Professional Documents
Culture Documents
human nutrition
Also in the Oily Press Lipid Library:
Volume 22. Phospholipid Technology and Applications
Edited by Frank D. Gunstone
Volume 21. Long-Chain Omega-3 Specialty Oils
Edited by Harald Breivik
Volume 20. Antioxidants in Food and Biology: Facts and Fiction
Written by Edwin N. Frankel
Volume 19. Lipids: Structure, Physical Properties and Functionality
Written by Kåre Larsson, Peter Quinn, Kiyotaka Sato and Fredrik Tiberg
Volume 18. Lipid Oxidation (second edition)
Written by Edwin N. Frankel
Volume 17. Bioactive Lipids
Edited by Anna Nicolaou and George Kokotos
Volume 16. Advances in Lipid Methodology – Five
Edited by Richard O. Adlof
Volume 15. Lipid Analysis (third edition)
Written by William W. Christie
Volume 14. Confectionery Fats Handbook
Written by Ralph E. Timms
Volume 13. Lipids for Functional Foods and Nutraceuticals
Edited by Frank D. Gunstone
Volume 12. Lipid Glossary 2
Written by Frank D. Gunstone and Bengt G. Herslöf
Volume 11. Lipids in Nutrition and Health: A Reappraisal
Written by Michael I. Gurr
Volume 9. Trans Fatty Acids in Human Nutrition (first edition)
Edited by Jean Louis Sébédio and William W. Christie
Volume 8. Advances in Lipid Methodology – Four
Edited by William W. Christie
Volume 7. Advances in Lipid Methodology – Three
Edited by William W. Christie
Volumes 1– 6 and 10. Out of print
Woodhead Publishing in Food Science, Technology and Nutrition
Edited by
FRÉDÉRIC DESTAILLATS
Nestlé Research Center, Lausanne, Switzerland
JEAN-LOUIS SÉBÉDIO
UMR 1019, Plateforme d’exploration du métabolisme, INRA centre
de Theix, St Genès Champanelle, France
FABIOLA DIONISI
Nestlé Research Center, Lausanne, Switzerland
JEAN-MICHEL CHARDIGNY
UMR 1019 INRA Université Clermont I,
Clermont-Ferrand, France
Woodhead Publishing India Private Limited, G-2, Vardaan House, 7/28 Ansari Road,
Daryaganj, New Delhi – 110002, India
www.woodheadpublishingindia.com
This book contains information obtained from authentic and highly regarded sources.
Reprinted material is quoted with permission, and sources are indicated. Reasonable
efforts have been made to publish reliable data and information, but the authors and the
publisher cannot assume responsibility for the validity of all materials. Neither the authors
nor the publisher, nor anyone else associated with this publication, shall be liable for any
loss, damage or liability directly or indirectly caused or alleged to be caused by this book.
Neither this book nor any part may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopying, microfilming and recording, or
by any information storage or retrieval system, without permission in writing from
Woodhead Publishing Limited.
The consent of Woodhead Publishing Limited does not extend to copying for general
distribution, for promotion, for creating new works, or for resale. Specific permission
must be obtained in writing from Woodhead Publishing Limited for such copying.
The chemistry of fats and oils has enjoyed a long and successful history. The
first evidence of the occurrence of trans fatty acids (TFA) in edible fats was
demonstrated by direct chemical analysis more than 80 years ago by Bertram in
1928 (Biochem. Z., 197, 433–441). In studying ruminant fats, Bertram discov-
ered the trans-11 18:1 acid and named it vaccenic acid. It was shown later that
vaccenic acid is not the only TFA found in ruminant fats and more recent
research revealed that vaccenic acid is further metabolized in ruminants as well
as in other animals.
Over the last thirty years numerous studies have been carried out in a number
of fields including analytical chemistry, food science, nutrition and epidemiol-
ogy to understand the composition, physical properties and health implications
of TFA found in partially hydrogenated vegetable oils. The basic chemical
information gained was that partial hydrogenation of vegetable oils generates
a very complex and diverse profile of TFA isomers. These TFA were found to
be conspicuously more stable toward chemical oxidation reactions compared
to their polyunsaturated precursors and to exhibit distinct physical properties.
However, the most recent research over the past two decades has documented
various detrimental effects of consumption of TFA on risk factors of vascular
health.
Since the 1990s there has been increasing regulatory concern about the
health effects of the trans mono-ethylenic acid isomers formed during partial
hydrogenation of vegetable oils. Consequently public health policies have
been implemented in various countries including Denmark, USA and Canada,
to ban the use or limit the consumption of TFA from industrial origin. However,
debate still rages around the world as to agreeing the most appropriate policies,
determining which specific chemicals are deleterious and by what mechanisms
and in what quantities, and deciding how regulatory agencies should guide the
public to appropriate food choices based on their TFA contents. To frame this
debate, scientific knowledge must take a central role. Therefore the editors
undertook to produce a state-of-the-art book that assembles the scientific
knowledge of trans fats – what is known and what needs to be determined.
An earlier book carried out a very similar task for the state of our knowledge
in the late 1990s; this was Trans Fatty Acids in Human Nutrition co-edited by
one of us ( Jean-Louis Sébédio) together with William W. Christie who also
contributes to the present book. Also like the present book, the earlier volume
was published by The Oily Press. Therefore it was decided that the new book
v
vi PREFACE
should become the Second Edition of Trans Fatty Acids in Human Nutrition
even though the rapid expansion and progress in the subject meant that it would
be completely re-written and be expanded from the original 9 to the present 15
chapters.
In this book, authors who are recognized international authorities in their
field have addressed various domains of TFA research such as consumption,
analysis, biochemistry, synthesis and natural TFA biosynthesis, health effects,
food formulation, and also regulation and consumer perception.
Each chapter contains the latest references and major advances and break-
throughs in the different areas of scientific research. Furthermore, the book also
includes a discussion of a major question on the health effects of the ‘natural
trans isomers’, comparing their effects to those observed for the industrially
produced TFA. We hope that the availability of so much information in a single
volume will help to clarify the major effects of TFA in human nutrition
discovered over the last two decades and guide the next generation of scientists
to the important opportunities for making further progress in this challenging
field of research.
12 Dietary trans fatty acids: from the mother’s diet to the infant 319
JEAN-MICHEL CHARDIGNY AND NICOLE COMBE
A. Introduction 319
B. TFA in infant nutrition 320
1. TFA in human milk
2. TFA in infant formulas
C. TFA: from the mother to foetus and infant 322
1. Placental transfer
2. Incorporation into cord tissues
D. Effects of TFA on infant growth and development 323
E. TFA and essential fatty acid metabolism in newborns 324
F. Conjugated linoleic acid (CLA) isomers in infant nutrition 325
G. Conclusion 325
References 325
1. North America
2. UK and Ireland
3. Continental Europe
4. Nordic Countries
5. Australia
6. Asian/Pacific Region
7. Central and South America
8. Iran
C. Consumption of trans fatty acids: future trends 372
Acknowledgement 374
References 374
Fabiola Dionisi, Nestlé Research Center, PO Box 44, CH-1000 Lausanne 26,
Switzerland
Mark Waldron, Nestlé Research Center, PO Box 44, CH-1000 Lausanne 26,
Switzerland
Prologue
J. BRUCE GERMAN
edible fats (butter, lard, tallow) acted to have the cheaper and inferior hydro-
genated substitutes banned, regulated and labelled.
During the 20th century continuous improvements in the chemistry of hydro-
genation, from improved selectivity of catalysts and reactors to blending of
diverse feedstocks, provided considerable control over the overall hydrogena-
tion process and its products. The quality of hydrogenated oil products increased
dramatically and led to the development of highly tailored functional edible
fats in shortening, confectionery and baking applications. One by-product of
the reaction conditions designed to induce addition of hydrogen to fatty acids,
though largely unnoticed, was critical to their success. Within the reactor,
interaction of fatty acids with the metal catalyst surface led to the isomerization
of cis double bonds to trans configuration. Interestingly, the consequences for
the physical properties of the products were not undesired. As important
components of partially hydrogenated oils, the trans fatty acids in oleomarga-
rine and shortenings actually provided the industrial chemists with a valuable
functionality. The apparently simple structural differences between the cis and
trans isomer configuration of a double bond have profound physical conse-
quences for the triacylglycerol molecules containing them. For C18 fatty acids
in triacylglycerols, monounsaturated fatty acids with a cis double bond are
liquid at room temperature whereas monounsaturated fatty acids with a trans
double bond are solid fats. The trans fatty acids allowed the fats to match the
melting points and crystal habits specified for particular product applications.
Hence, the trans structure was part of the physical properties that propelled
these hydrogenated oils into the global food supply. Nonetheless, as well as
being an effective means of conferring the physical properties of traditional
fats, these hydrogenated fats provided the additional benefit of being an
inexpensive substitute for the more valuable animal products such as butter,
lard and tallow. This situation was changed by a new factor in the marketing of
edible fats: the risk of causing disease.
Research in the mid-1900s began to link diet to long-term health. In
particular, early studies that examined the newly developed biomarker of heart
disease risk, serum cholesterol, found that consuming very high levels of
saturated fatty acids and animal fats caused an elevation of plasma cholesterol.
Over time, even though the risk of heart disease was clearly a highly complex
and multi-faceted problem, saturated fats became viewed as the villain at the
‘heart’ of heart disease. In the media and the marketing of ingredients in the
food marketplace, worldwide attention turned to the composition of edible fats
and oils. For animal fats and their saturated fatty acids there was nowhere to
hide. The public, agriculture and the food industry were advised to lower the
abundance in foods, and the dietary intakes of, saturated fats. Into this oppor-
tunity, partially hydrogenated oils came running. Previously perceived as a
poor substitute for edible fats, the lower proportions of saturated fatty acids in
the hydrogenated vegetable fats was marketed as a nutritional advantage. The
PROLOGUE xxi
lack of routine chemical methods at the time to distinguish cis double bonds
from trans double bonds was not considered an issue: they were simply
unsaturated. Remarkably, little human clinical research was pursued to chal-
lenge the assumption that these hydrogenated alternatives to saturated fats,
with or without trans fatty acids, did not adversely affect blood lipids. Marga-
rines and shortenings were assumed to be healthier alternatives to animal fats
simply because they were not saturated. As a result of this perceived nutritional
advantage of hydrogenated vegetable oils, their health advantages were ag-
gressively marketed by producers and recommended by health agencies and
there was a rapid growth in the use of vegetable shortenings and margarines at
the expense of saturated fat commodities.
The scientific and public health perception that the presence of trans fatty
acids in the diet provided a beneficial, or at worst neutral, contribution to blood
lipoprotein cholesterol was changed dramatically by a single study published
by Mensink and Katan (1990). These investigators for the first time had access
to a source of trans fatty acids at relatively high concentration and could
examine trans fats as an independent experimental variable in a large human
clinical study. This first report of a well-designed and well-executed human
clinical study examining trans fatty acids as a single, independent variable
while measuring both low-density lipoprotein (LDL) and high-density lipopro-
tein (HDL) cholesterol in blood as discrete end points was striking. It became
one of the most public and widely read scientific papers on food ever published.
Results were contrary to the assumption that trans fats were harmless. In fact,
Mensink and Katan documented that the consumption of trans fatty acids by
normal humans raised blood levels of LDL and lowered HDL, both deleterious
actions in terms of the risk of heart disease. This landmark study was made
possible by the availability of experimental oils containing high concentrations
of trans fatty acids. An interesting historical note is that this experimental
material, provided by , was manufactured not by hydrogenation, but
by an isomerization reaction to ensure that the levels of trans isomers were
sufficiently high to test their biological effects.
The publication of a scientific study documenting the deleterious conse-
quences of trans fatty acids for risk factors of heart disease was met with some
debate from the various organizations associated with edible oils and fats. Not
surprisingly, nutrition and health scientists called for the removal of trans fatty
acids from foods. However, the relevance of the study to the normal human
population was questioned due to the high content of trans fatty acids in the
experimental diets in the Mensink and Katan study. Debate continued until the
1992 study by Zock and Katan that repeated the same basic clinical trial
protocols of the previous work, but included trans fatty acids in the diet at
approximately half the concentration of the previous study. This new study,
again using normal men and women, reported that the effects of a lower trans
fatty acid intake were consistent with a linear dose effect of trans fatty acids on
xxii PROLOGUE
raising LDL and lowering HDL cholesterol levels in blood. With these con-
vincing studies establishing that trans fatty acids were clearly not producing
metabolic effects similar to those of unsaturated fatty acids, considerable
scrutiny began to focus on whether they were truly deleterious to health as the
initial studies implied.
Various epidemiologic databases were rapidly polled to determine whether
or not the predictions that trans intakes would be deleterious to risk of heart
disease were consistent with human health. Indeed, trans intakes as an epide-
miological variable appeared to coincide not with a decrease in heart disease
risk but with an increase as predicted from the now known consequences to
lipoprotein concentrations. Unfortunately, to date no prospective clinical trial
has yet been conducted that could define unequivocally the effects of trans fats
alone. Because the use of these fats commercially is not simply to achieve a
health effect but also to obtain functionality in a wide variety of food products
from baked goods to fried foods, it remains practically impossible to distin-
guish trans fats in the diet as a truly independent variable. Those who have high
trans fat intakes also have distinct dietary habits in many ways. Nonetheless,
although definitive scientific evidence was lacking, and in fact may be impos-
sible to obtain in a normal human population, regulatory agencies and public
health agencies worldwide began to act.
The tale of trans fats is not simply related to chemical hydrogenation.
Scientists examining the determinants of fat compositions in animals had
discovered that one of the interesting microbial effects of rumen digestion and
fermentation in ruminant animals was the biological hydrogenation of polyun-
saturated fatty acids. Grazing ruminants consume a low fat intake but the fatty
acids in plants, especially leafy matter, are highly unsaturated. Why then are the
fat depots of these animals not polyunsaturated? The polyunsaturated fatty
acids are hydrogenated in the rumen prior to digestion and absorption by the
animal, hence the animal’s diet is effectively low in polyunsaturated fats. Even
more strikingly, a small proportion of the polyunsaturated fatty acids are
converted by microbial fermentation into the trans isomer of oleic acid,
vaccenic acid (trans 18:1n–11). This observation would have remained an
obscure fact of interest only to ruminant lipid biochemists except for one
completely unexpected finding. In studies attempting to identify potential pro-
carcinogenic substances in processed foods, Michael Pariza and colleagues
(Ha et al., 1989) made an astonishing discovery — an isomer of linoleic acid
(9c,12c, 18:2n–6) containing two double bonds but in the cis,trans conjugated
conformation, a substance found in foods from ruminants, was apparently anti-
carcinogenic in animals. In now hundreds of cellular and animal studies, the
molecule identified as conjugated linoleic acid CLA has shown significant
anti-carcinogenic properties. CLA is in fact a misnomer as the fatty acid is not
linoleic acid. As Dale Bauman and colleagues at Cornell University have
shown (Griinari et al., 2000), CLA is instead the biological product formed
PROLOGUE xxiii
when the rumen trans fatty acid vaccenic acid is converted to a polyunsaturated
acid 18:2 9c,11t, octadecadienoic acid, by the action of the Δ9 desaturase in
animals.
With these results emerging, a major dilemma began to shape up for
regulatory agencies. Not all trans fats are equal. The chemistry and chemical
composition of trans fats from industrial hydrogenation differ from those of
trans fats produced by biological hydrogenation in ruminants. Even more
perplexing, their health effects appear to be quite different as well. Discourag-
ingly for regulatory agencies, the complexity and small magnitude of effects on
health, combined with the invariably confounding co-occurrence of the differ-
ent trans fats with different dietary choices and even lifestyles, makes it
practically impossible to develop convincing scientific evidence of the inde-
pendent effects of the different trans isomers on actual human health outcomes.
The agencies must decide how to act in the absence of definitive scientific
evidence.
From the mid-1990s forward, trans fats became the subject of increasing
calls from scientists and public health officials for the regulation of their
content in foods and their inclusion on product labels. In 2003, the food
regulatory agency of Denmark banned the use of all hydrogenated fats from
food products, but at the same time made an explicit exception allowing the use
of animal fats containing natural trans fatty acids as these were viewed as
chemically different. In 2006, the USA’s mandatory labelling of all trans fats
irrespective of source came into effect. While it is too early to conclude
unequivocally, trans fats obtained by chemical hydrogenation are rapidly
disappearing from the industrial food supply. As for trans fats from ruminants,
the future is much more complex. It is not yet known whether the properties of
the different trans isomers of ruminants will have an important different effect
on the incidence of heart disease relative to chemically hydrogenated fats. Nor
is it yet known whether the remarkable anti-cancer properties of the trans
isomers in ruminant fats that have been observed in animals will translate into
similar benefits in lowering the risk of human cancers. Scientific research in
these areas will be important, and prior to its completion it is impossible to
know – time alone will tell.
The issue of trans fats has provided a controversy encompassing all of the
sciences related to diet and health. There are villains, heroes, wars and even
life and death. This book provides a unique opportunity to gain a detailed
and relatively comprehensive overview of the full spectrum of the issues of
trans fats from many of the scientists who have made the key discoveries
and who have brought the science to the level where it is today. The book is
very enjoyable reading for all those who appreciate science, mystery and
drama.
xxiv PROLOGUE
References
Mensink, RP, and Katan, MB (1990) Effect of dietary trans fatty acids on high-density and
low-density lipoprotein cholesterol levels in healthy subjects. New Eng. J. Med., 323, 439–
445.
Zock, P, and Katan, MB (1992) Hydrogenation alternatives: effects of trans fatty acids and
stearic acid versus linoleic acid on serum lipids and lipoproteins in humans. J. Lipid Res.,
33, 399–410.
Griinari, JM, Corl, BA, Lacy, SH, Chouinard, PY, Nurmela, KV and Bauman, DE (2000)
Conjugated linoleic acid is synthesized endogenously in lactating dairy cows by Delta(9)-
desaturase. J. Nutr., 130, 2285-2291.
Ha, YL, Grimm, NK and Pariza, MW (1989) Newly recognized anticarcinogenic fatty acids:
identification and quantification in natural and processed cheeses. J. Agric. Food Chem.,
37, 75–81.
CHAPTER 1
Biosynthesis of trans fatty acids in ruminants
1
Université de Toulouse, INPT, ENVT, UMR 1289 Tandem, Tissus Animaux,
Nutrition, Digestion, Ecosystème et Métabolisme, ENVT, Toulouse, France
2
INRA, UMR 1289 Tandem, Tissus Animaux, Nutrition, Digestion, Ecosystème
et Métabolisme, Auzeville, Castanet-Tolosan, France
A. Introduction
Meat and milk from ruminants constitute an important natural source of trans
fatty acids (TFA) for humans. These TFA are formed from the lipids contained
in ruminant diets. Common forages and concentrates used to feed ruminants
typically contain a moderate amount of fat (i.e. under 5% of dry matter). The
main fatty acids (FA) found in ruminant diets are palmitic (16:0), oleic
(c9-18:1), linoleic (c9,c12-18:2) and α-linolenic (c9,c12,c15-18:3) acids. The
main lipids found in forages are pigments, and FA, the latter being normally
present in galactolipids and phospholipids and representing about 50% of total
fat. In grains and seeds, lipids are present mainly as triacylglycerols. In silages
(Lee et al., 2006a) and crushed or extruded feedstuffs stored for a long time
(Dierick & Decuypere, 2002), the main lipids are free FA formed by hydrolysis
of acyl ester linkages. In grains and peas, the level of fat is lower than 5%
whereas in oilseeds it ranges from 20% (dry matter basis) for soybean to 45%
for canola (rapeseed) and sunflower. In forages, c9,c12,c15-18:3 acid is the
most important FA except in corn silage where c9,c12-18:2 acid is the most
prevalent FA. In oilseeds, the main FA found depends on species. Indeed, in
canola and some cultivars of sunflower the main FA is c9-18:1 acid and in other
seeds (e.g. sunflower, soybean), c9,c12-18:2 is the main FA. In linseed
(flaxseed), the main FA found is c9,c12,c15-18:3 acid. Edible oils such as palm
oil (rich in 16:0 and c9-18:1), fats from terrestrial animals (rich in 16:0, 18:0
and c9-18:1), or marine oils rich in eicosapentaenoic (c5,c8,c11,c14,c17-20:5)
and docosahexadecenoic (c4,c7,c10,c13,c16,c19-22:6) acids can be used as
pure fat-based feedstuffs.
In ruminants, dietary FA undergo rumen microbial digestion. The two main
steps are lipolysis and biohydrogenation of unsaturated FA (UFA). During
1
2 TRANS FATTY ACIDS IN HUMAN NUTRITION
lipolysis, free FA are released from acyl lipids and during biohydrogenation,
UFA are sequentially reduced to saturated FA, and the final end product is
stearic (18:0) acid. The number of biohydrogenation steps depends on the
initial structure of the UFA and some intermediate TFA are formed. Due to the
continuous rumen outflow, some biohydrogenation intermediates can be
absorbed in the small intestine, metabolized and deposited in tissues or
excreted into milk fat.
c9,c12,c15-18:3
c9,t11,c15-18:3
c9,c12-18:2
c13-18:1
c12-18:1
c11-18:1
c15-18:1
t10-18:1 interconversions t11-18:1
t4-; t5-; t6+7+8-; t12-; t1 3+14-; t15- ; t16- 18:1
c9-18:1 t9-18:1
18:0
Figure 1. Major biohydrogenation pathways of oleic (c9 18:1), linoleic (c9,c12 18:2) and α-linolenic
(c9,c12,c15-18:3) acids in ruminants.
c9-18:1 acid, than soybean oil, which contains mainly PUFA. A strong rela-
tionship has been observed between c9-18:1 supply and ruminal t10-18:1
formation (Loor et al., 2002a). These biohydrogenation pathways are also
consistent with the positive correlation between c9-18:1 intake and concentra-
tion of t7,c9-18:2 conjugated acid isomer in milk fat (Collomb et al., 2004).
The t7,c9-18:2 acid almost exclusively originates from mammary desaturation
of ruminally derived t7-18:1 acid (Piperova et al., 2002, Corl et al., 2002).
It has been shown that the c9-18:1 biohydrogenation isomeric profile is
affected by the mass of the 13C labelled c9-18:1 acid (Mosley et al., 2006a).
However differences were not highly significant and therefore do not make
previous results (Mosley et al., 2002; Proell et al.; 2002, and AbuGhazaleh
et al., 2005) questionable.
Despite the formation of biohydrogenation intermediates, the concentrations
of 18:0 are much higher than those of trans-18:1 when c9-18:1 is incubated
with ruminal microorganisms (Mosley et al., 2002). These observations sug-
gest either a rapid reduction of biohydrogenation trans intermediates, or that
reduction of c9-18:1 acid is mainly direct. The hypothesis of a mainly direct
reduction is probable since the ruminal biohydrogenation of cis isomers is more
rapid than those of trans isomers (Kemp et al., 1984).
In the rumen, c9-18:1 can also be hydrated to 10-hydroxystearic acid
(Hudson et al., 1995). It has been shown that hydroxystearic acid is almost
exclusively formed from c9-18:1 acid and it is further converted to 10-
ketostearic acid but not to TFA (Jenkins et al., 2006).
increase of c9,t11-18:2 acid (Loor et al., 2004). Other isomers have been
shown to be related to c9,c12,c15-18:3 acid biohydrogenation in vitro (Jouany
et al., 2007) and in milk (Kraft et al., 2003; Collomb et al., 2004). In the study
by Collomb and co-workers (2004), the most abundant intermediates found
were 12,14-18:2 and t11,c13-18:2 CLA isomers and they hypothesized that
these conjugated isomers are formed by isomerization of t11,c15-18:2 acid.
This isomerization has been recently demonstrated by Hino & Fukuda (2006).
In addition, other geometrical isomers such as c11,c13-, c11,t13-and t11,t13-
18:2 acids, have been identified in the duodenal or omasal content of cows or
in in vitro rumen cultures (Duckett et al., 2002; Piperova et al., 2002; Sackman
et al., 2003; Shingfield et al., 2003; Loor et al. 2004, 2005c; Jouany et al.,
2007). The t11,t13-18:2 CLA isomer is more abundant when the diet contains
linseed oil than sunflower oil (Loor et al, 2005c).
The second reduction step results in the formation of t11-18:1 and c15-18:1
acid isomers from t11,c15-18:2 acid (Figure 1). Vaccenic acid (t11-18:1) is also
formed during biohydrogenation of c9,c12-18:2 acid, and is usually the most
important trans-18:1 isomer produced in the rumen, independently of the fat
source. The c15-18:1 acid and its geometrical isomer t15-18:1 have been
reported in higher concentrations in the digesta of cows receiving linseed,
compared to cows receiving diets without fat (Loor et al., 2004) or diets with
sunflower oil or fish oil (Loor et al., 2005c). These observations are consistent
with the study performed by White and co-workers (1970) who showed in vitro
that incubation of c9,c12,c15-18:3 acid with rumen bacteria results in the
formation of large amounts of c15-18:1 acid, but also t15-, t13- and t14-18:1 acid
isomers. In lactating cows, high linseed diets also result in high proportions of
t12-, t16- and t13+t14-18:1 acid isomers, the latter being the most abundant
trans-18:1 following vaccenic acid (Loor et al., 2004; Loor et al., 2005c; Akraim
et al. 2006a). The separation of the t13- from t14-18:1 by gas-liquid chromatog-
raphy is very critical and it is difficult to identify further if there is a difference
between the production of t13 and t14-18:1 acids. It has been reported that 18:2
isomers having a t13 double bond are much more abundant in the rumen or the
duodenum than 18:2 acid isomers with a t14 double bond (Palmquist et al.,
2005), therefore it is probable that t13-18:1 acid isomer is produced at a higher
level than t14-18:1 acid isomer. However, White and co-workers (1970), using
labelled FA, observed that in vitro incubations of c9,c12,c15-18:3 produced four
times more t14-18:1 than t13-18:1 acid isomers.
Based on the data reported by Kemp and co-workers (1975), in the
biohydrogenation pathway proposed by Harfoot and Hazlewood (1997), c15-
18:1 and t15-18:1 acid are not further hydrogenated, but only limited data are
available to ascertain this fact. The kinetics in vitro study by Akraim and co-
workers (2006b) showed that all trans-18:1 isomers, including t15-18:1,
reached a maximal proportion of FA at 8 hours of incubation, and decreased
thereafter, which could be due to either a hydrogenation or an isomerization.
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 7
et al., 2008). Analytical problems such as coelution of dietary UFA with their
isomers can also result in biased estimations. In vitro measurements usually
result in slower biohydrogenation rates and lower biohydrogenation extents
than in vivo studies, with high variability among studies (Moate et al., 2004).
Lipolysis is rapid and in earlier in vitro studies lipolysis was considered to
happen completely (Garton et al., 1958). Later it was shown that between 50
and 70% of dietary galactolipids are hydrolysed within 1 hour after a meal in
sheep (Dawson et al., 1974), and then that about 5% of lipids entering the
duodenum of goats are mono-, di- or triaclylglycerols or phospholipids
(Bickerstaffe et al., 1972). Similarly, Bauchart and co-workers (1990a) showed
that more than 90% of dietary triacylglycerols disappear before the duodenum.
Overall, using results from published in vivo trials, Moate et al. (2004) esti-
mated that more than 80% of lipids are hydrolysed in the rumen.
In vitro data using incubation of triacylglycerols do not suggest that the rate
of lipolysis can limit the biohydrogenation rate (Hawke & Silcock, 1970; Beam
et al., 2000). Similarly, based on eight experiments, Moate and co-workers
(2004) demonstrated that lipolysis rates were in most cases higher than
Observations 13 12 6 4 20 15
Biohydrogenation extent, %
c9-18:1*** 56.9 73.1 80.9 69.8 67.7 69.4
c9,c12-18:2 83.5 90.2 86.4 88.8 89.2 84.7
c9,c12,c15-18:3 89.2 90.6 92.0 92.0 92.2 89.1
Biohydrogenation completion, %
trans-18:2/ 1.27 0.87 8.91 8.16 2.51 4.59
disappeared PUFA
trans-18:1/ 15.9 17.3 31.7 55.5 22.2 27.7
disappeared UFA
appeared 18:0/ 86.5 55.4 46.6 49.2 68.9 64.8
disappeared UFA
*Biohydrogenation measured using fatty acid profiles. **Either sole fat source or associated with
another fat source. ***Values from Lee et al. (2006a), which were unexpectedly low due to a diet with
a low concentration of c9-18:1, were not taken into account. PUFA, polyunsaturated fatty acids; UFA,
unsaturated fatty acids.
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 9
Table 2. Average percentages of main trans fatty acids (g/100 g of total fatty acids (number of observations)) in the duodenal or abomasal
flow, reported by Duckett et al. (2002), Piperova et al. (2002), Sackmann et al. (2003), Shingfied et al. (2003), Loor et al. (2004), Lundy et
al. (2004), Lee et al. (2005), Loor et al. (2005c), Akraim et al. (2006a) and Lee et al. (2006a).
t6+t7+t8-18:1 0.35(12) 1.16 (2) 0.87 (6) 1.16 (4) 0.57 (17) 0.97 (7)
t9-18:1 0.25 (13) 0.64 (12) 0.55 (6) 0.80 (4) 0.39 (20) 0.64 (15)
t10-18:1 0.86 (13) 8.35 (12) 2.89 (3) 4.98 (4) 1.53 (17) 7.60 (15)
t11-18:1 3.90 (13) 5.22 (12) 10.44 (3) 16.85 (4) 7.96 (17) 5.11 (15)
t12-18:1 0.50 (13) 1.04 (13) 1.01 (6) 1.50 (4) 0.89 (20) 0.88 (15)
t13+t14-18:1 1.16 (12) 2.41 (11) 3.76 (6) 2.49 (2) 1.76 (14) 2.76 (7)
t15-18:1 0.71 (12) 0.94 (2) 1.82 (6) 1.09 (4) 1.02 (17) 1.19 (7)
c9,t11-18:2 0.112 (9) 0.093 (9) 0.154 (6) 0.244 (4) 0.180 (15) 0.080 (13)
t9,t11-18:2 0.034 (10) 0.034 (6) 0.122 (3) 0.068 (12) 0.013 (7)
t10,c12-18:2 0.061 (13) 0.114 (9) 0.016 (3) 0.059 (4) 0.072 (14) 0.073 (15)
c11,t13-18:2 0.009 (7) 0.017 (8) 0.011 (3) 0.050 (1) 0.013 (4) 0.016 (15)
t11,t13-18:2 0.038 (7) 0.021 (6) 0.160 (3) 0.059 (2) 0.063 (6) 0.051 (12)
t11,c15-18:2 0.41 (3) 0.86 (1) 2.97 (6) 3.30 (2) 1.68 (7) 2.95 (5)
c9,t11,c15-18:3 0.163 (3) 0.0163 (3)
TRANS FATTY ACIDS IN HUMAN NUTRITION
lost (Van de Vossenberg & Joblin, 2003) and further studies of their character-
istics are not possible. These results have been extensively reviewed by
Harfoot and Hazlewood (1997). During the last 10 years, some other strains
have been isolated (Kim et al., 2002: Megasphaera elsdenii YJ-4; Van de
Vossenberg & Joblin, 2003: Butyrivibrio hungatei Su6; Fukuda et al., 2005:
B. fibrisolvens TH1; Fukuda et al., 2006a: B. fibrisolvens MDT-5; Fukuda
et al., 2006b: B. fibrisolvens MDT-10; Maia et al., 2007: Clostridium
proteoclasticum B316 and P-18; Wallace et al., 2007: Propionibacterium
acnes), and Paillard et al. (2007a) recently demonstrated the biohydrogenation
capacity of many Butyrivibrio-like bacteria.
Butyrivibrio fibrisolvens is the most extensively investigated bacterium
(Polan et al., 1964; Kepler et al., 1966). Some strains are also cellulolytic, but
their capacity to digest cellulose is limited (Halliwell & Bryant, 1963). Polan
and co-workers (1964) showed that Butyrivibrio fibrisolvens is able to hydro-
genate c9,c12-18:2 to 18:1 acid isomers but not to 18:0 acid, concluding that
another biohydrogenation system was necessary for complete reduction. Moreo-
ver, they demonstrated an enhanced biohydrogenation activity when
Butyrivibrio fibrisolvens was incubated with two species of rumen bacteria:
Megaspahaera elsdenii (old name Peptostreptococcus elsdenii) and
Selenomonas strain 233 (Polan et al., 1964). Butyrivibrio fibrisolvens cannot
metabolize c5,c8,c11,c14,c17-20:5 or c4,c7,c10,c13,c16,c19-22:6 acids
(Wasowska et al., 2006; Maia et al., 2007).
Kemp and Lander (1984) proposed the classification of hydrogenating
bacteria into two groups: group A bacteria, which hydrogenate c9,c12-18:2
and c9,c12,c15-18:3 mostly to t11-18:1 or t11,c15-18:2, but not 18:0, and
group B bacteria which hydrogenate c9,c12,c15-18:3 to t11,c15-18:2, t15- or
c15-18:1, and hydrogenate c9,c12-18:2, c9-18:1 and t11-18:1 to 18:0. Earlier
studies isolated only a few group B bacteria, including two bacteria of the
genus Fusocillus (Harfoot, 1978). The Butyrivibrio hungatei Su6 strain is able
to complete the biohydrogenation of both c9-,c12-18:2 and c9,c12,c15-18:3 to
18:0, and therefore does not completely fit with this classification of bacteria
into A and B groups (Van de Vossenberg & Joblin, 2003). Fusocillus spp. and
Butyrivibrio hungatei are phenotypically similar. Phylogenetic analysis based
on 16S rRNA indicates that Butyrivibrio hungatei clusters with Clostridium
proteoclasticum, a specific branch of the Butyrivibrio tree (Wallace et al.,
2006; Paillard et al., 2007a; Jenkins et al., 2008; Wallace, 2008). Clostridium
proteoclasticum population represents between 2 and 9% of the rumen
eubacterial community; variations can be due to diets and to a higher extent to
inter-individual variations (Paillard et al., 2007b).
Hydration of UFA in the rumen is due to facultative anaerobic bacteria, the
main type of which is Sreptococcus bovis, but several strains of Streptococcus,
Staphylococcus, Lactobacillus, Enterococcus, and Pediococcus can also cata-
lyse this reaction (Hudson et al., 2000).
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 13
6. Intestinal digestion
FA that leave the rumen can be absorbed in the small intestine. Some data have
been published on the intestinal digestibility of total 18:1 TFA, with a range
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 15
c9,c12-18:2
Promoted by
corn silage
HC and/or low pH Decreased by
HC x high LA hay vs grass or grass silage a
HC x high LA x adaptation fat protection a
monensin HC and/or low pH b
high LAc (in vitro)
t10,c12-18:2 c9,t11-18:2
Decreased by
high LA substratec (in vitro)
extrusion
t10-18:1 t11-18:1
Decreased by
HC and/or low pH b
high LA substratec (in vitro)
extrusion
fish oil
Figure 2. Major effects of diet on the three steps of rumen biohydrogenation of linoleic (c9,c12 18:2,
LA) acid and the ratio of t11- to t10 18:1 acid isomers (adapted from Troegeler-Meynadier et al., 2006a).
HC, high concentrate diet.
from 82 to 96%, both when measured in the small intestine (Enjalbert et al.,
1997) or in the small intestine plus the hindgut (Romo et al., 2000; Loor et al.,
2004; Loor et al., 2005c). This is similar or slightly greater than values
observed for c9-18:1 acid. Loor and co-workers (2004; 2005c) reported similar
digestibility levels for various 18:1 and 18:2 TFA isomers ranging from 32 to
100%, with the lowest values for c9,t11-18:2 acid. In a recent review, Glasser
and co-workers (2008) reported apparent digestibility levels of 82.4% (45
observations) and 44.8% (17 observations) for t11-18:1 and c9,t11-18:2 acids,
respectively.
It has been reported that desaturation of 18:0 to c9-18:1 acid can happen in the
intestinal mucosa (Bickerstaffe et al., 1972), but Mosley and co-workers (2006b)
failed to demonstrate intestinal conversion of t11-18:1 to c9,t11-18:2 acid.
steps. The outflow of TFA increases with the efficiencies of lipolysis and
isomerization of dietary UFA, and decreases when the efficiency of the last
reduction step increases. Moreover, dietary factors can modulate the isomeric
profile of TFA. These major effects of diet on rumen biohydrogenation are
summarized in Figure 2.
concentrate diet, Latham and co-workers (1972) and Gerson and co-workers
(1985) showed that the lipolysis and biohydrogenation of PUFA are slower
when the donor cows receive a high concentrate diet, and explained this effect
by the decreased number of Butyrivibrio spp.. The inhibition of lipolysis at low
pH increases with oil concentration (Van Nevel & Demeyer, 1996b). Overall,
these results indicate that both ruminal bacteria adapted to high concentrate
diets and ruminal bacteria adapted to a ruminal pH over 6.0 or lower have a low
ability to isomerize PUFA. However, Choi and co-workers (2005) showed only
minor effects of culture pH on biohydrogenation, but found that
biohydrogenation was more active with rumen bacteria from cows receiving a
high concentrate diet and having a 5.6 ruminal pH than with rumen bacteria
from cows receiving a low concentrate diet and having a 6.8 ruminal pH.
In vivo, the biohydrogenation extent of PUFA decreases with increasing
proportion of concentrate (Table 1). Sackmann and co-workers (2003) ob-
served, within a narrow range of concentrate (64 to 88%), a decrease of
biohydrogenation for c9,c12-18:2 but not for c9,c12,c15-18:3 acids. Similarly,
Lee and co-workers (2006b) observed when increasing concentrate level from
20 to 40% a decrease of c9,c12-18:2 biohydrogenation but not c9,c12,c15-
18:3; further increase of concentrate did not affect the extent of
biohydrogenation. This effect of high concentrate levels is prevented by
addition of sodium bicarbonate, which limits the drop of pH due to high
concentrate diets (Kalscheur et al., 1997a). However, sodium bicarbonate is
also known to increase ruminal outflow rate (Hart & Polan, 1984), and with
continuous in vitro cultures, a high outflow rate could abolish the effects of a
low pH on the biohydrogenation extent (Martin & Jenkins, 2002). The effect of
high concentrate diets could also result from a lack of large particles in the
rumen. Indeed, when increasing the size of particles in vitro from 0.1–0.4 mm
to 1–2 mm the rate of lipolysis increases by 25% and that of c9,c12-18:2
biohydrogenation by 60% (Gerson et al., 1988).
As opposed to PUFA, in most experiments, the biohydrogenation extent of
c9-18:1 was not significantly affected by high concentrate diets (Kalsheur
et al., 1997a; Loor et al., 2004) except when a high level of concentrates was
investigated (Kucuk et al., 2001). This difference between monounsaturated
FA and PUFA could be due to a difference in the effect of low pH on the
isomerase needed for the disappearance of PUFA and on the reductase which
accounts for most of c9-18:1 acid disappearance. Moreover, most hydrating
bacteria are lactic acid bacteria, which develop at low rumen pH, so that
hydration could be a major fate of UFA during ruminal acidosis (Hudson et al.,
2000). To our knowledge, this hypothesis has not yet been investigated, and
whether such a shift of FA metabolism toward hydration would affect c9-18:1
rather than c9,c12-18:2 acid is unknown.
The third effect of the diet is through its composition of nutrients. The
carbohydrate content of silages (Lee et al., 2006a) or grass (Scollan et al.,
18 TRANS FATTY ACIDS IN HUMAN NUTRITION
2003) has been shown not to affect the extent of biohydrogenation. However,
in these experiments, the diets contained only forage so that increasing sugars
could have failed to modify the pH or the equilibrium between ruminal
bacterial species. On the contrary, addition of sucrose to continuous ruminal
cultures at constant pH linearly decreases the biohydrogenation extent of FA
from alfalfa hay (Ribeiro et al., 2005). Sucrose addition also decreased fibre
digestibility and, therefore, the authors attributed this effect on biohydrogenation
to a decrease in cellulolytic microorganisms which have biohydrogenation
activities. Gerson and co-workers (1985) observed an increased c9,c12-18:2
biohydrogenation rate when sucrose was added to the culture medium. The
effect of soluble carbohydrates supplements may depend on adaptation of
bacteria or on the time pattern of addition because in vitro an effect of sucrose
on biohydrogenation rates was not observed (Ribeiro et al., 2007).
Gerson and co-workers (1983) demonstrated that both lipolysis and
biohydrogenation rates increase in vitro when the nitrogen content increases
from 0.72 to 2.5% of dry matter. Since, the latter concentration is in the range
of nitrogen concentrations in most ruminant diets, nitrogen is not likely to be a
limiting factor in practice.
comparisons across experiments do not show any clear trend (Table 1). The
relative concentrations of c9,c12-18:2, c5,c8,c11,c14,c17-20:5 and
c4,c7,c10,c13,c16,c19-22:6 acids could account for these discrepancies, due to
complex interactions between c9,c12-18:2 acid and fish oil fatty acids
(Wasowska et al., 2006).
Technological treatment of the fat source also affects biohydrogenation. The
fat sources are strongly modified in the rumen, but rumen microorganisms also
can be affected by fat addition, as early in vitro and in vivo studies by Brooks
et al. (1954) demonstrated. Due to these reciprocal actions of microorganisms
and fat, specific treatments have been proposed to limit the inhibitory effect of
added fat on microorganisms, and the biohydrogenation of their FA. Other
processes applied to fat sources can incidentally affect rumen biohydrogenation.
Encapsulation of fat sources in a formaldehyde treated protein matrix was
first investigated in Australia (Scott et al., 1971). Formaldehyde-treated pro-
teins are disrupted only in the abomasum, and giving sheep a
formaldehyde-treated mixture of soybean oil and casein increases the duodenal
flow of c9,c12-18:2 + c9,c12,c15-18:3 acids from 2.5 to 13.0 g/d (Clapperton,
1978). Similarly, when applied to a synthetic mixture of c9,t11-18:2 and
t10,c12-18:2 acids, this method of protection strongly increases their abomasal
flows (Gulati et al., 2000). The efficiency of the protection decreases with the
oil/casein ratio (Clapperton, 1980), and is low when formaldehyde treatment is
applied without casein to oilseeds either directly (Bitman et al., 1973) or after
a formic acid treatment (Sinclair et al., 2005b). However, due to the cost of
casein and/or regulatory limitations on the use of formaldehyde in cattle
feeding in some countries, this method is not commercially used.
Recently, whey-protein complexes have proved to be efficient to protect
soybean oil, thus increasing the proportions of PUFA and decreasing the
proportions of TFA biohydrogenation intermediates in milk fat and plasma
triacylglycerols (Carroll et al., 2006; Heguy et al., 2006).
The use of FA calcium salts (also referred to as calcium soaps) was
developed two decades ago (Jenkins & Palmquist, 1982, 1984) based on
previous work performed by Davison & Woods (1963). The main purpose of
using FA Ca salts was to prevent the negative effect of FA on rumen microor-
ganisms. The protection of FA by Ca salts against biohydrogenation is moderate
(Wu et al., 1991; Wu & Palmquist, 1991; Ferlay et al., 1992; Enjalbert et al.,
1994, 1997) or absent (Fotouhi & Jenkins, 1992; Ferlay et al., 1993; Harvatine
& Allen, 2006). The use of CLA Ca salts instead of formaldehyde-protected
CLA results in much lower increase of milk CLA (De Veth et al., 2005). FA Ca
salts need to be dissociated prior to biohydrogenation. Sukhija and Palmquist
(1990) showed in vitro that dissociation of FA Ca salts can occur, with an extent
that increases when pH decreases and when the unsaturation level increases. At
pH 6.0, the percentage dissociation was around 50% for soybean oil FA and
20% for palm oil FA (Sukhija & Palmquist, 1990). Van Nevel and Demeyer
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 21
vivo (Gonthier et al., 2004). The reasons for the discrepancy between different
investigations remain unclear. The extrusion temperature has only minor
effects on the c9,c12-18:2 acid disappearance (Chouinard et al., 1997b), but
preconditioning and particle size seem to interact with the effects of extrusion
(Akraim et al., 2006a and 2006b). Particle size of fat-containing seeds has
minor effects on rumen biohydrogenation (Tice et al., 1994).
c. Use of additives
Some antimicrobials can inhibit lipolysis, specially ionophores and amoxicillin,
but have very limited effects on the disappearance of PUFA (Van Nevel &
Demeyer, 1995). Amoxicillin has a broad antibiotic spectrum, but ionophores
inhibit only gram-positive bacteria, therefore ionophores cannot inhibit lypolitic
activities of Anaerovibrio spp. (gram-negative bacteria). Recent data suggest
that in high-starch diets, monensin does not suppress classical gram-positive
bacteria but affects Megasphaera elsdenii and Butyrivibrio fibrisolvens (Weimer
et al., 2008). The effect of different ionophores could differ because monensin
results in a lower biohydrogenation rate than lasalocid (Martineau et al., 2008).
Treatment of linseed with quebracho condensed tannin reduced
biohydrogenation of c9,c12,c15-18:3 acid, but did not result in an improve-
ment of c9,c12,c15-18:3 transfer in the plasma lipids of steers (Kronberg et al.,
2007).
trate diets, amount of added fat, type of fat, technological treatment and quality
of the fat source, and additives.
An example of inter-individual variation is the large variations of milk CLA
content that have been observed between animals receiving the same diet
within the same herd. Peterson and co-workers (2002) showed that individual
animals were consistent over time when the diet remained unchanged, and that
the hierarchy of cows for milk c9,t11-18:2 production levels remained un-
changed when the diet was modified, suggesting that animals differ in both
their rumen outflow of TFA and their ability to produce CLA via mammary
desaturation.
High concentrate diets affect the ratios. In vitro, a low ruminal pH results in
an inhibition of the reduction of trans-18:1 to 18:0 acids, resulting in higher
concentrations of trans-8:1 isomers (Troegeler-Meynadier et al., 2006a). Simi-
larly, increasing in vivo the percentage of concentrate increases the
disappearance of PUFA in the rumen and the total 18:1 TFA isomers in the
duodenal flow (Piperova et al., 2002; Loor et al., 2004). Piperova and co-
workers (2002) showed that a high concentrate diet also increases CLA level in
the duodenum, suggesting an inhibition of the CLA reduction, which could not
be shown in vitro (Troegeler-Meynadier et al., 2006a). These effects of a high
concentrate diet can be prevented by dietary supply of sodium bicarbonate
(Piperova et al., 2002).
The amount of added fat influences ratios of the acids and, due to the interest
of CLA for nutritional applications, the formation of c9,c12-18:2 acid
biohydrogenation intermediates has been extensively investigated. In vitro,
increasing the initial concentration of c9,c12-18:2 increases the proportions of
conjugated 18:2 or trans-18:1 acid isomers but limits the formation of 18:0 acid
(Harfoot et al., 1973b; Troegeler-Meynadier et al., 2003). This modulation in
the formation of c9,c12-18:2 acid biohydrogenation intermediates is due to a
lowered reduction of conjugated 18:2 to trans-18:1 acid isomers and of trans
18:1 to 18:0. It has been shown that increased concentrations of CLA can
inhibit the growth of Butyrivibrio fibrisolvens cultures (Kim et al., 2000). An
effect on the enzyme activities catalysing the reduction of CLA to trans-18:1
isomers can explain the effect of initial c9,c12-18:2 acid concentration
(Troegeler-Meynadier et al., 2006a; Moate et al., 2008).
In vitro, high concentrations of c9,c12-18:2 acid inhibit trans-18:1 reduc-
tion. Polan and co-workers (1964) indicated that the reduction of trans-18:1
acid isomers only begins when their concentration is higher than c9,c12-18:2
acid concentration. However, Harfoot and co-workers (1973b) argued that the
inhibition due to a high initial concentration of c9,c12-18:2 is irreversible. The
inhibition threshold was determined to be about 1 mg of c9,c12-18:2/ml of
rumen contents (Harfoot et al., 1973b). More recently, Troegeler-Meynadier
and co-workers (2003), observed that a high initial concentration of c9,c12-
18:2 acid effectively decreases the 18:0/trans-18:1 ratio, but results in a linear
24 TRANS FATTY ACIDS IN HUMAN NUTRITION
increase over time of 18:0 acid. The rate of 18:0 acid increase was found to be
higher with high initial c9,c12-18:2 acid concentration (1.87 mg/ml) than with
low concentration (0.62 mg/ml) (Troegeler-Meynadier et al., 2003). These
authors hypothesized that there is a maximal rate for the reduction of trans-18:1
acids, therefore a high production of trans-18:1 acids results in an accumula-
tion. (Troegeler-Meynadier et al., 2003). Using in vitro kinetics, Moate and
co-workers (2008) observed that t11-18:1 acid biohydrogenation is inhibited
when its concentration is over 0.5 mg/ml. A direct inhibition of t11-18:1 acid
on its own reduction can explain why the reduction of t11-18:1 acid is limited
even after disappearance of CLA (Moate et al., 2008), and explain the apparent
irreversibility of the inhibition stated by Harfoot (1973b). CLA could also
inhibit the reduction of trans-18:1 acids (Troegeler-Meynadier et al., 2006a).
In vivo, the effect of graded supplies of c9,c12-18:2 acid have not been
extensively studied: Sackmann and co-workers (2003) observed that increas-
ing sunflower oil supply to steers does not increase the duodenal proportion of
total trans-18:2 and trans-18:1 acid isomers despite an increased c9,c12-18:2
acid disappearance in the rumen. In dairy cows, Shingfield and co-workers
(2008) showed that increasing dietary sunflower oil increased the concentra-
tion of both trans-18:1 and CLA in omasal FA flow. Modelling experiments
obtained from several in vivo data showed that the biohydrogenation rate of
trans-18:1 isomers is negatively affected by the concentration of free FA in the
rumen (Moate et al., 2004).
Adaptation of cows to fat supplementation can modify the completion of
c9,c12-18:2 acid biohydrogenation. Indeed, the concentration of milk CLA
decreases over time when oil supplements are given (Bauman et al., 2000;
Chilliard & Ferlay, 2004). Palmquist and co-workers (2005) related this
decrease of milk CLA content to the positive effect of adaptation on the
tolerance of Butyrivibrio fibrisolvens to c9,c12-18:2 acid observed earlier
(Kim et al., 2000). However, Roy et al. (2006) showed that total milk trans-
18:1 acid isomers do not decrease over time when fat is added to the diet. This
suggests that the decrease of milk CLA is not due to an increased efficiency of
the reduction reactions but to the t11 to t10 shift (see later).
The effect of fat type is illustrated by the body of evidence showing that
dietary fish oil modifies ruminal biohydrogenation (see Tables 1 and 2). The
main reported effect is a large increase (about 4 fold) in the rumen outflow of
trans-18:1 acid isomers (Shingfield et al., 2003, Wonsil et al., 1994). A dietary
supplementation with 2.5% of fish oil results in an enhanced formation of
trans-18:1 acid isomers in the rumen as observed with 5% of sunflower oil
(Loor et al., 2005b and 2005c). In steers supplemented with a constant amount
of sunflower oil and gradual addition of fish oil, increases of duodenal flows of
trans-18:1 acid isomers were observed (Lee et al., 2005). The effects of fish oil
are isomer specific, t10-18:1 and t11-18:1 acids showing the greatest increases
(Shingfield et al., 2003; Lee et al., 2005). Fish oil addition to a diet without
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 25
16
linseed
14
12
10
% of C18 FA
0
0 6 12 18 24
Incubation time, h
16
canola
14
12
10
% of C18 FA
0 6 12 18 24
Incubation time, h
Figure 3. In vitro synthesis of 9c,11t 18:2 () and t10+t11 18:1 (¸) during incubation of raw (——)
or extruded (- - - - -) linseed or canola (adapted from Enjalbert et al., 2003 and Akraim et al., 2006b).
that the effects are of limited extent and depended on animal breeds and on the
addition of dietary fat. Therefore further research is needed to better understand
the effect of copper supplementation.
Recently, Fukuda and co-workers (2006b) showed in vitro that Butyrivibrio
fibrisolvens MDT-10 strain multiplies by four the concentration of t11-18:1
acid isomers when added to mixed ruminal cultures. Simultaneous addition of
Bifidobacterium adolescentis HF-11 strain, which has a high capacity to
incorporate t11-18:1 acid, further increases t11-18:1 acid level in the cultures
(Fukuda et al.,2006b). These authors concluded that these strains, used as
dietary additives, could possibly be used to increase the amount of t11-18:1
acid absorbed from the small intestine of ruminants (Fukuda et al.,2006b).
30 30
25
25 t11-C18:1
t11-C18:1
t10-C18:1 t10-C18:1
20
20
% of FA
% of FA
15 15
10 10
5 5
0 0
C Cu4 0
40
88u2
4
C C i0 i3
C i0 3
6488
Li M0
M 25
L Li0
4 5
7564
S Li
Su Su
L Li
C C7
L L
0
C C i3
3
C
C2
S
C
M
C
35
35 35
65 65
35 65
2
M
C
C
C
65
C40 and C75: 40 and 75% of concentrate, duodenal flow (Piperova et al., 2002);
C64 and C88: 64 and 88% of concentrate, duodenal flow (Sackmann et al., 2003);
Su2 and Su4: 2 and 4% of added sunflower oil (Sackmann et al., 2003); C35, C65,
Li0 and Li3: 35 and 65% of concentrate, 0 and 3% of added linseed oil, duodenal
flow (Loor et al., 2004); M0 and M25: 0 or 25 mg of monensin /kg of dry matter,
in vitro continuous culture, (Jenkins et al., 2003).
Figure 4. Effects of the level of concentrate (% of dry matter), fat addition and monensin on t10- and
t11 18:1 acid levels (% of total FA) in the duodenal flow or in in vitro cultures. C40 and C75: 40 and 75%
of concentrate, duodenal flow, Piperova et al. (2002). C64 and C88: 64 and 88% of concentrate, duodenal
flow, Sackmann et al. (2003). Su2 and Su4: 2 and 4% of added sunflower oil, Sackmann et al. (2003).
C35, C65, Li0 and Li3: 35 and 65% of concentrate, 0 and 3% of added linseed oil, duodenal flow, Loor
et al. (2004). M0 and M25: 0 or 25 mg of monensin per kg of dry matter, in vitro continuous cultures,
Jenkins et al. (2003).
2002; Sackmann et al., 2003) but not in the experiment of Loor and co-workers
(2004). The same effect has been observed in ewes (Kucuk et al., 2001). The
effect of high concentrate diets on t10-18:1 and t10,c12-18:2 acids production
can be alleviated by the addition of sodium bicarbonate to the diet, suggesting
a pH dependent effect (Piperova et al., 2002).
It has been shown in vitro, using the same donor cow, that lowering the pH
of rumen fluid cultures only slightly modifies the t10/t11 ratio (Troegeler-
Meynadier et al., 2003), which contrasts with the effect observed in vivo after
adaptation of cows. However, it has been shown that whatever the pH, in vitro
cultures contained more t10,c12-18:2 when the donor cow received a high
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 29
concentrate diet (Choi et al., 2005). These results suggest that the effect of
concentrate is mediated by a modification of the ruminal ecosystem, but not by
a modification of enzyme activities. Dietary starch with a high degradation rate
results in a higher level of t10-18:1 acid. This has been demonstrated in dairy
cows switched from dry ground corn to high moisture corn (Bradford & Allen,
2004) or from potatoes to wheat (Jurjanz et al., 2004).
Dietary addition of linseed oil results in lower duodenal flows of t10-18:1
acid than addition of soybean oil (Loor et al., 2004) or sunflower oil (Loor et
al, 2005c), and graded increase of dietary sunflower oil increases t10 acid
isomers (Sackmann et al., 2003; Shingfield et al., 2008), confirming that
c9,c12-18:2 acid is the main precursor of the t10-18:1 and t10,c12-18:2 acid
isomers (Figure 1).
The ruminal production of t10 isomers is higher with c9,c12-18:2 acid in oils
than with c9,c12-18:2 acid seed sources (Duckett et al., 2002), which shows
that the availability of c9,c12-18:2 acid for bacteria can play an important role
in the t10 shift. Many experimental data have been obtained with oil supple-
mentation; however in practice oilseeds supplementation is more current than
oils supplementation. Limited data suggest that soybean amides increase
ruminal production of t10 isomers (Lundy et al., 2004).
The results related to the effect of fish oil on the production of t10 isomers are
conflicting. Shingfield and co-workers (2003) reported that fish oil addition
(1.6%) does not impact t10/t11 ratio. However, Loor and co-workers (2005c)
observed a t10/t11 ratio that was around 1 with either 2.5% of fish oil or 5% of
sunflower oil, compared to 0.3 with 5% of linseed oil suggesting that fish oil
can result in a high production of t10 isomers. These observations are consist-
ent with the higher t10/t11 ratio obtained with the addition of 3% fish oil
compared to 1% (Kim et al., 2008). In vitro, low pH or a low forage level result
in high concentrations of t10-18:1 acid when fish oil is added to the culture
(AbuGhazaleh & Jacobson, 2007a and 2007b).
It has been shown in lactating cows that the duration of c9,c12-18:2
supplementation affects the t10/t11 ratio. It has been found that t11-18:1 and
c9,t11-18:2 acids reach a maximal concentration 4–7 days after the beginning
of oil supplementation, whereas t10-18:1 reaches a plateau from 10 to 20 days
of supplementation (Bauman et al., 2000; Roy et al., 2006; Shingfield et al.,
2006). Such an effect on t11/t10 ratio was not observed when linseed oil was
added to a grass hay-based diet (Roy et al., 2006). However, it has been
observed when linseed oil was added to a corn silage based diet (Pottier et al.,
2006).
The interaction between fat addition and diet composition has been seen to
influence the isomeric profile. In the experiment of Griinari et al. (1998),
switching from 50 to 20% of forage in the diet of lactating cows increased t10-
18:1 acid by 27% in milk fat when the diet contained saturated fat, but a 314%
increase was observed when the diet contained corn oil.
30 TRANS FATTY ACIDS IN HUMAN NUTRITION
Similarly, a high level of t10-18:1 acid isomer (> 10% of total FA) has been
observed in the duodenal flow when high concentrate diets are associated with
unsaturated fat addition (Sackmann et al., 2003; Loor et al., 2005c; Lundy
et al., 2004). However, experiments investigating the interaction of concen-
trate level and oil addition on TFA duodenal flow failed to demonstrate such
interaction (Sackmann et al., 2003, Loor et al., 2004).
Additives may affect isomeric profile and Fellner and co-workers (1997)
demonstrated that ionophores increased, in vitro, the production of trans-18:1
isomers, with or without addition of c9,c12-18:2 acid. It was shown later that
this modulation is mainly due to increase in the formation of t10-18:1 acid
isomer (Jenkins et al., 2003). In this study, complex interactions were reported:
monensin increased the effect of soybean oil on t10-18:1 acid production when
the starch source was barley but not corn (Jenkins et al., 2003). This observa-
tion suggests that ionophore can impact fermentation of starch (Jenkins et al.,
2003). These biohydrogenation changes affect milk fat composition, monensin
exacerbating the positive effects of sunflower oil on the proportions of t10-18:1
and t10,c12-18:2 acid (Bell et al., 2006; Cruz-Hernandez et al., 2006).
Pottier and co-workers (2006) showed that a supplementation of cows with
a high dosage of vitamin E prevents the shift toward the t10 pathway for at least
3 weeks but does not reverse the shift once it has occurred.
E. Conclusion
Ruminal biohydrogenation of unsaturated FA is responsible for the synthesis of
TFA, mainly t11-18:1 and t10-18:1 acid isomers and various positional iso-
mers. Current knowledge allows to channel quantitatively and qualitatively the
rumen TFA outflow by modifying forage and concentrate sources, and propor-
tion of concentrate. The level and the profile of the different fat sources alone
or in combination, or via the use of feed additives also affect the quality of TFA
produced during biohydrogenation. The most specific way of control is via fat
supplementation: most experiments have used addition of oil. Current know-
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 31
ledge on the effects of lipids supplied not as oil but as processed oilseeds on the
duodenal flow of TFA is limited.
The formation of a limited number of TFA isomers (i.e. t11, t10-18:1,
t10,c12- and c9,t11-18:2 acids) during biohydrogenation has been studied for
their nutritional properties. The biological properties of the various conjugated
and non-conjugated 18:2 acid isomers remain unknown and their formation
during biohydrogenation unclear.
References
AbuGhazaleh, AA and Jacobson, BN (2007a) The effect of pH and polyunsaturated C18 fatty
acid source on the production of vaccenic acid and conjugated linoleic acids in ruminal
cultures incubated with docosahexaenoic acid. Anim. Feed Sci. Technol., 136, 11–22.
AbuGhazaleh, AA and Jacobson, BN (2007b) Production of trans C18:1 and conjugated
linoleic acid in continuous culture fermenters fed diets containing fish oil and sunflower
oil with decreasing levels of forage. Animal, 1, 660–665.
AbuGhazaleh, AA and Jenkins, TC (2004a) Disappearance of docosahexaenoic and
eicosapentaenoic acids from cultures of mixed ruminal microorganisms. J. Dairy Sci., 87,
645–651.
AbuGhazaleh, AA and Jenkins, TC (2004b) Short communication: docosahexaenoic acid
promotes vaccenic acid accumulation in mixed ruminal cultures when incubated with
linoleic acid. J. Dairy Sci., 87, 1047–1050.
AbuGhazaleh, AA, Riley, MB, Thies, EE and Jenkins, TC (2005) Dilution rate and pH effects
on the conversion of oleic acid to trans C18:1 positional isomers in continuous cultures.
J. Dairy Sci., 88, 4334–4341.
Akraim, F, Nicot, MC, Juaneda, P and Enjalbert, F (2007) Conjugated linolenic acid (CLnA),
conjugated linoleic acid (CLA), and other biohydrogenation intermediates in plasma and
milk fat of cows fed raw or extruded linseed. Animal, 1, 835–843.
Akraim, F, Nicot, MC, Weill, P and Enjalbert, F (2006a) Effects of preconditioning and
extrusion of linseed on the ruminal biohydrogenation of fatty acids. 1. In vivo studies.
Anim. Res., 55, 83–91.
Akraim, F, Nicot, MC, Weill, P and Enjalbert, F (2006b) Effects of preconditioning and
extrusion of linseed on the ruminal biohydrogenation of fatty acids. 2. In vitro and in situ
studies. Anim. Res., 55, 261–271.
Annison, EF, Linzell, JL, Fazakerley, S and Nichols, BW (1967) The oxidation and utilization
of palmitate, stearate, oleate and acetate by the mammary gland of the fed goat in relation
to their overall metabolism, and the role of plasma phospholipids and neutral lipids in
milk-fat synthesis. Biochem. J., 102, 637–647.
Ashes, JR, Siebert, BD, Gulati, SK, Cuthbertson, AZ and Scott, TW (1992) Incorporation of
n–3 fatty acids of fish oil into tissue and serum lipids or ruminants. Lipids, 27, 629–631.
Atkinson, RL, Scholljegerdes, EJ, Lake, SL, Nayigihugu, V, Hess, BW and Rule, DC (2006)
Site and extent of digestion, duodenal flow, and intestinal disappearance of total and
esterified fatty acids in sheep fed a high-concentrate diet supplemented with high-
linoleate safflower oil. J. Anim. Sci., 84, 387–396.
Bauchart, D, Legay-Carmier, F and Doreau, M (1990a) Ruminal hydrolysis of dietary
triglycerides in dairy cows fed lipid-supplemented diets. Reprod. Nutr. Dev., 30, suppl.,
187s.
Bauchart, D, Legay-Carmier, F, Doreau, M and Gaillard, B (1990b) Lipid metabolism of
liquid-associated and solid-adherent bacteria in rumen contents of dairy cows offered
lipid-supplemented diets. Br. J. Nut., 63, 563–578.
32 TRANS FATTY ACIDS IN HUMAN NUTRITION
Bauman, DE, Barbano, DM, Dwyer, DA and Griinari, JM (2000) Technical note: production
of butter with enhanced conjugated linoleic acid for use in biomedical studies with animal
models. J. Dairy Sci., 83, 2422–2425.
Bayourthe, C, Enjalbert, F and Moncoulon, R (2000) Effects of different forms of canola oil
fatty acids plus canola meal on milk composition and physical properties of butter.
J. Dairy Sci., 83, 690–696.
Beam, TM, Jenkins, TC, Moate, PJ, Kohn, RA and Palmquist, DL (2000) Effects of amount
and source of fat on the rates of lipolysis and biohydrogenation of fatty acids in ruminal
contents. J. Dairy Sci., 83, 2564–2573.
Bell, JA, Griinari, JM and Kennelly, JJ (2006) Effect of safflower oil, flaxseed oil, monensin
and vitamin E on concentration of conjugated linoleic acid in bovine milk fat. J. Dairy
Sci., 89, 733–748.
Bickerstaffe, R, Noakes, DE and Annison, EF (1972) Quantitative aspects of fatty acids
biohydrogenation, absorption and transfer into milk in the lactating goat, with special
reference to the cis- and trans-isomers of octadecenoate and linoleate. Biochem. J., 130,
607–617.
Bitman, J, Dryden, LP, Goering, HK, Wrenn, TR, Yoncoskie, RA and Edmondson, LF (1973)
Efficiency of transfer of polyunsaturated fats into milk. J. Amer. Oil Chem. Soc., 50, 93–
98.
Boeckaert, C, Fievez, V, Van Hecke, D, Verstraete, W and Boon, N (2007a) Changes in
rumen biohydrogenation intermediates and ciliate protozoa diversity after algae supple-
mentation to dairy cattle. Eur. J. Lipid Sci. Technol., 109, 767–777.
Boeckaert, C, Vlaeminck, B, Mestdagh, J and Fievez, V (2007b) In vitro examination of
DHA-edible micro algae. 1. Effect on rumen lipolysis and biohydrogenation of linoleic
and linolenic acids. Anim. Feed Sci. Technol., 136, 63–79.
Bradford, BJ and Allen, MS (2004) Milk fat responses to a change in diet fermentability vary
by production level in dairy cattle. J. Dairy Sci., 87, 3800–3807.
Brooks, CC, Garner, GB, Gehrke, CW, Muhrer, ME and Pfander, WH (1954) The effect of
added fat on the digestion of cellulose and protein by ovine rumen micoorganisms. J.
Anim. Sci., 13, 758–764.
Carroll, SM, DePeters, EJ and Rosenberg, M (2006) Efficacy of a novel whey protein gel
complex to increase the unsaturated fatty acids composition of bovine milk fat. J. Dairy
Sci., 89, 640–650.
Chilliard, Y and Ferlay, A (2004) Dietary lipids and forages interactions on cow and goat milk
fatty acid composition and sensory properties. Reprod. Nutr. Dev., 44, 467–492.
Choi, NJ, Imm, JY, Oh, S, Kim, BC, Hwang, HJ and Kim, YJ (2005) Effect of pH and oxygen
on conjugated linoleic acid (CLA) production by mixed rumen bacteria from cows fed
high concentrate and high forage diets. Anim. Feed Sci. Technol., 123–124, 643–653.
Chouinard, PY, Corneau, L, Butler, RW, Chilliard, Y, Drackley, JK and Bauman, DE (2001)
Effect of dietary lipid source on conjugated linoleic acid concentration in milk fat. J.
Dairy Sci., 84, 680–690.
Chouinard, PY, Girard, V, Brisson and GJ (1997a) Performance and profiles of milk fatty
acids of cows fed full fat, heat-treated soybeans using various processing methods. J.
Dairy Sci., 80, 334–342.
Chouinard, PY, Levesque, J, Girard, V and Brisson, GJ (1997b) Dietary soybeans extruded
at different temperatures: milk composition and in situ reactions. J. Dairy Sci., 80, 2913–
2924.
Chow, TT, Fievez, V, Moloney, AP, Raes, K, Demeyer, D and De Smet, S (2004) Effect of
fish oil on in vitro lipolysis, apparent biohydrogenation of linoleic and linolenic acids, and
accumulation of biohydrogenation intermediates. Anim. Feed Sci. Technol., 117, 1–12.
Clapperton, JL (1978) Biohydrogenation of protected soya-bean oil in sheep fed a low-fibre
diet. Proc. Nutr. Soc., 37, 65A.
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 33
Engle, TE, Spears, JW, Fellner, V and Odle, J (2000) Effects of soybean oil and dietary copper
on ruminal and tissue lipid metabolism in finishing steers. J. Anim. Sci., 78, 2713–2721.
Enjalbert, F (1995) Les lipides dans l’alimentation des ruminants. 2. Particularités de
l’utilisation digestive. Rev. Méd. Vét., 146, 383–392.
Enjalbert, F, Akraim, F, Juaneda, P, Troegeler-Meynadier, A and Nicot, MC (2006) Cis-15
intermediates of biohydrogenation in the duodenal flow of cows receiving linseed. Fourth
Euro Fed Lipid Congress, Madrid.
Enjalbert, F, Eynard, P, Nicot, MC, Troegeler-Meynadier, A, Bayourthe, C and Moncoulon,
R (2003) In vitro versus in situ ruminal biohydrogenation of unsaturated fatty acids from
a raw or extruded mixture of ground canola seed/canola meal. J. Dairy Sci., 86, 351–359.
Enjalbert, F, Nicot, MC, Bayourthe, C, Vernay, M and Moncoulon, R (1997) Effects of
dietary calcium soaps of unsaturated fatty acids on digestion, milk composition and
physical properties of butter. J. Dairy Res., 64, 181–195.
Enjalbert, F, Nicot, MC, Vernay, M, Moncoulon, R and Griess, D (1994) Effect of different
forms of polyunsaturated fatty acids on duodenal and serum fatty acids profiles in sheep.
Can. J. Anim. Sci., 74, 595–600
Fellner, V, Sauer, FD and Kramer, JKG (1997) Effect of nigericin, monensin, and tetronasin
on biohydrogenation in continuous flow-through ruminal fermenters. J. Dairy Sci., 80,
921–928.
Ferlay, A, Chabrot, J, Elmeddah, Y and Doreau, M (1993) Ruminal lipid balance and
intestinal digestion by dairy cows fed calcium salts of rapeseed oil fatty acids or rapeseed
oil. J. Anim. Sci., 71, 2237–2245.
Ferlay, A, Chilliard Y and Doreau M (1992) Effects of calcium salts differing in fatty acid
composition on duodenal and milk fatty acid profiles in dairy cows. J. Sci. Food Agric.,
60, 31–37.
Fievez, V, Vlaeminck, B, Jenkins, T, Enjalbert, F and Doreau, M (2007) Assessing rumen
biohydrogenation and its manipulation in vivo, in vitro and in situ. Eur. J. Lipid Sci.
Technol., 109, 740–756.
Fotouhi, N and Jenkins, TC (1992) Ruminal biohydrogenation of linoleoyl methionine and
calcium linoleate in sheep. J. Anim. Sci., 70, 3607–3614.
Franklin, ST, Martin, KR, Baer, RJ, Shingoethe, DJ and Hippen, A (1999) Dietary marine
algae (Schizochytrium sp.) increases concentrations of conjugated linoleic,
docosahexaenoic and transvaccenic acids in milk of dairy cows. J. Nutr., 129, 2048–
2052.
Fukuda, S, Furuya, H, Suzuki, Y, Asanuma, N and Hino, T (2005) A new strain of
Butyrivibrio fibrisolvens that has high ability to isomerize linoleic acid to conjugated
linoleic acid. J. Gen. Appl. Microbiol., 51, 105–113.
Fukuda, S, Suzuki, Y, Komiri, T, Kawamura, K, Asanuma, N and Hino, T (2006a)
Purification and gene sequencing of conjugated linoleic acid reductase from a
gastrointestinal bacterium, Butyrivibrio fibrisolvens. J. Appl. Microbiol., 103, 365–371.
Fukuda, S, Suzuki, Y, Murai, M, Asanuma, N and Hino, T (2006b) Isolation of a novel strain
of Butyrivibrio fibrisolvens that isomerizes linoleic acid to conjugated linoleic acid
without hydrogenation, and its utilization as a probiotic for animals. J. Appl. Microbiol.,
100, 787–794.
Fukuda, S, Suzuki, Y, Murai, M, Asanuma, N and Hino, T (2006c) Augmentation of
vaccenate production and suppression of vaccenate biohydrogenation in cultures of
mixed ruminal microbes. J. Dairy Sci., 89, 1043–1051.
Garton, GA, Hobson PN and Lough AK (1958) Lipolysis in the rumen. Nature, 182, 1511–
1512.
Gerson, T, John, A and King, ASD (1985) The effects of dietary starch and fibre on the in vitro
rates of lipolysis and biohydrogenation by sheep ruminal digesta. J. Agric. Sci. (Camb.),
105, 27–30.
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 35
Gerson, T, John, A and Sinclair, BR (1983) The effect of dietary N on in vitro lipolysis and
fatty acid hydrogenation in rumen digesta from sheep fed diets high in starch. J. Agric.
Sci. (Camb.), 101, 97–101.
Gerson, T, King, ASD, Kelly, KE and Kelly, WJ (1988) Influence of particle size and surface
area on in vitro rates of gas production, lipolysis of triglycerol and hydrogenation of
linoleic acid by sheep rumen digesta or Ruminococcus flavefaciens. J. Agric. Sci.
(Camb.), 110, 31–37.
Girard, V and Hawke, JC (1978) The role of holotrichs in the metabolism of dietary linoleic
acid in the rumen. Biochim. Biophys. Acta, 528, 17–27.
Glasser, F, Schmidely, P, Sauvant, D and Doreau, M (2008) Digestion of fatty acids in
ruminants: a meta-analysis of flows and variation factors: 2. C18 fatty acids. Animal, 2,
691–704.
Gonthier, C, Mustafa, AF, Berthiaume, R, Petit, HV and Ouellet, DR (2004) Feeding
micronized and extruded flaxseed to dairy cows: effects on digestion and ruminal
biohydrogenation of long-chain fatty acids. Can. J. Anim. Sci., 84, 705–711.
Gonthier, C, Mustafa, AF, Ouellet, DR, Chouinard, PY, Bertiaume, R and Petit, HV (2005)
Feeding micronized and extruded flaxseed to dairy cows: effects on blood parameters and
milk fatty acid composition. J. Dairy Sci,. 88, 748–756.
Griinari, JM and Bauman, DE (1999) Biosynthesis of conjugated linoleic acid. In: Ad-
vances in conjugated linoleic acid research, Vol.1 (MP Yurawecz, MM Mossoba,
JKG Kramer, MW Pariza and GJ Nelson, eds), AOCS Press, Champaign, Illinois,
USA, pp.180–200.
Griinari, JM, Dwyer, DA, McGuire, MA, Bauman, DE, Palmquist, DL and Nurmela, KV
(1998) Trans-octadecenoic acids and milk fat depression in lactating dairy cows. J. Dairy
Sci., 81, 1251–1261.
Gulati, SK, Ashes, JR and Scott, TW (1999) Hydrogenation of eicosapentaenoic and
docosahexaenoic acids and their incorporation into milk fat. Anim. Feed Sci. Technol.,
79, 57–64.
Gulati, SK, Kitessa, SM, Ashes, JR, Fleck, E, Byers, EB, Byers, YG and Scott, TW (2000)
Protection of conjugated linoleic acids from ruminal hydrogenation and their incorpora-
tion into milk fat. Anim. Feed Sci. Technol., 86, 139–148.
Halliwell, G and Bryant MP (1963) The cellulolytic activity of pure stains of bacteria from
the rumen of cattle. J. Gen. Microbiol., 32, 441–448.
Harfoot, CG (1978) Lipid metabolism in the rumen. Prog. Lipid Res., 17, 21–54.
