Endothelial Cullin3 Mutation Causes Vascular Dysfunction, Arterial Stiffening, and HypertensionΔ

Frame‐shift mutations in Cullin‐3 (causing skipping of exon 9, CUL3Δ9) were previously identified in pseudohypoaldosteronism type II (PHAII) patients that exhibit increased renal NaCl reabsorption, hyperkalaemia, metabolic acidosis and hypertension. The hypertension phenotypes were initially found to be driven by renal tubular mechanisms, but we have recently shown that smooth muscle expression of Cul3Δ9 causes vascular dysfunction and elevation of arterial pressure via augmented RhoA/Rho‐kinase signaling. Whether other extra‐tubular mechanisms contribute remains unknown. We generated a model of selective and inducible expression of CUL3Δ9 in the endothelium using a construct (CAG‐Cul3Δ9), which induces the expression of Cul3Δ9 (and TdTomato reporter) in response to Cre‐recombinase. In primary aortic endothelial cells isolated from CAG‐CUL3Δ9 mice, CUL3Δ9 expression was robustly induced by adenovirus carrying cre recombinase gene. Cul3Δ9 acted in a dominant negative manner by interfering with expression and function of endogenous Cullin‐3, leading to impaired turnover of Cullin‐3 substrate protein phosphatase 2A (PP2A), a marked decrease in phosphorylated eNOS, and reduced nitric oxide (NO) bioavailability. Treatment with PP2A inhibitor Calyculin A (5 nM) rescued CUL3Δ9‐induced impairment of eNOS activity in these cells. Likewise, inhibition of Cullin activity using MLN4924 (1 μM) caused a marked reduction in phosphorylated eNOS and NO production in mouse lung endothelial cells (MLECs), while these changes were prevented by Calyculin A. To test the importance of endothelial Cul3 in vivo, we bred CAG‐Cul3Δ9 mice with Tek‐CRE ERT2 mice specifically expressing a tamoxifen inducible Cre‐recombinase in the endothelium. The specificity of transgene expression was confirmed by western blot analysis of sorted endothelial and non‐endothelial fractions of the aortas from the resultant mice (termed E‐Cul3Δ9) 4 weeks following tamoxifen injection. E‐Cul3Δ9 mice exhibited impaired endothelial‐dependent relaxation in the carotid artery (relaxation to acetylcholine at 30 μM: 69.9 ± 5.7 % in E‐Cul3Δ9 vs. 85.6 ± 4.3 % in control littermates, p<0.05, mean ± SEM), while no differences were seen in non‐endothelial dependent relaxation. Moreover, E‐Cul3Δ9 mice exhibited nocturnal hypertension as determined by radio telemetry (systolic pressure at 11pm, 143.7 ± 2.8 mmHg E‐Cul3Δ9 vs. 118.5 ± 3.4 control littermates, p<0.01) and arterial stiffening as indicated by elevated pulse wave velocity (4.0 ± 0.5 E‐Cul3Δ9 vs. 2.9 ± 0.2 control littermates, p<0.05). These data define a novel pathway involving Cullin‐3/PP2A/phopho‐eNOS in the regulation of endothelial function. Selective expression of the CUL3Δ9 mutation in the endothelium partially phenocopies the hypertension observed in CUL3Δ9 patients, suggesting that mutations in Cullin‐3 cause human hypertension in part through a vascular mechanism featured by impaired eNOS activity in endothelium. Support or Funding Information Grants from NIH and AHA