Exactly how the AD risk variant near the clusterin gene affects the clusterin protein, and how the allele contributes to risk, remains a puzzle. Some recent work implicates the variant in altered brain connectivity (see ARF related news story). Clusterin, also known as apolipoprotein J, bears similarities to ApoE, the most important genetic risk factor for sporadic AD. Some studies suggest that, like ApoE, clusterin modulates clearance of Aβ or other misfolded proteins from the brain (see, e.g., Nuutinen et al., 2007, and Wyatt et al., 2011).

To investigate clusterin’s effect on Aβ, Klenerman and colleagues approached the question from a molecular, quantitative level. First author Priyanka Narayan labeled synthetic Aβ40 monomers with two different-colored fluorophores, put them in solution at near physiological concentrations (30 nM), and allowed the peptides to aggregate into fibrils over several hours. The authors analyzed the resulting amyloid species using confocal two-color coincidence detection. In this technique, molecules pass one by one through a detector, with oligomers easily distinguished from monomers by their two-color signal. In the absence of clusterin, about 1 percent of the molecules formed oligomers, and oligomers were present in a range of sizes up to 50-mers, the authors report. Dimers were the most common species, with oligomer quantity falling off exponentially as the size increased.

The addition of human clusterin changed the behavior of the solution, however. Clusterin added in equimolar amounts with monomeric Aβ prevented oligomer formation, the authors report. In solutions in which Aβ was already partially aggregated, clusterin bound oligomers of all sizes and blocked further growth. Clusterin-Aβ complexes persisted for 50 hours or more. The authors also studied disaggregation, placing Aβ fibrils in buffer and observing the release of soluble particles using total internal reflection fluorescence microscopy, which allows scientists to see molecules deposited on a surface. Under these conditions, clusterin bound and stabilized oligomers released from fibrils. The upshot was more oligomers and fewer monomers, compared to solutions without clusterin. Although all the experiments were performed in vitro, Narayan and colleagues note that Aβ and clusterin are present in human cerebrospinal fluid (CSF) at concentrations that would allow these complexes to form. In addition, in-vivo Aβ clearance or endocytosis take about 15 hours (see Bateman et al., 2006; Mawuenyega et al., 2010; and Nielsen et al., 2009)—well less than the lifespan of the complexes—suggesting that clusterin could affect physiological processes.

How would that change AD pathology? One possibility is that by encasing oligomers, clusterin prevents them from interacting with cell membranes or other molecules, Klenerman told ARF. Since oligomers are widely believed to be the cytotoxic form of Aβ, this implies clusterin could be neuroprotective. However, existing in-vivo data are mixed on this point. Work led by David Holtzman at Washington University School of Medicine, St. Louis, Missouri, showed that PDAPP AD mice crossed with clusterin knockouts deposit fewer Aβ fibrils and have healthier neurites than those with clusterin, suggesting the protein exacerbates pathology (see ARF related news story on DeMattos et al., 2002). Complicating the picture, however, Holtzman and colleagues found that AD mice lacking both clusterin and ApoE accumulate amyloid deposits faster than control mice, implying the two chaperones may cooperate to clear Aβ (see ARF related news story on DeMattos et al., 2004). To add to the confusion, recent work from Holtzman’s group belies the idea that ApoE clears plaques, suggesting it slows clearance, instead (see ARF related news story). Together, these studies suggest that clusterin’s effect varies with the cellular context.

“One cannot do in vivo the type of beautiful in-vitro assessment of oligomers as was done in the Narayan paper, so we do not know what the status of oligomers is in the presence or absence of clusterin in vivo,” Holtzman wrote to ARF. “Further studies will be required to sort out whether human clusterin is actually promoting or inhibiting toxic forms of Aβ in vivo.”

The effect of clusterin may depend on the relative ratio of the protein to Aβ, Klenerman noted. In previous work, coauthor Wilson reported that when Aβ is present at 500 times the molar concentration of clusterin, the latter increases amyloid toxicity. However, at Aβ concentrations only 10 times that of clusterin, the chaperone decreases toxicity (see Yerbury et al., 2007). Small amounts of clusterin may stabilize oligomers, but not completely encase them, allowing the molecules to exert toxic effects, Klenerman speculated. AD model mice typically overexpress Aβ, but in human CSF, clusterin concentration exceeds that of Aβ40 by several-fold, the authors point out.

Klenerman and colleagues are currently investigating how clusterin-oligomer complexes affect primary neuronal cultures, which may shed some light on in-vivo effects. They are also studying clusterin’s interaction with Aβ42, which aggregates more readily than Aβ40. In addition, they plan to examine the clusterin variant linked to AD to see if they can discover differences between it and the protective variant.—Madolyn Bowman Rogers.

Reference:
Narayan P, Orte A, Clarke RW, Bolognesi B, Hook S, Ganzinger KA, Meehan S, Wilson MR, Dobson CM, Klenerman D. The extracellular chaperone clusterin sequesters oligomeric forms of the amyloid-β(1-40) peptide. Nat Struct Mol Biol. 2011 Dec 18. Abstract