Harfoot, CG and Hazlewood, GP (1997) Lipid metabolism in the rumen. In: The rumen
microbial ecosystem (PN Hobson and DS Stewart), Chapman and Hall, London, UK,
pp.382–426.
Harfoot, CG, Noble, RC and Moore, JH (1973a) Food particles as a site for biohydrogenation
of unsaturated fatty acids in the rumen (short communication). Biochem. J., 132, 829–
832.
Harfoot, CG, Noble, RC and Moore, JH (1973b) Factors influencing the extent of
biohydrogenation of linoleic acid by rumen micro-organisms in vitro. J. Sci. Food. Agric.,
24, 961–970.
Hart, SP and Polan, CE (1984) Effect of sodium bicarbonate and disodium phosphate on
animal performance, ruminal metabolism, and rate of passage in ruminating calves. J.
Dairy Sci., 67, 2356–2368.
Harvatine, KJ and Allen, MS (2006) Fat supplements affect fractional rates of ruminal fatty
acid biohydrogenation and passage in dairy cows. J. Nutr., 136, 677–685.
Hawke, JC and Silcock, WR (1969) Lipolysis and hydrogenation in the rumen. Biochem. J.,
112, 131–132.
Hawke, JC and Silcock, WR (1970) The in vitro rate of lipolysis and biohydrogenation in
rumen contents. Bioch. Bioph. Acta, 218, 201–212.
36 TRANS FATTY ACIDS IN HUMAN NUTRITION
Hazlewood, GP, Kemp, P, Lander, D and Dawson, RMC (1976) C18 unsaturated fatty acid
hydrogenation patterns of some rumen bacteria and their ability to hydrolyse exogenous
phospholipids. Br. J. Nutr., 35, 293–297.
Heguy, JM, Juchem, SO, DePeters, EJ, Rosenberg, M, Santos, JEP and Taylor, SJ (2006)
Whey protein gel composites of soybean and linseed oils as a dietary method to modify
the unsaturated fatty acid composition of milk lipids. Anim. Feed Sci. Technol., 131, 370–
388.
Henderson, C (1971) A study of the lipase produced by Anaerovibrio lipolytica, a rumen
bacterium. J. Gen. Microbiol., 65, 81–89.
Hino, T and Fukuda, S (2006) Biohydrogenation of linoleic and linolenic acids, and
production of their conjugated isomers by Butyrivibrio fibrisolvens. Fourth Euro Fed
Lipid Congress, Madrid.
Hobson, PN (1965) Continuous culture of some anaerobic and facultatively anaerobic rumen
bacteria. J. Gen. Microbiol., 38, 167–180.
Hudson, JA, Cai, Y, Corner, RJ, Morvan, B and Joblin, KN (2000) Identification and
enumeration of oleic acid and linoleic acid hydrating bacteria in the rumen of sheep and
cows. J. Appl. Microbiol., 88, 286–292.
Hudson, JA, MacKenzie, CAM and Joblin, KN (1995) Conversion of oleic acid to 10-
hydroxystearic acid by two species of ruminal bacteria. Appl. Microbiol. Biotechnol., 44,
1–6.
Hudson, JA, Morvan, B and Joblin, KN (1998) Hydration of linoleic acid by bacteria isolated
from ruminants. FEMS Microbiol. Letters, 169, 277–282.
Hughes, PE, Hunter, WJ and Toce, SB (1982) Biohydrogenation of unsaturated fatty acids.
Purification and properties of cis9,trans11 octadecadienoate reductase. J. Biol. Chem.,
257, 3643–3649.
Jenkins, TC (1997) Ruminal fermentation and nutrient digestion in sheep fed
hydroxyethylsoyamide. J. Anim. Sci., 75, 2277–2283.
Jenkins, TC (1998) Fatty acid composition of milk from Holstein cows fed oleamide or canola
oil. J. Dairy Sci., 81, 794–800.
Jenkins, TC, AbuGhazaleh, AA, Freeman, S and Thies, EJ (2006) The production of 10-
hydroxystearic and 10-ketostearic acids is an alternative route of oleic acid transformation
by the ruminal microbiota in cattle. J. Nutr., 136, 926–931.
Jenkins, TC and Adams, CS (2002) The biohydrogenation of linoleamide in vitro and its
effects on linoleic acid concentration in duodenal content of sheep. J. Anim. Sci., 80, 533–
540.
Jenkins, TC and Bridges, WC Jr (2007) Protection of fatty acids against ruminal
biohydrogenation in cattle. Eur. J. Lipid Sci. Technol., 109, 778–789.
Jenkins, TC, Fellner, V and McGuffey, RK (2003) Monensin by fat interaction on trans fatty
acids in cultures of mixed ruminal microorganisms grown in continuous fermentors fed
corn or barley. J. Dairy Sci., 86, 324–330.
Jenkins, TC and Palmquist, DL (1982) Effect of added fat and calcium on in vitro formation
of insoluble fatty acid soaps and cell wall digestibility. J. Anim. Sci., 55, 957–963.
Jenkins, TC and Palmquist, DL (1984) Effects of fatty acids or calcium soaps on rumen and
total nutrient digestibility of dairy rations. J. Dairy Sci., 67, 978–986.
Jenkins, TC and Thies, E (1997) Plasma fatty acids in sheep fed hydroxyethylsoyamide, a
fatty acid amide that resists biohydrogenation. Lipids, 32, 173–178.
Jenkins, TC, Thompson, CE and Bridges, WC Jr (2000) Site of administration and duration
of feeding oleamide to cattle on feed intake and ruminal fatty acid concentrations. J. Anim.
Sci., 78, 2745–2753.
Jenkins, TC, Wallace, RJ, Moate, PJ and Mosley, EE (2008) Board-invited review: Recent
advances in biohydrogenation of unsaturated fatty acids within the rumen microbial
ecosystem. J. Anim. Sci., 86, 397–412.
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 37
Jouany, JP, Lassalas, B, Doreau, M and Glasser, F (2007) Dynamic features of the rumen
metabolism of linoleic acid, linolenic acid and linseed oil measured in vitro. Lipids, 42,
351–360.
Jurjanz, S, Monteils, V, Juaneda, P and Laurent, F (2004) Variations of trans octadecenoic acid
in milk fat induced by feeding different starch based diets to cows. Lipids, 39, 19–24.
Kalscheur, KF, Teter, BB, Piperova, LS and Erdman, RA (1997a) Effect of dietary forage
concentration and buffer addition on duodenal flow of trans-C18:1 fatty acids and milk
fat production in dairy cows. J. Dairy Sci., 80, 2104–2114.
Kalscheur, KF, Teter, BB, Piperova, LS and Erdman, RA (1997b) Effect of fat source on
duodenal flow of trans-C18:1 fatty acids and milk fat production in dairy cows. J. Dairy
Sci., 80, 2115–2126.
Kay, JK, Roche, JR, Solver, ES, Thomson, NA and Baumgard, LH (2005) A comparison
between feeding systems (pasture and TMR) and the effect of vitamin E supplementation
on plasma and milk fatty acid profiles in dairy cows. J. Dairy Res., 72, 322–332.
Kemp, P, Dawson and RMC (1968) Isomerization of linolenic acid by rumen micro-
organisms. Biochem. J., 109, 477–478.
Kemp, P and Lander, DJ (1984) Hydrogenation in vitro of alpha-linolenic acid to stearic acid
by mixed cultures of pure strains of rumen bacteria. J. Gen. Microbiol., 130, 527–533.
Kemp, P, Lander, DJ and Gunstone, FD (1983) The biohydrogenation of gamma-linolenic
acid to stearic acid by pure cultures of two rumen bacteria. Biochem. J., 216, 519–522.
Kemp, P, Lander, DJ and Gunstone, FD (1984) The hydrogenation of some cis- and trans-
octadecenoic acids to stearic acid by a rumen Fusocillsu sp. Brit. J. Nutr., 52, 165–170.
Kemp, P, White, RR and Lander, DJ (1975) The hydrogenation of unsaturated fatty acids by
five bacterial isolates from the sheep rumen, including a new species. J. Gen. Microbiol.,
90, 100–114.
Kepler, CR, Hirons, KP, McNeill, JJ and Tove, SB (1966) Intermediates and products of the
biohydrogenation of linoleic acid by Butryrivibrio fibrisolvens. J. Biol. Chem., 241,
1350–1354.
Kepler, CR and Tove, SB (1967) Biohydrogenation of unsaturated fatty acids: purification
and properties of a linoleate Δ12 cis- Δ11 trans isomerase from Butyrivibrio fibrisolvens.
J. Biol. Chem., 242, 5686–5692.
Kepler, CR, Tucker, WP and Tove, SB (1970) Biohydrogenation of unsaturated fatty acids.
IV. Substrate specificity and inhibition of linoleate delta12-cis, delta11-trans isomerase
from Butyrivibrio fibrisolvens. J. Biol. Chem., 245, 3612–3620.
Kim, EJ, Huws, SA, Lee, MRF, Wood, JD, Muetzel, SM, Wallace, RJ and Scollan, ND (2008)
Fish oil increases the duodenal flow of long chain polyunsaturated fatty acids and trans-
11 18:1 and decreases 18:0 in steers via changes in the rumen bacterial community. J.
Nutr., 138, 889–896.
Kim, EJ, Sanderson, R, Dhanoa, MS and Dewhurst, RJ (2005) Fatty acids profiles associated
with microbial colonization of freshly ingested grass and rumen biohydrogenation. J.
Dairy Sci., 88, 3220–3230.
Kim, YJ, Liu, RH, Bond, DR and Russel, JB (2000) Effect of linoleic acid concentration on
conjugated linoleic acid production by Butyrivibrio fibrisolvens. Appl. Environ. Microbiol.,
66, 5226–5230.
Kim, YJ, Liu, RH, Rychlik, JL and Russel, JB (2002) The enrichment of a ruminal bacterium
(Megasphaera elsdenii YJ-4) that produces the trans-10, cis-12 isomer of conjugated
linoleic acid. J. Appl. Microbiol., 92, 976–982.
Kraft, J, Collomb, M, Mockel, P, Sieber and R,Jarheis, G (2003) Differences in CLA isomer
distribution of cow’s milk lipids. Lipids, 38, 657–664.
Kramer, JPG, Parodi, PW, Jensen, RG, Mossoba, MM, Yurawecz, MP and Adlof, RO (1998)
Rumenic acid: a proposed common name for the major conjugated linoleic acid isomer
found in natural products. Lipids, 33, 835.
38 TRANS FATTY ACIDS IN HUMAN NUTRITION
digestion of orchardgrass or red clover with three levels of ground corn supplementation.
J. Anim. Sci., 81, 1611–1627.
Loor, JJ, Ueda, K, Ferlay, A, Chilliard, Y and Doreau, M (2004) Biohydrogenation, duodenal
flow, and intestinal digestibility of trans fatty acids and conjugated linoleic acids in
response to dietary forage:concentrate ratio and linseed oil in dairy cows. J. Dairy Sci.,
87, 2472–2485.
Loor, JJ, Ueda, K, Ferlay, A, Chilliard, Y and Doreau, M (2005c) Intestinal flow and
digestibility of trans fatty acids and conjugated linoleic acids (CLA) in dairy cows fed a
high-concentrate diet supplemented with fish oil, linseed oil, and sunflower oil. Anim.
Feed Sci. Technol., 119, 203–225.
Lundy, FP, Block, E, Bridges, WC Jr, Bertrand, JA and Jenkins, TC (2004) Ruminal
biohydrogenation in Holstein cows fed soybean fatty acids as amides or calcium salts. J.
Dairy Sci., 87, 1038–1046.
Maia, MRG, Chaudhary, LC, Figueres, L and Wallace, RJ (2007) Metabolism of polyunsatu-
rated fatty acids and their toxicity to the microflora of the rumen. Antonie Van
Leeuwenhoek, 91, 303–314.
Martin, SA and Jenkins, TC (2002) Factors affecting conjugated linoleic acid and trans-
C18:1 fatty acid production by mixed ruminal bacteria. J. Anim. Sci., 80, 3347–3352.
Martineau, R, Petit, HV, Benchaar, C, Lapierre, H, Ouellet, DR, Pellerin, D and Berthiaume,
R (2008) Effects of lasalocid or monensin on in situ biohydrogenation of flaxseed and
sunflower seed unsaturated fatty acids. Can. J. Anim. Sci., 88, 335–339.
Moate, PJ, Boston, RC, Jenkins, TC and Lean, IJ (2008) Kinetics of ruminal lipolysis of
triacylglycerol and biohydrogenation of long-chain fatty acids: new insights from old
data. J. Dairy Sci., 91, 732–742.
Moate, PJ, Chalupa, W, Jenkins, TC and Boston, RC (2004) A model to describe ruminal
metabolism and intestinal absorption of long-chain fatty acids. Anim. Feed. Sci. Technol.,
112, 79–105.
Morales, MS, Palmquist, DL and Weiss, WP (2000) Effects of fat source and copper on
unsaturatuon of blood and milk triacylglycerol fatty acids in Holstein and Jersey cows.
J. Dairy Sci., 83, 2105–2111.
Morris, LJ (1970) Mechanisms and stereochemistry in fatty acid metabolism. Biochem. J.,
118, 681–693.
Morvan, B and Joblin, KN (2001) Ruminal bacteria and hydration of long-chain unsaturated
fatty acids. Reprod. Nutr. Dev., 40, 201 (Abstr.).
Mosley, EE, Nuda, A, Corato, A, Rossi, E, Jenkins, TC and McGuire, MA (2006a) Diffrential
biohydrogenation and isomerization of [U-13C]oleic and [1-13C]oleic acids by mixed
ruminal microbes. Lipids, 41, 513–517.
Mosley, EE, Powell, GL, Riley, MB and Jenkins, TC (2002) Microbial biohydrogenation of
oleic acid to trans isomers in vitro. J. Lipid Res., 43, 290–296.
Mosley, EE, Shafi, B, Moate, PJ and McGuire, MA (2006b) Cis-9, trans-11 conjugated
linoleic acid is synthesized directly from vaccenic acid in lactating dairy cattle. J. Nutr.,
135, 571–575.
Nam, IS and Garnsworthy, PC (2007) Biohydrogenation of linoleic acid by rumen fungi
compared with rumen bacteria. J. Appl. Microbiol., 103, 551–556.
Noble, RC, Moore, JH and Harfoot, CG (1974) Observations on the pattern of biohydrogenation
of esterified and unesterified linoleic acid in the rumen. Br. J. Nutr., 31, 99–108.
Ogawa, J, Kishino, S, Ando, A, Sugimoto, S, Mihara, K and Shimizu, S (2005) Production
of conjugated fatty acids by lactic acid bacteria. J. Bioscience Bioengeneering, 100, 355–
364.
Or-Rashid, MM, Odongo, NE and McBride, BW (2007) Fatty acid composition of ruminal
bacteria and protozoa, with emphasis on conjugated linoleic acid, vaccenic acid, and odd-
chain and branched-chain fatty acids. J. Dairy Sci., 85, 1228–1234.
40 TRANS FATTY ACIDS IN HUMAN NUTRITION
Paillard, D, McKain, N, Chaudhary, LCL, Walker, ND, Pizette, F, Koppova, I, McEwan, NR,
Kopecny, J, Vercoe, PE, Louis, P and Wallace, RJ (2007a) Relation between phylogenetic
position, lipid metabolism and butyrate production by different Butyrivibrio-like bacteria
from the rumen. Antonie Van Leeuwenhoek, 91, 417–422.
Paillard, D, McKain, N, Rincon, MT, Shingfield, KJ, Givens, DI and Wallace, RJ (2007b)
Quantification of ruminal Clostridium proteoclasticum by real-time PCR using a
molecular beacon approach. J. Applied Microbiol., 103, 1251–1261.
Palmquist, DL, Lock, AL, Shingfield, KJ and Bauman, DE (2005) Biosynthesis of conjugated
linoleic acid in ruminants and humans. In: Advances in food and nutrition research (S
Taylor, ed) Elsevier, London, UK, pp.179–218.
Park, SJ, Park, KA, Park, CW, Park, WS, Kim, JO and Ha, YL (1996) Purification and amino
acid sequence of the linoleate isomerase produced from Butyrivibrio fibrisolvens A38. J.
Food Sci. Nutr., 1, 244–251.
Parodi, PW (1976) Distribution of isomeric fatty acids in milk fat. J. Dairy Sci., 59, 1870–
1873.
Peterson, DG, Kelsey, JA and Bauman, DE (2002) Analysis of variations of cis-9, trans-11
conjugated linoleic acid (CLA) in milk fat of dairy cows. J. Dairy Sci., 85, 2164–2172.
Piperova, LS, Sampugna, J, Teter, BB, Kalscheur, KF, Yurawecz, MP, Ku, Y, Morehouse,
KM and Erdman, RA (2002) Duodenal and milk trans octadecenoic acid and conjugated
linoleic acid (CLA) isomers indicate that postabsorptive synthesis is the predominant
source of cis-9containing CLA in lactating dairy cows. J. Nutr., 132, 1235–1241.
Pires, AV, Eastridge, ML, Firkins, JL and Lin, YC (1997) Effects of heat treatment and
physical processing of cottonseed on nutrient digestibility and production performance
by lactating cows. J. Dairy Sci., 80, 1685–1694.
Polan, CE, McNeill, JJ and Tove, CB (1964) Biohydrogenation of unsaturated fatty acids by
rumen bacteria. J. Bacteriol., 88, 1056–1064.
Pottier, J, Focant, M, Debler, C, De Buysser, G, Goffe, C, Mignolet, E, Froidmond, E and
Larondelle, Y (2006) Effect of dietary vitamin E on rumen biohydrogenation pathways
and milk fat depression in dairy cows fed high fat diets. J. Dairy Sci., 89, 685–692.
Prins, RA, Lankhorst, A, Van Der Meer, P and Van Nevel, CJ (1975) Some characteristices of
Anaerovibrio lipolytica, a rumen lipolytic organism. Antonie Van Leeuwenhoek, 41,1–11.
Proell, JM, Mosley, EE, Powell, GL and Jenkins, TC (2002) Isomerization of stable
isotopically labeled elaidic acid to cis and trans monoenes by ruminal microbes. J. Lipid
Res., 43, 2072–2076.
Reddy, PV, Morril, JL, Nagaraja, TG (1994) Release of free fatty acids from raw or processed
soybeans and subsequent effects on fiber digestibilities. J. Dairy Sci., 77, 3410–3416.
Ribeiro, CVDM, Eastridge, ML, Firkins, JL, St-Pierre, NR and Palmquist, DL (2007)
Kinetics of fatty acid biohydrogenation in vitro. J. Dairy Sci., 90, 1405–1416.
Ribeiro, CVDM, Karnati, SKR and Eastridge, ML (2005). Biohydrogenation of fatty acids
and digestibility of fresh alfalfa or alfalfa hay plus sucrose in continuous cultures. J. Dairy
Sci., 88, 4007–4017.
Romo, GA, Erdman, RA, Teter, BB, Sampugna, J and Casper, DP (2000) Milk composition
and apparent digestibilities of dietary fatty acids in lactating dairy cows abomasally
infused with cis or trans fatty acids. J. Dairy Sci., 83, 2609–2619.
Roy, A, Ferlay, A, Shingfield, KJ and Chilliard, Y (2006) Examination of the persistency of
milk fatty acid composition responses to plant oils in cows fed different basal diets, with
particular emphasis on trans-C18:1 fatty acids and isomers of conjugated linoleic acid.
Anim. Sci., 82, 479–492.
Sackman, JR, Duckett, SK, Gillis, MH, Realini, CE, Parks, AH and Eggelston, RB (2003)
Effects of forage and sunflower oil levels on ruminal biohydrogenation of fatty acids and
conjugated linoleic acid formation in beef steers fed finishing diets. J. Anim. Sci., 81,
3174–3181.
BIOSYNTHESIS OF TRANS FATTY ACIDS IN RUMINANTS 41
Van Nevel, CJ and Demeyer, DI (1995) Lipolysis and biohydrogenation of soybean oil in the
rumen in vitro: inhibition by antimicrobials. J. Dairy Sci., 78, 2797–2806.
Van Nevel, CJ and Demeyer, DI (1996a) Effect of pH on biohydrogenation of polyunsatu-
rated datty acids and their Ca-salts by rumen microorganisms in vitro. Arch. Anim. Nutr.,
49, 151–157.
Van Nevel, CJ and Demeyer, DI (1996b) Influence of pH on lipolysis and biohydrogenation
of soybean oil by rumen contents in vitro. Reprod. Nutr. Dev., 36, 53–63.
Vázquez-Añón, M and Jenkins, TC (2007) Effects of feeding oxidized fat with or without
dietary antioxidants on nutrient digestibility, microbial nitrogen, and fatty acid metabo-
lism. J. Dairy Sci., 90, 4361–4367.
Vlaeminck, B, Mengitsu, G, Fievez, V, de Jonge, L and Dijkstra, J (2008) Effect of in vitro
docosahexaenoic acid supplementation to marine algae-adapted and unadapted rumen
inoculum on the biohydrogenation of unsaturated fatty acids in freeze-dried grass. J.
Dairy Sci., 91, 1122–1232.
Wallace, RJ (2008) Gut microbiology – broad genetic diversity, yet specific metabolic niches.
Animal, 2, 661–668.
Wallace, RJ, Chaudhary, LC, McKain, N, McEwan, NR, Richardson, AJ, Vercoe, PE,
Walker, ND and Paillard, D (2006) Clostridium proteoclasticum: a ruminal bacteria that
forms stearic acid from linoleic acid. FEMS Microbiol. Lett., 265, 195–201.
Wallace, RJ, McKain, N, Shingfield, KJ and Devillard, E (2007) Isomers of conjugated
linoleic acids are synthesized via different mechanisms in ruminal digesta and bacteria.
J. Lipid Res., 48, 2247–2254.
Ward, PFV, Scott, TW and Dawson, RMC (1964) The hydrogenation of unsaturated fatty
acids in the ovine digestive tract. Biochem. J., 92, 60–68.
Wasowska, I, Maia, MRG, Niedzwiedzka, KM, Czauderna, M, Ramalho-Ribeiro, JMC,
Devillard, E, Shingfield, KJ and Wallace, RJ (2006) Influence of fish oil on ruminal
biohydrogenation of C18 unsaturated fatty acids. Br. J. Nutr., 95, 1199–1211.
Weimer, PJ, Stevenson, DM, Mertens, DR and Thomas, EE (2008) Effect of monensin
feeding and withdrawal on populations of individual bacterial species in the rumen of
lactating dairy cows fed high-starch rations. Applied Microbial Cell Physiol., 80, 135–
145.
White, RW, Kemp, P and Dawson, MC (1970) Isolation of a rumen bacterium that
hydrogenates oleic acid as well as linoleic acid and linolenic acid. Biochem. J., 116, 767–
768.
Winkler, K and Steinhart, H (2001) Identification of conjugated isomers of linolenic acid and
arachidonic acid in cheese. J. Separation Sci., 24, 663–668.
Wonsil, WB, Herbein, JH and Watkins, BA (1994) Dietary and ruminally derived trans-18:1
fatty acids alter bovine milk lipids. J. Nutr., 124, 556–565.
Wu, Z, Ohajuruka, A and Palmquist, DL (1991) Ruminal synthesis, biohydrogenation, and
digestibility of fatty acids by dairy cows. J. Dairy Sci., 74, 3025–3034.
Wu, Z and Palmquist, DL (1991) Synthesis and biohydrogenation of fatty acids by ruminal
microorganisms in vitro. J. Dairy Sci., 74, 3035–3046.
Yáñez-Ruiz, DR, Scollan, ND, Merry, RJ and Newbold, CJ (2006) Contribution of rumen
protozoa to duodenal flow of nitrogen, conjugated linoleic acid and vaccenic acid in steers
fed silages differing in their water-soluble carbohydrate content. Br. J. Nutr., 96, 861–
869.
CHAPTER 2
Formation of trans fatty acids during catalytic
hydrogenation of edible oils
1
Nestlé Product Technology Center, Marysville, Ohio, USA
2
Carbougnères, St Eutrope-de-Born, France
In the early 20th century, with the increasing world demand for solid, edible
fats, especially for bakery fats, margarine and shortenings, much attention was
given to the development of alternatives for lard and tallow that maintained a
good quality during prolonged storage. In 1897, catalytic hydrogen addition to
the double bond of organic material was demonstrated by Sabatier and, based
on Normann’s patent (1903), the hydrogenation of edible oils was quickly
developed on an industrial scale in the UK (Hastert, 1998). Progressively
during the 20th century, this technology became one of the most important
processes in fats and oils transformation, allowing available fats and oils to be
modified in accordance with demands for specific physical properties (such as
melting point and consistency), with improved oxidative and thermal stability
at an acceptable price.
In comparison with most industrial hydrogenation processes, the hydrogena-
tion of vegetable oils (also known as fat hardening) is exceptional. The former
aim at completing the reduction or saturation reactions, but in the case of
triacylglycerols, the reaction aims at partial conversion. Triacylglycerol spe-
cies in the starting material can have one, two or three monounsaturated or
polyunsaturated fatty acids. During hydrogenation, triacylglycerol ethylenic
double bonds progressively disappear by being saturated by hydrogen. Before
disappearing, their positions can shift along the fatty acid chain (positional
isomerization) and/or their geometry can change from cis to trans configura-
tion and back leading to various trans fatty acids (TFA) in the final product.
The TFA levels (up to 80% of the ethylenic double bonds) and the isomeric
profile can vary tremendously according to operational condition and the
starting material.
It is worth noting that high temperatures applied during refining of oils can
also cause geometrical isomerization of ethylenic double bonds. In particular
during the deodorization step, vegetable oils are exposed to temperatures
ranging from 180 to 260ºC depending on the type of oil and process conditions.
43
44 TRANS FATTY ACIDS IN HUMAN NUTRITION
and content could be determined accurately. This greatly assisted the observa-
tion of what happens on a molecular scale during the hydrogenation reaction,
but did not reveal it in detail.
The literature tends to be rather vague about the maximum TFA content and
mentions, for example, a nickel sub sulphide catalyst (Baltes, 1970; 1972) that
causes close to 100% of the double bonds to have a trans configuration. This is
not correct. The cis/trans equilibrium is governed by the enthalpy difference of
ΔHiso = – 4 kJ /mol. Accordingly, its position is temperature dependent and
shows, for instance, 79% trans at 100ºC and 72% trans at 250ºC (Dijkstra,
2006).
The literature regularly mentions ‘shunt reactions’ taking place during the
hydrogenation of polyunsaturated triacylglycerols. The ‘oleate shunt’, being a
‘direct-through’ reaction of linolenic acid to oleic/elaidic acid, was suggested
by Bailey (1949). Subsequent authors (Mounts & Dutton, 1967) even sug-
gested stearate shunts in which linolenic acid and linoleic acid would react
straight through to stearic acid. Their conclusion that these reaction paths
existed was based on otherwise inexplicable deviations from kinetic models. In
retrospect, the validity of these models is doubtful, since they do not take into
account triacylglycerol selectivity or the order with respect to hydrogen of the
various fatty acid moieties. Besides, the only difference in reaction rate
between fatty acid isomers taken into account is between linoleic acid and
isolinoleic (cis-9,cis-15 18:2) acid. Accordingly, the models used are oversim-
plified and do not provide valid evidence for the existence of shunt reactions.
When an oil like sunflower oil is hydrogenated and elaidic acid is observed
in the reaction product, it is impossible to say whether this trans isomer of oleic
acid results from the isomerization of oleic acid or from the hydrogenation of
linoleic acid. On the other hand, when high-erucic-acid rapeseed (HEAR) oil is
hydrogenated, the isomerization of monounsaturated acid can be followed by
looking at just the erucic (22:1) acid since HEAR oil does not contain a large
amount of dodecadienoic acids. Coenen and Boerma (1968) studied the
formation of brassidic (trans-22:1) acid during hydrogenation of HEAR oil at
100ºC. They noted the formation of this trans isomer always coincided with the
formation of behenic (22:0) acid and concluded that monounsaturated fatty
acids cannot isomerize without some of them being reduced at the same time.
This turned out to be an unwarranted generalization. When the experiment with
HEAR oil was repeated at a much higher temperature and under selective
conditions, the erucic acid was observed to isomerize without behenic acid
being formed (W.L.J. Meeussen, personal communication).
As will be explained later on, the reason for the different outcome of the two
experiments lies in the hydrogen concentration, which was much higher in the
low temperature experiment than in the high temperature one. During most
hydrogenation experiments, the hydrogen concentration varies and this is
something most authors, with one exception (Coenen, 1978), do not take into
account or even deny (Jonker, 1999). In the beginning of an industrial hydro-
genation experiment or its laboratory mimic, the iodine value and the reactivity
of the reaction mixture will still be high. This high reactivity causes the
FORMATION OF TFA DURING CATALYTIC HYDROGENATION 47
1 2
H2 H2* 2 H*
c-M t-M
13
12
14 15
c-M*+ H* MH* t-M* + H*
+
H*
16
S
Figure 1. Horiuti-Polanyi mechanism (see text for explanation).
step (10) will be favoured and if this is low, steps (4) and (5) will be favoured.
Accordingly, the hydrogen concentration determines how many double bonds
are isomerized per double bond being saturated. This ratio is commonly
referred to as the ‘isomerization index’.
The hydrogen concentration on the catalyst’s surface [H*] can be low for two
reasons. The catalyst can be poisoned (Baltes, 1972; Rijnten & Eikema, 1975)
so that a given concentration of hydrogen in the oil [H2] corresponds to a lower
concentration of atomic hydrogen on the catalyst’s surface [H*], or the
molecular hydrogen concentration [H2] in the bulk of the oil can be low. This
latter phenomenon can have a number of causes such as the high reactivity of
the hydrogenation substrate, a low pressure in the system, or a small gas-liquid
volumetric gas transfer coefficient kla due to, for instance, slow agitation.
Other reasons can be a high temperature, a relatively high amount of catalyst
and a relatively active catalyst, all of which increase the rate of the reaction and
thus lower the molecular hydrogen concentration [H2] in accordance with the
dynamic equilibrium mentioned above. According to the mechanism shown in
Figure 1, a monoene (c-M or t-M) can adsorb onto the catalyst surface via steps
(12) or (13) and react with an adsorbed hydrogen atom to form a half-
hydrogenated monoene (MH*). As before, this intermediate can react via step
FORMATION OF TFA DURING CATALYTIC HYDROGENATION 49
(16) with a further hydrogen atom and become fully saturated, or dissociate.
This dissociation may lead to isomerization that can be geometrical, positional
or both.
In the partial hydrogenation of edible oils, it is often the intention to saturate
polyenes but to refrain from saturating monoenes and forming saturated fatty
acids. The reason for this intention is simply that saturated fatty acids lead to an
increase in triacylglycerol melting point and this can cause a sticky mouthfeel.
Accordingly, the ratio of the rates of reaction of dienes and monoenes (also
referred to as the ‘linoleic acid selectivity’) is an important process character-
istic that merits a detailed discussion. It has already been mentioned that this
ratio is not constant during the course of a hydrogenation process but decreases
when the reactivity of the substrate decreases. This has been tentatively
explained (Dijkstra, 1997; 2002a), by assuming that for diene hydrogenation,
the rate determining step is the formation of the half-hydrogenated intermedi-
ate (cDH*); accordingly this rate is proportional to the concentration of the
atomic hydrogen [H*]. For monoene hydrogenation on the other hand, the
saturation of the half-hydrogenated intermediate (MH*) determines the overall
rate of stearic acid formation. Since the equilibrium concentration of the half-
hydrogenated intermediate MH* is proportional to the concentration of adsorbed
atomic hydrogen [H*] and the rate of saturation of this intermediate is also
proportional to this adsorbed atomic hydrogen concentration [H*], the rate of
stearic acid formation is proportional to [H*]2 or [H2].
Accordingly, when the concentration of molecular hydrogen in the bulk of
the oil increases because the reactivity of the substrate decreases, the concen-
tration of the atomic hydrogen [H*] adsorbed increases. This favours monoene
saturation over diene saturation and causes the linoleic acid selectivity to
decrease. However, in addition to the hydrogen concentration, there is another
reason why monoenes may react differently from dienes. This reason has to do
with the catalyst. If the catalyst would only adsorb dienes and would show no
affinity for monoenes whatsoever, no monoenes would be hydrogenated and
no stearic acid would be formed. This would correspond to an infinite linoleic
acid selectivity. Much effort has gone into developing catalysts with this kind
of adsorption preference, but no nickel catalyst with an absolute linoleic acid
selectivity has as yet resulted.
On the other hand, copper catalysts that do not hydrogenate monoenes have
been developed and studied in depth. These catalysts do not cause monoenes to
be saturated since they only catalyse the hydrogenation of conjugated double
bonds. Accordingly, the hydrogenation process commences with this conjuga-
tion and it is likely that this conjugation is initiated by the abstraction of a
bis-allylic hydrogen atom rather than by the addition of a hydrogen atom to one
of the double bonds (Dijkstra, 2002b). As only to be expected, this hydrogen
abstraction and the subsequent addition of a hydrogen atom are reversible
reactions and the intermediates involved lose their original geometrical con
50 TRANS FATTY ACIDS IN HUMAN NUTRITION
Table 2. Trans octadecenoic acid isomers level (g per 100 g of fatty acids) and profile
(% of total trans octadecenoic acid isomers) in commercial partially hydrogenated
vegetable oils (source: Nestlé, 2005)
Sample 1 2 3 4 5 6 7 8 9 10
60
Soya bean oil
50 Fish oil
Palm oil
TFA content (%)
40
30
20
10
0
180 160 140 120 100 80 60 40 20 0
Iodine value
Figure 2. Trans isomer formation in three different oils under classical hydrogenation conditions
(adapted from Engelhard).
52 TRANS FATTY ACIDS IN HUMAN NUTRITION
have TFA isomer contents up to 60%. Most of the TFA isomers found in
partially hydrogenated vegetable oils are monounsaturated fatty acids, which
are partly responsible for the physical properties (plasticity and melting behav-
iour). As depicted in Figure 2, TFA content varies with the degree of
unsaturation. The level depends strongly on the nature (iodine value) of the
starting material. The higher the starting iodine value, the higher will be the
level of TFA isomers at any point in the conventional hydrogenation process.
Hydrogen
Venturi
mixer
Heat exchanger
Oil with
suspended
catalyst
Circulating pump
reactor (Buss AG, Switzerland, see Figure 4) through which the reaction
mixture is continuously circulated at great velocities causing intense, intimate
mixing of the oil with the hydrogen being sucked into the venturi (Duveen &
Leuteritz, 1982).
An important aspect in hydrogenation is the design of the agitator, which has
to ensure a proper gas dispersion and good mass transfer during the reaction.
Radial flow flat-blade and axial flow flat-blade turbines are commonly used
providing high shear rates and vortex formation. However, it has been shown
that mixing intensities achievable with these rotary stirrers in industrial-scale
reactors are not high enough to control, significantly, the selectivity and
geometrical isomerization rate occurring during nickel-catalysed hydrogena-
tion of oils (Ackman & Mag, 1998).
Despite the interest in continuous hydrogenation, few continuous plants are
in operation.
2. Hydrogenation parameters
As described previously, the rate and selectivities of the hydrogenation reac-
tion are mainly the result of process parameters such as temperature, hydrogen
pressure, stirring conditions, catalyst concentration and catalyst type. By
varying these parameters, a great variety of finished products with different
physical properties can be obtained from the same starting material. To
illustrate this variety, various curves for solid fat content of partially hydrogen-
ated palm oil obtained using different hydrogenation conditions are given in
Figure 5. As shown in this Figure, the reaction time is a process parameter that
gives rise to different degrees of saturation as well as different TFA contents
and solid fat content profiles. The reaction time influences the final physical
properties, in this case the melting characteristics of the oils.
Referring back to the mechanism of hydrogenation, the availability of
hydrogen at the catalyst/reaction site is a crucial parameter in the cis/trans
isomerization process. Like nearly all chemical reactions, hydrogenation is
temperature dependent: the reaction rate constants increase with the tempera-
ture. Indeed, an increase in temperature directly induces higher solubility of
hydrogen and a decrease in viscosity which improves mass transfer during the
reaction. While increased temperature increases the hydrogenation reaction
rate it also tends to decrease the concentration of hydrogen at the surface of the
catalyst (Ariaansz, 2006). Under these hydrogen-starved conditions the level
of TFA will increase (Larsson, 1983). As depicted in Figure 6, during partial
hydrogenation of soybean oil with the same catalyst concentration and under
the same low pressure (3 bars), higher temperatures induce higher TFA
contents.
Hardening is usually carried out at pressures between 2 and 5 bars, but the
rate and selectivity of the reaction can be affected by modifying hydrogen
FORMATION OF TFA DURING CATALYTIC HYDROGENATION 55
120
Hydrogenation time
100
80
Solid (%)
60
40
20
0
10 20 30 40 50 60
Temperature (°C)
Figure 5. Solids content of hardened palm oil dependent on reaction time. õ palm oil, IV 55, 0% trans;
palm oil, IV 48, 35% trans; ¸ palm oil, IV 42, 30% trans; ü palm oil, IV 1.5, 1.2% trans.
Selective conditions
40
180°C, 3 bar 35
150°C, 3 bars
30
120°C, 20 bars
120°C , 3 bars 25
% trans
20
15
Non-selective conditions
10
5
0
140 120 100 80 60
Iodine value
Figure 6. Trans fatty acid level in hardened soybean oil as a function of reaction temperature and
hydrogen pressure (nickel catalyst, 0.01%).
Unfortunately, the final TFA level remains quite important even under these
non-selective conditions.
Nickel catalysts are still the most commonly used catalyst for vegetable oil
hydrogenation (Farr, 2005). The amount of nickel used in conventional indus-
trial processes is between 0.005 and 0.01% (w/w % of oil) (Ackman & Mag,
1998). Even if these catalysts present several advantages including low cost,
high activity, potential tailored linolenic and linoleic selectivities, they also
promote the geometrical and positional isomerization of cis-ethylenic double
bonds during the hardening process. Increasing the catalyst concentration will
increase the hydrogen consumption that reinforces hydrogen-starved condi-
tions at the catalyst site leading to increased formation of TFA isomers. The
selectivity of the nickel catalysts is influenced by the particle size and porosity
(Beckmann, 1983). As previously mentioned, wide pore and small particle size
tend to favour higher selectivity (high linoleic and linolenic selectivities, SI,
SII respectively) because this allows shorter residence time at the catalyst site
(Beckmann, 1983). Under conventional process parameters (low pressure
3 bars, high temperature 180ºC) differences between catalyst structures have
only a small influence on TFA formation (Ariaansz, 2006, personal communi-
cation). However, hydrogenating soybean oil with the same iodine value using
a selective catalyst (wide pore) at low temperature and high pressure, results in
a significant reduction of 55% of TFA content compared to using non-selective
catalysts which leads to only 37% reduction (Ariaansz, 2006, personal commu-
nication).
Impurities in the feedstock can also have an influence on TFA formation, by
altering the activity of the catalyst. Sulphur compounds present in rapeseed oil
for instance tend to decrease the active sites in the catalyst surface thereby
increasing the likelihood of isomerization of cis-ethylenic double bonds
(Drozdowski & Szukalska, 2000). On the other hand, the presence of
phosphatides in the starting material tends to reduce the formation of TFA
(Beckmann, 1983). However, it appears that despite a decrease in the level of
TFA by optimizing temperature, pressure, and nickel catalyst structure, the
existing process does not allow operation under optimal conditions. For
example, to achieve a final TFA level below 10%, high hydrogen pressures
(above 50–60 bars) are required but this pressure cannot be achieved using the
existing hydrogenation plants. The overall influence of processing conditions
on hydrogenation characteristics is summarized in Table 3.
D. Conclusion
An understanding of the mechanism and key parameters involved in the
formation of TFA during catalytic hydrogenation of vegetable oils has been
acquired during recent decades. The performance of catalysts and plant designs
have been investigated in order to identify the necessary operating conditions
to limit the formation of TFA during partial hydrogenation of edible oils.
However, for the time being, the production of ‘zero trans’ hydrogenated fats
is elusive due, essentially, to cost limitations. Alternatives to partial hydro-
genation such as interesterification and fractionation become increasingly
popular. These processes allow the effective formulation of fats and oils having
physical properties that make them suitable to replace hydrogenated fats in
food products. The food industry has already started to reformulate food
60 TRANS FATTY ACIDS IN HUMAN NUTRITION
products using these trans fat alternatives and recent dietary surveys indicate
that the TFA intake has decreased in many European Union countries (Morin,
2005). A new challenge for food manufacturers nowadays is to overcome the
increased content in food products of pro-atherogenic saturated fatty acids
introduced by these new fractionated or interesterified fats.
References
Ackman, RG and Mag, TK (1998) Trans fatty acid and the potential for less in technical
products. In: Trans Fatty Acids in Human Nutrition (J-L Sébédio and WW Christie, eds),
Oily Press, Bridgwater, UK, pp.35–36.
Alouche, A, Lambert, DC, Hubaut, R, Davies, P and Hertoghe, P (1994) Procédé
d’hydrogénation sélective d’une huile végétale avec rétention des doubles liaisons de
configuration cis et utilisation des huiles obtenues par ce procédé. FR Patent 2 694 015,
assigned to Société de la Raffinerie BP et ELF de Dunkerque.
An, W, Hong, JK, Pintauro, PN, Warner, K and Neff, W (1998) The electrochemical
hydrogenation of edible oils in a solid polymer electrolyte reactor. I. Reactor design and
operations. J. Am. Oil Chem. Soc., 75, 917–924.
An, W, Hong, JK, Pintauro, PN, Warner, K and Neff, W (1999) The electrochemical
hydrogenation of edible oils in a solid polymer electrolyte reactor. II. Hydrogenation
selectivity studies. J. Am. Oil Chem. Soc., 76, 215–222.
Ariaansz, RF (1996) Hydrogenation with minimum trans acids. In: Lipids & Nutrition:
Current Hot Topics (KG Berger, ed) PJ Barnes & Associates, Bridgwater, UK, pp.81–
103; proceedings of SCI conference held on 30 April 1996, London, UK.
Bailey, AE (1949) Theory and mechanics of the hydrogenation of edible fats. J. Am. Oil
Chem. Soc., 26, 596–601.
Bailey, AE and Fisher, GS (1946a) Modifications of vegetable oils. V. Relative reactivities
toward hydrogenation of the mono- di- and triethenoid acids in certain oils. Oil & Soap,
23, 14–18.
Baltes, J (1970) Selektive Härtung von Fetten und Fettsäuren mit Sulfiden der Übergangsmetalle
als Katalysatoren. Fette Seifen Anstrichm., 72, 425–432.
Baltes, J (1972) Process for the selective hydrogenation of fats and fatty acids. US Patent
3,687,989.
Beckmann, HJ (1983) Hydrogenation practice. J. Am. Oil Chem. Soc., 60, 282–290.
Beers, AEW, Berben, PH, Groen, C, Jagta, R, Lazar, G and Mangnus, G (2004) Lowering the
trans content in edible oils. Paper presented at the 3rd EuroFedLipid Congress,
Edinburgh, UK, page 62 of the abstracts.
Beers, AEW, Berben, PH and Okonek, DV (2005) Low trans hydrogenation of edible oils —
new developments by Engelhard. Paper presented at the 26th ISF World Congress and
Exhibition, Prague, page 78 in the abstracts.
Beers, AEW and Mangnus, G (2004) Hydrogenation of edible oils for reduced trans fatty acid
content. Inform, 15, 404–405.
Beers, AEW (2007) Low trans hydrogenation of edible oils. Lipid Technology, 19, 56–58.
Beers, AEW, Ariaansz, RF, Okonek, D (2008) Trans isomer control in hydrogenation of
edible oils. In: Trans Fatty Acids (AJ, Dijkstra, RJ Hamilton, W, Hamm, eds) Blackwell
Publishing Ltd, Oxford, UK, pp.147–180.
Beyens, Y and Dijkstra, AJ (1983) Positional and triglyceride selectivity of hydrogenation
of triglyceride oils. In: Fat Science 1983, Proceedings of 16th ISF congress, Budapest (J
Holló, ed), Akadémiai Kiadó, Budapest, Hungary, pp.425–432.
Bockisch, M (1998) Fats and Oils Handbook, AOCS Press, Champaign, Illinois, USA.
FORMATION OF TFA DURING CATALYTIC HYDROGENATION 61
Toor, H van, Rossum, GJ van, and Kruidenberg, M (2005) Low trans fatty acid compositions;
low-temperature hydrogenation, e.g. of edible oils. US Patent Application Publication
2005/0027136 A1, assigned to Cargill Incorporated.
Veldsink, JW, Bouma, MJ, Schöön, N-H, and Beenackers, AACM (1997) Heterogeneous
hydrogenation of vegetable oils: a literature review. Catal. Rev. -Sci. Eng., 39, 253–318.
Warner, K, Neff, WE, List, GR, and Pintauro, PN (2000) Electrochemical hydrogenation of
edible oils in a solid polymer electrolyte reactor. Sensory and compositional character-
istics of low trans soybean oils. J. Am. Oil Chem. Soc., 77, 1113–1117.
Willett, WC and Ascherio, A (1994) Trans fatty acids: are the effects only marginal? Am. J.
Public Health, 85, 722–744.
Wright, AJ, Mihele, AL, and Diosady, LL (2003a) Ni-catalyst promotion of a cis selective
Pd catalyst for canola oil. Food Res. Int., 36, 797–804.
Wright, AJ, Wong, A, and Diosady, LL (2003b) Cis selectivity of mixed catalysts systems
in canola oil hydrogenation. Food Res. Int., 36, 1069–1072.
CHAPTER 3
Formation of trans fatty acids during
deodorization of edible oils
1
Nestlé Product Technology Center, Marysville, Ohio, USA
2
Nestlé Research Center, Lausanne, Switzerland
CRUDE OIL
C P
H H
E Y
M Degumming Degumming S
I I
C C
A Neutralisation FFA A
L Bleaching L
R Bleaching R
E FFA E
F Deodorisation Deodorisation F
I I
N volatiles contaminants N
I I
N N
G REFINED OIL G
under normal operating conditions at 250°C, the TFA content can reach 3%
(Pudel & Denecke, 1997; Wolff,1992).
Deodorization consists of injecting live steam into oil maintained under
vacuum (2 to 4 mbar) and at high temperature (220°C–260°C). It has been
shown by Ackman et al. (1974) in the early 1970s that small amounts of
geometrical (trans) isomers of the essential fatty acids can be formed. Geo-
metrical isomerization, also called stereomutation, refers to modification of the
configuration of the ethylenic double bond. Almost all the vegetable oils
contain linoleic (cis-9,cis-12 18:2) acid, and some of them such as soybean or
rapeseed oils also contain significant amount of α-linolenic (cis-9,cis-12,cis-
15 18:3) acid. The nutritional value of these vegetable oils is very important
since both linoleic and α-linolenic acids are essential fatty acids, required for
the endogenous synthesis of long-chain polyunsaturated fatty acids. Activation
energies for cis to trans geometrical isomerization of ethylenic double bonds of
linoleic and α-linolenic acids are 178 kJ/mole and 144–148 kJ/mole respec-
tively (Greyt and Kellens, 2005). Exposure to elevated temperatures during the
deodorization triggers the formation of trans isomers of polyunsaturated
linoleic and α-linolenic acids. It has been shown that the consumption of trans
isomers of α-linolenic acid increases the ratios of plasma LDL-cholesterol to
HDL-cholesterol and total cholesterol to HDL-cholesterol in healthy men
(Vermunt et al., 2001).
5 Total TFA
C18:3 trans
TFA content (%)
4 C18:2 trans
C18:1 trans
3
0
200 220 240 260
Deodorisation temperature (°C)
Figure 2. Formation of trans isomers of oleic, linoleic and α-linolenic acids during low-erucic
rapeseed oil deodorization (adapted from Bertoli and co-workers, 1998)
HO
O O
HO HO
HO
Figure 3. Representation of the geometrical isomers formed from linoleic acid during deodorization.
The star symbol and plain arrows indicate the main isomers formed under conventional deodorization
conditions.
HO
68
O O O
HO HO HO
O O
HO HO
HO
TRANS FATTY ACIDS IN HUMAN NUTRITION
HO
Figure 4. Representation of the geometrical isomers formed from α-linolenic acid during deodorization. The star symbol and plain arrows indicate the main
isomers formed under conventional deodorization conditions.
FORMATION OF TFA DURING DEODORIZATION 69
(1992a, see Figure 4). Out of the 8 geometrical isomers that can be produced by
geometrical isomerization of α-linolenic acid, only 4 isomers are found in
conventionally refined vegetable oils (Wolff, 1992). The kinetics of isomeriza-
tion of α-linolenic acid during deodorization are shown in Figure 5. The main
isomers formed are the three mono-trans (trans-9,cis-12,cis-15; cis-9,cis-
12,trans-15; and cis-9,trans-12,cis-15) and the di-trans isomer trans-9,cis-
12,trans-15 (Figures 4 and 5). Less cis-9,trans-12,cis-15 18:3 acid isomer is
produced (about 6–7 times less, Wolff 1992a) compared to the other mono-
trans isomers having a trans double bond (Δ 9 or Δ15) at the extremity of the
methylene interrupted system (Wolff 1992a). This fact clearly shows that some
conformations are thermodynamically preferred. The level of trans-9,trans-
12,cis-15, cis-9,trans-9,trans-15 and trans-9,trans-12,trans-15 18:3 acid
isomers is usually very low or not detectable (Wolff 1992a; Wolff 1993).
It has been observed that the amount of geometrical isomers derived from
linoleic acid increases linearly with deodorization time (Pudel and Denecke,
1997) but below 220°C very small amounts of trans linoleic are formed. It has
been observed at high temperatures, for example 275°C (conditions that are
never applied in industrial refining), that about 10% of (all-cis) linoleic acid is
isomerized into trans isomers (Pudel and Denecke, 1997). Also, below 220°C,
the isomerization rate of α-linolenic acid is low and the level of geometrical
isomers formed increases linearly with time (Pudel and Denecke, 1997).
Bertoli and co-workers (1998) showed that the final TFA content in rapeseed
oil stripped below 220°C for 6 hours did not exceed 1% (see Figure 5). At
higher temperature, it appears that geometrical isomerization of polyunsatu-
rated fatty acids becomes exponential; at 255°C, about half of the initial
quantity of the α-linolenic acid is isomerized (see Figure 5).
10 2
c,c,c linolenic acid (%)
8 1.6
t, linolenic acid (%)
6 1.2
4 0.8
2 0.4
0 0
180 200 220 240 260 280
Temperature (°C)
100
ARA
90
EPA
w3 DPA
80
w6 DPA
70 DHA
60
Relative %
50
40
30
20
10
0
Control 180 220 250
Deodorization Temperature (°C)
Figure 6. Relative decrease in concentration of the main long-chain polyunsaturated fatty acids found
in tuna oil submitted to deodorization at various temperatures for 2 hours (adapted from Fournier et al.,
2006a).
Mjøs, 2008). The theoretical number of geometrical isomers formed from EPA
(25 = 32) or DHA (26 = 64) are important but not all isomers are formed during
heat treatment of these long-chain polyunsaturated fatty acids. The chromato-
graphic separation of EPA and DHA geometrical isomers can be achieved
using the high-polarity open-tubular capillary columns conventionally used for
separation of fatty acid methyl ester derivatives (Fournier and co-workers,
2006a,b; Fournier and co-workers, 2007; Mjøs and Solvang, 2006; Mjøs 2005)
Interesting separations have also been obtained on long columns with
polyethylene glycol stationary phase (Mjøs 2008).
Identification of geometrical isomers formed during deodorization has been
achieved using elaidinized pure methyl EPA and DHA as previously done by
Wolff for the identification of geometrical isomers of α-linolenic acid (1992b).
It has been shown that mono- and di-trans isomers are mainly formed during
fish oil deodorization (Fournier and co-workers, 2006a,b; Mjøs and Solvang,
2006). Further isomerization can occur when EPA or DHA are exposed to
temperatures above 200°C as reported by Fournier and co-workers (2006a,b)
and illustrate in Figure 8. The main results obtained from these different studies
is that deodorization temperatures above 200°C lead to an important (> 10%)
72 TRANS FATTY ACIDS IN HUMAN NUTRITION
400
Cyclic fatty acid monomers (CFAM)
Polar compounds
300
250
200
150
100
50
0
Control 180 220 250
Deodorization Temperature (°C)
Figure 7. Distribution of degradation products formed from long-chain polyunsaturated fatty acids
(LC-PUFA) during deodorization of fish (tuna) oil at various temperatures for 2 hours. Results are
expressed as mg per g of oil (adapted from Fournier et al., 2006a).
loss of all cis EPA and DHA and the formation of geometrical isomers (mainly
mono-trans) (Fournier and co-workers, 2006a; Mjøs and Solvang, 2006).
Analysis of commercially available refined marine oil samples revealed that
levels of geometrical isomers rarely exceed 1% (Fournier et al., 2007). The
effects of these trans fatty acids on health are not known. However, the
consumer exposure levels are low and marine oils are incorporated at low
levels in food formulations (Kolanowski and Laufenberg, 2006). The forma-
tion of polar compounds seems to be more important than that of geometrical
isomers in the case of fish oil, which differs from the case of vegetable oils
(Figure 7).
Di-trans EPA
Tri-trans DHA
Ag-TLC Fraction 4
Ag-TLC Fraction 3
Mono-trans DHA
All-cis EPA trans-10
trans-13 + trans-16
trans-19
All-cis DPA
Ag-TLC Fraction 2
All-cis DHA
Ag-TLC Fraction 1
40.0 51.0
Retention Time (min)
Figure 8. Partial gas chromatograms of Ag-TLC fractions of fatty acid methyl ester derivatives of
eicosapentaenoic (cis-5,cis-8,cis-11,cis-14,cis 17 20:5, EPA) and docosahexaenoic (cis-4,cis-7,cis-
10,cis-13,cis-16,cis 19 22:6, DHA) geometrical acid isomers obtained from fish oil deodorized at 220°C.
Fractions 1, 2, 3 and 4 correspond respectively to the 6 cis, 5 cis, 4 cis and 3 cis polyenoic fractions. DPA
stands for docosapentaenoic acid and, as an example, trans 19 indicates the cis-4,cis-7,cis-10,cis-13,cis
16,trans-19 22:6 acid isomer. Figure adapted from Fournier et al., 2006b.
single vessel. This system allows different options such as deodorization at two
different temperatures, and deep or shallow bed deodorization. Likewise, the
development of more efficient vacuum systems (dry condensing) allows the
reduction of the deodorization temperature without affecting the stripping
efficiency (Greyt & Kellens, 2005).
E. Conclusion
The formation of geometrical isomers from all-cis polyunsaturated fatty acids
happens when oils are deodorized at high temperature. The main consequence
of this phenomenon is the loss of nutritive value. The development of analytical
methods in the early 1990s provided a better understanding of the isomers
formed from linoleic and α-linolenic acids and measures were taken to better
control the deodorization operation at the plant. The formation of trans isomers
of EPA and DHA during the deodorization of fish oil was investigated recently.
Overall, it can be concluded that the residual content of geometrical isomers
formed from polyunsaturated fatty acids during deodorization is nowadays
very low and does not seem to represent a public health issue.
References
Ackman, RG, Hooper, SN and Hooper, DL (1974) Linolenic acid artifacts from the
deodorisation of oils. J. Am. Oil Chem. Soc., 51, 42–49.
Ackman, RG and Mag, TK (1998) Trans fatty acid and the potential for less in technical
products. In: Trans Fatty Acids in Human Nutrition (J-L Sébédio and WW Christie, eds),
Oily Press, Bridgwater, UK, pp.35–36.
Berdeaux, O, Fournier, V, Lambelet, P, Dionisi, F, Sébédio, J-L and Destaillats, F (2007).
Isolation and structural analysis of the cyclic fatty acid monomers formed from
eicosapentaenoic and docosahexaenoic acids during fish oil deodorisation. J. Chrom. A,
1138, 216–224.
Bertoli, C, Delvecchio, A, Durand, P, Gumy, D, Bellini, A and Stancanelli, M (1998)
Formation of trans fatty acids during deodorisation of low erucic acid rapeseed oil. In:
World Conference on Oilseed and Edible OIls Processing (SS Köseolu, KC Rhee and RF
Wilson, eds), AOCS Press, Champaign, Illinois, USA, pp.67–71.
Fournier, V, Destaillats, F, Juaneda, P, Dionisi, F, Lambelet, P, Sébédio, J-L and Berdeaux,
O (2006a) Thermal degradation of long-chain polyunsaturated fatty acids (LC-PUFAs)
during deodorisation of fish oil. Eur. J. Lipid Sci. Technol., 108, 33–42.
Fournier, V, Juanéda, P, Destaillats, F, Dionisi, F, Lambelet, P, Sébédio, J-L and Berdeaux,
O (2006b) Analysis of eicosapentaenoic and docosahexaenoic acid geometrical isomers
formed during fish oil deodorisation. J. Chrom. A, 1129, 21–28.
Fournier, V, Destaillats, F, Hug, B, Golay, P-A, Joffre, F, Juanéda, P, Sémon, E, Dionisi, F,
Lambelet, P, Sébédio, J-L and Berdeaux, O. (2007) quantification of eicosapentaenoic
(EPA) and docosahexaenoic (DHA) acid geometrical isomers formed during fish oil
deodorisation by gas-liquid chromatography, J. Chrom. A, 1154, 353–359.
Kolanowski, W and Laufenberg, G (2006) Enrichment of food products with polyunsaturated
fatty acids by fish oil addition. Eur. Food Res. Technol., 222, 472–477.
Mjøs, SA (2005) Properties of trans isomers of eicosapentaenoic acid and docosahexaenoic
acid methyl esters on cyanopropyl stationary phases. J Chrom A, 1100, 185–192.
FORMATION OF TFA DURING DEODORIZATION 75
1
Université de Toulouse; INP; LCA (Laboratoire de Chimie AgroIndustrielle);
ENSIACET, Toulouse, France.
2
INRA; LCA; Toulouse, France.
The US Food and Drug Administration defines trans fatty acids (TFA) as fatty
acids containing one or more non-conjugated double bonds in trans configura-
tion. This definition excludes conjugated linoleic acids (CLA) that have at least
one trans ethylenic double bond.
Pure trans fatty acids are currently not available in large quantities and their
study is therefore limited. In order to improve our understanding of TFA, it is
necessary to achieve large-scale production of individual isomers in high
purity and, for each TFA, a different chemical operating process has to be
developed. All the processes are built according to an integrated approach,
each step providing a value-added molecule. The common process for TFA
synthesis can be broken down into three steps as follows.
• Double bond creation in specific geometrical configuration and position:
the Wittig olefination between a carbonyl compound and a phosphonium
ylide is the preferred method that leads to cis-trans fatty acid mixtures.
• Cis-trans isomerization: as Wittig olefination provides geometrical isomers
mixtures, it is necessary to convert the cis fatty acids into trans fatty acids
by double bond isomerization.
• Fractionated crystallization: a purification process in dry or solvent condi-
tions that allows high purity to be achieved. Such technology is well-known
and already developed at industrial scale for stearin/olein purification.
Each step will be developed in this chapter with the presentation of the
chemical reaction, the key parameters of multi-step processes, the technologies
and technical barriers, the limits, and the products obtained with the objective
of helping readers in their choice of reaction pathways.
–
X
PPh 3 + X CH R Ph 3P+ CH R Ph 3P+ C
–
R Ph 3P C R
base
R' R' R' R'
Phosphonium salt Ylide
various functional groups, such as OH, OR, NR2, aromatic nitro or halo, acetal
or even ester group. The stereochemistry of the product is thought to be
influenced by the reversibility of the formation of the isomeric erythro and
threo oxaphosphetanes which undergo specific loss of triphenylphosphine
oxide to give the trans (E) and cis (Z) respectively (Abell & Edmonds, 2004).
Factors that enhance the reversibility of this step favour the threo intermediate.
O O
-
R2CH2P(OR1)2 R2CH2P(OR1)2
base
+ - + R1CHO
PR3 CH R2
R1 R2
O PR3 O PR3
Oxaphosphetanes
R1 R2 R1 R2
+ +
O PR3 O PR3
- -
R2
R1 R2 R1 R2
R1 R2 Zwitterions R1
Figure 5. Temperature influence on the stereochemistry of the Wittig olefination (Vinczer et al.,
1988).
Table 1. Usual solvents in the Wittig olefination and some of their physical and
empirical parameters: bp, boiling point; εr, dielectric constant; µ, dipole moment; ENT,
normalized acid Lewis solvent parameter (Zalewski & Kokocinske, 1989).
R1CHO
Ph 3P CHR R1HC CHR
+ - NaOH
Ph 3P CH2RX
+ -
Ph 3P CH2ROH Ph 3PO + RCH3
Figure 6. Wittig reaction in phase transfer catalysis (PTC) conditions using aqueous sodium hydroxide.
the use of polar solvents. Table 1 provides physical and empirical parameters
of usual solvents for the Wittig synthesis. The knowledge of these parameters
could ease the choice for controlling the stereoselectivity.
Table 2. Influence of the reaction temperature and reaction time on the Wittig Horner
reaction using a phase transfer catalysis (PTC) liquid-liquid system (Villieras &
Rambaud, 1983). T, temperature.
Table 3. Influence of the aliphatic chains of the aldehyde and of the phosphonium salt
on the Wittig reaction in 1,4-dioxane and methanol (Moussaoui et al., 2006).
CH3 C2H5 72 68 30 20
n-C4H9 74 68 28 20
C2 H 5 C2H5 70 68 28 –
n-C4H9 70 66 30 –
C3 H 7 C2H5 68 60 26 18
n-C4H9 72 62 26 –
n-C7H15 C2H5 64 54 22 –
n-C4H9 70 54 28 –
n-C8H17 C2H5 58 50 26 18
n-C4H9 60 52 26 18
+ -
M2CO3 solid + 2 (C4H9)4N Br ((C4H9)4N+)2 CO32- + 2 MBr
O O
O O O O
M2CO3 solid + M+ CO3
2-
O O O O
O O 2
Figure 8. Use of phase transfer catalysts with carbonates in the Wittig olefination.
aldehyde. This could be due to the donating effect of the alkyl group placed on
the carbonyl bond, increasing thereby the electron density on the carbon of the
carbonyl group which is detrimental to the attack of the ylide. On the other
hand, there is very little effect of phosphonium chain length, confirming
previous studies (Vinczer et al., 1988).
The reaction temperature has also an influence on the yield. Increasing
temperature results in the enhancement of the basicity by the weakening of the
anion-cation link of the base.
In solid-liquid PTC systems, reactions using bases are usually carried out in
the presence of phase transfer catalysts such as crown ethers or tetra-
alkylammonium salts which are beneficial for reactivity in organic solvent. The
catalyst leads to an increase of the basicity of the anion via ionic exchange
equilibria in the case of tetra-alkylammonium salts or via complexation of the
metal cations with crown ethers (Figure 8).
In solid-liquid systems, another synthesis option involves transforming
triphenylalkylphosphonium bromides, chlorides or iodides into more reactive
phosphonium fluorides by the use of NaF or KF coupled with dibenzo-18-crown-
6 ether as catalyst in organic solvent (Kossmehl & Nuck, 1979) (Figure 9).
For Wittig-Horner reactions, solid KOH or NaOH, K2CO3, Ba(OH)2 and
Cs2CO3 can be used (Dehmlow & Barahona-Naranjo, 1981; Sinisterra et al.,
1991). For mild bases, the determination of the carbanionic structures and the
effect of water on carbanion formation have been investigated for the conver-
sion of furfural and benzaldehyde into α,β-unsaturated compounds by the
reaction of triethyl-phosphonoacetate in the presence of Ba(OH)2.H2O,
K2CO3.1.5H2O and Cs2CO3.3H2O (Mouloungui et al., 1989). The quantity of
water needed to accelerate the reaction depends upon the nature of the cation.
Apparently, water decreases the reticulation energy of the crystalline structure
at the interface level. The interaction between water and the solid base
corresponds to the solid-liquid equilibrium in a binary system. The reactions
under the specified conditions proceed through three distinct steps: phosphonate
adsorption on the base active centre and carbanion formation; reaction between
carbanion and carbonyl substrate with the formation of the final product on the
surface of the base catalyst; and desorption of the reaction product.
86 TRANS FATTY ACIDS IN HUMAN NUTRITION
R3R4C=O
R1R2C=CR3R4 (R)3P-C-R1R2 (R)3P=CR1R2
-R3PO
- HF + KF
-
(R)3P+CHR1R2X- (R)3P+CHR1R2F
- + + -
F K K X Liquid phase
Solid phase
KF KX
Figure 9. Wittig reaction in phase transfer catalysis conditions; in situ synthesis of phosphonium
fluorides (Kossmehl & Nuck, 1979).
Figure 10. Wittig reaction in phase transfer catalysis conditions for large-scale production of ethyl
vaccenate (Mouloungui et al., 2007)
B. Cis-trans isomerization
Fatty acid double bonds can be found in two different configurations, namely
cis or Z, and trans or E (Figure 11). The olefination by Wittig reaction mainly
R2
R1 R2 R1
Z, cis E, trans
Figure 11. Unsaturated fatty acid geometrical isomers.
88
Table 4. Examples of unsaturated fatty acid synthesis via the Wittig reaction. T, temperature; α, ylide activation temperature; β, reaction
temperature.
O H3C O
(Foglia & Vail, 40 α 3 Methyl 13- 87 70
1993) Ph3P+, I– 20 β methyloctadec-11- (95/5)
H3CO C10H20
H3C H enoate
(Prakash et al., (CH2)3COOCH3 0α 2 (Z,Z,Z,Z) methyl 5,8, 60 <1
1989) C5H11 Ph3P+, Br– H –78 β 11,14 eicosatetra-
enoate
O
O
(Tranchepain α,β unsaturated THF/ –78 α,β 4 Methyl-13-benzoyl- 55 <1
Ph3P+, Br–
et al., 1989) H3CO C8H16
Hexane/ 25 β oxy-9(Z),11(E)- (85/15)
TRANS FATTY ACIDS IN HUMAN NUTRITION
HMPA octadecadienoate
O O
(Labelle et al., +
Ph3P , I – C8H17 –78 α,β 2 (Z,Z,Z,E)methyl 62 <1
1990) nBuLi/ 20 β 5,8,11,13
H3CO C10H20 H HMDS eicosatetraenoate
O
(Genard & Patin, + – –78 α,β 6 Methyl n-methyl- 80 <1
Ph3P , Br H3COOC
1991) C7H14 C7H14 H –10 β octadec-9-enoate (97/3)
O
O
(Duffy et al., –78 α,β 48 Vaccenic acid 70 10
C6H13
2006) Ph3P+, Br– 20 α,β (82/18)
HO C10H20 H
O O
(Duffy et al., THF/ –78 α,β 16 Vaccenic acid 89 5
+ – C6H13
2006) Ph3P , Br Toluene 20 α,β (88/12)
HO C10H20 H KHMDS
O O
(Klein et al., C9H19
DMSO 20 α,β 24 Ethyl 6-hexa- 35 40
Ph3P+, Br–
2002) C2H5O C6H12
NaHMDS decenoate (95/5)
H
O O
(Mouloungui et al., + –
Cyclo- Reflux 12 Ethyl octadec- 50 570
Ph3P , Br C7H15
2007) C2H5O C9H18
hexane 10-enoate (86/14)
H
O
K2CO3
O
Ethyl vaccenate 82 315
CHEMICAL SYNTHESIS OF MONOUNSATURATED TRANS FATTY ACIDS
C6H13
Ph3P+, Br– (84/16)
C2H5O C10H20 H
89
90 TRANS FATTY ACIDS IN HUMAN NUTRITION
70
60
50
Melting point (°C)
40
cis
30 trans
20
10
0
0 2 4 6 8 10 12 14 16 18
Figure 12. Melting points of 18:1 fatty acid positional and geometrical isomers (Barve & Gunstone,
1971).
affords cis configured molecules. As our objective is trans fatty acid, the
olefination should be followed by an isomerization reaction.
The energy required to activate the transition from one configuration to
another is relatively high (30 kcal/mole) and both configurations are thermally
stable. A cis-trans equilibrium is reached and isomers can be separated
according to their different physical and chemical properties and most often
according to their melting points. As illustration, we can represent octadece-
noic acid melting points versus the position and the configuration of the double
bond (Barve & Gunstone, 1971) (Figure 12).
Cis-trans isomerization involves a rotation at the double bond level. Isomeri-
zation consists in using experimental conditions that cause loss of sp2 character
at the ethylenic carbons. Several reactants have been found to catalyse geo-
metrical isomerization of double bonds, for example selenium, nitrous acids,
sulphinic acids, and thiyl and phosphonyl radicals.
Selenium produces cis-trans isomerization through π complexes (Figure 13) .
Yield is limited by a 0.70 trans / 0.30 cis equilibrium reached under very
strong conditions (190–210°C) (Fitzpatrick & Orchin, 1957). Positional isomeri-
zation and the formation of highly polar substances are drawbacks.
The main isomerization technique rests upon nitrous oxides (NO and NO2),
commonly named ‘nitrous vapours’. Molar fractions of 0.75 trans / 0.25 cis
are obtained at equilibrium. Isomerization proceeds through a radical mecha-
nism but also through an electrophilic addition-elimination mechanism
(Litchfield et al., 1965). Nitric acid reacts with sodium nitrite to produce
nitrous acid in situ which decomposes in the presence of stronger nitric acid.
CHEMICAL SYNTHESIS OF MONOUNSATURATED TRANS FATTY ACIDS 91
Se6 3Se2
Figure 13. Geometrical isomerization mechanism using selenium (Fitzpatrick & Orchin, 1957).
N2O3 • NO
2 + NO
• NO ((aqueous phase) • NO
2 2 (organic phase)
NO2 R2
• NO • •
2 + NO2 +
R1 R2 R1 R2 R1
• NO
2
R1 R2 R1 R2
or
Figure 14. Geometrical isomerization mechanism using nitrous oxides (Litchfield et al., 1965).
Nitrous acid anhydride (N2O3) then decomposes to give the free radical •NO2.
After transition from aqueous to fatty acid phase, the radical adds to the double
bond of the fatty acid, creating a freely rotating C–C bond. This addition is
reversible and either a cis or a trans double bond can be formed when nitrogen
dioxide leaves. The addition of a second •NO2 molecule produces side reaction
products and a fall in yield (Figure 14).
Another isomerization method, leading to 77–80% trans ethylenic com-
pounds, uses p-toluenesulphinic acid (Gibson & Strassburger, 1976; Snyder &
Scholfield, 1982). This acid acts under reflux in 1,4-dioxane. The mechanism
has not been fully established yet. The homolytic fission of the acid depends on
92 TRANS FATTY ACIDS IN HUMAN NUTRITION
O O O O
2 ArSO2H Ar S S Ar + H2O Ar S • + Ar S•
O O
Figure 15. Geometrical isomerization mechanism using p-toluenesulphinic acid (Gibson & Strassburger,
1976).
Radical source X•
X • + RSH XH + RS •
R2
RS
RS • + • + RS
R1 R2 R2
R1 R1
A•
A• + RSH AH + RS•
2 RS • RSSR
Figure 16. Geometrical isomerization mechanism using thiols (Chatgilialoglu et al., 2005).
a duplication reaction with water elimination. One of the two species formed
can add to the double bond and remove it (Figure 15).
Finally, thiols, mainly 2-mercaptoethanol but also thioacetic acid, alkylthiols
and cysteine chlorhydrate undergo homolytic fission in the presence of an
initiator (e.g. peroxide) (Figure 16). Formation of additional substances is
unavoidable (Sgoutas & Kummerow, 1969). Molar fractions of 0.84 trans /
0.16 cis are obtained at equilibrium, using 2-mercaptoethanol under photo-
lytic conditions (Chatgilialoglu et al., 2005).
The usual experimental conditions are summarized in Table 5. Experiments
were done first on fatty acids and fatty acid methyl esters and then applied to the
triacylglycerols of linseed oil (Wolff, 1992) and of borage, rice and olive oils
(Samadi et al., 2004).
All the described methods allow the transition cis → trans. In fact, for all of
these methods, the equilibrium is more rapidly reached starting from cis isomer
than from trans isomer. So, during the same reaction, the conversion of cis
isomer has finished whereas that of trans has just begun. That is why the trans
→ cis conversion is more difficult and must pass through a stabilized reaction
intermediate at the level of the ethylenic bond: epoxide, glycol, thiocarbonate
or dihalide. The stereochemistry of addition and elimination will determine the
configuration of the reaction product.
Table 5. Comparison of the experimental conditions used for cis-trans geometrical isomerization. T, temperature.
Reference Raw material Catalyst Solvent Time (h) T (°C) Final trans/cis ratio
(Fitzpatrick & Orchin, 1957) oleic acid selenium solvent-free 1.0–1.5 190–210 70/30
(Gibson & Strassburger, 1976) methyl oleate p-toluenesulfinic acid 1,4-dioxane 0.2–1.0 Reflux 79/21
(Litchfield et al.,1965) oleic acid HNO3/NaNO2 solvent-free 0.2–0.5 40–65 75/25
(Chatgilialoglu et al.,2005) methyl oleate 2-mercaptoethanol tert-butanol 0.5–1.0 20 84/16
+ di-tert-butyl-ketone Photolytic
CHEMICAL SYNTHESIS OF MONOUNSATURATED TRANS FATTY ACIDS
93
94 TRANS FATTY ACIDS IN HUMAN NUTRITION
C. Fractionation techniques
Investigation of the clinical effects of individual trans fatty acids requires that
they be available at a purity of at least 90%, and this is why the different fatty
acids in a mixture should be separated. That is the aim of the two fractionation
techniques, thermal fractionation and fractional crystallization. Thermal sepa-
ration aims to separate different chain length fatty acids and is based on the
difference in their boiling points. Fractional crystallization leads to the separa-
tion of fatty acids having the same chain length but with different unsaturation
(saturated vs unsaturated, cis vs trans) and is based on the difference in their
melting points. The purification of trans fatty acids is therefore achieved using
fractional crystallization. The difference in melting point between cis and trans
18:1 fatty acids illustrates this choice (Figure 12).
Fractional crystallization has already been studied widely and optimized to
fractionate oils into high melting stearin fractions and lower melting olein
fractions. This was the original aim of the purification technique and its
principal application. Contrary to fatty acids which are characterized by a
single melting point, triacylglycerols exist as three polymorphic crystalline
forms: α (metastable), β' (intermediate) and β (stable), differing from each
other in the hydrocarbon chain packing. Table 6 summarizes the melting points
for the three different crystalline forms for several homogeneous triacyl-
glycerols.
60
50
Melting point (°C)
40
30 cis
trans
20
10
0
0 2 4 6 8 10 12 14 16
Double Bond Position
Figure 17. Homogeneous triacylglycerol β melting point versus double bond position in the 18:1 chain
(Hagemann et al., 1975).
FatMixture
Fat Mixturetoto be
be Solvent:
crystallized
crystallized acetone and hexane
Solvent + Fat
Crystallization:
Nucleation, crystal growth
Separation
Separation
Lowmelting
Low melting fraction
fraction
in solvent
solvent Highmelting
High melting fraction
fraction
in
Concentration:
Solvent evaporation
Solvent evaporation
Solvent distillation
Low
Low melting
melting High melting
High melting
Solvent
Solvent fraction
fraction fraction
fraction
of each phase to remove solvent. Repetition of this cycle and the processing of
materials finally allows a very high quality product to be obtained (Figure 18).
Recent studies on solvent fractionation demonstrate that, occasionally, a
phase separation occurs before crystallization. The liquid phases are solutions
of the material in solvent but they differ in their concentrations. This is known
as ‘oiling out’ (Smith et al., 2007). This liquid-liquid phase separation prior to
crystallization raises the nucleation temperature and has an influence on the
separation of the different molecules of the mixture.
