Role Played by Paxillin and Paxillin Tyrosine Phosphorylation in Hepatocyte Growth Factor/Sphingosine-1-Phosphate-Mediated Reactive Oxygen Species Generation,Lamellipodia Formation,and Endothelial Barrier Function

Reagents

Human lung microvascular ECs (HLMVECs) and endothelial basal medium (EBM-2) were obtained from Lonza (San Diego, CA). Gold antifade mounting medium, 4′,6-diamidino-2-phenylindole dihydro-chloride, Hoechst and precast Tris-glycine polyacrylamide gel electrophoresis (PAGE) gel, Alexa Fluor 488, Alexa Fluor 568, and Alexa Fluor phalloidin 568 were procured from Life Technologies (Eugene, OR). HGF was obtained from PeproTech (Rocky Hill, NJ). Antibodies to phospho-c-Abl, c-Abl, phospho-paxillin, and paxillin as well as cell lysis buffer were obtained from Cell Signaling Technology (Danvers, MA). Antibody to p47 was obtained from Millipore (Bedford, MA). Scrambled siRNA and siRNA for paxillin, c-Abl, and p47 ^phox were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). pHyPer-cyto plasmid was purchased from Evrogen (Moscow, Russia). RNA transfection reagent Gene Silencer was obtained from Genlantis (San Diego, CA). In vivo RNA transfection reagent jetPEI was obtained from Polyplus Transfection (Illkirch, France). FuGENE HD transfection reagent was obtained from Promega (Madison, WI). Electrical cell-substrate impedance sensing electrodes 8W1E were procured from Applied Biophysics (Troy, NY).

EC culture

HLMVECs cultured in complete medium (EBM-2) were maintained at 37°C in 5% CO₂ and grown to contact-inhibited monolayers that revealed typical cobblestone morphology. Cells were then detached with 0.05% trypsin, suspended in fresh medium, and cultured on gold electrodes for electrical resistance determinations, on glass cover-slips for fluorescent microscopy studies, or in 60–100-mm culture dishes for preparation of cell lysates and Western blot analysis.

Transfection of siRNA and complementary DNA (cDNA)

Endogenous paxillin, c-Abl, and p47 in HLMVECs were depleted using gene-specific siRNA, as described elsewhere. ²⁰ In brief, predesigned human paxillin, c-Abl siRNA, p47, or nonspecific/non-targeting siRNA was used to transfect HLMVECs (5–7 passages). Before transfection, cells were starved in 2% fetal bovine serum (FBS) EBM-2 medium overnight. The next day, 50 nM scrambled, paxillin, c-Abl, or p47 siRNA complexes were prepared in Gene Silencer transfection reagent according to the manufacturer's recommendations, cells were transfected in serum-free medium for 4 hours, and the medium was replaced with fresh complete EBM-2 medium supplemented with 10% FBS and growth factors. Cells were used 72 hours after transfection. To determine intracellular hydrogen peroxide generation, HLMVECs were transfected with 1 μg/mL pHyPer-cyto plasmid. The cDNA of control plasmid or pHyPer-cyto was incubated with FuGene HD transfection reagent in serum-free EGM-2 medium according to the manufacturer's recommendations. After 3 hours, the medium of transfected cells was replaced with complete EGM-2 medium containing 10% FBS, and the cells were used after 72 hours. Similarly, HLMVECs were transfected with control plasmid or with nonphosphorylatable paxillin mutants for study of the role played by paxillin phosphorylation in lamellipodia formation.

