SIRT6 protects vascular smooth muscle cells from osteogenic transdifferentiation via Runx2 in chronic kidney disease

Abstract

Vascular calcification (VC) is regarded as an important pathological change lacking effective treatment and associated with high mortality. Sirtuin 6 (SIRT6) is a member of the Sirtuin family, a class III histone deacetylase and a key epigenetic regulator. SIRT6 has a protective role in patients with chronic kidney disease (CKD). However, the exact role and molecular mechanism of SIRT6 in VC in patients with CKD remain unclear. Here, we demonstrated that SIRT6 was markedly downregulated in peripheral blood mononuclear cells (PBMCs) and in the radial artery tissue of patients with CKD with VC. SIRT6-transgenic (SIRT6-Tg) mice showed alleviated VC, while vascular smooth muscle cell-specific (VSMC-specific) SIRT6 knocked-down mice showed severe VC in CKD. SIRT6 suppressed the osteogenic transdifferentiation of VSMCs via regulation of runt-related transcription factor 2 (Runx2). Coimmunoprecipitation (co-IP) and immunoprecipitation (IP) assays confirmed that SIRT6 bound to Runx2. Moreover, Runx2 was deacetylated by SIRT6 and further promoted nuclear export via exportin 1 (XPO1), which in turn caused degradation of Runx2 through the ubiquitin-proteasome system. These results demonstrated that SIRT6 prevented VC by suppressing the osteogenic transdifferentiation of VSMCs, and as such targeting SIRT6 may be an appealing therapeutic target for VC in CKD.

Keywords: Cardiovascular disease; Cell Biology; Chronic kidney disease; Ubiquitin-proteosome system; Vascular Biology.

Figure 1. Low level of SIRT6 expression was associated with increased risk of vascular calcification.

(A) The qPCR showed expression of SIRT1-7 in WT VSMCs with different calcification statuses. SIRT4 was not detected in VSMCs (n = 4 per group). Data were expressed as mean ± SD, *P < 0.05. (B) SIRT6 mRNA levels in PBMCs from patients with CKD with (n = 27) or without (n = 12) VC. Data were expressed as mean ± SD. (C) Correlation between the SIRT6 mRNA level and VC scores in patients with CKD (n = 39, the Pearson’s correlation coefficient R value and the P value are shown). (D) von Kossa assay and IF staining for SIRT6 in radial arteries sections from hemodialysis patients with CKD (n = 4 per group). Scale bars: von Kossa 100 μm; IF 50 μm. (E) The bars showing SIRT6 protein expression (mean ± SD; n = 4 per group; AU) in nuclei of aortic tissues between patients with CKD with and without VC. Statistical significance was assessed using 1-way ANOVA followed by Dunnett’s test (A) and 2-tailed t tests (B and E).

Figure 2. SIRT6 attenuated VC.

(A) Computed tomography (CT) images showing calcification in the abdominal aorta. The green arrows and circle indicated the calcification in abdominal aorta of the WT mouse (n = 12 per group). The bar chart shows the relative VC Agatston score (fold change) of mouse aortas. Scale bars: 10 mm. (B) Representative von Kossa staining of abdominal aorta sections (n = 12 per group). Scale bars: 100 μm. (C) Western blot shows SIRT6 protein in abdominal aorta was reduced in VC. (D and E) VSMCs were exposed to Pi (3.0 mM) for 7 days and then stained for mineralization by Alizarin red (D), and the quantitative analysis of calcium content (E) and ALP (F) were detected (n = 3 per group). (G) SIRT6 protein expression was reduced in WT and SIRT6-Tg VSMCs in response to Pi (3.0 mM) treatment (n = 4 per group). (H–J) WT and SIRT6-Tg VSMCs were pretransfected with siSIRT6 or si-negative control (siNC) and then exposed to Pi (3.0 mM) for 7 days. VSMCs were stained for mineralization by Alizarin red S (H), and calcium content (I) and ALP (J) were quantified (n = 3 per group). Statistical significance was assessed using 1-way ANOVA followed by Dunnett’s test (A, C–F, I, and J). *P < 0.05. All values are mean ± SD.