Apart from obtaining a high-quality product, the other advantages of using a
solvent are faster nucleation and growth, easier filtration and easier heat
transfer. However, the Emersol process also presents several drawbacks. The
efficiency of the filter depends on the characteristics of the fatty acid to be
crystallized. Moreover, the use of solvents such as methanol requires that
safety precautions be taken. Alternative technological solutions have been
developed to overcome these drawbacks.
2. Dry fractionation
In contrast to the classical fractionation in solvent, the second technology is
CHEMICAL SYNTHESIS OF MONOUNSATURATED TRANS FATTY ACIDS 97
TEMPERATURE
TEMPERATURE
Liquid
T1 a b c
Solid +
Liquid
Solid
A COMPOSITION B
10
0.5
2.5
10
1.0
8
12.5
Solid content index (%)
6 Saturation
curve
4 Metastable Stable
0
20 40 60 80
Temperature (F)
Figure 20. Saturation-supersaturation diagram for crystallization of partially hardened soybean oil.
Effect of cooling rate on metastable/unstable boundary (dashed line). Numbers are rate of cooling (°F/
min) at 120 rev/min (Singh, 1976).
must be separated from the liquid molecules. The separation step can be
achieved by centrifugation, vacuum filtration or pressing (Gibon & Tirtiaux,
2002). This last technique is nowadays the preferred choice in new fractionation
plants.
Fat
Crystallization Surfactantsolution
Surfactant solution
Mixture
Mixture
Separation
Separationby
bycentrifugation
centrifugation
Olein Pseudo
Pseudo-solution
-solution ofofstearin
stearin
Washing Separation
Separation by heating
heating
Olein
Surfactant
Surfactant
Stearin solution
solution
emulsion provides saturated fatty acids and surfactant solution. This separation
uses an Alfa-Laval ‘Hyfran’ system. This process is simple, compact safe and,
above all, the surfactant solution wets all the solid particles, irrespective of the
size and form of the crystals.
Table 7. Comparison of fractionation processes for the case of palm oil (Bockisch,
1998).
Advantages Disadvantages
References
Abell, AD and Edmonds, MK (2004) The Wittig and related reactions. In: Organophospho-
rus reagents: a practical approach in chemistry, Oxford University Press, Oxford, UK,
99–127.
Anderson, RJ and Henrick, CA (1975) Stereochemical control in Wittig olefination synthesis.
Preparation of the pink bollworm sex pheromone mixture, Gossyplure. J. Am. Chem.
Soc., 97, 4327–4334.
Barve, JA and Gunstone, FD (1971) Fatty acids. 33. Synthesis of all the octadecynoic acids
and all the trans-octadecenoic acids. Chem. Phys. Lipids, 7, 311–323.
Berdeaux, O, Voinot, L, Angioni, E, Juaneda, P and Sebedio, JL (1998) A simple method of
preparation of methyl trans-10,cis-12- and cis-9,trans-11-octadecadienoates from me-
thyl linoleate. J. Am. Oil Chem. Soc., 75, 1749–1755.
Bernas, A, Kumar, N, Maki-Arvela, P, Holmbom, B, Salmi, T and Murzin, DY (2004)
Heterogeneous catalytic production of conjugated linoleic acid. Organic Process Re-
search & Development, 8, 341–352.
Bockisch, M (1998) Fats and oils handbook, AOCS Press, Champaign, Illinois, USA.
Boden, RM (1975) A mild method for preparing trans-alkenes; crown ether catalysis of the
Wittig reaction. Synthesis, 784.
Chatgilialoglu, C, Samadi, A, Guerra, M and Fischer, H (2005) The kinetics of Z/E
isomerization of methyl oleate catalyzed by photogenerated thiyl radicals. Chemphyschem,
6, 286–291.
Cui, JG, Lin, CW, Zeng, LM and Su, JY (2002) Synthesis of polyhydroxysterols (III):
synthesis and structural elucidation of 24-methylenecholest-4-en-3β,6α-diol. Steroids,
67, 1015–1019.
Dehmlow, EV and Barahona-Naranjo, S (1981) Applications of phase transfer catalysis. Part
17. Wittig reactions in various two-phase systems. J. Chem. Res., Synopses, 5, 142.
DeJarlais, WJ and Emken, EA (1978) Synthesis of di-, tetra-, and hexadeuterated 11-
octadecenoates. Journal of Labelled Compounds and Radiopharmaceuticals, 15, 451–460.
Demmerle, RL (1947) Emersol process. Journal of Industrial and Engineering Chemistry,
39, 126–131.
Duffy, PE, Quinn, SM, Roche, HM and Evans, P (2006) Synthesis of trans-vaccenic acid and
cis-9-trans-11-conjugated linoleic acid. Tetrahedron, 62, 4838–4843.
Dupin, JFE and Chenault, J (1985) Phase transfer catalyzed Wittig-Horner synthesis.
Preparation of hydroxycinnamic esters from ortho hydroxy aromatic aldehydes. Obten-
tion of hydroxycinnamic acids. Synthetic Communications, 15, 581–586.
Fitzpatrick, JD and Orchin, M (1957) The selenium catalyzed cis-trans isomerization of 9-
octadecenoic (oleic-elaidic) acids. J. Am. Chem. Soc., 79, 4765–4771.
Foglia, TA and Vail, PD (1993) An efficient, large-scale synthesis of triisopentadecanoin.
Organic Preparations and Procedures International, 25, 209–213.
Genard, S and Patin, H (1991) Studies related to fatty acid desaturation. Part I. Preparation
of methyl-substituted nonanols, nonyl bromides and total synthesis of new branched oleic
acids. Bulletin de la Societe Chimique de France, 397–406.
Gibon, V and Tirtiaux, A (2002) Latest trends in dry fractionation. Lipid Technology, 14, 33–
36.
Gibson, TW and Strassburger, P (1976) Sulfinic acid catalyzed isomerization of olefins. J.
Org. Chem., 41, 791–793.
Gravier-Pelletier, C, Dumas, J, Le Merrer, Y and Depezay, JC (1990) Methods for the total
synthesis of acyclic hydroxylated fatty acids. Prog. Lipid Res., 29, 229–276.
Hagemann, JW, Tallent, WH, Barve, JA, Ismail, IA and Gusntone, FD (1975) Polymorphism
in single-acid triglycerides of positional and geometric isomers of octadecenoic acid.
J. Am. Oil Chem. Soc., 52, 204–207.
102 TRANS FATTY ACIDS IN HUMAN NUTRITION
Hsu, N, Diosady, LL and Rubin, LJ (1989) Catalytic behavior of Palladium in the hydrogena-
tion of edible oils. II. Geometrical and positional isomerization characteristics. J. Am. Oil
Chem. Soc., 66, 232–236.
Klein, D, Ernst, H and Koppenhofer, J (2002) Process for the preparation of cis-6-
hexadecenoic acid. US patent 2002/0107412 A1.
Kodali, DR, Atkinson, D, Redgrave, TG and Small, DM (1987) Structure and polymorphism
of 18-carbon fatty acyl triacylglycerols: effect of unsaturation and substitution in the 2-
position. J. Lipid Res., 28, 403–413.
Kossmehl, G and Nuck, R (1979) Phase-transfer-catalyzed Wittig reactions without base
addition. Chemische Berichte, 112, 2342–2346.
Labelle, M, Falgueyret, JP, Riendeau, D and Rokach, J (1990) Synthesis of two analogues of
arachidonic acid and their reactions with 12-lipoxygenase. Tetrahedron, 46, 6301–6310.
Larock, RC, Dong, X, Chung, S, Reddy, CK and Ehlers, LE (2001) Preparation of conjugated
soybean oil and other natural oils and fatty acids by homogeneous transition metal
catalysis. J. Am. Oil Chem. Soc., 78, 447–453.
Le Bigot, Y, Delmas, M and Gaset, A (1982) A simplified Wittig synthesis using solid/liquid
transfer processes. II. The use of potassium carbonate for the synthesis of alkenes from
aromatic and aliphatic aldehydes. Synthetic Communications, 12, 107–112.
Le Pivert, M, Poisson, S, Laur, J, Coustille, JL and Pages, X (1995) Synthesis of a branched
fatty acid triglyceride at the laboratory and pilot plant level. Oleagineux, Corps Gras,
Lipides, 2, 369–374.
Litchfield, C, Harlow, RD, Isbell, AF and Reiser, R (1965) Cis-trans isomerization of oleic
acid by nitrous acid. J. Am. Oil Chem. Soc., 42, 73–78.
Märkl, G and Merz, A (1973) Styrene derivatives by aromatic aldehyde reaction with
nonstabilized phosphonium ylides in an aqueous system. Synthesis, 5, 295–297.
Maryanoff, BE and Reitz, AB (1989) The Wittig olefination reaction and modifications
involving phosphoryl-stabilized carbanions. Stereochemistry, mechanism, and selected
synthetic aspects. Chemical Reviews, 89, 863–927.
Mikolajczyk, M, Grzejszcak, S, Midura, W and Zatorski, A (1976) Organosulfur compounds.
X. Horner-Wittig reactions in a two-phase system in the absence of a typical phase-
transfer catalyst. Synthesis, 6, 396–398.
Mouloungui, Z, Candy, L and Destaillats, F (2007) A simplified Wittig method for large scale
molecule production. Unpublished results.
Mouloungui, Z, Delmas, M and Gaset, A (1989) The Wittig-Horner reaction in weakly
hydrated solid/liquid media: structure and reactivity of carbanionic species formed from
ethyl (diethyl phosphono)acetate by adsorption on solid inorganic bases. J. Org. Chem.,
54, 3936–3941.
Moussaoui, Y, Said, K and Ben Salem, R (2006) Anionic activation of the Wittig reaction
using a solid-liquid phase transfer: examination of the medium-, temperature-, base- and
phase-transfer catalyst effects. ARKIVOC, 1–22.
Nichols, PL, Jr., Herb, SF and Riemenschneider, RW (1951) Isomers of conjugated fatty
acids. I. Alkali-isomerized linoleic acid. J. Am. Chem. Soc., 73, 247–252.
Nicolaou, KC, Harter, MW, Gunzner, JL and Nadin, A (1997) The Wittig and related
reactions in natural product synthesis. Liebigs Annalen/Recueil, 1283–1301.
Odinokov, VN (1999) Olefination of aliphatic aldehydes in the synthesis of mono-, di-, and
trienic pheromones of linear structure. Chemistry of Natural Compounds, 35, 260–285.
Pascariu, A, Ilia, G, Bora, A, Iliescu, S, Popa, A, Dehelean, G and Pacureanu, L (2003) Wittig
and Wittig-Horner reactions under phase transfer catalysis conditions. Central European
Journal of Chemistry, 1, 491–542.
Piechucki, C (1974) Horner reaction in an aqueous two-phase system using phase-transfer
catalysis. Synthesis, 869–870.
CHEMICAL SYNTHESIS OF MONOUNSATURATED TRANS FATTY ACIDS 103
Prakash, C, Saleh, S, Sweetman, BJ, Taber, DF and Blair, IA (1989) A synthon for C-20
trideuterated eicosanoids: preparation of [2H3]-arachidonic acid. J. Labelled Compd.
Radiopharm., 27, 539–551.
Samadi, A, Andreu, I, Ferreri, C, Dellonte, S and Chatgilialoglu, C (2004) Thyil radical-
catalyzed isomerization of oils: an entry to the trans lipid library. J. Am. Oil Chem. Soc.,
81, 753–758.
Sgoutas, DS and Kummerow, FA (1969) Cis-trans isomerization of unsaturated fatty acid
methyl esters without double bond migration. Lipids, 4, 283–287.
Singh, G (1976) Analysis of crystallization systems with applications to continuous fractional
crystallization of fatty acid triglycerides AIChE Symposium Series, 72, 100–109.
Sinisterra, JV, Barrios, J, Mouloungui, Z, Delmas, M and Gaset, A (1991) New cesium
carbonate (Cs2CO3) catalyst for the Wittig-Horner synthesis of acrylates in interfacial
solid liquid conditions. Bulletin des Societes Chimiques Belges, 100, 267–275.
Smith, KW, Cain, FW, Favre, L and Talbot, G (2007) Liquid-liquid phase separation in
acetone solutions of palm olein: implications for solvent fractionation. Eur. J. Lipid Sci.
Technol., 109, 350–358.
Snyder, JM and Scholfield, CR (1982) Cis-trans isomerization of unsaturated fatty acids with
p-toluenesulfinic acid. J. Am. Oil Chem. Soc., 59, 469–470.
Tagaki, W, Inoue, I, Yano, Y and Okonogi, T (1974) Wittig reaction in benzene-aqueous
alkaline solution. Tetrahedron Letters, 30, 2587–2590.
Thompson, SK and Heathcock, CH (1990) Effect of cation, temperature, and solvent on the
stereoselectivity of the Horner-Emmons reaction of trimethyl phosphonoacetate with
aldehydes. J. Org. Chem., 55, 3386–3388.
Timms, RE (2005) Fractional crystallisation — the fat modification process for the 21st
century. Eur. J. Lipid Sci. Technol., 107, 48–57.
Tranchepain, I, Le Berre, F, Dureault, A Le Merrer, Y and Depezay, JC (1989) Total
enantiospecific syntheses of 13(S)-hydroxy 9Z, 11E-octadecadienoic (coriolic) acid and
13(S)-N-tosylamino analogue. Tetrahedron, 45, 2057–2065.
Tucker, WP, Tove, SB and Kepler, CR (1965) The synthesis of 11,11-Diteurerooleic acid.
Journal of Labelled Compounds, 7, 137–143.
Tyrwhitt-Walker, GTG (1985) Unsaturated carboxylic acid. GB patent 2153826.
Vedejs, E and Peterson, MJ (1994) Stereochemistry and mechanism in the Wittig reaction.
Topics in Stereochemistry, 21, 1–157.
Vedejs, E and Snoble, KAJ (1973) Direct observation of oxaphosphetanes from typical Wittig
reactions. J. Am. Chem. Soc., 95, 5778–5780.
Villieras, J and Rambaud, M (1983) Wittig-Horner reactions in heterogeneous media. 2. A
convenient synthesis of α,β˜-unsaturated esters and ketones using weak bases in water.
Synthesis, 300–303.
Vinczer, P, Juvancz, Z, Novak, L and Szantay, C (1988) Synthesis of pheromones. IV.
Chemistry of the Wittig reaction. I. Effects of reaction conditions on the stereoselectivity
and yield of the Wittig reaction. Acta Chimica Hungarica, 125, 797–820.
Wolff, RL (1992) Resolution of linolenic acid geometrical isomers by gas-liquid chromatog-
raphy on a capillary column coated with a 100% cyanopropyl polysiloxane film
(CPTMSil 88). J. Chromatogr. Sci., 30, 17–22.
Yang, T-S and Liu, T-T (2004) Optimization of production of conjugated linoleic acid from
soybean oil. J. Agric. Food Chem., 52, 5079–5084.
Zalewski, R and Kokocinske, H (1989) Non-hierarchical classification of organic solvents
using characteristic vector analysis of physical and empirical solvent parameters. Journal
of Physical Organic Chemistry, 2, 232–242.
CHAPTER 5
Analysis of trans fatty acids of partially
hydrogenated vegetable oils and dairy products
1
Nutrition Research Division, Food Directorate, Health Products and Food
Branch, Health Canada, Ottawa, Ontario, Canada
2
Nestlé Research Center, Lausanne, Switzerland
A. Introduction
It is now well recognized that the consumption of trans fats, especially
industrially produced trans fats via partial hydrogenation of vegetable oils
(PHVO), increases the risk of coronary heart disease and that trans fats may do
even more harm than saturated fats (for a review see Mozaffarin et al., 2006).
During the last five years, professional health organizations have taken notice
of the negative effects of industrially produced trans fats and some govern-
ments agencies have taken actions to reduce the trans content in processed
foods. In 2002, the Panel on Macronutrients of the US National Academies of
Science, Institute of Medicine, recommended that trans fat consumption be as
low as possible (Institute of Medicine, 2002). In recent years, the Food and
Agricultural Organization/World Health Organization (FAO/WHO) (2002)
and the American Heart Association (Lichtenstein et al., 2006) have recom-
mended that the trans fat content in the human diet be less than 1% of total
energy. In 2003, Denmark became the first country to set an upper limit on the
percentage of trans fats in foods, prohibiting the sale of foods containing more
than 2% industrially produced trans (as percent total fat) (Stender & Dyerberg,
2003). In December 2005, Canada became the first country to regulate the
mandatory labelling of trans fats on pre-packaged foods (Health Canada,
2003). Subsequently, the USA introduced mandatory declaration of trans fats
in foods (Food and Drug Administration, 2006).
These actions aimed at reducing the trans fat content in foods have prompted
a renewed interest in reliable analytical methods for accurate determination of
the total trans fat content, the proportions of individual trans isomers and the
total fatty acid profile of dietary fats. Several good methods are now available
for the determination of trans fats. The advances in the methods up to the mid
1990s were comprehensively reviewed by Firestone & Sheppard (1992), the
105
106 TRANS FATTY ACIDS IN HUMAN NUTRITION
For pure oil samples, such as refined vegetable oils, which contain primarily
triacylglycerols with no other lipid classes (free fatty acids, phospholipids,
sterol esters, wax esters etc.), base-catalysed transesterification is recom-
mended for simplicity and speed (Christopherson & Glass, 1969; Christie,
1989): sodium or potassium hydroxide or methoxide in methanol are the most
commonly used transesterification bases. Treatment of 19 parts of 10% solu-
tion of the pure fat sample in pertroleum ether or hexane with 1 part of either 2N
methanolic potassium hydroxide or sodium hydroxide at room temperature
results in the almost instantaneous formation of FAME. The solution may be
analysed by GC at once.
stipulated in the AOAC 996.06 (AOAC 2005a) could isomerize CLA isomers
present in dairy products. Therefore, this official method is not suitable for
analysis of ruminant fat containing CLA isomers. Special procedures, which
are discussed in section B of this chapter, need to be used.
The AOAC official method 996.01 also meets the current definition of fat for
food labelling purposes (AOAC, 2005b). This official method is similar to
AOAC 996.06, but is intended for cereal products containing 0.5–13% total fat.
The primary difference between the two official methods is in the recom-
mended internal standard; AOAC 996.01 recommends tritridecanoin
(triacylglycerol 13:0) as the internal standard, whereas AOAC 996.06 recom-
mends triacylglycerol 11:0. Considering the lower volatility of 11:0 FAME
compared to 13:0 FAME, it might be an advantage to use triacylglycerol 13:0
as the internal standard. Ali et al. (1997) demonstrated that the AOAC acid
hydrolysis/GC procedure (996.01) is not only applicable to cereal products, but
also to many other foods, including salad oils, partially hydrogenated fats,
margarines, fish fillets and chocolate bars.
and unnatural cis isomeric fatty acids. The position of the double bond of these
dietary trans 18:1 isomers, counted from the carboxylic carbon, usually varies
from 4 to 16 carbon atoms of the fatty acid molecule. Very often in PHVO, the
trans-18:1 isomers form a Gaussian distribution that centres around the 9 or 10
double bond. The distribution depends on the fatty acid composition of the
starting vegetable oils and on the extent of hydrogenation. For example, in
mildly hydrogenated canola oil, elaidic acid (i.e., 9t-18:1) is the major isomer
whereas 10t-18:1 is the major isomer in moderately and heavily hydrogenated
canola oils (Table 1). On the other hand, 10t-18:1 and vaccenic acid (11t-18:1)
are the major isomers in mildly and moderately hydrogenated soybean oils and
9t-18:1 is the major isomer in heavily hydrogenated soybean oil (Table 1).
In addition to the trans-18:1 isomers, PHVO contain several cis-octadece-
noic isomers (cis-18:1), whose double bond position generally ranges from 6 to
16 (Table 2). The naturally occurring cis isomer, namely oleic acid (i.e., 9c-
18:1) is always the predominant isomer followed by 10c-18:1 and cis-vaccenic
acid (11c-18:1).
Very often, the human diet contains a number of positional and geometrical
isomers of linoleic (18:2n–6) and α-linolenic acids (18:3n–3). Both partially
hydrogenated and non-hydrogenated unsaturated vegetable oils are sources of
these isomers (Ratnayake & Pelletier, 1992; Ackman et al.,1974). They are
often detected in large quantities in mildly hydrogenated vegetable
110 TRANS FATTY ACIDS IN HUMAN NUTRITION
Table 3. Composition (wt% of total fatty acids) of linoleic acid isomers in some
hydrogenated canola oil basestocks.
oils (up to 6% of total fatty acids) and refined, but unhydrogenated oils (up to
3%), whereas they are hardly detectable in heavily hydrogenated oils. PHVO
contain at least fifteen different isomers of linoleic acid (Ratnayake & Pelletier,
1992), the major ones being 9c,13t-18:2, 9c,12t-18:2 and 9t,12c-18:2 (Table 3).
The isomers present in non-hydrogenated oils or in many common dietary oils
are the results of exposure of the fat to some form of heat treatment such as
steam deodorization, stripping during refining of oils (Ackman et al., 1974), or
simple heating in deep fat frying (Grandgirard et al., 1984). In these processes,
the double bonds do not shift in position, but isomerize from cis to trans
ANALYSIS OF TRANS FATTY ACIDS OF PHVO 111
9c
9t
7c 12c
12t 11t
11c
7t
22.5 25 min
Figure 1. Gas chromatogram of a standard FAME mixture of 7t, 9t-, 11t-, 12t-, 7c-, 9c-,11c- and 12c-
18:1. Analysis on a SP 2560 fused silica capillary column (100 m × 0.25 mm i.d × 20 m film thickness),
operated isothermally at 180°C. Hydrogen carrier gas flow rate 1 ml/min.
18:0
9c-18:1
10t-18:1
11t-18:1
18:2n-6
(13+14)t +(6-8)c-18:1
10c+15t-18:1
6t-8t-18:1
9t-18:1
11c-18:1
12c-18:1
9c,13t+8t,12c-18:2
12t-18:1
14c-18:1
15c-18:1
16t-18:1
9c,12t-18:2
9c,15c-18:2
9t,12c-18:2
9t,12t-18:2
13c-18:1
4t-18:1
5t-18:1
tt-18:2
24 26 28 30 min
Figure 2. The C18 region of gas chromatogram of FAME from a margarine sample analysed on a SP
2560 fused silica capillary column (100 m × 0.25 mm id × 20 m film thickness), operated isothermally
at 180°C. Hydrogen carrier gas, flow rate 1 ml/min.
counts
CP-S11 88 (100 m × 0.25 mm)
14000 180°C isothermal
18:0 9
18:2N-6
12000
10000 10
11
12
8000 6 13
5 21
14
8 26 18:3N-3
15 20
16 19 25
6000 4 7 27
17 24
3
23 29
4000 18 34+20:1
22 20:0 32
1 2 28 30 31 33
2000
24 26 28 30 32 34 36 38 40 min
Figure 3. The C18 region of a GC chromatogram of FAME from a margarine sample analysed using a
CP Sil-88 capillary column (100 m × 0.25 mm i.d. × 20 µm film thickness), operated isothermally at
180°C. Hydrogen carrier gas, flow rate 1 ml/min. Peak identification: 1, 4t-18:1; 2, 5t-18:1; 3, (6t-8t)-
18:1; 4, 9t-18:1; 5, 10t-18:1; 6, 11t-18:1; 7,12t-18:1; 8, (13t+14t)-18:1 + (6c-8c)-18:1; 9, 9c-18:1; 10,
(15t+10c)-18:1; 11, 11c-18:1; 12, 12c-18:1; 13, 13c-18:1; 14, 16t-18:1; 15,14c-18:1; 16, 15c-18:1; 17-
20, tt-18:2; 21, 9t,12t-18:2; 22, (9c,13t+8t,12c)-18:2; 23, 9c,12t-18:2; 24, 8c,13c-18:2; 25, 16c-18:1; 26,
9t,12c-18:2; 27, (9t,15c+10t,15c)-18:2; 28, 9c,13c-18:2; 29, 9c,14c-18:2; 30, 9c,15c-18:2; 31, 9t,12c,15t-
18:3; 32, 9c,12c,15t-18:3; 33, 9c,12t,15c-18:3, 34; 9t,12c,15c-18:3 (Ratnayake, 2004) (Reproduced with
the kind permission of J. Assoc. Off. Analyt. Chem. Int.).
8000 25
5
6 18:3n-3
12
6000 4 8 13+14 19
7 11 18 26
3
20 25 27 34+11c-20:1
17 30
4000 16 24
21 28 29 20:0
1 2 15 23 31 32 33
22
2000
20 22 24 26 28 min
Retention Time (min)
Figure 4. The C18 region of a GC chromatogram of FAME from a margarine sample analysed using a
SP 2560 capillary column (100 m × 0.25 mm i.d. × 20 µm film thickness), operated isothermally at
190°C. Hydrogen carrier gas flow rate 1 ml/min. For peak identifications see Figure 3. (Ratnayake, 2004)
(Reproduced with the kind permission of J. Assoc. Off. Analyt. Chem. Int.).
the 16t-18:1 (peak 14) now overlaps with the 14c-18:1 isomer (peak 15). The
only advantage at lower operating temperature is the improved separation of
the individual trans-18:1 isomers compared to that at 180°C or other higher
temperatures. These overlap problems are also seen with column temperature
programme analysis (Figure 8).
12000
11
10000
5 6 12 18:3n-3
13+14
8000
4 15
7 16 28
3 8 17 30
6000 27 29
18 20 26
25 34
19
11c-20:1
4000 21 24
2 23 20:0 31
1 22 32 33
2000
20 22 24 26 28 30 min
Retention time (min)
Figure 5. The C18 region of a GC chromatogram of FAME from a margarine sample analysed using a
CP Sil 88 capillary column (100 m × 0.25 mm i.d. × 20 µm film thickness), operated isothermally at
190°C. Hydrogen carrier gas flow rate 1 ml/min. For peak identifications see Figure 3. (Ratnayake, 2004)
(Reproduced with the kind permission of J. Assoc. Off. Analyt. Chem. Int.).
7000 18:2n-6
6000
5 18:3n-3 + 11c-20:1
6
4 10
5000 8 14+15 21
11 20
7
19
12 18
3 17 26
4000 13 25
16 24 27 20:0 32 33
22 28 29 30 31 34
1 2 23
3000
7000
6000 18:3n-3
56
5000 8 10
4 21
7 11 20 11c-20:1
14+15
3 12 19 28
4000 18 25
13 17 27
24
30
16 23 26
3000 34
12 29 20:0
31
32
22 33
2000
35 40 45 50 55 min
Retention time (min)
Figure 7. The C18 region of a GC chromatogram of FAME from a margarine sample analysed using a
CP Sil 88 capillary column (100 m × 0.25 mm i.d. × 20 µm film thickness), operated isothermally at
170°C. Hydrogen carrier gas flow rate 1 ml/min. For peak identifications see Figure 3. (Ratnayake, 2004)
(Reproduced with the kind permission of J. Assoc. Off. Analyt. Chem. Int.).
(13+14)t-18:1+ (6-8)c-18:1
12c-18:1 10c-18:1+15t-18:1
9c-18:1
10t-18:1
FID1
11t-18:1
PA
11c-18:1
18:0
9t-18:1
20
6t-8t-18:1
18
12t-18:1
16
14
14c-18:1
16t-18:1
15c-18:1
12
13c-18:1
10
5t-18:1
4t-18:1
8
6
4
38 40 42 44
Figure 8. The C18 region of a GC chromatogram of FAME from a margarine sample analysed using a
SP 2560 capillary column (100 m × 0.25 mm i.d. × 20 µm film thickness). Analysis performed by the
following column temperature program. Initial temperature 45°C (hold 4 min), programmed to 175°C at
13°C/min, hold at 175°C for 27 min, programmed to 215°C at 4°C/min, hold at 215°C for 3 min.
Hydrogen carrier gas flow rate 1 ml/min.
118 TRANS FATTY ACIDS IN HUMAN NUTRITION
18:2n-6
9c-18:1
11c-18:1
9c,12t-18:2
9t,12t-18:2
9t,12c-18:2
13t-18:1
25 30 min
min
Figure 9. The 18:2 region of a GC chromatogram of FAME from a refined canola oil sample analysed
using a SP 2560 capillary column (100 m × 0.25 mm i.d. × 20 µm film thickness), operated isothermally
at 180°C. Hydrogen carrier gas flow rate 1 ml/min.
9c,13t-18:2 (major) + 8t,12c-18:2 (minor)
13c-18:1
18:2n-6
9t,15c+10t,15c-18:2
20:0
14c-18:1
9c,12t-18:2
16t-18:1
16c-18:2
9c,13c-18:2
15c-18:1
9t,12c-18:2
9c,14c-18:2
8c,13c-18:2
9c,15c-18.2
8c,13t-18:2
tt-18:2
ct/tc-18:2
9t,12t-18:2
28 30 32 min
Figure 10. The 18:2 region of a GC chromatogram of FAME from a margarine sample analysed using
a SP 2560 capillary column (100 m × 0.25 mm i.d. × 20 µm film thickness), operated isothermally at
180°C. Hydrogen carrier gas flow rate 1 ml/min.
ANALYSIS OF TRANS FATTY ACIDS OF PHVO 119
18:3n-3
20:0
9c,12c,15t-18:3
9t,12c,15c-18:3
9c,12t,15c-18:3
11c-20:1
9t,12c,15t-18:3
Figure 11. The 18:3 region of a GC chromatogram of FAME from a refined canola oil sample analysed
using a SP 2560 capillary column (100 m × 0.25 mm i.d. × 20 µm film thickness), operated isothermally
at 180°C. Hydrogen carrier gas flow rate 1 ml/min.
190°C or with column temperature programs, these two isomers are not well
resolved. In addition, the unavailability of a commercial 9c,13t-18:2 standard
means misidentification of this isomer is common in such instances. It is
recommended that a FAME mixture from a PHVO with a well-established fatty
acid composition be used as the secondary standard, especially for the identi-
fication of the peaks in the 18:2 region of the chromatogram. Such a secondary
standard mixture is now available from the American Oil Chemists’ Society
(Champaign, Illinois, USA).
The four common geometric isomers of α-linolenic acid, i.e., 9t,12c,15t-
18:3; 9c,12c,15t-18:3; 9c,12t,15c-18:3 and 9t,12c,15c-18:3, which are also
commonly found in refined liquid vegetable oils give peaks that can be readily
recognized in GC analyses on cyanosilicone capillary columns and are eluted
from the column in the order stated above (Figure 11).
temperature. Generally, on the SP-2560 capillary column, as for the 18:1 cis
and trans isomer separation, isothermal operation at 180°C gives the best
separation of 11c-20:1 from α-linolenic acid and its geometric isomers
(Ratnayake et al., 2002) (Figure 11). At column temperature above 180°C, the
11c-20:1 isomer co-elutes with 9t,12c,15c-18:3 (Figure 4) and below 180°C it
overlaps with α-linolenic acid (Figure 6). In contrast, the CP-Sil 88 column
operated at 180°C does not cleanly separate 11c-20:1 from the α-linolenic
isomers, but co-elutes it with 9t,12c,15c-18:3 (Figure 3). However, isothermal
operations of CP-Sil 88 column at 10°C above or below 180°C give baseline
separation of 11c-20:1 from α-linolenic acid and its geometrical isomers. At
190°C, 11c-20:1 elutes before 9t,12c,15c-18:3 and 9c,12t,15c-18:3 isomers
(Figure 5), whereas at 170°C, it elutes after 9t,12c,15c-18:3 but before
α-linolenic acid (Figure 7). It is recommended that pure standards of 11c-20:1
and 18:3n–3 be used for correct location of these fatty acids in the analysis of
dietary fat samples.
d. Effect of the type of GC carrier gas and flow rate on cis and trans
isomer resolution and fatty acid quantification
Recently, Ratnayake et al. (2006) tested the performances of hydrogen and
helium as carriers at different flow rates on the separation and quantification of
FAME of dietary fats containing TFA. Their data suggested that hydrogen
carrier gas at 1.0 ml/min and isothermal column temperature operation at
180°C are optimal for the analysis of fatty acid profiles of vegetable and animal
fats and oils containing TFA.
Table 4 shows the normalized area percentages concentration for total trans,
trans-18:1, trans-18:2, trans-18:3, conjugated linoleic acid (CLA), 16:0 and
18:0 of a margarine sample when analysed on different columns (100 m SP-
2560 and CP-Sil 88) with different carrier gases (hydrogen and helium) and
flow rates (0.6, 0.8 and 1.0 ml/min). The results were essentially identical at all
different conditions for both columns. Even though the quantification differ-
ences were minimal, the data demonstrate a few advantages of using hydrogen
as the carrier gas at a flow rate of 1 ml/min. The use of hydrogen at the same
flow rate as helium results in shortening the GC run time by 20 minutes. For
example, on the 100 m SP-2560 column with hydrogen carrier and a flow rate
of 1 ml/min, the 24:0 peak, usually the last to elute in many dietary fats, comes
off in about 60 minutes. With helium as the carrier gas at the same flow rate, the
same peak elutes at over 80 minutes. This time-saving could be of benefit to
most lipid analytical laboratories where run time is very important. In addition,
hydrogen carrier gas gives a better resolution of the critical pairs of 13t+14t-
18:1 and 9c-18:1, 16t-18:1 and 14c-18:1 and of 11c-20:1 and 9c,12c,15c-18:3.
Furthermore, hydrogen produced sharper peaks than helium. It was also noted
that there was no loss of resolution between flow rates of 0.6, 0.8 and 1.0 ml/
min in the early-eluting peaks for either hydrogen or helium (Ratnayake et al.,
ANALYSIS OF TRANS FATTY ACIDS OF PHVO 121
Table 4. Effect of helium and hydrogen carrier gas flow rates on the fatty acid
composition (% total fatty acids) of a partially hydrogenated margarine sample on 100 m
CP-Sil 88 and SP-2560 columns (Ratnayake et al. 2006). CLA, conjugated linoleic acids.
2006). However, for the separation of the peaks in the 18:3 region, a flow rate
1.0 ml/min provided the most satisfactory separation.
Figure 12. AgNO3-TLC fractionation of FAME from a margarine. Development in toluene. Bands
were visualized after spraying with 0.1% solution of 2',7'-dichlorofluorescein in ethanol and examining
under UV light (234 nm).
treat the trans-18:1 isomers with the double bond closer to the carboxyl group
(from Δ6 to Δ11) as the internal standard. This is possible because, as discussed
earlier, the 6t-18:1 to 11t-18:1 isomers are always well resolved from the cis-
18:1 isomers. In this method, the proportion of the trans-18:1 isomers from Δ12
to Δ16 that overlap with the cis-18:1 isomer peak on capillary GC is calculated
by comparing the 18:1 region of the GC chromatogram of the isolated trans-
18:1 with that of the parent FAME prior to AgNO3-TLC fractionation. This
calculation is done with respect to the well-separated trans-18:1 isomers (i.e.,
sum of 6t-18:1 to 11t-18:1). The total trans-18:1 content is then calculated by
summing the proportion of the trans-18:1 isomers (12t-18:1 to 16t-18:1) that
overlaps with the cis isomers with the well-separated trans-18:1 isomers (from
6t-18:1 to 11t-18:1). This also allows calculation of the total cis-18:1 content.
This approach eliminates the errors resulting from sample application, scraping
losses, incomplete extraction, weighing of small quantities of internal standard
and isolated bands. The AgNO3-TLC fractionation offers several other advan-
tages, especially the separation of the linoleic acid isomers into trans,trans and
cis,trans fractions (see Figure 12) that can be used to provide confirmatory
evidence for the identification of the 18:2 isomers. Ratnayake & Pelletier
(1992) used AgNO3-TLC fractionation in combination with GC mass
spectrometry (GC-MS) to determine the structures of linoleic acid isomers of
PHVO.
As an alternative to glass plates, pre-coated plastic-backed sheets with a
0.2 mm layer of silica gel impregnated with 10% AgNO3 in acetonitrile can be
124 TRANS FATTY ACIDS IN HUMAN NUTRITION
12t 13t+14t
9t
6t+7t+8t
18:0 15t
16t
4t 5t
24 26 min
11c
10c
13c
6c+7c+8c 15c
12c 14c
24 26 28 min
Figure 13. GC chromatograms of the trans- and cis-18:1 AgNO3-TLC bands of the margarine sample
shown in Figure 13. Analysis on a SP 2560 fused silica capillary column (100 m × 0.25 mm id × 20 µm
film thickness), operated isothermally at 180°C. Hydrogen carrier gas flow rate 1 ml/min.
used (Uberth & Henninger, 1992). The development of the sheet is similar to
that of glass plates with n-hexane-diethyl ether (9:1, v/v), which is followed by
spraying with 0.05% rhodamine B in ethanol. The separated bands are cut off
with a pair of scissors, and the cuttings are placed in a small bottle containing
5 ml diethyl ether. Extraction is complete in 30 minutes. These AgNO3-sheets
give separation of trans and cis isomers similar to that of conventional AgNO3-
TLC fractionation.
Figure 14. Analysis of partially hydrogenated vegetable oil FAME by Ag+-HPLC. Sample size, 20 µg;
flow rate, 1.0 ml/min; solvent system, 0.15% acetonitrile in hexane; ultraviolet detection at 206 nm.
Fractions: A, saturates; B, trans-18; C, cis-18:1; D, 18:2 (Adlof et al. 1995). (Reproduced with kind
permission of J. Am. Oil Chem. Soc.).
(Adlof et al., 1995). The separation was performed with a solvent system of
0.15% acetonitrile in hexane at a flow rate of 1 ml/min and UV detection at
204 nm. The separation is completed in 45 minutes. The saturated, trans-18:1
isomers, cis-18:1 isomers and the 18:2 were completely separated without any
overlaps. Collection of the trans-18:1 isomers with the saturated fatty acid
fraction and subsequent analysis by GC would allow quantitation of all the
trans-18:1 isomers with respect to the total saturated fatty acids or just with
respect to methyl stearate. By using 0.08% acetonitrile in hexane and injecting
a very small sample size (< 0.5 µg), individual cis and trans-positional FAME
isomers can be resolved (Figure 15, Adlof et al., 1995). The elution order of the
positional isomers is the reverse of those obtained by capillary GC. The high
delta value isomers elute before the low delta value isomers in Ag-HPLC.
The Ag+-HPLC method offers some advantages over AgNO3-TLC, includ-
ing rapid analysis (45 min/run) and complete separation of cis- and trans-18:1
isomers with no cross contamination. However, a severe draw back is the
difficulty of obtaining consistent separation pattern from one day to another.
126 TRANS FATTY ACIDS IN HUMAN NUTRITION
Δ8 + Δ9
Δ10
Δ11
Δ8 + Δ9
Δ12
Δ13 Δ10
Δ11
Δ12
Δ14
Δ13
Figure 15. Analysis of partially hydrogenated vegetable oil FAME by Ag+-HPLC. Sample size,
0.4 µg; flow rate, 1.0 ml/min 0.08% acetonitrile in hexane; ultraviolet detection at 206 nm. Fractions: A,
saturates; B, trans-18:1; C, cis-18:1 (Adlof et al., 1995). (Reproduced by permission of Journal of the
American Oil Chemists’ Society).
Figure 16. Mass spectra of Δ8, Δ9 and Δ10 trans 18:1 DMOX positional isomers; a gap of 12 amu was
observed between the pairs of adjacent ions at m/z 182 and 194; 196 and 208; and 210 and 222,
respectively (Mossoba et al. 1997). (Reproduced by permission of J. Am. Oil Chem. Soc.).
Figure 17. Mass spectra of Δ13 and Δ14 trans 18:1 DMOX positional isomers (Mossoba et al. 1997).
(Reproduced by permission of J. Am. Oil Chem. Soc.).
128 TRANS FATTY ACIDS IN HUMAN NUTRITION
1. Ruminant Fats
Cow’s milk usually contains 3.4 to 5.1% fat (Jensen, 2002). The lipid fraction
is represented by 96 to 98% triacylglycerol. Minor lipid components in milk
include diacylglycerols, monoacylglycerols, phospholipids, free fatty acids,
sterols and sterol esters (Gurr, 2002). Triacylglycerols are present in the form
of small globules surrounded by a thin membrane that is composed of protein
and phospholipids. More than 400 different fatty acids have been identified
(Jensen et al., 1991), with the potential to create over 6 million triacylglycerol
molecular species. This makes the lipid fraction of milk one of the most
ANALYSIS OF TRANS FATTY ACIDS OF PHVO 129
complex matrices known to food science (Jensen, 2002). Milk fatty acids
include saturated fatty acids of different chain-length groups, namely short-
chain, medium-chain, long-chain, and branch-chain fatty acids, and also a
variety of cis-and trans-monounsaturated FA, polyunsaturated FA, geometric
and positional trans-polyunsaturated fatty acids isomers, CLA, hydroxy fatty
acids and cyclic fatty acids (Christie, 1995; Walstra, 1999). Although only 14
fatty acids – including 14:0, 16:0 and 18:1 – represent about 96% of the total
fatty acid content, milk fat also has an exceptionally high content of short-chain
fatty acids, especially butyric acid (4:0) which makes milk fat a distinctive food
product (Christie, 1995; Gurr, 2002). Butyric acid is considered to play a role
in cancer prevention (German, 1999). It is also of considerable commercial
importance as a raw material in the manufacture of esters of lower alcohols
which are used as flavouring agents. Another minor, but very important group
of fatty acids present in ruminant fats is the CLA isomers. Studies conducted
over the last 10 years, especially with small laboratory animals, have shown
that CLA isomers, in particular rumenic acid (9c,11t-18:2), can provide many
health benefits including anti-carcinogenic, anti-obesogenic, and anti-athero-
genic effects (Beppu et al., 2006; Bhattacharya et al., 2006; De La Torre et al.
2006; Belury et al., 2002).
The fatty acid profile of dairy fats depends to a large extent on the diet of the
cow, the vast microflora activity present in the rumen, and the cellular fat
synthesis (Chilliard et al., 2001; 2000). If diets contain a high content of
polyunsaturated fatty acids protected by protein, the final product (milk or
meat) will also contain them (Beauleiu & Palmquist, 1995). Fatty acids such as
trans-18:1 are produced as a result of incomplete bio-hydrogenation of dietary
polyunsaturated fatty acids in the rumen. It has recently been demonstrated that
feeding of cows with grain diets results in milk and meat with unusually high
levels on trans-18:1 fatty acids (Bauman & Griinari, 2003).
Over the last 25 years, a number of good analytical methods have appeared
in the literature for the analysis of fatty acid profile in dietary fats (Albertyn
et al., 1982; Precht & Molkentin 1996; Ratnayake, 1998; Kramer et al., 1999;
Precht & Molkentin, 1999; 2000; Precht et al., 2001; Ratnayake, 2001; Wolff
et al., 2001; Destaillats & Angers, 2002; LeDoux et al., 2002; Wolff and
Precht, 2002; Christie, 2003; Ratnayake, 2004; and Aldai et al., 2005). Some of
these methods, however, are not suitable for analysis of ruminant fats. Because
of the presence of a very complex mixture of a large number of fatty acids, the
analysis of ruminant fats requires special analytical procedures. The objective
of this section is to propose a comprehensive methodology to completely
characterize the fatty acid profile of ruminant fats with enough accuracy and
precision to detect all the TFA and CLA isomers. The methodology is a
combination of complementary techniques, including lipid extraction, meth-
ylation of the extracted lipids to FAME and subsequent analysis of FAME
using various forms of chromatography such as Ag+-TLC, Ag+-HPLC and GC.
130 TRANS FATTY ACIDS IN HUMAN NUTRITION
2. Lipid extraction
There are certain basic principles and precautions that should be followed
during extraction of lipids from dairy products and ruminant meat. The
extraction should ideally be carried out using a solvent system that yields both
neutral and polar lipids. Since CLA is naturally present in both meat and dairy
products, it is also important to avoid acid digestion, which can lead to
isomerization of CLA isomers. When cheese is analysed, samples should be
entirely ground prior to lipid extraction in order to ensure a complete solvent
interaction with the sample. Butter should be melted, homogenized and sub-
sampled before extraction.
Total lipids can be extracted in two stages from dairy samples using a
chloroform/methanol/water mixture (Bligh & Dyer, 1959). Samples are first
well homogenized for 2 minutes in the mono-phase mixture chloroform/
methanol/water (1:2:0.8; by vol.), taking into account the water that is already
present in milk (Kramer et al., 2001). For this reason, adding water last when
preparing the mono- and biphasic systems is suggested. A solvent to sample
ratio greater than 15:1 is recommended to ensure better recovery (Kramer
et al., 2001; Cruz-Hernandez et al., 2006). After the addition of chloroform/
water (1:1; v/v) samples are again homogenized in the bi-phase system
(chloroform/methanol/water; 2:2:1.8, by vol.) for an additional 2 minutes. The
lipid fraction (including neutral and polar lipids) will be partitioned into the
chloroform phase, which is at the bottom of the flask. The lipid/chloroform
phase is collected and the lipids are then recovered by removing the chloroform
using a rotary evaporator as described by Cruz-Hernandez et al. (2006). In
order to obtain a dry lipid sample, the addition of a couple of drops of acetone
is recommended at the end of the extraction to completely remove all traces of
water. Small amounts of solvent can be removed by flushing with nitrogen and
working in a fume hood. Finally, the total lipid content is determined by weight
differences, and lipids should be stored in chloroform at very low temperatures
(preferably below –20°C) until further analysis.
For analysis of meat, the fresh sample should be rinsed immediately with
water and frozen (either on dry ice or dropped into liquid N2), and stored at
–70°C until analysed. Sample preparation is slightly different from dairy
samples. Meat samples should be pulverized at dry ice temperature in order to
restrain the action of lipases and phospholipases during homogenization of
tissues (Kramer and Hulan, 1978, Kramer et al., 1999). Briefly, the stored
frozen meat at –70°C (2 to 5 g) is directly pulverized using a stainless steel
mortar and pestle kept at dry ice temperature. The content is then transferred to
a beaker containing 30 ml chloroform/methanol (2:1) at room temperature. The
mixture is gently homogenized, allowed to stand for one hour, and then filtered
through a sintered glass funnel. Subsequently, the residue is washed two more
times with chloroform/methanol (2:1). The combined solvents are removed
ANALYSIS OF TRANS FATTY ACIDS OF PHVO 131
using a rotary evaporator as described above to obtain total lipids. This same
procedure is also recommended for the extraction of lipids from tissues of
laboratory animals.
Other solvent systems recommended for matrices such as milk or tissues are
those described by Folch et al. (1957). These systems utilize different ratios of
chloroform/methanol/ water or alternatively use mixtures containing either
hexane/isopropanol (Hara & Radin, 1978) or dichloromethane/methanol
(Marmer & Maxwell, 1981; Maxwell et al., 1986).
In meat fat, some lipid classes such as phospholipids and other individual
lipid classes are also important to consider. Analyses of individual lipid classes
could be helpful in understanding different functional properties of meat fat
(Kramer et al.,1998, Yurawecz et al., 1999, Cruz-Hernandez et al., 2006).
3. Methylation
Although base methylations are believed to be the best derivatization method
for the analysis of CLA isomers, other lipids present in the matrix must be taken
into consideration. Base catalysts specifically transesterify acyl lipids, but they
are not capable of transesterifying N-acyl and alk-1-enyl acyl lipids (Kramer &
Zhou 2001; Cruz-Hernandez et al., 2004; Kramer et al., 2004). For example, if
free fatty acids are also present and need to be analysed, then the methylation
has to be carried out using an acid catalyst such as HCl/methanol (Kramer
et al., 2001; Cruz-Hernandez et al., 2006).
Methylation using sodium methoxide, NaOCH3 (e.g., 0.5N methanolic base
#33080, Supelco Inc, Bellefonte, PA, USA) is recommended for the analysis of
CLA-containing lipids because the methylation is complete within 15 to
20 minutes and this reagent does not isomerize CLA isomers (Kramer et al.,
1997). A water-free system is preferred for the methylation of milk fat lipids
(Christie, 1999; Chouinard et al., 1999). Briefly, 2 mg of total fat is added to a
2 ml autosampler vial, the solvent is removed with a stream of N2, 1.7 ml of
hexane and 40 µl of methyl acetate are added and thoroughly mixed before the
addition of 100 µl of NaOCH3. The vial is securely capped, the contents shaken
and allowed to react for 20 minutes at room temperature with occasional
mixing. The vial is cooled at –20°C for 10 minutes, followed by addition of
60 µl of oxalic acid (0.5 g in 15 ml diethyl ether) and thoroughly shaken. The
vial is centrifuged to settle the Na oxalate precipitate, and the upper phase is
passed through a Pasteur pipette (5.75 inch) column containing a glass wool
plug (glass wool previously washed with 1:1 chloroform/methanol and dried),
and a 2 cm bed of anhydrous Na2SO4, directly into a 2 ml autosampler vial. The
FAME solution can be used directly for GC analyses.
Reagents other than NaOCH3, such as TMS-DAM (trimethylsilyl diazo-
methane), can also be used for methylation of lipids containing CLA isomers.
This reagent does not isomerize CLA. However, the primary drawback is that
132 TRANS FATTY ACIDS IN HUMAN NUTRITION
very often this reagent is not pure and, consequently, the FAME produced
should be purified by TLC, which is a time-consuming procedure. When
working with dairy products, purification of the FAME leads to loss of some
fatty acids (Kramer et al., 2001).
Figure 18. A partial GC chromatogram of a dairy fat 18:1 FAME region using a 100 m CP Sil 88
capillary column, hydrogen as a carrier gas, and a typical temperature program from 45 to 215°C.
Figure 19. A partial GC chromatogram of a dairy fat 18:1 FAME region using a 100 m CP Sil 88
capillary column, hydrogen as a carrier gas, and an isothermal temperature at 180°C.
remarkable difference in the way isomers such as 13t to 15t are separated from
9c 18:1 (Precht & Molkentin, 1996 and 2003). Different elution patterns for cis
and trans C18 isomers were recognized when the column temperature was
operated at 150°C versus 130°C (Precht & Molkentin 2003). Effects of column
temperature on the relative order of elution of 18:3 and 20:1 isomers have also
been evaluated (Wolff & Precht, 1998). In Figures 18 and 19 different column
temperatures are compared. Figure 18 shows the typical separation obtained
with the ramped temperature program already described. Figure 19 shows the
typical separation obtained with 180°C isothermal as recommended in this
chapter for PHVO. As mentioned at the beginning of this chapter, 180°C
134 TRANS FATTY ACIDS IN HUMAN NUTRITION
Figure 20. A partial GC chromatogram of a dairy fat 18:1 FAME region using a 100 m CP Sil 88
capillary column, hydrogen as a carrier gas, and a typical temperature program from 45 to 215°C. High
sample load.
isothermal temperature is appropriate for the analysis of the fatty acid profile,
including all the trans isomers of PHVO (Ratnayake, 2001). Isothermal
analysis at 180°C is also suitable for the analysis of trans isomers of milk fat.
However, it may not be appropriate for the complete analysis of the fatty acid
profile of dairy fats. At 180°C, the short-chain fatty acids, especially those from
4:0 to 8:0 would elute with the solvent front. In addition, the CLA isomers are
not adequately separated (Kramer et al., 2001; Cruz-Hernandez et al., 2004).
The GC sample load is an important parameter in GC analysis that is not
usually taken into consideration. The 18:1 cis and trans isomers will resolve
better if a low sample load is used, but a higher sample load is better for the
analysis of 16:1, CLA and polyunsaturated fatty acids. Therefore, any sample
solution that is too concentrated should be appropriately diluted. Generally,
two sample loads are analysed: a 1 µl solution containing about 1 to 2 µg is
injected to determine FAME present at low concentration, and a 3:1 or 5:1
dilution is necessary to gain a better resolution of FAME of the 18:1 region. In
the case of the CLA region, it will not be possible to separate the two major
isomers of CLA (7t,9c- and 9c,11t-CLA). CLA isomers are best analysed by
combining GC analysis with Ag+-HPLC fractionation. The procedure is dis-
cussed later in this section.
Figure 20 shows a chromatogram for the 18:0, 18:1 and 18:2 FAME regions
of a dairy sample. It illustrates why it is not possible to resolve some of the
trans-18:1 isomers from the cis-18:1 isomers at high sample loads. Especially,
the 12t-, 13-14t/6-8c-18:1 isomers overlap with the peak for 9c-, 10c-,15t-18:1.
At low sample load (Figures 18 and 19), there is an improved separation of 13-
14t/6-8c, appearing as a leading shoulder to the 9c peak. In Figures 18 and 19,
ANALYSIS OF TRANS FATTY ACIDS OF PHVO 135
many of the different isomers present in the 18:1 region can be observed. The
predominant trans isomer is 11t.
A B
5
6
Origin
Figure 21. AgNO3-TLC separation of the total milk fat FAME and FAME standards using hexane/
diethyl ether (90:10) as a developing solvent. Section A - Dairy fat FAME: 1, SFA; 2, trans-MUFA; 3,
cis-MUFA; 4, 18:2; 5, PUFA 1; 6, PUFA 2. Section B - FAME Standards: 1, 21:0; 2, 11t-18:1; 3, 9c-
18:1; 4, 18:2n–6; 5, 20:4n–6; 6, 20:5n–3.
Figure 22. A partial GC chromatogram of the trans-and cis-8:1 FAME fractions isolated by AgNO3-
TLC and analysed by a 100 m CP Sil 88 capillary column, hydrogen as carrier gas, and a typical
temperature program from 45 to 215°C.
separation with the total fat (Figure 18) and to identify some of the peaks that
were not there before the separation. The drawback of the short temperature
program is that 6t to 11t and 12t to 14t peaks elute close together (Figure 22,
upper) and it is difficult to separate them. Overlaps are also observed for the cis
isomers (Figure 22, lower). The relative concentration of 9c is too high, and
this was the reason for its co-elution with 10c.
Figure 23. A partial GC chromatogram of the CLA standard FAME region using a 100 m CP Sil 88
capillary column, hydrogen as carrier gas, and a typical temperature program from 45 to 215°C.
ANALYSIS OF TRANS FATTY ACIDS OF PHVO 139
Figure 24. A partial GC chromatogram of the CLA FAME region of a milk fat using a 100 m CP Sil
88 capillary column, hydrogen as carrier gas, and a typical temperature program from 45 to 215°C.
ether, is supplied isocratically at a flow rate of 1.0 ml/min. The mobile phase
should be continuously mixed using a magnetic stirring bar. The diode array
detector is operated at 233 nm for the identification of conjugated FAME, at
205 nm for unsaturated (non-conjugated) FAME, and at 268 nm for conjugated
trienoic acids if present (Cruz-Hernandez et al., 2004). The elution order of
the CLA isomers was described previously (Sehat et al., 1998; Eulitz et
al., 1999; Kramer et al., 1999). The elution order of the CLA isomers by GC
(c/t < c,c < t,t) and Ag+-HPLC (t,t < c/t < c,c) are different. The identification
of the individual CLA isomers has been established previously (Eulitz et al.,
1999; Kramer et al., 1999; Sehat et al., 1998). The columns are cleaned daily
by flushing them with 1% acetonitrile in hexane (v/v) for one hour, followed by
flushing with the mobile phase for one hour.
9c11t
25 30 35 40 min
Figure 25. A partial Ag+-HPLC separation of the cis/trans and trans/trans CLA region of a milk fat
sample using three Ag+-HPLC columns in series and UV detection at 233 nm.
D. Summary
Satisfactory analyses of the fatty acid profile of fats containing TFA are
achieved by GC using capillary columns coated with highly polar
cyanopolysiloxane stationary phases. In capillary GC methods, the key limita-
tion has been the incomplete separation of trans-18:1 isomers from their cis
isomers. However, almost complete separation of cis and trans-18:1 isomers of
PHVO origin is attainable with the use of 100 m cyanopolysiloxane capillary
columns operated isothermally at 180°C using hydrogen as the carrier gas at a
flow rate of 1 ml/min. Under these conditions, there is very little overlap of cis
and trans isomers. For obtaining detailed data on the individual cis and TFA
isomers, chromatographic analysis in conjunction with either AgNO3-TLC or
Ag+-HPLC has to be used.
Dairy and meat fats are more challenging lipid matrices to analyse. Milk fat
contains more than 400 different fatty acids, hence it is important to have
comprehensive analytical techniques to better evaluate their complex matrices.
In the past, the separation obtained by any kind of GC analysis was considered
satisfactory, and most researchers were not interested in separating and identi-
fying TFA isomers. Detailed fatty acid analysis is not accomplished with a
single chromatographic technique, and many factors (i.e., extraction, methyl-
ation, GC parameters, and unavailability of commercial standards) influence
ANALYSIS OF TRANS FATTY ACIDS OF PHVO 141
the outcome of the analysis. GC remains, by far, the most convenient method
to characterize fatty acids, as explored in this chapter. Recent studies have
described the advantages and disadvantages of different kinds of columns and
chromatographic conditions (Molkentin and Precht, 1994 and 2000; Molkentin
and Precht 1995; Kramer et al., 1998; Kramer et al. 2002; Precht and Molkentin
2003). Even when a good resolution of many fatty acids is accomplished by
GC, some fatty acids are not separated or resolved. That is the case with trans-
18:1 isomers where overlaps of isomers are observed. Underestimation of these
peaks is sometimes reported due to inaccurate or incomplete analysis and as a
consequence, the real isomers present are incorrectly reported. The AgNO3-
TLC/GC method has been highly recommended and in many cases proved to be
the best way to obtain a complete analysis of the 18:1 isomers. Eleven t-18:1
isomers have been separated by such techniques (Wolff et al., 1998; Precht and
Molkentin, 1999; Kramer et al., 2001). To complete the separation and resolu-
tion of trans-18:1 isomers, a longer GC temperature program is also
recommended (Precht and Molkentin, 1999; Kramer et al., 2001).
References
Ackman, RG, Hooper, SN and Hooper, DL (1974) Linolenic acid artifacts from the
deodorization of the oils. J. Am. Oil Chem. Soc., 51, 42–49.
Adlof, R and List, G (2007) Analysis and characterization of trans isomers by silver-ion
HPLC. In: Trans Fats in Foods (List, G, Ratnayake, WMN and Kricthevsky, D, eds),
AOCS Press, Champaign, Ilinois, USA, pp.193–222.
Adlof, RO, Copes, LC and Emken EA (1995) Analysis of the monoenoic fatty acid
distribution in hydrogenated vegetable oils by silver-ion high-performance liquid chro-
matography. J. Am. Oil Chem. Soc., 72, 571–574.
Albertyn, DE, Bannon, CD, Craske, JD, Hai, NT, O’Rourke, KL and Szonyi, C (1982)
Analysis of fatty acid methyl esters with high accuracy and reliability. I. Optimization of
flame-ionization detectors with respect to linearity. J. Chromatog., 247, 47–61.
Aldai, N, Murray, BE, Nájera, AI, Troy, DJ and Osoro, K. (2005) Review. Derivatization of
fatty acids and its application for conjugated linoleic acid studies in ruminant meat lipids.
J. Sci. Fd Agric., 85, 1073–1083.
Ali, LH, Angyal, G, Weaver, CM and Rader, JL (1997) Comparison of capillary column gas
chromatographic and AOAC gravimetric procedures for total fat and distribution of fatty
acids in foods. Food Chem., 58, 149–190.
AOAC 2005 (a). AOAC Official Method 996.06. Fat (total, saturated, and unsaturated) in
foods, hydrolytic extraction gas chromatographic method, Revised 2001. In: Official
Methods of Analysis of AOAC International, 18th Edition (Horwitz, W, ed.).
AOAC 2005 (b). AOAC Official Method 996.01. Fat (total, saturated, and unsaturated) in
products, acid hydrolysis capillary gas chromatographic method, first action. In: Official
Methods of Analysis of AOAC International 18th Edition (Horwitz, W, ed).
AOCS 2005. Determination of cis-, trans-, saturated, monounsaturated and polyunsaturated
fatty acids in vegetable or non-ruminant animal oils and fats by capillary GLC method.
In: Official and Recommended Practices of the AOCS, 5th ed., Revisions and Corrections,
AOCS Press, Champaign, Official Method Ce 1h-05, approved 2005, revised 2005.
Bauman, DE and Griinari, JM (2003) Nutritional regulation of milk fat synthesis. Ann. Rev.
Nutr., 23, 203–227.
142 TRANS FATTY ACIDS IN HUMAN NUTRITION
Bauman, DE, Barbano, DM, Dwyer, DA and Griinari JM (2000) Technical note: production
of butter with enhanced conjugated linoleic acid for use in biomedical studies with animal
models. J. Dairy Sci., 83, 2422–2425.
Beaulieu, AD and Palmquist, DL (1995) Differential effects of high fat diets on fatty acid
composition in milk of Jersey and Holstein cows. J. Dairy Sci., 78, 1336–1344.
Belury, MA (2002) Dietary conjugated linoleic acid in health: physiological effects and
mechanisms of action. Annual Review of Nutrition, 22, 505–531.
Beppu, F, Hosokawa, M, Tanaka, L, Kohno, H, Tanaka, T, Miyashita, K (2006) Potent
inhibitory effect of trans 9, trans 11 isomer of conjugated linoleic acid on the growth of
human colon cancer cells. The Journal of Nutritional Biochemistry, 17, 830–836.
Bhattacharya, A, Banu, J, Rahman, M, Causey, J and Fernandes, G (2006) Biological effects
of conjugated linoleic acids in health and disease (Review.) J. Nutr. Biochem., 17, 789–
810.
Bligh, EG and Dyer, WJ (1959) A rapid method of total lipid extraction and purification, Can.
J. Biochem. Physiol., 37, 911–917.
British Nutrition Foundation. Trans Fatty Acids, The Report of the British Nutrition
Foundation Task Force, July 1995. Holborn House, London, UK.
Chilliard, Y, Ferlay, A, Mansbridge, RM and Doreau, M (2000) Ruminant milk fat plasticity:
nutritional control of saturated, polyunsaturated, trans and conjugated fatty acids.
Annales de Zootechnie, 49, 181–205.
Chilliard, Y, Ferlay, A and Doreau, M (2001) Effect of different types of forages, animal fat
or marine oils in cow’s diet on milk fat secretion and composition, especially conjugated
linoleic acid (CLA) and polyunsaturated fatty acids. Livestock Production Science, 70,
31–48.
Chouinard, PY, Corneau, L, Barbano, DM, Metzger, LE. and Bauman, DE (1999) Conju-
gated linoleic acids alter milk fatty acid composition and inhibit milk fat secretion in dairy
cows. J. Nutr., 129, 1579–1584.
Christie, WW (1989) Gas Chromatography and Lipids. The Oily Press, Bridgwater, UK.
Christie, WW (1995) Composition and structure of milk lipids. In: Advanced Dairy Chemistry,
Volume 2 (Fox, P.F., ed.) 2nd ed., Chapman and Hall, London, UK, pp. 1–36.
Christie, WW (1999) The analysis of evening primrose oil. Industrial Crops and Products,
10, 73–83.
Christie, WW (2003) Lipid Analysis. Isolation, separation, identification and structural
analysis of Lipids. 3rd Edition, The Oily Press, Bridgwater, UK.
Christie, WW (2006a) Methylation of fatty aids. Online: www.lipidlibrary.co.uk/topics/
methests/index.htm
Christie, WW (2006b) Preparation of lipid extracts from tissues. Online:
www.lipidlibrary.co.uk/topics/extract2/index.htm
Christie, WW (2007) Mass spectrometry of fatty acid derivatives. Online:
www.lipidlibrary.co.uk/topics/masspec.html
Christopherson, SW and Glass, RL. (1969) Preparation of milk fat methyl esters by
alcoholysis in an essentially non-alcoholic solution. J. Dairy Sci., 52, 1289–1290.
Cruz-Hernandez, C, Deng, Z, Zhou, J, Hill, AR, Yurawecz, MP, Delmonte, P, Mossoba, MM,
Dugan, MER and Kramer, JKG (2004) Methods for analysis of conjugated linoleic acids
and trans-18:1 isomers in dairy fats by using a combination of gas chromatography,
silver-ion thin-layer chromatography/gas chromatography, and silver-ion liquid chroma-
tography. J. Assoc. Off. Analyt. Chem. Int., 87, 545–562.
Cruz-Hernandez, C, Kramer, JKG, Kraft, J, Santercole, V, Or-Rashid, M, Deng, Z, Dugan,
MER, Delmonte, P and Yurawecz, MP (2006) Systematic analysis of trans and conju-
gated linoleic acids in the milk and meat of ruminants, chapter 4. Pages 45–93 in Advances
in Conjugated Linoleic Acid Research. Volume III. (M.P Yurawecz, J.K.G. Kramer, O.
Gudmundsen, M.P. Pariza, and S. Banni, eds) AOCS Press, Champaign Ilinois, USA.
ANALYSIS OF TRANS FATTY ACIDS OF PHVO 143
(CLA) isomers in tissue lipid classes of pigs fed a commercial CLA mixture determined
by gas chromatography and silver ion-high performance liquid chromatography. Lipids,
33, 549–558.
Kramer, JKG, Sehat, N, Fritsche, J, Mossoba, MM, Eulitz, K, Yurawecz, MP and Ku, Y
(1999) Separation of conjugated linoleic acid isomers. In: Advances in Conjugated
Linoleic Acid Research, Volume 1. (Yurawecz, M.P., Mossoba, M.M., Kramer, J.K.G.,
Pariza, M.W. and Nelson, G.J., eds), pp.83–109, AOCS Press, Champaign, Illinois, USA.
Kramer, JKG and Zhou, J (2001) Conjugated linoleic acids and octadecenoic acids: extraction
and isolation of lipids. Eur. J. Lipid Sci. Technol., 103, 594–600.
Kramer, JKG, Cruz-Hernandez C and Zhou J (2001) Conjugated linoleic acids and octade-
cenoic acids: analysis by GC. Eur. J. Lipid Sci. Technol., 103, 600–609.
Kramer, JKG, Blackadar, CB and Zhou, J (2002) Evaluation of two GC columns (60-m
Supelcowax 10 and 100-m CP Sil 88) for analysis of milk fat with emphasis on CLA,
18:1, 18:2 and 18:3 isomers, and short- and long-chain FA. Lipids. 37, 823–835.
Kramer, JKG, Cruz-Hernandez, C, Deng, Z, Zhou, J, Jahreis, G and Dugan, MER (2004) The
analysis of conjugated linoleic acid and trans 18:1 isomers in synthetic and animal
products. Am. J. Clin. Nutr., 79 (Suppl.), 1137S–1145S.
LeDoux, M, Rouzeau, A, Bas, P and Sauvant, D (2002) Occurrence of trans-C18:1 fatty acid
isomers in goat milk: effect of two dietary regimens. J. Dairy Sci., 85, 190–197.
Lichtenstein, AH, Appel, LH, Brands, M, Carnethon, M, Daniels, S, Franch, HA, Franklin,
B, Kris-Etherton, P, Harris, WS, Howard, B, Karanja, N, Lefevre, M, Rudel, L, Sacks,
F, Van Horn, L, Winston, M, and Wylie-Rosett, J (2006) Diet and lifestyle recommen-
dations revision 2006: a scientific statement from the American Heart Association
Nutrition Committee (AHA Scientific Statement). Circulation 114, 82–96.
Marmer, WN and Maxwell, RJ (1981) Dry column method for the quantitative extraction and
simultaneous class separation of lipids from muscle tissue. Lipids 16, 365–371.
Maxwell, RJ, Mondimore, D and Tobias, J (1986) Rapid method for the quantitative
extraction and simultaneous class separation of milk lipids. J. Dairy Sci., 69, 321–325.
Molketin, J and Precht, D (1994) Optimized analysis of trans-octadecenoic acids in edible
fats. Chromatographia, 41, 267–272.
Molkentin, J and Precht, D (2000) Occurrence and biochemical characteristics of natural
bioactive substances in bovine milk lipids. Brit. J. Nutr., 84, Suppl. 1, S47–S53.
Mossoba, MM, McDonald, RE, Roach, JAG, Fingerhurt, DD, Yurawecz, MP and Sehat, N
(1997) Spectral confirmation of trans monounsaturated C18 fatty acid positional isomers.
J. Am. Oil Chem. Soc., 74, 125–130.
Mossoba, MM, Yurawecz, MP, Delmonte, P and Kramer, JKG (2004) Overview of
infrared methodologies for trans fat determination. J. Assoc. Off. Analyt. Chem. Int.,
87, 540–544.
Mozaffarin, D, Katan, MB, Ascherio, A, Stampfer, MJ and Willett, WC (2006) Trans fatty
acids and cardiovascular disease. New Engl. J. Med., 354, 1601–1613.
Nikolova-Damyanova, B (2003) Lipid analysis by silver ion chromatography. In: Advances
in Lipid Methodology-Five (Adlof, R.O., ed.). The Oily Press, Bridgwater, UK, pp.43–
123.
Precht, D and Molkentin, J (1996) Rapid analysis of the isomers of trans-octadecenoic acid
in milk fat, International Dairy Journal, 6, 791–809.
Precht, D and Molkentin, J (1999) C18:1, C18:2 and C18:3 trans and cis fatty acid isomers
including conjugated cis Δ9, trans Δ11 linoleic acid (CLA) as well as total fat composition
of German human milk lipids. Nahrung 43, 233–244.
Precht, D and Molkentin, J (2000) Identification and quantitation of cis/trans C16:1 and
C17:1 fatty acid positional isomers in German human milk lipids by thin-layer chroma-
tography and gas chromatography/mass spectrometry. Eur. J. Lipid Sci. Technol., 103,
102–113.
ANALYSIS OF TRANS FATTY ACIDS OF PHVO 145
Precht, D and Molkentin, J (2003) Overestimation of linoleic acid and trans-C18:2 isomers
in milk fats with emphasis on Δtrans 9, Δtrans 12-octadecadienoic acid. Milchwissenschaft,
58, 30–34.
Precht, D and Molkentin, J (2000a) Frequency and distributions of conjugated linoleic acid
and trans fatty acid contents in European bovine milk fats. Milchwissenschaft, 55, 687–
691.
Precht, D, Molkentin, J, Destaillats, F and Wolff, RL (2001) Comparative studies on
individual isomeric 18:1 acids in cow, goat, and ewe milk fats by low-temperature high-
resolution capillary gas-liquid chromatography. Lipids, 36, 827–832.
Ratnayake, WMN (1998) Analysis of trans fatty acids. In: Trans fatty acids in human
nutrition (Sebedio, JL and Christie, WW eds), The Oily Press, Bridgwater, UK, pp.115–
161.
Ratnayake, WMN (2001) Analysis of dietary trans fatty acids. Journal of Oleo Science, 50,
339–352.
Ratnayake, WMN (2004) Overview of methods for the determination of trans fatty acids by
gas chromatography, silver-ion thin-layer chromatography, silver-ion liquid chromatog-
raphy, and gas chromatography/mass spectrometry. J. Assoc. Off. Analyt. Chem. Int., 87,
523–539.
Ratnayake, WMN and Beare-Rogers, JL (1990) Problems of analyzing C18 cis- and trans-
fatty acids of margarine on the SP-2340 capillary column. J. Chromatogr. Sci., 28,
633–639.
Ratnayake, WMN and Pelletier G (1992) Positional and geometrical isomers of linoleic acid
in partially hydrogenated oils. J. Am. Oil Chem. Soc., 69, 95–105.
Ratnayake, WMN, Chen, ZY, Pelletier, G and Weber, D (1994) Occurrence of 5c,8c,11c,15t-
eicosatetraenoic acid and other unusual polyunsaturated fatty acids in rats fed partially
hydrogenated canola oil. Lipids 29, 707–714.
Ratnayake, WMN, Plouffe, LJ, Pasquier, E and Gagnon, C (2002) Temperature-sensitive
resolution of cis- and trans-fatty acid isomers of partially hydrogenated vegetable oils on
SP-2560 and CP-Sil 88 capillary columns. J. Assoc. Off. Analyt. Chem. Int., 85, 1112–
1118.
Ratnayake, WMN, Hansen, SL, Kennedy, MP (2006) Evaluation of the CP-Sil 88 and SP-
2560 GC columns used in the recently approved AOCS Official Method Ce 1h-05:
determination of cis-, trans-, saturated, monounsaturated, and polyunsaturated fatty acids
in vegetable or non-ruminant animal oils and fats by capillary GLC method. J. Am. Oil
Chem. Soc., 83, 475–488.
Roach, JAG, Yurawecz, MP, Kramer, JKG, Mossoba, MM, Eulitz, K and Ku, Y (2000) Gas
chromatography-high resolution selected-ion mass spectrometric identification of trace
21:0 and 20:2 fatty acids eluting with conjugated linoleic acid isomers. Lipids 35, 797–
802.
Sebedio, JL and Ratnayake, WMN (2007) Analysis of trans mono- and polyunsaturated fatty
acids. In: Trans Fatty Acids (Hamilton R., Hamm, W. and Dijkstra, A.J. eds), Blackwell
Publishers, New York, USA, pp.102–131.
Sehat, N, Yurawecz, MP, Roach, JAG, Mossoba, MM, Kramer, JKG and Ku, Y (1998)
Silver-ion high-performance liquid chromatographic separation and identification of
conjugated linoleic acid isomers. Lipids, 33, 217–221.
Stender, S and Dyerber, J (2003) The influence of trans fatty acids on health. A report from
the Danish Nutrition Council. 4th edition, Publication no. 34.
Uberth, F and Henninger, M (1992) Simplified method for the determination of trans
monoenes in edible fats by TLC-GLC. J. Am. Oil Chem. Soc., 69, 829–831.
Walstra, P, Geurts, TJ, Noomen, A, Jellema, A and van Boekel, MAJS (1999) Dairy
Technology, Principles of Milk Properties and Processes, Marcel Dekker Inc, New York,
USA.
146 TRANS FATTY ACIDS IN HUMAN NUTRITION
Wolff, RL, Bayard, CC and Fabien, RJ (1995) Evaluation of sequential methods for the
determination of butterfat fatty acid composition with emphasis on trans-18:1 acids.
Application to the study of seasonal variations in French butters. J. Am. Oil Chem. Soc.,
72, 1471–1483.
Wolff, RL and Precht, D (1998) Comments on the resolution of individual trans-18:1 isomers
by gas-liquid chromatography, J. Am. Oil Chem. Soc., 75, 421–422.
Wolff, RL, Combe, NA, Precht, D, Molkentin, J, and Ratnayake, WMN (1998) Accurate
determination of trans-18:1 isomers by capillary gas-liquid chromatography on cyanoalkyl
polysiloxane stationary phases. a Oleagineux, Corps Gras Lipids 5, 295–300.
Wolff, RL, Precht, D, Nasser, B and El Kebbaj, MS (2001) Trans- and cis-octadecenoic acid
isomers in the hump and milk lipids from Camelus dromedarius. Lipids, 36, 1175–1178.
Wolff, RL and Precht, D (2002) A critique of 50-m CP Sil 88 capillary columns used alone
to assess trans-unsaturated fatty acids in foods: the case of the TRANSFAIR study,
Lipids, 37, 627–629.
Yurawecz, MP, Mossoba, MM, Kramer, JKG, Pariza, MW and Nelson, GJ (1999) Advances
in Conjugated Linoleic Acid Research, Volume 1 (Yurawecz, M.P., Mossoba, M.M.,
Kramer, J.K.G., Pariza, M.W. and Nelson, G.J., eds), AOCS Press, Champaign, Illinois,
USA.
