Endothelial C-type natriuretic peptide/guanylyl cyclase-B signaling prevents pulmonary arterial hypertension

Fig. 1. Endothelial cell-derived CNP prevents the progression of PH in mouse models. RVSP (a) and RV/(LV + IVS) ratios (b) in CNP ecKO mice and Tie2-Cre controls following MCTp treatment (10-week-old; n = 5 per group). Whole-lung mRNA expression of Nppc (c) and Npr2 (d) in Tie2-Cre mice with or without MCTp treatment (10-week-old; n = 5 per group). Representative pulmonary artery sections stained with hematoxylin-eosin and Victoria blue (e; scale bar, 50 μm) and quantification of percent medial wall thickness (f) in MCTp-treated CNP ecKO and Tie2-Cre mice. RVSP (g) and RV/(LV + IVS) ratios (h) in CNP ecKO and Tie2-Cre mice exposed to Hx (12-week-old; n = 6 per group). Whole-lung mRNA expression of Nppc (i) and Npr2 (j) in Tie2-Cre mice under normoxia or Hx (12-week-old; n = 6 per group). Representative pulmonary artery sections stained with hematoxylin-eosin and Victoria blue (k; scale bar, 50 μm) and quantification of medial wall thickness (l) in normoxic and Hx mice (12-week-old; n = 5 for normoxia, n = 10 for Hx). Data are presented as the mean ± SEM. P values were calculated using one-way ANOVA with Tukey’s multiple comparisons test.

Fig. 2. GC-B signaling in endothelial cells, but not smooth muscle cells, prevents the progression of pulmonary hypertension in the MCTp-induced PAH model. RVSP in GC-B ecKO versus Tie2-Cre mice (a; 10-week-old; n = 5 per group) and in GC-B smcKO versus SM22α-Cre mice (b; 10-week-old; n = 4 per group) following MCTp treatment. c–e GC-B ecKO versus Tie2-Cre mice following MCTp treatment. RV/(LV + IVS) ratios (c), representative pulmonary artery sections stained with hematoxylin–eosin and Victoria blue (d; scale bar, 50 μm), and quantification of medial wall thickness (e) (10-week-old; n = 5 for MCTp (–), n = 10 for MCTp (+)). f–h GC-B smcKO versus SM22α-Cre mice following MCTp treatment. RV/(LV + IVS) ratios (f), representative pulmonary artery sections stained with hematoxylin–eosin and Victoria blue (g; scale bar, 50 μm), and quantification of medial wall thickness (h) (10-week-old; n = 5 for MCTp (–), n = 10 for MCTp (+)). Data are presented as the mean ± SEM. P values were calculated using one-way ANOVA with Tukey’s multiple comparisons test.

Fig. 3. GC-B signaling in endothelial cells, but not smooth muscle cells, prevents the progression of pulmonary hypertension in Hx-induced model. a–d GC-B ecKO versus Tie2-Cre mice following Hx exposure. RVSP (a), RV/(LV + IVS) ratios (b), representative pulmonary artery sections stained with hematoxylin-eosin and Victoria blue (c; scale bar, 50 μm), and quantification of medial wall thickness (d) (12-week-old; n = 5 per group). e–h GC-B smcKO versus SM22α-Cre mice following Hx exposure. RVSP (e), RV/(LV + IVS) ratios (f), representative pulmonary artery sections stained with hematoxylin-eosin and Victoria blue (g; scale bar, 50 μm), and quantification of medial wall thickness (h) (12-week-old; n = 5 for normoxia, n = 10 for Hx). Data are presented as the mean ± SEM. P values were calculated using one-way ANOVA with Tukey’s multiple comparisons test.

