Recent Findings in Alzheimer Disease and Nutrition Focusing on Epigenetics^,

Abstract

Alzheimer disease (AD) is a chronic neurodegenerative disease with no effective cure so far. The current review focuses on the epigenetic mechanisms of AD and how nutrition can influence the course of this disease through regulation of gene expression, according to the latest scientific findings. The search strategy was the use of scientific databases such as PubMed and Scopus in order to find relative research or review articles published in the years 2012–2015. By showing the latest data of various nutritional compounds, this study aims to stimulate the scientific community to recognize the value of nutrition in this subject. Epigenetics is becoming a very attractive subject for researchers because it can shed light on unknown aspects of complex diseases like AD. DNA methylation, histone modifications, and microRNAs are the principal epigenetic mechanisms involved in AD pathophysiology. Nutrition is an environmental factor that is related to AD through epigenetic pathways. Vitamin B-12, for instance, can alter the one-carbon metabolism and thus interfere in the DNA methylation process. The research results might seem ambiguous about the clinical role of nutrition, but there is strengthening evidence that proper nutrition can not only change epigenetic biomarker levels but also prevent the development of late-onset AD and attenuate cognition deficit. Nutrition might grow to become a preventive and even therapeutic alternative against AD, especially if combined with other antidementia interventions, brain exercise, physical training, etc. Epigenetic biomarkers can be a very helpful tool to help researchers find the exact nutrients needed to create specific remedies, and perhaps the same biomarkers can be used even in patient screening in the future.

Introduction

Dementia is an overall term for diseases and conditions characterized by a decline in memory or other thinking skills, which finally affect a person’s ability to perform everyday activities. Dementia is caused by damage to neurons, which can no longer function normally and may die.

Alzheimer disease (AD)⁷ is the most common type of dementia. It accounts for an estimated 60–80% of dementia cases (1). According to the World Alzheimer Report 2015 (2), an estimated 46.8 million people worldwide were living with dementia in 2015, and this number will double every 20 y. The estimated incidence of dementia in 2015 was 9.9 million new cases every year, which means one new case every 3.2 s (2).

At the diagnostic stage, the AD brain is characterized by abundant senile amyloid plaques—more than naturally expected—formed by extracellular aggregates of amyloid β (Aβ) peptides, small peptides, 39–43 amino acids in length, and by neurofibrillary tangles (NFTs) consisting of intracellular aggregates of abnormally phosphorylated tau protein, which disintegrate the neuron’s transport system (3). Aβ plaques are believed to interfere with the neuron-to-neuron communication at synapses, whereas NFTs block the transport of nutrients and other essential molecules in the neurons (1). Aβ is generated from β-amyloid precursor protein (APP) through sequential cleavage by β-secretase and the γ-secretase complex. Alternatively, APP can be cleaved by α-secretase within the Aβ domain to release soluble APPα, which precludes Aβ generation (4).

AD is associated with initial memory loss, followed by impairment in other cognitive functions, such as language, visuospatial skills, and executive function, coupled with behavioral changes. Terminally, patients may become bedridden, incontinent, and unable to communicate (5).

The neuropathology and the etiology of AD, according to recent studies (6), indicate that complex molecular mechanisms are involved in this neurodegenerative disease. Epigenetics is a scientific field that may manage to shed some light on this complexity. DNA methylation, histone modifications, and microRNAs are the principal epigenetic mechanisms involved in AD pathophysiology. Nutrition is an environmental factor, which seems to be strongly related to AD through epigenetic pathways, and this is the topic this review discusses.

Search Methods

This study consists of a narrative review that includes the basic principles of a systematic review. The data were collected independently by 2 reviewers, and every effort was made to eliminate bias as much as possible. The reason our study cannot be characterized as systematic is that there was no quality evaluation of the literature used. This was a decision that we made because of the small amount of literature relative to our subject.

The literature search was performed on reviews addressing the topics below in the databases PubMed, Google Scholar, and Scopus. Some literature was searched through Google. The search terms (concept) included the following topics: 1) dementia or AD, 2) epigenetics, and 3) nutrition or nutrients or food. The search language was English. There was no specific time limit, but the purpose was to include the most recent literature. The most recent citation in the current review is dated in December 2015. The search results are described in Figure 1.

FIGURE 1.

Flow diagram of search results.

Current Status of Knowledge

Epigenetics

Epigenetics is the study of cellular and physiologic trait variations that are not caused by changes in the DNA sequence; in layman’s terms, epigenetics is essentially the study of external or environmental factors that turn genes on and off and affect how cells read genes. Hence, epigenetic research seeks to describe dynamic alterations in the transcriptional potential of a cell. Epigenetic marks drive much of the expression and provide diversity to the phenotype via chromatin alteration, which affects gene transcription. An epigenome is the chromatin state found across the genome at a certain time point and cell type; therefore, thousands of epigenomes can exist for a single given genome (6) (Figure 2).

