Mitochondrial dysfunction is one of the earliest hallmarks of Alzheimer’s disease (AD), but the association of amyloid-β (Aβ) toxicity with mitochondrial dysfunction remains largely elusive. Here, we explored early-stage AD pathogenic mechanisms in APP/PS1 mice and neuronal cells, and identified malic enzyme 3 (Me3) as a critical bridge linking Aβ aggregation to mitochondrial dysfunction. Me3 involved in the tricarboxylic acid (TCA) cycle was consistently upregulated during super early (month 3) and early stage (month 6) of AD. The upregulation of Me3 served as a compensation response to the TCA cycle disruption by Aβ aggregation. Under pathological conditions, Me3 aggregated in mitochondria and co-localized with Aβ aggregates, leading to excessive reactive oxygen species (ROS) production and mitophagy blockage. The hydrogen-deuterium exchange and cross-linking mass spectrometry further confirmed the interaction between Me3 and Aβ, resulting in the binding of Aβ42 to the catalytic domain of Me3, thereby inhibiting Me3 catalytic function. Knockdown of Me3 in neuronal cells attenuated ROS accumulation, normalized mitochondrial morphology, and rescued mitophagy blockage. Our findings highlight the great potential of Me3 as a therapeutic target for early intervention against AD.