 2001) of vegetarians might also have contributed to the resistance against oxidative stress from DHA.
Conclusion
This study shows that DHA supplementation at a dose of
2.14 g/day for 42 days decreases plasma cholesterol, but does
not evidence to change the breast cancer risk marker
measured as the ratio of urinary 2-OHE1 to 16a-OHE1, in
postmenopausal vegetarian women. The in vivo antioxidant
or oxidative stress markers measured as plasma a-tocopherol,
and urinary F2-isoprostane are not deleteriously affected.
However, this study had some limitations such as being a
single-blind study with relative short duration, and small
number of subjects. Therefore, longer period and larger
number of subjects are needed to confirm these observations.
Acknowledgements
This study was supported by a grant (Grant# DOH89-TD1062) from Department of Health, Executive Yuan, Taiwan.
DHA and postmenopausal vegetarians
WH Wu et al
391
We thank reviewers for the suggestion in statistical analysis
and Professor Sieh-Hwa Lin from National Taiwan Normal
University for assistance in statistical analysis.
References
Bernard-Gallon D, Vissac-Sabatier C, Antoine-Vincent D, Rio PG,
Maurizis JC, Fustier P et al. (2002). Differential effects of n–3 and n–
6 polyunsaturated fatty acids on BRCA1 and BRCA2 gene
expression in breast cell lines. Br J Nutr 87, 281–289.
Bradlow HL, Sepkovic DW, Klug T, Osborne MP (1998). Application
of an improved ELISA assay to the analysis of urinary estrogen
metabolites. Steroids 63, 406–413.
Brooks JD, Ward WE, Lewis JE, Hilditch J, Nickell L, Wong E et al
(2004). Supplementation with flaxseed alters estrogen metabolism
in postmenopausal women to a greater extent than does
supplementation with an equal amount of soy. Am J Clin Nutr
79, 318–325.
Cave WT (1997). Omega-3 polyunsaturated fatty acids in rodent
models of breast cancer. Breast Cancer Res Treat 46, 239–246.
Chen HW, Tsai CW, Yang JJ, Liu CT, Kuo WW, Lii CK (2003). The
combined effects of garlic oil and fish oil on the hepatic
antioxidant and drug-metabolizing enzymes of rats. Br J Nutr 89,
189–200.
Conquer JA, Holub BJ (1996). Supplementation with an algae source
of docosahexaenoic acid increases (n-3) fatty acid status and alters
selected risk factors for heart disease in vegetarian subjects. J Nutr
126, 3032–3039.
Dos Santos Silva I, Mangtani P, McCormack V, Bhakta D, Sevak L,
McMichael AJ (2002). Lifelong vegetarianism and risk of breast
cancer: a population-based case–control study among South Asian
migrant women living in England. Int J Cancer 99, 238–244.
Dougherty RM, Galli C, Ferro-Luzzi A, Iacono JM (1987). Lipid and
phospholipid fatty acid composition of plasma, red blood cells,
and platelets and how they are affected by dietary lipids: a study of
normal subjects from Italy, Finland, and the USA. Am J Clin Nutr
45, 443–455.
Fraser GE (1999). Associations between diet and cancer, ischemic
heart disease, and all-cause mortality in non-Hispanic white
California Seventh-day Adventists. Am J Clin Nutr 70, 532–538.
Harris WS (1997). n-3 fatty acids and serum lipoproteins: human
studies. Am J Clin Nutr 65 (Suppl), 1645S–1654S.
Higdon JV, Liu J, Du SH, Morrow JD, Ames BN, Wander RC (2000).
Supplementation of postmenopausal women with fish oil rich in
eicosapentaenoic acid and docosahexaenoic acid is not associated
with greater in vivo lipid peroxidation compared with oils rich in
oleate and linoleate as assessed by plasma malondialdehyde and
F(2)-isoprostanes. Am J Clin Nutr 72, 714–722.