Immunoblotting and immunoprecipitation

Immunoblotting and immunoprecipitation were performed as described elsewhere. ³⁶ In brief, after appropriate treatments, cells were pelleted in ice-cold phosphate-buffered saline, lysed in standard lysis buffer (Cell Signaling, Beverly, MA), and sonicated. Lysates were then centrifuged at 1,000 g for 10 minutes at 4°C, supernatants were collected, and protein was assayed using a Pierce BCA protein assay kit. For immunoprecipitation experiments, equal amounts of protein (1 mg) from each sample were precleared with control immunoglobulin G conjugated to A/G agarose beads at 4°C for 1 hour, and supernatants were collected and incubated overnight with primary antibody conjugated to A/G agarose beads at 4°C. The next day, samples were centrifuged at 1,000 g for 1 minute in a microfuge centrifuge, and the pellets containing agarose beads were washed three times with lysis buffer at room temperature. After centrifugation at 1,000 g for 1 minute, the beads were collected by removing supernatant buffer, and 40 μL of sodium dodecyl sulfate (SDS) sample buffer (100 mM Tris-HCl [pH 6.8], 4% SDS, 0.1% bromophenol blue, 20% glycerol, 200 mM dithiothreitol) was added to the beads and boiled. Lysates were then subjected to 10% SDS-PAGE followed by Western blotting.

Proteins were detected by immunoblotting using appropriate primary antibodies and horseradish peroxidase–conjugated antirabbit or anti-mouse secondary antibodies. Band intensities were quantified by densitometry using ImageJ software.

Measurement of endothelial permeability by transendothelial electrical resistance

Endothelial permeability changes were measured using the highly sensitive biophysical assay with an electrical cell-substrate impedance sensing system (Applied Biophysics, Troy, NY), as described elsewhere. ²⁰ Transendothelial resistance was measured dynamically across the monolayer, and the effect of HGF challenge was monitored over a period of 6 hours. Resistance was normalized to the initial voltage and expressed as a fraction of the normalized resistance value.

Quantification of lamellipodia and colocalization

Lamellipodia were quantified as described elsewhere. ²⁰ In brief, for each image background signal was subtracted by drawing a region of interest around the cell periphery of individual cells. All areas outside the cell were cleared to best visualize the leading edges, including cell periphery, and the fluorescence intensity within the entire cell was summed using the MBF ImageJ bundle (Tony Collins, McMaster University, Hamilton, Ontario, and Wayne Rasband, National Institutes of Health, Bethesda, MD). Pearson's correlation coefficient (PCC) was used as a statistic for quantifying colocalization. PCC was defined using the following formula:

PCC = ∑ i ( R i - R ˉ ) × ( G i - G ˉ ) ∑ i ( R i - R ˉ ) 2 × ∑ i ( G i - G ˉ ) 2 ,

where R_i and G_i refer to the intensity values of the red and green channels, respectively, of pixel i, and R̄ and R̄ refer to the mean intensities of the red and green channels, respectively, across the entire image. Values near 1 reflect the fluorescence intensities of two images that are perfectly and linearly related to one another. Values near 0 reflect distribution of probes that are uncorrelated with each one another.

In vivo transfection of mouse lungs with paxillin siRNA

Mice were housed in a pathogen-free barrier facility maintained by the University of Illinois at Chicago Biologic Resources Laboratory. All experiments involving mice were conducted under protocols approved by the Institutional Animal Care and Use Committee of the University of Illinois at Chicago. In vivo paxillin knockdown in mouse lungs was achieved by intratracheal paxillin siRNA transfection (5 mg/kg of mouse body weight) using jetPEI as a transfecting agent, as described elsewhere. ³⁷ Adult male C57BL/6J mice, 8–10 weeks old with an average weight of 20–25 g, were anesthetized, intubated with a 20-G × 1″ catheter (Exel International Medical Products), and administered 20–30 μL of siRNA working solution; mice were then analyzed 72 hours after transfection.

In vivo model of ventilator-induced lung injury (VILI)

To explore the effect of paxillin on pulmonary permeability, scrambled RNA (scRNA)– or paxillin siRNA–treated mice were anesthetized with an intraperitoneal injection of ketamine (75 mg/kg) and acepromazine (1.5 mg/kg). Tracheotomy was performed, and the trachea was cannulated with a 20-G intravenous catheter. Ventilation was performed at a tidal volume of 30 mL/kg, 75 breaths per minute, and 0 positive end-respiratory pressure for 4 hours. Intravenous fluid boluses of 0.1 mL of 0.90% NaCl saline solution were given to maintain a mean arterial pressure >65 mmHg. Bronchial alveolar lavage (BAL) fluid was obtained at the end of the experiment by intratracheal injection of 1 mL of sterile Hanks balanced salt solution followed by gentle aspiration. The BAL fluid was used for measurement of cell count and protein concentration.