FIGURE 2.

Alzheimer disease pathology is influenced by genetics and epigenetics, along with environmental influences on these complex networks. Environmental influences include but are not limited to oxidative stress, inflammation, and chemical exposure. In addition, lifestyle factors such as diet and nutrition, traumatic brain injury, and smoking may have an effect. This can impact AD pathology through malfunction in regulation of genes, noncoding RNAs, and changes to an individual’s epigenome (6). AD, Alzheimer disease; APOE, apolipoprotein E; APP, amyloid precursor protein; DNMT, DNA methyltransferase; HAT, histone acetyltransferase; HDAC, histone deacetylase; miRNA, microRNA; NCRNA, noncoding RNA; PSEN, presenilin.

DNA methylation.

One of the best-characterized chromatin modifications is DNA methylation. The human genome is predominantly methylated on cytocine-phosphate-guanine motifs, which are enriched in promoters, where their methylation results in gene silencing (7). The mechanism of DNA methylation is carried out by a set of proteins named DNA methyltransferases (DNMTs) (8).

A study on humans investigated the association between DNMT1 and DNMT3B polymorphisms and AD. For both genes, the polymorphisms were in strong linkage disequilibrium. Carriers of the DNMT3B thymine-guanine-guanine haplotype were associated with AD. No significant differences between the AD and control groups were observed for DNMT1 polymorphisms (9).

Furthermore, quantitative studies showed that there is a decrease in DNA methylation in AD patients. A quantitative immune-histochemical study assessed the 2 most important markers of DNA methylation and hydroxyl-methylation—5-methylcytidine (5-mC) and 5-hydroxymethylcytidine (5-hmC)—in the hippocampus of AD patients and compared these with nondemented, age-matched controls. The amount of 5-hmC in the hippocampus of a pair of monozygotic twins discordant for AD was also assessed. The amounts of 5-mC and 5-hmC were furthermore analyzed in a cell type– and hippocampal subregion–specific manner and were correlated with Aβ plaque load and NFT load. The results showed robust decreases in the hippocampal concentrations of 5-mC and 5-hmC in AD patients. Similar results were obtained for the twin with AD compared with the nondemented twin. Moreover, the concentrations of 5-mC and 5-hmC showed a significant negative correlation with amyloid plaque load in the hippocampus (10). In addition to this, Condliffe et al. (11) used immune fluorescence to confirm the presence of cytosine modifications in the human brain and investigate their cross-tissue abundance in AD patients and elderly control samples. The result showed a significant AD-associated decrease in global 5-hmC in entorhinal cortex and cerebellum and differences in 5-fluorocytosine concentrations between brain regions (11).

Histone modifications.

Eukaryotic cells organize their genomic material into chromatin, which consists of both DNA and DNA-associated proteins such as histones. The basic structural elements of chromatin are nucleosomes, which comprise a histone core, around which DNA is wrapped. The histone core is an octamer containing 2 units from histone, family 2A (H2A), H2B, H3, and H4, whereas the H1 linker is involved with packing of the bead-like nucleosomes into a higher order structure. Each histone protein consists of a central globular domain and an N-terminal tail that contains multiple sites for potential modifications. These modifications are divided into 2 types: rapid sliding movements that occur locally by use of ATP hydrolysis to expose specific DNA regions, such as ATP-dependent chromatin remodeling complexes, and posttranslational modifications of amino acids on histone tails (12). Among others, these modifications include acetylation and methylation. Each of these modifications is bidirectionally catalyzed or removed by a specific set of enzymes (13).

Histone acetylation is a modification associated with transcriptional activity by promoting access of the transcription machinery to the genes and thus activating genes transcription (8). An important attribute of histone acetylation is that it is reversible. Histone acetyltransferases (HATs) catalyze the transfer of an acetyl group from acetyl coenzyme A to the ε-amino group of lysine residues, whereas histone deacetylases (HDACs) remove the acetyl group. In both mice and humans, 18 proteins with HDAC activity have been identified and are divided into 4 classes. Class I, II, and IV HDACs share a related catalytic mechanism that requires a zinc metal ion but does not involve the use of a cofactor. In contrast, class III HDACs [sirtuin (SIRT) family] are nicotinamide adenine dinucleotide-dependent and can be nuclear (SIRT1, 2, 6, and 7), mitochondrial (SIRT3, 4, and 5), or cytoplasmic (SIRT1 and 2) (12).