Huang Z, Guengerich FP, Kaminsky LS (1998). 16a-Hydroxylation
of estrone by human cytochrome P4503A4/5. Carcinogenesis 19,
867–872.
Kabat GC, Chang CJ, Sparano JA, Sepkovic DW, Hu XP, Khalil A et al.
(1997). Urinary estrogen metabolites and breast cancer: a case–
control study. Cancer Epidemiol Biomaker Prev 6, 505–509.
Kall MA, Vang O, Clausen J (1996). Effects of dietary broccoli on
human in vivo drug metabolizing enzymes: evaluation of caffeine,
oestrone and chlorzoxazone metabolism. Carcinogenesis 17,
793–799.
Kaplan LA, Miller JA, Stein EA (1987). Simultaneous measurement of
serum retinal, tocopherols, carotenes and carotenoids by high
performance liquid chromatography. J Clin Lab Anal 1, 147–152.
Karmali RA (1989). n-3 fatty acids and cancer. J Intern Med 225 (Suppl
1), 197–200.
Lee HY, Woo J, Chen ZY, Leung SF, Peng XH (2000). Serum fatty acid,
lipid profile and dietary intake of Hong Kong Chinese omnivores
and vegetarians. Eur J Clin Nutr 54, 768–773.
Lepage G, Roy CC (1986). Direct transesterification of all classes of
lipids in a one-step reaction. J Lipid Res 27, 114–120.
Li D, Sinclair A, Mann N, Turner A, Ball M, Kelly F et al. (1999). The
association of diet and thrombotic risk factors in healthy male
vegetarians and meat-eaters. Eur J Clin Nutr 53, 612–619.
Lu LJW, Cree M, Josyula S, Nagamani M, Grady JJ, Anderson KE
(2000a). Increased urinary excretion of 2-hydroxyestrone but not
16a-hydroxyestrone in premenopausal women during a soya diet
containing isoflavones. Cancer Res 60, 1299–1305.
Lu SC, Wu WH, Lee CA, Chou HF, Lee HR, Huang PC (2000b). LDL of
Taiwanese vegetarians are less oxidizable than those of omnivores.
J Nutr 130, 1591–1596.
Manjari V, Suresh Y, Sailaja Devi MM, Das UN (2001). Oxidant stress,
anti-oxidants and essential fatty acids in South Indian vegetarians
and non-vegetarians. Prostaglandins Leukot Essent Fatty Acids 64,
53–59.
Meydani M, Natiello Goldin B, Free N, Woods M, Schaefer E,
Blumberg JB et al (1991). Effect of long-term fish oil supplementation on vitamin E status and lipid peroxidation in women. J Nutr
121, 484–491.
Mezzano D, Kosiel K, Martinez C, Cuevas A, Panes O, Aranda E et al.
(2000). Cardiovascular risk factors in vegetarians. Normalization
of hyperhomocysteinemia with vitamin B (12) and reduction of
platelet aggregation with n-3 fatty acids. Thromb Res 100, 153–160.
Michnovicz JJ, Adlercreutz H, Bradlow HL (1997). Changes in levels
of urinary estrogen metabolites after oral indole-3-carbinol
treatment in humans. J Natl Cancer Inst 89, 718–723.
Mori TA, Beilin LJ (2001). Long-chain omega 3 fatty acids, blood
lipids and cardiovascular risk reduction. Curr Opin Lipidol 12,
11–17.
Mori TA, Woodman RJ, Burke V, Puddey IB, Croft KD, Beilin L (2003).
Effect of eicosapentaenoic acid and docosahexaenoic acid on
oxidative stress and inflammatory markers in treated-hypertensive
type 2 diabetic subjects. Free Rad Biol Med 35, 772–781.
Morrow JD, Roberts II LJ (1996). The isoprostanes. Current
knowledge and directions for future research. Biochem Pharmacol
51, 1–9.
