Biol Rex 32: 273-287 , 1999

Trans fatty acid isomers in human health and in the food industry

ALFONSO VALENZUELA and NORA MORGADO

Laborator io de Lípidos y Ant ioxidantes , INTA, Univers idad de Chi le , Casil la 138-11, Sant iago, Chi le . Phone: 56-2-678-1446. FAX: 56-2-221-4030. E-mail : avalenzu@uec. in ta .uchi le .c l

A B S T R A C T

Trans fatty acids are unsaturated fatty acids with at least one double bond in the trans configuration. These fatty acids occur naturally in dairy and other natural fats and in some plants. However, industrial hydrogenation of vegetable or marine oils is largely the main source of trans fatty acids in our diet. The metabolic effect of trans isomers are today a matter of controversy generating diverse extreme positions in light of biochemical, nutritional, and epidemiological studies. Trans fatty acids also have been implicated in the etiology of various metabolic and functional disorders, but the main concern about its health effects arose because the structural similarity of these isomers to saturated fatty acids, the lack of specific metabolic functions, and its competition with essential fatty acids. The ingestion of trans fatty acids increases low density lipoprotein (LDL) to a degree similar to that of saturated fats, but it also reduces high density lipoproteins (HDL), therefore trans isomers are considered more atherogenic than saturated fatty acids. Trans isomers increase lipoprotein(a), a non-dietary-related risk of atherogenesis, to levels higher than the corresponding chain-length saturated fatty acid. There is little evidence that trans fatty acids are related to cancer risk at any of the major cancer sites. Considerable improvement has been obtained with respect to the metabolic effect of trans fatty acids due the development of analytical procedures to evaluate the different isomers in both biological and food samples. The oleochemical food industries have developed several strategies to reduce the trans content of hydrogenated oils, and now margarine and other hydrogenated-derived products containing low trans or virtually zero trans are available and can be obtained in the retail market. The present review provides an outline of the present status of trans fatty acids including origin, analytical procedures, estimated ingestion, metabolic effects, efforts to reduce trans isomers in our diet, and considerations for future prospects on trans isomers.

KEY WORDS: trans fatty acid isomers, positional and geometric isomerization of fatty acids, health effects of trans isomers, hydrogenation and trans fatty acids, conjugated linoleic acid.

I-INTRODUCTION

1-A Basic structural background

Wherever there is a double bond in a fatty acid, there is a possibility for the formation of both positional and/or geometric isomers. Under conditions of partial hydrogenation (see below), a double bond may change from a cis to a trans c o n f i g u r a t i o n (geometric isomerization) or move to other positions in the carbon chain (positional isomerization) (Dutton, 1979). Both types of isomerization may frequently occur in the same fatty acid molecule. The two hydrogen atoms at a cis double bond are on the same side of the carbon chain, a situation

that produces a bend in the carbon chain. The two hydrogen atoms at a trans double bond are diagonally opposite each other and thereby straighten the carbon chain. The trans structure of a fatty acid increases its melting point when compared to the cis i s o m e r and a p p r o a c h e s tha t of the corresponding saturated form. Therefore, the i s o m e r may be r e g a r d e d as an i n t e r m e d i a t e be tween an o r ig ina l cis u n s a t u r a t e d fa t ty ac id and a m o r e completely saturated fatty acid. Trans fatty acids (TFAs) are unsaturated fatty acids with at least one double bond in the trans configuration. The most common TFAs are monounsaturated, but various diunsaturated cis, trans and trans, cis isomers, and even

Corresponding Author: Alfonso Valenzuela . Laborator io de Lípidos y Ant ioxidantes , INTA, Univers idad de Chi le , Casi l la 138-1 1, Sant iago, Chi le . Phone: 56-2-678-1446 - FAX: 56-2-221-4030. E-mail : avalenzu@uec. in ta .uchi le .c l

Received September 25, 1999. Accepted October 14, 1999.

274 V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287

triunsaturated cis, trans isomers may occur formed from the processing of oils with complex fatty acid composition, such as marine oils (Almendingen et al., 1995).

1-B Historical background

In 1993, a report published in Lancet by Willett et al. took the American public and the rest of the western world by surprise. An extensive study of more than 85,000 nurses concluded that women who ate four or more teaspoons of margarine a day had more heart attacks than women who rarely ate margar ine . The main goal of this controversial study was the correlation of dietary vegetable oil-based TFA intake with coronary heart disease. The results of the publication received wide media coverage, and consumers who had switched from butter to margarine as part of a "heart healthy diet," became confused and even angry. The results also started biochemical, toxicological and epidemiological research aiming to elucidate the real nutritional and h e a l t h i m p a c t of trans fa t ty a c i d s . Controversy has arisen about the potential health hazards of TFAs in the western diet. Extreme positions range from calling for the elimination of TFAs from the diet by avoiding specific foods, to removing them from processed foods and/or including their amounts on food labels, to the denial of any adverse effects from trans isomers, or to the need to modify food processing and production, food labels, health claims, or individual intakes.

2- NOMENCLATURE FOR TRANS FATTY ACIDS

According to official International Union of Pure and Applied Chemistry (IUPAC) nomenclature, the position of double bonds in fatty acids is counted from the carboxyl end of the molecule. Thus the vaccenic acid, the major trans isomer in milk (Hay & Morrison, 1970), is trans-l 1 or trans-Al 1-o c t a d e c e n o i c ac id . B i o c h e m i s t s and physiologists find it more advantageous to use the n- ("n minus") terminus, which counts from the methyl end of the molecule,

so that numbers remain the same when the carbon atoms are added to or removed from the carboxyl end during metabolism and when double bonds are inserted or remo­ved. The addition of the n- designation to the formulas for TFAs underscores the relationship between cis fatty acids and the trans isomers formed from them. Thus, vaccenic acid, which arises in the rumen of the cow through b io-hydrogena t ion of linoleic acid, is designated as trans-C\&:\ A l l according to IUPAC nomenclature, and as trans-C\8:l n-7 according to the n-system. Linoleic acid must be cis, cis C18:2 n - 6 , n -9 a c c o r d i n g to th i s l a t t e r nomenclature.

3- CHEMICAL AND PHYSICAL PROPERTIES OF TRANS

FATTY ACIDS

In a trans double bond in an unsaturated fatty acid the two hydrogen atoms bound to the carbon atom that form the double bond are located on opposite sides of the carbon chain. In contrast to the more typical cis isomeric configuration, the double bond angle of the TFAs is smaller and the acyl chain is more linear, resulting in a more rigid and straight molecule with a higher melting point. In the cis configuration, the hydrogen atoms are on the same side of the carbon chain, resulting in a "kink" in the acyl chain and a more flexible molecule. The spatial structure of TFAs is between that of sa tu ra t ed fatty ac ids and cis unsaturated fatty acids. As a result, oleic acid or cis CI 8:1 n-9 melts at 13°C ; its trans isomer elaidic acid (trans C18:1 n-9) melts at 44°C; and stearic acid (CI 8:0), which is straight and saturated, melts at 72°C. Therefore, the number, geometry, and position of double bonds in fatty acids affect the melting behavior of the fats of which they form part. Triglycerides high in saturated fatty acids will easily stack in a crystal lattice and are therefore solid at room temperature. Bends in the triglyceride molecule that hinder crystal formation introduced by cis unsaturated double bonds e x p l a i n why oi ls are l iquid at room temperature or at even lower.

