Present trends in vitamin E research
Nearly after one century of research and thousands of publications, the physiological function(s) of vitamin E remain unclear. Available evidence suggests a role in cell homeostasis that occurs through the modulation of specific signaling pathways and genes involved in proliferative, metabolic, inflammatory, and antioxidant pathways. Vitamin E presence in the human body is under close metabolic control so that only α-tocopherol and, to a lower extent, γ-tocopherol are retained and delivered to tissues. Other vitamin E forms that are not retained in the body in significant amounts, exhibit responses in vitro that are different form those of α-tocopherol and may include tumor cell specific toxicity and apoptosis. These responses provide a therapeutic potential for these minor forms, either as such or metabolically modified, to produce bioactive metabolites. These cellular effects go beyond the properties of lipophilic antioxidant attributed to α-tocopherol particularly investigated for its alleged protective role in atherosclerosis or other oxidative stress conditions. Understanding signaling and gene expression effects of vitamin E could help assign a physiological role to this vitamin, which will be discussed in this review. Besides vitamin E signaling, attention will be given to tocotrienols as one of the emerging topics in vitamin E research and a critical re-examination of the most recent clinical trials will be provided together with the potential use of vitamin E in disease prevention and therapy.
1. Cellular effects and signaling of vitamin E
Within the cell body α-tocopherol shows well defined biochemical and molecular effects that indicate the existence of specific signaling events associated with the function of this vitamin.
Signaling of vitamin E could originate within the different lipid environments of plasmalemma and organelles, where it is delivered consequently to subcellular distribution targets that may discriminate between this vitamin and other fat soluble factors . These are represented by cytosolic proteins that bind with their hydrophobic domains vitamin E and regulate its trafficking and subcellular localization. Sec14p-like proteins are prototype components of this cell vitamin E regulation system that may play also a key role in the signaling of vitamin E and other lipids . TAP/Sec14L2 is one of the members of this family that is highly expressed in tissues that have been identified to have the highest responsiveness to the anticancer activity of vitamin E, that is prostate and breast, and the evidence of a reduced expression in breast cancer cells  may suggest a role for this protein in combination with vitamin E as tumor suppression factor.
Intracellular players in signaling expected to be influenced by the interaction between these proteins and vitamin E are numerous (reviewed in [4–7]) and include kinases such as PKC, Akt/PKB, some members of MAPK family and cell cycle related kinases, and the recently proposed downstream components of the death domain and proteins of the endoplasmic reticulum stress signaling . Tocotrienols have been reported to be ligands of the estrogen receptor β (ERβ)  that could thus contribute some aspects of vitamin E signaling that will be further discussed later in the section dedicated to tocotrienols.
The fact that vitamin E is effective in scavenging lipid radicals has suggested that its distribution may strategically follow that of PUFA species in membranes . This view, however, is not universally accepted and an association of α-tocopherol with lipid rafts has been recently described . This association was observed for α-tocopherol but not for γ-tocopherol. 7-ketocholesterol and α-tocopherol presence in the sphingolipid/cholesterol-enriched domains appeared to be mutually exclusive. Akt-PKB dephosphorylation was prevented by α-tocopherol, but not γ-tocopherol pretreatment. Such a subtle difference between α-and γ-tocopherol appears to be related to a protein recognition event, although the authors are still debating the issue of a complex local antioxidant property. The importance for this association can be inferred from the role that lipid rafts have been proposed to play in cell signaling and ROS flux control through the activity of the ROS-generating enzyme NADPH-oxidase (reviewed in ).