Muti P, Bradlow HL, Micheli A, Krogh V, Freudenheim JL, Schunemann HJ et al. (2000). Estrogen metabolism and risk of breast
cancer: A prospective study of the 2:16a-hydroxyestrone ratio in
premenopausal and postmenopausal women. Epidemiology 11,
635–640.
Nelson GJ, Schmidt PC, Bartolini GL, Kelley DS, Kyle D (1997). The
effect of dietary docosahexaenoic acid on plasma lipoproteins and
tissue fatty acid composition in humans. Lipids 32, 1137–1146.
Nettleton JA, Greany KA, Thomas W, Wangen KE, Adlercreutz H,
Kurzer MS (2005). The effect of soy consumption on the urinary
2:16-hydroxyestrone ratio in postmenopausal women depends on
equol production status but is not influenced by probiotic
consumption. J Nutr 135, 603–608.
Nourooz-Zadeh J, Halliwell B, Anggard EE (1997). Formation of a
novel class of F3-isoprostanes during peroxidation of eicosapentaenoic acid (EPA). Adv Exp Med Biol 433, 185–188.
Nourooz-Zadeh J, Liu EH, Anggard E, Halliwell B (1998). F4isoprostanes: a novel class of prostanoids formed during peroxidation of docosahexaenoic acid (DHA). Biochem Biophys Res Commun
242, 338–344.
Oostenbrug GS, Mensink RP, Hornstra G (1994). Effects of fish oil
and vitamin E supplementation on copper-catalysed oxidation
of human low density lipoprotein in vitro. Eur J Clin Nutr 48,
895–898.
Osborne MP, Karmali RA, Hershcopf RJ, Bradlow HL (1988). Omega-3
fatty acids: modulation of estrogen metabolism and potential for
breast cancer prevention. Cancer Invest 6, 629–631.
Service RF (1998). New role for estrogen in cancer? Science 279,
1631–1633.
Suzukawa M, Abbey M, Howe PR, Nestel PJ (1995). Effects of fish oil
fatty acids on low density lipoprotein size, oxidizability, and
uptake by macrophages. J Lipid Res 36, 473–484.
European Journal of Clinical Nutrition
DHA and postmenopausal vegetarians
WH Wu et al
392
Takahashi M, Tsuboyama-Kasaoka N, Nakatani T, Ishii M, Tsutsumi S,
Aburatani H et al. (2002). Fish oil feeding alters liver gene
expressions to defend against PPARalpha activation and ROS
production. Am J Physiol-Gastr L 282, G338–G348.
Terry PD, Rohan TE, Wolk A (2003). Intakes of fish and marine fatty
acids and the risks of cancers of the breast and prostate and of
other hormone-related cancers: a review of the epidemiologic
evidence. Am J Clin Nutr 77, 532–543.
Ursin G, Wilson M, Henderson BE, Kolonel LN, Monroe K, Lee HP
et al. (2001). Do urinary estrogen metabolites reflect the
differences in breast cancer risk between Singapore Chinese and
United States African-American and white women? Cancer Res 61,
3326–3329.
European Journal of Clinical Nutrition
Valdes E, Vega P, Avalos N, Orellana M (1995). Dietary fish oil and
cytochrome P-450 monooxygenase activity in rat liver and kidney.
Lipids 30, 955–958.
Wang HH, Hung TM, Wei J, Chiang AN (2004). Fish oil increases
antioxidant enzyme activities in macrophages and reduces
atherosclerotic lesions in apoE-knockout mice. Cardiovasc Res 61,
169–176.
Yamazaki H, Shaw PM, Guengerich FP, Shimada T (1998). Role of
cytochromes P450 1A2 and 3A4 in the oxidation of estradiol and
estrone in human liver microsomes. Chem Res Toxicol 11, 659–665.
Zock PL, Mensink RP, Harryvan J, de Vries JH, Katan MB (1997). Fatty
acids in serum cholesteryl esters as quantitative biomarkers of
dietary intake in humans. Am J Epidemiol 145, 1114–1122.