A clear picture of the perivascular cell landscape and its implication in tissue functions

Abstract

The vasculature of every organ is adapted to the specific tissue functions with activity-dependent blood flow control. While the lumen of blood vessels is formed by endothelial cells (ECs), their abluminal surface is covered by morphologically and functionally distinct perivascular cells (PVCs). Currently known PVCs encompass vascular smooth muscle cells (vSMCs), pericytes, perivascular fibroblasts (PVFs) and perivascular macrophages (PVMs). However, their distribution in relation to each other along the arteriovenous (AV) axis remains poorly characterized. Determining the distribution of PVCs along the AV axis would help to better understand the vascular architecture to infer their functions.

During my doctoral studies, I developed a tissue-clearing based pipeline enabling multi-scale 3D imaging of diverse PVCs to probe their organization along the AV axis. Moreover, I designed an open-source volumetric image handling framework to systematically store and analyze the data. Using this workflow, I characterized the distribution and morphology of vMSCs, pericytes, PVFs and PMVs throughout the central nervous system (CNS). Probing the distributional difference of PVFs and pericytes along the AV axis in the spinal cord, I contributed to the understanding on the heterogenous origin of the fibrotic scar. Furthermore, I analyzed how penile erection alters the contact between PVFs and other cells and the results help to understand the role of PVFs in penile blood flow regulation.

In paper I, I presented the tissue-clearing based pipeline and the volumetric image handling framework. Utilizing various transgenic mouse lines and antibody labeling, we probed the distribution and organization of vSMCs, pericytes, PVFs, and PVMs along the AV axis. Being able to visualize the vasculature landscape deep within the brain and spinal cord, we could determine the major arterial and venous network in various CNS regions to act as reference for PVC mapping. We found that morphological heterogeneity along AV axis was conserved throughout the CNS. In the brain, PVFs and PVMs were predominately located along arterioles and only a few large venules but in the spinal cord, they were found in virtually all arterioles and venules. We also found distinct morphological differences of PVFs and PVMs along the AV axis and described a novel ‘thin-stranded’ PVF morphology at the entry or exit of the capillary network which extended long processes along multiple capillary branches. This study provides a comprehensive analysis of the perivascular landscape along the AV axis which infers functional specialization.

Previous work from our lab has shown that scar-forming stromal fibroblasts in the injured spinal cord and brain originate from a small population of glutamate aspartate transporter (GLAST)-expressing PVCs. In paper II, we investigated the heterogeneity of these scar-forming PVCs in mice. Based on single-cell RNA sequencing analysis, we discovered that GLAST-expressing PVCs in the spinal cord encompassed two types of cells that could be transcriptionally defined as PVFs and pericytes. Based on strong and specific expression of COL1A1 in PVFs, we selected a COL1A1-CreERT2 mouse line to target PVFs. Using 3D imaging analysis, we compared the distribution of GLAST+ and COL1A1+ PVCs using GLAST- CreERT2;R26-tdTomato and COL1A1-CreERT2;R26-tdTomato mouse lines and found that GLAST+COL1A1- PVFs distributed along larger penetrating vessels enriched in white matter regions, while GLAST+COL1A1- pericytes were in the microvasculature enriched in grey matter regions. Upon injuries, both populations were recruited locally, proliferated, and contributed to fibrotic scar formation. The contribution to scar forming stromal fibroblasts was dependent on the anatomical location of the lesion. The results of this study highlight the functional relevance of the spatial distribution of PVCs in response to injury.

In paper III, we examined the role of PVFs in penile blood flow regulation which is important for erectile function. This study resulted in the discovery of a novel function of corpora cavernosa PVFs in supporting vasodilation by reducing norepinephrine availability. We found that more frequent erections led to higher number of PVFs and improved blood flow, and vice versa. Employing volumetric analysis of cleared tissue, we determined that tissue structural changes led to altered contacts between PVFs and other cells in the corpora cavernosa which prompted us to investigate the Notch signaling pathway which involved physical contact between cells. Our results demonstrated that downregulation of Notch signaling in PVFs increased the numbers of PVFs. The study highlights the coupling of PVC organization and their physiological functions.

In summary, I presented a pipeline to interrogate the morphology and organization of diverse PVCs along the 3D vascular network. Using this pipeline, we refined the organization of PVCs along the AV axis in the CNS. Furthermore, we revealed the linkage between 3D PVC organization and (patho)physiological responses as demonstrated in the case of fibrotic scar formation and penile erection.