Another example of proteins that respond to vitamin E in the proximity of the membrane is the complex Nrf2/Keep1 . This complex stabilized by disulphide bridges is associated with the membrane, where it represents the inactive form of a redox sensitive component with early effects on cell signaling and protection responses elicited during stress conditions. Phenolics and several xenobiotics with electrophilic properties are capable to activate this complex releasing Nrf2 that after nuclear translocation activate the transcription of several antioxidant and phase II detoxifying genes of the ARE sequence . Accordingly, it has been reported that vitamin E influences the AhR/Nrf2/NFkB pathway in the lung of α-TTP deficient animals exposed to cigarette smoke  and in the in vivo anticancer activity of dietary γ-tocopherol . Thus Nrf2 together with other receptors and master controllers of cell metabolism, could contribute to vitamin E signaling and function in several tissues. This aspect is particularly relevant in the GI tract where beside to Ahr vitamin E has been reported to bind some members of the family of nuclear receptors [16, 17]. Vitamin E has been proposed to be a ligand of PXR that after activation and heterodimerization with RXR translocates into the nucleus to promote the transcription of genes primarily involved in the metabolism of vitamin E and other lipids. Other orphan receptors could be involved in vitamin E signaling. Likely to other lipid ligands, tocopherols may sustain the transactivation of RXR with other receptors (such as PPARγ) that through homo or heterodimerization can control different groups of genes such as antioxidant and cell protection genes associated with the metabolism of glutathione  and hemeoxygenase 1 . PPARγ activity is influenced by vitamin E  with possible role in the control of inflammatory and cell survival pathways. Also the effect that vitamin E can have on COX-2 activity could sustain PPARγ-dependent signals through a lowered generation of cell prostaglandin E2α (PGE-2α), which is a physiological PPARγ inhibitor .
Besides its physiological meaning, the binding to nuclear receptors might produce a competition between the metabolism of vitamin E and that of drugs, xenobiotics, and nutrients with particular regard to lipid metabolism. Although the activity of the main drug metabolizing gene CYP3A4 does not seem to be influenced by the supplementation with high doses of α-tocopherol in humans , other receptors and genes should be taken into account and further investigated. For instance, the responses elicited by oxylipids on lipid metabolism and particularly on cholesterol transport through receptors such as PPARγ and LXRα are of potential relevance in the risk of toxicity by chronic exposure to high doses of α-tocopherol. In THP-1 macrophages α-tocopherol was observed to inhibit the constitutive and oxLDL stimulated activity of LXRα that together with the expression of CD36 and some isoforms of ABC transporters is responsible of the control of the inverse cholesterol transport . On the other hand, some common therapy protocols and generally used drugs could interfere with vitamin E metabolism and levels. This is the case for instance of lipid lowering drugs and particularly of statin therapy , indicating the need of a constant evaluation of the vitamin E status in some group of patients to avoid complications and the risk for a relative vitamin E deficiency .
Evidence was provided of the fact that vitamin E signaling could be coupled with the control of enzymes involved in lipid signaling. The formation of a complex and a possible competitive inhibition by α-tocopherol were described for phospholipase A2 isoenzymes [24, 25], and other mechanisms that include the control of substrate and cofactor availability could be responsible of the modulation of COX-2, 5-lipoxygenase and cPLA2 (reviewed in [19, 26]). The γ form of vitamin E and more recently long chain metabolites were confirmed to be the most efficient COX-2 inhibitors in vitro and in vivo .
2. α- and γ-carboxyethyl-6-hydroxychromans (CEHC) and long chain metabolites
Liver uptake and blood transfer of α-tocopherol are less effective than in the case of α-tocotrienol and γ-tocopherol, which have been extensively demonstrated to be the most potent in vitro and in vivo modulators of cell signaling and gene expression [27–29]. Tocotrienols have been repeatedly reported to be more effective than α-tocopherol as neuroprotective, hypocholesterolemic, and anticancer agents  and further support to the importance of minor tocopherol forms is coming from the clinical side, where supplementation with γ-tocopherol alone or in combination with α-tocopherol, but not α-tocopherol alone, was found to improve antioxidant and anti-inflammatory parameters of patients with metabolic syndrome .
A relationship between this molecular activity and the extent of their metabolic processing was hypothesized  and could be explained on the bases that both α–tocopherol and minor vitamers could be effectively recognized by signaling proteins other than PXR that stimulate beside metabolizing genes other genes responsible for the many-sided response to vitamin E.
In the last two decades the discovery of hepatic metabolites and the identification of the genes and proteins involved in it have brought about significant advancements in the understanding of bioavailability and physiological functions of vitamin E. Furthermore one or more hepatic metabolites could represent active forms of the vitamin.
