This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: 98/1/73    most recent 01.RES.0000198387.44395.e9v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Broman, M. T. Articles by Malik, A. B. Search for Related Content PubMed PubMed Citation Articles by Broman, M. T. Articles by Malik, A. B. Related Collections Pulmonary circulation and disease Endothelium/vascular type/nitric oxide

(Circulation Research. 2006;98:73.)
© 2006 American Heart Association, Inc.

Cellular Biology

Cdc42 Regulates Adherens Junction Stability and Endothelial Permeability by Inducing -Catenin Interaction With the Vascular Endothelial Cadherin Complex

Michael T. Broman, Panos Kouklis, Xiaopei Gao, Ramaswamy Ramchandran, Radu F. Neamu, Richard D. Minshall, Asrar B. Malik

From the Department of Pharmacology and Center for Lung and Vascular Biology, The University of Illinois College of Medicine, Chicago.

Correspondence to Asrar B. Malik, Department of Pharmacology, 808 S Wood St, M/C 868, Chicago, IL 60612. E-mail abmalik{at}uic.edu

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
The endothelial adherens junctions (AJs) consist of trans-oligomers of membrane spanning vascular endothelial (VE)-cadherin proteins, which bind ß-catenin through their cytoplasmic domain. ß-Catenin in turn binds -catenin and connects the AJ complex with the actin cytoskeleton. We addressed the in vivo effects of loss of VE-cadherin interactions on lung vascular endothelial permeability and the role of specific Rho GTPase effectors in regulating the increase in permeability induced by AJ destabilization. We used cationic liposomes encapsulating the mutant of VE-cadherin lacking the extracellular domain (EXD) to interfere with AJ assembly in mouse lung endothelial cells. We observed that lung vascular permeability (quantified as microvessel filtration coefficient [K_f,c]) was increased 5-fold in lungs expressing EXD. This did not occur to the same degree on expression of the VE-cadherin mutant, EXDß, lacking the ß-catenin–binding site. The increased vascular permeability was the result of destabilization of VE-cadherin homotypic interaction induced by a shift in the binding of ß-catenin from wild-type VE-cadherin to the expressed EXD mutant. Because EXD expression in endothelial cells activated the Rho GTPase Cdc42, we addressed its role in the mechanism of increased endothelial permeability induced by AJ destabilization. Coexpression of dominant-negative Cdc42 (N17Cdc42) prevented the increase in K_f,c induced by EXD. This was attributed to inhibition of the association of -catenin with the EXD–ß-catenin complex. The results demonstrate that Cdc42 regulates AJ permeability by controlling the binding of -catenin with ß-catenin and the consequent interaction of the VE-cadherin/catenin complex with the actin cytoskeleton.

Key Words: adhesion molecules • gene transfer • catenins • Cdc42 • VE-cadherin

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Adherens junctions (AJs) are dynamic structures mediating endothelial cell–cell adhesion¹ and thereby regulate endothelial barrier function and tissue fluid homeostasis.^2,3 Endothelial cell adhesion to contiguous cells, forming the endothelial monolayer, is determined in part by the transmembrane protein vascular endothelial (VE)-cadherin, which forms homotypic adhesive interactions^2,4 and binds 2 members of the catenin family of cytosolic proteins (ß and catenins).⁵ ß-Catenin binds -catenin and connects junctional cadherins with the actin cytoskeleton. Permeability-increasing mediators such as histamine and thrombin increase junctional permeability in part by inducing the disassembly of AJs.^6–8 The end result is a shift in fluid and plasma proteins to the extravascular space and development of protein-rich tissue edema.

Phosphorylation of AJ proteins contributes to the mechanism of destabilization of AJs.⁷ Studies showed that histamine induced the phosphorylation of VE-cadherin and ß- and -catenins within 60 sec consistent with the rapidly increased microvascular permeability secondary to the formation of 100 to 400 nm-wide interendothelial gaps.⁸ It was also shown that thrombin-induced activation of PKC regulated the phosphorylation of VE-cadherin and catenins and thereby contributed to disruption of AJs and the increase endothelial permeability.⁹ Phosphatases such as protein tyrosine phosphatase VE-PTP may also regulate the phosphorylation of VE-cadherin/catenin components¹⁰ and thus AJ integrity.

In addition, Rho GTPases RhoA, Rac1, and Cdc42 are important in regulating AJ assembly.^11,12 Inhibition of RhoA prevented both thrombin- and histamine-induced disassembly of AJs¹³ and the increase in endothelial permeability. RhoA is also directly linked to AJs through p120 catenin, a Src substrate that binds to the juxta-membrane domain of VE-cadherin.¹⁴ Overexpression of p120 catenin led to inhibition of RhoA in a GDP dissociation inhibitor–like manner.¹⁵ Up- or downregulation of p120 catenin in endothelial cells influenced VE-cadherin levels in endothelial cells.¹⁶ In addition, activation of Cdc42 and Rac1 regulated the post-Golgi transport of endothelial cadherin to the AJs.¹⁷ Other studies showed that Cdc42 can signal the activation of actin polymerization at the plasma membrane leading to formation of filopodia¹⁸ and thus contributing to assembly of AJs.¹⁹ Cdc42 and Rac1 may also affect AJ function through their interactions with the scaffold protein IQGAP1.^20,21 Although the role of IQGAP1 in regulating AJ assembly has not been studied in endothelial cells, it was shown to colocalize with members of the AJ complex in other cell types²¹ and thus may be an important determinant of junctional permeability downstream of Cdc42 and Rac1.

