Alzheimer’s disease (AD) is a progressive neurodegenerative brain disorder that gradually destroys a patient’s memory and ability to learn, make judgments, communicate effectively, and perform day-to-day tasks. The short-term memory is affected first, caused by neuronal dysfunction and degeneration in the hippocampus and amygdala. As the disease progresses further, neurons also degenerate and die in other cortical regions of the brain (Stuchbury and Münch, 2005). Sufferers then often experience dramatic changes in personality and behavior, such as anxiety, paranoia, or agitation, as well as delusions or hallucinations (Cummings, 2004). The prevalence of AD in the age bracket of 65–69 years is 1%; 70–74 years, 3%; 75–79 years, 6%; 80–84 years, 12%, and for people aged 85 and over the prevalence is 25%.
AD is further characterized by two major neuropathological hallmarks. The deposition of neuritic, 0-amyloid (AP) peptide-containing senile plaques in hippocampal and cerebral cortical regions of AD patients is accompanied by the presence of intracellular neurofibrillary tangles that occupy most of the cytoplasm of pyramidal neurons. Inflammation, as evidenced by the activation of microglia and astroglia, is another hallmark of AD. Inflammation, including superoxide production (“oxidative burst”), is a significant source of oxidative stress in AD patients (Münch et al., 1998; Retz et al., 1998). The inflammatory process occurs mainly around the amyloid plaques and is characterized by pro-inflammatory substances that are released from activated microglia and astroglia (Wong et al., 2001b). Cytokines, including interleukin(IL)-1o, IL-6, macrophage colony-stimulating factor (M-CSF), and tumor necrosis factor (TNF-a), are the prominent signaling molecules in the inflammatory process, being responsible for a vicious cycle of microglial and astroglial activation, resulting in the secretion of neurotoxins including superoxide and nitric oxide (Griffin et al., 1995). In addition to those morphological and physiological alterations, AD is also associated with a markedly impaired cerebral glucose metabolism as detected by reduced cortical [18F]-deoxyglucose utilization in positron emission tomography (Ishii and Minoshima, 2005).
AD patients show a progressive neuronal cell loss that is associated with region-specific brain atrophy. In particular, the earliest and most severely affected pathway is the cholinergic projection from the nucleus basalis of Meynert to areas of the cerebral cortex (Nordberg et al., 1987). Loss of basal forebrain cholinergic neurons is demonstrated by a reduction in the expression of choline acetyltransferase (ChAT), lower numbers of muscarinic and nicotinic acetylcholine receptors, and lower levels of acetylcholine (ACh) itself (Nordberg and Winblad, 1986). These changes are highly correlated with the degree of dementia in AD. ACh is derived from acetyl-CoA, the final product of the glycolytic pathway. Pyruvate, derived from glycolytic metabolism serves as an important energy source in neurons. Therefore, the inhibition of pyruvate production, for example, by glucose depletion, is considered a crucial factor that leads to acetyl-CoA deficiency in AD brains. ACh is hydrolyzed by acetylcholine esterase (AChE) and since AD patients have reduced levels of ChAT and ACh, compared to healthy elderly people, acetylcholine esterase inhibitors were introduced for the treatment of AD, but they do not delay the progression of the disease.
Currently, only symptomatic treatments with AChE inhibitors are approved for mild to moderate forms of AD. The need for an all-encompassing therapy that not only improves cholinergic transmission but also targets other pathological processes in AD is urgent, and we propose Lipoic acid (LA) to be a promising candidate for such multi-target treatment (Holmquist et al., 2007).
In vitro and in vivo studies suggest that LA also acts as a powerful micronutrients with diverse pharmacological and antioxidant properties (Packer et al., 1995). LA naturally occurs only as the R-form (RLA), but pharmacological formulations in the past have also extensively used a racemic mixture of RLA and S-lipoic acid (SLA) as stereoselective synthesis methods have not been available.
