Pathological Alzheimer?s disease (AD) is a major cause of dementia characterized by memory loss and aggregation of insoluble beta amyloid plaques and tau tangles. Memories are stored at synapses, and it is thought that an early driver of dementia may be synapse pruning occurring even before plaque deposition. Extensive activity-dependent synaptic pruning also occurs during developmental critical periods when learning and experience strengthen and stabilize actively used synapses, while others weaken and are pruned. In an unbiased in vivo screen for genes regulated by neural activity during visual system development, my lab made the unexpected discovery that specific Major Histocompatibility Class I (MHCI) molecules, famous for their immune system roles, are expressed in neurons and at synapses. Next, we identified an innate immune MHCI receptor expressed in neurons: PirB (Paired immunoglobulin-like receptor B). Functional studies in mice reveal that the MHCI - PirB axis is required for synapse pruning during normal development. Genetic deletion of PirB selectively in cortical pyramidal neurons, or pharmacologic blockade using a recombinant protein, rapidly generates new spines and functional synapses even in adult cerebral cortex. In the APP/PS1 transgenic model of autosomal dominant AD, mice lacking PirB are protected from memory loss at 9 months of age despite high levels of beta amyloid. Remarkably, PirB is a receptor for soluble beta amyloid oligomers, with high affinity saturable binding. This interaction hyperactivates cofilin signaling which drives actin depolymerization and contributes to synapse pruning in the APP/PS1 AD mouse model. In human the LilrB (leukocyte immunoglobulin- like receptor B) family of 5 related molecules are PirB homologs. Similar to PirB, LilrB1 and LilrB2 are known to bind MHCI ligands, including HLA-A, B and C alleles, which are implicated in human GWAS and gene expression studies of AD. We discovered that LilrB2 binds soluble beta amyloid oligomers with nanomolar affinity, and LilrB2 protein is expressed in human frontal lobe. A crystal structure of the interaction between beta amyloid and LilrB2 has been solved, confirming genuine structural interactions and pointing to novel drug targets for AD. A major goal of this research is to test the hypothesis that innate immune signaling via MHCI-PirB/LilrB at the synapse is disrupted by pathological oAbeta, and to connect observations in mice to human neurobiology by (1) studying MHCI-PirB dependent signaling in neurons using RiboTag cell type- specific transcription profiling in AD model mice, and (2) by identifying and studying the function of human homologs, the HLA Class I and LilrB receptor families, in 3-dimensional forebrain organoids derived from human iPSCs, followed by validation in brain samples. Results from these studies will build a bridge between mouse models of AD and human neurons. They should also provide mechanistic information about how nervous and immune systems communicate at the synapse and open up new therapeutic avenues for treating synapse pruning disorders in development and in Alzheimer?s disease.

Public Health Relevance

Memories are stored at synapses, and synapse loss is hypothesized to be an early hallmark of Alzheimer?s disease (AD). The major aims of this research are to determine if molecular mechanisms of synapse pruning mediated by an innate immune receptor expressed in neurons and its MHC class I ligands are disrupted in the presence of soluble oligomers of beta amyloid, and to link findings in mouse models of AD to human neurons and pathology by studying brain organoids derived from human iPS cells, as well as autopsy material. Because the homologous receptor in humans also binds soluble oligomers of beta amyloid, this research can offer mechanistic explanations for synapse and memory loss in human AD and may lead to new drug targets for treatment.

National Institute of Health (NIH)
National Institute on Aging (NIA)
Research Project (R01)
Project #
Application #
Study Section
Cellular and Molecular Biology of Neurodegeneration Study Section (CMND)
Program Officer
Wise, Bradley C
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Stanford University
Schools of Arts and Sciences
United States
Zip Code