25th Anniversary Linus Pauling Institute at Oregon State University banner


A Primer on Dietary Fat: The Good, the Bad and the Unknown

Rosemary Wander, Ph.D.
Associate Professor of Nutrition
and Food Management
LPI/OSU Faculty Affiliate Investigator

photo of Rosemary Wander

Dietary fat plays a significant role in the maintenance of health and well-being. The consumption of saturated fatty acids (SFAs), especially lauric acid and myristic acid, both found in palm and coconut oil, and palmitic acid from palm oil and animal fat is associated with an increased risk of cardiovascular disease (CVD), while the consumption of monounsaturated fatty acids (MUFAs), specifically oleic acid found in olive, canola, and sunflower oils, and polyunsaturated fatty acids (PUFAs) from corn, safflower, soybean, and sunflower oils may decrease CVD risk. Less clear, although of equal importance, is the unique role that individual PUFAs have on health.

Fatty acid structure

To appreciate the specific roles that different fatty acids have on health, it is first useful to understand their basic chemical structure. Fatty acids are long chains of carbon atoms bonded to each other; hydrogen atoms are bonded to each carbon atom. One end of the chain ends with a carbon atom to which three hydrogen atoms are attached. This end is called the methyl-, or omega-, end. The other end of the carbon chain terminates with an acid. The fatty acid molecules are grouped as saturated or unsaturated. A saturated fatty acid has one bond between each pair of carbon atoms in the chain. An unsaturated fatty acid has two bonds, called double bonds, between some of its carbon-atom pairs. An unsaturated fatty acid becomes "partially hydrogenated", called a "trans" fatty acid, when hydrogen is added to alter its characteristics. (Trans fatty acids, found in many processed foods, are possibly more detrimental to health than ordinary saturated fats.) Unsaturated fatty acids can be further divided into two groups: monounsaturated and polyunsaturated fatty acids. Monounsaturated fatty acids have a double bond between only one carbon-atom pair. Polyunsaturated fatty acids have a double bond between more than one carbon-atom pair.
chemical structure diagram of fatty acids

Nutritionally important polyunsaturated fatty acids belong primarily to two families: the (n-6), or omega-6, fatty acids and the (n-3), or omega-3, fatty acids, where n is the number of the carbon atom on which the first double bond occurs starting with the methyl end of the fatty acid molecule. Common dietary sources of (n-6) fatty acids are safflower, corn, soybean, and sunflower oils. The (n-3) PUFAs are found in small amounts in soybean and canola oils; eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are found in marine fish. Mammals can use linoleic acid to make other fatty acids from the (n-6) family and linolenic acid to make those of the (n-3) family, but cannot interconvert fatty acids from the two families.

Scientific evidence indicates that the functions of fatty acids from these two families differ. For more than four decades it has been known that (n-6) fatty acids lower serum cholesterol, an independent risk factor for CVD. More recently, however, high intakes of (n-6) fatty acids have been shown to have some adverse health effects, such as increasing tumor growth, elevating the formation of spontaneous thrombosis, and heightening inflammatory response. Health benefits of the (n-3) fatty acids, especially EPA and DHA found in fish oil, often oppose the deleterious effects of the (n-6) fatty acids.

(n-3) Fatty acids in heart disease

Numerous epidemiological studies have shown that the frequent consumption of fish leads to a decrease in CVD risk. This association has been attributed to EPA and DHA in fish oil because of their antithrombotic, antivasorestrictive, antihypertensive, and antiarrhythmic influences. These fatty acids may also inhibit atherosclerosis. In addition to lessening the risk of a first myocardial infarction, most studies have shown that (n-3) fatty acids prevent the early restenosis that occurs in 25-40% of patients who undergo angioplasty. While these fatty acids lower serum triglyceride concentrations, their effect on low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C) concentrations is variable. We showed that diets containing large amounts of fatty fish fed to normotensive males actually increased concentrations of serum LDL-C, and other studies have reached similar conclusions. The effect of (n-3) fatty acids on serum lipid concentrations appears to depend upon characteristics of the individual and whether the (n-3) fatty acids replace dietary saturated fatty acids.

Other therapeutic applications

(n-3) Fatty acids also have anti-inflammatory properties--joint pain in patients with rheumatoid arthritis was lessened when EPA and DHA were given along with nonsteroidal antirheumatic drugs. In patients with psoriasis, the oral intake of (n-3) fatty acids improved skin lesions. These fatty acids also appear to have a beneficial effect on patients with ulcerative colitis. In cancer studies with animal models, they appear to decrease the number and size of tumors and to impede their growth. Other research has indicated that (n-3) fatty acids may depress the immune response. For example, we measured a smaller immune response in aged Beagles fed diets containing large quantities of (n-3) fatty acids. Because of these demonstrated clinical properties of (n-3) fatty acids, there is an intense research focus on understanding their mechanisms of action.


