Probing the in vivo economy of amyloid beta-protein during the development of Alzheimer's disease-type pathology.

摘要

Despite intense therapeutic and diagnostic focus on dyshomeostasis of amyloid beta-peptide (Abeta) in Alzheimer's disease (AD), we still lack insight into the in vivo economy of Abeta in the normal and diseased brain. Thus, my thesis research focused on understanding the dynamics of Abeta in the living brain during the development of AD-type pathology. Using in vivo microdialysis, I showed that the steady-state level of Abeta that remains diffusible in the hippocampal interstitial fluid (ISF) of awake, behaving hAPP transgenic mice falls as Abeta steadily accumulates in the brain parenchyma. In accord, I observed distinct dispositions of microinjected radiolabeled Abeta in plaque-rich versus plaque-free mice, suggesting that cerebral amyloid deposits rapidly sequester newly released Abeta. This provides the first in vivo evidence from controlled animal experiments for the hypothesis that soluble Abeta42 in human cerebrospinal fluid (CSF) falls in AD because it is sequestered into insoluble parenchymal deposits as the disease develops. My data further show that the association of Abeta with insoluble parenchymal deposits is not irreversible, as acute inhibition of gamma-secretase in plaque-rich mice failed to lower ISF Abeta42, whereas it did in plaque-free mice. Hence, the ISF in plaque-rich mice seems to be a reservoir for both newly produced Abeta and Abeta that diffuse off of cell membrane- and plaque-bound deposits.;Finally, I showed that Abeta dimers, which are known to be potent synaptic neurotoxins, are undetectable in the aqueous compartments of the central nervous system, i.e., the brain ISF and CSF, in hAPP transgenic mice. Acute injection of Abeta dimers into living wild-type mice showed a rapid sequestration of the dimers away from the hippocampal ISF pool and a higher recovery in the membrane-bound pool than in the cytosolic pool of the brain homogenates. Interestingly, I found that the Abeta recovered in the membrane-bound pool was tightly associated with endogenous GM1 ganglioside. Taken together, my results suggest that Abeta dimers, and probably higher oligomers, are rapidly sequestered away from the ISF and bind to GM1 ganglioside-enriched lipid membranes, such as raft-like microdomains of secreted vesicles or on the plasma membranes of neurons and other cells.