Mary J. Savage

Characterization of plasma β-secretase (BACE1) activity and soluble amyloid precursor proteins as potential biomarkers for Alzheimer's disease

Reduction in cerebrospinal fluid (CSF) amyloid β42 (Aβ42) and elevation in total tau and phospho‐thr181 tau consistently differentiate between Alzheimers disease (AD) and age‐matched control subjects. In contrast, CSF β‐site APP‐cleaving enzyme activity (BACE1) and soluble amyloid precursor proteins α and β (sAPPα and sAPPβ) are without consistent patterns in AD subjects. Plasma sampling is much easier, with fewer side effects, and is readily applied in primary care centers, so we have developed and validated novel plasma BACE activity, sAPPβ, and sAPPα assays and investigated their ability to distinguish AD from age‐matched controls. Plasma BACE activity assay was sensitive and specific, with signal being immunodepleted with a specific BACE1 antibody and inhibited with a BACE1‐specific inhibitor. Plasma sAPPβ and sAPPα assays were specific, with signal diluting linearly, immunodepleted with specific antibodies, and at background levels in APP knockout mice. In rhesus monkeys, BACE1 but not γ‐secretase inhibitor led to significant lowering of plasma sAPPβ with concurrent elevation of plasma sAPPα. AD subjects showed a significant increase in plasma BACE1 activity, sAPPβ, sAPPα, and Aβ42 (P < 0.001) compared with age‐matched controls. In conclusion, plasma BACE activity and sAPP endpoints provide novel investigative biomarkers for AD diagnosis and potential pharmacodynamic biomarkers for secretase inhibitor studies.

Decrease in brain soluble amyloid precursor protein β (sAPPβ) in Alzheimer's disease cortex

Amyloid‐β peptide (Aβ) is generated by sequential cleavage of the amyloid precursor protein (APP) by β‐site amyloid precursor protein cleaving enzyme 1 (β‐secretase, or BACE1) and γ‐secretase. Several reports demonstrate increased BACE1 enzymatic activity in brain and cerebrospinal fluid (CSF) from Alzheimers disease (AD) subjects, suggesting that an increase in BACE1‐mediated cleavage of APP drives amyloid pathophysiology in AD. BACE1 cleavage of APP leads to the generation of a secreted N‐terminal fragment of APP (sAPPβ). To relate BACE1 activity better to endogenous APP processing in AD and control brains, we have directly measured brain sAPPβ levels using a novel APP β‐site specific enzyme‐linked immunosorbent assay. We demonstrate a significant reduction in brain cortical sAPPβ levels in AD compared with control subjects. In the same brain samples, BACE1 activity was unchanged, full‐length APP and sAPPα levels were significantly reduced, and Aβ peptides were significantly elevated. In conclusion, a reduction in cortical brain sAPPβ together with unchanged BACE1 activity suggests that this is due to reduced full‐length APP substrate in late‐stage AD subjects. These results highlight the need for multiparameter analysis of the amyloidogenic process to understand better AD pathophysiology in early vs. late‐stage AD.

Induction of Alzheimer's-like changes in brain of mice expressing mutant APP fed excess methionine.

J. Neurochem. (2011) 116, 82–92.

Exploring the effect of combined β- & γ-secretase inhibitor treatment on βAPP processing in a transfected SHSY5Y cell line

Background: Inhibition of bor g-secretase as a means of reducing amyloid b (Ab) levels to slow progression of Alzheimer’s disease (AD) has been a primary focus of many drug discovery efforts. The challenge for b-secretase (BACE) has been to design a small molecule capable of crossing the blood-brain barrier with sufficient potency to block the enzyme’s large catalytic pocket while for g-secretase it has been has been to balance Ab-lowering efficacy with selectivity against Notch inhibition. Given these challenges, we examined whether partially inhibiting both enzymes simultaneously could achieve Ab reduction comparable to maximal doses of each inhibitor individually. Methods: Using SHSY5Y cells over-expressing bAPP with a mutation favoring BACE cleavage, a study was designed to compare, at multiple doses, the individual effects of the BACE inhibitor, Merck 3 and a previously published gsecretase (GSI) inhibitor with their additive effects on bAPP processing. Following IC50 determinations, a matrix experiment was conducted to compare levels of secreted bAPP analytes at concentrations above, at and below the IC50 of each inhibitor as well as vehicle control. Results: At the GSI’s IC50, increasing concentrations of the BACE inhibitor significantly reduced the secretion of both Ab species while increasing sAPPa secretion to a level greater than BACE inhibition alone across several doses. This effect was also observed at the highest, but not lowest GSI dose. There was no additive effect on the secretion of sAPPb, as the extent of dose-dependent inhibition driven by the BACE inhibitor was equivalent, with or without g-secretase inhibition. Conclusions: These results support further consideration of combined therapy utilizing g-secretase and BACE inhibitors as a possible treatment modality for AD. It is possible that other BACE inhibitors in combination with a sub-optimal Ab-lowering and notch-sparing GSI dose could achieve comparable efficacy observed with a maximal dose of a GSI alone. Secretase inhibitors co-administered in low doses not only may reduce Ab production sufficiently to ameliorate the cascade of events triggered by its aggregation but also considerably boost sAPPa levels which could have a profound effect on the preservation of neuronal and synaptic viability and integrity.