|Home | About | Journals | Submit | Contact Us | Français|
To investigate the global effects of vitamin E supplementation on aging, we used high-density oligonucleotide arrays to measure transcriptional alterations in the heart and brain (neocortex) of 30-month-old B6C3F1 mice supplemented with α- and γ-tocopherol since middle age (15 months). Gene expression profiles were obtained from 5- and 30-month-old controls and 30-month-old mice supplemented with α-tocopherol (1g/kg), or a mixture of α- and γ-tocopherol (500mg/kg of each tocopherol). In the heart, both tocopherol supplemented diets were effective in inhibiting the expression of genes previously associated with cardiomyocyte hypertrophy and increased innate immunity, while having a moderate effect on age-related transcriptional alterations linked to the stress response and protein synthesis. In the brain, induction of genes encoding ribosomal proteins and proteins involved in ATP biosynthesis was observed with aging and was markedly prevented by the mixture of α- and γ-tocopherol supplementation, but not by α-tocopherol alone. These results demonstrate that middle age-onset dietary supplementation with α- and γ-tocopherol can partially prevent age-associated transcriptional changes and that these effects are tissue- and tocopherol-specific.
Reactive oxygen species (ROS) generated as a result of aerobic metabolism may play a role in normal aging of tissues that rely heavily on oxidative phosphorylation (1,2). Since ROS can cause damage to virtually all cellular macromolecules, aerobic organisms have developed elaborate antioxidant defense systems for the prevention of ROS-mediated cellular damage. ROS can be reduced enzymatically by antioxidant enzymes, such as catalase, superoxide dismutase (SOD) and glutathione peroxidase (3,4), or non-enzymatically by antioxidants, such as vitamin E, vitamin C and glutathione (5). Previous studies suggest that the antioxidant vitamin E may inhibit some aspects of the aging process by preventing lipid peroxidation. Supplemental vitamin E has been shown to have immunoenhancing effects on naive T cells, a T cell subset that exhibits the greatest age-related functional impairment in old mice (6). High dose vitamin E supplementation (5g/kg of α-tocopherol) was also associated with increased median and maximum life span and improved brain mitochondrial function in male mice from a short-lived strain (7).
Vitamin E consists of two major forms, α-tocopherol and γ-tocopherol, that differ structurally only by a methyl group substitution at the 5-position. α-Tocopherol is quantitatively the major form of vitamin E in human and animal tissues including blood plasma and is a more potent antioxidant in vitro than γ-tocopherol (8). Therefore, α-tocopherol is the primary form found in human nutritional supplements. In contrast, γ-tocopherol is the most abundant form of vitamin E in foods; however, there is no current recommended dietary allowance (RDA) for γ-tocopherol, since its bioavailability and bioactivity are lower than that of α-tocopherol (9). Recent studies suggest that γ-tocopherol may have biological activities that are not shared by α-tocopherol, such as the inhibition of cyclooxygenase activity (10), detoxification of electrophiles (11), and natriuretic activity (12). In normal aging there is a significant decline in the plasma concentration of γ-tocopherol but not of α-tocopherol, whereas platelet concentrations of both tocopherols decrease (13). The sum of serum α- and γ-tocopherol, but neither tocopherol alone, is inversely associated with the incidence of age-related nuclear cataracts (14). A study of antioxidant activity of vitamin E in vitro suggests that there is synergism between α- and γ-tocopherol (15). However, the relative contribution of these tocopherols to antioxidant activity in vivo is unknown.
Cardiac tissue is largely postmitotic and relies heavily on mitochondrial oxidative energy metabolism (16,17). Age-related changes in human and rodent hearts include a reduction in the number of myocytes (18), myocyte hypertrophy (18,19), cardiac fibrosis (20), lipofucin pigment accumulation (21), a reduction in calcium transport across sarcoplasmic reticulum membrane (22) and alterations in the response to β-adrenergic stimulation (23). Although the mechanisms leading to age-related heart diseases are not well understood, it is thought that increased oxidative stress significantly contributes to myocardial dysfunction with age (24,25). Antioxidant vitamins reduce cardiac oxidative stress and cardiomyocyte apoptosis and attenuate cardiac dysfunction in pacing-induced congestive heart failure (26). However, a recent study in humans suggested an unexpected increased risk of heart failure associated with long-term supplementation of α-tocopherol (27).
Aging of the brain is associated with subtle morphological and functional alterations in specific neuronal circuits, rather than large-scale neuronal loss (28). The impairment of functional capacity of the brain increases susceptibility to several common neurological disorders such as Alzheimer disease, Parkinson’s disease and amyotrophic lateral sclerosis. Although there may be several mechanisms involved in brain aging, oxidative damage caused by ROS may contribute to functional impairment in the aged brain (29). The brain is more susceptible to oxidative stress owing to its high oxygen consumption per unit weight and scarcity of antioxidant defense systems (30). Cognitive impairment associated with aging can be prevented by vitamin E supplementation in rats (31) and dogs (32).