In the present study, we addressed the role of Cdc42 in the mechanism of increased endothelial permeability. Studies were made in the mouse lung microcirculation, so that inferences could be drawn concerning the in vivo regulation of endothelial permeability. We used the VE-cadherin mutant lacking the extracellular domain (EXD) and another mutant lacking both extracellular domain and cytosolic distal ß-catenin–binding domain (EXDß).²² We tested the hypothesis that expression of EXD promotes a shift in the VE-cadherin–bound ß-catenin as well as -catenin toward the expressed mutant and thus would destabilize AJs. Because studies have shown that expression of EXD in endothelial cells induced Cdc42 activation,²² we also addressed whether destabilization of AJs secondary to the loss of VE-cadherin homotypic interaction increases endothelial permeability through a Cdc42-dependent mechanism. We observed an increase in lung microvascular permeability on EXD expression, which did not occur to the same degree with EXDß expression. The increased endothelial permeability was the result of loss of normal VE-cadherin homotypic interaction induced by the binding of ß-catenin and -catenin to the expressed EXD mutant. We observed that coexpression of the dominant-negative (dn) Cdc42 mutant (N17Cdc42) significantly reduced the increase in lung vascular permeability induced by EXD. This was the result of preventing the association of -catenin with the EXD–ß-catenin complex. Thus, activation of Cdc42 plays an important role in the mechanism of EXD-induced AJ disruption and increased endothelial permeability by promoting the interaction of -catenin with the EXD–ß-catenin complex.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
VE-Cadherin and Rho GTPase Mutants
The plasmid vector pcDNA3 was cleaved at its multicloning site and specific cDNAs corresponding to truncated sequences of VE-cadherin were inserted downstream of cytomegalovirus promoter. We constructed, as described,²² a VE-cadherin mutant lacking the extracellular domain (EXD) and a mutant lacking both the extracellular domain and cytosolic distal ß-catenin–binding domain (EXDß) (Figure 1A). Vector-control studies were performed with empty pcDNA3 plasmid vector (Invitrogen, Carlsbad, Calif). Cytomegalovirus-driven dominant active Cdc42 (V12), dnCdc42 (N17), Rac1 (N17), and RhoA (N19) plasmid DNAs were obtained from Drs Tohru Kozasa and Tatyana Voyno-Yasenetskaya (University of Illinois, Chicago).

View larger version (41K): [in this window] [in a new window]   Figure 1. Expression of EXD mutant in junctions of lung endothelial cells in situ. A, Wild-type VE-cadherin protein is shown. S indicates signal peptide; P, pre-peptide; CR, cadherin repeat; JMD, juxta-membrane domain; DBD, distal ß-catenin–binding domain. The first mutant, EXD, lacks the extracellular domain and contains a FLAG epitope. The second mutant, EXDß, lacks both the extracellular domain and DBD. B, Expression of EXD and N17Cdc42 in mouse lung endothelial cells. Endothelial cell–enriched lung lysates were analyzed for expression of constructs. The second and third lanes show the expression of FLAG-EXD (36 kDa). The third and fourth lanes show the coexpression of FLAG-EXD with myc-N17Cdc42 and myc-N19Rho, respectively (25 kDa). Expression of these constructs did not affect expression of total VE-cadherin (140 kDa) and Cdc42 (23 kDa). C, Expression of EXD was also determined by immunofluorescence in lung endothelial cells isolated from liposome–EXD cDNA-injected mice. Lung endothelial cells were obtained 24 hours after IV injection of constructs and stained for FLAG-EXD with anti-FLAG (green) and anti–VE-cadherin (red) antibodies. Maximum expression was seen at this time period (bottom) (scale=10 µmol/L). Top, Control endothelial cells showing only VE-cadherin staining (scale=20 µmol/L). Results are representative of 4 experiments.

Cell Culture and Transfection
HMEC-1 (Human Microvascular Endothelial Cells) and HPAEC (Human Pulmonary Artery Endothelial Cells) were grown²² and transfected by electroporation as described in the online data supplement, available at http://circres.ahajournals.org.

Immunofluorescence and Confocal Microscopy
Endothelial cells were either seeded onto glass coverslips posttransfection and allowed to grow for 24 hours in complete medium or freshly obtained from collagenase-digested lungs and plated onto coverslips for 1.5 hour (described below). Cells were immunostained and visualized as described²³ in the online data supplement.

Preparation of Cationic Liposomes for In Vivo Studies
Liposomes composed of dimethyldioctadecyl ammonium bromide (Sigma) in a 1:1 molar ratio with cholesterol (Sigma) were prepared as described,^24–28 except that the dried lipid film was resuspended in 5% dextrose in water and then sonicated for 20 minutes, followed by incubation at 42°C for 20 minutes and 0.45-µm filtration. Liposomes were extruded through a 50-nm pore polycarbonate filter (Avestin). CD1 mice (Charles River, Wilmington, Mass), weighing 20 to 25 g, were injected IV with 50 µg of plasmid DNA, which was mixed with 100 µL of liposome suspension and allowed to equilibrate for 20 minutes before injection. Protein expression constructs was assessed 24 hours thereafter. All experimental procedures complied with Institutional and National Institutes of Health guidelines for animal use, and approvals were obtained from the Institutional Animal Care Committee.

Lung Endothelial Fractionization
CD1 mice were anesthetized²⁹ and excised lungs were perfused with Hanks’ balanced salt solution through the pulmonary artery for 5 minutes, then with endothelial fractionation buffer (0.2% Triton X-100, 50 mmol/L Tris-Cl pH 7.9, 1x Protease Inhibitor Cocktail, Phosphatase Inhibitor Cocktails 1 and 2, Sigma) at 0.5 mL/min for 30 sec. Eluate was collected from the left atrial cannula (200 µL) for Western blotting, and remaining lung tissue was homogenized in tissue lysis buffer (1.5% Triton X-100, 0.1% sodium dodecyl sulfate, 0.5% dideoxycholate, 100 mmol/L phenylmethylsulfonyl fluoride, 1x Protease Inhibitor Cocktail in PBS), followed by centrifugation at 3000g at 4°C for 10 minutes to remove insoluble material. Remaining lung tissue was compared with endothelial-rich lysate by Western blotting using endothelial cell markers (eg, angiotensin-converting enzyme, VE-cadherin). Endothelial-rich lysate stained strongly positive for endothelial antigens (data not shown); the purity of this lysate was 90%.

