Project Goals

Our primary goal is to identify molecular and connectional disruptions that occur in AD and to understand how they contribute to the manifestation and progression of the disease. In AD, the neurons most vulnerable to early degeneration are located in the entorhinal cortex (ENT), subiculum, and hippocampal CA1. The ENT provides a communication gateway between the hippocampus and other structures like the cortex and thalamus, and is essential for memory functions and spatial navigation.  The dorsolateral zone of the ENT and its adjacent perirhinal cortical area, homologous to the transentorhinal cortex in humans, is considered the anatomical ground zero of AD from which pathologies then propagate across limbic and associated cortices.  Specifically, ENT layer 2/3 pyramidal neurons may be the first to develop AD pathology including tau aggregates, Aβ deposits, and neuronal loss. Moreover, in early AD, ENT neurons show prominent synapse loss correlated with cognitive impairment, prominent dendritic morphological pathology, and axonal swellings. Thus, the ENT is a critical brain region to study molecular and cellular mechanisms of selective neuropathogenesis in AD.

Selection of Mouse Models

Recently developed gene replacement mouse models of tau pathology will be utilized to examine these age-related connectopathies, as well as alterations in tau phosphorylation and aggregation, neurodegeneration, and gene expression.
There are notable limitations in current transgenic and early onset AD mouse models, particularly for preclinical AD studies. Existing models express high and non-physiological levels of disease proteins (e.g. >10 fold) and lack endogenous regulatory elements for proper gene expression patterns. Thus, these models are insufficient for recapitulating the late-onset, slowly progressive, brain regional, and cell-type selective pathogenesis of AD. Through their NIA-funded Model-AD project, MPIs Lamb and Co-I Oblak are developing the next generation of knock-in, and “humanized” LOAD mouse models that use endogenous genomic constructs incorporating multiple LOAD risk variants. The consortium is performing longitudinal -omics, imaging, biomarker, and neuropathological phenotyping of these new LOAD models to define their progressive phenotypes across the lifespan, and to provide rigorous platforms for AD pathogenic and therapeutic studies. Recent advancements in APP knock-in models carrying multiple familial AD (FAD) mutations have produced FAD models with better construct validity. A recently developed APP knock-in model, co-developed by Denali and JAX, shows slowly progressive Aβ deposition, progressive neurite dystrophy, robust microglial activation, and transcriptomic and lipomic deficits. A key advantage of APPSAA KI, in contrast to other APP KI mice (e.g. APPNL-G-F) is that it will be publicly available to the research community (JAX #34711) and free of IP restrictions. 

Further, Dr. Michael Koob’s laboratory has generated humanized tau MAPT knock-in mice. The mice were created by: 1) deletion of the MAPT gene in a mouse embryonic stem cell line, and 2) insertion of the syntenic sequence from the human genome. The experimental mouse models used in these studies are MAPT(H1)*N279K (a frontotemporal dementia Tau mutation; unpublished data, Koob lab) and their H1 haplotype controls [MAPT(H1)]. 

These recently developed APPSAA KI and MAPT(H1)*N279K mouse models will be used to study molecular, morphological, and connectomic deficits of ENT neurons AD. 

Team Members

UCLA (Yang group)

---

Indiana University (Lamb group)

---

UCLA (Dong group)