V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287 275

4- NATURAL SOURCES OF TRANS FATTY ACIDS

TFAs occur naturally in dairy and other natural fats and in some plants. It has long been known that the rumen and feces of animals and humans contain relatively high amoun t s of T F A s . Appa ren t ly hydro ­génation of multiple unsaturated fatty acids from plant material by the ruminai and intestinal bacterial flora is responsible for this occurrence (Kepler et al., 1966). It is well documented that ruminai bacteria can reduce different unsaturated fatty acids, such as linoleic and linolenic acid, to yield largely stearic acid and some trans acyl derivatives (Mackie et al., 1991). TFAs arise in the first stomach of ruminants as in te rmedia tes of the hydrogéna t ion of die tary unsa tura ted fatty acids dur ing bacterial fermentation. The first step in this biohydrogenation is the isomerization of linoleic acid to mainly cis, trans C I8 :2 n -9 , n-11 ca ta lyzed by the anaerob ic bacterium Butyrivibriofibrisolvens (Kepler et al., 1966). These intermediates are then hydrogenated to form a mixture of mainly trans vaccenic acid and elaidic acid (trans CI8:1 n-9). The physiological reasons for these fatty acid t ransformations in the anaerobic environment are still unknown. It has been reported that the polyunsaturated fatty acids are more toxic to the anaerobic b a c t e r i a t han are the p r o d u c t of hydrogénation. Therefore, one function of bacterial fatty acid transformations may be to destroy growth-inhibi t ing substances (Verhulst et al., 1986). Another suggestion is that this process enables the strictly anaerobic bacteria dispose of the excess of reducing power (Mackie et al., 1991). This bacterial activity is also responsible for the occurrence of TFAs in cow's milk and milk products such as butter (Hay & Morrison, 1970). Depending on the diet, milk fat has 2-9% total TFA isomers (Parodi, 1976). These isomers are also found in the body fat of ruminants such as cattle and sheep, with concentrations ranging from 4% to 11 % (similar to butter fat). The major TFAs in milk, butter, and beef fat are trans 18:1 n-1 1 (vaccenic acid). TFAs are also found in goats, deer, and marsupials. In addition to the consumption of artificial-derived fats

as hydrogenated vegetable or marine oils, uptake of ruminal bacterial transformation products leads to the presence of TFAs in humans.

In contrast to the TFAs formed by rumen bacteria, the membrane const i tuents of aerobic bacteria are synthesized by a direct isomerization of the complementary cis configuration of the double bond without a shift of the position (Keweloh & Heipieper, 1996). Anaerobes always shift the position of the double bond dur ing fatty acid conversion (Seltzer, 1972). This system of isomerization is located in the cytoplasmic m e m b r a n e . The c o n v e r s i o n of cis unsaturated fatty acids to trans changes the m e m b r a n e f l u id i t y in r e s p o n s e to environmental stimuli, particularly where growth is inhibited due the presence of high concentrations of toxic substances. Under these conditions, lipid synthesis also stops so that cells are not able to modify their membrane fluidity by any other mechanism. The quantity of TFAs in bacte­ria can often be greater than that of cis unsaturated fatty acids. Similar to other fatty acids in bacteria, the relative amounts of TFAs depend on the phys io log ica l condition of the cells such as growth rate, medium composition, and environmental factors.

The chloroplast membranes of higher plants and algae also contain some fatty acids with a double bond with of trans configuration (Lamberto & Ackman, 1994). TFAs can be detected in minor amounts in m a r i n e a n i m a l s ( f i sh , s h e l l f i s h and mammals ) . These occur rences may be caused by the input and incorporation of fatty acids of microbial or plant origin into membrane lipids.

5- THE DEVELOPMENT OF HYDROGENATED FATS

The types of edible fats harvested, prepared, and sold as fats (the so-called "visible fats") have been influenced by changes in the relative cost of rendering fat from different sources, in the need for fats with specific qualities, and in the growing interest in the healthfulness of the population. In the early 1 9 0 0 ' s , the pr imary fat ava i lab le for

276 V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287

commerc ia l use was lard , which was rendered easily and inexpensively from pork fat. It had a good shelf life and excellent shortening properties, although w i t h o u t the b e n e f i t s of m o d e r n refrigeration, it became semi-solid at warm temperatures. Lard is also high in saturated fat, and growing health concerns about the dangers of excess dietary saturated fat stimulated efforts to find an alternative source of fat.

Pressing seeds and vegetables could pro­duce liquid fat, but this oil lacked the stability and shortening properties of lard. The hydrogenation process was therefore developed to modify liquid vegetable oils so these fats would be suitable substitutes for lard ( P a t t e r s o n , 1996) . The first hydrogenated products were a blend of totally hydrogenated cottonseed oil and refined liquid cottonseed oil. This created a product that had the consistency of lard but which was less likely to liquefy at warmer temperatures. The technique of partial hydrogenation was developed in the 1930's and complemented the development of a high-yield solvent extraction method to render fats from vegetables and seeds. In the process of partial hydrogenation, using pressure, temperature, and a metal catalyst (most frequently nickel), hydrogen gas is bubbled through liquid vegetable oil. Under these condi t ions , the double bonds on monounsaturated and polyunsaturated fatty acids in the liquid oil are subjected to structural modification. Three different modifications can occur: (1) A double bond can be changed to a single bond, e.g., changing a 2-polyunsaturated fatty acid into a m o n o u n s a t u r a t e d fa t ty ac id or a monounsaturated fatty acid into a saturated fatty acid. (2) The location of the double bond can be moved up or down the fatty acid chain, and/or (3) the configuration of the double bond can be changed to either cis o trans. Highly polyunsaturated fatty acids are the most susceptible to the process of hydrogenation as they contain more double bonds than other fatty acids. Because hydrogenation raises the melting point of the fat, the usual product is a fat that is a solid at 25°C, although there is a wide variation in physical properties, depending

on the s tar t ing fat and the degree of hydrogenation.

The hydrogenation process is controlled by selecting the appropriate temperature, pressure, duration, catalyst, and source of fat or fats to achieve a product of the desired c o m p o s i t i o n and func t ion . Pa r t i a l hydrogenation of fats and oils results in a mixture of fatty acids: polyunsaturated (including trans isomers), monounsaturated (also including trans isomers), and saturated fatty acids. As the degree of hydrogenation increases, the proportion of polyunsaturates d e c r e a s e , m o n o u n s a t u r a t e s and TFAs increase, and saturates increase only slightly. If hydrogenation is taken to completion, this process results in a fully-hydrogenated fat that contains neither cis nor TFAs, but which is made up completely of saturated fatty acids, primarily stearic acid. The basic Food and Drug Adminis t ra t ion (FDA, USA) definition of a hydrogenated fat is one that is solid at room temperature. Such fats typically contain 15-25% trans fatty acids. Partially hydrogenated oils are defined by the FDA as liquid at room temperature and are lower in trans fatty acids. Many vegetable fats and oils consumed in United States, Europe, and some countries of Latin America have been s l i gh t l y h y d r o g e n a t e d or moderately hydrogenated. Fats such as margarine frequently represent mixtures of fats that are hydrogenated (slightly or moderately) and fats that have not been processed. The types of fats added to a product and their processing methods must be described on the composition label.

6- QUANTIFICATION OF TRANS FATTY ACIDS IN

BIOLOGICAL SAMPLES OR IN FOODS

The two major assay methods for TFAs are inf rared abso rp t i on s p e c t r o s c o p y and c a p i l l a r y g a s - l i q u i d c h r o m a t o g r a p h y . Infrared spectroscopy is the classical method routinely used to determine TFAs in foods over the last two decades. Generally infrared spectroscopy measurements are carried out using TFA methyl ester-derivatives. A major problem is that samples analyzed by infrared spectroscopy as methyl esters produce trans levels that are 1.5%-3.0% lower for trans

V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287 277

va lue s from 1 to 1 5 % ( F i r e s t o n e & LaBoul iere , 1965). The Associat ion of Official Analy t ica l Chemis t s (Official Methods of Ana lys i s , 1994) therefore proposed correction factors to compensate for the lower absorption of methyl esters.

Gas-liquid chromatography is now the mos t popu la r t e c h n i q u e . It is wide ly available and allows identification of indi­v idua l fat ty ac ids when the su i t ab l e standards are available, whereas infrared absorption spectroscopy does not. However, in complex mixtures of isomeric fatty acids such as those presen t in some foods containing partially hydrogenated vegetable or fish oil, all fatty acid isomers are rarely resolved by gas chromatography. As a result, cis and trans isomers may overlap, and results may be biased. Pre-separation of cis and trans isomers by argentation chromatography may solve the problem but is e x t r e m e l y l a b o r i o u s ( U l b e r t h & Henninger , 1992) The problem is even further pronounced by the multitude of isomers of C20:1 , C20:2, C22:2, and C22:3 unsaturated fatty acids that can be formed in hydrogenated and partially hydrogenated fish oil ( M o r g a d o et al., 1998) . Gas chromatography is therefore unsuitable for routine assessment of the TFA content of foods containing partially hydrogenated fish oil, but can be applied successfully to biological samples (e.g.: tissue samples) due to the restricted variety of trans isomers frequently present in such samples.