Previously, we have performed microarray analyses of gene expression in the aging brain (33) and heart of mice (33,34). Heart aging was associated with transcriptional alterations consistent with a metabolic shift from fatty acid to carbohydrate metabolism, increased expression of genes encoding structural proteins involved in age-related cardiomyocyte hypertrophy and reduced protein synthesis. The global transcriptional profile of brain aging was suggestive of marked immune and inflammatory responses, oxidative stress accompanied by the accumulation of altered or misfolded proteins and reduced neuronal plasticity and neurotrophic support. In the present study, we investigated the effects of α- and γ-tocopherol on age-related transcriptional changes in the heart and brain on a genomic scale by monitoring mRNA levels using high-density oligonucleotide arrays.
Male B6C3F1 mice were purchased from Harlan Sprague-Dawley at 6–7 weeks of age. Mice were housed singly in a specific pathogen-free facility and provided acidified water ad libitum. The control group was fed 98 kcal/week of AIN-93M diet, which contains 100mg/kg of vitamin E as α-tocopherol (Table 1). The two vitamin E-supplemented groups received either a diet supplemented with 1000mg/kg of α-tocopherol or a diet containing 500mg/kg of α- and γ-tocopherol each, starting at 15 months of age. Mice were euthanized at 30 months of age by rapid cervical dislocation and autopsied to exclude animals showing overt disease. Tissues were immediately flash-frozen in liquid nitrogen and stored at −80°C until use. All aspects of animal care were approved by the appropriate university committees and conformed to institutional guidelines.
Tissues were weighed and homogenized in 1.0 ml ethanol containing 1.2% pyrogallol using a PowerGen 125 polytron (Fisher Scientific). The sample was then vortexed for 20 seconds and heated at 70°C for 2 minutes. After 100 µL of 30% KOH being added, the sample was vortexed for 20 seconds and heated at 70°C for 30 minutes for saponification. After cooling for a few minutes, 1.0 ml distilled water was added, followed by 100 µL internal standard (approximately 4ng/mL Tocol in ethanol). 2.0 ml hexane containing 0.02% BHT was added to the sample. The sample was then vortexed for 2 minutes and centrifuged at 3,000rpm for 5 minutes at 4°C. The top hexane layer was transferred to a glass vial where it was dried down under nitrogen gas. The dried sample was reconstituted in 100 µl of ethanol containing 0.02 % BHT and injected into HPLC. The concentration of tocopherols was calculated by the HPLC Millennium software by generating a standard curve by injecting known amounts of α- and γ-tocopherol standard containing internal standard Tocol.
Total RNA was extracted from frozen tissue and used to synthesize biotin-labeled cRNA. cRNA was then hybridized to the gene chips as previously described (35). Following hybridization, the hybridization solutions were removed and the gene chips were installed in a fluidics system for washes and staining. The gene chips were read using a Hewlett Packard GeneArray Scanner (Affymetrix). The averaged images collected from two scanned images were used for the analysis. We used five animals per group, and hybridized each sample to independent DNA chips, because previous work from our laboratory suggests that variability between individuals is higher than variability observed in replicate hybridizations of the same samples (36).
Detailed protocols for data analysis of Affymetrix microarrays and extensive documentation of the sensitivity and quantitative aspects of the method have been described (34). To evaluate the effects of vitamin E on aging, we conducted a statistical evaluation procedure. In the first step, we used nonparametric bootstrap hypothesis testing (37) to obtain the most frequent p-value for each gene. Part of our reason for having five animals per group is that n = 5 in a two-group comparison offers a sufficient number (>15,000) of unique bootstrap resamples. We used 10,000 resamples to compute p-values to the level of 10−4. Gene expression change was called significant when the p-value was < 0.05. In the second step, we used a mixture modeling approach (38) that estimates the posterior probabilities (pp) that a gene is a true positive for each gene. The pp is the Bayesian probability that a gene with a given the most frequent p-value or smaller is true different between the two conditions being studied. The mixture modeling approach also provides an omnibus test of whether there is any overall effect of treatment on gene expression levels, an estimate of the total number of genes that have their expression levels altered, and a probabilistic model predicting the effects of treatment on gene expression (38).
