Nutrition, Brain Aging, and Neurodegeneration

Abstract

The onset of age-related neurodegenerative diseases superimposed on a declining nervous system could enhance the motor and cognitive behavioral deficits that normally occur in senescence. It is likely that, in cases of severe deficits in memory or motor function, hospitalization and/or custodial care would be a likely outcome. This means that unless some way is found to reduce these age-related decrements in neuronal function, health care costs will continue to rise exponentially. Applying molecular biological approaches to slow aging in the human condition may be years away. So, it is important to determine what methods can be used today to increase healthy aging, forestall the onset of these diseases, and create conditions favorable to obtaining a “longevity dividend” in both financial and human terms. Recent studies suggest that consumption of diets rich in antioxidants and anti-inflammatory components such as those found in fruits, nuts, vegetables, and spices, or even reduced caloric intake, may lower age-related cognitive declines and the risk of developing neurodegenerative disease.

Introduction

By the year 2050, 30% of the total population will be over 65 years of age. As the aged population expands, the economic burden of care and treatment of those with age-related health disorders also increases, necessitating measures to prevent or even reverse age-related health disorders. One such potential option is the use of nutritional substances such as berry fruit and fatty acids from walnuts and fish oils. Research has recently shown that consumption of the aforementioned substances can dramatically impact the aging brain, possibly leading to improved cognition and motor abilities. It has been postulated that these behavioral and neuronal declines are the result of an increasing vulnerability to oxidative and inflammatory insults, thus creating a “fertile environment” (Yu, 1994; Sadoul, 1998; Gilissen et al., 1999; Joseph et al., 2001) for the subsequent development of age-related neurodegenerative disease such as Alzheimer disease (AD) (Markesbery, 1997; Behl, 1999; Praticò and Delanty, 2000; Sultana et al., 2006). The destructive properties of oxidative stress in the aged brain are evidenced by reductions in redox active iron (Gilissen et al., 1999; Joseph et al., 2001), as well as increases in B-cell lymphoma 2 (Bcl-2) (Sadoul, 1998) and membrane lipid peroxidation (Yu, 1994). Studies have also shown that there are significant increases in cellular hydrogen peroxide (Cavazzoni et al., 1999). Additionally, there is significant lipofuscin accumulation (Gilissen et al., 1999) along with alterations in membrane lipids (Denisova et al., 1998), indicating the involvement of lipid rafts with oxidative stress sensitivity (Shen et al., 2004). Importantly, the consequences of these increases in oxidative stress at several levels may result in reduced calcium homeostasis, alterations in cellular signaling cascades, and changes in gene expression (Dalton et al., 1999; Annunziato et al., 2002), which combine to contribute to the increased vulnerability to oxidative stress seen in the aging population (Halliwell, 2001; Rego and Oliveira, 2003) and which is elevated in AD (Smith et al., 1991; Lovell et al., 1995; Marcus et al., 1998) and Parkinson disease (PD) (Dexter et al., 1994; Spencer et al., 1998). Together, these findings indicate that oxidative stress increases during aging, leading to widespread damage to cellular components, and ultimately manifesting in declines in motor and cognitive abilities.

It is clear that the incidence of many of the major disorders, such as Alzheimer's disease, vascular dementia, and cardiovascular disease, increase as a function of age, and that their etiology may partially involve lifestyle determinants such as obesity, decreased sensitivity to insulin, and the metabolic syndrome. Unfortunately, there is a distinct lack of knowledge of these issues among clinicians and physicians, where nutritional recommendations could be used with traditional approaches in neurogerontology. Research described below suggests that polyphenolic compounds contained in berry fruits, walnuts, curcumin, and fish oils exhibit potent antioxidant and anti-inflammatory activities that may reduce the age-related sensitivity to oxidative stress or inflammation and may alter neurodegeneration. Interestingly, the results found with respect to these nutrients are similar to those seen with caloric restriction and caloric restriction mimetics, suggesting a final common pathway among these various interventions.

