FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by LIU, F.
Right arrow Articles by GARCIA, J. G. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by LIU, F.
Right arrow Articles by GARCIA, J. G. N.
(The FASEB Journal. 2002;16:950-962.)
© 2002 FASEB

Hepatocyte growth factor enhances endothelial cell barrier function and cortical cytoskeletal rearrangement: potential role of glycogen synthase kinase-3ß

FENG LIU, KANE L. SCHAPHORST, ALEXANDER D. VERIN, KERI JACOBS, ANNA BIRUKOVA, REGINA M. DAY*, NATALIA BOGATCHEVA, D. P. BOTTARO{ddagger} and JOE G. N. GARCIA

Division of Pulmonary and Critical Care Medicine, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA;
* Tufts University/New England Medical Center, Boston, Massachusetts, USA; and
{ddagger} National Cancer Institute, Bethesda, Maryland, USA


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The stabilization of endothelial cell (EC) barrier function within newly formed capillaries is a critical feature of angiogenesis. We examined human lung EC barrier regulation elicited by hepatocyte growth factor (HGF), a recognized angiogenic factor and EC chemoattractant. HGF rapidly and dose-dependently elevated transendothelial electrical resistance (TER) of EC monolayers (>50% increase at 100 ng/ml), with immunofluorescence microscopic evidence of both cytoplasmic actin stress fiber dissolution and strong augmentation of the cortical actin ring. HGF rapidly stimulated phosphatidylinositol 3'-kinase, ERK, p38 mitogen-activated protein kinase, and protein kinase C activities. Pharmacological inhibitor studies demonstrated each pathway to be intimately involved in HGF-induced increases in TER, cortical actin thickening, and phosphorylation of the Ser/Thr glycogen synthase kinase-3ß (GSK-3ß), a potential target for the HGF barrier-promoting response. GSK-3ß phosphorylation was strongly correlated with reductions in both HGF-induced TER and enhanced ß-catenin immunoreactivity observed at cell-cell junctions. Our data suggest a model in which HGF-mediated EC cytoskeletal rearrangement and barrier enhancement depend critically on the activation of a complex kinase cascade that converges at GSK-3ß to increase the availability of ß-catenin, thereby enhancing endothelial junctional integrity and vascular barrier function.—Liu, F., Schaphorst, K. L., Verin, A. D., Jacobs, K., Birukova, A., Day, R. M., Bogatcheva, N., Bottaro, D. P., Garcia, J. G. N. Hepatocyte growth factor enhances endothelial cell barrier function and cortical cytoskeletal rearrangement: potential role of glycogen synthase kinase-3ß.


Key Words: ß-catenin • MAP kinases • endothelial permeability • cytoskeleton • transendothelial electrical resistance


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
HEPATOCYTE GROWTH FACTOR (HGF), also known as scatter factor, is a heparin-binding glycoprotein originally identified as a fibroblast product that induces scattering of contiguous epithelial sheets into isolated cells (1) . Subsequently, HGF was recognized as a multifunctional cytokine secreted by several cell types (2) displaying diverse biological effects including mitogenesis, motogenesis, morphogenesis, organogenesis, and cell survival (3 , 4) . HGF also elicits potent angiogenic activities (5 , 6) , with direct stimulation of endothelial cell (EC) motility, proliferation, protease production, invasion, and organization into capillary-like tubes (2) . These complex biological functions occur via ligation of the HGF tyrosine kinase receptor known as c-Met, which is composed of a 50-kDa extracellular {alpha} subunit and a 145-kDa transmembrane ß subunit (7) . The ß subunit contains tyrosine kinase domains, tyrosine phosphorylation sites, and tyrosine-docking sites (8) . Binding of HGF with the receptor stimulates receptor tyrosine kinase activity, leading to autophosphorylation of the receptor, followed by the recruitment of multiple SH2 domain-containing signaling molecules including Grb2-associated binder-1 (Gab-1), Grb2, phosphatidylinositol 3'-kinase (PI-3'-kinase), phospholipase C{gamma}, p60src, Shc, and Shp2 (9 , 10) . These signaling components are likely involved in diverse responses, which include prevention of apoptosis (11) , activation of mitogen-activated protein kinase (MAPK) pathways, and branching morphogenesis (10) .

The complex angiogenic effects of HGF have not been studied in the pulmonary circulation where the pulmonary vascular endothelium functions as a semiselective barrier regulating the exchange of fluid, macromolecules, and cells between blood vessels and the surrounding lung tissues. Vascular barrier regulation is now recognized to be involved in the multifaceted process of angiogenesis (12 , 13) , as newly formed capillaries are leaky and therefore not fully functional (14) . Several angiogenic factors regulate vascular barrier function including vascular endothelial growth factor (VEGF), formerly known as vascular permeability factor (15) , and both angiopoietin-1 and -2 (14) . Increases in VEGF are observed in inflammatory lung syndromes (15) and in the ischemic lung (16) and may contribute to EC activation, formation of intercellular gaps, and increased vascular permeability and life-threatening edema (17) in patients with acute lung injuries. We previously reported that the platelet phospholipid growth factor and complete angiogenic factor sphingosine 1-phosphate (12) participates in the terminal angiogenic effect characterized by barrier stabilization of the newly formed but leaky vessels (18) via ligation of Edg-1 receptors (12) . Targeted disruption of the Edg-1 gene in mice leads to embryonic lethality with progressive edema formation and hemorrhage (19) . Thus, the maintenance of the normal EC barrier and the integrity of the microcirculation are also final processes during new blood vessel formation. The potentially important role of HGF in modulating pulmonary EC barrier properties has not been previously addressed, however.

In the present study, we defined a role for HGF in human pulmonary vascular endothelial barrier regulation and identified signaling pathways that contribute to HGF-evoked barrier alterations. Our data demonstrate that HGF potently enhances EC barrier integrity; that is, HGF reduces permeability as reflected by increases in transendothelial electrical resistance (TER). These changes occur in association with increased cortical actin rearrangement, and improved adherens junction integrity as defined by vascular endothelial (VE)-cadherin/ß-catenin association with the cytoskeleton. Both physiological and immunofluorescent events depended on PI-3'-kinase, MAPK, and protein kinase C (PKC) activities. HGF-mediated EC barrier protection may be a critical biological property of this important angiogenic factor.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Reagents
HGF was purchased from R&D Systems (Minneapolis, MN). Anti-phospho-Akt, anti-phospho-GSK3ß, anti-Akt antibodies, and LY-294002 were purchased from Cell Signaling (Beverly, MA). Anti-GSK3ß antisera and Rac activity assay kit were obtained from Upstate Biotechnology (Lake Placid, NY). Anti-PKC{alpha} and anti-ß-catenin antisera were from Transduction Labs (Lexington, KY). Anti-phospho-pan-PKC, anti-pan-ERK, anti-phospho-p44/42 ERK, anti-p38 MAPK, and anti-phospho-p38 MAPK antibodies were purchased from New England Biolabs (Beverly, MA), and SB-203580, UO126, PP2, and protease inhibitory mixture were from Calbiochem (La Jolla, CA). Myosin light chain antibody was produced in rabbits against baculovirus-expressed and purified smooth muscle myosin light chain by Biodesign International (Kennebunk, ME). Protein G Sepharose 4 Fast Flow was purchased from Amersham Pharmacia Biotech (Piscataway, NJ), and the enhanced chemiluminescence detection system (ECL) was from Amersham (Little Chalfront, Buckinghamshire, England). Reagents used for immunofluorescent staining were purchased from Molecular Probes (Eugene, OR), and all other common reagents were obtained from Sigma Chemical (St. Louis, MO). NK2 was produced and purified as described previously (20) .

