Atherosclerosis 200 (2008) 396402

Seafood diets: Hypolipidemic and antiatherogenic effects

of taurine and n-3 fatty acids
Edel O. Elvevoll a, , Karl-Erik Eilertsen b , Jan Brox c , Bjrn T. Dragnes a , Pal Falkenberg a ,
Jan O. Olsen b , Bente Kirkhus d , Amandine Lamglait d , Bjarne sterud b
a Department of Marine Biotechnology, Norwegian College of Fishery Science, University of Troms, 9037 Troms, Norway
b Department of Biochemistry, Institute of Medical Biology, Faculty of Medicine, University of Troms, Norway
c Department of Medical Biochemistry, University Hospital of North Norway, Norway
d Mills DA, Oslo, Norway

Received 29 October 2007; received in revised form 12 December 2007; accepted 18 December 2007
Available online 1 February 2008

Abstract
Background: Health aspects of seafood have primarily been linked to n-3 polyunsaturated fatty acids (PUFA), eicosapentaenoic acid (EPA)
and docosahexaenoic acid (DHA). Although animal studies have suggested beneficial contributions from taurine, highly abundant in seafood,
its effect in humans is obscure. This study evaluates the combined effects of n-3 PUFA and taurine.
Methods: Healthy volunteers (n = 80) were recruited to a 7-week double-blind and parallel intervention trial. One group (n = 39) received fish
pate (36 g/day) enriched in n-3 (1.1 g EPA + DHA/day) and the second (n = 41) an identical pate enriched both in n-3 and taurine (425 mg/day).
Results: Total cholesterol (TC) (5%, P < 0.001), low-density lipoprotein (LDL)-cholesterol (8%, P < 0.001) and Apo B (4%, P < 0.001)
decreased more in the n-3 + taurine compared to the n-3 group. A significant within-group enhancement of high-density lipoprotein (HDL)-
cholesterol was demonstrated in the n-3 + taurine group (6%, P < 0.0001). Reductions in triacylglycerol (TG) (16%, P < 0.05 in n-3; 14%,
P < 0.05 in n-3 + taurine), thromboxane B2 (TxB2 ) (21%, P < 0.001 in n-3; 15%, P < 0.05 in n-3 + taurine), tumor necrosis factor (TNF)
(24%, P < 0.001 in n-3; 12%, P < 0.05 in n-3 + taurine) and monocyte chemotactic protein (MCP-1) (12%, P < 0.05 in n-3; 6%,
P < 0.0001 in n-3 + taurine) were evident in both groups. Reductions in interleukin (IL)-6 (16%, P < 0.05) and LTB4 (18%, P < 0.05) were
only significant in the n-3 group.
Conclusions: The effects, particularly on blood lipids, of combining n-3 PUFAs and taurine proved superior to those of n-3 alone.
2008 Elsevier Ireland Ltd. All rights reserved.

Keywords: Seafood; n-3 PUFA; Taurine; Healthy humans; Hypolipidemic; Antiatherogenic

Scientific and technological developments in the field of
food have led to a marked shift in the way people deal with
food and health. There is growing awareness that the dietary
source and degree of processing may affect the overall health
of the consumer. Dietary changes targeted to prevent disease
have exceptional value in that the cost of these modifications
is minimal and they can be applied to entire populations as a
public health measure. The encouragement of consumption
of seafood is such an important public health initiative [1].
Corresponding author. Tel.: +47 776 46146; fax: +47 776 46020.
E-mail address: edel.elvevoll@nfh.uit.no (E.O. Elvevoll).
The health aspects of seafood have primarily, and since
the discovery of the low incidence of coronary vascular
disease (CVD) in Greenland Eskimos, been linked to high
intakes of marine n-3 polyunsaturated fatty acids (PUFA)
in particular eicosapentaenoic acid (EPA) (20:5, n-3) and
docosahexaenoic acid (DHA) (22:6, n-3) [2]. Possible contri-
butions from other beneficial substances have at least partly
been neglected.
Significant vascular benefits from modest fish consump-
tion have been observed. Recent meta-analysis assessing the
effects of consumption of fish, found that people who ate
white or oily fish at least once a week had a significantly
reduced risk of cardiovascular events [3].

