Amyloid-β plaque load poorly associates with cognition in Alzheimer’s disease (AD) and in people with mild cognitive impairment. Synapse loss, on the other hand, is believed to be a major correlate of cognitive decline. At this year’s annual meeting of the Society for Neuroscience, held 13-17 November 2010 in San Diego, California, several presentations and posters focused on links between synapse health and apolipoprotein E (ApoE), a major risk factor for late-onset AD. Researchers presented evidence that in both animal models and the human brain, ApoE conspires with Aβ oligomers to destroy synapses. The findings expand the modi operandi of ApoE, which already include promoting Aβ aggregation and preventing clearance.

Work from Tara Spires-Jones’s lab at Massachusetts General Hospital, Charlestown, pinpointed ApoE and oligomeric Aβ together in synapses in postmortem tissue samples from AD patients. In people who carried an ApoE4 gene, which is the major genetic risk factor for late-onset AD, oligomeric Aβ-positive synapses were small, and synapse number was down (see Part 1 of this series). The results suggest that the two proteins conspired to damage synapses, a theory supported by work from Karen Gylys and colleagues at the University of California, Los Angeles. Gylys also presented data on synaptic localization of ApoE. By ELISA and flow cytometry, Gylys and colleagues detected increased levels of ApoE in synaptosomes isolated from postmortem brain tissue of AD patients and of 20-month-old APP/PS1 transgenic rats (Flood et al., 2009) compared with aged-matched controls.

The team also found isoform-dependent differences in synaptic ApoE levels. Analyzing 18-month-old ApoE targeted replacement (TR) mice developed by Patrick Sullivan of Duke University, Durham, North Carolina, the UCLA researchers found that synaptic ApoE levels were highest and Aβ levels lowest in E2/4- compared to E3/3- and E4/4-expressing TR mice. In addition, Gylys reported that synaptic cholesterol levels were “dramatically increased in the E2/4 animals, with E3/3 and E4/4 levels similar to each other but much lower,” Gylys noted in an e-mail to ARF (see also Part 3 of this story). The data hint that less cholesterol is available in E3- and E4-containing synapses than in E2-containing synapses. “This suggests that poorly lipidated E4 may impair Aβ clearance, and that insufficient cholesterol delivery may be a central problem in AD synapses.” In a more general sense, she wrote, “the modest ApoE deficit in AD synapses likely contributes over time to synaptic dysfunction and loss, and suggests the relevance of ApoE- and lipid-related therapeutic targets.”

Gylys’s results complement “our array tomography data showing that synaptic ApoE is found at large, presumably healthy, synapses,” Spires-Jones noted. “We propose that ApoE alone is trophic for synapses, and only when it targets Aβ to synapses, as in the case of E4, is it a bad thing.”

On a separate SfN poster, coauthor Tadafumi Hashimoto of the Hyman group presented additional biochemical data suggesting that ApoE promotes formation of Aβ42 oligomers in an isoform-dependent manner. Using a luciferase assay that allows specific and quantitative measurement of Aβ oligomerization in the presence of different ApoE isoforms, the researchers found the highest levels of Aβ oligomers in E4-transfected HEK293 cells, less in the E3 transfectants, and the lowest levels of oligomeric Aβ in E2-containing cells, as reported earlier this year (see ARF related conference story). Now, Hashimoto shows that cells transfected with an ApoE4 mutant (R61T) that mimics E3 structure had Aβ oligomer levels comparable to E3-transfected cells and lower than in the E4 transfectants, offering further support that ApoE enhances Aβ oligomerization in an isoform-dependent fashion. Studies in HEK293 cells transfected with various ApoE fragments showed that the C-terminal lipid-binding region of ApoE is needed for facilitation of Aβ oligomer formation, Hashimoto and colleagues reported.