Infrared absorption spectroscopy is less s e n s i t i v e then gas c h r o m a t o g r a p h y (although sensitivity may be increased with Fourier transforming equipment) and does not distinguish individual TFAs or detect posi t ional cis i somers. This method is nonetheless invaluable as a check on results obtained by gas chromatography. In gene­ral, the routine assay of TFAs is more complicated than that of other fatty acids, and figures in food tables and consumption figures for TFAs may be unreliable.

7- CONSUMPTION OF TRANS FATTY ACIDS

TFA ingestion has been estimated from d ie ta ry q u e s t i o n n a i r e s or reca l l data ,

analysis of adipose tissue or milk fat data, or food content analysis data. The basis of TFA intake calculation commonly varies, resulting in varying intake levels in the literature. Another reason for differences are individual and cultural eating habits. Different analytical techniques such as gas-chromatography determination and infrared absorption spectroscopy measurements are also responsible for widely varying intake estimations.

Edible oils, such as refined or unrefined o l ive , sunf lower , saff lower, r apeseed , soybean, peanut oil or coconut fat, contain only negligible amounts of TFAs. Refined o i l s have h i g h e r TFA c o n t e n t s than unrefined oils. The amount of TFAs in refined oils is influenced by the duration and temperature of refining. Refined and b l e a c h e d s o y b e a n oil t r e a t e d wi th temperatures between 160°C and 240°C was found to contain trans 18:3 isomers (O'Keefe et al., 1984). Rapeseed oil heated at 2 7 5 ° C for 3 h o u r s c o n t a i n e d approx imate ly 4% trans 18:3 i somers ( D e n e c k e , 1 9 9 5 ) . D u r i n g l o w e r deodorization conditions, e.g.: 220°C for 3 hours, the isomerization rate of rapeseed oil was approximately 25% lower. One reason for the increased isomerization rate at 275°C might be the more intense oil mixing resulting from the use of steam, which rules out local overheating with its strong influence on trans isomerization (Deneke, 1995).

Historically, the consumption of TFAs has increased steadily since the 1920's , in parallel with the increasing commercial production of margarine and shortenings. TFA consumption within a given population will differ greatly depending on lifestyle and socioeconomic status. The estimated a v e r a g e c o n s u m p t i o n in d e v e l o p e d countries is approximately 7-8 g per capita per day, or 6% of the total fatty acid intake (Report of expert panel, 1995). However, such average intake figures mask the wide range of isomeric fat intakes that actually occur. Although the average absolute intake of trans isomers may have declined slightly in recent years due to changes in the extent of hardening of edible oils and a move toward the consumption of softer, less

278 V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287

hydrogenated margarine, average intake is predicted to remain at a relatively constant 7-8 g per capita per day over the next few years. The recent replacement of tallow (3-5% trans) by hydrogenated vegetable oil (30% trans) for frying in the fast food industry can result in products (e.g.: french fried potatoes) with 24%-35% of the total fatty acids in the trans form. This suggests that there may be marked increases in the intake of TFA isomers in the present fast-food generation, and average intakes within the popu la t i on could r ise ra ther than remaining relatively constant. It is difficult to ascertain the absolute intake of isomeric fatty acids in population groups using food availability data based on the figures, and more detailed studies are needed to assess intakes in population groups. The metabolic aspec ts of i somer ic fatty acids re la te primarily to trans positional isomers of mono- and dienoic acids with the cis isomeric having been largely neglected, despite the fact that they can represent 20% or more of the total cis fatty acids and have distinct metabolic effects.

8- METABOLISM OF TRANS FATTY ACIDS

Dietary TFAs are readily absorbed and incorporated into most mammalian tissues, including human tissues, at concentrations that apparently reflect their content in the diet. The extent of incorporation of indivi­dual cis and trans isomers into tissue lipids is different for specific tissues and between neutral and phosphol ip ids in the same tissues. The brain and placenta apparently discriminate most readily against isomer inclusion compared with heart, liver, and adipose tissue, and the isomer content of phospholipids is generally lower than that of triglycerols (Ohlrogge etal., 1981). The level of dietary polyunsaturated or essential fatty acids will also influence trans isomer incorporation into tissue lipids. Inadequate intakes allow greater isomer incorporation (Beare-Rogers, 1988).

Previous research reported a marked discrimination against placental transfer of TFAs to fetal tissues (Ohlrogge etal, 1982). More recent observations, however, clearly

show a significant incorporation of trans isomers into umbilical cord blood lipids (Koletzko, 1991), although at lower levels than in maternal blood, thus indicating a degree of d iscr iminat ion against these isomers in the placenta. Low levels of trans i somers in premature fetal brain c o m p a r e d wi th o the r t i s s u e s is a l so indicative of discrimination by this organ. B r e a s t - f e d i n f a n t s in i n d u s t r i a l i z e d countries have a greater intake of TFAs through their mother ' s milk (2-5% of total mi lk fa t ty a c id s ) than t h o s e in less developed countries ( < 1 % of total milk fatty acids) which reflects the maternal consumption of these fatty acids (Koletzko, 1991). The high content of TFAs in human milk, even in the upper socioeconomic groups in industrialized countries, reflects their widespread occurrence in human foods (Koletzko et al., 1988). A reduction in the content of these isomers in human milk wi th in days of l ow er ing the i r in take indicates that they are readily mobilized and catabolized or excreted.

9- BIOLOGICAL EFFECTS OF TRANS FATTY ACIDS

9-A TFAs and the physical behavior of biological membranes

The phys ica l p rope r t i e s of b io log ica l membranes, which can occur either in a highly-ordered or a fluid crystalline phase state, are determined by the composition of the membrane lipids and fatty acids (Cevc, 1 9 9 1 ) . The c e l l u l a r b a r r i e r is simultaneously the matrix for enzymes, whose activi ty depends on its fluidity conditions (Carruthers & Melchior, 1983). On the other hand, some environmental factors, such as the temperature and organic so lven t s , have a s t rong in f luence on m e m b r a n e f l u i d i t y , and thus on the physiological properties of a membrane. Microorganisms, however, can adapt to changes of their membrane fluidity; they react to externally dictated changes in their environment by modifying their membrane to maintain a constant degree of fluidity. This mechanism is called "homeoviscosic

V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287 279

adapta t ion" (Suutari & Laakso , 1994). Eukaryotic cells mainly regulate the fluidity of their membranes by changes in the cholesterol content. Bacteria, which do not generally posses steroids, must rely on other mechanisms. Most bacteria regulate the fluidity of their membranes by varying the degree of saturation of the membranes ' fatty acids (Keweloh et al., 1991).

In membranes with a high content of saturated fatty acids, the acyl chains of the fatty acids produce the optimal hydrophobic interaction with each other, which leads to a highly-packed, rigid membrane. Due to the free ro t a t ion a round C-C b o n d s , saturated acyl chains can occupy an all-trans conformation, which allows the chain to be m a x i m a l l y e x t e n d e d . T h i s conformation is preferred if the fluidity of a membrane is low. The temperature for the transition from the gel to the liquid-crystal membrane phase rises with an increasing ratio of saturated toward unsaturated fatty acids. Contrary to this, the double bond of cis unsaturated fatty acids possesses a non-mobile bend with an angle of 30° in the acyl chain. Due to this bend, the highly ordered package of the acyl chains is disturbed, which leads to a lower phase transition temperature of these membranes (Cronan & Gelman, 1975). The transition temperature of a phosphatidylethanol amine bilayer decreases by about 42°C if only one saturated acyl chain per lipid molecule is replaced by a cis unsaturated one (Okuyama et al., 1991). The trans configuration has a completely different effect on the fluidity of the phospholipid bilayer than the cis one (Cevc, 1991). TFAs can take up a long extended structure, which is similar to that of saturated fatty acids. Membranes built of TFAs can therefore form a more rigid p a c k a g e of the p h o s p h o l i p i d s t han m e m b r a n e s with cis fatty ac ids . The m e a s u r e m e n t of p h a s e t r a n s i t i o n tempera tures of membranes built from identical phospholipids with trans rather than of cis unsaturated fatty acids results in significantly higher transition temperatures (Wever et al., 1994). The replacement of one cis acyl chain by a trans fatty acid in phosphatidylethanol amine increases the t r a n s i t i o n t e m p e r a t u r e by 1 8 - 3 1 ° C ,

depending on the structure of the other acyl chain of the lipid molecule. Therefore, the conversion of cis unsaturated fatty acids into their trans configuration results in a significant reduct ion of the membrane fluidity which is, however, smaller than the replacement of cis by saturated fatty acids.