0021-9150/$ see front matter 2008 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.atherosclerosis.2007.12.021
 E.O. Elvevoll et al. / Atherosclerosis 200 (2008) 396402
Seafood is considered to be a valuable source of proteins.
Both individual amino acids and the total amino acids pro-
file in fish differ from those found in other sources. Seafood,
fish and especially invertebrates such as mollusks and crus-
taceans, are high in taurine (2-aminoethanesulfonic acid) [4].
The capacity for taurine biosynthesis varies remarkably
among species and since humans only have limited ability
for biosynthesis of taurine by the liver and reabsorption by
the kidneys, taurine needs to be acquired with the diet. As
a consequence, vegans and parenterally fed individuals not
supplemented with taurine have a low plasma taurine content.
Taurine is remarkable for its high intracellular concentrations
but is not incorporated into proteins. A variety of functions
have been ascribed to taurine, ranging from bile acid synthesis
to neurotransmission [5]. Beneficial effects of dietary taurine
have been observed in animal and human studies [5,6] and
suggestions of a reduced CVD risk through taurine alone or
in combination with n-3 PUFA have been put forward [79].
Another aspect of the favorable traditional Eskimo
and Japanese diets, not fully explored, is the effects of
the consumption of less processed marine food items. In
contrast, Western populations consuming little seafood and
high quantities of processed foods have a high prevalence of
CVD. The main task of modern food processing is to ensure
tasty, edible and stable products. Processing and preparation
of seafood causes huge losses of water-soluble compounds
like taurine [4,10].
In this trial we investigated whether moderate intake of a
combination of taurine and marine n-3 (EPA + DHA), obtain-
able through dietary changes reflecting a diet high in seafood,
might be effective in improving lipid metabolism and inflam-
matory markers in healthy subjects.
1. Methods
1.1. Study participants and design
The 80 healthy adult volunteers used as test subjects, 43
women and 37 men, were recruited by local advertising in
the community of Troms, Norway. The regional Ethical
Committee approved the study and all participants signed an
informed consent form before entering the study. The sub-
jects were asked to follow their normal diets and instructed
not to vary these or their lifestyle during the intervention
period. Exclusion criteria were treatment for cardiovascular
disease, pregnancy or severe obesity (body mass index (BMI)
> 35 kg/m2 ). Blood samples were taken before and at the end
of the test period.
Moderate supplementation of taurine (and n-3) was
equated with a level possibly obtained by changing to a diet
high in seafood [11,12]. Mean daily taurine intake in Europe
has been reported to be 58 mg/day (range 40400 mg)
[13,14].
It was recognized from our previous studies that a sample
size of at least 30 subjects was required to achieve a sta-
397
tistical power of at least 80%. The subjects were randomly
divided into two groups (n-3 group, n = 39; n-3 + taurine
group, n = 41) receiving a diet supplemented with approxi-
mately 1 g/day EPA and DHA, with the use of an enriched
fish pate as carrier. In addition, the fish pate distributed to
the subjects in the n-3 + taurine group was enriched with tau-
rine (approximately 425 mg/day). The randomization code
and the products were not accessible for the study-personnel
during the trial.
The two pates, or spreads, were administered as pre-
pared products in 125 g small aluminium tubs. Each subject
received 14 tubs at the start of the study and was asked to cau-
tiously consume the required amount of pate, two tubs a week,
evenly distributed during the test period of 7 weeks. They
were asked to report on their intake at the last blood sampling
appointment when the amount of leftover pate was recorded.
The participants also received a questionnaire where they
were asked to report on their normal diets and changes in
these during the intervention period. Assessment of compli-
ance was based on information from the subjects and the
amount of pate left after the experimental period.
1.2. Preparation of the sh pates
Prepared (salted and smoked) fish muscle of farmed
Atlantic salmon (Salmo salar) and cod liver oil (CLO)
(DenomegaTM , Borregaard Ingredients, Sarpsborg, Norway)
was blended with rapeseed oil based (egg-free) mayonnaise.
Taurine (Sogo Pharmaceutical Co. Ltd. Tokyo, Japan) was
added in one batch of the product. The fish pates were made
by Mills DA, Oslo, Norway.
1.3. Total lipids and fatty acids in sh pates
The lipids of the fish pates were extracted using a
modified Bligh and Dyer [15] method (n = 3) and fatty acid
(FA) methylation was done using a modified procedure as
described in [16].
1.4. Free amino acids in sh pates
The samples were subjected to analysis of free amino acids
(FAA) using a procedure as described in [10].
1.5. Blood collection
Blood was drawn as previously reported [16], at the start
of the intervention period and at the end of the study 7 weeks
later, after an overnight fast between 08:00 and 10:00 a.m. to
minimize the effect of ingested food.
1.6. Serum fatty acids
The FAs were extracted from serum as previously
described [17] and analyzed using a modified procedure
described in a previous study [16].
398 E.O. Elvevoll et al. / Atherosclerosis 200 (2008) 396402