Berry fruit polyphenols

All plants, including fruit-, nut-, or spice-bearing plants, synthesize a vast array of chemical compounds that are not necessarily involved in the plant's metabolism but instead serve a variety of functions that enhance the plant's survivability. These include combating oxidative stress and inflammation. In this respect, previous studies have found that crude blueberry (BB) or strawberry (SB) extracts significantly attenuated age-related motor and cognitive deficits in aged rodents. Thus, these supplementations reversed age-related deficits in neuronal and behavioral (cognitive) function in aged (19 months) Fischer 344 rats (Joseph et al., 1999). The rodents in all diet groups, but not the control group, showed improved working memory (short-term memory) performance in the Morris water maze (MWM). A later study has suggested that, in addition to MWM performance, blueberry supplementation was also effective in reversing cognitive declines in object recognition (Goyarzu et al., 2004). Furthermore, the beneficial effects of BB were seen even when superimposed on an already healthy rodent diet, which was more representative of a balanced human diet (Youdim et al., 2000). BB supplementation also improved performance on tests of motor function that assessed balance and coordination (e.g., rod walking and the accelerating rotarod) (Joseph et al., 1999). Thus far, only blueberry, cranberry (Shukitt-Hale et al., 2005), strawberry (Shukitt-Hale et al., 2006b), Concord grape juice (Shukitt-Hale et al., 2006a), blackberry (Shukitt-Hale et al., 2009), or walnut supplementations (Willis et al., 2009) have been effective in reversing motor behavioral deficits. Rats on the BB diet have generally shown the greatest increases in motor performance, as well as increases in carbachol-stimulated GTPase activity and oxotremorine-enhanced dopamine (DA) release (both markers of muscarinic receptor sensitivity).

Importantly, even though these diets were supplemented based on equal antioxidant activity [as determined by the oxygen radical absorbance capacity (ORAC) assay], they were not equally effective in preventing or reversing age-related changes (Joseph et al., 1999). Therefore, antioxidant activity alone was not predictive in assessing the potency of these compounds against certain disorders affected by aging. In fact, oxidative stress markers (as measured by DCF fluorescence and glutathione levels in the brain) were only modestly reduced by the diets (Joseph et al., 1999), suggesting that berry fruit polyphenols may possess a multiplicity of actions aside from antioxidant activity. Other possible mechanisms for the berry fruit's positive effects include: direct effects on signaling to enhance neuronal communication (Joseph et al., 2003), the ability to buffer against excess calcium (Joseph et al., 2004), enhancement of neuroprotective stress shock proteins (Galli et al., 2006), and reduction of stress signals such as nuclear factor κ B (NF-κB) (Goyarzu et al., 2004). Additionally, the anthocyanins contained in blueberries have been shown to enter the brain, and their concentrations were correlated with cognitive performance (Andres-Lacueva et al., 2005).

Research has demonstrated the involvement of various signaling molecules in the protective effects of berries in both cell culture systems and animal models. BB treatment of dopamine (DA)-exposed COS-7 cells (Joseph et al., 2006) or primary hippocampal neurons (Joseph et al., 2007a) significantly increased protective mitogen activated protein kinase (MAPK) expression. Additionally, BB-supplemented APP/PS1 mice exhibited greater levels of hippocampal extracellular signal regulated kinase (ERK), as well as striatal and hippocampal protein kinase C (PKC) α, than seen in the transgenic mice maintained on the control diet (Joseph et al., 2003). ERK has been shown to be involved in diverse forms of memory, such as: contextual fear conditioning [33]; long-term potentiation (English and Sweatt, 1997); striatum-dependent learning and memory (Mazzucchelli and Brambilla, 2000); hippocampus-dependent spatial memory (Selcher et al., 1999); and inhibitory avoidance (Schafe et al., 1999). PKC has been implicated in the regulation of synaptic plasticity and modulation of short- to long-term memory. Studies have shown that PKC activity is important in spatial memory formation (for review, see Micheau and Riedel, 1999). Research has also shown that BB treatment was effective in protecting against amyloid β (Aβ₄₂)- or DA-induced decrements in intracellular calcium clearance (Joseph et al., 2007b) following depolarization in M1 muscarinic receptor (MAChR)-transfected COS-7 cells or neonatal hippocampal neurons. This protection involved increases in phosphorylated MAPK and decreases in PKCγ and phosphorylated cAMP response element binding protein (CREB). Blueberry supplementation has also been shown to impact cell signaling molecule expression in vivo. BB supplementation reduced the toxicity of kainic acid on hippocampal cells concomitant with reduced expression of stress signals and increased expression of protective signals (Shukitt-Hale et al., 2008a). These alterations in stress signaling were associated with enhanced behavioral performance (Morris water maze) and reduced microglial activation (Bodles and Barger, 2004). An additional study showed that BB-supplemented aged animals had increased ERK and insulin growth factor-1 (IGF-1) activation in the dentate gyrus that was associated with increased neurogenesis and enhanced cognitive ability (Casadesus et al., 2004). Similar findings were observed with respect to microglial activity, BB and stress signaling wherein BB treatment dose-dependently inhibited the production NO, as well as the cytokines Il-1β and TNF-α in lipopolysaccharide- (LPS-) activated BV2 microglia (Lau et al., 2007). Overall, as shown in Table 1, we have proposed the following schemes for the BB effects on stress and protective signaling.

