21

Trans-fatty acids in heated hemp seed oil

H. Mölleken

Bergische Universität, Physiologische Chemie der Pflanzen, Gauss-Str. 20, 42119 Wuppertal, Germany

        Mölleken, H. 1998 Trans-fatty acids in heated hemp seed oil. Journal of the International Hemp Association 5(1): 21-23. Prompted by the frequently discussed question of whether trans-fatty acids are formed when hemp oil is used for cooking, we have analysed hemp oil and several other oils for their content after exposing them to various stress conditions. The results demonstrate that high temperatures do not change the configuration of the fatty acids.

Figure 1. The cis and trans configurations of a fatty acid with one double bond.

Introduction
        Fats and oils have been of interest to biochemists, nutritionists and health professionals for many years, and one of the points of discussion is the presence of trans-fatty acids (t-FAs). The double bonds of essential fatty acids (FAs) are confined to certain positions on the aliphatic chain and are all of the cis configuration. Relatively large amounts of t-FAs have been found only in a few plant species, mainly in the seeds and leaves. The highly polyunsaturated FA 18:2 (10 trans, 12 trans), for example, occurs to the extent of 5-10% in the Mexican bush Chilopsis linearis , and there is as much as 40% of FA 18:3 (9 trans, 11 trans, 13 trans) in the triglycerides of various Catalpa species (Belitz & Grosch 1994, Sommerfeld 1983, Steinhardt & Pfalzgraf 1994, Steinhardt 1996). These are exceptions however, as t-FAs normally occur to a minimal extent in plants, if at all.
        In contrast, t-FAs are widely found in animal fats. They largely originate from their diet, although in marsupials and ruminants t-FAs are synthesized in the rumen or stomach by microbial hydrogenation of the polyunsaturated FAs from plant food. Thus some t-FAs in the human diet are traced to milk and meat. In human metabolism, they are also formed during the degradation of unsaturated FAs, though the cis-configuration is of physiological importance. Polyunsaturated FAs, for example, loosen the packing of the phospholipids within the hydrophobic zone of biological membranes.
        In the human diet, t-FAs arise mainly from isomerisation of the natural cis double bonds of vegetable oils during their industrial hydrogenation to margarine and vegetable shortenings (Figure 1). The change in the configuration of the double bond means that the acyl chain more closely resembles that of saturated FAs. The accompanying hydrogenation of some of the double bonds also means, of course, that an important property of the food oils - the high content of polyunsaturated FAs - is sacrificed in favor of increased stability and a higher melting point (Belitz and Grosch 1994, Steinhardt 1996).
        Many foods made with milk or hydrogenated oils (for example: deep-fry, chips, butter, cheese, margarine) contain t-FAs. Normally, they cannot be found in products made from non-hydrogenated vegetable oils. Table 1 compares the content of t-FAs in various foods. Whereas products of milk and hydrogenated oils have a high content of t-FAs, the vegetable oils and their products e.g., mayonaise, have no more than 0.4 % and often have none.
        The physiological effects of these hydrogenated dietary fats, and especially the influence of t- FAs, on human metabolism have therefore been of great scientific interest for a long time. For example, t-FAs have been connected with arteriosclerosis and cancer, but up to now, these investigations are incomplete and controversial (Ascherio et al. 1997, Ärzte Zeitung 1994, Christiansen et al. 1997, Kohlmeier et al. 1997, Katan et al. 1994, Koga et al. 1997, Lichtenstein 1993, VDD-Mitteilungen 1994, Shapiro et al. 1997, Wolfram 1994).
        The presence of t-FAs also becomes a matter for discussion when vegetable oils, such as hemp seed oil, are used for cooking or frying (Huppertz et al. 1997). We have therefore heated hemp oil at several temperatures and analysed its deterioration and isomerisation.

