• Principal Investigator: Patrick R. Hof, M.D.
• Co-Investigator: John H. Morrison, Ph.D.
• Co-Investigator: Cheuk Tang, Ph.D.

Much evidence points to tau- and amyloid-related cellular alterations as major contributing factors to the pathogenesis of Alzheimer's disease (AD). Although numerous studies of neuronal pathology have been carried out in the human brain, they have not been designed to assess directly the role these proteins play in the progression of degeneration in identified neuronal subpopulations that are vulnerable. This project investigates the process by which neurons die in AD, and tries to determine the point at which the death of these neurons leads to dementia. Four groups of human postmortem specimens from patients with differing levels of impairment are analyzed:

1. neurologically normal elderly cases,
2. cases with mild cognitive impairment and early AD,
3. cases with moderate dementia, and
4. severe AD cases.

We also study a "humanized" mouse model that expresses the human tau gene instead of the normal mouse tau gene, resulting in neuron death that models what occurs in AD.

We expect the early AD cases to emerge as a particularly interesting group of brains that will permit us to pinpoint the earliest changes in neocortical neurons that are known to be at risk to the degenerative process of AD. Particularly, a small subgroup of large neocortical neurons enriched in neurofilament proteins are the first to display dendritic changes that precede that stage at which pathologic tau proteins are accumulating and dementia becomes evident. These analyses focus on the prefrontal cortex, which is critically important to cognitive function and is affected severely and early in AD. We use quantitative techniques that reveal every detail of the neuron's structure, as well as the repercussions of tau- and amyloid-related pathology at the level of individual neuron morphology. The analyses in mice will permit us to follow the dynamic changes in live animals, obtain very high resolution magnetic resonance microscopy datasets prior to sectioning these specimens for detailed morphologic analyses, and provide quantitative analyses of neurons potentially at risk of degeneration with a much higher level of resolution than has been achievable to date. Altogether this project will provide a quantitative assessment, in AD cases of different severity, of the contribution of neuritic pathology and amyloid deposition to the progressive demise of selectively vulnerable neurons that are so critically important to normal cognitive function.

• Principal Investigator: Joseph Buxbaum, Ph.D.

The deposition of Beta-amyloid (ABeta) in the Alzheimer's brain has long been regarded as a potential causative agent in the progression of Alzheimer's disease (AD) pathology. Over the past 10 years, characterization of the biochemical events leading to ABeta generation has been the object of rigorous and exhaustive investigation. It is now understood that processing of the amyloid precursor protein (APP) by a group of enzymes known as the secretases ultimately underlies the liberation of ABeta. As more is known about these enzymes, it becomes increasingly clear that they likely regulate many important biochemical processes and inhibiting them may have multiple biological effects. Which intercellular signals regulate the cleavage of APP, i.e. the formation of ABEta? The processing of APP resembles that of other transmembrane proteins. A newly discovered protein, Notch, goes through similar processing that results in generation of an important fragment, the Notch Intracellular Domain (NICD). NICD is then translocated to the nucleus where it acts as a regulator of transcription. With Notch, it is the binding of specific small molecules that activates the secretases. It appears likely that specific activation of APP cleavage may be induced by the binding of a ligand to APP, much the same way the process works for Notch.

The specific aims are as follows:

1. Determine the identity of ligands that bind APP and regulate cleavage in a manner consistent with Notch processing;
2. Determine the potential for growth-factor receptors to regulate cleavage of APP;
3. Elucidate the potential molecular pathways through which APP ligands or growth-factor receptors regulate APP cleavage; and;
4. Determine the levels and distribution of first messenger pathways (including APP ligands and heterologous receptors and their ligands) that regulate APP processing and this could be important in AD.

• Principal Investigator: Anastasios Georgakopoulos, Ph.D.
• Co-Principal Investigator: Nikolaos Robakis, Ph.D.

Mutations in Presenilin-1 (PS1) are the most common cause of familial Alzheimer's disease (FAD). PS1 controls the g-secretase cleavage of many type I transmembrane proteins. EphrinB proteins are type I transmembrane proteins that function as ligands for the ephrinB receptors (EphBs). The ephrinB-EphB interactions regulate important cellular processes in development and adulthood including cell migration, axon guidance, angiogenesis and synaptic plasticity. This system participates in the regulation of two forms of long-term synaptic plasticity that are important for information storage in the brain, long-term potentiation (LTP) and the long-term depression (LTD), both of which are involved in learning and memory. We found that PS1 controls the processing of both ephrinB and ephB proteins by a g-secretase-like activity. Our data shows that ephrinB and ephB proteins are processed to produce a fragment that is subsequently cleaved by the PS1/g-secretase system to produce molecules that can affect gene expression patterns in neurons, as well as other biochemical processes. Some of these processes are involved in establishing synaptic contracts between neurons. Thus PS1 might work through the EphrinB/EphB system to maintain key circuits, and its disruption through mutation might be linked to the circuit breakdown evident in AD.