Role of Runx2 in Calcific Aortic Valve Disease in Mouse Models

Abstract

Background: Calcific aortic valve disease is common in the aging population and is characterized by the histological changes of the aortic valves including extracellular matrix remodeling, osteochondrogenic differentiation, and calcification. Combined, these changes lead to aortic sclerosis, aortic stenosis (AS), and eventually to heart failure. Runt-related transcription factor 2 (Runx2) is a transcription factor highly expressed in the calcified aortic valves. However, its definitive role in the progression of calcific aortic valve disease (CAVD) has not been determined. In this study, we utilized constitutive and transient conditional knockout mouse models to assess the molecular, histological, and functional changes in the aortic valve due to Runx2 depletion. Methods: Lineage tracing studies were performed to determine the provenance of the cells giving rise to Runx2+ osteochondrogenic cells in the aortic valves of LDLr^-/- mice. Hyperlipidemic mice with a constitutive or temporal depletion of Runx2 in the activated valvular interstitial cells (aVICs) and sinus wall cells were further investigated. Following feeding with a diabetogenic diet, the mice were examined for changes in gene expression, blood flow dynamics, calcification, and histology. Results: The aVICs and sinus wall cells gave rise to Runx2+ osteochondrogenic cells in diseased mouse aortic valves. The conditional depletion of Runx2 in the SM22α+ aVICs and sinus wall cells led to the decreased osteochondrogenic gene expression in diabetic LDLr^-/- mice. The transient conditional depletion of Runx2 in the aVICs and sinus wall cells of LDLr^-/-ApoB¹⁰⁰ CAVD mice early in disease led to a significant reduction in the aortic peak velocity, mean velocity, and mean gradient, suggesting the causal role of Runx2 on the progression of AS. Finally, the leaflet hinge and sinus wall calcification were significantly decreased in the aortic valve following the conditional and temporal Runx2 depletion, but no significant effect on the valve cusp calcification or thickness was observed. Conclusions: In the aortic valve disease, Runx2 was expressed early and was required for the osteochondrogenic differentiation of the aVICs and sinus wall cells. The transient depletion of Runx2 in the aVICs and sinus wall cells in a mouse model of CAVD with a high prevalence of hemodynamic valve dysfunction led to an improved aortic valve function. Our studies also suggest that leaflet hinge and sinus wall calcification, even in the absence of significant leaflet cusp calcification, may be sufficient to cause significant valve dysfunctions in mice.

Keywords: CAVD; RUNX2; aortic valve stenosis; calcification; osteochondrogenic differentiation; valve interstitial cells.

Figure 1

Lineage tracing of runt-related transcription factor 2 (Runx2)+ valvular interstitial cells (VICs) and sinus wall cells. (A) Schematic showing how Cre recombination leads to a permanent activation of the LacZ gene in SM22α+ VICs. (B–G) LDLr^−/−:β-Gal^SM22 mice were fed a diabetogenic (T2DM) diet for 28 weeks. Cartilaginous matrices (yellow collagen staining) in aortic valves were determined by Movat pentachrome staining (B,C). VICs and sinus wall cells were stained blue with X-gal (D–G). (D,E) Cell nuclei were counterstained with nuclear fast red. Note macrophage foam cells in the aortic side of the thickened valve leaflet that was not stained by X-gal (dashed line; D,E). CM, cardiomyocyte cells (E). Higher magnification (60x) image of boxed area in (D). Immunohistochemical staining of X-gal stained (blue) hinge (F) and cusps (G) of the aortic valve for Runx2 (brown, arrowheads). Inset in (F,G) are enlarged images of the boxed area. C, Cusp; H, Hinge.

Figure 2

Runx2 in SM22α+ VICs and sinus wall cells contributes to osteochondrogenic differentiation in the aortic valve. (A) LDLr^−/−:Runx2^ΔSM22 mice and LDLr^−/−:Runx2^f/f littermates were fed with a T2DM diet for 26 weeks. Aortic valves (dashed lines) were collected for Runx2 immunofluorescent staining. Graphs show the quantification of Runx2 intensity in leaflets (cusp + hinge) normalized to the area. (B) LDLr^−/−:Runx2^f/f and LDLr^−/−:Runx2^ΔSM22 mice were fed with T2DM for 26 weeks. Four to ten aortic valves were pooled to generate RNA. Runt-related transcription factor 2, osteocalcin (OCN), alkaline phosphatase (ALP), osteopontin (OPN), and chondrocyte markers, Collagen X (ColX), and MMP13, gene expression was determined in triplicate via real-time quantitative PCR (qRT-PCR). Differences between groups were analyzed by student's t-test. *p < 0.05. Data shown are mean ± SD. C, Cusp; H, Hinge.