9-B Effects of TFAs on cellular and tissue function

TFAs have been implicated in the etiology of v a r i o u s m e t a b o l i c and func t iona l d i s o r d e r s . They i n c r e a s e e r y t h r o c y t e fragility (Decker & Mertz, 1967), produce mitochondrial swelling, thereby reducing its oxygen consumption and ATP synthesis (Zevenbergen et al., 1988), and enhance the arrhytmogenicity of cardiac miocytes in expe r imen ta l an imals , p robably by modulating or by diminishing membrane f lu id i ty ( W e n z e l & K l o e p e l l , 1980) . Because of the effects of TFAs on the m e t a b o l i s m of g a m m a - l i n o l e n i c and arachidonic acid (Kinsella et al., 1981), ingestion of trans isomers can affect the metabolism of prostaglandins and other e i c o s a n o i d s and may a l t e r p l a t e l e t aggregation and vascular function as well (Asherio et al., 1994). TFAs also show competitive interactions with essential fatty acid metabolism (EFA) by inhibiting its incorporation into membrane phospholipids and by reducing the conversion of EFAs to e i c o s a n o i d s in d i f ferent an ima l ce l l s (Peacock & Wahle, 1989). Trans isomer ingestion results in essential fatty acid deficiency which may be prevented by increasing EFA availability (Kinsella et al., 1981 ; Zevenbergen & Haddeman , 1989). In addition, the incorporation of TFAs into membrane phospholipids may influence the physical properties of the membrane as well as the activities of the membrane-associated enzymes (Holman et al., 1991). Trans isomers have a weaker inh ib i to ry effect on co l l agen - induced platelet aggregation than do cis isomers (Peacock & Wahle, 1988). Further support for a specific effect of isomeric fatty acids derives from evidence that dietary TFAs

280 V A L E N Z U E L A & M O R G A D O Biol Res 32 , 1999, 273-287

inhibit the activities of membrane-bound enzymes such as Na+/K+ ATPase and adenylate cyclase, and reduce the density of beta adrenergic receptors in the plasma m e m b r a n e of heart t i ssue taken from animals that were not deficient in EFAs (Alam et al., 1989). However, TFAs do not appea r to have s i gn i f i c an t effect on reproduction, longevity, or the incidence of cancer (Senti, 1985).

Perhaps the primary concerns about the health effects of TFAs have arisen because these isomers are structurally similar to saturated fats, lack the essential metabolic functions of their parent polyunsaturated fats, and compete with EFAs in many complex metabolic pathways (Mann, 1994). Recently important new data on the health effects of TFAs have become available. It has been demonstrated that TFA ingestion increases low density lipoprotein (LDL) cholesterol to a degree similar to that of saturated fats (Mensink & Katan, 1990). The increase in LDL concentration has been attributed in part to the down-regulation of the LDL receptor (Hayashi et al., 1993). In contrast to other forms of fats, trans isomers decrease high-density lipoprotein (HDL) cholesterol. The mechanism of this decrease has been pos tu la ted to be due to the stimulation of cholesteryl ester transfer l i p o p r o t e i n a c t i v i t y , wh ich t r ans f e r s cholesteryl esters from HDL to VLDL and LDL (Abbey & Nestel, 1994). However, TFA feeding is also known to decrease HDL in animal species such as pigs (Jackson et al., 1977) and rats (Sano & Privett, 1980; Morgado et al, 1999), which do not have cholesteryl ester transfer protein activity, suggesting that additional mechanisms may be operative. Recently Subbaiah et al., (1998) suggested that TFAs may decrease H D L t h r o u g h the i n h i b i t i o n of lecithin:cholesterol acyltransferase. This enzyme esterifies free cholesterol in plasma by t r a n s f e r r i n g an acyl g r o u p from phosphatidylcholine (Glomset & Norum, 1973) . Both h u m a n and rat l ec i th in : cholesterol acyl t ransferases , which are normally specific for the sn-2 acyl group of phosphatidylcholine, exhibit an alteration in their positional specificity when TFAs containing-phosphatidylcholine is used as a

substrate. This may induce an inhibition of the enzyme and may allow the formation of more saturated cholesteryl esters, which are more atherogenic due to their increased deposit ion and reduced clearance from arterial tissue (Phillips et al., 1987). Thus the increase in the ratio of LDL cholesterol to H D L c h o l e s t e r o l for T F A s is approximately double that of saturated fatty acids (Mens ink & Katan, 1990). And therefore the effect of TFAs on the serum lipoprotein profile is at least as unfavorable as that of the cholesterol-raising saturated fatty acids, because they not only raise LDL-cholesterol levels, but also lower HDL-cholesterol levels. Similar adverse effects were confirmed in other studies. Unlike other fa t s , T F A s were found to i n c r e a s e l ipopro te in (a ) , another r isk factor for coronary heart disease (Mensink et al., 1992).

Partially hydrogenated marine oils, which are used extensively in the manufacture of cheaper margarines, may contain up to 50% trans bonds and considerably high amounts of 20 and 22 monoenes isomers. These isomers, whose tissue distribution is quite different from those found in natural fats, are not incorporated into the membrane phospho l ip id s , but induce pe rox i some proliferation and are beta-oxidized to chain l eng ths of 18 and 16 ca rbons in the peroxisomes (Hoy, 1991). The marine oils hydrogenated fatty acids trans isomers interfere with delta-5 desaturases in the liver (Rosenthal and Doloresco, 1984), resulting in an accumulation of 18:2 n-6 and 20:3 n-6 fatty acids (Hoy, 1991).

9-C TFAs and lipoprotein(a).

Lipoprotein(a) (Lp(a)) is a macromolecular c o m p l e x , m a d e up of a p o p r o t e i n B , cholesterol and other lipids, and a protein cal led apo(a) . Apo(a) shows sequence homology to plasminogen (Mc Lean et a/., 1987). Lp(a) concentration in the blood is largely under genetic control and does not change significantly with age (Dahlen, 1990). Most subjects have Lp(a) levels be low 150 mg/L , but levels in some individuals may exceed 400 mg/L. Such

V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287 281

subjects have a markedly increased risk for coronary heart disease, a relationship that is different and must not be confused with s e r u m l o w - d e n s i t y or h i g h - d e n s i t y l ipoprote in (LDL or HDL) choles terol levels (Sandkamp et al., 1990). In spite of the structural resemblance between Lp(a) and LDL, determinants of serum Lp(a) levels are distinctly different from those of LDL. Although niacin used with neomycin has been reported to decrease Lp(a) levels (Guraker et al., 1985), attempts to modify Lp(a) by drugs (Vessby et al., 1982) or diet (Katan & Beynen, 1987) have not been successful. Horns t raera / . (1991), however, h a v e s h o w n tha t d i e t a r y fa t ty ac id compos i t i on may affect Lp(a) l eve l s . Although the Lp(a) levels are under genetic regulation and, unlike LDL, Lp(a) levels are insensi t ive to diet, this suggest ion appears to be correct as far as dietary c h o l e s t e r o l is c o n c e r n e d . Die ta ry fat c o m p o s i t i o n d o e s a f fec t L p ( a ) concentrations in a way that correlates with the saturated fatty acid and the trans fatty acid composition of the diet. Mensink et al. ( 1 9 9 2 ) d e m o n s t r a t e d t ha t trans monounsaturated fatty acids (trans-CIS: 1) increased Lp(a) levels relative to three other fatty acids with 18 carbon atoms: stearic acid (CI 8:0), oleic acid (cis-C\S:\), and linoleic acid (cis, c/.s-C18:2). Results of Mensink et al., are also in good agreement with those of Nestel et al. (1992), who reported that a diet rich in the TFA elaidic acid (trans-Cl8:1) elevated the levels of Lp(a) in middle hypercholesterolemic men. The general conclusion of these studies is that short-term dietary experiments suggest that diets high in trans monounsaturated fatty acids may increase human serum levels of Lp(a).

