The role of endoglin in vascular patterning and malformation across diverse vascular beds

Abstract

During development, numerous processes must be finely orchestrated to construct a functional vasculature, essential to sustain life. Hereditary haemorrhagic telangiectasia is a disease characterised by the development of arteriovenous malformations and it is caused in over 80% of the cases by mutations in either endoglin (ENG) or ALK1. So far, studies have demonstrated the relevance of ENG in angiogenesis during embryonic and postnatal development. However, its roles during adulthood and in the context of vascular regression had not yet been as explored. In addition, traditional methods to study the vasculature in general, and angiogenesis in particular, usually involve sectioning of tissues or retinal whole mounts, not completely allowing to analyse the vasculature in relation to its surroundings. Therefore, we optimised a clearing protocol to use light sheet fluorescence microscopy (LSFM) and study the vasculature of intact mouse eyes.

In Paper I, we study the role of ENG in the adult vasculature and show that loss of Eng leads to high output heart failure in adult mice due to malformations in a distal location, the pubic symphysis. Inhibition of VEGFR signalling completely blocked AVM establishment and the resulting heart failure. In Paper III, we show that ENG is also required for the normal development and regression of the hyaloid vessels. Together, we demonstrate that ENG is also needed during adult stages and nonangiogenic conditions to maintain a healthy vasculature.

In Paper II, we provide an optimised protocol to clear mouse eyes and analyse their multiple vascular beds using LSFM. Using this protocol, we overcame the challenging dissection of the hyaloid vasculature and were able to describe its regression as well as the remodelling of the iris vasculature. We also applied the protocol to a model of laser induced choroidal neovascularization to show the potential and advantages that 3D imaging combined with in silico dissection have over the traditional techniques to study this model.

In Paper IV, we apply single cell RNA sequencing to our mouse model of HHT1 to further characterise the vascular malformations that these mice develop and find potential cell autonomous effects derived from the loss of ENG. We were able to identify five clusters that correspond to known EC subpopulations and found multiple genes that are differentially regulated upon loss of Eng, such as those relating to the BMP/TGF-β pathway, Vwf and Klf4, as well as cluster-specific differentially regulated genes.