1.7. Serum taurine
Thawed serum (0.36 mL) was deproteinized by adding
0.054 mL of 35% (1.6 M) SSA (with 0.04 mM cysteic acid as
internal standard) at 4 C for 1 h. The mixture was centrifuged
at 20.000 g for 15 min, 5 C and the supernatant was used
for analysis. The concentration of taurine was determined
using a Knauer A220 amino acid analyzer (Knauer GmbH,
Berlin, Germany) with a lithium citrate equilibrated column.
1.8. Cell counts and standard hematological parameters
The procedure of measuring cell counts and hematological
parameters are given in a recent paper [16].
1.9. Determination of serum lipids, lipoproteins and
apolipoproteins
Fasting serum concentrations of total cholesterol (TC),
HDL cholesterol, LDL cholesterol, triacylglycerol (TG),
apolipoprotein A1 (Apo A), and apolipoprotein B-100 (Apo
B) were measured as previously described [18].
1.10. Ex vivo stimulation of whole blood with
lipopolysaccharide (LPS)
phosphatase (ALP) and the kidney marker creatinine were
analyzed by a Hitachi 917 autoanalyzer with reagents from
Roche Diagnostics Mannheim, Germany. The analytical
median coefficient of variation (CV) was <4% for all assays.
1.12. Statistical methods
All of the analyses were performed using SPSS for Win-
dows (release 14.0.2; SPSS, Inc., Chicago, IL, USA). Values
are presented as mean standard deviation unless otherwise
stated. Normality was tested and skewed data was log trans-
formed before statistical analysis. The significance of any
differences between the groups at baseline or the impact of
treatment on the changes in each group was determined using
independent samples t-test or MannWhitney U-test. Within
group, changes were analyzed using a paired Students t-test
or Wilcoxon signed-rank test. To correct for bias, analysis of
covariance (ANCOVA) was enlisted to eliminate the effect
of difference at baseline level. Differences at the level of
P < 0.05 were considered statistically significant.
2. Results
2.1. Compliance

The procedures for LPS activation of whole blood and
quantification of tissue factor (TF), TNF, IL-6, LTB4 and
TxB2 are previously described [16].
1.11. Other laboratory procedures
Monocyte chemotactic protein-1 (MCP-1) in plasma of
non-stimulated blood was determined by means of ELISA
(Amersham Pharmacia Biotech, Little Chalfont, UK). Plas-
minogen activator inhibitor-1 (PAI-1) was measured in
citrated plasma using the PAI-1 Antigen ELISA Reagent Kit
(Technoclone GmbH, Vienna, Austria). Fibrinogen plasma
levels were measured according to Inada et al. [19]. The
toxicity indicators (liver enzymes) aspartate aminotrans-
ferase (ASAT), alanine aminotransferase (ALAT), alkaline
Compliance was assessed based on information from the
subjects and the amount of pate left after the experimental
period. All the 80 subjects recruited completed the study,
whereas five subjects (two in n-3 and three in n-3 + taurine)
were removed from the results due to lack of compli-
ance. There were no differences in baseline demographics
(supplemental Table I) and serum lipid levels (Table 1) for
the study groups.
2.2. Fish pate composition
The total lipid contents were 23.9 0.4 g/100 g in the n-
3 and 24.4 0.3 g/100 g in the n-3 + taurine pate. The fatty
acid composition of the total lipids (g/100 g) and the pates
(mg/g) are given (supplemental Table II). The intake was

Table 1
Mean (S.D.) values of serum lipids (mmol/L), Apo A (g/L) and Apo B (g/L) of the subjects at baseline and change during the intervention (t7weeks tbaseline )
n-3 (n = 37) n-3 + taurine (n = 38) Pa Pb

Baseline Change Baseline Change

Total cholesterol (mmol/L) 5.53 1.25 0.04 0.43 5.54 1.13 0.25 0.41c 0.004 0.01
Triacylglycerol (mmol/L) 1.17 0.63 0.19 0.53c 1.14 0.55 0.16 0.44c 0.84 0.27
LDL-cholesterol (mmol/L) 3.38 1.09 0.09 0.40 3.38 1.01 0.26 0.42c 0.001 0.02
HDL-cholesterol (mmol/L) 1.61 0.33 0.05 0.18 1.64 0.43 0.10 0.14c 0.16 0.32
Apo A (g/L) 1.53 0.25 0.01 0.11 1.53 0.26 0.01 0.13 0.81 0.83
Apo B (g/L) 0.93 0.26 0.03 0.09c 0.93 0.23 0.03 0.08c 0.001 0.02
a Independent samples 2-tailed t-test for the change between groups.
b To correct for bias, analysis of covariance (ANCOVA) was enlisted to eliminate the effect of differences in the baseline levels of the fatty acid EPA
(log-transformed values used as covariates) in the two groups on the changes in serum lipid and lipoprotein levels (TC, TG, LDL-C, HDL-C, Apo,
and Apo B: dependent variables) after treatment (independent variable).
7weeks tbaseline ; paired samples t-test or Wilcoxon signed-rank test).
c P < 0.05 within-group changes (t
 E.O. Elvevoll et al. / Atherosclerosis 200 (2008) 396402 399

