A small pilot study led by Suzanne Craft, Wake Forest University School of Medicine, Winston-Salem, North Carolina, hints that a diet low in simple sugars and saturated fats reduces toxic forms of Aβ in the cerebrospinal fluid (CSF). The findings appeared online June 17 in JAMA Neurology and support prior work from numerous labs suggesting a possible link between a healthy diet and risk for AD (see ARF related news story and ARF news story).

“Although the study is preliminary, it is a noteworthy effort to help translate findings from large-scale epidemiologic studies into practical effects on disease outcomes,” wrote Deborah Blacker, Harvard School of Public Health, Boston, Massachusetts, in an accompanying editorial.

Two years ago, Craft and colleagues reported that a four-week diet low in simple sugars and saturated fats either reduced or raised CSF Aβ42 levels, depending on whether the person was cognitively normal or had mild cognitive impairment (see ARF related news story). In the current study, the researchers, including first author Angela Hanson, University of Washington School of Medicine, Seattle, conducted additional assays on those same CSF samples to see whether the fraction of Aβ that is free of lipoproteins (lipid-depleted, or LD-Aβ) changed with diet. The scientists took ApoE4 status into account. Previous studies had suggested that LD-Aβ aggregates more quickly than peptides associated with lipoproteins (see Ly et al., 2013).

To separate the lipid-rich from the lipid-poor Aβ, the researchers added potassium bromide to the CSF, then centrifuged it at high speed. They analyzed samples from 27 people with amnestic MCI (aMCI) and 20 age-matched, cognitively healthy controls. They found that at baseline, people with aMCI, especially ApoE4 carriers, had more LD-Aβ than did cognitively normal people. In all volunteers a low sugar/low saturated fat diet reduced LD-Aβ over four weeks, whereas a diet high in those dietary components boosted LD-Aβ levels.

Hanson and colleagues suggest a model whereby lipidated ApoE binds to Aβ, shuttling it away to be degraded, while lipid-depleted Aβ remains free to oligomerize and become neurotoxic. In keeping with this idea, they found that in people carrying the ε4 variant of ApoE, a major genetic risk factor for AD, lipid-depleted ApoE in the CSF was almost threefold higher than in people without ApoE4 but with at least one copy of the protective ApoE2 allele. Curiously, while the low sugar/low fat diet improved Aβ lipidation, it made no difference to the amount of lipid-depleted ApoE in the CSF.

The authors acknowledge that the size of the sample population was small, and that since they manipulated both sugar and fat, they cannot tell if one or both causes the effects. Craft said her group is conducting a larger study with about 150 older adults to verify the pilot results. They have made the diet more extreme in terms of glycemic index and fat content, and have differentiated between good and bad saturated fats.

Mary Jo LaDu and Leon Tai, University of Illinois at Chicago, pointed out that this is the first study to look at the effect of diet on Aβ lipidation in humans. However, they raised several concerns. For one thing, the centrifuge method used to isolate the lipid-depleted fraction knocks proteins off lipids, so the authors may have underestimated the amount of lipidated protein. Other researchers contacted by Alzforum raised the same issue. Hanson said that she used the same method at both baseline and at the end of the intervention, which should control for limitations in this method. LaDu and Tai also wondered how much total Aβ was isolated from these individuals' CSF and how much appeared in the lipid fraction, as Hanson and colleagues reported only non-lipidated Aβ levels.

Other commentators noted that the authors did not report total CSF ApoE, which could be influenced by diet, nor did they attempt to analyze different lipid states of the protein. One researcher suggested that biochemical analysis of the ApoE recovered in the non-lipid fraction would enable a proper interpretation of the findings. Some claimed that almost all Aβ in the CSF is free from lipoproteins under normal conditions. In relying on Aβ binding to ApoE, the model reported here is inconsistent with a recent, high-profile study arguing that in physiological conditions, this interaction does not occur to a significant extent (see ARF related news story).—Gwyneth Dickey Zakaib


  1. This is an interesting study showing that both soluble Aβ and apolipoprotein E (ApoE) exist in lipid-depleted cerebrospinal fluid (CSF), and that ApoE4 is linked to cognitive impairment. The results fit well with what we have hypothesized for 20 years: Patients with Alzheimer’s disease bear disease-related metabolic conditions in the central nervous system (CNS), where interaction between lipoproteins and soluble Aβ is impaired, leading to Aβ assembly. According to our experiments, the conversion of lipoprotein-free monomeric soluble Aβ42 into oligomers preferentially occurs in AD CSF. Furthermore, the accumulation of de-lipidated soluble 12-mers precedes the appearance of neuronal loss or cognitive impairment and is enhanced as the Braak neurofibrillary tangle stages progress, confirming that the entorhinal cortex of AD patients establishes metabolic conditions that accelerate Aβ assembly. Although direct evidence is not available in the current paper, the authors propose that lipoprotein-free ApoE may preferentially interact with lipoprotein-free soluble Aβ, inducing and/or maintaining an abnormal β-sheet conformation that initiates a cascade favoring Aβ assembly in an ApoE isoform-dependent manner.

    Another important finding is that diet can improve the above-mentioned risk environment in the brain via the modulation of insulin, which regulates cholesterol metabolism in the CNS. Since cholesterol metabolism in the CNS is quite different from that in systemic circulation, therapeutic intervention for disease-related lipidic environments via insulin appears sound.

    View all comments by Etsuro Matsubara

Make a Comment

To make a comment you must login or register.


News Citations

  1. Pass the Bordeaux—Mediterranean Diet, Exercise Reduce AD Risk
  2. Mediterranean Diet Slims Down Risk of AD
  3. ApoE Does Not Bind Aβ, Competes for Clearance

Paper Citations

  1. . Binding of Apolipoprotein E Inhibits the Oligomer Growth of Amyloid-β Peptide in Solution as Determined by Fluorescence Cross-correlation Spectroscopy. J Biol Chem. 2013 Apr 26;288(17):11628-35. PubMed.

Further Reading


  1. . Liver X receptor agonist treatment ameliorates amyloid pathology and memory deficits caused by high-fat diet in APP23 mice. J Neurosci. 2010 May 19;30(20):6862-72. PubMed.
  2. . Binding of Apolipoprotein E Inhibits the Oligomer Growth of Amyloid-β Peptide in Solution as Determined by Fluorescence Cross-correlation Spectroscopy. J Biol Chem. 2013 Apr 26;288(17):11628-35. PubMed.
  3. . Diet intervention and cerebrospinal fluid biomarkers in amnestic mild cognitive impairment. Arch Neurol. 2011 Jun;68(6):743-52. PubMed.

Primary Papers

  1. . Food for Thought. JAMA Neurol. 2013 Jun 17;:1-3. PubMed.
  2. . Effect of apolipoprotein e genotype and diet on apolipoprotein e lipidation and amyloid peptides: randomized clinical trial. JAMA Neurol. 2013 Aug 1;70(8):972-80. PubMed.