Review
Neuroprotective action of omega-3 polyunsaturated fatty acids against neurodegenerative diseases: Evidence from animal studies

https://doi.org/10.1016/j.plefa.2007.10.019Get rights and content

Abstract

Studies in animals clearly show that oral intake of docosahexaenoic acid (DHA) can alter brain DHA concentrations and thereby modify brain functions. This provides us with an opportunity to use DHA as a nutraceutical or pharmaceutical tool in brain disorders such as Alzheimer disease (AD) and Parkinson disease (PD). Most of the published epidemiological studies are consistent with a positive association between high reported DHA consumption or high DHA blood levels and a lower risk of developing AD later in life. Such observations have prompted the investigation of DHA in three different transgenic models of AD. These analyses show that animal models of AD are more vulnerable to DHA depletion than controls and that DHA exerts a beneficial effect against pathological signs of AD, including Aβ accumulation, cognitive impairment, synaptic marker loss, and hyperphosphorylation of tau. Multiple mechanisms of action can be associated with the neuroprotective effects of DHA and include antioxidant properties and activation of distinct cell signaling pathways. Although the first randomized clinical assays have yet failed to demonstrate convincing beneficial effects of DHA for AD patients, the knowledge gathered in recent years holds out a hope for prevention and suggests that the elderly and people bearing a genetic risk for AD should at least avoid DHA deficiency.

Introduction

Fatty acids comprise over 10% (w/w) of the dry weight of the brain [1], [2], [3]. The role of these fatty acids in the brain ranges from cell membrane building blocks to highly bioactive compounds. Among the major brain fatty acids, docosahexaenoic acid (DHA; 22:6n-3, or cervonic acid) is one of major brain fatty acids and represents between 12% and 16% of total fatty acids in gray matter lipids [1], [2]. Quite remarkably, the main source of DHA for our brain remains our diet. Despite a strong capacity of the central nervous system to retain its DHA, a wealth of animal studies reveal that the dietary intake of DHA can significantly alter levels of DHA in the brain [2], [4], [5], [6], [7], [8]. This vulnerability to an environmental factor might seem surprising given the importance of DHA in brain function. However, from a therapeutic point of view, this peculiar characteristic rather represents an opportunity because it opens the door to the use of exogenous DHA or other long-chain polyunsaturated fatty acids (PUFA) to directly influence brain function. Since omega-3 (n-3) PUFAs are widely available and have an excellent safety profile, they could stand up as nutraceutical defenses against brain diseases. Alzheimer disease (AD) and Parkinson disease (PD) are the most common neurodegenerative diseases that deprive millions of people worldwide of enjoying successful aging. Recent studies have tried to decipher the role of DHA in these prevalent disorders and to probe for a therapeutic effect of DHA. In the present review, we will provide an overview of the literature on the role of DHA in AD and PD with a focus on investigations performed in animal models of these neurodegenerative diseases.

Prospective studies were designed to investigate the links between the dietary intake of PUFA and the risk of developing AD [9]. In a questionnaire-based 4-year study, which was part of the Chicago Healthy Aging Project (CHAP), patients who consumed high amounts of DHA (median: 0.10 g/day) had a lower risk of developing AD compared with those who ate low amounts of DHA (median: 0.03 g/day) [10]. This positive association was found for total n-3 PUFA and for DHA, but not with eicosapentaenoic acid (EPA) [10]. An earlier prospective questionnaire-based study reported lower relative risk of developing dementia in individuals who consumed fish or seafood at least once a week; they had a significantly lower risk of being diagnosed as having dementia in the seven subsequent years [11]. However, this association did not show up as significant specifically for AD and was lost after adjustment for education [11]. Another large study, better known as the Rotterdam study, did not highlight any positive association between n-3 PUFA ingestion and the risk of developing AD [12], in contrast with an earlier Rotterdam study report [13]. As it is impossible to determine exactly the DHA intake of the study participants, the investigators have to rely on questionnaires and assumptions on the nutrient content of listed types of food, which obviously deliver imprecise information.

