SUMMARY The continuing saga of anti-Ab antibody aducanumab has produced the first positive phase 3 outcome for Alzheimer's disease (AD) since memantine was approved in 2003. This promising, albeit controversial, result has breathed new life into the therapeutic potential for Ab-lowering strategies and legitimized the ongoing exploration of other means to chronically and safely mitigate Ab. Past work had shown that small peptide inhibitors can be readily tailored to prevent Ab aggregation, but in vivo delivery of peptide-based drugs was limited by short half-life and poor brain penetration. We have identified two Ab sequence variants that meet criteria for potential therapeutic use, as they 1) prevent aggregation of WT Ab in vitro, 2) promote disassembly of Ab existing fibrils, 3) mitigate toxicity of Ab oligomers, and importantly, 4) do not self-aggregate. To deliver these peptides in vivo, we have developed a novel mini-gene to express our variant peptides at the plasma membrane where they are released into the extracellular space by g-secretase cleavage. By packaging this minigene into an AAV vector that is injected into APP/PS1 mice, our pilot data show that viral expression of variant Ab lowers Ab load and delays plaque formation. The current proposal will build on these results through the following specific aims. First, we will decipher the biophysical mechanism of interactions between variant and wild type Aβ peptides. We will use analytical methods of CD spectroscopy, SEC chromatography, EM, and antibody profiling to define the structural mechanism by which our variants prevent/reverse aggregation of wild-type Ab. Second, we will determine how dosage, timing and route of variant Aβ administration influence efficacy in vivo. We will use viral strategies to compare interventional treatment after amyloid onset with preventative treatment starting at birth, determine the lowest effective ratio of variant:wild- type Ab needed to modify plaque formation and cognitive function, and test whether delivery through the CSF can match the effect of neuronal transduction. Third, we will interrogate the neuroimmune reaction to variant Aβ as an accomplice to plaque reduction. We will use histological and transcriptional profiling to assess whether a neuroimmune response to either the variant peptide or its AAV carrier contribute to plaque prevention in vivo, and if the neuroimmune response changes with age. Finally, we will test the potential for variant Aβ to slow AD aggregate seeding. We anticipate that variant Ab will slow seeding by AD Ab extracts, but will more importantly test whether variant Ab can assuage cross-seeding of tau in amyloid-bearing mice. If successful, this strategy for self-inhibition may also be applicable to other protein misfolding diseases where peptide treatments have been eschewed for technical reasons that can now be overcome through expression engineering and viral technology.