This metabolism centered in the liver (described in detail in ) efficiently transforms vitamin E to chromanol metabolites through a process of enzymatic shortening of the pythyl chain with an initial step of cyt P450-catalyzed ω oxidation of the terminal methyl group, which occurs in microsomes with a main role of the CYP4F2 isoform , which is followed by a β-oxidation process that rapidly proceed to the formation of the final product 2,2′-charboxyethylhydroxy chroman acid (CEHC). This latter phase of vitamin E metabolism was originally suggested to occur in peroxisomes, but more recent evidence obtained in rats supplemented with α-tocopherol suggests that it may localize also to liver mitochondria where CEHC were found to form to the highest extent .
As a result of this metabolism, CEHC are produced in the liver and excreted in bile and urine as main vitamin E metabolites . Metabolites with side chain of different length have been also identified to transiently accumulate in liver cells and also in other cell types loaded with vitamin E . Prostate cancer cells (PC3) were also demonstrated to produce high intracellular levels of CEHCs when exposed to tocotrienols  and these metabolites were tentatively identified to possess some biological activities resembling those of their vitamer precursors, such as that of antioxidant, anti-inflammatory, and antiproliferative agents, and in the case of γ CEHC metabolite also of mild natriuretic factor. These functions occur through signaling effects that CEHC metabolites produce with a close analogy with their vitamin E precursors (reviewed in [7, 27, 29]). Other than for the final products of vitamin E metabolism CEHCs, signaling effects have been more recently described in the case of long chain metabolites that are present in the circulation and are formed preferentially in some cell lines such as A549 alveolar type II epithelial cells .
These aspects regarding differences on the molecular and cell events promoted by the vitamers and their metabolites clearly suggest that vitamin E physiological responses are based to a large extent upon redox independent mechanisms. In fact the antioxidant features of the different vitamers are essentially the same . This is also the case of CEHCs and long-chain metabolites that produce the same type of chromanoxyl radicals of the vitamer precursors  with the exception that these have different degrees of hydrophobicity that may affect their partitioning within cell lipid environments and the cytosolic water.
3. Phosphorylation: a way to activate vitamin E signaling and function
Moreover, a recent series of studies has demonstrated the existence of a phosphorylation pathway for the main forms of vitamin E (reviewed in  and more recently in ). This pathway may serve to activate vitamin E and might represent a biochemical event, analogous to other signaling molecules, at the basis of the biological effects of this vitamin at the cellular level. α-Tocopherol can become phosphorylated both in isolated cells and in vivo in experimental animals; tocopheryl phosphate has been also shown to be in enzymatically dephosphorylated. Because of the poor reactivity of α-tocopherol and the chemical stability of tocopheryl phosphate, the existence of enzymes with tocopherol kinase or tocopheryl phosphate phosphatase activity must be postulated. The vitamin E (α-tocopherol) derivative, α-tocopheryl phosphate is present in measurable amounts in plasma, tissues, and cells. Animals and humans supplementation has shown that tocopheryl phosphate can reach plasma concentrations similar to tocopherol in rabbits  and in humans (Zingg et al, unpublished). Furthermore, its dephosphorylation is not efficient. Tocopheryl phosphate modulates the activity of several enzymes; in cell culture, it affects proliferation, signal transduction, and gene expression at concentrations lower than tocopherol (Zingg and coworkers, ). In high cholesterol fed rabbits and LDL knockout mice tocopheryl phosphate prevents atherosclerosis at much lower concentrations than tocopherol . In the ischemia/reperfusion injury of rat heart , tocopheryl phosphate may act at the molecular level as an active lipid mediator similar to phosphatidylinositol therefore influencing signal transduction and gene expression.
4. Distribution and signaling in target tissues
A-tocopherol reaches peripheral tissues thanks to transport mechanisms common to other lipoprotein-associated lipids (reviewed in [28, 43]). Similarly to other fat soluble micronutrients such as carotenoids, its concentrations in plasma are clearly under the control of genetic factors; this results in different phenotypes in terms of tocopherol plasma concentrations and response to dietary supplements [44, 45]. Specific polymorphisms of the apolipoprotein C-II, CEPT and hepatic lipase have been proposed to influence fasting plasma concentrations of vitamin E , and the Hp 2-2 polymorphism is associated with reduced levels of vitamin E and C . Genetic factors are now identified to play a role in the physiological control of the intestinal absorption and transfer of vitamin E to chylomicrons and intestinal HDL [47, 48], and thus in the liver uptake and incorporation of vitamin E into nascent VLDL for peripheral tissues distribution . Genetic factors are also at the basis of vitamin E catabolism and biliary elimination by the ABC transporters .