The importance of HDAC inhibition in AD can be shown via a study that involves 2 novel HDAC inhibitors with improved pharmacologic properties, such as a longer half-life and greater penetration of the blood-brain barrier: mercaptoacetamide-based class II histone deacetylase, class I (HDACI; coded as W2) and hydroxamide-based class I and II HDACI (coded as I2), and that investigates how they affect Aβ concentration and cognition. It was demonstrated that HDACI W2 decreased gene expression of γ-secretase components and increased the Aβ degradation enzyme matrix metallopeptidase 2. Similarly, HDACI I2 decreased expression of β- and γ-secretase components and increased mRNA levels of Aβ degradation enzymes. HDACI W2 also significantly decreased Aβ concentration and rescued learning and memory deficits in aged human APP 3× transgenic AD mice. Furthermore, they found that the novel HDACI W2 decreased tau phosphorylation at threonine 181 (Thr181), an effect previously unknown for HDACIs (14).

Histone methylation is associated with multiple processes, such as transcriptional activation and repression, depending on the modified amino acid residue. This modification occurs mainly in arginine and lysine residues. In addition, these residues can be methylated multiple times, giving different signals depending on how many times the residues are methylated, making their analysis difficult. In this regard, the current literature has shown that lysine residues can be methylated 3 times; meanwhile, arginine residues can only be methylated 2 times (8). The transcriptional effect of histone methylation is dictated in a site-specific manner and by the number of methylation marks at a specific residue (10).

Noncoding RNAs.

Other important epigenetic players are the noncoding RNAs (ncRNAs). Recent genome-wide studies (the Encyclopedia of DNA Elements project) have shown that <2% of the human genome codes for proteins, but the genome is pervasively transcribed and produces many thousands of regulatory ncRNAs, including small ncRNAs (∼20 nucleotides in length), such as microRNAs, small interfering RNAs, piwi-interacting RNAs, and various classes of long ncRNAs (7).

microRNAs are the best-characterized members of the ncRNAs. microRNAs bind to the 3′-untranslated regions of target mRNAs, which will be silenced or degraded (7). microRNAs repress translation by forming RNA-induced silencing complex and induce the degradation of mRNA by binding imperfectly to the 3′-untranslated region. Some microRNAs positively regulate gene expression by enhancing mRNA translation and inducing gene expression via binding to the promoter of the target gene (12). In recent studies in short postmortem interval AD brain tissues compared with age-matched controls and in proinflammatory cytokine- and Aβ42 peptide-stressed human neuronal-glial cells in primary culture, several brain-abundant microRNA species were found to be significantly up-regulated, including microRNA-125b and microRNA-146a. Both these NF-κB-activated cells and 22 nucleotide small noncoding RNAs target the mRNA of the key innate-immune– and inflammation-related regulatory protein, complement factor H (chr 1q32), resulting in significant decreases in complement factor H expression. It is indicated that human neuronal-glial cells respond to IL-1β + Aβ42-peptide–induced stress by significant NF-κB–modulated up-regulation of microRNA-125b and microRNA-146a (15). More examples of microRNAs related to AD are shown in Table 1.

TABLE 1.

microRNA-targeting genes in AD¹

Action	microRNAS
Downregulate APP	106a, 520c, 106b, 17-5p
Influence BACE1	29a, 29b-1, 107, 298, 328
Potent negative regulator of innate immunity	146
Direct inhibition of human tau	34a
Enhance tau phosphorylation through downregulation of retinoblastoma	Increased microRNA-26b

¹In sporadic AD, clinical and research evidence indicates that aberrant regulation of microRNA-dependent gene expression is closely associated with molecular events responsible for Aβ peptide production, neurofibrillary tangle formation, and neurodegeneration. The current table shows some microRNA examples according to the action they perform (16). AD, Alzheimer disease; APP, amyloid precursor protein; BACE1, β-site APP cleaving enzyme 1.

Risk factors

Lead.

Lead is known to be one of the environmental factors promoting the development of cognition deficits in the brain. Various studies in humans and animals have shown that low-dose lead exposure is associated with cognitive deficits and that the highest concentrations of lead in the brain have been found in the hippocampus and cerebral cortex, areas responsible for learning and memory functions (17). A study in mice found that early lead exposure causes latent but persistent epigenetic changes leading to DNA hypermethylation and thus down-regulation of gene expression (18). Animal and cell studies show that lead can specifically reduce Dnmt1 and Dnmt3A expression and thereby cause hypomethylation of AD-related genes such as App and β-site APP cleaving enzyme 1 (Bace1), resulting in overexpression and abnormal amyloid processing (19–21). There is also research, both in animals and in humans, that correlates higher long interspersed nuclear element (LINE-1) methylation with better cognitive performance and also lead exposure with hypomethylation of LINE-1 (21, 22). A study in adolescent and old mice focused on the effects on global expression and DNA methylation by developmental lead exposure (23). The study showed that lead exposure causes a global downshift of total gene expression through DNA methylation. In normal aging, there are some up-regulated genes perhaps as a compensatory response to stressors acting on the aging brain. Prior developmental lead exposure leads to an overwhelming repression of these genes. As a result, early-life lead exposure may interfere with the methylation pattern of genes, which is then sustained throughout life and has an impact on an animal’s ability to respond in old age (23). Trying to put the above observations together, we can assume that lead exposure leads to global DNA hypermethylation, which in turn down-regulates specific genes such as Dnmt1 and Dnmt3A. These 2 genes might be responsible for the conflicting hypomethylation of some other genes such as App and Bace1.

