Alzheimer's disease (AD) affects over 5.4 million Americans. AD is believed to result in whole or in part by accumulation of the toxic amyloid-β peptide (Aβ), making its generation a candidate for treatment. Aβ is generated from the Aβ-precursor protein (APP) via cleavage with β-site APP cleaving enzyme 1 (BACE1), followed by γ-secretase complex cleavage. BACE1 activity is the rate-limiting step in production of Aβ. Recent human trials of BACE1-targeting drugs were discontinued due to liver toxicity. This has been explained by invoking “off-target effects”. We disagree: Such “failed” drugs may have been molecularly “on-target” but acting “off-site” (OnTOS). Specifically for BACE1: In brain, the primary substrate is APP, but in liver, the primary substrate is α-2,6-sialyltransferase (ST6Gal1). ST6Gal1 is an important enzyme in the body's defense against radiation and oxidative damage. Interference in its processing may have produced hepatotoxicity similar to that of the “failed” anti-BACE1 drug trials. Animal model-based tests may miss such risks because they often use transgenic knockout (KO) animals, which may have compensatory mechanisms to lacking BACE1 and potentially important uncharacterized artifacts. These animals also have such short lifespans under sheltered conditions. Radiation and oxidative load will not approach that faced by “wildype (WT) and in the field” humans. Aim 1: Measure levels of BACE1 and orthologs and of APP and ST6Gal1 processing in brain, kidney, and liver samples from BACE1 KO and WT mice treated with vehicle or BACE1 inhibitors (LY2811376 and MK8931) for potential compensating effects in KO vs. drugs. We will use Westerns for total APP and processing of ST6Gal1 and four other BACE1 substrates (distinguished by band migration) and ELISA for soluble APPβ. Impact: Establish empirical multi-organ comparisons to demonstrate OnTOS as an “induced lack” (drug) effect vs. “congenital” (KO) effect. Aim 2: Determine parallels between mouse models and human samples to lay groundwork for human medical studies. We will obtain at least 15 donor-matched samples of human brain, liver and kidney from AD and non-AD subjects. Proteins and miRNA will be measured as in SA1. We will employ western or quantitative ELISA as in 1-1. RNA-enriched extracts to be used for RT-qPCR of different miRNAs. Impact: Any medical research requires translation to humans. The proposed experiments will lay groundwork for such translation. Aim 3: Measure levels of different miRNAs implicated in regulating BACE1 levels. This will provide a frame of reference for future work to take advantage of different endogenous levels of miRNAs to sidestep OnTOS. We will then validate potentially useful miRNA species in primary mouse brain and liver cultures and in iPSC-derived human brain and liver cultures. Impact: The knowledge generated will provide concrete, mechanistic explanations of a pervasive problem in drug design—ignoring OnTOS. It may also suggest a viable alt...