Isolation of Endothelial Cells by Collagenase Digestion of Mouse Lungs
To assess the expression of transfected proteins in endothelial cells, mice were euthanized 20 hours postinjection of liposome–DNA mixture. Lungs were perfused as above, diced into 2-mm cubes and transferred to 1% collagenase A (Roche)/Hanks’ balanced salt solution (Gibco) and mixed at 37°C for 1 hour. Lung tissue was aspirated 10x through a serological pipette and allowed to settle at room temperature. The supernatant was removed and centrifuged at 3000g for 1 minute. The supernatant was discarded, and cells were resuspended in modified EGM-2 endothelial cell medium (penicillin/streptomycin instead of gentamicin/amphotericin B; Clonetics), and allowed to adhere to gelatinized coverslips for 1.5 hour. Cells were subsequently analyzed by immunofluorescence.

Pulmonary Microvascular Filtration Coefficient and Isogravimetric Lung Water Determinations
CD1 mice were anesthetized and lungs were removed, ventilated, and perfused ex vivo using our methods^2,29 to obtain stable pulmonary artery pressure (7 cm H₂O) and lung wet weight over a minimum of a 90-minute period. Microvessel filtration coefficient (K_f,c) was measured by applying a 10 cm H₂O brief step increase in left atrial pressure at 20 minutes postextraction (during the isogravimetric period).^2,29

Immunoprecipitation and Western Blotting
Cell and tissue lysates were subjected to immunoprecipitation and Western blotting as described in the online data supplement.

Cdc42 Activation Assay
Cdc42 activity was measured using the Cdc42 activation assay Biochem Kit (BK034, Cytoskeleton, Denver, Colo). Whole cell lysates were also analyzed for Cdc42 expression and total protein.

Data Analysis
Data were analyzed using the 2-tailed Student’s t test as well as ANOVA. Values are reported as mean±SEM. Values were considered significant at P<0.05. Densitometry measurements of Western blots were performed using the ImageJ program (NIH).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Expression of EXD in Junctions of Lung Endothelial Cells In Situ
Figure 1B shows the expression of FLAG-EXD as well as myc-N17Cdc42 in mouse lung endothelial cells. Endothelial cell-enriched lysates from mouse lungs were analyzed for expression of the constructs. The second and third lanes show the expression of FLAG-EXD (36 kDa), and the third lane shows the coexpression of myc-N17Cdc42 (25 kDa) (Figure 1B). Expression of FLAG-EXD construct did not affect the expression of wild-type VE-cadherin and Cdc42. Figure 1C shows the expression of EXD in lung endothelial cells isolated from the liposome–DNA–injected mice. Lung endothelial cells were obtained 24 hours postinjection of construct and stained with anti-FLAG (green) and anti–VE-cadherin (red) antibodies. Maximum expression of EXD at the junctions seen at 24 hours was colocalized with the endogenous VE-cadherin (bottom) in contrast to control endothelial cells (top) (Figure 1C).

Expression of EXD in Confluent Endothelial Cells Induces Junctional Destabilization
Figure 2A shows results of confocal microscopy using anti-FLAG and anti–ß-catenin antibodies in the EXD-expressing (top) and EXDß-expressing (bottom) confluent HPAEC monolayers. ß-Catenin was localized to the cell junctions. Expression of EXD caused the formation of numerous intercellular gaps and filopodia, whereas this did not occur with expression of EXDß (Figure 2A). Similar studies were performed in HMEC-1 cells (supplemental Figure I). EXD, but not EXDß, colocalized with ß-catenin. Figure 2B shows the results of anti–ß-catenin immunoprecipitation of HMEC lysates expressing EXD or EXDß followed by immunoblotting. A decrease of total ß-catenin–bound VE-cadherin was observed on expression of EXD (25±5% decrease, second lane), but not with EXDß (top, second and third lanes). ß-Catenin interacted with EXD but not with EXDß (third panel). The VE-cadherin truncation mutants showed equal expression (bottom) and did not alter the endogenous VE-cadherin expression (fourth panel). Figure 2C shows inhibition of interaction of VE-cadherin with ß-catenin following different levels of EXD expression. HMECs were transfected with 12 µg or 6 µg EXD cDNA (third and fourth lanes), and lysates were immunoprecipitated with anti–VE-cadherin antibody. Decrease in binding between VE-cadherin and ß-catenin (bottom) in the presence of EXD was proportional to the level of expression of EXD (middle). The first lane shows immunoprecipitation with an isotype-matched control goat antibody.

