HIPPO-Integrin-linked Kinase Cross-Talk Controls Self-Sustaining Proliferation and Survival in Pulmonary Hypertension

Abstract

Rationale: Enhanced proliferation and impaired apoptosis of pulmonary arterial vascular smooth muscle cells (PAVSMCs) are key pathophysiologic components of pulmonary vascular remodeling in pulmonary arterial hypertension (PAH).

Objectives: To determine the role and therapeutic relevance of HIPPO signaling in PAVSMC proliferation/apoptosis imbalance in PAH.

Methods: Primary distal PAVSMCs, lung tissue sections from unused donor (control) and idiopathic PAH lungs, and rat and mouse models of SU5416/hypoxia-induced pulmonary hypertension (PH) were used. Immunohistochemical, immunocytochemical, and immunoblot analyses and transfection, infection, DNA synthesis, apoptosis, migration, cell count, and protein activity assays were performed in this study.

Measurements and main results: Immunohistochemical and immunoblot analyses demonstrated that the HIPPO central component large tumor suppressor 1 (LATS1) is inactivated in small remodeled pulmonary arteries (PAs) and distal PAVSMCs in idiopathic PAH. Molecular- and pharmacology-based analyses revealed that LATS1 inactivation and consequent up-regulation of its reciprocal effector Yes-associated protein (Yap) were required for activation of mammalian target of rapamycin (mTOR)-Akt, accumulation of HIF1α, Notch3 intracellular domain and β-catenin, deficiency of proapoptotic Bim, increased proliferation, and survival of human PAH PAVSMCs. LATS1 inactivation and up-regulation of Yap increased production and secretion of fibronectin that up-regulated integrin-linked kinase 1 (ILK1). ILK1 supported LATS1 inactivation, and its inhibition reactivated LATS1, down-regulated Yap, suppressed proliferation, and promoted apoptosis in PAH, but not control PAVSMCs. PAVSM in small remodeled PAs from rats and mice with SU5416/hypoxia-induced PH showed down-regulation of LATS1 and overexpression of ILK1. Treatment of mice with selective ILK inhibitor Cpd22 at Days 22-35 of SU5416/hypoxia exposure restored LATS1 signaling and reduced established pulmonary vascular remodeling and PH.

Conclusions: These data report inactivation of HIPPO/LATS1, self-supported via Yap-fibronectin-ILK1 signaling loop, as a novel mechanism of self-sustaining proliferation and apoptosis resistance of PAVSMCs in PAH and suggest a new potential target for therapeutic intervention.

Keywords: HIPPO/LATS1; ILK; PAH; proliferation/apoptosis imbalance; vascular smooth muscle.

Figure 1.

HIPPO/large tumor suppressor 1 (LATS1) is inactivated in pulmonary arterial vascular smooth muscle cells in human pulmonary arterial hypertension (PAH) that is required for increased proliferation and impaired apoptosis. (A) Histochemical analysis of human lung tissues. Left to right: hematoxylin and eosin (H&E) staining; P-T1079 LATS1 (brown); smooth muscle α-actin (SMA) (green); proliferating cell nuclear antigen (PCNA) (red); PCNA (red) + SMA (green) merge. Images are representative of three control subjects and three subjects with idiopathic PAH; minimum of 10 PA/subject. Scale bar = 50 µm. (B–D) Pulmonary arterial vascular smooth muscle cells from four idiopathic PAH and four control (Contr) subjects were subjected to immunoblot (B and C) or DNA synthesis (bromodeoxyuridine [BrdU] incorporation assay) (D) analyses. (B) Two bands = two alternatively spliced human LATS1 isoforms (50). (C and D) Data are P/total LATS1 ratio; folds to control (C) and percentage of BrdU-positive cells per total number of cells (D). Data are means ± SE; n = 4 subjects/group; P < 0.05 by Mann–Whitney U test. (E–L) Cells were transfected with indicated mammalian vectors or empty plasmid (−) (E–J) or small interfering RNA (siRNA) LATS1 and control siRNA GLO (−) (K and L) for 48 hours followed by DNA synthesis (BrdU) (E, I, and K), immunoblot (F, J, and L), apoptosis (terminal deoxynucleotidyl transferase dUTP nick end labeling [TUNEL]) (G), and migration (Boyden chamber assay) (H) analyses. (F, J, and L) Representative immunoblots from three (F and J) and four (L) subjects per group (see Figure E3 for statistical analysis). Data represent percentage of BrdU- (E, I, and K) or TUNEL-positive cells (G) per total number of cells taken as 100% and fold changes to control (H). Data are means ± SE from three (E and G–I) and four (K) subjects per group. *P < 0.05 for LATS1 T1079D versus empty vector (−) (E and G), LATS1 T1079A and LATS1 kinase-dead (KD) versus empty vector (−) (I), and siRNA LATS1 versus control siRNA (−) (K) by Mann–Whitney U test. WT = wild type.