CHAPTER 6
Replacement of partially hydrogenated oils in
food products: a technological challenge
1
Nestlé Product Technology Center, Marysville, Ohio, USA
2
Nestlé Research Center, Lausanne, Switzerland
A. Introduction
Vegetable fats with tailor-made melting properties are used in several food
applications, such as margarine, ice cream, bakery, confectionery, frying and
culinary. There are four basic ways to modify edible fats for obtaining the
necessary food functionality: fractionation, interesterification, hydrogenation
and blending.
Due to economic reasons, the hydrogenation process is the most commonly
used method to produce fats with the desired functionality and long shelf-life.
A decade ago, it was estimated that about one-third of the edible oils produced
worldwide were hydrogenated (Fitch-Haumann, 1994). Nevertheless, the hy-
drogenation process is under attack because it produces high levels of harmful
trans fatty acids (TFA) (up to 50% of total fat depending on the starting oil and
the hydrogenation conditions). As a consequence, efforts have been made to
reduce or eliminate TFA in food products.
The food industry is using or developing four strategies to reduce or
eliminate TFA in its products. These are modification of the hydrogenation
process; interesterification; modification of the food formulation; and the use
of oil with modified composition.
These four strategies are not used alone, but often in combination. Some of
the most important challenges faced by the food industry in replacing TFA in
foods are to: develop formulation options that provide the same functionality
without increasing costs; maintain relatively low levels of saturated fatty acids;
and comply with food regulations (Hunter, 2005).
industrial hydrogenation of edible oils in the 1920s in Europe and the USA,
which has remained basically unchanged until recently. By the mid-1990s
approximately 25 million tonnes of fats, oils and fatty acids were being
hydrogenated every year for the food, cosmetics and lubricant industries (Fitch
Haumann, 1994). Despite the recent decline in the demand for partially
hydrogenated vegetable oils containing high concentrations of TFA, hydro-
genation of edible vegetable oils is still an important process among other oil
modification technologies.
Partial hydrogenation in the food industry is traditionally used to produce
semi-solid, oxidatively-stable fats from vegetable oils for multiple food appli-
cations. The partial hydrogenation of vegetable oils, mainly from soybean,
palm, rapeseed (canola) and cottonseed oils, is the major source of TFA in our
food supply (for example Van Poppel et al., 1998; Wijesundera et al., 2007).
The concentration of TFA in partially hydrogenated vegetable fats can vary
from about 3% in lightly hydrogenated (brushed) oils, to more than 50%,
depending on the type of base oil, process conditions and degree of hydrogena-
tion. In many cases, the desired functionality of the hydrogenated fat, such as
a particular solid fat content and melting profile, is to a large extent a function
of the concentration and type of triacylglycerols containing TFA. For example,
high levels of elaidic acid (9-trans-18:1), with a melting point of 43°C are
frequently required in many applications in order to achieve the desired
hardness or melting profile. In those cases a reduction of the concentration of
TFA will be accompanied by a reduction of the desired functionality of the fat.
This is typically the case in many high TFA content confectionery products
(Timms, 2003) and in culinary fats derived from the partial hydrogenation of
palm, soybean and cottonseed oils. On the other hand, in cases where the
desired functionality is mainly the function of the content and distribution of
saturated and cis unsaturated fatty acids in the triacylglycerols, changes in the
parameters and conditions of hydrogenation that favour the reduction of
unsaturation, and at the same time minimizing trans isomerization, are of
interest.
Formation of trans fatty acids during the hydrogenation process is described
in Chapter 2.
fat without changing its fatty acid composition was rapidly recognized as an
important tool for the development of low-TFA alternatives (Rozendaal &
Macrae, 1997).
In the context of applications to the edible fat and oil industry,
interesterification involves one form or another of rearrangement of the fatty
acids in the triacylglycerol molecules. Natural fats and oils have not only
specific fatty acid compositions, but also specific fatty acid distribution in the
triacylglycerol structure. This specific distribution is the response of organisms
to the need to maintain appropriate cellular function and to adjust to environ-
mental variables (Ucciani & Debal 1996). Vegetable fats and oils tend to have
most of the unsaturated fatty acids esterifying the position sn–2 of the
triacylglycerol, whereas saturated fatty acid occupies positions sn–1 and sn–3.
The two basic types of fatty acid rearrangement with respect to applications
for the edible fat and oil industry are chemical interesterification and enzymatic
interesterification.
1. Chemical interesterification
Chemical interesterification of lipids proceeds spontaneously at temperatures
> 250°C. However, chemical interesterification of edible fats is achieved at
much lower temperatures, usually 80–100°C and in the presence of sodium
methoxide powder (0.05–0.1% w/w) as a catalyst. Industrial chemical
interesterification of fats and oils produces a statistical or random distribution
of the fatty acids over the sn–1, 2 and 3 positions of the glycerol molecules as
shown schematically in Figure 1.
The effect of chemical interesterification of a fat or oil varies depending on
the degree of order and arrangement of the fatty acids present in the initial
material. The effect on melting point, for example on refined, bleached and
deodorized (RBD) palm oil, is relatively small, but for other oils with a more
marked non-random distribution of fatty acids, such as cocoa butter, the effect
of chemical interesterification is more pronounced.
The reduction in the melting point obtained by chemical interesterification
of a blend of vegetable oil with a fully hydrogenated vegetable oil depends on
the ratio of the components in the original blend and on the decrease in the
concentration of high melting point components, mainly di- and tri-saturated
triacylglycerols. Reduction in the melting point of 20–25ºC after chemical
interesterification of vegetable oil blends (sunflower and rapeseed oils) with
fully hydrogenated vegetable oils is possible (Schmidt et al., 1996).
In a special case of interesterification (directed chemical interesterification),
the temperature of the process can be controlled by inducing the crystallization
of the newly formed tri-saturated triacylglycerols of palmitic and stearic acids.
These solid tri-saturated triacylglycerols no longer take part in the
interesterification reaction, remaining as such and significantly increasing the
melting point and solid fat content of the final product (Sreenivasan, 1978).
2. Enzymatic interesterification
Enzymatic interesterification typically produces triacylglycerols with a non-
random distribution, selectively cleaving and re-attaching the fatty acids in the
sn–1 and sn–3 positions, leaving the fatty acids in the sn–2 position virtually
untouched (Figure 1). The enzymes used for this type of interesterification are
lipases isolated from yeast and other fungi such as Candida antarctica,
Thermomyces lanuginose and Rhizomucor miehei, which are immobilized in
an inert substrate. Enzymatic interesterification is now widely used for the
production of low-TFA plastic fats for numerous applications, but it is not a
new technique. It was patented by in 1977 and its use was very limited
until immobilized enzymes became affordable for large-scale industrial use.
Initially, lipase-catalysed interesterification was used for high-value speciality
structured lipids. Still today, enzymatic interesterification is required for
special applications where positional specificity of the fatty acids in the
triacylglycerol molecule is important or necessary to achieve the intended
functionality. Examples are the production of symmetric triacylglycerols of the
type SUS or USU, such as cocoa butter equivalents (Timms, 2003) and breast
milk substitutes (Lien et al., 1997), respectively.
The operational costs of enzymatic interesterification are higher than those
of chemical interesterification. In recent years, however, efficiency of the
immobilized enzymes has increased significantly making enzymatic
interesterification more competitive. More recently, the availability of cost-
effective immobilized enzymes has allowed its use for the production of plastic
fats of massive consumption for margarines, spreads and shortenings that
traditionally used partially hydrogenated vegetable oils.
Additional advantages of enzymatic interesterification are those associated
REPLACEMENT OF PARTIALLY HYDROGENATED OILS 151
3. Applications
Chemical and enzymatic interesterification are applied to edible fats and oils
for many different purposes. Most applications of interesterification to edible
fats are aimed at modifying the melting point of blends or single fats by
reducing the concentration of tri-saturated or tri-unsaturated triacylglycerol
species (Sonntag, 1982). Interesterification is also performed to induce the
formation of a preferred crystal polymorphism, or to increase the miscibility
between ‘incompatible’ fats (e.g., lauric and non-lauric fats), thus reducing an
unwanted eutectic softening effect (Noor Lida et al., 2006). Due to the current
health and nutritional trends, those changes in physicochemical characteristics
of fats and oils induced by interesterification are extensively used for produc-
ing the ‘zero’- and low-TFA fats and oils now required in many food products.
These mainly include margarines, shortening, spreads, confections and many
other products that require a fat phase with the plasticity provided by the
previously ubiquitous high-TFA partially hydrogenated vegetable oils.
The versatility of interesterified palm and palm kernel oil blends for the
formulation of low-TFA fat products have been summarized and reviewed by
Berger and Idris (2005). These authors covered a number of common applica-
tions for the production of reduced-TFA or low-TFA foods using both chemical
and enzymatic interesterification (e.g. shortenings, margarines and spreads),
but also referred to less common fat products obtained by interesterification,
such as frying fats, low-fat spreads and whipped cream substitutes.
Most low-TFA margarines and spreads are currently produced by
interesterification of vegetable oils with fully hydrogenated vegetable oils as
base stocks. The purpose of interesterification for manufacturing low-TFA
152 TRANS FATTY ACIDS IN HUMAN NUTRITION
margarines is not only to obtain plastic fat with the suitable spreading and
melting characteristics, but also to promote the formation of small and stable β'
crystals that confer the desired smoothness and integrity of the product, in
blends that otherwise would tend to form coarse β crystals.
Soybean and canola oils are used as common base stocks for manufacturing
shortening and margarines in the USA. Interesterification blends of palm or
canola oils with lauric fats are preferentially used in Europe. Fully hydrogen-
ated palm or canola oils interesterified with a lauric fat can produce a range of
spread products of different hardness and low in TFA (< 2%). The
interesterification also significantly sharpens the melting profile of the original
physical blend, producing a fat with more than 90% solids at temperatures
below 20°C and much less solids at ambient and near-body temperatures
(Allen, 1996).
In an effort to improve the oxidation stability of interesterified hard stocks
used for the production of shortenings, soybean oil was replaced by high- or
mid-oleic/low-linoleic varieties of sunflower and canola oils (Lampert, 2000).
Blends of high-oleic canola oil and fully hydrogenated soybean oil (55% and
45% respectively) can produce low-TFA fats that would reproduce the func-
tionality required for baking applications of a conventional all-purpose
high-TFA (>20%) shortening. Furthermore, it is important to note that the
TFA + SFA (SFA = saturated fatty acids) of these interesterified blends re-
mains well below those of the conventional all purpose shortenings (~35% vs
55% respectively).
Enzymatic interesterification of a blend of palm stearin and coconut oil
(75:25 w/w) was used to study the production of TFA-free margarine in a
bench-scale reactor using the Rhizomucor miehei lipase immobilized in an ion
exchange resin (Zhang et al., 2000). Manipulating the reaction conditions,
such as temperature, reaction time and especially water content, it was possible
to obtain an enzymatically interesterified fat that was similar in characteristics
to that obtained by chemical interesterification.
The enzyme stability and retention of activity was a major concern during the
early stages of development of enzymatic interesterification. Rapid lost of
enzymatic activity was a major factor limiting the use and cost competitiveness
of enzymatic interesterification of commodity oils. However, significant
progress has been made in the immobilization and retention of enzyme activity.
Zhang et al. (2000) measured effective and stable activity of immobilized
Rhizomucor miehei lipase after re-using it for a total of ten cycles. Effective
activity for the enzymatic interesterification of palm stearin and soybean oil of
immobilized lipase suitable for margarine production was maintained for
21 days in a continuous reactor with an immobilized lipase from Candida
antartica (Osorio et al., 2005).
Vanaspati, a popular alternative to ghee in India and South Asia, is manu-
factured with partially or fully hydrogenated vegetable oils and can contain
REPLACEMENT OF PARTIALLY HYDROGENATED OILS 153
a major component in the fat raw material may not represent, or may not be
perceived as, a significant nutritional advantage for the replacement of high-
TFA hydrogenated fats.
Reduced fat spreads made with partially hydrogenated oils can contain up to
20% TFA. Chemically interesterified ternary blends of sunflower, palm and palm
kernel oils and oleins were studied for the production of tub and block spreads
(Noor Lida & Md. Ali, 1998). It is generally accepted that a solid fat content (SFC)
value between 4° and 10°C determines the spreadability of the product at
refrigeration temperature, and that an SFC < 30% at 10°C is necessary to achieve
spreadability at that temperature. The SFC at refrigerator temperatures of the
tertiary blends reported here ranged from about 10% to 38%, depending on their
composition, making most of them suitable for the intended applications.
A study was conducted to follow the changes in physical properties of
enzymatically interesterified hard stocks during a storage time of 12 weeks
(Zhang et al., 2005). These hard stocks were intended to be used in the
formulation of table margarines and the parameters measured included hard-
ness, dropping point, crystal form and colour. Margarine stability increased
with the degree of conversion and, not surprisingly, margarine produced with
the fat fully converted had overall physical properties similar to that produced
with chemically interesterified fat.
Fattahi-far and co-workers (2006) studied the use of chemical inter-
esterification of hydrogenated and non-hydrogenated tea seed oil as a hard
stock for the production of table margarines. Physical and sensory properties of
these experimental margarines (i.e., melting point, SFC, texture, colour, fla-
vour, spreadability) were favourable compared with commercial types. It was
claimed that interesterification is viewed as a natural process for the production
of low-TFA foods and tea seed oil is considered a healthy ingredient. One
wonders whether the positive health attributes of the tea seed oil can be
maintained after subjecting the oil to such a drastic process of hydrogenation
and chemical interesterification.
According to food legislation in different countries, interesterification,
whether chemical or enzymatic, does not need to be declared as an oil
modification technique. However, there are some concerns over long-term
acceptance of chemical interesterification. These concerns are related to envi-
ronmental (water effluents) and safety issues for plant workers (flammable
catalyst) and consumers (side-reaction products).
In order to obtain oils more resistant to oxidation and with an improved health
profile, many efforts have been put into ‘designing’ food oils, for instance
agronomically suitable crops, with a high content of oleic acid and low content
of polyunsaturated and saturated fatty acids.
Several oilseeds with modified fatty acid compositions have been developed
(List, 2004). Breeding programmes have used available germplasm to develop
crops, for example sunflower and canola, that have a more desirable oil content
or fatty acid composition (McKeon, 2005). Mutagenic processes and genetic
engineering have been intensively used to produce genetically-modified seed
oils (e.g. sunflower and canola seed oils).
Currently the four genetically-engineered (transgenic) crops that have been
adopted are all oilseed crops: sunflower, soybean, corn, cotton and canola. The
mutagenic approach is geared to eliminating genes; as a consequence this
approach decreases levels of undesired fatty acids or increases levels of a
desired fatty acid (Spiller et al., 1998). An example of induced mutagenesis
applied to oil crops is the conversion of the non-edible high α-linolenate
oilseed, linseed, to a high-linoleate variety. The high α-linolenate seeds were
treated with the mutagen ethyl methane sulphonate to disrupt the fatty acid
desaturase responsible for the conversion of linoleate to α-linolenate. By
crossing single mutants, a population of double mutants was obtained that
yielded linseed oil with similar acyl composition to that of sunflower oil
(Murphy, 2006).
Crop genetic engineering can lead to undesirable consequences, such as the
need for chemical treatments, by introducing the genes encoding herbicide
tolerance and/or insect resistance (Murphy, 2006 ), and it can introduce
desirable traits such as high levels of monounsaturated and low levels of
saturated fatty acids to improve nutritive value (Murphy, 2006). The most
common approach involves understanding plant lipid metabolism followed by
identification and cloning of the enzyme producing desired fatty acids and
further expression in the breeding varieties.
There follow some examples of oils with modified composition used in the
food industry.
F. Conclusions
Current efforts to replace or reduce TFA in food products focus on the
following strategies: modification of the hydrogenation process; use of
interesterified fat; modification of the food formulation; and use of oil with
modified composition
Currently, there is no hydrogenation technology available that could elimi-
nate TFA formation. Modification of the food formulation is not always
possible and would affect taste and texture with consequent consumer com-
plaints. Breeding programmes take a long time; traditional breeding techniques
REPLACEMENT OF PARTIALLY HYDROGENATED OILS 159
References
Albright, LF (1987) Hydrogenation of triglycerides. In: Hydrogenation: Proceedings of the
AOCS Colloquium (Hastert, R, ed.) American Oil Chemists’ Society, Champaign
Illinois, USA, p.12.
Allen, DA (1996) Interesterification. A vital tool for the future? Lipid Technology, 11, 15.
An, W, Hong, JK, Pintauro, PN, Warnr, K and Neff, WE (1998) The electrochemical
hydrogenation of edible oils in a solid polymer electrolyte reactor: I. Reactor design and
operation. J. Am. Oil. Chem. Soc., 75, 917–925.
Berben, P (2002) Hydrogenation of fats and oils: trends in catalyst development. Lipid
Technology Newsletter, 12, 133–137.
Berger, KG and Idris, NA (2005) Formulation of zero-trans acids shortening and margarines
and other food fats with products of the oil palm. J. Am. Oil. Chem. Soc., 82, 775–782.
Bhattacharyya, S, Bhattacharyya, DK and De, BK (2000) Modification of tallow fractions in
the preparation of edible fat products. Eur. J. Lipid Sci. Technol., 102, 323–328.
Chang, T, Lai, X, Zhang, H, Sondergaard, I and Xu, X (2005) Monitoring lipase-catalyzed
interesterification for bulky fat modification with FT-IR spectroscopy. J. Agric. Food
Chem., 53, 9841–9847.
Cook, R (2000) Chemical interesterification – Process and safety management. SCI Lecture
Papers Series 116. Ebortec, Whale Bridge Park, UK.
Crupkin, M, and Zambelli, A (2008) Detrimental impact of trans fats on human health: stearic
acid-rich fats as possible substitutes. Comprehensive Reviews in Food Science and Food
Safety, 7 (3), 271–279.
Dijkstra, AJ (2002) Hydrogenation and fractionation. In: Fats in Food Technology (Rajah,
KK, ed.) CRC Press, Boca Raton, Florida, USA, pp.123–157.
Dijkstra, AJ (2006) Revisiting formation of trans isomers during partial hydrogenation of
triacylglycerol oils. Eur. J. Lipid Sci. Technol., 108, 249–264.
160 TRANS FATTY ACIDS IN HUMAN NUTRITION
Dijkstra, AJ, Toke, ER, Kolonitis, P, Recseg, K, Kovari, K, and Poppe, L (2006) The base
catalyzed, low temperature interesterification mechanism revisited. Eur. J. Lipid Sci.
Technol., 107, 912–921
Dow Agrosciences, information available at: www.dowagro.com
Drozdowski, B and Szukalska, E (2000) Effect of rapeseed oil hydrogenation conditions on
trans isomer formation. Eur. J. Lipid Sci. Technol., 102, 642–645.
Farr, WE (2005) Hydrogenation: processing technologies. In: Bailey’s Industrial Oil and Fat
Products. Sixth edition, volume 5 (Shahidi, F, ed.) Wiley-Interscience, Hoboken, New
Jersey, USA, pp.385–96.
Fattahi-far, E, Sahari, MA and Barzegar, M (2006) Interesterification of tea seed oil and its
application in margarine production. J. Am. Oil. Chem. Soc., 83, 841–845.
Fitch-Haumann, B (1994) Tools: hydrogenation, interesterification. Inform, 5, 668–678.
Hastert, RC (2000) Hydrogenation. In: Introduction to Fats and Oils Technology (O’Brien,
RD, Farr, WE and Wan, PJ, eds) American Oil Chemists’ Society, Champaign Illinois,
USA, pp.179–193.
Hunter, JE (2005) Dietary levels of trans-fatty acids: basis for health concerns and industry
efforts to limit use. Nutrition Research, 25, 499–513.
Karabulut, I, Turan, S and Ergin, G (2004) Effects of chemical interesterification on solid fat
content and slip melting point of fat/oil blends. Eur. Food Res. Technol., 218, 224–229.
Kellens, M (2000) Oil modification processes. In: Edible Oil Processing (Hamm, W and
Hamilton, RJ, eds) CRC Press. Boca Raton, Florida, USA, pp.129–173.
Khatoon, S and Reddy, SRY (2005) Plastic fats with zero trans fatty acids by interesterification
of mango, mahua and palm oils. Eur. J. Lipid Si. Technol., 107, 786–791.
King, JW, Russell, LH List, GR and Snyder, JM (2001) Hydrogenation of vegetable oils using
mixtures of supercritical carbon dioxide and hydrogen. J. Am. Oil. Chem. Soc., 78, 107–
113.
Kinney, AJ (1997) In: Physiology, Biochemistry and Molecular Biology of Plant Lipids
(Williams JP, Khan MU and Lem NW, eds) Kluwer Academic, Dordrecht, The Nether-
lands, 298–300.
Koritala, S, Friedrich, JP and Dutton, HJ (1980) Selective hydrogenation of soybean oil. X.
Ultra high pressure and low pressure. J. Am. Oil. Chem. Soc., 57, 1–5.
Lampert, D (2000) Processes and products of interesterification. In: Introduction to Fats and
Oils Technology, Second edition (O’Brien, RD, Farr, W E and Wan, PJ, eds) American
Oil Chemists’ Society, Champaign Illinois, USA, pp.208–234.
Lien, E, Boyle, FG, Yuhas, R, Tomarelli, RM, Quinlan, P (1997) J. Pe d. Gastroenterol. Nutr.,
25, 167–174.
List, GR (2004) Decreasing trans and saturated fatty acids content in food oils. Food
Technology, 58 (1), 23–31.
List, GR, Warner, K, Pintauro, P and Gil, M (2007) Low-trans shortening and spread fats
produced by electrochemical hydrogenation. J. Am. Oil. Chem. Soc., 84, 497–501.
Liu, KS (1999) Soy oil modification: products, applications. Inform, 10 (9), 868–878.
Macher, M-B, Högberg, J, Moller, P, and Härröd, M (1999) Partial hydrogenation of fatty
acids methyl esters at supercritical conditions. Fett/Lipid, 101, 301–305.
McKeon, TA (2003) Genetically modified crops for industrial products and processes and
their effect on human health. Trend in Food Science & Technology, 14, 229–241.
McKeon, TA (2005) Transgenic oils. In: Bailey’s Industrial Oil & Fat Products, Sixth
Edition, Volume 3 (F Shahidi, ed.) Wiley-Interscience, J. Wiley & Sons, Inc., Hoboken,
New Jersey, USA.
Moller, P, Harrod, M (1997) Hydrogenation of oils at supercritical conditions. Abstract of
88th American Oil Chemists’ Society Annual Meeting & Expo, Seattle, Washington,
USA, p.8.
Monsanto, information available at: www.monsanto.com
REPLACEMENT OF PARTIALLY HYDROGENATED OILS 161
Mounts, TL, Warner, K, List, GR, Neff, WE and Wilson, RF (1994) J. Am. Oil. Chem. Soc.,
71, 495–499.
Murphy DJ (2006) Plant breeding to change lipid composition for use in food. In: Modifying
Lipids for Use in Food. (Gunstone, FD, ed.) CRC Press, Cambridge, UK.
National Sunflower Association, information available at: www.sunflowernsa.com
Noor Lida, HMD and Ali, AR (1998) Physico-chemical characteristics of palm-based oil
blends for the production of reduced fat spreads. J. Am. Oil. Chem. Soc., 75, 1625–
1631.
Noor Lida, HMD, Sundram, K and Idris, NA (2006) DSC study on the melting properties of
palm oil, sunflower, and palm kernel olein blends before and after chemical
interesterification. J. Oil Am. Chem. Soc., 83, 739–745.
Osorio, NM, Gusmao, JH da, Fonseca, MM and Ferreira-Diaz, S (2005) Lípase-catalyzed
interesterification of palm stearin with soybean oil in a continuous fluidised-bed reactor.
Eur. J. Lipid Sci. Technol., 107, 455–463.
Ronne, TH, Lars, SP and Xu, X (2005) Triglyceride selectivity of immobilized Thermomyces
lanuginose lipase in interesterification. J. Am. Oil. Chem. Soc., 82, 737–743.
Ronne, TH, Yang, T, Mu, H, Jacobsen, C and Xu, X (2005) Enzymatic interesterification of
butterfat with rapeseed oil in a continuous packed bed reactor. J. Agric. Food Chem., 53,
5617–5624.
Rozendaal, A and Macrae, AR (1997) Interesterification of oils and fats. In: Lipid Technolo-
gies and Applications (Gunstone, FD and Padley, FB, eds) Marcel Deckker, New York,
USA, pp.223–263.
Rubin, LJ, Koseoglu, SS, and Graydon, WF (1986) Hydrogenation of canola oil in the
presence of nickel and methyl benzoate-chrome-carbonyl complex. J. Am. Oil Chem.
Soc., 63, 1551–1557.
Schmidt, S, Hurtova, S, Zemanovic, J, Sekretar, S, Simon, P and Ainsworth, P (1996)
Preparation of modified fats from vegetable oil and fully hydrogenated vegetable oil by
randomization with alkali catalysis. Food Chemistry, 55, 343–348.
Schööm, N-H (1995) Is a low trans content attainable by conventional hydrogenation of
vegetable oils. In: Oil-Fats-Lipids (Castenmiller, WAM, ed.) PJ Barnes & Associates,
Bridgwater, UK, pp.155–158.
Smith, SA, King RE, Min DB (2007) Oxidative and thermal stabilities of genetically modified
high oleic sunflower oil. Food Chemistry, 102, 1208–1213
Sonntag, NOV (1982) Fat splitting, esterification and interesterification. In: Bailey’s Indus-
trial Oil and Fat Products, Volume 2, Fourth Edition (Swern, D, ed.) John Wiley & Sons,
New York, USA, pp.97–173.
Sreenivasan, B (1978) Interesterification of fats. J. Am. Oil Chem. Soc., 55, 796–805.
Sundram, K, Karapaiah, T and Hayes, KC (2007) Stearic acid-rich interesterified fat and
trans-rich fat raise the LDL/HDL ration and plasma glucose relative to palm olein in
humans. Nutrition & Metabolism, 4 (3), online: doi:10.1186/1743-7075-4-3
Timms, R (2003) Confectionery Fats Handbook. Properties, Production and Application.
See Chapter 4. The Oily Press, Bridgwater, UK.
Ucciani, E and Debal, A (1996) The effect of triglyceride positional distribution on fatty acid
absorption in rats. chemical properties of fats. In: Oils and Fats Manual (Karlesking, A
and Wolff J-P, eds) Lavoisier Publishing, Paris, France, pp.325–332.
Van Poppel, G, Van Erp-Baart, MA, Kafatos, DA and Aro, A (1998) Trans fatty acids in foods
in Europe: The Transfair study. J. Food Com. Anal., 11, 112–136.
Warner, K, Neff, WE, List GR and Pintauro, P (2000) Electrochemical hydrogenation of
edible oils in solid polymer electrolyte reactor. Sensory and compositional characteristics
of low trans soybean oils. J. Am. Oil. Chem. Soc., 77, 1113–1118.
Wassel, P and Young, NWG (2007) Food applications of trans fatty acid substitutes.
International Journal of Food Science and Technology, 42, 503–517.
162 TRANS FATTY ACIDS IN HUMAN NUTRITION
Wijesundera, C, Richard, A and Ceccato, C (2007) Industrially produced trans fat in foods
in Australia. J. Am. Oil Chem. Soc., 84, 433–442.
Wright, AJ, Mihele, AL and Diosady, LL (2003) Cis selectivity of mixed catalyst systems in
canola oil hydrogenation. Food Res. Intl., 36, 77–804.
Yusoff MSA and Dian NHM (1995) Trans free formulation: a short review. Palm Oil
Developments, 22, 33–37.
Zhan, H, Jacobsen, C and Adler-Nilsen, J (2005) Storage stability study of margarines
produced from enzymatically interesterified fats compared to margarines produced by
conventional methods. Eur. J. Lipid Sci. Technol., 107, 530–539.
Zhang, H, Smith, P and Adler-Nilssen, J (2004) Effect of degree of enzymatic interesterification
on the physical properties of margarine fats: solid fat content, crystallization behaviour,
crystal morphology, and crystal network. J. Agric. Food Chem., 52, 44323–4431.
Zhang, H, Xu, X, Mu, H Nilsson, J, Adler-Nissen, J and Höy, C-E (2000) Lipozyme IM-
catalyzed interesterification for the production of margarine fats in a 1 kg scale stirred
tank reactor. Eur. J. Lipid Sci. Technol., 102, 411–418.
CHAPTER 7
Metabolism of trans fatty acid isomers
1
UMR 1019, Unité de Nutrition Humaine, Plateforme d’exploration du
métabolisme, INRA centre de Theix, St Genes Champanelle, France
2
Scottish Crop Research Institute (and Mylnefield Lipid Analysis), Invergowrie,
Dundee, Scotland
A. Introduction
Trans fatty acids (TFA) include monounsaturated and polyunsaturated fatty
acids having methylene interrupted double bonds but also isomers having
conjugated double bonds such as conjugated isomers of linoleic acid or CLA.
These may be natural fatty acids but they may also be formed during techno-
logical treatments such as hydrogenation, refining, or frying of oils (Sebedio &
Juaneda, 2007). This review will only deal with isomers not having a conju-
gated double bond system. A comprehensive review on the metabolism of
non-conjugated isomers was recently published by Emken (2007). Earlier
reviews deal with trans monoenes (Hølmer, 1998) and trans polyenes (Sébédio
& Chardigny, 1998) separately.
Trans monounsaturated fatty acids are formed during hydrogenation of oils
to produce margarine and shortening (Dutton, 1979), and they are also present
in ruminant meat and milk as a result of biohydrogenation in the rumen (Precht
et al., 2001; Wolff, 1995). These quantitatively and nutritionally are the most
important sources of TFA in the human diet. The position of the double bond
of these dietary trans 18:1 isomers, counted from the carboxylic carbon,
usually varies from Δ4 to Δ16 carbon atoms of the fatty acid molecule. Very
often in partially hydrogenated vegetable oils, the trans 18:1 isomers form a
Gaussian distribution that centres on the Δ9 or Δ10 double bond. The distribu-
tion depends on the fatty acid composition of the starting vegetable oils and on
the extent of hydrogenation. The trans 18:1 isomer distribution of ruminant fats
is distinctly different from that of partially hydrogenated vegetable oils. In the
former, 11t-18:1 is always the major isomer, whereas 9t-18:1 and 10t-18:1
isomers occur in relatively small amounts (Wolff et al., 1998).
There is also evidence that TFA can be produced within animal tissues by
thiyl-radical catalysed cis/trans isomerization in vitro and in vivo (Zambonin
163
164 TRANS FATTY ACIDS IN HUMAN NUTRITION
et al., 2006), but the contribution from this source in humans is not known.
Similarly, isomerization of one of the double bonds in polyunsaturated fatty
acids and catalysed by nitrous oxide is possible (Balazy, 2000; Zghibeh et al.,
2004).
Dietary fats may also contain a number of positional and geometrical
isomers of linoleic and α-linolenic acids, which are frequently present in low
concentrations in both partially hydrogenated and non-hydrogenated dietary
fats (Ratnayake, 2004). The linoleic and α-linolenic acid isomers present in
non-hydrogenated fats or in many common food fats are the result of exposure
of these polyunsaturated fatty acids to heat treatment; such as steam
deodorization (Ackman et al., 1974), or deep fat frying (Grandgirard et al.,
1984). α-Linolenic acid is more prone to isomerization than linoleic acid,
whereas oleic acid is hardly isomerized at all. In many non-hydrogenated
dietary fats, usually the two mono-trans isomers of linoleic (i.e., 9c,12t-18:2
and 9t,12c-18:2) are present at similar levels and generally higher than the all-
trans isomer, 9t,12t-18:2. Eight geometric isomers are possible for α-linolenic
acid, but usually only four are present in industrially refined oils or oils subject
to mild heat treatments. They have been identified as 9t,12c,15t-18:3, 9c,12c,15t-
18:3, 9c,12t,15c-18:3 and 9t,12c,15c-18:3 (Ackman et al., 1974; Grandgirard
et al., 1984) Geometrical isomers of eicosapentaenoic acid (EPA) and of
docosahexaenoic acid (DHA) are also formed during the deodorization process
of fish oils. In this case, the mono-trans isomers are the major TFA formed
(Fournier et al., 2006; Fournier et al., 2007)
Besides the trans polyunsaturated fatty acids resulting from technological
processing of fats and oils, some unusual trans polyunsaturated fatty acids are
present in certain seeds from which it is possible to extract oil. For example, one
fatty acid of this type is columbinic acid (5-trans,9-cis,12-cis-18:3), which is
linoleic acid with addition of a trans Δ5 bond (Houtsmüller, 1981). However,
these are rarely if ever encountered in the diet.
distributions when these are fed as part of a normal diet. The reviews by
(Hølmer, 1998) and Emken (2007) cover the literature exhaustively, and some
general conclusions can be drawn. For example, much of the TFA are found in
triacylglycerols in tissues, with somewhat less in phospholipids, especially in
liver, kidney, testes, heart and adipose tissue. Very little TFA is found in testes
or the brain. With the exception of the adrenals, cholesterol esters tend to have
the lowest concentrations of TFA. Amongst the main glycerophospholipid
classes, there is greater trans monoene incorporation into phosphatidyl-
ethanolamine, phosphatidylserine and phosphatidylinositol than into
phosphatidylcholine.
In a study with rats on a diet containing 15% partially hydrogenated
safflower oil (50%) for 5 weeks, Wood (1979) observed that all positional
isomers of TFA were incorporated into triacylglycerols of all tissues examined
in proportions similar to those fed, with a slight discrimination in favour of the
8- and to some extent the 9-18:1 isomers. In addition, some trans-hexadec-
enoates, especially trans-8-16:1, were found in triacylglycerols, indicating
chain shortening of trans-10-18:1. In general, trans-12, 13- and 14-18:1
isomers were preferentially incorporated. Comparable results have been ob-
tained by others (see Hølmer, 1998, and Emken, 2007).
Acyltransferases are responsible for triacylglycerol, phospholipid and cho-
lesterol esterification, the specificities of these controlling tissue fatty acid
composition, membrane fluidity and cell function. In all phospholipid classes,
trans monoenoic fatty acids like the saturated fatty acids tend to be concen-
trated in the primary positions relative to position 2, in contrast to cis monoenes.
For example in one study, 11-, 12-, 13- and 14-18:1 were the main trans
isomers found in position sn–1 of the phosphatidylcholine of liver, heart and
plasma, with evident discrimination against trans-10-18:1. In contrast, small
amounts only of trans-9-, 11- and 12-18:1 were present in position sn-2, where
cis-9- and 11-18:1 were the main monoenoic components (Reichwald-Hacker
et al., 1979). In a series of papers published between 1979 and 2002, Emken
and colleagues studied the specificity of the incorporation of specific cis and
trans positional isomers of monoenoic fatty acids into the phosphatidylcholine
and phosphatidylethanolamine of human plasma (reviewed by Emken, 2007),
and some comparable data were obtained by Wood (1979) for phospholipids in
other tissues. Relative to oleate, they observed discrimination against 8t- and
10t-18:1 and the preferential incorporation of 9t- and 12t-18:1.
In relation to trans-polyenes, it was demonstrated that di-trans 18:2 was
incorporated preferentially into position 1 of phosphatidylcholine (Selinger &
Holman, 1965). Comparable results were obtained by Privett et al. (1966).
Furthermore, trans-monoenes, di-trans 18:2, palmitic and stearic acids were
incorporated mainly at the sn–1 and sn–3 positions of the triacylglycerols while
the 9-cis,12-trans-18:2, like linoleic acid was incorporated in the sn-2 position.
A preferential incorporation of 9-cis,12-cis,15-trans-18:3 into rat mitochon-
METABOLISM OF TRANS FATTY ACID ISOMERS 167
Retinal phospholipids
cis DHA (dark squares) trans DHA (white squares)
35 1.4
30 1.2
25 1.0
20 0.8
15 0.6
10 0.4
5 0.2
0 0
0 6 12 18 21 months of diet
Cerebral phospholipids
cis DHA (dark squares) trans DHA (white squares)
35 1.4
30 1.2
25 1.0
20 0.8
15 0.6
10 0.4
5 0.2
0 0
0 6 12 18 21 months of diet
Figure 1. Incorporation of cis and trans isomers of DHA into retinal and cerebral phospholipids of rats
fed for 21 months with cis and trans isomers of 18 :3n–3 (n=6). Adapted from Acar et al. (2006).
what was observed for the liver (4.5% of the total 22:6). Along with the
retina, synaptosomes and brain microvessels were shown to contain the
highest levels of trans 22:6 (Grandgirard et al., 1994). This fatty acid was
also observed in myelin and sciatic nerve but to a lesser extent. However, the
ratios of trans 22:6 to cis 22:6 were similar in all the tissues studied. When
the diet was deficient in linolenic acid, the incorporation of trans 22:6 was
doubled. However, the decrease in the amount of 22:6 in the tissue was not
balanced by the increase in 19-trans-22:6. Instead, a large increase in 22:5
METABOLISM OF TRANS FATTY ACID ISOMERS 171
(n–6) was observed, as has previously been described in (n–3) fatty acid
deficiency (Bourre et al., 1984).
When comparing incorporation of trans PUFA isomers in rat retina and brain
during a 21 months feeding study with trans 18:3, Acar et al. (2006) showed
that, after 6 months, the incorporation of trans DHA was higher in retinal
phospholipids since trans DHA represented 0.5% of total fatty acids whereas
it was only 0.2% in cerebral cortex. After 21 months of feeding, the level of
trans DHA in retina was 1.2% of the total fatty acids while that in the cerebral
cortex increased to 0.3% (Figure 1). Concomitantly, the amount of DHA
decreased in retina from 32% to 26% after 21 months, while it was fairly
constant in the cerebral cortex (14% of the total fatty acids). At the same time,
the amount of 22:5n–6 doubled in the retina while only increasing slightly in
the cerebral cortex. Consequently, the n–6/n–3 ratio increased in the retina of
trans 18:3 fed animals compared to controls, which were fed cis 18:3. On the
contrary, no significant modification of this ratio occurred in the cerebral
cortex. These results suggest that the mechanisms leading to the incorporation
of cis and trans fatty acids are quite different in the retina compared to the brain,
the retina being more susceptible to modification of dietary lipids.
3. β-oxidation
In experiments in which 1-14C-labelled oleic, elaidic and palmitic acids were
fed to rats, the rates of catabolism as measured by the respiratory CO2 were the
same (Coots, 1964), and similar results were found by others after intravenous
injections as albumin complexes (Ono & Fredrickson, 1964). Subsequently, in
experiments with fasted rats, uniformly 14C-labelled oleate in a soybean oil
vehicle was reportedly catabolized to CO2 to a slightly greater extent than was
elaidate but the differences were trivial (Anderson & Coots, 1967). Also in the
latter work, the 14CO2 production from mono-trans 18:2 (70%) and 9-trans,12-
trans-18:2 (68%) was found to be slightly higher than that from linoleic acid
(64%, P < 0.10). This difference was correlated with a preferential esterifica-
tion of cis,cis-18:2 (linoleic acid) into liver phospholipids. It is well established
from a number of feeding experiments with animal models that partial β-
oxidation of trans-18:1 to trans-16:1 fatty acids occurs, but this process
appears not to have been studied in a formal manner (see Wood, 1979).
More recently, the oxidative metabolism of linoleic and α-linolenic acids
and of their synthetic mono-trans isomers, 9-cis,12-trans-18:2, 9-trans,12-cis-
18:2 and 9-cis,12-cis,15-trans-18:3, 9-trans,12-cis,15-cis-18:3, was studied in
fasting rats by Bretillon et al. (1998b). A single dose of 18.5 MBq of each 1-14C
labelled fatty acid (260 mg) was orally given to the animals. The 14CO2 expired
was monitored at regular intervals over 24 h. The 14CO2 production 24 h after
oral administration of the fatty acid ranged from 55.5% to 68.7% of the
radioactivity administered for the 18:2 isomers and from 69.7% to 73.5% for
172 TRANS FATTY ACIDS IN HUMAN NUTRITION
40
A
35
production per 100g of rat
9c, 12t
30 9t, 12c
9c, 12c
25
20
15
2
14CO
10
0
0 2 4 6 8 10 12 14 16 18 20 22 24
Time
Figure 2. 14CO2 recovery after oral administration of [1 14C] linoleic acid and its mono trans isomers.
Adapted from Bretillon et al. (1998).
the 18:3 fatty acids. From 6 to 24 h, 14CO2 recovery was significantly higher
after oral administration of 9-cis,12-trans-18:2 than after giving both of the
other octadecadienoic isomers (Figure 2). The 14C retention per gram of tissue
in the liver and in the heart was significantly lower after feeding 9-cis,12-trans-
18:2 than after administration of both other 18:2 isomers. The 14C retention per
gram of tissue in the muscle was significantly lower after administration of both
trans 18:2 isomers in comparison to linoleic acid.
Neither 14CO2 recoveries nor 14C retentions were significantly different after
administration of the three octadecatrienoic acids. The difference observed in
14
CO2 recovery within the dienes was probably not due to a higher specificity
of the enzymes involved in the β-oxidation sequence for the Δ12 trans double
bond. Indeed, due to the labelling of the fatty acids on the carboxyl end, 14C
values recorded in the CO2-trapping agent were only due to the first cycle of β-
oxidation. No major differences were found for the 18:3 geometrical isomers,
neither for 14C recovery in rat tissues nor for 14CO2 expired.
human diet. More recent studies of Berdeaux et al. (1996) also confirmed that
when fed at low levels in the diet, the mono-trans isomers of linoleic acid did
not induce any modification in the arachidonic acid content in different organs.
However, in the presence of linoleic acid at 1.1% of the calories, dietary di-
trans 18:2 may affect essential fatty acid metabolism by reducing the Δ6
desaturase activity (Schimp et al., 1982).
When fed to animals in the form of heat-isomerized linseed oil (HLO), 18:3
isomers resulted in a decrease in 20:4(n–6) and in the 20:4(n–6)/18:2(n–6) ratio
in liver phospholipids compared to what was observed for the animals fed the
fresh linseed oil (LO) (Blond et al., 1990). In vitro assays using the liver
microsomes of the rats fed the HLO and the LO diets showed that the Δ6
desaturase activity [18:2 → 18:3(n–6)] was not significantly altered. In con-
trast, the Δ5 desaturase activity [20:3 → 20:4(n–6)] was higher in the HLO
group compared to the LO group. However, the 18:3 geometrical isomers fed
to animals as a purified fraction were shown to increase the Δ6 desaturase
activity of linolenic acid when rats were fed a (n–3) deficient diet (Blond et al.,
1995). This may indicate that the presence of trans-18:3 in the diet induces
increased rates of 18:3(n–3) desaturation, related to a requirement of the liver
for cis (n–3) fatty acids.
cnts
120 18:3 Δ9c,12t
100
80
60
18:3 Δ6c,9c,12t
40
20
0
0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0
min
cnts
500
18:2 Δ9t,12c
400
300
200
0
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0
min
Figure 3. Top: GC analysis (radioactive signal) of the FAME from liver microsomes (5 mg protein)
incubated for 15 min with [1 14C] 9c, 12t 18:2. Bottom: GC analysis (radioactive signal) of the FAME
from liver microsomes (5 mg protein) incubated for 30 min in anaerobic conditions with [1 14C] 9t, 12c
18:2. Adapted from Berdeaux et al. (1998).
METABOLISM OF TRANS FATTY ACID ISOMERS 177
3. β-Oxidation
Animals degrade fatty acids by β-oxidation systems that operate separately in
mitochondria and peroxisomes. These organelles utilize different enzyme
systems, which have some specificity for particular fatty acid structures. The
mitochondrial system deals with most of the more conventional fatty acids and
is the more important quantitatively. It has been estimated that about 70 to 85%
of the oxidation of dietary fatty acid occurs by this route (Masters and Crane,
1984; Osmundsen et al., 1991). Oxidation usually goes nearly, if not entirely,
to completion with carbon dioxide as the end product. In contrast, the peroxi-
somal β-oxidation system oxidizes those fatty acids longer than 24 carbons
together with some less abundant substrates including branched-chain and
dicarboxylic fatty acids. In this instance, chain-shortening of fatty acids by only
two or three β-oxidation cycles tends to occur so that trans-18:1 fatty acids
178 TRANS FATTY ACIDS IN HUMAN NUTRITION
yield trans-14:1 and 12:1 products. The latter can then be subjected to
elongation with formation of 16:1 isomers. Emken (2007) has comprehen-
sively reviewed the enzymology of the process in respect to TFA.
Somewhat conflicting results of the relative rates of oxidation of oleate and
elaidate have been obtained in liver perfusion experiments, for which no
convincing explanation is available. On the other hand, experiments with
mitochondrial preparations from rat heart, strongly suggested that elaidic acid
was oxidized more slowly than oleic acid, a finding that was attributed to the
fact that part of the normal mechanism of β-oxidation involves a cis-3-12:1
intermediate, which is isomerized prior to the next step. In contrast, a trans
intermediate requires other enzymatic reactions before oxidation can continue.
In a classic study, the cis and trans isomers of 4-18:1 to 16-18:1, all of which
can be present in partially hydrogenated soybean oil, were compared as
substrates for β-oxidation in the form of the CoA esters by rat heart and liver
mitochondria in vitro (Lawson & Holman, 1981). With a few exceptions only,
both heart and liver mitochondria oxidized the cis isomers, especially 9- and
11-18:1, significantly more rapidly than their respective trans isomers. The cis
isomers were found to be catabolized in a similar manner by heart and liver,
with the even-positional cis isomers being oxidized much more slowly than
adjacent odd-positional isomers, which were in fact oxidized at approximately
the same rate. Very different results were obtained with the trans isomers, on
the other hand. In this instance, liver mitochondria tended to oxidize the even-
positional trans isomers much more rapidly than the adjacent odd-positional
isomers, and at almost the same rate as stearoyl-CoA. This was only true for the
trans isomers in which the double bond was located near the middle of the acyl
chain with heart mitochondria. Again, these results have been interpreted in
terms of the activities of two key enzymes in the β-oxidation pathway, the 3-
hydroxyacyl-CoA epimerase and Δ3-cis-Δ2-trans-enoyl-CoA isomerase.
However, it was recently demonstrated that when elaidoyl-CoA was incu-
bated with rat mitochondria from liver or heart a major metabolite,
5-trans-tetradecenoyl-CoA accumulated in the mitochondrial matrix, although
no analogous metabolites were detected with oleoyl-CoA or stearoyl-CoA as
substrates. Following its conversion to 5-trans-tetradecenoylcarnitine, this
partially degraded fatty acid was able to escape from mitochondria (Yu et al.,
2004). Whether the magnitude of this reaction is sufficient to explain the
reduced rate of conversion of elaidate to CO2 relative to oleate by mitochon-
dria, which is observed in vitro, is not yet known.
Increased insulin production in islet cells exposed to trans as opposed to cis-
monoenoic fatty acids may have been a consequence of the reduced rate of
oxidation of the former (Alstrup et al., 2004).
It is well established that peroxisomes have a preference for the oxidation of
the less common fatty acids, and it appears that trans monoenoic fatty acids
may fall into this category. Certainly rat liver peroxisomes in vitro have a
METABOLISM OF TRANS FATTY ACID ISOMERS 179
distinct preference for the oxidation of trans in comparison to cis fatty acids.
For example, a recent study demonstrated that the rate of peroxisomal oxida-
tion of elaidic acid was about 2.5-fold greater than that of oleic acid. This
appears to reflect the substrate specificity of the peroxisomal β-oxidation
apparatus as opposed to the specificity of entry into the peroxisomal compart-
ment (Guzmán et al., 1999). Peroxisomal oxidation is especially important for
the oxidation of the longer-chain TFA of the kind found in partially hydrogen-
ated fish oils (Flatmark et al., 1988).
In order to explain the preferential oxidation of 9c,12t-18:2 compared to
linoleic acid and to the 9t, 12c-18:2 isomer, Bretillon and colleagues (1998a,
1998b) studied all the different steps involved in the mitochondrial oxidation.
In a first step, hepatic mitochondria from rats were incubated in the presence of
the three radio-labelled fatty acids studied (9c,12c-, 9c,12t- and 9t,12c-18:2).
Data revealed that during the 40 min incubation, fatty acid oxidation ranged
from 60 to 80%, the 9c,12t-18:2 being more oxidized than the other two
isomers as was also demonstrated in vivo. The measure of the acyl-CoA
synthase activity indicated that this enzyme had more affinity for the mono-
trans isomers compared to linoleic acid. However the carnitine palmitoyl
transferase I had the same affinity for 9c,12c-acyl CoA, the 9c,12t-acyl CoA
and the 9t,12c-acyl CoA. These results did not permit an explanation of the
difference between the two TFA geometrical isomers.
dietary 9c,12t-18:2 and 9t,12c-18:2 did not modify the levels of 5c,8c,11c,14c-
20:4 in rat liver lipids but induced changes in the Δ6 desaturation of the
9c,12c-18:2 by liver microsomes in vitro (Berdeaux et al., 1996). Furthermore,
assays using liver microsomes in vitro from rats fed a fat-free diet showed that
9c,12t-18:2 was a more effective inhibitor than 9t,12c-18:2 (Berdeaux et al.,
1998).
Blond et al. (1990) used liver microsomes of rats fed a heated oil which
contained geometrical isomers of linolenic acid and showed that hepatic Δ6
desaturase activity was not significantly modified while the Δ5 desaturase
activity was increased. However, great care must be taken in interpreting these
results considering that rats were fed a heated fat which usually contains a
variety of components. For the 18:3 geometrical isomers, it was shown that the
9t,12c,15c-18:3 was a weak inhibitor of the conversion of linolenic into
stearidonic acid in vitro as were other geometrical isomers having a trans bond
in the Δ9 position as 9t-18:1 and 9t,12c-18:2 (Chardigny et al.,, 1995).
200
(A) a
pmol 6-keto PGF1α/mg cell
150
a
O-diene
100
H-diene
50
a
a
0
Basal Stimulated
Figure 4. Prostacyclin release by endothelial cells from endogenous arachidonic acid (mean ± SD).
Cells were incubated with 100 µM of O diene or H diene for 3 days. Values having the same superscript
are significantly different. O diene is a diene fraction isolated from a soybean oil and H diene is a diene
fraction isolated from a partially hydrogenated soybean oil. Adapted from Kummerow et al. (2007).
that only very minor quantities of this di-trans isomer have been detected in
commercial fats and oils while the major isomers in the human diet were the
9c,12t- and the 9t,12c-18:2.(Entressangles, 1986). Consequently, the large
quantities of the fatty acids used in these studies did not reflect the human diet.
Effect of trans dienoic isomers present in hydrogenated fat on prostacyclin
release by endothelial cells in the presence of high level of linoleic acid was
recently reported by Kummerow et al. (2007). A mixture of fatty acids isolated
from partially hydrogenated soybean oils (diene fraction) were shown to inhibit
the release of prostacyclin from endothelial cells compared to cells incubated
with a diene fraction isolated from a soybean oil (Figure 4). The diene fraction
isolated from soybean oil contained 96% of linoleic acid, 4% 18:1 and no trans
18:2 isomers while the mixture isolated from partially hydrogenated soybean
contained 69.4% of linoleic acid, 3.7% 18:1 and 26.7% of a mixture of trans
18:2 isomers. The data were explained by an inhibition of the conversion of
linoleic to arachidonic acid, as the quantity of arachidonic acid decreased and
that of linoleic acid increased in the cells incubated with the diene fraction
isolated from the partially hydrogenated oil. Some effects of a trans-monoene
fraction on prostacyclin release were also observed, but the effect was much
smaller than that for the trans-diene fraction.
Mono-trans isomers of 18:2(n–6) and 18:3(n–3) are converted into higher
metabolites (see above), including C20 and C22 PUFA containing one (or two)
trans double bonds, which may then be incorporated in phospholipids. Conse-
182 TRANS FATTY ACIDS IN HUMAN NUTRITION
Figure 5. Left: inhibition of rat platelet aggregation by increasing amount of arachidonic acid,
20:3n–9, or 14t 20:4. Right: effect of 14t 20:4 on [1 14C] arachidonic acid metabolism by rat platelets.
Adapted from Berdeaux et al. (1996).
6000
4000
6-keto PGF1α
2000
PGD2
0
1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 ml
(dpm)
C20:5 Δ17t
5000
B
4000
3000
2000
P1
1000
CP2
0
1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 ml
Figure 6. HPLC profile of radiolabelled eicosanoids from endothelial cells incubated with A: [1 14C]
arachidonic acid, and B: [18 14C] 20:5 Δ17t. Adapted from Loi et al. (2000).
E. Human studies
In this part of the chapter, we will not describe the effects of trans monoethylenic
fatty acids on lipoprotein metabolism in humans as this is dealt with elsewhere
in this volume, but we will discuss the presence and the transformation of these
monoethylenic isomers in the tissues as well as their effects on the metabolism
of other fatty acids. The reader will find most of the earlier literature in reviews
by Emken (2007) and Holmer (1998). To summarize, TFA have been detected
in different tissues such as adipose, liver, heart, aorta and milk, as a conse-
quence of their presence in the human diet (Aitchison et al., 1977; Jensen,
METABOLISM OF TRANS FATTY ACID ISOMERS 185
1999). Feeding deuterated fatty acid isomers has shown that plasma cholesteryl
esters content exhibited a 4 to 1 ratio of oleic to elaidic acid while all plasma
phospholipids selectively incorporated elaidic compared to oleic acid.
Studies of human milk composition of two cohorts comprising 103 mothers
and 769 mothers respectively showed that milk TFA were inversely related to
milk linoleic and linolenic acid contents (Innis & King, 1999; Szabo et al.,
2007). Furthermore, in the larger cohort, milk TFA were also inversely related
to the availability of long-chain PUFA (arachidonic and docosahexaenoic
acids) in mature human milk. It was also reported that there could be a PUFA
deficiency in infants due to the presence of TFA (reviewed by Holmer, 1998).
For example, the effects of TFA, including trans mono- and di-enoic isomers
on the biosynthesis of long-chain polyunsaturated fatty acids in premature
infants were studied earlier by Koletzko (1992). Trans octadecenoic acid and
total trans isomers, including 9-cis,12-trans-18:2, 9-trans,12-cis-18:2 and 9-
trans,12-trans-18:2 in plasma lipid fractions were not related to either linoleic
or linolenic acids but these were correlated inversely to the long-chain (n–3)
and (n–6) polyunsaturated acids. These data may indicate a potential impair-
ment of essential fatty acid metabolism.
More recent data confirmed the earlier observations as high exposure to TFA
is related to lower levels of docosahexaenoic acid, and consequently, TFA may
have adverse effects on growth and development through interfering with
essential fatty acid metabolism (Decsi, 2003; Innis, 2006). Results of two birth
cohorts in the Netherlands demonstrated that birth length and head circumfer-
ence were significantly and negatively related to the trans-18:1 concentration
in phospholipids isolated from umbilical plasma, umbilical arterial walls, or
umbilical venous walls (Hornstra et al., 2006).
Different studies have also shown that TFA are not only incorporated in
human tissue but also that a conjugated linoleic acid isomer, the 9c,11t-18:2
(CLA) can be synthesized from vaccenic acid, the 11t-18:1 isomer present in
large quantities in ruminant milk and fat (Turpeinen et al., 2002; Mosley et al.,
2006, Kuhnt et al., 2006a). Vaccenic acid was shown to be converted into
CLA, the conversion rate being 20% on average. This indicates that this
endogenous synthesis is likely to contribute significantly to the total amount of
CLA in the body (Turpeinen et al., 2002).
Using 11t vaccenic-1-13C acid, Mosley et al., (2006) confirmed this in vivo
synthesis of rumenic acid in a population of lactating women. However, it
would be necessary to carry out a larger study to define more accurately the
contribution of the Δ9 desaturase activity in mammary and non-mammary
tissues to the endogenously synthesized 9c,11t-18:2. The present results seem
to indicate that about 10% of the CLA present in milk would come from a
synthesis in the mammary gland. In a recent study, Kuhnt et al. (2006a)
evaluated this in humans, the endogenous conversion of vaccenic acid into
9c,11t-CLA contributes approximately 25% to the human CLA pool. Further-
186 TRANS FATTY ACIDS IN HUMAN NUTRITION
more, a conversion of another trans isomer the 12t-18:1 into 9c, 12t-18:2 was
not detected.
Interestingly, the urinary excretion of 8-isoprostaglandin F2α, corresponding
to arachidonic acid oxidation, increased significantly for the volunteers in-
cluded in the intervention study of Turpeinen et al. (2002). A similar effect on
the production of 8-isoprostaglandin F2α was observed by Kuhnt et al. (2006b)
after feeding a large quantity of a mixture of 11t- and 12t-18:1 isomers.
Association between higher intake of TFA and higher concentration of urine
F isoprostanes was also found in the Study of Women’s health Across the
Nation (SWAN), which included 1610 participants (Tomey et al., 2007).
Geometrical linoleic acid isomers were reported in human tissues, including
milk (Boatella et al., 1993; Chardigny et al., 1995, Chen et al., 1995, Koletzko
et al., 1988) adipose tissue (Adlof & Emken, 1986; Boué et al., 2000; Baylin
et al., 2002), kidney, heart and liver (Adlof & Emken, 1986). Adlof and Emken
(1986) reported that adipose tissue had the highest concentration of trans 18:2
(9-cis,12-trans and 9-trans,12-cis) while the amount present in the brain was
too small to be quantified. Only traces of the di-trans isomers were detected.
While 18:3 geometrical isomers were reported in human milk (Chardigny
et al., 1995), none of the 18:3 isomers were detected, neither by Boué (2000)
nor by Adlof and Emken, in the other human tissues. Longer-chain (n–3)
polyunsaturated fatty acids, especially the metabolites of 9-cis,12-cis,15-trans-
18:3, were also detected in human platelets (Chardigny et al., 1993).
The metabolism in humans of the 12-cis,15-trans-18:2, which can be formed
from linolenic acid by catalytic hydrogenation, was studied by Emken
et al.(1987) using triacylglycerols containing deuterium-labelled fatty acids.
Analyses performed on two human subjects indicated that the turnover for
triacylglycerols, cholesterol esters, phosphatidylethanolamine and
phosphatidylcholine of all the deuterated fatty acids tested (16:0, 18:0, 18:1,
9-cis,12-cis-18:2 and 9-cis,15-trans-18:2) were similar. Furthermore, the select-
ivity ratios for the deuterium-labelled isomers incorporated into the
triacylglycerol and free fatty acid classes were small compared to what was
observed for the phospholipids. It appeared that the 9-cis,15-trans-18:2 must
have been utilized mainly for energy, considering that most of the selectivity
ratios for the phospholipids are negative and that no large positive ratios for the
neutral lipids were observed.
To our knowledge, only one intervention study using linolenic acid geo-
metrical isomers has been carried out on humans (Sebedio et al., 2000;
Vermunt et al., 2001). After a wash out period without TFA, human volunteers
either received or not some isomerized rapeseed oil which contained trans
isomers of linolenic acid at the level of 0.6 % of the total energy for 6 weeks.
At the end of the nutritional intervention, parameters connected to cardiovas-
cular risk factors were measured. It was shown that the trans isomers of
linolenic acid may increase the ratios of plasma LDL-cholesterol to HDL-
METABOLISM OF TRANS FATTY ACID ISOMERS 187
Figure 7. Fatty acid composition of platelets from human volunteers who received (high trans) or not
(low trans) 0.6% energy as trans isomers of linolenic acid for six weeks. T0–T6: 6 weeks wash out period
and T6–T12: experimental period. Adapted from Sebedio et al. (2000).
–5
–10
Expired 13CO2
12cis-18:2
–15 12trans-18:2
–20
–25
Time (min)
–30
0 600 1200 1800 2400 3000
Figure 8. 13CO2 enrichment (δ ‰ vs PDB) after ingestion of linoleic acid (12 cis 18:2) or the mono-
trans isomer (12 trans 18:2). Adapted from Bretillon et al. (2001).
References
Acar, N, Bonhomme, B, Joffre, C, Bron, AM, Creuzot-Garcher, C, Doly, M and Chardigny,
JM (2006) The retina is more susceptible than the brain and the liver to the incorporation
of trans isomers of DHA in rats consuming trans isomers of alpha-linolenic acid. Reprod.
Nutr. Dev., 5, 515–525.
Ackman RG, Hooper SN and Hooper DL (1974) Linolenic acid artefacts from the deodorisation
of oils. J. Am. Oil Chem. Soc., 51, 42–49.
Adlof, RO and Emken, EA (1986) Distribution of hexadecenoic, octadecenoic and octadeca-
dienoic acid isomers in human tissue lipids. Lipids, 21, 543–547.
Aitchison, JM, Dunkley, WL, Canolty, NL and Smith, LM (1977) Influence of diet on trans
fatty acids in human milk. Am. J. Clin. Nutr., 30, 2006–2015.
Alstrup, KK, Brock, B and Hermansen, K (2004) Long-term exposure of INS-1 cells to cis
and trans fatty acids influences insulin release and fatty acid oxidation differentially.
Metabolism, 53, 1158–1165.
Anderson, RL and Coots, RH (1967) The catabolism of the geometric isomers of uniformly
14C-labeled δ9-octadecenoic acid and uniformly 14C-labeled δ9,12-octadecadienoic acid
Chardigny, JM, Wolff, RL, Mager, E, Sebedio, JL, Martine, L, Juaneda P and Grandgirard,
A (1995) Trans mono and polyunsaturated fatty acids in human milk. Eur. J. Clin. Nutr.,
49, 523–531.
Chen, ZY, Pelletier, G, Hollywood, R and Ratnayake WMN (1995) Trans fatty acid isomers
in Canadian human milk. Lipids, 30, 15–21.
Cook, HW (1981) The influence of trans acids on desaturation and elongation of fatty acids
in developing brain. Lipids, 16, 920–926.
Cook, HW and Emken EA (1990) Geometric and positional fatty acid isomers interact
differently with desaturation and elongation of linoleic and linolenic acids in cultured
glioma cells. Biochem. Cell Biol., 68, 653–660.
Coots, RH (1964) A comparison of the metabolism of elaidic, oleic, palmitic, and stearic acids
in the rat. J. Lipid Res., 5, 468–472.
Decsi, T (2003) Nutritional relevance of trans isomeric fatty acids in human milk. Acta
Paediatr., 92, 1369–1371.
Dutton, HJ (1979) Hydrogenation of fats and its significance In: Geometrical and Positional
Fatty Acid Isomers (EA Emken and HJ Dutton, eds), AOCS Press, Champaign, Illinois,
USA, pp.1–16.
Emken, EA (1983) Biochemistry of unsaturated fatty acid isomers. J. Am. Oil Chem. Soc.,
60, 995–1004.
Emken, EA (1984). Nutrition and biochemistry of trans and positional fatty acid isomers in
hydrogenated oils. Ann. Rev. Nutr., 4, 339–376.
Emken, EA (1986) Absorption of deuterium-labeled cis- and trans-octadecenoic acid
positional isomers in man. In: Fat Absorption, Volume 1 (A K uksis, ed), CRC Press, Boca
Raton, Florida, USA, pp.159–175.
Emken, E (2007) Metabolism of trans and cis fatty acid positional isomers compared to non-
isomeric fatty acids. In: Trans Fats in Foods (GR List, WMN Ratnayake and D
Kritchevsky, eds), AOCS Press, Champaign, Illinois, USA, pp. 59–95.
Emken, EA, Rohwedder, WK, Adlof, RO, Rakoff, H and Gulley, RM. (1987) Metabolism in
humans of cis-12,trans-15-octadecadienoic acid relative to palmitic, stearic, oleic and
linoleic acids. Lipids, 22, 495–504.
Entressangles, B (1986) Survey of edible trans fatty acids. Rev. Fr. Corps Gras, 33, 47–58.
Flatmark, T, Nielsson, A, Kvannes, J, Eikhom, TS, Fukami, MH, Kryvi, H and Christiansen,
EN (1988) On the mechanism of induction of the enzyme systems for peroxisomal β-
oxidation of fatty acids in rat liver by diets rich in partially hydrogenated fish oil. Biochim.
Biophys. Acta, 962, 122–130.
Fournier, V, Destaillats, F, Hug, B, Golay, PA, Joffre, F, Juaneda, P, Semon, E, Dionisi, F,
Lambelet, P, Sebedio, JL, and Berdeaux, O (2007) Quantification of eicosapentaenoic
and docosahexaenoic acid geometrical isomers formed during fish oil deodorization by
gas-liquid chromatography. J. Chromatogr. A, 1154, 353–359.
Fournier, V, Juaneda, P, Destaillats, F, Dionisi, F, Lambelet, P, Sebedio, JL, and Berdeaux,
O (2006) Analysis of eicosapentaenoic and docosahexaenoic acid geometrical isomers
formed during fish oil deodorization. J. Chromatogr. A, 1129, 21–28.
Grandgirard, A (1992) Transformations des lipides au cours des traitements thermiques.
Effets nutritionnels et toxicologiques. les Cahiers de l’ENSBANA, 8, 49–67.
Grandgirard, A, Bourre, JM, Julliard, F, Homayoun, P, Dumont, O, Piciotti, M and Sébédio,
JL (1994) Incorporation of trans long-chain n–3 polyunsaturated fatty acids in rat brain
structures and retina. Lipids, 29, 251–258.
Grandgirard, A, Piconneaux, A, Sébédio, JL, O’Keefe, SF, Sémon, E and Le Quéré, JL (1989)
Occurrence of geometrical isomers of eicosapentaenoic and docosahexaenoic acids in
liver lipids of rats fed heated linseed oil. Lipids, 24, 799–804.
Grandgirard, A, Sebedio, JL and Fleury, J (1984) Geometrical isomerization of linolenic acid
during heat treatment of vegetable oils J. Am. Oil Chem. Soc., 61, 1563–1568.
METABOLISM OF TRANS FATTY ACID ISOMERS 191
Guzmán, M, Klein, W, Gómez del Pulgar, T and Geelen, MJH (1999) Metabolism of trans
fatty acids by hepatocytes. Lipids, 34, 381–386.
Hølmer, G (1998) Biochemistry of trans monoenoic fatty acids. In: Trans Fatty Acids in
Human Nutrition (JL Sébédio and WW Christie, eds), Oily Press, Bridgwater, UK,
pp.163–189.
Hornstra, G, van Eijsden, M, Dirix, C and Bonsel, G (2006) Trans fatty acids and birth
outcome: Some first results of the MEFAB and ABCD cohorts. Atheroscler. Suppl., 7,
21–23.
Houtsmüller, UMT (1981) Columbinic acid, a new type of essential fatty acid. Prog. Lipid
Res., 20, 889–896.
Hwang, DH and Kinsella, JE (1978) The effects of trans linoleic acid on the concentration
of serum prostaglandin F2 alpha and platelet aggregation. Prostaglandins Med., 1, 121–
30.
Innis, SM (2006) Trans fatty acid intakes during pregnancy, infancy and early childhood.
Atheroscler. Suppl., 7, 17–20.
Innis, SM and King DJ (1999) Trans fatty acids in human milk are inversely associated with
concentrations of essential all cis n–6 and n–3 fatty acids and determine trans but not n–
6 and n–3 fatty acids in plasma lipids of breast-fed infants. Am. J. Clin. Nutr., 70, 383–390.
Jensen, RG (1999) Lipids in human milk. Lipids, 34, 1243–1271.
Kameda, K, Valicenti, AJ and Holman, RT (1980) Chain elongation of trans-octadecenoic
acid isomers in rat liver microsomes. Biochim. Biophys. Acta, 618, 13–17
Karney, RI and Dhopeshwarkar, GA (1978) Metabolism of all trans 9,12-[1-
14C]octadecadienoic acid in the developing brain. Biochim. Biophys. Acta, 531, 9–15.
Loï, C, Chardigny, JM, Almanza, S, Leclere, L, Ginies C, and Sébédio, JL (2000) Incorpo-
ration and metabolism of dietary trans isomers of linolenic acid alter the fatty acid profile
of rat tissues. J. Nutr., 130, 2550–2555.
Mahfouz, MM and Kummerow, FA (1999) Hydrogenated fat high in trans monoenes with
an adequate level of linoleic acid has no effect on prostaglandin synthesis in rats. J. Nutr.,
129, 15–24.
Mahfouz, MM, Johnson, S and Holman, RT (1980) The effect of isomeric trans-18:1 acids
on the desaturation of palmitic, linoleic and eicosa-8,11,14-trienoic acids by rat liver
microsomes. Lipids, 15, 100–107.
Marchand, CM and Beare-Rogers, JL (1978) The acylation of 1-palmityglycerol 3-phosphate
with cis and trans C-16 to C-22 monoenoic fatty acids in rat liver microsomes. Lipids, 13,
329–333.
Masters, C and Crane, D (1984) The role of peroxisomes in lipid metabolism. Trends
Biochem. Sci., 9, 314–317.
Mosley, EE, McGuire, MK, Williams, JE and McGuire, MA (2006) Cis-9,trans-11 conju-
gated linoleic acid is synthesized from vaccenic acid in lactating women. J. Nutr., 136,
2297–2301.
Mu, H and Høy, CE (2004) The digestion of dietary triacylglycerols. Prog. Lipid Res., 43,
105–133.
Normann, PT, Thomassen, MS, Christiansen, EN and Flatmark, T (1981) Acyl-CoA
synthetase activity of rat liver microsomes substrate specificity with special reference to
very-long-chain and isomeric fatty acids. Biochim. Biophys. Acta, 664, 416–427.
O’Keefe, SF, Lagarde, M, Grandgirard, A and Sebedio, JL (1990) Trans n–3 isomers exhibit
different inhibitory effects on arachidonic acid metabolism in human platelets compared
to the respective cis fatty acids. J. Lipid Res., 31, 1241–1246.
Okuyama, H, Lands, WEM, Gunstone, FD and Barve, JA (1972) Selective transfers of trans-
ethylenic acids by acyl coenzyme A:phospholipid acyltransferases. Biochemistry, 77,
4392–4398.
Ono, K and Fredrickson, DS (1964) The metabolism of 14C-labeled cis and trans isomers of
octadecenoic and octadecadienoic acids. J. Biol. Chem., 239, 2482–2488.
Osmundsen, H, Bremer, J and Pedersen JI (1991) Metabolic aspects of peroxisomal β-
oxidation. Biochim. Biophys. Acta, 1085, 141–158.
Piconneaux, A (1987) Etude de la désaturation et de l’élongation in vivo d’isomères
géométriques de l’acide linolenique. PhD thesis, University of Dijon, France.
Precht D, Molkentin J, Destaillats F, and Wolff, RL (2001) Comparative studies on individual
isomeric 18:1 acids in cow, goat, and ewe milk fats by low-temperature high-resolution
capillary gas-liquid chromatography. Lipids, 36, 827–32.
Privett, OS, Nutter, LJ and Lightly, FS (1966) Metabolism of trans acids in the rat: influence
of the geometric isomers of linoleic acid on the structure of liver triglycerides and
lecithins. J. Nutr., 89, 257–264.
Privett, OS, Stearns, EM and Nickell, EC (1967) Metabolism of the geometric isomers of
linoleic acid in the rat. J. Nutr., 92, 303–310.
Ratnayake WMN (2004) Overview of methods for the determination of trans fatty acids by
gas chromatography, silver-ion thin-layer chromatography, silver-ion liquid chromatog-
raphy, and gas chromatography/mass spectrometry. J. AOAC Int., 87, 523–539.
Ratnayake, WMN, Chen, ZY, Pelletier, G and Weber, D (1994) Occurrence of 5c,8c,11c,15t-
eicosatetraenoic acid and other unusual polyunsaturated fatty acids in rats fed partially
hydrogenated canola oils. Lipids, 29, 707–714.
Reichwald-Hacker, I, Grosse-Oetringhaus, S, Kiewitt, I and Muhkerjee, KD (1979) Incorpo-
ration of positional isomers of cis- and trans-octadecenoic acids into acyl moieties of rat
tissue lipids. Biochim. Biophys. Acta, 575, 327–334.
Reitz, RC, El-Sheikh, M, Lands, WEM, Ismail, IA and Gunstone, FD (1969) Effects of
METABOLISM OF TRANS FATTY ACID ISOMERS 193
Wolff, RL, Combe, NA, Entressangles, B, Sébédio, JL and Grandgirard, A (1993) Preferen-
tial incorporation of dietary cis-9,cis-12,trans-15 18:3 acid into rat cardiolipins. Biochim.
Biophys. Acta, 1168, 285–291.
Wolff, RL, Precht, D and Molkentin, J (1998) Occurrence and distribution profiles of trans-
18:1 fatty acids in edible fats of natural origin. In: Trans Fatty Acids in Human Nutrition
(JL Sébédio and WW Christie, eds), Oily Press, Bridgwater, UK, pp.1–33.
Wood, R (1979) Distribution of dietary geometrical and positional isomers in brain, heart,
kidney, liver, lung, muscle, spleen, adipose and hepatoma. In: Geometrical and Positional
Fatty Acids Isomer, (EA Emken and HJ Dutton, eds) AOCS Press,, Champaign, Illinois,
USA, pp.213–281.
Yu, W, Liang, YW, Ensenauer, RE, Vockley, J, Sweetman, L and Schulz, H (2004). Leaky
β-oxidation of a trans-fatty acid: incomplete β-oxidation of elaidic acid is due to the
accumulation of 5-trans-tetradecenoyl-coa and its hydrolysis and conversion to 5-trans-
tetradecoylcarnitine in the matrix of rat mitochondria. J. Biol. Chem., 279, 52160–52167.
Zambonin, L, Ferreri, C, Cabrini, L, Prata, C, Chatgilialoglu, C and Landi, L (2006)
Occurrence of trans fatty acids in rats fed a trans-free diet: A free radical-mediated
formation? Free Rad. Biol. Med., 40, 1549–1556.
Zevenbergen, JL and Haddeman, E (1989) Lack of effects of trans fatty acids on eicosanoid
biosynthesis with adequate intakes of linoleic acid. Lipids, 24, 555–563.
Zevenbergen, JL, Houtsmuller, UM and Gottenbos, JJ (1988) Linoleic acid requirements of
rats fed trans fatty acids. Lipids, 23, 178–186.
Zghibeh, CM, Gopal, VR, Poff, CD, Falck, JR and Balazy, M (2004) Determination of trans-
arachidonic acid isomers in human blood plasma. Anal. Biochem., 332, 137–144.
CHAPTER 8
Biosynthesis and biological activity of rumenic
acid: a natural CLA isomer
1
Department of Animal Science, University of Vermont, Burlington, Vermont,
USA
2
Department of Animal Science, Cornell University, Ithaca, New York, USA
A. Introduction
Nutritional quality is increasingly an important consideration in food choices
because of the growing consumer awareness of the link between diet and
health. Many foods contain micro components that have beneficial effects
beyond those associated with their traditional nutrient content, and these are
often referred to as ‘bioactive’ or ‘functional food’ components. These func-
tional food components can play an important role in health maintenance and
the prevention of chronic diseases. Cancer and coronary heart disease (CHD)
are two examples, and epidemiological studies suggest that the risk for these
diseases can be substantially reduced by diet modification. Functional food
components from plants are often highlighted in popular press articles as the
basis for lowered risk of these diseases resulting from consumption of fruits
and vegetables. Foods of animal origin also contain functional food compo-
nents, but these have been less extensively investigated. In the case of dairy
products, there has been a general perception that a food containing saturated
fat is unlikely to be beneficial to health. Yet, over the last decade, evidence has
accumulated that the form of dietary fat is very important in determining the
relative risk to diseases like cancer and CHD, and that milk-derived fat may
offer significant health benefits compared to some common sources of dietary
fats (National Research Council, 1996; Parodi, 2004; Bauman et al., 2006).