LA has been suggested to have the following properties relevant to AD:
Increase of ACh production by activation of ChAT (“LA: An activator of ChAT” section)Chelation of redox-active transition metals, thereby inhibiting the formation of hydrogen peroxide and hydroxyl radicals (“LA: A potent metal chelator” section)Scavenging of Reactive oxygen species (ROS) (thus sparing glutathione) and downregulation of redox-sensitive inflammatory signals (“LA: An anti-inflammatory antioxidant and modulator of redox-sensitive signaling” section)Scavenging of reactive carbonyl compounds including lipid peroxidation products (“LA: A carbonyl scavenger” section)Increase of glucose uptake and utilization (“LA: A stimulator of glucose uptake and utilization (“insulinomimetic”)” section)Induction of enzymes for GSH synthesis and other antioxidant protective enzymes (“Upregulation of glutathione synthesis via activation of NRF-2” section)DHLA, the reduced form of LA, is formed by reduction of LA by the pyruvate dehydrogenase (PDH) complex. Haugaard et al. have demonstrated that DHLA strongly increases the activity of a purified preparation of ChAT (Haugaard and Levin, 2000). In a further publication, the authors showed that removal of DHLA by dialysis from purified ChAT (from rabbit bladder, rat brain, and heart extracts) causes complete disappearance of enzyme activity. The addition of DHLA restored activity toward normal levels while the addition of reduced ascorbic acid or reduced nicotinamide adenine dinucleotide was not effective (Haugaard and Levin, 2002). The authors concluded that DHLA serves an essential function in the action of this enzyme and that the ratio of reduced to oxidized LA plays an important role in ACh synthesis. From these data the authors further conclude that DHLA (1) may act as a coenzyme in the ChAT reaction or (2) is able to reduce an essential functional cysteine residue in ChAT, which cannot be reduced by any other physiological antioxidant, including reduced GSH (Haugaard and Levin, 2002).
There is now compelling evidence that Aß, the main component of amyloid plaques in the AD-affected brain, does not spontaneously aggregate as was originally thought. Evidence suggests that there is an age-dependent reaction with excess metal ions in the brain (copper, iron, and zinc), which induces the peptide to precipitate and form plaques. Furthermore, the abnormal combination of Aß with copper or iron ions induces the production of hydrogen peroxide from molecular oxygen (Huang et al., 1999), which subsequently produces the neurotoxic hydroxyl radical by Fenton or Haber–Weiss reactions. Because LA is a potent chelator of divalent metal ions in vitro, the effect of an RLA inclusive diet on cortical iron levels and antioxidant status was investigated in aged rats (Suh et al., 2005). Results show that cerebral iron levels in old LA-fed animals were lower when compared to controls and were similar to levels seen in young rats. These results thus show that chronic LA supplementation may be a means to modulate the age-related accumulation of cortical iron content, thereby lowering oxidative stress associated with aging (Suh et al., 2005).
Since amyloid aggregates have been shown to be stabilized by transition metals such as iron and copper, it was also speculated that LA could inhibit aggregate formation or potentially dissolve existing amyloid deposits. Fonte et al. successfully resolubilized Aß with transition metal ion chelators and showed that LA enhanced the extraction of Aß from the frontal cortex in a mouse model of AD, suggesting that like other metal chelators, it could reduce amyloid burden in AD patients (Fonte et al., 2001). A potential side effect of a long-term therapy with high doses of a metal chelator such as LA could be the inhibition of metal containing enzymes such as insulin degrading enzyme or superoxide dismutase. Suh et al. investigated whether LA and DHLA remove copper or iron from the active sites of Cu, Zn superoxide dismutase, and aconitase. They found that even at millimolar concentrations neither LA nor DHLA altered the activity of these enzymes (Suh et al., 2004b), providing promising results for the long-term use of LA in AD.
