Impaired Vitamin D Signaling in Endothelial Cell Leads to an Enhanced Leukocyte-Endothelium Interplay: Implications for Atherosclerosis Development

Abstract

Endothelial cell activation leading to leukocyte recruitment and adhesion plays an essential role in the initiation and progression of atherosclerosis. Vitamin D has cardioprotective actions, while its deficiency is a risk factor for the progression of cardiovascular damage. Our aim was to assess the role of basal levels of vitamin D receptor (VDR) on the early leukocyte recruitment and related endothelial cell-adhesion-molecule expression, as essential prerequisites for the onset of atherosclerosis. Knockdown of VDR in endothelial cells (shVDR) led to endothelial cell activation, characterized by upregulation of VCAM-1, ICAM-1 and IL-6, decreased peripheral blood mononuclear cell (PBMC) rolling velocity and increased PBMC rolling flux and adhesion to the endothelium. shVDR cells showed decreased IκBα levels and accumulation of p65 in the nucleus compared to shRNA controls. Inhibition of NF-κB activation with super-repressor IκBα blunted all signs of endothelial cell activation caused by downregulation of VDR in endothelial cells. In vivo, deletion of VDR led to significantly larger aortic arch and aortic root lesions in apoE-/- mice, with higher macrophage content. apoE-/-VDR-/-mice showed higher aortic expression of VCAM-1, ICAM-1 and IL-6 when compared to apoE-/-VDR+/+ mice. Our data demonstrate that lack of VDR signaling in endothelial cells leads to a state of endothelial activation with increased leukocyte-endothelial cell interactions that may contribute to the more severe plaque accumulation observed in apoE-/-VDR-/- mice. The results reveal an important role for basal levels of endothelial VDR in limiting endothelial cell inflammation and atherosclerosis.

Fig 1. Knockdown of vitamin D receptor in endothelial cells leads to endothelial cell activation.

Responses of leukocyte rolling velocity (A), rolling flux (B) and adhesion (C) to endothelial monolayer were monitored using 40x objective lens of an inverted microscope connected to a video camera. Data presented are mean ± SEM, n≥4, *p<0.05, **p<0.01 vs. corresponding shc002 cells. (D) Whole cell lysates were immunoblotted with antibodies against VCAM-1, ICAM-1 and VDR. The same samples were reprobed with tubulin to ensure equal loading. Representative Western blot (D) and quantitative analysis (E) demonstrate decrease in VDR expression in EA.hy926 infected with pLKO.1-VDR (shVDR) compared with pLKO.1-SHC002 (shc002 control). Data presented are mean ± SEM from 3 independent experiments, ***p<0.005 vs. shc002. Downregulation of VDR in endothelial cells resulted in increased VCAM-1 (D, F) and ICAM-1 (D, G) levels compared to control cells. *p<0.05 vs. control.

Fig 2. VDR knockdown-induced endothelial cell activation is mediated via NF-κB signaling.

EA.hy926 cells (shc002 or shVDR) were untreated (A, B, C) or treated with IκB kinase (IKK) inhibitor PS-1145 (D) for 24 hours. (A, B, D) Whole cell lysates were immunoblotted with antibodies against VDR, IκB-α, VCAM-1 and ICAM-1. The same samples were reprobed with tubulin to ensure equal loading. (C) Western blot analysis of p65 levels in cytosolic and nuclear extracts isolated from shc002 and shVDR endothelial cells. The nuclear protein Histone 1, which is absent in the cytosolic fraction, served as a nuclear protein loading control. (D) Representative Western blot and quantitative analysis (G, H), *p<0.05 vs. shc002 or shVDR, respectivey (E) Real time PCR analysis of IL-6 mRNA in shc002 and shVDR endothelial cells. Data presented are mean ± SEM from 3 independent experiments. **p<0.01 vs. shc002. (F) Determination of IL-6 secretion into the medium by ELISA. Endothelial cells were grown in normal growth media for 72 hours. IL-6 production was determined using a human IL-6 HS ELISA kit. Data presented are mean ± SEM from 3 independent experiments. *p<0.05 vs. shc002.

Fig 3. Super-repressor IκBα reduces VDR knockdown-induced endothelial cell activation.