View larger version (75K): [in this window] [in a new window]   Figure 2. Expression of EXD in confluent endothelial cells disrupts adherens junctions. A, Confocal microscopy using anti-FLAG and anti–ß-catenin antibodies in EXD-transfected (top) and EXDß-transfected (bottom) confluent HPAEC monolayers at 24 hours posttransfection. ß-Catenin was prominently localized at the plasma membrane. Expression of EXD caused widespread formation of intercellular gaps and filopodia (arrows) (top, middle), whereas this did not occur with the expression of EXDß (bottom, middle). Also EXD colocalized with ß-catenin (yellow in merged image, top), and this did not occur to the same degree following expression of EXDß (bottom). Results are representative of 4 experiments (scale=10 µmol/L). B, Interactions of VE-cadherin mutants and ß-catenin. Lysates from HMECs expressing EXD and EXDß were immunoprecipitated with anti–ß-catenin antibody and immunostained for endogenous VE-cadherin, ß-catenin, and FLAG-EXD or -EXDß. In the first panel of B, EXD-expressing cells (second lane) showed decreased interaction between ß-catenin and endogenous VE-cadherin compared with control cells (first lane). In contrast, EXDß failed to coimmunoprecipitate with ß-catenin (third lane). Quantification of results indicated 25±5% decrease in association from control between ß-catenin and endogenous VE-cadherin following EXD expression and no change following EXDß expression. In the second panel of B, immunoblotting for ß-catenin showed similar levels of ß-catenin expression in the EXD- or EXDß-expressing endothelial cells. In the third panel of B, immunoblotting for FLAG showed association with ß-catenin in the EXD- but not in EXDß-expressing endothelial cells. The fourth and fifth panels of B show immunoblots of whole cell lysates with VE-cadherin and expressed FLAG-tagged EXD or EXDß mutants. All results are representative of 4 separate experiments. C, ß-Catenin binding to EXD is dependent on level of EXD expression. HMECs were transfected with 0, 6, or 12 µg of EXD cDNA (second, third, and fourth lanes) and immunoprecipitated with control antibody (first lane) or anti–VE-cadherin antibody (second, third, and fourth lanes). Equal loading was measured by anti–VE-cadherin immunoblot (top). Middle, The two different levels of EXD expression (third and fourth lanes). VE-cadherin/ß-catenin interaction was reduced to a greater extent with the higher level of EXD expression (decrease of 34% vs 17%) (bottom, third lane vs fourth lane).

Cdc42 Regulates the Association of -Catenin With VE-Cadherin
The top 2 panels of Figure 3A show results of immunoprecipitation of HMEC lysates with anti-FLAG antibody followed by immunoblotting with anti–-catenin or ß-catenin monoclonal antibody (mAb). In EXD-expressing cells, both - and ß-catenin were associated with EXD (second lane), but not with EXDß (third lane) (Figure 3A). Coexpression of N17Cdc42 interfered with the association of -catenin with EXD, whereas the association of ß-catenin with EXD was unaffected (fourth lane) (Figure 3A). In contrast, coexpression of N19RhoA had no effect on the association of -catenin or ß-catenin with EXD (fifth lane). The bottom 3 panels of Figure 3A show expression of similar amounts of FLAG-tagged EXD or EXDß (in second through fifth lanes) as well as similar expression of myc-tagged dnRhoA, dnCdc42, and endogenous VE-cadherin. Probing these immunoblots for the transfected myc-Cdc42 and myc-RhoA showed no interaction with the FLAG-EXD construct (data not shown).

View larger version (58K): [in this window] [in a new window]   Figure 3. Cdc42 regulates the association of -catenin with ß-catenin. The first and second panels of A show results of immunoprecipitation of HMEC lysates with anti-FLAG antibody followed by immunoblotting with anti–-catenin or ß-catenin antibody. In EXD-expressing cells, both - and ß-catenin were immunoprecipitated (second lane); however, this failed to occur after expression of EXDß (third lane). Coexpression of N17Cdc42 prevented the association of -catenin, but not ß-catenin, with EXD (fourth lane). In contrast, coexpression of N19Rho had no effect on the association of -catenin or ß-catenin with EXD (fifth lane). The third and fourth panels of A show similar levels of expression of EXD and EXDß as well as wild-type VE-cadherin in all lanes. The fifth panel of A shows similar levels of myc expression in cells coexpressing N17Cdc42 or N19RhoA. B, Results of immunoprecipitation of lysates with ß-catenin mAb followed by immunoblotting with anti–-catenin Ab. First panel of B shows increased -catenin association with ß-catenin in the EXD-expressing cells (3-fold increase over control, second lane) but not in EXDß-expressing cells (third lane). N17Cdc42 prevented the EXD-mediated increase (1.3-fold, fourth lane); in contrast, expression of N19RhoA had no effect (3-fold, fifth lane). The second panel of B shows results of immunoblotting for ß-catenin following ß-catenin immunoprecipitation. The third panel of B shows ß-catenin immunoblot in whole-cell lysates.

Figure 3B shows the results of immunoprecipitation of lysates with ß-catenin mAb followed by immunoblotting with anti–-catenin antibody. In EXD-expressing cells, we found increased -catenin association with ß-catenin (second lane) (Figure 3B). This failed to occur with the expression of EXDß (third lane). Coexpression of N17Cdc42 with EXD prevented the increased association of -catenin with ß-catenin (fourth lane), whereas coexpression of N19RhoA had no effect on -catenin association with ß-catenin (fifth lane). Expression of EXD and Rho GTPases also had no effect on endogenous ß-catenin expression (bottom 2 panels) (Figure 3B).

Figure 4A shows results of Cdc42 GTPase activity. Expression of EXD increased the amount of activated Cdc42 (second lane), whereas coexpression of N17Cdc42 abrogated this response (fourth lane). N19RhoA had no effect on Cdc42 activity induced by EXD expression (fifth lane). Transfection of EXDß also increased Cdc42 activity relative to control, but to a lesser degree than EXD (third lane). Total Cdc42 levels in the lysate were assessed using Western blotting (Figure 4A, bottom); the upper band in Lane 3 represents the myc-tagged N17Cdc42. The expression of myc-tagged Cdc42 and RhoA was comparable (Figure 4A).