9-D Conjugated linoleic acid, a special class of trans fatty acid isomer.

Conjugated linoleic acid (CLA) is formed, as are other trans isomers, in the first stomach of ruminants as an intermediate of the hydrogenation of dietary unsaturated fatty acids during bacterial fermentation. The first step in this biohydrogenation is

the isomerization of linoleic acid to mainly cis, trans C I8 :2 n-9, n -11 , a conjugated form of linoleic acid, catalyzed by the a n a e r o b i c b a c t e r i u m Butyrivibrio fibrisolvens (Kepler et al., 1966). These intermediates are then hydrogenated to form a mix tu re cons i s t ing main ly of trans vaccenic and elaidic acid. However, a small fraction remains as CLA. This fatty acid can also be formed through the autoxidation of linoleic acid by free radicals, followed by reprotonation of the pentadienyl radical by proteins , for instance whey protein (Fritsche & Steinhart, 1998). Chin et al. ( 1 9 9 4 ) i n v e s t i g a t e d the a b i l i t y of nonruminants (rats) to produce CLA. They supplemented the diet with 5% free linoleic acid or 8.6% corn oil (equivalent to 5% free linoleic acid as triglyceride) and observed higher tissue CLA concentrations in rats fed free linoleic acid than in control animals. These investigators concluded that the in­testinal bacterial flora of rats is capable of converting free linoleic acid but not linoleic acid esterified in triglycerides, to cis, trans, C 18:2 n-9, n-11 and trans, cis C I8 :2 n-9, n-11.

Strong hydrogenat ion of soybean oil allows the formation of variable small amounts of CLA (Mossoba et al.,1991). Chin et al. (1992) found CLA in coconut and olive oil in an order of approximately 0.02 g/lOOg fat. Another source of CLA was reported by Spitzer et al. (1991) who identified this trans isomer in exotic seed oils, e.g.: in Acioa edulis oil, which is used for cooking in areas of Brazi l . Dairy products are also a major source of CLA. The content in milk fat ranged from 0.24% to 1.77% (Fritsche & Steinhart, 1998). This wide variability is caused by the range of CLA content in the raw material, which is probably influenced by different dairy cow breeds and feeding systems as well as processing parameters. The CLA content in meat from ruminants is higher than in the meat of non-ruminants (e.g.: 1.20% in lamb and 0.12% in pork). In the case of non­ruminants, CLA may occur from dietary sources such as feeding powdered meat and tallow (Fritsche & Steihart, 1998).

CLA shows un ique c h e m o p r o t e c t i v e actions. It has been shown to protect against

282 V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287

some types of cancer in both animals and in humans (Ha et al, 1990; Ip et al., 1991; Visonneau et al., 1997). In addition to the chemoprotective properties of CLA, this trans isomer has also been linked to an influence on the growth and development of rats (Belury & Kempa-Steczko, 1997). A beneficial influence on the development of atherosclerosis has also been associated with CLA (Lee et al., 1994). Another remarkable property of this trans isomer is its antioxidant action. In some experimen­tal models, CLA has been shown to be a more potent antioxidant than a-tocopherol and almost as effective as BHT (Ha et al., 1990). To date, all physiological studies on CLA have been carried out with mixtures of CLA isomers (mainly cis, trans CI 8:2 n-9, n-11). Therefore, the biologically active isomer(s) remain(s) unknown. Although it is controversial to directly extrapolate from animal studies or cell cultures studies to human beings, CLA may be able to render lasting protection against subsequent cancer risk. It might be desirable to enhance the CLA concentrations in foods to obtain a beneficial level. A supplementation of CLA should therefore be considered (Fritsche & Steinhart, 1998).

9-E Trans fatty acids and cancer.

Numerous epidemiological studies have reported a posit ive correlation between increased dietary fat intake and cancers of the breast, colon and prostate (BNF report, 1987). A more specific association between these forms of cancer in humans and the intake of hydrogenated vegetable fats has also been reported (Enig et al., 1978). However, a restricted number of research papers about TFAs and cancer have been publ i shed . These papers represent the literature available on trans fat and cancer in animal models. The general conclusion is that increasing the intake of TFAs (at expense of cis fat) does not produce an adverse outcome with respect to cancer risk. Furthermore, it has even been reported that h igh trans fat i n g e s t i o n is less tumorigenic than a blend of fat with a lower trans c o n t e n t w h i c h s i m u l a t e d t he

composition of dietary fat commonly found in the western diet (Hunter & Applewhite, 1991). Inhib i t ion of p ros tag land in E2 synthesis reduces tumor proliferation and metastas is (Karmal i , 1980). Therefore , considering the reported inhibitory effects of TFAs on essential fatty acid metabolism and eicosanoid synthesis (Koletzko, 1991), it is conceivable that isomeric trans fatty acids could actually reduce tumor growth and metastasis. Although the most recent paper on the subject was published in 1996 (Ip & Marshall) , there has been no evidence to date to indicate that TFA intake presents a risk factor for cancer under properly-controlled conditions. The evidence that breast cancer is related to diet stems less from epidemiological evidence than from an imal e x p e r i m e n t a t i o n and from the analysis of ecological data (Selenskas et al., 1984). The epidemiological evidence is that the impact of fat intake in general upon breast cancer risk is slight to negligible (Erickson et al., 1984). There is at present no strong evidence that the intake of TFAs is related to an increased risk of breast cancer (van den Brandt et al., 1993).

There is however, good and increasing evidence that dietary fat is related to the risk of cancer of the colon and rectum (F reudenhe in & Graham, 1989) . This evidence centers on the intake of saturated fats or animal fat as increasing risk and the intake of fiber and vegetable products as decreasing risk. The clinical trials that are presently underway are evaluating dietary restriction of fat intake and supplementation of fiber and fruit and vegetable intake (Greenberg et al., 1994). No evidence indicates that the intake of TFAs is related to increased risk of cancer of the colon and rectum (Giovannuci et al., 1994). There is good evidence that fat intake might be re la ted to the risk of p ros ta te cancer (Freudenhein & Graham, 1989). Most of th i s e v i d e n c e has f o c u s e d on the consumption of saturated fat and fats of animal origin. The intake of TFAs has not been correlated to the risk of prostate c a n c e r . The one s tudy tha t d i r e c t l y examined the relevance of TFAs reported that they are not related to risk (Giovannuci et al., 1993). In summary, there is little

V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287 283

evidence that TFAs are related to the risk of cancer at any of the major cancer sites (lp & Marshall , 1996).

10- EFFORTS TO REDUCE THE CONSUMPTION OF

TRANS FATTY ACIDS

In response to the controversy provoked by studies of TFA consumption and human health, some companies are now taking steps to develop TFA-reduced or TFA-free products. As margarines are the primary target of criticism, the food industry has f o c u s e d i ts e f fo r t s on r e d u c i n g or e l iminat ing the trans content of these products. One strategy has been to blend fully-hydrogenated vegetable oils without trans isomers with unhydrogenated liquid oils, which naturally have no trans isomers. The hardness and the spreadability of the final product must be adjusted by varying the proportion of the solid and the liquid portions of the blend. These procedures have made it possible to reduce the TFA content of stick and tub margarine from 50% to 10% or less. "Low fat," "diet," or "light" margarines, with less than 40% fat, have a trans content ranging from 5% to 10%.

Another strategy that has not yet been fully developed is the interesterification of highly-hydrogenated fats with liquid oils by means of chemical or enzymatic-drive p r o c e d u r e s . U n l i k e h y d r o g e n a t i o n , interesterification neither affects the degree of saturation nor causes isomerization of the double bonds. Thus it does not change the fatty acid profile of the starting mate­rial. Instead, it rearranges the fatty acids in the g l y c e r o l m o l e c u l e . R a n d o m interesterification changes the crystal form of an oil or blend to produce the desired solid fat content curves or to produce blends with high levels of polyunsaturated fatty a c i d s . By th i s m i x e d p r o c e d u r e , hydrogenation-interesterification, it will be possible to obtain margarine with virtually zero trans content. However, to achieve improved results further developments are necessary (Fitch-Haumann, 1994).