Fig. 4. Whole-lung mRNA expression demonstrates that endothelial CNP/GC-B signaling regulates vasoactive and inflammatory genes in mouse models of pulmonary hypertension. RT-qPCR analyses of the whole lung expression of Nppc (a, b), Npr2 (c, d), Npr1 (e, f), Edn1 (g, h), Il6 (i, j), Ccl2 (k, l), and Tgfb1 (m, n) in Tie2-Cre, SM22α-Cre, CNP ecKO, GC-B ecKO and GC-B smcKO mice in the MCTp-induced model (10-week-old, a, c, e, g, i, k, m) and Hx-induced model (12-week-old, b, d, f, h, j, l, n) compared with untreated controls (10-week-old of MCTp (–) and 12-week-old normoxia; n = 8 per group). GAPDH was used as the internal control. Data are presented as the mean ± SEM. P values were calculated using one-way ANOVA with Tukey’s multiple comparisons test.

Fig. 5. CNP/GC-B signaling suppresses Hx-induced proinflammatory gene expression in HPAECs. Quantitative real-time RT-PCR analyses of NPPC (a) and NPR2 (b) mRNA expression in HPAECs exposed to Hx. mRNA expression of EDN1 (c), IL6 (d), CCL2 (e), and TGFB1 (f) in HPAECs transfected with control siRNA or siGC-B under normoxic and Hx conditions, with or without CNP-53 treatment. Hx-induced upregulation of these genes was abolished by CNP-53 in control cells, but not in siGC-B-transfected cells. g Knockdown efficiency of GC-B shown as reduced NPR2 mRNA levels in siGC-B–transfected HPAECs compared to scramble controls. (n = 8 biological replicates, independent culture wells, per group). GAPDH was used as the internal control. Data are presented as the mean ± SEM. P values were calculated using one-way ANOVA with Tukey’s multiple comparisons test.

Fig. 6. CNP/GC-B/PKG signaling corrects pathological alterations in SMAD signaling pathways in endothelial cells. Representative western blots of phosphorylated SMAD2/3 and total SMAD2/3 (a), and quantitative analysis of phosphorylated SMAD2/3 normalized to total SMAD2/3 (b) in HPAECs exposed to Hx with or without CNP or the PKGi (n = 4 biological replicates, independent culture wells, per group). Representative blots of phosphorylated SMAD1/5/9 and total SMAD1 (c) and quantitative analysis of phosphorylated SMAD1/5/9 normalized to total SMAD1 (d) under the same conditions (n = 4 biological replicates, independent culture wells, per group). e Effect of CNP and GC-B knockdown on Hx-induced expression of ID1 mRNA in HPAECs (n = 8 biological replicates, independent culture wells, per group). f Effect of CNP on Hx-induced expression of BMPR2 mRNA in HPAECs (n = 8 biological replicates, independent culture wells, per group). Representative blots of BMPRII and GAPDH (g) and quantitative analysis of BMPRII normalized to GAPDH (h) in HPAECs exposed to Hx with or without CNP (n = 3 per biological replicates, independent culture wells, group). Expression of Id1 (i, j) and Bmpr2 (k, l) mRNA in…

Fig. 7. Endothelial CNP/GC-B signaling restrains PASMC proliferation and migration via paracrine mechanisms. a PASMC proliferation assessed using CCK-8 assays under hypoxia. CM from GC-B-deficient HPAECs (EC+siGC-B) significantly increased PASMC proliferation compared with CM from control HPAECs (EC control). Proliferation was quantified in PASMCs (n = 12 biological replicates, independent culture wells, per group). b Schematic representation of the Boyden chamber co-culture assay. PASMCs (blue) were seeded into the upper insert, while HPAECs (red) were cultured in the lower chamber to evaluate endothelial-smooth muscle paracrine interactions. c PASMC migration measured by Boyden chamber assays under hypoxia. Migrated cells were quantified based on fluorescence intensity. Migration was significantly increased in response to CM from EC+siGC-B compared with control CM (n = 6 biological replicates, independent culture wells, per group). Representative Ki67 staining (d) and quantification (e) of PASMC proliferation in small pulmonary arteries from CNP ecKO mice in the MCTp-induced PAH model, treated with vehicle (Control), MCTp (PAH model) or MCTp + CNP (PAH model + CNP). Arrowheads indicate…