Table 2
Mean (S.E.M.) values of serum fatty acids (mmol/L), basal levels and changes during the intervention are given
n-3 (n = 37) n-3 + Taurine (n = 38)

Baseline Change Baseline Change

EPA 20:5, n-3 0.144 0.016 0.098 0.0192 0.206 0.024a 0.133 0.026b
DHA 22:6, n-3 0.298 0.022 0.050 0.020b 0.339 0.023 0.014 0.019
AA 20:4, n-3 0.611 0.031 0.104 0.031b 0.583 0.025 0.095 0.019b
a P < 0.05 between-group differences at baseline.
b P < 0.05 within-group changes (t7weeks tbaseline ; Wilcoxons signed-rank test).

1.09 g (506 mg EPA and 585 mg DHA/day) in the n-3 group
and 1.15 mg (527 mg EPA and 626 mg DHA per day) in the
n-3 + taurine group. The taurine concentrations in the pates
were 0.17 0.0 mg/g in the n-3 and 11.9 0.2 mg/g in the n-
3 + taurine. The groups had a daily intake of taurine from pate
of 6 mg (n-3) (prepared salmon) and 425 mg (n-3 + taurine)
per day, respectively.
2.3. Serum fatty acid composition
The daily intake of EPA and DHA was 1.09 and 1.15 g/day
in intervention groups and the relative changes in serum EPA
and DHA were only slightly different in the two groups.
The baseline level of EPA was significantly higher in the
n-3 + taurine group. A significant enhancement of EPA was
observed within both groups. The increases in DHA did not
reach significant levels for the n-3 + taurine group (Table 2)
(supplemental Table III).
2.4. Serum taurine concentration
No significant differences were observed in serum
taurine concentrations between the groups at baseline
((mol/L) 51.5 13.0, mean of all subjects; 51.8 12.8,
n-3; 50.6 12.0, n-3 + taurine). The changes in taurine con-
centration were significantly different between the groups.
The n-3 + taurine group resulted, as expected, in a significant
increase from 50.6 mol/L (S.D. 12.0) to 67.7 mol/L (S.D.
19.3) or 34% in serum taurine concentration. This was calcu-
lated after the removal of one subject with an extreme increase
in serum taurine concentration (from 65.2 to 353.9 mol/L).
significant enhancement in total HDL-cholesterol (6%) was
observed in the n-3 + taurine group. Fig. 1 shows the changes
from baseline values for the serum lipids. After treatment,
serum total cholesterol, LDL cholesterol, and the ratio of
total-to-HDL cholesterol were observed to decrease signifi-
cantly more in the n-3 + taurine group compared to the n-3
group. Noteworthily, the ratio of total-to-HDL cholesterol
did not decrease significantly within the n-3 group (1.86%,
n.s.) whereas the ratio decreased significantly within the n-
3 + taurine group (9.23%, P < 0.001).
2.6. Cell counts
No significant changes were seen in WBC or RBC. An
examination of within-group changes demonstrated a signif-
icant enhancement of MPV in both groups and reduction of
number of platelets in the n-3 group (results not shown).
2.7. Toxicity indicators
None of the markers of liver enzyme activity (ASAT,
ALAT, creatinine) were significantly changed (results not
shown). A within-group analysis of changes in ALP activ-
ity revealed a significant reduction in the n-3 + taurine group
(68.2 17.0; 2.8 6.4; 4%, P < 0.01).

2.5. Serum lipids and apolipoproteins

No significant differences were observed in the serum lipid

levels at baseline (Table 1).
Total cholesterol (5%), LDL-cholesterol (8%) and
Apo B (4%) all decreased significantly more in the n-
3 + taurine group compared to the n-3 group (Table 1). To
correct for possible bias, analysis of covariance (ANCOVA)
was enlisted to eliminate the effect of difference at base-
Fig. 1. Changes from baseline values for the serum lipids. After treatment,
line level of the fatty acids EPA and DHA (n.s.) in the two
significant decreases were observed for serum total cholesterol, LDL choles-
groups on the changes in serum lipid and lipoprotein lev- terol, and for the total-to-HDL cholesterol ratio (Table 1). Bars and error bars
els after treatment. A significant within-group reduction in indicate mean and S.E.M., respectively. *P < 0.05 for treatment differences
TG of (16% and 14%) was evident in both groups and a between n-3 and n-3 + taurine groups.
400 E.O. Elvevoll et al. / Atherosclerosis 200 (2008) 396402