Cell culture
Bovine pulmonary artery ECs were purchased from American Type Culture Collection (ATCC, Rockville, MD) and utilized at passages 19–24. Cells were maintained in Medium 199 (Life Technologies, Rockville, MD) supplemented with 20% (v/v) colostrum-free bovine serum (Irvine Scientific, Santa Ana, CA), 15 µg/ml EC growth supplement (Upstate Biotechnology), 1% antibiotic and antimycotic, and 0.1 mM nonessential amino acids (Life Technologies). Human pulmonary artery ECs were purchased from Clonetics (Walkersville, MD) and were cultured in EBM-2 complete medium (Clonetics) and utilized at passages 5–10. Human alveolar epithelial cells (A549) were acquired from ATCC and cultured in the same medium as were bovine ECs except for omitting the EC growth factor. All cells were maintained at 37°C in a humidified atmosphere of 5% CO2 and 95% air. Both EC types grew to contact-inhibited monolayers with the typical cobblestone morphology (21 , 22) .

Measurement of transendothelial monolayer electrical resistance
Electrical resistance across EC monolayers was measured by using an electrical cell impedance sensor system (Applied Biophysics, Troy, NY), as we previously described (23) . Cells grown on gold microelectrodes (10-3 cm2) in polycarbonate wells act as insulating particles, and the resistance across the monolayers (TER) is measured in real time. TER increases as cells adhere on the microelectrode and intercellular cell contacts are formed, or in response to agents that increase cell-cell adhesive interactions (12) . In contrast, cell retraction, rounding, or loss of adhesion is reflected by decreases in TER (23) . These measurements provide a highly sensitive biophysical assay that indicates the state of cell shape, focal adhesion, and endothelial barrier function (24 , 25) . All electrical resistance data are presented as normalized values. Briefly, current was applied across the electrodes by a 4000-Hz AC voltage source with an amplitude of 1 V in series with a 1-M{Omega} resistance to approximate a constant current source (~1 µA). The small gold electrode and the larger counter electrode (1 cm2) are connected to a phase-sensitive lock-in amplifier (5301A; EG&G Instruments, Princeton, NJ) with a built-in differential preamplifier (5316A; EG&G Instruments). The in-phase and out-of-phase voltages between the electrodes were monitored in real time with the lock-in amplifier and converted to scalar measurements of transendothelial impedance, of which resistance was the primary focus. TER was monitored for 30 min to establish a baseline resistance (R0), and the mean baseline resistance for each cell type was determined by pooling electrode data (n = 8–14). For human pulmonary artery endothelium on 10 electrode/well arrays R0 = 1300.3 ± 123.4 {Omega}, and for A549 cells R0 = 1240.2 ± 35.2 {Omega}. Subsequent wells whose baselines exceeded 2 standard deviations from the pooled means were rejected from analysis. For some experiments, total TER was vectorially resolved into components reflecting resistance to current flow beneath the cell layer ({alpha}) and resistance to current flow between adjacent cells (Rb) as we previously described (23) utilizing the method of Giaiver and Keese, which models the endothelial monolayer mathematically (24) . Thus, changes in {alpha} reflect alterations in the net state of cell-matrix adhesion, whereas changes in Rb reflect alterations in the integrity of cell-cell adhesion. TER values from each microelectrode were pooled at discrete time points and plotted vs. time as the mean ± standard error (23) .

Western immunoblotting
Endothelial cell monolayers grown to confluence in 12-well plates and challenged with HGF were lysed with 100 µl of 2 x sodium dodecyl sulfate (SDS) sample buffer and cell lysates were transferred into microcentrifuge tubes and boiled for 5 min. After a brief spin, proteins from 10-µl cell lysates were separated on 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose (Schleicher & Schuell, Keene, NH) (30 V, 18 h). After blocking the samples with PBST (PBS with 0.1% Tween 20) containing 5% nonfat milk for 1 h, nitrocellulose blots were reacted with primary antibodies diluted in PBST containing 5% bovine serum albumin for 1 h, washed with PBST (three times for 10 min each time), incubated with peroxidase-conjugated secondary antibodies (goat anti-rabbit IgG, 1:10,000 dilution, Sigma; or goat anti-mouse IgG, 1:10,000 dilution, Bio-Rad Laboratories, Richmond, CA) diluted in PBST with 5% nonfat milk for 1 h and again washed with PBST (three times for 10 min each time). Finally, immunoreactive proteins were detected using ECL. The relative intensities of the protein bands were quantified via scanning densitometry.

Differential detergent fractionation of subcellular components
Endothelial cells were fractionated into cytosolic, membrane, and nuclear/cytoskeleton fractions as we previously described (26) . Briefly, endothelial monolayers were incubated with cytosolic buffer (0.01% digitonin, 10 mM piperazine-N,N'-bis[2-ethanesulfonic acid] (PIPES), pH 6.8, 300 mM sucrose, 100 mM NaCl, 3 mM MgCl2, 5 mM EDTA, 5 µM phallicidin) and protease inhibitory mixture with agitation for 10 min at 4°C. The digitonin-soluble faction (cytosolic fraction) was collected, and the residual material was incubated with membrane buffer (0.5% Triton X-100, 10 mM PIPES, pH 7.4, 300 mM sucrose, 100 mM NaCl, 3 mM EDTA, 5 µM phallicidin, and protease inhibitory mixture) with agitation for 20 min at 4°C. The Triton-soluble (membrane) fraction was collected, and the materials remaining on dishes were scraped in SDS buffer (0.5% Triton X-100, 0.5% SDS, 10 mM Tris-HCl, pH 6.8, and protease inhibitory mixture). The mixture was then sonicated, boiled, and centrifuged, and the supernatants (cytoskeletal fraction) together with other two fractions were subjected to SDS-PAGE and Western immunoblotting.