To classify the gene expression profile according to functional annotation of each gene and test the statistical significance of age-related changes in each category, we used web-based DAVID (Database for Annotation, Visualization and Integrated Discovery) (39) and EASE (the Expression Analysis Systemic Explorer) (40). DAVID allows us to access a relational database of functional annotation. EASE facilitates the biological interpretation of gene lists derived from the results of microarray and provides statistical methods for discovering enriched biological themes within gene lists. First, we uploaded the Affymetrix probe set ID of genes that were significantly changed with aging to yield functional annotations using DAVID, then EASE analysis generated gene annotation tables with Fisher Exact Score, which represents the statistical significance of specific class of genes.
In order to validate the results of oligonucleotide arrays, we perfomed real-time quantitative RT-PCR analyses. mRNA quantification was performed using ABI prism 7000 Sequence Detection System (TaqMan), which uses the 5’ nuclease activity of Taq DNA polymerase to generate a real-time quantitative DNA analysis assay. Gene-specific TaqMan probes contain a fluorescent reporter (FAM) at the 5’ end of the probe and a quencher dye (TAMRA) at the 3’ end of the probe. As the PCR cycle progresses, the degradation and release of the fluorescent reporter results in fluorescence at 518nm. The accumulation of PCR products, therefore, is detected directly by monitoring the increase in fluorescence during the complete amplification process.
We measured the level of vitamin E in several tissues of 30-month-old mice that received the experimental diets to confirm that dietary supplementation with α- and γ-tocopherol increases the tissue levels of these tocopherols (Table 2). Vitamin E concentrations were measured in heart, liver, quadriceps, cerebellum, and neocortex of 30-month-old mice. Tissue levels of α-tocopherol were significantly increased in both α-tocopherol-supplemented and the mixture of α- and γ-tocopherol-supplemented groups. We also determined if α-tocopherol supplementation affects tissue γ-tocopherol concentration. In α-tocopherol-supplemented mice, tissue levels of γ-tocopherol were only increased in liver. γ-Tocopherol concentration was significantly higher in tissues from mice supplemented with the mixture of α- and γ-tocopherol as compared to the mice fed control diet (Table 2).
We compared gene expression patterns between young (5-month-old) and aged (30-month-old) hearts using the Affymetrix Murine Genome U74Av2 arrays. Among 12,422 transcripts, 1,258 (10%) were significantly altered in expression (p-value < 0.05) with heart aging. Functional classes were assigned according to the biological function of each gene using DAVID and EASE (Table 3). Aging resulted in induction of genes involved in energy metabolism and cytoskeleton organization in the heart. As previously reported (34), as a result of heart there was transcriptional evidence of a metabolic shift from the oxidation of fatty acids, the major energy source in the adult heart, to oxidation of carbohydrate. Several genes involved in fatty acid oxidation were decreased in expression with aging in the heart. With aging in the heart, genes encoding cytosolic ribosomal proteins and translation initiation factors were down-regulated, whereas genes encoding mitochondrial ribosomal proteins were up-regulated (Table 3). Heart aging was also characterized by up-regulation of genes involved in heat shock response and apoptosis.
In the neocortex, of 45,037 transcripts surveyed using the Affymetrix Mouse Genome 430 2.0 arrays, 8,523 (20%) displayed significant changes in expression with aging. There were also age-related inductions of genes involved in energy metabolism and cellular immune and inflammatory responses, as observed in heart aging (Table 4). However, genes involved in protein synthesis showed opposite age-related expression patterns in the neocortex as compared to the heart. mRNA levels of genes encoding ribosomal proteins and translation initiation and elongation factors were increased and several genes involved in mRNA processing were decreased in expression with aging. In addition, several genes involved in neurotransmitter transport were induced, whereas genes involved in chromosome organization, and anti-apoptotic activity were reduced in expression with aging in the brain (Table 4).
To determine the effect of vitamin E on aging, we compared 30-month-old mice receiving supplementation of vitamin E from 15 months of age with 30-month-old control mice. Vitamin E prevented many age-related gene expression changes in heart either completely or partially. Among 1,258 genes that were significantly changed with aging, 1,015 genes (81%) were transcriptionally regulated in the opposite direction as normal aging by α-tocopherol supplementation (Fig. 1A) and 1,021 genes (81%) by α-and γ-tocopherol supplementation (Fig. 1). Of these, the number of genes that differed in a statistically significant manner between aged control and vitamin E supplemented group was 533 (53%) and 634 (64%) in α-tocopherol supplemented group and α-and γ-tocopherol supplemented group, respectively. The average % inhibition effect of α-tocopherol on individual age-related genes was 68.9% and that of α- and γ-tocopherol was 55.1% in the heart.