Table 1.

Proposed signal transduction pathways possibly affected by berry fruit polyphenols

DA/Aβ42 → ROS → PKCγ/PKCα → p38 MAPK → CREB/ NfκB → increased ROS and calcium dysregulation 2 DA/Aβ42 → ROS → JNK → p53 → increased ROS and calcium dysregulation)

ROS, Reactive oxygen species.

Subsequent studies in progress have suggested that scheme two may be the pathway most likely to be involved in the DA-induced stress signaling and that BBs appear to be able to act at multiple points in the pathway to prevent calcium dysregulation.

Walnut and fish polyunsaturated fatty acids

In addition to plant derived polyphenols, polyunsaturated fatty acids (PUFAs) represent another potential dietary intervention to forestall age-related neuronal and cognitive decline. PUFAs are critical components of neuronal cell membranes, maintaining membrane fluidity that is essential for synaptic vesicle fusion and neurotransmitter communication within neural networks. In addition, membrane PUFAs serve as precursors for lipid messengers, which can participate in signaling processes to promote neuronal protection or induce neuronal dysfunction (Bazan, 2005). In the aged brain, studies have shown a deficit in the amount of PUFAs in the hippocampus, cortex, and cerebellum, all areas involved in cognitive and motor function (Little et al., 2007). These deficits may be further increased in AD.

Walnuts are well known for their high levels of PUFAs, specifically the ω-6 fatty acid linoleic acid (LA) and the ω-3 fatty acid α-linolenic acid (ALA). LA and ALA can either exist as membrane components or can be metabolized via the arachidonic acid cascade to generate numerous lipid messengers including prostaglandins, eicosapentaenoic acid (EPA), and docosahaenoic acid (DHA), which are ω 3 oils. Indeed, a recent study has shown that walnuts improved cognitive function in aged rodents, much as was seen with respect to berry fruit (Willis et al., 2009). Importantly, the aged animals on walnuts also showed reduced microglial activity in the hippocampus. In addition to PUFAs, walnuts also contain other bioactive constituents which have been shown to influence brain function, including vitamin E, melatonin, and antioxidant polyphenols such as ellagic acid (Venkatachalam and Sathe, 2006) that could act synergistically with the PUFAs to increase dietary polyphenolic absorption and uptake following consumption (Huo et al., 2007).

PUFAs from fish oils may have similar beneficial effects. A 2005 ω-3 meta-analysis assessing the quality of available epidemiology and preclinical studies concluded clinical trials were warranted (Maclean et al., 2005). To date, nine epidemiological studies associate increased fish consumption with reduced AD, while 8/10 studies associate higher blood n-3 with reduced cognitive decline. Moreover, three studies indicate limited protection in ApoE4 carriers (Cole et al., 2009). Four small completed trials with n-3 (typically fish oil) suggest protection, but only in patients showing mild cognitive impairment (MCI). Two trials with n-3 and other nutrients (α lipoate or B vitamins and uridine-5′-monophosphate, a putative enhancer of DHA incorporation) seem to show some effect in AD (Cole et al., 2009). A larger recently completed 6 month trial (MIDAS, 485 subjects with mild memory complaints) reported improvements in mild memory complaints with 900 mg of DHA (Yurko-Mauro et al., 2009). A National Institutes of Health cooperative DHA trial in mild to moderate AD (Quinn, 2009) reported possible but nonsignificant slowing of progression in ApoE3, but not ApoE4 subjects. Additional trials are in progress.