One such component in foods derived from ruminants is cis-9,trans-11 con-
jugated linoleic acid.
The term conjugated linoleic acids (CLA) refers to a mixture of positional
and geometric isomers of linoleic acid (cis-9,cis-12 octadecadienoic acid) with
a conjugated double bond system. The presence of CLA isomers in ruminant fat
195
196 TRANS FATTY ACIDS IN HUMAN NUTRITION
Table 1. Range of positional and geometric isomers of conjugated linoleic acids (CLA)
in milk and dairy products. Adapted from Lock & Bauman (2004).
trans-7,cis-9 1.2–8.9
trans-7,trans-9 < 0.1–2.4
trans-8,cis-10 < 0.1–1.5
trans-8,trans-10 0.2–0.4
cis-9,trans-11 72.6–91.2
trans-9,trans-11 0.8–2.9
trans-10,cis-12 < 0.1–1.5
trans-10,trans-12 0.3–1.3
cis-11,trans-13 0.2–4.7
trans-11,cis-13 0.1–8.0
trans-11,trans-13 0.3–4.2
cis-12,trans-14 < 0.01–0.8
trans-12,trans-14 0.3–2.8
cis, cis isomers 0.1–4.8
Figure 1. Chemical structures of (A) linoleic acid (cis-9,cis-12 18:2), (B) rumenic acid (cis-9,trans-11
conjugated linoleic acid), and (C) vaccenic acid (trans-11 18:1). Arrows indicate location of double
bonds. Adapted from Bauman et al. (2006).
198 TRANS FATTY ACIDS IN HUMAN NUTRITION
major source (see review by Parodi, 2003). As illustrated by data from the USA,
dairy products account for about 70% of total food intake of CLA and ruminant
meats account for another 25% (Ritzenthaler et al., 2001). Differences in food
preferences result in a wide range in individual daily intakes. In general, the
average dietary intake of rumenic acid ranges from 100 to 300 mg/d (Parodi,
2003). This intake estimate, however, does not include the contribution of
endogenous synthesis of rumenic acid from dietary vaccenic acid. Based on
kinetic data obtained from studies with humans (Turpeinen et al., 2002; Kuhnt
et al., 2006), Parodi (2003) has emphasized that the endogenous synthesis of
rumenic acid needs to be considered to obtain an estimate of the ‘effective
rumenic acid’ provided by ruminant-derived food products.
In this chapter we will first review the origin of rumenic acid present in milk
fat, developing the interrelationships between biohydrogenation intermediates
produced in the rumen, synthesis of rumenic acid in tissues and its presence in
milk fat. Secondly, we will highlight the nutritional and physiological factors
that affect the level of rumenic acid in milk fat and discuss milk quality
considerations for dairy products that have a naturally enhanced content of
rumenic acid. Finally, we will review the biological effects of rumenic acid in
relation to human health. This will include the significance of rumenic acid in
dairy products and its potential as a functional food component that may benefit
health maintenance and disease prevention.
Figure 2. Pathways for ruminal and endogenous synthesis of rumenic acid (cis-9,trans 11-CLA) in the
lactating dairy cow. Pathways for biohydrogenation of linoleic and linolenic acids yielding vaccenic acid
(trans-11 18:1) are shown in the rumen box and endogenous synthesis by Δ9-desaturase is shown in the
mammary gland box. Potential strategies to increase milk fat content of CLA are indicated in the white
boxes. Adapted from Bauman & Lock (2008).
Table 2. Range of double bond positions in trans-18:1 and conjugated linoleic acid
(CLA) isomers and their ruminal outflow in growing and lactating cattle. Adapted from
Bauman & Lock (2006).
with the concept that these two fatty acid intermediates had escaped complete
biohydrogenation in the rumen and were subsequently absorbed from the
digestive tract and used for milk fat synthesis. Lines of evidence, however,
highlighted a number of inconsistencies with this premise. First, the kinetics of
rumen biohydrogenation are such that rumenic acid represents only a transitory
product, and vaccenic acid is the major biohydrogenation intermediate that
accumulates in the rumen (Harfoot & Hazlewood, 1997). Second, nutrition
studies demonstrated that increases in the milk fat content of rumenic acid
occurred when linseed oil and other dietary sources of linolenic acid were fed
(e.g., Kelly et al., 1998; Lock & Garnsworthy, 2002). As previously discussed,
rumenic acid is not an intermediate in the biohydrogenation of linolenic acid,
but the biohydrogenation of both linoleic and linolenic acids produces vaccenic
acid as an intermediate. Third, the ratio of vaccenic acid to rumenic acid is
> 50:1 in rumen fluid but only about 3:1 in milk fat. Based on these considera-
tions, Griinari and Bauman (1999) proposed that endogenous synthesis could
be an important source of the rumenic acid found in milk fat, with synthesis
involving the enzyme Δ9-desaturase and vaccenic acid as the substrate (see
Figure 2). Previous investigations with Δ9-desaturase from rat liver established
that while the preferred reaction was the conversion of stearic acid to oleic acid,
this enzyme could also desaturate positional isomers of trans-octadecenoic
acids (Mahfouz et al., 1980; Pollard et al., 1980).
The first study to show directly that milk fat CLA could originate via
endogenous synthesis was Griinari et al. (2000); they infused 12.5 g/d of
BIOSYNTHESIS AND BIOLOGICAL ACTIVITY OF RUMENIC ACID 201
Inhibition of Δ9-desaturase2
Hay/concentrate TMR 4 64% Griinari et al. (2000)
Hay/concentrate TMR 7 ⎫
plus PHVO 8 ⎭⎬ 78% Corl et al. (2001)
Pasture 16 > 91% Kay et al. (2004)
Rumen Output3
Grass silage/concentrate 8–11 > 80% Lock & Garnsworthy
plus various plant oils (2002)
Corn silage/alfalfa hay TMR 5–7 > 93% Piperova et al. (2002)
Grass silage/concentrate 19 > 74% Shingfield et al. (2003)
plus fish oil
Modeling of isotope kinetics4
Alfalfa silage and 4.8 80 Mosely et al. (2006)
hay/concentrate
1
Diet abbreviations are TMR (total mixed ration) and PHVO (partially hydrogenated vegetable oil).
2
Direct estimates of endogenous synthesis determined by use of sterculic oil as a source of cyclopropoic
fatty acids to inhibit Δ9-desaturase.
3
Indirect estimates determined by measuring rumen outflow of cis-9,trans-11 CLA and comparing this
to the quantity of cis-9,trans-11 CLA secreted in milk fat.
4
Based on modeling of isotope kinetics.
vaccenic acid into the abomasum of dairy cows and observed a 31% increase in
the concentration of rumenic acid in milk fat. This investigation clearly
demonstrated the potential for endogenous synthesis, but additional studies
were needed to determine its actual importance. To address this, three ap-
proaches were used (see Table 3). One approach was to inhibit Δ9-desaturase
directly, and this involved the use of sterculic oil which contains two
cyclopropene fatty acids, sterculic acid and malvalic acid, which are specific
inhibitors of Δ9-desaturase (Jeffcoat & Pollard, 1977). Griinari et al. (2000),
using this approach, estimated that 64% of the rumenic acid in milk fat was of
endogenous origin in cows fed a diet based on alfalfa hay and corn grain. This
represented the first direct demonstration that endogenous synthesis was the
major source of rumenic acid in milk fat. Subsequent investigations using the
same approach extended results to other dietary situations (total mixed diets
with or without plant oils and pasture) and in all cases endogenous synthesis
was the predominant source of the rumenic acid in milk fat (Corl et al., 2001;
Kay et al., 2004). Results from grazing cows are of special note because
pasture is high in linolenic acid and endogenous synthesis accounted for > 91%
of the total rumenic acid in milk fat (Kay et al., 2004).
202 TRANS FATTY ACIDS IN HUMAN NUTRITION
have been identified in mice, two in rats, and three in hamsters (Ntambi &
Miyazaki, 2004; Lengi & Corl, 2008).
Our understanding of the regulation of Δ9-desaturase in ruminants is limited,
with current knowledge coming mainly from investigations on rodents. Δ9-
desaturase has no known allosteric or feedback inhibition involving its substrates
or products; rather, it is regulated by dietary factors such as glucose and long-
chain PUFA, and by hormones such as insulin and glucagon (Ntambi &
Miyazaki, 2004). The enzyme protein has a relatively short half-life (~4 h) and,
thus, gene transcription is its major point of regulation (Ozols, 1997). Both,
PUFA and trans-10,cis-12 CLA, down-regulate its gene expression but rumenic
acid has no effect (Ntambi & Miyazaki, 2004). Interestingly, the cyclopropene
fatty acids in sterculic oil do not affect Δ9-desaturase gene expression or protein
synthesis, but they directly inhibit the activity of the enzyme (Gomez et al.,
2003).
The relationship between substrate and product for Δ9-desaturase is reflected
by the desaturase index, defined as the ratio [product of desaturase ÷ (product
+ substrate for desaturase)] (Kelsey et al., 2003). The desaturase index in milk
fat represents a proxy for Δ9-desaturase and a several-fold range is observed
among individuals providing a strong indication that there are genetic differ-
ences among individuals with respect to this enzyme. This is discussed in more
detail in Section B.5.
(Palmquist et al., 2005; Bauman & Lock, 2006), and estimates of digesta flow
indicate that rumen output is more than adequate to account for the trace
amounts secreted in milk fat (Piperova et al., 2002; Shingfield et al., 2003).
Furthermore, there has been no demonstration that other mammalian desaturases
act in a manner analogous to Δ9-desaturase to synthesize additional CLA
isomers endogenously from other trans-18:1 fatty acids. Thus, these CLA
isomers found at trace levels in milk fat are of rumen origin and represent
intermediates formed in the biohydrogenation of PUFA.
Information on the effect of diet on the production of minor isomers of CLA
in the rumen and alterations in their content in milk fat is limited. Best
described are diet-induced changes in trans-10,cis-12 CLA, and its biological
effects in the dairy cow and impact on milk fat synthesis (see reviews by
Bauman & Lock, 2006; Shingfield & Griinari, 2007; Bauman et al., 2008). The
ability to better characterize rumen biohydrogenation pathways and establish
their relationship to specific rumen bacteria and diets are important areas for
future research.
of milk fat. These changes are often associated with increased rumen outflow
of trans-10 18:1 and trans-10,cis-12 CLA which results in dramatic reductions
in milk fat synthesis (Bauman & Lock, 2006).
The most effective dietary treatments for increasing the rumenic acid content
of milk fat are those that increase both the supply of C18 PUFA and modify the
rumen environment. The most widely studied of these is the use of fresh
pasture, with numerous studies indicating that fresh pasture results in a 2-fold
to 3-fold increase in the rumenic acid content of milk fat (e.g., Stanton et al.,
1997; Dhiman et al., 1999). The degree of response, however, decreases as the
pasture matures and the proportion in the diet decreases. Correspondingly,
seasonal effects on milk rumenic acid content have been reported, with the
trend that content is greatest when fresh pasture is plentiful, and decreases
throughout the growing season (Auldist et al., 2002; Lock & Garnsworthy,
2003). These results cannot be explained fully in terms of the fatty acid
composition and supply of PUFA that grass provides; therefore, there must be
additional factors or components of grass that promote the production of
vaccenic acid in the rumen, and these lessen in effect as the pasture matures
(Lock & Bauman, 2004). Presumably, these factors inhibit the conversion of
vaccenic acid to stearic acid, as discussed previously. The effect of different
farming systems has also been investigated, with systems differentiated by the
amount and type of forages typically fed to cows. In general, production
systems with the greatest proportion of fresh forage in the diet often result in the
highest level of rumenic acid in milk fat.
Although the use of fresh pasture has striking effects on enhancing the
rumenic acid content of milk fat, a similar or even greater increase is possible
using standard dietary ingredients such as plant oils/oilseeds and fish oil/fish
meal supplements. Further, there is some indication that dietary regimes
involving a combination of supplements can have an additive effect on
increasing the level of rumenic acid in milk; for example Whitlock et al. (2002)
observed higher levels with a combination of plant oil and fish oil than when
either was fed alone. In all of the dietary situations designed to enhance the
level of rumenic acid in milk fat, it is vital that the normal vaccenic acid
pathway of biohydrogenation is maintained. If shifts in biohydrogenation
occur, then the pattern of trans-18:1 fatty acids is altered resulting in a
reduction in the rumen output of vaccenic acid, and as a consequence a
reduction in the level of rumenic acid in milk fat.
The final method to enhance the level of rumenic acid in milk is to increase
endogenous synthesis and this must be the basis for the variation among cows
in a herd fed the same diet. Undoubtedly, the variation in Δ9-desaturase among
individuals has a genetic basis but this has not been examined directly.
However, an indirect evaluation is possible because milk fat contains four
major fatty acid pairs that represent a product/substrate relationship for Δ9-
desaturase, myristoleic/myristic acid, palmitoleic/palmitic acid, oleic/stearic
208 TRANS FATTY ACIDS IN HUMAN NUTRITION
acid and rumenic acid/vaccenic acid. As mentioned earlier, ratios for these
pairs of fatty acids, referred to as a desaturase index, represent a proxy for Δ9-
desaturase activity. In the largest study to examine this, Kelsey et al. (2003)
reported a 3-fold difference in both milk fat content of rumenic acid and
desaturase index among individuals consuming the same diet. Other investiga-
tions have also observed a 2-fold to 3-fold range in desaturase index among
cows in the same herd (e.g. Lock & Garnsworthy, 2002; Peterson et al., 2002).
Peterson et al. (2002) also demonstrated a consistency in the individual hierar-
chy in desaturase index over time when cows were fed the same diet and a
consistency in the individual hierarchy when cows were switched between
diets. Furthermore, the same studies also observed a similar 2-fold to 3-fold
range in the milk fat content of rumenic acid among individual animals (Lock
& Garnsworthy, 2002; Peterson et al., 2002).
In a structured genetic study involving 2400 cows, the heritability in the
desaturase index (using myristoleic/myristic acid ratio) was approximately 0.3
(Garnsworthy & Royal, 2005). Presumably this variation reflects individual
differences in the activity of Δ9-desaturase involving the regulation of gene
expression, primary or tertiary structure of the enzyme due to gene
polymorphisms, post-translational modifications, or other factors affecting the
interaction between the enzyme and the substrate or product. Furthermore, it
was recently demonstrated that the myristoleic/myristic acid ratio correlates
well with concentration of Δ9-desaturase mRNA in somatic cells extracted
from milk, and that between-cow variability seems to account for more
variation in Δ9-desaturase mRNA than does stage of lactation or diet (Feng
et al., 2007). Thus, cows could potentially be selected for high mammary
expression of Δ9-desaturase. Whether this would have a significant impact on
milk fatty acid composition, however, remains to be established.
Several specific physiological factors have been examined for effects on the
level of CLA in milk fat, but because of the large impact of diet and the wide
range among individuals, it is important that these comparisons involve a
reasonable number of cows fed a common diet. Studies meeting these condi-
tions found that the rumenic acid content of milk fat and the desaturase index
had no relationship to parity or stage of lactation (days in milk) (Kelsey et al.,
2003; Lock et al., 2005a). Likewise, they observed that milk fat content of
rumenic acid and desaturase index also had no relationship to milk yield, milk
fat percent or yield of milk fat (Kelsey et al., 2003; Lock et al., 2005a). The
investigation by Kelsey et al. (2003) involved over 200 cows fed the same diet
and found no difference between Holstein and Brown Swiss breeds. In contrast,
several studies have reported breed differences in the rumenic acid content of
milk fat (e.g. Lawless et al., 1999; Whitlock et al., 2002), which may reflect
differences in desaturase index among breeds. However, these studies often
involved very few animals or were confounded by diet, or both. Using a larger
data set, DePeters et al. (1995) reported breed differences in the desaturase
BIOSYNTHESIS AND BIOLOGICAL ACTIVITY OF RUMENIC ACID 209
index in milk fat of dairy cows, consistent with the suggestion that the activity
of Δ9-desaturase is higher in Holstein than in Jersey mammary tissue (Beaulieu
& Palmquist, 1995). If breed differences exist they would, however, appear to
be minor compared with the magnitude of dietary effects and variation among
cows in terms of both the rumenic acid content of milk fat and desaturase index.
levels of rumenic acid and vaccenic acid. Two studies have reported adverse
effect of rumenic acid enrichment on sensory and storage properties of milk;
however, one of these studies used extremely high levels of fish oil in the cows’
diet (Lacasse et al., 2002) and the other study involved a post-harvest fortifica-
tion of milk fat with synthetic CLA during the manufacturing process (Campbell
et al., 2003). Overall, results to date indicate that manufacturing and quality
characteristics were similar for dairy products naturally-enriched with CLA
compared with regular dairy products, and consumer acceptability was compa-
rable to un-enriched dairy products.
De la Torre et al., 2005; Beppu et al., 2006; Kim et al., 2006). However, in
most of those studies, with the exception of those conducted in colon cancer
cell lines, the trans-10,cis-12 CLA isomer was much more efficacious than
rumenic acid. Yet, other studies showed that rumenic acid did not repress cell
proliferation in the colon cancer cells Caco-2 and HT-29 (Kim et al., 2002; Cho
et al., 2003; Lee et al., 2006). On the contrary, when human colon cancer cells
(SW480) were treated with milk fat triacylglycerol-bound CLA (consisting
primarily of rumenic acid) cell growth was significantly suppressed (Miller
et al., 2003b).
One study evaluated the effectiveness of rumenic acid on RLh-84 (rat
hepatoma cell line) where rumenic acid was found to be a growth promoter
rather than an inhibitor (Yamasaki et al., 2002). The data suggest that rumenic
acid may elicit different effects in various cell lines.
Taken together, the inhibitory potency of rumenic acid varies considerably
among different cell lines, depending also on experimental conditions, such as
concentration of rumenic acid and duration of the treatment. Clearly, in vitro
data with cell cultures demonstrate that the pure single isomer rumenic acid was
only moderately effective against treatment of cancer especially when com-
pared with other pure CLA isomers. However, it is conceivable that rumenic
acid derived from milk is more proficient in suppressing tumour cell growth
than the synthetic single isomer rumenic acid, indicating that its precursor
vaccenic acid (or other active components in milk) are involved in the modu-
lation of tumour cell growth.
2. Animal studies
Because of the difficulty in obtaining pure rumenic acid for animal studies,
previous studies on animal models in cancer research focused on CLA mix-
tures, mainly rumenic acid and trans-10,cis-12 in a ratio of 1:1. Thus, there
have been only a few animal studies documenting effects of rumenic acid in
decreasing the risk of various cancer types, many of which utilized rumenic
acid enriched milk fat.
Diets, containing either rumenic acid enriched butter fat (0.8% rumenic acid)
or 1% purified rumenic acid, fed to rats 30 days prior to N-nitroso-N-methyl
nitrosourea (MNU) treatment, were found to be equally effective in inhibiting
mammary tumorigenesis (reduced mammary epithelial mass, decreased size of
the terminal end bud population, suppressed proliferation of terminal end bud
population cells, and inhibited mammary tumour yield (Table 5) (Ip et al.,
1999). In a subsequent study, the same group verified that dietary rumenic acid
at a dose level of 0.5% fed to rats for 6 or 24 weeks after MNU administration
was also effective in inhibiting the development of premalignant lesions (33–
36%) and the total number of carcinomas (35–40%) in the mammary gland (Ip
et al., 2002). In a similar study, 1% rumenic acid fed to rats for 6 or 20 weeks
BIOSYNTHESIS AND BIOLOGICAL ACTIVITY OF RUMENIC ACID 213
3. Human studies
Human studies of rumenic acid and cancer have been scarce. Evaluating the
specific role of rumenic acid in health maintenance and the prevention of
cancer in humans is difficult, since cancer takes many years to develop. Thus,
BIOSYNTHESIS AND BIOLOGICAL ACTIVITY OF RUMENIC ACID 215
the atheroma formation. This leads to a proliferation of certain cell types within
the arterial wall that gradually impinge on the vessel lumen and impede blood
flow. Atherosclerosis is the primary cause of CHD around the world, and in
North America, is the underlying cause of nearly half of all deaths (Lusis,
2000).
1. Animal studies
Changes in both total plasma cholesterol and individual lipoprotein cholesterol
concentrations have been implicated as major determinants of atherosclerosis
disease risk. This has led to a number of studies specifically investigating the
effects of rumenic acid on cholesterol and lipoprotein metabolism. The ability
of rumenic acid to alter blood lipid profiles has been tested in various animal
models, such as the apoE knock-out mouse, the Golden Syrian hamster, and the
New Zealand White rabbit. In apoE–/– mice, diets rich in rumenic acid de-
creased triacylglycerol and triacylglycerol-rich lipoproteins and raised
high-density lipoprotein cholesterol (HDL-C) in diabetic and non-diabetic
mice, respectively (Nestel et al., 2006). In Golden Syrian hamsters, rumenic
acid has been demonstrated to have neutral and/or beneficial effects on blood
lipids. Mitchell et al. (2005) showed that when supplemented in amounts as
high as 1% of diet, rumenic acid had no effect on blood lipid profile. However,
Ledoux et al. (2007) determined that adding 1% rumenic acid to the diet
decreased total cholesterol, low-density lipoprotein cholesterol (LDL-C), and
HDL-C, when compared to animals fed control diets. Differences in the control
diets used in these studies should be noted. In the study in which rumenic acid
had neutral effects, it was compared to a diet high in non-conjugated linoleic
acid isomers. In the study in which rumenic acid had beneficial effects, it was
compared to a diet rich in oleic acid. It is noteworthy that both oleic and linoleic
acid have been shown to reduce total cholesterol and LDL-C (Mensink et al.,
2003). These results indicate that consuming rumenic acid may improve
cholesterol levels, and certainly not adversely affect them. While both of the
above studies were conducted with pure rumenic acid sources, other studies
have been conducted using whole foods to deliver the experimental rumenic
acid. Valeille et al. (2005) supplemented butter with ~1% rumenic acid which
decreased the non-HDL-C to HDL-C ratio in Syrian Golden hamsters com-
pared to hamsters fed regular butter. In another study, when butter naturally
enriched with rumenic acid (~1%) and vaccenic acid was fed to Golden Syrian
hamsters, total cholesterol, LDL-C, and non-HDL-C to HDL-C ratio were
lower than in those fed a control diet (Valeille et al., 2006). This is in agreement
with Lock et al. (2005b) who also used the Golden Syrian hamster to examine
the potential of rumenic acid when fed as a component of a functional food
(butter enriched in vaccenic acid and rumenic acid) as part of a diet that was
high in cholesterol (0.2%) and fat (20%). Compared with the control animals,
218 TRANS FATTY ACIDS IN HUMAN NUTRITION
12 a
Control
Cholesterol (mmol/litre) 10 VA/RA-enriched butter
b
8
4
a
2 b
a
b
0
Total LDL HDL LDL/HDL
Plasma Cholesterol
Figure 3. Effect of butter enriched in vaccenic and rumenic acids (VA/RA) on plasma cholesterol
levels in the cholesterol-fed Golden Syrian hamster. Adapted from Lock et al. (2005b). Standard error is
indicated by error bars over each column. P < 0.01 for treatment effects with significance differences
indicated by letter differences (a, b) over the columns.
those fed the enriched butter showed a number of beneficial effects, including
reduced total plasma cholesterol, and very-low-density lipoprotein cholesterol
(VLDL) and LDL, suggesting that CLA may modify the production of athero-
genic lipoproteins by the liver (Figure 3). In addition, the enriched butter
produced a less atherogenic profile than an equivalent diet in which the
enriched butter was replaced with trans fatty acids from partially hydrogenated
vegetable oil (Lock et al., 2005b). A similar study in New Zealand White
rabbits determined that rumenic acid enriched butter had a neutral effect on
plasma lipoprotein cholesterol (Bauchart et al., 2007). Finally, rumenic acid
was reported to have neutral effects on plasma and hepatic cholesterol, as well
as hepatic esterified cholesterol, in rabbits, when compared to animals fed a
cholesterol-free diet (Kritchevsky et al., 2004).
Although rumenic acid has been widely recognised for its ability to improve
plasma cholesterol and lipoprotein profiles, perhaps more impressive is evi-
dence demonstrating its ability to reduce the size of atherosclerotic lesions, and
halt the progression of such lesions in animal models. Supplementation of a
high-fat, high-cholesterol diet with 1% rumenic acid resulted in fewer fatty
streak lesions in the aortas of Golden Syrian hamsters when compared with
those fed isomers of non-conjugated linoleic acid (Mitchell et al., 2005).
Although findings were not statistically significant, there were nearly 20%
fewer fatty lesions present than in control animals. New Zealand White rabbits
fed a 50:50 mixture of rumenic acid and trans-10,cis-12 CLA isomer mixture
BIOSYNTHESIS AND BIOLOGICAL ACTIVITY OF RUMENIC ACID 219
at 0.1% of diet exhibited a lower plaque diameter in the abdominal aorta than
animals fed a CLA-free diet (Kritchevsky et al., 2000). When the isomer blend
was increased to 0.5% and 1.0%, plaque thickness was decreased in both the
thoracic aorta and aortic arch (Kritchevsky et al., 2000), indicating a decrease
in the progression of atherosclerosis. Rumenic acid, alone, significantly re-
duced the size of lesions in both the thoracic aorta and aortic arch of New
Zealand White rabbits fed an atherogenic diet (Kritchevsky et al., 2004). In a
recent study in which butter was naturally-enriched either with vaccenic acid
and rumenic acid or with trans-10 18:1, the trans-10 18:1-enriched butter was
found to have detrimental effects as evidenced by increases in plasma concen-
trations of total cholesterol, LDL-C, VLDL-C, and non-HDL-C:HDL-C ratio
(Roy et al., 2007). In contrast, vaccenic/rumenic enriched butter had neutral
effects on blood lipid profiles and tended to reduce aortic lipid deposition.
It is important to note that an elevated and/or altered plasma lipid level is
only one of a wide range of risk factors that contribute to the clinical manifes-
tations of atherosclerosis in humans (Lusis, 2000). Consequently, in some
studies, the reduced incidence of atherosclerosis in animals fed CLA isomers
was not accompanied by an improvement in the plasma lipid profile during the
CLA-feeding phase (e.g. Wilson et al., 2000). Reasons for these effects are not
fully understood. Atherosclerosis, however, is a chronic inflammatory disease
(Libby, 2002) and several important anti-inflammatory effects have been
associated with rumenic acid.
As previously stated, when Golden Syrian hamsters were fed butter supple-
mented with rumenic acid, they showed improved cholesterol profiles and
lower total fatty streak development in the aorta (Valeille et al., 2005). These
hamsters also had lower inflammatory serum amyloid A levels and improved
anti-oxidized LDL paraoxonase activity; as well as down-regulated expression
of proinflammatory cytokines such as tumour necrosis factor-α (TNF-α) and
interleukin-1β (IL-1β); decreased cyclooxygenase 2 (COX-2) activation of the
transcription factor peroxisome proliferator-activating receptor (PPAR)/liver X
receptor (LXR)-α signalling cascade; and increased ATP-binding cassette
subfamily A1 expression in the aorta (Valeille et al., 2005). Similar results
were demonstrated in a study in which hamsters were fed a diet supplemented
with up to 9 g/kg of rumenic acid as rumenic acid enriched milk fat. Hamsters
fed a high-fat diet (67 g/100 g saturated fatty acids), supplemented with 2.6%
of lipids from rumenic acid, had increased ATP-binding cassette subfamily A-
1 gene expression in the aorta, decreased LDL-peroxidability index, and
down-regulation of the proinflammatory IL-1β gene in the aorta when com-
pared to hamsters fed the control diet without rumenic acid supplementation
(Valeille et al., 2006).
In apo E–/– mice, a high-fat, high-cholesterol diet containing 7% rumenic acid
increased five post-translationally modified forms of heat shock protein 70kD
(HSP 70kD) (de Roos et al., 2005). In humans, elevated levels of this protein
220 TRANS FATTY ACIDS IN HUMAN NUTRITION
are associated with low coronary artery disease risk, perhaps due to its inverse
association with the activity of COX-2, a proinflammatory cytokine (de Roos
et al., 2005). HSP 70kD attenuates NF-kB activation, an important transcrip-
tion factor in the expression of proinflammatory genes (de Roos et al., 2005),
which lends insight into the mechanism behind the ability of rumenic acid to
reduce inflammation.
2. Human studies
Since results from studies with biomedical models indicate potential beneficial
effects of rumenic acid, there is obvious interest in the effects of rumenic acid
consumption in foods on the risk of atherosclerosis in humans. The use of
surrogate biomarkers for disease risk is more readily achievable for atheroscle-
rosis than for cancer in humans and a number of genetic and environmental risk
factors have been identified, with the relative abundance of the different
lipoproteins being of primary importance (Lusis, 2000). To date, there have been
no epidemiological studies that have examined the relationship between intake
of rumenic acid derived from foods and risk of atherosclerosis. Several human
intervention studies involving dietary supplements of rumenic acid in the form of
capsules have reported plasma lipid variables as secondary observations, but
most used mixed isomers of CLA and found inconsistent results (see Bauman
et al., 2006). However, recent studies examined the specific effects of rumenic
acid on blood lipids in healthy subjects. In normal healthy men and women, a
capsule containing a mixture of rumenic acid and trans-10,cis-12 CLA, admin-
istered three times daily with food, resulted in reduced triacylglycerol levels
when administered in a 50:50 ratio, and decreased VLDL-C concentrations when
administered at an 80:20 ratio, respectively (Noone et al., 2002). When rumenic
acid and trans-10,cis-12 CLA where examined against each other in a dose-
response study in healthy men, rumenic acid supplementation consistently
resulted in lower total cholesterol, LDL-C, and LDL-C:HDL-C, with a dose-
response observed for total cholesterol and LDL-C (Tricon et al., 2004). In an
attempt to examine the effects of rumenic acid on lipoprotein profiles in the
context of the human experience, foods naturally enriched with rumenic acid
were fed to healthy men. At the conclusion of 6 weeks of eating naturally-
enriched rumenic acid milk, butter, and cheese alongside a normal diet, there
were no significant changes in total cholesterol, LDL-C, or HDL-C (Tricon et al.,
2006), although it is worth noting that subjects consumed twice the level of dairy
fat compared with baseline values. The enriched dairy foods changed LDL-C
fatty acid composition, but had no significant effect on LDL particle size or
susceptibility to oxidation (Tricon et al., 2006). These studies indicated that
human consumption of naturally-occurring rumenic acid, such as from dairy
products, had a neutral effect on lipoprotein profiles, with the potential for a
protective effect, dependent on dose.
BIOSYNTHESIS AND BIOLOGICAL ACTIVITY OF RUMENIC ACID 221
E. Summary
Research in animal science has traditionally focused on the productivity and
well-being of food-producing animals. However, there is also a growing
recognition of the consumer desire for healthier and more nutritious foods. The
contributions of animal-derived foods in supplying essential nutrients have
been recognized for some time, but consumers are increasingly aware that
foods also contain components that can have positive effects on health mainte-
nance and the prevention of chronic diseases. Referred to as ‘bioactive food
components’ or ‘functional foods’, CLA in milk fat is such a component. Milk
fat contains many isomers of CLA, but rumenic acid predominantly, and this is
the CLA isomer with bioactive potential in relation to cancer and atherosclero-
sis. The uniqueness of rumenic acid in ruminant-derived foods is related to
rumen biohydrogenation of PUFA. Research over the last decade has estab-
lished that rumenic acid in milk fat originates mainly from endogenous
synthesis by mammary Δ9-desaturase from vaccenic acid, a biohydrogenation
intermediate produced in the rumen. Thus, an understanding of dietary lipid
supply, rumen fermentation and mammary synthesis of fat is required and
strategies to increase milk fat rumenic acid centre on enhancing rumen output
of vaccenic acid and increasing tissue activity of Δ9-desaturase.
The anti-carcinogenic activity of rumenic acid has been demonstrated with
animal models and in vitro studies for a wide range of cancer types. The anti-
cancer effects of rumenic acid have been particularly impressive in biomedical
models of breast cancer. By comparison, investigations of rumenic acid effects
on atherosclerosis are more limited. A number of studies using animal models
222 TRANS FATTY ACIDS IN HUMAN NUTRITION
have reported that dietary supplementation with rumenic acid can improve the
plasma profile of lipids and lipoproteins, reduce the development of atheroscle-
rotic lesions, and indeed even induce the regression of pre-existing lesions. The
rumenic acid used for most studies of cancer and atherosclerosis has been
synthetically produced from vegetable oils. Of special importance are recent
results showing that beneficial effects were also achieved when the dietary
supply of rumenic acid was provided by a natural food. These studies are
among the first to demonstrate the feasibility of a functional food approach
using the natural form of rumenic acid (esterified) in a naturally-enriched food
(dairy products). Furthermore, these investigations also established that the
vaccenic acid in milk fat is a functional food component with beneficial effects
on cancer and atherosclerosis because it can be used for endogenous synthesis
of rumenic acid in humans.
Extending the results from biomedical models to implications of rumenic
acid and vaccenic acid as functional food components in humans is limited and
problematic. These evaluations are challenging because the development of
cancer and atherosclerosis has long latency periods and there is a lack of
consensus biomarkers, especially for cancer. Further, there have been signifi-
cant advances in the analysis of rumenic acid and specific trans fatty acids, and
these techniques have not yet been applied to updating values in food databases.
Finally, successful application of bioactive food components found in milk fat
includes the education of the public that not all fatty acids have the same
biological effects. This is of special importance with the introduction of food
labelling of trans fatty acid content as undesirable and the fact that both
vaccenic acid and rumenic acid are trans fatty acids. Overall, studies summa-
rized in this review demonstrate the exciting potential of rumenic acid and
vaccenic acid in dairy products as bioactive food components that may provide
benefits in health maintenance and disease prevention as well as improve the
public perception of milk and dairy products.
References
Abu-Ghazaleh AA, Schingoethe DJ and Hippen AR (2001) Conjugated linoleic acid and
other beneficial fatty acids in milk from cows fed soybean meal, fish meal, or both. J.
Dairy Sci., 84, 1845–1850.
Abu-Ghazaleh AA and Jenkins TC (2004) Short communication. Docosahexaenoic acid
promotes vaccenic acid accumulation in mixed ruminal cultures when incubated with
linoleic acid. J. Dairy Sci., 87, 1047–1050.
Adlof RO, Duval S and Emken EA (2000) Biosynthesis of conjugated linoleic acid in humans.
Lipids, 35, 131–135.
Aro A, Mannisto S, Salminen I, Ovaskainen ML, Kataja V and Uusitupa M (2000) Inverse
association between dietary and serum conjugated linoleic acid and risk of breast cancer
in postmenopausal women. Nutr. Cancer, 38, 151–157.
Auldist MJ, Kay JK, Thomson NA, Napper AR and Kolver ES (2002) Brief communication.
Concentrations of conjugated linoleic acid in milk from cows grazing pasture or fed a total
mixed ration for an entire lactation. Proc. New Zealand Soc. Anim. Prod., 62, 240–247.
BIOSYNTHESIS AND BIOLOGICAL ACTIVITY OF RUMENIC ACID 223
Baer RJ, Ryali J, Schingoethe DJ, Kasperson KM, Donovan DC, Hippen AR and Franklin ST
(2001) Composition and properties of milk and butter from cows fed fish oil. J. Dairy Sci.,
84, 345–353.
Banni S, Angioni E, Murru E, Carta G, Melis MP, Bauman DE, Dong Y and Ip C (2001)
Vaccenic acid feeding increases tissue levels of conjugated linoleic acid and suppresses
development of premalignant lesions in rat mammary gland. Nutr. Cancer, 41, 91–97.
Banni S, Heys SD and Wahle KWJ (2003) Conjugated linoleic acids as anticancer nutrients:
studies in vivo and cellular mechanisms. In: Advances in Conjugated Linoleic Acid
Research, Volume 2 (JL Sebedio, WW Christie and RO Adlof, eds) AOCS Press,
Champaign, Illinois, USA.
Barber MC, Ward RJ, Richards SE, Salter AM, Buttery PJ, Vernon RG and Travers MT
(2000) Ovine adipose tissue monounsaturated fat content is correlated to depot-specific
expression of the stearoyl-CoA desaturase gene. J. Anim. Sci., 78, 62–68.
Bauchart D, Roy A, Lorenz S, Chardigny JM, Ferlay A, Gruffat D, Sebedio JL, Chilliard Y
and Durand D (2007) Butters varying in trans 18:1 and cis-9,trans-11 conjugated linoleic
acid modify plasma lipoproteins in the hypercholesterolemic rabbit. Lipids, 42, 123–133.
Bauman DE, Barbano DM, Dwyer DA and Griinari JM (2000) Technical note. Production
of butter with enhanced conjugated linoleic acid for use in biomedical studies with animal
models. J. Dairy Sci., 83, 2422–2425.
Bauman DE and Lock AL (2006) Conjugated linoleic acid: biosynthesis and nutritional
significance. In: Advanced Dairy Chemistry, Volume 2: Lipids, 3rd Edition, (PF Fox and
PLH McSweeney, eds) Kluwer Academic/Plenum Publishers, New York, USA, pp.93–
136.
Bauman DE and Lock AL (2008) Modifying animal fat to enhance animal and human health.
Proc. 29th Canadian Western Nutrition Conference, pp.265–277.
Bauman DE, Lock AL, Corl BA, Salter AM, Ip C and Parodi PW (2006) Milk fatty acids and
human health potential role of conjugated linoleic acid and trans fatty acids. In: Ruminant
Physiology: Digestion, Metabolism and Impact of Nutrition on Gene Expression, Immu-
nology and Stress (K Sejrsen, T Hvelplund and MO Nielsen, eds) Wageningen Academic
Publishers, Wageningen, The Netherlands, pp.523–555.
Bauman DE, Perfield JW, 2nd, Harvatine KJ and Baumgard LH (2008) Regulation of fat
synthesis by conjugated linoleic acid: lactation and the ruminant model. J. Nutr., 138,
403–409.
Beaulieu AD and Palmquist DL (1995) Differential-effects of high-fat diets on fatty-acid
composition in milk of Jersey and Holstein cows. J. Dairy Sci., 78, 1336–1344.
Bell JA, Griinari JM and Kennelly JJ (2006) Effect of safflower oil, flaxseed oil, monensin,
and vitamin E on concentration of conjugated linoleic acid in bovine milk fat. J. Dairy
Sci., 89, 733–748.
Belury MA (2002) Dietary conjugated linoleic acid in health: physiological effects and
mechanisms of action. Annu. Rev. Nutr., 22, 505–531.
Beppu F, Hosokawa M, Tanaka L, Kohno H, Tanaka T and Miyashita K (2006) Potent
inhibitory effect of trans9,trans11 isomer of conjugated linoleic acid on the growth of
human colon cancer cells. J. Nutr. Biochem., 17, 830–836.
Bickerstaffe R and Annison EF (1969) Glycerokinase and desaturase activity in pig, chicken
and sheep intestinal epithelium. Comp. Biochem. Physiol., 31, 47–54.
Bickerstaffe R and Annison EF (1970) The desaturase activity of goat and sow mammary
tissue. Comp. Biochem. Physiol., 35, 653–665.
Booth RG, Dann WJ, Kon SK and Moore T (1933) A new variable factor in butter fat. Chem.
Ind., 52, 270.
Campbell W, Drake MA and Larick DK (2003) The impact of fortification with conjugated
linoleic acid (CLA) on the quality of fluid milk. J. Dairy Sci., 86, 43–51.
Chang JHP, Lunt DK and Smith SB (1992) Fatty acid composition and fatty acid elongase
224 TRANS FATTY ACIDS IN HUMAN NUTRITION
and stearoyl-CoA desaturase activities in tissues of steers fed high oleate sunflower seed.
J. Nutr., 122, 2074–2080.
Chen BQ, Yang YM, Gao YH, Liu JR, Xue YB, Wang XL, Zheng YM, Zhang JS and Liu
RH (2003) Inhibitory effects of c9,t11-conjugated linoleic acid on invasion of human
gastric carcinoma cell line SGC-7901. World J. Gastroenterol., 9, 1909–1914.
Chilliard Y, Ferlay A and Doreau M (2001) Effect of different types of forages, animal fat
or marine oils in cow’s diet on milk fat secretion and composition, especially conjugated
linoleic acid (CLA) and polyunsaturated fatty acids. Livest. Prod. Sci., 70, 31–48.
Cho HJ, Lee HS, Chung CK, Kang YH, Ha YL, Park HS and Park JH (2003) Trans-10, cis-
12 conjugated linoleic acid reduces insulin-like growth factor-II secretion in HT-29
human colon cancer cells. J. Med. Food, 6, 193–199.
Chujo H, Yamasaki M, Nou S, Koyanagi N, Tachibana H and Yamada K (2003) Effect of
conjugated linoleic acid isomers on growth factor-induced proliferation of human breast
cancer cells. Cancer Lett., 202, 81–87.
Chung M, Ha S, Jeong S, Bok J, Cho K, Baik M and Choi Y (2000) Cloning and
characterization of bovine stearoyl CoA desaturase 1 cDNA from adipose tissues. Biosci.
Biotechnol. Biochem., 64, 1526–1530.
Corl BA, Barbano DM, Bauman DE and Ip C (2003) cis-9,trans-11 CLA derived endogenously
from trans-11 18:1 reduces cancer risk in rats. J. Nutr., 133, 2893–2900.
Corl BA, Baumgard LH, Dwyer DA, Griinari JM, Phillips BS and Bauman DE (2001) The
role of delta(9)-desaturase in the production of cis-9,trans-11 CLA. J. Nutr. Biochem., 12,
622–630.
De la Torre A, Debiton E, Durand D, Chardigny JM, Berdeaux O, Loreau O, Barthomeuf C,
Bauchart D and Gruffat D (2005) Conjugated linoleic acid isomers and their conjugated
derivatives inhibit growth of human cancer cell lines. Anticancer Res., 25, 3943–3949.
de Roos B, Rucklidge G, Reid M, Ross K, Duncan G, Navarro MA, Arbones-Mainar JM,
Guzman-Garcia MA, Osada J, Browne J, Loscher CE and Roche HM (2005) Divergent
mechanisms of cis-9,trans-11- and trans-10, cis-12-conjugated linoleic acid affecting
insulin resistance and inflammation in apolipoprotein E knockout mice: a proteomics
approach. FASEB Journal, 19, 1746–1748. Epub.
Delmonte P, Kataoka A, Corl BA, Bauman DE and Yurawecz MP (2005) Relative retention
order of all isomers of cis/trans conjugated linoleic acid FAME from the 6,8 to 13,15
positions using silver ion HPLC with 2 different elution system. Lipids, 40, 509–514.
DePeters EJ, Medrano JF and Reed BA (1995) Fatty acid composition of milk fat from three
breeds of dairy cattle. Can. J. Anim. Sci., 75, 264–269.
Dhiman TR, Anand GR, Satter LD and Pariza MW (1999) Conjugated linoleic acid content
of milk from cows fed different diets. J. Dairy Sci., 82, 2146–2156.
Fellner V, Sauer FD and Kramer JKG (1997) Effect of nigericin, monensin, and tetronasin
on biohydrogenation in continuous flow-through ruminal fermenters. J. Dairy Sci., 80,
921–928.
Feng S, Salter AM, Parr T and Garnsworthy PC (2007) Extraction and quantitative analysis
of stearoyl-coenzyme A desaturase mRNA from dairy cow milk somatic cells. J. Dairy
Sci., 90, 4128–4136.
Fite A, Goua M, Wahle KW, Schofield AC, Hutcheon AW and Heys SD (2007) Potentiation
of the anti-tumour effect of docetaxel by conjugated linoleic acids (CLAs) in breast
cancer cells in vitro. Prostaglandins Leukot. Essent. Fatty Acids, 77, 87–96.
Franklin ST, Martin KR, Baer RJ, Schingoethe DJ and Hippen AR (1999) Dietary marine
algae (Schizochytrium sp.) increases concentrations of conjugated linoleic, docosa-
hexaenoic and trans vaccenic acids in milk of dairy cows. J. Nutr., 129, 2048–2054.
Garnsworthy PC and Royal MD (2005) Heritability of delta-9 desaturase enzyme activity in
dairy cows. Proc. 26th World Congress of the International Society for Fat Research,
Prague, Czech Republic, pp.55.
BIOSYNTHESIS AND BIOLOGICAL ACTIVITY OF RUMENIC ACID 225
Glaser KR, Wenk C and Scheeder MRL (2002) Effects of feeding pigs increasing levels of
C18:1 trans fatty acids on fatty acid composition of backfat and intramuscular fat as well
as backfat firmness. Arch. Anim. Nutr., 56, 117–130.
Gomez FE, Bauman DE, Ntambi JM and Fox BG (2003) Effects of sterculic acid on stearoyl-
CoA desaturase in differentiating 3T3-L1 adipocytes. Biochem. Biophys. Res. Commun.,
300, 316–326.
Gonzalez S, Duncan SE, O’Keefe SF, Sumner SS and Herbein JH (2003) Oxidation and
textural characteristics of butter and ice cream with modified fatty acid profiles. J. Dairy
Sci., 86, 70–77.
Griinari JM and Bauman DE (1999) Biosynthesis of conjugated linoleic acid and its
incorporation into meat and milk in ruminants. In: Advances in Conjugated Linoleic Acid
Research (MP Yurawecz, MM Mossoba, JKG Kramer, MW Pariza and G Nelson, eds)
AOCS Press, Champaign, Illinois, USA, pp.180–200.
Griinari JM, Corl BA, Lacy SH, Chouinard PY, Nurmela KVV and Bauman DE (2000)
Conjugated linoleic acid is synthesized endogenously in lactating dairy cows by delta(9)-
desaturase. J. Nutr., 130, 2285–2291.
Ha YL, Grimm NK and Pariza MW (1987) Anticarcinogens from ground beef: heat altered
derivatives of linoleic acid. Carcinogenesis, 8, 1881–1887.
Harfoot CG and Hazlewood GP (1997) Lipid metabolism in the rumen. In: The Rumen
Microbial Ecosystem (Second Edition) (PN Hobson and DS Stewart, editors) Chapman
& Hall, London, UK, pp.382–426.
Hubbard NE, Lim D and Erickson KL (2003) Effect of separate conjugated linoleic acid
isomers on murine mammary tumorigenesis. Cancer Lett., 190, 13–19.
Ip C, Banni S, Angioni E, Carta G, McGinley J, Thompson HJ, Barbano D and Bauman DE
(1999) Conjugated linoleic acid-enriched butter fat alters mammary gland morphogen-
esis and reduces cancer risk in rats. J. Nutr., 129, 2135–2142.
Ip C, Dong Y, Ip MM, Banni S, Carta G, Angioni E, Murru E, Spada S, Melis MP and Saebo
A (2002) Conjugated linoleic acid isomers and mammary cancer prevention. Nutr.
Cancer, 43, 52–58.
Ip MM, McGee SO, Masso-Welch PA, Ip C, Meng X, Ou L and Shoemaker S (2007) The
t10,c12 isomer of conjugated linoleic acid stimulates mammary tumorigenesis in
transgenic mice overexpressing erbB2 in the mammary epithelium. Carcinogenesis, 28,
1269–1276.
Jahreis G, Fritsche J and Steinhart H (1997) Conjugated linoleic acid in milk fat: high
variation depending on production system. Nutr. Res., 17, 1479–1484.
Jeffcoat R and Pollard MR (1977) Studies on the inhibition of the desaturases by cyclopropenoid
fatty acids. Lipids, 12, 480–485.
Jenkins TC, Fellner V and McGuffey RK (2003) Monensin by fat interactions on trans fatty
acids in cultures of mixed ruminal microorganisms grown in continuous fermentors fed
corn or barley. J. Dairy Sci., 86, 324–330.
Jiang J, Bjoerck L, Fonden R and Emanuelson M (1996) Occurrence of conjugated cis-
9,trans-11-octadecadienoic acid in bovine milk: effects of feed and dietary regimen. J.
Dairy Sci., 79, 438–445.
Kay JK, Mackle TR, Auldist MJ, Thomson NA and Bauman DE (2004) Endogenous
synthesis of cis-9,trans-11 conjugated linoleic acid in dairy cows fed fresh pasture. J.
Dairy Sci., 87, 369–378.
Kelly ML, Berry JR, Dwyer DA, Griinari JM, Chouinard PY, Van Amburgh ME and Bauman
DE (1998) Dietary fatty acid sources affect conjugated linoleic acid concentrations in
milk from lactating dairy cows. J. Nutr., 128, 881–885.
Kelsey JA, Corl BA, Collier RJ and Bauman DE (2003) The effect of breed, parity, and stage
of lactation on conjugated linoleic acid (CLA) in milk fat from dairy cows. J. Dairy Sci.,
86, 2588–2597.
226 TRANS FATTY ACIDS IN HUMAN NUTRITION
Kim EJ, Holthuizen PE, Park HS, Ha YL, Jung KC and Park JH (2002) Trans-10, cis-12-
conjugated linoleic acid inhibits Caco-2 colon cancer cell growth. Am. J. Physiol.
Gastrointest. Liver Physiol., 283, G357–367.
Kim EJ, Shin HK, Cho JS, Lee SK, Won MH, Kim JW and Park JH (2006) Trans-10,cis-12
conjugated linoleic acid inhibits the G1-S cell cycle progression in DU145 human
prostate carcinoma cells. J. Med. Food, 9, 293–299.
Kinsella JE (1972) Stearyl CoA as a precursor of oleic acid and glycerolipids in mammary
microsomes from lactating bovine: possible regulatory step in milk triglyceride synthe-
sis. Lipids, 7, 349–355.
Kraft J, Collomb M, Mockel P, Sieber R and Jahreis G (2003) Differences in CLA isomer
distribution of cow’s milk lipids. Lipids, 38, 657–664.
Kraft J, Hanske L, Mockel P, Zimmermann S, Hartl A, Kramer JKG and Jahreis G (2006) The
conversion efficiency of trans-11 and trans-12 18:1 by Δ9-desaturation differs in rats.
J. Nutr., 136, 1209–1214.
Kramer JKG, Parodi PW, Jensen RG, Mossoba MM, Yurawecz MP and Adlof RO (1998)
Rumenic acid: a proposed common name for the major conjugated linoleic acid isomer
found in natural products. Lipids, 33, 835.
Kritchevsky D, Tepper SA, Wright S, Czarnecki SK, Wilson TA and Nicolosi RJ (2004)
Conjugated linoleic acid isomer effects in atherosclerosis: growth and regression of
lesions. Lipids, 39, 611–616.
Kritchevsky D, Tepper SA, Wright S, Tso P and Czarnecki SK (2000) Influence of conjugated
linoleic acid (CLA) on establishment and progression of atherosclerosis in rabbits. J. Am.
Coll. Nutr., 19, 472S–477S.
Kuhnt K, Kraft J, Moeckel P and Jahreis G (2006) Trans-11-18:1 is effectively Δ9-
desaturated compared with trans-12-18:1 in humans. Br. J. Nutr., 95, 752–761.
Lacasse P, Kennelly JJ, Delbecchi L and Ahnadi CE (2002) Addition of protected and
unprotected fish oil to diets for dairy cows. I. Effects on the yield, composition and taste
of milk. J. Dairy Res., 69, 511–520.
Lavillonniere F, Chajes V, Martin JC, Sebedio JL, Lhuillery C and Bougnoux P (2003)
Dietary purified cis-9,trans-11 conjugated linoleic acid isomer has anticarcinogenic
properties in chemically induced mammary tumors in rats. Nutr. Cancer, 45, 190–194.
Lawless F, Stanton C, L’Escop P, Devery R, Dillon P and Murphy JJ (1999) Influence of
breed on bovine milk cis-9,trans-11-conjugated linoleic acid content. Livest. Prod. Sci.,
62, 43–49.
Ledoux M, Laloux L, Fontaine JJ, Carpentier YA, Chardigny JM and Sebedio JL (2007)
Rumenic acid significantly reduces plasma levels of LDL and small dense LDL
cholesterol in hamsters fed a cholesterol- and lipid-enriched semi-purified diet. Lipids,
42, 135–141.
Lee SH, Yamaguchi K, Kim JS, Eling TE, Safe S, Park Y and Baek SJ (2006) Conjugated
linoleic acid stimulates an anti-tumorigenic protein NAG-1 in an isomer specific manner.
Carcinogenesis, 27, 972–981.
Lengi AJ and Corl BA (2008) Comparison of pig, sheep and chicken SCD5 homologs:
evidence for an early gene duplication event. Comp. Biochem. Physiol. B. Biochem. Mol.
Biol., 150, 440–446.
Libby P (2002) Inflammation in atherosclerosis. Nature, 420, 868–874.
Lock AL and Bauman DE (2004) Modifying milk fat composition of dairy cows to enhance
fatty acids beneficial to human health. Lipids, 39, 1197–1206.
Lock AL, Bauman DE and Garnsworthy PC (2005a) Effect of production variables on the cis-
9,trans-11 conjugated linoleic acid content of cows’ milk. J. Dairy Sci., 88, 2714–2717.
Lock AL, Corl BA, Barbano DM, Bauman DE and Ip C (2004) The anticarcinogenic effect
of trans-11 18:1 is dependent on its conversion to cis-9,trans-11 CLA by Δ9-desaturase
in rats. J. Nutr., 134, 2698–2704.
BIOSYNTHESIS AND BIOLOGICAL ACTIVITY OF RUMENIC ACID 227
Lock AL and Garnsworthy PC (2002) Independent effects of dietary linoleic and linolenic
fatty acids on the conjugated linoleic acid content of cows’ milk. Anim. Sci., 74, 163–176.
Lock AL and Garnsworthy PC (2003) Seasonal variation in milk conjugated linoleic acid and
delta(9)-desaturase activity in dairy cows. Livest. Prod. Sci., 79, 47–59.
Lock AL, Horne CAM, Bauman DE and Salter AM (2005b) Butter naturally enriched in
conjugated linoleic acid and vaccenic acid alters tissue fatty acids and improves the
plasma lipoprotein profile in cholesterol-fed hamsters. J. Nutr., 135, 1934–1939.
Lusis AJ (2000) Atherosclerosis. Nature, 407, 233–241.
Lynch JM, Lock AL, Dwyer DA, Norbaksh R, Barbano DM and Bauman DE (2005) Flavor
and stability of pasteurized milk with elevated levels of conjugated linoleic acid and
vaccenic acid. J. Dairy Sci., 88, 489–498.
Mahfouz MM, Valicenti AJ and Holman RT (1980) Desaturation of isomeric trans-
octadecenoic acids by rat liver microsomes. Biochim. Biophys. Acta, 618, 1–12.
McCann SE, Ip C, Ip MM, McGuire MK, Muti P, Edge SB, Trevisan M and Freudenheim JL
(2004) Dietary intake of conjugated linoleic acids and risk of premenopausal and
postmenopausal breast cancer, Western New York Exposures and Breast Cancer Study
(WEB study). Cancer Epidemiol. Biomarkers Prev., 13, 1480–1484.
Mensink RP, Zock PL, Kester ADM and Katan MB (2003) Effects of dietary fatty acids and
carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and
apolipoproteins: a meta-analysis of 60 controlled trials. Am. J.Clin. Nutr., 77, 1146–1155.
Miller A, McGrath E, Stanton C and Devery R (2003a) Vaccenic acid (t11-18:1) is converted
to c9,t11-CLA in MCF-7 and SW480 cancer cells. Lipids, 38, 623–632.
Miller A, Stanton C, Murphy J and Devery R (2003b) Conjugated linoleic acid (CLA)-
enriched milk fat inhibits growth and modulates CLA-responsive biomarkers in MCF-7
and SW480 human cancer cell lines. Br. J. Nutr., 90, 877–885.
Miller A, Stanton C and Devery R (2002) Cis 9,trans 11- and trans 10,cis 12-conjugated
linoleic acid isomers induce apoptosis in cultured SW480 cells. Anticancer Res., 22,
3879–3887.
Mitchell PL, Langille MA, Currie DL and McLeod RS (2005) Effect of conjugated linoleic
acid isomers on lipoproteins and atherosclerosis in the Syrian Golden hamster. Biochim.
Biophys. Acta, 1734, 269–276.
Moore T (1939) Spectroscopic changes in fatty acids VI: General. Biochem. J., 33, 1635–
1638.
Mosley EE, Shafii B, Moate PJ and McGuire MA (2006) Cis-9,trans-11 conjugated linoleic
acid is synthesized directly from vaccenic acid in lactating dairy cattle. J. Nutr., 136, 570–
575.
National Research Council (1996) Carcinogens and anticarcinogens in the human diet.
National Academy Press, Washington, DC, USA.
Nestel P, Fujii A and Allen T (2006) The cis-9,trans-11 isomer of conjugated linoleic acid
(CLA) lowers plasma triglyceride and raises HDL cholesterol concentrations but does not
suppress aortic atherosclerosis in diabetic apoE-deficient mice. Atherosclerosis, 189,
282–287.
Noone EJ, Roche HM, Nugent AP and Gibney MJ (2002) The effect of dietary supplemen-
tation using isomeric blends of conjugated linoleic acid on lipid metabolism in healthy
human subjects. Br. J. Nutr., 88, 243–251.
Ntambi JM (1999) Regulation of stearoyl-CoA desaturase by polyunsaturated fatty acids and
cholesterol. J. Lipid Res., 40, 1549–1558.
Ntambi JM and Miyazaki M (2004) Regulation of stearoyl-CoA desaturases and role in
metabolism. Prog. Lipid Res., 43, 91–104.
Nugent AP, Roche HM, Noone EJ, Long A, Kelleher DK and Gibney MJ (2005) The effects
of conjugated linoleic acid supplementation on immune function in healthy volunteers.
Eur. J. Clin. Nutr., 59, 742–750.
228 TRANS FATTY ACIDS IN HUMAN NUTRITION
Ochoa JJ, Farquharson AJ, Grant I, Moffat LE, Heys SD and Wahle KW (2004) Conjugated
linoleic acids (CLAs) decrease prostate cancer cell proliferation: different molecular
mechanisms for cis-9,trans-11 and trans-10,cis-12 isomers. Carcinogenesis, 25, 1185–
1191.
O’Shea M, Devery R, Lawless F, Murphy J and Stanton C (2000) Milk fat conjugated linoleic
acid (CLA) inhibits growth of human mammary MCF-7 cancer cells. Anticancer. Res.,
20, 3591–3601.
Offer NW, Marsden M, Dixon J, Speake BK and Thacker FE (1999) Effect of dietary fat
supplements on levels of n–3 poly-unsaturated fatty acids, trans acids and conjugated
linoleic acid in bovine milk. Anim. Sci., 69, 613–625.
Ozols J (1997) Degradation of hepatic stearyl CoA delta9-desaturase. Mol. Biol. Cell, 8,
2281–2290.
Palmquist DL, Lock AL, Shingfield KJ and Bauman DE (2005) Biosynthesis of conjugated
linoleic acid in ruminants and humans. In: Advances in Food and Nutrition Research (SL
Taylor, ed.) Elsevier Inc., San Diego, California, USA, pp.179–217.
Palombo JD, Ganguly A, Bistrian BR and Menard MP (2002) The antiproliferative effects
of biologically active isomers of conjugated linoleic acid on human colorectal and
prostatic cancer cells. Cancer Lett., 177, 163–172.
Pariza MW, Ashoor SH, Chu FS and Lund DB (1979) Effects of temperature and time on
mutagen formation in pan-fried hamburger. Cancer Lett., 7, 63–69.
Pariza MW (1999) The biological activities of conjugated linoleic acid. In: Advances in Con-
jugated Linoleic Acid Research, Volume 1 (MP Yurawecz, MM Mossoba, JKG Kramer,
MW Pariza and G Nelson, eds) AOCS Press, Champaign, Illinois, USA, pp.12–20.
Park HS, Chun CS, Kim S, Ha YL and Park JH (2006) 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-dimethylhydrazine. J. Med. Food, 9, 22–27.
Parodi PW (1977) Conjugated octadecienoic acids of milk fat. J. Dairy Sci., 60, 1550–1553.
Parodi PW (1994) Conjugated linoleic acid — an anticarcinogenic fatty acid present in milk
fat. Aust. J. Dairy Tech., 49, 93–97.
Parodi PW (2003) Conjugated linoleic acid in food. In: Advances in Conjugated Linoleic Acid
Research, Volume 2 (JL Sebedio, WW Christie and RO Adlof, eds) AOCS Press,
Champaign, Ilinois, USA, pp.101–122.
Parodi PW (2004) Milk fat in human nutrition. Aust. J. Dairy Tech., 59, 3–59.
Parodi PW (2006) Nutritional significance of milk lipids. In: Advanced Dairy Chemistry,
Volume 2: Lipids, 3rd Edition (PF Fox and PLH McSweeney, eds) Kluwer Academic/
Plenum Publishers, New York, USA, pp.601–639.
Peterson DG, Kelsey JA and Bauman DE (2002) Analysis of variation in cis-9,trans-11
conjugated linoleic acid (CLA) in milk fat of dairy cows. J. Dairy Sci., 85, 2164–2172.
Piperova LS, Sampugna J, Teter BB, Kalscheur KF, Yurawecz MP, Ku Y, Morehouse KM
and Erdman RA (2002) Duodenal and milk trans octadecenoic acid and conjugated
linoleic acid (CLA) isomers indicate that postabsorptive synthesis is the predominant
source of cis-9-containing CLA in lactating dairy cows. J. Nutr., 132, 1235–1241.
Pollard MR, Gunstone FD, James AT and Morris LJ (1980) Desaturation of positional and
geometric isomers of monoenoic fatty acids by microsomal preparations from rat liver.
Lipids, 15, 306–314.
Rajakangas J, Basu S, Salminen I and Mutanen M (2003) Adenoma growth stimulation by
the trans-10,cis-12 isomer of conjugated linoleic acid (CLA) is associated with changes
in mucosal NF-kappaB and cyclin D1 protein levels in the Min mouse. J. Nutr., 133,
1943–1948.
Ramaswamy N, Baer RJ, Schingoethe DJ, Hippen AR, Kasperson KM and Whitlock LA
(2001a) Composition and flavor of milk and butter from cows fed fish oil, extruded
soybeans, or their combination. J. Dairy Sci., 84, 2144–2151.
BIOSYNTHESIS AND BIOLOGICAL ACTIVITY OF RUMENIC ACID 229
Ramaswamy N, Baer RJ, Schingoethe DJ, Hippen AR, Kasperson KM and Whitlock LA
(2001b) Short communication. Consumer evaluation of milk high in conjugated linoleic
acid. J. Dairy Sci., 84, 1607–1609.
Ritzenthaler KL, McGuire MK, Falen R, Shultz TD, Dasgupta N and McGuire MA (2001)
Estimation of conjugated linoleic acid intake by written dietary assessment methodolo-
gies underestimates actual intake evaluated by food duplicate methodology. J. Nutr., 131,
1548–1554.
Roy A, Chardigny JM, Bauchart D, Ferlay A, Lorenz S, Durand D, Gruffat D, Faulconnier
Y, Sebedio JL and Chilliard Y (2007) Butters rich either in trans-10-C18:1 or in trans-
11-C18:1 plus cis-9,trans-11 CLA differentially affect plasma lipids and aortic fatty
streak in experimental atherosclerosis in rabbits. Animal, 1, 467–476.
Santora JE, Palmquist DL and Roehrig KL (2000) Trans-vaccenic acid is desaturated to
conjugated linoleic acid in mice. J. Nutr., 130, 208–215.
Scimeca JA (1999) Cancer inhibition in animals. In: Advances in Conjugated Linoleic Acid
Research, Volume I (MP Yurawecz, MM Mossoba, JKG Kramer, MW Pariza and G
Nelson, eds) AOCS Press, Champaign, Illinois, USA, pp.420–443.
Shantha NC, Ram LN, O’Leary J, Hicks CL and Decker EA (1995) Conjugated linoleic acid
concentrations in dairy products as affected by processing and storage. J. Food Sci., 60,
695–697.
Shingfield KJ, Ahvenjarvi S, Toivonen V, Arola A, Nurmela KVV, Huhtanen P and Griinari
JM (2003) Effect of dietary fish oil on biohydrogenation of fatty acids and milk fatty acid
content in cows. Anim. Sci., 77, 165–179.
Shingfield KJ, Ahvenjarvi S, Toivonen V, Vanhatalo A and Huhtanen P (2007) Transfer of
absorbed cis-9,trans-11 conjugated linoleic acid into milk is biologically more efficient
than endogenous synthesis from absorbed vaccenic acid in lactating cows. J. Nutr., 137,
1154–1160.
Shingfield KJ and Griinari JM (2007) Role of biohydrogenation intermediates in milk fat
depression. Eur. J. Lipid Sci. Tech., 109, 799–816.
Stanton C, Lawless F, Kjellmer G, Harrington D, Devery R, Connolly JF and Murphy J (1997)
Dietary influences on bovine milk cis-9,trans-11-conjugated linoleic acid content. J.
Food Sci., 62, 1083–1086.
Stanton C, Murphy J, McGrath E and Devery R (2003) Animal feeding strategies for
conjugated linoleic acid enrichment of milk. In: Advances in Conjugated Linoleic Acid
Research, Volume 2 (JL Sebedio, WW Christie and RO Adlof, eds) AOCS Press,
Champaign, Illinois, USA, pp.123–145.
Steck SE, Chalecki AM, Miller P, Conway J, Austin GL, Hardin JW, Albright CD and
Thuillier P (2007) Conjugated linoleic acid supplementation for twelve weeks increases
lean body mass in obese humans. J. Nutr., 137, 1188–1193.
Tanmahasamut P, Liu J, Hendry LB and Sidell N (2004) Conjugated linoleic acid blocks
estrogen signaling in human breast cancer cells. J. Nutr., 134, 674–680.
Tricon S, Burdge GC, Jones EL, Russell JJ, El-Khazen S, Moretti E, Hall WL, Gerry AB,
Leake DS, Grimble RF, Williams CM, Calder PC and Yaqoob P (2006) Effects of dairy
products naturally enriched with cis-9,trans-11 conjugated linoleic acid on the blood lipid
profile in healthy middle-aged men. Am. J. Clin. Nutr., 83 744–753.
Tricon S, Burdge GC, Kew S, Banerjee T, Russell JJ, Jones EL, Grimble RF, Williams CM,
Yaqoob P and Calder PC (2004) Opposing effects of cis-9,trans-11 and trans-10,cis-12
conjugated linoleic acid on blood lipids in healthy humans. Am. J. Clin. Nutr., 80, 614–
620.
Turpeinen AM, Mutanen M, Aro A, Salminen I, Basu S, Palmquist DL and Griinari JM (2002)
Bioconversion of vaccenic acid to conjugated linoleic acid in humans. Am. J. Clin. Nutr.,
76, 504–510.
Valeille K, Férézou J, Amsler G, Quignard-Boulang‚ A, Parquet M, Gripois D, Dorovska-
230 TRANS FATTY ACIDS IN HUMAN NUTRITION
Taran V and Martin JC (2005) A cis-9,trans-11 conjugated linoleic acid-rich oil reduces
the outcome of atherogenic process in hyperlipidemic hamster. Am. J. Physiol. Heart
Circ. Physiol., 289, H652–H659.
Valeille K, Ferezou J, Parquet M, Amsler G, Gripois D, Quignard-Boulange A and Martin
JC (2006) The natural concentration of the conjugated linoleic acid, cis-9,trans-11 in milk
fat has antiatherogenic effects in hyperlipidemic hamsters. J. Nutr., 136, 1305–1310.
Voorrips LE, Brants HAM, Kardinaal AFM, Hiddink GJ, van den Brandt PA and Goldbohm
RA (2002) Intake of conjugated linoleic acid, fat, and other fatty acids in relation to
postmenopausal breast cancer: the Netherlands cohort study on diet and cancer. Am. J.
Clin. Nutr., 76, 873–882.
Ward RJ, Travers MT, Richards SE, Vernon RG, Salter AM, Buttery PJ and Barber MC
(1998) Stearoyl-CoA desaturase mRNA is transcribed from a single gene in the ovine
genome. Biochim. Biophys. Acta, 1391, 145–156.
Werner SA, Luedecke LO and Shultz TD (1992) Determination of conjugated linoleic acid
content and isomer distribution in three cheddar-type cheeses: effects of cheese cultures,
processing and aging. J. Agric. Food Chem., 40, 1817–1821.
Whitlock LA, Schingoethe DJ, Hippen AR, Kalscheur KF, Baer RJ, Ramaswamy N and
Kasperson KM (2002) Fish oil and extruded soybeans fed in combination increase
conjugated linoleic acids in milk of dairy cows more than when fed separately. J. Dairy
Sci., 85, 234–243.
Wilson TA, Nicolosi RJ, Chrysam M and Kritchevsky D (2000) Conjugated linoleic acid
reduces early aortic atherosclerosis greater than linoleic acid in hypercholesterolemic
hamsters. Nutr. Res., 20, 1795–1805.
Yamasaki M, Chujo H, Koga Y, Oishi A, Rikimaru T, Shimada M, Sugimachi K, Tachibana
H and Yamada K (2002) Potent cytotoxic effect of the trans10,cis12 isomer of conjugated
linoleic acid on rat hepatoma dRLh-84 cells. Cancer Lett., 188, 171–180.
Yurawecz MP, Roach JAG, Sehat N, Mossoba MM, Kramer JKG, Fritsche J, Steinhart H and
Ku Y (1998) A new conjugated linoleic acid isomer, 7 trans,9 cis-octadecadienoic acid,
in cow milk, cheese, beef and human milk and adipose tissue. Lipids, 33, 803–809.
CHAPTER 9
Biosynthesis, synthesis and biological activity
of trans-10,cis-12 conjugated linoleic acid
(CLA) isomer
1
Nestlé Purina PetCare, St Louis, Missouri, USA
2
Nestlé Research Center, Lausanne, Switzerland
A. Introduction
Conjugated linoleic acids (CLA) consist of several positional and geometric
isomers. Although there are 28 known isomers, CLA is not one molecule, but
rather is a mixture of primarily cis-9,trans-11 18:2 and trans-10,cis-12 18:2
CLA isomers. In fact it is these two isomers that are responsible for all known
biological effects of CLA. The cis-9,trans-11 isomer is the most abundant form
found in food, the trans-10,cis-12 isomer being the second most abundant form
(Fritsche, 1999). Until recently, investigations of the effects of CLA on bio-
logical processes used a mixture of these two isomers and thus it was impossible
to ascertain which of the primary isomers was eliciting the biological effect.
Numerous biological effects and health benefits have been attributed to CLA
including weight loss, improved body composition, enhanced immune func-
tion, anti-carcinogenic and anti-atherosclerotic effects, improved glucose control
and insulin sensitivity, and others. There are several outstanding reviews of the
literature reviewing the health benefits of CLA (Bhattacharya et al. 2006;
Wahle et al., 2004; Wang & Jones, 2004; Pariza et al.; and Sébédio, Christie &
Adlof, 2003). However, nearly all reviews of CLA include reports in which
both isomers were fed or tested in vitro. Recent advances in technology have
led to the availability of the purified isomers and thus a great deal of studies
have been designed and conducted investigating the biological effects of the
individual isomers. However, until recently scientists were unable to assess
which isomer was eliciting these reported biological effects. There is a growing
body of literature indicating that the trans-10,cis-12 isomer may be primarily
responsible for several of these observed effects. A complete review of CLA
and its biological effects are beyond the scope of this chapter. Rather this
231
232 TRANS FATTY ACIDS IN HUMAN NUTRITION
chapter will focus on the biosynthesis, synthesis and biological effects of the
trans-10,cis-12 CLA isomer and focus on the newer and more relevant findings
with respect to the biological activity of the trans-10,cis-12 CLA isomer.
B. Biosynthesis
1. Biosynthesis in mammals
The investigations on CLA isomers showed that two main isomers, cis-9,trans-
11 and trans-10,cis-12 isomers occur naturally in foods derived from ruminants
(Chin et al., 1992; Kraft et al., 2003; Kramer et al., 1998; Kramer et al., 2004;
Parodi, 1977; Parodi, 1999). Two routes for CLA synthesis in ruminants were
described and occur in two different parts of the ruminant animal (Bauman
et al., 1999; Griinari & Bauman, 1999). The first route is a ruminal biohydro-
genation of dietary linoleic acid performed by the ruminal bacterial population.
The second route, occurring in the animal’s tissue, is the desaturation of
vaccenic acid by endogenous Δ9-desaturase (Griinari & Bauman, 1999; Mosley
et al., 2006b), which is also active in producing cis-9,trans-11 isomer in
mammary glands of other mammals such as human and rat (Lock et al., 2004;
Mosley et al., 2006a).
In humans, no conversion of trans-10 18:1 into trans-10,cis-12 18:2 CLA by
the endogenous Δ12-desaturase was detected (Adlof et al., 2000). The predomi-
nant isomer in ruminant and milk fat is cis-9,trans-11 18:2 representing 80 to
90% of the total milk CLA (Chin, et al., 1992; Parodi, 1977) and more than
90% of total CLA in the subcutaneous and intra-muscular fat of steers (Fritsche
& Fritsche, 1998). However, the content of trans-10,cis-12 CLA isomer can be
increased in lean beef or milk fat by specific dietary conditions (Griinari &
Bauman, 1999; Peterson et al., 2003, Xu et al., 2006). Some studies have
demonstrated that the rumen fermentation pathway was altered by feeding a
low-fibre diet resulting in an increase in trans-10 octadecenoic acid content in
the milk fat. Griinari and Bauman (1999) proposed that this new route,
alternative to the biohydrogenation process and induced by the diet, would
involve a specific bacterial cis-9,trans-10 isomerase that would form a trans-
10,cis-12 conjugated double bond structure as a first reaction.
Sieber and co-workers (2004) showed that the trans-10,cis-12 CLA isomer
is biosynthesized by some bacteria in the rumen only from linoleic acid.
Furthermore, protecting unsaturated fatty acids from bacterial hydrogenation
by thermal treatment of the diet increased the production of trans-10,cis-12
isomer in lean beef and the level of this CLA isomer in the intra-muscular fat
(Xu et al., 2006). In addition, in vitro studies showed that acidic pH conditions
(pH from 6 to 7) as well as aerobic conditions influence drastically the
production of trans-10,cis-12 CLA isomer by ruminal bacteria (Choi et al.,
2007; Kawahara et al., 2007).
BIOSYNTHESIS, SYNTHESIS AND BIOACTIVITY OF TRANS-10,CIS-12 CLA 233
* Adapted from Alonso et al., 2003; Ando et al., 2003; Ando et al., 2004; Coakley et al., 2003; Jiang et
al., 1998; Kim et al., 2002; Kim, 2003; Kishino et al. 2002; Lee et al., 2006; Lee et al., 2003; Ogawa et
al., 2001; Ogawa et al., 2005 and Sieber et al., 2004. Results obtained using linoleic acid, ricinoleic acid
or castor oil as substrates.
234 TRANS FATTY ACIDS IN HUMAN NUTRITION
the active isomers is required. The purification of these isomers was performed
successfully using an enzymatic method with lipases from either Candida
rugosa or Geotrichum candidum (McNeill et al., 1999; Nagao et al., 2002,
2003). The process comprises lipase-catalysed selective esterification with
lauryl alcohol, molecular distillation, and urea adduct fractionation under
strictly controlled conditions in ethanol. As a result, purification of 1.0 kg of
the CLA mixture yielded pure cis-9,trans-11 (93.1%) and trans-10,cis-12
(95.3%) CLA isomers.