Since aging of the heart is associated with induction of genes involved in structural roles, cell adhesion and extracellular matrix, which are suggestive of hyperthrophy, we examined the influence of vitamin E supplementation on these specific transcriptional classes (Table 5). Up-regulation of many procollagens observed in normal heart aging was completely suppressed in both vitamin E-supplemented diets. Vitamin E suppressed age-related increases in mRNA levels of genes involved in cytoskeleton organization, including some actin proteins and myosin light chains. We also detected the suppression of age-related increases in gene expression by vitamin E for genes encoding junctional proteins, such as claudin 5 and gap junction membrane channel protein alpha 1. In addition to preventing age-related changes in gene expression of structural proteins, vitamin E down-regulated other genes encoding structural proteins that were not changed in expression in response to aging in the hearts of the control group (data not shown). These observations suggest profound transcriptional shifts related to structural components in response to vitamin E supplementation in the heart.
Many age-related diseases are associated with a chronic inflammatory state, including local infiltration of inflammatory cells and higher circulatory levels of pro-inflammatory cytokines, complement components and cell adhesion molecules (41). Age-related up-regulation of genes involved in the complement cascade and histocompatibility complex factors were largely prevented by both α-tocopherol and the mixture of α- and γ-tocopherol supplementation (Table 5). These findings suggest that vitamin E may lower the pro-inflammatory state observed in the aging heart.
Aging in the heart of B6C3F1 mice results in down-regulation of genes involved in fatty acid metabolism and a transcriptional shift toward carbohydrate metabolism (34). Three key enzymes involved this metabolic shift, pyruvate dehydrogenase kinase 4, uncoupling protein 3 and malonyl coA decarboxylase, are down-regulated in the hearts of 30-month old animals. Vitamin E had no, or marginal effects in preventing age-related decreases in expression of these three genes (Table 5). Age-related changes in expression of other genes involved in fatty acid metabolism were only marginally prevented by vitamin E. These findings suggest that unlike caloric restriction (34), vitamin E supplementation does not prevent age-related shifts in mitochondrial energy metabolism in the heart. There was no significant difference in the global effect on heart aging between α-tocopherol supplementation and α- and γ-tocopherol supplementation. These results suggest that supplemental γ-tocopherol does not enhance the beneficial effects that result from α-tocopherol supplementation in heart aging.
The average % inhibition effect on 8,523 transcripts significantly changed in brain aging was 13.2% and 31.6% by α-tocopherol and the mixture of α- and γ-tocopherol, respectively. Accordingly, unlike what we observed in heart tissue, the effect of supplementation on age-associated alterations in gene expression of the brain was influenced by γ-tocopherol supplementation. In the comparison between old control and vitamin E supplemented groups (Fig. 1C and D), both α-tocopherol and the mixture of α- and γ-tocopherol appeared to prevent global age-related changes (66% by α-tocopherol and 74% by α- and γ-tocopherol). However, the number of genes whose expression was statistically significantly inhibited by vitamin E (p-value < 0.05) was much higher in the α- and γ-tocopherol-supplemented group (89%) as compared to the α-tocopherol-supplemented group (29%) (Fig. 1C and D). In fact, when we averaged the % inhibition effect of individual genes, there was no significant effect of α-tocopherol alone on age-related changes in expression in the neocortex. However, the mixture of α- and γ-tocopherol showed significant attenuation of age-associated transcriptional alterations in some specific transcriptional classes (Table 6). For example, the supplementation of α-and γ-tocopherol markedly prevented the age-related up-regulation of genes involved in glycolysis, the tricarboxylic acid (TCA) cycle and ATP biosynthesis. However, it failed to suppress the age-related induction of genes involved in cellular immune and inflammatory responses. These findings suggest that in the brain, γ-tocopherol may influence some aspects of the aging process by its unique biochemical activities.
It is well documented that glucose is the primary substrate for oxidative metabolism in the adult brain. In rats, there is an age-related increase in activities of several enzymes of the glycolytic pathway, suggesting an enhancement of the glucose-dependence of brain with aging (42,43). Many genes involved in glycolysis and the TCA cycle were induced in the aged neocortex. These included hexokinase 1, phosphofructokinase, pyruvate kinase, and isocitrate dehydrogenase (Table 6). Among them, phosphofructokinase and isocitrate dehydrogenase are key enzymes that regulate the rate of glycolysis and the TCA cycle, respectively. Compared with age-matched controls, only the mixture of α-and γ-tocopherol significantly suppressed the age-related up-regulation of genes in this class. Gene expression profiling also revealed that there is an age-associated increase in expression of genes encoding the mitochondrial F0F1-ATP synthase complex and most of these transcriptional inductions were prevented by supplementation of the mixture of α-and γ-tocopherol (Table 6). Because the supplementation with α-tocopherol alone has no effect on aging in the brain, these results suggest that middle age-onset supplementation of γ-tocopherol may inhibit age-associated changes in metabolic pathways in the aged brain.
