Project summary Alzheimer's disease (AD) is a leading cause of dementia, which is characterized by memory loss and thinking problems interfering with daily life. Although the importance of beta amyloid and Tau in AD pathogenesis has been well appreciated, unfortunately no effective treatment is available to prevent dementia progress so far due to the lack of understanding of neurodegeneration in AD, which encourages researchers in the field to further identify novel risk factors and to explore the complexity of AD pathogenesis. Recent integrative transcriptome analyses of the aging brain have revealed that aberrant alternative splicing events are reproducibly associated with AD and multiple mitochondrial proteins have been identified as new risk factors, implicating the importance of mitochondria dysfunction and pre-mRNA splicing in AD pathogenesis. Apoptosis-inducing factor (AIF) is an X-chromosome linked mitochondrial flavoprotein serving as a free radical scavenger and plays a vital function in mitochondrial bioenergetics. Spontaneous genetic mutations of AIF have been observed in both human and mouse and provided strong association of Aif gene with the etiology of neuropathy and cognitive impairment. Recently, we unexpectedly discovered a hitherto unknown AIF splicing isoform lacking two exons at its N- terminus, defined as AIF3 distinct to other two known isoforms. AIF3 was undetectable in normal human or mouse brain, but induced in human AD patients. Induction of AIF3 splicing in mouse brain caused human AD- like phenotypes, including mitochondrial dysfunction, phospho-Tau and beta amyloid aggregation, neurofibrillary tangle structure formation, and neuron loss in cortex and hippocampus. Our extensive preliminary data as well as genetic evidence provide the strong scientific premise and lead us to hypothesize that AIF3 splicing plays an essential role in neurodegeneration and AD pathogenesis. The goals of this R01 project are to 1) obtain a comprehensive understanding of AIF3 functions in cognitive deficits and neurodegeneration in AD; 2) decipher the molecular mechanisms of AIF3-meidated neurodegeneration; 3) dissect the molecular and cellular mechanisms of AIF3 splicing regulation in AD using unbiased approaches. In addition to the assembly of a strong AD research team to ensure the success of the proposed project, two tamoxifen inducible AIF3 mouse models have been established successfully in the lab, which provide valuable tools to understand the role of AIF3 in neurodegeneration in AD by gain-of-function and loss-of-function approaches. Successful completion of this project will discover significant new functions of AIF3 in AD pathogenesis, which may lead to identify a new therapeutic target for AD dementia, and also yield a valuable new mouse model for understanding AD pathogenesis and a platform for AD drug development.