Recently, a novel and efficient method by crystallization in acetone of the
two CLA isomers in the presence of medium-chain fatty acid additives was
reported (Uehara et al., 2006). Crystals containing mainly trans-10,cis-12
isomer were obtained by addition of lauric and decanoic acids, whereas
octanoic acid yielded crystals containing predominantly cis-9,trans-11 isomer
(Uehara et al., 2006).
isomer that protects against cancer and has anti-cancer properties. However,
this is not conclusive and in some cases either isomer is effective or, as in one
report, long-term supplementation of the trans-10,cis-12 isomer actually in-
creased tumour formation. The mechanisms by which the trans-10,cis-12
isomer elicits these anti-cancer effects include inhibition of cellular prolifera-
tion, increased apoptosis, and changes in COX-2 expression as well as other
proposed mechanisms. However, given the recent report by Ip and co-workers
(2007), more coordinated long-term research is needed to ensure the safety and
efficacy of the trans-10,cis-12 isomer as an agent that protects against cancer.
E. Conclusion
Conjugated linoleic acid (CLA) is the common name for a group of positional
and geometric isomers of linoleic acid. The two primary isomers are the cis-9,
trans-11 and trans-10,cis-12 linoleic acid. The earliest health observations of
CLA go back to 1992 (Chin et al., 1992). Since those early observations
considerable research has been conducted on the biological and health effects
of CLA. The reported health benefits attributed to CLA consumption include:
anti-obesity, anti-atherosclerotic/anti-atherogenic, anti-cancer, and enhanced
immune function. As CLA preparations are most commonly a mixture of the
two isomers, the question remains as to which isomer elicited which biological
effect? And by what mechanisms are these effects elicited? Techniques have
been developed for the synthesis and purification of the two isomers allowing
for studies addressing the question of which isomer elicited the observed
biological effects. There is a vast amount of literature using in vitro cell culture
systems and in non-human animal models supporting the efficacy of CLA
supplementation for a variety of health benefits. However, conclusive evidence
of the specific benefits of CLA supplementation for humans is lacking. With
respect to which biological effect is attributable to which isomer, recent studies
have clarified to a great degree which isomer is responsible for which biologi-
cal effect as well as elucidating the mechanisms by which these biological
effects are achieved.
In this regard, it is clear that supplementation with the trans-10,cis-12 isomer
BIOSYNTHESIS, SYNTHESIS AND BIOACTIVITY OF TRANS-10,CIS-12 CLA 247
profoundly decreases fat mass and improves body composition in rodents. The
effect of CLA supplementation on reducing fat mass and improving body
composition is less clear in humans as clinical trials have failed to demonstrate
uniform results. Why are the effects observed in rodents conclusive yet not in
humans? As indicated previously, the reasons may range from experimental
design, dose, and duration of the study, age of the study population, gender, or
genetic predisposition. Furthermore, some reports have indicated that the
trans-10,cis-12 isomer may contribute to insulin resistance and hyper-
insulinaemia.
Thus, future studies are necessary to confirm these observations and en-
sure that long term supplementation with the trans-10,cis-12 isomer is safe.
It should be noted that Pariza et al. (2004) suggest that the lack of uniformity
amongst these observations is due to experimental conditions. CLA supple-
mentation modulates the immune response and indeed each isomer appears
to have differential effects, yet as with weight loss the literature is inconclu-
sive with respect to modulation of immune function in humans. Thus, further
research needs to be conducted to confirm both the effect of CLA supple-
mentation as well as which isomer elicits which effect. For example, in
rodents supplementation with the trans-10,cis-12 results in greater IgA and
IgM production compared to the cis-9,trans-11 CLA isomer. Another study
testing both the mixture of the isomers and the isomers individually observed
a decrease in proinflammatory eicosanoids (leukotriene B 4 and
prostaglandin E2). Thus, the potential exists for CLA, a mixture of the iso-
mers or alone as an anti-inflammatory application, to address other specific
needs with respect to immune function.
The trans-10,cis-12 isomer also appears to have potent anti-cancer activity
in vitro and in animal trials. To our knowledge, there are no clinical trials
demonstrating the ability of CLA to reduce tumorigenesis or tumour growth
in humans. And although the trans-10,cis-12 CLA isomer appears to have
potent anti-cancer activity in vitro and in vivo, some studies indicate that the
cis-9,trans-11 CLA isomer also has potent anti-cancer potential. Further
research in a more controlled manner needs to be conducted to verify the
effects and mechanisms by which the independent isomers elicit their bio-
logical effects. As suggested by others, such research could lead to the
development of therapeutic applications to address obesity, inflammatory
conditions, and improve the host response to infection, as well as other
potential applications.
Of course safety is of utmost importance. A recent report indicates that
during long term supplementation the trans-10,cis-12 isomer may actually be
deleterious. Thus, careful trials must also be conducted to confirm both the
safety and efficacy of supplementation with the trans-10,cis-12 isomer.
248 TRANS FATTY ACIDS IN HUMAN NUTRITION
References
Ackman, RG (1998) Laboratory preparation of conjugated linoleic acids. J. Am. Oil Chem.
Soc., 75, 1227–1469.
Adlof, RO, Duval, S and Emken, EA (2000) Biosynthesis of conjugated linoleic acid in
humans. Lipids, 35, 131–135.
Albers, R, van der Wielen, RPJ, Brink, EJ, Hendriks, HFJ, Dorovska-Taran, VN and Mohede,
ICM (2003) Effects of cis-9, trans-11 and trans-10, cis-12 conjugated linoleic acid (CLA)
on immune function in healthy men. Eur. J. Clin. Nutr., 57, 595–603.
Alonso, L, Cuesta, EP and Gilliland, SE (2003) Production of free conjugated linoleic acid
by Lactobacillus acidophilus and Lactobacillus casei of human intestinal origin. J. Dairy
Sci., 86, 1941–1946.
Ando, A, Ogawa, J, Kishino, S and Shimizu, S (2003) CLA production from ricinoleic acid
by lactic acid bacteria, J. Am. Oil Chem. Soc., 80, 889–894.
Ando, A, Ogawa, J, Kishino, S and Shimizu, S (2004) Conjugated linoleic acid production
from castor oil by Lactobacillus plantarum JCM 1551. Enzyme and Microbial Technol.,
35, 40–45.
Azain, MJ, Hausamann, DB, Sisk, MB, Flatt, WP and Jewell, DE (2000) Dietary conjugated
linoleic acid reduces rat adipose tissue size rather than cell number. J. Nutr., 130, 1548–
1554.
Bauman, DE, Baumgard, LH and Griinari, JM (1999) Biosynthesis of conjugated linoleic
acid in ruminants, Proceedings of the American Society of Animal Science. ASAS
Annual meeting, Conjugated Linoleic Acid: Effects on Metabolism and Health
Minisymposium (Indianapolis, Indiana, USA) pp.1–15 (www.asas.org/jas/symposia/
proceedings/0937.pdf)
Battacharya, A, Banu, J, Rahman, M, Causey, J and Fernandes, G (2006) Biological effects
of conjugated linoleic acids in health and disease. J. Nutr. Biochem., 17, 789–810.
Bissonauth, V, Chouinard,Y, Marin, J, Leblanc, N, Richard, D and Jacques, H (2006) The
effects of t10,c12 CLA isomer compared with c9,t11 CLA isomer on lipid metabolism
and body composition in hamsters. J. Nutr. Biochem., 17, 597–603.
Blankson, H, Stakkestad, JY, Fagertun, H, Thorn,E, Wadstein, J and Gudmundsen, O (2000)
Conjugated linoleic acid reduces body fat mass in overweight and obese humans. J. Nutr.,
130, 2943–2948.
Chin, SF, Liu, W, Storkson, J, Ha, YL and Pariza, M (1992) Dietary sources of conjugated
dienoic isomers of linoleic acid, a newly recognized class of anticarcinogens. J. Food
Comp.Anal., 5, 185–197.
Chipault, JR and Hawkins, JM (1959) The determination of conjugated cis-trans and trans-
trans methyl octadecadienoates by infrared spectrometry, J. Am. Oil Chem. Soc., 36,
535–539.
Cho, HJ, Kim, WK, Jung, JI, Kim, EJ, Lim, SS, Kwon, DY and Park, JH (2005) Trans-10,12-
cis, not cis-9,trans-11, conjugated linoleic acid decreases ErbB3 expression in HT-29
human cancer cells. World J. Gasteroenterol., 11, 5142–5150.
Cho, HJ, Lee, HS, Chung, CK, Kang, YH, Ha, YL, Park, HS and Park, JH (2003) Trans-
10,cis-12 conjugated linoleic acid reduces insulin-like growth factor-II in HT-29 human
cancer cells. J. Med. Food, 6, 193–199.
Choi, Y, Kim, YC, Han, YB, Park, Y, Pariza, MW and Ntambi, JM (2000) The trans-10,cis-
12 isomer of conjugated linoleic acid downregulates stearoyl-CoA desaturase 1 gene
expression in 3T3-L1 adipocytes. J. Nutr, 130, 1920–1924.
Choi, JS, Jung, MH, Park, HS and Song, J (2004) Effect of conjugated linoleic acid isomers
on insulin resistance and mRNA levels of genes regulating energy metabolism in high-
fat-fed rats. Nutrition, 20, 1008–10017.
Choi, N-J, Imm, JY, Oh, S, Kim, B-C, Hwang, H-J and Kim, YJ (2007) Effect of pH and
BIOSYNTHESIS, SYNTHESIS AND BIOACTIVITY OF TRANS-10,CIS-12 CLA 249
oxygen on conjugated linoleic acid production by mixed rumen bacteria from cows fed
high concentrate and high forage diets, Animal Feed Science and Technology, 123–124,
643–653.
Christie, WW, Dobson, G and Gunstone, FD (1997) Isomers in commercial samples of
conjugated linoleic acid. Lipids, 32, p.1231.
Clement, L, Poirier, H, Niot, I, Bocher, V, Guerre-Millo, M, Krier, S, Staels, B and Besnard,
P (2002) Dietary trans-10,cis-12 conjugated linoleic acid induces hyperinsulinemia and
fatty liver in the mouse. J. Lipid Research, 43, 1400–1409.
Coakley, M, Ross, RP, Nordgren, M, Fitzgerald, G, Devery, R and Stanton, C (2003)
Conjugated linoleic acid biosynthesis by human-derived Bifidobacterium species. J. Appl.
Microbiol., 94, 138–145.
Degner, SC, Kemp, MQ, Bowden, T and Romangnolo, DF (2006) Conjugated linoleic acid
attenuates cyclooxygenase-2 transcriptional activity via an anti-Ap-1 mechanism in
MCF-7 breast cancer cells. J. Nutr, 136, 421–427.
DeLaney, JP and West, DB (2000) Changes in body composition wit conjugated linoleic acid.
J. Am. Coll. Nutr., 19, 487S–493S.
Devillard, E, McIntosh, FM, Duncan, SH and Wallace, RJ (2007) Metabolism of linoleic acid
by human gut bacteria: different routes for biosynthesis of conjugated linoleic acid.
J. Bacteriol., 189, 2566–2570.
Eder, K, Schlesser, S, Becker, K and Korting, R (2003) Conjugated linoleic acids lower the
release of eicosanoids and nitric oxide from human aortic endothelial cells. J. Nutr., 133,
4083–4089.
Evans, M, Geigermann, C, Cook, J, Curits, L, Kuebler, B and MacIntosh, M (2000)
conjugated linoleic acid suppresses triglyceride accumulation and induces apoptosis in
3T3-L1 preadipocytes. Lipids, 35, 134–138.
Fite, A, Goua, M, Wahle, KWJ, Schofield, AC, Hutcheson, AW and Heys, SD (2007)
Potentiation of the anti-tumor effect of docetexel by conjugated linoleic acids (CLAs) in
breast cancer cells in vitro. Pros. Leuko. Essent. Fatty Acids, 77, 87–96.
Fritsche, S and Fritsche, J (1998), Occurrence of conjugated linoleic acid isomers in beef.
J. Am. Oil Chem. Soc., 75, 1449–1451.
Fritsche, JRR and Steinhard, H (1999) Formation, contents and estimation of daily intake of
conjugated linoleic acid isomers and trans-fatty acids in food. Advances in Comjugated
Linoleic Acid Research, 1, 378–396.
Fukuda, S, Furuya, H, Suzuki, Y, Asanuma, N and Hino, T (2005) A new strain of
Butyrivibrio fibrisolvens that has high ability to isomerize linoleic acid to conjugated
linoleic acid. J. Gen. Appl. Microbiol., 51, 105–113.
Gavino, VC, Gavio, G, Leblanc, M-J and Tuchweber, B (2000) An isomeric mixture of
conjugated linoleic acid but not pure cis-9,trans-11-octadecadienoic acid affects body
weight gain and plasma lipids in hamsters. J. Nutr., 130, 27–29.
Griinari, JM and Bauman, DE (1999) Biosynthesis of conjugated linoleic acid and its
incorporation into meat and milk in ruminants. In: Advances in Conjugated Linoleic Acid
Research, First edition (Yurawecz, MP, Mossaba, MM, Kramer, JKG, Pariza, MW,
Nelson, GJ, eds) AOCS Press, Champaign, Illinois, USA, pp.180–200.
Ha, YI, Grimm, NK and Pariza, MW (1987) Anticarcinogens from fried ground beef: heat-
altered derivatives of linoleic acid. Carcinogenesis, 8, 1881–1887.
Ha, YI, Storkson, J and Pariza, MW (1990) Inhibition of benzo(a)pyrene-induced mouse
forestomach neoplasia by conjugated dienoic derivatives of linoleic acid. Cancer Res.,
15, 1097–1101.
Houseknecht, KL, Vanden, Heuvel, JP, Moya-Camarena, SY (1998) Dietary conjugated
linoleic acid normalizes impaired glucose tolerance in Zucker diabetic fatty fa/fa rat.
Biochem. Biophys. Res. Commun., 244, 678–682.
Hornung, E, Krueger, C, Pernstich, C, Gipmans, M, Porzel, A and Feussner, I (2005)
250 TRANS FATTY ACIDS IN HUMAN NUTRITION
Kramer, JKG, Parodi, PW, Jensen, RG, Mossoba, MM, Yurawecz, MP and Adlof, RO (1998)
Rumenic acid: a proposed common name for the major conjugated linoleic acid isomer
found in natural products. Lipids, 33, 835.
LaRosa, CP, Miner, J, Xia, Y, Zhou, Y, Kachman, S and Fromm, ME (2006) Trans-10,cis-
12 conjugated linoleic acid causes inflammation and delipidation of white adipose tissue
in mice: a microarray and histologic analysis. Physiol. Genomics, 27, 282–294.
LaRosa, PC, Riethoven, JJ, Chem H, Xia, Y, Chen, M, Miner, JL and Fromm, ME (2007)
Trans-10,cis-12 conjugated linoleic acid activates the integrated stress response pathway
in adipocytes. Physiol. Genomics, 31, 544–553.
Lee, HY, Park, JH, Seok, SH, Baek, MW, Kim, DJ, Lee, KE, Paek, KS, Lee, Y and Park, JH
(2006) Human originated bacteria, Lactobacillus rhamnosus PL60, produce conjugated
linoleic acid and show anti-obesity effects in diet-induced obese mice, Biochim. Biophys.
Acta, 1761, 736–744.
Lee, SO, Hong, GW and Oh, DK (2003) Bioconversion of linoleic acid into conjugated
linoleic acid by immobilized Lactobacillus reuteri. Biotechnology Progress, 19, 1081–
1084.
Lee, SO, Kim, CS, Cho, SK, Choi, HJ, Ji, GE and Oh, DK (2003) Bioconversion of linoleic
acid into conjugated linoleic acid during fermentation and by washed cells of Lactoba-
cillus reuteri. Biotechnology Letters, 25, 935–938.
Li, G, Barnes, D, Butz, D, Bjorling, D and Cook, ME (2005) 10t,12c-conjugated linoleic acid
inhibits lipopolysaccharide-induced cyclooxygenase expression in vitro and in vivo.
J. Lipid Res., 46, 2134–2142.
Liavonchanka, A, Hornung, E, Feussner, I and Rudolph, MG (2006) Structure and mecha-
nism of the Propionibacterium acnes polyunsaturated fatty acid isomerase. Proc. Natl.
Acad. Sci. USA, 103, pp.2576–2581.
Lim-do, Y, Tyner, AL, Park, JB, Lee, Jy, Choi, YH and Park, JH (2005) Inhibition of colon
cancer cell proliferation by the dietary compound conjugated linoleic acid is mediated by
the CDK inhibitor p21CIP1/WAF1. J. Cell Physiol., 205, 107–113.
Lock, AL, Corl, BA, Barbano, DM, Bauman, DE and Ip, C (2004) The anticarcinogenic effect
of trans-11 18:1 is dependent on its conversion to cis-9, trans-11 CLA by delta9-
desaturase in rats. J. Nutr., 134, 2698–2704.
Malpuech-Brugére, C, Verboeket-van de Venne, W PHG, Mensink, RP, Arnal, M-A, Morio,
B, Brandolini, M, Saebo, A, Lassel, TS, Chardigny, JM, Sébédio, JL and Beaufrere
(2003) Effects of two conjugated linoleic acid isomers on body fat mass in overweight
humans. u Obes. Res., 12, 591–598.
Martin, JC, Gregoire,S and Siess, MH (2000) Effects of conjugated linoleic acid isomers on
lipid-metabolizing enzymes in male rats. Lipids, 35, 91–98.
McNeill, GP, Rawlins, C and Peilow, AC (1999) Enzymatic enrichment of conjugated
linoleic acid isomers and incorporation into triglycerides. J. Am. Oil Chem. Soc., 76,
1265–1268.
Miller, A, Snanton, C and Devenery, R (2002) Cis 9, trans 11- and trans 10, cis 12-conjugated
linoleic acid isomers induce apoptosis in cultured SW480 cells. Anticancer Res., 22,
3879–3887.
Moon, HS, Lee, HG, Seo, JH, Chung, CS, Guo, DD, Kim, TG, Choi, YJ and Cho, CS (2007)
Leptin-induced matrix matelloproteinase-2 secretion is suppressed by trans-10,cis-12
conjugated linoleic acid. Biochem. Biophys. Res. Commun., 356, 955–960.
Mosley, EE, McGuire, MK, Williams, JE and McGuire, MA (2006a) Cis-9,trans-11
conjugated linoleic acid is synthesized from vaccenic acid in lactating women. J. Nutr.,
136, 2297–2301.
Mosley, EE, Shafii, DB, Moate, PJ and McGuire, MA (2006b) Cis-9,trans-11 conjugated
linoleic acid is synthesized directly from vaccenic acid in lactating dairy cattle. J. Nutr.,
136, 570–575.
252 TRANS FATTY ACIDS IN HUMAN NUTRITION
Murphy, EF, Hooivled, GJ, Muller, M, Calogero, RA and Cashman, KD (2007) Conjugated
linoleic acid alters global gene expression in human intestinal-like Caco-2 cells in an
isomer-specific manner. J. Nutr, 137, 2359–2365.
Nagao, T, Shimada, Y, Yamauchi-Sato, Y, Yamamoto, T, Kasai, M, Tsutsumi, K, Sugihara,
A and Tominaga, Y (2002) Fractionation and enrichment of CLA isomers by selective
esterification with Candida rugosa lipase. J. Am. Oil Chem. Soc., 79, 303–308.
Nagao, T, Yamauchi-Sato, Y, Sugihara, A, Iwata, T, Nagao, K, Yanagita, T, Adachi, S and
Shimada, Y (2003) Purification of conjugated linoleic acid isomers through a process
including lipase-catalyzed selective esterification. Biosci. Biotechnol. Biochem., 67,
1429–1433.
Nichols, PL, Herb, SF and Riemenschneider, RW (1951) Isomers of conjugated fatty acids.
I. Alkaline-isomerized linoleic acid. J. Chem. Soc., 73, 247–252.
Ogawa, J, Kishino, S ando, A, Sugimoto, S, Mihara, K and Shimizu, S (2005) Production of
conjugated fatty acids by lactic acid bacteria, J. Biosci. Bioeng., 100, 355–364.
Ogawa, J, Matsumura, K, Kishino, S, Omura, Y and Shimizu, S (2001) Conjugated linoleic
acid accumulation via 10-hydroxy-12 octadecaenoic acid during microaerobic transfor-
mation of linoleic acid by Lactobacillus acidophilus. Appl. Environ. Microbiol., 67,
1246–1252.
Ogawa, J, Sugimoto, S, Ito, T, Kishino, S, Ando, A and Shirasaka, N (2004) Screening and
analysis of microbial reactions useful for CLA production. Abstracts of AOCS 95th
Annual Meeting & Expo, 9–12 May 2004, Cincinnati, Ohio, USA.
Oh, DK, Hong, GH, Lee, Y, Min, SG, Sin, HS and Cho, SK (2003) Production of conjugated
linoleic acid by isolated Bifidobacterium strains. World Journal of Microbiology &
Biotechnology, 19, 907–912.
O’Shea, M, Bassaganya-Riera, J and Mohede, ICM (2004) Immunomodulatory properties of
conjugated linoleic acid. Am. J. Clin. Nutr., 79(S), 1199S–2006S.
Ostrowska, E, Muralitharan, M, Cross, RD, Bauman, DE and Dunshea, FR (1999) Dietary
conjugated linoleic acids increase lean tissue and decrease fat deposition in growing pigs.
J. Nutr., 129, 2037–2042.
Palombo, JD, Ganguly, A, Bistrian, BR and Menard, MP (2002) The antiproliferative effects
of biologically active isomers of conjugated linoleic acid on human colorectal and
prostate cancer cells. Cancer Lett., 28, 163–172.
Pariza, MW (1988) Dietary fat and cancer risk; evidence and research needed. Ann. Rev.
Nutr., 167–183.
Pariza, MW (2004) Perspective on the safety and effectiveness of conjugated linoleic acid.
Am. J. Clin. Nutr., 79 (suppl), 1132S–1136S.
Pariza, MW, Park, Y, Cook, ME, (2000) Mechanisms of action of conjugated linoleic acid:
evidence and speculation. Proc Soc Exptl Biol Med, 223, 8–13.
Pariza, MW, Park, Y and Cook, ME (2001) The biologically active isomers of conjugated
linoleic acid. Prog. Lipid Res., 40, 283–298.
Pariza, MW, Park, Y, Xu, X, Ntambi, J and Kang, KC (2003) Speculation on the mechanisms
of action of conjugated linoleic acid. In: Advances in Conjugated Linoleic Acid Research,
Volume 2 (Sébédio, JL, Christie, WW and Adolf, R, eds), AOCS Press, Champaign,
Illinois, USA, pp.251–258.
Park, Y, Albright, KJ, Liu, W, Storkson, JM, Cook, ME,and Pariza, MW (1997) Effect of
conjugated linoleic acid on body composition in mice. Lipids, 32, 853–858.
Park,Y, Storkson, JM, Albright, KJ, Liu, W and Pariza, MW (1999) Evidence that the trans-
10,cis-12 isomer of conjugated linoleic acid induces body composition changes in mice.
Lipids, 34, 235–241.
Park, Y and Pariza, MW (2001) Lipoxygenase inhibitors inhibit heparin releasable lipopro-
tein lipase activity in 3T3-L1 adipocytes and enhance bodyweight reduction mice by
conjugated linoleic acid. Biochim. Biophys. Acta, 1534, 27–33.
BIOSYNTHESIS, SYNTHESIS AND BIOACTIVITY OF TRANS-10,CIS-12 CLA 253
Park, Y and Pariza, MW (2001) The effects of dietary conjugated nonadecadienoic acid on
body composition in mice. Biochim. Biophys. Acta, 1533, 171–174.
Parodi, PW. 1977, Conjugated octadecadienoic acids of milk fat. J. Dairy Sci., 60, 1550–
1553.
Parodi, PW. 1999, Conjugated linoleic acid and other anticarcinogenic agents of bovine milk
fat. J. Dairy Sci., 82, 1339–1349.
Plourde, M, Jew, S, Cunnane, SC and Jones PJH (2008) Conjugated linoleic acids: why the
discrepancy between animal and human studies? Nutr. Rev., 66, 415–421.
Poulos, SP, Sisk, M, Hausmann, DB, Azain, MJ and Hausamann, GJ, (2001) Pre- and
postnatal dietary conjugated linoleic acid alters adipose development, body weight, gain
and body composition in Sprague-Dwley rats. J. Nutr., 131, 2722–2733.
Peterson, DG, Matitashvili, EA and Bauman, DE (2003) Diet-induced milk fat depression in
dairy cows results in increased trans-10,cis-12 CLA in milk fat and coordinate suppres-
sion of mRNA abundance for mammary enzymes involved in milk fat synthesis. J. Nutr.,
133, 3098–3102.
Reaney, MJT, Liu, YD and Westcott, ND (1999) Commercial production of conjugated
linoleic acid. In: Advances in Conjugated Linoleic Acid Research, First edition (Yurawecz,
MP, Mossoba, MM, Kramer, JKG, Pariza, MW, and Nelson, GJ, eds) AOCS Press,
Champaign, Illinois, USA, pp.39–54.
Riserus, U, Berglund, L and Vessby, B (2001) Conjugated linoleic acid (CLA) reduced
abdominal adipose tissue in obese middle-aged men with signs of metabolic syndrome:
a randomized control trial. J. Obes. Relat. Metab. Disord., 25, 1129–1135.
Riserus, U, Amer, P, Brismar, K and Vessby, B (2002) Treatment with dietary trans-10,cis-
12 conjugated linoleic acid isomers causes isomer-specific insulin resistance in obese
men with the metabolic syndrome. Diabetes Care, 25, 1516–1521.
Riserus, U, Vessby, B, Amer, P and Zethelius, B. (2004) Supplementation with trans-10,cis-
12 conjugated linoleic acid induces hyperproinsulinemia in obese men: close association
with impaired insulin sensitivity. Diabetologia, 47, 1016–1019.
Roche, HM, Noon, E, Sewter, C, Savage, D, Gibney, MJ, O’Rahilly, S and Vidal-Puig, AJ
(2002) Isomer-independent metabolic effects of conjugated linoleic acid: insights from
molecular markers and sterol regulatory element-binding protein-1c and LXR alpha.
Diabetes, 51, 2037–2044.
Rosberg-Cody, E, Johnson, MC, Fitzgerald, GF, Ross, PR and Stanton, C (2007) Heterolo-
gous expression of linoleic acid isomerase from Propionibacterium acnes and
anti-proliferative activity of recombinant trans-10,cis-12 conjugated linoleic acid. Micro-
biology, 153, 2483–2490.
Rosson, RA, Deng, MD, Grund, AD and Peng, SS (2001) Linoleate isomerase. World Patent
WO 01/00846 1-118.
Ryder, JW, Portocarrero, CP, Song XM, Cui, L, Yu, M, Combatsiaris, T, Galuska, D,
Bauman, DE, Barbano, DM, Charron, MJ, Zierath, JR and Housenecht, KL (2001)
Isomer-specific antidiabetic properties of conjugated linoleic acid. Improved glucose
tolerance, skeletal muscle insulin action and UCP-2 gene expression. Diabetes, 50, 1149–
1157.
Sébédio, J-L, Christie, WW and Adlof, R (eds) (2003) Advances in Conjugated Linoleic Acid
Research, Volume 2. AOCS Press, Champaign, Illinois, USA.
Sieber, R, Collomb, M, Aeschlimann, A, Jelen, P and Eyer, H (2004) Impact of microbial
cultures on conjugated linoleic acid in dairy products — a review, International Dairy
Journal, 14, 1–15.
Simon, E, Macarulla, MT, Fernandez-Quinteal, A, Rodriguez,VM and Portillo, MP (2005)
Body fat-lowering effect of conjugated linoleic acid is not due to increased lipolysis.
J. Phys. Biochem, 61, 363–370.
Stachowska, E, Baskiewicz-Masiuk, M, Dziedziejko, V, Adler, G, Bober, J, Machalinski, B
254 TRANS FATTY ACIDS IN HUMAN NUTRITION
and Chlubek, D (2007) Conjugated linoleic acids can change phagocytosis of human
monocytes/macrophages by reduction of Cox-2 expression. Lipids, 42, 707–716.
Steck, SE, Chalecki, AM, Miller, P, Conway, J, Austin, GL, Hardin, JW, Albright, CD and
Thuillier, P (2007) Conjugated linoleic acid supplementation for twelve weeks increases
lean body mass in obese humans. J. Nutr., 137, 1188–1193.
Tricon, S, Burdge, GC, Kew, S, Banerjeee, T, Russel, JJ, Grimble, RF, Williams, CM, Calder,
PC and Yaqoob, P (2004) Effects of cis-9,trans-11 and trans-10,cis-12 conjugated
linoleic acid on immune cell function in healthy humans. Am. J. Clin. Nutr., 80, 1626–
1633.
Tricon, S, Burdge, GC, Kew, S, Banerjeee, T, Russel, JJ, Jones, EL, Grimble, RF, Williams,
CM, Yaqoob, P and Calder, PC (2004) Opposing effects of cis-9, trans-11 and trans-
10,cis-12 conjugated linoleic acid on blood lipids in healthy humans. Am. J. Clin. Nutr.,
80, 614–620.
Tsuboyama-Kasaoka N, Takahashi, M, Tanemura, K, Kim, HJ, Tange, T, Okuyama, H,
Kasai, M, Ikemoto, S and Ezaki, O (2000) Conjugated linoleic acid supplementation
reduces adipose tissue by apoptosis and develops lipodystrophy in mice. Diabetes, 49,
1534–1542.
Uehara, H, Suganuma, T, Negishi, S, Satoru, U and Sato, K (2006) A novel method for solvent
fractionation of two CLA isomers., J. Am. Oil Chem. Soc., 83, 261–267.
Vahvaselka, M, Lehtinen, P, Sippola, S and Laakso, S (2004) Enrichment of conjugated
linoleic acid in oats (Avena sativa L.) by microbial isomerization, J. Agric. Food Chem.,
52, 1749–1752.
Verhulst, A (1987) Isomerization of polyunsaturated long chain fatty acids by Propionibac-
teria. Syst. Appl. Microbiol., 9, 12–15.
Wahle, KWJ, Heys, SD and Rotondo, D (2004) Conjugated linoleic acids; are they beneficial
or detrimental to health? Prog. Lipid Res., 43, 553–587.
Wallace, RJ, McKain, N, Shingfield, KJ and Devillard, E (2007) Isomers of conjugated
linoleic acids are synthesized via different mechanisms in ruminal digesta and bacteria.
J. Lipid Res., 48, pp.2247–2254.
Wang, LS, Huang, YW, Sugimoto, Y, Liu, S, Chang, HL, Ye, W, Shu, S and Lin, YC (2006)
Conjugated linoleic acid (CLA) up-regulates the estrogen-regulated cancer suppressor
gene, protein tyrosine phosphatase gamma (PTPgamma), in human breast cells. Antican-
cer Res., 26, 27–34.
Wang, YW and Jones, PJH (2004) Conjugated linoleic acid and obesity control: efficacy and
mechanisms. Intl J. Obesity, 28, 941–955.
West, DB, DeLaney, JP, Camet, PM, Blohm, FY, Truett, AA and Sciemaca, J (1998) Effects
of conjugated linoleic acid on body fat and energy metabolism in the mouse. Am. J.
Physiol., 275, R667–R672.
West, DB, Blohm, FY, Truett, AA and DeLaney, JP (2000) Conjugated linoleic acid
persistently increases total energy expenditure in AKR/J mice without increasing
uncoupling protein gene expression. J. Nutr., 130, 2471–2477.
Yamasaki, M, Chujo, H, Hiarao, A, Koyanagi, N, Okamoto, T, Tojo, N, Oishi, A, Iwata, T,
Yamauchi-Sato, Y, Yamamoto, T, Tsutsumi, K, Tachibana, H and Yamada, K (2003)
Immunoglobulin and cytokine production from spleen lymphocytes is modulated in
C57BL/6J mice by dietary cis-9,trans-11 and trans-10,cis-12 conjugated linoleic acid.
J. Nutr., 133, 784–788.
Xu, C, Lee, H, Lee, B, Wang, J, Hong, Z, Kim, T, Kang, S, Choi, N, Roh, S and Choi, Y (2006)
Production of lean beef containing a high content of trans-10,cis-12 conjugated linoleic
acid by feeding a high-temperature-micro-time-treated diet with extruded soybean.
Biosci. Biotechnol. Biochem., 70, 2589–2597.
CHAPTER 10
Observational epidemiological studies on
intake of trans fatty acids and risk of ischaemic
heart disease
A. Background
3. Aim
The aim of this chapter is to summarize and evaluate the results from observa-
tional epidemiological studies on the intake of TFA and the risk of IHD.
B. Methods
The selection criteria for the observational epidemiological studies were that
they included data on TFA intake and development of IHD in the general
population. Measures of TFA intake included both direct measures of dietary
TFA intakes and proportions of TFA in adipose tissue, whole blood, plasma, or
erythrocytes which are considered to provide biomarkers of TFA intake
(Baylin et al., 2002; Baylin et al., 2005; Sun et al., 2007a).
258 TRANS FATTY ACIDS IN HUMAN NUTRITION
C. Results
1. Included studies
In total, 31 potential studies on the association between TFA intake and risk of
IHD emerged from the PubMed search (Aro et al., 1995; Ascherio et al., 1994;
Ascherio et al., 1996; Baylin et al., 2003; Bolton-Smith et al., 1996; Clifton
et al., 2004; Colon-Ramos et al., 2006; Dlouhy et al., 2003; Fritsche et al.,
1998; Ghahremanpour et al., 2008; Harris et al., 2007; Heckers et al., 1977;
Hodgson et al., 1996; Jakobsen et al., 2008; Kromhout et al., 1995; Lemaitre
et al., 2002; Lemaitre et al., 2006; Lopes et al., 2007; Mozaffarian et al., 2007;
Oomen et al., 2001; Pedersen et al., 2000; Pietinen et al., 1997; Roberts et al.,
1995; Siguel & Lerman, 1993; Singh et al., 1996; Sun et al., 2007b; Sun et al.,
2007c; Thomas et al., 1983a; Thomas et al., 1983b; van de Vijver et al., 1996;
Willett et al., 1993). Seven studies were excluded as the selection criteria were
not fulfilled (Fritsche et al., 1998; Heckers et al., 1977; Hodgson et al., 1996;
Mozaffarian et al., 2007; Singh et al., 1996; Thomas et al., 1983b; van de
Vijver et al., 1996). Four studies were included after search in references (Hu
et al., 1997; Hu et al., 1999; Oh et al., 2005; Thomas & Scott, 1981). In total,
28 studies met the selection criteria.
trend = 0.11); for ruminant TFA the OR for highest compared with lowest
quintile was 0.96 (95% CI: 0.58–1.59, p value for trend = 0.78); and for
industrial TFA the OR for highest compared with lowest quintile was 1.26
(95% CI: 0.92–1.72, p value for trend = 0.11). Intake of ruminant TFA was
negatively associated with IHD after control for dietary and non-dietary IHD
risk factors among men who had undiagnosed IHD at baseline (OR for highest
compared with lowest quintile, OR = 0.65 (95% CI: 0.41–1.04, p value for
trend = 0.02) (Bolton-Smith et al., 1996). Intakes of total and industrial TFA
were not associated with IHD; for total TFA the OR for highest compared with
lowest quintile was 0.87 (95% CI: 0.60–1.24, p value for trend = 0.48); for
industrial TFA the OR for highest compared with lowest quintile was 1.08
(95% CI: 0.79–1.48, p value for trend = 0.99) (Bolton-Smith et al., 1996).
3. Case-control studies
In total, 18 case-control studies met the selection criteria; one study using direct
measures of dietary TFA intakes (Ascherio et al., 1994), 15 studies using
biomarkers of TFA intakes (Aro et al., 1995; Baylin et al., 2003; Colon-Ramos
et al., 2006; Dlouhy et al., 2003; Ghahremanpour et al., 2008; Harris et al.,
2007; Lemaitre et al., 2002; Lemaitre et al., 2006; Pedersen et al., 2000; Roberts
et al., 1995; Siguel & Lerman, 1993; Sun et al., 2007b; Sun et al., 2007c; Thomas
& Scott, 1981; Thomas et al., 1983a), and two studies using both direct measures
of dietary TFA intakes and biomarkers of TFA intakes (Clifton et al., 2004;
Lopes et al., 2007). Three of the case-control studies were nested case-control
studies (Lemaitre et al., 2006; Sun et al., 2007b; Sun et al., 2007c). Two of these
studies (Sun et al., 2007b; Sun et al., 2007c) were nested in the Nurses’ Health
Study and therefore these two studies cannot be considered independent of the
follow-up studies by Willett et al. (1993), Hu et al. (1997), Hu et al. (1999) and
Oh et al. (2005) also based on data from the Nurses’ Health Study (described
below). The studies by Baylin et al. (2003) and Colon-Ramos et al. (2006) were
partly based on the same participants and therefore these studies cannot be
considered independent. In the 2003 study (Baylin et al., 2003), the investigators
reported results on adipose tissue total TFA; trans 16:1n–7; total trans 18:1; and
total trans 18:2 before a reduction of TFA in foods. In the 2006 study (Colon-
Ramos et al., 2006), the investigators reported results on adipose tissue total
TFA, total trans 18:1, and total trans 18:2 before and after a reduction of TFA in
foods. For this review, the 2003 study (Baylin et al., 2003) was selected for
results on adipose tissue trans 16:1n–7 and the 2006 study (Colon-Ramos et al.,
2006) was selected for results on adipose tissue total TFA, total trans 18:1, and
total trans 18:2 before and after a reduction of TFA in foods. Characteristics of the
case-control studies included are shown in Tables 1 and 2. Cases and controls
were sampled in two independent time periods before and after a reduction of
TFA in foods. The results from the two different time periods can be considered
independent and are referred to as two separate studies.
Table 1. Case-control studies investigating the association between intake of trans fatty acids and the risk of ischaemic heart disease.
260
Ascherio 239 cases F&M 58±10 FFQ g/d Total TFA: Total TFA: Sex, age, BMI, family ↑‡ →‡ ↑
(1994) of first AMI 4.7±2.6 Q1(median)=1.7: history of IHD,
(no history OR=1.00 (reference) physical activity,
of hyper- Q2(median)=2.5: smoking status,
choleste- OR=0.63(0.31–1.30) alcohol consumption,
rolaemia, Q3(median)=3.4: intakes of SFA, MUFA,
diabetes, or OR=1.03 (0.53–2.00 linoleic acid, and chole-
angina Q4(median)=4.5: sterol, and hypertension;
pectoris) OR=1.35 (0.69–2.63) further adjustment for
Q5(median)=6.5: education, occupation,
OR=2.02 (1.03–3.93) marital status, personal-
p value for trend=0.002 ity type, use of multi-
vitamin supplement,
R-TFA: R-TFA:
use of aspirin, and
NA Q1(median)=0.5:
intakes of fibre, caro-
OR=1.00 (reference)
tene, vitamin E, and
TRANS FATTY ACIDS IN HUMAN NUTRITION
Q2(median)=0.7:
vitamin C did not
OR=0.64 (0.33–1.23)
change the results of
Q3(median)=1.0:
intake of total TFA
OR=0.87 (0.46–1.66)
Q4(median)=1.2:
OR=0.86 (0.43–1.71)
Q5(median)=1.8:
OR=1.23 (0.60–2.50)
p value for trend=0.09
I-TFA: I-TFA:
NA Q1(median)=0.8:
OR=1.00 (reference)
Q2(median)=1.6:
OR=1.46 (0.74–2.88)
Q3(median)=2.3:
OR=1.16 (0.57–2.34)
Q4(median)=3.3:
OR=1.26 (0.61–2.61)
Q5(median)=5.0:
OR=3.05 (1.48–6.31)
p value for trend=0.001
OR=0.81 (0.42–1.57)
Q3(median)=NA:
OR=0.40 (0.19–0.83)
Q4(median)=NA:
261
OR=0.72 (0.36–1.48)
Table 1. continued
262
Q5(median)=NA:
OR=2.03 (0.98–4.22)
p value for trend=0.0001
R-TFA:
Q1(median)=NA:
OR=1.00 (reference)
Q2(median)=NA:
OR=1.17 (0.59–2.34)
Q3(median)=NA:
OR=1.12 (0.53–2.34)
Q4(median)=NA:
OR=1.00 (0.45–2.21)
Q5(median)=NA:
OR=1.02 (0.43–2.41)
p value for trend=0.57
TRANS FATTY ACIDS IN HUMAN NUTRITION
I-TFA:
Q1(median)=NA:
OR=1.00 (reference)
Q2(median)=NA:
OR=0.74 (0.38–1.46)
Q3(median)=NA:
OR=0.43 (0.21–0.90)
Q4(median)=NA:
OR=0.63 (0.30–1.32)
Q5(median)=NA:
OR=1.94 (0.93–4.04)
p value for trend=0.0001
Controls NA
Clifton 79 cases F&M 56±13 FFQ g/d NA Total TFA: Intakes of energy and SFA →
(2004) of first IHD Q1(median)=1.6:
(no history OR=1.00 (reference)
of hyper Q2(median)=2.4:
triglyce- OR=1.37 (95% CI NA)
ridaemia, Q3(median)=3.0:
hyper- OR=1.51 (95% CI NA)
cholesterol- Q4(median)=3.7:
aemia, or OR=0.81 (95% CI NA)
diabetes) Q5(median)=5.5:
OR=0.98 (95% CI NA)
p value for trend=NS
174 controls 56±11 NA
matched with
the cases for
sex, age, and
postal code
(no history of
hypertriglyce-
ridaemia,
hyperchole-
sterolaemia,
diabetes, or
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS
IHD)
263
Table 1. continued.
264
Lopes (2007) 297 cases of M 57±11 FFQ g/d NA Total TFA: Age, BMI, family →
first AMI Q1(median)=0.3: history of AMI,
OR=1.00 (reference) education, physical
Q2(median)=0.6: activity, smoking
OR=0.79 (0.47–1.33) status, and intake of
Q3(median)=1.2: energy; further
OR=0.66 (0.38–1.14) adjustment for
Q4(median)=1.4: dyslipidaemia, hyper-
OR=0.81 (0.48–1.37) tension, diabetes,
p value for trend=0.34 and angina pectoris
did not change the results
310 controls 57±11 NA
(no history
of AMI)
TRANS FATTY ACIDS IN HUMAN NUTRITION
AMI, acute myocardial infarction; BMI, body mass index; CI, confidence interval; F, Female; FFQ, food frequency questionnaire; I-TFA, industrial trans fatty acids;
M, Male; MUFA, monounsaturated fatty acids; NA, not available; NS, not statistically significant; OR, odds ratio; Q, quintile/quartile; R-TFA, ruminant trans fatty
acids; SFA, saturated fatty acids; SD, standard deviation; TFA, trans fatty acids.
†↑ Statistically significant positive association.
‡→ No statistically significant association.
Table 2. Case-control studies investigating the association between biomarkers of trans fatty acids and the risk of ischaemic heart disease.
Thomas (1981)
136 cases of IHD M 53±NA Subcu- % of FA NA OR NA; total TFA was not →
death taneous statistically different between
AT (abdo- cases and controls
minal)
95 controls of death 52±NA NA
from unrelated causes
matched with the cases
for sex, age, and area
Thomas(1983a)
136 cases of M 53±NA Subcu- % of FA NA OR NA; total trans 16:1 was ↑‡ →§
IHD death taneous significantly higher in cases
AT (abdo- than in controls (p value=
minal) < 0.005); total trans 18:1
was not statistically different
between cases and controls
95 controls of death 52±NA NA
from unrelated causes
matched with the cases
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS
0.40±0.14 IHD
Table 2. continued
266
Q2=0.5–0.5:
OR=1.26 (0.49–3.21)
Q3=0.6–0.6:
OR=1.05 (0.40–2.78)
Q4=0.6–0.7:
OR=1.37 (0.53–3.56)
Q5=≥0.7:
OR=0.99 (0.35–2.84)
p value for trend=NA
286 controls matched 57±7 Total TFA:
with the cases for sex 2.9±0.7
and age Total trans18:1:
2.3±0.7
Total trans 18:2:
0.6±0.2
TRANS FATTY ACIDS IN HUMAN NUTRITION
Pedersen (2000)
100 cases of first F&M 62±8 Subcu- % of FA NA Total TFA: Sex, age, waist to hip →
AMI (no history taneous AT Q1(cut-off)=<3.4: ratio, family history of
of diabetes) (buttock) OR=1.00 (reference) IHD, smoking status,
Q2(cut-off)=>3.4: and AT linoleic acid;
OR=1.04 (0.32–3.37) further adjustment
Q3(cut-off)=>3.8: for AT α-linolenic
OR=1.07 (0.33–3.48) acid did not alter
Q4(cut-off)=>4.4: the associations
OR=0.54 (0.14–2.07)
Q5(cut-off)=>4.8:
OR=1.49 (0.47–4.69)
p value for trend=0.19
98 controls matched 62±8 NA
with the cases for sex,
age, and geographical
location (no history
of diabetes or AMI)
Lemaitre (2002)
179 cases of SCD F&M 60±10 Erythro % of FA Total TFA: Total TFA: Age, weight, height, ↑ → ↑
(nohistory of -cytes 2.1±0.6 IQ range=0.9: family history of AMI
IHD OR=1.47 (1.01–2.13) or sudden death,
Total Total trans 18:1: education, physical
trans 18:1: IQ range=0.8: activity, smoking
1.7±0.5 OR=0.77 (0.48–1.24) status, intake of SFA,
Total Total trans 18:2: hypertension, and
trans 18:2: IQ range=0.1: diabetes; trans 18:1
0.2±0.1 OR=3.05 (1.71–5.44) and trans 18:2 were
also adjusted for erythro-
cyte long-chain n-3
PUFA and were assessed
simultaneously
285 controls matched 58±10 Total TFA:
with the cases for 2.0±0.5
sex and age (no Total trans 18:1:
history of IHD) 1.6±0.5
Total trans 18:2:
0.2±0.1
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS
269
Table 2. continued
270
Baylin (2003)#
482 cases of first F&M 57±10 Subcuta- % of FA NA Trans 16:1, n–7: Income, years living ↑
AMI (no history neous AT Q1(median)=0.0: in the house, physical
of CVD) (buttock) OR=1.00 (reference) activity, smoking status,
Q2(median)=0.1: alcohol consumption,
OR=1.57 (0.83–2.98) AT α-linolenic acid,
Q3(median)=0.1: intakes of energy, SFA,
OR=1.39 (0.73–2.66) and vitamin E, hyper-
Q4(median)=0.1: tension, and diabetes
OR=1.34 (0.65–2.79) (conditioned on the
Q5(median)=0.1: matching factors: sex,
OR=2.58 (1.22–5.43) age, and area of residence);
p value for trend=0.025 further adjustment for
BMI, waist to hip ratio,
use of multivitamin
TRANS FATTY ACIDS IN HUMAN NUTRITION
Total trans18:2:
Q1(median)=0.8:
OR=1.00 (reference)
Q2(median)=1.0:
OR=1.04 (0.60–1.80)
Q3(median)=1.2:
OR=2.16 (1.18–3.96)
Q4(median)=1.5:
OR=2.86 (1.50–5.46)
Q5(median)=2.0:
OR=4.76 (2.24–10.11)
p value for trend=≤0.001
477 controls 57±11 NA
matched with the
cases for sex, age,
and area of residence
(no history of AMI)
Colon-Ramos
(2006) 2000–03#
1,320 cases of first F&M 59±11 Subcuta- % of FA NA Total TFA: Income, physical → → →
AMI (no history of neous AT Q1(median)=1.8: activity, smoking status,
of CVD) (buttock) OR=1.00 (reference) alcohol consumption,
Q2(median)=2.3: hypertension, AT
OR=0.78 (0.58,1.04) α-linolenic acid, intakes
Q3(median)=2.6: of energy, SFA, and
OR=1.03 (0.80,1.43) vitamin E, and diabetes
Q4(median)=2.9: (conditioned on the
OR=0.88 (0.65,1.18) matching factors: sex,
Q5(median)=3.4: age, and area of resi-
OR=1.03 (0.75,1.42) dence); further adjustment
p value for trend=0.65 for weight, height, waist
Total trans 18:1: to hip ratio, education,
Q1(median)=0.9: other AT FA, and intakes
OR=1.00 (reference) of protein, cholesterol,
Q2(median)=1.1: fibre, folate, margarine,
OR=0.94 (0.71,1.24) fish, and dairy products
Q3(median)=1.3: did not change the results
OR=0.92 (0.69,1.23)
Q4(median)=1.5:
OR=0.97 (0.72,1.30)
Q5(median)=1.9:
OR=1.02 (0.75,1.37)
p value for trend=0.80
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS
OR=1.09 (0.81–1.46)
Table 2. continued
274
Q3(median)=1.1:
OR=1.10 (0.81–1.50)
Q4(median)=1.2:
OR=1.05 (0.75–1.46)
Q5(median)=1.4:
OR=1.15 (0.80–1.64)
p value for trend=0.56
1,320 controls 59±11 NA
matched with the
cases for sex, age,
and area of residence
(no history of AMI)
Lemaitre (2006)
214 cases of fatal F&M 77±6 Plasma % of FA Total TFA: Total TFA: Education, smoking → → ↓** ↑
IHD NA IQ range =1.4:OR=0.94 status, plasma long-chain
TRANS FATTY ACIDS IN HUMAN NUTRITION
Harris (2007)
94 cases of ACS F&M 46±5 Whole % of FA Total TFA: Total TFA: Sex, race, age, BMI, → → →
blood 2.1±0.7 OR(1 SD increase)=0.87 education, smoking
(0.55–1.37) status, alcohol con-
Total Total trans 18:1: sumption, TG, LDL-
trans 18:1: OR(1 SD increase)=0.80 cholesterol, HDL-
1.7±0.6 (0.51–1.25) cholesterol, diabetes,
Total Total trans 18:2: and PCE
trans 18:2: OR(1 SD increase)=1.41
0.4±0.2 (0.84–2.39)
94 controls 47±5 Total TFA:
matched with the 2.0±0.9
cases for sex, race, Total trans 18:1:
and age 1.6±0.8
Total trans 18:2:
0.4±0.2
Lopes (2007)
49 cases of first M 55±9 Subcuta- % of FA Total TFA: Total TFA: Age, BMI, family history ↓
AMI neous AT 0.6±0.1 T1:OR=1.00 (reference) of AMI, education,
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS
results
Table 2. continued
276
Sun (2007c)
166 cases of IHD (no F 61±6 Plasma % of FA Trans Trans 16:1, n–7: BMI, menopausal status, →
history of CVD) 16:1 n–7: T1(mean)=0.1: postmenopausal hormone
0.1±0.0 OR=1.00 (reference) use, family history of
T2(mean)=0.2: AMI, physical activity,
OR=0.79 (0.46–1.36) alcohol consumption, use
T3(mean)=0.2: of aspirin, plasma linoleic
OR=0.79 (0.44–1.42) acid and total TFA,
p value for trend=NS hypertension, hyper-
cholesterolaemia, and
diabetes (conditioned on
the matching factors: age,
smoking status,fasting
status at blood draw, and
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS
cholesterolaemia, and
diabetes (conditioned on
the matching factors: age,
smoking status, fasting
status at blood draw,
and date of blood draw);
further adjustment for
intakes of fibre, calcium,
vitamin D, folate,
fruits, and vegetables
did not change the results
Controls Trans 16:1 n–7:
0.1±0.0
Ghahremanpour (2008)
105 cases of IHD F&M 54±9 Subcuta- % of FA Total TFA: Total TFA: Smoking status, AT ↑ → ↑ →
(no history of hyper- neous AT 8.9±2.5 IQ range=13.7: PUFA, hypertension,
cholesterolaemi, (buttock) OR=1.41 (1.0–1.8) and TG
diabetes, or CVD) Total Total trans 16:1:
trans 16:1: IQ range=1.5:
0.5±0.3 OR=0.80 (0.3–1.9)
Total Total trans 18:1:
trans 18:1: IQ range=11.3:
6.6±2.1 OR=1.32 (1.0–1.8)
Total Total trans18:2:
trans 18:2: IQ range=4.6:
1.8±0.8 OR=1.85 (0.6–4.8)
68 controls matched 52±7 Total TFA:
with the cases for 8.0±2.2
sex and age (no Total trans 16:1:
history of hyper- 0.5±0.3
cholesterolaemi, Total trans 18:1:
diabetes, or CVD) 6.0±1.8
Total trans 18:2:
1.5±0.7
ACS, acute coronary syndrome; AMI, acute myocardial infarction; AT, adipose tissue; BMI, body mass index; CE, cholesterol esters; CI, confidence interval; CVD,
cardiovascular disease; F, Female; FA, fatty acids; FFQ, food frequency questionnaire; HDL, high-density lipoprotein; IQ, interquintile; LDL, low-density lipoprotein; M,
Male; MUFA, monounsaturated fatty acids; NA, not available; NS, not statistically significant; OR, odds ratio; PCE, percutaneous coronary intervention; PUFA,
polyunsaturated fatty acids; Q, quintile/quartile; SCD, sudden cardiac death; SFA, saturated fatty acids; SD, standard deviation; T, tertile; TG, triacylglycerol; TFA, trans fatty
acids.
† Trans 16:1n–7 or the sum of trans 16:1n–7 and 16:1n–9.
‡ ↑ Statistically significant positive association or statistically significant higher in cases than in controls.
§ → No statistically significant association or no statistically significant difference between cases and controls.
# The studies by Baylin et al. (2003) and Colon-Ramos et al. (2006) were partly based on some of the same participants and therefore these studies cannot be considered
independent. In the study by Baylin et al. (2003), the investigators reported results on AT total TFA, trans 16:1n–7, total trans 18:1, and total trans 18:2 before a reduction
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS
of TFA in foods. In the study by Colon-Ramos et al. (2006), the investigators reported results on AT total TFA, total trans 18:1, and total trans 18:2 before and after a reduction
of TFA in foods. For this review, the study by Baylin et al. (2003) was selected for results on trans 16:1n–7 and the study by Colon-Ramos et al. (2006) was selected for total
TFA, total trans 18:1, and total trans 18:2 before and after a reduction of TFA in foods. Cases and controls were sampled in two independent time periods before and after
a reduction of TFA in foods. The results from the two different time periods can be considered independent and are referred to as two separate studies (Colon-Ramos (2006)
279
(Clifton et al., 2004; Ghahremanpour et al., 2008) (Table 2). Among the two
studies which showed no significant associations, one study, however, indi-
cated a trend for a negative association between adipose tissue trans 16:1 and
risk of IHD (Ghahremanpour et al., 2008) (Table 2). One study showed a
significant positive association between adipose tissue total trans 18:1 and risk
of IHD (Ghahremanpour et al., 2008) and six studies showed no significant
associations (Aro et al., 1995; Colon-Ramos et al., 2006; Dlouhy et al., 2003;
Roberts et al., 1995; Thomas et al., 1983a) (Table 2). However, among the six
studies which showed no significant associations, three studies indicated
trends for positive associations between adipose tissue trans 18:1 and risk of
IHD (Aro et al., 1995; Colon-Ramos et al., 2006; Dlouhy et al., 2003) and one
study indicated a trend for a negative association (Roberts et al., 1995)
(Table 2). In the study by Aro et al. (1995), the trend for a positive association
between adipose tissue total trans 18:1 appeared after exclusion of participants
from the two Spanish centres for whom proportions of adipose tissue TFA were
very low compared with participants from other countries. One study showed
a significant positive association between adipose tissue total trans 18:2 and
risk of acute myocardial infarction (Colon-Ramos et al., 2006) and three
studies showed no significant associations between adipose tissue total trans
18:2 and risk of IHD (Colon-Ramos et al., 2006; Ghahremanpour et al., 2008;
Roberts et al., 1995) (Table 2). Among the three studies which showed no
significant associations, two studies, however, indicated trends for positive
associations between adipose tissue trans 18:2 and risk of IHD (Colon-Ramos
et al., 2006; Ghahremanpour et al., 2008) (Table 2). In the study by Clifton
et al. (2004), adipose tissue trans 18:1n–7 was positively associated with risk
of IHD after control for dietary IHD risk factors, whereas other positional
isomers within trans 18:1 (18:1n–8 and 18:1n–9) were not significantly
associated with risk of IHD.
By controlling for other adipose tissue fatty acids in analyses of adipose
tissue TFA and risk of IHD, the models address the effects of the latter
independently of the former. One study reported results from multiple analyses
without and with control for other adipose tissue fatty acids (Pedersen et al.,
2000). In that study, investigating the association between total adipose tissue
TFA and risk of acute myocardial infarction the significant positive association
became non-significant after additional control for adipose tissue linoleic acid
(cis 18:2n–6) (Pedersen et al., 2000).
Five studies controlled for dietary variables (energy intake, saturated fatty
acids, or vitamin E) in multiple analyses of adipose tissue TFA and risk of IHD
(Baylin et al., 2003; Clifton et al., 2004; Colon-Ramos et al., 2006; Lopes
et al., 2007) (Table 2). By controlling for these dietary variables, the models
address the effect of adipose tissue TFA independent of these dietary variables.
However, as the proportion of adipose tissue TFA is expressed as percentage of
total adipose tissue fatty acids the proportion of TFA also depends indirectly on
282 TRANS FATTY ACIDS IN HUMAN NUTRITION
for, example, the saturated fatty acids intake (i.e. because adipose tissue TFA
is expressed as percentage of total fatty acids, any increase or decrease in other
fatty acids will result in reciprocal changes in the TFA percentage).
In five studies reporting results from multiple analyses, the analyses in-
cluded control for plasma total cholesterol or diabetes mellitus which may be
considered as potential metabolic effects of intake of TFA (Baylin et al., 2003;
Clifton et al., 2004; Colon-Ramos et al., 2006; Roberts et al., 1995) (Table 2).
within trans 18:1 (18:1n–7, 18:1n–9 and 18:1n–12) and within trans 18:2
(18:2n–6tt, 18:2n–6ct and 18:2 n–6tc) were significantly positively associated
with risk of IHD after control for dietary and non-dietary IHD risk factors.
One study reported results from multiple analyses without and with control for
other fatty acids (Sun et al., 2007b). In that study, investigating the association
between erythrocyte total TFA, total trans 18:1, and total trans 18:2 and risk of
IHD, the significant positive associations became stronger after additional
control for erythrocyte n–6 PUFA and long-chain n–3 PUFA (Sun et al., 2007b).
In two studies, total trans 18:1 and total trans 18:2 were assessed simultaneously
(Lemaitre et al., 2002, 2006) (Table 2). In the study by Lemaitre et al. (2002),
erythrocyte total trans 18:1 was non-significantly positively associated with risk
of sudden cardiac death in analysis without control for erythrocyte total trans
18:2; however, when assessed simultaneously, erythrocyte total trans 18:1
became non-significantly negatively associated with risk of sudden cardiac
death. Erythrocyte total trans 18:2 was significantly positively associated with
risk of sudden cardiac death in analyses without and with control for erythrocyte
total trans 18:1 (Lemaitre et al., 2002). In the study by Lemaitre et al. (2006),
plasma total trans 18:1 was significantly negatively associated with risk of fatal
IHD in analyses without and with control for plasma total trans 18:2. Plasma total
trans 18:2 was non-significantly positively associated with risk of fatal IHD in
analysis without control for plasma total trans 18:1; however, when assessed
simultaneously, plasma total trans 18:2 became significantly positively associ-
ated with risk of fatal IHD (Lemaitre et al., 2006).
One study controlled for intake of saturated fatty acids in multiple analyses
of erythrocyte total TFA, total trans 18:1, and total trans 18:2 and risk of
sudden cardiac death (Lemaitre et al., 2002) (Table 2). By controlling for
intake of saturated fatty acids, the models address the effect of erythrocyte TFA
independent of saturated fatty acids intake. However, as the proportion of
erythrocyte TFA is expressed as percentage of total erythrocyte fatty acids the
proportion of TFA also depends indirectly on the saturated fatty acids intake.
In five studies reporting results from multiple analyses, the analyses in-
cluded control for LDL-cholesterol, HDL-cholesterol, hypercholesterolaemia
or diabetes mellitus which may be considered as potential metabolic effects of
intake of TFA (Harris et al., 2007; Lemaitre et al., 2002; Lemaitre et al., 2006;
Sun et al., 2007b; Sun et al., 2007c) (Table 2).
Willett (1993) F, NA FFQ, g/d Non-fatal 356 8 R-TFA: Age, BMI, family →‡ ↑§
69,181 (no (energy- AMI or Q1:RR=1.00 history of AMI, meno-
history of adjusted), fatal IHD (reference) pausal status, post-
hypercholes- NA Q2:RR=0.76 menopausal hormone
terolaemia, (0.51–1.12) use, smoking status,
diabetes, stroke, Q3:RR=0.69 alcohol consumption,
angina pectoris, (0.43–1.10) use of multivitamin
or AMI) who Q4:RR=0.55 supplement, intakes
reported no (0.31–0.96) of energy, SFA,
change in margar- Q5:RR=0.59 MUFA, and linoleic
ine intake in (0.30–1.17) acid, and hypertension
previous 10 y p value for trend=0.23
I-TFA:
Q1:RR=1.00
(reference)
Q2:RR=1.43
TRANS FATTY ACIDS IN HUMAN NUTRITION
(1.00–2.04)
Q3:RR=1.11
(0.74–1.66)
Q4:RR=1.39
(0.91–2.13)
Q5:RR=1.78
(1.12–2.83)
p value for trend=0.009
Ascherio (1996) M, 40–75 FFQ, g/d Non-fatal 734 6 Total TFA: Age, BMI, family →
43,757 (no (energy- AMI or Q1(median)=1.5: history of AMI,
history of diabetes, adjusted), fatal IHD RR=1.00 (reference) profession, physical
peripheral arterial NA Q2(median)=2.2: activity, smoking
disease, stroke, RR=1.12 (0.86–1.44) status, alcohol con-
angina pectoris, Q3(median)=2.7: sumption, intake of
transient ischaemic RR=1.12 (0.87–1.46) fibre, hypertension,
attack, coronary Q4(median)=3.3: and hypercholeste-
artery surgery, RR=1.12 (0.86–1.46) rolaemia
or AMI) Q5(median)=4.3:
RR=1.21 (0.93–1.58)
p value for trend=0.20
E%, NA Total TFA: Age, BMI, family →
RR(2 unit)=1.13 history of AMI,
(0.81–1.58) profession, physical
activity, smoking
status, alcohol con-
sumption, intakes of
energy, fat, and fibre,
hypertension, and
hypercholeste-
rolaemia
g/d Fatal 229 Total TFA: →
(energy- IHD Q1(median)=1.5:
adjusted), RR=1.00 (reference)
NA Q2(median)=2.2:
RR=1.63 (1.01–2.62)
Q3(median)=2.7:
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS
RR=1.18 (0.71–1.96)
Q4(median)=3.3:
RR=1.59 (0.98–2.60)
Q5(median)=4.3:
285
Table 3. continued
286
RR=1.41 (0.86–2.32)
p value for trend=0.42
E%, NA Total TFA: →
RR(2 unit)=0.93
(0.52–1.69)
Pietinen (1997) M, 50–69 FFQ, g/d Non-fatal 1,399 6 Total TFA: Age, BMI, education, →
21,930 smokers (energy- AMI or Q1(median)=1.3: physical activity,
(no history of adjusted), fatal IHD RR=1.00 (reference) smoking status, alcohol
diabetes, stroke, NA Q2(median)=1.7: consumption, intakes
exercise-related RR=1.10 (0.93–1.31) of energy and fibre,
chest pain, angina Q3(median)=2.0: blood pressure, and
pectoris, or AMI) RR=0.97 (0.82–1.16) treatment group;
Q4(median)=2.7: further adjustment
RR=1.07 (0.90–1.28) for intake of SFA,
Q5(median)=6.2: linoleic acid, and
TRANS FATTY ACIDS IN HUMAN NUTRITION
RR=0.94 (0.73–1.20)
Q5(median)=5.1:
RR=1.23 (0.97–1.55)
p value for trend=0.004
287
Table 3. continued
288
Oomen (2001) M, 71±5 Dietary Non-fatal 98 10 Total TFA: Age, BMI smoking ↑ → →
667 (no history history, E%, AMI or RR(2 unit)=1.28 status, alcohol con-
of angina pectoris Total TFA: fatal IHD (1.01–1.61) sumption, use of
or AMI) 4.3±2.2 R-TFA: vitamin supplement,
R-TFA: RR(0.5 unit)=1.17 intakes of energy,
0.7± 0.2 (0.69–1.98) SFA, MUFA, PUFA,
I-TFA I-TFA (trans 18:1): cholesterol, and fibre
(trans 18:1): RR(0.5unit)=1.05
2.1±1.2 (0.94–1.17)
I-TFA I-TFA (other TFA):
(other TFA): RR(0.5unit)=1.07
1.6±1.4 (0.99–1.15)
Fatal IHD 49 Total TFA: →
RR(2 unit)=1.33
(0.96–1.86)
TRANS FATTY ACIDS IN HUMAN NUTRITION
Oh (2005) F, NA FFQ, E%. Non-fatal 1,766 20 Total TFA: Age, BMI, family ↑
78,778 (no NA AMI or Q1(median)=1.3: history of AMI, meno-
history of hyper- fatal IHD RR=1.00 (reference) pausal status, post-
cholesterolaemia, Q2(median)=1.6: menopausal hormone
diabetes, CVD, RR=1.08 (0.92–1.26) use, physical activity,
or cancer) Q3(median)=1.9: smoking status,
RR=1.29 (1.09–1.53) alcohol consumption,
Q4(median)=2.2: use of multivitamin
RR=1.19 (0.99–1.44) supplement, use of
Q5(median)=2.8: vitamin E supplement,
RR=1.33 (1.07–1.66) use of aspirin, intakes
p value for trend=0.01 of energy, protein,
SFA, MUFA, PUFA,
α-linolenic, long-chain
n–3 PUFA, chole-
sterol , fibre, fruits,
and vegetables, and
hypertension
Jakobsen (2008) F&M, Food record, Non-fatal 374 18 R-TFA: Sex, age, family →
3,686 (no history 30–71 (80% g/d, AMI or RR(0.5 unit)=0.97 history of AMI,
of diabetes or central) R-TFA: fatal IHD (0.91–1.04) education, smoking
IHD) 90% central status, alcohol con-
range:0.6–3.6 sumption, intakes of
protein, SFA, MUFA,
PUFA, cholesterol,
fibre, and weighted
intake of foods con-
taining I-TFA, blood
pressure, and cohort
identification**
g/d(energy- R-TFA: Sex, age, BMI, family →
adjusted), RR(0.5 unit)=1.05 history of AMI,
NA (0.92–1.19) education, physical
activity, smoking
status, alcohol con-
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS
sumption, intakes of
energy, protein, SFA,
MUFA, PUFA, chole-
sterol, fibre, and
289
290
Table 3. continued
weighted intake of
foods containing
I-TFA, blood
pressure, and cohort
identification**
AMI, acute myocardial infarction; BMI, body mass index; CI, confidence interval; E%, percentage of energy intake; F, Female; FFQ, food frequency questionnaire;
I-TFA, industrial trans fatty acids; M, Male; MUFA, monounsaturated fatty acids; NA, not available; PUFA, polyunsaturated fatty acids; Q, quintile; R-TFA,
ruminant trans fatty acids; RR, relative risk; SD, standard deviation; SFA, saturated fatty acids; TFA, trans fatty acids.
† Relative association; in most studies hazard ratio.
‡ → No statistically significant association.
§ ↓ Statistically significant positive association.
# ↑ Statistically significant negative association.
** A variable was included if it changed the beta coefficient for the R-TFA variable 10% or more.
TRANS FATTY ACIDS IN HUMAN NUTRITION
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS 291
4. Follow-up studies
Eight follow-up studies met the selection criteria (Ascherio et al., 1996; Hu
et al., 1997; Hu et al., 1999; Jakobsen et al., 2008; Oh et al., 2005; Oomen
et al., 2001; Pietinen et al., 1997; Willett et al., 1993). The studies by Willett
et al. (1993), Hu et al. (1997), Hu et al. (1999), and Oh et al. (2005) were based
on the same participants and therefore these studies cannot be considered
independent. In the 1993 study by Willett et al., the investigators reported
results on intakes of total TFA and TFA obtained from ruminant and industrial
sources after 8 years follow-up. In the Hu et al. 1997 and 1999, and Oh et al.
2005 studies, the investigators reported results on intake of total TFA after 14,
14, and 20 years follow-up, respectively. For this review, the 1993 paper by
Willett et al. was selected for results on intakes of TFA obtained from ruminant
and industrial sources, and the 2005 Oh et al. paper was selected for results on
intake of total TFA due to the long follow-up. Characteristics of the follow-up
studies included are shown in Table 3.
TFA and risk of fatal IHD (Pietinen et al., 1997) and two studies showed no
significant associations between energy-adjusted intake of total TFA and risk of
fatal IHD (Ascherio et al., 1996; Oomen et al., 2001) (Table 3).
Four studies investigated the associations between intake of TFA obtained
from ruminant sources and risk of IHD (Jakobsen et al., 2008; Oomen et al.,
2001; Pietinen et al., 1997; Willett et al., 1993) (Table 3). One study showed a
significant negative association between energy-adjusted intake of ruminant
TFA and risk of fatal IHD (Pietinen et al., 1997) and three studies showed no
significant associations between energy-adjusted intake of ruminant TFA and
risk of IHD (Jakobsen et al., 2008; Oomen et al., 2001; Willett et al., 1993)
(Table 3). Among the three studies which showed no significant associations,
one study, however, indicated a trend for a negative association (Willett et al.,
1993) and one study indicated a trend for a positive association (Oomen et al.,
2001) (Table 3). Three studies investigated the associations between intake of
TFA obtained from industrial sources and risk of IHD (Oomen et al., 2001;
Pietinen et al., 1997; Willett et al., 1993) (Table 3). Two studies showed
significant positive associations between energy-adjusted intake of industrial
TFA and risk of IHD (Pietineno et al., 1997; Willett et al., 1993) and one study
showed no significant association (Oomen et al., 2001) (Table 3). In the study
by Oomen et al. (2001), however, the results indicated a trend for a positive
association (Table 3).
In the study by Pietinen et al. (1997) energy-adjusted intake of trans 18:1n–9
was significantly positively associated with risk of fatal IHD after control for
dietary and non-dietary IHD risk factors (relative risk, RR, for highest compared
with lowest quintile = 1.37, 95% CI: 1.07–1.75, p value for trend = 0.002).
Saturated fatty acids intake is a potential confounder in analyses of TFA
intake and risk of IHD. Of the six studies (Ascherio et al., 1996; Jakobsen et al.,
2008; Oh et al., 2005; Oomen et al., 2001; Pietinen et al., 1997; Willett et al.,
1993) on TFA intake and risk of IHD, all but that of Ascherio et al. were
controlled for saturated fatty acids intake in multiple analyses (Table 3).
In one study reporting results from multiple analysis, the analysis included
control for hypercholesterolaemia which may be considered as a potential
metabolic effect of intake of TFA (Ascherio et al., 1996) (Table 3).
Whether intake of TFA should be expressed as absolute intake or energy-
adjusted intake depends on the mechanisms behind a potential association
between TFA intake and IHD. In the study by Jakobsen et al. (2008), both
approaches were used (Table 3).
(Table 3). One study was carried out both among women and men (Jakobsen
et al., 2008). In that study there were no overall associations between energy-
adjusted ruminant TFA intake and risk of IHD (Table 3). Among women,
however, an indication of a negative association between energy-adjusted
ruminant TFA intake and risk of IHD was found (RR per 0.5 g intake = 0.77, 95%
CI: 0.55–1.09); no association was found among men (RR per 0.5 g intake = 1.05,
95% CI: 0.94–1.17). The p value for effect modification by sex was 0.52.
In two studies that investigated potential effect modification by age (Jakobsen
et al., 2008; Oh et al., 2005) the associations between TFA intake and risk of
IHD differed by age category. In the study by Oh et al. (2005), a positive
significant association between the percentage of energy intake obtained from
total TFA intake and risk of IHD was found among women less than age 65 years
(RR for highest compared with lowest quintile = 1.50, 95% CI: 1.13–2.00,
p value for trend = 0.01) but not among women aged 65 years or more (RR for
highest compared with lowest quintile = 1.15, 95% CI: 0.80–1.66, p value for
trend = 0.49, p value for effect modification by age = 0.03). In the study by
Jakobsen et al. (2008), an indication of a negative association between energy-
adjusted ruminant TFA intake and risk of IHD was found among women less
than age 60 years (RR per 0.5 g intake, intake, = 0.59, 95% CI: 0.35–1.01);
among women aged 60 years or more the RR per 0.5 g intake was 0.81 (95% CI:
0.57–1.15). The p value for effect modification by age was 0.22. Among men,
less than age 60 years the RR per 0.5 g intake was 0.94 (95% CI: 0.78–1.14) and
among men aged 60 years or more the RR per 0.5 g intake was 1.09 (95% CI:
0.97–1.22). The p value for effect modification by age was 0.17.
Table 4. Summary of the results of the case-control and follow-up studies investigating the association between intake of trans fatty acids
and risk of ischaemic heart disease.
* A bracket indicates that the association was not statistically significant. TFA, trans fatty acid.
TRANS FATTY ACIDS IN HUMAN NUTRITION
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS 295
total TFA, ruminant TFA, and industrial TFA were not significantly associated
with IHD among women who had undiagnosed IHD at baseline. Among men
who had undiagnosed IHD at baseline, there was a significant negative
association between intake of ruminant TFA and IHD; intakes of total and
industrial TFA were not significantly associated with IHD (Bolton-Smith
et al., 1996).
In summary, the studies using direct measure of intake of TFA suggest that
intake of TFA derived from industrial sources is associated with a higher risk
of IHD. There is no observational epidemiological evidence of a harmful effect
on IHD risk of intake of TFA obtained from ruminant sources.
D. Discussion
1. Summary of the study results
One ecologic study (Kromhout et al., 1995), one cross-sectional study (Bolton-
Smith et al., 1996), 19 case-control studies (Aro et al., 1995; Ascherio et al.,
1994; Baylin et al., 2003; Clifton et al., 2004; Colon-Ramos et al., 2006;
Dlouhy et al., 2003; Ghahremanpour et al., 2008; Harris et al., 2007; Lemaitre
et al., 2002; Lemaitre et al., 2006; Lopes et al., 2007; Pedersen et al., 2000;
Roberts et al., 1995; Siguel & Lerman, 1993; Sun et al., 2007b; Sun et al.,
2007c; Thomas & Scott, 1981; Thomas et al., 1983a), and six follow-up studies
(Ascherio et al., 1996; Jakobsen et al., 2008; Oh et al., 2005; Oomen et al.,
2001; Pietinen et al., 1997; Willett et al., 1993) met the selection criteria and
were included in this review of observational epidemiological studies on the
intake of TFA and the risk of IHD. The studies using direct measure of intake
of TFA suggest that intake of TFA obtained from industrial sources is associ-
ated with a higher risk of IHD. There is no observational epidemiological
evidence of a harmful effect on IHD risk of intake of TFA obtained from
ruminant sources. The studies using biomarkers of intakes of TFA suggest that
intake of trans 18:2 is associated with a higher risk of IHD whereas results on
total trans 16:1 and total trans 18:1 are not consistent.
b. Information problems
Measures of TFA intake included both direct measures of dietary TFA intakes
and proportions of TFA in adipose tissue, whole blood, plasma, or erythrocytes
as biomarkers of TFA intakes.
The primary method used for quantification of dietary TFA intakes was food
frequency questionnaires. A potential source of random measurement error
298 TRANS FATTY ACIDS IN HUMAN NUTRITION
of TFA also depends indirectly on the saturated fatty acids intake. Moreover,
by controlling for saturated fatty acids intake, the variation in food sources
contributing to variation in tissue TFA is then restricted to variation in food
sources not directly overlapping food sources contributing to variation in
saturated fatty acids intake which are mostly the same foods. Therefore,
inclusion of fatty acids intake in multiple analyses should be considered
carefully to avoid models for which the results have no clear interpretation.
