PDGF-dependent β-catenin activation is associated with abnormal pulmonary artery smooth muscle cell proliferation in pulmonary arterial hypertension

Abstract

Pulmonary arterial hypertension (PAH) is characterized by excessive pulmonary arterial smooth muscle cells (PASMCs) growth, partially in response to PDGF-BB but whether this is dependent on β-catenin (βC) activation is unclear. Compared to healthy cells, PAH PASMCs demonstrate higher levels of proliferation both at baseline and with PDGF-BB that correlate with GSK3β dependent βC activation. We show that βC knockdown but not Wnt5a stimulation reduces PDGF-BB dependent growth and normalizes PAH PASMCs proliferation. These findings support that cross-talk between PDGF and Wnt signaling modulates PASMC proliferation and suggest that βC targeted therapies could treat abnormal vascular remodeling in PAH.

Keywords: PDGF; Wnt signaling; pulmonary disease; pulmonary hypertension; smooth muscle cells; vascular remodeling.

Figure 1. PDGF-BB stimulation increases levels of active βC, which are already elevated at baseline in PAH PASMCs

β-catenin activation and GSK3β phosphorylation (p) were measured in lysates from healthy donor (HD) and PAH PASMCs stimulated with 20 ng/mL PDGF-BB over a period of 24h. Corresponding densitometry values are shown. Loading was compared against α-tubulin. Cells were starved before stimulation for 48h in 0.1% FBS. CON, unstimulated control. Bars represent means ± SEM. ***P <0.0001 compared to unstimulated healthy donor, ###P<0.0001 vs. corresponding healthy donor using one-way ANOVA with Bonferroni post-test, (N=3).

Figure 2. PDGF-BB stimulation increases nuclear translocation of βC and transcriptional activity

(A) Immunofluorescence images showing active βC in healthy (top) and PAH (bottom) PASMCs. Cells were starved 48h in 0.1% FBS and then stimulated with 20 ng/mL PDGF-BB over a period of 6h. DAPI (blue) stain was used to label cell nuclei. Green nuclear fluorescence was measured and compared using one-way ANOVA with Bonferroni post-test. Bar=10 μm. (B) TOPflash luciferase assays in PDGF-stimulated healthy donor (HD) and PAH PASMCs. Cells were transfected with TOPflash or FOPflash (negative control) luciferase reporter plasmids, or Renilla as a control for transfection efficiency. Cells were stimulated with 20 ng/mL PDGF-BB for 6h. Lysates were analyzed for luciferase activity relative to Renilla (Relative luciferase activity or RLU: a measure of βC mediated changes in transcriptional activity). HD=healthy donors. Bars represent means ± SEM. *P<0.01, **P<0.001, ***P <0.0001 compared to unstimulated controls, ###P<0.0001 vs. corresponding HD using one-way ANOVA with Bonferroni post-test (N=3).

Figure 3. RNAi knockdown of βC reduces the PDGF-mediated growth response in PASMCs and normalizes baseline growth in PAH PASMCs

PASMCs from healthy and PAH patients were transfected for RNAi specific for βC or scrambled RNA as a negative control. Cells were seeded in 24-well plates at 25,000 cells per well and starved for 48h in 0.1% FBS. Cells were then grown 72h in either starvation media alone or supplemented with 20 ng/mL PDGF-BB. SC=scrambled RNAi control. Bars represent means ± SEM. *** P<0.0001 compared to unstimulated SC, ##P<0.001 vs. corresponding SC using one-way ANOVA with Bonferroni post-test (N=3).

Figure 4. Wnt5a stimulation prevents proliferation in PDGF BB-stimulated control PASMCs and fails to stimulate a response in PAH PASMCs

Cell proliferation assays using healthy donor (A) and PAH (B) PASMCs. Cells were seeded in 24-well plates at 25,000 cells per well and starved for 48h in 0.1% FBS. Cells were then grown 72h in either starvation media alone or supplemented with 20 ng/mL PDGF-BB and/or concentrations of Wnt5a described in the figure. Cells stimulated with Wnt5a and PDGF-BB were pre-incubated with Wnt5a for 30 minutes prior to the addition of PDGF-BB. CON, control. Bars represent means ± SEM from three different experiments. ***P<0.0001 compared to unstimulated controls, ###P <0.0001 compared to PDGF-BB stimulation using one-way ANOVA test with Bonferroni post-test.

Figure 5. Wnt5a stimulation prevents PDGF-mediated βC activation and transcriptional activity in healthy PASMCs and fails to stimulate a response in PAH PASMCs

(A and C) β-catenin activation was measured in lysates from healthy donor (top) and PAH (bottom) PASMCs stimulated with 50 ng/mL Wnt5a alone or supplemented with 20 ng/mL PDGF-BB over a period of 6h. Corresponding densitometry values against α-tubulin are shown below. (B and D) TOPflash luciferase assays in healthy donor (top) and PAH (bottom) PASMCs. Cells were stimulated with 50 ng/mL Wnt5a alone or supplemented with 20 ng/mL PDGF-BB for 6h. All co-stimulation samples were pre-incubated for 30 minutes with Wnt5a before the addition of PDGF-BB. Bars represent means ± SEM. ***P <0.0001 compared to unstimulated controls using one-way ANOVA with Bonferroni post-test.