Phosphatidylinositol 3-kinase class II α-isoform PI3K-C2α is required for transforming growth factor β-induced Smad signaling in endothelial cells

Abstract

We have recently demonstrated that the PI3K class II-α isoform (PI3K-C2α), which generates phosphatidylinositol 3-phosphate and phosphatidylinositol 3,4-bisphosphates, plays crucial roles in angiogenesis, by analyzing PI3K-C2α knock-out mice. The PI3K-C2α actions are mediated at least in part through its participation in the internalization of VEGF receptor-2 and sphingosine-1-phosphate receptor S1P1 and thereby their signaling on endosomes. TGFβ, which is also an essential angiogenic factor, signals via the serine/threonine kinase receptor complex to induce phosphorylation of Smad2 and Smad3 (Smad2/3). SARA (Smad anchor for receptor activation) protein, which is localized in early endosomes through its FYVE domain, is required for Smad2/3 signaling. In the present study, we showed that PI3K-C2α knockdown nearly completely abolished TGFβ1-induced phosphorylation and nuclear translocation of Smad2/3 in vascular endothelial cells (ECs). PI3K-C2α was necessary for TGFβ-induced increase in phosphatidylinositol 3,4-bisphosphates in the plasma membrane and TGFβ receptor internalization into the SARA-containing early endosomes, but not for phosphatidylinositol 3-phosphate enrichment or localization of SARA in the early endosomes. PI3K-C2α was also required for TGFβ receptor-mediated formation of SARA-Smad2/3 complex. Inhibition of dynamin, which is required for the clathrin-dependent receptor endocytosis, suppressed both TGFβ receptor internalization and Smad2/3 phosphorylation. TGFβ1 stimulated Smad-dependent VEGF-A expression, VEGF receptor-mediated EC migration, and capillary-like tube formation, which were all abolished by either PI3K-C2α knockdown or a dynamin inhibitor. Finally, TGFβ1-induced microvessel formation in Matrigel plugs was greatly attenuated in EC-specific PI3K-C2α-deleted mice. These observations indicate that PI3K-C2α plays the pivotal role in TGFβ receptor endocytosis and thereby Smad2/3 signaling, participating in angiogenic actions of TGFβ.

Keywords: Endosome; Endothelial Cell; PI3K-C2alpha; Phosphatidylinositol Kinase (PI Kinase); Receptor Endocytosis; SARA; SMAD Transcription Factor; TGF-B Receptor; Transforming Growth Factor Beta (TGF-B); Vascular Endothelial Growth Factor (VEGF).

FIGURE 1.

PI3K-C2α is required for TGFβ1-induced Smad2/3 phosphorylation in EC. A, siRNA-mediated knockdown of PI3K isoforms. HUVECs were transfected with PI3K-C2α#1 (C2α#1)-, PI3K-C2β (C2β)-, p110α-, Vps34-, and Smad4-specific siRNA or scrambled (sc)-siRNA, and the expression of the PI3K proteins, Smad4, and β-actin as a loading control were analyzed with immunoblotting. Upper panel, representative blots. Lower panel, quantified data. B, time-dependent phosphorylation of Smad3 in response to TGFβ1 in C2α#1, C2α#2, or sc-siRNA transfected HUVECs. Serum-starved cells were stimulated with TGFβ1 (5 ng/ml) for the indicated time periods. The cell lysates were subjected to immunoblot analysis for p-Smad3 and total Smad2 and Smad3. Upper left panel, effects of siRNA-mediated knockdown of PI3K-C2α. Lower left panel, representative blots of Smads. Right panel, quantified data. C, effects of knockdown of PI3K isoforms on TGFβ1-induced Smad2 and Smad3 phosphorylation. HUVECs that had been transfected with either of C2α#1-, C2β-, p110α-, and Vps34-specific siRNAs or sc-siRNA were stimulated with TGFβ1 (5 ng/ml) for 30 min, followed by the immunoblot analysis for p-Smad2, p-Smad3, and total Smad2 and Smad3. Upper panel, representative blots. Lower panel, quantified data. D, time-dependent phosphorylation of Smad1/5/8 in response to TGFβ1 in HUVECs. The cells were treated as in B. The cell lysates were subjected to immunoblot analysis for p-Smad1/5/8 and total Smad5. E, time-dependent phosphorylation of Smad1/5/8 in response to BMP9 in HUVECs. Serum-starved cells were stimulated with BMP9 (10 ng/ml) for the indicated time periods. The cell lysates were subject to immunoblot analysis for p-Smad1/5/8 and total Smad5. In A–E, the data are means ± S.E. of three or four determinations (n = 3 or 4.). In all figures, the asterisks indicate statistical significance between the indicated groups at the levels of p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***). ns, statistically not significant.

