Does Lowering Gamma-tocopherol Explain the Failure of Vitamin E Supplementation Trials?

Abstract & Commentary

By Diane L. McKay, PhD, FACN. Dr. McKay is an Assistant Professor at the Friedman School of Nutrition Science and Policy, and conducts research at the Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University; she reports no financial relationship to this field of study.

Synopsis: Several large observational studies have suggested that vitamin E supplementation lowers the risk of coronary heart disease. However, the results of large randomized controlled trials failed to show a clear benefit of a-tocopherol supplementation on cardiovascular outcomes. In this study, supplementation with high-, medium-, and low-doses of a-tocopherol, plus vitamin C, decreased plasma g-tocopherol in type 2 diabetics, but did not affect markers of oxidative stress, inflammation, or hypercoagulation following an atherogenic test meal. The findings suggest that a-tocopherol supplementation suppresses the beneficial effects of the g-tocopherol isomer. Although the health impact of lowering plasma g-tocopherol is an issue that requires further investigation, the authors' overall conclusions are unwarranted due to the marked limitations of their study.

Source: Gutierrez AD, et al. The response of gamma vitamin E to varying doses of alpha vitamin E plus vitamin C. Metabolism 2009;58:469-478.

A randomized, crossover study was conducted to determine if supplementation with a-tocopherol and vitamin C lowered plasma levels of g-tocopherol in a dose-dependent manner, and whether these changes affected surrogate markers of atherosclerotic risk following a high-fat meal. Subjects, including six men and six women (mean age, 53 years; BMI, 29 kg/m2) with well-controlled type 2 diabetes (mean duration, 9 years), were randomized to receive either no vitamins, low- (200 IU a-tocopherol + 250 mg vitamin C), medium- (400 IU a-tocopherol + 500 mg vitamin C), or high-dose vitamins (800 IU a-tocopherol + 1 g vitamin C) for a period of two weeks each. At the end of each treatment period, subjects were given an atherogenic high-fat test lunch, equivalent to a McDonald's Big Mac meal. Plasma levels of a- and g-tocopherols (lipid-adjusted), as well as markers of oxidative stress (oxidized LDL, malondialdehyde, erythrocyte glutathione peroxidase), atherosclerotic risk (non-esterified fatty acids), inflammation (C-reactive protein, adiponectin, interleukin-6), and hypercoagulation (plasminogen activator inhibitor-1, fibrinogen) were assessed just prior to the meal and for five hours during the postprandial period.

After two weeks, a dose-dependent elevation in lipid-adjusted a-tocopherol levels was observed with the high-, medium-, and low-dose vitamins compared with the no-vitamin control. However, a 50% reduction in lipid-adjusted plasma g-tocopherol was observed at all three doses of a-tocopherol, indicating a threshold effect at or below the 200 IU dose. No significant between-group differences were observed for any surrogate markers, measured at the end of the two-week treatment period and again postprandially, when compared with the no-vitamin control. The authors conclude that the reason a-tocopherol supplementation had no effect on markers of atherosclerotic risk can be explained by the suppressed levels of circulating g-tocopherol.


Vitamin E is not a single molecule, but rather a group of eight isomers, including four tocopherols (a-, b-, g-, d-tocopherol) and four tocotrienols (a-, b-, g-, d-tocotrienol). Most vitamin E dietary supplements contain only a-tocopherol, while approximately 70% of the vitamin E from food sources in the United States is in the g-tocopherol form, largely due to the high intake of soybean and other vegetable oils, including canola oil, in the American diet.1 Although the ratio of g- to a-tocopherol is relatively high in the diet, it is essentially reversed in plasma. This is attributable to the preferential uptake of a-tocopherol by the hepatic a-tocopherol transfer protein. Because a-tocopherol is the predominant form of vitamin E found in plasma and tissue, it has been studied more extensively, both in vitro and in vivo, and serves as the "reference" form for achieving vitamin E requirements in humans.

All forms of vitamin E have some antioxidant activity. Recent data indicate that the different vitamin E forms also have other bioactivities unrelated to their antioxidant activity, including anti-inflammatory and immunostimulatory properties, which may also be important for maintaining and improving human health.2 In fact, several distinct molecular targets are thought to be involved in the anti-inflammatory effects mediated by a- and g-tocopherol. For example, g-tocopherol is a stronger inhibitor of cyclooxygenase in vitro, and possibly lipoxygenase, than a-tocopherol. Furthermore, g-tocopherol traps reactive nitrogen species in vitro more efficiently than a-tocopherol. While both tocopherols exhibit anti-inflammatory activity in vitro and in vivo, supplementation with mixed tocopherols seems to be more potent than supplementation with a-tocopherol alone.2

