G-protein-coupled receptor kinase interacting protein-1 is required for pulmonary vascular development

Abstract

Background: The G-protein-coupled receptor kinase interacting protein-1 (GIT1) is a multidomain scaffold protein that participates in many cellular functions including receptor internalization, focal adhesion remodeling, and signaling by both G-protein-coupled receptors and tyrosine kinase receptors. However, there have been no in vivo studies of GIT1 function to date.

Methods and results: To determine essential functions of GIT1 in vivo, we generated a traditional GIT1 knockout mouse. GIT1 knockout mice exhibited approximately 60% perinatal mortality. Pathological examination showed that the major abnormality in GIT1 knockout mice was impaired lung development characterized by markedly reduced numbers of pulmonary blood vessels and increased alveolar spaces. Given that vascular endothelial growth factor (VEGF) is essential for pulmonary vascular development, we investigated the role of GIT1 in VEGF signaling in the lung and cultured endothelial cells. Because activation of phospholipase-Cgamma (PLCgamma) and extracellular signal-regulated kinases 1/2 (ERK1/2) by angiotensin II requires GIT1, we hypothesized that GIT1 mediates VEGF-dependent pulmonary angiogenesis by modulating PLCgamma and ERK1/2 activity in endothelial cells. In cultured endothelial cells, knockdown of GIT1 decreased VEGF-mediated phosphorylation of PLCgamma and ERK1/2. PLCgamma and ERK1/2 activity in lungs from GIT1 knockout mice was reduced postnatally.

Conclusions: Our data support a critical role for GIT1 in pulmonary vascular development by regulating VEGF-induced PLCgamma and ERK1/2 activation.

Figure 1. Generation of GIT1 KO mice

A. Strategy to generate GIT1 KO mice by homologous recombination. Exons 2-7 were replaced with the targeting vector containing a neomycin-resistance cassette. B. Analysis of GIT1 and GIT2 transcripts by RT-PCR. C. GIT1 protein expression in different tissues of WT and KO mice.

Figure 2. Survival, gross anatomy and histopathology in GIT1 WT and KO mice

A. Kaplan-Meier survival analysis. B-C. GIT1 WT and KO neonates (P4): WT neonates (B) and two KO neonates appeared normal (C, #1 and 4) while others (C, #2 and 3) showed varying degrees of cyanosis and respiratory distress. D-E. Fetal lungs (P4) from WT mice (D) had a normal gross appearance, whereas lungs from KO mice (E) showed scattered hemorrhages. F-I. Hematoxylin and eosin (H&E) staining of lung sections from P5 GIT1 WT and KO mice. There was obvious hemorrhage in parenchyma of GIT1 KO mice (G, I) compared to WT mice (F, H). J-M. GIT1 WT and KO mice lungs were perfused as described in methods, then the tissues were embedded and stained with H&E. GIT1 WT mice show normal, well-developed saccular and alveolar airway structures (J, L), whereas GIT1 KO showed abnormally large airspaces (K; M). N-O H&E staining of lung sections from embryos (E14.5) of GIT1 WT and KO mice. There were no obvious differences in histological appearance between WT and KO embryonic lungs.

Figure 3. Imaging of lungs shows reduced microvasculature in GIT1 KO mice

X-ray (A-B), fluorescein microangiography (C-D) micro-CT (E-F) were performed as described in the methods. WT mice displayed numerous small pulmonary arterioles extending into the capillary circulation (A, C, E), whereas there were strikingly fewer arterioles and capillaries in lungs from GIT1 KO mice (B, D, F) G. Quantification of micro-CT data according to the size of the vessels (mean ±SE; n =3,*P<0.05, compared with WT).

Figure 4. Decreased EC number and vessel density in GIT1 KO mice

A-B. vWF staining of lungs of GIT1 WT (A) and KO (B) at P7. Black arrows indicate vWF positive vessels (brown staining). Bar=20 μm C. Quantification of vWF staining. 5 fields were chosen from each sample (n=3 per group). D. VEGFR2 expression in lungs of GIT1 WT and KO mice at different ages. E. Quantitation of the relative expression of VEGFR2 normalized to actin (WT P0 = 1.0). *P< 0.01 compared with WT groups (mean ±SE; n =4).

Figure 5. GIT1 is required for activation of MEK1, ERK1/2 and PLCγ signaling pathways in EC and lung

A. GIT1 and control siRNA were transfected for 48 h into HUVECs. After serum starvation for 4 h, the cells were stimulated with 20 ng/ml VEGF for the indicated times and phosphorylation of the indicated proteins was determined. B-E. Right panels are quantitation of relative increase of PLCγ, MEK1, ERK1/2 and VEGFR2 phosphorylation compared to control siRNA group without VEGF stimulation. P< 0.05 compared with control siRNA groups (mean ±SE; n =3). F. Phosphorylation of PLCγ and ERK1/2 as well as expression of GIT1 in lungs of GIT1 WT and KO mice were detected by immunoblot. Total PLCγ, ERK1/2 and actin were assayed as loading controls. G-I. Quantitation of relative changes of PLCγ and ERK1/2 phosphorylation, as well as GIT1 expression (normalized to actin in WT P0 group). P< 0.01 compared with WT (mean ±SE; n =4).

Figure 6. Effect of GIT1 on EC proliferation in vitro and in vivo

A. Microvascular EC were transfected with GIT1 siRNA. After serum starvation for 24 h, the cells were stimulated with 5% FBS and DNA synthesis was evaluated by ³H-thymidine incorporation. Note that FBS was used as microvascular EC respond weakly to VEGF. B. Expression of PCNA in lungs of GIT1WT and KO mice were detected by immunoblot. Total actin was assayed as loading control. C. Quantitation of relative changes of PCNA expression (normalized to actin in WT P3-5 group). *P<0.01compared with WT groups of same time point (mean ±SE; n =4).

Figure 7. Effect of GIT1 on EC tube formation

HUVEC were transfected with control siRNA or GIT1 siRNA. After serum starvation for 24 h, the cells were seeded on 6-well plates at 1×10⁵ cells/well precoated with Matrigel and maintained in serum free medium, or with VEGF (50 ng/ml). Tube formation was visualized after 6 h (A) and analyzed using ImageJ (B).