Functional Definition of Progenitors Versus Mature Endothelial Cells Reveals Key SoxF-Dependent Differentiation Process

BACKGROUND: During adult life, blood vessel formation is thought to occur via angiogenic processes involving branching from existing vessels. An alternate proposal suggests that neovessels form from endothelial progenitors able to assemble the intimal layers. We here aimed to define vessel-resident endothelial progenitors in vivo in a variety of tissues in physiological and pathological situations such as normal aorta, lungs, and wound healing, tumors, and placenta, as well. METHODS: Based on protein expression levels of common endothelial markers using flow cytometry, 3 subpopulations of endothelial cells could be identified among VE-Cadherin+ and CD45- cells. RESULTS: Lineage tracing by using Cdh5cre(ERt2)/Rosa-YFP reporter strategy demonstrated that the CD31-/loVEGFR2lo/intracellular endothelial population was indeed an endovascular progenitor (EVP) of an intermediate CD31intVEGFR2lo/intracellular transit amplifying (TA) and a definitive differentiated (D) CD31hiVEGFR2hi/extracellular population. EVP cells arose from vascular-resident beds that could not be transferred by bone marrow transplantation. Furthermore, EVP displayed progenitor-like status with a high proportion of cells in a quiescent cell cycle phase as assessed in wounds, tumors, and aorta. Only EVP cells and not TA and D cells had self-renewal capacity as demonstrated by colony-forming capacity in limiting dilution and by transplantation in Matrigel plugs in recipient mice. RNA sequencing revealed prominent gene expression differences between EVP and D cells. In particular, EVP cells highly expressed genes related to progenitor function including Sox9, Il33, Egfr, and Pdfgra. Conversely, D cells highly expressed genes related to differentiated endothelium including Ets1&2, Gata2, Cd31, Vwf, and Notch. The RNA sequencing also pointed to an essential role of the Sox18 transcription factor. The role of SOX18 in the differentiation process was validated by using lineage-tracing experiments based on Sox18Cre(ERt2)/Rosa-YFP mice. Besides, in the absence of functional SOX18/SOXF, EVP progenitors were still present, but TA and D populations were significantly reduced. CONCLUSIONS: Our findings support an entirely novel endothelial hierarchy, from EVP to TA to D, as defined by self-renewal, differentiation, and molecular profiling of an endothelial progenitor. This paradigm shift in our understanding of vascular-resident endothelial progenitors in tissue regeneration opens new avenues for better understanding of cardiovascular biology.