Table 3
Baseline and changes in inflammatory parameters (mean S.D.) after the 7 weeks intervention period
n-3 (n = 37) n-3+Taurine (n = 38) Pa Pb
Baseline Change Baseline Change
hsCRP (mg/L) 1.66 2.15 0.12 2.35 2.85 4.64 0.71 5.14 0.42 0.58
MCP-1 (pg/mL) 415 81.8 23.2 68.9c 435 113 54.2 77.9c 0.07 0.05
TF (mU/106 cells) 48.1 24.5 0.40 25.4 51.2 25.0 2.11 24.8 0.63 0.56
IL-6 (pg/mL) 1208 555 195 546c 1101 558 58.4 408c 0.22 0.13
TNF (pg/Ml) 2284 871 549 955c 2296 1093 278 781 0.18 0.29
TxB2 (ng/mL) 8.81 11.5 1.83 3.68c 5.78 3.42 0.85 2.32c 0.59 0.71
LTB4 (pg/mL) 404 472 71.7 180c 349 378 26.3 179 0.04 0.27
hsCRP, hyper-sensitive C-reactive protein; MCP-1, monocyte chemotactic protein -1; TF, tissue factor; IL-6, interleukin-6; TNF, tumor necrosis factor ;
LTB4 , leukotriene B4; TxB2 , thromboxane B2 .
a Independent samples 2-tailed t-test for the change between groups. Non-parametric variables where analyzed by MannWhitney U-test.
b ANCOVA was used to correct the effect of differences in the baseline levels of the fatty acids EPA (log-transformed; covariate) in the two groups.

Non-parametric variables were log-transformed.

7weeks tbaseline ; paired samples t-test or Wilcoxon signed-rank test).
c P < 0.05 within-group changes (t