View larger version (59K): [in this window] [in a new window]   Figure 4. A, Expression of EXD induces Cdc42 GTPase activity. Expression of EXD and EXDß increased the amount of activated Cdc42 in HMECs, but a reduced response was observed following EXDß expression (second lane vs third lane). Coexpression of N17Cdc42 abrogated this response (fourth lane), whereas N19RhoA had no effect (fifth lane). Total Cdc42 levels were assessed by Western blotting (middle); the upper band in Lane 4 represents myc-tagged N17Cdc42. The expression levels of myc-tagged Cdc42 and RhoA were comparable (bottom). B. Dominant-active Cdc42 mutant promotes increased binding of -catenin and ß-catenin. Results of immunoprecipitation of HMEC lysates with anti–ß-catenin mAb followed by immunoblotting with anti–-catenin mAb. The first panel shows transfection with dominant-active (V12) Cdc42 caused an increase in the affinity of -catenin for ß-catenin (third lane), whereas dominant-negative (N17) Cdc42 decreased this interaction by 75%. (fourth lane). Expression levels of immunoprecipitated ß-catenin (second panel), lysate -catenin and myc-tagged Cdc42 were comparable (second and third panels). All results shown above are representative of 3 experiments.

Figure 4B shows the direct effect of Cdc42 activation and inhibition on the interaction between -catenin and ß-catenin. Expression of V12 dnCdc42 in endothelial cells increased this interaction (top, third lane), whereasN17 dnCdc42 decreased it by 75% (top, fourth lane). Immunoprecipitation with mouse nonspecific IgG did not immunoprecipitate -catenin (first lane), and lysate levels of -catenin and myc-Cdc42 constructs were comparable (second and third lanes, Figure 4B).

Cdc42-Dependent Increase in Lung Vascular Permeability Following EXD Expression
Figure 5A shows the gravimetric analysis of lungs obtained from mice expressing EXD and EXDß mutants in lung endothelia. The EXD-expressing lungs became markedly edematous at 60 minutes postextraction in contrast to control lungs, which gained weight only at later time points (Figure 5A). This early weight increase is indicative of compromised endothelial barrier function. EXDß-expressing lungs were less edematous than EXD-expressing lungs such that the values between the 2 groups were different (P<0.05) at the time points shown. Figure 5B and 5C show that in EXD lungs the coexpression of N17Cdc42, in contrast to coexpression of N19RhoA, reduced the edema formation (P<0.05). Coexpression of N17Rac1 also failed to inhibit pulmonary edema (data not shown). Figure 5D shows K_f,c measurements. Expression of EXD increased vascular permeability compared with control lungs (P<0.05). The increase in permeability was markedly less in the EXDß-expressing lungs relative to the expression of EXD (P<0.05). Coexpression of N17Cdc42, in contrast to coexpression of N19RhoA or N17Rac1, reduced the increase in vascular permeability seen in the EXD-expressing lungs (P<0.05). Expression of N17Cdc42 alone had no effect on basal pulmonary vascular permeability values (Figure 5D).

View larger version (29K): [in this window] [in a new window]   Figure 5. Cdc42-dependent increase in lung vascular permeability induced by EXD. A, Gravimetric analysis of isolated-perfused mouse lungs expressing EXD or EXDß mutant in endothelia 24 hours postinjection of construct. The EXD-expressing lungs became markedly edematous in contrast to control vector-transfected lungs (*P<0.05). EXDß-expressing lungs were less edematous than EXD-expressing lungs at 70 to 90 minutes after initiation of perfusion (P<0.05; n=4). B, Coexpression of N17Cdc42, in contrast to N19RhoA, reduced edema formation in EXD-expressing lungs between 80 and 90 minutes (*P<0.05). Coexpression of N17Rac1 also failed to inhibit pulmonary edema at these time points (data not shown). C, Analysis of lung wet weight gain from A and B during the 60- to 90-minute perfusion period. Lungs transfected with EXD and EXD+N19RhoA showed differences in rate of weight increase compared with EXD and EXD+N17Cdc42 groups (**P<0.05). All values were different from control vector group (*P<0.05). D, Lung microvessel filtration coefficient (Kf,c). Expression of EXD, EXDß, and EXD+N17Cdc42 increased Kf,c compared with control lungs (*P<0.05; n=4 in each group). The increase in Kf,c was less in EXDß-expressing lungs than in EXD-expressing lungs (**P<0.05). Coexpression of N17Cdc42, in contrast to N19RhoA or N17Rac1 (P<0.05), reduced the increase in Kf,c seen in EXD-expressing lungs (**P<0.05, n=3 in each group). Expression of N17Cdc42 alone did not increase Kf,c. All values are reported as mean±SEM.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
We performed these studies to address the effects of destabilizing the VE-cadherin homotypic interaction by the in vivo expression of the VE-cadherin cytoplasmic domain on lung vascular permeability and to determine the role of Rho GTPases in mediating the alterations in permeability. We showed that destabilization of VE-cadherin junctions induced by the expression of the VE-cadherin mutant, EXD, increased pulmonary vascular endothelial permeability in mice. Studies in confluent endothelial cells showed that expression of EXD resulted in the formation of intercellular gaps, indicating that the EXD competitively disrupted the normal VE-cadherin homotypic interactions, and thus increased vascular permeability. Previous studies have shown that cadherin proteins display a strong homophilic interaction in the presence of catenins bound to the cytoplasmic domain of cadherin.³⁰ We show herein that the EXD-expressing cells disrupted normal AJ assembly because of the redistribution of - and ß-catenins away from the endogenous VE-cadherins localized at the AJs. This notion is supported by the finding that the expression of EXD promoted its association with - and ß-catenins, whereas these effects were not observed after transfection of the EXDß mutant lacking the ß-catenin–binding domain.

Our mouse lung studies were performed using cationic liposomes to transduce the expression of the EXD mutant in lung vascular endothelial cells in vivo. We have shown that proteins are encoded in lung endothelial cells following the IV injection of the liposome–DNA complex.^25,31 Using endothelial-rich fractions and endothelial cells obtained by collagenase digestion of mouse lungs, we demonstrated the expression of the FLAG-tagged EXD construct in endothelial cells. The expression of the EXD mutant was maximal at 24 hour after the liposome–DNA construct injection when all of the physiological measurements were made.