Measurement of Rac GTPase activity
Rac GTPase activity was assessed as we recently described (12) . ECs grown in 100-mm dishes were incubated with agonists in serum-free Medium 199. Cells were lysed in 500 µl of Mg2+ lysis buffer (Upstate Biotechnology) and homogenized by pipetting. After a brief centrifugation to remove cell debris, 300-µl samples of supernatants were incubated with the agarose-conjugated Rac-binding domain of human PAK-1 (10 µg, 30 min, Upstate Biotechnology). The agarose beads were washed five times with 1 ml of lysis buffer and were resuspended in 30 µl of 2 x SDS buffer. After 10 min of centrifugation at 14,000g, 15 µl of supernatant from each sample was subjected to electrophoresis in 15% PAGE. After Western transfer, active Rac was detected by using anti-Rac monoclonal antibody. For total Rac protein measurement, 5-µl samples of the original cell lysates were used for electrophoresis and Western analysis.

Immunofluorescence microscopy
EC monolayers grown on gelatinized coverslips were rinsed with Medium 199 and incubated with agonists in the same medium in a 37°C incubator (5% CO2). Monolayers were then rinsed with PBS (three times for 2 min each time), fixed in 4% paraformaldehyde for 10 min, again rinsed with PBS (three times for 2 min each time), and permeabilized with 0.25% Triton X-100 for 5 min. Cells were then washed briefly with PBS (three times for 2 min each time), blocked with PBS containing 2% bovine serum albumin for 30 min, and incubated with 1 U/ml Texas Red-X phalloidin (Molecular Probes), ß-catenin antibody (Transduction Laboratories), glycogen synthase kinase-3ß (GSK-3ß) (Santa Cruz Biotechnology, Santa Cruz, CA), or monophosphorylated myosin light chain antisera (see ref 12 ) for 1 h. After samples were washed with PBS (three times for 2 min each time), coverslips were mounted on slides using SlowFade mounting medium (Molecular Probes). Cells were analyzed using a 60 x oil objective on a Nikon Eclipse TE 300 microscope. Images were captured by Sony Digital Photo camera DKC 5000. The same exposure time was applied to all samples within one experiment.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Hepatocyte growth factor increases TER
Human and bovine pulmonary artery EC monolayers, grown on gold microelectrodes to monitor real-time TER, were challenged with serial doses of HGF (from 1 to 100 ng/ml). HGF increased TER in a dose-dependent manner, with an elevation in TER clearly evident after 1–2 min (Fig. 1 A). HGF-induced barrier enhancement peaked 15–20 min after exposure to 100 ng/ml HGF, with an increase in resistance from about 1600 {Omega} (baseline TER) to about 2400 {Omega}, reflecting an approximate 50% enhancement in barrier function; this enhancement was sustained above baseline values for several hours. No further significant increases in TER were observed with concentrations of HGF greater than 100 ng/ml. HGF-mediated elevations in TER across bovine pulmonary artery ECs (Table 1 ) were similar to those in human cells with regard to time and concentration dependence. HGF-mediated barrier protection, however, appears to be specific for lung ECs because HGF did not alter TER values in an immortalized A549 human alveolar epithelial cell line, whereas sphingosine 1-phosphate (12) , another recently described barrier-enhancing angiogenic agent, enhanced lung epithelial cell integrity (Fig. 1B ), consistent with tissue- and stimulus-specific TER responses.



View larger version (33K):
[in this window]
[in a new window]
  		Figure 1. HGF increased human TER. A) Human pulmonary artery ECs were grown to confluence on gelatinized gold microelectrodes. Two hours prior to TER measurement, growth medium was replaced with serum-free Medium 199. Serial-diluted HGF was added to cells at indicated concentrations, and TER was monitored for 2.5 h. HGF increased TER a dose-dependent manner consistent with barrier enhancement. The result shown is a representative TER tracing of three independent experiments. B) Similar to results shown in A, human alveolar epithelial cells (A549) were grown on gold microelectrodes and challenged with vehicle, HGF (100 ng/ml), or sphingosine 1-phosphate (Sph 1-P) (1 µM). Depicted is the differential sensitivity to sphingosine 1-phosphate; HGF was completely without effect. These results indicate that HGF increases in electrical resistance and that enhanced paracellular integrity is specific to endothelium. C) In these experiments, HGF (1, 10, or 100 ng/ml) was added to human EC monolayers prior to subsequent restimulation at 2 h with HGF (10 ng/ml). Whereas the GF barrier-protective response was not altered by prior HGF stimulation at 1 ng/ml, pretreatment with HGF at 10 and 100 ng/ml significantly reduced the subsequent HGF responses, consistent with receptor desensitization. D) Human pulmonary ECs were pretreated with cytochalasin B (5 µg/ml) and challenged with HGF (20 mg/ml). The dramatic enhanced TER response observed with HGF was abolished by prior microfilament disruption evoked by cytochalasin B pretreatment, which directly reduces TER. Furthermore, cytochalasin B added to HGF-challenged endothelium rapidly abolished the HGF barrier enhancement.


View this table:
[in this window]
[in a new window]
  		Table 1. Effect of angiogenic factors on transendothelial electrical resistance across bovine pulmonary artery endotheliuma

HGF is known to signal through its specific tyrosine kinase receptor, c-Met, and we next examined whether c-Met signaling might be modulated by HGF challenge. Human pulmonary artery endothelium exhibited the expected barrier-enhancing response to an initial challenge with HGF (1–100 ng/ml) (Fig. 1C ). However when HGF pretreated endothelium were rechallenged with 10 ng/ml HGF at 2 h, the TER response was reduced in those monolayers pretreated with 10 ng/ml HGF and completely absent in monolayers pretreated with 100 ng/ml HGF (Fig. 1C ). Thus, we observed that HGF stimulation produced dose-dependent attenuation of TER values in response to subsequent HGF challenge, consistent with receptor desensitization. HGF/NK2 is a naturally occurring 28-kDa truncated HGF isoform derived from an alternatively spliced HGF transcript, which in specific cellular systems binds c-Met with high affinity and functions as a partial agonist (27) . Conversely, NK2 is capable of functionally antagonizing HGF effects on HGF-induced mitogenesis (28 29 30) . We failed to observe a significant direct response of human endothelium to NK2 (1–100 ng/ml) (data not shown). Furthermore, subsequent HGF challenge in NK2-pretreated endothelium resembled the effect of HGF in vehicle-treated monolayers, suggesting that the barrier-protective response of HGF is not affected by its truncated splice variant (data not shown).