AD is accompanied by a chronic inflammatory process around amyloid plaques, characterized by the activation of microglia and astrocytes and increased levels of radicals and pro-inflammatory molecules such as iNOS, IL-1 â, IL-6, and TNF-á (Griffin et al., 1995). AD patients also show increased cytokine levels (e.g., IL-1 0 and TNF-a) in the cerebral spinal fluid (CSF), with TNF-a being a good predictor for the progression from mild cognitive impairment to AD. Recently, much attention has been paid to ROS as mediators in signaling processes, termed “redox-sensitive signal transduction.” ROS modulate the activity of cytoplasmic signal transducing enzymes by at least two different mechanisms: oxidation of cysteine residues or reaction with iron–sulfur clusters. One widely investigated sensor protein is the p21Ras protein (Lander et al., 1997). Activation of Ras by oxidants is caused by oxidative modification of a specific cysteine residue (Cys118). Ras interacts with PI3- kinase, protein kinase C, diacylglycerol kinase, and MAP-kinase-kinase-kinase, regulating expression of IL-1 0, IL-6, and iNOS. LA can scavenge intracellular free radicals (acting as second messengers), downregulate pro-inflammatory redox-sensitive signal transduction processes including NF-xB translocation, and thus attenuate the release of more free radicals and cytotoxic cytokines (Bierhaus et al., 1997; Wong et al., 2001a).
Cellular and mitochondrial membranes contain a significant amount of arachidonic acid and linoleic acid, precursors of lipid peroxidation products 4-hydroxynonenal (HNE) and acrolein that are extremely reactive. Acrolein decreases pyruvate dehydrogenase (PDH) and a-ketoglutarate dehydrogenase (KGDH) activities by covalently binding to LA, a component in both the PDH and KGDH complexes. Acrolein, which is increased in AD brains, may be partially responsible for the dysfunction of mitochondria and loss of energy found in the AD-affected brain through its inhibition of PDH and KGDH activities, potentially contributing to neurodegeneration (Pocernich and Butterfield, 2003). In a further study, levels of lipid peroxidation, oxidized glutathione (GSSG), and nonenzymatic antioxidants and the activities of mitochondrial enzymes were measured in liver and kidney mitochondria of young and aged rats before and after LA supplementation. In both the liver and kidney, a decrease in the activities of mitochondrial enzymes was observed in aged rats. LA supplemented aged rats showed a decrease in the levels of lipid peroxidation and inhibition of the activities of mitochondrial enzymes like isocitrate dehydrogenase, KGDH, succinate dehydrogenase, NADH dehydrogenase, and cytochrome C oxidase. The authors conclude that LA reverses the age-associated decline in mitochondrial enzymes and therefore may lower the increased risk of oxidative damage that occurs during aging (Arivazhagan et al., 2001).
Increased prevalence of insulin abnormalities and insulin resistance in AD may contribute to the disease pathophysiology and clinical symptoms. Insulin and insulin receptors are densely but selectively expressed in the brain, including the medial temporal regions that support the formation of memory. It has recently been demonstrated that insulin-sensitive glucose transporters are localized to the same regions and that insulin plays a role in memory functions. Collectively, these findings suggest that insulin contributes to normal cognitive functioning and that insulin abnormalities may exacerbate cognitive impairments, such as those associated with AD (Watson and Craft, 2003). This view is further supported by the finding that higher fasting plasma insulin levels and reduced CSF: plasma insulin ratios (suggestive of insulin resistance) have also been observed in patients with AD. When AD patients were treated with insulin in a glucose clamp approach, a marked enhancement in memory was observed, whereas normal adults’ memory was unchanged (Craft et al., 2003). As previously mentioned, AD is associated with a markedly impaired cerebral glucose metabolism in affected regions. Impaired glucose uptake (partially mediated by insulin resistance) in vulnerable neuronal populations not only compromises production of ACh but also renders neurons vulnerable to excitotoxicity and apoptosis. There is abundant evidence that LA can ameliorate insulin resistance and impaired glucose metabolism in the periphery in type II diabetes mellitus. One study examined the beneficial effects of LA on glucose uptake using soleus muscles derived from nonobese, insulin-resistant type II diabetic Goto-Kakizaki rats, a genetic rat model for human type II diabetes. In this model, chronic administration of LA moderately improved the diabetes-related deficit in glucose metabolism and protein oxidation, as well as the activation of Akt/PKB and PI3K by insulin (Bitar et al., 2004).
