BIN1 genetic risk factor for Alzheimer is sufficient to induce early structural tract alterations in entorhinal-hippocampal area and memory-related hippocampal multi-scale impairments

Late Onset Alzheimer Disease (LOAD) is the most common form of dementia and one of the most challenging diseases of modern society. Understanding the preclinical stages of AD that begins in the brain at least 2 to 3 decades before evidence of episodic memory defects in patients is pivotal for the design of successful approaches to delay or reverse the transition from normal brain function to cognitive impairments. Our working hypothesis is that LOAD genetic risk factors can be sufficient to generate early phenotypical changes before any changes in either Abeta or Tau. To test this hypothesis, we generated an hBIN1 mouse model based on the human BIN1 gene overexpression that we found in post-mortem brain samples from LOAD patients. BIN1 is the second important risk factor for AD, following the APOE gene. We identified co-deregulated gene repertoires common to both 7-week mouse hippocampus sub-regions and post-mortem brain samples from LOAD patients, demonstrating the validity of this hBIN1 model. We evidenced an early phenotype of neurodegeneration starting at 3 months with structural impairment fiber pathways quantified by high resolution (17.2T) (MRI DTI) and related functional impacts. We found structural changes in entorhinal cortex-dentate gyrus (EC-DG) pathway known to be the earliest brain region impacted in LOAD. Similarly, the function of this pathway was impaired both in vitro and in vivo, with the changes in spine density and dendritic simplification of DG neurons, impaired EC-DG long-term potentiation (LTP) and behavioral deficits linked to object recognition episodic memory. As expected for a neurodegenerative model, we evidenced the progression of dysfunction at the morphological, functional and behavioral levels with age. Structural spreading involved impairment of fibers in somatosensory and temporal associative cortexes at month 15. Functional and behavioral spreading was characterized by impact on pattern separation of spatial episodic memory. Moreover, this neurodegeneration occurred without any detectable changes in Abeta 1-42 and tau. Overall, these data show the possibility to identify a repertoire of molecular changes occurring both in patients and in hBIN1 mice and whose further manipulation can be instrumental to rescue or delay episodic memory defects.