HGF enhances cortical actin ring formation: role of Rac GTPases
We and other investigators have shown EC barrier regulation to depend critically on the dynamics of EC actin cytoskeleton organization (17 , 21) . As recently shown for sphingosine 1-phosphate (12) , an intact microfilamentous cytoskeleton is also required for HGF-mediated enhancement of EC barrier function. When human ECs were pretreated with cytochalasin B, an agent that disrupts F-actin filaments, subsequent HGF challenge failed to elicit an increase in TER (Fig. 1D ). We next examined the effect of HGF on the spatial localization of polymerized actin in human endothelial monolayers by immunofluorescence microscopy. Consistent with the evoked increases in EC TER, HGF (20 ng/ml) produced rapid enhancement of F-actin staining, which was spatially confined to the cortical cytoskeletal ring, with reproducible increases in monophosphorylated myosin light chains in the same distribution (Fig. 2 ); we previously noted such results with the barrier enhancement induced by sphingosine 1-phosphate (12) . In many cell systems, cytoskeletal rearrangements are tightly regulated by Rac GTPases, signaling effectors whose activities are intimately involved with dramatic alterations in the endothelial cortical cytoskeleton and cytoplasmic stress fibers (12) . Consistent with Rac GTPase-mediated cytoskeletal rearrangement, both HGF and sphingosine 1-phosphate produced rapid (1 min) Rac activation as determined by Rac GTP-binding to the Rac-binding domain of the p21-associated kinase PAK (Fig. 3 ).



View larger version (120K):
[in this window]
[in a new window]
  		Figure 2. HGF enhanced cortical polymerized actin immunofluorescence and myosin light chain (MLC) phosphorylation. Cells were treated with either vehicle or HGF (20 ng/ml for 5 min). F-actin staining was assessed with Texas Red-X phalloidin, and myosin light chain staining was evaluated with anti-monophosphorylated myosin light chain polyclonal antibody. HGF significantly enhanced cortical actomyosin staining, which correlates with enhancement of EC barrier function.



View larger version (60K):
[in this window]
[in a new window]
  		Figure 3. HGF rapidly activated Rac GTPases involved in EC barrier enhancement. Pulmonary artery ECs were incubated with HGF (10 ng/ml) or sphingosine 1-phosphate (Sph 1-P) (1 µM) for the indicated times. Cells were lysed, supernatants were collected, and activated GTP-bound Rac was precipitated by agarose-conjugated human PAK-1 p21-binding domain and subsequently immunoblotted by use of anti-Rac monoclonal antibody. Total Rac protein was detected via cell lysates. Both sphingosine 1-phosphate and HGF rapidly and transiently increased Rac activity in ECs.

HGF-mediated endothelial TER enhancement involves PI-3'-kinase activity
HGF elicits stimulus/coupling responses in a number of cells, which often depend on PI-3'-kinase activation. To identify key signaling mediators involved in HGF-induced barrier protection, we first examined the role of PI-3'-kinase in HGF-stimulated barrier improvement and characterized HGF-dependent phosphorylation of the serine/threonine kinase Akt, a well-accepted method of defining PI-3'-kinase activity. Figure 4 A demonstrates the rapid increase in Akt phosphorylation beginning at 2 min, with the maximal effect reaching a plateau by 5–30 min and with activation sustained up to 2 h, in response to concentrations as low as 5 ng/ml (Fig. 4B ). The highly specific PI-3'-kinase inhibitor LY294002 (25 µM, 30 min) abolished HGF-mediated Akt phosphorylation, confirming that PI-3'-kinase activity is the key effector in this response (Fig. 4B ). In the next series of experiments, human EC monolayers were grown on gold microelectrodes and preincubated with LY294002, followed by stimulation with HGF (20 ng/ml). PI-3'-kinase inhibition with LY294002 reduced the elevation of HGF-induced TER by more than 50% (Fig. 5 A, B), a fundamental difference between HGF and sphingosine 1-phosphate signaling because sphingosine 1-phosphate, which ligates G protein-coupled Edg receptors, does not require PI-3'-kinase for either EC migration (22) or barrier enhancement (12) . Similarly, LY294002 abolished the enhanced cortical actin ring formation elicited by HGF (Fig. 5C ) but did not affect sphingosine 1-phosphate-induced actin reorganization (data not shown). These results suggest that PI-3'-kinase plays a critical role in the HGF-mediated signaling pathway leading to EC cytoskeleton reorganization and subsequent barrier enhancement.



View larger version (39K):
[in this window]
[in a new window]
  		Figure 4. HGF rapidly activated PI-3'-kinase in human pulmonary endothelium. Endothelial monolayers were incubated with HGF (20 ng/ml) for indicated time periods (A) or were incubated with indicated concentrations of HGF for 15 min (B). In some cases, cells were pretreated with LY294002 (25 µM, 1 h) followed by incubation with HGF (B). Cell homogenates were analyzed by Western immunoblotting with anti-phospho-AktSer473 antibody. HGF stimulated phosphorylation of Akt, a downstream kinase of PI-3'-kinase, in a time- and dose-dependent manner. LY294002 completely abolished HGF-induced Akt phosphorylation.



View larger version (63K):
[in this window]
[in a new window]
  		Figure 5. Effect of PI-3'-kinase inhibition on HGF-induced EC cortical actin rearrangement and barrier enhancement. A) Human pulmonary artery EC monolayers were pretreated with LY294002 (25 µM, 1 h), or vehicle control, followed by stimulation with HGF (20 ng/ml). TER was monitored for 2.5 h. The maximal increases in TER elicited by HGF were expressed as the percent increase over vehicle control (data were collected at 15 min after HGF addition). The reductions of HGF-induced TER increases by LY294002 were expressed as a percentage of the maximal TER increases by HGF in the absence of the inhibitor. LY294002 significantly attenuated increases in TER stimulated by HGF. Data represent the mean ± SD from three independent experiments (two wells each). B) The electrical resistance tracing is a representative experiment (n = 3) demonstrating the effect PI-3'-kinase inhibition by LY294002 on the increase in TER induced by HGF. Data are presented as normalized resistance. C) These immunofluorescent images depict EC monolayers pretreated with LY294002 (25 µM, 1 h), or vehicle control, followed by stimulation with HGF (20 ng/ml, 15 min). Cells were fixed, permeabilized, and stained with Texas Red-X phalloidin to visualize polymerized actin bundles. HGF induced dissolution of cytosolic F-actin with robust increases in cortical ring polymerized actin, which was attenuated by LY294002.

MAPKs are involved in HGF-stimulated EC barrier enhancement
The MAP family of kinases (ERK1/2 and p38) is actively involved in agonist-induced EC actin reorganization and barrier regulation (12 , 31) and has been noted to participate in HGF-mediated cell activation (32) . These reports were confirmed for human EC monolayers incubated with HGF (20 ng/ml), in which activation of p42/44 ERK (Fig. 6 A) and p38 MAPK (Fig. 6B ) by HGF was detected by immunoblotting with antibodies that recognize only the phosphorylated (thereby activated) forms of ERK or p38 MAPK. The activation of ERK was evident at 5 min and was maximal after 10–15 min, with a gradual decline thereafter but remaining sustained above basal levels for more than 2 h. Pretreatment with the specific ERK inhibitor UO126 (10 µM, 30 min) completely abolished HGF-stimulated ERK activity. The onset of HGF-mediated activation of p38 MAPK was similar to ERK, with a plateau at 10–15 min. However, the duration of this response was much more truncated than ERK activation, beginning to decline by 20 min and returning to the basal value by 1 h. To determine whether MAPK signaling events are important in the barrier enhancement mediated by HGF, EC monolayers were preincubated with UO126 or the p38 MAPK inhibitor SB203580 (20 µM, 30 min), followed by stimulation with HGF (20 ng/ml) or sphingosine 1-phosphate, again used as a negative control (12 , 22) . HGF-mediated barrier protection (Fig. 7 A, B) and actin cytoskeletal remodeling (data not shown) were significantly attenuated by p38 MAPK inhibition. Attenuation of HGF-induced TER increases occurred to a lesser extent with MEK inhibition; however, consistent with Fig. 5B , the coadministration of UO126 and LY294002 essentially abolished the HGF-induced increases in TER (Fig. 7C ). These data indicate important roles for both p38 MAPK and ERK signaling pathways in HGF-mediated EC barrier protection.