In a further study, the incorporation of 14C-2-deoxyglucose (2DG) into areas of basal ganglia was investigated in rats treated acutely or for 5 days with RLA or SLA. Following acute administration, RLA was more effective than SLA in increasing 14C-2DG incorporation. For example, acute administration of RLA caused an approximate 40% increase in 14C-2DG incorporation in the substantia nigra while SLA was without effect. However, the effects observed were dependent on basal 14C-DG incorporation in different rat strains. Following subacute administration, the pattern of change in 14C-2DG incorporation was altered and both isomers were equally effective. The effects of RLA were largely maintained with increasing animal age, but the ability of the S-isomer to alter 14C-2DG incorporation was lost by 30 months of age. The authors conclude that RLA has the ability to increase glucose utilization in vivo, which may be relevant to the treatment of neurodegenerative disorders (Seaton et al., 1996). Based on this and similar studies it is quite conceivable that LA might increase glucose uptake in insulin-resistant neurons and thus provide more glycolytic metabolites including acetyl-CoA for these neurons. Since ACh synthesis depends on the availability of acetyl-CoA provided from glucose metabolism, LA might additionally be able to directly increase the concentration of the substrate acetyl-CoA for ACh synthesis (Hoyer, 2003).
The tri-peptide y-L-glutamyl-L-cysteinyl-glycine or glutathione (GSH) is the most abundant nonprotein thiol in animal cells. GSH is required for the maintenance of the thiol redox status of the cell, protection against oxidative damage, detoxification of endogenous and exogenous reactive metal ions and electrophiles, storage and transport of cysteine, as well as protein and DNA synthesis, cell cycle regulation, and cell differentiation (Butterfield et al., 2002). GSH and GSH enzymes play a key role in protecting the cell against the effects of ROS. The key functional element of GSH is the cysteinyl moiety, which provides the reactive thiol group. ROS are reduced by GSH through the enzymatic activity of glutathione peroxidase (GSH-Px) (Butterfield et al., 2002). As a result, GSH is oxidized to GSSG, which is rapidly reduced back to GSH by glutathione reductase (GR) at the expense of NADPH. This is a redox-cycling mechanism to prevent GSH loss (Pocernich et al., 2000). The key role of GSH is that it is a cofactor of glutathione peroxidase and glyoxalase I, important for the detoxification of ROS and methylglyoxal, respec-tively. In the de novo synthesis, GSH is synthesized from its constituent amino acids by the sequential action of two enzymes, y-glutamylcysteine synthetase (y-GCS), which is the rate-limiting enzyme, and glutathione synthetase. y-GCS catalyzes the formation of the dipeptide y-glutamylcysteine, which is the rate-limiting substrate in this reaction.
Nuclear factor E2–related factor 2 (Nrf2) is a transcription factor known to induce expression of a variety of cytoprotective and detoxification genes. In recent years, Nrf2 has become a promising novel drug target. Activators of Nrf-2-mediated tran-scription increase the expression of enzymes involved in GSH synthesis, to main-tain sustainable high GSH production and provide protection to neurons against oxidative stress. Nrf-2 activators (a class of potential GSH “boosters”) include tert-butylhydroquinone (TBH), sulforaphane (from broccoli), resveratrol, a variety of polyphenols, and a-lipoic acid (Karelson et al., 2001).
In 2002, it was suggested that lipoic acid, like the structurally related dithiolethi-ones, such as anethole dithiolethione (ADT), induces phase II detoxification enzymes (which are involved in conjugation reactions) in cultured astroglial cells. LA, like ADT, induced a highly significant, time and concentration dependent, increase in the activity of NAD(P)H dehydrogenase (NQO1) and glutathione-S-transferase (GST) in C6 astroglial cells. The LA- or ADT-mediated induction of NQO1 was further confirmed by quantitative PCR and Western blot analysis. This work for the first time unequivocally demonstrates LA-mediated upregulation of phase II detoxification enzymes, which may highly contribute to the neuroprotective potential of LA. Moreover, the data support the notion of a common mechanism of action of LA and ADT (Flier et al., 2002).
