AJP - Lung Track the topics, authors and articles important to you
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Lung Cell Mol Physiol 294: L686-L697, 2008. First published February 15, 2008; doi:10.1152/ajplung.00417.2007
1040-0605/08 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
294/4/L686    most recent
00417.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rentsendorj, O.
Right arrow Articles by Pearse, D. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rentsendorj, O.
Right arrow Articles by Pearse, D. B.

Role of vasodilator-stimulated phosphoprotein in cGMP-mediated protection of human pulmonary artery endothelial barrier function

Otgonchimeg Rentsendorj,1 Tamara Mirzapoiazova,2 Djanybek Adyshev,2 Laura E. Servinsky,1 Thomas Renné,3 Alexander D. Verin,4 and David B. Pearse1

1Department of Medicine, Division of Pulmonary and Critical Care Medicine, The Johns Hopkins University, Baltimore, Maryland; 2Department of Medicine, Section of Pulmonary and Critical Care, The University of Chicago Center for Integrative Science, Chicago, Illinois; 3Institute for Clinical Biochemistry and Pathobiochemistry, Julius-Maximilians-University Würzburg, Würzburg, Germany; and 4Vascular Biology Center, Medical College of Georgia, Augusta, Georgia

Submitted 9 October 2007 ; accepted in final form 13 February 2008


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Increased pulmonary endothelial cGMP was shown to prevent endothelial barrier dysfunction through activation of protein kinase G (PKGI). Vasodilator-stimulated phosphoprotein (VASP) has been hypothesized to mediate PKGI barrier protection because VASP is a cytoskeletal phosphorylation target of PKGI expressed in cell-cell junctions. Unphosphorylated VASP was proposed to increase paracellular permeability through actin polymerization and stress fiber bundling, a process inhibited by PKGI-mediated phosphorylation of Ser157 and Ser239. To test this hypothesis, we examined the role of VASP in the transient barrier dysfunction caused by H2O2 in human pulmonary artery endothelial cell (HPAEC) monolayers studied without and with PKGI expression introduced by adenoviral infection (Ad.PKG). In the absence of PKGI expression, H2O2 (100–250 µM) caused a transient increased permeability and pSer157-VASP formation that were both attenuated by protein kinase C inhibition. Potentiation of VASP Ser157 phosphorylation by either phosphatase 2B inhibition with cyclosporin or protein kinase A activation with forskolin prolonged, rather than inhibited, the increased permeability caused by H2O2. With Ad.PKG infection, inhibition of VASP expression with small interfering RNA exacerbated H2O2-induced barrier dysfunction but had no effect on cGMP-mediated barrier protection. In addition, expression of a Ser-double phosphomimetic mutant VASP failed to reproduce the protective effects of activated PKGI. Finally, expression of a Ser-double phosphorylation-resistant VASP failed to interfere with the ability of cGMP/PKGI to attenuate H2O2-induced disruption of VE-cadherin homotypic binding. Our results suggest that VASP phosphorylation does not explain the protective effect of cGMP/PKGI on H2O2-induced endothelial barrier dysfunction in HPAEC.

protein kinase G; cAMP; protein kinase A; small interfering RNA; H2O2

AN INCREASE IN PULMONARY ENDOTHELIAL cGMP concentration has been shown to prevent endothelial barrier dysfunction caused by reactive oxygen species (ROS) both in vitro (19, 22, 36) and in vivo (15, 21, 23, 24). In human pulmonary artery endothelial cells (HPAEC), we (19) recently demonstrated that the protective effect of a membrane-permeable cGMP analog on the increased permeability caused by H2O2 was significantly dependent on the expression of protein kinase G (PKGI), the serine/threonine family kinase activated by cGMP. Multiple phosphorylation targets for PKGI have been identified in smooth muscle cells (20), where it plays a prominent role in mediating smooth muscle relaxation. Less is known about PKGI-dependent signaling in endothelial cells (6, 7, 16). Although PKGI has been shown to inhibit calcium influx in endothelial cells (11, 12), we (19) found that PKGI-mediated endothelial barrier protection occurred without inhibition of an increased intracellular calcium concentration caused by H2O2. Thus, the protective effect of activated PKGI appeared to occur downstream from calcium signaling (19). Endothelial barrier protection secondary to PKGI activation was associated with Ser239 and Ser157 phosphorylation of vasodilator-stimulated phosphoprotein (VASP), a well-known PKGI target (19). These data suggested a possible role for this cytoskeletal protein in the protective effect of PKGI.

VASP has previously been proposed as the likely effecter responsible for the modulation of endothelial barrier function by PKGI. VASP is the founding member of the Ena/VASP family of proteins (26, 34). In endothelium, VASP can be found in association with actin stress fibers, cell-cell adherens junctions, and cell-matrix focal adhesions (7, 30, 34). VASP simultaneously binds to F-actin microfilaments and key cell junction proteins including vinculin, zyxin, and profilin, which promotes actin polymerization by presenting G actin monomers for assembly (25, 31). VASP has three functional domains: an NH2-terminal domain (EVH1) that binds to vinculin and zyxin, a central proline-rich domain that binds profilin, and a COOH-terminal domain (EVH2) that binds F-actin (31, 34). Both PKGI and protein kinase A (PKA; the downstream kinase of cAMP) regulate VASP function. PKGI preferentially phosphorylates Ser239, whereas PKA prefers Ser157 (33), although both Ser sites can be phosphorylated by either kinase. Neither kinase phosphorylates Thr278 in intact cells (2). Unphosphorylated VASP promoted polymerization and bundling of F-actin filaments leading to F-actin stress fiber formation and cell membrane ruffling in endothelial cells (25). Phosphorylation of VASP Ser239 inhibited actin binding and reversed these effects (8). On the basis of the critical location of VASP between the actin cytoskeleton and key cell junction proteins as well as its promotion of F-actin filament construction, it has been hypothesized that VASP phosphorylation may regulate endothelial permeability and explain the barrier-enhancing effects of both PKGI (7, 18, 25, 35) and PKA (5).

The goal of the present study was to determine if the attenuating effects of cGMP/PKGI on H2O2-induced endothelial barrier dysfunction could be explained by Ser phosphorylation of VASP in HPAECs. In our previous study, we chose HPAECs instead of human lung microvascular endothelial cells because HPAECs were found to reproducibly lack PKGI expression in vitro allowing the ability to control PKGI expression through adenoviral expression (19). We therefore utilized the same well-characterized system to examine the role of VASP phosphorylation. We report that potentiation of VASP phosphorylation at Ser157 by either phosphatase inhibition or PKA activation accentuated, rather than inhibited, the transient increased endothelial permeability caused by H2O2. Marked inhibition of VASP expression with small interfering RNA (siRNA) aggravated H2O2-induced barrier dysfunction but had no effect on the cGMP/PKGI-mediated barrier protection. Expression of a phosphorylation-mimicking mutant VASP construct at Ser157 and Ser239 also failed to reproduce the protective effects of activated PKGI. Finally, expression of a phosphorylation-resistant VASP mutant failed to interfere with cGMP/PKGI-mediated protection of VE-cadherin homotypic binding. We conclude that VASP phosphorylation does not explain the protective effect of cGMP/PKGI on ROS-induced endothelial barrier dysfunction in HPAEC.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Cell culture conditions and reagents. HPAEC (lot 4F1355) were obtained from Clonetics (Walkersville, MD) at the third passage and were studied between passages 5 and 7. The cells were cultured as previously described (19) in EBM-2 complete medium (Clonetics) at 37°C in a humidified atmosphere of 5% CO2, 95% air. Cells from each primary flask were detached with 0.05% trypsin, resuspended in fresh culture medium, and passaged into eight-well electrical cell-substrate impedance sensing system (ECIS) electrode arrays for transendothelial electrical resistance (TER) determinations or 12-well plates (on 18-mm #1D coverslips; Fisher Scientific) for fluorescence microscopy. Reagents for SDS-PAGE and Immun-Blot transfer membrane with filter papers were purchased from Bio-Rad (Richmond, CA). Nonstabilized 30% H2O2 was purchased from Fisher Chemical (Pittsburgh, PA). Phorbol myristate acetate (PMA) and Ro-31-8220 were purchased from Calbiochem (San Diego, CA) and dissolved in DMSO. Forskolin and cyclosporin A were purchased from Sigma-Aldrich (St. Louis, MO). The membrane-permeable cGMP analog 8-pCPT-cGMP was purchased from Biolog Life Science Institute (La Jolla, CA). The following antibodies were commercially obtained: rabbit anti-myristoylated alanine-rich C kinase substrate (MARCKS) antibody from Santa Cruz Biotechnology (Santa Cruz, CA), rabbit anti-phospho-MARCKS (Ser152/156) from Cell Signaling Technology, (Danvers, MA), rabbit anti-PKGI antibody from Stressgen Bioreagents (Victoria, BC, Canada), rabbit anti-VASP from Calbiochem, mouse anti-pSer239-VASP from A. G. Scientific (San Diego, CA), mouse monoclonal anti-FLAG antibody from Sigma, mouse anti-human CD144 (cadherin-5) antibody from BD Biosciences-Transduction Laboratories (Lexington, KY), and goat polyclonal anti-actin antibody from Santa Cruz Biotechnology (Santa Cruz, CA). Texas red-X phalloidin was purchased from Molecular Probes (Eugene, OR). The secondary antibody used for immunostaining was Alexa Fluor 594 conjugated to goat anti-mouse IgG (H + L) from Invitrogen-Molecular Probes, and secondary horseradish peroxidase (HRP)-labeled antibodies for Western detection were purchased from Bio-Rad.