Similarly, DHA reduced amyloid β 42 (Aβ42) in AD mice (Lim et al., 2005; Oksman et al., 2006; Green et al., 2007; Hooijmans et al., 2007) and production by cultured human neurons (Lukiw et al., 2005). The mechanism involved in the DHA-induced reductions in Aβ42 may be due to multiple effects, such as: changes in lipid raft structure (Stillwell et al., 2005), alterations in APP processing (Ehehalt et al., 2003), induction of anti-amyloidogenic chaperones for APP (Ma et al., 2007), and Aβ transthyretin (Schwarzman et al., 1994; Puskás et al., 2003). As with the polyphenols, proposed neuroprotective mechanisms also include increasing survival signaling. For example, insulin/neurotrophin signaling, defective in AD, protects against Aβ oligomer toxicity (Cole and Frautschy, 2007). AD and preclinical models show synaptic and dendritic loss from a postsynaptic attack by Aβ oligomers, which DHA blocks in vitro, consistent with protection of the synaptic protein, drebrin, in APP Tg mice.

Insulin receptor substrate (IRS), an adaptor protein, couples insulin/trophic factor signaling to PI3-K/Akt. Thus, DHA reduces Aβ42 production and protects against its toxicity via multiple mechanisms. This is important since DHA is enriched in neuronal phospholipids where it may be in as high as 35% of phosphatidylethanolamine (Salem et al., 2001). Because archidonic acid (AA) and DHA compete for esterification into the labile SN-2 phospholipid position, DHA reduces proinflammatory AA available for cyclooxygenase and lipoxygenase enzymes, an anti-inflammatory NSAID-like property contributing to interest in DHA and AD prevention (Cole et al., 2009). In summary, it appears that DHA reduces Aβ₄₂ production and protects against its toxicity via multiple mechanisms.

Curcuminoids

In addition to the polyphenols found in walnuts and berry fruits, polyphenols (curcuminoids) found in the curry spice turmeric may have similar effects. The biophenolic curcumin was isolated as the active yellow component of turmeric, a food preservative inhibitor of lipid peroxidation with potent anti-inflammatory and anti-cancer activities and a long history of use in Asian traditional medicines (Aggarwal et al., 2007). In AD models, curcumin reduced proinflammatory cytokines, oxidative damage, Aβ42 and cognitive deficits (Frautschy et al., 2001). Like Congo red, it is an amyloid binding dye and direct inhibitor of Aβ oligomer and fibril formation that can enter the brain to directly label plaques and markedly reduce Aβ42 and plaques even in old APP Tg mice, suggesting a possible “vaccine-like” clearance (Cole et al., 2003). Direct evidence for curcumin stimulation of amyloid plaque clearance and dystrophic neurite reduction was provided by elegant in vivo imaging before and after curcumin (Garcia-Alloza et al., 2007). Curcumin has other pleiotropic anti-AD activities including limiting the tau kinase JNK (c-Jun N-terminal protein kinase) and stimulating neurogenesis and BDNF (Cole et al., 2007). Cucumin synergized with fish oil in reducing insulin signaling defects in triple Tg AD model mice (Ma et al., 2009). These pleiotropic anti-AD activities led to studies assessing curcumin/DHA combinations in aging tau transgenic mice and positive effects were seen against cognitive deficits. A major obstacle with curcumin in the clinic has been limited bioavailability of supplements, but this problem has been solved with new lipidated formulations (Begum et al., 2008), currently in clinical trials. While there are many new treatment approaches, the major advantages of these nutritional interventions is their safety, broad spectrum utility, low cost, and suitability for prevention, especially in diets that contain polyphenols with more “traditional” antioxidants such as vitamin C and vitamin E. A subset of these studies is described in the next section.

The canine antioxidant diet

In this respect it has been shown (Zandi et al., 2004) that the dietary intake of antioxidants in foods is superior to supplements in human studies on cognition and risk of developing AD (Morris et al., 2002; Barberger-Gateau et al., 2007). Furthermore, the addition of mitochondrial cofactors that target mitochondrial function and reduce reactive oxygen species may enhance the effects of cellular antioxidants such as vitamin E. These considerations led to studies involving administering an antioxidant diet to aged beagles. Dogs are particularly useful because they naturally develop cognitive decline with age, accumulate oxidative damage and AD-like neuropathology, and absorb dietary nutrients in a similar manner as humans (Cotman and Head, 2008).