FIGURE 2.

PI3K-C2α is required for TGFβ-induced Smad2/3 phosphorylation in EC but not smooth muscle cells or epithelial cells. A, TGFβ1-induced time-dependent phosphorylation of Smad2/3 and Smad1/5/8 in HMVECs. Serum-starved cells were stimulated with TGFβ1 (5 ng/ml) for the indicated time periods. The cell lysates were subjected to immunoblot analysis for p-Smad1/5/8 and p-Smad2/3. B, effects of Cre-mediated deletion of C2α on TGFβ1-induced Smad3 phosphorylation in MLECs. The cells were infected with adenoviruses encoding LacZ (Ad-LacZ) or Cre recombinase (Ad-Cre), stimulated as in A, and analyzed for p-Smad3. C, effects of Cre-mediated deletion of C2α on TGFβ1-induced Smad3 phosphorylation in MASM cells. The cells were treated and analyzed as in A. D, effects of knockdown of PI3K isoforms on TGFβ1-induced phosphorylation of Smad2/3. Caco2 were treated and analyzed as in A. In A–D, upper panels indicate representative blots, and lower panels indicate quantified data (n = 3–5). The data are expressed as multiples relative to the values in TGFβ1-nonstimulated sc-siRNA transfected cells or Ad-Cre-transfected cells.

FIGURE 3.

TGFβ1-induced nuclear translocation of Smad2/3 depends on PI3K-C2α in EC. A and B, immunofluorescent staining of Smad2/3 (A) and p-Smad3 (B) in TGFβ1-stimulated HUVECs. The cells were transfected with C2α#1 siRNA or sc-siRNA and stimulated with TGFβ1 (5 ng/ml) for 30 min, followed by anti-Smad2/3 antibody or anti-p-Smad3 antibody staining. Nuclei were stained by DAPI. Left panels, representative confocal images of the stained cells. Right panels, quantified data. The data were obtained from 48 cells per group. Scale bar, 20 μm. C, HUVECs were transfected with either wtC2α or C2α-siRNA-resistant C2α (C2α^r) and either C2α#1 siRNA or sc-siRNA. The cells were stimulated with TGFβ1 (5 ng/ml) for 30 min, followed by immunoblot analysis for p-Smad3. Left panel, representative blots. Right panel, quantified data (n = 3). D, HUVECs were transfected with either GFP-wtC2α, GFP-C2α^r, or GFP-tagged kinase deficient mutant of C2α^r (GFP-kdC2α^r) and either C2α-specific siRNA or sc-siRNA. The cells were stimulated with TGFβ1 (5 ng/ml) for 30 min, followed by immunofluorescent staining with anti-Smad2/3. Nuclei were stained by DAPI. Scale bar, 20 μm. #, transfected cells.

FIGURE 4.

TGFβ signaling requires PI3K-C2α distally to TGFβ1-induced ALK5 phosphorylation and endocytosis. A, effects of C2α depletion on TGFβ1-induced ALK5 phosphorylation. Left panel, representative blots. Right panel, quantified data. HEK293T cells were transfected with either FLAG-wtALK5 or FLAG-caALK5 and either C2α#1 siRNA or sc-siRNA. The cell lysates were immunoprecipitated with anti-Flag antibody, followed by immunoblotting (IB) using anti-Flag and anti-phosphoserine (p-Ser) antibodies. A portion of the cell lysates was analyzed for the expression of the indicated proteins with IB. IP, immunoprecipitation. The data are from three-independent experiments, which yielded comparable results, and are expressed as multiples over the values in wtALK5 expressed in sc-siRNA-transfected cells. B, effects of the endocytosis inhibitor dynasore and the ALK5 inhibitor (iALK5) on TGFβ1induced-Smad3 phosphorylation. The cells were prepretreated or not with dynasore (80 μm) or ALK5 inhibitor (5 μm) for 30 min and stimulated with TGFβ1 (5 ng/ml) for 30 min (n = 5). C, effects of the expression of RFP wild-type dynamin2 (RFP-wtDyn2) or RFP dominant negative dynamin2 (RFP-dnDyn2) on TGFβ1-induced nuclear translocation of Smad2/3. The cells were transfected with the RFP-wtDyn2 or RFP-dnDyn2 expression vectors and stimulated with TGFβ1 (5 ng/ml) for 30 min, followed by anti-Smad2/3 immunostaining. Nuclei were stained by DAPI. The arrowheads denote the transfected cells. Scale bar, 20 μm.