View larger version (68K):
[in this window]
[in a new window]
  		Figure 6. HGF rapidly activated p42/44 ERK and p38 MAPK. A) Human EC monolayers were treated with HGF (20 ng/m) for the indicated times. For samples pretreated with the MEK inhibitor UO126 (the last two lanes), cell monolayers were incubated with UO126 (10 µM, 30 min) followed by incubation with HGF (20 ng/ml, 15 min) or vehicle control. The levels of ERK1/2 phosphorylation were analyzed by Western immunoblotting of cell homogenates with specific anti-phospho-p44/42 MAPK (1 µg/ml). The blots were then stripped and reprobed with anti-pan-ERK antibody (50 ng/ml) to detect total ERK protein content. HGF induced ERK activation in a time-dependent manner, which was abolished by UO126. Data shown represent three independent experiments. B) Similar to findings shown in A, human ECs were exposed to HGF for the depicted times and lysates were probed for phospho-p38 as an indication of p38 MAPK activity. The same blots were stripped and probed with a pan-p38 MAPK antibody. HGF stimulated p38 MAPK in a time-dependent manner, but this activity was not sustained when compared with that of p42/p44 MAPK.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 7. Effect of MAPK inhibitors on increases in TER induced by HGF. A) Human pulmonary artery EC monolayers were pretreated with ERK kinase (MEK) inhibitor UO126 (10 µM, 1 h), p38 inhibitor SB203580 (20 µM, 1 h), or vehicle control, followed by stimulation with HGF (20 ng/ml). TER was continuously monitored for 2.5 h. UO126 and SB203580 significantly blocked HGF-induced increases in TER. Data are means ± SE, n = 3 for the UO126 experiment, n = 4 for the SB203580 experiment. B) Depicted is the HGF-mediated TER response in the presence and absence of p38 MAPK inhibition with SB203580. Inhibition of p38 MAP kinase produced a marked reduction in the HGF-mediated increases in TER. C) Similar to the experiments in shown in A, human ECs were exposed to a combination of LY7294002 (25 µM) and UO126 (10 µM), which produced near total abolishment of the HGF-mediated increase in TER.

PKC activity is required for the enhancement of EC barrier function evoked by HGF
We and others recently noted PKC isotype-specific regulation of EC barrier function, which evolves in an agonist-specific manner (21 , 31 , 33 , 34) . As HGF stimulates PKC activity in certain cell types (35) , we tested whether PKC{alpha} is involved in HGF-mediated TER increases in human endothelium. Initial experiments confirmed HGF-mediated PKC{alpha} activation detailed by increases in phospho-PKC immunoreactivity (Fig. 8 A) as well as by rapid (5 min) translocation to the membrane fraction after HGF (Fig. 8B ). Endothelial cell monolayers were next pretreated with a highly specific pan-PKC inhibitor, Ro-31–2880 (10 µM, 30 min), which preferentially inhibits membrane-bound PKC isoforms. As shown in Fig. 8C , treatment with Ro-31–2880 produced an 80% reduction in HGF (20 ng/ml)-evoked increases in TER, implying a major role for PKC in barrier enhancement mediated by HGF.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 8. Involvement of PKC activities in HGF-induced barrier enhancement. A) EC lysates were blotted with antisera immunoreactive with phosphorylated PKC{alpha}, an index of enzymatic activation. HGF-mediated PKC{alpha} activation was detectable at 15 min. B) PKC{alpha} immunoblots of HGF-treated human endothelial subcellular fractions obtained by differential detergent fractionation as described in Materials and Methods. PKC{alpha} immunoreactivity translocated from the cytosol to the membrane fraction, reflecting enzymatic activation. C) EC monolayers grown on gold microelectrodes were pretreated with the specific pan-PKC inhibitor Ro-31–2880 (10 µM, 30 min), or vehicle control, followed by stimulation with HGF (20 ng/ml). TER was continuously monitored for 2.5 h. Ro-31–2880 significantly attenuated HGF-induced increases in TER. Data represent the mean ± SD from four independent experiments. D) The effect of Ro-31–2880 on HGF-mediated increases in TER is depicted. PKC inhibition produced significant elevation in TER alone but blunted the HGF response.

Role of GSK-3ß in HGF-induced endothelial barrier enhancement
Increases in barrier function are conceptualized as reflecting either enhanced cell-matrix adhesion via focal adhesions or strong increases in cell-cell tethering produced by homotypic cadherin linkage via catenins to the actin cytoskeleton (17) . For example, ß-catenin is a critical component of the adherens junction (36) and is essential for EC monolayer integrity and paracellular barrier regulation (17) . Increased ß-catenin availability has been postulated to increase intercellular tethering and thus enhance cell-cell adhesion (37) . Consistent with this cell-cell tethering paradigm, partitioning of electrical resistance vectors across human ECs grown on gold microelectrodes identified sharp increases in paracellular junction resistance (Rb) after HGF (data not shown), results that were identical to the TER vectorial-derived responses to sphingosine 1-phosphate (12) . Consistent with enhanced paracellular resistance, HGF-stimulated human ECs examined by immunofluorescent microscopy demonstrate increased colocalization of ß-catenin immunoreactivity along cell borders with the cortical actin cytoskeleton (Fig. 9 ). These two events depended on PI-3'-kinase activation, as LY294002 diminished this response. Differential detergent fractionation revealed enhanced ß-catenin and VE-cadherin partitioning to the Triton-insoluble cytoskeletal fraction (Fig. 10 A), and immunoprecipitation of {alpha}-catenin after HGF challenge showed enhanced association with VE-cadherin (Fig. 10B ), results that indicate increased tethering of the cytoskeleton to zonula adherens proteins.