Aged beagles (between ∼8–12 years) were used in this study. An antioxidant-enriched diet was formulated to include a broad spectrum of antioxidants and two mitochondrial cofactors (Milgram et al., 2005), which were well within those used in human clinical trials. The daily doses for each compound were 800 IU or 210 mg/d (21 mg/kg/d) of vitamin E, 16 mg/d (1.6 mg/kg/d) of vitamin C, 52 mg/d (5.2 mg/kg/d) of carnitine, and 26 mg/d (2.6 mg/kg/d) of lipoic acid. Fruits and vegetables were also incorporated at a 1 to 1 exchange ratio for corn, resulting in 1% inclusions of each of the following: spinach flakes, tomato pomace, grape pomace, carrot granules, and citrus pulp. This was equivalent to raising fruits and vegetable servings from 3 to 5–6/d. Vitamin E was increased ∼75% by the antioxidant diet in treated dogs (Milgram et al., 2002). A second intervention included a behavioral enrichment condition consisting of: (1) additional cognitive experience (20–30 min/d, 5 d/week), (2) an enriched sensory environment (housing with a kennel-mate, rotation of play toys in kennel once/week), and (3) physical exercise (2 × 20 min walks/week outdoors) (Milgram et al., 2005).

The dogs were evaluated over a 2.8 year period. Treatment with the antioxidant diet lead to cognitive improvements in learning that were rapid and within 2 weeks of beginning the diet; aged animals showed significant improvements in spatial attention (landmark task) (Milgram et al., 2002). Subsequent testing of animals with a more difficult complex learning task, oddity discrimination, also revealed benefits of the diet (Cotman et al., 2002). Improved visual discrimination and reversal (frontal function) learning ability was maintained over time with the antioxidant treatment while untreated animals showed a progressive decline (Milgram et al., 2005). Interestingly, the antioxidant diet benefitted from the inclusion of behavioral enrichment and aged dogs receiving both treatments were superior to either treatment alone (Milgram et al., 2004, 2005). As predicted, oxidative damage was reduced in antioxidant-fed dogs and in particular within the group of animals receiving the combination of antioxidants and behavioral enrichment (Opii et al., 2008). Endogenous antioxidant activity was also increased (Opii et al., 2008). Interestingly, behavioral enrichment but not the antioxidant diet protected against neuron loss in the hilus of the hippocampus of treated dogs (Siwak-Tapp et al., 2008). These results suggest that cognitive benefits of cellular antioxidants and mitochondrial cofactors can be further enhanced with the addition of behavioral enrichment due to different yet synergistic mechanisms of action in the brain (reducing oxidative damage, maintaining neuron health). These findings suggest that studies in humans may be more efficacious if combinations of antioxidants are administered and dietary intake of antioxidants considered in treatment protocols.

Caloric restriction

Furthermore, however, as compelling as the data concerning the various antioxidant/anti-oxidant diets may be with respect to reducing behavioral deficits in aging, a large literature exists to support the view that caloric restriction rather than caloric selection may provide an additional approach for reducing age-related behavioral deficits. Epidemiological studies have reported the inverse relationship between caloric intake and risk of AD and PD (Luchsinger et al., 2002; Mattson et al., 2002; Mattson, 2003). These findings fit well within the context of the calorie restriction (CR) paradigm, one of the most robust in gerontology (Weindruch and Sohal, 1997; Weindruch and Walford, 1998; Masoro, 2005; Piper and Bartke, 2008). As demonstrated in numerous animal models, CR has proven to be the most effective means to retard aging, including brain aging. Reducing intake of a nutritious diet by 20–50% can increase lifespan, reduce incidence and retard onset of chronic diseases, enhance stress protection, and maintain youthful behavioral function accommodated by preserved features of neural anatomy and activity (Weindruch and Sohal, 1997; Weindruch and Walford, 1998; Masoro, 2005; Piper and Bartke, 2008). Recent studies in mouse models of AD confirm that restricting caloric intake 30–40% from normal levels can markedly slow pathogenesis of the disease (Qin et al., 2006b; Halagappa et al., 2007). Aβ deposition in squirrel monkeys on a CR regimen is also reduced (Qin et al., 2006a). Similarly long-term studies of rhesus monkeys conducted at the National Institute on Aging and the University of Wisconsin have produced data indicating that CR animals (30% less than controls) are healthier than fully fed counterparts based on reduced incidence of various diseases, on exhibition of better indices of predisposition to disease, and slower rates of aging based on analysis of several biomarkers (Ramsey et al., 2000; Roth et al., 2004; Mattison et al., 2007; Raman et al., 2007). A recent report also indicates a significant increase in survival in CR monkeys as well as attenuation of the age-related declines in brain volume in selected regions (Colman et al., 2009). Thus, in a species closely related to humans, CR has shown promise as an intervention that could retard brain aging and neurodegenerative disease. In fact, recent reports of persons electing to practice CR close to levels applied in nonhuman primate studies have also noted many indices of reduced risk of age-related diseases, such as improved blood lipids, cardiac function, enhanced insulin sensitivity, and reduced measures of inflammation (Fontana et al., 2004; Meyer et al., 2006). Formal clinical studies of CR lasting only 6 months in duration have also documented positive impact on many indices of health and risk factors for chronic disease (Civitarese et al., 2007).