Cadmium.

Cadmium is another toxic heavy metal associated with neurologic alterations including memory loss and mental retardation, whereas cadmium exposure is linked with Aβ overproduction. These effects are attributed to a reduced expression of genes, such as a-disintegrin and metalloproteinase domain-containing protein 10 (ADAM10) (24), which is an enzyme important to α-secretase activity. Cadmium seems to influence DNMT activity. According to cell studies, short exposure to cadmium is related to DNA hypomethylation, whereas chronic exposure is related to DNA hypermethylation (20).

Arsenic.

Arsenic is another risk factor for AD development and is known to impair the neurodevelopment and cognitive function (17, 24). Arsenic can lead to DNA hypomethylation via depletion of the methyl donor pool of the cell, because it is extensively methylated by the arsenic methyltransferase that utilizes S-adenosylmethionine (20). Apart from DNA methylation, it has also been found that arsenic influences histone modifications. According to an animal study, arsenic can cause reduction of global acetylation on lysine 9 of histone 3 (H3K9ac) in the cortex and hippocampus, as well as demethylation of arginine residues on histones. In addition, treatment with a histone deacetylase inhibitor, sodium butyrate, can attenuate the cognitive deficits observed in arsenic-exposed animals (25).

Aluminum.

Aluminum is another potent neurotoxin involved in neurodegeneration and late-onset AD (26). In addition, chronic aluminum exposure induces oxidative stress, which might be the mechanism responsible for the role of aluminum in AD (27). There is evidence that aluminum influences NF-κB cells, certain microRNAs, oxidation mechanisms, and gene expression in AD models. Interestingly, treatment with aluminum and iron together seems to have stronger genotoxic effects than either aluminum or iron alone (28).

Ethanol (alcohol).

Excessive acute and chronic alcohol consumption is related to lower cognitive function and brain damage. Human alcoholics show significant volume loss in cortical and subcortical structures, whereas alcohol abstinence seems to reverse these effects. There are cumulative data suggesting that ethanol-induced neuronal death and hippocampal neurogenesis impairment is related to oxidative stress coinciding with the induction of proinflammatory cytokines and transcriptional changes (29). In relation to epigenetics, alcohol consumption influences S-adenosylmethionine metabolism, leading to degradation of DNMTs. In chronic alcohol consumption, this may result in either significant DNA hypomethylation or histone modification (demethylation). In addition, chronic consumption of ethanol can lead to histone demethylation (20).

Potential risk factors (both deficiency and excessive intake can be harmful)

Zinc.

Zinc is related to increased amyloid plaques deposition and the development of late-onset AD (17, 30). Zinc deficiency, however, is also proposed as risk factor for AD, as shown by an animal study, but according to a systematic review, there are not enough data to give answers regarding the complexity of zinc metabolism in the human brain (31). Concerning epigenetics, dietary deficiency of zinc is capable of reducing the utilization of methyl groups enzymatically, leading to one-carbon metabolism modifications and thus DNA methylation changes. In addition, the role of zinc in histone modifications is crucial, as all histone deacetylases (HDACs), except for HDAC class III members, require zinc for their activity (32).

Copper.

In vitro and animal models have demonstrated that copper is a causative factor in Aβ toxicity, whereas numerous studies on living humans have shown the association of copper with AD. However, copper deficiency can also be harmful (33). A research study on pheochromocytoma cell line 12 (PC12) cells treated with copper confirmed the cell damage caused by the oxidative stress. The same study found that the mRNA expression of some epigenetic factors (DNMT1, DNMT3A, SIRT1, and SIRT2) tended to be upregulated when the cells were exposed to copper and that this upregulation was even greater when H₂O₂ was added (34).

Risk-reduction factors

B vitamins.

B vitamins are substances well associated with the methionine cycle (one-carbon cycle metabolism), which includes the molecules homocysteine, methionine, and S-adenosylmethionine and utilizes cofactors, such as vitamin B-12, B-9 (folate), and B-6. The methionine cycle and the reaction chain, beginning from folate and ending with DNA methylation, are described in Figures 3 and 4. The hypothesis is that nutritional deficits in B vitamins can lead to hyperhomocysteinemia and, consequently, to decreased S-adenosylmethionine concentration. The decrease in S-adenosylmethionine, which is a methyl donor, can induce demethylation of DNA, resulting in overexpression of genes involved in AD pathology (19).

FIGURE 3.