It is well known that supplementation with a-tocopherol alone lowers plasma levels of g-tocopherol, and studies exploring the potential consequences of this relationship are warranted. In this study by Gutierrez et al, type 2 diabetics, with presumably elevated oxidative stress levels, were supplemented with RRR-a-tocopherol at levels eight- to 36-fold higher than the RDA for two weeks to determine whether there is a dose-response relationship with plasma g-tocopherol levels. The fact that all doses used in this study lowered plasma g-tocopherol equally suggests a threshold effect and that the range of doses was insufficient to draw any meaningful conclusions regarding whether a true dose-response effect exists. Moreover, the authors report only post-supplementation levels relative to the no-vitamin control group, a critical flaw, and did not report whether there were any changes in plasma vitamin E levels between the pre- and post-supplementation period. Nor did they include any washout period between each arm of this crossover study. This, too, is a serious flaw as steady-state vitamin E status in tissues typically takes at least four weeks to achieve. Furthermore, no information on the vitamin E content of subjects' background diets prior to supplementation, or changes during the supplementation period, is provided. Previous studies have shown that subjects' response to vitamin E supplementation depends on their initial status, as well as their dietary intake, neither of which was considered in this study design.

A second objective of the Gutierrez et al study was to determine whether several biomarkers of atherogenic risk were affected after challenging supplemented subjects with a high atherogenic meal to further elevate their oxidative stress. An earlier study demonstrated that a single high-fat meal significantly reduced flow-mediated dilation (FMD), a measure of endothelial function, for up to four hours in healthy subjects, and no reduction was observed when vitamins C (1 g) and E (800 IU) were administered with the same meal, suggesting a potential oxidative mechanism.3 Gutierrez et al correctly note that type 2 diabetic patients show greater oxidative stress in response to a standardized meal when compared with healthy subjects. However, they fail to consider that these patients may also require a higher dose of antioxidant supplements, or a longer duration of treatment, to counteract the elevated stress. Interestingly, it was recently determined that vitamin E in doses of a least 1,600 IU/d for 16 weeks were necessary to suppress F2-isoprostane levels, a validated measure of oxidative stress, in individuals at risk for cardiovascular disease.4

Gutierrez et al conclude that the absence of an effect on all biomarkers with every level of a-tocopherol supplementation used in their study was due to the lowering of plasma g-tocopherol levels. Within the context of their study this conclusion is inappropriate as it could not be drawn from the experimental design, e.g., it is just as likely that their findings were due to the increase in plasma or tissue a-tocopherol or vitamin C. However, it is more likely that their subjects were well-nourished, and that the magnitude of change in their vitamin E status after two weeks of supplementation was minimal and had little influence on these biomarkers in the short duration of this intervention. This is supported by the fact that there were no significant differences between groups, including the no-vitamin control group, with regard to any of the oxidative stress biomarkers prior to administering the high-fat meal.

In a pooled analysis of several cohort studies, the increased consumption of total vitamin E and dietary vitamin E were each associated with a significantly lower risk of CHD, while vitamin E supplementation was not.5 Likewise, the results of large randomized controlled trials failed to show a clear benefit of a-tocopherol supplementation on cardiovascular outcomes.6 There are several plausible explanations for these discrepant findings, and the role of plasma g-tocopherol should certainly be considered. However, based on the findings of Gutierrez et al, it cannot be concluded that the failure of the large trials was due to a reduction in g-tocopherol. The authors appropriately point out that the lack of well-standardized g-tocopherol supplements, absent the a-isomer, limits research into this relationship, and that further studies are warranted.


1. Dietrich M, et al. Does gamma-tocopherol play a role in the primary prevention of heart disease and cancer? A review. J Am Coll Nutr 2006;25:292-299.

2. Reiter E, et al. Anti-inflammatory properties of alpha- and gamma-tocopherol. Mol Aspects Med 2007;28: 668-691.

3. Plotnick GD, et al. Effect of antioxidant vitamins on the transient impairment of endothelium-dependent brachial artery vasoactivity following a single high-fat meal. JAMA 1997;278:1682-1686.

4. Roberts LJ 2nd, et al. The relationship between dose of vitamin E and suppression of oxidative stress in humans. Free Radic Biol Med 2007;43:1388-1393.

5. Mente A, et al. A systematic review of the evidence supporting a causal link between dietary factors and coronary heart disease. Arch Intern Med 2009;169: 659-669.

6. Sesso HD, et al. Vitamins E and C in the prevention of cardiovascular disease in men: The Physicians' Health Study II randomized controlled trial. JAMA 2008; 300:2123-2133.