FIGURE 5.

PI3K-C2α is required for TGFβ1-induced internalization of TGFβ receptor into the early endosomes in EC. A, effects of C2α depletion or dynasore on TGFβ1-induced internalization of endogenous ALK5. The cells were either transfected with C2α#1 siRNA or sc-siRNA or pretreated with dynasore (80 μm) for 30 min and stimulated with TGFβ1 (5 ng/ml) for 30 min. Left panel, representative confocal images of the stained cells. Right panel, quantified data of fluorescence intensity per cell that were obtained from 24 cells per group. Nuclei were stained by DAPI. Scale bar, 20 μm. B, effects of ALK5 knockdown on the nuclear staining in anti-ALK5 immunostaining. HUVECs were transfected with ALK5-specific siRNA (ALK5-siRNA) or sc-siRNA, followed by anti-ALK5 immunofluorescent staining. Nuclei were stained by DAPI. Scale bar, 20 μm. #, nonspecific nuclear signals. C, effects of the expression of a kinase-deficient C2α mutant on TGFβ1-induced ALK5 internalization. HUVECs were transfected with either GFP-C2α^r or GFP-kdC2α^r and either C2α#1 siRNA or sc-siRNA. The cells were stimulated with TGFβ1 (5 ng/ml) for 30 min, followed by immunofluorescent staining with anti-ALK5 antibody. Nuclei were stained by DAPI. #, transfected cells. D, double immunofluorescent staining of ALK5 (green) and EEA1 (red) in TGFβ1-stimulated HUVECs. The cells were transfected with C2α#1 siRNA or sc-siRNA and stimulated with TGFβ1 (5 ng/ml) for 30 min. Left panel, representative confocal images of the stained cells. Magnified views of the boxed areas are also shown. Nuclei were stained by DAPI. Scale bar, 20 μm. Right panel, quantified data of the ALK5/EEA1-double positive vesicle numbers per cell that were obtained from 24 cells per group. E, PLA staining of ALK5 and EEA1 (green) in TGFβ1-stimulated HUVECs. The cells were transfected with C2α#1 siRNA or sc-siRNA and stimulated with TGFβ1 (5 ng/ml) for 30 min. Nuclei were stained by DAPI. Upper panel, representative confocal images of the stained cells. Scale bar, 20 μm. Lower panel, quantified data of the numbers of ALK5/EEA1 interactions per cell that were obtained from 24 cells per group.

FIGURE 6.

PI3K-C2α is required for TGFβ receptor internalization into SARA-containing endosomes in EC. A, effects of SARA depletion on TGFβ1-induced Smad2/3 phosphorylation in HUVECs. The cells were transfected with SARA-specific siRNA or sc-siRNA. Serum-starved cells were stimulated with TGFβ1 (5 ng/ml) for the indicated time periods and subjected to immunoblot analysis for p-Smad2, p-Smad3, and total Smad2 and Smad3. Left panel, representative blots. Right panel, quantified data (n = 3). B, double immunofluorescent staining of EEA1 (red) and SARA (green) in TGFβ1-stimulated HUVECs. The cells were transfected with C2α#1 siRNA or sc-siRNA and stimulated with TGFβ1 (5 ng/ml) for 30 min. Nuclei were stained by DAPI. Right panel, quantified data of the numbers of SARA (upper panel), EEA1 (middle panel), and SARA/EEA1-double (lower panel) positive vesicle numbers per cell that were obtained from 48 cells per group. Scale bar, 20 μm. C, PLA staining (green) of ALK5 and SARA and anti-EEA1 immunostaining (red) in TGFβ1-stimulated HUVECs. The cells were transfected with C2α#1 siRNA or sc-siRNA and stimulated with TGFβ1 (5 ng/ml) for 30 min. Left panel, representative confocal images of the stained cells. Green and red denote PLA signals and immunostaining signals, respectively. Nuclei were stained by DAPI. Scale bar, 20 μm. Right panel, quantified data of the numbers of PLA signals (left graph) and PLA signal/anti-EEA1-double positive vesicle numbers per cell (right graph) that were obtained from 24 cells per group.