Tissue vitamin E delivery and uptake mechanisms in the extra-hepatic tissues remain poorly characterized and appear to be essentially associated to lipoprotein receptor expression and function (reviewed in ). In some tissues, specific delivery mechanisms have been proposed and could contribute to differentiate uptake rate and content of the diverse vitamers. For instance, γ-tocopherol has different distribution in human tissues and is higher for instance in skin, skeletal muscle, vein, and adipose tissue . HDL may play a role in the distribution and reuptake of vitamin E in several tissues possibly mediated by the function of the SR-BI lipoprotein receptor that has a main role also in intestinal absorption (reviewed in ). It could play a role in the differential distribution of α-tocopherol and γ-tocopherol in peripheral tissues could be and in the different extent of their metabolic processing in women and men [52, 53]. SR-BI lipoprotein receptor might contribute also to the selective lipid, and thus vitamin E uptake of brain  where recent evidence has suggested that also apoE could be involved in α-tocopherol transport. Indeed, brains of apolipoprotein E deficient mice fed vitamin E deficient diets show alteration in handling α-tocopherol injected into the cerebral ventricles .
In vivo studies in α-TTP deficient animals suggest that the brain is between the most vulnerable organs in the presence of defective vitamin E availability. Brain seems to resist more than other tissues to vitamin E depletion, but in the presence of a chronic reduction of vitamin E availability by either lowered intake or deficient α–TTP expression, it clearly shows signs of modified mitochondrial metabolism and an apparent paradoxical decreased oxidative stress . In a mouse model of Alzheimer's disease the genetic deficiency of vitamin E was found to stimulate amyloid β peptide accumulation, a key biochemical abnormality of this condition . These aspects suggest that transport and signaling by α-TTP and possibly other vitamin E binding proteins could be far more important than providing a high dietary intake of vitamin E. As α-TTP and other binding proteins are widely expressed in extra-hepatic tissues where they are involved not only in uptake and trafficking, but also in vitamin E signaling (see earlier for a more detailed discussion), it is not excluded that a loss in this latter function could represent a risk factor for developing a neurodegenerative phenotype.
Delivery and function of vitamin E in lung tissue and lining fluids have been recently proposed to be influenced in vivo by the expression of apolipoprotein E genotype . The distribution of cytosolic proteins such as TAP/Sec14L2 varies in the different tissues . This protein and others that bind with low affinity vitamin E and contribute to its trafficking and signaling might also play a role to determine steady state levels of vitamin E in tissues.
As introduced earlier, a new molecular mechanism for tocotrienol has been proposed recently that involves estrogen receptor signaling . Computer simulations and in vitro binding analyses indicated a high affinity of tocotrienols for the estrogens receptor β but not for the α form. Tocotrienols increase the estrogens receptor β translocation into the nucleus, with consequent activation of estrogen-responsive genes. These genes include COX-2 and telomerase-related antiapoptotic enzymes , which might explain some of the anti-inflammatory and cell cycle regulatory effects of tocotrienols described in this sections. But emerging evidence suggests a even more complex series of effects of this estrogen receptor form that responds to ligands of the EGFR/ErbB family of receptor tyrosine kinases and contains phosphorylation sites for the kinases MAPK/Erk and Akt/PKB , which may further support a role of this receptor in tocotrienol signaling. This receptor is expressed also in several other cell types that include prostate cancer cells , and thus it could be involved in a proposed anticancer role of tocotrienols .
Different studies have, in fact, suggested that tocotrienols suppress proinflammatory markers and COX-2 expression in macrophages in vitro Tocotrienols have been also observed to affect the insulin-dependent signaling in differentiating preadipocytes . This signaling effect of tocotrienols seems to be particularly pronounced in the case of the γ form and is related with a specific control mechanism that involves Akt/CEBP and PPARγ.
