Alternative Theory of AD Pathogenesis: Membrane Delimitation of the Histone Acetyltransferase Tip60/Kat5

Abstract Background For decades, mounting genetic and physiological data have pointed to the APP proteolysis as a seminal event in AD pathogenesis. This data has been viewed as supporting the amyloid hypothesis. However, growing data from multiple disciplines disputes that amyloid is a causal mechanism of neurodegeneration. Evidence suggests that FAD mutations are loss of function, inhibiting normal proteolytic processing of APP. APP extracellular and intracellular domains exhibit a broad array of binding interactions. Tip60 (a histone acetyltransferase, also known as Kat5) is an intracellular binding partner that supports the long‐standing speculation that APP functions as a signaling receptor. Recently, a conditional knockout (cKO) of Tip60 demonstrated neurodegenerative pathology and massive reactive gliosis, supporting a role for Tip60 in neurodegeneration. Reports that Tip60 is restricted from the nucleus in AD patient brains bolsters this idea. Method To assess the overlap of the cKO Tip60 model and human AD pathogenesis, we compared the dysregulated gene sets (DEGs) in the cKO Tip60 mouse and late‐onset AD postmortem brains. We break DEGs from mouse and human models into AD “biological domains” (BDs), and perform GSEA of GO enrichment within each BD. We then perform downstream analyses using network reconstruction applications and by examining the enriched BDs with pseudotime measures of disease progression in LOAD patients. Result We find significant overlap between DEGs in LOAD patients and the Tip60 cKO in multiple BDs, including, mitochondrial metabolism, immune response, structural stabilization, synaptic function, and endolysosomal trafficking. GSEA of LOAD and cKO Tip60 further implicates similar deficits in both models. The network analysis shows a convergence of gene dysregulation at core nodes, further implicating similar processes of dysregulation may impact particularly significant genes in human and mouse models. The pseudotime analysis shows that this dysfunction is an early signature of AD, suggesting Tip60 dysregulation may trigger early‐stage homeostatic destabilization. Conclusion We propose that aberrant APP proteolysis traps Tip60 at the membrane, preventing the nuclear localization of Tip60, and altering both positive and negative gene regulation that is necessary for neuronal and glial homeostasis, and hypothesize that this is an explanation for the genetic linkage of APP proteolytic processing and AD.