Examination of temporal lobe structures from Alzheimer patients reveals a specific cellular pattern of pathology of the subiculum of the hippocampal formation and layers II and IV of the entorhinal cortex. The affected cells are precisely those that interconnect the hippocampal formation with the association cortices, basal forebrain, thalamus, and hypothalamus, structures crucial to memory. This focal pattern of pathology isolates the hippocampal formation from much of its input and output and probably contributes to the memory disorder in Alzheimer patients.

An impairment of energy metabolism may underlie slow excitotoxic neuronal death in neurodegenerative diseases. We therefore examined the effects of intrastriatal, subacute systemic, or chronic systemic administration of the mitochondrial toxin 3-nitropropionic acid (3-NP) in rats. Following intrastriatal injection 3-NP produced dose-dependent striatal lesions. Neurochemical and histologic evaluation showed that markers of both spiny projection neurons (GABA, substance P, calbindin) and aspiny interneurons (somatostatin, neuropeptide Y, NADPH-diaphorase) were equally affected. Subacute systemic administration of 3-NP produced age-dependent bilateral striatal lesions with a similar neurochemical profile. However, in contrast to the intrastriatal injections, striatal dopaminergic afferent projections were spared. Both freeze-clamp measurements and chemical shift magnetic resonance spectroscopy showed that 3-NP impairs energy metabolism in the striatum in vivo. Microdialysis showed no increase in extracellular glutamate concentrations after systemic administration of 3-NP. The lesions produced by intrastriatal injection or systemic administration of 3-NP were blocked by prior decortication. However, the NMDA antagonist MK-801 did not block the effects of intrastriatal 3-NP, consistent with a non-NMDA excitotoxic mechanism. In contrast to subacute systemic administration of 3-NP, chronic (1 month) administration produced lesions confined to the striatum in which there was relative sparing of NADPH-diaphorase interneurons, consistent with an NMDA excitotoxic process. Chronic administration showed growth-related proliferative changes in dendrites of spiny neurons similar to changes in Huntington's disease (HD). These results are consistent with in vitro studies showing that mild metabolic compromise can selectively activate NMDA receptors while more severe compromise activates both NMDA and non-NMDA receptors. Chronic administration of 3-NP over 1 month produces selective striatal lesions that replicate many of the characteristic histologic and neurochemical features of HD.

These data suggest that APOE epsilon2 and epsilon4 might promote CAA-related hemorrhage through separate mechanisms: epsilon4 by enhancing amyloid deposition and epsilon2 by causing amyloid-laden vessels to undergo the vasculopathic changes that lead to rupture.

The mechanisms of delayed onset and cell death in Huntington's disease (HD) are unknown. One possibility is that a genetic defect in energy metabolism may result in slow excitotoxic neuronal death. Therefore, we examined the effects of age on striatal lesions produced by local administration of the mitochondrial toxin 3-nitropropionic acid in rats. In vivo chemical shift magnetic resonance imaging showed marked increases in striatal lactate concentrations that significantly correlated with increasing age. Histologic and neurochemical studies showed a striking age dependence of the lesions, with 4- and 12-month-old animals being much more susceptible than 1-month-old animals. Continuous systemic administration of low doses of 3-nitropropionic acid for 1 month resulted in striatal lesions showing growth-related changes in dendrites of striatal spiny neurons using the Golgi technique. These results show that a known mitochondrial toxin can produce selective axon-sparing striatal lesions showing both the age dependence and striatal spiny neuron dendritic changes that characterize HD.

Alz-50 is a monoclonal antibody directed against an antigen present in Alzheimer's disease. Unlike conventional pathological stains that reveal only neurons with neurofibrillary tangles (NFTs), Alz-50 recognizes neurons that contain NFTs and additional neurons that do not. Our investigation of the topographical distribution of both NFT-containing and non-NFT-containing neurons recognized by Alz-50 reveals that they are both found primarily in certain cytoarchitectural areas and lamina that have been established as consistent sites for NFT pathology. Some of the neurons recognized by Alz-50 appear histopathologically normal. We suggest that Alz-50 recognizes an antigen in neurons that both precedes and accompanies NFT formation. Thus, it may mark a point early in cellular pathology before irreversible cytoskeletal and degenerative changes occur.