Statistical analysis

Results are expressed as means ± SD of three to five independent experiments. Statistical significance was assessed using analysis of variance, followed by Bonferroni multiple-comparison post hoc tests. Differences with P < 0.05 were considered statistically significant.

Down-regulation of paxillin with siRNA attenuates HGF- and S1P-induced lamellipodia formation and barrier function in lung ECs

We have previously demonstrated that HGF- and S1P-stimulated reorganization and colocalization of actin and cortactin in lamellipodia of lung ECs was dependent on PI3K/Akt signaling. ²⁰ As actin regulates formation of cell protrusions at the leading edge of cells and paxillin is a cytoskeletal protein, we therefore determined the role played by paxillin in lamellipodia formation and barrier function in lung ECs. Paxillin was down-regulated (~80% of expression) in HLMVECs by siRNA (Fig. 1C), and treatment of these cells with HGF or S1P significantly suppressed colocalization of actin and cortactin to lamellipodia (Fig. 1A). In parallel experiments, paxillin depletion with siRNA reduced endothelial barrier enhancement by HGF or S1P (Fig. 1B). Taken together, these results demonstrate a key role for paxillin in HGF- or S1p-induced lamellipodia formation and barrier enhancement in lung ECs.

OPEN IN VIEWER

Figure 1.

Paxillin mediates sphingosine-1-phosphate (S1P)– or hepatocyte growth factor (HGF)-induced endothelial cell lamellipodia formation and barrier function. Human lung microvascular endothelial cells were transfected with scrambled RNA (scRNA) or paxillin small interfering RNA (siRNA; 50 nM) for 72 hours. Cells were then subjected to either immunofluorescent staining for lamellipodia formation with cortactin antibody and phalloidin after S1P (1 μM) or HGF (20 ng/mL) treatment for 15 minutes (A) or electrical cell-substrate impedance measurement for barrier function (B), as described in “Methods.” At least 20 cells were analyzed for each condition; results shown are representative of three independent experiments. Knockdown of paxillin was confirmed by Western blotting, as shown in C. *P < 0.01, compared with vehicle control; ^#P < 0.05, compared with scRNA plus S1P or scRNA plus HGF. Scale bar = 10 μm. Pxn: paxillin; siPxn: paxillin siRNA.

Tyrosine phosphorylation of paxillin at Y31 and Y118 is essential for HGF- or S1P-induced lamellipodia formation and barrier function in lung ECs

Earlier studies have shown that paxillin is tyrosine phosphorylated by growth factors and lipopolysaccharide (LPS) in ECs;^37,38 however, the role played by Y31 and Y118 paxillin tyrosine phosphorylation in lamellipodia formation and barrier function is yet to be defined. Exposure of HLMVECs to HGF or S1P stimulated paxillin tyrosine phosphorylation at Y31 and Y118 in a time-dependent manner (Fig. 2A). To address the role played by paxillin tyrosine phosphorylation in lamellipodia formation and endothelial barrier function, HLMVECs were transfected with Y31F and Y118F paxillin mutants (1 μg cDNA/Fugene 6, 72 h) prior to HGF or S1P challenge. As shown in Figure 2B and 2C, overexpression of nonphosphorylatable Y31 and Y118 paxillin mutants significantly reduced HGF- or S1P-mediated redistribution of actin and cortactin to lamellipodia and endothelial barrier enhancement response, respectively. However, overexpression of these two mutants had no effect on HGF- or S1P-induced FAK phosphorylation (data not shown). Interestingly, expression of the Y31 and Y118 paxillin mutants modulated cortactin tyrosine phosphorylation at Y446, suggesting a role for Y31 and Y118 paxillin tyrosine phosphorylation in regulating cortactin phosphorylation (data not shown).