In 2004, Hagen’s group at the Linus Pauling Institute at Oregon State University discovered that R-lipoic acid is an in vivo inducer of Nrf2 and increases the enzymatic activity of gamma-glutamylcysteine ligase on a transcriptional level. They observed that the rate-controlling enzyme in GSH, gamma-glutamylcysteine ligase (GCL) loses enzymatic activity with age. With age, the expression of the catalytic (GCLC) and modulatory (GCLM) subunits of GCL decrease by about 50%. In addition, approximately 50% age-related loss in total and nuclear Nrf2 levels was observed, suggesting attenuation of Nrf2-dependent gene transcription. To determine whether the constitutive loss of Nrf2 transcriptional activity also affects the inducible nature of Nrf2 nuclear translocation, old rats were treated with RLA. LA administration increased nuclear Nrf2 levels in old rats and induced Nrf2 binding to the ARE and consequently, higher GCLC levels and GCL activity were observed after LA injection (Suh et al., 2004a).
A13, the major component of senile plaques, contributes to neuronal degeneration in AD by stimulating the formation of free radicals. Zhang et al. have investigated the potential efficacy of LA against cytotoxicity induced by A13 (30 µM) and hydrogen peroxide (100 µM) in primary neurons of rat cerebral cortex and found that treat-ment with LA protected cortical neurons against cytotoxicity induced by both toxins (Zhang et al., 2001). In a similar study, Lovell et al. investigated the effects of LA and DHLA on neuronal hippocampal cultures treated with A13 (25–35) and iron/ hydrogen peroxide (Fe/H2O2) (Lovell et al., 2003).
In a further study, Müller and Krieglstein tested whether pretreatment with LA can protect cultured neurons against injury caused by cyanide, glutamate, or iron ions. Neuroprotective effects were only significant when the pretreatment with LA occurred for >24 h. The authors conclude that neuroprotection occurs only after pro-longed pretreatment with LA and is probably due to the radical scavenger properties of endogenously formed DHLA (Muller and Krieglstein, 1995).
In summary, data from these studies suggest that pretreatment of neurons with LA (or application of DHLA) before exposure to A13 or Fe/H2O2 significantly reduces oxidative stress and increases cell survival. However, concomitant application of A13 or Fe/H2O2 with LA can temporarily increase oxidative stress as the reduction of LA by the PDH complex consumes reducing equivalents and inhibits energy production.
Protective effects of LA against cognitive deficits have been shown in several studies in aged rats and mice. In one study, a diet supplemented with RLA was fed to aged rats to determine its efficacy in reversing the decline in metabolism seen with age. Young (3–5 months) and aged (24–26 months) rats were fed for 2 weeks. Ambulatory activity, a measure of general metabolic activity, was almost threefold lower in untreated old rats vs. controls, but this decline was reversed in old rats fed with RLA (Hagen et al., 1999). In a combination treatment study, the effects on cognitive function, brain mitochondrial structure, and biomarkers of oxidative damage were studied after feeding old rats a combination of acetyl-L-carnitine (ALCAR) and/or RLA. Dietary supplementation with ALCAR and/or RLA improved memory, the combination being the most effective for tests of spatial memory and temporal memory. The authors suggest that feeding ALCAR and RLA to old rats improves performance on memory tasks by lowering oxidative damage and improving mito-chondrial function. Feeding the substrate ALCAR with RLA restores the velocity of the reaction (K(m)) for ALCAR transferase and mitochondrial function. The principle appears to be that, with age, increased oxidative damage to protein causes a deformation of structure of key enzymes with a consequent lessening of affinity (K(m)) for the enzyme substrate (Liu et al., 2002).
Similar experiments were performed in the senescence accelerated prone mouse strain 8 (SAMP8), which exhibits age-related deterioration of memory and learning along with increased oxidative markers, and provides a good model for disorders with age-related cognitive impairment. In one study, the ability of LA (and also N-acetyl-L-cysteine) to reverse the cognitive deficits found in the SAMP8 mouse was investigated. Chronic administration of LA improved cognition of 12 month old SAMP8 mice in the T-maze foot-shock avoidance paradigm and the lever press appetitive task. Furthermore, treatment of 12 month old SAMP8 mice with LA reversed all three indexes of oxidative stress. These results provide further support for a therapeutic role for LA in age- and oxidative stress–mediated cognitive impair-ment, including that associated with AD (Farr et al., 2003).