We also observed a concerted induction of the complement cascade genes and histocompatibility genes involved in innate immunity in the brain (Table 6). Expression of these genes leads to the generation of proinflammatory peptide fragments and has been observed in the striatum of aged rats (44). Neither experimental diet showed significant inhibitory effect on age-related changes in expression of these genes. These data indicate that expression of immune-related transcripts in aged mouse brain, which was attenuated by caloric restriction (33), may not be prevented by supplementation of vitamin E.
We also measured transcriptional changes that were induced by the experimental diets, but not by aging. Vitamin E supplementation up-regulated anti-apoptotic genes and down-regulated genes involved in the activation of apoptosis in the aged heart (Table 7). Bcl2-associated athanogene 3, an anti-apoptotic gene that was not changed in expression in normal aging, was up-regulated by both α-tocopherol-supplemented and α- and γ-tocopherol-supplemented groups. Genes, which are involved in induction of apoptosis and show no age-related changes in mRNA levels, such as Bcl2-like 11 and 13, were down-regulated by vitamin E supplementation. The gene encoding Bcl2-associated X protein (Bax), which activates caspases via cytochrome C (45), was up-regulated in normal aging (p-value = 0.009, pp = 0.81). However, its expression level remained at the level of young controls in the aged hearts of the vitamin E-supplemented groups. Caspase 3 and caspase 7, intracellular proteases that are the key mediators of apoptosis, showed reduced expression (0.4-fold in both genes) in the α-tocopherol-supplemented groups. Our results at the transcriptional level are consistent with the hypothesis that cytochrome C release from mitochondria and alterations in Bcl2 and caspase activation with age could be underlying mechanisms for the age-related increase in cardiomyocyte apoptosis (46) and that vitamin E supplementation results in a transcriptional pattern consistent with protection of cardiomyocytes from apoptosis.
Several genes encoding proteins of known antioxidant function were down-regulated in aging hearts, including catalase, metallothionein 1, metallothionein 2, heme oxygenase 1 and glutathione S-transferase 1. In both vitamin E-supplemented groups, the age-related down-regulation of heme oxygenase 1 and glutathione S-transferase 1 was partially prevented (data not shown). Other genes involved in the antioxidant response also increased in expression levels in vitamin E-supplemented groups as compared to the control group. Both vitamin E-supplemented groups showed induction of SOD1, glutathione S-transferase 2 and glutathione peroxidases, the expression levels of which were not changed in normal aging (Table 7). These observations suggest that vitamin E induces the expression of antioxidant response genes, and prevents the age-related decrease in expression in some genes encoding antioxidant proteins. Vitamin E supplementation also suppressed the expression of genes involved in DNA repair (Table 7). Nine genes, which were not changed in expression in normal aging, were significantly reduced by α-tocopherol and three genes were affected by the mixture of α- and γ-tocopherol. Previous studies indicate that vitamin E can decrease the formation of oxidative DNA damage by lipid peroxidation (47,48). The reduced expression of genes involved in DNA repair by vitamin E may result from the inhibitory effect of vitamin E on oxidative DNA damage.
Vitamin E supplementation was associated with a reduced expression of pro-apoptotic factors and modest induction of anti-apoptotic factors in the brain (Table 8). These included genes encoding proteins having anti-apoptotic activity, such as Bcl2, BCL2/adenovirus E1B 19kDa-interacting protein 1 and BCL2-associated athanogene 4, and genes involved in the initiation of apoptosis, including Bcl10, Bcl2 binding component 3, capase 3 and programmed cell death proteins. Previous studies indicate that oxidative stress may activate mitochondrial-mediated apoptosis in brain, and that neuronal loss due to apoptosis may play a role in normal brain aging (28,49,50). Our data suggest that vitamin E supplementation may decrease oxidative stress in the aged brain as a result of its antioxidant activity and thus could prevent age-related induction of apoptosis. A state of lowered endogenous oxidative stress in vitamin E supplemented mice is also suggested by reduced expression of several heat shock proteins that were not changed in expression in aged control mice (Table 8).
The synthesis of phospholipids and steroids is decreased in the aged brain (51,52) and excessive oxidative stress associated with aging appears to contribute to the age-related decline in steroid production (53). Steroids exert neuroprotective effect in the nervous system, preserving neural function and promoting neural survival (52). Compared with age-matched controls, vitamin E increased the expression of eight genes that are involved in phospholipid and steroid biosynthesis (Table 8). For example, 3-hydroxy-3-methylglutaryl-Coenzyme A (HMG-CoA) reductase, which converts HMG-CoA to mevalonate, catalyzes the rate-limiting step in cholesterol biosynthesis. The expression of HMG-CoA reductase was significantly increased in α-tocopherol-supplemented mice (1.3-fold change). CDP-diacylglycerol synthase 2 is a key regulator of the amount of phosphatidylinositol 4,5-bisphosphate, which plays a central role in phosphoinosite-mediated signaling pathway (54). Both forms of vitamin E supplementation induced upregulation of this gene in the brain (1.4- to 1.8-fold).