OPEN IN VIEWER

Figure 2.

Paxillin tyrosine phosphorylation is essential for sphingosine-1-phosphate (S1P)– or hepatocyte growth factor (HGF)–induced endothelial cell lamellipodia formation. A, Human lung microvascular endothelial cells (HLMVECs) grown in 35-mm dishes (90% confluence) were treated with S1P (1 μM) or HGF (20 ng/mL) for varying time points, as indicated. Phosphorylation of paxillin at Y31 and Y118 was determined by Western blotting using phospho-specific paxillin antibodies. Results shown are representative of three independent experiments. *P < 0.05, compared with vehicle control; **P < 0.01, compared with vehicle control. B, HLMVECs grown on slide chambers (60% confluence) were transfected with either control plasmid or nonphosphorylatable paxillin mutants (Y31F, Y181F), as described in “Methods.” After 72 hours, cells were treated with S1P (1 μM) or HGF (20 ng/mL) for 15 minutes and subjected to immunofluorescent staining for lamellipodia with cortactin antibody. At least 20 cells were analyzed for each condition; examples shown are representative of three independent experiments. *P < 0.05, compared with control plasmid plus S1P or control plasmid plus HGF; **P < 0.01, compared with control plasmid vehicle control. C, HLMVECs grown on electrical cell-substrate impedance sensing plates were transfected with either control plasmid or nonphosphorylable paxillin mutants (Y31F, Y181F), as described in “Methods.” After 72 h, transendothelial resistance reflecting endothelial cell monolayer barrier properties was monitored before and after S1P and HGF treatment. *P < 0.05, compared with plasmid control; ^#P < 0.05, compared with S1P- or HGF-treated control plasmid-transfected cells. Pxn: paxillin.

Role played by c-Abl in HGF- or S1P-mediated tyrosine phosphorylation of paxillin and cortactin, lamellipodia formation, and endothelial barrier function

There is evidence for the role played by Src family kinases in agonist-mediated tyrosine phosphorylation of cortactin and paxillin in mammalian cells;^24,27 however, the role played by c-Abl in tyrosine phosphorylation of paxillin is limited. We have recently shown a role for c-Abl, but not for Src or FAK, in LPS-mediated tyrosine phosphorylation of paxillin in lung ECs. We first investigated whether S1P and HGF treatment activates c-Abl. As shown in Figure 3A, treatment of HLMVECs with S1P or HGF shows obvious phosphorylation of c-Abl at Y245 in a time-dependent manner. To further investigate the role played by c-Abl in tyrosine phosphorylation of paxillin, we down-regulated c-Abl using specific siRNA. Both HGF- and S1P-stimulated Y245 phosphorylation of c-Abl in scRNA-transfected HLMVECs and down-regulation of c-Abl with siRNA attenuated paxillin phosphorylation at Y31 and Y118 residues (Fig. 3B), which was accompanied by decreasedlocalization of actin and cortactin in the lamellipodia (Fig. 4A). Furthermore, knockdown of paxillin with siRNA also attenuated HGF- or S1P-induced barrier enhancement of lung ECs (Fig. 4B). These results demonstrate a key role for c-Abl in paxillin tyrosine phosphorylation, lamellipodia formation, and endothelial barrier function.

OPEN IN VIEWER

Figure 3.

c-Abl tyrosine kinase mediates paxillin phosphorylation and lamellipodia formation in human lung endothelial cells. A, Human lung microvascular endothelial cells (HLMVECs; 90% confluence) were treated with sphingosine-1-phosphate (S1P; 1 μM) or hepatocyte growth factor (HGF; 20 ng/mL) for varying times. Phosphorylation of c-Abl at tyrosine Y245 was analyzed by Western blotting. Results shown are representative of three independent experiments. *P < 0.05, compared with vehicle control; **P < 0.01, compared with vehicle control; P < 0.001, compared with vehicle control. B, HLMVECs (60% confluence in 35-mm dishes) were transfected with scrambled RNA (scRNA) or c-Abl small interfering RNA (siRNA; 50 nM) for 72 hours, as described in “Methods.” After 72 hours of transfection, cells were treated with S1P (1 μM) or HGF (20 ng/mL) for 30 minutes, phosphorylation of paxillin was done at Y31 and Y118, and c-Abl expression was analyzed by Western blotting. Results shown are representative of three independent experiments. *P < 0.05, compared with scRNA vehicle; ^#P < 0.05, compared with scRNA plus S1P or scRNA plus HGF. Pxn: paxillin; sicAbl: c-Abl siRNA.