Plasmid constructions of wild-type and mutant derivatives of human VASP. Plasmid cDNAs of the four phosphorylation mutants of human VASP (34), AST, SAT, DST, and SDT, were generated as previously described (33). All four constructs were NH2 terminally tagged by a hexahistidin (6xHis) in frame and were cloned into the EcoRI site of a modified pcDNA3 expression vector. Using these mutants, we generated wild-type VASP-SST, Ser phosphorylation-resistant mutant VASP-AAT (S157A/S239A), or Ser-double phosphorylation-mimetic VASP-DDT (S157D/S239D) in the expression vector pEGFP-C1 (Clontech, Palo Alto, CA) to generate a green fluorescence protein (GFP) tag. The constructs VASP-AST (S157A), VASP-SAT (S239A), VASP-DST (S157D), and VASP-SDT (S239D) harboring single phospho-resistant mutations or single phospho-mimetic mutants were generated as follows. The 1.3-kb HindIII (blunted)-XbaI fragment harboring the hexahistidine tag in frame with the cDNA of the mutant VASPs were recloned into the HindIII (blunted) and XbaI site of the pEGFP-C1 to produce the desired in frame clones. The VASP-SST construct harboring the wild-type cDNA of VASP was generated as follows. The 5.3-kb PstI fragment from pESAT harboring the vector backbone and first half of the VASP cDNA that covers the functional Ser157 site was ligated to the 0.5-kb PstI fragment containing the functional Ser239 site of VASP-AST. VASP-AAT was generated by ligating the 5.3-kb PstI fragment of VASP-AST to the 0.5-kb PstI fragment of VASP-SAT, whereas VASP-DDT was generated by ligating the 5.3-kb PstI fragment of VASP-SDT to the 0.5-kb PstI fragment of VASP-DST. Proper directional insertions were verified by restriction mapping and nucleotide sequencing.

Transient expression of plasmid constructs. To ectopically express wild-type and mutant VASP in HPAEC, the cells were transfected using Effectene transfection reagent (Qiagen, Valencia, CA) at a 1:20 DNA:Effectene (µg:µl) ratio. Optimal transfection efficiency was determined with a control plasmid pEGFP-C1 (Clontech) using an Olympus IX51 inverted fluorescence microscope (excitation at 485 nm, emission at 520 nm). The day before transfection, HPAEC were plated in 12-well plates (2 x 105 cells/well) and transfected with 0.3 µg of each construct for 6 h. DNA:Effectene complexes were removed and replaced with fresh complete media, and the cells were incubated for an additional 20–42 h.

Adenoviral infection. Recombinant adenovirus Ad.PKG-FLAG containing the full-length human cDNA for PKGI with an NH2-terminal FLAG epitope (DYKDDDDK) was the generous gift of Dr. K. D. Bloch (Harvard Univ.). The viral stock solution was 3 x 1010 pfu/ml, and an adenovirus-mediated PKG gene transfer at multiplicity of infection (MOI) = 500 pfu/cell was used for all experiments.

Determination of Ser157 and Ser239 phosphorylation of wild-type and mutant VASP proteins. The expression and phosphorylation of the wild-type and mutant derivatives of VASP were detected in HPAECs coinfected with Ad.PKG-FLAG. Briefly, after removal of DNA:Effectene complexes as described above, Ad.PKG-FLAG (MOI = 500 pfu/cell) in complete media was added to the cells and incubated for an additional 20–24 h. To monitor the phosphorylation function or dysfunction of transgenic VASPs at Ser157 and Ser239 by PKGI, cells were preincubated with serum-free media for 30–60 min followed by treatment with 8-pCPT-cGMP (100 µM) for 30 min at 37°C. The experiment was terminated by washing the cells three times with ice-cold PBS before the cells were lysed and harvested for Western analysis.

Gel electrophoresis and immunoblot analysis. The HPAEC were lysed in a Laemmli sample buffer (Bio-Rad) containing an additional 4 mM Na3VO4, 40 mM NaF, 1% mercaptoethanol and 1:200 dilution of Sigma protease inhibitor cocktail. Samples were then subjected to SDS-PAGE in 8% polyacrylamide gels, processed, and quantified as previously described (19).

Transfection of siRNA. Predesigned VASP-specific siRNA was ordered from Ambion in PAGE-purified, desalted, unprotected, and annealed double-stranded form. The following two pair of 21-bp duplexes were used: duplex 1 (siVASP1), sense 5'-GGAAAUAAGAUGCUGUAACtt-3' and antisense 5'-GUUACAGCAUCUUAUUUCCtc-3'; duplex 2 (siVASP2), sense 5'-GGAUGAAGUCGUCUUCUUCtt-3' and antisense 5'-GAAGAAGACGACUUCAUCCtt-3'. HPAEC cells were transfected with siRNA using Cell Signaling transfection reagent (Beverly, MA) according to the protocol provided by the manufacturer. Non-specific, non-silencing, and fluorescently conjugated siRNA from Cell Signaling was used as a control. Cells grown to 70–90% confluence in 12-well plates or ECIS plates were serum starved for 1 h followed by incubation with 100 nM target siRNA (or control siRNA or no siRNA) for 6 h in serum-free media. Complete media with serum was then added (1% serum final concentration) for 42 h before Western blot analysis was conducted. In parallel studies, to measure TER of VASP-depleted HPAEC, 24-h incubated transfection complexes were removed and replaced with Ad.PKG-FLAG (MOI = 500 pfu/cell) containing complete media and incubated for an additional 18–20 h.

Measurement of TER. The TER of HPAEC monolayers was measured with the ECIS technique as previously described (19). The effects of 8-pCPT-cGMP (100 µM), H2O2 (250 µM, or, where indicated, 100 µM), and the combination of 8-pCPT-cGMP pretreatment followed by H2O2 were examined in HPAEC transfected with wild-type or mutant VASP constructs or VASP siRNA. Coinfection with Ad.PKG-Flag was performed to introduce PKG1 expression as needed.

Cell migration assay. HPAECs were transiently transfected in 12-well plates with either control GFP plasmid, VASP-SST, or VASP-DDT as described above. After achieving confluence, a wound was introduced in each well with a pipette tip. Photomicrographs were then obtained of each wound at time 0 using the x10 objective, and at least two pairs of GFP-expressing cells found at the leading wound edge were selected randomly. The cell pairs were then followed for 11 h during wound healing as described (27). The change in distance measured in micrometers between paired cells determined (Image-Pro Plus 5.1) after 11 h was divided by 11 and then by 2 to obtain the average migration velocity in µm/h per transfected cell.