2.8. Changes in inammatory markers
MCP-1 was significantly reduced in both the groups,
12% in the n-3 + taurine and 6% in the n-3 group. When
using ANCOVA to correct for the difference in baseline lev-
els of EPA in the two groups, the difference between the
MCP-1 levels in the two groups was significant (Table 3).
The ANCOVA analysis revealed no significant difference
between the two groups in the LPS-induced activation prod-
ucts even though LPS-induced TNF was reduced 24% in
the n-3 group and 12% in the n-3 + taurine group, whereas
IL-6 reduction was only significant in the n-3 group (16%)
versus 12.3% in the n-3 + taurine group (n.s.). LPS-induced
TxB2 was significantly reduced in both groups (21% in n-3
and 15% in n-3 + taurine). These changes were parallel to a
significant reduction in plasma arachidonic acid (AA) (20:4,
n-6), the metabolic precursor for these eicosanoids, in both
groups (17% in the n-3 group and 16% in the n-3 + taurine
group). A comparison of the changes between the n-3 and the
n-3 + taurine group showed a significant difference in LPS-
induced LTB4 (Table 3). A within-group analysis showed
that LTB4 was significantly reduced (18%) in the n-3 group
whereas the enhancement (8%) in the n-3 + taurine group was
not significant. No significant changes were observed in the
coagulation factors, fibrinogen, tissue factor (LPS-induced)
and the fibrinolytic test parameter PAI-1 (results not shown).
Similarly, no significant changes were found in hsCRP during
this intervention trial.
3. Discussion
This trial was designed to study if hypolipidemic and
antiatherogenic effects from a diet high in seafood may be
attributed to n-3 (EPA + DHA) alone or to combined effects of
n-3 and taurine. The outcome was measured by the changes in
blood lipids and lipoproteins, as well as various inflammatory
markers.
Dietary uptake is the major method of taurine supply
as humans have limited ability for biosynthesis of taurine,
and it may therefore be regarded as conditionally essen-
tial [5]. The baseline concentrations of serum taurine in
this experiment (51.8 12.8 mol/L) were in close agree-
ment with results (48.6 4.9 mol/L) previously reported
for healthy human controls on a western diet [20]. Con-
sumption of fish pate enriched in n-3 + taurine resulted in
a significant increase in serum taurine. Fish muscle con-
sumption is reported to result in increased concentrations of
serum taurine when compared with beef and chicken mus-
cle [21] and women from Korean fish farming areas are
reported to be high in serum taurine (169.7 41.5 mol/L)
[12].
Significantly stronger reductions were observed in total
cholesterol, LDL-cholesterol and Apo B in the n-3 + taurine
group compared to the n-3 group (Table 1). Human and ani-
mal intervention trials have demonstrated beneficial effects
of taurine on serum lipids [6]. One of the well-documented
biological activities of taurine is bile salt formation. Tau-
rine (and glycine) conjugates with cholesterol derivates
to form taurocholate (and glycocholate). Taurocholate is
the major bile salt that extracts cholesterol from plasma
in humans and a decreased taurine content is associ-
ated with a lower cholesterol extraction and subsequent
accumulation and increased risk of atherosclerosis. The
antiatherosclerotic effects of taurine have been studied
in different hypercholesterolaemic and hyperlipidemic ani-
mals, but the exact mechanism of action is still unclear
[5,6]. Taurine supplemented diets have been shown to
improve bile faecal excretion and to suppress the devel-
opment of atherosclerotic lesions in mice and rabbits
[22].
A further positive effect of n-3 + taurine was observed in
the total-to-HDL cholesterol ratio. This was not found for
the n-3 group. Our results indicate that taurine may have an
additive effect of n-3 fatty acids to enhance HDL-cholesterol.
Whether the effect is mediated through n-3 fatty acids cannot
be resolved by this study.
 E.O. Elvevoll et al. / Atherosclerosis 200 (2008) 396402
The effects measured in this study may result from a com-
bined effect of taurine and n-3 fatty acids [8]. Two Japanese
studies in rats on hypercholesterolemic diets suggest such
combined effects. In one study effects are attributed to both
the lipid and the non-lipid fractions of oysters [23], known
to be high in taurine [4], or in the other study to sulphur con-
taining amino acids in general, and taurine in particular, in
combination with EPA and DHA [24].
The changes in blood lipids in the n-3 group, although
not significant, were in agreement with data reported from
several parallel studies reviewed by Harris [25]. There were
no changes on total cholesterol and small enhancements of
LDL- cholesterol and HDL-cholesterol in this group. The rel-
atively low amount of EPA and DHA (1.1 g/day), compared
to 34 g/day may explain the lack of significant enhance-
ments. Within-group analysis revealed that serum TG was
significantly reduced in both groups. Elevated levels of serum
TG are commonly thought to affect atherogenesis directly
and indirectly. Fish oils are considered among the most effi-
cient dietary means of reducing TG levels, with optimal daily
intakes of 1.5 g EPA and DHA per day [25].
Taurine is known to have antioxidant activity and has been
suggested to reduce the production of inflammatory media-
tors [26]. MCP-1 was reduced in the n-3 + taurine group and
to a lesser extent in the n-3 group. Daily ingestion of 7 g/day