Recent studies on neuroprotective effects of tocotrienols showed the existence of a mechanism that preferentially enrich the neuronal cells of γ-tocotrienol, which is also responsible for a cooperative effect in the uptake of α-tocopherol . As introduced in the previous section, cell uptake efficacy is higher for the minor forms of vitamin E as compared to α-tocopherol and it represents a leading aspect to discriminate between the potency as signaling effectors of individual tocols . Moreover, the preferential uptake of tocotrienols could sustain the cell uptake of α-tocopherol through a so far unidentified cooperative mechanism already described for γ-tocopherol  thus providing a mechanism for the physiological control of vitamin E status in the cell. Similar findings on a preferential uptake of tocotrienols were reported in other cell lines such as Jurkat cells  and were associated with the survival promoting (anti-apoptotic) activity of these forms of vitamin E under stress conditions. Das et al.  demonstrated that tocotrienols prevent cardiomiocytes death induce by ischemia-reperfusion injury. This effect seems to occur through the inhibition of pp-60(c-Src) kinase, which is implicated also in the α-tocotrienol induced neuronal protection against different cell stresses , and also involves other players, such as prosurvival components of MAPK family, Akt signaling, and proteasome stabilization. The preferential uptake of tocotrienols may thus produce a cell protection equivalent to that of other phenolics at submaximal toxicity .
During the last years, the anticancer effects have been the most investigated biological aspect of tocotrienols; in particular, in the hormone-dependent malignancies, tocotrienols have shown several and apparently distinct cell mechanisms [69, 70]. Among these the suppression of mevalonate synthesis by the inhibition of the 3-hydroxy-3-methyl- glutaryl-CoA reductase activity, directly responsible for the cholesterol lowering activity of tocotrienols, also has been suggested to play a role in the antiproliferative activity of these vitamers. In fact, γ-tocotrienol was found to arrest the cell cycle transition to G1 phase synergistically with statins in mammary tumor cells . Interestingly, the antiproliferative pathway of statins and γ-tocotrienol is selectively mediated by the inhibition of the cyclin D1 pathway through p27 phosphorylation and inactivation of the downstream cyclin kinase cdk2. The same pathway is involved also in the antiproliferative effect that γ-tocopherol exerts in diverse cancer cell lines such as PC3 prostate carcinoma and C6 murine glioma cells [36, 63, 72]. This aspect suggests that tocopherols and tocotrienols have similar and possibly overlapping mechanisms.
In another study by Wali et al.,  in breast cancer cells other mechanisms were tentatively identified as involved in the anticancer effect of this form of vitamin E that was found to induce apoptosis through endoplasmic reticulum stress signaling.
Studies on palm oil derived tocotrienol-rich fractions (TRF) have suggested that the in vitro and in vivo anticancer activity of tocotrienols may depend also on antiangiogenetic effects by decreasing the production of vascular endothelial growth factor (VEGF) .
6. Vitamin E and cardiovascular disease
A critical reconsideration of the rationale at the basis of the effort made to identify a therapeutic role for vitamin E (mainly as α-tocopherol) in chronic and age-related diseases has occurred in the last years. Cardiovascular disease (CVD), following the original work of Fred Gey [74, 75] showing a geodistribution of this ailment that inversely correlated with the vitamin E intake of the studied populations, has been the object of a large series of in vitro and in vivo studies that initially suggested a role of this vitamin in the prevention of early atherosclerotic lesions (recently reviewed in  and ) and in the regulation of immuno-inflammatory, coagulative and cell protection mechanisms (reviewed in ). Observational evidence of a role of vitamin E in the primary prevention of CVD came mostly from prospective cohorts such as the Nurses' Health Study and the Health Professionals' Follow-Up Study (reviewed in ).
In the last two decades, a series of randomized double-blind intervention trials have been carried out. Apart from the CHAOS and the SPACE studies, both the first series of trials carried out essentially before 2000 (reviewed in ) and more recent trials [77–79] have failed to provide evidence of a beneficial effect on the primary endpoints of CVD, even if effects on secondary endpoints were identified in some trials.
