Endothelial Nox1 oxidase assembly in human pulmonary arterial hypertension; driver of Gremlin1-mediated proliferation

Abstract

Pulmonary arterial hypertension (PAH) is a rapidly degenerating and devastating disease of increased pulmonary vessel resistance leading to right heart failure. Palliative modalities remain limited despite recent endeavors to investigate the mechanisms underlying increased pulmonary vascular resistance (PVR), i.e. aberrant vascular remodeling and occlusion. However, little is known of the molecular mechanisms responsible for endothelial proliferation, a root cause of PAH-associated vascular remodeling. Lung tissue specimens from PAH and non-PAH patients and hypoxia-exposed human pulmonary artery endothelial cells (ECs) (HPAEC) were assessed for mRNA and protein expression. Reactive oxygen species (ROS) were measured using cytochrome c and Amplex Red assays. Findings demonstrate for the first time an up-regulation of NADPH oxidase 1 (Nox1) at the transcript and protein level in resistance vessels from PAH compared with non-PAH patients. This coincided with an increase in ROS production and expression of bone morphogenetic protein (BMP) antagonist Gremlin1 (Grem1). In HPAEC, hypoxia induced Nox1 subunit expression, assembly, and oxidase activity leading to elevation in sonic hedgehog (SHH) and Grem1 expression. Nox1 gene silencing abrogated this cascade. Moreover, loss of either Nox1, SHH or Grem1 attenuated hypoxia-induced EC proliferation. Together, these data support a Nox1-SHH-Grem1 signaling axis in pulmonary vascular endothelium that is likely to contribute to pathophysiological endothelial proliferation and the progression of PAH. These findings also support targeting of Nox1 as a viable therapeutic option to combat PAH.

Keywords: NADPH oxidase; endothelial cells; gremlin1; hedgehog; pulmonary hypertension; reactive oxygen species.

Figure 1. Human PAH is associated with increased Nox1 and Grem1 expression and elevated ROS production that is paralleled in rat PAH

(A) Real-time qPCR analysis of Nox1 and Grem1 mRNA levels normalized to 18S in lung tissue samples from patients with PAH compared with non-PAH controls (n=3–5). (B) Western blot analysis of Nox1, Grem1, and β-actin protein levels in lung tissue samples from patients with PAH compared with non-PAH controls. Bar graph shows densitometry quantitation of protein expression normalized to β-actin protein expression (n=7–9). (C,D) Expression of Nox1 (C, red) and Grem1 (D, red) in PAH and non-PAH human lung sections by co-immunofluorescent staining. Sections were counterstained with α-smooth muscle actin (green) to visualize the medial smooth muscle layer of blood vessels, and DAPI (blue) to visualize the nuclei; the endothelial layer (intima) was defined as the layer delimited circumferentially by α-smooth muscle actin staining; L refers to vessel lumen. Scale bars represent 50 µm. (E) Increased DPI-inhibitable total homogenate comprehensive tissue ROS levels (direct H₂O₂ production plus dismuted O₂^•−) in lungs of PAH patients compared with non-PAH controls as measured by Amplex Red fluorescence for evaluation of H₂O₂ (n=4). *P<0.05 compared with non-PAH by Student’s t test. (F) Increased O₂^•− in total lung homogenates of SuHx-treated rats compared with normoxia-treated controls measured by the cytochrome c reduction assay. **P<0.01 compared with normoxia (n=5). The results are means ± S.E.M. of indicated sample size.

Figure 2. Hypoxia increases expression of Nox1 oxidase and its subunits in HPAEC

(A) Analysis by qPCR of Nox1, NoxA1, NoxO1, and p47^phox mRNA levels normalized to 18S in HPAEC in response to 24 h of hypoxia depicted as fold from normoxic control (n=3–6). (B,C) Western blot analysis of Nox1 and NoxO1 protein levels in response to 24 h of hypoxia in HPAEC; bar graphs depict densitometry quantitation normalized to β-actin (n=6–8). *P<0.05; **P<0.01; ***P<0.001 compared with normoxia by Student’s t test. (D,E) Confocal immunofluorescence images of Nox1 showing increased expression and colocalization with NoxO1 and p47^phox following 24 h hypoxia in HPAEC. HPAEC were double immunostained for NoxO1 (green, D) or p47^phox (green, E), and Nox1 (red, D and E). Inset shows magnification of areas of colocalization; n=3.