In studies using confluent endothelial cells, we observed that EXD colocalized with ß-catenin at the plasma membrane. Expression of EXD caused widespread formation of intercellular gaps and filopodia, whereas these changes did not occur with the expression of EXDß. The impairment of normal VE-cadherin/catenin interactions induced by EXD likely interfered with AJ formation because of the loss of connection to the actin cytoskeleton through the ß- and -catenins.³² We observed a marked shift of ß-catenin from native VE-cadherin to the expressed EXD mutant in support of this concept. In contrast, the expression of EXDß had a less pronounced effect on lung vascular permeability compared with EXD. We noted that the transient transfection of EXD in the present study did not reduce endogenous VE-cadherin levels, an effect seen previously in studies using adenoviral transfection of a VE-cadherin mutant similar to EXD.¹⁶ On stable transfection of EXD in HMEC-1 using a retroviral vector; however, we have also observed a decrease in endogenous VE-cadherin expression (data not shown). Taken together, these findings support the hypothesis that translocation of ß-catenin from endogenous VE-cadherin to the expressed EXD mutant resulted in junctional instability and increased vascular permeability.

We have shown that expression of the EXD mutant in endothelial cells induces the specific activation of Cdc42 as compared with RhoA and Rac1²²; thus, we addressed the possibility that Cdc42 may be involved in regulating AJ disassembly induced by EXD expression. These experiments were made by cotransfecting dominant-negative mutants of Cdc42, RhoA, and Rac1 along with EXD. Our goal in using these constructs was to minimize the potential nonspecific effects of Rho GTPase-modifying reagents³³ and ensure the simultaneous delivery of both VE-cadherin and Rho GTPase constructs to assess protein–protein interactions and the role of Rho GTPases in mediating AJ disassembly. We identified an important role of Cdc42 in regulating the association of -catenin with ß-catenin following the binding of ß-catenin to the expressed EXD. Strikingly, the coexpression of dnCdc42 (N17Cdc42) prevented the association of -catenin with the EXD–ß-catenin complex, whereas dnRhoA or dnRac had no effect. In a study involving the immunoprecipitation of endothelial cell lysates with ß-catenin mAb followed by immunoblotting with anti–-catenin antibody, we observed that EXD expression induced the association of - with ß-catenin, whereas this did not occur on expression of EXDß. The coexpression of dnCdc42 with EXD prevented the association of -catenin with ß-catenin; in contrast, coexpression of dnRhoA or dnRac had no effect on this association. Our findings demonstrate the key role of Cdc42 in regulating the binding of -catenin to EXD–ß-catenin complex without altering the interaction of ß-catenin to EXD. Thus, the mechanism of AJ instability induced by the Cdc42-mediated -catenin association with the EXD–ß-catenin complex involves a shift of actin filaments from the endogenous VE-cadherin/catenin complex to the expressed EXD.

The mutant EXDß served as a useful control for EXD expression in these studies because it did not significantly disrupt AJs in endothelial cells. Additionally, the level of Cdc42 activation following EXDß expression was significantly lower than EXD. We observed that EXDß expression in the mouse lung increased vascular permeability, although the response was significantly attenuated compared with EXD. The basis of this diminished but persistent effect of EXDß in increasing endothelial permeability is not clear, but it may be attributed to the reported interaction of EXDß with p120 catenin,^22,34 which could result in actin cytoskeletal rearrangement at the level of AJs^32,35 and promote AJ alterations, albeit to a lesser degree than EXD.

We observed that the increase in vascular permeability following expression of EXD in mouse lung endothelial cells was attributed to Cdc42. Gravimetric analysis of lungs obtained from mice expressing the VE-cadherin mutants in lung endothelia showed that the EXD-expressing lungs became markedly edematous and coexpression of dnCdc42 significantly reduced both edema formation and increase in K_f,c in these lungs. However, expression of dnCdc42 did not completely inhibit these effects of EXD, suggesting there may also be a Cdc42-independent effect on AJs. Another explanation for the incomplete inhibition of permeability is that dnCdc42 expression may not have fully blocked Cdc42 activity.

Although our results show that Cdc42 is involved in the mechanism of increased vascular permeability induced by the loss of homotypic VE-cadherin interactions, we cannot rule out the possibility that Cdc42 has additional other regulatory effects on endothelial barrier function. Our previous studies have shown that the reannealing of endothelial AJs occurring 1 to 2 hours following the thrombin-induced increase in permeability depends on Cdc42 activation.²³ These dual actions of Cdc42 (ie, promoting the binding of -catenin to ß-catenin shown in the present study and previously described reannealing of AJs^23,36) suggest that Cdc42 activation regulates both AJ disassembly as well as reassembly. This concept is consistent with the versatility of Rho GTPases as effectors^37,38; thus, the Cdc42-GTP "switch" may integrate both events depending on its spatial and temporal activation in the endothelial cell membrane. An important pathophysiological implication of our studies is that the Cdc42-mediated binding of -catenin with ß-catenin may serve to dampen the increase in endothelial permeability in response to inflammatory mediators.

