Toll-like receptor-3 contributes to the development of aortic valve stenosis

Figures

• 

  Fig. 1. TLR3 is expressed in human aortic valves and TLR3 stimulation activates VICs: Toll-like-receptor 3 (TLR3) expression in human stenotic aortic valves (AS) was visualized via immunofluorescence staining (A) and quantitative PCR compared to valves explanted due to aortic valve regurgitation (AR) (B) (two-sided T test, mean ± SEM). Human valvular interstitial cells (VICs) were stimulated with the TLR3 agonist Polyinosinic:polycytidylic acid (polyIC). Representative image (C) and quatification of VICs upon polyIC incubation shows decreased cell numbers (D, number of cells per field of view (FOV)) and increased cell budding (E). Cell viability was quantified with AlamarBlue assay (F) (****p < 0.0001, **p < 0.01, *p < 0.05, two-sided T test, mean ± SEM). For qPCR analysis, VICs were cultured either in basal medium (BM) or pro-calcifying medium (PCM) and TLR3 knockdown was performed with TLR3-siRNA, scramble-si RNA (scr.) was used as control. qPCR analysis of TLR3, interleukin-6 (IL-6), bone-morphogenic protein (BMP) and runt-related transcription factor 2 (RUNX2) (G–J). (****p < 0.0001, two-way ANOVA with Tukey multiple comparisons test, mean ± SEM)

• 

  Fig. 2. TLR3 stimulation enhances AS development in mice: AS was induced in C57BL6/J mice via wire injury. The mice received daily injections of polyIC or vehicle control (NaCl) respectively. Ultrasound was performed every two weeks after surgery (A). Serum concentration of pro- and anti-inflammatory cytokines after 6 weeks (B, C). Peak velocity levels (D, E) and mean pressure gradient (F) over the aortic valve was minimally increased upon polyIC stimulation, left ventricular function remained unchanged (G) (n = 7–8). (****p < 0.0001, ***p < 0.001, mixed-effect model with Bonferroni multiple comparisons test (E–G) or two-sided T tests (B, C), mean ± SEM)

• 

  Fig. 3. AS development is impaired in TLR3-deficient mice: AS was induced in WT and TLR3−/− mice via mechanical wire injury (A). Trans-aortic valve blood flow peak velocity (B) and mean gradient (C) did not increase in the same degree in TLR3−/− animals as in wildtype mice after surgery. Left ventricular function remained unchanged in both groups (D) (n = 19–20, *** p < 0.001, **p < 0.01, *p < 0.05, mixed-effect model with Bonferroni multiple comparisons test, mean ± SEM). Plasma concentrations of pro- and anti-inflammatory cytokines (E). Histologic analysis of the aortic valve (F): H.E.-staining with quantitative analysis of the valve cusp area (G), CD68 staining: intra-valvular monocyte infiltration (H). FACS analysis of monocytes in peripheral blood of polyIC-treated mice (experimental design in Fig. 2A) and TLR3−/− mice vs. control (I) (*p < 0.05, two-sided T tests, mean ± SEM). Quantification of valvular fibrosis (J) and calcification (K). (n = 14–15, **p < 0.01, two-sided T tests, mean ± SEM)

• 

  Fig. 4. C4a reduces inflammation and calcification in valvular cells: Valvular interstitial cells (VICs) were treated with polyIC and/or C4a. TLR3 inhibition prevented upregulation of TLR3, IL-6 and BMP. No significant difference could be observed for RUNX2 (A) (***p < 0.001, **p < 0.01, *p < 0.05, two-way ANOVA with Tukey multiple comparisons test, mean ± SEM). Extracellular calcium deposits visualized by Alizarin red S staining in VIC cultivated in basal medium (BM) or pro-calcifying medium (PCM). Concomitant C4a incubation prevented VIC calcification upon polyIC stimulation (B). Relative length of valvular endothelial cells (VECs) and microscopic images 24 h after incubation with polyIC ± C4a (C, D). mRNA expression of TLR3 (E), CD31 (F), von Willebrand factor (VWF) (G), vascular cell adhesion molecule 1 (VCAM1) (H), alpha actin 2 (ACTA) (I) by VEC following stimulation with polyIC ± C4a (****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05, two-way ANOVA with Tukey multiple comparisons test, mean ± SEM)

• 

  Fig. 5. Pharmacological TLR3 inhibition reduces AS formation in mice: Serum concentration of Regulated on activation, normal T cell expressed and secreted (RANTES) in mice 6 h after 100 µg polyIC injection and concomitant dose escalation C4a treatment (A) (n = 2). RANTES serum concentration after C4a treatment and injection of increasing polyIC dose (B) (***p < 0.001, *p < 0.05, two-way ANOVA with Tukey multiple comparisons test, mean ± SEM, n = 3). C4a or vehicle was injected in WT mice every second day after AS induction (n = 20) (C). C4a inhibited increase in blood flow peak velocity (D) and mean gradient (E) compared to vehicle-treated mice. Left ventricular ejection fraction remained unchanged (F) (**** p < 0.0001, ***p < 0.001, mixed-effect model with Bonferroni multiple comparisons test, mean ± SEM). Aortic valve cusp area (G), infiltration of CD68 positive cells (H) and fibrosis (J) was significantly increased in C4a treated mice. Calcification was similar in both groups (H) (n = 12–16, ***p < 0.001, **p < 0.01, *p < 0.05, two-sided T tests, mean