Cell adhesion assay. Recombinant human VE-cadherin-Fc (VEC-Fc) was purchased from R&D Systems (Minneapolis, MN). Human immunoglobulin (Ig)G Fc protein was purchased from Jackson ImmunoResearch Laboratories (West Grove, PA). Ninety-six-well tissue culture plates were coated with 1.5 µg of VEC-Fc or Fc protein/0.1 ml in PBS-Ca2+/Mg2+ overnight at 4°C followed by blocking with 1% heat-inactivated BSA in PBS (inactivated at 85°C for 12 min) for 1 h at room temperature. HPAEC pretreated with plasmid vectors (control vector, wild type, and mutant VASPs as described above) followed by Ad.PKG-FLAG infection (as described above) were seeded per 2 x 104 cells in 100 µl into VEC-Fc- or Fc-coated 96-well dishes and incubated at 37°C. After 1–2 h, the cells were washed with PBS to remove unattached cells, and cell attachment was quantified by counting all cells in three random photographs obtained with an Olympus IX51 fluorescence microscope using an x10 objective and a Cooke digital camera. The cells were then serum starved for 1 h and treated with H2O2 (250 µM) in the presence or absence of 8-pCPT-cGMP (100 µM) pretreatment (30 min). After exposure to the desired experimental treatment, the plate was rinsed twice with PBS-Ca2+/Mg2+ to remove nonadherent cells before repeat quantification of total and GFP-expressing adherent cells. For each condition, triplicate wells transfected with VASP plasmids were studied in each of three independent experiments. In two experiments, control plasmid wells were included to evaluate possible nonspecific effects of plasmid transfection.

Immunofluorescence microscopy. HPAECs were grown on gelatinized coverslips before being exposed to various conditions (plasmid DNA or siRNA transfections coinfected with Ad.PKG-FLAG) as described for individual experiments. The cells were then fixed in 4% paraformaldehyde for 10 min, washed three times with Tris-buffered saline solution containing 0.1% Tween 20 (TBS-T, Sigma), permeabilized with 0.2% Triton X-100 in PBS for 5 min, and blocked with 2% BSA in PBS for 30 min. Actin filaments were visualized by staining cells with Texas red-conjugated phalloidin for 1 h at room temperature. In most cases, cells were counterstained with a VE-cadherin antibody (BD Biosciences-Transduction Laboratories) at a 1:200 dilution and incubated for 1 h at room temperature. After three washes with TBS-T, the coverslips were mounted with a Slow Fade kit (Molecular Probes) and analyzed using an Olympus IX51 inverted fluorescent microscope (x60 oil objective) and a Cooke digital camera.

Statistics. The effects of H2O2 and other drug treatments on the time course of TER were analyzed by two-way split-plot ANOVA. The effects of 8-pCPT-cGMP and H2O2 on the maximal change in TER ({Delta}TERmax) following siRNA or mutant VASP transfections were analyzed by randomized two-way (drug treatment, transfection) ANOVA. The effect of mutant VASP transfection on cell motility was analyzed by randomized one-way ANOVA. When significant variance ratios were obtained, least significant differences were calculated to allow comparison of individual means. The effect of VASP siRNA on basal TER was determined by unpaired two-tailed t-test. Values presented in the text are means ± SE. Differences were considered significant when P ≤ 0.05.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Effect of H2O2-induced VASP phosphorylation in HPAEC on endothelial barrier function. We previously found that H2O2 alone caused Ser157 VASP phosphorylation that was detectable at 30 min in HPAEC in the presence or absence of PKGI expression, although PKGI expression appeared to increase the signal (19). Because PKC has been shown to directly phosphorylate VASP at Ser157 in smooth muscle cells (4), and H2O2 activates endothelial PKC (32), we measured the early time course of H2O2-induced VASP phosphorylation and the phosphorylation of the PKC phosphorylation target MARCKS as an index of PKC activity (Fig. 1A). Additional HPAEC monolayers were exposed to the same H2O2 concentration to demonstrate the typical time course of the H2O2-induced decrease in TER (Fig. 1B). H2O2 caused a rapid, self-limited threefold phosphorylation of VASP at Ser157 and MARCKS that peaked after 5 min of H2O2 exposure (P < 0.01). This phosphorylation peak corresponded to a rapid decrease in TER that reached a nadir at 30 min when pSer157-VASP had returned to basal levels. There was no basal expression of pSer239-VASP, and H2O2 had no effect on Ser239 phosphorylation (data not shown).


Figure 1
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 1. A: time course of pSer157-vasodilator-stimulated phosphoprotein (VASP), unphosphorylated VASP, pMARCKS, and unphosphorylated MARCKS expression by Western blot and mean densitometry (n = 4) following exposure to H2O2 (250 µM) as indicated by the black bar that indicates the timing of H2O2 administration for both parts of this figure. B: time course of normalized transendothelial electrical resistance (TER) in human pulmonary artery endothelial cell (HPAEC) monolayers (n = 4–6) following exposure to H2O2 (250 µM). Values are means ± SE. *P < 0.01 vs. basal expression (A) or diluent treatment (B).

 
To determine if the formation of pSer157-VASP correlated in any way with the change in endothelial barrier function, we first examined the effect of PKC inhibition on both pSer157-VASP and TER. Figure 2A demonstrates that PMA, a direct PKC activator, potently phosphorylated VASP Ser157 in HPAEC. Additional experiments with the PKC inhibitor Ro-31-8220 (Fig. 2B) showed that Ro-31-8220 attenuated H2O2-mediated phosphorylation of MARCKS and VASP Ser157. Specifically, Ro-31-8220 decreased H2O2-induced VASP Ser157 phosphorylation by 21 ± 9% (P < 0.05, n = 3). As shown in Fig. 2C, Ro-31-8220 also significantly blocked the H2O2-induced decrease in TER (P < 0.0001). Of note, Ro-31-8220 transiently decreased basal TER before the administration of H2O2.


Figure 2
View larger version (26K):
[in this window]
[in a new window]

  		Fig. 2. A: effect of 30- and 60-min exposure of HPAEC to phorbol myristate acetate (100 nM) on pSer157-VASP phosphorylation by Western blot. B: effect of the PKC inhibitor Ro-31-8220 (100 nM) on H2O2-induced (250 µM) pMARCKS and pSer157-VASP generation at 10 min. C: effect of Ro-31-8220 (100 nM) or Ro diluent on the time course of normalized TER in HPAEC monolayers (n = 6–9) before and after exposure to H2O2 (250 µM). Values are means ± SE. *P < 0.0001 vs. H2O2 alone by 2-way (treatment, time) ANOVA interaction.

 
Because it remained unclear whether the early phosphorylation of VASP by H2O2 contributed to or attenuated the simultaneous change in permeability, we next attempted to dissociate pSer157-VASP formation and the effect of H2O2 on TER. In pilot experiments, we found that cyclosporin, a phosphatase 2B (PP2B) inhibitor (39), both increased and prolonged H2O2-induced Ser157 VASP phosphorylation. If VASP phosphorylation serves to oppose H2O2-induced endothelial barrier dysfunction, we reasoned that cyclosporin should attenuate the decrease in TER secondary to H2O2 by maintaining more VASP in the phosphorylated state. As shown in Fig. 3A, cyclosporin treatment resulted in a marked increase in both pre- and post-H2O2 pSer157-VASP formation. Despite the enhancement of VASP phosphorylation, cyclosporin had no effect on TER before H2O2 and did not attenuate the decreased TER caused by H2O2. Rather, cyclosporin significantly prolonged the barrier-disruptive effects of H2O2 (P < 0.03).


Figure 3
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 3. A: effect of H2O2 (250 µM) on pSer157-VASP formation in HPAEC at 10 min in the absence and presence of cyclosporin (200 nM) pretreatment by Western blot. B: effect of cyclosporin (200 nM) or diluent on the time course of normalized TER in HPAEC monolayers (n = 2–3) before and after exposure to H2O2 (250 µM). Values are means ± SE. *P < 0.03 vs. H2O2 alone by 2-way (treatment, time) ANOVA interaction.