	Footnotes

 
Original received May 16, 2005; revision received October 17, 2005; accepted November 10, 2005.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Caveda L, Martin-Padura I, Navarro P, Breviario F, Corada M, Gulino D, Lampugnani MG, Dejana E. Inhibition of cultured cell growth by vascular endothelial cadherin (cadherin-5/VE-cadherin). J Clin Invest. 1996; 98: 886–893.[Medline] [Order article via Infotrieve]
2. Gao X, Kouklis P, Xu N, Minshall RD, Sandoval R, Vogel SM, Malik AB. Reversibility of increased microvessel permeability in response to VE-cadherin disassembly. Am J Physiol Lung Cell Mol Physiol. 2000; 279: L1218–L1225.[Abstract/Free Full Text]
3. Venkiteswaran K, Xiao K, Summers S, Calkins CC, Vincent PA, Pumiglia K, Kowalczyk AP. Regulation of endothelial barrier function and growth by VE-cadherin, plakoglobin, and beta-catenin. Am J Physiol Cell Physiol. 2002; 283: C811–C821.[Abstract/Free Full Text]
4. Bazzoni G, Dejana E. Endothelial cell-to-cell junctions: molecular organization and role in vascular homeostasis. Physiol Rev. 2004; 84: 869–901.[Abstract/Free Full Text]
5. Lampugnani MG, Corada M, Caveda L, Breviario F, Ayalon O, Geiger B, Dejana E. The molecular organization of endothelial cell to cell junctions: differential association of plakoglobin, beta-catenin, and alpha-catenin with vascular endothelial cadherin (VE-cadherin). J Cell Biol. 1995; 129: 203–217.[Abstract/Free Full Text]
6. Stevens T, Garcia JGN, Shasby DM, Bhattacharya J, Malik AB. Mechanisms regulating endothelial cell barrier function. Am J Physiol Lung Cell Mol Physiol. 2000; 279: L419–L422.[Abstract/Free Full Text]
7. Konstantoulaki M, Kouklis P, Malik AB. Protein kinase C modifications of VE-cadherin, p120, and {beta}-catenin contribute to endothelial barrier dysregulation induced by thrombin. Am J Physiol Lung Cell Mol Physiol. 2003; 285: L434–L442.[Abstract/Free Full Text]
8. Shasby DM, Ries DR, Shasby SS, Winter MC. Histamine stimulates phosphorylation of adherens junction proteins and alters their link to vimentin. Am J Physiol Lung Cell Mol Physiol. 2002; 282: L1330–L1338.[Abstract/Free Full Text]
9. Sandoval R, Malik AB, Minshall RD, Kouklis P, Ellis CA, Tiruppathi C. Ca2+ signalling and PKC{alpha} activate increased endothelial permeability by disassembly of VE-cadherin junctions. J Physiol (Lond). 2001; 533: 433–445.[Abstract/Free Full Text]
10. Nawroth R, Poell G, Ranft A, Kloep S, Samulowitz U, Fachinger G, Golding M, Shima DT, Deutsch U, Vestweber D. VE-PTP and VE-cadherin ectodomains interact to facilitate regulation of phosphorylation and cell contacts. EMBO J. 2002; 21: 4885–4895.[CrossRef][Medline] [Order article via Infotrieve]
11. Fukata M, Kaibuchi K. Rho-family GTPases in cadherin-mediated cell-cell adhesion. Nat Rev Mol Cell Biol. 2001; 2: 887–897.[CrossRef][Medline] [Order article via Infotrieve]
12. Zigmond S. Formin’ adherens junctions. Nat Cell Biol. 2004; 6: 12–14.[Medline] [Order article via Infotrieve]
13. Wojciak-Stothard B, Potempa S, Eichholtz T, Ridley A. Rho and Rac but not Cdc42 regulate endothelial cell permeability. J Cell Sci. 2001; 114: 1343–1355.[Abstract]
14. Reynolds AB, Roczniak-Ferguson A. Emerging roles for p120-catenin in cell adhesion and cancer. Oncogene. 2004; 23: 7947–7956.[CrossRef][Medline] [Order article via Infotrieve]
15. Anastasiadis PZ, Moon SY, Thoreson MA, Mariner DJ, Crawford HC, Zheng Y, Reynolds AB. Inhibition of RhoA by p120 catenin. Nat Cell Biol. 2000; 2: 637–644.[CrossRef][Medline] [Order article via Infotrieve]
16. Xiao K, Allison DF, Buckley KM, Kottke MD, Vincent PA, Faundez V, Kowalczyk AP. Cellular levels of p120 catenin function as a set point for cadherin expression levels in microvascular endothelial cells. J Cell Biol. 2003; 163: 535–545.[Abstract/Free Full Text]
17. Wang B, Wylie FG, Teasdale RD, Stow JL. The polarized trafficking of E-cadherin is regulated by Rac1 and Cdc42 in MDCK cells. Am J Physiol Cell Physiol. 2005; 288: C1411–C1419.[Abstract/Free Full Text]
18. Nobes CD, Hall A. Rho, Rac, and Cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell. 1995; 81: 53–62.[CrossRef][Medline] [Order article via Infotrieve]
19. Wojciak-Stothard B, Entwistle A, Garg R, Ridley AJ. Regulation of TNF-alpha-induced reorganization of the actin cytoskeleton and cell-cell junctions by Rho, Rac, and Cdc42 in human endothelial cells. J Cell Physiol. 1998; 176: 150–165.[CrossRef][Medline] [Order article via Infotrieve]
20. Kuroda S, Fukata M, Kobayashi K, Nakafuku M, Nomura N, Iwamatsu A, Kaibuchi K. Identification of IQGAP as a putative target for the small GTPases, Cdc42 and Rac1. J Biol Chem. 1996; 271: 23363–23367.[Abstract/Free Full Text]
21. Briggs MW, Sacks DB. IQGAP1 as signal integrator: Ca2+, calmodulin, Cdc42 and the cytoskeleton. FEBS Lett. 2003; 542: 7–11.[CrossRef][Medline] [Order article via Infotrieve]
22. Kouklis P, Konstantoulaki M, Malik AB. VE-cadherin-induced Cdc42 signaling regulates formation of membrane protrusions in endothelial cells. J Biol Chem. 2003; 278: 16230–16236.[Abstract/Free Full Text]
23. Kouklis P, Konstantoulaki M, Vogel S, Broman M, Malik AB. Cdc42 regulates the restoration of endothelial barrier function. Circ Res. 2004; 94: 159–166.[Abstract/Free Full Text]
24. Xu N, Gao X-P, Minshall RD, Rahman A, Malik AB. Time-dependent reversal of sepsis-induced PMN uptake and lung vascular injury by expression of CD18 antagonist. Am J Physiol Lung Cell Mol Physiol. 2002; 282: L796–L802.[Abstract/Free Full Text]
25. Zhou M-Y, Lo SK, Bergenfeldt M, Tiruppathi C, Jaffe A, Xu N, Malik AB. In vivo expression of neutrophil inhibitory factor via gene transfer prevents lipopolysaccharide-induced lung neutrophil infiltration and injury by a beta 2 integrin-dependent mechanism. J Clin Invest. 1998; 101: 2427–2437.[Medline] [Order article via Infotrieve]
26. Gao X, Huang L. Cationic liposome-mediated gene transfer. Gene Ther. 1995; 2: 710–722.[Medline] [Order article via Infotrieve]
27. Liu Y, Liggitt D, Zhong W, Tu G, Gaensler K, Debs R. Cationic liposome-mediated intravenous gene delivery. J Biol Chem. 1995; 270: 24864–24870.[Abstract/Free Full Text]
28. Thurston G, McLean JW, Rizen M, Baluk P, Haskell A, Murphy TJ, Hanahan D, McDonald DM. Cationic liposomes target angiogenic endothelial cells in tumors and chronic inflammation in mice. J Clin Invest. 1998; 101: 1401–1413.[Medline] [Order article via Infotrieve]
29. Tiruppathi C, Freichel M, Vogel SM, Paria BC, Mehta D, Flockerzi V, Malik AB. Impairment of store-operated Ca2+ entry in TRPC4^–/– mice interferes with increase in lung microvascular permeability. Circ Res. 2002; 91: 70–76.[Abstract/Free Full Text]
30. Navarro P, Caveda L, Breviario F, Mndoteanu I, Lampugnani M-G, Dejana E. Catenin-dependent and -independent functions of vascular endothelial cadherin. J Biol Chem. 1995; 270: 30965–30972.[Abstract/Free Full Text]
31. Ong ES, Gao X-P, Xu N, Predescu D, Rahman A, Broman MT, Jho DH, Malik AB. E. coli pneumonia induces CD18-independent airway neutrophil transalveolar migration in the absence of increased lung vascular permeability. Am J Physiol Lung Cell Mol Physiol. 2003; 285: L879–L888.[Abstract/Free Full Text]
32. Stockton RA, Schaefer E, Schwartz MA. p21-activated kinase regulates endothelial permeability through modulation of contractility. J Biol Chem. 2004; 279: 46621–46630.[Abstract/Free Full Text]
33. Kowluru A, Amin R. Inhibitors of post-translational modifications of G-proteins as probes to study the pancreatic beta cell function: potential therapeutic implications. Curr Drug Targets Immune Endocr Metabol Disord. 2002; 2: 129–139.[CrossRef][Medline] [Order article via Infotrieve]
34. Iyer S, Ferreri DM, DeCocco NC, Minnear FL, Vincent PA. VE-cadherin-p120 interaction is required for maintenance of endothelial barrier function. Am J Physiol Lung Cell Mol Physiol. 2004; 286: L1143–L1153.[Abstract/Free Full Text]
35. Lee TY, Gotlieb AI. Microfilaments and microtubules maintain endothelial integrity. Microsc Res Tech. 2003; 60: 115–127.[CrossRef][Medline] [Order article via Infotrieve]
36. Birukov KG, Bochkov VN, Birukova AA, Kawkitinarong K, Rios A, Leitner A, Verin AD, Bokoch GM, Leitinger N, Garcia JGN. Epoxycyclopentenone-containing oxidized phospholipids restore endothelial barrier function via Cdc42 and Rac. Circ Res. 2004; 95: 892–901.[Abstract/Free Full Text]
37. Aspenstrom P. Effectors for the Rho GTPases. Curr Opin Cell Biol. 1999; 11: 95–102.[CrossRef][Medline] [Order article via Infotrieve]
38. Aspenstrom P, Fransson A, Saras J. Rho GTPases have diverse effects on the organization of the actin filament system. Biochem J. 2004; 377: 327–337.[CrossRef][Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				R. Ramchandran, D. Mehta, S. M. Vogel, M. K. Mirza, P. Kouklis, and A. B. Malik Critical role of Cdc42 in mediating endothelial barrier protection in vivo Am J Physiol Lung Cell Mol Physiol, August 1, 2008; 295(2): L363 - L369. [Abstract] [Full Text] [PDF]