In a number of the included studies, the analyses included control for
potential metabolic effects of intake of TFA. As these factors (plasma cho-
lesterol concentration and history of diabetes mellitus) may represent steps
in the causal chain between intake of TFA and risk of IHD (i.e. the effect of
intake of TFA on risk of IHD is mediated through the effect of these interme-
diate factors) they should not be treated simply as potential confounding
factors, but instead analysed in more detail taking into account their inter-
mediate nature.
Acknowledgements
The authors collaborate within the project ‘Hepatic and adipose tissue and
functions in the metabolic syndrome’ (HEPADIP, www.hepadip.org), which is
supported by the European Commission as an Integrated Project under the 6th
Framework Program (Contract LSHM-CT-2005-018734) and within the
research programme of the ‘Danish obesity research centre’ (DanORC,
(www.danorc.dk), which is supported by the Danish Council for Strategic
Research (Contract 2101-06-0005).
References
Aro, A, Kardinaal, AF, Salminen, I, Kark, JD, Riemersma, RA, Delgado-Rodriguez, M,
Gomez-Aracena, J, Huttunen, JK, Kohlmeier, L, Martin, BC, Martin-Moreno, JM,
Mazaev, VP, Ringstad, J, Thamm, M, van’t Veer, P and Kok, FJ (1995) Adipose tissue
isomeric trans fatty acids and risk of myocardial infarction in nine countries: the
EURAMIC study. Lancet, 345, 273–278.
Ascherio, A, Hennekens, CH, Buring, JE, Master, C, Stampfer, MJ and Willett, WC (1994)
Trans-fatty acids intake and risk of myocardial infarction. Circulation, 89, 94–101.
Ascherio, A, Rimm, EB, Giovannucci, EL, Spiegelman, D, Stampfer, M and Willett, WC
EPIDEMIOLOGICAL STUDIES ON INTAKE OF TRANS FATTY ACIDS 303
(1996) Dietary fat and risk of coronary heart disease in men: cohort follow up study in
the United States. Br. Med. J., 313, 84–90.
Baylin, A, Kabagambe, EK, Ascherio, A, Spiegelman, D and Campos, H (2003) High 18:2
trans-fatty acids in adipose tissue are associated with increased risk of nonfatal acute
myocardial infarction in Costa Rican adults. J. Nutr., 133, 1186–1191.
Baylin, A, Kabagambe, EK, Siles, X and Campos, H (2002) Adipose tissue biomarkers of
fatty acid intake. Am. J. Clin. Nutr., 76, 750–757.
Baylin, A, Kim, MK, Donovan-Palmer, A, Siles, X, Dougherty, L, Tocco, P and Campos,
H (2005) Fasting whole blood as a biomarker of essential fatty acid intake in epide-
miologic studies: comparison with adipose tissue and plasma. Am. J. Epidemiol., 162,
373–381.
Bolton-Smith, C, Woodward, M, Fenton, S and Brown, CA (1996) Does dietary trans fatty
acid intake relate to the prevalence of coronary heart disease in Scotland? Eur. Heart J.,
17, 837–845.
Chardigny, JM, Destaillats, F, Malpuech-Brugere, C, Moulin, J, Bauman, DE, Lock, AL,
Barbano, DM, Mensink, RP, Bezelgues, JB, Chaumont, P, Combe, N, Cristiani, I, Joffre,
F, German, JB, Dionisi, F, Boirie, Y and Sebedio, JL (2008) Do trans fatty acids from
industrially produced sources and from natural sources have the same effect on cardio-
vascular disease risk factors in healthy subjects? Results of the Trans Fatty Acids
Collaboration (TRANSFACT) study. Am. J. Clin. Nutr., 87, 558–566.
Clifton, PM, Keogh, JB and Noakes, M (2004) Trans fatty acids in adipose tissue and the food
supply are associated with myocardial infarction. J. Nutr., 134, 874–879.
Colon-Ramos, U, Baylin, A and Campos, H (2006) The relation between trans fatty acid
levels and increased risk of myocardial infarction does not hold at lower levels of trans
fatty acids in the Costa Rican food supply. J. Nutr., 136, 2887–2892.
Craig-Schmidt, MC (2006) World-wide consumption of trans fatty acids. Atheroscler.
Suppl., 7, 1–4.
Delmonte, P and Rader, JI (2007) Evaluation of gas chromatographic methods for the
determination of trans fat. Analyt. Bioanalyt. Chem., 389, 77–85.
Dlouhy, P, Tvrzicka, E, Stankova, B, Vecka, M, Zak, A, Straka, Z, Fanta, J, Pachl, J,
Kubisova, D, Rambouskova, J, Bilkova, D and Andel, M (2003) Higher content of 18:1
trans fatty acids in subcutaneous fat of persons with coronarographically documented
atherosclerosis of the coronary arteries. Ann. Nutr. Metab, 47, 302–305.
Fagt, S and Trolle, E (2001) 1. Forsyningen af fødevarer 1955–1999. Udviklingen i
danskernes kost — forbrug, indkøb og vaner (Food statistics 1955–1999. Development
in dietary intake in the Danish population — consumption, purchase and habits). (In
Danish). National Food Agency, Søborg, Denmark.
Fritsche, J, Steinhart, H, Kardalinos, V and Klose, G (1998) Contents of trans-fatty acids in
human substernal adipose tissue and plasma lipids: relation to angiographically docu-
mented coronary heart disease. Eur. J. Med. Res., 3, 401–406.
Ghahremanpour, F, Firoozrai, M, Darabi, M, Zavarei, A and Mohebbi, A (2008) Adipose
tissue trans fatty acids and risk of coronary artery disease: a case-control study. Ann. Nutr.
Metab., 52, 24–28.
Harnack, L, Lee, S, Schakel, SF, Duval, S, Luepker, RV and Arnett, DK (2003) Trends in the
trans-fatty acid composition of the diet in a metropolitan area: the Minnesota Heart
Survey. J. Am. Diet. Assoc., 103, 1160–1166.
Harris, WS, Reid, KJ, Sands, SA and Spertus, JA (2007) Blood omega-3 and trans fatty acids
in middle-aged acute coronary syndrome patients. Am. J. Cardiol., 99, 154–158.
Heckers, H, Korner, M, Tuschen, TW and Melcher, FW (1977) Occurrence of individual
trans-isomeric fatty acids in human myocardium, jejunum and aorta in relation to
different degrees of atherosclerosis. Atherosclerosis, 28, 389–398.
Hodgson, JM, Wahlqvist, ML, Boxall, JA and Balazs, ND (1996) Platelet trans fatty acids
304 TRANS FATTY ACIDS IN HUMAN NUTRITION
Oomen, CM, Ocke, MC, Feskens, EJM, van Erp-Baart, MJ, Kok, FJ and Kromhout, D
(2001) Association between trans fatty acid intake and 10-year risk of coronary heart
disease in the Zutphen Elderly Study: a prospective population-based study. Lancet,
357, 746–751.
Pedersen, JI, Ringstad, J, Almendingen, K, Haugen, TS, Stensvold, I and Thelle, DS (2000)
Adipose tissue fatty acids and risk of myocardial infarction – a case-control study. Eur.
J. Clin. Nutr., 54, 618–625.
Pietinen, P, Ascherio, A, Korhonen, P, Hartman, AM, Willett, WC, Albanes, D and Virtamo,
J (1997) Intake of fatty acids and risk of coronary heart disease in a cohort of Finnish men.
The Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study. Am. J. Epidemiol., 145,
876–887.
Precht, D and Molkentin, J (1995) Trans fatty acids: implications for health, analytical
methods, incidence in edible fats and intake (a review). Nahrung, 39, 343–374.
Precht, D and Molkentin, J (1996) Rapid analysis of the isomers of trans-octadecenoic acid
in milk fat. Int. Dairy J., 6, 791–809.
Roberts, TL, Wood, DA, Riemersma, RA, Gallagher, PJ and Lampe, FC (1995) Trans
isomers of oleic and linoleic acids in adipose tissue and sudden cardiac death. Lancet, 345,
278–282.
Shai, I, Rimm, EB, Hankinson, SE, Curhan, G, Manson, JE, Rifai, N, Stampfer, MJ and Ma,
J (2004) Multivariate assessment of lipid parameters as predictors of coronary heart
disease among postmenopausal women: potential implications for clinical guidelines.
Circulation, 110, 2824–2830.
Siguel, EN and Lerman, RH (1993) Trans-fatty acid patterns in patients with angiographically
documented coronary artery disease. Am. J. Cardiol., 71, 916–920.
Singh, RB, Niaz, MA, Ghosh, S, Beegom, R, Rastogi, V, Sharma, JP and Dube, GK (1996)
Association of trans fatty acids (vegetable ghee) and clarified butter (Indian ghee) intake
with higher risk of coronary artery disease in rural and urban populations with low fat
consumption. Int. J. Cardiol., 56, 289–298.
Sun, Q, Ma, J, Campos, H, Hankinson, SE and Hu, FB (2007a) Comparison between plasma
and erythrocyte fatty acid content as biomarkers of fatty acid intake in US women. Am.
J. Clin. Nutr., 86, 74–81.
Sun, Q, Ma, J, Campos, H, Hankinson, SE, Manson, JE, Stampfer, MJ, Rexrode, KM, Willett,
WC and Hu, FB (2007b) A prospective study of trans fatty acids in erythrocytes and risk
of coronary heart disease. Circulation, 115, 1858–1865.
Sun, Q, Ma, J, Campos, H and Hu, FB (2007c) Plasma and erythrocyte biomarkers of dairy
fat intake and risk of ischemic heart disease. Am. J. Clin. Nutr., 86, 929–937.
Thomas, LH and Scott, RG (1981) Ischaemic heart disease and the proportions of hydrogen-
ated fat and ruminant-animal fat in adipose tissue at post-mortem examination: a
case-control study. J. Epidemiol. Community Health, 35, 251–255.
Thomas, LH, Winter, JA and Scott, RG (1983a) Concentration of 18:1 and 16:1 transunsaturated
fatty acids in the adipose body tissue of decedents dying of ischaemic heart disease
compared with controls: analysis by gas liquid chromatography. J. Epidemiol. Commu-
nity Health, 37, 16–21.
Thomas, LH, Winter, JA and Scott, RG (1983b) Concentration of transunsaturated fatty acids
in the adipose body tissue of decedents dying of ischaemic heart disease compared with
controls. J. Epidemiol. Community Health, 37, 22–24.
van de Vijver, LP, van Poppel, G, van Houwelingen, A, Kruyssen, DA and Hornstra, G (1996)
Trans unsaturated fatty acids in plasma phospholipids and coronary heart disease: a case-
control study. Atherosclerosis, 126, 155–161.
van Poppel, G (1998) Intake of trans fatty acids in western Europe: the TRANSFAIR study.
Lancet, 351, 1099.
van Toll, A, Zock, PL, van Gent, T, Scheek, LM and Katan, MB (1995) Dietary trans fatty
acids increase serum cholesterylester transfer protein activity in man. Atherosclerosis,
115, 129–134.
Willett, WC, Stampfer, MJ, Manson, JE, Colditz, GA, Speizer, FE, Rosner, BA, Sampson,
LA and Hennekens, CH (1993) Intake of trans fatty acids and risk of coronary heart
disease among women. Lancet, 341, 581–585.
CHAPTER 11
Dietary Trans Fatty Acids and Cardiovascular
Disease Risk
1
Clermont Université, UFR Médecine, UMR 1019 Nutrition Humaine,
Clermont-Ferrand, France
2
INRA, UMR 1019 Nutrition Humaine, Saint Genès Champanelle, France
3
Maastricht University, Nutrition and Toxicology Research Institute Maastricht,
Department of Human Biology, Maastricht, The Netherlands
In this chapter, the clinical evidence linking the consumption of the different
trans fatty acids (TFA) to cardiovascular disease risk factors is discussed. The
main results from epidemiological studies on the relationship between TFA
intake and cardiovascular risk are summarized in the first section (for detailed
information see Chapter 10).
between the intakes of TFA and the risk of CHD in a cohort of 21,930 smoking
men aged 50 to 69 years. At the start of the study, subjects were free of
diagnosed CVD. After 6.1 years of follow-up, 1399 major coronary events and
635 coronary deaths were reported. After adjustment for several potential
confounding variables, a significant positive relationship was observed be-
tween the intake of TFA and the risk of coronary death. For men in the top
quintile of TFA intake (median = 6.2 g/d), the relative risk of coronary death
was 1.39 as compared with men in the lowest quintile of intake (median = 1.8 g/
d). For all major coronary events, associations were in the same direction, but
in general somewhat less pronounced.
In the Zutphen Elderly Study from the Netherlands, 667 men (aged 64–84
years) and free of coronary heart disease at baseline, were studied prospec-
tively. After 10 years of follow-up, there were 98 cases of fatal or non-fatal
CHD. After multivariate analyses, a positive relationship was found between
TFA intake at baseline with the 10-year risk of CHD. It was estimated that a
difference of 2% of energy in TFA intake at baseline increased the risk for CHD
by 28% (Oomen et al., 2001).
Oh and co-workers (Oh et al., 2005) examined the relationship between TFA
intake with the risk of CHD among 78,778 US women, who were initially free
of cardiovascular disease and diabetes. During 20 years of follow-up, 1766
incident CHD cases were reported. TFA intake was positively associated with
CHD risk. The multivariate relative risk for the highest vs the lowest quintile of
intake was 1.33. Median intakes of TFA intakes in these two quintiles were
respectively 2.8 and 1.3% of energy. This association was particularly evident
among women younger than age 65 years.
No such associations have been found between the intakes of TFA from
ruminant sources with CVD risk (Oomen et al., 2001; Willett et al., 1993). In
one study, even a negative relationship was reported (Pietinen et al., 1997).
Also, a recent study (Jakobsen et al., 2008) did not suggest that the intake of
ruminant TFA is associated with a higher risk of CVD. For women, even a
negative relationship was reported. It should be noted however that the intake
of TFA from animal origin is low, which may have masked any relationship. In
fact, based on the data of prospective cohort studies, Weggemans and co-
workers (2004) suggested that in prospective cohort studies no differences in
the risk of CHD is evident between total, ruminant and industrial TFA when the
intake is below 2.5 g/d.
CRP, SAA, fibrinogen (acute phase proteins) Liver, adipose tissue, macrophages, smooth CRP associated with production of
muscle cells, endothelial cells inflammatory cytokines, chemokines, tissue
factor expression, chemotaxis of monocytes.
Downregulation of eNOS and
prostacyclin. Predisposition to a state of
hypercoagulability (increased PAI-1, and
decreased tPA).
Cytokines (IL-1β, IL-6, TNF, IL-7, TNF-α, Endothelial cells, macrophages, adipose tissue Pro-atherogenic and augment monocyte-
IL-18 etc) endothelial adhesion. IL-7 activates
monocytes and enhances production
of inflammatory cytokines. IL-18, a strong
independent predictor of mortality in patients
with coronary artery disease.
Chemokines (MCP-1, IL-8) Endothelial cells, macrophages Stimulate chemotaxis.
Adhesion molecules (ICAM, VCAM, Endothelial cells Promote monocyte–endothelial adhesion.
E-selectin, P-selectin)
PAI-1 (inhibitor of fibrinolysis) Endothelial cells, adipose tissue, macrophages Reduced plasma fibrinolysis.
Promotes atherothrombosis.
ICAM, intercellular adhesion molecule; MCP-1, monocyte chemoattractant protein-1; PAI-1, plasminogen activator inhibitor-1; SAA, serum amyloid A; VCAM,
vascular cell adhesion molecule.
DIETARY TRANS FATTY ACIDS AND CARDIOVASCULAR DISEASE RISK
311
312 TRANS FATTY ACIDS IN HUMAN NUTRITION
molecule that mediates the initial rolling of inflammatory cells along endothe-
lial cells and which is preferentially expressed on the endothelium overlaying
the atherosclerotic plaques. Intracellular adhesion molecules-1 (ICAM-1) and
vascular adhesion molecule-1 (VCAM-1) are thought to regulate attachment
and transendothelial migration of leukocytes (Ross, 1999, Basu et al., 2006).
Both macrophages and endothelial cells produce ICAM-1 in response to
inflammatory cytokines such as interleukin-1 (IL-1), and TNF-α (Steffens &
Mach, 2004). Biomarkers of endothelial dysfunction found to be higher in the
quintile of subjects with the highest consumption of trans fatty acids (3.7 g/d)
compared to those in the lowest quintile (1.5 g/d) were: E-selectin 20%, and
soluble cell adhesion molecules (sICAM-1 and sVCAM-1) both 10%. E-selectin,
sICAM-1 and sVCAM-1 are adhesion proteins whose concentrations are
increased in heart diseases and the early stage of atherosclerosis. TFA intake
was positively related to E-selectin (P = 0.003), sICAM-1 (P = 0.007) and
sVCAM-1 (P = 0.001) in linear regression models after controlling for age,
BMI, physical activity, smoking status, alcohol consumption, intake of mono-
, poly- and saturated fatty acids, and postmenopausal hormone therapy. They
concluded that trans fatty acids adversely affect endothelial function, which
might explain in part the association of these fatty acids with risk of cardiovas-
cular disease and a higher intake of trans fatty acids could favour inflammation
and adversely affect endothelial function (Lopez-Garcia et al., 2005).
the Nurses Health Study which, over 14 years, prospectively followed 84 204
healthy women aged 34–59 years in 1980. The data suggested that total lipid
intake, compared with equivalent energy intake from carbohydrates, was not
associated with risk of type 2 diabetes in women (Salmeron et al., 2001). By
contrast, TFA increased this risk whereas PUFA reduced it. Indeed, for a 5%
increase in energy from PUFA, the relative risk was 0.63 (0.53, 0.76; P < 0.0001)
and for a 2% increase in energy from TFA the relative risk was 1.39 (1.15, 1.67;
P < 0.001). In addition, substituting non-hydrogenated PUFA for TFA sub-
stantially reduced the risk of type 2 diabetes. The authors estimated that
replacing 2% of energy from TFA isoenergetically with PUFA would lead to a
40% lower risk. Subgroup analyses revealed that effects were primarily ob-
served in obese women, possibly due to the fact that these women were already
more insulin resistant compared with non-obese women. Regarding the other
fatty acids, no relationships were observed between incidence of type 2 diabe-
tes and intakes of saturated fatty acids or MUFA (Salmeron et al., 2001). The
Iowa Women’s Health Study showed that diabetes incidence was negatively
associated with dietary PUFA, vegetable fat, and trans fatty acids and posi-
tively associated with omega-3 fatty acids (Meyer et al., 2001). Relative risks
for diabetes among quintiles of intake of TFA were 1.00, 1.01, 0.93, 1.09, 0.86
and 0.88 (P = 0.03) (Meyer et al., 2001). In contrast, the intakes of oleic acid,
trans fatty acids, long-chain n–3 fat, and linolenic acid were not associated with
risk of diabetes after multivariate adjustment in men included in the Health
Professionals Follow-up Study (van Dam et al., 2002). The authors explained
that the differences between these three studies came from the different
amounts of TFA consumed by the subjects. Indeed the median intake at
baseline was 1.2% of energy in their study vs 2.2% in the Nurses Health Study,
suggesting that adverse effects of TFA on insulin have been observed at high
levels of TFA (Christiansen et al., 1997). In addition, the sources of TFA were
not examined, so no conclusions could be brought regarding the respective
effects of dairy and hydrogenated fats. However, differences in the relative
abundance of each trans MUFA isomer in food products could probably lead
to different metabolic effects.
Short term intervention studies have not been very conclusive either. We
reported in Wistar rats that an 8 week diet enriched (4% of energy intake) in
either dairy, industrial, or control MUFA did not alter insulin and glucose
responses to an intraperitoneal glucose tolerance test (1 g/kg). Furthermore, we
found that in C2C12 myotubes, vaccenic (dairy TFA) and elaidic (industrial
TFA) acids did not modify insulin sensitivity compared with oleic acid. In
addition, four to five week dietary challenges have shown that dietary trans
MUFA from both sources (industrial vs ruminant) do not impair insulin
sensitivity in healthy subjects, even if there is an alteration of the lipoprotein
metabolism (Louheranta et al., 1999; Lovejoy et al., 2001; Tholstrup et al.,
2006). Therefore, increasing evidence suggests that dietary trans MUFA of
316 TRANS FATTY ACIDS IN HUMAN NUTRITION
dairy or industrial origin at nutritional levels (< 4% energy intake) may not
impair insulin sensitivity in healthy subjects. However, this question remains to
be elucidated in subjects at risk for diabetes, because Christiansen and co-
workers (1997) observed the deleterious effect of a diet high in industrial trans
MUFA in obese and insulin resistant subjects.
C. Conclusion
To conclude, while trans MUFA from hydrogenated oils have been clearly
implicated in the pathogenesis of cardiovascular diseases, those from dairy
products may be less deleterious. Furthermore, dietary intakes of trans MUFA
from dairy origin are far from being high enough to constitute a serious threat
for public health. Therefore, regular moderate consumption of dairy fats could
be tolerated with respect to cardiovascular risks. Regarding the role of trans
MUFA from both origins on the incidence of type 2 diabetes, studies carried
out on the topic are still controversial. Mechanistic and intervention studies
showed no significant effect of trans MUFA on muscle insulin resistance,
suggesting that these fatty acids may not increase the risk for diabetes.
However, further intervention studies are required to examine the impact of
trans MUFA in individuals at risk for metabolic syndrome and diabetes.
References
Almendingen, K, Jordal, O, Kierulf, P, Sandstad, B and Pedersen, JI (1995) Effects of
partially hydrogenated fish oil, partially hydrogenated soybean oil, and butter on serum
lipoproteins and Lp[a] in men. J. Lipid Res., 36, 1370–1384.
Almendingen, K, Seljeflot, I, Sandstad, B and Pedersen, JI (1996) Effects of partially
hydrogenated fish oil, partially hydrogenated soybean oil, and butter on hemostatic
variables in men. Arterioscler. Thromb. Vasc. Biol., 16, 375–380.
Armstrong, RA, Chardigny, JM, Beaufrere, B, Bretillon, L, Vermunt, SH, Mensink, RP,
Macvean, A, Elton, RA, Sebedio, JL and Riemersma, RA (2000) No effect of dietary
trans isomers of alpha-linolenic acid on platelet aggregation and haemostatic factors in
european healthy men. The TRANSLinE study. Thromb. Res., 100, 133–141.
Ascherio, A, Rimm, EB, Giovannucci, EL, Spiegelman, D, Stampfer, M and Willett, WC
(1996) Dietary fat and risk of coronary heart disease in men: cohort follow up study in
the United States. Brit. J. Med., 313, 84–90.
Baer, DJ, Judd, JT, Clevidence, BA and Tracy, RP (2004) Dietary fatty acids affect plasma
markers of inflammation in healthy men fed controlled diets: a randomized crossover
study. Am. J. Clin. Nutr., 79, 969–973.
Basu, A, Devaraj, S and Jialal, I (2006) Dietary factors that promote or retard inflammation.
Arterioscler. Thromb. Vasc. Biol., 26, 995–1001.
Chardigny, JM, Destaillats, F, Malpuech-Brugere, C, Moulin, J, Bauman, DE, Lock, AL,
Barbano, DM, Mensink, RP, Bezelgues, JB, Chaumont, P, Combe, N, Cristiani, I, Joffre,
F, German, JB, Dionisi, F, Boirie, Y and Sebedio, JL (2008) Do trans fatty acids from
industrially produced sources and from natural sources have the same effect on cardio-
vascular disease risk factors in healthy subjects? Results of the Trans Fatty Acids
Collaboration (TRANSFACT) study. Am. J. Clin. Nutr., 87, 558–566.
DIETARY TRANS FATTY ACIDS AND CARDIOVASCULAR DISEASE RISK 317
Chavez, JA, Knotts, TAe , Wang, LP, Li, G, Dobrowsky, RT, Florant, GL and Summers,
SA (2003) A role for ceramide, but not diacylglycerol, in the antagonism of insulin signal
transduction by saturated fatty acids. J. Biol. Chem., 278, 10297–10303.
Christiansen, E, Schnider, S, Palmvig, B, Tauber-Lassen, E and Pedersen, O (1997) Intake
of a diet high in trans monounsaturated fatty acids or saturated fatty acids. Effects on
postprandial insulinemia and glycemia in obese patients with NIDDM. Diabetes Care,
20, 881–887.
Han, SN, Leka, LS, Lichtenstein, AH, Ausman, LM, Schaefer, EJ and Meydani, SN (2002)
Effect of hydrogenated and saturated, relative to polyunsaturated, fat on immune and
inflammatory responses of adults with moderate hypercholesterolemia. J. Lipid Res., 43,
445–452.
Hu, FB, Manson, JE, Stampfer, MJ, Colditz, G, Liu, S, Solomon, CG and Willett, WC (2001)
Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. N. Engl. J. Med., 345,
790–797.
Jakobsen, MU, Overvad, K, Dyerberg, J and Heitmann, BL (2008) Intake of ruminant trans
fatty acids and risk of coronary heart disease. Int. J. Epidemiol., 37, 173–182.
Lennie, TA, Chung, ML, Habash, DL and Moser, DK (2005) Dietary fat intake and
proinflammatory cytokine levels in patients with heart failure. J. Card. Fail., 11, 613–
618.
Libby, P (2002) Inflammation in atherosclerosis. Nature, 420, 868–874.
Lopez-Garcia, E, Schulze, MB, Meigs, JB, Manson, JE, Rifai, N, Stampfer, MJ, Willett, WC
and Hu, FB (2005) Consumption of trans fatty acids is related to plasma biomarkers of
inflammation and endothelial dysfunction. J. Nutr., 135, 562–566.
Louheranta, AM, Turpeinen, AK, Vidgren, HM, Schwab, US and Uusitupa, MI (1999) A
high-trans fatty acid diet and insulin sensitivity in young healthy women. Metabolism,
48, 870–875.
Lovejoy, JC, Champagne, CM, Smith, SR, DeLany, JP, Bray, GA, Lefevre, M, Denkins, YM
and Rood, JC (2001) Relationship of dietary fat and serum cholesterol ester and
phospholipid fatty acids to markers of insulin resistance in men and women with a range
of glucose tolerance. Metabolism, 50, 86–92.
Madge, LA and Pober, JS (2001) TNF signaling in vascular endothelial cells. Exp. Mol.
Pathol., 70, 317–325.
Mensink, RP, Zock, PL, Katan, MB and Hornstra, G (1992) Effect of dietary cis and trans
fatty acids on serum lipoprotein[a] levels in humans. J. Lipid Res., 33, 1493–1501.
Mensink, RP, Zock, PL, Kester, AD and Katan, MB (2003) Effects of dietary fatty acids and
carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and
apolipo-proteins: a meta-analysis of 60 controlled trials. Am. J. Clin. Nutr., 77, 1146–1155.
Meyer, KA, Kushi, LH, Jacobs, DR, Jr and Folsom, AR (2001) Dietary fat and incidence of
type 2 diabetes in older Iowa women. Diabetes Care, 24, 1528–1535.
Motard-Belanger, A, Charest, A, Grenier, G, Paquin, P, Chouinard, Y, Lemieux, S, Couture,
P and Lamarche, B (2008) Study of the effect of trans fatty acids from ruminants on blood
lipids and other risk factors for cardiovascular disease. Am. J. Clin. Nutr., 87, 593–599.
Mozaffarian, D, Pischon, T, Hankinson, SE, Rifai, N, Joshipura, K, Willett, WC and Rimm,
EB (2004) Dietary intake of trans fatty acids and systemic inflammation in women. Am.
J. Clin. Nutr., 79, 606–612.
Mutanen, M and Aro, A (1997) Coagulation and fibrinolysis factors in healthy subjects
consuming high stearic or trans fatty acid diets. Thromb. Haemost., 77, 99–104.
Ndumele, CE, Pradhan, AD and Ridker, PM (2006) Interrelationships between inflammation,
C-reactive protein, and insulin resistance. J. Cardiometab. Syndr., 1, 190–196.
Nestel, P, Noakes, M, Belling, B, McArthur, R, Clifton, P, Janus, E and Abbey, M (1992)
Plasma lipoprotein lipid and Lp[a] changes with substitution of elaidic acid for oleic acid
in the diet. J. Lipid Res., 33, 1029–1036.
318 TRANS FATTY ACIDS IN HUMAN NUTRITION
Oh, K, Hu, FB, Manson, JE, Stampfer, MJ and Willett, WC (2005) Dietary fat intake and risk
of coronary heart disease in women: 20 years of follow-up of the nurses’ health study. Am
J Epidemiol, 161, 672–679.
Oomen, CM, Ocke, MC, Feskens, EJ, van Erp-Baart, MA, Kok, FJ and Kromhout, D (2001)
Association between trans fatty acid intake and 10-year risk of coronary heart disease in
the Zutphen Elderly Study: a prospective population-based study. Lancet 357, 746–751.
Parker, DR, Weiss, ST, Troisi, R, Cassano, PA, Vokonas, PS and Landsberg, L (1993)
Relationship of dietary saturated fatty acids and body habitus to serum insulin concen-
trations: the Normative Aging Study. Am. J. Clin. Nutr. 58, 129–136.
Pietinen, P, Ascherio, A, Korhonen, P, Hartman, AM, Willett, WC, Albanes, D and Virtamo,
J (1997) Intake of fatty acids and risk of coronary heart disease in a cohort of Finnish men.
The Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study. Am. J. Epidemiol., 145,
876–887.
Popa, C, Netea, MG, van Riel, PL, van der Meer, JW and Stalenhoef, AF (2007) The role of
TNF-alpha in chronic inflammatory conditions, intermediary metabolism, and cardio-
vascular risk. J. Lipid Res., 48, 751–762.
Rauchhaus, M, Doehner, W, Francis, DP, Davos, C, Kemp, M, Liebenthal, C, Niebauer, J,
Hooper, J, Volk, HD, Coats, AJS, Stefan, DM, Anker, D (2000) Plasma cytokine
parameters and mortality in patients with chronic heart failure. Circulation, 102, 3060–
3067.
Ross, R (1999) Atherosclerosis is an inflammatory disease. Am. Heart J., 138, S419–420.
Ridker, PM (2007) C-reactive protein and the prediction of cardiovascular events among
those at intermediate risk: moving an inflammatory hypothesis toward consensus. J. Am.
Coll. Cardiol., 49, 2129–2138.
Salmeron, J, Hu, FB, Manson, JE, Stampfer, MJ, Colditz, GA, Rimm, EB and Willett, WC
(2001) Dietary fat intake and risk of type 2 diabetes in women. Am. J. Clin. Nutr., 73,
1019–1026.
Sanders, TA, de Grassi, T, Miller, GJ and Morrissey, JH (2000) Influence of fatty acid chain
length and cis/trans isomerization on postprandial lipemia and factor VII in healthy
subjects (postprandial lipids and factor VII). Atherosclerosis, 149, 413–420.
Steffens, S and Mach, F (2004) Inflammation and atherosclerosis. Herz, 29, 741–748.
Tholstrup, T, Miller, GJ, Bysted, A and Sandstrom, B (2003) Effect of individual dietary fatty
acids on postprandial activation of blood coagulation factor VII and fibrinolysis in
healthy young men. Am. J. Clin. Nutr., 77, 1125–1132.
Tholstrup, T, Raff, M, Basu, S, Nonboe, P, Sejrsen, K and Straarup, EM (2006) Effects of
butter high in ruminant trans and monounsaturated fatty acids on lipoproteins, incorpo-
ration of fatty acids into lipid classes, plasma C-reactive protein, oxidative stress,
hemostatic variables, and insulin in healthy young men. Am. J. Clin. Nutr., 83, 237–243.
van Dam, RM, Willett, WC, Rimm, EB, Stampfer, MJ and Hu, FB (2002) Dietary fat and meat
intake in relation to risk of type 2 diabetes in men. Diabetes Care, 25, 417–424.
Vermunt, SH, Beaufrere, B, Riemersma, RA, Sebedio, JL, Chardigny, JM, Mensink, RP and
TransLinE Investigators (2001) Dietary trans alpha-linolenic acid from deodorised
rapeseed oil and plasma lipids and lipoproteins in healthy men: the TransLinE Study. Br.
J. Nutr., 85, 387–392.
Weggemans, RM, Rudrum, M and Trautwein, EA (2004) Intake of ruminant versus industrial
trans fatty acids and risk of coronary heart disease — what is the evidence? Eur. J. Lipid
Sci. Technol.,106, 390–397.
Willett, WC, Stampfer, MJ, Manson, JE, Colditz, GA, Speizer, FE, Rosner, BA, Sampson,
LA and Hennekens, CH (1993) Intake of trans fatty acids and risk of coronary heart
disease among women. Lancet, 341, 581–585.
Yu, H and Rifai, N (2000) High-sensitivity C-reactive protein and atherosclerosis: from
theory to therapy. Clin. Biochem., 33, 601–610.
CHAPTER 12
Dietary trans fatty acids: from the mother’s
diet to the infant
1
UMR 1019 INRA Université Clermont I, Clermont-Ferrand, France ; CRNH
Auvergne, Clermont-Ferrand, France.
2
ITERG, Département de Nutrition, Bordeaux, France.
A. Introduction
The occurrence of trans fatty acids (TFA) in human tissues depends on dietary
intake, as humans do not synthesize these fatty acid isomers. It is now well known
that TFA are provided by three different dietary sources : the ruminant fats (milk
and dairy products, meat, tallow) (Wolff et al., 1998) partially hydrogenated
vegetable oils (margarines produced with partially hydrogenated vegetable oils,
shortenings, bakeries and chips) (Craig-Schmidt, 1992) and to a lesser extent,
refined vegetable oils (Wolff, 1993). In the latter, TFA are mainly PUFA isomers,
whereas the other sources contain mainly trans monoenes.
Several in vivo animal studies (Anderson et al., 1975; Privett et al., 1967;
Berdeaux et al., 1998) and in vitro experiments in animal tissues (Mahfouz
et al., 1984; Cook and Emken, 1990; Berdeaux et al., 1998; Blond et al., 1995),
as well as human fibroblasts studies (Rosenthal and Doloresco, 1984), have
demonstrated that TFA impair the microsomal desaturation and chain elonga-
tion of linoleic acid and α-linolenic acid to long-chain polyunsaturated fatty
acids. Among all trans isomers, the 9t12t-18:2 isomer exerts the strongest
impairment of Δ6 desaturase activity (Anderson et al., 1975; Shimp et al.,
1982; Hwang and Kinsella, 1979; Brenner and Peluffo, 1969). In adults, the
inhibitory effect of TFA does not cause serious undesirable health hazards
when dietary linoleic acid is sufficient (Zevenbergen et al., 1988; Zevenbergen
and Haddeman, 1989). However, the infant and the foetus are sensitive to
factors that can interfere with the synthesis of long-chain polyunsaturated fatty
acids, due to an extensive need for arachidonic and docosahexaenoic acids for
the development of the neurological structures of the brain and retina (Innis
et al., 1999; Clandinin, 1999; Carlson and Neuringer, 1999) and for the
synthesis of eicosanoids (Sellmayer & Koletzko, 1999).
319
320 TRANS FATTY ACIDS IN HUMAN NUTRITION
Table 1. Concentration of trans 18:1 acid isomers (% of total fatty acids) in human milk
in various countries (adapted from Mosley et al., 2005).
Table 2. Profile of trans-18:1 isomers (% of total trans-18 acid isomers) in breast milk
analysed in the Aquitaine study (adapted from Combe et al., 2005). SD, standard
deviation.
Clearly, it can be stated that TFA content in human milk depends on dietary
habits, with a higher content found in countries were consumption of TFA is
greatest. The trans 18:1 isomeric profile could also differ according to dietary
habits, such as the major source of TFA (partially hydrogenated vegetable oils
versus ruminant fat) as observed in the Aquitaine study (Combe et al., 1995).
In 1997, the mean trans 18:1 content of 27 mature breast milk samples was
1.78% of total fatty acids; dietary records of correspondent mothers showed
that ruminant fat was the main source of trans 18:1 (66%) compared with
(33%). Complete analysis of the trans positional 18:1 isomer profile of 11 milk
samples indicated a high proportion of Δ11 trans 18:1, i.e. 38% of total trans
18:1 isomers, the most prevalent isomer derived from ruminant fat (Table 2)
(Combe et al., 1995).
Recent data have shown that the TFA content of French human milk is
significantly lower in 2007 than in 1997 (Vaysse et al., 2008). In 2007, the
average trans 18:1 content of 142 mature milk samples collected from eight
human milk banks was 0.96 +/– 0.30% of total fatty acids versus 1.78 +/–
0.86% in 1997. This decreasing level of TFA observed in 2007 is probably
explained by a parallel decrease in level of TFA in margarine composition
(< 1% TFA) during this last decade.
available in France and in Canada. The range of trans 18:1 content was 0.6 to
3.1% in Canadian formulas, with a higher content in liquid items when
compared to powdered ones, and 0.2 to 4.3% in French formulas (Chardigny
et al., 1996). However, current data are lacking and the TFA content has
probably decreased since the 1990s, as a result of the abundant literature now
available describing adverse effects.
1. Placental transfer
In the early 1980s, it was demonstrated that [1-14C]-elaidic acid crosses the
placental barrier in rats (Moore & Dhopeshwarkar, 1980). However, contradic-
tory data were obtained in piglets from mothers fed a high trans diet (Pettersen
& Opstvedt, 1989). In humans, Johnston and co-workers (Johnston et al.,
1958) first reported that the transfer was limited. More recently, Koletzko and
Müller (1990) reported that cord blood in term infants presented the same TFA
content than maternal plasma lipids. Similar data were obtained in the Nether-
lands, with a direct correlation between trans 18:1 in maternal plasma and
foetal tissues (Van Houwelingen & Hornstra, 1994). The ILSI expert panel
further concluded that: “TFA are transferred by the placenta to the foetus and
incorporated into foetal tissues” (Carlson et al., 1997). TFA have been reported
to represent 1–2% of total fatty acids in blood lipid fractions in preterm and
term newborns as well as 1–12 month infants (Koletzko, 1992; Decsi &
Koletzko, 1994). The transfer is probably limited as illustrated by the low
content compared to TFA level in the maternal diet (Pettersen & Opstvedt,
1989). Comparison of the TFA content in maternal and umbilical plasma from
French pregnant women (Boue et al., 2001) was in accordance with this point
of view. TFA levels were significantly higher (p = 0.001) in maternal (0.9% of
total fatty acids) than in umbilical plasma total lipids (0.6%). Moreover, the
trans isomer pattern in cord plasma lipids was different from that of the mother
(Figure 1). The trans 18:1 acid level in maternal plasma (0.64% of total fatty
acids) exceeded significantly (p < 0.001) that found in umbilical plasma
(0.24%), while trans 18:2 acid percentage was significantly higher (p < 0.001)
in umbilical (0.30%) than in maternal plasma (0.16%). More precisely, the
9t12c-18:2 isomer and the sum 9c13t+9t12t-18:2 were significantly higher
( p < 0.001) in umbilical (0.15% and 0.08% of total fatty acids respectively)
than in maternal plasma (0.05%), whereas no difference was observed between
these two compartments with respect to 9c12t-18:2 isomer level (0.06% of total
fatty acids in maternal plasma and 0.07% in umbilical plasma, respectively).
DIETARY TFA: FROM THE MOTHER’S DIET TO THE INFANT 323
1.2
1.1 Maternal plasma (n=90)
1.0 Umbilical plasma (n=77)
0.9 *
% of total fatty acids
0.8
0.7
*
0.6
0.5
0.4
0.3 *
0.2
*
0.1 *
0.0
TFA trans trans trans 9c13t 9c12t 9t12c
16:1 18:1 18:2 + 9t12t 18:2 18:2
*: p<0.001 18:2
Figure 1. Trans fatty acid composition (% total fatty acids) of maternal and umbilical plasma total
lipids (adapated from Boué et al., 2001).
The prevalence of trans 18:2 isomers and more precisely of 9t12c-18:2 isomer
in umbilical plasma may be explained either by a preferential transfer of this
isomer across the placenta, as a consequence of sequestration process, or by an
accumulation in cord plasma linked to poor metabolism of this isomer by the
foetus compared with the other trans 18:2 isomers (Combe et al., 1997).
Similarly, an inverse correlation has been reported between TFA and birth
weight in term born babies (Decsi & Koletzko, 1995), which may be correlated
with metabolism of essential fatty acids (see below). Thus, it has been sug-
gested that exposure to high levels of TFA during pregnancy may impair the
growth of the foetus.
However, the recent Aquitaine study failed to demonstrate any correlation
(Boue et al., 2000) between either newborn weight or head circumference
and the TFA level of maternal adipose tissue, as an indicator of the usual
TFA intake level. This intake was assessed at 3.1 g/day as mean value,
ranging from 0.9 to 6.5 g/day (or 0.5 to 2.5% of total energy intake). More-
over, no relationship was observed between the TFA present in maternal
plasma and either newborn weight or head circumference, even after multi-
factorial corrections.
G. Conclusion
High intakes of TFA have been reported to have deleterious effects on infant
development. This could be related to interference with the metabolism of
essential fatty acids. However, under nutritional conditions, the impact seems
to be limited, except in the case of CLA supplementation. As a result of the
decreased use of hydrogenated fats in food products and the consequent fall in
the dietary intake of TFA, interference of dietary trans fatty acids with infant
development is not a major issue nowadays.
References
Anderson, RL, Fullmer, CSJ and Hollenbach, EJ (1975) Effects of the trans isomers of
linoleic acid on the metabolism of linoleic acid in rats. J. Nutr., 105, 393–400.
Berdeaux, O, Blond, JP, Bretillon, L, Chardigny, JM, Mairot, T, Vatèle, JM, Poullain, D and
Sébédio, JL (1998) In vitro desaturation or elongation of monotrans isomers of linoleic
acid by the rat liver microsomes. Mol. Cell. Biochem., 85, 17–25.
326 TRANS FATTY ACIDS IN HUMAN NUTRITION
Blond, JP, Chardigny, JM, Sébédio, JL and Grandgirard, A (1995) Effects of dietary 18:3 n–
3 trans isomers on the Δ6 desaturation of α-linolenic acid. J. Food Lipids, 2, 99–106.
Boue, C, Combe, N, Billeaud, C and Entressangles, B (2001) Nutritional implications of trans
fatty acids during perinatal period in French pregnant women. Ol. Corps Gras, Lipides,
8, 68–72.
Boue, C, Combe, N, Billeaud, C, Mignerot, C, Entressangles, B, Thery, G, Geoffrion, H,
Brun, JL, Dallay, D and Leng, JJ (2000) Trans fatty acids in adipose tissue of French
women in relation to their dietary sources. Lipids, 35, 561–566.
Brenner, RR and Peluffo, RO (1969) Regulation of unsaturated fatty acid biosyntheses. I.
Effect of unsaturated fatty acid of 18 carbons on the microsomal desaturation of linoleic
acid into γ-linolenic acid. Biochim. Biophys. Acta, 176, 471–479.
Carlson, SE, Clandinin, MT, Cook, HW, Emken, EA and Filer, LJ (1997) Trans fatty acids:
infant and fetal development Am. J. Clin. Nutr., 66, S717–S736.
Carlson, SE and Neuringer, M (1999) Polyunsaturated fatty acid status and neurodevelopment:
a summary and critical analysis of the literature Lipids, 34, 171–178.
Chardigny, JM, Wolff, RL, Mager, E, Bayard, CC, Sébédio, JL, Martine, L and Ratnayake,
WMN (1996) Fatty acid composition of French infant formulas with emphasis on the
content and detailed profile of trans fatty acids J. Am. Oil Chem. Soc., 73, 1595–1601.
Chardigny, JM, Wolff, RL, Mager, E, Sébédio, JL, Martine, L and Juanéda, P (1995) Trans
mono- and poly-unsaturated fatty acids in human milk Eur. J. Clin. Nutr., 49, 523–531.
Chen, ZY, Pelletier, G, Hollywood, R and Ratnayake, WM (1995) Trans fatty acid isomers
in Canadian human milk Lipids, 30, 15–21.
Clandinin, MT (1999) Brain development and assessing the supply of polyunsaturated fatty
acid Lipids, 34, 131–137.
Combe, N, Billeaud, C, Mazette, B, Entressangles, B and Sandler, B (1995) TFAs in human
milk reflect animal or vegetable TFA consumption in France. Pediatr. Res., 37, 304A.
Combe, N, Judde, A, Billeaud, C, Boue, C, Turon, F, Entressangles, B, Dallay, D and Leng,
JJ (1997) In: Essential Fatty Acids and Eicosanoids (Riemersma, RA, Armstrong, RA,
Kelly, RW and Wilson, R, eds) AOCS Press, Champaign, illinois, USA, pp.239–242.
Cook, HW and Emken, EA (1990) Geometric and positional fatty acid isomers interact
differently with desaturation and elongation of linoleic acid in cultured glioma cells.
Cellular Biology, 68, 653–660.
Craig-Schmidt, MC (1992) In: Fatty Acids in Foods and their Health Implications (Chow,
CK, ed.) Marcel Dekker, New York, USA, pp.365–398.
De Schrijver, R and Privett, OS (1982) Interrelationship between dietary trans fatty acids and
the 6- and 9-desaturases in the rat. Lipids, 17, 27–34.
Decsi, T, Boehm, G, Tjoonk, HM, Molnar, S, Dijck-Brouwer, DA, Hadders-Algra, M,
Martini, IA, Muskiet, FA and Boersma, ER (2002) Trans isomeric octadecenoic acids are
related inversely to arachidonic acid and DHA and positively related to mead acid in
umbilical vessel wall lipids. Lipids, 37, 959–965.
Decsi, T, Burus, I, Molnar, S, Minda, H and Veitl, V (2001) Inverse association between trans
isomeric and long-chain polyunsaturated fatty acids in cord blood lipids of full-term
infants. Am. J. Clin. Nutr., 74, 364–368.
Decsi, T and Koletzko, B (1994) Fatty acid composition of plasma lipid classes in healthy
subjects from birth to young adulthood. Eur. J. Pediatr., 153, 520–525.
Decsi, T and Koletzko, B (1995) Do trans fatty acids impair linoleic acid metabolism in
children? Ann. Nutr. Metab., 39, 36–41.
Elias, SL and Innis, SM (2001) Infant plasma trans, n–6, and n–3 fatty acids and conjugated
linoleic acids are related to maternal plasma fatty acids, length of gestation, and birth
weight and length. Am. J. Clin. Nutr., 73, 807–814.
Friesen, R and Innis, SM (2006) Trans fatty acids in human milk in Canada declined with the
introduction of trans fat food labeling. J. Nutr., 136, 2558–2561.
DIETARY TFA: FROM THE MOTHER’S DIET TO THE INFANT 327
Glew, RH, Herbein, JH, Moya, MH, Valdez, JM, Obadofin, M, Wark, WA and Vanderjagt,
DJ (2006) Trans fatty acids and conjugated linoleic acids in the milk of urban women and
nomadic Fulani of northern Nigeria. Clin. Chim. Acta, 367, 48–54.
Hayashi, AA, de Medeiros, SR, Carvalho, MH and Lanna, DP (2007) Conjugated linoleic
acid (CLA) effects on pups growth, milk composition and lipogenic enzymes in lactating
rats. J. Dairy Res., 74, 160–6.
Hwang, DH and Kinsella, JE (1979) The effects of trans trans methyl linoleate on the
concentration of prostaglandins and their precursors in rat. Prostaglandins, 17, 543–59.
Innis, SM (2005) Essential fatty acid transfer and fetal development. Placenta, 26 Suppl A,
S70–S75.
Innis, SM, Sprecher, H, Hachey, D, Edmond, J and Anderson, RE (1999) Neonatal
polyunsaturated fatty acid metabolism. Lipids, 34, 139–149.
Jensen, RG (1999) Lipids in human milk. Lipids, 34, 1243–1271.
Jensen, RG, Lammi-Keefe, CJ, Hill, DW, Kind, AJ and Henderson, R (1998) The
anticarcinogenic conjugated fatty acid, 9c, 11t-18:2, in human milk: confirmation of its
presence. J. Hum. Lact., 14, 23–27.
Johnston, PV, Kummerow, FA and Walton, CH (1958) Origin of the trans fatty acids in
human tissue. Proc. Soc. Exp. Biol. Med., 99, 735–736.
Koletzko, B (1992) Trans fatty acids may impair biosynthesis of long-chain polyunsaturates
and growth in man. Acta Paediatr., 81, 302–306.
Koletzko, B, Mrotzek, M and Bremer, HJ (1988) Fatty acid composition of mature human
milk in Germany. Am. J. Clin. Nutr., 47, 954–959.
Koletzko, B and Muller, J (1990) Cis- and trans-isomeric fatty acids in plasma lipids of
newborn infants and their mothers. Biol. Neonate, 57, 172–178.
Koletzko B, Thiel I and Abiodun PO (1991) Fatty acid composition of mature human milk
in Nigeria. Z. Ernahrungswiss, 30, 289–297.
Koletzko, B, Thiel, I and Springer, S (1992) Lipids in human milk: a model for infant
formulae? Eur. J. Clin. Nutr., 46, S45–S55.
Kurata, N and Privett, O (1980) Effects of dietary trans acids on biosynthesis of arachidonic
acid in rat liver microsomes. Lipids, 15, 1029–1036.
Larque, E, Perez-Llamas, F, Puerta, V, Giron, MD, Suarez, MD, Zamora, S and Gil, A (2000)
Dietary trans fatty acids affect docosahexaenoic acid concentrations in plasma and liver
but not brain of pregnant and fetal rats. Pediatr. Res., 47, 278–283.
Luna, P, Juarez, M and De la Fuente, MA (2007) Fatty acid and conjugated linoleic acid
isomer profiles in human milk fat. Eur. J. Lipid Sci. Technol, 109, 1160–1166.
Mahfouz, MM, Smith, TL and Kummerow, FA (1984) Effect of dietary fats on desaturase
activities and the biosynthesis of fatty acids in rat-liver microsomes. Lipids, 19, 214–
222.
Masters, N, McGuire, MA, Beerman, KA, Dasgupta, N and McGuire, MK (2002) Maternal
supplementation with CLA decreases milk fat in humans. Lipids, 37, 133–138.
Mc Guire, MK, Park, Y, Behre, RA, Harrison, LY, Shultz, TD and Mcguire, MA (1997)
Conjugated linoleic acid concentrations of human milk and infant formula. Nutrition
Research, 17, 1277–1283.
Mojska, H, Socha, P, Socha, J, Soplinska, E, Jaroszewska-Balicka, W and Szponar, L (2003)
Trans fatty acids in human milk in Poland and their association with breastfeeding
mothers’ diets. Acta Paediatr., 92, 1381–1387.
Moore, CE and Dhopeshwarkar, GA (1980) Placental transport of trans fatty acids in the rat.
Lipids, 15, 1023–1028.
Mosley, EE, Wright, AL, McGuire, MK and McGuire, MA (2005) Trans fatty acids in milk
produced by women in the United States. Am. J. Clin. Nutr., 82, 1292–1297.
Permanyer, JJ, Pinto, BA, Hernandez, N and Boatella, J (1990) Trans isomer content in infant
milk formulas. Lait, 70, 307–311.
328 TRANS FATTY ACIDS IN HUMAN NUTRITION
Pettersen, J and Opstvedt, J (1989) Trans fatty acids. 3. Fatty acid composition of the brain
and other organs in the newborn piglet. Lipids, 24, 616–624.
Pettersen, J and Opstvedt, J (1992) Trans fatty acids. 5. Fatty acid composition of lipids of
the brain and other organs in suckling piglets. Lipids, 27, 761–769.
Precht, D and Molkentin, J (1999) C18:1, C18:2 and C18:3 trans and cis fatty acid isomers
including conjugated cis delta 9, trans delta 11 linoleic acid (CLA) as well as total fat
composition of German human milk lipids. Nahrung, 43, 233–244.
Privett, OS, Stearns, EM and Wickell, EC (1967) Metabolism of the geometrical isomers of
linoleic acid in the rat. J. Nutr., 92, 303–310.
Ringseis, R, Saal, D, Muller, A, Steinhart, H and Eder, K (2004) Dietary conjugated linoleic
acids lower the triacylglycerol concentration in the milk of lactating rats and impair the
growth and increase the mortality of their suckling pups. J. Nutr., 134, 3327–34.
Rosenthal, MD and Doloresco, MA (1984) The effects of trans fatty acids on fatty acyl delta 5
desaturation by human skin fibroblasts. Lipids, 19, 869–874.
Sellmayer, A and Koletzko, B (1999) Long-chain polyunsaturated fatty acids and eicosanoids
in infants — physiological and pathophysiological aspects and open questions. Lipids,
34, 199–205.
Shimp, JL, Bruckner, G and Kinsella, JE (1982) The effects of dietary trilinelaidin on fatty
acid and acyl desaturases in rat liver. J. Nutr., 112, 722–735.
Szabo, E, Boehm, G, Beermann, C, Weyermann, M, Brenner, H, Rothenbacher, D and Decsi,
T (2007) Trans octadecenoic acid and trans octadecadienoic acid are inversely related to
long-chain polyunsaturates in human milk: results of a large birth cohort study. Am. J.
Clin. Nutr., 85, 1320–1326.
Tinoco, SM, Sichieri, R, Setta, CL, Moura, AS and Tavares do Carmo, MD (2008) Trans fatty
acids from milk of Brazilian mothers of premature infants. J. Paediatr. Child Health, 44,
50–56.
Van Houwelingen, AC and Hornstra, G (1994) Trans fatty acids in early human development.
World Rev. Nutr. Diet, 75, 175–178.
Vaysse, C, Couedelo, L, Guesnet, P, Alessandri, JM, Billeaud, C, Putet, G and Combe, N
(2008). Abstract of a paper presented at the 6th EuroFedLipid Congress, 7–10 September
2008, Athens, Greece.
Wolff, RL (1993) Heat-induced geometrical isomerization of linolenic acid: effect of
temperature and heating time on the appearance of individual isomers. J. Am. Oil Chem.
Soc., 70, 425–430.
Wolff, RL, Precht, D and Molkentin, J (1998) In: Trans Fatty Acids in Human Nutrition, First
Edition (Sebedio, JL and Christie, WW, eds) The Oily Press, Bridgwater, UK, pp.1–33.
Zevenbergen, JL and Haddeman, E (1989) Lack of effects of trans fatty acids on eicosanoid
biosynthesis with adequate intakes of linoleic acid. Lipids, 24, 555–63.
Zevenbergen, JL, Houtsmuller, UM and Gottenbos, JJ (1988) Linoleic acid requirement of
rats fed trans fatty acids. Lipids, 23, 178–86.
CHAPTER 13
Evolution of worldwide consumption of trans
fatty acids
Trans fatty acids (TFA) from industrial processing of vegetable oils entered the
food supply in the USA in 1911 with the commercialization of ‘Crisco’, the first
partially hydrogenated shortening marketed for frying and baking. This prod-
uct was developed in part from the work of Paul Sabatier, a French chemist who
won the 1912 Nobel Prize in Chemistry for his method of hydrogenating
organic compounds in the presence of metal catalysts. Wilhelm Normann, a
German chemist, extended Sabatier’s work and in 1902 patented the process by
which liquid oils could be hydrogenated. In 1909, the Proctor and Gamble
company acquired the US rights to Normann’s patent and within two years
began marketing Crisco. This product became widely used in households in the
USA as a result of the distribution of free cookbooks in which every recipe
called for Crisco.
Partial hydrogenation in which liquid vegetable oils are converted to solid or
semi-solid fats, such as margarine and shortening, had several advantages for
the food industry. The consistency of the fat could be controlled by the degree
of hydrogenation, and use of these semi-solid fats resulted in desirable qualities
for baked goods. The partially hydrogenated fats were less susceptible to
oxidation and thus afforded a longer shelf life to commercial products.
Prior to the early 1900s, the only sources of trans isomers in the food supply
were meat and dairy products. Biohydrogenation in the rumen of animals
results in the formation of vaccenic acid (trans-11 18:1), the predominant trans
fatty acid in milk and meat, as well as smaller amounts of other TFA. Ruminant
products contain 1 to 8% of the fat as TFA.
Industrial processing of soybean and other vegetable oils results in products
such as margarine, shortening and frying fats that contain TFA at levels up to
40 to 50% of total fatty acids (Craig-Schmidt & Teodorescu, 2008). The
predominant trans isomers in partially hydrogenated vegetable fat are trans-9
18:1 and trans-10 18:1. In addition to trans isomers of 18:1 in the diet, smaller
amounts of trans isomers of 18:2 and even smaller amounts of trans-16:1 have
329
330 TRANS FATTY ACIDS IN HUMAN NUTRITION
been reported (Lemaitre et al., 1998). Marine oils have also been hydrogen-
ated, resulting in a variety of trans isomers. Deep-fried fast foods, as well as
bakery goods, snack foods and other commercial products made with industri-
ally hydrogenated fat, are major sources of dietary TFA (Craig-Schmidt and
Teodorescu, 2008). Thus, populations consuming relatively large amounts of
commercially hydrogenated fat have a greater dietary TFA intake than do
populations whose dietary fat is primarily derived from ruminant sources.
Populations who use olive oil and other non-hydrogenated vegetable oils as the
primary dietary fat have relatively minor amounts of TFA in the diet.
Health professionals originally promoted margarine and other partially hydro-
genated fats as part of a diet to lower the risk for cardiovascular disease and other
chronic diseases because these products were derived from vegetable rather than
animal sources. However, since the early 1990s, evidence has accumulated that
TFA have a detrimental effect on risk for cardiovascular disease and other
chronic diseases. The landmark work of Mensink and Katan (1990) and Zock and
Katan (1992) led to a revision in thinking about TFA. These investigators
demonstrated that dietary TFA had an adverse effect on lipoprotein profile,
decreasing high-density lipoproteins or HDL and increasing low-density lipo-
proteins or LDL. This finding, as well as a myriad of other results on the harmful
effects of TFA (as reviewed by Hunter, 2006; Mozaffarian, et al. 2006; Stender
& Dyerberg, 2003; and many others), has led to efforts to remove industrially
produced TFA from the food supply either by regulation of industry, by changing
food labelling laws, or by local regulation of restaurants and fast food chains. A
variety of ‘trans-free’ foods are now available to the consumer and alternative
fats for frying and baked goods are beginning to enter the market place.
The purpose of this review is to summarize data on the worldwide consump-
tion of TFA from an historical perspective, with an emphasis on trends in
consumption since the 1990s. Evidence is presented that trans fatty acid
consumption in North America and Europe has decreased over the last 10
to15 years. Earlier reviews or government reports which summarize estimates
of TFA in the diet include a report by the Life Sciences Research Office for the
US Food and Drug Administration compiled by Senti (1985), the International
Life Sciences Institute Expert Panel on Trans Fatty Acids and Coronary Heart
Disease (1995), the American Society for Clinical Nutrition/American Insti-
tute of Nutrition Task Force on Trans Fatty Acids (1996), the International Life
Sciences Institute Expert Panel on Trans Fatty Acids and Early Development
(1997) and Aro (1998). A comprehensive review of worldwide consumption of
TFA through the late 1990s was published by Craig-Schmidt (1998), and the
present chapter seeks to update this earlier review. In addition, a brief overview
of worldwide consumption was published as part of a report on the ‘First
International Symposium on Trans Fatty Acids and Health’ (Craig-Schmidt,
2006). A review dealing with trans fatty acid consumption in only North
America has been published recently (Craig-Schmidt, 2007).
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS 331
Methods used to estimate TFA in the diet include: (1) estimates based on ‘food
disappearance’ or market share data; (2) analysis of dietary consumption data
of a representative population; (3) laboratory analysis of duplicate portion or
composite diets; and (4) biomarkers. Each of these methods has inherent
advantages and disadvantages; however, collectively these methods result in
reasonable estimates of trans fatty acid consumption which are useful in
comparing consumption among various populations or identifying the relation-
ship between trans fatty acid intake and heath effects.
Estimates based on food disappearance or market share data tend to overes-
timate trans fatty acid consumption. This method has the advantage that it
accounts for fat in the total population. However, estimates from market share
data can vary widely depending on the ‘typical’ value for the trans fatty acid
content of each commodity or on the ‘wastage’ factor used in the estimates.
Data from US government commodity sales information were used in an early
estimation of trans fatty acid intake in the US population (Senti, 1985). The per
capita estimates at this time were 8 g/day from vegetable fats and 2.21 g/day
from animal and dairy fats, giving a total of 10.21 g/day for total trans fatty acid
intake. After a wastage factor was applied, trans fatty acid intake in the 1980s
was believed to be 8.3 g/day/person in the USA. Estimates using this method
are not commonly reported in the literature at the present time; however,
Johansson et al. (2006) used this method to estimate per capita trans fatty acid
consumption in Norway as 1.6 g/day based on a survey among Norwegian food
manufacturers and importers in 2003.
Laboratory analysis of duplicate portion or composite diets can also be used
to estimate trans fatty acid intake of a population. In this method, diets are
collected from a representative population or formulated based on a diet
composition considered to be ‘typical’ of a population and then analysed in a
laboratory. The advantage of this method is that dietary TFA can be determined
with a high degree of accuracy, especially if sophisticated analytical method-
ology is used; a disadvantage is that laboratory analyses are very expensive
and, thus, only a small number of diets can be used. Additionally, this method
is useful only to the extent that the subset of the population from which the diets
are collected is representative of the entire population or only to the extent that
the composite diets are reflective of the diet of the population as a whole. This
method was used successfully in the Seven Countries Study (Kromhout et al.,
1995) in which the trans fatty acid content of food composites was determined
for The Netherlands, Finland, the USA, former Yugoslavia, Italy, Greece and
Japan. The differences in consumption of TFA among these countries were
clearly illustrated. Kromhout and colleagues (1995) showed that in 1987 intake
of TFA was greater (4 to 6 g/day/person) in northern European countries and
the USA than in eastern European or Mediterranean countries (less than 1.5 g/
332 TRANS FATTY ACIDS IN HUMAN NUTRITION
day/person) and much greater than in Japan (less than 0.5 g/day/person). Thus,
this method appears to be particularly useful in cross-country comparisons in
which the diets of the countries vary widely.
The most commonly used method to estimate TFA consumption in the
decade from 1995 to 2005 was based on analysis of dietary consumption data
of a representative population. Diet recalls, diet records, and food frequency
questionnaires can all be used to collect data on the diets of individuals. The
amounts of various food items consumed can then be combined with data from
food composition tables to give the average daily consumption of a nutrient or
other component in the diet. This method has the advantage that consumption
by the individual rather than the population is assessed. The main problem with
this method in light of a rapidly changing food supply is that values for trans
fatty acid content in the food composition tables are outdated quickly. These
values are subject to inaccuracies not only because of changes in food process-
ing, but also because of the wide variability in trans fatty acid content of similar
types of foods and differences in the analytical methods used to determine the
trans fatty acid content of each food item. The inherent problems with
assessing the diet of an individual, as well as the limitations of food composi-
tion databases for the trans fatty acid values, tend to result in underestimations
of the trans fatty acid content of the diet compared to values obtained by other
methods.
Innis et al. (1999) have clearly illustrated the problems with dietary method-
ology in assessing the intake of TFA. By analysing over 200 Canadian food
items, these investigators demonstrated the variability in trans fatty acid
content among food items or brands within a product category. For example, in
17 brands of crackers, trans fatty acid content ranged from 23 to 51% of total
fatty acids or differences of 1 to 13 g TFA per 100 g of crackers. The
significance of this variability to the estimation of trans fatty acid consumption
from dietary intake data was illustrated by analyses of a single diet record.
Using minimum and maximum values for TFA within a given food category,
the investigators reported that values for trans fatty acid intake could range
from a low of 1.4 to a high of 25.4 g a day. Thus, the wide variability in trans
fatty acid content of different food items within a category may result in large
errors in the estimation of trans fatty acid intake of individuals and groups.
Because of these unavoidable inaccuracies for trans fatty acid values in food
composition tables and the inherent problems in assessing the diet of individu-
als, the use of methods requiring a food composition database to estimate
dietary intake of TFA has decreased. Instead, the use of biomarkers is believed
to reflect better the actual individual consumption of TFA.
In recent years, the use of biomarkers of trans fatty acid consumption has
become increasingly popular in studies designed to relate the consumption of
TFA to health risk for various diseases. Several biological markers have been
used to estimate trans fatty acid intake. Adipose tissue is believed to reflect
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS 333
long-term consumption, i.e., that of the previous year (Beynen et al., 1980;
Hirsch et al., 1960), whereas the fatty acid composition of erythrocytes reflects
consumption over the previous months and that of plasma or serum over the
previous weeks. Human milk, on the other hand, reflects the trans fatty acid
composition of the maternal diet of the previous day (Craig-Schmidt et al.,
1984). The tissue values can be used directly in making comparisons between
groups; however, if one wants a value for dietary intake, regression equations
relating tissue values to dietary values must be used. Enig et al. (1990) derived
an equation for estimating trans fatty acid intake from the concentration of
trans-18:1 in adipose tissue. In a similar manner, Craig-Schmidt et al.(1984)
and Wolff (1995) derived equations applicable to converting human milk
values to dietary values for TFA.
Advantages for using biomarkers as surrogate values for dietary intake
include the fact that they are directly reflective of the diet of the individual from
which the sample is obtained, estimates are not dependent on food composition
tables which are constantly changing, and biomarkers are good for compari-
sons among countries or for relating trans fatty acid consumption to risk for
various diseases. One disadvantage is that the amounts of TFA in human milk
or adipose tissues may be influenced by factors such as caloric intake. Com-
parisons are best if the analytical methods are the same, i.e., if the samples are
analysed in the same laboratory. Differences in analytical methodology may
make data difficult to compare among different laboratories. Additionally, the
equations used to convert tissue values to dietary values may have limited
applicability.
Aro et al. (1995) used the fatty acid composition of adipose tissue as a
biomarker of consumption in the EURAMIC study for a cross-country com-
parison of nine countries in studying the relationship of TFA to risk of
myocardial infarction. In 1991–92, trans-18:1 in adipose tissue in men from
The Netherlands, Norway, UK and Israel was over 2%, compared to about
1.5% for Switzerland, Finland, Russia and Germany. Men from Spain, on the
other hand, had much less trans-18:1 in their adipose tissue (less than 0.5%).
Thus, the trans-18:1 content of adipose tissue seems to reflect consumption of
TFA in various countries.
Other investigators have used the trans fatty acid content of adipose tissue
as a marker of trans fatty acid consumption in relationship to the risk for
sudden cardiac death (Roberts et al., 1995) and myocardial infarction
(Clifton et al., 2004; Baylin et al., 2002, 2003). Adipose tissue aspirates can
be used also to indicate the intake of exogenous fatty acids, distinguishing
between the consumption of vegetable fats and those of animal origin (Gar-
land et al. 1998). For example, Combe et al. (1998) used the trans isomer
distribution in adipose tissue of pregnant French women to study the relative
amounts of ruminant fats versus industrially produced hydrogenated fats in
the French diet compared with that of other Western countries. Thus, adipose
334 TRANS FATTY ACIDS IN HUMAN NUTRITION
tissue can serve as a valuable biomarker of both the amount and type of TFA
in the habitual diet.
Human milk as a biomarker has been particularly useful in cross-country
comparisons of trans fatty acid consumption (Craig-Schmidt, 2006; Wolff,
1995) as well as in documenting a decrease in trans fatty acid intake response
to an effort to decrease trans fat in the food supply (Friesen & Innis, 2006).
Also, distribution of positional trans isomers of the fatty acids in breast milk,
like adipose tissue, appears to reflect the predominant source of TFA in the
maternal diet. The presence of relatively large amounts of vaccenic acid (trans-
11 18:1) and the relative amount of trans-16 18:1 indicate that milk and other
dairy products are the major sources of TFA in the maternal diet (Chardigny
et al. 1995; Wolff, 1995). On the other hand, an almost equal distribution of
trans-11 18:1 and trans-10 18:1 indicates that the TFA in the maternal diet are
derived primarily from commercially hydrogenated fats (Chen et al., 1995a,b).
Values for the trans fatty acid content of human milk in various countries
reported prior to 1998 have been summarized in several reviews (Wolff et al.,
1998; Jensen, 1999; Craig-Schmidt, 2001). More recent values for TFA in
human milk as percent of total fatty acids (wt/wt) include 7.1 ± 0.32% in 103
Canadian women (Innis & King, 1999), 3.81 ± 0.97% with a range of 2.38 to
6.03% in 40 German women (Precht & Molkentin, 1999), 2.80 ± 1.75% in 19
Kuwaiti mothers (Hayat et al., 1999), 2.81 ± 0.61% in 35 women from Prague,
Czech Republic (Dlouhy et al., 2002), 1.00 to 4.31% depending on individual
variation, season of the year, and stage of lactation in 19 Polish women (Mojska
et al., 2003), 11.3 ± 3.4 in 52 women from Western Iran (Bahrami & Rahimi,
2005), 7.0 ± 2.3% with a range of 2.5–13.8% in 81 women in southwestern
USA (Mosley, et al. 2005), 0.22 and 0.34% in 41 nomadic Fulani and 41 urban
women, respectively, in Nigeria (Glew et al., 2006), and 6.2 ± 0.48, 5.3 ± 0.49
and 4.6 ± 0.32 in samples collected in three consecutive periods from late 2004
to early 2006 in 87 Canadian women (Friesen & Innis, 2006). Differences in
these values probably reflect cultural differences in diet and changes in the food
supply over time, as well as differences in the methodology used in collection
and analysis of the samples.
The concentration of trans fatty acid isomers in various fractions of the blood
has been used increasingly as a biomarker to relate the incidence of disease to
trans fatty acid consumption. In some of these studies, correlations have been
found with some of the minor trans isomers and not with trans-18:1 or overall
TFA. For example, Lemaitre et al. (2006) reported that greater trans-18:2, but
not trans-18:1 in plasma phospholipid was associated with greater risk of fatal
ischaemic heart disease and sudden cardiac death in older adults. Also, van de
Vijver et al. (1996) have used TFA in plasma phospholipids as a biomarker for
dietary trans fatty acid consumption in studying risk for cardiovascular dis-
ease. Other investigators, such as Lemaitre, et al. (2002), Mozaffarian et al.
(2004a,b), and Sun et al. (2007a,b), have studied TFA in the red blood cell
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS 335
1. North America
Estimates of trans fatty acid intake in the USA and Canada published from the
late 1990s to the present are summarized in Table 1. In the mid 1990s several
groups in the USA published ‘consensus’ estimates of TFA in the US diet.
These values for trans fatty acid intake were: ILSI Expert Panel on Cardiovas-
cular Disease (1995), 2–4% of energy and 4–12% of total fatty acids; ASCN/
AIN Task Force (1996), 2–4% of energy and 8.1–12.8 g/day; and ILSI Expert
Panel on Development (1997), 8% of total fatty acids and 6.4 g/day. Values
published since that time are generally somewhat less.
In results published since 1997, the semi-quantitative food frequency ques-
tionnaire, particularly that of Willett and colleagues, is the most commonly
used method to collect dietary data for determination of trans fatty acid intake
in the USA (Table 1). Typically, mean or median values using food frequency
methodology range from 2 to 3 g/person/day (Garland et al., 1998; Lemaitre
et al., 1998; Lopez-Garcia et al., 2004; Mozaffarian et al., 2004a; Lopez-
Garcia et al., 2005; Tsai et al., 2005) and 1 to 2% of calories (Hu et al., 1997;
Hu et al., 2000; Lopez-Garcia et al., 2005; Lu et al. 2005; Sun et al., 2007a),
whereas values obtained using diet recalls or diet records for the adult pop-
ulation are generally higher, 4 to 8 g/person/day (Allison et al., 1999; Harnack,
et al., 2003) or 2 to 3% of energy (Harnack et al., 2003; Zhou et al., 2003;
Table 1. Estimates of trans fatty acid consumption in North America published late-1990s to present.
336
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
USA Habitual diet Subjects (n = 51) Self-administered food Lemaitre et al., 1998
(1996) from greater Seattle, frequency question-
Washington area, naire modified from
aged 51–78 y NCI/Block combined
with revised US
Men (n = 24) Department of
total TFA 2.57 ± 0.69 5.46 ± 1.68 Agriculture Food
16:1t 0.02 ± 0.002 Composition Database
18:1t 2.24 ± 0.60
18:2t 0.29 ± 0.08
Women (n = 27)
total TFA 1.99 ± 0.88 4.98 ± 2.39
16:1t 0.01 ± 0.002
18:1t 1.78 ± 0.78
18:2t 0.24 ± 0.09
USA Individual diet Subjects (n = 11,258) 24-hour recall and Allison et al., 1999
(1989–1991) from Continuing 2-day diet record
Survey of Food methodology combined
Intakes by Individuals with US Department
(CSFII) of Agriculture
Average intake for 5.3 ± 0.08 2.6 ± 0.02 7.4 ± 0.06 Database
US population
Boys & girls, 3–5 y 4.1 ± 0.12 2.6 ± 0.05 7.5 ± 0.13
(n = 615)
Boys & girls, 6–11 y 5.3 ± 0.13 2.7 ± 0.05 7.7 ± 0.13
(n = 1,172)
Boys, 12–19 y 7.1 ± 0.37 2.8 ± 0.06 7.6 ± 0.14
(n = 618)
Girls, 12–19 y 5.1 ± 0.16 2.7 ± 0.06 7.6 ± 0.13
(n = 672)
Men, 20–49 y 6.6 ± 0.16 2.6 ± 0.03 7.1 ± 0.08
(n = 2,076)
Women, 20–49 y 4.6 ± 0.10 2.6 ± 0.03 7.4 ± 0.09
(n = 2,787)
Men, 50–69 y 5.8 ± 0.19 2.6 ± 0.06 7.3 ± 0.14
(n = 879)
Women, 50–69 y 4.2 ± 0.13 2.6 ± 0.05 7.5 ± 0.13
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS
(n = 1,226)
Men & women, 70+ 4.5 ± 0.12 2.6 ± 0.05 7.9 ± 0.12
(n = 1,213)
337
338
Table 1. continued
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Table 1. continued
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Canada Typical Sample Canadian ‘Market basket’ diet Innis et al., 1999
Vancouver diet diet used to show analysed with database
variability in of 200 food items pur-
estimated trans chased in Vancouver,
fatty acid intake British Columbia,
depending on brand Canada, analysed in
or database chosen Innis’ laboratory or
total TFA 1.4 – 25.4 FOOD PROCESSOR
database (ESHA
Research, Salem,
Oregon)
Canada Individual diets Canadian lactating 3-day diet records; Innis and King, 1999
women (n = 21) analysed using FOOD
total TFA 6.87 ± 1.07 2.46 ± 0.38 PROCESSOR database
(ESHA Research, Salem,
Oregon) modified to
include 300 brand name
TRANS FATTY ACIDS IN HUMAN NUTRITION
Data expressed as total trans fatty acids, and as mean ± SD or as mean (range) unless otherwise specified.
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS
343
344 TRANS FATTY ACIDS IN HUMAN NUTRITION
Lee et al., 2007). An exception is Liu et al., 2007, who reported total trans fatty
acid consumption values of 4 to 6 g/day even though a food frequency
questionnaire was used to obtain dietary data. In general, these estimates of
trans fatty acid consumption using both types of dietary data are less than the
‘consensus’ values published in the mid 1990s.