FIGURE 7.

PI3K-C2α is not required for PtdIns(3)P enrichment or the localization of SARA in the endosomes but for TGFβ-induced increase in PtdIns(3,4)P2 in lamellipodia. A, GFP-FYVE^SARA fluorescence. The HUVECs were co-transfected with the GFP-FYVE^SARA expression vector and either C2α#1 siRNAs or sc-siRNA and stimulated with TGFβ1 (5 ng/ml) for 30 min or left untreated. Nuclei were stained by DAPI. Left panel, representative confocal images. Right panel, quantified data of the numbers of GFP-FYVE fluorescence-positive vesicles per cell that were obtained from 48 cells per group. B, anti-SARA immunofluorescent staining. The HUVECs were transfected with either C2α-siRNAs or sc-siRNA and stimulated with TGFβ1 (5 ng/ml) for 30 min or left untreated. Nuclei were stained by DAPI. Left panel, representative confocal images. Right panel, quantified data of anti-SARA-positive vesicles per cell that were obtained from 48 cells per group. Scale bar, 20 μm. C, GFP-PH^TAPP1 fluorescence. The HUVECs were co-transfected with the GFP-PH^TAPP1 expression vector and either C2α-siRNAs or sc-siRNA and stimulated with TGFβ1 (5 ng/ml) for 30 min or left untreated. Nuclei were stained by DAPI. Confocal images at 2, 5, and 10 min after the additions of TGFβ1 or vehicle are shown.

FIGURE 8.

SARA-Smad3 complex formation is dependent on PI3K-C2α. A–C, analyses of complex formation between SARA and Smad2/3 by co-immunoprecipitation-immunoblotting in HEK293T cells. A, the cells were transfected with the expression vectors for either wtSARA or ΔSBD-SARA and either wtALK5 or caALK5, with or without Smad2/3 expression vectors. The cell lysates were immunoprecipitated with anti-SARA antibody, followed by immunoblotting (IB) using anti-SARA, anti-Smad3, or anti-Smad2 antibody. B and C, the cells were co-transfected with the expression vectors for either Smad2 or Smad3, either wtSARA or ΔSBD-SARA, either wtALK5 or caALK5, and either C2α#1 siRNA or sc-siRNA. The cell lysates were immunoprecipitated with anti-SARA antibody, followed by immunoblotting using anti-SARA, anti-Smad3 antibody in B, and anti-Smad2 antibody in C. Portions of the cell lysates were analyzed for the expression of the indicated proteins with immunoblotting (Input). IP, immunoprecipitation; Ctrl, control. Left panel, representative blots. Right panel, quantified data of the amounts of Smad3 and Smad2 in immunoprecipitates from the cells transfected as indicated. The data are means ± S.E. from four independent experiments, which yielded comparable results, and expressed as multiples over the values in wtALK5- and sc-siRNA-transfected cells. D, PLA staining of SARA and Smad2/3 (green) and anti-EEA1 immunostaining (red) in TGFβ1-stimulated HUVECs. The cells were transfected with C2α#1 siRNA or sc-siRNA and stimulated with TGFβ1 (5 ng/ml) for 30 min. Left panel, representative confocal images of the stained cells. Green and red denote PLA signals and immunostaining signals, respectively. Nuclei were stained by DAPI. Scale bar, 20 μm. Right panel, quantified data of the numbers of PLA signals (left graph) and PLA signal/anti-EEA1 staining-double positive vesicle numbers per cell (right graph) that were obtained from 24 cells per group.

FIGURE 9.