The latest development in Europe has been the introduction of a fish oil spread

manufactured by FDB Viby Fabrikker of Denmark. The spread consists of 20-25% highly deodorized liquid fish oil and 70-80% of highly hydrogenated vegetable fat. The product called Tobis has trans isomer contents lower than 15%. It contains no water, has a slightly yellow color, and is s table to oxidat ion (Inform, 1990). A product with different characteristics was d e v e l o p e d by U n i l e v e r NV of the N e t h e r l a n d s . T h i s is a b l e n d of unhydrogenated marine oil and vegetable oil in a ratio of marine to vegetable below 1:3. This product, which has non-trans isomers, can be used for frying and baking as well as for making margarine and soft s p r e a d s ( I n f o r m , 1 9 9 0 ) . As a n o t h e r approach, non-caloric or hypocaloric fat substitutes are now used to develop non-trans and n o n - c h o l e s t e r o l bu t t e r and margarine (Singhal et al., 1991).

Meanwhile, Lipton Canada has obtained two temporary marketing authorizations for including a TFA declaration on food product labels. The company began using the first t emporary market ing author iza t ion for TFAs on its Becel l ine of marga r ine products in late 1997. Declarations on two of the Becel products - Becel RSF, which contains 26% fat, and its Light version containing 40% fat - list trans fat content as zero because of the low fat level per serving. The other two Becel products - its full-fat version and a salt-free margarine -list the trans fatty acid content as 0.1 gram per 10 gram (two teaspoon) serving. Also from Lipton is the Fleischmann margarine line, whose products also list their TFA content as 0.1 gram per 10-gram serving. Products containing 0.1 gram of trans fat per serving include the statement "Virtually no trans fat." The declarations are part of the nutrition listing label.

The r e p l a c e m e n t of p a r t i a l l y hydrogenated soybean oil (or part ial ly hydrogena ted fish oil) by pa lm oil in m a r g a r i n e is a n o t h e r t e c h n o l o g i c a l alternative used to considerably reduce the trans content of the final product. Palm oil, unlike soybean or fish oil, can be used without hydrogenation to achieve a certain hardness of margarine products because of its semisolid texture at room temperature.

284 V A L E N Z U E L A & M O R G A D O Biol Res 32 , 1999, 273-287

Palm oil consists of 50% saturated and 50% unsaturated fatty acids and has a content of 44% palmitic acid and 10.6% linoleic acid. Muller et al. (1998) recently demonstrated that a margarine based on palm oil was less atherogenic (based on examination of LDL-cholesterol and HDL-cholesterol profiles) than a soybean-oil , partially-hydrogenated ?ra«.y-containing margarine. The product, however, is less favorable than one based on a more polyunsaturated vegetable oil.

11 - FUTURE PROSPECTS FOR TRANS FATTY ACIDS

The food industry could voluntarily phase out production of TFAs, but at present producers resist even acknowledging that their products have adverse effects. A voluntary phase-out is therefore unlikely, although some of the largest producers have publicly committed themselves to reducing the trans isomer content of their products. An alternative is to develop policies to regulate TFAs in foods by labeling their contents. Some have suggested that the trans fatty acids be listed along with the saturated fat on the label, but this must be accompanied by efforts to educate the public about TFAs and the potential risks associated with consuming hydrogenated fats. Warning labels are already applied to alcohol and tobacco products; such action is even further just if iable for products containing partially-hydrogenated fats, as their presence is not readily visible to the consumer.

As a matter of fact, some conclusions may be drawn from the current information and summarized as follows:

• Intake of TFAs cannot be avoided when the diet includes milk fat, ruminant fat, and/or hydrogenated fats in margarine and shortenings.

• The average dietary intake of TFAs will tend to decline due to a general reduction in fat content in foods.

• N u t r i t i o n a l l y , TFAs should be of secondary priori ty after saturated fats. However, high intakes may have adverse effects on the plasma LDL/HDL cholesterol ratio.

• There appear to be many ways to avoid TFAs, but the healthiest requires changes in individual eating habits.

• The U.S. Food and Drug Administration (FDA) is still working on provisions for possible TFA labeling on foods, but does not anticipate releasing a proposal during 199.8 (Inform, 1998).

ACKNOWLEDGMENTS

The preparation of this review was partially supported by FONDECYT grants 1960988 (A.V.) and 2960044 (N.M.) and by Malloa-Alimentos SA.

REFERENCES

A B B E Y M, N E S T E L PJ (1994) P la sma choles teryl ester t ransfer protein act ivi ty is increased when trans-elaidic acid is subst i tuted for ci's-oleic acid in the diet . Atherosc le ros i s 106: 99-107

A L A M S Q , R E N YF, A L A M BS (1989) Effect of dietary trans fatty ac ids on some m e m b r a n e a s soc i a t ed enzymes and receptors in the rat heart . Lipids 24: 39-44

A L M E N D I N G E N K, J O R D A L O , K I E R U L F P, S A N D S T A D B, P E D E R S E N J (1995) Effects of part ial ly hydrogenated fish oil , part ial ly hydrogenated soybean oil , and butter on serum l ipopro te ins and Lp(a) in men. J Lipid Res 36: 1370-1384

A O A C ( 1 9 9 4 ) Off ic ia l M e t h o d s of A n a l y s i s of the Assoc ia t ion of Official Analyt ica l Chemis t s . Method 965 .34 , H E L R I C H K (ed) , Ar l ing ton , VA

A S H E R I O A, H E N N E K E N S C, B U R I N G L M A S T E R C. S T A M P E R M, W1LLETT W (1994) Trans fatty acids in take and risk of myocard ia l infarct ion. Circula t ion 89: 94-101

B E A R E - R O G E R S J L ( 1 9 8 8 ) Trans and posi t ional i somers of common fatty ac ids . Adv Nutr Res 5: 171-200

B E L U R Y MA, K E M P A - S T E C Z K O A (1997) Conjuga ted l inoleic acid modula tes hepat ic lipid compos i t ion in mice . Lipids 32: 199-204

B N F R E P O R T (1987) Trans fatty ac ids . London: Brit ish Nutr i t ion Founda t ion

C A R R U T H E R S A, M E L C H I O R D (1983) S tudies on the re la t ionship be tween b i layer and water permeabi l i ty and bi layer physical s ta te . Biochemis t ry 22: 5797-5807

C E V C G (1991) How membrane cha in -mel t ing phase-t ransi t ion t empera tu re is affected by the lipid chain a symmet ry and degree of sa tura t ion- an effective cha in- length model . B iochemis t ry 30: 7186-7193

C H I N SF, LIU W, S T O R K S O N JM, HA YL, PARIZA MW (1992) Dietary sources of conjugated dienoic i somers of l i n o l e i c a c i d , a n e w l y r e c o g n i z e d c l a s s an t i ca rc inogens . J Food C o m p Anal 5: 185-197

V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287 285

CHIN SF, STORKSON JM, ALBRIGHT KJ, COOK ME, P A R I Z A - M W (1994) Conjugated linoleic acid is a growth factor (or rats as shown by enhanced weight gain and improved feed efficiency. J Nutr 124: 2344-2349

C R O N A N J, G E L M A N E (1975) Physica l p roper t ies of membrane l ipids: Biological re levance and regulat ion. Bacter io l Rev 39: 232-256

D A H L E N G H ( 1 9 9 0 ) I n c i d e n c e of L p ( a ) a m o n g popu la t ions . In: S C A N U AM (ed) L ipopro te in (a ) . San Diego: Academic Press Inc . pp : 151-173

D E C K E R WJ. M E R T Z W (1967) Effect of dietary e la idic acid on membrane function in rat mi tochondr ia and e ry th rocy tes . J Nutr 9 1 : 324-330

D E N E C K E P (1995) About format ion of trans fatty acids dur ing deodor iza t ion of rapeseed oil . Eur J Med Res 17: 109-114

D U T T O N HJ ( 1 9 7 9 ) H y d r o g e n a t i o n of fa ts and its s i g n i f i c a n c e . In : E r a k e n E A , D u t t o n HJ ( e d s ) G e o m e t r i c a l and p o s i t i o n a l fa t ty ac id i s o m e r s . Champaign , IL: American Oil Chemis ts Society: 1-16