To perform an independent validation of the oligonucleotide array data, we used a quantitative RT-PCR method for a subset of genes the expression of which changes significantly with age, and for a few other genes of interest. α-Globin and TATA-box binding protein (Tbp), the expression of which did not change with age, were used as a controls for data normalization. We observed similar FC with quantitative RT-PCR and DNA microarray analysis (Fig. 2). In the heart, the age-related increase in expression of actin, alpha 1 (Acta1), mitochondrial ribosomal protein L34 (Mrpl34), and procollagen type XVIII, alpha 1 (Col18a1), which were significantly prevented by vitamin E as determined by the microarray analysis (Table 4), were also markedly prevented as determined by quantitative RT-PCR (Fig. 2A). Phenylalanine hydroxylase (55) demonstrated the largest fold change in normal heart aging (7.7 fold), and the alteration was partially prevented (46%) by α-tocopherol as determined by both DNA microarray and quantitative RT-PCR data. Cardiotrophin 1 (Ctf1), a molecular marker of cardiac hypertrophy (56), was not changed in expression levels with normal aging, but showed decreased expression in both vitamin E-supplemented groups as determined by both DNA microarray analysis (p-value < 0.05) and quantitative RT-PCR (Fig. 2A).
Complement component 4 (C4), which was induced in the aged brain according to micorarray data, also showed an age-related increase in expression as determined by quantitative RT-PCR (Fig. 2B). P lysozyme structural (Lzp-s) and glial fibrillary acidic protein (Gfap), known to have neuroprotective activity in brain (57,58), were up-regulated with normal brain aging as determined by both DNA microarray analysis and quantitative RT-PCR. The age-related expression of Lzp-s was significantly reduced by α-tocopherol (51% inhibition). Apolipoprotein D (ApoD), a glycoprotein induced after stress or injury of the nervous tissues (59), showed increased level of mRNA in normal brain aging (Fig. 2B). Our micorarray data indicated that vitamin E supplementation had no effect on the age-related up-regulation of ApoD transcription. Interestingly, the expression of ApoD was increased significantly as compared to old control by both forms of vitamin E supplementation. In the brain, age-related up-regulation of genes involved in cellular energy metabolism was significantly prevented by the mixture of α- and γ-tocopherol, but not by α-tocopherol alone (Table 6). We observed age-related increases in expression of ATPase, H+ transporting, V0 subunit B (Atp6vob), a gene involved in ATP biosynthesis, by quantitative RT-PCR, similar to results obtained through microarray analysis (Fig. 2B). Only the mixture of α- and γ-tocopherol supplementation showed significant inhibitory effect of the age-related transcriptional change of this gene (101% inhibition as determined by microarray and 119% inhibition in quantitative RT-PCR).
Increasing evidence suggests that p53-induced effects play a significant role in organismal aging and longevity (60,61). We investigated the effect of aging and vitamin E supplementation on the expression of p53 in heart and brain using the real-time quantitative RT-PCR method. There was an age-related up-regulation of p53 in the heart (p-value < 0.05) and this age-related induction was completely prevented by vitamin E (Fig. 2A). The % inhibition effect of α-tocopherol and the α- and γ-tocopherol was 95% and 100%, respectively. However, the expression of p53 was not changed with aging or vitamin E in the aged neocortex (Fig. 2B). We also monitored the expression level of p16, one of the downstream components of p53 and a marker of senescence (62). We detected an age-related increase in expression of p16 in both heart and brain, and this induction was partially inhibited by vitamin E supplementation (Fig. 2).
ROS are generated in mitochondria throughout the life span, and evidence exists to support the causal involvement of ROS in aging (4,63,64). In this study, we determined the effects of vitamin E, the major lipid antioxidant, on age-related transcriptional alterations in the heart and brain, and also attempted to obtain evidence for unique biological roles of α- and γ-tocopherol. As previously reported (34), heart aging resulted in induction of genes encoding structural proteins and down-regulation of genes involved in fatty acid metabolism and protein synthesis. An age-related transcriptional shift toward carbohydrate metabolism was also observed. In addition, we observed age-related induction of genes involved in immune and inflammatory responses, which is suggestive of a heightened pro-inflammatory state in aged heart. We also observed induction of genes involved in heat shock response, suggesting increased protein damage with aging. Induction of heat shock protein with aging was previously observed in Caenorhabditis elegans and Drosophila (65,66). Heart aging was associated with a concerted decline of mRNA encoding ribosomal proteins, such as ribosomal protein L10, L10a, L19, L21, S3, S3a, S5 and S28, and translation factors, including several eukaryote translation initiation, elongation and termination factors. Decreased protein biosynthesis with aging in the heart was reported previously in rats (67), suggesting that this may be a general feature of heart aging in laboratory rodents. Previously, we found no evidence that aging in the heart is associated with a transcriptional profile indicative of an oxidative stress response (34). In this study, however, several genes involved in the antioxidant response were down-regulated with age, including catalase (0.5-fold), metallothionein 1 (0.6-fold), and metallothionein 2 (0.4-fold). Because these genes are inducible, these observations suggest either a decrease in ROS production in the aged heart, or that the enzymatic defense system for ROS-induced oxidative stress may be compromised.