 
To complement these experiments, we examined the effect of increasing Ser157 VASP phosphorylation by an entirely different mechanism using the adenylyl cyclase activator forskolin, which causes cAMP-stimulated PKA to preferentially phosphorylate VASP at this site (3). Similar to the cyclosporin effect, forskolin increased pSer157-VASP before H2O2 and maintained increased phosphorylation well beyond the normal recovery time following H2O2 (Fig. 4A). Also similar to the cyclosporin results, the presence of enhanced VASP phosphorylation on Ser157 did not attenuate the {Delta}TER from H2O2 (Fig. 4B) despite the expected increased basal TER caused by forskolin before H2O2 was added (14, 19). In fact, the {Delta}TER from H2O2 in the forskolin-treated monolayers was greater than that from H2O2 alone by a significant ANOVA (treatment, time) interaction (P < 0.0001).


Figure 4
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 4. A: effect of forskolin (50 µM) pretreatment on pSer157-VASP phosphorylation time course by Western blot in HPAEC after exposure to H2O2 (100 µM). B: time course of normalized TER in HPAEC monolayers (n = 3–5) pretreated with forskolin (50 µM) or forskolin diluent before exposure to H2O2 (100 µM). Values are means ± SE. *P < 0.0001 vs. H2O2 without forskolin pretreatment by 2-way (treatment, time) ANOVA interaction performed after normalizing TER to value just before H2O2 administration.

 
Effect of VASP siRNA on H2O2-induced barrier dysfunction in HPAEC. We next investigated the effect of reducing total VASP expression on the time course of TER following H2O2. The effect of two 21-bp VASP siRNA duplexes (siVASP1 and siVASP2) on VASP expression after 24 h is shown in Fig. 5A. The individual or combination of siVASP1 and siVASP2 resulted in >80% decrease in VASP protein expression in HPAEC monolayers. There was no effect of either siRNA on the expression of either VE-cadherin or actin, key adherens junction and cytoskeletal proteins. Fluorescently conjugated control siRNA used to track transfection efficiency showed that 90–95% of the cells were transfected.


Figure 5
View larger version (27K):
[in this window]
[in a new window]

  		Fig. 5. A: dose-response Western blot analysis of depletion of VASP after treatment with 2 different VASP small interfering RNA (siRNA) separately and the combination of both. VASP expression is normalized to VE-cadherin as a loading control and expressed as fold change relative to the addition of GeneSilencer transfection reagent without siRNA (0 nM siRNA). B: effect of VASP depletion with siVASP1 (100 nM) on the time course of TER following exposure of HPAEC monolayers with 250 µM H2O2 (n = 3–8). Values are means ± SE. *P < 0.0001 vs. siControl + H2O2 by 2-way (treatment, time) ANOVA interaction.

 
Treatment with siVASP1 did not significantly affect basal TER (control siRNA = 1,194 ± 100 {Omega}; siVASP1 = 1,003 ± 56 {Omega}; P = 0.10; n = 10–11, data not shown) but did significantly exacerbate H2O2-induced barrier dysfunction as evidenced by a larger, more prolonged decrease in TER (P < 0.0001) compared with monolayers treated with control siRNA and H2O2 (Fig. 5B).

Effect of VASP siRNA on cGMP/PKGI barrier protection in HPAEC. To determine the role of VASP phosphorylation in the barrier-protective effects mediated by cGMP and PKGI, we first introduced PKGI expression with Ad.PKG-FLAG and examined the effect on H2O2-induced barrier dysfunction. HPAEC were infected with increasing concentrations of Ad.PKG-FLAG (MOI = 200, 500, 800, 1,000, and 2,000 pfu/cell), and the expression of the recombinant PKG protein was measured by immunoblotting (Fig. 6A). The recombinant protein was detectable as early as 20 h and peaked at 48–72 h; however, cytotoxic effects were observed with longer incubations. Next, to identify the MOI required to observe maximal PKGI-induced barrier protection, HPAEC were infected with Ad.PKG-FLAG at increasing MOI for 20 h before exposure to the cGMP analog 8-pCPT-cGMP (100 µM) or diluent (media) followed by H2O2 (250 µM) while monitoring TER. As shown in Fig. 6B, a nearly 50% inhibition of the H2O2-induced TER decrease was detected in monolayers infected with an MOI of 500 pfu/cell. Using this infection protocol, PKGI construct protein expression was observed by FLAG immunoreactivity in a cytoplasmic distribution with an infection efficiency of >95% (Fig. 6C). Subsequent experiments utilized this protocol to examine the role of VASP phosphorylation in the PKGI-dependent barrier protection.


Figure 6
View larger version (34K):
[in this window]
[in a new window]

  		Fig. 6. A: dose-response effect of Ad.PKG-FLAG infection of HPAECs on expression of PKGI-FLAG by Western blot. B: dose-response effect of PKG-FLAG infection of HPAECs on the 8-pCPT-cGMP-induced inhibition of the change in TER ({Delta}TER) secondary to H2O2 exposure (n = 2). Values are means ± SE. C: immunofluorescent staining of HPAEC infected with Ad.PKG-FLAG utilizing anti-FLAG antibody.

 
To quantify the effects of siVASP treatment on PKGI-mediated barrier protection, we examined the ability of activated PKGI to prevent H2O2-mediated barrier dysfunction in monolayers in which VASP had been depleted with siVASP1+2 treatment. In Ad.PKG-FLAG-infected monolayers, the combination of siVASP1 and siVASP2 significantly decreased baseline levels of TER (P < 0.01), indicating an increase in permeability secondary to loss of VASP expression (Fig. 7, inset). There was no effect, however, on the PKGI-mediated inhibition of the increase in permeability caused by H2O2 as evidenced by a similar inhibition of the {Delta}TER by 8-pCPT-cGMP when comparing control and VASP siRNA-treated monolayers (Fig. 7). If anything, PKGI-mediated protection appeared magnified in VASP-depleted monolayers, although the ANOVA interaction term did not reach statistical significance (P = 0.24).


Figure 7
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 7. Mean maximal change in TER ({Delta}TERmax) in HPAEC infected with Ad.PKG-FLAG and treated with either control siRNA (open bars) or VASP siRNA (black bars) following H2O2 (250 µM) or H2O2 diluent with or without pretreatment (30-min incubation) with 100 µM 8-pCPT-cGMP (n = 5–12). Inset shows basal TER before administration of 8-pCPT-cGMP or H2O2 but after infection with Ad.PKG and transfection with siRNA (n = 15–16). Values are means ± SE. #P < 0.01; *P < 0.02 vs. H2O2 by 2-way (siRNA, cGMP) ANOVA treatment effect.

 
We next determined if the continued ability of activated PKGI to attenuate oxidant barrier dysfunction in VASP-depleted monolayers included an attenuation of the actin cytoskeletal rearrangement and loss of junctional VE-cadherin staining that occurs in the presence of H2O2 (19). Figure 8 shows representative monolayers cotransfected with Ad.PKG-FLAG and control or VASP siRNAs (Fig. 8, A–I) staining for actin (red) and VE-cadherin (blue). The effect of H2O2 and 8-pCPT-cGMP on uninfected HPAEC (lacking PKGI) is shown for comparison (Fig. 8, J–L). We previously showed that uninfected HPAECs responded in an identical fashion to HPAECs infected with a control adenovirus (19). In Fig. 8, A–C, Ad.PKG-FLAG-infected cells were cotransfected with control siRNA tagged with GFP. H2O2 exposure caused a subtle increase in actin stress fiber formation and cellular elongation with minimal intercellular gaps in VASP-deficient (Fig. 8, E and H) compared with VASP-sufficient monolayers (Fig. 8B). The uninfected control HPAEC lacking PKGI expression demonstrated a marked increase in intercellular gaps with loss of VE-cadherin staining after application of H2O2 (Fig. 8K) that were minimally prevented by 8-pCPT-cGMP pretreatment (Fig. 8L). The PKG-FLAG construct has constitutive PKGI activity (19) explaining the relative preservation of cellular architecture after H2O2 treatment alone compared with uninfected cells. Nevertheless, VASP-independent protective effects of PKGI activation by 8-pCPT-cGMP on the H2O2 effects can be observed in Fig. 8, C, F, and I, by the absence of actin stress fibers (magnified inset), the rounded cellular morphology, and the abundant intercellular VE-cadherin staining. Thus, severe VASP depletion from treatment with siVASP1 (Fig. 8, D–F) or the combination of siVASP1 and siVASP2 (Fig. 8, G–I) did not appear to alter the protective effects of PKGI (either constitutive or cGMP-stimulated) on the cytoskeletal rearrangement caused by H2O2.