In recent years, one of the most interesting findings about the (n-3) fatty acids is that DHA is essential for the normal functional development of the brain. DHA and arachidonic acid are the predominant structural fatty acids in the gray matter of the brain. Human brain growth takes place from the 25th week of gestation until two years after birth. During this period, DHA and arachidonic acid are supplied by the placenta in utero and in the diet after birth. Consequences of a deficiency of in utero exposure to these fatty acids are likely to be more severe in premature infants.
photo of alert baby

The retina, a tissue developed in the embryo from the same source as neural cells, is commonly used to investigate the role of DHA in brain development. Retinal function can be assessed by the visual acuity test. Visual acuity has been shown to develop more slowly during the first four months of life in infants fed formula that had a low concentration of long-chain PUFA than in those given human breast milk, which contains significant quantities of these same fatty acids. Even at the age of three, the breast-fed infants had better visual acuity. In preterm infants visual acuity also developed more slowly when they were given formula low in long-chain (n-3) fatty acids. When DHA was added to the formula, the visual acuity deficit could be normalized.

The impact of (n-3) fatty acids on brain development is also evident in intellectual performance. Infants fed (n-3) fatty acids later demonstrated better cognitive development than those not fed these fatty acids. One study measured the IQ of eight-year old children who had been given, as premature infants, (n-3)-poor formula or human milk. Those who received breast milk had scores 8 points higher.

Unfortunately, infant formula in the Western world is virtually lacking in long-chain fatty acids like DHA. Often they contain very little linolenic acid but large amounts of linoleic acid, resulting in a large (n-6) to (n-3) fatty acid ratio. Much debate has centered around the issue of adding (n-3) fatty acids to infant formula, a practice which has not yet been adopted in the United States.


One of the most exciting findings about (n-3) fatty acids to emerge from recent epidemiological studies is the association between the dietary intake of (n-3) fatty acids and major depression. Consistent with this observation, blood plasma concentrations of (n-3) fatty acids predict concentrations of a metabolite of serotonin, a biomarker for depression and aggression. Patients with major depression have a higher ratio of long-chain (n-6) to (n-3) fatty acids in plasma than do normal patients. The higher the ratio, the more severe are the depressive symptoms. Therefore, dietary advice to lower serum cholesterol by substituting (n-6) fatty acids for saturated fatty acids, thus increasing the (n-6) to (n-3) fatty acid ratio in the diet, is disquieting. A high intake of (n-6) fatty acids may explain the correlation between low plasma cholesterol concentrations and increased mortality due to suicide, homicide, and accidents.

Possible negative effects of (n-3) fatty acids

Despite the known beneficial effects of (n-3) fatty acids, there have been concerns that they may also have negative health effects. Researchers have speculated that oxidative stress in our bodies will increase as the content of fatty acids containing double bonds in the diet increases. Thus, consumption of EPA and DHA may increase in vivo (in the body) lipid peroxidation and the quantity of free radicals, which have been implicated in carcinogenesis, inflammation, atherosclerosis, aging, and various other diseases and disorders. However, very little information is available from human studies that relate the extent of in vivo lipid peroxidation to (n-3) fatty acids in the diet.

The determination of lipid peroxidation in the body induced by consumption of highly unsaturated fatty acids is hampered by the lack of a definitive measurement. The assay that has been used most frequently is called "thiobarbituric acid reactive substances" (TBARS). Although simple to use, it has many well-recognized limitations, such as its lack of chemical specificity, i.e. many molecules other than (n-3) fatty acids may cause an increase in TBARS. Other assays, although procedurally more difficult than the TBARS test, have recently been developed and may lead to a better understanding of the consequences of increased amounts of dietary EPA and DHA.

Previous animal and human studies that used TBARS to measure lipid peroxidation showed that diets rich in long-chain (n-3) fatty acids resulted in increased plasma concentrations of lipid peroxides. We recently conducted a study in which 48 postmenopausal women were given supplements of 4.3 g of long chain (n-3) fatty acids (2.5 g EPA and 1.8 g DHA) daily for five weeks. This dose of EPA/ DHA is comparable to that found in a 5-6 ounce serving of Chinook salmon. Since this amount is about the largest quantity of salmon one would be expected to consume, it would produce the maximum concentration of EPA and DHA in plasma and tissues obtainable by normal dietary means. We measured TBARS in plasma and urine, as well as four more specific measures of lipid peroxidation, and the results were equivocal: TBARS increased, two of the other measures increased, and two were unchanged.