View larger version (97K):
[in this window]
[in a new window]
  		Figure 9. The effect of HGF on the immunocytochemical colocalization of ß-catenin with actin. Confluent endothelium grown on gelatinized coverslips were treated with HGF (20 ng/ml) for 15 min and then were fixed, permeabilized, and costained for F-actin (Texas Red-X phalloidin) and ß-catenin (monoclonal anti-ß-catenin green fluorescence), as described in Materials and Methods. A and B show F-actin and ß-catenin distribution in vehicle treated with or without pretreatment with the PI-3'-kinase inhibitor LY294002 (25 µM, 1 h). HGF produced an enhancement of the colocalization of the ß-catenin and F-actin fluorescence (arrowheads in C) as compared with the vehicle-treated control (A). The enhanced colocalization of actin and ß-catenin fluorescence induced by HGF was attenuated by pretreatment of cells with LY294002 (D). For each panel, the inset represents ß-catenin fluorescence without actin colocalization.



View larger version (26K):
[in this window]
[in a new window]
  		Figure 10. HGF increased VE-cadherin and ß-catenin linkage to the endothelial cytoskeleton. A) Human pulmonary artery ECs were grown to confluence in 100-mm2 dishes and then were challenged with either vehicle (PBS) or HGF (20 ng/ml). The monolayers were then partitioned by fractionation with a buffer containing Triton X-100 (1%), were separated by SDS-PAGE, and were immunoblotted with pan-cadherin antiserum or a monoclonal anti-ß-catenin antibody. The positions of cadherin (top) and ß-catenin (bottom) are indicated (arrows). The Triton-insoluble pellet contains the great majority of cellular F-actin (not shown). The relative densitometric unit (R.D.) denotes the optical density of HGF stimulated VE-cadherin and ß-catenin translocation compared with control. B) Lysates prepared from HGF-challenged human pulmonary artery endothelium (for the times indicated) were incubated overnight with a monoclonal antibody to {alpha}-catenin (1 µg/100 µg of lysate, 4°C). Immune complexes were collected with a protein A/G resin, washed five times with excess lysis buffer, and then boiled in electrophoresis sample buffer. The collected proteins were separated by SDS-PAGE and were then immunoblotted for {alpha}-catenin and VE-cadherin. The positions of VE-cadherin (top) and {alpha}-catenin (bottom) are indicated (arrows). HGF treatment induced a significant increase in the immunoreactive VE-cadherin coprecipitating with {alpha}-catenin.

HGF has been reported to increase the phosphorylation status of GSK-3ß, a multifunctional enzyme involved in glycogen synthesis and protein synthesis regulation in mammary epithelial cells (38) . Of potential importance to barrier regulation, GSK-3ß phosphorylation results in enzymatic inactivation and increases the level of uncomplexed cellular ß-catenin (39) . GSK-3ß phosphorylation can be catalyzed by multiple pathways including via the PI-3'-kinase-activated Akt (40) , by the p38 MAPK-activated protein kinase 1 (MAPKAP-K1), by MEK-dependent pathways (41) , or by PKC (42) . Given that our results indicate that each of these signaling paradigms is involved in HGF-stimulated barrier enhancement, we next examined the phosphorylation status of GSK-3ß mediated by HGF in human ECs by Western immunoblot analysis with antisera specific for the phosphorylated N-terminal Ser-9 of this enzyme (41) . Concomitant with HGF-induced TER augmentation, increases in phosphorylation of GSK-3ß after HGF were detected by increased fluorescence at immunofluorescence microscopy (Fig. 11 ) and prominent by Western blotting at 2 min, with peak intensity leveling off at 15–30 min, although the increases remained above basal levels for more than 2 h (Fig. 12 A). Phosphorylation of GSK-3ß induced by HGF was dramatically attenuated by pharmacological inhibitors of PI-3'-kinase (Fig. 12B, C ), ERK and p38 MAPK (Fig. 12C ), and PKC (Fig. 12D ).



View larger version (94K):
[in this window]
[in a new window]
  		Figure 11. HGF stimulated immunofluorescence microscopic detection of GSK-3ß phosphorylation. To detect HGF-mediated increases in GSK-3ß phosphorylation, human pulmonary artery endothelial cells were treated with 20 ng/ml for 15 min, then fixed and stained with phospho-GSK-3ß antibody (Ser-9) as described in Materials and Methods. Pictures were taken under 60 x magnification.



View larger version (57K):
[in this window]
[in a new window]
  		Figure 12. Regulation of HGF-induced GSK-3ß phosphorylation by PI-3'-kinase, ERK, and p38 MAPK. Depicted are the time course (A) and dose responses (B) of HGF-induced GSK-3ß phosphorylation. EC monolayers were incubated with HGF (20 ng/ml) for indicated time periods or were incubated with indicated concentrations of HGF for 15 min. Cell homogenates were analyzed by Western immunoblotting with anti-phospho-GSK-3ßSer9 (1 µg/ml) antibody. HGF stimulated GSK-3ß phosphorylation in a time- and dose-dependent manner. Data shown represent three independent experiments. C) EC monolayers were pretreated with the PI-3'-kinase inhibitor LY294002 (25 µM), MEK/ERK inhibitor UO126 (10 µM), p38 MAPK inhibitor SB203580 (20 µM), or vehicle control for 30 min, followed by stimulation with HGF (20 ng/ml, 15 min). Cell homogenates were analyzed by Western immunoblotting with anti-phospho-GSK-3ßSer9 (1 µg/ml) antibody and anti-GSK antibody. HGF-induced GSK-3ß phosphorylation was differentially attenuated by the reduction of PI-3'-kinase and p38 MAPK, but not by UO126, although this agent augmented the decrease in GSK-3ß phosphorylation in the presence of LY294002. D) Similar to C, pan-PKC inhibition with Ro-31–2880 abolished HGF-induced GSK-3ß phosphorylation.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
HGF, or scatter factor, originally cloned as a hepatocyte mitogen, has proved to be a multifunctional protein that influences tissue repair, survival, and angiogenesis (43) . Given the dual observations that alterations in vascular barrier function are essential to both early and late angiogenic processes (12 , 13) and that increased HGF levels are found in acute inflammatory lung syndromes of increased vascular permeability (44) , it was of interest to further examine HGF effects on human lung EC barrier properties. Our results clearly demonstrate dramatic HGF-mediated, dose-dependent increases in TER in primary human lung endothelial cultures. Our results further indicate that the enhancement of barrier function is specific to vascular ECs, as HGF did not enhance TER values in human alveolar epithelium. Jiang et al. have reported HGF to increase the paracellular permeability of human endothelium (45) . These results are in obvious direct conflict with our studies but may be attributable to the apparent differences between our primary cells and the cell line (ECV304) utilized in that report (45) . Although initially believed to be a putative human vascular EC line, ECV304 has subsequently proved to be derived from a bladder carcinoma epithelial cell line (46 , 47) . Also as an important difference, these cells do not express VE-cadherin, rendering them quite dissimilar from normal endothelial barrier regulation given that VE-cadherin is a major adherens protein regulating cell-cell interactions and barrier function in vascular ECs (17 , 36) . HGF has also recently been noted to decrease electrical resistance across epithelial cell monolayers (48 , 49) , whereas we failed to detect any effect of HGF on TER across a human alveolar epithelial cell line known to express c-Met (50 , 51) . This fundamental discrepancy in the response to HGF can be easily attributed to the different epithelial cell systems utilized. For example, the barrier-disruptive effect of HGF was observed in nonpulmonary epithelium, e.g., gastric and intestinal epithelial monolayers (48 , 49) . In contrast, Shen et al. described the lack of effect of HGF on primary tracheal epithelial TER responses (52) , findings that are consistent with our studies with an immortalized human alveolar epithelial cell line (Fig. 1B ).