As one might expect, there are documented differences in trans fatty acid
consumption among various groups. Men, because of greater caloric intake,
have a greater trans fatty acid consumption than do women (Lemaitre et al.,
1998; Allison et al., 1999; Harnack et al, 2003). Expressed as energy percent,
values for men and women are somewhat similar (Zhou et al., 2003; Lee et al.,
2007). In a comparison of various age and gender groups, Allison et al. (1999)
found that teenage boys had the greatest consumption of TFA (7.1 g/day) of
any age group studied. Children had a surprisingly high level of TFA in the diet
(4.1 g/day for 3–5 year old boys and girls and 5.3 g/day in 6–11 year old boys
and girls). Also, trans fatty acid consumption appears to be greater in African
Americans than in Caucasians (Liu et al., 2007).A direct comparison of TFA
intake in the USA with that of three other countries was done by Zhou et al.
(2003). In this study, consumption in the USA and the UK (2.0 and 1.6% of
energy, respectively) was considerably higher than that in China or Japan (0.2
and 0.3% of energy, respectively).
There is evidence that dietary intake of TFA in the USA has decreased since
about 1980. For example, Hu et al. (2000) reported decreases in total TFA
consumption of 2.20, 1.90, 1.68 and 1.52% of calories for 1980, 1984, 1886,
and 1990, respectively. Similarly, Harnack et al. (2003) documented a de-
crease in consumption of TFA in US men from 8.4 g/day in 1980 to 1982 to
6.4 g/day in 1995 to 1997 and in US women from 5.4 g/day to 4.7 g/day in the
same periods of time. This trend was confirmed by Lee et al. (2007) who
reported a drop in TFA consumption from 3.0 to 2.3% of energy in men and 2.8
to 2.2% of energy in women in the period from 1980 to 1982 until 2000 to 2002.
The decrease in TFA consumption in the USA is due to the response of
industry to food label regulations and increased consumer awareness of the
deleterious effects of TFA in foods. In 2003, the US Food and Drug Adminis-
tration (FDA) issued a ruling requiring the declaration of trans fat on food
labels (DHHS/FDA 2003; Schrimpf-Moss & Wilkening, 2007). Manufacturers
were given until 1 January 2006, to implement the regulation. ‘Trans-free’
packaged foods are now prevalent in the US marketplace, although baked
goods and fried foods in restaurants may still contain relatively high amounts
of TFA. Stender and colleagues (Stender et al., 2006) demonstrated that the
TFA content of fast food in the USA was among the greatest in the 20 countries
from which samples were collected. French fries and chicken nuggets, pur-
chased from McDonald’s or Kentucky Fried Chicken establishments between
November, 2004, and September, 2005, were analysed by capillary gas chro-
matography in a single laboratory using a standard method (Leth et al, 2003).
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS 345
For fast food from McDonald’s restaurants, the USA was the greatest in TFA
(10 g or 23% of total fat in French fries and 11% in chicken nuggets) with
Denmark being the least (less than 1 g or 1% of total fat in both products). It is
likely that recent reformulation of fats (Tarrago-Trani et al., 2006; Eckel et al.,
2007) has resulted in decreased use of fat and oils rich in TFA by the fast food
industry in the USA, but this change has not been well documented in the
scientific literature.
Trends in TFA consumption in Canada are roughly parallel to that of the
USA (Craig-Schmidt, 2007). Brisson (1981) was one of the first to estimate
TFA intake in the Canadian population, calculating that a typical Canadian
person would consume an average of 9.1 g/day, with a maximal value of
17.5 g/day. These estimates, based on the average TFA content of selected food
items and Canadian government consumption data from 1977, were in the same
range as early estimates of TFA consumption by the population of the USA.
Estimates in the early 1990s for TFA consumption in Canada were also fairly
high. Using the equation of Craig-Schmidt et al. (1984) in combination with
human breast milk analysis of 198 samples collected in 1992, Ratnayake and
colleagues (Chen et al., 1995a) calculated that TFA intake by lactating women
ranged from 3.0 g/day to 20.3 g/day, with the majority of women consuming
about 10 g/day. These same investigators (Ratnayake and Chen, 1995) also
estimated TFA intake to be 10.4% of the total dietary fat or 5.2 g/day for elderly
women to 12.5 g/day for young men in Nova Scotia.
More recently, Innis and King (1999) reported the average TFA intake by
Canadian lactating women (n = 21) to be 6.87 ± 1.07 g/person/day or 2.46 %
of calories (Table 1), a value somewhat less than that reported earlier by Chen
et al. (1995a). Moreover, in a cross-sectional prospective study of healthy,
pregnant women in Vancouver, Canada, Elias and Innis (2002) reported values
of 3.8 ± 0.3 g/day for women in the second trimester of pregnancy and
3.4 ± 0.3 g/day for women in the third trimester of pregnancy. These values
were based on a food intake questionnaire designed specifically to estimate
TFA intakes from a database of food items analysed in the investigator’s own
laboratory. Using a similar approach, Innis et al. (2004) reported values for the
TFA intake of Canadian children aged 18–60 months to be 4.81 ± 3.12 g/day
for this vulnerable population. This intake for Canadian children is similar to
the value of 4.1 g/day reported by Allison et al. (1999) for US children, 3–5
years old.
The recent values for TFA consumption reported by Innis’ group (Elias &
Innis, 2002; Innis et al., 2004) are less than those reported in the 1990s. Using
Canadian human milk as a biomarker and the equation of Craig-Schmidt et al.
(1984), Innis and colleagues (Friesen & Innis, 2006) confirmed that the
consumption of TFA had declined with the introduction of trans fat food
labelling regulations in Canada (Table 1). Like the USA, most Canadian food
products must carry the amount of trans fat in grams on the food label. In
346
Table 2. Estimates of trans fatty acid consumption in UK and Republic of Ireland published late 1990s to present.
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
UK Household UK men and women 7-day household con- Hulshof et al., 1999
consumption (n = 7921) in sumption survey
(1996) TRANSFAIR study, combined with Market
aged 0–75+ y Basket analyses of 100
total TFA 2.8 1.3 foods collected
14:1t9 0.11 1995–1996 plus other
16:1t9 0.18 country-specific data
18:1t 2.00
18:2t 0.28
18:3t +20:1 0.17
20:2t11,14 0.02
22:1t 0.06
UK Habitual diet UK population in the 4 standardized 24-h Zhou et al., 2003
(1997–1999) INTERMAP study, dietary recalls
TRANS FATTY ACIDS IN HUMAN NUTRITION
Data expressed as total trans fatty acids, and as mean ± SD or as mean (range) unless otherwise specified.
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS
347
348 TRANS FATTY ACIDS IN HUMAN NUTRITION
Canada, there must also be the percent Daily Value for trans fat and saturated
fat combined (Canadian Department of Health, 2003). Increased consumer
awareness of TFA in processed foods and reformulation of products by the food
industry will likely result in further decline of TFA in the Canadian food
supply, as well as in that of the USA.
2. UK and Ireland
Estimates of TFA intake in the UK and Republic of Ireland published from the
late 1990s to the present are summarized in Table 2. As part of the TRANSFAIR
study, Hulshof et al. (1999) estimated that TFA consumption for men and
women in the UK was 2.8 g/person/day. This value was less than earlier values
reported for the UK (Craig-Schmidt, 1998), but comparable to some European
countries, such as France, included in the TRANSFAIR study. In the
INTERMAP study, Zhou et al. 2003 reported that consumption of TFA in the
UK was 1.3–1.6en%, values which were in the same range as the TRANSFAIR
study. However, Irish adults appeared to consume greater amounts of TFA,
5.42 g/day or 1.89en%, than their counterparts in the UK (Cantwell et al.,
2005a,b).
Cantwell et al. (2005a,b) have studied the proportion of ruminant or ‘natu-
ral’ TFA compared to industrially produced TFA in the diets of healthy Irish
adults (Table 2). These investigators estimate that about 30% of the TFA
consumption in Ireland is from ruminant fats. Irish men consumed 1.8 g/day
from meat or dairy fats out of a total of 6.26 g/day, whereas Irish women
consumed 1.34 g/day from meat or dairy fats out of a total of 4.21 g/day as
dietary TFA.
The Food Standards Agency of the UK makes declaration of trans fat on food
labels mandatory if a claim is made regarding trans fat (Food Standards
Agency, 2003; Schrimpf-Moss & Wilkening, 2007). In the UK, as in North
America, it is probable that food labelling regulations eventually will make
industrially synthesized TFA in foods a thing of the past.
3. Continental Europe
Estimates of TFA intake in continental Europe published from the late 1990s to
the present are summarized in Table 3. The TRANSFAIR study (Hulshof et al.,
1999; van de Vijver, 2000) is the major study comparing TFA consumption in
European countries. Hulshof et al. (1999) reported that intakes for continental
Europe were low in the Mediterranean/southern European countries with
Greece, Italy, Portugal and Spain having values of 1.2 to 2.1 g/day for men
compared to northern European countries, with values ranging from 2.4 g/day
in Germany to 4.8 g/day in The Netherlands. Intermediate values were reported
for men in France (2.7 g/day) and Belgium (4.4 g/day). Other investigators
Table 3. Estimates of trans fatty acid consumption in continental Europe published late-1990s to present.
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Belgium Individual diet Belgium population 3-day diet records Hulshof et al., 1999
(1991–1992) (n = 492) in combined with Market
TRANSFAIR study, Basket analyses of
aged 18–65 y 100 foods collected
Men (n = 323) 1995–1996 plus other
total TFA 4.4 ± 2.0 1.4 ± 0.5 country-specific data
14:1t9 0.19 ± 0.10
16:1t9 0.30 ± 0.16
18:1t 3.2 ± 1.6
18:2t 0.47 ± 0.19
18:3t +20:1 0.16 ± 0.08
20:2t11,14 0.04 ± 0.06
22:1t 0.02 ± 0.02
Women (n = 169)
total TFA 3.6 ± 1.8 1.5 ± 0.5
14:1t9 0.16 ± 0.09
16:1t9 0.27 ± 0.16
18:1t 2.6 ± 1.3
18:2t 0.37 ± 0.17
18:3t +20:1 0.14 ± 0.08
20:2t11,14 0.03 ± 0.04
22:1t 0.02 ± 0.02
France Individual diet French population in 7-day diet records Hulshof et al., 1999
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Women (n = 881)
total TFA 1.9 ± 0.7 0.9 ± 0.2
14:1t9 0.26 ± 0.12
16:1t9 0.29 ± 0.13
18:1t 1.07 ± 0.40
18:2t 0.28 ± 0.10
18:3t +20:1 0.03 ± 0.02
20:2t11,14 0.02 ± 0.01
22:1t 0.00 ± 0.00
Greece Individual diet Greek population 24-h recall combined Hulshof et al., 1999
(1995) in TRANSFAIR with Market Basket
study, aged 23–64y analyses of 100 foods
Men (n = 141) collected 1995–1996
total TFA 1.2 ± 1.2 0.5 ± 0.4 plus other country-
14:1t9 0.04 ± 0.04 specific data
16:1t9 0.12 ± 0.14
18:1t 0.82 ± 0.89
18:2t 0.18 ± 0.18
TRANS FATTY ACIDS IN HUMAN NUTRITION
18:2t 0.23
18:3t +20:1 0.02
20:2t11,14 0.02
353
22:1t 0.00
Table 3. continued
354
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
The Individual diet Dutch population 2-day diet record Hulshof et al., 1999
Nether- (1992) in TRANSFAIR combined with Market
lands study, aged 19–64 y Basket analyses of 100
Men (n = 1822) foods collected 1995–
total TFA 4.8 ± 2.3 1.5 ± 0.6 1996 plus other
14:1t9 0.05 ± 0.00 country-specific data
16:1t9 0.14 ± 0.08
18:1t 3.9 ± 2.0
18:2t 0.55 ± 0.22
18:3t +C20:1 0.09 ± 0.06
20:2t11,14 0.0 ± 0.0
22:1t 0.02 ± 0.04
Women (n = 2203)
total TFA 3.8 ± 2.0 1.6 ± 0.7
14:1t9 0.04 ± 0.04
16:1t9 0.11 ± 0.06
18:1t 3.1 ± 1.8
18:2t 0.43 ± 0.19
18:3t +C20:1 0.07 ± 0.05
TRANS FATTY ACIDS IN HUMAN NUTRITION
Table 3. continued
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Spain Household Spanish men and 7d household con- Hulshof et al., 1999
consumption women (n = 21,155) sumption combined
(1991) in TRANSFAIR with Market Basket
study, aged 0–70+ y analyses of 100 foods
total TFA 2.1 0.7 collected 1995–1996
14:1t9 0.10 plus other country-
16:1t9 0.17 specific data
18:1t 1.40
18:2t 0.25
18:3t +C20:1 0.06
20:2t11,14 0.48
22:1t 0.03
Spain Dietary history Spanish healthy Dietary history com- van de Vijver et al.,
population as part bined with Market 2000
of TRANSFAIR Basket analyses of
study, aged 50–65 y 100 foods collected
Men (n = 36) 1995–1996 plus other
TRANS FATTY ACIDS IN HUMAN NUTRITION
Portugal Dietary history Portuguese healthy Dietary history com- van de Vijver et al.,
population in bined with Market 2000
TRANSFAIR Basket analyses of
study, aged 50–65 y 100 foods collected
Men (n = 41) 1995–1996 plus other
Total TFA 0.54 ± 0.20 country-specific data
14:1t9 0.04 ± 0.02
16:1t9 0.05 ± 0.02
18:1t 0.37 ± 0.14
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS
Table 3. continued
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Women (n = 39)
Total TFA 0.57 ± 0.26
14:1t9 0.04 ± 0.02
16:1t9 0.06 ± 0.03
18:1t 0.39 ± 0.18
18:2t 0.05 ± 0.02
22:1t 0.00 ± 0.01
Portugal Habitual diet Portuguese men Semi-quantitative food Lopes, et al., 2007
aged ≥40 y frequency ques-
(*data = % total fat tionnaire combined
intake) with Portuguese food
Myocardial infarction composition table
cases (n = 297) 1.27 ± 0.83*
TRANS FATTY ACIDS IN HUMAN NUTRITION
Population Control
(n = 310) 1.16 ± 0.78*
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS 359
4. Nordic Countries
Estimates of TFA intake in the Nordic countries published from the late 1990s
to the present are summarized in Table 4. The TRANSFAIR study (Hulshof
et al., 1999; van de Vijver, 2000) is the major study comparing TFA consump-
tion in these countries. Hulshof et al. (1999) reported that intakes for men in the
Nordic countries ranged from 2.3 g/day in Finland to 6.7 g/day in Iceland.
Intermediate values were reported for men in Denmark (2.9 g/day) and Norway
(4.8 g/day).
Jakobsen, et al. (2006, 2008) have studied the amount of ruminant fat
consumed by the Danish population. They report that adult men consume 2.0 g/
day and adult women consume 1.5 to 1.6 g/day of TFA from meat and dairy
products. With recent legislation targeting elimination of industrially produced
TFA from the Danish diet (Leth et al. 2006), these values could represent total
TFA consumption in Denmark.
In addition to Denmark, TFA consumption has decreased in recent years in
several of the other Nordic countries. Pedersen et al. (1998) have documented
this decrease in the Norwegian population. Trans fatty acid consumption has
decreased in Norway from 15 g/day/person in 1958 to 4 g/day/person in 1996.
A 2003 market survey conducted by Johansson et al. (2006) indicates that TFA
consumption from industrially produced sources is 1.6 g/day/person or 0.6en%.
5. Australia
Estimates of TFA intake in Australia published from the late 1990s to the
present are summarized in Table 5. Clifton et al. (2004) reported that in an
Australian adult population with a mean age of 56 years, the TFA intake was
Table 4. Estimates of trans fatty acid consumption in Nordic countries published late-1990s to present.
360
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Denmark Individual diet Danish population 7-day diet records Hulshof et al., 1999
(1995) in TRANSFAIR combined with Market
study, aged 19–64 y Basket analyses of
Men (n = 650) 100 foods collected
total TFA 2.9 ± 1.3 1.0 ± 0.4 1995–1996 plus other
14:1t9 0.21 ± 0.09 country-specific data
16:1t9 0.14 ± 0.07
18:1t 2.0 ± 1.0
18:2t 0.30 ± 0.13
18:3t + 20:1 0.09 ± 0.05
20:2t11,14 0.10 ± 0.06
22:1t 0.02 ± 0.01
Women (n = 702)
total TFA 2.3 ± 1.2 1.0 ± 0.5
14:1t9 0.16 ± 0.07
16:1t9 0.11 ± 0.06
18:1t 1.6 ± 0.8
18:2t 0.22 ± 0.10
18:3t + 20:1 0.06 ± 0.03
TRANS FATTY ACIDS IN HUMAN NUTRITION
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Women (n = 991)
total TFA 1.9 ± 0.9 0.9 ± 0.3
14:1t9 0.11 ± 0.05
16:1t9 0.11 ± 0.05
18:1t 1.3 ± 0.7
18:2t 0.21 ± 0.09
18:3t +20:1 0.06 ± 0.03
20:2t11,14 0.05 ± 0.08
22:1t 0.00 ± 0.00
Finland Dietary history Finnish healthy Dietary history com- van de Vijver et al.,
population in bined with Market 2000
TRANSFAIR study, Basket analyses of
aged 50–65 y 100 foods collected
Men (n = 38) 1995–1996 plus other
Total TFA 0.88 ± 0.29 country-specific data
14:1t9 0.04 ± 0.02
16:1t9 0.04 ± 0.02
18:1t 0.67 ± 0.25
TRANS FATTY ACIDS IN HUMAN NUTRITION
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Women (n = 45)
Total TFA 1.65 ± 0.82
14:1t9 0.08 ± 0.03
16:1t9 0.21 ± 0.09
18:1t 1.05 ± 0.65
18:2t 0.16 ± 0.05
22:1t 0.04 ± 0.05
Norway Norwegian population Pedersen et al., 1998
1958 15 5.5
1996 4
Norway Habitual diet Norwegian population Quantitative food Hulshof et al., 1999
(1993–1994) in TRANSFAIR frequency questionnaire
study, aged 19–64 y combined with Market
Men (n = 1257) Basket analyses of
total TFA 4.8 ± 2.9 1.5 ± 0.6 100 foods collected
14:1t9 0.17 ± 0.10 1995–1996 plus other
16:1t9 0.46 ± 0.26 country-specific data
18:1t 3.49 ± 2.40
18:2t 0.24 ± 0.12
TRANS FATTY ACIDS IN HUMAN NUTRITION
20:2t11,14 <0.05
22:1t <0.05
365
Table 4. continued
366
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Sweden Dietary history Swedish healthy Dietary history com- van de Vijver et al.,
population as part bined with Market 2000
of TRANSFAIR Basket analyses of
study, aged 50–65 y 100 foods collected
Men (n = 42) 1995–1996 plus other
Total TFA 0.95 ± 0.32 country-specific data
14:1t9 0.05 ± 0.02
16:1t9 0.06 ± 0.03
18:1t 0.72 ± 0.26
18:2t 0.09 ± 0.03
22:1t 0.00 ± 0.00
Women (n = 38)
Total TFA 1.01 ± 0.44
14:1t9 0.05 ± 0.02
16:1t9 0.06 ± 0.02
18:1t 0.77 ± 0.40
18:2t 0.09 ± 0.03
22:1t 0.00 ± 0.00
TRANS FATTY ACIDS IN HUMAN NUTRITION
Data expressed as total trans fatty acids, and as mean ± SD or as mean (range) unless otherwise specified.
Table 5. Estimates of trans fatty acid consumption in Australia published late-1990s to present (FFQ, food-frequency questionnaire; WFR,
weighed food record).
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Australia Individual diet Australian population, 300-item food- Clifton et al., 2004
(1995–1997) mean age 56 y frequency questionnaire
median intake and Australian food
3rd quintile 3.03 composition database
` (range) (1.55–5.46)
Myocardial infarction
cases (n = 209)
Total TFA 3.52 ± 1.82
18: 1t 1.21 ± 0.4
Control (n = 174)
Total TFA 3.01 ± 1.29
18:1t 1.17 ± 0.4
Australia Habitual/ Queensland men 129-item FFQ/WFR McNaughton
Individual diet (n = 18) and women and Australian fatty et al., 2007
(n = 25), aged 28–75 y acids in foods com-
FFQ 0.11 ± 0.31 position database
WFR 0.01 ± 0.01 (Mann, et al., 2003)
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS
367
368 TRANS FATTY ACIDS IN HUMAN NUTRITION
3.03 g/day in 1995 to 1997. Toward the end of the study, ‘no-trans’ margarines
were introduced in Australia, and these investigators had evidence that TFA
consumption decreased in their study population after 1996. McNaughton
et al. (2007) reported very low TFA intakes in men and women of Queensland,
Australia, based on a food frequency questionnaire or a weighed food record
combined with an Australian food database by Mann, et al. (2003). These
investigators report 0.11 and 0.01 g/day for TFA intake based on a food
frequency questionnaire or on weighed food records, respectively. Recent
regulations for food labelling of trans fat have been instituted in Australia and
New Zealand (Schrimpf-Moss & Wilkening, 2007).
6. Asian/Pacific Region
Estimates of TFA intake in the Asian/Pacific region published from the late
1990s to the present are summarized in Table 6. In the study of Zhou et al.
(2003), intakes of TFA in China and Japan were low, 0.2en% and 0.3en%,
respectively. The populations in both countries were middle-aged (40 to 59
years old), and it is possible that some younger people would have a greater
TFA intake as they adopt a more Western diet.
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
China Individual diet Chinese population 4 standardized 24-h Zhou et al., 2003
in the INTERMAP dietary recalls combined
study, aged 40–59 y with country-specific
Men (n = 416) 0.2 ± 0.4 food composition data-
Women (n = 423) 0.2 ± 0.3 base updated by the
Nutrition Coordinating
Center, University of
Minnesota
Japan Individual diet Japanese population 4 standardized 24-h Zhou et al., 2003
in the INTERMAP dietary recalls combined
study, aged 40–59 y with country-specific
Men (n = 574) 0.3 ± 0.2 food composition data-
Women (n = 571) 0.5 ± 0.3 base updated by the
Nutrition Coordinating
Center, University of
Minnesota
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS
369
Table 7. Estimates of trans fatty acid consumption in Central and South America published late-1990s to present.
370
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Costa Habitual diet Costa Rican people 135-item food- Baylin et al., 2002
Rica (n = 503) frequency question-
Total TFA 3.98 ± 1.85 4.76 ± 2.24 naire combined with
Trans 18:1 2.47 ± 1.23 2.94 ± 1.43 modified USDAfood
composition tables
Costa Individual diet Costa Rican 3-day food record and Monge-Rojas et al.,
Rica adolescents fatty acid content of 2005
(n = 275), Costa Rican foods
aged 12–19 y
Male (n = 144)
Rural (n = 75)
Total TFA 4.04 ± 0.78
16:1t 0.07 ± 0.01
18:1t 3.14 ± 0.44
18:2t 0.57 ± 0.69
Urban (n = 69)
Total TFA 4.96 ± 0.82
16:1t 0.10 ± 0.03
TRANS FATTY ACIDS IN HUMAN NUTRITION
8. Iran
A recent estimate of TFA intake in Iran is summarized in Table 8. The intake
of industrially produced TFA has been estimated by Mozaffarian et al. (2007)
to be 4.2% of energy. Even though this estimate does not include ruminant fats,
it is quite high compared to other countries, such as the USA, in which values
are about 2% of energy (Table 1). That TFA consumption in Iran is greater than
in most countries is confirmed by an analysis of human milk samples from 52
Western Iranian women. The trans-18:1 content of human milk was 11.3% of
total milk fatty acids, a value much higher than currently observed in North
American or European women (Craig-Schmidt, 2001).
Country Dietary data Population Trans fatty acid (TFA) intake Method used Reference
g/person/day energy % % total FA
Iran Habitual diet 7158 urban and Estimates of con- Mozaffarian et al.,
rural households sumption of industrial 2007
containing 35,924 TFA by detailed in-
individuals home assessments of
Industrial TFA 4.2 dietary intake and lab-
oratory analysis of
commonly consumed fats
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS
373
374 TRANS FATTY ACIDS IN HUMAN NUTRITION
in Hungary, Russia, Germany, The Netherlands, Portugal and Spain were more
moderate (averaging 6 to 8%). Denmark was the least with only 1–2% of the
total fat as TFA in the French fries. It is anticipated that as appropriate frying
fats become available and as fast-food chains respond to consumer pressure,
the trans fat content of fast-foods will decrease as have the trans contents of
packaged foods.
Within a few years, the TFA in the food supply should be limited to the
‘natural’ ruminant fats in meat and dairy products. This lower limit for Western
countries should be in the neighbourhood of 1 to 2 g/day and even lower in
countries where dairy products are not routinely consumed. The data of
Hulshof et al. (1999) can be used to calculate trans fat consumption from
ruminant sources (milk, cheese, meat and butter) as 0.9 in Finland to 2.1 in
Iceland. Other estimates, such as 1.7 g/day for ruminant fat consumption in the
Danish population (Jakobsen, et al., 2006) are within this range. Whether or
not TFA isomers in ruminant fats have the same effects as trans isomers in
partially hydrogenated vegetable oils is currently under investigation
(Weggemans et al., 2004; Chardigny et al., 2006; Willett and Mozaffarian,
2008; Motard-Belanger, et al., 2008; Chardigny, et al., 2008) and the issue is
discussed elsewhere in this book.
The trans fatty acid saga is near its end. These food ingredients have had a
long journey from being the new, popular shortening in bakery products in the
early twentieth century to being one of the ‘Bad Fat Boys’ of the American
Heart Association in the twenty-first century.
Acknowledgement
Appreciation is expressed to Elias J. Bungenstab for translation of Portuguese
material into English.
References
Allison, DB, Egan, K, Barraj, LM, Caughman, C, Infante, M and Heimbach, JT (1999)
Estimated intakes of trans fatty and other fatty acids in the US population. J. Am. Diet.
Assoc., 99, 166–174.
American Society for Clinical Nutrition /American Institute for Nutrition (ASCN/AIN) Task
Force on Trans Fatty Acids (1996), Position paper on trans fatty acids. Am. J. Clin. Nutr.,
63, 663–670.
ANVISA (Agência Nacional de Vigilância Sanitária), National Agency for Sanitary Moni-
toring Brazil. Mercosul Regulamento Téchnico Sobre Rotulagem Nutricional de Alimentos
Embalados. RDS No.360, 23 December 2003. (In Portuguese.)
Aro, A (1998) Epidemiology of trans fatty acids and coronary heart disease in Europe. Nutr.
Metab. Cardiovasc. Dis., 8, 402–407.
Aro, A, Kardinaal, AF, Salminen, I, Kark, JD, Riemersma, RA, Delgado-Rodriguez, M,
Gomez-Aracena, J, Huttunen, JK, Kohlmeier, L, Martin, BC, Martin-Moreno JM,
Mazaev, VP, Ringstad, J, Thamm, M, Van’t Veer, P, Kok, FJ (1995) Adipose tissue
isomeric trans fatty acids and risk of myocardial infarction in nine countries: the
EURAMIC study. Lancet, 345, 273–278.
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS 375
Bahrami, G and Rahimi, Z (2005) Fatty acid composition of human milk in Western Iran. Eur.
J. Clin. Nutr., 59, 494–497.
Baylin, A, Kabagambe, EK, Siles, X and Campos, H (2002) Adipose tissue biomarkers of
fatty acid intake. Am. J. Clin. Nutr., 76, 750–757.
Baylin, A, Kabagambe, EK, Ascherio, A, Spiegelman, D and Campos, H (2003) High 18:2
trans-fatty acids in adipose tissue are associated with increased risk of nonfatal acute
myocardial infarction in Costa Rican adults. J. Nutr., 133, 1186–1191.
Bertolino, CN, Castro, TG, Sartorelli, DS, Ferreira, SR and Cardoso, MA (2006) Dietary
trans fatty acid intake and serum lipid profile in Japanese-Brazilians in Bauru, São Paulo,
Brazil. Cadernos de Saúde Pública, 22, 357–364. (In Portuguese.)
Beynen, AC, Hermus, RJ and Hautvast, JG (1980) A mathematical relationship between the
fatty acid composition of the diet and that of the adipose tissue in man. Am. J. Clin. Nutr.,
33, 81–85.
Boué, C, Combe, N, Billeaud, C, Mignerot, C, Entressangles, B, Thery, G, Geoffrion, H,
Brun, JL, Dallay, D and Leng, JJ (2000) Trans fatty acids in adipose tissue of French
women in relation to their dietary sources. Lipids, 35, 561–566.
Brisson, J (1981) Lipids in Human Nutrition: An Appraisal of Some Dietary Concepts. Jack
K. Burgess Inc, Eaglewood, New Jersey, USA, pp.41–71.
Canadian Department of Health. (2003) Regulations amending and food and drug regulations
(Nutrition labeling, nutrient content claims and health claims). Canada Gazette, Part II,
137 (1), p.154, 1 January 2003.
Cantwell, MM, Flynn, MA, Cronin, D, O’Neill, JP and Gibney, MJ (2005a) Contribution of
foods to trans unsaturated fatty acid intake in a group of Irish adults. J. Hum. Nutr. Diet.,
18, 377–385.
Cantwell, MM, Gibney, MJ, Cronin, D, Younger, KM, O’Neill, JP, Hogan, L and Flynn, MA
(2005b) Development and validation of a food-frequency questionnaire for the determi-
nation of detailed fatty acid intakes. Public Health Nutrition, 8, 97–107.
Chardigny, JM, Wolff, RL, Mager, E, Sébédio, J-L, Martine, L and Juanéda, P (1995)
Trans mono- and polyunsaturated fatty acids in human milk. Eur. J. Clin. Nutr., 49,
523–531.
Chardigny, JM, Malpuech-Brugere, C, Dionisi, F, Bauman, DE, German, B, Mensink, RP,
Combe, N, Chaumont, P, Barbano, DM, Enjalbert, F, Bezelgues, J-B, Cristiani, I, Moulin,
J, Boirie, Y, Golay, P-A, Giuffrida, F, Sebedio, J-L, Destaillats, F (2006) Rationale and
design of the TRANSFACT project phase I: a study to assess the effect of the two different
dietary sources of trans fatty acids on cardiovascular risk factors in humans. Contempo-
rary Clinical Trials, 27, 364–373.
Chardigny, JM, Destaillats, F, Malpuech-Brugere, C, Moulin, J, Bauman, DE, Lock, AL,
Barbano, DM, Mensink, RP, Bezelgues, J-B, Chaumont, P, Combe, N, Cristiani, I, Joffre,
F, German, JB, Dionisi, F, Boirie, Y, and Sebedio, J-L. (2008) Do trans fatty acids from
industrially produced sources and from natural sources have the same effect on cardio-
vascular disease risk factors in healthy subjects? Results of the trans Fatty Acids
Collaboration (TRANSFACT) study. Am. J. Clin. Nutr., 87, 558–566.
Chen, ZY, Pelletier, G, Hollywood, R and Ratanayke, WM (1995a) Trans fatty acid isomers
in Canadian human milk. Lipids, 30, 15–21.
Chen, ZY, Ratnayake, WM, Fortier, L, Ross, R and Cunnane, SC (1995b) Similar distribution
of trans-fatty acid isomers in partially hydrogenated vegetable oils and adipose tissue of
Canadians. Can. J. Physiol. Pharmacol., 73, 718–723.
Clifton, PM, Keogh, JB and Noakes, M (2004) Trans fatty acids in adipose tissue and the food
supply are associated with myocardial infarction. J. Nutr., 134, 874–879.
Colón-Ramos, U, Baylin, A and Campos, H (2006) The relation between trans fatty acid
levels and increased risk of myocardial infarction does not hold at lower levels of trans
fatty acids in the Costa Rican food supply. J. Nutr., 136, 2887–2892.
376 TRANS FATTY ACIDS IN HUMAN NUTRITION
Combe, N, Boué, C and Entressangles, B (2000) Consommation en acides gras trans et risqué
cardio-vasculaire: Étude Aquitaine. Ol. Corps Gras, Lipides, 7, 17–24. (In French.)
Combe, N, Judde, A, Boué, C, Billeaud, C, Entressangles, B, Dallay, D, Leng, JJ and Baste,
JC (1998) Composition en acides gras trans du tissu adipeux d’une population francaise
et origines alimentaires de ces acides gras trans. Ol. Corps Gras, Lipides, 5, 142–148. (In
French.)
Craig-Schmidt, MC (1998) Worldwide consumption of trans fatty acids. In: Trans-Fatty
Acids in Human Nutrition (JL Sébédio and WW Christie, eds), The Oily Press,
Bridgwater, UK, pp.59–113.
Craig-Schmidt, MC (2001) Isomeric fatty acids: evaluating status and implications for
maternal and child health. Lipids, 36, 997–1006.
Craig-Schmidt, MC (2006) World-wide consumption of trans fatty acids. Atherosclerosis
Supplements, 7, 1–4.
Craig-Schmidt, MC (2007) Consumption of trans fatty acids in North America. In: Trans Fats
in Foods (GR List, D Kritchevsky, and N Ratnayake, eds), AOCS Press, Urbana. Illinois,
USA, pp.127–153.
Craig-Schmidt, MC, Weete, JD, Faircloth, SA, Wickwire, MA and Livant, EJ (1984) The
effect of hydrogenated fat in the diet of nursing mothers on lipid composition and
prostaglandin content of human milk. Am. J. Clin. Nutr., 39, 778–786.
Craig-Schmidt, MC and Teodorescu, CA (2008) Trans-fatty acids in foods. In: Fatty Acids
in Foods and Their Health Implications, Third edition (CK Chow, ed.), CRC Press, New
York, USA, pp.377–437.
DHHS/FDA, Department of Health and Human Services, Food and Drug Administration.
(2003) 21 CFR Part 101 (Docket No. 94P-0036). Food labeling: Trans fatty acids in
nutrition labeling, nutrient content claims and health claims. Washington, DC, USA, July
2003, p.254.
Dlouhý, P, Tvrzická, E, Stanková, B, Buchtiková, M, Pokorný, R, Wiererová, O, Bílková, D,
Rambousková, J and Andel, M (2002) Trans fatty acids in subcutaneous fat of pregnant
women and in human milk in the Czech Republic. Ann. NY Acad. Sci., 967, 544–547.
Eckel, RH, Borra, S, Lichtenstein, AH, Yin-Piazza, SY and Trans Fat Conference Planning
Group. (2007) Understanding the complexity of trans fatty acid reduction in the
American diet: American Heart Association Trans Fat Conference 2006: report of the
Trans Fat Conference Planning Group. Circulation, 115, 2231–2246.
Elias, SL and Innis, SM (2002) Bakery foods are the major dietary source of trans-fatty acids
among pregnant women with diets providing 30 percent energy from fat. J. Am. Diet.
Assoc., 102, 46–51.
Enig, MG, Atal, S, Keeney, M and Sampugna, J (1990) Isomeric trans fatty acids in the US
Diet. J. Am. Coll. Nutr., 9, 471–486.
Friesen, R and Innis, SM (2006) Trans fatty acids in human milk in Canada declined with the
introduction of trans fat food labeling. J. Nutr., 136, 2559–2561.
FSA, UK Food Standards Agency (2003) The Food Labeling Regulations 1996: Guidance
Notes. 2003. Internet address: www.food.gov.uk/foodindustry/guidancenotes/
labelregsguidance/foodlabelregsguid
Garland, M, Sacks, FM, Colditz, GA, Rimm, EB, Sampson, LA, Willett, WC and Hunter, DJ
(1998) The relation between dietary intake and adipose composition of selected fatty
acids in US women. Am. J. Clin. Nutr., 67, 25–30.
Glew, RH, Herbein, JH, Moya, MH, Valdez, JM, Obadofin, M, Wark, WA and Vanderjagt,
DJ (2006) Trans fatty acids and conjugated linoleic acids in the milk of urban women and
nomadic Fulani of northern Nigeria. Clin. Chim. Acta, 367, 48–54.
Harnack, L, Lee, S, Schakel, SF, Duval, S, Luepker, RV and Arnett, DK (2003) Trends in the
trans-fatty acid composition of the diet in a metropolitan area: the Minnesota Heart
Survey. J. Am. Diet. Assoc., 103, 1160–1166.
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS 377
Hayat, L, al-Sughayer, MA and Afzal, M (1999) Fatty acid composition of human milk in
Kuwaiti mothers. Comp. Biochem. Physiol. Part B: Biochemistry & Molecular Biology,
124, 261–267.
Hirsch, J, Farquhar, JW, Ahrens Jr, EH, Peterson, ML and Stoffel, W (1960) Studies of
adipose tissue in man: a microtechnic for sampling and analysis. Am. J. Clin. Nutr., 8,
499–511.
Hu, FB, Stampler, MD, Manson, JE, Rimm, E, Colditz, GA, Rosner, BA, Hennekens, CH and
Willett, WC (1997) Dietary fat intake and the risk of coronary heart disease in women.
New Eng. J. Med., 337, 1491–1499.
Hu, FB, Stampfer, MJ, Manson, JE, Grodstein, F, Colditz, GA, Speizer, FE and Willett, WC
(2000) Trends in the incidence of coronary heart disease and changes in diet and lifestyle
in women. New Eng. J. Med., 343, 530–537.
Hulshof, KF, van Erp-Baart, MA, Anttolainen, M, Becker, W, Church, SM, Couet, C,
Herman-Kunz, E, Kesteloot, H, Leth, T, Martins, I, Moreiras, O, Moschandreas, J,
Pizzoferrato, L, Rimestad, AH, Thorgeirsdottir, H, van Amelsvoort, JM, Aro, A, Kafatos,
AG, Lanzmann-Petithory, D and van Poppel, G (1999) Intake of fatty acids in Western
Europe with emphasis on trans fatty acids: the TRANSFAIR Study. Eur. J. Clin. Nutr.,
53, 143–157.
Hunter, JE (2006) Dietary trans fatty acids: review of recent human studies and food industry
responses. Lipids, 41, 967–992.
Innis, SM, Green, TJ and Halsey, TK (1999) Variability in the trans fatty acid content of foods
within a food category: implications for estimation of dietary trans fatty acid intake.
J. Am. Coll. Nutr., 18, 255–260.
Innis, SM and King, DJ (1999) Trans fatty acids in human milk are inversely associated with
concentrations of essential all-cis n–6 and n–3 fatty acids and determine trans, but not n–
6 and n–3, fatty acids in plasma lipids of breast-fed infants. Am. J. Clin. Nutr., 70,
383–390.
Innis, SM, Vaghri, Z and King, DJ (2004) n–6 Docosapentaenoic acid is not a predictor of
low docosahexaenoic acid status in Canadian preschool children. Am. J. Clin. Nutr., 80,
768–773.
International Life Sciences Institute (ILSI) Expert Panel on Trans Fatty Acids and Coronary
Heart Disease (1995) Trans fatty acids and coronary heart disease risk. Am. J. Clin. Nutr.,
62, 655S–708S.
International Life Sciences Institute (ILSI) Expert Panel on Trans Fatty Acids and Early
Development (1997) Trans fatty acids: infant and fetal development. Am. J. Clin. Nutr.,
66, 715S–736S.
Jakobsen, MU, Bysted, A, Andersen, NL, Heitmann, BL, Hartkopp, HB, Leth, T, Overvad,
K and Dyerberg, J (2006) Intake of ruminant trans fatty acids in the Danish population
aged 1–80 years. Eur. J. Clin. Nutr., 60, 312–318.
Jakobsen, MU, Overvad, K, Dyerberg, J and Heitmann, BL (2008) Intake of ruminant trans
fatty acids and risk of coronary heart disease. Int. J. Epidemiology, 37, 173–182.
Jensen, RG (1999) Lipids in human milk. Lipids, 34, 1243–1271.
Johansson, L, Borgejordet, A and Pedersen, JI (2006) Trans fatty acids in the Norwegian diet.
Tidsskrift for den Norske Legeforening,126, 760–763. (In Norwegian.)
Katan, MB (2006) Regulation of trans fats: the gap, the Polder, and McDonald’s French fries.
Atherosclerosis Supplements, 7, 63–66.
King, IB, Kristal, AR, Schaffer, S, Thornquist, M and Goodman, GE (2005) Serum trans-
fatty acids are associated with risk of prostate cancer in beta-carotene and retinol efficacy
trial. Cancer Epidemiology, Biomarkers & Prevention, 14, 988–992.
Kromhout, D, Menotti, A, Bloemberg, B, Aravanis, C, Blackburn, H, Buzina, R, Dontas, AS,
Fidanza, F, Giampaoli, S, Jansen, A, Karvonen, M, Katan, M, Nissinen, A, Nedeljkovic,
S, Pekkanen, J, Pekkarinen, M, Punsar, S, Rasanen, L, Simic, B and Toshima, H (1995)
378 TRANS FATTY ACIDS IN HUMAN NUTRITION
Dietary saturated and trans fatty acids and cholesterol and 25-year mortality from
coronary heart disease: The Seven Countries Study. Preventive Medicine, 24, 308–315.
Lee, S, Harnack, L, Jacobs, DR Jr, Steffen, LM, Luepker, RV and Arnett, DK (2007) Trends
in diet quality for coronary heart disease prevention between 1980–1982 and 2000–2002:
The Minnesota Heart Survey. J. Am. Diet. Assoc., 107, 213–222.
Lemaitre, RN, King, IB, Mozaffarian, D, Sotoodehnia, N, Rea, TD, Kuller, LH, Tracy, RP
and Siscovick, DS (2006) Plasma phospholipid trans fatty acids, fatal ischemic heart
disease, and sudden cardiac death in older adults: the cardiovascular health study.
Circulation, 114, 209–215.
Lemaitre, RN, King, IB, Patterson, RE, Psaty, BM, Kestin, M and Heckbert, SR (1998)
Assessment of trans-fatty acid intake with a food frequency questionnaire and validation
with adipose tissue levels of trans-fatty acids. Am. J. Epidemiology, 148, 1085–1093.
Lemaitre, RN, King, IB, Raghunathan, TE, Pearce, RM, Weinmann, S, Knopp, RH, Copass,
MK, Cobb, LA and Siscovick, DS (2002) Cell membrane trans-fatty acids and the risk
of primary cardiac arrest. Circulation, 105, 697–701.
Leth, T, Bysted, A, Hansen, K and Ovesen, L (2003) Trans fatty acid content in Danish
margarines and shortenings. J. Am. Oil Chem. Soc., 80, 475–478.
Leth, T, Jensen, HG, Mikkeksen, AA, and Bysted, A (2006) The effect of the regulation on
trans fatty acid content in Danish food. Atherosclerosis Supplements 7, 53–56.
Liu, X, Schumacher, FR, Plummer, SJ, Jorgenson, E, Casey, G and Witte, JS (2007) Trans-
fatty acid intake and increased risk of advanced prostate cancer: modification by
RNASEL R462Q variant. Carcinogenesis, 28, 1232–1236.
Lopes, C, Aro, A, Azevedo, A, Ramos, E and Barros, H (2007) Intake and adipose tissue
composition of fatty acids and risk of myocardial infarction in a male Portuguese
community sample. J. Am. Diet. Assoc., 107, 276–286.
Lopez-Garcia, E, Schulze, MB, Manson, JE, Meigs, JB, Albert, CM, Rifai, N, Willett, WC
and Hu, FB (2004) Consumption of (n–3) fatty acids is related to plasma biomarkers of
inflammation and endothelial activation in women. J. Nutr., 134, 1806–1811.
Lopez-Garcia, E, Schulze, MB, Meigs, JB, Manson, JE, Rifai, N, Stampfer, MJ, Willett, WC
and Hu, FB (2005) Consumption of trans fatty acids is related to plasma biomarkers of
inflammation and endothelial dysfunction. J. Nutr., 135, 562–566.
Lu, M, Taylor, A, Chylack, LT Jr, Rogers G, Hankinson, SE, Willett, WC, Jacques, PF (2005)
Dietary fat intake and early age-related lens opacities. Am. J. Clin. Nutr., 81, 773–779.
Mann, NJ, Sinclair, AJ, Percival, P, Lewis, JL, Meyer, BJ and Howe, PRC (2003) Develop-
ment of a database of fatty acids in Australian foods. Nutr. Diet., 60, 42–45.
McNaughton, SA, Hughes, MC and Marks, GC (2007) Validation of a FFQ to estimate the
intake of PUFA using plasma phospholipid fatty acids and weighed foods records. Brit.
J. Nutr., 97, 561–568.
Mensink, RP and Katan, MB (1990) Effect of dietary trans fatty acids on high-density and
low-density lipoprotein cholesterol levels in healthy subjects. New Eng. J. Med., 323,
439–445.
Mojska, H, Socha, P, Socha, J, Soplinska, E, Jaroszewska-Balicka, W and Szponar, L (2003)
Trans fatty acids in human milk in Poland and their association with breastfeeding
mothers’ diets. Acta Paediatrica, 92, 1381–1387.
Monge-Rojas, R, Campos, H and Fernández Rojas, X (2005) Saturated and cis- and trans-
unsaturated fatty acids intake in rural and urban Costa Rican adolescents. J. Am. Coll.
Nutr., 24, 286–293.
Mosley, EE, Wright, AL, McGuire, MK and McGuire, MA (2005) Trans fatty acids in milk
produced by women in the United States. Am. J. Clin. Nutr., 82, 1292–1297.
Motard-Belanger, A, Charest, A, Grenier, G, Paquin, P, Chouinard, Y, Lemieux, S, Couture,
P, and Lamarche, B (2008) Study of the effect of trans fatty acids from ruminants on blood
lipids and other risk factors for cardiovascular disease. Am. J. Clin. Nutr., 87, 593–599.
EVOLUTION OF WORLDWIDE CONSUMPTION OF TRANS FATTY ACIDS 379
KOENRAAD DUHEM
“Trans fatty acids means the total number of unsaturated fatty acids where one or more
of the double bonds are in the trans configuration and declared as trans fat.”
381
382 TRANS FATTY ACIDS IN HUMAN NUTRITION
2. France
In France too, a similar solely chemical definition was adopted in February
2005 by the Agence Française de Sécurité Sanitaire des Aliments (AFSSA,
2005).
4. Denmark
In Denmark, TFA are defined as:
“The sum of all fatty acid isomers with 14, 16, 18, 20 or 22 carbon atoms and one or
more trans double bonds, i.e. C14:1, C16:1, C18:1, C18:2, C18:3, C20:1, C20:2,
C22:1, C22:2 fatty acid trans isomers, but only polyunsaturated fatty acids with
methylene interrupted double bonds.”
5. Canada
In Canada, TFA means:
“Unsaturated fatty acids that contain one or more isolated or non-conjugated double
bonds in a trans configuration”
6. USA
The definition of TFA in the USA is:
LEGISLATION RELATING TO TRANS FATTY ACIDS 383
“Unsaturated fatty acids that contain one or more isolated (i.e, non-conjugated) double
bonds in a trans configuration.”
“all geometrical isomers of mono- and poly-unsaturated fatty acids having non
conjugated carbon-carbon double bounds interrupted by at least one methylene group
in trans configuration”.
On its 26th session in November 2004 this definition evolved slightly and in
addition incorporated a mistake because the methylene group was designated
–CH2–CH2– where it should have been –CH2–.
Many regulatory definitions of TFA, while not specifically excluding rumi-
nant TFA, exclude fatty acids with conjugated bonds, even though these acids
have double bonds in trans configuration. These definitions stem from the view
that regulatory definitions try to a certain extent to identify adequately the fatty
acids targeted by the regulation.
“Trans fatty acids (TFA) are unsaturated fatty acids that have at least one double bond
in the trans configuration. Some polyunsaturated TFA have a conjugated structure (e.g.
conjugated linoleic acid [CLA] in milk fat), i.e. have double bonds which are not
separated by a methylene group, but most have isolated (non-conjugated) double
bonds”.
At the European Commission level the TFA issue is currently being dis-
cussed in the context of the ‘Nutrient profiles’ and for labelling purposes, but
no regulatory definition has been proposed yet (as at November 2007). How-
ever, release has been announced. The fact that regulatory definitions exclude
CLA acknowledges differences between TFA, but only to a limited extent.
B. Regulatory decisions
1. CODEX/WHO/FAO
The Codex Alimentarius Commission (Codex) was created in 1963 to develop
food standards, guidelines and related texts such as codes of practice under the
Joint Food and Agriculture Organization/World Health Organisation Food
384 TRANS FATTY ACIDS IN HUMAN NUTRITION
“If the provisions for labelling of, and claims for, trans-fatty acids do not affect a
marked reduction in the global availability of foods containing trans-fatty acids
produced by processing of oils and by partial hydrogenation, consideration should be
given to the setting of limits on the content of industrially produced trans-fatty acids
in foods.”
“…to assess the risks and the health effects of excessive intakes of fat and fatty acids,
in particular total fat, saturated fatty acids and trans fatty acids”.
• General historical background of the work related to TFA and the Global
Strategy.
LEGISLATION RELATING TO TRANS FATTY ACIDS 385
3. UK
In the UK, the Food Labelling Regulations require hydrogenated fat to be
386 TRANS FATTY ACIDS IN HUMAN NUTRITION
identified as such in the ingredient list on the label when it has been used as an
ingredient in food. However, if hydrogenated fat is part of a compound
ingredient that makes up less than 25% of the finished product it is not required
to be mentioned in the ingredient list. In the UK, it appears there are no plans
at this stage to undertake regulatory action over and above current promotion
of a balanced diet within which edible oils of all types should be consumed
sparingly.
4. European Union
In reaction to the Danish decision and taking notice that it fostered observations
from other EU member states with diverging views upon the issue, the
European Commission decided to request the EFSA to issue a scientific
opinion on the presence of TFA in foods and on the effect on human health of
the consumption of TFA. In this context the Authority was asked:
• to take into account all trans fatty acids in foods, including the ingredients
of food products, both those that are naturally present, such as in certain
animal fats, and those occurring as a result of manufacturing processes,
such as hydrogenation of oils;
• to advise whether the evidence indicates any specific effects on health of
trans fatty acids, whether the effects, if any, differ according to the food
source, and how the effects, if any, compare to effects on health of other
types of fatty acids;
• to advise, if there are effects on health, whether the effects are associated
with a specific level of intake of trans fatty acids in the context of the overall
diet.
• to advise if there are any methods of analysis that can distinguish between
trans fatty acids which are naturally present in fats and those formed during
the processing of fats, oils or foods.
At the time of the opinion formulation the experts stated:
“In most of the human intervention studies monounsaturated TFA from hydrogenated
vegetable oils were evaluated. No human intervention studies have been carried out to
evaluate the effects of TFA from ruminant fat, and indeed such studies are not
practicable. Thus it is not possible to determine whether there are differences between
TFA from ruminant fat and TFA from hydrogenated vegetable oils in their effects on
metabolic risk parameters such as LDL-C or HDL-C”.
This prudent statement prompted the European Commission to wait for more
scientific evidence before taking any decisions on labelling or banning TFA.
In early 2007, Members of the European Parliament worked out a ‘Written
declaration on trans fatty acids’ calling on the European Commission and
Council to increase awareness of the dangers of TFA by introducing a manda-
LEGISLATION RELATING TO TRANS FATTY ACIDS 387
5. USA
On 11 July 2003, the US Food and Drug Administration (FDA) issued a
regulation requiring that TFA be declared on the nutrition label of conventional
foods and dietary supplements on a separate line immediately under the line for
the declaration of saturated fatty acids on the Nutrition Facts panel of foods and
some dietary supplements (Food and Drug Administration, 2003). The new
labeling rule allowed for immediate voluntary compliance with mandatory
compliance by 1 January 2006. The regulation allows trans fat levels of less
than 0.5 g per serving to be labeled as 0 g per serving. The FDA did not approve
nutrient content claims such as “trans fat free” or “low trans fat”, as they could
not determine a “recommended daily value”.
It is worthwhile relating that this amendment to FDA regulation on nutrition
labelling came in response to a citizen petition from the Center for Science in
the Public Interest (CSPI) which was received by the FDA dated 14 February
1994. Until 1992, CSPI had defended TFA in preference to saturated fatty
acids.
The US regulatory authorities decided not to exclude TFA derived from
rumen hydrogenation from labelling requirements; consequently, dairy prod-
ucts must be appropriately labelled. However, TFA with conjugated bonds are
not included because they do not meet the US regulatory definition of TFA.
The state of California has recently announced a new legislation which will
388 TRANS FATTY ACIDS IN HUMAN NUTRITION
ban from 1 January 2010 the use of trans fats in oil, shortening and margarine used
in spreads or for frying in restaurants, and by 2011 in all retail baked goods.
Packaged foods will be exempt. Several major US towns, such as New York and
Seattle, have already adopted similar bans on trans fat use in restaurants.
6. Canada
Trans fats in Canadian foods have been reviewed by the Trans Fat Task Force
in its report ‘TRANSforming the food supply’ submitted to Canada’s Minister
of Health in June 2006 (Health Canada, 2006).
The Canadian Food and Drug Regulations (FDR) specifically prescribe what
information must be displayed on a food label. In December 2002, the Food and
Drug Regulations were amended to make nutrition labelling mandatory on
most food labels, to update requirements for nutrient content claims, and to
permit the use of diet-related health claims on foods (Canada Gazette, 2003) .
As a result, nutrition labelling became mandatory for most pre-packaged foods
in December 2005, with smaller businesses having until December 2007 to
comply with the new regulations. The TFA content of a food is considered core
nutrition information and must be declared in a Nutrition Facts table.
With regard to TFA, the Danish example triggered the Canadian House of
Commons to introduce and vote a motion to ban TFA. In December 2005,
Health Canada required that food labels list the amount of trans fat in the
nutrition facts section for most foods. Products with less than 0.2 g of trans fat
per serving may be labeled as free of trans fats. Other claims are allowed such
as “Reduced in trans fatty acids” provided the food is processed, formulated,
reformulated or otherwise modified, without increasing the content of satu-
rated fatty acids, so that it contains at least 25% less trans fatty acid. The claim
“Lower in trans fatty acids” is also permitted provided the food contains at least
25% less TFA and the content of saturated fatty acids is not higher per reference
amount of the food, than the reference amount of the reference food of the same
food group. These labelling allowances are not widely known and controversy
over truthful labelling has arisen; they are not believed to protect consumers
from high exposure.
Presently, TFA quantities on labels include naturally-occurring trans fats.
CLA isomers are not included in the label declaration because they fall outside
the definition.
In mid-2007, the federal government announced its intention to regulate
trans fats to new standards. These would be based on the June 2006 outcomes
of a ‘Trans Fats Task Force” co-chaired by Health Canada and the Heart and
Stroke Foundation of Canada. The Task Force had a mandate to develop
recommendations and strategies to effectively eliminate or reduce processed
trans fats in Canadian foods to the lowest level possible. It recommended the
following.
LEGISLATION RELATING TO TRANS FATTY ACIDS 389
Health Canada website. The second set of Trans Fat Monitoring data shows an
encouraging downward trend.
7. Australia
8. Asia
In 2007, the Korea Food and Drug Administration adopted plans to designate
“zero-trans fat” “labeling on food products that contain less than 0.5 g of trans
fatty acids per portion and “non-trans fat” for those with less than 0.2 g.
Korea and Japan, whose citizens eat less TFA than Western countries, have
been less active in legislation efforts. But as the harmful effects of TFA have
been widely publicized around the world, Korea has revised food labeling
regulations to specify cholesterol, trans fats and saturated fat content and is the
first Asian country to do so.
The law takes effect in December 2007. By this date the Korean FDA had
also developed criteria for claims of “low trans fatty acids” and “free of trans
fatty acids”.
In 2007, the Taiwanese Department of Health implemented mandatory
labeling of TFA on pre-packaged foods. Trans fat must be expressed in g per
100 g or per serving. Nutrients content for trans fat may be labeled “0” if
nutrient content is less than 0.3 g for 100 g of solid (semi-solid) food or 100 ml
of liquid food.
target the issue at its main source. The trend is also toward considering
naturally occurring TFA as not having the same deleterious effects as the
industrial isomers, at the concentration at which they are found naturally.
Scientific investigation of the biological role of TFA started only very
recently. TFA exist as natural compounds, albeit in small amounts compared to
the cis isomers; they seem to be subject to selection by nature, judging from the
very specific pattern they display compared to the random pattern produced by
catalytic hydrogenation of vegetable oil. Some families of those fatty acids are
present in animal products or vegetables in amounts that compare to the levels
of some biologically active fatty acid families such as omega-3 fatty acids. The
experimental study of their effects – in vitro, on animal models, or in humans
– have so far been hindered by the limited availability of the pure isomers. Most
studies have so far used mixtures of isomers, which resulted in limited or
confusing conclusions.
New, forthcoming studies using pure isomers, at physiological levels, are
starting to demonstrate some coherent effects of those trans compounds. Some
positions of the trans double bounds in the carbon chain of fatty acids seem to
be favoured by biological selection (trans-11 double bound) as are also specific
non-methylene-interrupted double bonds.
It is hoped that these studies will shed light on a broader range of physiologi-
cally active and beneficial fatty acids and help the regulatory authorities take
decisions on a more scientifically founded basis.
References
AFSSA (2005) Risques et bénéfices pour la santé des AG trans apportés par les aliments —
recommandations. Agence Française de Sécurité Sanitaire des Aliments. Online:
www.afssa.fr
Canada Gazette (2003) Part II, Vol.137, No.1 (1 January 2003).
Commission of European Communities (2008) Com2008(40) Final. Proposal for a Regula-
tion of the European Parliament and of the council on the provision of food information
to consumers. Online: ec.europa.eu/food/food/labellingnutrition/foodlabelling/publica-
tions/proposal_regulation_ep_council.pdf
European Food Safety Authority (2004) Opinion of the Scientific Panel on Dietetic Products,
Nutrition and Allergies on a request from the Commission related to the presence of trans
fatty acids in foods and the effect on human health of the consumption of trans fatty acids
(adopted on 8 July 2004). EFSA Journal, 81 , 1–49. Online: www.efsa.eu.int/science/
nda/nda_opinions/588/opinion_nda09_ej81_tfa_en1.pdf
Food and Drug Administration, USA (2003) CFR Part 101. Food Labeling: Trans Fatty Acids
in Nutrition Labeling: Consumer Research to Consider Nutrient Content and Health
Claims and Possible Footnote or Disclosure Statements: Final Rule and Proposed Rule.
National Archives and Records Administration. Federal Register, 68 (133), pages
41433–41506, 11 July 2003.
Food Standards Australia New Zealand, FSANZ (2007) Trans fatty acids in the New Zealand
and Australian food supply. Review Report, FSANZ. Online: www.foodstandards.gov.au/
_srcfiles/Transfat%20report_CLEARED.pdf
394 TRANS FATTY ACIDS IN HUMAN NUTRITION
Health Canada (2006) TRANSforming the food supply. Report by Canada’s Trans Fat Task
Force (ISBN 0-662-43689-X). Online: www.hc-sc.gc.ca/fn-an/alt_formats/hpfb-dgpsa/
pdf/nutrition/tf-gt_rep-rap_e.pdf (Accessed 11 November 2007)
Stender S, Dyerberg J and Astrup A (2006) High levels of industrially produced trans fat in
popular fast foods. New Engl. J. Med., 354, 1650–1652.
Stender, S, Dyerberg, J and Astrup, A (2006). Consumer protection through a legislative ban
on industrially produced trans fatty acids in foods in Denmark. Scandinavian Journal of
Food & Nutrition, 50, 155–160.
World Health Organisation (2007) Draft action plan for implementation of Global Strategy
on diet, Physical activity and Health. Alinorm 07/30/26 and CL/2006/44 -CAC
CHAPTER 15
Consumer concerns and risk perception related
to trans fatty acids
CLOTILDE AUBERTIN
Only ten years ago few people would have anticipated or even imagined the
enormous impact on our society of worldwide communication over the Internet,
and the profound effect that it has since had. Nowadays, consumers use this
technology daily to help them to gauge potential risks. In the case of trans fatty
acids (TFA), it would be relevant to measure the impact of Internet traffic
pertaining to its consumers. In this chapter, we are mainly concerned with the
following questions. What is the definition of ‘consumer’ in the increasing
information flow? What are consumers’ concerns in terms of foods and health?
How do they clarify such concerns? As information can now be obtained so
rapidly and easily, could the Internet be considered as a relevant way to identify
and measure risk perception by consumers? And how are TFA perceived in this
context?
Figure 1. Interaction between entities generating information perceived and conveyed by consumer.
group relationships as found in the real world (Huffaker, 2004). Some web sites
become influential as they are the most read, the most quoted, and the most
discussed. Blogs created by nutritionists have more weight in the ‘blogosphere’;
as an example, ‘Youtube’ allows dieticians and nutritionists to make their own
video programmes on health benefits or bad effects of nutrients. Later in this
chapter we will analyse how blogs tackle TFA.
The Internet plays an important role in risk communication and the forma-
tion of consumers’ views on an issue. It is an information source that helps the
public to form an opinion on risk (Yankelovich, 1991). Bennett (1999) sug-
gests that while media coverage may in fact amplify the public’s interest in an
issue, it does not create it. Bennett goes on to say that “a good story is one in
which public and media interests reinforce each other”.
The Internet tends to highlight existing concerns, uncertainties and con-
flicts. It rarely questions the legitimacy of any source and presents all sources
on a somewhat equal footing. In this sense, the Internet’s role might be
considered to be ‘non-judgemental’. However, some universities such as the
University of Kentucky, USA, and more particularly its College of Agricul-
ture, allow internet readers the opportunity to use a variety of learning modes
and information retrieval methods. The University’s Agripedia project and
website offer not only instructional material, but also increase the abilities of
users to consider information with more objectivity (see Agripedia refer-
ence).
Most online communities grow slowly at first, due in part to the fact that
the strength of motivation for contributing is usually proportional to the size
of the community. As the size of the potential audience increases, so does
the attraction of writing and contributing. This, coupled with the fact that
organizational culture does not change overnight, means that creators can
expect slow progress at first with a new virtual community. However, as
more people begin to participate, it creates a virtuous cycle in which more
participation brings more participants. Many online community members
describe their participation as “addictive” (Bishop, 2007). The growth in
community adoption is often forecast (estimating the number of users in the
community) by use of the Bass diffusion model. This is a mathematical
formula originally conceived by Frank Bass (1969) to describe the process
by which new products get adopted as an interaction between users and
potential users.
able to deal with the TFA problem. It took ten years to clarify the situation and
reduce the risk perception of consumers.
During the CSPI story, several consumer groups gave warnings about the
side-effects of consuming too much TFA and explained physiological mecha-
nisms and the regulatory situation using creditable experts on their websites.
Furthermore, heart health foundations grant research projects on this topic in
order to understand the impacts of TFA on human health. Since the 1990s,
several governing bodies, scientific associations and consumer interest groups
have expressed concerns about TFA (e.g. the International Association of
Consumer Food Organizations, IACFA; the European Food Information Coun-
cil, EUFIC; the Trans Atlantic Consumer Dialogue; the Australian Consumers
Association). Turning to science, policy formation should be underpinned by
reliable scientific evidence. However, this remains a potential problem as the
scientific community is still often viewed as being remote from the general
public. In the pursuit of scientific progress, the common interests between
science and society become vital for the benefit of public health.
Consumers International (London, UK) is an independent global campaign-
ing voice for consumers. With over 220 member organizations in 115 countries,
its goal is to build a powerful international consumer movement to help protect
and empower consumers everywhere. Through the lobbying of countries by
Consumers International and its member organizations, it has achieved a
resolution that has committed the World Health Organization (WHO) to:
“promote responsible marketing including the development of a set of recom-
mendations on marketing of foods and non-alcoholic beverages to children”.
Consumers International takes over the message from WHO for explaining that
the exposure to the commercial promotion of foods high in saturated fat, TFA,
sugars or salt has a direct effect on children wanting and eating these unhealthy
foods.
• Consumers are aware of certain food safety issues, especially the conse-
quences of TFA on heart health.
• Although various laws regulate types and amounts that are allowed in
402 TRANS FATTY ACIDS IN HUMAN NUTRITION
Figure 2. Analysis and interpretation of the most frequently occurring words cited on Internet blogs
since 2007 related to trans fatty acids.
foods, consumers say TFA are one of the greatest food-associated hazards
at present.
• Awareness of consequences is dependent on education and age; the more
educated and older the consumers are, the more aware they are of food
safety issues.
• According to some bloggers, exposure to TFA may be diminished by
consuming organic food, purchasing food at a reliable source or consuming
food of known origin.
We can conclude that the understanding of food safety issues among
consumers is low, which calls for information campaigns to raise consumer
education.
In addition to these blogs, there are forums, chatrooms or discussion lists,
which offer the opportunity to exchange information and remain an important
Internet activity. Surely, consumers do not access these virtual spheres for the
sole pleasure of denouncing abuse (related to consumer rights, for instance).
Online communities have been identified by the Internet founders as “common
interest communities” (Licklider & Taylor, 1990). These communities, very
numerous and often enduring on the Internet, may exchange experiences or
expertise. In the case of TFA, heterogeneous typology of online communities
is observed. In this context, typology is the study or systematic classification of
CONSUMER CONCERNS AND RISK PERCEPTION RELATED TO TFA 403
first find ways of segregating individual differences and needs and include the
public’s real concerns in the information provided (Frewer et al., 1998).
Consumption
1. 2. 3. 4.
No information Information Decreasing media Persisting anxiety
communicated communicated attention or Food safety
concerns
Figure 4. Comparison of consumer reaction to a food scare and the perception of adverse effects of
trans fatty acids consumption (adapted from Beardworth and Keil, 1996 and Mazzochi, 2004).
to understand the context and its consequences. Online debates do not meet the
standards of debates in the public sphere. In other words, a debate is sought
between equals where rational arguments prevail over a common position.
Internet debates would only match the first feature (Poster, 1997). However,
discussions about TFA by Web users do indeed exchange on an equal footing
but this exchange may not be reasoned with creditable evidence.
For a wider topic than TFA, the debate usually does not move towards the
elaboration of a common position but rather splinters into many contradictory
viewpoints. This splintering is further reinforced by the fact that users’ identi-
ties are vague and shifting. Not only do contributors use aliases and create for
themselves a virtual identity but they may also change their identity or have
several different identities. The coexistence of identities, which was studied by
Turkle (1997), appears to be one of the major causes of difficulty in reaching a
common viewpoint by online communities. In real life, when dealing with each
other face to face, each speaker can perceive the complex arguments of the
other and still reach an agreement. By contrast, virtual communities foster a
multiplicity of fixed viewpoints rather than flexibility (Kolko & Reid, 1998).
Fortunately, TFA offers a simpler situation because the topic is precise and
it concerns specific communities even if the problem affects everybody.
Opinions on the Internet converge on a similar sense and there emerges a
common wish and action tailored to the problem itself.
Figure 5. A timeline of trans fats from the introduction of hydrogenation to the launch of zero-trans
fats.
opportunity to open new markets. They explain that the effects of TFA on
human health represent the biggest issue facing the edible oils and fats industry
today. Edible oils and fats will also be used on a wide scale in diet therapy,
where they can halt the progress of diseases or reverse the damage caused by
illnesses.
Scientific research is growing faster partly because the scientific community
has benefited form easier information access in recent years.
Media coverage is sometimes difficult to predict and explain. Two effects
are well known: the term ‘snowball effect’ has been used to describe a story that
has become established and where increased frequency of coverage results in
media competition for interest; the ‘ripple effect’ is a form of risk amplification
and it often occurs when stories with an original focus on a single specific risk
are expanded to include related issues. The example of TFA illustrates how risk
perception could be ‘amplified’, or extended, by media coverage to create new
concerns about other fats and oils. Often these effects, which originate from
routine media coverage, can lead to what is termed by some as a ‘media event’.
It has been suggested that media coverage is driven largely by rarity, novelty
and commercial viability but, unfortunately, very little by risk evaluation.
and growth in usage since 2000 is 306%. The Internet is used by 74% of the
population in North America, 60% in Oceania/Australia and 48% in Europe.
The majority of Internet users interested in nutrition are active on the Web.
Groups, such as scientists, authorities, foundations and associations, as well as
individuals, share their expertise, opinions and web video and thus amplify the
mass of information on health effects of TFA. The TFA issue, in spite of the
relative level of knowledge expected from consumers, was taken into account
without significant drawbacks. In conclusion, the Internet might increasingly
constitute one way of evaluating risk as perceived by consumers. The impact on
them is potentially measurable.
Analysis of consumer perception regarding a specific food issue through the
Internet could offer one approach. For example, the issue of intense sweeteners
might be examined. In fact, more than ever, people are consuming large
amounts of sugar in their daily diet, which adds extra calories, which in turn
could cause weight gain. Artificial sweeteners are the subject of several stories
or comments, presented in the press and on the Internet, claiming that they
cause a variety of health problems, including cancer. Numerous studies con-
firm that artificial sweeteners are safe for the general population. However
‘Aspartame’ does carry a cautionary note, particularly for people who have the
rare hereditary disease phenylketonuria.
New scientific investigations are, or could be, undertaken with natural
intense sweeteners extracted from plants. Moreover science is progressing and
food companies are working to integrate these into products. Thus, it might be
interesting to measure consumer perceptions on the consumption of artificial
sweeteners or introduction of natural intense sweeteners into food products.
Finally, we can now see how risk perception data regarding these issues might
be captured from the Internet as already done for TFA.
References
Agripedia (1996) College of Agriculture, University of Kentucky, USA. Online:
www.ca.uky.edu/agripedia
Akrich, M and Méadel, C (2002) Prendre ses médicaments, prendre la parole: les usages des
médicaments par les patients dans les listes de discussion électroniques [To take a pill,
to take a stand: arguing about medication on electronic patient discussion lists], Sciences
sociales et santé, 1, 89–116
Auray, N (2004) La régulation de la connaissance: arbitrage sur la taille et gestion aux
frontières dans la communauté Debian [The regulation of knowledge: size and scope
control for the Debian community], Revue d’économie politique, March 2004, pp.160–
182.
Auray, N (2005) Le sens du juste dans un noyau d’experts: debian et le puritanisme civique
[Debian and civic puritanism]. In: Internet, une utopie limitée. Nouvelles régulations,
nouvelles solidarités [Internet a Limited Utopia. New Regulations, New Solidarities]
(Proulx, S, Massit-Folléa, F and Conein, B, eds) Les presses de l’Université Laval,
Québec, Canada, pp.71–94.
412 TRANS FATTY ACIDS IN HUMAN NUTRITION
Bass, F (1969). A new product growth model for consumer durables. Management Science,
15 (5), 215–227.
Bennett, P (1999) Understanding responses to risk: some basic findings. In: Risk Communi-
cation and Public Health (Bennett, P and Calman, K, eds), Oxford University Press, New
York, USA, pp.3–19.
Bishop, J (2007) Increasing participation in online communities: a framework for human–
computer interaction. Computers in Human Behavior, 23, 1881–1893.
Brashers, DE, Goldsmith, D J, and Hsieh, E (2002) Information seeking and avoiding in
health contexts. Human Communication Research, 28, 258–271.
Consumers International, London, UK; Online: www.consumersinternational.org
EFSA (2006) ‘The Metabolic Syndrome Summit’, European Food Safety Authority, London,
UK.
Fein, SB, Jordan, C and Levy, AS (1995) Foodborne illness: perceptions, experience, and
preventive behaviors in the United States. Journal of Food Protection, 58, 1405–1411.
Fischhoff, B, Watson, SR, and Hope, C (1984) Defining risk. Policy Sciences, 17, 123–139.
Flichy, P (2007) Internet Imaginaire, Chapter 7: On Cyberdemocracy. MIT Press, Cam-
bridge, Massachusetts, USA.
Flynn, L, Slovic, P and Mertz, CK (1993) Decidedly different: expert and public views of
risks from a radioactive waste repository. Risk Analysis, 13, 643 648.
Food Institute (1996) Population trends 1996–2050. Food Institute, Fair Lawn, New Jersey,
USA, p.31.
Frewer, LJ, Howard, C, Hedderley, D and Shepherd, R (1998) Methodological approaches
to assessing risk perceptions associated with food-related hazards. Risk Analysis, 18, 95–
102.
Frewer, LJ, Raats, MM and Shepherd, R (1994) The interrelationship between perceived
knowledge, control and risk associated with a range of food related hazards targeted at
the self, other people and society. Journal of Food Safety, 14, 19–40.
Howlett E, Burton S and Kozup J (2008) How modification of the nutrition facts panel
influences consumers at risk for heart disease. Journal of Public Policy & Marketing, 27
(1), 83–97.
Huffaker, D (2004) Spinning yarns around a digital fire: storytelling and dialogue among
youth on the Internet. First Monday, 9 (1) (January 2004). Online: www.firstmonday.org/
issues/issue9_1/huffaker
ILSI (2007) ‘Consumer understanding of health claims’, ILSI Report Series. International
Life Sciences Institute, Brussels, Belgium.
Institute of Food Technologists’ Expert Panel on Food Safety and Nutrition (1995) Scientific
status summary, foodborne illness: role of home food-handling practices. Food Technol-
ogy,49, 119–131.
Internet World Stats. Online: www.internetworldstats.com/stats.htm
Kasperson, RE, Renn, O, Slovic, P, Brown, HS, Emel, J, Goble, R, Kasperson, JK and Ratick,
S (1998) The social amplication of risk: a conceptual framework. Risk Analysis, 8, 177–
187.
King, BS, Tietyen, JL and Vickner, SS (2000) Food and Agriculture: Consumer Trends and
Opportunities Fats, Oils, and Sweets, Cooperative Extension Service, University of
Kentucky, College of Agriculture, USA. Online: www.ca.uky.edu/agc/pubs/ip/ip58g/
ip58g.htm
Kolko, B and Reid, E (1998) Dissolution and fragmentation: problems in on-line communi-
ties. In Cybersociety 2.0 (Steven Jones, ed.), Sage, Thousand Oaks, California, USA,
p.212–229.
Latouche K, Rainelli, P and Vermersch D (1998) Food safety issues and the BSE scare: some
lessons from the French case. Food Policy, 23, 347–356.
Licklider, J, and Taylor, R (1990) The computer as a communication device. In: Science and
CONSUMER CONCERNS AND RISK PERCEPTION RELATED TO TFA 413
soybean oil 4, 18, 20–21, 27, 29–30, 50, 54–56, linseed oil 5–6, 19, 27–29, 92, 156, 169, 174–
98, 109–110, 152, 157–158, 171, 173, 178, 175, 200
181, 236, 238, 310 non-genetically modified 156–157
stearidonic acid 180 palm oil 1, 20, 50–51, 54–55, 65, 98–100,
stearoyl-CoA desaturase 202 148–149, 151–155, 157
sterculic acid 201 rapeseed (canola) oil 1, 3, 20–21, 25–27, 50,
sunflower oil 5–6, 19, 24, 28–30, 46, 156–157 56, 58–59, 65, 67, 69, 108–110, 113, 118–
119, 122, 148, 152–153, 156–157, 168, 186,
Transgenic (see genetically-modified) 400, 407
transmethylation 107 safflower oil 9, 166
type 2 diabetes 314–316 soybean oil 4, 18, 20–21, 27, 29–30, 50, 54–
56, 98, 109–110, 152, 157–158, 171, 173,
UK and Ireland, trans fatty acids consump- 178, 181, 236, 238, 310
tion 348 sunflower oil 5–6, 19, 24, 28–30, 46, 156–157
venturi loop reactor 53
Vaccenic acid 2, 6, 30, 86, 88–89, 109, 167, Wadsworth-Emmons reaction 80
185, 196–222, 232, 234, 255, 301, 315, 329, Wittig reaction 77–100
334, 381 Wittig-Horner reaction 80–81, 83, 85
vegetable oil
genetically-modified 156–159, 234 ylides 78–81, 83