Materials and Methods
        Various samples of a hemp seed oil were held in an electric heater: (1) for 30 min. at a constant temperature between 170°C and 350°C and (2) for 16 hours between 200°C and 220°C. A 10-ml sample was then homogenised with 1 ml methylene chloride (CH 2 Cl 2 ). 100 ml trimethylsulfonium hydroxide (TMSH) was added for quantitative hydrolysis of the triglycerides (Mölleken and Theimer 1997a,b) and conversion of the resulting FAs to FA methyl esters (FAMEs). The resulting FAMEs were analyzed on a HP 5890 gas chromatograph equipped with an FID detector, using compound standards from Sigma (Deissenhofen, Germany) for comparison.

Results
        Table 2 compares the influence of various temperatures on the t-FA content of hemp seed oil. It is clear that cooking temperatures of about 170-250°C do not lead to an increase of the t-FAs. Similar investigations with other hemp seed oils underscore these results. Only two oils had a t-FA content of 0.85% after being heated at 220°C for half an hour (Mölleken and Melchior 1998). In contrast, a temperature of 350°C deteriorates the hemp oil, and leads to the formation of significant amounts of t-FAs.
        This fact is confirmed by the results in Table 3. Some t-FAs can be detected in hemp oil that is held at 200 to 220°C for a long period (16 hours). Thus, isomerisation does take place under prolonged moderate stress, so that t-FAs are formed and the amounts of the unsaturated FAs decrease.

Discussion and Conclusion
        Both experiments demonstrate that isomerisation to t-FAs does not occur when native hemp oil is used under normal cooking conditions, though the contrary seems to be generally accepted (Huppertz et al. 1997). Experiments with various vegetable oils (for example from sunflower, safflower, soy bean or walnut) give similar results (Mölleken and Melchior 1998). The comparison of Table 1 and Table 2 also supports this conclusion. It is obvious that hydrogenated vegetable oils like margarine and deep-fry fats contain t-FAs in higher amounts than heated hemp oil.
        This does not mean that the polyunsaturated FAs of hemp oils are stable towards all negative influences. Under especially stressful conditions they do change their configuration or get destroyed, and then t-FAs, hydroxy FAs, hydroperoxides, aldehydes and ketones can be formed (Mölleken and Melchior 1998).
        Heated native hemp oils are quite stable under high-temperature conditions. Temperatures of 170-250°C do not stress this oil, so that high concentrations of t- FAs are scarcely formed. Presumably antioxidants, such as tocopherols, stabilise the oil, since hemp oils contain enough gamma-tocopherol to have a strong antioxidant effect (Belitz and Grosh 1994, Gordon and Kourimská 1994, Mölleken and Andersen 1998, Pryzybylski et al. 1997). Further investigations show that the influence of temperature on various vegetable oils leads to degradation of the tocopherols (Mölleken & Melchior 1998).
        While as much as 6-10 g of t-FAs per person per day is thought not to affect health adversely (Ärzte Zeitung 1994, Wolfram 1994), the current mean human uptake through normal nutrition (milk, cheese, margarine and deep-fried products, cf. Table 1) has been estimated at only 3-4 g per day (Gertz 1996). It seems reasonable, nevertheless, to keep the amounts of t-FAs in the diet to a minimum for general health purposes, and in accordance with this, many people, including scientists, recommend: "Use high-quality vegetable oil for frying, deep-frying and salads" (Ärzte Zeitung 1994, Wolfram 1994). Hemp oil fulfills this requirement, since like other vegetable oils, as we show here, it does not form significant amounts of t-FAs on being heated at usual cooking temperatures.