Figure 2.

HIPPO/large tumor suppressor 1 (LATS1) inactivation increases human pulmonary arterial hypertension (PAH) pulmonary arterial vascular smooth muscle cell (PAVSMC) proliferation and survival via Yap. (A and B) Immunoblot analysis of distal PAVSMCs from four nondiseased (control [Contr]) and four subjects with idiopathic PAH; data represent fold changes in Yap/Taz/actin ratio to control; data are means ± SE; four subjects per group; *P < 0.05 by Mann–Whitney U test. (C) Endogenous Yap/Taz activity accessed by dual-luciferase reporter assay. Data represent fold changes in 8 × GTIIC-luc/hRluc ratio to control; data are means ± SE; four subjects per group; *P < 0.05 by Mann–Whitney U test. (D) Representative images from three subjects per group. Yap, red; nuclei (DAPI), blue. Arrows indicate cells with increased nuclear localization of Yap (purple). (E) Immunoblot analysis of cells transfected with indicated LATS1 constructs or empty vector (−) for 48 hours. Immunoblots are representative of three separate experiments, each performed on cells from a different subject (see Figure E4 for statistical analysis). (F–K) Immunoblot (F and I), DNA synthesis (bromodeoxyuridine [BrdU] incorporation) (G and J), and apoptosis (terminal deoxynucleotidyl transferase dUTP nick end labeling [TUNEL]) (H and K) analyses of PAH PAVSMCs infected with shYap-producing or control adenovirus (−) for 72 hours (F–H) or treated for 24 hours with 1 µM VP or diluent (−). (F and I) Data are fold changes in Taz/tubulin ratio to controls; immunoblots are representative of three experiments, each on cells from a different subject with PAH. *P < 0.05 by Mann–Whitney U test. (G, H, J, and K) Data represent percentage of BrdU- (G and J) or TUNEL-positive cells (H and K) per total number of cells taken as 100%. Data are means ± SE, n = 3 subjects per condition (F, G–I, and K); n = 4 subjects per condition (J); *P < 0.05 by Mann–Whitney U test. DAPI = 4′,6-diamidino-2-phenylindole; KD = kinase-dead; sh = short hairpin RNA; VP = verteporfin; WT = wild type.

Figure 3.

HIPPO/large tumor suppressor 1 (LATS1) inactivation up-regulates mTOR-Akt signaling. Cells were transfected with indicated LATS1 constructs or empty vector (−) (A and B), or treated with diluent (−) or 1 µM VP for 24 hours (C) followed by immunoblot analysis to detect indicated proteins. Representative images and statistical analysis of three separate experiments, each performed on cells from a different subject. Data are fold changes in phospho/total ratios to control (−) (A and B) or diluent (C). Data are means ± SE; *P < 0.05 versus − by Mann–Whitney U test. KD = kinase-dead; mTOR = mammalian target of rapamycin; PAH = pulmonary arterial hypertension; VP = verteporfin; WT = wild type.

Figure 4.