The effects of LA were also investigated in various animal models of familial AD. For example, 10 month old Tg2576 and wild-type mice were fed an LA-containing diet for 6 months and then assessed for the diet’s influence on memory and neuropathology. LA-treated Tg2576 mice exhibited significantly improved learning and memory retention in the Morris water maze task compared to untreated Tg2576 mice. Twenty-four hours after contextual fear conditioning, untreated Tg2576 mice exhibited significantly impaired context-dependent freezing. Assessment of brain soluble and insoluble Ap levels revealed no differences between LA-treated and untreated Tg2576 mice. The authors conclude that chronic dietary LA can reduce hippocampal-dependent memory deficits of Tg2576 mice without affecting Ap levels or plaque deposition (Quinn et al., 2007).
Although LA has been used for the treatment of diabetic polyneuropathy in Germany for more than 30 years, no epidemiological study has taken advantage of this large patient population and investigated whether the incidence of AD in the LA-treated patients is lower than in the untreated diabetic and/or untreated nondiabetic population. Therefore, the first indication for a beneficial effect of LA in AD and related dementias came from a rather serendipitous case study. In 1997, a 74 year old patient presented herself at the Department of Medical Rehabilitation and Geriatrics at the Henriettenstiftung Hospital in Hannover with signs of cognitive impairment. Diabetes mellitus and a mild form of polyneuropathy were her main concomitant diseases. With clinical criteria of Diagnostic and Statistical Manual of Mental Disorders (DSM)-III-R, deficits in the neuropsychological tests, an MRI without signs of ischemia, and a typical single photon emission computed tomography show-ing a decreased bi-temporal and bi-parietal perfusion, early stage AD was diagnosed. Treatment with AChE inhibitors was initiated and the patient received 600 mg LA each day for treatment of her diabetic polyneuropathy. Since 1997, several retests were performed, which showed no substantial decline in the patient’s cognitive functions. Therefore, the diagnosis of mild AD was reevaluated several times, but the diagnosis did not change and the neuropsychological tests showed an unusually slow progression of her cognitive impairment. This observation inspired an open pilot trial at the Henriettenstiftung Hospital in Hannover.
LA was administered at a dose of 600 mg once daily (in the morning 30 min before breakfast) to nine patients with probable AD (age: 67 ± 9 years, mini-mental state examination (MMSE) score at first visit/start of AChE inhibitor therapy: 23 ± 2 points) receiving a standard treatment with AChE inhibitors over an observation period of 337 ± 80 days. The cognitive performance of the patients before and after addition of LA to their standard medication was compared. A steady decrease in cognitive performance (a 2 points/year decrease in scores in the MMSE and a 4 points/year increase in the AD assessment scale, cognitive subscale (ADAScog) was observed before initiation of the LA regi-men. Treatment with LA led to a stabilization of cognitive function, demonstrated by constant scores in two neuropsychological tests for nearly a year. This study was continued and in the end included 43 patients who were followed for an observation period of up to 48 months. In patients with mild dementia (ADAScog < 15), the disease progressed extremely slowly (ADAScog: +1.2 points = year, MMSE: -0.6 points/year). In patients with moderate dementia the disease progressed at approximately twice the rate (Hager et al., 2007). However, this study was small and not ran-domized. In addition, patients were diagnosed with “probable” AD, and the diagnosis was (in nearly all cases) not confirmed by a neuropathological postmortem analysis. Therefore, a double-blind, placebo-controlled phase II trial is urgently needed before LA could be recommended as a therapy for AD and related dementias. The first randomized, double-blind, placebo-controlled 12 month pilot study involving LA was conducted with 39 subjects diagnosed with mild to moderate AD in Portland, Oregon, USA. Inclusion criteria consisted of the following: aged 55 years or older, diagnosis of probable AD, MMSE score 15–26, Clinical Dementia Rating Scale 0.5–1.0, not depressed, general health status that would not interfere with patients ability to participate and complete the study. Subjects were allowed to continue stable doses of medications for cognitive impairment (e.g., AChE inhibitors, memantine) and stable doses of dietary supplements (e.g., ginkgo biloba, vitamin E). Subjects were excluded if they were eating fish more than one serving per week, taking omega-3 fatty acid supplementation, or taking LA supplementation. Subjects that met inclusion criteria were randomized to one of three groups: (1) placebo, (2) omega-3 fatty acids (cO-3), and (3) omega-3 fatty acids plus lipoic acid (co-3+LA).