A second set of studies directly assessed blood concentration of n-3 PUFA in relation with the diagnosis of AD at the same time (case–control studies) or years after sampling (prospective studies) [9]. In patients already diagnosed with AD or other dementias, Conquer et al. found decreased DHA in total phospholipids, in phosphatidylcholine (PC), in phosphatidylethanolamine (PE) but not in lysophosphatidylcholine fractions of plasma in comparison to age-matched control volunteers [14]. Consistent with this, Tully et al. reported lower levels of EPA and DHA in the cholesteryl ester fractions of the plasma drawn from AD patients compared with controls [15]. However, such an effect on blood DHA levels may likely be a consequence of dementia rather than a cause of it. On the other hand, at least two independent prospective studies found that lower plasma DHA levels increased the risk of developing AD later in life [16], [17]. In the recent study of Schaefer et al., patients with blood DHA concentrations in the highest quartile had a lower risk of developing dementia compared with the lowest three quartiles during a mean follow-up of 9 years [17]. This positive association is only detected with DHA and not EPA, in good agreement with the prospective study of Morris et al. [10]. Altogether, these analyses suggest that dietary intake of DHA can alter the risk of developing AD over the long term and that low blood DHA concentrations might be an important risk factor for AD. This hypothesis could be strengthened by investigating DHA levels in mild cognitive impairment (MCI), a cognitive diagnosis presumed to include prodromal AD patients, but such data are not available yet [18]. It must be noted, however, that another large prospective study has been published showing no inverse relationship between blood DHA and the incidence of cognitive decline or dementia [19]. Interestingly, the studies from Schaefer et al., Heude et al., and Laurin et al. have measured DHA in PC fractions, erythrocyte membranes and total plasma phospholipids, respectively [16], [17], [19]. It is possible that these differences in methodology can account for the discrepancies between the results of the studies. If one can draw a parallel with blood cholesterol measurements, the “neuroprotective” role of blood DHA might depend on its incorporation into a specific blood compartment. For instance, subtypes of blood DHA can be more easily transported into the brain than another form [20], [21]. Blood DHA pools might also differ in their long-term stability and might be more or less reliable assessment of chronic dietary consumption or fatty acid metabolism.

In contrast to AD, the literature linking fat intake and PD risk is very limited. Two recent questionnaire-based prospective studies reported an association between high dietary consumption of saturated fat and low intake of unsaturated fatty acids with higher risk of developing PD [22], [23]. This observation remains in agreement with some of the older retrospective studies [24], [25], [26]. However, no study to date has directly associated n-3 PUFA intake and the risk of suffering from PD.

An even more direct way to assess the importance of DHA in AD or PD is to assess the postmortem brain fatty acid profile of AD patients compared with controls, if the available tissue meets certain quality standards. Soderberg first reported in 1991 a gas chromatography analysis from 8 to 10 AD patients and 8 to 10 controls showing decreased levels of DHA in the PE fractions of the frontal cortex and the hippocampus [27]. This observation was partly confirmed by a study from Prasad et al. showing decreased DHA in the PE fraction of the hippocampus in 9 AD patients compared with the same number of controls [28]. However, Prasad et al. found no significant difference in the cerebral cortex nor in PC or phosphatidylinositol fractions [28]. Two additional gas chromatography investigations did not report significant changes in DHA concentrations in the cerebral cortex of 15 AD patients [29] or in the hippocampus of 8 AD patients [30]. In a more recent publication, Lukiw showed a 50% decrease in DHA in the hippocampus and temporal cortex of AD patients using HPLC coupled with MS detection [31]. Interestingly, products of DHA enzymatic metabolism (e.g. neuroprotectin 1) are also decreased in AD [31] whereas other oxidized DHA products such as neuroprostanes are increased [32]. A single study has sought to determine the fatty acid profile of levodopa-treated PD patients compared to control individuals using gas chromatography. This report revealed no significant difference in n-3 PUFA concentration in the brain cortex between PD patients and controls [1]. Overall, important discrepancies between studies preclude us from firmly concluding that AD or PD are associated with altered brain levels of DHA [33]. In any case, the identification of specific pools of DHA that could prove to be more closely linked to the progression of the disease probably would shed some light on this issue. Finally, because DHA is heavily enriched in neurons and synapses relative to glia [34], observed losses of DHA may be difficult to interpret.

