Augmented osteogenic responses in human aortic valve cells exposed to oxLDL and TLR4 agonist: a mechanistic role of Notch1 and NF-κB interaction

Abstract

Aortic valve calcification causes the progression of calcific aortic valve disease (CAVD). Stimulation of aortic valve interstitial cells (AVICs) with lipopolysaccharide (LPS) up-regulates the expression of osteogenic mediators, and NF-κB plays a central role in mediating AVIC osteogenic responses to Toll-like receptor 4 (TLR4) stimulation. Diseased aortic valves exhibit greater levels of oxidized low-density lipoprotein (oxLDL). This study tested the hypothesis that oxLDL augments the osteogenic responses in human AVICs through modulation of NF-κB and Notch1 activation. AVICs isolated from normal human aortic valves were treated with LPS (0.1 µg/ml), oxLDL (20 µg/ml) or LPS plus oxLDL for 48 h. OxLDL alone increased cellular bone morphogenetic protein-2 (BMP-2) levels while it had no effect on alkaline phosphatase (ALP) levels. Cells exposed to LPS plus oxLDL produced higher levels of BMP-2 and ALP than cells exposed to LPS alone. Further, LPS plus oxLDL induced greater NF-κB activation, and inhibition of NF-κB markedly reduced the expression of BMP-2 and ALP in cells treated with LPS plus oxLDL. OxLDL also induced Notch1 activation and resulted in augmented Notch1 activation when it was combined with LPS. Inhibition of Notch1 cleavage attenuated NF-κB activation induced by LPS plus oxLDL, and inhibition of NF-κB suppressed the expression of BMP-2 and ALP induced by the synergistic effect of Jagged1 and LPS. These findings demonstrate that oxLDL up-regulates BMP-2 expression in human AVICs and synergizes with LPS to elicit augmented AVIC osteogenic responses. OxLDL exerts its effect through modulation of the Notch1-NF-κB signaling cascade. Thus, oxLDL may play a role in the mechanism underlying CAVD progression.

Figure 1. OxLDL and LPS synergize in the induction of the osteogenic responses in human AVICs.

Human AVICs were treated with LPS (0.10 µg/ml), oxLDL (20 µg/ml) or LPS plus oxLDL for 48 h. A. Representative immunoblots and densitometric data show that oxLDL up-regulates BMP-2 expression. Protein levels of BMP-2 are much higher in cells treated with LPS plus oxLDL than those in cells treated with LPS alone or oxLDL alone. B. OxLDL alone has a minimal effect on ALP levels, but cells treated with LPS plus oxLDL produce higher levels of ALP than cells treated with LPS alone. Similarly, ALP activity is higher in cells treated with LPS plus oxLDL. n = 4 separated cell isolates in each group, *P<0.05 vs. untreated control; †P<0.05 vs. cells treated with LPS alone or oxLDL alone.

Figure 2. NF-κB plays an important role in the augmented osteogenic responses.

A. Human AVICs were treated with LPS (0.10 µg/ml), oxLDL (20 µg/ml) or LPS plus oxLDL for 1–8 h. Representative immunoblots and densitometric data (n = 4) show that oxLDL and LPS each induces NF-κB phosphorylation at 4 and 8 h. NF-κB phosphorylation is enhanced when oxLDL and LPS are combined. *P<0.05 vs. untreated control; †P<0.05 vs. cells treated with LPS alone. B and C. Human AVICs were treated with LPS plus oxLDL for 48 h in the absence or presence of NF-κB inhibitor SN50 or BAY11-7082. Representative immunoblots and densitometric data (n = 4) show that inhibition of NF-κB with either of the inhibitors markedly reduces levels of BMP-2 (B) and ALP (C) in cells treated with LPS plus oxLDL. *P<0.05 vs. cells treated with LPS plus oxLDL.

Figure 3. Notch1 is involved in the induction of the osteogenic responses by oxLDL and LPS plus oxLDL.

A. Human AVICs were treated with LPS (0.10 µg/ml), oxLDL (20 µg/ml) or LPS plus oxLDL for 1–8 h. Representative immunoblots and densitometric data (n = 4) show increased NICD1 generation in cells treated for 8 h with LPS or oxLDL. Higher levels of NICD1 are detected in cells treated with LPS plus oxLDL. Time course data show that NICD1 is detectable at 4 h after exposing to LPS plus oxLDL and accumulates over time. *P<0.05 vs. untreated control; †P<0.05 vs. cells treated with LPS alone; ‡ P<0.05 vs. cells treated with oxLDL alone. B and C. Human AVICs were treated with LPS plus oxLDL in the presence or absence of γ-secretase inhibitor DAPT or GSI1. Representative immunoblots and densitometric data (n = 4) show that inhibition of Notch1 with DAPT or GSI1 reduces levels of BMP-2 (B) and ALP (C). *P<0.05 vs. cells treated with LPS plus oxLDL. C. Human AVICs were treated with oxLDL alone in the presence or absence of γ-secretase inhibitor DAPT or GSI1. Representative immunoblots of 3 separate experiments show that inhibition of Notch1 with DAPT or GSI1 reduces BMP-2 levels.

Figure 4. Notch1 knockdown reduces the expression of BMP-2 and ALP following stimulation with LPS plus oxLDL.

Human AVICs were treated with Notch1-specific siRNA and then stimulated with LPS plus oxLDL. Representative immunoblots of 2 separate experiments show that knockdown of Notch1 results in decreased expression of BMP-2 and ALP following stimulation with LPS plus oxLDL that is associated with reduced levels of NICD1.

Figure 5. Inhibition of Notch1 cleavage attenuates NF-κB activation induced by LPS plus oxLDL.

Human AVICs were treated with LPS plus oxLDL in the presence or absence of γ-secretase inhibitor DAPT. A representative immunoblot of 3 separate experiments and ELISA data (n = 4 in each group) show that NF-κB phosphorylation and DNA-binding activity are reduced when DAPT is present during stimulation with LPS plus oxLDL. *P<0.05 vs. untreated control; # P<0.05 vs. cells treated with LPS plus oxLDL.

Figure 6. Inhibition of NF-κB suppresses the osteogenic responses induced by Jagged1 plus LPS.

A. AVICs were treated with LPS (0.10 µg/ml), oxLDL (20 µg/ml) or LPS plus oxLDL for 4 or 8 h. ELISA data (n = 4 in each group) show that both LPS and oxLDL up-regulate Jagged1 secretion. Cells treated with LPS plus oxLDL release higher levels of Jagged1. *P<0.05 vs. untreated control; †P<0.05 vs. cells treated with LPS alone; ‡ P<0.05 vs. cells treated with oxLDL alone.