As one of the most recent examples, in the recent evaluation of the Physicians' Health Study II (PHS-II) trial, vitamin E as well as vitamin C was not found to promote CVD prevention . In this large and long-term (up to 10 years), randomized, double-blind, placebo-controlled factorial trial, daily supplements of 400 IU and 500 mg of vitamin E and vitamin C, respectively, in 14,641 US male physicians that at the evaluation were 50 years of age or older, did not significantly affect the composite end point of major cardiovascular events (nonfatal myocardial infarction, nonfatal stroke, and cardiovascular disease death) during a mean follow-up of 8 years. The Authors concluded that “neither vitamin E nor vitamin C supplementation reduced the risk of major cardiovascular events. These data provide no support for the use of these supplements for the prevention of cardiovascular disease in middle-aged and older men”.
The unambiguous conclusion from these trials appears to be that using vitamin E supplements, as well as any other antioxidant supplement, to prevent CVD progression and mortality is not recommended, and rather it should be discouraged. However, an optimal intake of natural vitamin E, introduced with a well-balanced diet, appears to be best way to obtain prevention effects, if any, on myocardial infarction, stroke and other atherothrombotic conditions and is now a well accepted health recommendation that has been extended also to all the natural antioxidants in food .
7. Vitamin E and cancer
The same conclusion reported in the previous section on CVD prevention seems to apply for vitamin E, or better α-tocopherol, in the primary and secondary prevention of cancer. After waves of optimistic observational and cohort studies , large randomize trials have provided now robust evidence on the absence of any clinical advantage in the tumors so far most investigated. This is the case in particular of prostate cancer that was found to be positively influenced in terms of incidence and number of deaths in the α-Tocopherol, B-Carotene Cancer Prevention (ATBC) trial, but not in other trials that include the HOPE-TOO, PLCO, and HPFS trials (reviewed in ) and in the more recent and largest trials on vitamin E, that is the SELECT study  and the Physician's Health Study II (PHS-II) . It is noteworthy that the SELECT study has been interrupted at the first end-point evaluation, set after 7 years of the planned 14 years, due to the lack of differences between the two groups under investigation. In the PHS-II that recently completed its follow-up neither the risk of prostate cancer nor that of total cancer was affected by the supplementation with vitamin E or C.
Clinical trials on vitamin E prevention in other types of cancers are fewer than in the case of prostate cancer, but the final conclusion reported in reviewing the available literature is again the same . According with this poor outcome of vitamin E supplementation in cancer, this vitamin as well as other antioxidants was found to be ineffective also in the secondary prevention of precancerous gastric lesions .
However, some positive evidence suggests a possible role of vitamin E in protection and primary prevention of specific cancers. In a recent study, indeed it was reported a lowered incidence of gastric noncardia adenocarcinoma in subjects taking high doses of supplemental α-tocopherol . In this study the gastric cardia adenocarcinoma was not affected by the vitamin E intake and also in one of the evaluations of the ATBC cohort study vitamin E intake did not correlate with the incidence of cancer in the upper aero-digestive tract . Secondary analysis of the data collected in the ATBC study has shown that male smokers with higher levels of α-tocopherol (and not β-carotene or retinol), besides a higher overall cancer survival , have a lower incidence of pancreatic cancer . Again, the levels of γ-tocopherol and α-tocopherol inversely correlated with the risk of ovarian cancer in another study .
Breast cancer was found to be one of the most responsive forms of cancer in preclinical in vitro and in vivo studies . In a study on postmenopausal women, in which blood micronutrients were actually determined, γ-tocopherol levels were associated with diminished risk of breast cancer  whereas no association of α-tocopherol and γ-tocopherol intake determined by dietary questionnaire was found in another study on Hispanic and nonHispanic women .
8. Neurodegenerative and other neurological conditions
At present, there is no substantial evidence that Alzheimer's disease and other forms of dementia such as mild cognitive impairment, as well as Parkinson's disease and other neurodegenerative conditions can be successfully prevented to any extent by the supplementation with vitamin E or other antioxidants [95–98]. In Alzheimer's disease patients, both preclinical evidence [99, 100] and small clinical studies  suggested the premise for a therapeutic perspective that needs, however, further substantiation. Idiopathic neurological disorders associated with symptoms resembling those of AVED should be further investigated at the clinical level and could be considered at least in certain cases for a symptomatic first-approach treatment with vitamin E. These may include some forms of inherited ataxias , and patients with tardive dyskinesia have been proposed as candidate population that may benefit of such a prophylactic treatment . Some symptoms in rare neurodegenerative conditions such as familial dysautonomia seem to be positively affected by tocotrienol supplementation , and vitamin E supplementation is recommended in secondary neurological disturbance that may rise over the long-term as a consequence of vitamin E deficiency in gastrectomy patients .