DNA methylation. The reaction pathway by which the methyl group is transferred from tetrahydrofolate to DNA is shown. The figure also indicates the involvement of folate and vitamin B-12 in this process. Hcy, homocysteine; MTHF, methyltetrahydrofolate; SAM, S-adenosylmethionine; THF, tetrahydrofolate.

FIGURE 4.

Methionine cycle (one-carbon metabolism). SAH, S-adenosylhomocysteine; SAM, S-adenosylmethionine.

Vitamin B-12 deficiency might be related to neurologic problems, poor cognition, and AD (35). In addition, supplementation of vitamins B-12, B-6, and folate appears to reduce homocysteine concentrations and to slow cognitive decline in people with mild cognitive impairment, in particular those with elevated homocysteine, as shown by a randomized controlled trial in humans (36). A study found direct associations between AD and low vitamin B-12 concentrations in humans with normal homocysteine and folate concentrations, as well as between AD and low vitamin B-12 and folate concentrations in humans with normal homocysteine concentrations (37). Interestingly, vitamin B-12 deficiency can induce presenilin (PSEN1) and β-site APP cleaving enzyme (β-secretase) (BACE) up-regulation via hypomethylation of their gene promoters leading to Aβ deposition—both PSEN1 and BACE are AD-related genes. PSEN1 encodes presenilin 1, which is one of the γ-secretase complex proteins. BACE encodes β-secretase. More specifically, the activity of some DNMTs (DNMT1, 3a, and 3b) appears to be influenced by vitamin B-12 deficiency (19).

There are several population-based studies indicating that low folate concentrations and hyperhomocysteinemia might be associated with dementia (38). A recent study on human neuroblastoma cells correlated folate deficiency with NFT formation via the deregulation of protein phosphatase 2A (39). A study that used subjects with type 2 diabetes indicated a relation between mild cognitive impairment, lower S-adenosylmethionine, and folate concentrations, in addition to a correlation between serum folate concentrations, plasma S-adenosylmethionine concentrations, and methylation potential (40). Interesting data also emerged from a study investigating the association between mandatory folate fortification and leukocyte global DNA methylation, as well as between global DNA methylation, RBC folate, and other biomarkers of the one-carbon cycle metabolism in postmenopausal women. These data include, among others, the finding that folate and global DNA methylation are associated through a reverse U-shaped curve rather than a linear relation, because excess folate intake leads to accumulation of unmetabolized folate and consequently to abnormal folate metabolism and lower global DNA methylation (41).

Vitamins A, C, and E.

Vitamins A, C, and E are well known for their antioxidant properties, which are important for epigenetics, because oxidative lesions interfere with the ability of DNA to serve as an optimal substrate for the DNMTs, thus leading to global hypomethylation (42). Oxidation may also cause alterations in the histone acetylation mechanisms. The balance of histone and protein acetylation can be tipped toward increased acetylation in an oxidative environment caused by increased protein-tyrosine nitration and other redox modifications to HDACs, which lead to decreased deacetylase activity (43). So, vitamins A, C and E may reduce such abnormal histone acetylation modifications through their antioxidant properties.

Vitamin A is a group of nutritional compounds including retinol, retinoic acid, and β-carotene (provitamin A). β-carotene might have beneficial effects on cognitive functions such as memory (35). Concerning epigenetics, dietary deficiencies in retinoic acid may decrease DNA methylation by altering the availability of methyl groups (32). A recent study investigated the effects of postweaning high multivitamin and high vitamin A diets on gene expression in offspring from dams fed a high multivitamin diet. The results showed increased expression of retinoic acid receptor α in the hippocampus in both high-vitamin A and high-multivitamin groups. However, the effect of high-vitamin A diet was greater than in the high-multivitamin group maybe because of the competition between vitamin A and vitamin D in binding to retinoic X response elements via its receptors (44).

Vitamin E is a group of compounds including tocopherols and tocotrienols. Increased intake of vitamin E from foods might be associated with decreased risk for developing AD. Interestingly enough, among vitamins E and C and β-carotene, vitamin E has shown the most significant protective effect (35). In addition, there are data showing that the combination of dietary tocopherols, rather than the individual tocopherols, has the strongest protective effect on AD. According to research on humans, however, not all kinds of tocopherols are beneficial for the brain. The non-α-tocopherol forms of vitamin E exhibit the most neuroprotective results; however, high brain concentrations of α-tocopherol may be associated with increased AD neuropathology (45).

Maintaining healthy vitamin C plasma concentrations might be protective against AD (35). Vitamin C is another essential nutrient that acts as a reducing agent and possibly induces epigenetic changes via its antioxidant function. Apart from that, vitamin C is required for recycling of the α-tocopherol radical, and thus the beneficial effects of vitamin E are often at least partially dependent on vitamin C concentration (46). It should be noticed, however, that food-derived vitamin C is considered preferable to synthetic vitamin C because of the concomitant consumption of numerous other nutrients, which may synergize with vitamin C and confer additional health benefits (47).