HIPPO/large tumor suppressor 1 (LATS1) inactivation promotes fibronectin production and up-regulates integrin-linked kinase 1 (ILK1). (A–C) Immunoblot analysis of distal pulmonary arterial vascular smooth muscle cells from five subjects with idiopathic pulmonary arterial hypertension (PAH) and five nondiseased (control [Contr]) subjects. Data represent fold changes in fibronectin (Fn)/tubulin (B) and ILK1/tubulin ratios (C) to control subjects. Data are means ± SE; n = 5 subjects per group; *P < 0.01, **P < 0.05 versus control by Mann–Whitney U test. (D and E) Cells were maintained in cultural media for 48 hours; then supernatants were collected, normalized by total cell protein content, and subjected to immunoblot analysis to assess Fn levels. (D) Representative immunoblots. (E) Fn protein levels were normalized to the nonsignificant band and presented as fold changes versus control (taken as one fold). Data are means ± SE from n = 4 subjects per group; *P < 0.05 versus control by Mann–Whitney U test. (F and G) Cells were transfected with mammalian vectors expressing indicated LATS1 constructs or empty vector (−), or treated with 1 µM VP for 24 hours followed by immunoblot analysis to detect Fn and tubulin. Data are means ± SE from n = 3 subjects per group; data represent fold changes in Fn/tubulin ratio to LATS1 WT or diluent (−), respectively. **P < 0.05 by Mann–Whitney U test. (H–K) PAH pulmonary arterial vascular smooth muscle cells were transfected with small interfering RNA fibronectin (siFn) or control siRNA GLO (−), and immunoblot (I), DNA synthesis (bromodeoxyuridine [BrdU]), or apoptosis (terminal deoxynucleotidyl transferase dUTP nick end labeling [TUNEL]) analyses were performed. Data represent ILK1/tubulin ratio folds to control (I), percentage of BrdU- (J) or TUNEL-positive cells (K) per total number of cells taken as 100%. Data are means ± SE from three subjects per group. **P < 0.05 versus control siRNA GLO (−) by Mann–Whitney U test. (L) Histochemical analysis of human lung tissues. Left to right: hematoxylin and eosin (H&E) staining; ILK1 (red); ILK1 (red) + smooth muscle α-actin (SMA) (green); proliferating cell nuclear antigen (PCNA) (red); PCNA (red) + SMA (green). Images are representative of three control subjects and three subjects with idiopathic PAH; minimum of 10 PAs per subject. Scale bar = 50 µm. KD = kinase-dead; NS = nonspecific; VP = verteporfin; WT = wild type.

Figure 5.

Integrin-linked kinase 1 down-regulates HIPPO/large tumor suppressor 1 (LATS1), leading to increased pulmonary arterial vascular smooth muscle cell (PAVSMC) proliferation and survival. (A–C) Cells, treated with diluent (−) or 5 µM Cpd22 were subjected to cell count (5-d treatment) (A), DNA synthesis (bromodeoxyuridine, 24-h treatment) (B), and apoptosis (TUNEL, 24-h treatment) (C) analyses. Data are means ± SE; (A) n = 4 subjects per group; (B and C) n = 3 subjects per group. *P < 0.05 for diluent-treated pulmonary arterial hypertension (PAH) versus diluent-treated control (Contr) cells; **P < 0.05 for Cpd22- versus diluent-treated PAH PAVSMCs by Mann–Whitney U test. (D and E) Cells were treated with 1 µM Cpd22 for 30 minutes (D) or 24 hours (E). Immunoblots are representative of three experiments, each performed on cells from a different subject. See Figures E11A and E11B for statistical analysis. (F) PAH PAVSMCs (four subjects per group), cotransfected with 8 × GTIIC-luc and pGL4 Renilla hRluc, were treated with 5µM Cpd22 or diluent (−) (24 h), and endogenous Yap/Taz activity accessed by dual-luciferase reporter assay. *P < 0.05 for Contr versus PAH cells; **P < 0.05 for diluent- versus Cpd22-treated PAH cells by Mann–Whitney U test. (G) Immunohistochemical analysis of PAH PAVSMCs treated with indicated concentrations of Cpd22 for 24 hours. Red, Yap/Taz; blue, DAPI. Images are representative of three separate experiments, each on cells from a different subject. (H and I) PAH PAVSMCs were transfected with indicated LATS1 constructs or empty vector for 48 hours, treated with 1 µM Cpd22 or diluent (−) for 24 hours, and immunoblot (H) or DNA synthesis analyses (I) were performed. (H) Representative immunoblots from three experiments, each on cells from a separate subject. See Figure E11C for statistical analysis. (I) *P < 0.05 versus empty vector–transfected diluent-treated cells; **P < 0.05 versus empty vector-transfected Cpd22-treated cells by Mann–Whitney U test. DAPI = 4′,6-diamidino-2-phenylindole; KD = kinase-dead; mTOR = mammalian target of rapamycin; TUNEL = terminal deoxynucleotidyl transferase dUTP nick end labeling; WT = wild type.

Figure 6.

Increased substrate stiffness, fibronectin, and platelet-derived growth factor-BB (PDGF-BB) down-regulate HIPPO/large tumor suppressor 1 (LATS1) signaling in nondiseased pulmonary arterial vascular smooth muscle cells. (A–G) Immunoblot analysis of nondiseased (control) human pulmonary arterial vascular smooth muscle cells maintained on the Softwell hydrogels with 0.2-kPa and 25-kPa stiffness (A and B), plates covered with indicated matrix proteins (C and D), or treated with 10 ng/ml PDGF-BB for 0.5 and 24 hours (E–G; see Figure E10 for IL-6, ET-1, tumor necrosis factor-α, and insulin-like growth factor-1). Immunoblots are representative of three separate experiments, each performed on cells from a different subject. Data are means ± SE from three subjects per group; *P < 0.05 versus control (−) (B, F, and G); *P < 0.05 versus control and versus poly-d-lysine (D) by Mann–Whitney U test.

