Specificity Protein 1-Mediated Promotion of CXCL12 Advances Endothelial Cell Metabolism and Proliferation in Pulmonary Hypertension

Abstract

Pulmonary arterial hypertension (PAH) is a rare yet devastating and incurable disease with few treatment options. The underlying mechanisms of PAH appear to involve substantial cellular proliferation and vascular remodeling, causing right ventricular overload and eventual heart failure. Recent evidence suggests a significant seminal role of the pulmonary endothelium in the initiation and promotion of PAH. Our previous work identified elevated reactive oxygen species (ROS)-producing enzyme NADPH oxidase 1 (NOX1) in human pulmonary artery endothelial cells (HPAECs) of PAH patients promoting endothelial cell proliferation in vitro. In this study, we interrogated chemokine CXCL12's (aka SDF-1) role in EC proliferation under the control of NOX1 and specificity protein 1 (Sp1). We report here that NOX1 can drive hypoxia-induced endothelial CXCL12 expression via the transcription factor Sp1 leading to HPAEC proliferation and migration. Indeed, NOX1 drove hypoxia-induced Sp1 activation, along with an increased capacity of Sp1 to bind cognate promoter regions in the CXCL12 promoter. Sp1 activation induced elevated expression of CXCL12 in hypoxic HPAECs, supporting downstream induction of expression at the CXCL12 promoter via NOX1 activity. Pathological levels of CXCL12 mimicking those reported in human PAH patient serum restored EC proliferation impeded by specific NOX1 inhibitor. The translational relevance of our findings is highlighted by elevated NOX1 activity, Sp1 activation, and CXCL12 expression in explanted lung samples from PAH patients compared to non-PAH controls. Analysis of phosphofructokinase, glucose-6-phosphate dehydrogenase, and glutaminase activity revealed that CXCL12 induces glutamine and glucose metabolism, which are foundational to EC cell proliferation. Indeed, in explanted human PAH lungs, demonstrably higher glutaminase activity was detected compared to healthy controls. Finally, infusion of recombinant CXCL12 into healthy mice amplified pulmonary arterial pressure, right ventricle remodeling, and elevated glucose and glutamine metabolism. Together these data suggest a central role for a novel NOX1-Sp1-CXCL12 pathway in mediating PAH phenotype in the lung endothelium.

Keywords: CXCL12; NOX1; Sp1; pulmonary arterial hypertension.

Figure 1

CXCL12 expression is elevated in PAH and is NOX1-dependent in hypoxic pulmonary endothelial cells. (A) Western blot analysis of CXCL12 expression in explanted lung tissue samples from PAH or non-PAH patients (n = 6–10). (B) ELISA analysis of CXCL12 levels in explanted human lung tissue samples from PAH and non-PAH patients (n = 6–10). Mean values are indicated by a dotted red line. (C) Cultured human pulmonary artery endothelial cells (hPAECs) were exposed to 1% O₂ (hypoxia) or normoxia for 48 h and CXCL12 expression was assessed by Western blot (n = 7). D-E. hPAECs were transfected with scrambled or NOX1-targeted siRNAs for 16 h. Cells were exposed to 1% O₂ or normoxia and CXCL12 expression was assessed by (D). Western blot or (E). flow cytometry (n = 3–6). Representative flow cytometry histograms depict mean fluorescent intensity of FITC along the x axis, with a rightward shift of the peak indicating increased CXCL12 expression. Results analyzed by one-tailed t-test (A–C), and one-way ANOVA followed by Holm–Sidak post hoc analysis to determine differences between treatment groups (D,E) * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 2

NOX1 activates the transcription factor specificity protein 1 (Sp1), Sp1 consensus DNA binding, and CXCL12 gene promotion. (A) Western blot analysis of human lung lysates for p-Sp1/β-actin (n = 6). (B) Sp1 consensus DNA binding activity in human lung lysates measured by Active Motif TransAM assay (n = 6). (C) hPAECs transfected with scrambled or NOX1-targeted siRNA for 24 h were exposed to 1% O₂ and analyzed by Western blot for Sp1 activation, p-Sp1/β-actin (n = 3–6). (D) hPAECs transfected with scrambled or NOX1-targeted siRNA for 24 h were exposed to 1% O₂ and analyzed by Active Motif Trans AM assay. (E) Sp1 transcriptional activation was tested in HUVECs transfected with SpRE luciferase plasmid in addition to scrambled or NOX1-targeted siRNA. Cells were placed in normoxia or 1% O₂ for 48 h and luciferase activity was measured by plate reader luminometer (n = 3–6). (F) hPAECs transfected with scrambled or Sp1-targeted siRNA for 24 h were exposed to 1% O₂ 48 h. CXCL12 expression was measured by flow cytometry (n = 3–9). Representative flow cytometry histograms depict mean fluorescent intensity of FITC along the x axis, with a rightward shift of the peak indicating increased CXCL12 expression. (G) HUVECs transfected with luciferase reporter construct containing 1 kb of the human CXCL12 promoter were transfected with scrambled or Sp1-targeted siRNA. Cells were then exposed to 48 h hypoxia (1% O2) and Sp1′s ability to promote luciferase activity measured by plate reader luminometer (n = 3–4). Results were analyzed by Student t-test (A,B), one-way ANOVA followed by the Holm–Sidak test (C–G) * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 3

