Smooth muscle cell-specific runx2 deficiency inhibits vascular calcification

Abstract

Rationale: Vascular calcification is a hallmark of atherosclerosis, a major cause of morbidity and mortality in the United States. We have previously reported that the osteogenic transcription factor Runx2 is an essential and sufficient regulator of calcification of vascular smooth muscle cells (VSMC) in vitro.

Objective: To determine the contribution of osteogenic differentiation of VSMC to the pathogenesis of vascular calcification and the function of VSMC-derived Runx2 in regulating calcification in vivo.

Methods and results: SMC-specific Runx2-deficient mice, generated by breeding SM22α-Cre mice with the Runx2 exon 8 floxed mice, exhibited normal aortic gross anatomy and expression levels of SMC-specific marker genes. Runx2 deficiency did not affect basal SMC markers, but inhibited oxidative stress-reduced expression of SMC markers. High-fat-diet-induced vascular calcification in vivo was markedly inhibited in the Runx2-deficient mice in comparison with their control littermates. Runx2 deficiency inhibited the expression of receptor activator of nuclear factor κB ligand, which was accompanied by decreased macrophage infiltration and formation of osteoclast-like cells in the calcified lesions. Coculture of VSMC with bone marrow-derived macrophages demonstrated that the Runx2-deficient VSMC failed to promote differentiation of macrophages into osteoclast-like cells.

Conclusions: These data have determined the importance of osteogenic differentiation of VSMC in the pathogenesis of vascular calcification in mice and defined the functional role of SMC-derived Runx2 in regulating vascular calcification and promoting infiltration of macrophages into the calcified lesion to form osteoclast-like cells. Our studies suggest that the development of vascular calcification is coupled with the formation of osteoclast-like cells, paralleling the bone remodeling process.

Figure 1. SMC-specific Runx2 deficiency inhibits VSMC calcification in vitro

(A) Western blot analysis of the production of Runx2 in VSMC from Runx2^f/f and Runx2^{ΔE8/ΔE8 SMC} mice (Runx2^Δ/Δ). Lamin B was used as a loading control. (B) In vitro calcification of VSMC from Runx2^f/f and Runx2^{ΔE8/ΔE8 SMC} mice. VSMC were exposed to osteogenic media with or without H₂O₂ (0.4mM) for 3 weeks, and calcification was determined by Alizarin Red and Von Kossa staining. Representative images of stained dishes from 3 independent experiments are shown. (C) Quantification of calcium content. In parallel experiments, VSMC were exposed to osteogenic media with or without H₂O₂ (0.4mM) for 3 weeks, and total calcium content in the cell lysates was quantified by Arsenazo III method. Results shown are normalized by total protein amount (n=3, *p=0.003). (D) Real-time PCR analysis of osteogenic markers in VSMC exposed to osteogenic media with or without H₂O₂ (0.4mM) for 2 weeks. The expression of each gene in Runx2^f/f VSMC at control conditions (first bar in each group) is defined as 1(n=3, *p <0.05 compared with control conditions).

Figure 2. SMC-specific Runx2 deficiency inhibits atherosclerotic calcification in vivo

Runx2^f/f and Runx2^{ΔE8/ΔE8 SMC} (Runx2^Δ/Δ) mice in the ApoE^−/− background were fed normal chow (NC) or HFD for 30 weeks. (A) Vascular calcification in the aortic roots. Frozen aortic root sections were obtained throughout the aortic valve area. Representative images of sections stained with Alizarin Red, to detect calcification, are shown. (B) Quantification of calcification at the aortic root sections. Calcification was quantified using ImageJ software (NIH Bethesda, MD). Results presented are the percentage of positively stained areas in the total area of aortic roots. Bar values are means ± SD. The numbers of mice analyzed in each group are shown in each bar (*p<0.001). (C) ALP activity in the aortic roots. Representative images of ALP staining on consecutive sections shown in (A) are shown. Eosin staining was used for histology (pink). Arrows indicate the areas that are positive for ALP (dark purple). (D) Total calcium content in the descending aortic tissues. The calcium content was quantified by Arsenazo III method. Results shown are normalized by total protein amount (*p<0.005).