ENIG MG, M U N N RJ, K E N N E Y M (1978) Die tary fat and cancer t rends - a cr i t ique . Fed Proc 37: 2215-2220

E R I C K S O N K L , S C H L A N G E R D S , A D A M S D A , F R E G E A U DR, STERN JS (1984) Influence of dietary fatty acid concent ra t ion and geomet r ic conf igura t ion on mur ine mammary tumor igenes i s and exper imenta l metas tas i s . J Nut r 114: 1834-1842

F I R E S T O N E D, L A B O U L I E R E P (1965) Dete rmina t ion of isolated trans i somers by infrared spec t roscopy . J Assoc Off Anal Chem 48: 437-443

F I T C H - H A L M A N N B (1994) T o o l s : H y d r o g e n a t i o n , in teres ter i f ica t ion . I N F O R M 5: 668-678

F R E U D E N H E I N JL, G R A H A M S (1989) Toward dietary prevent ion of cancer . Epidemiol Rev 1 1: 229-235

FRITSCHE J, STEINHART H (1998) Analys is , occurrence, and physiological proper t ies of trans fatty acids (TFA) with par t icular emphas i s on conjugated l inoleic acid isomers (CLA) - a rev iew. Fet t /Lipid 100: 190-210

G I O V A N N U C 1 E, R I M M E, C O L D I T Z GA (1993) A prospect ive study of dietary fa( and risk of pros(ate cancer . J Natl Cancer Ins 85: 1571-1579

G I O V A N N U C I E, R I M M E, S T A M P F E R MJ (1994) Intake of fat, meat and fiber in relat ion to risk of colon cancer in men. Cance r Res 54 : 2390-2397

G L O M S E T J A, N O R U M KR (1973) The metabol ic role of l ec i t h inxho le s t e ro l acyl t ranferase : perspect ives from pa thology . Adv Lipid res 1 1: 1-65

G R E E N B E R G ER, B A R O N JA, T O S T E N S O N TD (1994) Cl in ica l t r ial of an t iox idan t v i t amins to p reven t co lorec ta l adenoma. N Engl J Med 3 3 1 : 141-147

G U R A K E R A J , H O E G M, K O S T N E R G, P A P A D O P O U L O S M, B R E W E R HB (1985) Levels of l ipoprote in(a) decl ine with neomycin and niacin t rea tment . Atherosc le ros i s 57 : 293-301

HA YL, S T O R K S O N J, P A R I Z A M W (1990) Inhibi t ion of benzo(a ) -pyrene- induced mouse fores tomach neopla­sia by conjugated der iva t ives of l inoleic acid. Cancer Res 50: 1097-1 101

HAY J D . M O R R I S O N WR (1970) I somer ic monoeno ic fatty acids in bovine milk fat. Biochem Biophys Acta 202: 237-243

H A Y A S H I K, H I R A T A Y. K U R U S H I M A H. SAEKI M. A M I O K A H, N O M U R A S, K U G A Y, O H K U R A Y, O H T A M I H, K A J I Y A M A G (1993) Effect of dietary hydrogena ted corn oil ( / ranx-oc tadecenoate rich oil) on p lasma and hepat ic choles terol metabol i sm in the hamster . Atherosc le ros i s 99 : 97-106

H O L M A N RT, P U S H F . S V I N G E N B, D U T T O N H (1991) Unusual i somer ic po lyunsa tura ted fatty acids in liver phospho l ip ids of rats fed hydrogena ted oil. Proc Natl Acad Sei USA 88: 4830 -4834

H O R N S T R A G, V A N H O U W E L I N G E N A C , KESTER AD, S U D R A M K (1991) A palm oi l -enr iched diet lowers serum l ipopro te in(a) in no rmocho les t e ro l emic vo lun tee rs . Atherosc le ros i s 90: 91-93

HOY CH (1991) Conference : Trans fatty ac ids -Analys i s and metabol i sm. Lip idforum, Hels ink i , F in land , Oct 21-22

H U N T E R JE, A P P L E W H I T E TH (1991) Reasses smen t of trans fatty acid avai labi l i ty in the US diet . Am J Clin Nutr 54: 363-369

I N F O R M (1990) New ideas for margar ine . INFORM 1: 174-184

I N F O R M (1998) FDA: no trans label proposal this year. I N F O R M 10: 967

IP C, M A R S H A L L J (1996) Trans fatty acids and cancer . Nut Revs 54: 138-145

IP C. S I N G H M, T H O M P S O N HJ, S C I M E C A JA (199 1) C o n j u g a t e d l i n o l e i c ac id s u p p r e s s e s m a m m a r y c a r c i n o g e n e s i s and p r o l i f e r a t i v e ac t iv i ty of the mammary gland in the rat. Cancer Res 5 1 : 61 18-6124

J A C K S O N RL, M O R R I S E T T JD, P O W N A L L HJ, G O T T O A M , K A M I O A, IAMI H, T R A C Y R, K U M M E R O W F (1977) Inf luence of dietary trans fatty acids on swine l ipoprote in compos i t ion and s t ruc ture . J Lipid Res 18: 182-190

K A R M A L I RA (1980) Pros tag land ins and cancer . Prost Leukot r Med 5: 1 1-28

KAT AN M B , B E Y N E N AC (1987) Charac te r i s t i cs of h u m a n h y p o - and h y p e r r e s p o n d e r s to d i e t a r y cho les te ro l . Am J Epidemio l 125: 387-399

KEPLER C, H I R O N S K, M C N E I L L J, T O V E S (1966) In te rmedia tes and produc ts of the b iohydrogena t ion of l inoleic acid by Butyrivibrio fibrisolvens. J Biol Chem 2 4 1 : 1350-1354

K E W E L O H H, D I E F E N B A C H R, REHM H (1991) Increase of phenol to lerance of Echer ich ia coli by al terat ion of the fatty acid compos i t ion of the membrane l ipids . Arch ives Microbio l 157:, 49-53

K E W E L O H H, H E I P I E P E R H (1996) Trans unsa tura ted fatty acid in bacter ia . L ip ids 3 1 : 129-137

K I N S E L L A J, B R U C K E R G, MAI J. S C H R I M E J (1981) Me tabo l i sm of trans fatty acids with emphas i s on the e f f e c t s of r r a f l j ' - o c t u d e c a d i e n o a t e on l i p i d composi t ion , essential fatty acids , and pros tag landins : an overv iew. Am J Clin Nutr 34: 2307-2318

K O L E T Z K O B, M R O T Z E K M, ENG B, B R E M E R HJ (1988) Fatty acid compos i t ion of mature human milk in Germany . Am J Clin Nutr 47 : 954-959

K O L E T Z K O B ( 1 9 9 1 ) Zufuhr, Stoffweschel und biologische Wirkungen trans isomerer Fettsauren bei Säuglingen. Die Nahrung 35 : 229-283

L A M B E R T O M, A C K M A N R (1994) Conf i rmat ion by gas ch roma tog raphy /mass spec t romet ry of two unusual f ran .s-3-monoethylenic fatty acids from the Nova Scot ia seaweeds Palmaria palmata and Chororidrus crispus. Lipids 29 : 441-444

L E E KN. K R I T C H E V S K Y D, P A R I Z A M W (1994) Conjugated linoleic acid and atherosclerosis in rabbits . Atherosc le ros i s 108: 19-25

M A C K I E R. W H I T E B, B R Y A N T M ( 1 9 9 1 ) L ip id me tabo l i sm in anaerob ic sys t ems . CRC Crit Rev Microbio l 17: 449-479

286 V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287

M A N N G (1994) Metabo l i c consequences of dietary trans fatty ac ids . Lance t 3 4 3 : 1268-1271

M C L E A N J W , T O M 1 L S O N JE, K U A N G WJ (1987) C D N A sequence of human apo l ipopro te in (a ) is homologous to p lasminogen . Nature 300: 132-137

M E N S I N K RP. K A T A N MB (1990) Effect of dietary trans fatty acids on high-densi ty and low-densi ty l ipoprotein choles te ro l levels in heal thy subjec ts . N Engl J Med 323 : 439-445

M E N S I N K RP, Z O C K PL, K A T A N M B , H O R N S T R A G (1992) Effect of dietary cis and trans fatty acids on serum l ipopro te in(a) levels in humans . J Lipid Res 3 3 : 1493-1501