• Ascherio, A. and Willett, W. C. 1997. Health effects of trans-fatty acids. Am. J. Clin. Nutr. 66 (suppl.): 10006S-10010S.
• Ärzte, Zeitung 1994. Wenig trans-Fettsäuren in deutscher Margarine. Stellungnahme zu Berichten über schädliche trans-Fettsäuren. June.
• Belitz, H.-D. and W. Grosch 1994. Lehrbuch der Lebensmittelchemie. Springer Verlag Berlin, Heidelberg, New York , London.
• Christiansen, E. et al. 1997. Intake of a diet high in trans monosaturated fatty acids or saturated fatty acids. Effects on postprandial insulinemia and glycemia in obese patients with NIDDM. Diab. Care 20(5): 881-887.
• Demmelmair, H. et al. 1996. Trans fatty acid contents in spreads and cold cuts usually consumed by children. Ernährungswiss. 35(3): 235-240.
• Fernandez San Juan, P. M. 1996. Study of isomeric trans-fatty acid content in the commercial Spanish foods. Int. J. Food Sci. Nutr. 47(5): 399-403.
• Gertz, C. Z. 1997. Trans-Fettsäuren in Lebensmitteln. Lebensmittel-Unters., -Forsch. 50: 50-52.
• Gordon, M. H. and L. Kourimská, 1995. Effect of antioxidants on losses of tocopherols during deep-fat frying. Food Chem. 52: 175-177.
• Huppertz, R., M. Karus and F. Grotenhermer 1997. Hanfsamen und Hanföl. Ernährungs-physiologischer und therapeutischer Wert. nova-Institut 3/97.
• Katan, M. B., W. Zockx and R. P. Mensink 1994. Effects of fats and fatty acids on blood lipids in humans: An overview. Am. J. Clin. Nut. 60 (Suppl.): 1017S-1022S.
• Koga, T. et al. 1997. Linoleic and alpha-linolenic acids differently modify the effects of elaidic acid on polyunsaturated fatty acid metabolism and some immune indices in rats. Br. J. Nutr. 77(4): 645-656.
• Kohlmeier, L. et al. 1997. Adipose tissue trans fatty acids and breast cancer in the European Community Multicenter Study on antioxidants, myocardial infarction, and breast cancedial infarction, and breast cancer. Cancer Epidemiol. Biomarkers Prev. 6(9): 705-710.
• Lichtenstein, A. 1993. Trans fatty acids, blood lipids and cardiovascular risk: where do we stand? Nutr. Rev. 51(11): 340-343.
• Mölleken, H. and G. Andersen 1998. Tocopherols in different hemp extracts. (in preparation)
• Mölleken, H. and D. Melchior 1998. Vegetable oils under oxygen and temperature stress (in preparation).
• Mölleken, H. and R. R. Theimer 1997a. Evaluierung von C. sativa Saatgutherkünften im Hinblick auf eine verbesserte Ölqualität. Biorohstoff Hanf (2nd edition): 485-499.
• Mölleken, H. and R. R. Theimer 1997b. Survey of minor fatty acids in C. sativa L. Fruits of various origins. Journal of the International Hemp Association 4(1): 13-17.
• Mörsel, J.-T., C. Mörsel and R. Godeck 1997. Methoden der trans-Fettsäurebestimmung. Lab. Prax. Febr.: 80-85.
• Ovensen, L., T. Leth and K. Hansen 1996. Fatty acid composition of Danish margarines and shortenings, with special emphasis on trans fatty acids. Lipids 31(9): 971-975.
• Pryzybylski, R., J. Moes and A. Sturko 1997. Effect of growing conditions on composition of hemp oils. Biorohstoff Hanf (2nd edition): 505-514.
• Sommerfeld, M. 1983. Trans-unsaturated fatty acids in natural products. Prog. Lipid Res. 22: 221- 233.
• Steinhardt, C. 1996. Conjugated linoleic acid - the good news about animal fat. J. Chem. Educ. 73 (12): A302-A303.
• Steinhardt, H. and A. Pfalzgraf 1994. Trans-Fettsäuren in Lebensmitteln. Fat. Sci. Technol. 96(2): 42-44.
• Ulberth, F. and M. Henninger 1994. Quantitation of trans fatty acids in milk fat using spectroscopic and chromatographic methods. J. Dairy Res. 61: 547-527.
• VDD-Mitteilungen 1994. Diätetik in der Praxis. Trans-Fettsäuren und Herz-Kreislauf-Erkrankungen. Ernäh-rungs-Umschau 41(2).
• Wolfram, G. 1994. Transfettsäuren sind kein besonderes Ernährungsrisiko. Ärzte Zeitung October.
• Woollett, L., C. M. Daumerie and J. M. Dietschy 1994. Trans-9-octatetradecenoic acid is biologically neutral and does not regulate the low density lipoprotein receptor as the cis isomer does in hamster. J. Lipid Res. 35(9): 1661-1673.