Omega-3 fatty acids were administered in the form of fish oil concentrate at 3 g/day, containing a daily dose of 675 mg DHA and 975 mg EPA (Lynne Shinto, OHSU, personal communication). LA was given as the racemic form with a daily dose of 600 mg. The primary outcome measure was peripheral F2-isoprostane levels to measure lipid peroxidation. Of the 39 randomized subjects, 32 completed their 12 month outcomes visit. There was no baseline difference between groups on medication for the treatment of cognitive impairment (p = 0.50, mean use range 77%–92%) and no difference between groups on 12 month change in peripheral F2-isoprostane levels (p = 0.10). For secondary clinical measures, a significantly delayed decline in MMSE score between groups favoring co-3 + LA (p = 0.04) over 12 months but no difference between the groups in the ADAS-cog (p = 0.54) was found. In summary, the promising results found for the co-3+LA group in the double-blind placebo control trial and for LA alone in the open trial for delayed cognitive decline warrant further investigation of LA in combination with other anti-inflammatory compounds for the treatment of AD, but should include longer studies and patients in different stages of the disease (Münch et al., 2010).
Since AD is a multifactorial disease, it has been suggested that a combination, rather than a single drug, might be most beneficial for AD patients. Among many suggested add-on treatments to LA, nutraceuticals with antioxidant and anti-inflammatory properties could potentially be promising candidates (Steele et al., 2007). Nutraceuticals may be broadly defined as any food substance that offers health or medical benefits (Ferrari and Torres, 2003; Ferrari, 2004). Given that plant foods are derived from biological systems, they contain many compounds in addition to traditional nutrients that can elicit biological responses and are also termed phytonutrients. One of the largest groups of phytonutrients that may confer beneficial health effects are polyphenols (Shanmugam et al., 2008). Over the past decade, polyphenols, which are abundant in fruits and vegetables, have gained recognition for their antioxidant properties and their roles in protecting against chronic diseases such as cancer and cardiovascular diseases (Hertog et al., 1997; Liu, 2004). Consequently, diet is now considered to be an important environmental factor in the development of late-onset AD (Solfrizzi et al., 2003). Polyphenols are therefore beginning to attract increasing interest. Numerous epidemiological studies have suggested a positive association between the consumption of polyphenol-rich foods and the prevention of diseases.
A recent epidemiological study reported that consumption of fruit and vegetable juices (high in polyphenols) more frequently than three times a week resulted in a 76% reduction in the risk of developing probable AD over a 9 year period (Dai et al., 2006). Another epidemiological study of 1010 subjects aged 60–93 reported that individuals who consumed curry (containing curcumin) “often” and “very often” had significantly better cognitive test scores as measured via MMSE (Ng et al., 2006). As antioxi-dants, polyphenols may protect cells against oxidative damage, thereby limiting the risk of AD, which is associated with oxidative stress. Converging epidemiological data also suggests that a low dietary intake of omega-3 (cO-3) essential fatty acids is a candidate risk factor for AD (Morris et al., 2003; Maclean et al., 2005). Docosahexaenoic acid (DHA) is one of the major co-3 fatty acids in the brain where it is enriched in neurons and synapses. DHA is associated with learning and memory and is also required for the structure and function of brain cell membranes. In the AD-affected brain DHA levels are known to be decreased (Soderberg et al., 1991; Prasad et al., 1998), while people who ingest higher levels of DHA are less likely to develop AD (Morris et al., 2003; Tully et al., 2003). As many Western diets have been reported to be deficient in DHA and also low in polyphenolic content, supple-mentation with DHA and polyphenols may offer potential preventative treatments for AD. Several of these nutraceuticals/phytonutrients (e.g., (-)-epigallocatechin gallate (EGCG) from green tea, curcumin from the curry spice turmeric and the omega-3 DHA from fish oils) have shown promising results, when used as single therapies in animal studies.