Figure 3. SIRT6 suppresses osteogenic transdifferentiation of VSMCs via regulation of Runx2.

(A) Expression levels of α-SMA and OPN in abdominal arteries of indicated groups were determined by IF staining (n = 4 per group). Scale bars: 50 μm. (B) Western blot analysis of osteogenic and contractile property factors expression in abdominal arteries (n = 3 per group). (C) Analysis of osteogenic and contractile property factor expression in WT and SIRT6-Tg VSMCs after Pi (3.0 mM) treatment by Western blot (n = 4 per group). (D) VSMCs were pretransfected with siSIRT6 or siNC, and then incubated with Pi (3.0 mM) for 7 days, and the downstream osteogenic markers (OPN, OCN) and contractile property markers (α-SMA, SM22α) were analyzed by Western blot (n = 4 per group). (E) Runx2 expression was analyzed in WT and SIRT6-Tg VSMCs after Pi (3.0 mM) treatment by Western blot (n = 4 per group). (F–H) SIRT6-Tg VSMCs were pretransfected with Runx2 plasmid or vector plasmid, and then exposed to Pi (3.0 mM) for 7 days. The expression of SIRT6 and Runx2 were analyzed by Western blot (F). VSMCs were stained for mineralization by Alizarin red S (G), and osteogenic markers (OPN, OCN) and contractile property markers (α-SMA, SM22α) were analyzed by qPCR (n = 3 per group) (H). Statistical significance was assessed using 1-way ANOVA followed by Dunnett’s test (H). *P < 0.05. All values are mean ± SD.

Figure 4. SIRT6 deacetylates Runx2.

(A) Representative IF images showing the colocalization of SIRT6 and Runx2. Scale bars: 50 μm. (B) Anti-SIRT6 IP followed by Western blot with anti-Runx2 or anti-SIRT6 antibody in SIRT6-Tg VSMCs after treatment with Pi (3.0 mM) for 7days. Anti-rabbit IgG IP was used as a negative control. (C) Anti-Runx2 IP in SIRT6-Tg VSMCs after treatment with Pi (3.0 mM) for 7days. Western blot was carried out with anti-SIRT6 or anti-Runx2 antibody. Anti-mouse IgG IP was used as a negative control. (D) The anti-HA IP and anti- flag IP followed by Western blot with anti-HA or anti-flag antibody in HEK-293T cells infected with HA-Runx2 plasmid, flag-SIRT6 plasmid, or both. Anti-rabbit IgG IP was used as a negative control. (E) WT and SIRT6-Tg VSMC lysates were immunoprecipitated with anti-Runx2 antibody and immunoblotted with anti-acetylated lysine antibody. (F) HEK-293T cells were infected with HA-Runx2 plasmid, flag-SIRT6 plasmid, or both. The anti-HA IP followed by Western blot with anti-acetylated lysine antibody and anti-HA antibody. (G) SIRT6-Tg VSMCs were pretransfected with siSIRT6 or siNC together with Pi (3.0 mM) for 7 days and OSS-128167 or DMSO were incubated with Pi (3.0 mM) for 7 days. The cell lysates were immunoprecipitated with anti-Runx2 antibody and immunoblotted with anti-acetylated lysine antibody and anti-Runx2 antibody. All the above experimental processing were duplicated 3 times.

Figure 5. SIRT6 promotes Runx2 degradation via the ubiquitin-proteasome system.