We then conducted a second study in postmenopausal women using supplements of monounsaturated, (n-6), or (n-3) fatty acids. Using more specific measures of lipid peroxidation than previously used, our preliminary analysis suggests that supplements of the long-chain (n-3) fatty acids do not lead to increased oxidation. Similarly, other laboratories have shown that diets enriched in EPA and DHA do not cause increased in vivo peroxidation. The clearest evidence to show that EPA and DHA cause an increase in lipid peroxidation has come from in vitro studies in which the concentration of EPA and DHA often exceeded that which could be found in the body, even when large amounts of EPA and DHA are consumed.

If increased in vivo lipid peroxidation occurs when fish or fish oils are eaten, it may be prevented by a sufficient dietary intake of the antioxidant vitamin E. It is recommended that for each gram of linoleic acid, an (n-6) PUFA, consumed, 0.4 mg of RRR-a-tocopherol is needed to protect against peroxidation. However, there is little information available to suggest how the amount of vitamin E should be adjusted if the concentration of (n-3) fatty acids like EPA and DHA is increased. In another recent study, we gave postmenopausal women supplements of fish oil and either maintained their dietary intake of vitamin E or gave them supplements of 100, 200, or 400 mg of a-tocopheryl acetate per day. Lipid peroxidation in blood was measured by TBARS as well as more specific measures. Only the TBARS measurement decreased consistently with increasing doses of vitamin E.

Although further, more precise clinical testing is needed, these data collectively suggest that the consumption of (n-3) fatty acids may not increase in vivo lipid peroxidation. Until more studies are conducted, the possible negative effects of the consumption of (n-3) fatty acids remain speculative.

Historical perspective and dietary implications

Studies of paleolithic nutrition and modern-day hunter-gatherer populations have indicated that humans evolved on a diet that was much lower in saturated fatty acids than is the modern diet. The primitive diet contained small but roughly equal amounts of the (n-3) and (n-6) fatty acids. Over the past 50 years the intake of (n-6) fatty acids has increased markedly while that of (n-3) fatty acids has not. This change was stimulated by the Dietary Guidelines from the United States Department of Agriculture. The guidelines recommended that the ratio of the intake of PUFA be twice that of SFA to reduce cholesterol. At the time the Guidelines were developed, serum cholesterol was the major risk factor for CVD. Consequently, the use of soybean, safflower, corn, and, more recently, canola oils has increased. Increased consumption of vegetable oils was made possible because of two major factors. Technological developments occurred at the turn of the century that marked the beginning of the modern vegetable oil industry, and modern agriculture practices began to place an increased emphasis on grain feeds, rich in (n-6) fatty acids, for domestic livestock.

Vegetable oils contain large amounts of linoleic acid, which is the major PUFA in the American diet. For instance, soybean oil supplies about 70% of our vegetable oil market and is almost three-quarters linoleic acid. Consequently, the (n-6) to (n-3) fatty acid ratio of the Western diet has risen to about 20 or 25 to 1, with some estimates as high as 40 to 1.

The major source of (n-3) fatty acids in most American diets is linolenic acid, also from soybean oil. However, the influence of the linolenic acid content of soybean oil on the ratio of (n-6) to (n-3) fatty acids in the American diet is small. Much of the vegetable oil used in the United States is not consumed directly, but is hydrogenated in the process of making products like shortening and margarine. These products contain little linolenic acid. With the advent of genetic engineering, the fatty acid profiles of oilseeds are more easily modified than by traditional breeding practices. Oilseed producers are currently considering decreasing precipitously the (n-3) fatty acid content of soybeans because this fatty acid oxidizes more readily than other fatty acids, thereby shortening the shelf life of products that contain it.

The fatty acid content of the diet has changed without the benefit of a thorough understanding of the metabolic and nutritional impact. A more detailed understanding of the roles of (n-3) and (n-6) fatty acids is critical in order to make proper dietary adjustments.

Although the optimal ratio of (n-6) to (n-3) fatty acids in the diet is unknown, current evidence suggests that a goal of 5 to 1 may be appropriate. The United States has not made a formal recommendation, but Britain and Canada have recommended that the (n-3) fatty acid content of the diet should be about 0.5% of the total calories, or 1.0-1.5 grams per day. A 3.5 ounce serving of a fatty fish (mackerel, tuna, salmon, bluefish, mullet, sturgeon, menhaden, anchovy, herring, trout, or sardines) can generally provide this amount. Dietary advice about fats has been confusing, but as research progresses and we acquire better understanding, more consistent and accurate recommendations will emerge.

Last updated November, 1998

Honoring a Scientific Giant with Research Toward Longer, Better Lives

Return to Fall/Winter 1998 Table of Contents Return to LPI Home Page Please send any comments, suggestions, or questions about The Linus Pauling Institute to lpi@oregonstate.edu