26 October 2012. How does apolipoprotein E (ApoE) boost risk for late-onset Alzheimer’s disease (LOAD)? While prior work suggests it accelerates amyloid-β deposition, a study in the October 24 Journal of Neuroscience examines how ApoE, the top LOAD risk gene, influences the formation of oligomers, widely seen as the most toxic forms of Aβ. Researchers led by Brad Hyman at Massachusetts General Hospital, Boston, show that ApoE purified from AD patient brains enhances Aβ oligomerization in an isoform-dependent manner, with E4 having the strongest effect. The data “add to the evidence that ApoE is a critical modulator of Aβ metabolism,” noted Cheryl Wellington of the University of British Columbia, Vancouver, Canada, who was not involved in the study.
Research has long indicated that ApoE4 drives up Aβ fibrillization. AD model mice that overexpress amyloid precursor protein (APP) develop fewer plaques if they lack endogenous ApoE (Holtzman et al., 2000; Fagan et al., 2002) or express just one copy (ARF related news story on Kim et al., 2011). Among late-onset AD patients, people with two copies of the E4 isoform rack up more brain Aβ than E3 homozygotes do (Rebeck et al., 1993; Gomez-Isla et al., 1996), and clear the peptide more slowly (Castellano et al., 2011). Furthermore, ApoE binds Aβ in vitro (Strittmatter et al., 1993; LaDu et al., 1994), as well as in human brain (Näslund et al., 1995) and cerebrospinal fluid (Strittmatter et al., 1993). Since research has fingered soluble multimeric forms of Aβ as being most dangerous to neurons (see, e.g., ARF related news story), scientists have wondered if and how ApoE might directly influence oligomerization of the peptide.
In the current study, first author Tadafumi Hashimoto and colleagues measured Aβ oligomers in postmortem brain extracts of AD patients who had various ApoE genotypes. They separated peptides in soluble fractions by size-exclusion chromatography and probed them with Aβ antibodies on immunoblots. Even when matched for amyloid burden, E4/E4 samples had nearly three times more oligomers than E3 homozygote brains did, and about seven times more than samples from E2 carriers.
To look at oligomer formation, the researchers purified ApoE lipid particles from immortalized mouse astrocytes expressing human ApoE2, ApoE3, or ApoE4, then mixed them with synthetic Aβ. Again, immunoblots showed oligomers forming most robustly in the ApoE4 samples and least in the presence of ApoE2. The team confirmed the E4->E3->E2 effect on oligomers formed from luciferase-tagged Aβ. In this quantitative assay, HEK293 cells were transfected with Aβ genes that had been fused with a gene for either the N- or C-terminal ends of luciferase. Formation of Aβ dimers or larger oligomers enabled cells to express functional luciferase and glow (Hashimoto et al., 2011, and ARF related news story). When introduced into these cells, lipidated ApoE from human brain also promoted Aβ oligomerization in an isoform-dependent manner.
Hashimoto and colleagues went on to determine that the ApoE’s C-terminal lipid-binding domain is necessary and sufficient for enhancing oligomerization. Plus, a mutation (R61T) that makes ApoE4 structurally more similar to ApoE3 brought Aβ oligomer levels down in the HEK293 split-luciferase assay, suggesting that differences in protein conformation may underlie ApoE’s effects.
A key strength of the study was its use of physiologically lipidated ApoE, “unlike innumerable previous studies that used either unlipidated ApoE or ApoE reconstituted with non-physiological lipids,” Wellington suggested (see full comment below). On the flipside, Joachim Herz of the University of Texas, Southwestern, Dallas, noted that HEK293-secreted ApoE particles are poorly lipidated (LaDu et al., 2006). In his view, the present data suggest that lipidation status matters little in ApoE’s enhancement of Aβ oligomerization, since similar effects were seen using lipidated and HEK293-derived forms of ApoE.
The situation seems more complex in vivo, though. In a recent study, AD mice accumulated less soluble brain Aβ and fared better cognitively if fed a cancer drug (bexarotene) that drives up ApoE production (ARF related news story on Cramer et al., 2012). The drug also increased ApoE lipidation, which apparently confers benefits that outweigh ApoE’s enhancement of Aβ deposition, Herz said. Compounds that raise ApoE lipidation have been reported to curb plaque formation in APP-overexpressing mice (ARF related news story on Jiang et al., 2008). Bexarotene had no effect on plaques. Taken together, the evidence to date suggests that ApoE4 drives up Aβ oligomerization, but whether it needs to be lipidated to achieve that remains unclear.—Esther Landhuis.
Hashimoto T, Serrano-Pozo A, Hori Y, Adams KW, Takeda S, Banerji AO, Mitani A, Joyner D, Thyssen DH, Bacskai BJ, Frosch MP, Spires-Jones TL, Finn MB, Holtzman DM, Hyman BT. Apolipoprotein E, Especially Apolipoprotein E4, Increases the Oligomerization of Amyloid β Peptide. J Neurosci. 24 Oct 2012;32(43):15181-92. Abstract