In the Neuron paper, a collaboration led by David Holtzman at Washington University School of Medicine, St. Louis, Missouri, and Steven Paul at Lilly Research Laboratories, Indianapolis, report the effects of knocking out ApoE and/or clusterin in the PDAPP transgenic mouse model, which expresses mutant human Aβ precursor protein and develops amyloid deposits. The same group has used this model previously to study the effect of knocking out only clusterin (see ARF Stockholm news story). The present study recapitulates the earlier data and adds a new twist because double ApoE/ApoJ knockouts did not behave as predicted.

If knocking out either ApoE or ApoJ alone results in reduced deposition of immunoreactive Aβ, then knocking out both should be even better, right? Wrong. Joint first authors Ronald DeMattos, John Cirrito, and colleagues found that the double KOs actually accumulated Aβ deposits faster than did controls, and much faster than single KO mice. At six months, only the double knockouts had accumulated deposits, and at 12 months, the ApoE/ApoJ-negative mice had by far the greatest burden of all strains studied.

Trying to describe the nature of the Aβ, the authors next examined the immunoreactive deposits by staining them with thioflavin S, which distinguishes fibrillar amyloid deposits from soluble Aβ. The double knockouts had almost three times as many thioflavin-S-positive plaques than did control animals, which, in turn, had more plaques than animals with single knockouts of ApoE or ApoJ. What could this puzzling result mean? Are the lipoproteins protective or not?

The findings suggested to the authors that an effect distinct from that on fibrillogenesis per se was coming unmasked in the double knockouts. Perhaps apolipoproteins affect the dynamics of Aβ metabolism? When the authors used ELISAs to quantify soluble Aβ40 and 42 in hippocampus of three-month-old ApoE/ApoJ-negative animals, these levels were elevated by about 1.5- and twofold, respectively, compared to controls. Furthermore, the cerebrospinal fluid (CSF)-to-plasma ratio of Aβ40 was significantly higher in either double knockout or ApoJ-negative mice, indicating a possible role for this protein in clearing Aβ40 from the brain. Oddly enough, the ApoE knockout was the only one to have a statistically higher ratio of CSF-to-plasma Aβ42.

“The current study strongly demonstrates that ApoE and clusterin cooperatively suppress Aβ deposition,” write the authors. In support of this conclusion, they found that neither protein influences AβPP processing, leaving clearance the most likely avenue by which they can influence the dynamics of plaque formation.

In the PNAS paper, scientists including Holtzman, Paul, and Guojun Bu at Washington University, among others, report that LRP may also influence the dynamics of soluble Aβ metabolism. To test the role of LRP, first author Celina Zerbinatti and colleagues expressed a mini-version of the gene (mLRP) in PDAPP mice. mLRP contains the cytoplasmic tail, transmembrane domain, and second ligand-binding domain of the full-length protein, and has been shown to bind most of LRP's physiological ligands. In the cortex of 22-month-old mLRP mice, the authors found increased levels of soluble and insoluble Aβ. The increases were small (about 60 percent and 20 percent for soluble and insoluble, respectively), but statistically significant. In the hippocampus, an increase in only soluble Aβ was also found; however, the authors failed to find any changes in levels of Aβ plaques or fibrillar Aβ. But is the elevated Aβ physiologically significant?

To test this, the authors subjected mLRP mice to spatial learning and memory tasks. In two different maze tests, older mLRP mice performed poorly in comparison to their younger littermates and controls, but the results were, for the most part, not statistically significant. The authors did, however, find a significant relationship between learning and the amount of soluble Aβ in the hippocampus. Mice with more of the peptide needed to swim longer to find the hidden platform in a Morris water maze.

This data conflicts with a previous study showing that mice deficient in LRP were protected from Aβ deposition (see ARF related news story). However, Zerbinatti and colleagues suggest that this protection may have been afforded by loss of other lipoproteins because van Uden used knockouts of RAP, a lipoprotein chaperone, to ablate LRP.

Overall, the two latest papers strengthen the argument that apolipoproteins play a role in clearing Aβ from the brain, and that the dynamics of this process must be finely tuned to prevent learning and memory losses.—Tom Fagan.

References:
DeMattos RB, Cirrito JR, Parsadanian M, May PC, O'Dell MA, Taylor JW, Harmony JAK, Aronow BJ, Bales KR, Paul SM, Holtzman D. ApoE and clusterin cooperatively suppress Abeta levels and deposition: evidence that ApoE regulates extracellular Abeta metabolism in vivo. Neuron 2004 January 22;41:193-202. Abstract

Zerbinatti CV, Wozniak DF, Cirrito J, Cam JA, Osaka H, Bales KR, Zhuo M, Paul SM, Holtzman DM, Bu G. Increased soluble amyloid-{beta} peptide and memory deficits in amyloid model mice overexpressing the low-density lipoprotein receptor-related protein. PNAS 2004 January 27;101:1075-1080. Abstract