However, it is evident implementation of such a stringent regimen would be problematic due to difficulties of compliance, as well as other quality of life issues impacted by CR (McCaffree, 2004; Dirks and Leeuwenburgh, 2006). This has engendered increased attention on the development of calorie restriction mimetics (CRM) (Hursting et al., 2003; Ingram et al., 2004, 2006; Chen and Guarente, 2007), compounds which can mimic CR by targeting metabolic and stress response pathways affected by CR, but without restricting caloric intake. One of these candidates is resveratrol (Baur and Sinclair, 2006; Knutson and Leeuwenburgh, 2008; Markus and Morris, 2008). This polyphenol is found in high concentrations in red grapes and was noted to activate SIRT2 in invertebrates. SIRT1, its homolog in mammals, has actions similar to that of CR. This class of sirtuins represent NAD-dependent histone deacetylases that regulate a variety of stress responses, including CR (Guarente, 2007; Michan and Sinclair, 2007; Lavu et al., 2008). Knock-out of sirt2 in invertebrates eliminated the lifespan extension induced by CR while overexpression increased lifespan similar to CR (Guarente, 2007; Michan and Sinclair, 2007; Lavu et al., 2008). Results from a variety of recent studies in rodent models show that resveratrol can produce a remarkable range of beneficial effects including protection against high fat diets, neurodegeneration and age-related pathologies, such as cardiac function and cataracts (Baur et al., 2006; Fukuda et al., 2006; Kim et al., 2007; Lu et al., 2008; Pearson et al., 2008), and motor declines (Pearson et al., 2008), but did not significantly increase lifespan in mice on a normal diet (Pearson et al., 2008).

The fact that resveratrol is a naturally occurring polyphenol that is produced in response to fungal attack has created an interesting connection to CR (Sinclair, 2005; Howitz and Sinclair, 2008) since, as described above, diets rich in fruits and vegetables have also been related to enhanced health and longevity in human studies (Ferrari, 2004; Heber, 2004; Stanner et al., 2004), while animal studies demonstrate anti-aging effects of such diets paralleling those observed in CR (Shukitt-Hale et al., 2008b). It may be that the convergence point for CR and polyphnolic research may actually involve hormetic-enhanced stress protection (Gems and Partridge, 2008). Thus, moving beyond their demonstrated actions as antioxidants, plant polyphenols appear to have direct actions on signaling pathways involved in stress protection in neurons as described in other parts of this summary. This view provides further support for the potential for effective nutritional interventions to attenuate brain aging, neurodegeneration, and functional declines.

Conclusions

Together, the findings discussed in the previous sections provide compelling evidence to suggest lifestyle changes involving caloric selection through alterations in berry fruit, nut, fish oil, and curcumin intake, and caloric restriction mimetics may provide beneficial effects in aging and prevent or delay the onset of neurodegenerative diseases such as AD. These changes along with those not discussed in this review, such as environmental enrichment, may provide the most efficacious methods thus far for increasing “health span.” Interestingly, while many of the mechanisms for the beneficial effects of these nutritional interventions have yet to be discerned, it is clear that they involve decreases in oxidative/inflammatory stress signaling, increases in protective signaling, and may even involve hormetic effects to protect against the two major villains of aging, oxidative and inflammatory stressors.