OPEN IN VIEWER

Figure 4.

c-Abl mediates sphingosine-1-phosphate (S1P)– and hepatocyte growth factor (HGF)–induced lamellipodia formation and barrier enhancement. A, Human lung microvascular endothelial cells (HLMVECs) grown on slide chambers (60% confluence) were transfected with scrambled RNA (scRNA) or c-Abl small interfering RNA (siRNA; 50 nM, 48 hours) prior to S1P (1 μM) or HGF (20 ng/mL) challenge for 30 minutes. Cells were fixed, and lamellipodia formation was visualized by immunofluorescent staining with cortactin antibody or phalloidin for actin. At least 20 cells were analyzed for each condition. Shown is an immunofluorescence image representative of three independent experiments. *P < 0.05, compared with scRNA transfection groups. B, HLMVECs were transfected with scRNA or c-Abl siRNA for 48 hours as described in A. Cells were then reseeded in 8-well electrical cell-substrate impedance sensing plates for another 24 hours, and transendothelial electrical resistance was monitored before and after S1P or HGF treatment and normalized to initial resistance. *P < 0.05, compared with scRNA control; ^#P < 0.05, compared with scRNA plus S1P or scRNA plus HGF. sicAbl: c-Abl siRNA.

c-Abl siRNA attenuates HGF-induced localization of p47 ^phox and actin in lamellipodia

Having established that HGF- or S1P-mediated paxillin tyrosine phosphorylation is c-Abl dependent in lung ECs, we next determined whether c-Abl regulates localization of p47 ^phox , a NADPH oxidase component, with actin in lamellipodia. As shown in Figure 5, knockdown of c-Abl with siRNA attenuated HGF-mediated redistribution and colocalization of p47 ^phox and actin in lamellipodia.

OPEN IN VIEWER

Figure 5.

c-Abl mediates hepatocyte growth factor (HGF)–induced p47 ^phox localization at lamellipodia. Human lung microvascular endothelial cells were transfected with scrambled RNA (scRNA) or c-Abl small interfering RNA (siRNA; 50 nM) for 72 hours and then treated with HGF (20 ng/mL) for 30 minutes. After treatment, cells were stained with anti-p47 ^phox antibody and phalloidin for actin. At least 20 cells were analyzed for each condition. Shown is an immunofluorescence image representative of three independent experiments. *P < 0.05, compared with scRNA transfection groups. sicAbl: c-Abl siRNA.

Paxillin and paxillin tyrosine phosphorylation are involved in HGF- or S1P-induced p47 ^phox /ROS localization in lamellipodia of lung ECs

We have previously shown that HGF stimulated p47 ^phox translocation and ROS generation in lamellipodia of lung ECs. ³⁹ Having established a role for paxillin and paxillin tyrosine phosphorylation in HGF- or S1P-mediated lamellipodia formation and endothelial barrier function, we next determined the potential link between paxillin and paxillin tyrosine phosphorylation in accumulation of ROS in lamellipodia. To elucidate the role played by paxillin in HGF- or S1P-induced ROS in lamellipodia, ECs were transfected with p-Hyper-cytosol, and living cells were analyzed by confocal microscopy. As shown in Figure 6A, 0HGF or S1P stimulated hydrogen peroxide accumulation in the cell periphery, which was significantly inhibited by paxillin siRNA. Next, to determine the role played by paxillin tyrosine phosphorylation in ROS accumulation in lamellipodia, HLMVECs were transfected with the paxillin mutant Y118F for 48 h prior to S1P challenge. As expected, S1P enhanced colocalization of actin and p47 ^phox in lamellipodia, which was attenuated by the paxillin Y118F mutant (Fig. 6B). These results establish a key role for HGF- or S1P-mediated paxillin and paxillin tyrosine phosphorylation in ROS accumulation in lamellipodia.