Vitamin D.

Important studies in living humans indicate that vitamin D deficiency might be related to AD pathology (48, 49) and moreover that vitamin D uptake could decrease the risk for developing AD (35). There are many mechanisms suggested that could explain the neuroprotective effects of vitamin D. Neurotrophic regulation, calcium homeostasis, Aβ metabolism, antioxidative, and anti-inflammatory action are some very well examined mechanisms (50). The epigenetic effects of vitamin D have not been studied specifically for AD yet. However, there are important data occurring from studying other diseases, such as cancer. Intriguingly, not only does vitamin D affect epigenetics, but also epigenetics influence vitamin D, because critical vitamin D tool genes can be silenced by DNA methylation [i.e., the activating enzymes cytochrome P450, family 27, subfamily A, polypeptide 1 (CYP27A1) and cytochrome P450, family 27, subfamily B, polypeptide 1 (CYP27B1); and the inactivating enzyme cytochrome P450, family 24 (CYP24)]. Vitamin D exerts its activity via the vitamin D receptor protein (VDR), which can also be modulated epigenetically by histone acetylation. VDR usually forms heterodimers with retinoid X receptors. The VDR-retinoid X receptor dimer can influence gene transcription by interacting with HATs leading, probably, even in modulations in DNA methylation. Thus, VDR ligands are considered to even regulate DNA methylation and modulate gene expression (51, 52). Finally, it is very important that these effects be confirmed in AD models as well.

Selenium.

There are data showing that selenium might have significant effects in the brain, and its essential role is played through various mechanisms, mostly through its antioxidant activity (53, 54). Concerning epigenetics, selenium supplementation can modify DNA methylation globally and at specific gene regions, possibly through DNMT inhibition (55). In addition, it has been found that dietary deficiency in selenium may decrease DNA methylation by enhancing the trans-sulfonation pathways (32). Selenium can also alter histone modifications via inhibition of HDAC activity by the selenium metabolism products seleno-a-keto acids (55).

Omega-3 fatty acids.

There is evidence supporting the role of PUFAs, which are fatty acids containing more than one double bond in their molecule and include the subcategory of ω-3 fatty acids, mostly DHA, in slowing cognitive decline in older individuals when treated before clinical manifestation of dementia (35). A study on peripheral blood mononuclear cells showed that the ability of producing specialized proresolving mediators by peripheral blood mononuclear cells decreases with time, as the AD progresses. Supplementation with ω-3 fatty acids for 6 mo can prevent this reduction in specialized proresolving mediators, indicating that a defined ω-3 fatty acid supplement can hinder age- and AD-related deterioration in proresolving signaling (56). One study on neuroblastoma cells showed that DHA can increase global H3K9ac and decrease concentrations of HDAC1, HDAC2, and HDAC3. In addition, DHA is involved in histone demethylation processes. It is shown that it decreases global concentrations of dimethyl-histone, family 3, lysine (H3K) 4, -H3K9, -H3K27, -H3K36, and -H3K79. Overall, treatment with DHA tended to induce histone changes consistent with active transcription and modifications to gene expression that suggest reduced apoptosis (57).

Resveratrol.

The possible beneficial effects of resveratrol in relation to AD are indicated by the facilitation of Aβ clearance via the AMP-phosphokinase signal pathway, in addition to the lysosome and autophagy pathways (58). The newest data associated with epigenetics of AD indicate that low doses of resveratrol reduce the expression of genes crucial for age-related diseases (59). Actually there is accumulative evidence that resveratrol activates SIRT1 (48, 58–60), which leads to increased expression of ADAM10 (58) and to decreased neuronal loss caused by chronic inflammation (60). Resveratrol also mimics the effects of calorie restriction (59), maybe through SIRT3 inhibition (48). Finally, polyphenols in general appear to counteract the modulation of microRNAs induced by knock-out of the APOE gene (61).

Oleuropein.

A recent study showed a substantial contribution of oleuropein against AD pathology by combating Aβ neurotoxicity and Aβ-induced cognitive impairment (62). One study investigated whether oleuropein affected histone acetylation in the brains of wild-type and transgenic mice. The results showed an increase of H3K9ac and of acetylation on lysine 4 of histone 5 in oleuropein-fed transgenic mice compared with untreated littermates in the cortex and in the hippocampus. In addition, there was an increase of HDAC2 in untreated transgenic mice compared with the untreated wild-type mice, whereas oleuropein significantly counteracted that increase in the treated transgenic mice (63).

Curcumin.