Responses of leukocyte rolling velocity (A), rolling flux (B) and adhesion (C) to endothelial monolayer were monitored using 40x objective lens of an inverted microscope connected to a video camera. Data presented are mean ± SEM, n≥4, **p<0.01, ***p<0.005, *#p<0.001. (D) Whole cell lysates were immunoblotted with antibodies against VCAM-1, ICAM-1, IκBα and VDR. The same samples were reprobed with tubulin to ensure equal loading. Representative Western blot (D) and quantitative analysis (E, F) *p<0.05 (VCAM-1) or **p<0.01 (ICAM-1) vs. shc002/Control; **p<0.01 (VCAM-1) or *#p<0.001 (ICAM-1) vs. shVDR/Control. (G) Real time PCR analysis of IL-6 mRNA. Data presented are mean ± SEM from 3 independent experiments. *p<0.05. (H) Determination of IL-6 secretion into the medium by ELISA. Endothelial cells were grown in normal growth media for 72 hours. IL-6 production was determined using a human IL-6 HS ELISA kit. Data presented are mean ± SEM from 3 independent experiments. *p<0.05, *#p<0.001, **#p<0.0001.

Fig 4. VDR deletion accelerates atherosclerosis in apoE^-/- mice.

apoE^-/- and apoE^-/-VDR^-/- mice were fed a high fat-rescue diet (HFRD) for 8 weeks. (A) Aortas were stained with Oil-Red O, opened longitudinally and pinned onto a black wax surface. Luminal side of the aorta is displayed, showing red lipid-rich atherosclerotic lesions within aortic arch and thoracic aorta. (B) Quantitative analysis of atherosclerotic lesion size within aortic arch and thoracic aorta. Data presented are mean ± SEM of 14–15 mice/group. *p<0.05 vs. apoE^-/-. (C) Quantitative analysis of atherosclerotic lesion size within aortic arch in males and females of apoE^-/- and apoE^-/-VDR^-/- mice. Data presented are mean ± SEM of 10 mice/group *p<0.05 vs. apoE^-/-. (D) Quantitative analysis of atherosclerotic lesion size within aortic root. Five cross sections (spanning 640 μm of the aortic root) were analyzed per mouse, as explained in Methods. Data presented are mean ± SEM of n = 10 mice/group. ##p<0.0005.

Fig 5. VDR ablation affects atheroma composition in apoE^-/- mice.

Tissue sections of aortic arch (A, B) and aortic root (D, E) from apoE^-/- and apoE^-/-VDR^-/- mice were stained with anti-Mac-3 to quantify plaque macrophage content. (C) Staining intensity and % of Mac-3 positive cells in the aortic arch were graded semiquantitatively and histological scores were obtained from every sample, as explained in Materials and Methods. (F) Quantification of Mac-3 staining in the aortic root is shown as percentage of stained area relative to the area occupied by atheroma. The photomicrographs (20x magnification) show representative images of aortic arch and aortic root sections. Differences between groups were evaluated by the Student t test. Data presented are mean ± SEM of 10 mice/group. *p<0.05, *#p<0.001 vs. apoE^-/- mice.

Fig 6. VDR deletion increases aortic and systemic inflammatory responses.

(A) Effects of VDR deletion on proinflammatory cytokine production. Circulating serum levels of interleukin (IL-6) were determined by ELISA. Data presented are mean ± SEM of n = 10 mice/group. *p<0.05 vs. apoE^-/-. (B, C, D, E, F) Real time PCR analysis of VCAM-1, ICAM-1 and IL-6 mRNA in the aortic tissue of apoE^-/- and apoE^-/-VDR^-/- mice (B, C, D) and VDR^+/+ and VDR^-/- mice (E, F). Data presented are mean ± SEM. (B, C) *p<0.05 vs. apoE^-/-; (E, F) *p<0.05 vs. VDR^+/+ mice. (G) Enhanced VCAM-1 and ICAM-1 expression in the vascular endothelium of apoE^-/-VDR^-/- mice. Aortic root lesions from apoE^-/- and apoE^-/-VDR^-/- mice were stained with anti-VCAM-1 and anti-ICAM-1. CD31 was used for endothelial cell staining. Representative images are shown in G. Magnification 20x.