Toll-like Receptor 3 Mediates the Onset of Calcific Aortic Valve Disease

Objectives: Calcific aortic valve disease (CAVD) is caused by an osteoblastic phenotype switch of valvular interstitial cells (VICs). The trigger for the phenotype switch remains unknown. Toll-like receptor 3 (TLR3) is part of the innate immune system activated by viral and endogenous RNA released from stressed cells. We hypothesized that mechanical strain leads to TLR3 activation leading to an osteoblastic phenotype switch of VICs with subsequent initiation of CAVD.

Methods: Aortic valves were obtained from patients undergoing aortic valve replacement or from explanted hearts. VICs were isolated and treated with TLR3 agonist poly (I:C) or a TLR3/dsRNA complex inhibitor. Cells were challenged with osteoblastic medium and analyzed for alkaline phosphatase activity and calcific nodule formation. Aortic valve morphology and function of aged wild-type (WT) and TLR3^−/− mice were analyzed via transthoracic echocardiography, microCT and histological evaluation. To confirm results in a second model, experiments were repeated in ApoE^−/− and ApoE^−/−/TLR3^−/− mice.

Results: Aortic valves and VICs showed abundant TLR3 expression. Mechanical stimulation of VICs resulted in TLR3 activation. Stimulation of TLR3 lead to expression of TNF-a, IL-6, IFN-y, IL-10, Runx2 and BMP2 and enhanced osteoblastic activity of treated cells. Mice showed age-dependent TLR3 expression. We found increased TLR3 expression in valves from CAVD patients. TLR3 Inhibition resulted in prevention of osteogenic phenotype switch in conditioned VICs. Aged WT mice showed increased mean gradients (mmHg: 3.14 ± 0.64 versus 1.16 ± 0.22, p = 0.025) and mean velocity (mm/s: 796 ± 77 versus 527 ± 55, p = 0.029) in echocardiographs. MicroCT and histological analyses revealed thickened valve leaflets and atherosclerotic plaques. These changes were missing in age-matched TLR3^−/− mice. ApoE^−/− mice under high fat diet exhibited clear signs of CAVD with decreased aortic valve opening diameters (mm: 1.07 ± 0.046 versus 1.22 ± 0.41, p = 0.023) and increased valve leaflet thickness (mm: 0.11 ± 0.01 versus 0.08 ± 0.001, p = 0.018). However, these findings were completely missing in ApoE^−/−/TLR3^−/− mice.

Conclusion: TLR3 stimulation leads to an osteoblastic phenotype switch of VICs, whereas inhibition of TLR3 prevents from osteoblastic activity. TLR3 expression is increased in aortic valves from CAVD patients. TLR3^−/− mice show no phenotype of CAVD. TLR3 could become an effective target for the pharmacological prevention of CAVD.