Semaphorin 3A and 3F overexpression in TIE2 hyperactive endothelial cells contribute to the pathological lumen expansion in venous malformation

Background

Venous malformation (VM) are developmental defects of the vasculature characterized by tremendously enlarged and dysfunctional veins. Gain-of-function somatic mutations in the endothelial-specific tyrosine kinase receptor TIE2 have been identified as leading driver of VM pathogenesis. The aim of this study was to determine whether the aberrant venous lumen expansion is caused by recruitment of wild-type endothelial cells (EC) to the lesion or by TIE2-mutant EC clonal expansion.

Methods

To investigate the contribution of TIE2-mutant EC and wild-type EC to the aberrant venous lumen expansion, we used a xenograft murine model of VM generated with a combination of TIE2-mutant EC and wild-type EC. To perform longitudinal studies, we employed a three-dimensional (3D) fibrin gel lumen formation assay and a migration assay, both using wild-type EC in competition or confrontation with TIE2-mutant EC. To investigate the mechanisms implicated in VM lumen expansion we used RNA-sequencing and short interference (sh)RNA in the TIE2-mutant EC.

Results

We demonstrate here that in the VM xenograft model, the aberrant blood vessels were lined almost exclusively by TIE2-mutant EC, and wild-type EC were rarely found. Functionally, the TIE2-mutant EC exerted a competitive advantage over wild-type EC by inhibiting wild-type EC sprouting. In line with these findings, TIE2-mutant EC promoted repulsion of wild-type EC. ShRNA-mediated silencing of Sema3A or Sema3F in TIE2-mutant EC rescued this chemorepellent phenotype and restored the ability of wild-type EC to migrate, sprout and form lumens. Furthermore, knock-down of Sema3A or 3F in TIE2-mutant EC normalized the blood vessel size in vivo.

Conclusions

Our results demonstrate that wild-type EC are not recruited to the aberrant veins suggesting VM pathogenesis is fueled by clonal expansion of TIE2-mutant EC. Mechanistically, we show that Sema3A and 3F are overexpressed in TIE2-mutant EC and play a crucial role in the pathological vascular lumen expansion in VM.

Abstract (not formatted)

Venous malformations (VM) are developmental defects of the vasculature characterized by tremendously enlarged and dysfunctional veins. Gain-of function somatic mutations in the endothelial-specific tyrosine kinase receptor TIE2 have been identified as one of the leading drivers of VM pathogenesis. The aim of this study was to determine whether the aberrant venous lumen expansion is caused by the recruitment of wild-type endothelial cells (EC) to the lesion or instead by TIE2-mutant EC clonal expansion.

In a xenograft murine model of VM generated with a combination of TIE2-mutant EC and wild-type EC, we demonstrated that the aberrant blood vessels were lined almost exclusively by TIE2-mutant EC, suggesting lesions form by clonal expansion, while wild-type EC were not recruited to the aberrant veins. Functionally, in a three-dimensional cell competition assay, we showed that TIE2-mutant EC exerted a competitive advantage over wild-type EC by inhibiting wild-type EC sprouting and ability to form vascular lumens. In line with these findings, TIE2-mutant EC repelled wild-type EC by reversing their migration direction in a cell confrontation assay.

In seeking to define the mechanism driving this repulsion phenotype, we detected elevated levels of the chemorepellent Semaphorin 3A (Sema3A) and Sema3F in the TIE2-mutant EC. ShRNA-mediated silencing of Sema3A or Sema3F rescued the chemorepellent phenotype and restored the ability of wild-type EC to migrate, sprout and form lumens. Furthermore, Sema3A or 3F knock-down in TIE2-mutant EC normalized the blood vessel morphology and size in vivo. Taken together, these data strongly indicates that Sema3A and 3F are important players in VM pathogenesis.