EGCG has previously been shown to prevent neuronal cell death caused by Ap neurotoxicity in cell cultures (Choi et al., 2001; Levites et al., 2003). A study by Rezai-Zadeh et al. (Rezai-Zadeh et al., 2005) reported that EGCG reduced Ap gen-eration in vitro in neuronal-like cells and primary neuronal cultures from Tg2576 mice, along with promotion of the nonamyloidogenic a-secretase proteolytic path-way. Furthermore when 12 month old Tg2576 mice were treated with 20 mg/kg EGCG via intra-peritoneal injections for 60 days, it was found that Ap levels and plaque load in the brain were decreased. Curcumin has been reported to be several times more potent than vitamin E as a free radical scavenger (Zhao et al., 1989) and there is also increasing evidence showing that curcumin can inhibit Ap aggregation (Yang et al., 2005). In a study by Lim et al. curcumin was tested for its ability to inhibit the combined inflammatory and oxidative damage in Tg2576 transgenic mice. In this study Tg2576 mice aged 10 months old were fed a cur-cumin diet (160 ppm) for 6 months. Their results showed that the curcumin diet significantly lowered the levels of oxidized proteins, IL-1p, the astrocyte marker glial fibrillary acidic protein (GFAP), soluble and insoluble Ap, and also plaque burden (Lim et al., 2001). Following on from this work, Yang et al. evaluated the effect of feeding a curcumin diet (500 ppm) to 17 month old Tg2576 mice for 6 months. When fed to the aged Tg2576 mice with advanced amyloid accumulation, curcumin resulted in reduced soluble amyloid levels and plaque burden. These data raise the possibility that dietary supplementation with curcumin may provide a potential preventative treatment for AD by decreasing Ap levels and plaque load via inhibition of Ap oligomer formation and fibrillization, along with decreasing oxidative stress and inflammation.
The interest in dietary DHA supplementation has arisen from the approach to protect neurons from neuronal degradation and therefore prevent neurological diseases like AD. Converging epidemiological data suggests that a low dietary intake of co-3 polyunsaturated fatty acids is a candidate risk factor for AD (Calon et al., 2005). In the AD brain, DHA is known to be decreased (Soderberg et al., 1991; Prasad et al., 1998), while people who ingest higher levels of DHA are less likely to develop AD (Conquer et al., 2000; Barberger-Gateau et al., 2002; Morris et al., 2003). A recent in vitro study by Florent et al. demonstrated that DHA provided cortical neurones with a higher level of resistance to the cytotoxic effects induced by soluble Ap oligomers (Florent et al., 2006). Lukiw et al. also demonstrated that DHA decreased Ap40 and Ap42 secretion from aging human neuronal cells (Lukiw et al., 2005). A study by Calon et al. showed that a reduction of dietary co-3 PUFA in Tg2576 transgenic mice resulted in a loss of post-synaptic proteins and behavioral deficits, while a DHA-enriched diet prevented these effects (Calon et al., 2004). Other studies have shown that DHA protects neurons from Ap accumulation and toxicity and ameliorates cognitive impairment in rodent models of AD (Hashimoto et al., 2005; Lim et al., 2005). A recent study by Cole and Frautschy showed that DHA supplementation in Tg2576 transgenic mice aged 17 months markedly reduced Ap accumulation, oxidative damage, and also improved cognitive function (Cole and Frautschy, 2006). Thus, dietary supplementation with DHA may also provide a potential preventative treatment for AD via prevention of cognitive deficits and reduction of Ap accumulation and oxidative stress.
These nutraceuticals have all been demonstrated to have varying mechanisms of action, relating to decreasing cognitive deficits, oxidative stress, inflammation, and Ap levels. Therefore, combination therapies of these nutraceuticals containing polyphenols (EGCG and curcumin), co-3 essential fatty acids (DHA), and LA have the potential to provide nutritional supplement therapies for the prevention of AD pathology and cognitive impairments as LA has been previously demonstrated in Tg2576 mice to prevent cognitive deficits(Quinn et al., 2007). The use of LA in addition to other nutraceuticals would provide the means to prevent cognitive deficits in combination with the benefits of curcumin, EGCG, and DHA to decrease oxidative stress, inflammation, Ap levels, and Ap plaque load.