(A) WT and SIRT6-Tg VSMCs were treated with Pi (3.0 mM) for 7 days and incubated with the protein translation inhibitor CHX (0.2 mM) for the indicated times before harvest, followed by immunoblotting with the anti-Runx2 antibody and anti-GAPDH anti-body. The curve shows the stability of Runx2 protein. (B and C) SIRT6 was decreased in primary VSMCs via siRNA (B) or specific inhibitor (C) together with Pi (3.0 mM) incubation for 7 days. The protein translation inhibitor CHX (0.2 mM) was added for indicated times before harvest, followed by immunoblotting with the anti-Runx2 antibody and anti-GAPDH antibody. The curve shows the stability of Runx2 protein. (D and E) SIRT6-Tg VSMCs were incubated with Pi (3.0 mM) together with the leupeptin (1.5 μM) (D) or MG132 (10 μM) (E) for 7 days, and then the protein translation inhibitor CHX (0.2 mM) was added for the indicated times before harvest, followed by immunoblotting with the anti-Runx2 antibody and anti-GAPDH antibody. The curve shows the stability of Runx2 protein. (F) WT and SIRT6-Tg VSMC lysates were immunoprecipitated with anti-Runx2 antibody and immunoblotted with anti-ubiquitin (anti-Ub) antibody. (G) HEK-293T cells were transfected with His-Ub together with HA-Runx2 plasmid, flag-SIRT6 plasmid, or both. The anti-HA IP was followed by Western blot with anti-Ub antibody and anti-HA antibody. (H) SIRT6-Tg VSMCs were pretransfected with siSIRT6 or siNC together with Pi (3.0 mM) for 7 days, and OSS-128167 or DMSO were incubated with Pi (3.0 mM) for 7 days. The cell lysates were immunoprecipitated with anti-Runx2 antibody and immunoblotted with anti-Ub antibody and anti-Runx2 antibody. Statistical significance was assessed using 2-way ANOVA (A–E). All the above experimental processing was duplicated 3 times.

Figure 6. SIRT6 mediates Runx2 nuclear export depending on XPO1.

(A) Runx2 IF staining was performed in abdominal arteries. Scale bar: 50 μm. Statistical significance was assessed using 2-tailed t tests, *P < 0.05. (B) VSMCs were incubated with Pi for 7 days. IF staining was performed for Runx2. Scale bars: 50 μm. (C) VSMCs were incubated with Pi for 7 days. Cells were harvested and immunoblotted for the indicated proteins. (D) SIRT6-Tg VSMCs were incubated with Pi for 7 days after posttransfection of siSIRT6. Cells were harvested and immunoblotted for the indicated proteins. (E) SIRT6-Tg VSMCs were incubated with Pi together with nicotinamide for 7 days. Cells were harvested and immunoblotted for the indicated proteins. (F) SIRT6-Tg VSMCs were transfected with shRNA targeting XPO1, XPO4, XPO7, or their vector control, and then incubated with Pi for 7 days after transfection. Nuclear extracts were immunoblotted for Runx2. (G and H) SIRT6-Tg VSMCs were incubated with Pi together with Leptomycin A (0.5 nM) for 7 days. Cells were harvested and immunoblotted for the indicated proteins (G). IF staining was performed for Runx2. Scale bars: 50 μm (H). (I) Anti-XPO1 IP followed by Western blot with anti-Runx2 or anti-XPO1 antibody in SIRT6-Tg VSMCs after treatment with Pi for 7 days. Anti-rabbit IgG IP was used as negative control. (J) Anti-Runx2 IP in SIRT6-Tg VSMCs after treatment with Pi for 7 days. Western blot was carried out with anti-XPO1 or anti-Runx2 antibody. Anti-mouse IgG IP was used as negative control. (K) SIRT6-Tg VSMCs were incubated with Pi together with Leptomycin A for 7 days, and then CHX (0.2 mM) was added for the indicated times before harvest, followed by immunoblotting for the indicated proteins. (L) Curve shows the stability of Runx2 and was assessed using 2-way ANOVA. Pi treatment is 3.0 mM. All the above experimental processing was duplicated 3 times.

Figure 7. Nuclear export of Runx2 is a key component of SIRT6 vascular calcification suppressor function.

(A–C) WT and SIRT6-Tg VSMCs were incubated with Pi (3.0 mM) together with Leptomycin A for 7 days. VSMCs were stained for mineralization by Alizarin red S (A), and calcium content (B) and ALP (C) were quantified (n = 3 per group). (D and E) The osteogenic markers (OPN, OCN) and the contractile property markers (α-SMA, SM22α) were analyzed by qPCR for the WT (D) and SIRT6-Tg VSMCs (E) mouse being incubated with Pi (3.0mM) together with Leptomycin A for 7 days (n = 3 per group). Statistical significance was assessed using 1-way ANOVA followed by Dunnett’s test (B–E). *P < 0.05. All values are mean ± SD.