The dramatic increase in TER observed was specific to HGF and not reproduced by the naturally occurring truncated HGF isoform NK2 (30) , a ligand for c-Met that failed to influence TER or to desensitize c-Met to subsequent HGF stimulation. The pleiotropic effects of HGF are known to be mediated by the SH2 domains within the tyrosine kinase receptor, which involve signaling pathways such as small G-protein GTPases, PI-3'-kinase, and MAPK family members (53) , with both PI-3'-kinase and ERK critically involved in HGF-mediated survival and angiogenesis (54) . In Madin-Darby canine kidney (MDCK) epithelial cells, HGF leads to transient activation of the Rho GTPases Rac and Cdc42, which correlates with the induction of filopodia and lamellipodia. Moreover, HGF-mediated Rac activation in MDCK cells depends on PI-3'-kinase activity (53) . We determined that HGF rapidly activates Rac GTPases and that PI-3'-kinase, ERK, p38 MAPKs, and PKC all participate in the pathways leading to EC barrier enhancement elicited by HGF. In specific growth factor signaling paradigms, ERK activation occurs via a PKC-dependent pathway (55) . Consistent with these results, inhibition of PKC with Ro-31–2880 (10 µM, 30 min) attenuated the phosphorylation of ERK induced by HGF (data not shown), suggesting that PKC is upstream of ERK in HGF-mediated signaling.

HGF-mediated increases in TER depend on enhanced actin cytoskeleton-tethering protein linkages. Unlike stimulated astrocytes, in which HGF rapidly produced stress fibers that disappeared after 30 min (35) , we found HGF to induce significant PI-3'-kinase-dependent increases in EC cortical actin polymerization. The marked increase in cortical polymerized actin after HGF is strikingly similar to our observations with sphingosine 1-phosphate (12) , which proceed in a Rac GTPase-dependent manner. We recently reported the sphingosine 1-phosphate-induced actin rearrangement to involve p21-associated kinase (PAK), LIM kinase, and the actin-severing protein cofilin. However, unlike HGF-induced changes, sphingosine 1-phosphate-induced increases in TER do not depend on PI-3'-kinase activity. Actin polymerization and rearrangement in response to sphingosine 1-phosphate and HGF were essential, as cytochalasin, which reduces actin polymerization, abolished the induced EC barrier enhancement.

Although the actual site of enhanced cortical cytoskeleton linkage to tethering structure is unknown, our TER studies designed to determine vectors by which electrical resistance is modulated indicated that HGF produces a primary increase in EC tethering via enhanced cell-cell adhesion and the bolstering of junctional integrity. Paracellular adherens protein constituents include VE-cadherin and {alpha}, ß, and {gamma}-catenins, which provide a linkage for the adherens junction to the cortical actin cytoskeleton. We explored one potential mechanism by which HGF may affect junctional integrity, i.e., via GSK-3ß, an enzyme initially characterized as an insulin-regulated kinase that participates in glycogen metabolism. GSK-3ß is generally found in a cytosolic complex with ß-catenin and the adenomatous polyposis coli protein (APC) (56) , and it phosphorylates APC in vitro within a region required for APC to regulate ß-catenin degradation (56) . A number of critical bioregulatory pathways are known to enhance GSK phosphorylation including the PI-3'-kinase target Akt (40) , the MAPK signaling cascade element MAPKAP-K1 (41) , and PKC (42) , signaling molecules that we have demonstrated to be involved in HGF-induced EC barrier enhancement (Figs. 4 5 6 7 8) . Phosphorylation of GSK-3ß on Ser (9) inhibits GSK-3ß kinase activity, and HGF treatment of mammary epithelial cells leads to a decrease in GSK-3ß activity that is accompanied by an increase in the uncomplexed free pool of ß-catenin and increased steady-state ß-catenin levels (38) . Our data, which showed PI-3'-kinase-dependent increases in ß-catenin immunoreactivity after HGF with strengthened linkage to the cytoskeleton, appear to be consistent with the notion that the increased availability and accumulation of ß-catenin potentially stabilize the binding of ß-catenin to the paracellular junctional adhesion protein VE-cadherin, resulting in the strengthening of calcium-dependent cell-cell adhesion (37) . However, HGF has also been reported to stimulate an increased association of E-cadherin and ß-catenin complex and colocalization of the complex with c-Met in prostate cancer cells (57) , suggesting a potentially direct effect of c-Met on the interaction of cadherin and ß-catenin.

Interestingly, the HGF receptor c-Met invokes many of the same signaling pathways also invoked by the VEGF receptor tyrosine kinase KDR/flk-1 (58 59 60 61 62) , whose activation, in contrast to c-Met, is characterized by increased endothelial permeability. The ability of c-Met to orchestrate enhanced barrier function utilizing similar signal transduction pathways that are employed by barrier-disrupting receptors is presently not well understood. A potential explanation for this differential response is the spatially restricted activation of signal transduction pathways resulting from receptor-specific signaling complexes that are assembled by particular scaffolding proteins. Recently, c-Met has been shown to form a protein complex with E-cadherin in tumor cells (57 , 63) , suggesting that the effects of HGF on cell-cell adhesion may involve the regional activation of effector molecules in the zonula adherens. Whether c-Met can form a complex with VE-cadherin is not presently known. One potential scaffolding protein candidate that binds c-Met to orchestrate endothelial-specific effects is Gab1, which belongs to the insulin receptor substrate-1-like (IRS-1) family of adapter molecules and binds to many of the signaling proteins that interact with the multisubstrate binding domain of the HGF receptor (64 , 65) . As Gab1 does not interact with the VEGF receptor, this docking protein is a plausible molecule for explaining the unique physiological effects of the activated c-Met receptor, a question that is currently under study.

In summary, we demonstrate here for the first time that HGF, a protein angiogenic factor, enhances EC barrier function as measured by increases in TER. Our studies define increases in EC cortical actin modulated by multiple complex signaling pathways, which include PI-3'-kinase, p38 MAPK, and PKC activation, with linkage to barrier-regulatory adherens junction proteins. One potential and common target for each of these signaling pathways is GSK-3ß, whose phosphorylation status regulates cell-cell adhesion through the modulation of the association of ß-catenin with cadherin (38) . HGF stimulated the phosphorylation of GSK-3ß, which was significantly attenuated by inhibition of PI-3'-kinase, MEK, p38 MAPK, and membrane-associated PKC activities. The relative effectiveness of the inhibition of these kinases on GSK-3ß phosphorylation induced by HGF was strongly correlated with the reduction of HGF-induced cortical actin rearrangement and barrier enhancement. Taken together, our data indicate that HGF stabilization of monolayer integrity is critically dependent on a complex kinase cascade, converging at GSK-3ß, to increase the availability of ß-catenin, thereby enhancing endothelial junctional integrity. HGF-stimulated EC barrier protection is consistent with an increasing literature of the critical role of barrier regulation by potent angiogenic factors.


