A new study adds support to the theory that a critical first step in the production of the Aβ peptide, i.e., the cleavage of amyloid precursor protein (AβPP) by the BACE enzyme, occurs primarily in specialized, cholesterol-rich areas of the plasma membrane called "lipid rafts." Anthony Turner, Joanna Cordy, and colleagues of the University of Leeds and GlaxoSmithKline Research in Essex, United Kingdom, report in the 30 September issue of PNAS that they were able to boost production of Aβ by directing most of the BACE in the plasma membrane to the lipid rafts.

These findings dovetail with various lines of recent evidence that decreasing cholesterol with statins decreases AD risk, and that increased cholesterol levels boost the production of Aβ in brain (see ARF related news story, and especially the extended comment by Ben Wolozin). Lipid rafts-or more officially, detergent-insoluble glycolipid-enriched membrane domains (DIGs)-differ from the rest of the cell membrane in that their concentrations of cholesterol, glycosphingolipids, and ganglioside lipids are particularly high (see ARF related news story). The rafts act as platforms for an ever-growing number of cellular signaling and trafficking processes.

Interest in rafts in the context of AD arose several years ago, when a number of studies found Aβ, BACE, APP, and PS-1 in the rafts Bouillot et al., 1996; Lee et al., 1998; Simons et al., 1998; Riddell et al., 2001). Reducing cholesterol levels decreases the number of lipid rafts; however, the vast majority of the BACE or APP in the membrane is not found in the rafts. Kai Simons and his colleagues have proposed a model whereby normal membrane conditions favor APP and BACE floating about in the membrane, where BACE is nudged out of the running by the preferred cleavage path of APP, i.e., first by α-secretase and then by γ-secretase, leading to nontoxic peptide products. When age-related changes in membrane conditions (e.g., more cholesterol) increase the number of rafts, APP and BACE find themselves together-and away from the α-secretase-more often, and hence produce more toxic Aβ.

In support of this model, Simons and colleagues recently demonstrated that when APP and BACE are bound together with antibodies in N2A cells, the production of Aβ increases (Ehehalt, et al., 2003). This appeared to be occurring in the rafts, and reducing cholesterol also abolished the Aβ increase. In their current PNAS paper, Turner and colleagues provide further inferential evidence for rafts as the site of the BACE-APP communion in neuroblastoma cells. The researchers anchored BACE to the rafts and substantially boosted the production of Aβ. Again, cholesterol reduction abolished the increase.

In their discussion, the authors bring up the question of the increased incidence of AD with aging. Because cholesterol levels tend to rise during aging, they write, it is conceivable that this increases the opportunities for BACE to interact with APP in lipid rafts, thereby increasing the brain burden of Aβ.—Hakon Heimer


  1. Despite the problems with chemical tractability, β-secretase (BACE) remains one of the most promising targets for the treatment of Alzheimer's disease. It has been known for some time that APP, BACE, and γ-secretase (presenilins and associated proteins) all localize, at least partially, to membrane domains enriched in cholesterol called rafts. This is of interest because the statins, blockbuster drugs that lower cholesterol, inhibit the generation of Aβ peptides in cultured cells. Kai Simons and coworkers have proposed that APP inside rafts may be "bad" because it is mainly cleaved by β-secretase, whereas APP outside rafts may be "good" because it is mainly processed by α-secretase. Cordy et al. provide additional evidence for this hypothesis. They demonstrate an increase of Aβ secretion upon transfection of cells with a BACE construct targeted to rafts by an artificial GPI anchor. Lovastatin treatment abolishes this increase. The current study was performed in a neuroblastoma cell line, but the effects of protein sorting may be even more severe in a mature neuron where transport has to occur over very long distances. Hence, it would be interesting to study the effect of the GPI-BACE in a transgenic mouse.

  2. I think this issue is worth exploring, but I am concerned that the ability of a GPI-BACE chimera to cleave APP may be too artificial a model to tell us much about how the authentic BACE behaves. GPI-linked peptides would be expected to partition in rafts or raft-like domains. The studies of the Simons group also claim that an endocytosis reaction is important for BACE function, a point not considered by Cordy et al.

  3. Increasing evidence from biochemical, epidemiological, and genetic studies links cholesterol levels to Aβ production and the onset of AD pathology. Total cellular cholesterol can regulate Aβ generation, as evidenced by decreased Aβ levels in cells treated with statins, which inhibit cholesterol intake and synthesis. Intracellular distribution and/or trafficking of cholesterol are also gaining more and more support as potentially valid targets for indirect antiamyloid therapy. Altered cholesterol trafficking through the Niemann-Pick C1 compartment and altered cholesterol distribution into cholesteryl-ester droplets and lipid rafts, all affect APP processing into Aβ. Anthony Turner, Joanna Cordy and colleagues show that targeting BACE1 to lipid rafts is sufficient to elevate secreted Aβ levels, further strengthening the relevance of intracellular compartmentation of cholesterol in Aβ generation. Although Kai Simons and colleagues, Kumar Sambamurti and colleagues, and others have previously shown that lipid rafts are important players in at least BACE1-mediated processing of APP, the new study by Cordy et al. elegantly employs recombinant GPI-anchored BACE1 to show that mislocalization of BACE1 alone to lipid rafts is sufficient to elevate Aβ generation in a neuroblasoma cell line. Whether this is a saturable increase or all APP contained in lipid rafts is necessarily processed by BACE1 remains to be established. Interestingly, APP695 was shown in this study to be a better substrate for GPI-anchored BACE1 processing than APP751. The authors attribute this effect to differential compartmentation of the APP isoforms, but it would also be interesting to speculate on potential activities of the KPI/OX-2 domains in lipid rafts.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Rotterdam Study Questions Links between Fat and Dementia Risk
  2. Membrane Rafts Revealed

Paper Citations

  1. . Axonal amyloid precursor protein expressed by neurons in vitro is present in a membrane fraction with caveolae-like properties. J Biol Chem. 1996 Mar 29;271(13):7640-4. PubMed.
  2. . A detergent-insoluble membrane compartment contains A beta in vivo. Nat Med. 1998 Jun;4(6):730-4. PubMed.
  3. . Cholesterol depletion inhibits the generation of beta-amyloid in hippocampal neurons. Proc Natl Acad Sci U S A. 1998 May 26;95(11):6460-4. PubMed.
  4. . Compartmentalization of beta-secretase (Asp2) into low-buoyant density, noncaveolar lipid rafts. Curr Biol. 2001 Aug 21;11(16):1288-93. PubMed.
  5. . Amyloidogenic processing of the Alzheimer beta-amyloid precursor protein depends on lipid rafts. J Cell Biol. 2003 Jan 6;160(1):113-23. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Exclusively targeting beta-secretase to lipid rafts by GPI-anchor addition up-regulates beta-site processing of the amyloid precursor protein. Proc Natl Acad Sci U S A. 2003 Sep 30;100(20):11735-40. PubMed.