UC San Diego

Prion diseases are rapidly progressive neurodegenerative diseases characterized by spongiform degeneration, gliosis, synaptic loss, and neuronal dystrophy. Prion disease is caused when the cellular glycosylphosphatidylinositol (GPI)-anchored prion protein, PrPC, misfolds into an aggregated conformer, known as PrPSc. PrPC has an unstructured flexible N-terminus and a structured C-terminus with two N-linked glycans. Several human prion diseases are caused by mutations in the N-terminus of PrPC. For example, some cases of Gerstmann-Straüssler- Scheinker disease or Creutzfeldt-Jakob disease are caused by insertions in the octapeptide repeat region of the N-terminus. Furthermore, the untethered N-terminus of PrPC can act as a neurotoxic effector domain. Mice expressing only the N-terminus of PrPC show altered interactions between the truncated PrPC and endoplasmic reticulum proteins, which results in neurotoxicity. To understand how the N-terminus impacts PrPC function and neurodegenerative disease progression, we generated a knock-in mouse model that modifies the N-terminus of PrPC by substituting glycine 93 to asparagine, allowing the N-terminus to be glycosylated and potentially untethered. This mutation causes spontaneous neurodegeneration without the presence of PrPSc aggregates. In vitro experiments support a role for excitotoxicity in the neurodegenerative phenotype of the Prnp93N mice. Hippocampal and cortical primary neuron cultures from Prnp93N mice display dendritic beading, which is characteristic of excitotoxic stress. The excessive dendritic beading in the Prnp93N neurons was rescued using a glutamate receptor antagonist implicating signaling through glutamate receptors that are driving the excitotoxic insult. In vivo experiments also indicate a role for excitotoxicity in a brain specific region. Hippocampi of p26 Prnp93N mice have higher phosphorylated Ca2+/calmodulin-dependent protein kinase II (CaMKII) at threonine 286 (Thr286). Calcium ion binding to CaMKII activates the kinase and promotes autophosphorylation at Thr286 to constitutively activate CaMKII. Hippocampi of p26 Prnp93N mice also show higher phosphorylated glutamate receptors, indicating increased channel conductance when compared to PrnpWT mice. Cortex samples do not show differences. Together, this data suggests neuronal activation differs between the cortex and the hippocampus in the Prnp93N mice. This study supports the hypothesis that enhanced excitotoxic signaling involving receptor contributes to this neurodegenerative phenotype.