Figure 8
View larger version (100K):
[in this window]
[in a new window]

  		Fig. 8. Composite immunofluorescent images of HPAEC monolayers infected with Ad.PKG-FLAG followed by transfection with siControl RNA (A–C), siVASP1 (D–F), or siVASP1&2 (G–I) and treated with diluent (A, D, and G), 250 µM H2O2 (B, E, and H), or 100 µM 8-pCPT-cGMP + 250 µM H2O2 (C, F, and I). Uninfected control HPAEC subjected to the same treatments are shown in the last column for comparison (J–L). Successful control siRNA transfection is shown by GFP fluorescence (A–C). The F-actin cytoskeleton is stained with Texas red-conjugated phalloidin. VE-cadherin staining is in blue. Arrows point to increased actin stress fiber formation. Inset solid-line rectangles show magnified views of dashed rectangular areas to highlight differences in actin stress fiber formation and VE-cadherin staining as a function of 8-pCPT-cGMP treatment and PKGI and VASP expression in H2O2-treated monolayers. Representative images are shown from 3 separate experiments.

 
Effect of expression of Ser non-phosphorylatable or phosphorylation-mimicking VASP mutants on cGMP/PKGI-mediated barrier protection and cell migration velocity in HPAEC. Because VASP Ser phosphorylation was specifically implicated in the barrier-protective effects of cGMP/PKGI (5), we generated VASP-GFP fusion plasmids that express wild-type VASP (VASP-SST), Ser phosphorylation-resistant mutant VASP (VASP-AAT), and Ser-double phosphorylation-mimetic VASP (VASP-DDT) in the expression vector pEGFP-C1 (Fig. 9A). As expected, 8-pCPT-cGMP treatment of HPAEC, which were previously infected with Ad.PKG-FLAG, resulted in the characteristic phosphorylation of endogenous VASP regardless of the absence or presence of mutant VASP transfection. Specifically, Ser157 phosphorylation resulted in a molecular mass shift of endogenous VASP from 46 kDa to 50 kDa (Fig. 9B, top blot, lanes 2, 4, 6, and 8), whereas the monoclonal anti-pSer239-VASP antibody detected a new band at 46 kDa (Fig. 9B, bottom blot, lanes 2, 4, 6, and 8). The second fainter band detected by this antibody represents VASP that is simultaneously phosphorylated by PKGI at both Ser sites. The ectopically expressed VASP constructs were detected at 75 kDa because of the presence of GFP. As expected, only the ectopically expressed VASP-SST was phosphorylated at Ser157 and Ser239 in response to 8-pCPT-cGMP in cells previously infected with Ad.PKG-FLAG (Fig. 9B, lane 4). The transfection efficiency of these plasmids was 20–25%.


Figure 9
View larger version (26K):
[in this window]
[in a new window]

  		Fig. 9. A: schema of GFP reporter constructs containing wild-type (WT) and mutant VASP genes. VASP-SST consists of WT VASP with intact serine (157 and 239) and threonine phosphorylation sites. The VASP-AAT mutant has alanine substituted for both serines rendering them resistant to phosphorylation. The VASP-DDT construct has aspartic acids replacing the serine sites to mimic the effects of serine phosphorylation. B: Western blots of VASP constructs ectopically expressed in HPAEC without (odd lanes) and with (even lanes) prior infection with Ad.PKG-FLAG and PKGI activation with 8-pCPT-cGMP.

 
We next transfected HPAEC with VASP-SST or VASP-DDT and compared the effect of H2O2 on TER to monolayers transfected with control plasmid. As shown in Fig. 10A, H2O2 treatment resulted in a decrease in TER that was not attenuated by the expression of VASP-DDT. To further confirm this negative effect, we examined the effect of H2O2 on the actin cytoskeleton and the integrity of the intercellular junctions between VASP-DDT-transfected cells, identified by the presence of GFP fluorescence. Diluent-treated cells expressing VASP-DDT (Fig. 10, B–D) had intact intracellular junctions demonstrating continuous VE-cadherin staining. The VASP-DDT was expressed in the intercellular junctions (arrowhead) and along the actin cytoskeleton in addition to the punctate staining of focal adhesions (Fig. 10C). Treatment with H2O2 (Fig. 10, E–G) resulted in the typical increase in actin stress fibers and intercellular gaps (arrows) that occurs in HPAEC in the absence of PKGI expression. Examination of the intercellular junctions between two VASP-DDT-expressing cells shows gaps and loss of VE-cadherin staining (Fig. 10, F and G) despite the presence of Ser phosphorylation-mimicking VASP expression in the junctions and actin cytoskeleton.


Figure 10
View larger version (44K):
[in this window]
[in a new window]

  		Fig. 10. A: mean {Delta}TERmax (n = 4–6) in HPAEC transfected with control plasmid, VASP-SST, or VASP-DDT, and treated with either diluent (white bars) or H2O2 (black bars). Values are means ± SE. B–G: immunofluorescent images of HPAEC monolayers transfected with VASP-DDT before and after treatment with H2O2 (250 µM). The F-actin cytoskeleton is stained with Texas red-conjugated phalloidin, VASP-DDT expresses GFP, and VE-cadherin is in blue. Arrowhead points to VASP expression in intercellular junction. Arrows indicate H2O2-induced intercellular gaps. Magnified image shows H2O2-induced disruption of cell junction VE-cadherin staining between 2 VASP-DDT-expressing cells. Representative images are shown from 3 separate experiments.

 
Because the HPAEC transfected with mutant VASP retained endogenous VASP expression, we exploited the known effects of VASP phosphorylation on cell motility to determine if the expression of mutant VASP was capable of overcoming the effect of endogenous VASP. As shown in Fig. 11, HPAECs expressing VASP-DDT exhibited significantly less (P < 0.05) cell velocity during in vitro wound healing compared with either control GFP or VASP-SST plasmid-transfected HPAECs.


Figure 11
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 11. Mean velocity of HPAECs transfected with control plasmid, VASP-SST, or VASP DDT during an 11-h wound healing assay (n = 2–8 pairs of cells in each of 4 independent experiments). *P < 0.05.

 
Effect of expression of Ser non-phosphorylatable VASP mutant on cGMP/PKGI-mediated protection of cell-cell adhesion in HPAEC. Because of the prominent role played by VE-cadherin homotypic interactions in adherens junctions, we examined the ability of activated PKGI to attenuate the H2O2-induced loss of VE-cadherin-dependent cell adhesion in the presence of VASP-AAT expression. HPAEC infected with Ad.PKG-FLAG and cotransfected with control plasmid, VASP-SST or VASP-AAT were allowed to bind to VE-cadherin-Fc chimeric fusion protein-coated wells. No cells attached to control Fc-coated dishes (data not shown), indicating that the adhesion of HPAEC to the fusion protein was specific for VE-cadherin. As shown in Fig. 12, nearly 80% of total adherent HPAECs was displaced when wells were treated with H2O2 alone, whereas wells that received 8-pCPT-cGMP pretreatment lost only 40% of total cells, demonstrating cGMP/PKGI-mediated protection (P < 0.0001). This protective effect was not significantly altered by transfection of either VASP-SST or VASP-AAT. Importantly, the same result was observed when the analysis was limited to the subset of plasmid-transfected cells that were identified by their green fluorescence.


Figure 12
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 12. Mean percent change in Ad.PKG-FLAG-infected HPAECs adherent to VE-cadherin fusion protein following exposure to H2O2 (250 µM) without and with pretreatment with 8-pCPT-cGMP (100 µM) to activate PKGI. All HPAEC were transfected with GFP-expressing control plasmid, VASP-SST, or VASP-AAT before treatment with H2O2 or 8-pCPT-cGMP. Data are shown for both non-GFP cells and the subpopulation of GFP-expressing plasmid-transfected cells (n = 3 independent experiments for VASP constructs and 2 experiments for control plasmid). Values are means ± SE. *P < 0.0001 vs. HPAEC treated with H2O2 alone.