9. Immune response
Vitamin E has been suggested to exert the physiological role of immuno-homeostatic factor with regulatory effects on leukocyte sub-families and inflammatory pathways [105, 106].
It has been shown that supplementation with vitamin E restores age associated declines in T cell function both in healthy elderly and aged mice. Vitamin E exerts its effect by improving the ability of T cells to produce IL-2 and to advance through the cell cycle. Moreover, it increases effective immune synapse formation with redistribution of the signaling proteins Zap70, LAT, and Vav to the immune synapse .
The effect of vitamin E on the physiological cytokine production in the elderly has been investigated also in the case of TNF-α, and has been discussed in terms of individual responsiveness and inflammatory risk . After supplementation, the lipopolysaccharide-stimulated production of this cytokine in whole blood was lowered in those subjects bearing a specific single nucleotide polymorphism on the TNF-α gene that is associated with higher TNF-α levels. This pilot intervention study suggests that individual immune responses to vitamin E supplementation are in part mediated by genetic factors and could be specific for those subjects genetically predisposed to higher inflammation.
10. Vitamin E toxicity in humans
The evidence discussed in the previous section has clearly demonstrated the absence of a relevant preventive role of vitamin E in CVD. However, a debate on the discrepancy between the expected results and the negative evidence provided by the clinical trials is still open. A further complication has come from metanalysis studies revealing possible vitamin E toxicity in patients treated with high doses of α-tocopherol [109, 110]. These studies may have suffered from the study selection bias and from the used statistic methods . However, as discussed in a commentary by Blumberg and Frei  some functions of this vitamin and particularly that of an in vivo antioxidant might need very high dosages to be achieved [113, 114], compatible with potential damage and the supposed increase mortality [109, 110].
Vitamin E toxicity in humans may be related with depletion by α-tocopherol of blood and tissues γ-tocopherol, as indeed demonstrated in experimental animals and humans supplemented with α-tocopherol . As an example of this concept, α-tocopherol supplementation has been suggested to interfere with the anticancer signaling and tumor growth arrest that γ-tocopherol caused in mice bearing human breast cancer xenografts .
11. Individual and genetic predictors of vitamin E responsiveness
Individual factors can affect the possibility to obtain a response to vitamin E in clinical trials. Actually, the effect of vitamin E supplementation on morbidity and mortality varies between different population groups. More careful evaluation by subject stratification of ATBC study data showed that vitamin E may either prevent or increase the overall risk of mortality depending on age and vitamin C intake . More in detail, when the dietary vitamin C intake was above the median of 90 mg/day, vitamin E increased mortality among those aged 50–62 years by 19% and decreased mortality among those aged 66–69 years by 41%.
Also other recent studies have suggested similar conclusions. Divergent effects of α-tocopherol and vitamin C on the generation of dysfunctional HDL, which is associated with diabetes and the haptoglobin 2-2 genotype have been described . In this study, it has been concluded that “choosing the correct antioxidant in the correct subset of patients may be critical to demonstrate benefit from antioxidant therapy”. The Hp 2-2 polymorphism is associated with increased production of oxygen free radicals and reduces levels of vitamin E and C; the consequent elevated risk for cardiovascular disease could be prevented by vitamin E supplementation . Other individual factors affecting vitamin E effects could include life stile such as smoking habit, which has been recently suggested to increase the risk of tuberculosis when associated with vitamin E supplementation and with a high intake of vitamin C .
A number of molecular and cellular studies appear to favor non antioxidant molecular functions of vitamin E (tocopherols and tocotrienols). At large concentrations, toxic effects of tocopherols may results. Metabolic products of tocopherols, especially α-tocopheryl phosphate, promise to provide interesting future developments. Clinical trials aimed at establishing CVD and tumor preventing effects of tocopherols have been, however, largely inconsistent with the molecular and cellular data. There is a possibility that in selected populations and with particular types of tumors more successful outcomes be obtained.