Figure 3. Hypoxia increases binding of Nox1 oxidase subunits in HPAEC

(A,B) PLA for Nox1 and p47^phox or Nox1 and NoxO1 (green punctates) following 24 h hypoxia in HPAEC. Nuclei were stained with DAPI (blue). (C) Quantitation of PLA results from A and B (n=3). *P<0.05; ***P<0.001 compared with normoxia; ^#P<0.05 compared with Nox1–NoxO1 interaction in hypoxia.

Figure 4. Hypoxia induces ROS in HPAEC, which is attenuated by Nox1 silencing

(A) Membrane fractions were prepared from HPAEC exposed to normoxia or hypoxia for the indicated times and assayed for O₂^•− levels by cytochrome c reduction assay (n=6); **P<0.01 compared with normoxia. (B) Representative cytochrome c absorbance over time plots. (C) HPAEC were exposed to hypoxia for 24 h and pretreated for 30 min with 10 µM DHE prior to lysis. Evaluation of O₂^•− levels was made using DHE-HPLC assay (n=5); *P<0.05 compared with normoxia. (D) Representative HPLC retention time plot. Values on plot indicate retention times and concentrations of 2-OH-E+ for one representative plot (E)Western blot analysis of Nox1 in HPAEC following transfection with scrambled (Scr) or Nox1 siRNA (n=3); **P<0.01 compared with Scr. (F) HPAEC transfected with scrambled (Scr) or Nox1 siRNA were exposed to hypoxia for 24 h, and then assayed for O₂^•− levels by cytochrome c reduction assay using membrane fractions (n=9); ***P<0.001 compared with Scr normoxia, ^#P<0.05 compared with Scr hypoxia by one-way ANOVA.

Figure 5. Hypoxia induces Nox1-dependent Grem1 expression in HPAEC

(A) Real-time qPCR analysis of Grem1 mRNA expression normalized to 18S in HPAEC subjected to hypoxia for the indicated times shown as fold from respective normoxic controls (n=6). (B) HPAEC were exposed to hypoxia for 24 h and then lysed and subjected to Western blotting for Grem1 and β-actin. Bar graph shows absorbance quantitation normalized to β-actin and depicted as fold from normoxia (n=6). (C) HPAEC transfected with Nox1 siRNA or scrambled siRNA (Scr) were exposed to hypoxia for 24 h, and then lysed and subjected to Western blotting for Grem1 and β-actin. Bar graph shows absorbance quantitation normalized to β-actin and depicted as fold from Scr normoxia control (n=6). (D) Western blot analysis of SHH expression in HPAEC transfected with Nox1 siRNA (siRNA) or scrambled control siRNA (Scr) and subjected to 24 h hypoxia; bar graph shows absorbance quantitation normalized to β-actin and depicted as fold from Scr normoxia control (n=3). ***P<0.001 compared with normoxia, **P<0.01 compared with respective normoxia group, ^#P<0.01 compared with Scr hypoxia, *P<0.05 compared with respective normoxia group by Student’s t test (A,B) or one-way ANOVA (C,D).

Figure 6. Loss of Nox1 or Grem1 attenuates hypoxia-induced HPAEC proliferation

HPAEC were transfected with Scr, Nox1 or Grem1 siRNA and exposed to hypoxia for 24 h. (A) HPAEC proliferation was assessed by BrdU incorporation following transfection of cells with Nox1 or Scr siRNA and subjecting them to 24 h hypoxia. Data are shown as means ± S.E.M. of fold change relative to Scr normoxia control (n=6). (B) Proliferation of HPAEC following transfection with Grem1 or Scr siRNA and 24 h hypoxia treatment was assessed by the CFSE FACS assay. Data are shown as percent proliferating cells. (C) Representative CFSE FACS tracings wherein the red spectrum indicates CFSE intensity in proliferating cells and gray spectrum indicates non-proliferating controls (n=3). *P<0.05 compared with Scr normoxia, ^#P<0.05 compared with Scr hypoxia.

Figure 7. Schematic diagram showing Nox1-Grem1 activation pathway in EC proliferation and RV hypertrophy

Hypoxia exposure of pulmonary ECs leads to Nox1-derived ROS production, which in turn activates SHH leading to Grem1 expression. This increase in Grem1 expression results in an unregulated proliferative process that ultimately leads to PAH and RV hypertrophy.