PI3K-C2α is required for TGFβ1-induced, Smad-dependent VEGF-A production in EC. HUVECs were transfected with either of C2α(#1), C2β, p110α, Vps34, and Smad4-specific siRNA or sc-siRNA and stimulated with TGFβ1 in the presence and absence of the indicated inhibitors. A, VEGF-A (VEGFA) mRNA expression in the cells stimulated with TGFβ1 (5 ng/ml) for 6 h. The VEGFA mRNA expression levels were determined with real time PCR and were corrected for 18 S rRNA level (n = 6). B, VEGFA mRNA expression in the cells were transfected with either of C2α#1, C2α#2, and sc-siRNA and stimulated with TGFβ1 (5 ng/ml) for 6 h. The VEGFA mRNA expression levels were determined with real time PCR and were corrected for 18 S rRNA level (n = 3). C, VEGF-A peptide concentrations in the media of the cells stimulated with TGFβ1 (5 ng/ml) for 12 h (n = 3). D, effects of dynasore and ALK5 inhibitor on VEGFA mRNA expression. The cells were prepretreated or not with dynasore (80 μm) or iALK5 (5 μm) for 30 min and stimulated with TGFβ1 (5 ng/ml) for 6 h (n = 5). In A–D, the data are expressed as multiples over the values in TGFβ1-nonstimulated sc-siRNA-transfected or vehicle-treated control cells.

FIGURE 10.

TGFβ1-induced endothelial cell migration and tube formation are VEGFR2-mediated and PI3K-C2α-dependent. A and B, effects of inhibitors of VEGFR2 and ALK5 on TGFβ1- and VEGF-induced cell migration. HUVECs were stimulated with TGFβ1 (5 ng/ml) or VEGF (50 ng/ml) in the presence and absence of iALK5 (5 μm) and the VEGFR2 inhibitor (iVEGFR2) (10 μm). Cell migration was determined with scratch wounding healing assay. A, representative microscopic views. B, quantified data (n = 4). C and D, effects of C2α-knockdown on TGFβ1- and VEGF-induced tube formation. siRNA-transfected cells were stimulated with VEGF-A (50 ng/ml) or TGFβ1 (5 ng/ml) for 12 h. C, representative microscopic views; D, quantified data (n = 4). E and F, effects of dynasore (80 μm), iVEGFR2 (10 μm), and iALK5 (5 μm) on TGFβ1- and VEGF-induced tube formation. E, representative microscopic views. F, quantified data (n = 4).

FIGURE 11.

Endothelial PI3K-C2α is necessary for TGFβ1-induced angiogenesis in vivo. Matrigels containing PBS (vehicle) or TGFβ1 were injected into the subcutaneous tissues on the back of EC-specific C2α-deleted (C2α^ΔEC), SMC-specific C2α-deleted (C2α^ΔSMC), and control (C2α^flox/flox) mice. Matrigel plugs were removed 10 days later and analyzed for microvascular formation by immunohistochemistry using anti-vWF antibody. A and D, representative views of anti-vWF immunostained sections of Matrigel plugs in C2α^ΔEC (A) and C2α^ΔSMC (D) mice. B and E, quantified data of neovessel formation in Matrigel plugs (seven mice per group). C and F, representative gross views of Matrigel plugs resected from mice. G, double immunofluorescent staining of CD31 (red) and p-Smad2 (green) in Matrigel plugs in C2α^ΔEC (left panel) and C2α^ΔSMC (right panel) mice. Nuclei were stained by DAPI. Scale bar, 50 μm.

FIGURE 12.

Receptor endocytosis and ALK5 are necessary for TGFβ1-induced angiogenesis in vivo. Matrigels containing of PBS (vehicle) or TGFβ1 and dynasore (200 μm) or iALK5 (20 μm) were injected into the subcutaneous tissues on the back. Matrigel plugs were removed 10 days later and analyzed for microvascular formation by immunohistochemistry using anti-vWF antibody. A, representative gross views of Matrigel plugs resected from mice. B and D, representative views of anti-vWF immunostained sections of Matrigel plugs containing dynasore (left panel) and iALK5 (right panel). C and E, quantified data of the effects of dynasore and iALK5 on neovessel formation in Matrigel plugs (six mice per group).