Hypoxia-induced proliferation and migration of pulmonary endothelial cells are NOX1-, Sp1- and CXCL12-dependent. (A,B) Normoxic vs. hypoxic hPAECs (1% O₂) were transfected with either scrambled or Sp1-targeted siRNA and analyzed by (A) Crystal Violet and (B). Wound closure assay; insets are wound images taken at 24 h. (C) hPAECs were subjected to normoxic vs. hypoxic (1% O₂) conditions following treatment with vehicle (DMSO) or Sp1 inhibitor mithramycin; proliferation was analyzed by Click-It^TMEdU incorporation and flow cytometry. (D,E) hPAECs transfected with scrambled or CXCL12-targeted siRNA analyzed by (D). Crystal Violet and (E). wound closure assay. (F) hPAECs were treated with scrambled or Nox1-selective inhibitory peptide NoxA1ds, in addition to recombinant CXCL12 or vehicle control (PBS),and were subjected to hypoxia (1% O₂) or normoxia. Proliferation was analyzed by Click-It EdU incorporation and flow cytometry. Representative flow cytometry histograms depict populations of EdU-positive and -negative cells in a sample along the x axis, with right hand peaks indicating proliferating cells taken as percentage to the total number of cells in the sample (see Figure S1F for gating). Results were analyzed by one-way ANOVA and Holm–Sidak post hoc analysis (A–F). * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 4

CXCL12 induces pro-proliferative metabolic activity. (A) hPAECs were treated with recombinant CXCL12 for 24 h and glucose uptake was analyzed by fluorescent NBGD using flow cytometry. Representative flow cytometry histograms depict mean fluorescent intensity of FITC along the x axis, with a rightward shift of the peak indicating increased glucose uptake. (B) Phosphofructokinase (with RFU/min kinetic activity, inset) and (C) glucose-6-phosphate dehydrogenase (G6PD) activity were tested in hPAECs treated with hrCXCL12 vs. control. (D) Wound closure was measured in hPAECs treated with hrCXCL12 along with the glucose-6-phosphate dehydrogenase inhibitor DHEA or vehicle (methanol); inset shows images at 24 h. (E) Lung lysates from age- and sex-matched PAH and non-PAH subjects were assessed for glutaminase activity (n = 6). (F) hPAECs treated with hrCXCL12 were assessed for glutaminase activity after 24 h stimulation. (G,H) hPAECs treated with hrCXCL12 plus or minus the glutaminase inhibitor C-968 or vehicle (DMSO) to measure (G). glucose uptake (fluorescent NBGD-flow cytometry) and (H). proliferation (Click-it Edu-flow cytometry). Representative flow cytometry histograms in H depict populations of EdU-positive and -negative cells in a sample along the x axis, with a righthand peak cluster indicating proliferating cells taken as percentage to the total number of cells in the sample (for gating see Figure S1F). Results were analyzed with Student t-test (A–C,E,F) and one-way ANOVA and Holm–Sidak post hoc analysis (D,G,H) * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 5

Subcutaneous CXCL12 infusion induced hallmark characteristics of PAH in female mice. (A) Mice infused with mrCXCL12 or vehicle control for 3 wks were analyzed for pulmonary arterial pressure changes (mPAP) via right heart catheterization. (B–E) Right and left ventricles were dissected away from the septum. Each were weighed to calculate relative heart chamber and total mass and Fulton Index. (n = 6–8). (F,G) Activity of (F) glutaminase and (G) phosphofructokinase (PFK) was quantified in lung lysates (n = 6–8). Results were analyzed by Student t-test. * p < 0.05, ** p < 0.01, **** p < 0.0001.

Figure 5

Subcutaneous CXCL12 infusion induced hallmark characteristics of PAH in female mice. (A) Mice infused with mrCXCL12 or vehicle control for 3 wks were analyzed for pulmonary arterial pressure changes (mPAP) via right heart catheterization. (B–E) Right and left ventricles were dissected away from the septum. Each were weighed to calculate relative heart chamber and total mass and Fulton Index. (n = 6–8). (F,G) Activity of (F) glutaminase and (G) phosphofructokinase (PFK) was quantified in lung lysates (n = 6–8). Results were analyzed by Student t-test. * p < 0.05, ** p < 0.01, **** p < 0.0001.

Figure 6

Working model of NOX1-mediated SP1 regulation of CXCL12 in hypoxia and PAH. Working model depicting NOX1 (alongside its critical subunits p22^phox, NOXO1, NOXA1, and Rac1/2) induction of CXCL12 via Sp1 following hypoxia. Resultant CXCL12 is postulated to be secreted from cells, leading to downstream promotion of vascular remodeling and PAH symptoms via pro-proliferative metabolic alteration and glucose uptake.