OPEN IN VIEWER

Figure 6.

Paxillin tyrosine phosphorylation regulates hepatocyte growth factor (HGF)– or sphingosine-1-phosphate (S1P)–induced p47 ^phox /reactive oxygen species accumulation in lamellipodia of human lung endothelial cells. A, Human lung microvascular endothelial cells (HLMVECs; 60% confluence) were transfected with pHyPer-cyto plasmid (1 μg/mL) for 24 hours and then transfected with scrambled RNA (scRNA) or paxillin small interfering RNA (siRNA; 50 nM) for 48 hours. At the end of transfection, cells were treated with S1P (1 μM) or HGF (20 ng/mL) for 15 minutes. Accumulation of pHyPer-cyto signal at the cell periphery is indicated by arrows. At least 20 cells were analyzed for each condition. Shown is an image representative of three independent experiments. *P < 0.01, compared with scRNA vehicle; ^#P < 0.05, compared with scRNA plus S1P or scRNA plus HGF. B, HLMVECs were transfected with control plasmid or PxnY118F mutant plasmid (1 μg/mL) for 72 hours, and cells were treated with S1P (1 μM) for 15 minutes. After treatment, cells were stained with anti-p47 ^phox antibody and phalloidin for actin. At least 20 cells were analyzed for each condition. Shown is an image representative of three independent experiments. *P < 0.05, compared with control vector. siPxn: paxillin siRNA.

Knockdown of paxillin with siRNA in mouse lungs inhibits VILI

Our in vitro results suggested a role for paxillin and its tyrosine phosphorylation in HGF- or S1P-mediated lamellipodia formation and endothelial barrier enhancement in HLMVECs; therefore, we next examined the role played by paxillin in vivo in VILI. Paxillin in mouse lungs was down-regulated by intratracheally instilling paxillin siRNA for 72 h prior to ventilator challenge. As shown in Figure 7A and 7B, LPS challenge induced significant leakage of protein in scRNA-treated mice and infiltration of inflammatory cells into the alveolar space. The above indicators of pulmonary leak and inflammation were significantly attenuated in paxillin knockdown lungs. Furthermore, histochemical analysis of paraffin-embedded mouse lungs revealed inflammatory cell recruitment and areas of alveolar hemorrhage indicative of vascular disruption in scrambled siRNA–treated mice. In contrast, there was a marked attenuation of lung injury in paxillin siRNA–treated mice (Fig. 7C). Knockdown of paxillin in mouse lungs was confirmed by Western blotting of lung tissue lysates obtained from mice administered scrambled and paxillin siRNA (Fig. 7A). These results suggest that paxillin is an important regulator of ventilator-induced pulmonary vascular permeability in vivo.

OPEN IN VIEWER

Figure 7.

Knockdown of paxillin in mouse lungs attenuates ventilator-induced lung permeability and injury. A, B, C57BL/6J mice were instilled with scrambled or paxillin small interfering RNA (siRNA; 5 mg/kg of body weight) for 48 hours prior to ventilation challenge. Lung permeability and injury were assessed by measurement of bronchial alveolar lavage (BAL) fluid cell count and BAL protein concentration. Values are means ± SD from 5 animals. *P < 0.01, compared with scrambled RNA (scRNA) control; **P < 0.05, compared with scRNA plus ventilator-induced lung injury (VILI). C, Hematoxylin-eosin staining of lung sections from scRNA- or paxillin siRNA–transfected mice after 4 hours of ventilation. Images are representative lung sections from 5 mice per group. Magnification = ×20; scale bar = 100 μm. Down-regulation of paxillin was confirmed by Western blotting, as shown in A. siPxn: paxillin siRNA.