Fig. 8. Reduced NPPC and NPR2 expression in endothelial cells from human PAH lung tissues. a UMAP of endothelial cells (EC1-3) from control and IPAH lung samples. FeaturePlots of NPPC (b) and NPR2 (c) expression within endothelial cells. d DotPlot showing NPPC and NPR2 expression across endothelial subclusters. Violin plots comparing NPPC (e) and NPR2 (f) expression in endothelial cells between control and IPAH samples. Data are presented as the mean ± SEM. P values were calculated using the two-sided Wilcoxon rank-sum test.

Fig. 9. CNP-53 administration suppresses the development of pulmonary hypertension in mouse models via endothelial GC-B. Effect of continuous subcutaneous infusion of CNP-53 (0.2 mg/kg/day) on systemic BP in Tie2-Cre control, GC-B ecKO and GC-B smcKO mice in the MCTp-induced model (a; 10-week-old) and Hx-induced model (b; 12-week-old) (n = 3 per group). Effects of CNP-53 infusion on RVSP (c, e) and RV/(LV + IVS) ratios (d, f) in Tie2-Cre mice under MCTp (c, d; 10-week-old) and Hx (e, f; 12-week-old) conditions (MCTp model: n = 5 for untreated [MCTp (–)/CNP (–)], n = 5 for MCTp (+)/CNP (–), n = 3 for MCTp (+)/CNP (+); Hx model: n = 5 for normoxia/CNP (–), n = 5 for Hx/CNP (–), n = 3 for Hx/CNP (+)). Effects of CNP-53 infusion on RVSP (g, i) and RV/(LV + IVS) ratios (h, j) in GC-B ecKO and GC-B smcKO mice under MCTp (g, h; 10-week-old) and Hx (i, j; 12-week-old) conditions (MCTp model: n = 5 for MCTp (–)/CNP (–), n = 5 for MCTp (+)/CNP (–), n = 3 for MCTp (+)/CNP (+); Hx model: n = 5 for normoxia/CNP (–), n = 5 for Hx/CNP (–), n = 3 for Hx/CNP (+)). Data are presented as the mean ± SEM. P values were calculated using one-way ANOVA with Tukey’s multiple comparisons test.

Fig. 10. Therapeutic effects of CNP-53 in SuHx rodent models and its additive benefits with established therapies. SuHx rat model. RVSP (a), RV/(LV + IVS) ratios (b) and BP (c) in SuHx rats treated with vehicle or CNP-53 (0.2 mg/kg/day) (n = 5 per group). d–g SuHx mouse model (wild-type). RVSP (d) and RV/(LV + IVS) ratios (e) in SuHx mice treated with vehicle, CNP-53 (0.2 mg/kg/day), macitentan (20 mg/kg/day) or the combination of macitentan and CNP-53. RVSP (f) and RV/(LV + IVS) ratios (g) in SuHx mice treated with vehicle, CNP-53, sotatercept (10 mg/kg twice weekly) or the combination of sotatercept and CNP-53 (controls and SuHx+vehicle groups, n = 5 each; macitentan group, n = 6; sotatercept group, n = 7). h–k SuHx mouse model (CNP ecKO). RVSP (h) and RV/(LV + IVS) ratios (i) in SuHx mice treated with vehicle, CNP-53 (0.2 mg/kg/day), macitentan (20 mg/kg/day) or the combination of macitentan and CNP-53. RVSP (j) and RV/(LV + IVS) ratios (k) in SuHx mice treated with vehicle, CNP-53, sotatercept (10 mg/kg twice weekly) or the combination of sotatercept and CNP-53 (controls and SuHx+vehicle groups, n = 5 each; macitentan group, n = 5; sotatercept group, n = 6). Data are presented as the…