The importance of DHA in the maintenance of cognitive capacity and other behavioral parameters was demonstrated by a large number of studies in animals and has been reviewed recently [35]. These major effects are likely to be explained by the finding that DHA is specifically enriched in synapses [34]. To investigate the effect of n-3 PUFA on the pathogenesis of AD, a first generation transgenic animal model for memory loss and Aβ deposition (Tg2576 mouse; APPswe) were exposed to low n-3 PUFA dietary intake along with very high n-6 fatty acid intake from 16 to 20 months of age. This DHA-depleting treatment induced massive losses of the postsynaptic markers drebrin, PSD-95, NMDA receptor subunits and CaM kinase II, without concomitant loss of presynaptic markers synaptophysin or SNAP-25 in the cortex of Tg2576 mice [36], [37]. In parallel, n-3 PUFA restriction increased caspase-cleaved actin content and oxidative stress markers in the brain cortex [37]. Such postsynaptic defects were not observed in non-transgenic mice but were present to various degrees in the brains of AD patients [36], [37]. Most importantly, Tg2576 mice fed with a high-DHA (∼1 g/kg/day) diet exhibited improved spatial memory and were protected against most of the molecular biomarkers changes listed above, suggesting a prominent neuroprotective role [36], [37]. DHA treatment also reduced the amyloid plaque burden as well as the cortical accumulation of insoluble Aβ40 and Aβ42 peptides [38]. These results suggest that, on one hand, the combination of a genetic risk factor of AD (mutant APP) and an environmental risk factor for AD (low n-3 PUFA dietary intake) potentiates AD pathology in the mouse brain and, on the other hand, that DHA exerts a neuroprotective action in this model.

In agreement with these data, DHA treatment (∼0.9 g/kg/day) lasting from 6 to 10 months decreased both Aβ40 and Aβ42 peptided in the hippocampus of a different transgenic animal model (APPswe/PS1dE9) [39]. More recently, Green et al. observed that high DHA dietary intake (∼2.3 g/kg/day) during 9 months exerted a protective effect restricted to intraneuronal Aβ (immunohistochemistry) and the soluble form of Aβ40 (ELISA) in the 3XTg-AD mouse model of AD [40]. Aside from this limited effect on Aβ pathology, DHA reduced the accumulation of tau and phosphorylated tau in the soluble fractions of protein extracted from the whole brain [40].

Consistent with the studies in transgenic animals, a series of publications from Hashimoto et al. have shown that dietary DHA exerted both a protective and a restorative action against cognitive impairment induced in rats by intracerebral infusion of Aβ40 peptides with metal injection [41], [42], [43]. They further reported that DHA decreased Aβ40 peptide and cholesterol levels [41], [42], [43].

These studies performed in various animal models for AD using different paradigms all suggest that DHA acts against a wide range of pathological signs. However, an important issue in the use of DHA in animal diets is to make sure that it is protected from oxidation to which long-chain PUFA are vulnerable [44]. Except the study performed in Tg2576 mice, which used microencapsulated DHA [37], most of the other reports did not provide details on how DHA concentration in the diet was maintained during the study. Thus, effects of DHA derivatives produced prior to vivo administration cannot be ruled out.

In the PD research field, it has been recently demonstrated that short-term administration of DHA (100 mg/kg) reduced by about 40% the extent of levodopa-induced dyskinesias in a non-human primate model of Parkinsonism [45]. The authors proposed that this effect of DHA can be explained by the activation of nuclear receptors that operate as transcription factors, such as retinoid X receptors. Another study reports that a 1-month treatment with levodopa decreased DHA concentrations in the brain cortex in a primate animal model of PD, but increased arachidonic acid and n-6 docosapentaenoic acid levels [1]. Data consistent with an upregulating effect of chronic levodopa treatment on post mortem brain arachidonic acid were also collected in levodopa-treated PD patients [1]. Thus, the most prescribed treatment against the symptoms of PD appears to modulate brain PUFA while its effect on motor activity appears to be regulated by DHA. Of further interest, recent data also demonstrate a protective effect of DHA against MPTP-induced neurotoxicity in a mouse model of acute parkinsonism, in terms of preserved dopamine levels, tyrosine hydroxylase-positive neurons and nurr-1 expression [46].

