Structure-function properties of the apoE-dependent COX-2 pathway in vascular smooth muscle cells

Abstract

Apolipoprotein (apoE) E is a multifunctional protein that plays a critical role in atherogenesis, in part by regulating the intimal proliferation of vascular smooth muscle cells. Recently, a novel cyclooxygenase (COX)-2 pathway was shown to contribute to the anti-proliferative action of human apoE3 in vascular smooth muscle cells (VSMC). Here, we provide insight into the structure-function properties by which apoE mediates these effects. ApoE3 is most effective in promoting COX-2 expression as a lipid-free protein and is less active after lipidation. Alterations in the stability of the helix bundle N-terminal domain of apoE that contains the binding site for the low density lipoprotein (LDL) receptor and heparin do not affect the up-regulation of the COX-2 pathway. In addition, the apoE2, 3, and 4 isoforms are all capable of up-regulating the COX-2 pathway. Finally, the effect of apoE on COX-2 was found to be independent of expression on the VSMC surface of the LDL receptor and heparan sulfate proteoglycans (HSPG). In summary, our data indicates that apoE, particularly in the lipid-free state, can up-regulate COX-2 in murine vascular smooth muscle cells apparently independently of binding to the LDLR, LRP or HSPG.

Figure 1. Influence of apoE3 domain structure on its ability to up-regulate the COX-2 signaling pathway

Quiescent primary murine VSMC were stimulated in 10% HI-FSC media with or without 2 µM of full length apoE3, apoE3-10kDa or apoE3-22kDa domains for 6 or 24 hours. Unstimulated or quiescent cells were maintained in serum-free medium as described in methods A). mRNA fold induction above control levels after 6 hours B–D). Fold-increase of 6-keto PGF_1α, PGE₂, and TXB₂ production compared to control after 24 hours. The control value of PGF_1α was 15.5±2.0 ng/mL, PGE₂ was 0.6±0.2 ng/mL, and TXB₂ was 1.2±0.0 ng/mL. All data represents the mean ± S.E.M of three independent experiments, statistical significance p< 0.05. * is compared with control.

Figure 2. Lipdation of apoE3 reduces its ability to up-regulate the COX-2 pathway

Relative COX-2 mRNA, protein expression levels and eicosanoids production after 6 or 24 hours stimulation of quiescent primary murine VSMC with 2 µM of apoE3 with or without DMPC (2:1 w/w ratio) A). mRNA fold induction above control after 6 hours B). COX-2 protein expression C–D). The fold induction of 6-keto PGF_1α and PGE₂ production compared to control after 24 hours. The control value of 6-keto PGF_1α was 3.6±0.7 ng/mL and 1.9±0.5 ng/mL for PGE₂. The data represents the mean ± S.E.M of three independent experiments except for the COX-2 protein expression, which is the mean from two independent experiments. Statistical significance is a p< 0.05. * is compared with control and # compared with apoE3 without DMPC.

Figure 3. The influence of apoE polymorphism

Quiescent primary murine VSMC were stimulated with 2 µM of apoE isoforms in 10% HI-FCS media for 6 or 24 hours. A). Fold-induction of COX-2 mRNA after 6 hours B). COX-2 protein after 24 hours C–D). Eicosanoids production after 24 hours. The control values were 8.2±2.8 ng/mL of PGF_1α and 0.5±0.2 ng/mL of PGE₂. The data in the figures represents the mean ± S.E.M of three independent experiments, statistical significance * p< 0.05 as compared to control.

Figure 4. Disruption of LDL receptor binding activity of apoE3

WT VSMC were quiescent for 48 hours followed by stimulation in 10% HI-FCS media supplemented with 2 µM of WT apoE3, 10kDa or 22kDa domains of apoE3, or a quadruple 22kDa apoE3 mutant. A). COX-2 mRNA fold induction after 6 hours stimulation. B–D). COX-2 protein expression and fold-induction of 6-keto PGF_1α and PGE₂ production after 24 hours. The control values were 16.4± 5.2 ng/mL of PGF_1α and 0.7 ±0.3 ng/mL of PGE₂. The data in the figures represents the mean ± S.E.M of three independent experiments, statistical significance p< 0.05. * is significant difference to control.

Figure 5. Effects of disruption of LDLR and HSPG binding of apoE3 on COX-2 signaling pathway

COX-2 mRNA and protein expression as well as eicosanoids production after stimulation of quiescent wild-type or LDLR null VSMC in 10% HI-FCS with and without 2 µM of apoE3 variants. A). COX-2 mRNA fold-induction after 6 hours stimulation in WT VSMC. B.) 24 hours COX-2 protein expression in WT VSMC. C–D). Eicosanoid fold-induction after 24 hours. The control value of 6-keto PGF_1α was 15.5±2.0 ng/mL and PGE₂ was 0.6±0.2 ng/mL. E–F). COX-2 mRNA and protein expression in apoE3-stimulated LDLR null VSMC. All data represents the mean ± S.E.M of three independent experiments, statistical significance * p< 0.05 as compared to control.

Figure 5. Effects of disruption of LDLR and HSPG binding of apoE3 on COX-2 signaling pathway

COX-2 mRNA and protein expression as well as eicosanoids production after stimulation of quiescent wild-type or LDLR null VSMC in 10% HI-FCS with and without 2 µM of apoE3 variants. A). COX-2 mRNA fold-induction after 6 hours stimulation in WT VSMC. B.) 24 hours COX-2 protein expression in WT VSMC. C–D). Eicosanoid fold-induction after 24 hours. The control value of 6-keto PGF_1α was 15.5±2.0 ng/mL and PGE₂ was 0.6±0.2 ng/mL. E–F). COX-2 mRNA and protein expression in apoE3-stimulated LDLR null VSMC. All data represents the mean ± S.E.M of three independent experiments, statistical significance * p< 0.05 as compared to control.