n-3 fatty acids by human volunteers has been reported to
reduce MCP-1 mRNA levels in ex vivo mononuclear cells,
and the authors suggested that the reduced expression of
MCP-1 to be involved in regression of established CVD [27].
Except for the pro-inflammatory product MCP-1, the changes
in inflammatory biomarkers did not suggest an additional
anti- inflammatory effect of taurine above that of the n-3
fatty acids. An enhanced reduction of the production of pro-
inflammatory mediators, through a secondary metabolite of
taurine, taurine chloramine, has been reported [28] but was
not observed in this study. The design of the present study
does not allow for calculation of the separate effects of taurine
and n-3. Our results indicate that the n-3 effect on MCP-
1 may be enhanced by taurine. LPS-stimulation of whole
blood has been used as an indicator of the potential of blood
cells to generate inflammatory products. LPS-TNF was
reduced in both groups with the highest reduction in the n-3
group.
The changes in serum fatty acids after intake of the n-3
enriched pate or n-3 + taurine enriched pate (Table 2) were
similar. Primarily EPA, but also DHA, is able to inhibit the
formation of important pro-inflammatory eicosanoids, TxA2
and LTB4 from AA, as well as function as metabolites for
the biosynthesis of less active eicosanoids such as TxA3
and LTB5 . The pro-inflammatory lipid mediator TxB2 was
significantly reduced in both groups. LTB4, which plays an
important role in atherogenesis, was significantly reduced in
the n-3 group but slightly enhanced in the n-3 + taurine group.
The reduction of TxB2 and LTB4 are known to be associated
with an increased intake of n-3 PUFA and a reduced incidence
of CVD [29].
401
A recent epidemiological study addressing male car-
diovascular mortality and dietary markers finds taurine
excretion to be the most significant single factor, correlat-
ing inversely with ischemic (IHD) heart disease mortality
[8]. This study also reports significant dependence of the
combined effects of taurine and n-3 PUFA and IHD mor-
tality.
3.1. Limitations of the study
The present study was designed to answer whether other
abundant beneficial components, such as taurine, in addi-
tion to n-3 PUFA, may contribute to the protective effects
of seafood towards CVD risk factors. Seafood is a well-
known source of n-3 PUFA, but the content taurine and
possible actions are not widely studied or known. It was
recognized from our previous studies that a sample size of
at least 30 subjects were required to achieve a statistical
power of at least 80%. Due to uncertainty of the voluntary
test subjects to fulfil 7 weeks of fish diet supplementation
it was decided to start with around 40 subjects in each
group. The limitations in funding and the goal of perform-
ing a realistic study on the effect of mimicking a diet rich
in seafood where the taurine is consumed together with
n-3 PUFA, and at the same time be able to compare the
effects of consumption of n-3 PUFA alone, led to this study
design.
Therefore, this study has the limitation that we cannot
decide whether measured effects on, for instance, the blood
lipids are exerted by taurine alone or the combined action
of taurine and n-3 PUFA. Consuming 1 g PUFA/day and
425 mg taurine/day from seafood would best be acquired
trough a mixed diet of fatty fish and shellfish. As an example,
the consumption of 3040 g/day of salmon and 6080 g/day
of shellfish will meet the requirement. This will represent a
consumption for instance of per capita annual intake of live
weight (as usually reported by FAO) at a level of 6075 kg,
at the same level as of high-fish consumer countries such
as Japan (65.7 kg) or Iceland (91.1 kg). Future research will
be needed to explore the effects of dietary advices with a
realistic frequency of seafood consumption. The possible
effects of long-term consumption over which the metabolic
effects might either increase or diminish, should also be stud-
ied.
There was no difference in baseline demographics and
serum lipids (Supplemental Table I). Dietary analysis was
completed, each subject was requested not to change dietary
and exercise habits during the study, and no such changes
were detected. No subjects were on any drugs. Subjects were
not screened or selected on the basis of metabolic risk fac-
tors, so regression to the mean could not have confounded
the results. However, the study may have several limitations
inherent and it may seem important to study the effects of tau-
rine with or without co-treatment with n-3 PUFA in a larger
study.
402 E.O. Elvevoll et al. / Atherosclerosis 200 (2008) 396402
4. Conclusion
This is the first human intervention trial to report positive
effects of taurine in combination with n-3 (EPA + DHA) in
healthy humans on a normal diet. In addition, the intake or
amounts supplemented of both taurine and n-3 (EPA + DHA)
are obtainable from a diet high in seafood or through changes
based on a western diet. Intake of the n-3 + taurine enriched
fish pate for 7 weeks by healthy subjects had additional ben-
eficial effects in particular upon blood lipids as the ratio of
total cholesterol over HDL-cholesterol was reduced by 9.2%
in the n-3 + taurine group compared to 1.8% with n-3 alone.
Preparation of seafood causes losses of water-soluble com-
pounds like taurine; thus, these effects may be expected to
be elevated when consuming less prepared foods [4,10]. Fur-