M O R G A D O N, G A L L E G U I L L O S A, S A N H U E Z A J, GARR1DO A, N I E T O S, V A L E N Z U E L A A (1998) Effect of the degree of hydrogena t ion of dietary fish oil on the trans fatty acid content and enzymat ic act ivi ty of rat hepat ic mic rosomes . Lipids 33 : 669-673

M O R G A D O N, S A N H U E Z A J, G A L L E G U I L L O S A, G A R R I D O A, N I E T O S, V A L E N Z U E L A A (1999) Effect of d ie tary hydrogena ted fish oil on the p l a sma l ipoprote in profile and on the fatty acid compos i t ion of different t i ssues of the rat. Ann Nutr Metab (In press)

M O S S O B A M M , M C D O N A L D RE, A R M S T R O N G DJ, PAGE SW (1991) Ident i f icat ion of minor C18 t r iene and conjugated diene isomers in hydrogenated soybean oil and margar ine by GC-MI-FTIR spec t roscopy . J Chroma tog r 29 : 324-330

M U L L E R H, J O R D A L O, K I E R U L F P, K I K H U S B, P E D E R S E N J ( 1 9 9 8 ) R e p l a c e m e n t of p a r t i a l l y hydrogena ted soybean oil by palm oil in margar ine without unfavorable effects on serum l ipopro te ins . Lipids 3 3 : 879-887

N E S T E L P, N O A K E S M, B E L L I N G B (1992) P la sma l ipoprote in l ipid and Lp(a) changes with subst i tu t ion of e laidic acid for oleic acid in the diet . J Lipid Res 33 : 1029-1036

O H L R O G G E JB , EM KEN EA, G U L L E Y RM (1981) Human t issue l ip ids : occur rence of fatty acid isomers from dietary hydrogena ted oi ls . J Lipid Res 22: 9 5 5 -960

O H L R O G G E JB , G U L L E Y R M , E M K E N EA (1982) Occur rence of oc tadecenoic fatty acid i somers from hydrogenated fats in human t issue lipid classes . Lipids 17: 551-557

O K U Y A M A H, O K A J I M A N, S A S A K I S, HIGASHI S, M U R A T A N (1991) The cisltrans i somer iza t ion of the double bond of a fatty acid as a s t ra tegy for adapta t ion to changes in ambien t t empera ture in the psychrophi l i c bac te r ium Vibrio sp. s train A B E - 1 . B iochem Biophys Acta 1084: 13-20

P A R O D I PW (1976) Dis t r ibut ion of isomeric oc tadecenoic fatty acids in milk fats. J Dai ry Sci 59: 1870-1873

P A T T E R S O N HB (1996) Hydrogena t ion of fats and oi ls : theory and prac t ice . A O C S Press , Champa ign , I l l inois

P E A C O C K LL. W A H L E KW (1988) I somer ic cis and trans fatty acids and porc ine platelet react ivi ty to co l lagen in vi t ro. Biochem Soc Trans 16: 291

P E A C O C K LL, W A H L E KW (1989) I somer ic cis and trans fatty acids and a rach idonic acid metabol i sm in porc ine pla te le t phospho l ip ids . Biochem Soc Trans 17: 679-680

PHILLIPS M C , J O H N S O N WJ, R O T H B L A T GH (1987) M e c h a n i s m s and consequences of ce l lu lar choles te ro l exchange and transfer . Biochem Biophys Acta 906: 223-276 . Repor t of the exper t panel on trans fatty acids and coronary heart d isease (1995) Am J Clin Nutr 62: 655S-708S

R O S E N T A H A L M, D O L O R E S C O M (1984) The effects of trans fatty acids on fatty acyl del ta-5 desa tura t ion by human skin f ibroblas ts . L ip ids 19: 869-874

S A N D K A M P M, F U N K E H, S C H U L T E H. K O H L E R E. A S S M A N N G (1990) Lipoprote in(a) is an independent r isk factor for myocard ia l infarct ion at a young age. Clin Chem 36: 20-23

S A N O M. P R I V E T T OS (1980) Effects of an essent ia l fatty acid def ic iency on serum l ipoprote ins in the rat. L ip ids 15: 337-344

S E L E N S K A S SL, IP M M , IP C (1984) Similar i ty between trans fat and sa tura ted fat in the modif icat ion of rat mammary ca rc inogenes i s . Cancer Res 44 : 1 321 -1326

S E L T Z E R S (1972) Cis-Trans i somer iza t ion . In: BOYER PD (ed) The E n z y m e s . Vol 6, New York: Academic Press , pp : 381-406

SENTI FR (1985) Heal th aspects of dietary trans fatty ac ids . Federa t ion of Amer ican Socie t ies for Exper i ­mental Bio logy , Be thesda , M D . (Contrac t FDA № 223 .83 -2020) .

S I N G H A L RS, G U P T A AK, K U L K A R N 1 PR (1991) Low calor ie fat subs t i tu tes . Trends Food Sci Techno l , Oc tober : 241-244

SPITZER V, M A R X F, MAIA JG, P F E I L S T I C K E R K (1991) Ident i f icat ion of conjugated fatty acids in the seed oil of Acioia edulis (Prance) and Couepia edu/is (Chrysoba lanaceae ) J Am Oil Chem Soc 68: 183-189

S U B B A I A H PV, S U B R A M A N I A N V S , LIU M (1998) Trans u n s a t u r a t e d f a t t y a c i d s i n h i b i t l e c i t h i n : c h o l e s t e r o l a c y l t r a n s f e r a s e and a l te r its pos i t ional specif ic i ty . J Lipid Res 39: 1438-1447

S U U T A R I M, L A A K S O S (1994) Microbia l fatty acids and thermal adapta t ion . Crit Rev Microb io logy 20: 285-238

U L B E R T H F, H E N N I N G E R M (1992) Simplif ied method for the de te rmina t ion of trans monoenes in edible fats by T L C - G L C . J A m Oil Chem Soc 69: 829-831

V A N DEN B R A N D T PA, V A N T VEER P, G O L D B A U M A (1993) A prospec t ive study on dietary fat and the risk of post menopausa l breast cancer . Cancer Res 5 3 : 75-82

V E R H U L S T A , P A R M E N T I E R G, J A N S S E N G, A S S E L B E R G H S S, E Y S S E N H ( 1 9 8 6 ) Bio t ransformat ion of unsa tura ted long-cha in fatty acids by Eubacterium lentum. Appl Environ Microbiol 5 1 : 532-538

VESSBY B, K O S T N E R G, L1THELL H, T H O M I S J (1982) D i v e r g i n g e f f e c t s of c h o l e s t y r a m i n e in hypercho les t e ro laemia . Atheros le ros i s 44 : 61-71

V I S O N N E A U S, C E S A N O A, T E P P E R SA, SC1MECA J, S A N T O L I D, K R I T C H E V S K Y D (1997) Conjugated l inoleic acid suppresses the growth of human breast adenoca rc inoma cel ls in SCID mice . Ant icancer Res 17: 969-974

W E N Z E L DG, K L O E P E L L JD (1980) Incorpora t ion of sa tura ted and cis and trans unsa tura ted long-chain fatty acids in rat miocytes and increased suscept ibi l i ty to a r rhy thmias . Tox ico logy 18: 27-36

V A L E N Z U E L A & M O R G A D O Biol Res 32, 1999, 273-287 287

W E V E R F, I S K E N S, DE B O N T J (1994) CislTrans i somer iza t ion of fatty acids as a defense mechan i sm of Pseudomonas putida starins to toxic concent ra t ions of to luene . Microbio logy 140: 2013-2017

W I L L E T T W C . S T A M P F E R MJ, M A N S O N JE (1993) Intake of trans fatty acids and risk of coronary heart d i sease among women . Lancet 3 4 1 : 581-585

Z E V E N B E R G E N J L , H O U T S M U L L E R V M . G O T T E N B O S JJ (1988) Linole ic acid requ i rements of. rats fed trans fatty ac ids . Lipids 2 3 : 178-186

Z E V E N B E R G E N JL, H A D D E M A N , E (1989) Lack of effects of trans fatty acids on e icosanoids b iosynthes is with adequa te intakes of l inoleic acid. Lipids 24: 555-563