 
Figure 13 shows the results of VASP-AAT transfection on HPAEC monolayers infected with Ad.PKG-FLAG to determine if the Ser phosphorylation-resistant VASP mutant would attenuate the protective effects of PKGI on H2O2-induced junctional disruption and cytoskeletal rearrangement. Compared with 8-pCPT-cGMP treatment alone (Fig. 13, A–C), H2O2 caused an increase in actin stress fibers and intercellular gaps, but the VE-cadherin staining between the two VASP-AAT-expressing cells appeared to be least affected by the H2O2 (Fig. 13, E and F). 8-pCPT-cGMP pretreatment attenuated the adverse effects of H2O2 as shown by a decrease in actin cytoskeletal rearrangement and intercellular gap formation. Importantly, the intercellular junction and VE-cadherin staining between the two VASP-AAT-expressing cells was completely without gaps despite the intercellular junction expression of VASP-AAT (Fig. 13, H and I).


Figure 13
View larger version (87K):
[in this window]
[in a new window]

  		Fig. 13. Immunofluorescent images of HPAEC monolayers infected with Ad.PKG-FLAG followed by transfection with VASP-AAT before treatment with 8-pCPT-cGMP (100 µM) (A–C), H2O2 (250 µM) (D–F), or H2O2 following pretreatment with 8-pCPT-cGMP (G–I). The F-actin cytoskeleton is stained with Texas red-conjugated phalloidin, VASP-AAT expresses GFP, and VE-cadherin is in blue. Arrowheads point to plasmid-derived VASP expression in intercellular junctions. Arrows indicate H2O2-induced intercellular gaps. Magnified image from dashed rectangular area shows preservation of cell junction VE-cadherin staining between 2 VASP-AAT-expressing cells following treatment with 8-pCPT-cGMP and H2O2. Representative images are shown from 3 separate experiments.

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
VASP has long been hypothesized to be involved in PKGI-mediated endothelial barrier regulation (7) because VASP is a phosphorylation target of PKGI known to be present in close association with both adherens (30, 38) and tight junction proteins (5). VASP promoted actin filament nucleation, bundling, and elongation, suggesting a role in cytoskeletal rearrangement that could alter cell shape and adhesion (10). Finally, it was shown that phosphorylation of VASP Ser239 by PKA inhibited VASP-actin binding (8, 40), prompting the notion that VASP phosphorylation could represent the mechanism behind the potent endothelial barrier protection mediated by activation of PKGI or PKA. In support of this theory, Comerford et al. (5) demonstrated coassociation of VASP with the tight junction proteins zonula occludins-1, occludin, and junctional adhesion molecule-1 in systemic endothelial cells. PKA activation resulted in simultaneous Ser157 phosphorylation of junctional VASP and decreased paracellular permeability. Moreover, expression of truncated VASP constructs missing the Ser phosphorylation and actin bundling domains decreased basal permeability suggesting that non-phosphorylated VASP served to maintain an open paracellular path (5). Alternatively, Schlegel et al. (29) found that the decrease in paracellular permeability caused by cAMP was similar in immortalized systemic endothelial monolayers derived from VASP knockout mice compared with wild-type cells suggesting that VASP phosphorylation was not involved.

Much of the published data attempting to link changes in VASP phosphorylation to endothelial permeability in primary endothelial cells are correlative rather than causative (5, 7). When we altered VASP Ser phosphorylation, decreased VASP expression with siRNA, or introduced VASP Ser phosphorylation mutant proteins, we found little evidence to support the hypothesis that VASP phosphorylation is barrier protective in HPAEC exposed to H2O2. We first examined whether the transient barrier dysfunction caused by H2O2 was associated with changes in VASP phosphorylation in HPAECs lacking PKGI expression. The absence of PKGI and pSer239-VASP expression in HPAECs in these experiments (19) allowed us to specifically isolate the potential role of Ser157 VASP phosphorylation. Consistent with previous results (5, 7, 19), we found little basal pSer157-VASP in HPAEC. H2O2 treatment resulted in an increase in permeability that correlated temporally with evidence of PKC activation and pSer157-VASP formation. PKC inhibition attenuated both the increase in permeability and the VASP Ser157 phosphorylation consistent with the known contribution of PKC to H2O2-induced pulmonary endothelial barrier dysfunction (32) and more recent published data in smooth muscle cells demonstrating that PKC can directly phosphorylate VASP at this site (4).

Based on the observations of Comerford et al. (5), the PKC-mediated pSer157-VASP formation could be functioning as a brake on the barrier-disruptive effects of PKC activation, although our time course data in Fig. 1 demonstrate that VASP phosphorylation has resolved before the recovery of barrier function following H2O2. To examine this possibility, we increased and prolonged pSer157-VASP by two pharmacologically independent methods but found no consistent association between VASP Ser157 phosphorylation and barrier protection. First, we utilized the phosphatase 2B inhibitor, cyclosporin, to increase VASP phosphorylation because the Ca2+/calmodulin-dependent phosphatase 2B was implicated as one of the phosphatases responsible for Ser dephosphorylation of VASP (1). Consistent with this, we found that cyclosporin markedly increased pSer157-VASP before and after H2O2 administration (Fig. 3A), but cyclosporin alone did not affect TER, and the H2O2-induced decrease in TER was identical in the absence and presence of cyclosporin treatment (Fig. 3B), suggesting that Ser157 VASP phosphorylation did not attenuate H2O2-induced barrier dysfunction. Cyclosporine did delay the recovery of TER following H2O2 exposure consistent with the previous observation that phosphatase 2B inhibition delayed TER recovery following thrombin secondary to increased PKC activation (17).

We also examined the effect of increasing cAMP with forskolin to generate pSer157-VASP by PKA activation similar to Comerford et al. (5). As expected (5, 19), forskolin markedly increased pSer157-VASP and TER, but the subsequent {Delta}TER from H2O2 was again not inhibited, indicating that VASP Ser157 phosphorylation did not prevent the increased permeability from H2O2 (Fig. 4). In fact, the relative {Delta}TER from the newly elevated baseline was actually greater compared with the {Delta}TER following H2O2 without forskolin pretreatment.

The inability of cAMP/PKA to attenuate the increased endothelial permeability from H2O2 differed from the results of Suttorp et al. (37) where forskolin was reported to prevent H2O2-induced hyperpermeability in pulmonary endothelium. This difference may be secondary to their use of porcine pulmonary artery endothelial cells or an H2O2 concentration (1 mM) that was shown to significantly decrease intracellular cAMP levels (37). Although we have not made cAMP measurements in HPAEC, we (unpublished observations) found no change in intracellular cAMP concentration in bovine PAEC monolayers following 30 min of exposure to 100 µM H2O2 (29.5 ± 10.1 vs. 21.1 ± 5.8 pmol/mg; n = 4; P = 0.63).

To complement the pharmacological manipulations of VASP phosphorylation, we also decreased VASP expression with two VASP-specific siRNAs. Unphosphorylated VASP was suggested to maintain an open paracellular pathway (5), but we found that decreased unphosphorylated VASP expression in HPAEC monolayers had no significant effect on basal TER but significantly exacerbated the decrease in TER from H2O2 (Fig. 5B). This result suggested that the actin polymerizing and bundling function of VASP was not necessary to mediate the endothelial barrier dysfunction from H2O2. The finding of an augmented decrease in TER after H2O2 observed in VASP-depleted monolayers was similar to the effects of increasing Ser157 VASP phosphorylation suggesting that unphosphorylated VASP may have a barrier-preserving function that counters the barrier-disrupting effects of H2O2. The effect of VASP on endothelial junction integrity may be dependent in part on the specific mechanism of endothelial activation. For example, we recently found that VASP depletion with siRNA increased the barrier-enhancing effects of ATP in HPAEC suggesting that VASP in some way interfered with the cell junction tightening process caused by ATP (9).

