Endothelial Regulation of Vascular Calcification

PROJECT SUMMARY:
Vascular calcification (VC) frequently complicates cardiovascular disease. It increases the morbidity and
mortality and constitutes a significant obstacle in interventions and surgeries. The vascular endothelium plays
an important role in VC. The intimal (luminal) endothelial cells (ECs) contribute to VC by providing
osteoprogenitor cells through endothelial-mesenchymal transitions (EndMTs). The adventitial ECs are known
to contribute to neo-angiogenesis in diseased vascular wall, but it is unknown whether such ECs support VC
and what defines them. A subset of ECs with high expression of the EC marker CD31 and the glycoprotein
Endomucin (Emcn) has been found in bone to support bone formation. It is possible that adventitial or other
peripheral ECs are recruited to diseased areas to promote calcification. Preliminary experiments, using the
Matrix Gla Protein null (Mgp-/-) mouse as a VC model, showed extensive EC involvement in the calcified aorta.
We identified two subtypes of ECs in the adventitial vs. the intimal endothelium (referred to as a-ECs and i-
ECs). The a-ECs were CD31+Emcm+ and correlated with the severity of the VC, whereas the i-ECs were
CD31+Emcn-. The two ECs had distinct transcriptional profiles with stem cell and osteogenic markers in the i-
ECs vs. enhanced Notch expression in the a-ECs. Endothelial deletion of Notch1 reduced the a-ECs and
limited VC while promoting cartilage formation and survival in the Mgp-/- mice. The bone transcription factor
Osterix was expressed in both types of ECs. We hypothesize that a-ECs are distinct from i-ECs, recruited to
nascent VC, and susceptible to Notch disruption. We also hypothesize that Osterix is protective of EC lineage.
In Aim 1, we will characterize the a-ECs (CD31+Emcn+) and compare to the i-ECs (CD31+Emcn-) in the Mgp-/-
model, and correlate with severity of VC and marker expression. We will identify unique markers for the
respective ECs, with comparison to bone, using transcriptional profiles from single cell RNA sequencing
(scRNAseq). We will test the concept that angiogenesis is required for VC using angiogenic inhibitors. In Aim
2, we will determine the effect of loss or gain of endothelial Notch signaling on the EC subtypes and VC. We
will examine the distribution of Notch components in relation to VC, and generate Mgp-/- mice with endothelial-
specific loss of Notch1 or the Notch receptor inactivator Fbxw7. We will use the mice to determine the effect on
the appearance of the endothelial subtypes, calcification and transcriptional profiles by scRNAseq. We will also
apply loss and gain of Notch to ECs in vitro and identify novel Notch targets and networks. In Aim 3, we will
determine if Osterix helps maintain EC lineage or promotes calcification in vitro and in vivo using human aortic
ECs and inducible endothelial-specific Osterix gene deletion in Mgp-/- mice. We will compare the transcriptional
profiles of ECs with and without Osterix by scRNAseq in order to clarify the effect on EC lineage, a-ECs vc. I-
ECs, and involved signaling networks. Our results may have a significant impact on the field of VC, in
particular on the understanding of the endothelial pathology and involvement in VC.

Grant Number: 5R01HL158053-04
NIH Institute/Center: NIH
Principal Investigator: Kristina Bostrom