Modeling Alzheimer's disease and other proteopathies in vivo: is seeding the key?

Protein misfolding and aberrant polymerization are salient features of virtually all central neurodegenerative disorders, including Alzheimer's disease (AD), Parkinson's disease, triplet repeat disorders, tauopathies, and prion diseases. In many instances, a single amino acid change can predispose to disease by increasing the production and/or changing the biophysical properties of a specific protein. Possible pathogenic similarities among the cerebral proteopathies suggest that therapeutic agents interfering with the proteopathic cascade might be effective against a wide range of diseases. However, testing compounds preclinically will require disease-relevant animal models. Numerous transgenic mouse models of beta-amyloidosis, tauopathy, and other aspects of AD have now been produced, but none of the existing models fully recapitulates the pathology of AD. In an attempt to more faithfully replicate the human disease, we infused dilute AD-brain extracts into Tg2576 mice at 3-months of age (i.e. 5-6 months prior to the usual onset of beta-amyloid deposition). We found that intracerebral infusion of AD brain extracts results in: 1). Premature deposition of beta-amyloid in eight month-old, beta-amyloid precursor protein ( betaAPP)-transgenic mice (Kane et al., 2000); 2). augmented amyloid load in the injected hemisphere of 15 month-old transgenic mice; 3). evidence for the spread of pathology to other brain areas, possibly by neuronal transport mechanisms; and 4). tau hyperphosphorylation (but not neurofibrillary pathology) in axons passing through the injection site. The seeding of beta-amyloid in vivo by AD brain extracts suggests pathogenic similarities between beta-amyloidoses such as AD and other cerebral proteopathies such as the prionoses, and could provide a new model for studying the proteopathic cascade and its neuronal consequences in neurodegenerative diseases.