Curcumin is an extremely strong antioxidant compared with other curcuminoids (48) having, thus, an effect on epigenetics and AD. There are many suggested pathways by which curcumin is beneficial for AD, but there is need for further research (64). Curcumin reduces the Aβ production by inhibiting glycogen synthase kinase 3-β-mediated PSEN1 activation (59). In addition, curcumin inhibits HDAC isoforms 1, 3, and 8 and is thus considered a pan-HDAC inhibitor that can participate in reprogramming neural stem cell–directed neurogenesis. Curcumin also inhibits HATs. Finally, there are data showing that curcumin induces microRNA-22, microRNA-186a, and microRNA-199a (58).

Catechins.

Catechins and epigallocatechin gallate have shown various effects on AD and epigenetics. A dietary intervention study provided evidence that regular catechin consumption can reduce some measures of age-related cognitive dysfunction, possibly through an improvement in insulin sensitivity, and these data suggest that the habitual intake of flavanols can support healthy cognitive function with age (65). There are data supporting that the green tea catechins are more effective antioxidants than vitamins C and E (48). In addition, treatment of cells with epigallocatechin gallate results in dose-dependent and time-dependent inhibition of class I HDACs (HDAC1, HDAC2, HDAC3, and HDAC8) (58).

Ginseng.

According to accumulating data, ginseng might be beneficial for cognition and AD (65–69). Interestingly enough, ginsenosides from Panax ginseng and American ginseng exhibit differential efficacies in interacting with the PPARγ and HDACs (58); PPARs are nuclear receptors that act as transcriptional factors in various metabolic pathways. PPARγ is associated with the insulin signaling pathway and, when activated, attenuates insulin resistance, mitochondrial dysfunctions, and cognitive decline according to animal studies (70).

Isoflavones.

Several studies, both animal and human, on isoflavones have indicated positive effects on AD and cognitive function through various mechanisms, such as their estrogenic activity or even the downregulation of PSEN1 (71–76). A study in rats showed that isoflavones can modulate DNA methylation of the promoter region of selected genes (77).

Anthocyanins.

A recent study in cells and rats showed the beneficial effects of anthocyanins on AD by reversing the neurotoxicity of Aβ (78). A positive indication of the epigenetic potential of anthocyanins comes from a study in worms, which found that although the lifespan of Caenorhabditis elegans extends when treated with apple procyanidins, treatment with procyanidins has no effect on the longevity of worms lacking the activity of sirtuin 2 (Sirt2) (48). In addition, a cell study about the effects of mulberry extract, which contains 8 different anthocyanins (mostly cyanidin), showed that it could substantially alleviate the cell injury induced by Aβ-25–35. In the same study, mulberry extract indicated a delay in AD progress through the negative regulation of apoptotic protease activating factor 1, BACE2, and 1-phosphatidylinositol-4,5-bisphosphate phosphodiesterase β-4 (79).

Other nutrients and bioactive compounds.

Iron has many effects associated with neurotoxicity similar to Al (28), but its epigenetic impact has not been studied yet. Iron accumulation has been demonstrated in cells associated with neuritic plaques in late-onset AD, and it seems to regulate α-secretase activity to influence APP cleavage (17).

A recent study about mangosteen pericarp, which is rich in xanthones (α-mangostin and γ-mangostin), showed that mangosteen pericarp treatment in mice had various effects, including neuroprotection and attenuation of cognitive impairment via brain-derived neurotrophic factor increase in hippocampal slices, increased anti-inflammation, and decreased tau protein concentration. Decreases of BACE1 were also observed (80).

Pterostilbene is a stilbenoid that, according to recent studies, is a more potent modulator of cognition and cellular stress than resveratrol, likely driven by increased expression of PPARα and increased lipophilicity of pterostilbene (48). Quercetin is a both pro-oxidant and antioxidant flavonol (81) and is considered to have great potential both in AD and epigenetics, because quercetin-3-O-glucuronide promotes neuroplasticity mechanisms in the brain (82), but more research is needed.

Recent studies suggest that adhering to caloric restriction or to the Mediterranean diet can positively influence epigenetic stability in adults and reduce incidence of pathologic phenotypes (83), including neurodegeneration and AD (32, 48, 58, 59, 84–87). However, there is reasonable evidence that other diets, such as the traditional diets in Africa, China, Japan, and India, may have even greater effects in lowering the risk of AD (88).

Some other studies have provided evidence about foods such as pomegranate (89, 90) and wheat grass (27). There are also data showing the effects of a high-fat diet (57, 91), insulin resistance (92), diabetes mellitus (93–95), obesity (95, 96), pesticides (20), souvenaid (97), and other combinatorial formulations (98). Finally, it is interesting that the combination of specific nutrients, such as vitamin B-12, vitamin D, and zinc, could have a better impact on AD biomarkers than when supplemented separately, as shown by a study in living humans (99). It is obvious, thereafter, that nutrition should be much further studied in order for its real value to be explored.