Figure 7.

P-T1079 large tumor suppressor 1 (Lats1) deficiency in murine animal models of pulmonary hypertension (PH). (A) Hematoxylin and eosin (H&E) and immunohistochemical analyses of lung tissues from 6- to-8-week-old male Sprague-Dawley rats with SU5416/hypoxia-induced PH (8 wk of PH induction). Upper panel (left to right): H&E staining; P-T1079 LATS1 (red); smooth muscle α-actin (SMA) (green) + 4′,6-diamidino-2-phenylindole (DAPI) (blue); P-T1079 LATS1 (red) + SMA (green) + DAPI (blue); proliferating cell nuclear antigen (PCNA) (red); PCNA (red) + SMA (green). Lower panel (left to right): H&E staining; integrin-linked kinase 1 (ILK1) (red); SMA (green) + DAPI (blue); ILK1 (red) + SMA (green) + DAPI (blue); PCNA (red); PCNA (red) + SMA (green). Images are representative of three animals per condition, minimum of 10 vessels per animal per condition. Scale bar = 50 µm. (B) H&E staining and immunohistochemical analysis of lung tissues from 6- to-8-week-old C57BL/6J mice with SU5416/hypoxia-induced PH (3 wk of PH induction). Images are representative of three animals per condition, 10 pulmonary arteries (PAs) per animal per condition. Red, P-T1079 Lats1 (left panel) and ILK1 (right panel); green, SMA; blue, DAPI. Scale bar = 25 µm. (C–F) PA medial thickness (MT) (C), mean systolic right ventricular pressure (sRVP) (D), mean pulmonary artery pressure (mPAP) (E), total pulmonary vascular resistance (PVR) (F), Fulton index (RV/[LV + S]) (G), and max dP/dT (H) of mice subjected to SU5416/hypoxia or normoxia exposure for 3 weeks. n = 5 mice/control (Contr) group; n = 6 mice per PH group. Data are means ± SE; *P < 0.01 by Mann–Whitney U test. See Figure E14 for systolic left ventricular pressure and mean arterial pressure data. LV = left ventricle; RV = right ventricle; S = septum.

Figure 8.

(A–F) Cpd22 reactivates large tumor suppressor 1 (Lats1), reduces established pulmonary vascular remodeling and pulmonary hypertension (PH). Six- to 8-week-old male C57BL/6J mice were subjected to SU5416/hypoxia exposure as described in the Methods section. Three weeks after PH induction, mice were treated with vehicle (Veh) or Cpd22 (20 mg/kg, intraperitoneally, 5 d/wk) for 2 weeks. Then hemodynamic measurements were performed, lung and heart tissues were collected for immunohistochemical and morphologic analyses; Fulton index was calculated as RV/(LV + S) ratio. Normoxia-maintained same-age male C57BL/6J mice were used as control (Contr) animals. (A) Hematoxylin and eosin (H&E) staining and immunohistochemical analysis of lung tissue sections. Images were representative of three mice per condition; minimum of 10 pulmonary arteries (PAs) per condition. Red, P-T1079 Lats1; green, smooth muscle α-actin (SMA); blue, DAPI. Scale bar = 50 µm. (B–F) PA medial thickness (MT) (B), systolic right ventricular pressure (sRVP) (C), mean pulmonary arterial pressure (mPAP) (D), max dP/dT (E), and Fulton index (F). Data are means ± SE; n = 5–8 mice per group. *P < 0.01 for PH Veh versus Contr (B–F) and for PH Cpd22 versus PH Veh (D); **P < 0.05 for PH Cpd22 versus PH Veh (B and C) by Mann–Whitney U test (see Figure E16 for systolic left ventricular pressure and mean arterial pressure). (G) Schematic representation of the potential function of HIPPO in nondiseased pulmonary arterial vascular smooth muscle cells (VSM) (left) and proposed mechanism of pulmonary VSM-specific HIPPO dysfunction in PH (right). DAPI = 4′,6-diamidino-2-phenylindole; Fn = fibronectin; ILK = integrin-linked kinase; LV = left ventricle; RV = right ventricle; S = septum.