A major goal of this study was to determine if VASP phosphorylation was specifically responsible for the endothelial barrier-protecting effects of cGMP/PKGI activation we previously observed in pulmonary endothelial cells (19, 22) in association with Ser239 and Ser157 VASP phosphorylation. We therefore generated PKGI-expressing HPAEC monolayers in which VASP was depleted with siRNA or modified with expression of mutant VASP constructs that would either mimic or block VASP Ser phosphorylation. Again, we found no evidence to support the theory that VASP phosphorylation of either Ser157 or Ser239 was necessary for cGMP-mediated barrier protection. Thus, marked inhibition of VASP expression had no effect on the ability of cGMP/PKGI to attenuate the increased permeability (Fig. 7), the decreased VE-cadherin junction expression, or the actin cytoskeletal rearrangement (Fig. 8) caused by H2O2.

Interestingly, VASP siRNA did significantly increase basal permeability in PKGI-expressing monolayers, whereas there was no significant effect in the absence of PKGI (see RESULTS). There was no basal Ser239 VASP phosphorylation in any of these monolayers (data not shown), suggesting a phosphorylation-independent PKGI-VASP interaction that served to enhance basal endothelial barrier function. Similar increases in basal permeability in systemic microvascular endothelial cells were reported in monolayers treated with VASP siRNA (28) or derived from VASP knockout mice (29), but PKGI expression was not measured.

Because inhibition of VASP expression with siRNA decreases both phosphorylated and unphosphorylated VASP, we also analyzed the effect of expressing VASP mutant constructs designed to either mimic or prevent Ser phosphorylation (Fig. 9). Expression of a phospho-mimicking VASP at both Ser sites failed to mimic the effects of cGMP/PKGI activation on the {Delta}TER (Fig. 10A). It is important to note that the transfection efficiency of the mutant VASP was <30%, so it is possible that an effect on TER could have been missed, although we would have expected at least a trend toward a protective effect. We examined the effect of H2O2 on the appearance of cell junction gaps and VE-cadherin expression between VASP-DDT-transfected cells and again saw no evidence of protection (Fig. 10, B–G).

An additional caveat in the interpretation of the VASP-DDT experiment was the fact that the transfected cells retained endogenous VASP expression (Fig. 9B). Thus, it was possible that the negative result occurred because the effects of mutant VASP were masked by the presence of endogenous VASP. On the basis of the results of the wound healing assay shown in Fig. 11, this did not appear to be the case. The VASP-DDT-transfected cells displayed significantly lower migration velocity compared with either control or VASP-SST-transfected cells, suggesting that the mutant VASP expression was sufficient to alter cellular function. This result is consistent with recent data showing a reduction in epithelial cell migration associated with VASP Ser239 phosphorylation (13).

Negative results were also obtained in experiments in which we attempted to block the protective effects of cGMP/PKGI on VE-cadherin-dependent cell adhesion (Fig. 12) and cell junction integrity (Fig. 13) by introducing a Ser phosphorylation-resistant VASP mutant. We utilized a VE-cadherin-Fc chimeric fusion protein in this experiment to allow a direct measurement of VE-cadherin homotypic binding that, unlike TER, could be limited to the subpopulation of mutant VASP-transfected cells. We found that the cells expressing the mutant VASP demonstrated cGMP/PKGI-mediated adherence protection in the presence of H2O2 that was identical to the whole population of cells, suggesting that VASP Ser phosphorylation was not involved.

In conclusion, VASP expression in HPAEC contributes to endothelial barrier function because decreased VASP expression decreased TER and exacerbated the loss of barrier function caused by H2O2. Moreover, VASP phosphorylation is temporally associated with both ROS-mediated endothelial barrier dysfunction and cyclic nucleotide-mediated signaling. However, we found no evidence to support the theory that the phosphorylation status of VASP at either Ser157 or Ser239 was responsible for the barrier-protective effects of cGMP/PKGI.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by National Heart, Lung, and Blood Institute Grants HL-67189 and HL-075236 and in part by Grant SFB 688 from the Deutsche Forschungsgemeinschaft.


   ACKNOWLEDGMENTS
 
We thank Dr. Kenneth D. Bloch for providing the PKGI adenovirus.