				S. Sehrawat, X. Cullere, S. Patel, J. Italiano Jr., and T. N. Mayadas Role of Epac1, an Exchange Factor for Rap GTPases, in Endothelial Microtubule Dynamics and Barrier Function Mol. Biol. Cell, March 1, 2008; 19(3): 1261 - 1270. [Abstract] [Full Text] [PDF]

				D. Vestweber VE-Cadherin: The Major Endothelial Adhesion Molecule Controlling Cellular Junctions and Blood Vessel Formation Arterioscler. Thromb. Vasc. Biol., February 1, 2008; 28(2): 223 - 232. [Abstract] [Full Text] [PDF]

				T. Mammoto, S. M. Parikh, A. Mammoto, D. Gallagher, B. Chan, G. Mostoslavsky, D. E. Ingber, and V. P. Sukhatme Angiopoietin-1 Requires p190 RhoGAP to Protect against Vascular Leakage in Vivo J. Biol. Chem., August 17, 2007; 282(33): 23910 - 23918. [Abstract] [Full Text] [PDF]

				J. Orrington-Myers, X. Gao, P. Kouklis, M. Broman, A. Rahman, S. M. Vogel, and A. B. Malik Regulation of lung neutrophil recruitment by VE-cadherin Am J Physiol Lung Cell Mol Physiol, October 1, 2006; 291(4): L764 - L771. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: 98/1/73    most recent 01.RES.0000198387.44395.e9v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services