Logo: to the web site of Uppsala University

Aims: Pericytes in the brain play important roles for microvascular physiology and pathology and are affected in neurological disorders and neurodegenerative diseases. Mouse models are often utilized for pathophysiology studies of the role of pericytes in disease; however, the translatability is unclear as brain pericytes from mouse and human have not been systematically compared. In this study, we investigate the similarities and differences of brain pericyte gene expression between mouse and human. Our analysis provides a comprehensive resource for translational studies of brain pericytes.

Methods: We integrated and compared four mouse and human adult brain pericyte single-cell/nucleus RNAsequencing datasets derived using two single-cell RNA sequencing platforms: Smart-seq and 10x. Gene expression abundance and specificity were analyzed. Pericyte-specific/enriched genes were assigned by comparison with endothelial cells present in the same datasets, and mouse and human pericyte transcriptomes were subsequently compared to identify species-specific genes.

Results: An overall concordance between pericyte transcriptomes was found in both Smart-seq and 10x data. 206 orthologous genes were consistently differentially expressed between human and mouse from both platforms, 91 genes were specific/up-regulated in human and 115 in mouse. Gene ontology analysis revealed differences in transporter categories in mouse and human brain pericytes. Importantly, several genes implicated in human disease were expressed in human but not in mouse brain pericytes, including SLC6A1, CACNA2D3, and SLC20A2.

Conclusions: This study provides a systematic illustration of the similarities and differences between mouse and human adult brain pericytes.

Claudin-5 (CLDN5) is an endothelial tight junction protein essential for blood-brain barrier (BBB) formation. Abnormal CLDN5 expression is common in brain disease, and knockdown of Cldn5 at the BBB has been proposed to facilitate drug delivery to the brain. To study the consequences of CLDN5 loss in the mature brain, we induced mosaic endothelial-specific Cldn5 gene ablation in adult mice (Cldn5^iECKO). These mice displayed increased BBB permeability to tracers up to 10 kDa in size from 6 days post induction (dpi) and ensuing lethality from 10 dpi. Single-cell RNA sequencing at 11 dpi revealed profound transcriptomic differences in brain endothelial cells regardless of their Cldn5 status in mosaic mice, suggesting major non-cell-autonomous responses. Reactive microglia and astrocytes suggested rapid cellular responses to BBB leakage. Our study demonstrates a critical role for CLDN5 in the adult BBB and provides molecular insight into the consequences and risks associated with CLDN5 inhibition.

Studying disease-related changes in the brain vasculature is warranted due to its crucial role in supplying oxygen and nutrients and removing waste and due to the anticipated vascular dysfunction in brain dis-eases. To this end, we have developed a protocol for fast and simple isolation of brain vascular fragments without the use of transgenic reporters. We used it to isolate and analyze 22,515 cells by single-cell RNA sequencing. The cells distributed into 23 distinct clusters corresponding to all known vascular and perivas-cular cell types in the brain. Western blot analysis also suggested that the protocol is suitable for proteomic analysis. We further adapted it for the establishment of primary cell cultures. The protocol generated highly reproducible results. In conclusion, we have developed a simple and robust brain vascular isolation proto-col suitable for different experimental modalities, such as single-cell analyses, western blotting, and pri-mary cell culture.