Figure 3. SMC-specific Runx2 deficiency inhibits HFD-induced RANKL expression and macrophage infiltration

Frozen aortic root sections from HFD-fed Runx2^f/f:ApoE^−/− and Runx2^{ΔE8/ΔE8 SMC}:ApoE^−/− mice (Runx2^Δ/Δ:ApoE^−/−) were stained with antibodies for (A) RANKL (a, b) and (B) CD68 (c, d). Representative images of stained sections are shown. Higher magnification images of the boxed areas of a, b, c & d are shown at the right of each image (a1, b1, c1 & d1). (C) Quantification of RANKL and CD68 immuno-positive areas in aortic root sections. Positively stained area was quantified using ImageJ software (NIH Bethesda, MD). Results presented are the percentage of positively stained areas as the total area of aortic roots. Bar values are means ± SD. The numbers of mice analyzed in each group are shown in each bar (*p<0.005). (D) Real-time PCR analysis of the expression of RANKL andCD68 in descending aortas from HFD fed Runx2^f/f:ApoE^−/− (n= 4 mice) and Runx2^{ΔE8/ΔE8 SMC}:ApoE^−/− mice (Runx2^Δ/Δ:ApoE^−/−, n=3 mice). The expression of RANKL and CD68 was normalized by the expression of GAPDH, and is defined as 1 in Runx2^f/f:ApoE^−/− (*p<0.001).

Figure 4. SMC-specific Runx2 deficiency inhibits the formation of TRAP-positive cells in calcified atherosclerotic lesions

(A) Immunofluorescent staining of RANKL and TRAP in frozen aortic root sections from HFD-fed Runx2^f/f:ApoE^−/−(upper two rows) and Runx2^{ΔE8/ΔE8 SMC}:ApoE^−/−(Runx2^Δ/Δ:ApoE^−/−) mice are shown. Images of Alizarin red staining indicating calcified areas (arrows) are shown in the left panel. Representative images of medium (first row) and high (second row) levels of calcification in the control mice are shown. (B) Quantification of TRAP-positive areas shown in (A). The positively stained areas were quantified using ImageJ software (NIH Bethesda, MD). Results presented are the percentage of positively stained areas in the total area of aortic roots. The numbers of mice analyzed in each group are shown in each bar (*p<0.005). (C) Correlation of TRAP-positive area with calcified area. Shapiro–Wilk normality test followed by Spearman correlation analysis was conducted to determine the correlation of percent TRAP-positive area with calcification area.

Figure 5. SMC-specific Runx2 deficiency inhibits formation of TRAP-positive cells in a SMC/macrophage co-culture system

(A) Real-time PCR analysis of RANKL expression in VSMC exposed to osteogenic media with or without H₂O₂ (0.4mM) for 2 weeks. The expression of RANKL in Runx2^f/f VSMC in the control condition (first bar) is defined as 1 (n=3, *p <0.05 compared with control condition). (B) RANKL protein expression in VSMC from separate sets of experiments as (A). The expression of RANKL protein in cell lysates was quantified by ELISA (n=3, *p <0.05 compared with control). (C) Co-culture of H₂O₂-treated VSMC with BMMs. BMMs were plated on top ofRunx2f/f (a & c) or Runx2^{ΔE8/ΔE8 SMC} (Runx2^Δ/Δ, b & d) VSMC that were pre-incubated in osteogenic media with H₂O₂ (0.4 mM) for 2 weeks in the absence (upper panel, a & b) or presence of RANKL (25 ng/ml, c & d) for 1 week, and stained for TRAP. Representative images of stained wells from 2 independent experiments performed in duplicate are shown. Higher magnification images of the boxed areas of a, b, c & d are shown at the right of each image (a1, b1, c1 & d1).

Figure 6. Model for the functional contribution of Runx2 to vascular calcification

Oxidative stress induces the expression of Runx2 in VSMC, which directly promotes osteogenic differentiation and calcification of VSMC in vitro and in vivo. In addition, increased Runx2 directly upregulates the expression of RANKL in VSMC by binding to the RANKL promoter , which in turn promotes infiltration of macrophages and formation of vascular osteoclasts in the calcified atherosclerotic lesions.