TNF-α regulates the proteolytic degradation of ST6Gal-1 and endothelial cell-cell junctions through upregulating expression of BACE1

Abstract

Endothelial dysfunction and monocyte adhesion to vascular endothelial cells are two critical steps in atherosclerosis development, and emerging evidence suggests that protein sialylation is involved in these processes. However, the mechanism underlying this phenomenon remains incompletely elucidated. In this study, we demonstrated that treatment with the proinflammatory cytokine TNF-α disrupted vascular endothelial cell-cell tight junctions and promoted monocyte endothelial cell adhesion. Western blotting and Sambucus nigra lectin (SNA) blotting analyses revealed that TNF-α treatment decreased α-2, 6-sialic acid transferase 1 (ST6Gal-I) levels and downregulated VE-Cadherin α-2, 6 sialylation. Further analysis demonstrated that TNF-α treatment upregulated β-site amyloid precursor protein enzyme 1 (BACE1) expression, thus resulting in sequential ST6Gal-I proteolytic degradation. Furthermore, our results revealed that PKC signaling cascades were involved in TNF-α-induced BACE1 upregulation. Together, these results indicated that the proinflammatory cytokine TNF-α impairs endothelial tight junctions and promotes monocyte-endothelial cell adhesion by upregulating BACE1 expression through activating PKC signaling and sequentially cleaving ST6Gal-I. Thus, inhibition of BACE1 expression may be a new approach for treating atherosclerosis.

Figure 1. TNF-α impaired vascular epithelial cell tight junctions and increased vascular epithelial cell adhesion.

(A) Cell viability was evaluated with CCK-8 assays. EA.hy926 cells were treated with 0, 10, 50, and 100 ng/ml TNF-α for 24 h and treated with WST-8 for 1 h at 37 °C. The absorbance at 450 nm was then measured with a microplate reader. ns, not significant versus untreated cells. (B) Transmission electron microscopy characterization of EA.hy926 cells with or without TNF-α treatment (50 ng/ml) for 24 h was performed (10000 × magnification). (C) Confocal immunofluorescence analysis of EA.hy926 cells treated with TNF-α using VE-Cadherin rabbit mAb (red) and DAPI (blue). (D) Monocyte adhesion to endothelial cells was quantified via monocyte adhesion assay. Calcein AM-labeled THP-1 cells were incubated with EA.hy926 cells in a 96-well plate for 1 h at 37 °C. The plate was then washed three times with PBS, and the fluorescence was measured. The data are the mean ± SD of three independent assays (**p < 0.01). Mock, EA.hy926 without treatment.

Figure 2. ST6Gal-I and total protein α-2, 6 sialylation levels were decreased in EA.hy926 cells after treatment with the proinflammatory cytokine TNF-α.

(A) Western blotting and lectin blotting assays were conducted to characterize the levels of ST6Gal-I and total protein α-2, 6 sialylation in EA.hy926 cells treated with different concentrations of TNF-α. Actin was used as a loading control. The relative changes in the protein bands were measured, and the mock control was set as 100%, as shown. One typical result from three independent experiments is shown. ns, not significant, *p < 0.05 and **p < 0.01. After 0, 10, and 50 ng/ml TNF-α treatment for 24 h, immunoblotting was performed with anti-ST6Gal-I or SNA. An equal amount (30 μg) of total lysate from each sample was resolved by 10% SDS-PAGE, with β-actin serving as a control. The relative changes in the protein bands were measured, and the control was set to 100%, as shown in the data. One typical result from three independent experiments is shown. (B) EA.hy926 cells were treated with TNF-α (0, 50 ng/ml) or left untreated for 24 h. The cells then were incubated with FITC-labeled SNA and DAPI for 1 h. α-2, 6 sialylation expression was examined by confocal fluorescence microscopy (200 × magnification). Mock, EA.hy926 without treatment. (C) After TNF-α (0,50 ng/ml) treatment for 24 h, the cells were incubated with FITC-labeled SNA for 1 h. Flow cytometry assay was then used to determine α-2, 6 sialylation expression. The data are the mean ± SD of three independent assays (*p < 0.05, **p < 0.01). Mean fluorescence intensity (MFI) was quantified using BD FACS Software. NC, EA.hy926 cells unstained with FITC-labeled SNA. Mock, EA.hy926 without treatment.

Figure 3. VE-Cadherin α-2, 6 sialylation levels were decreased under proinflammatory conditions.

(A) VE-cadherin levels in EA.hy926 cells treated with TNF-α (0, 10, 20, and 50 ng/ml) for 24 h were examined via western blot assays. (B) VE-Cadherin α-2, 6 sialylation levels in EA.hy926 cells were decreased after TNF-α (50 ng/ml) treatment for 24 h. VE-cadherin was immunoprecipitated with anti-human VE-cadherin polyclonal antibodies. VE-cadherin immunoprecipitates were analyzed via SNA blotting using biotinylated SNA and HRP-labeled Streptavidin. (C) Proteins modified by α-2, 6 sialylation in EA.hy926 cells were recognized via biotinylated SNA blotting and then precipitated with Dynabeads Streptavidin. VE-Cadherin levels in the precipitates were analyzed by western blotting using antibodies against VE-cadherin. The data are the mean ± SD of three independent assays (**p < 0.01).