What could all of this mean for therapeutic approaches? At SfN, Martin Sadowski, New York University School of Medicine, and colleagues reported on a poster that blocking ApoE-Aβ oligomer interactions can stem Aβ-induced synaptic toxicity in vitro. Since the structure of ApoE’s Aβ docking site is unclear, the researchers targeted ApoE-Aβ interplay using a peptide homologous to the fragment of Aβ that binds ApoE, i.e., Aβ12-28. This peptide (Aβ12-28P) crosses the blood-brain barrier and reduces amyloid plaques and vascular Aβ burden in AD transgenic mice (Sadowski et al., 2006 and ARF related news story).

The new data presented at SfN show how the peptide works at the level of neurons and synapses. The NYU scientists added synthetic Aβ1-40 and Aβ1-42 monomers to co-cultures of astrocytes and hippocampal neurons from wild-type mice. In this in-vitro system, lipidated ApoE secreted by astrocytes promoted buildup of intraneuronal Aβ oligomers, as detected by Aβ oligomer-specific ELISA and by dot-blot densitometry using oligomer-specific antibody (A11). ApoE also impaired Aβ degradation, as determined by pulse-chase experiments, and led to downregulation of post-synaptic proteins involved in synaptic plasticity and memory formation—namely, the NR1 subunit of NMDA receptor, PSD-95, and synaptophysin. Addition of Aβ12-28P peptide curbed these effects.

The researchers repeated the experiments using co-cultures of neurons and astrocytes from targeted replacement mice producing specific human ApoE isoforms. The E4 isoform had the strongest effect on promoting intraneuronal Aβ accumulation, whereas E3 showed a modest effect, and E2, no effect. Importantly, Aβ12-28P abolished the E4-induced effects, suggesting the approach may hold promise in E4 carriers, which constitute nearly half of AD patients, Sadowski noted.

Ongoing studies in the Sadowski lab are focused on developing an Aβ12-28P derivative that could be used in human studies. In the best of worlds, the researchers hope to tease out an active structure within Aβ12-28P and convert it into a brain-penetrant drug that could be taken by mouth. They will use peptidomimetic technology to replace particular amino acids with non-natural mimics that resist degradation, and test potential compounds in the astrocytic/neuronal cell culture model used in their SfN research, Sadowski told ARF.—Esther Landhuis.

This is Part 2 of a three-part series. See also Part 1 and Part 3.


Make a Comment

To make a comment you must login or register.

Comments on this content

No Available Comments

Comments on Primary Papers for this Article

No Available Comments on Primary Papers for this Article


News Citations

  1. San Diego: Aβ Oligomers Seen, With ApoE, at Synapses of Human Brain
  2. San Diego: The Curious Case of ApoE, Cholesterol, and Cognition
  3. St. Louis: ApoE—A Clearer View of its Role In AD?
  4. ApoE—Breaking Ties to Aβ Offers Potential Therapy

Paper Citations

  1. . A transgenic rat model of Alzheimer's disease with extracellular Abeta deposition. Neurobiol Aging. 2009 Jul;30(7):1078-90. PubMed.
  2. . Blocking the apolipoprotein E/amyloid-beta interaction as a potential therapeutic approach for Alzheimer's disease. Proc Natl Acad Sci U S A. 2006 Dec 5;103(49):18787-92. PubMed.

Further Reading


  1. San Diego: What—3 Percent? Money Woes Trump Science at SfN
  2. San Diego: Tau Oligomer Antibodies Relieve Motor Deficits in Mice
  3. San Diego: Pilin’ on the Pyro, Aβ Going Rogue
  4. San Diego: ALS Research Goes to the Dogs
  5. San Diego: TDP-43 Targets Loom Large—But Where’s the Bull’s Eye?
  6. San Diego: A New Tack on Insulin-Based Therapies?
  7. San Diego: Aβ Oligomers Seen, With ApoE, at Synapses of Human Brain
  8. San Diego: Subcortical Blues—Locus Ceruleus in AD, Neurodegeneration