Vitamin E markedly prevented the age-related up-regulation of genes encoding cellular structural proteins in the heart. Interestingly, we found that vitamin E down-regulates cardiotrophin-1, a cytokine that induces cardiomyocyte hypertrophy (56) and is induced in congestive heart failure (68). Based on these observations, it appears that vitamin E has multiple effects that suppress an age-related cardiomyocyte hypertrophy transcriptional program. Vitamin E also markedly suppressed the age-related increase in expression of genes encoding several complement components (complement component 1, q subcompoment, alpha polypeptide, complement component 1, q subcomponent, receptor 1, and complement component 4) (Table 5). Complement-related genes are involved in innate immunity and induced in response to ischemia and reperfusion (69). Our results suggest that vitamin E lowers the expression of immune-related genes, which in turn may confer a cardioprotective effect in the aged heart. In adult hearts, fatty acids are the major energy source, whereas fetal hearts primarily use glucose as metabolic fuel (70). We show here that the age-related down-regulation of genes involved in fatty acid β-oxidation is not opposed by vitamin E supplementation, a finding in contrast to the our previous observations with caloric restriction in the mouse heart.
Brain aging resulted in a gene expression profile indicative of increased energy metabolism and induced immune and inflammatory responses. We also observed age-related decreases in expression of genes involved in mRNA processing (Table 4). There was an age-related up-regulation of genes encoding ribosomal proteins and translation elongation factors in the aged brain. In the aged mouse neocortex, genes involved in base excision repair and DNA-damage responses were significantly decreased in expression, including DNA cross-link repair 1A, PSO2 homolog (0.66-fold), excision repair cross-complementing rodent repair deficiency, complementation group 5 (0.48-fold), and RAD50 homolog (0.54-fold). These results support the hypothesis that a decline in DNA repair activity in the aged brain may be an important mechanism contributing to the age-dependent accumulation of oxidative DNA lesions in brain mitochondria. Interestingly, DNA damage caused by oxidative stress is markedly increased in the promoters of genes that display reduced expression in the aged human cortex, and the base-excision DNA repair is reduced (Lu, 2004 #320). There is tissue-specific regulation of telomerase during aging (71) and telomere shortening, a marker of cellular senescence, occurs in the rat brain with increasing age (72). We observed an age-related down-regulation of several genes involved in telomere maintenance in the neocortex. Among them, Poly (ADP-ribose) polymerase family member 1, an enzyme that play a role in base excision repair, has been implicated in the maintenance of telomeres (73).
Previous studies demonstrated that several proteins involved in glycolysis and the TCA cycle, and subunits of mitochondrial ATP synthase increased progressively with aging in rat brain (42,74). At the mRNA level, we observed age-related induction of genes involved in glycolysis, the TCA cycle and ATP biosynthesis (Table 4). Surprisingly, the mixture of α- and γ-tocopherol, but not α-tocopherol alone, prevented most age-related gene expression changes in these functional classes either completely or partially (Table 6). As reported previously in brains from mice (33) and humans (Lu, 2004 #320), age-related increases in expression of genes involved in immune and inflammatory responses occur, which is suggestive of heightened immunity and pro-inflammatory status in aged tissues. Interestingly, transcriptional changes in this class were significantly inhibited by the combination of α- and γ-tocopherol in the heart, but not in the brain (Table 5 and Table 6).
It has been observed that γ-tocopherol has higher activity in trapping reactive nitrogen species (RNS) than α-tocopherol (75). According to the nitric oxide hypothesis of brain aging (76,77), accumulation of oxidative damage caused by RNS may lead to degeneration of brain cells. Compared with age-matched control, only the mixture of α- and γ-tocopherol showed significant aging-retarding effect on functional classes of genes in brain (Table 6). These observations suggest that γ-tocopherol can prevent the age-related changes in expression in brain, independent of α-tocopherol, and may provide evidence to support the nitric oxide hypothesis of brain aging. It was also reported recently that the protective effect of γ-tocopherol against neurodegenerative toxicity is stronger than that of α-tocopherol (78).