Figure 4. TNF-α regulates ST6Gal-I levels through BACE1.

(A) EA.hy926 cells were treated with TNF-α for 24 h, and then BACE1, ST6Gal-I and total protein α-2, 6 sialylation levels were detected by western blotting and SNA blotting. Actin was used as a loading control. The data are the mean ± SD of three independent assays (**p < 0.01, *p < 0.05). (B) The cells were pretreated with 1 μM BACE1 inh. for 24 h and then stimulated with TNF-α (50 ng/ml) for 24 h and analyzed by western blotting and SNA blotting. The data are the mean ± SD of three independent assays (**p < 0.01). ns, not significant. (C) The cells were pretreated with 1 μM BACE1 inh. for 24 h and then stimulated with TNF-α (50 ng/ml) for 24 h. The cells were incubated with FITC-labeled SNA and DAPI for 1 h. Proteins α-2, 6 sialylation levels were examined under confocal fluorescence microscopy (600 × magnification). (D) EA.hy926 cells were inoculated with a BACE1-overexpressing lentivirus and selected with puromycin for 4 weeks. BACE1, ST6Gal-I and total protein α-2, 6 sialylation levels were then detected by western blotting and SNA blotting. EA.hy926/Mock and EA.hy926 cells transduced with control lentivirus. EA.hy926/BACE1 and EA.hy926 cells transduced with BACE1-overexpressed lentivirus. (E) Total protein α-2, 6 sialylation levels in cells overexpressing BACE1 were examined by SNA staining. Because they were transduced with lentiviruses carrying an EGFP marker gene, the EA.hy926 cells were sequentially incubated with biotinylated SNA and Avidin-RBITC, which specifically binds biotinylated SNA. RBITC fluorescence signals were detected under confocal fluorescence microscopy. (F) Sialylated VE-cadherin levels were decreased after BACE1 overexpression. Cell lysates were immunoprecipitated with anti-VE-cadherin antibody and analyzed by SNA blotting. The data are the mean ± SD of three independent assays (**p < 0.01).

Figure 5. Tight junctions and cell adhesion were affected by BACE1.

(A) BACE1 inh. (1 μM) was administered as pretreatment for 24 h, and then the cells were stimulated with TNF-α (50 ng/mL) for 24 h. EA.hy926 cell tight junctions were detected by transmission electron microscopy (10000 × magnification). (B) EA.hy926 cell tight junctions were also detected by confocal immunofluorescence analysis using anti-VE-cadherin antibody (Red). Blue, DAPI. (C) After BACE1 inh. (1 μM) treatment, cell adhesion assay was performed in EA.hy926 cells (100 × magnification). The data are the mean ± SD of three independent assays (**p < 0.01). (D) Transmission electron microscopy analysis was used to identify BACE1-overexpressing EA.hy926 cell tight junctions (10000 × magnification). (E) Immunofluorescent analysis of the tight junction protein VE-cadherin. Red, VE-cadherin. Blue, DAPI. (F) Monocyte adhesion to endothelial cells was quantified by monocyte adhesion assay. Dil-labeled THP-1 cells were incubated with endothelial cells in 96-well plates for 1 h at 37 °C. The plates were then washed three times with PBS, and the fluorescence was measured. The data are the mean ± SD of three independent assays (**p < 0.01).

Figure 6. BACE1 expression induced by the proinflammatory cytokine TNF-α is PKC-dependent.

(A) EA.hy926 cells were treated with TNF-α (50 ng/ml) for the indicated time periods. Western blotting was conducted to characterize phospho-PKC δ levels. Quantification of normalized densities for p-PKC δ and PKC δ is shown. The graphs represent the relative activity of these kinases for three independent experiments. ns, ns, not significant versus untreated cells. **p < 0.01 versus untreated cells. (B) EA.hy926 cells were stimulated with TNF-α (50 ng/ml) with or without PKC inhibitor (10 μM) or ERK inhibitor (20 μM) for 24 h, and MEK1/2, phospho-MEK1/2, ERK1/2 and phospho-ERK1/2 levels were analyzed via western blotting assays. ns, not significant. **p < 0.01. (C,D) Effect of the PKC pathway on BACE1 expression in EA.hy926 cells. The cells were pretreated with a PKC activator (10 μM) for 6 h or PKC inhibitor (10 μM) for 30 min and then incubated with TNF-α (50 ng/ml) for 24 h. BACE1 expression levels in EA.hy926 cells were analyzed via western blotting. ns, not significant. **p < 0.01.