Deciphering the non-canonical function of the histone methyltransferase G9a in the etiology of AD

点击翻译按钮获取中文摘要

Abstract A prominent pathological feature of Alzheimer’s disease (AD) is extracellular accumulation of amyloid-beta (Aβ) peptides in neuritic plaques that activate microglia/macrophages. Using a quantitative proteomic approach we identified numerous proteins that showed increased secretion from Aβ-stimulated human monocyte-derived macrophages/microglia (hMDMs), including our newly characterized neurotoxic protein MMP9 that recapitulates neuritic tau beading for AD pathology. However, little is known about the AD-causing mechanisms underlying the translation and secretion of these neurotoxic proteins in Aβ-activated hMDMs. Clinicopathologic data show that AD onset is directly associated with an increased level of histone H3 lysine 9 dimethylation (H3K9me2) in prefrontal cortex tissue of AD patients. Consistent with this fact, enzymatic inhibition of the histone methyltransferases G9a that catalyze the increased H3K9me2 rescued synaptic and cognitive functions in AD mice, implicating constitutively active G9a and G9a-associated pathways in AD etiology. Our chemoproteomic approach with a biotinylated version of the same G9a inhibitor captured and identified G9a interactions with the N6-methyladenosine (m6A) RNA methylase METTL3 and other translation regulatory proteins in both Aβ-stimulated hMDMs and the hippocampus of AD mice. Considering that METTL3 promoted oncogene translation for cancer cell growth, our chemoproteomic discovery from Aβ-stimulated hMDMs indicated that, in addition to its canonical function in transcriptional silencing of ‘anti-AD’ genes, G9a activates translation of certain neurotoxic genes. Consequently, the objective of this project is to characterize this new translation regulatory function of G9a in AD etiology. We have found that, in the hMDMs with prolonged Aβ stimulation, (1) G9a showed a higher enzymatic activity and interacted with METTL3; (2) mRNAs of AD-related neurotoxic proteins were m6A-modified by METTL3; (3) depletion of G9a or METTL3 led to similarly reduced overexpression of these proteins, suggesting that G9a and METTL3 both work in the same translation regulatory pathways; (4) METTL3 is a non-histone substrate of G9a, and elimination of methylated lysines decreased METTL3 binding to translation initiation factor eIF3 that is otherwise critical for METTL3-mediated, cap-independent translation. We propose two Aims to test the central hypothesis that, via Aβ-induced interaction with METTL3, constitutively active G9a activates translation of certain neurotoxic inflammatory proteins whose secretion promotes AD. We will (1) Determine the molecular mechanism by which constitutively active G9a and METTL3 cooperate to activate the translation of Aβ-induced neurotoxic proteins, and (2) Determine how constitutively active G9a promotes METTL3-mediated translation of Aβ-induced neurotoxic proteins, which, in turn, contributes to AD pathogenesis. We will uncover this new non-canonical function of G9a in Aβ-induced, METTL3-mediated neurotoxic inflammation, a likely early-stage pathological mechanism that underlies AD etiology, for early therapeutic intervention of AD.