Besides the possibility of isolating variables such as dietary factors or terminal endpoints, another important advantage of studies performed in animal models is their capacity to uncover information to decipher the mechanisms by which DHA and other PUFAs induce their effects in the brain (Fig. 1). One of the first proposed mechanisms for the neuroprotective action of DHA is that it exerts anti-oxidative activity in vivo [37], [42], [47], [48], [49]. Indeed, evidence of DHA increasing glutathione reductase activity [42], decreasing the accumulation of oxidized proteins [37], [48] and levels of lipid peroxide and reactive oxygen species [42], [43] have been published. Data supporting DHA-induced inactivation of a cell-signaling pathway leading to caspase activation [36], [37] or to hyperphosphorylation of tau [40] have also been shown. Interestingly, the capacity of DHA to regulate the phosphatidylinositol-3 kinase (PI3-K)-Akt cascade has been elegantly demonstrated in vitro [50], [51], [52]. An action of DHA on beta-secretase or gamma-secretase complex activity has been partly ruled out, but recent data indicated that DHA can downregulate presenilin-1 in vitro and in vivo [38], [40]. Other potential mechanisms of action that remain to be thoroughly studied in animal models include regulation of inflammatory process, gene transcription or cell membrane properties [53], [54], [55].

The bulk of evidence from epidemiological and animal studies provides compelling arguments for a beneficial effect of DHA against AD. Whereas epidemiological data are by definition correlative, those gathered in transgenic animals of AD strongly suggest a cause–effect relationship between high dietary intake of DHA and reduced markers of AD pathology in addition to providing potential mechanisms. As a consequence, we have witnessed in the recent year the launch of the first randomized clinical trials (RCT) on DHA in AD. Later in 2006, Freund-Levi et al. published a placebo-controlled RCT studying the effect of long-chain n-3 PUFA on the performance of patients on cognitive tests. They reported beneficial effects of n-3 PUFA on cognitive decline restricted to a small group of patients with very mild cognitive dysfunction, suggesting that n-3 PUFA administration might become less useful as the disease progresses into significant neuron loss [56]. This is consistent with earlier open-labeled clinical studies showing cognition-enhancing properties of DHA-enriched phosphatidylserine in older adults with MCI, but not as consistently in AD patients [57], [58]. More recently, the same group described beneficial neuropsychiatric impacts of n-3 PUFA intake depending on the apoE e4 allele status [59]. ApoE may be relevant to more global DHA effects given evidence for opposite effects on lipid profiles in ApoE2 vs. ApoE4 carriers [60]. In contrast to those with ApoE3, ApoE4 carriers have been reported to lack risk reduction from fish consumption [61]. However, in the Freund-Levi et al. trial, 77% of the patients treated with n-3 PUFA were ApoE4 carriers [59]. Overall, these investigations involving over 200 patients do not highlight important beneficial effects of n-3 PUFA against AD symptoms in patients with established disease using a DHA/EPA formulation, EPAX1050TG marketed by Pronova. The bottom line is that we still need RCTs with larger numbers of carefully selected patients to draw a conclusions on the efficacy of n-3 PUFA in AD and MCI. In the US and Europe, governments are supporting fish oil and DHA trials in patients with established AD. However, available data rather suggest the best hope for a strong positive trial is likely to require costly primary prevention studies in MCI patients lacking ApoE4. Indeed, the combined observations that (i) the beneficial symptomatic effects of DHA were restricted to patients with very MCI [56], (ii) the drebrin loss occurs early (MMSE 26) in the superior temporal cortex of individuals diagnosed with MCI [62], and (iii) drebrin levels were sensitive to dietary intake of DHA in animal models of AD [37] both advise for earlier intervention with DHA. Such preventive clinical trials are difficult to conduct mainly because n-3 PUFAs are non-patentable and, consequently, the private sector cannot support their development [63]. Indeed, the most likely way to boost the development of drugs based on reported DHA effects in AD will probably be the development of patentable formulations either based on DHA chemical derivatives or on genuine combinations of compounds [63], [64]. In this regard, studies in animal models of AD can be particularly helpful in deciphering the mechanisms of action of DHA that could lead to the development of new potent chemical entities replicating the neuroprotective actions of DHA. Furthermore, disease-modifying treatments are difficult to study in clinical trials but are easier to investigate in animal models because postmortem neuropathological signs of AD can be directly assessed for a large range of compounds.