Discussion: Gaps in Knowledge

All the aforementioned data are only a sample of the possible effects of nutrition, and a concise summary of the most important nutrients is given in Table 2. Unfortunately, there are many gaps in current knowledge, and thus, we will try to hint at some points as future research directions. First, there is a lack of human trials, which are important because the human organism exhibits higher complexity in answering to external factors; therefore, it is not easy to translate the findings from animal models into human interventions. Second, there is a need for more epigenetics-related research, in order to achieve a better understanding of the epigenetic mechanisms. It is crucial to identify the exact epigenetic mechanisms involved in each different disease. Only then will it be possible to measure the appropriate epigenetic biomarkers on intervention studies for specific diseases. Third, it should be determined whether there are time windows in the lives of animals and humans when the subjects are more or less susceptible to changes upon certain interventions. Last, a global effort should be made for nutrition research to be more standardized. Otherwise, the results of various research studies cannot be easily compared with each other, because there might be many differences, such as the quality of the food administered to the subjects.

TABLE 2.

Nutrient summary¹

Nutrient or compound	Food sources	Main epigenetics mechanisms associated with AD	Favorable or unfavorable effect
Vitamin B-12	Fish, liver, meat, poultry, eggs, milk, milk products	DNA methylation	F
Folate	Vegetables, especially dark green leafy vegetables, fruits, nuts, beans, peas, dairy products, poultry, meat, eggs, seafood, grains	DNA methylation	F
Vitamin A	Liver, fish oils, milk, eggs, leafy green vegetables, orange and yellow vegetables, tomatoes, fruits, vegetable oils	DNA methylation, histone modifications	F
Vitamin C	Citrus fruits, tomatoes, potatoes, red and green peppers, broccoli, strawberries, Brussels sprouts, cantaloupe, fortified cereals, kiwifruit	DNA methylation, histone modifications	F
Vitamin E	Nuts, seeds, vegetable oils, green leafy vegetables, fortified cereals	DNA methylation, histone modifications	F
Selenium	Seafood, meat, nuts, cereals, dairy products	DNA methylation, histone modifications	F
Lead	Contaminated fruits and vegetables	DNA methylation	UF
Zinc	Oysters, red meat, poultry, beans, nuts, whole grains, dairy products	DNA methylation, histone modifications	UF (in deficiency and excess), F (in low doses)
Cadmium	Contaminated meat and plants (such as rice)	DNA methylation	UF
Arsenic	Contaminated meat, seafood, and plants (such as rice)	DNA methylation, histone modifications	UF
Ethanol	Alcoholic beverages	DNA methylation, histone modifications	UF
Isoflavones	Soybeans	DNA methylation	F
ω-3 fatty acids	Fish, organ meats	Histone modifications	F
Resveratrol	Red wine, peanuts, Itadori tea, grapes, soy, blueberries, raspberries	Histone modifications, noncoding RNA	F
Oleuropein	Extra virgin olive oil	Histone modifications	F
Curcumin	Turmeric	Histone modifications, noncoding RNA	F
Catechins	Peach, green tea, vinegar, cocoa	Histone modifications	F
Ginsenosides	Ginseng	Histone modifications	F
Anthocyanins	Elderberry juice, billberries, blackberries, mulberries, black grapes, aronia, Morello cherries, hazelnuts	Histone modifications	F
Aluminum	Processed foods with flour, baking powder, coloring, anticaking agents	Noncoding RNA	UF
Vitamin D	Fatty fish flesh such as salmon and tuna, fish liver oils, beef liver, cheese, egg yolks, fortified foods	DNA methylation, histone modifications	F
Copper	Oysters, shellfish, nuts, whole grains, beans, potatoes, organ meats, dark green leafy vegetables, dried fruits such as prunes, cocoa, black pepper, yeast	DNA methylation, histone modifications	UF (in deficiency and excess), F (in low doses)

¹The data in the 3rd and 4th columns were collected from the text. The data in the 2nd column derive from references 100–107. AD, Alzheimer disease; F, favorable; UF, unfavorable.

Conclusion

Alzheimer disease is a chronic neurodegenerative disorder, and its complexity makes its cure an unsolved problem for the time being. A novel pathophysiologic aspect, that of epigenetics, promises the explanation of many medical mysteries including etiology and pathophysiology of AD. Epigenetic biomarkers can be a very helpful tool for the researchers to find the exact nutrients needed to create specific remedies, and maybe the same biomarkers can even be used in patient screening in the future. In this way, nutrition might grow to become a useful preventive and even a therapeutic alternative against AD, especially if combined with other antidementia interventions, such as acetylcholinesterase inhibitors, brain exercise, physical training, etc. We hope that the present review will stimulate the scientific community to focus on nutrition and epigenetics more intensively and thus find a way toward prevention of and therapy for AD.