Impairment of hippocampal gamma oscillations, mitochondria and neurovascular function in CADASIL

Abstract Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is a small vessel disease caused by cysteine-altering NOTCH3 gene variants, leading to vascular smooth muscle cell degeneration, compromised cerebral blood flow, subcortical ischemic infarcts, cognitive decline, and often ultimately vascular dementia. Little is known about the cellular and molecular effects downstream of the cerebral ischemia in CADASIL, or whether brain regions known to be involved in dementia, such as the hippocampus, are particularly susceptible to such pathological downstream changes. In this study, we used a humanized CADASIL mouse model harbouring the p.(Arg182Cys) variant (R182C-TgN3), post-mortem human CADASIL brain sections with four different NOTCH3 gene variants and primary human cerebral vascular smooth muscle cells (VSMCs) harbouring the p.R133C NOTCH3 variant as primary cellular models to characterise the properties and contribution of mutant VSMCs to cognitive impairment. To specifically evaluate neuronal, mitochondrial and neurovascular function, we performed ex vivo electrophysiology, immunohistochemistry (confocal and iDISCO+ methods), western blotting, Seahorse assay, quantitative polymerase chain reaction (qPCR), and single-cell RNA sequencing. In the CADASIL mice, hippocampal gamma oscillation patterns were impaired along with significant decreases in neuronal fiber length and aberrant neuronal morphology. The latter two phenotypes were also observed in post-mortem brain tissue from CADASIL patients. Consistent with these findings, we noted significantly lower levels of mitochondrial respiratory complexes in the CADASIL mouse hippocampus, isolated mouse brain vessels and primary human cerebral VSMCs. The human cerebral VSMCs exhibited reduced oxygen consumption rates leading to reduced ATP production as well as decreased glycolytic capacity in conjunction with increased pro-inflammatory gene expression, suggesting a broader impact on cellular energy metabolism and a neuroinflammatory process. In the CADASIL mice, we also observed extensive accumulation of the NOTCH3 extracellular domain in hippocampal vessels. Light sheet imaging with iDISCO+ clearing demonstrated substantial VSMC loss and reduced vessel density in the hippocampus at 9 months of age. Additionally, 3D imaging showed increased microglial attachment to vessels and enlargement of the size of the vessel-associated microglia in CADASIL mice. Single-cell RNA sequencing revealed a microglial subcluster expressing genes involved in mitochondrial respiration and inflammation. Collectively, our results reveal how small vessel pathology in CADASIL leads to significant neuronal pathology in the hippocampus involving metabolic and neuroinflammatory changes and highlight the critical role of the neurovascular unit. Our findings pave the way for future research and potential therapeutic strategies.