In order to find possible mechanisms involved in the effect of vitamin E on the aging process, we analyzed genes that showed significant alterations in the vitamin E-supplemented mice, but were not affected by aging. Several genes involved in the inhibition of apoptosis were up-regulated by vitamin E supplementation, whereas genes involved in the activation of apoptosis were down-regulated as compared with the control group (Table 7 and Table 8). Phaneuf et al. (46) first demonstrated that cytochrome C release from mitochondria and alterations in the level of Bcl2, an anti-apoptotic protein, were correlated with cardiomyocyte apoptosis in the aging heart. Experimental evidence suggests that neuronal loss due to oxidative stress induced apoptosis may play a central role during normal brain aging (28,49). Qin et al. (26) observed that vitamin E feeding attenuates myocyte apoptosis in rabbits. These results support the role of ROS-induced apoptosis in aging tissues and suggest prevention of apoptosis in normal aging by vitamin E supplementation. We report here that vitamin E supplementation induced the up-regulation of several genes involved in the antioxidant response, including Cu/Zn-SOD and glutathione peroxidases, in the aging heart (Table 7). Thus, vitamin E may suppress the damaging effects of increased ROS in the aging heart not only through its own antioxidant activity, but also through its ability to increase the expression of genes encoding antioxidant genes. However, supplementation of a relatively high dose of α-tocopherol acted as a prooxidant showing no attenuation of oxidative stress in rat skeletal muscle (79). Therefore, it is possible that the induction of antioxidant genes by vitamin E may be due to its potential to become a pro-oxidant in the oxidative milieu of the heart. Vitamin E supplementation was also associated with a reduction in the expression of genes involved in DNA repair in the heart(Table 7), a finding consistent with reduced endogenous DNA damage. Our finding of a cardioprotective role of vitamin E contrasts with the recent finding of an apparent association between vitamin E intake and heart failure in diabetics (27).
Steroid hormones act on neurons and glial cells, regulating differentiation, survival and neuronal connectivity. A recent gene expression profiling study showed that the underlying molecular mechanisms of neuroprotection by vitamin E are associated with hormone metabolism and apoptosis (80). In our study, several genes involved in steroid biosynthesis were found to be induced by vitamin E supplementation (Table 8), suggesting that neuroprotective activity of vitamin E in brain may be partially contributed by the increased biosynthesis of steroids.
Recently biological functions of α-tocopherol that are independent of its antioxidant activity have been reported, including inhibition of cell proliferation, platelet aggregation and monocyte adhesion (81,82). At transcriptional level, a number of genes have been characterized to be regulated by the non-antioxidant activity of α-tocopherol. A non-antioxidant activity of α-tocopherol induces α-tropomyosin expression in muscle (83) and prevents age-related increase of collagenase expression in human skin fibroblasts (84). A non-antioxidant function of γ-tocopherol was also observed recently (85). Therefore, some effects of vitamin E supplementation observed in this study might be accounted for by non-antioxidant activities of vitamin E.
In summary, we report here that vitamin E can alter the transcriptional profiles associated with aging, and that this effect of vitamin E may be mediated through several mechanisms. First, vitamin E reduces the expression of structural proteins in the aging heart and thus could oppose age-related cardiomyocyte hypertrophy. Increased cardiomyocyte apoptosis associated with aging may be prevented by vitamin E, since it up-regulates genes encoding anti-apoptotic proteins and down-regulates genes involved in the induction of apoptosis in the aging heart (Table 7). Enhanced apoptosis in aged cardiomyocytes may be further attenuated by vitamin E through the suppression of the p53-induced apoptotic response (Fig. 2). In brain, both forms of vitamin E increased the expression of anti-apoptotic proteins and decreased the mRNA levels of pro-apoptotic proteins (Table 8). Surprisingly, only the mixture of α- and γ-tocopherol prevents age-related transcriptional alterations in the brain (Fig. 1D and Table 6). Genes involved in steroid biosynthesis were also changed in expression by vitamin E supplementation. Importantly, we observed a beneficial effect of γ-tocopherol on the transcriptional profile of brain aging, which was not observed in the aging heart. Therefore, the effect of vitamin E in aging is likely to be tissue and tocopherol-specific. Taken as a whole, the data from the present study demonstrate that the middle age-onset vitamin E supplementation can prevent age-related transcriptional alterations in post-mitotic tissues. Extension of this study to other post-mitotic or regenerating tissues should result in the identification of common mechanisms of action, facilitating the understanding the role of oxidative stress and the effect antioxidant supplementation in the aging process. Our general findings that vitamin E can modulate the aging process in two critical organs, if applicable to humans, may have major public health implications.