Figure 3

Time course of de novo smooth muscle α-actin (SMA) expression in disease aortic valves. (A) Aortic valve sections from LDLr^−/−ApoB¹⁰⁰ mice on the T2DM diet labeled for SMA (red) and 6-diamidino-2-phenylindole (DAPI) (blue). Cusp and hinge regions are outlined with dashed lines. The SMA expression is observed in the valve leaflets in sections from 6 weeks through 10 weeks of the T2DM diet mice (white arrowheads). (B) Quantification of fluorescence intensity in SMA labeled sections from LDLr^−/−ApoB¹⁰⁰ mice on the T2DM diet for 6–12 weeks (n = 3 per time point). Differences between groups were analyzed by one-way ANOVA with Bonferroni's correction. *p < 0.05, **p < 0.005.

Figure 4

Runx2 depletion in the aortic valve. (A) Schematic tamoxifen-induced Cre recombination that leads to a temporal inactivation of Runx2 in SMA+ VICs and sinus valve cells. (B) PCR genotyping to amplify the loxP flanked sites to confirm Runx2 depletion in the aortic valve. IF of aortic valve sections from LDLr^−/−ApoB¹⁰⁰Runx2^f/f (C) and LDLr^−/−ApoB¹⁰⁰Runx2^ΔSMA (D) mice for Runx2. Green borders highlight the sinus wall region, white borders highlight the hinge region and blue borders highlight the cusp region. (E) Quantification of Runx2 intensity in valve sections in the sinus wall, hinge, and cusp regions (n = 15 per group). Differences between groups were analyzed by Mann-Whitney non-parametric tests. *p < 0.05, **p < 0.005. Data shown are mean ± SD, n = 15.

Figure 5

Runx2 depletion in SMA+ VICs and sinus walls improves aortic valve function. Representative image of aortic velocity (AoV) flow measurements for blood flow parameters and B mode image used for aortic valve area measurements (A). (B) Quantitative analysis of the mean velocity, AoV peak velocity, mean gradient, peak gradient, and aortic valve area between the 26 wk LDLr^−/−ApoB¹⁰⁰Runx2^f/f normal chow (NC, n = 18), T2DM diet LDLr^−/−ApoB¹⁰⁰Runx2^f/f (n = 25), and T2DM diet LDLr^−/−ApoB¹⁰⁰Runx2^ΔSMA(n = 20). The differences between groups were analyzed by one-way ANOVA with Bonferroni's correction *p < 0.05, **p < 0.005. Data shown are mean ± SD.

Figure 6

SMA-specific depletion of Runx2 had no significant effect on ventricular function. Representative images of long-axis B-mode (A) and M-mode (B) images were used for the assessment of the left ventricular function. (C) Quantitative analysis of ejection fraction, fractional shortening, left ventricle mass (normalized to body weight), and relative wall thickness (RWT), between the 26 weeks LDLr^−/−ApoB¹⁰⁰Runx2^f/f normal chow (NC, n = 18), T2DM diet LDLr^−/−ApoB¹⁰⁰Runx2^f/f (n = 25) and T2DM diet LDLr^−/−ApoB¹⁰⁰Runx2^ΔSMA (n = 20) groups. The differences between groups were analyzed by one-way ANOVA with Bonferroni's correction **p < 0.005. Data shown are mean ± SD.

Figure 7

Runx2 depletion decreased calcification in the aortic valve. (A) Representative images of valve sections labeled with osteosense from the LDLr^−/−ApoB¹⁰⁰Runx2^f/f and LDLr^−/−ApoB¹⁰⁰Runx2^ΔSMA mice to assess calcification. Boxed areas (dashed lines) within (A) represent the sinus wall (green), hinge (red), and cusp (teal). Higher magnification of boxed area from (A), representing the sinus wall (B), hinge (C), and cusp (D) in the aortic valve and their respective intensity quantification graphs (n = 15 per group). The differences between groups were analyzed by Mann-Whitney non-parametric tests. *p < 0.05. Data shown are mean ± SD.