ther studies are warranted both to allow for the calculation of
the separate effects of taurine and to explore the dose depen-
dence of both the separate effects as well as the combined
effects and last but not least studies are needed to shed light
on long-term effects.
Acknowledgement
The technical assistance of H. Mhre and S.K. Stormo is
greatly appreciated.
Sources of funding: This project was a part of the DOC-
MAR project funded by Innovation Norway, RUBIN and
Norwegian Research Council.
Disclosures: There are no conflicts of interest.
Appendix A. Supplementary data
Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/
j.atherosclerosis.2007.12.021.
References
[1] Schmidt EB, Arnesen H, Christensen JH, et al. Marine n-3 polyunsatu-
rated fatty acids and coronary heart disease Part II: clinical trials and
recommendations. Thromb Res 2005;115:25762.
[2] Dyerberg J, Bang HO, Stoffersen E, Moncada S, Vane JR. Eicosapen-
taenoic acid and prevention of thrombosis and atherosclerosis? Lancet
1978;2:1179.
[3] He K, Song Y, Daviglus ML, et al. Accumulated evidence on fish
consumption and coronary heart disease mortality: a meta-analysis of
cohort studies. Circulation 2004;109:270511.
[4] Roe DA, Weston MO. Potential significance of free taurine in the diet.
Nature 1965;205:2878.
[5] Bouckenooghe T, Remacle C, Reusens B. Is taurine a functional nutri-
ent? Curr Opin Clin Nutr Metab Care 2006;9:72833.
[6] Militante JD, Lombardini JB. Dietary taurine supplementation:
hypolipidernic and antiatherogenic effects. Nutr Res 2004;24:787801.
[7] Mizushima S, Moriguchi EH, Ishikawa P, et al. Fish intake and car-
diovascular risk among middle-aged Japanese in Japan and Brazil. J
Cardiovasc Risk 1997;4:1919.
[8] Yamori Y, Liu L, Mizushima S, Ikeda K, Nara Y. Male cardiovascular
mortality and dietary markers in 25 population samples of 16 countries.
J Hypertens 2006;24:1499505.
[9] Yamori Y, Liu LJ, Ikeda K, et al. Distribution of twenty-four hour
urinary taurine excretion and association with ischemic heart disease
mortality in 24 populations of 16 countries: results from the WHO-
CARDIAC Study. Hypertens Res 2001;24:4537.
[10] Larsen R, Stormo SK, Dragnes BT, Elvevoll EO. Losses of taurine,
creatine, glycine and alanine from cod (Gadus morhua L.) fillet during
processing. J Food Compos Anal 2007;20:396402.
[11] Kibayashi E, Yokogoshi H, Mizue H, et al. Daily dietary taurine intake
in Japan. Adv Exp Med Biol 2000;483:13742.
[12] Kim ES, Kim JS, Yim MH, et al. Dietary thurine intake and serum
thurine levels of women on Jeju Island. In: Taurine 5: beginning the
21st century; 2003. p. 27783.
[13] Rana SK, Sanders TA. Taurine concentrations in the diet, plasma,
urine and breast milk of vegans compared with omnivores. Br J Nutr
1986;56:1727.
[14] Laidlaw SA, Grosvenor M, Kopple JD. The taurine content of common
foodstuffs. Jpen-Parenter Enter 1990;14:1838.
[15] Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purifi-
cation. Can J Biochem Physiol 1959;37:9117.
[16] Elvevoll EO, Barstad H, Breimo ES, et al. Enhanced incorporation of
n-3 fatty acids from fish compared with fish oils. Lipids 2006;41:1109
14.
[17] Osterud B, Elvevoll E, Barstad H, et al. Effect of marine oils supple-
mentation on coagulation and cellular activation in whole blood. Lipids
1995;30:11118.
[18] Eilertsen KE, Olsen JO, Brox J, Osterud L. Association of the
159C T polymorphism in the CD14 promoter with variations
in serum lipoproteins in healthy subjects. Blood Coagul Fibrinolysis
2003;14:66370.
[19] Inada Y, Okamoto H, Kanai S, Tamaura Y. Faster determination of
clottable fibrinogen in human plasma: an improved method and kinetic
study. Clin Chem 1978;24:3513.
[20] Brns C, Spohr C, Storgaard H, Dyerberg J, Vaag A. Effect of taurine
treatment on insulin secretion and action, and on serum lipid levels
in overweight men with a genetic predisposition for type II diabetes
mellitus. Eur J Clin Nutr 2004;58:123947.
[21] Uhe AM, Collier GR, Odea K. A comparison of the effects of beef,
chicken and fish-protein on satiety and amino-acid profiles in lean male-
subjects. J Nutr 1992;122:46772.
[22] Balkan J, Oztezcan S, Hatipoglu A, et al. Effect of a taurine treat-
ment on the regression of existing atherosclerotic lesions in rabbits
fed on a high-cholesterol diet. Biosci Biotechnol Biochem 2004;68:
10359.
[23] Tanaka K, Ikeda I, Kase A, et al. Effects of feeding oyster, Crassostrea
gigas, on serum and liver lipid levels in rats. J Nutr Sci Vitaminol
(Tokyo) 2003;49:1006.
[24] Aoki T, Matsukura M, Hokari K, Kunisaki N. Effect of simultane-
ous administration of epa, dha, and sulfur-containing amino-acids on
serum-lipids in rats fed hypercholesterolemic diets. Nippon Suisan
Gakkaishi 1994;60:6616.
[25] Harris WS. n-3 fatty acids and serum lipoproteins: human studies. Am
J Clin Nutr 1997;65:1645S54S.
[26] Niittynen L, Nurminen ML, Korpela R, Vapaatalo H. Role of argi-
nine, taurine and homocysteine in cardiovascular diseases. Ann Med
1999;31:31826.
[27] von Schacky C, Baumann K, Angerer P. The effect of n-3 fatty acids
on coronary atherosclerosis: results from SCIMO, an angiographic
study, background and implications. Lipids 2001;36(Suppl):S99
102.
[28] Schuller-Levis GB, Park E. Taurine: new implications for an old amino
acid. Fems Microbiol Lett 2003;226:195202.
[29] Simopoulos AP. Omega-3 fatty acids in inflammation and autoimmune
diseases. J Am Coll Nutr 2002;21:495505.