   FOOTNOTES
 

Address for reprint requests and other correspondence: D. B. Pearse, Johns Hopkins Bayview Medical Center, 5501 Hopkins Bayview Circle, Baltimore, MD 21224 (e-mail: dpearse{at}jhmi.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Abel K, Mieskes G, Walter U. Dephosphorylation of the focal adhesion protein VASP in vitro and in intact human platelets. FEBS Lett 370: 184–188, 1995.[CrossRef][Web of Science][Medline]
  2. Blume C, Benz PM, Walter U, Ha J, Kemp BE, Renne T. AMP-activated protein kinase impairs endothelial actin cytoskeleton assembly by phosphorylating vasodilator-stimulated phosphoprotein. J Biol Chem 282: 4601–4612, 2007.[Abstract/Free Full Text]
  3. Butt E, Abel K, Krieger M, Palm D, Hoppe V, Hoppe J, Walter U. cAMP- and cGMP-dependent protein kinase phosphorylation sites of the focal adhesion vasodilator-stimulated phosphoprotein (VASP) in vitro and in intact human platelets. J Biol Chem 269: 14509–14517, 1994.[Abstract/Free Full Text]
  4. Chitaley K, Chen L, Galler A, Walter U, Daum G, Clowes AW. Vasodilator-stimulated phosphoprotein is a substrate for protein kinase C. FEBS Lett 556: 211–215, 2004.[CrossRef][Web of Science][Medline]
  5. Comerford KM, Lawrence DW, Synnestvedt K, Levi BP, Colgan SP. Role of vasodilator-stimulated phosphoprotein in PKA-induced changes in endothelial junctional permeability. FASEB J 16: 583–585, 2002.[Free Full Text]
  6. Diwan AH, Thompson WJ, Lee AK, Strada SJ. Cyclic GMP-dependent protein kinase activity in rat pulmonary microvascular endothelial cells. Biochem Biophys Res Commun 202: 728–735, 1994.[CrossRef][Web of Science][Medline]
  7. Draijer R, Vaandrager AB, Nolte C, de Jonge HR, Walter U, van Hinsbergh VW. Expression of cGMP-dependent protein kinase I and phosphorylation of its substrate, vasodilator-stimulated phosphoprotein, in human endothelial cells of different origin. Circ Res 77: 897–905, 1995.[Abstract/Free Full Text]
  8. Harbeck B, Huttelmaier S, Schluter K, Jockusch BM, Illenberger S. Phosphorylation of the vasodilator-stimulated phosphoprotein regulates its interaction with actin. J Biol Chem 275: 30817–30825, 2000.[Abstract/Free Full Text]
  9. Kolosova IA, Mirzapoiazova T, Adyshev D, Usatyuk P, Romer LH, Jacobson JR, Natarajan V, Pearse DB, Garcia JG, Verin AD. Signaling pathways involved in adenosine triphosphate-induced endothelial cell barrier enhancement. Circ Res 97: 115–124, 2005.[Abstract/Free Full Text]
  10. Krause M, Dent EW, Bear JE, Loureiro JJ, Gertler FB. ENA/VASP proteins: regulators of the actin cytoskeleton and cell migration. Annu Rev Cell Dev Biol 19: 541–564, 2003.[CrossRef][Web of Science][Medline]
  11. Kwan HY, Huang Y, Yao X. Store-operated calcium entry in vascular endothelial cells is inhibited by cGMP via a protein kinase G-dependent mechanism. J Biol Chem 275: 6758–6763, 2000.[Abstract/Free Full Text]
  12. Kwan HY, Huang Y, Yao X. Regulation of canonical transient receptor potential isoform 3 (TRPC3) channel by protein kinase G. PNAS 101: 2625–2630, 2004.[Abstract/Free Full Text]
  13. Lindsay SL, Ramsey S, Aitchison M, Renne T, Evans TJ. Modulation of lamellipodial structure and dynamics by NO-dependent phosphorylation of VASP Ser239. J Cell Sci 120: 3011–3021, 2007.[Abstract/Free Full Text]
  14. Liu F, Verin AD, Borbiev T, Garcia JG. Role of cAMP-dependent protein kinase A activity in endothelial cell cytoskeleton rearrangement. Am J Physiol Lung Cell Mol Physiol 280: L1309–L1317, 2001.[Abstract/Free Full Text]
  15. Lofton CE, Baron DA, Heffner JE, Currie MG, Newman WH. Atrial natriuretic peptide inhibits oxidant-induced increases in endothelial permeability. J Mol Cell Cardiol 23: 919–927, 1991.[CrossRef][Web of Science][Medline]
  16. Lohmann SM, Walter U. Tracking functions of cGMP-dependent protein kinases (cGK). Front Biosci 10: 1313–1328, 2005.[Web of Science][Medline]
  17. Lum H, Podolski JL, Gurnack ME, Schulz IT, Huang F, Holian O. Protein phosphatase 2B inhibitor potentiates endothelial PKC activity and barrier dysfunction. Am J Physiol Lung Cell Mol Physiol 281: L546–L555, 2001.[Abstract/Free Full Text]
  18. Mehta D, Malik AB. Signaling mechanisms regulating endothelial permeability. Physiol Rev 86: 279–367, 2006.[Abstract/Free Full Text]
  19. Moldobaeva A, Welsh-Servinsky LE, Shimoda LA, Stephens RS, Verin AD, Tuder RM, Pearse DB. Role of protein kinase G in barrier-protective effects of cGMP in human pulmonary artery endothelial cells. Am J Physiol Lung Cell Mol Physiol 290: L919–L930, 2006.[Abstract/Free Full Text]
  20. Munzel T, Feil R, Mulsch A, Lohmann SM, Hofmann F, Walter U. Physiology and pathophysiology of vascular signaling controlled by cyclic guanosine 3',5'-cyclic monophosphate-dependent protein kinase. Circulation 108: 2172–2183, 2003.[Free Full Text]
  21. Pearse DB, Becker PM. Effect of time and vascular pressure on permeability and cyclic nucleotides in ischemic lungs. Am J Physiol Heart Circ Physiol 279: H2077–H2084, 2000.[Abstract/Free Full Text]
  22. Pearse DB, Shimoda LA, Verin AD, Bogatcheva N, Moon C, Ronnett GV, Welsh LE, Becker PM. Effect of cGMP on lung microvascular endothelial barrier dysfunction following hydrogen peroxide. Endothelium 10: 309–317, 2003.[Web of Science][Medline]
  23. Pearse DB, Wagner EM, Permutt S. Effect of ventilation on vascular permeability and cyclic nucleotide concentrations in ischemic sheep lungs. J Appl Physiol 86: 123–132, 1999.[Abstract/Free Full Text]
  24. Pinsky DJ, Naka Y, Chowdhury NC, Liao H, Oz MC, Michler RE, Kubaszewski E, Malinski T, Stern DM. The nitric oxide/cyclic GMP pathway in organ transplantation: critical role in successful lung preservation. Proc Natl Acad Sci USA 91: 12086–12090, 1994.[Abstract/Free Full Text]
  25. Price CJ, Brindle NP. Vasodilator-stimulated phosphoprotein is involved in stress-fiber and membrane ruffle formation in endothelial cells. Arterioscler Thromb Vasc Biol 20: 2051–2056, 2000.[Abstract/Free Full Text]
  26. Reinhard M, Halbrugge M, Scheer U, Wiegand C, Jockusch BM, Walter U. The 46/50 kDa phosphoprotein VASP purified from human platelets is a novel protein associated with actin filaments and focal contacts. EMBO J 11: 2063–2070, 1992.[Web of Science][Medline]
  27. Rodriguez LG, Wu X, Guan JL. Wound-healing assay. Methods Mol Biol 294: 23–29, 2005.[Medline]
  28. Rosenberger P, Khoury J, Kong T, Weissmuller T, Robinson AM, Colgan SP. Identification of vasodilator-stimulated phosphoprotein (VASP) as an HIF-regulated tissue permeability factor during hypoxia. FASEB J 21: 2613–2621, 2007.[Abstract/Free Full Text]
  29. Schlegel N, Burger S, Golenhofen N, Walter U, Drenckhahn D, Waschke J. The role of VASP in the regulation of cAMP- and Rac 1-mediated endothelial barrier stabilization. Am J Physiol Cell Physiol 294: C178–C188, 2008.[Abstract/Free Full Text]
  30. Scott JA, Shewan AM, den Elzen NR, Loureiro JJ, Gertler FB, Yap AS. Ena/VASP proteins can regulate distinct modes of actin organization at cadherin-adhesive contacts. Mol Biol Cell 17: 1085–1095, 2006.[Abstract/Free Full Text]
  31. Sechi AS, Wehland J. ENA/VASP proteins: multifunctional regulators of actin cytoskeleton dynamics. Front Biosci 9: 1294–1310, 2004.[Web of Science][Medline]
  32. Siflinger-Birnboim A, Goligorsky MS, Del Vecchio PJ, Malik AB. Activation of protein kinase C pathway contributes to hydrogen peroxide-induced increase in endothelial permeability. Lab Invest 67: 24–30, 1992.[Web of Science][Medline]
  33. Smolenski A, Bachmann C, Reinhard K, Honig-Liedl P, Jarchau T, Hoschuetzky H, Walter U. Analysis and regulation of vasodilator-stimulated phosphoprotein serine 239 phosphorylation in vitro and in intact cells using a phosphospecific monoclonal antibody. J Biol Chem 273: 20029–20035, 1998.[Abstract/Free Full Text]
  34. Smolenski A, Poller W, Walter U, Lohmann SM. Regulation of human endothelial cell focal adhesion sites and migration by cGMP-dependent protein kinase I. J Biol Chem 275: 25723–25732, 2000.[Abstract/Free Full Text]
  35. Sporbert A, Mertsch K, Smolenski A, Haseloff RF, Schonfelder G, Paul M, Ruth P, Walter U, Blasig IE. Phosphorylation of vasodilator-stimulated phosphoprotein: a consequence of nitric oxide- and cGMP-mediated signal transduction in brain capillary endothelial cells and astrocytes. Brain Res Mol Brain Res 67: 258–266, 1999.[Medline]
  36. Suttorp N, Hippenstiel S, Fuhrmann M, Krull M, Podzuweit T. Role of nitric oxide and phosphodiesterase isoenzyme II for reduction of endothelial hyperpermeability. Am J Physiol Cell Physiol 270: C778–C785, 1996.[Abstract/Free Full Text]
  37. Suttorp N, Weber U, Welsch T, Schudt C. Role of phosphodiesterases in the regulation of endothelial permeability in vitro. J Clin Invest 91: 1421–1428, 1993.[Web of Science][Medline]
  38. Vasioukhin V, Bauer C, Yin M, Fuchs E. Directed actin polymerization is the driving force for epithelial cell-cell adhesion. Cell 100: 209–219, 2000.[CrossRef][Web of Science][Medline]
  39. Verin AD, Cooke C, Herenyiova M, Patterson CE, Garcia JG. Role of Ca2+/calmodulin-dependent phosphatase 2B in thrombin-induced endothelial cell contractile responses. Am J Physiol Lung Cell Mol Physiol 275: L788–L799, 1998.[Abstract/Free Full Text]
  40. Walders-Harbeck B, Khaitlina SY, Hinssen H, Jockusch BM, Illenberger S. The vasodilator-stimulated phosphoprotein promotes actin polymerisation through direct binding to monomeric actin. FEBS Lett 529: 275–280, 2002.[CrossRef][Web of Science][Medline]



This article has been cited by other articles:


Home page
 J. Cell Sci.Home page
P. M. Benz, C. Blume, S. Seifert, S. Wilhelm, J. Waschke, K. Schuh, F. Gertler, T. Munzel, and T. Renne
Differential VASP phosphorylation controls remodeling of the actin cytoskeleton
J. Cell Sci., November 1, 2009; 122(21): 3954 - 3965.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
294/4/L686    most recent
00417.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rentsendorj, O.
Right arrow Articles by Pearse, D. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rentsendorj, O.
Right arrow Articles by Pearse, D. B.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2008 by the American Physiological Society.