NOTCH1 regulates matrix gla protein and calcification gene networks in human valve endothelium

Abstract

Valvular and vascular calcification are common causes of cardiovascular morbidity and mortality. Developing effective treatments requires understanding the molecular underpinnings of these processes. Shear stress is thought to play a role in inhibiting calcification. Furthermore, NOTCH1 regulates vascular and valvular endothelium, and human mutations in NOTCH1 can cause calcific aortic valve disease. Here, we determined the genome-wide impact of altering shear stress and NOTCH signaling on human aortic valve endothelium. mRNA-sequencing of primary human aortic valve endothelial cells (HAVECs) with or without knockdown of NOTCH1, in the presence or absence of shear stress, revealed NOTCH1-dependency of the atherosclerosis-related gene connexin 40 (GJA5), and numerous repressors of endochondral ossification. Among these, matrix gla protein (MGP) is highly expressed in aortic valve and vasculature, and inhibits soft tissue calcification by sequestering bone morphogenetic proteins (BMPs). Altering NOTCH1 levels affected MGP mRNA and protein in HAVECs. Furthermore, shear stress activated NOTCH signaling and MGP in a NOTCH1-dependent manner. NOTCH1 positively regulated endothelial MGP in vivo through specific binding motifs upstream of MGP. Our studies suggest that shear stress activates NOTCH1 in primary human aortic valve endothelial cells leading to downregulation of osteoblast-like gene networks that play a role in tissue calcification.

Keywords: Matrix gla protein; NOTCH signaling; NOTCH1; Valve calcification; Valve endothelium.

Figure 1

Activated NOTCH1 is detected in valve cells in a flow dependent manner. (A) Normal human aortic valve section stained with an antibody specific to the active form of NOTCH1 (NICD, red). NICD was found in both endothelial (arrows) and interstitial cells (arrowheads). Autofluorescence (AutoFI, green) of collagen and elastin highlight the fibrosa layer (F) and ventricularis layer (V), respectively. Nuclei, DAPI (blue). Scale bars indicate 100 μm. (B) Schematic of experimental procedure. Normal HAVECs were transfected with control or NOTCH1 siRNA, then cultured in static or fluid flow conditions. Gene expression was compared by qRT-PCR or mRNA-seq. (C) qRT-PCR analysis of HAVECs from four conditions. NOTCH1, two canonical direct targets, HES1 and HEY2 and a known flow responsive gene, KLF2, were analyzed. Graphs show mean gene expression relative to the static, control siRNA condition with error bars representing standard deviation. (n=3; *, p<0.05; ** p<0.01; *** p<0.001; NS, Not Significant).

Figure 2

Expression of endochondral ossification genes is affected by shear stress and NOTCH1 signaling. Graphs show mean gene expression relative to the static, control siRNA condition with error bars representing standard deviation. All numbers are shown in log₂ scale. (n=3; *, p< 0.05; **, p< 0.01; ***, p< 0.001, NS, Not Significant).

Figure 3

Heatmap and clustering analysis of RNA-seq and ChIP-seq from HAVECs reveal likely NOTCH1 direct targets. (A) The expression of each gene was normalized to the static, control siRNA condition and then Log₂ transformed. Displayed genes were selected as matching one of two patterns reflecting coordinate regulation by both NOTCH1 and shear stress: Pattern 1) at least 2-fold down upon NOTCH1 siRNA knockdown in both static and flow conditions, and at least 2-fold up in flow control siRNA condition compared to static, or Pattern 2) at least 2 fold up in NOTCH1 siRNA knockdown in both static and flow conditions, and at least 2 fold down in flow control siRNA condition compared to static. (B) ChIP-seq score shows the value for the highest ChIP peak +/− 20 kb from the transcriptional start site (TSS) of each gene. Annotated genes represent potential direct targets of NOTCH1 based on ChIP-seq score.

Figure 4

MGP expression is NOTCH1 dependent. (A) Normal human aortic valve sections stained with antibodies specific to: i. cMGP, the active, carboxylated form of MGP; ii. ucMGP, the inactive, uncarboxylated form of MGP; and iii. phosphoSMAD 1/5/8 (pSMAD1/5/8) cMGP, ucMGP and pSMAD1/5/8 (red) were found in both endothelial (arrows) and interstitial (arrowheads) cells. Autofluorescence (AutoFI, green) of collagen and elastin highlight the fibrosa layer (F) and ventricularis layer (V), respectively. Nuclei, DAPI (blue). Scale bars indicate 100 μm. (B) qRT-PCR analysis of MGP mRNA expression in HAVECs under four conditions. Graph shows mean gene expression relative to the no flow, no siRNA condition with error bars representing the standard deviation. (n=3; *, p< 0.05). (C) Western blot detecting NICD, active MGP (cMGP) and GAPDH from HAECs infected in vitro with increasing amounts of myc-tagged NOTCH1 intracellular domain (NICD). (D) Immunostaining for NICD (red) in cultured Human Aortic Endothelial Cells (HAEC) or Human Aortic Interstitial Cells (HAIC). (E) qRT-PCR to measure MGP mRNA in control or NOTCHI-overexpressing HAICs. (F) Western blot detecting active MGP in control or γ-secretase inhibitor (DAPT) treated HAECs. (G) Quantification of cMGP protein in (F) normalized to Actin. Error bars show standard deviation (n=3; **, p< 0.01; ***, p< 0.001; NS, Not Significant).

Figure 5

NOTCH directly regulates MGP through an endothelial enhancer. (A) NICD1-myc ChIP-qPCR. Relative enrichment is shown for four sites: validated HES1 NOTCH1/CSL binding site, validated HES1 non-CSL binding site, predicted CSL binding sites within the 821 bp putative MGP enhancer and an MGP non-CSL site. The HES1 and MGP non-CSL binding sites are located approximately 2 kb from the CSL binding sites. Error bars indicate standard deviation. (n=4; *, p< 0.05). (B): EMSA snowing CSL binding to putative MGP enhancer. A 46 bp oligo corresponding to the NICD1-myc ChIP-seq peak was labeled with ³²P-dCTP (MGP Probe). Incubation with in vitro transcribed and translated CSL (CSL TnT) shifted the probe (arrow) This interaction could be competed by addition of unlabeled MGP probe (Unlabeled MGP) or oligo corresponding to a validated HES1 enhancer containing CSL sites (Unlabeled HES1), but not with unlabeled oligos in which the CSL sites were mutated (Mut), demonstrating specificity. Reticulocyte lysate (Empty TnT) caused a higher, non-specific shift. (C) Images of whole mount or histological sections from E16.0 transgenic embryos containing the 821 bp MGP enhancer upstream of the Hsp68 minimal promoter driving LacZ with or without mutation of the three predicted CSL binding sites. Mice containing the wildtype (Wt) enhancer had strong endothelial cell expression of LacZ (blue) in the large vessels of the arterial system (8/13) and some weak expression in the aortic valve (arrows) (4/13 embryos). Mutation of CSL sites (Mut) abolished endothelial expression (8/8); the mutant enhancer had weak expression in the smooth muscle layer (arrowhead) of 3/8 embryos. Sections were counterstained with Eosin (red). Aorta, Ao; branchiocephalic artery, BC; left common carotid, LCC; aortic valve, AoV.

Figure 6

Model of NOTCH1 regulation of human endothelial cell calcification. Yellow shapes indicate different classes of molecules and their sub-cellular localization. Genes in black are regulated by shear stress and NOTCH1 signaling, while those in orange also have a significant NOTCH1 ChIP-seq peak within 20 kb of the transcriptional start site, suggesting potential direct transcriptional regulation by NOTCH1.