Clearly, the available data on the potential neuroprotective effect of DHA does not provide sufficient argument for any evidence-based recommendation [33]. However, due to the unique nature of DHA in term of safety and availability, it would be unwise to completely ignore the information compiled in the recent years. Therefore, it seems reasonable to suggest that DHA deficiency should be avoided at least in people at high risk of developing AD due to very old age or genetic factors.

Section snippets

Acknowledgments

This work was supported by grants from the NIH (GMC), the Canadian Institutes of Health Research (FC), and the Alzheimer Society Canada (FC).

References (64)

  • F. Calon et al.

    Docosahexaenoic acid protects from dendritic pathology in an Alzheimer's disease mouse model

    Neuron

    (2004)
  • M. Oksman et al.

    Impact of different saturated fatty acid, polyunsaturated fatty acid and cholesterol containing diets on beta-amyloid accumulation in APP/PS1 transgenic mice

    Neurobiol. Dis.

    (2006)
  • M. Hashimoto et al.

    Chronic administration of docosahexaenoic acid ameliorates the impairment of spatial cognition learning ability in amyloid {beta}-infused rats

    J. Nutr.

    (2005)
  • H.Y. Kim

    Novel metabolism of docosahexaenoic acid in neural cells

    J. Biol. Chem.

    (2007)
  • M.A. McDaniel et al.

    Brain-specific nutrients: a memory cure?

    Nutrition

    (2003)
  • T. Itoh et al.

    Synthesis of docosahexaenoic acid derivatives designed as novel PPARgamma agonists and antidiabetic agents

    Bioorg. Med. Chem.

    (2006)
  • N. Salem

    Omega-3 fatty acids: molecular and biochemical aspects

  • K.P. Whittall et al.

    In vivo measurement of T2 distributions and water contents in normal human brain

    Magn. Reson. Med.

    (1997)
  • C. Galli et al.

    Lipid alterations and their reversion in the central nervous system of growing rats deficient in essential fatty acids

    Lipids

    (1971)
  • R.J. Pawlosky et al.

    Ethanol consumption alters electroretinograms and depletes neural tissues of docosahexaenoic acid in rhesus monkeys: nutritional consequences of a low n-3 fatty acid diet

    Alcohol Clin. Exp. Res.

    (2001)
  • C.H. Maclean, A.M. Issa, S.J. Newberry, W.A. Mojica, S.C. Morton, R.H. Garland, L.G. Hilton, S.B. Traina, P.G....
  • M.C. Morris et al.

    Consumption of fish and n-3 fatty acids and risk of incident Alzheimer disease

    Arch. Neurol.

    (2003)
  • P. Barberger-Gateau et al.

    Fish, meat, and risk of dementia: cohort study

    Br. Med. J.

    (2002)
  • M.J. Engelhart et al.

    Diet and risk of dementia: does fat matter?: the Rotterdam study

    Neurology

    (2002)
  • S. Kalmijn et al.

    Dietary intake of fatty acids and fish in relation to cognitive performance at middle age

    Neurology

    (2004)
  • J.A. Conquer et al.

    Fatty acid analysis of blood plasma of patients with Alzheimer's disease, other types of dementia, and cognitive impairment

    Lipids

    (2000)
  • A.M. Tully et al.

    Low serum cholesteryl ester-docosahexaenoic acid levels in Alzheimer's disease: a case–control study

    Br. J. Nutr.

    (2003)
  • E.J. Schaefer et al.

    Plasma phosphatidylcholine docosahexaenoic acid content and risk of dementia and Alzheimer disease: the Framingham heart study

    Arch. Neurol.

    (2006)
  • D. Laurin et al.

    Omega-3 fatty acids and risk of cognitive impairment and dementia

    J. Alzheimers Dis.

    (2003)
  • S.I. Rapoport

    In vivo fatty acid incorporation into brain phosholipids in relation to plasma availability, signal transduction and membrane remodeling

    J. Mol. Neurosci.

    (2001)
  • J.A. Hamilton et al.

    A model for fatty acid transport into the brain

    J. Mol. Neurosci.

    (2007)
  • H. Chen et al.

    Dietary intakes of fat and risk of Parkinson's disease

    Am. J. Epidemiol.

    (2003)
  • Cited by (0)

    View full text