   ACKNOWLEDGMENTS
 
This work was supported by grants HL 58064, HL 50533, HL 69340, and HL 03666 from the National Heart, Lung, and Blood Institute. The authors wish to extend profound gratitude to Lakshmi Natarajan, Nicholas Shank, and Steve Durbin for their excellent technical assistance, and to Ellen G. Reather for expert manuscript preparation.

Received for publication November 8, 2001. Revision received March 20, 2002.
   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Stoker, M., Gherardi, E., Perryman, M., Gray, J. (1987) Scatter factor is a fibroblast-derived modulator of epithelial cell mobility. Nature (London) 327,239-242[CrossRef][Medline]
  2. Rosen, E. M., Goldberg, I. D. (1995) Scatter factor and angiogenesis. Adv. Cancer Res. 67,257-279[Medline]
  3. Matsumoto, K., Nakamura, T. (1996) Emerging multipotent aspects of hepatocyte growth factor. J. Biochem. 119,591-600[Abstract/Free Full Text]
  4. Zhang, L., Himi, T., Morita, I., Murota, S. (2000) Hepatocyte growth factor protects cultured rat cerebellar granule neurons from apoptosis via the phosphatidylinositol-3 kinase/Akt pathway. J. Neurosci. Res. 59,489-496[CrossRef][Medline]
  5. Bussolino, F., Di Renzo, M. F., Ziche, M., Bocchietto, E., Olivero, M., Naldini, L., Gaudino, G., Tamagnone, L., Coffer, A., Comoglio, P. M. (1992) Hepatocyte growth factor is a potent angiogenic factor which stimulates endothelial cell motility and growth. J. Cell Biol. 119,629-641[Abstract/Free Full Text]
  6. Grant, D. S., Kleinman, H. K., Goldberg, I. D., Bhargava, M. M., Nickoloff, B. J., Kinsella, J. L., Polverini, P., Rosen, E. M. (1993) Scatter factor induces blood vessel formation in vivo. Proc. Natl. Acad. Sci. USA 90,1937-1941[Abstract/Free Full Text]
  7. Bottaro, D. P., Rubin, J. S., Faletto, D. L., Chan, A. M., Kmiecik, T. E., Van de Woude, G. F., Aaronson, S. A. (1991) Identification of the hepatocyte growth factor receptor as the c-met proto-oncogene product. Science 251,802-804[Abstract/Free Full Text]
  8. Nguyen, L., Holgado-Madruga, M., Maroun, C., Fixman, E. D., Kamikura, D., Fournier, T., Charest, A., Tremblay, M. L., Wong, A. J., Park, M. (1997) Association of the multi-substrate docking protein Gab1 with the hepatocyte growth factor receptor requires a function Grb2 binding site involving tyrosine 1356. J. Biol. Chem. 272,20811-20819[Abstract/Free Full Text]
  9. Ponzetto, C., Bardelli, A., Zhen, Z., Maina, F., Dalla, Z. P., Giordano, S., Graziani, A., Panayotou, G., Comoglio, P. M. (1994) A multifunctional docking site mediates signaling and transformation by the hepatocyte growth factor/scatter factor receptor family. Cell 77,261-271[CrossRef][Medline]
  10. Schaeper, U., Gehring, N. H., Fuchs, K. P., Sachs, M., Kempkes, B., Birchmeier, W.C (2000) Coupling of Gab1 to c-Met, Grb2, and Shp2 mediates biological responses. J. Cell Biol. 149,1419-1432[Abstract/Free Full Text]
  11. Holgado-Madruga, M., Moscatello, D. K., Emlet, D. R., Dieterich, R., Wong, A. J. (1997) Grb2-associated binder-1 mediates phosphatidylinositol 3-kinase activation and the promotion of cell survival by nerve growth factor. Proc. Natl. Acad. Sci. USA 94,12419-12424[Abstract/Free Full Text]
  12. Garcia, J. G. N., Liu, F., Verin, A. D., Birukova, A., Dechert, M. A., Gerthoffer, W. T., Bamburg, J. R., English, D. (2001) Sphingosine 1-phosphate promotes endothelial cell barrier integrity by Edg-dependent cytoskeletal rearrangement. J. Clin. Invest. 108,689-711[CrossRef][Medline]
  13. Thurston, G., Suri, C., Smith, K., McClain, J., Sato, T. N., Yancopoulos, G. D., McDonald, D. M. (1999) Leakage-resistant blood vessels in mice transgenically over-expressing angiopoietin-1. Science 286,2511-2514[Abstract/Free Full Text]
  14. Thurston, G., Rudge, J. S., Ioffe, E., Zhou, H., Ross, L., Cross, S. D., Glazer, N., Holash, J., McDonald, D. M., Yancopoulos, G. D. (2000) Angiopoietin-1 protects the adult vasculature against plasma leakage. Nat. Med. 6,460-463[CrossRef][Medline]
  15. Connolly, D. T., Heuvelman, D. M., Nelson, R., Olander, J. V., Eppley, B. L., Delfino, J. J., Siegel, N. R., Leimgruber, R. M., Feder, J. (1989) Tumor vascular permeability factor stimulates endothelial cell growth and angiogenesis. J. Clin. Invest. 84,1470-1478[Medline]
  16. Becker, P. M., Verin, A. D., Liu, F., Booth, M. A., Birukova, A. A., Garcia, J. G. N. (2001) Differential dose-response and signaling in response to vascular endothelial growth factor in pulmonary vascular endothelial cells. Am. J. Physiol. 281,L1500-L1511[Abstract/Free Full Text]
  17. Dudek, S. M., Garcia, J. G. N. (2001) Cytoskeletal regulation of pulmonary vascular permeability. J. Cell. Physiol. 91,1487-1500
  18. Carmeliet, P. (2000) Mechanisms of angiogenesis and arteriogenesis. Nat. Med. 6,389-395[CrossRef][Medline]
  19. Liu, Y., Wada, R., Yamashita, T., Mi, Y., Deng, C. X., Hobston, J. P., Rosenfeldt, H.M., Nava, V. E., Chae, S. S., Lee, M. J., Liu, C. H., Hla, T., Spiegel, S., Proia, R. L. (2000) Edg-1, the G protein-coupled receptor for sphingosine-1-phosphate, is essential for vascular maturation. J. Clin. Invest. 106,951-961[Medline]
  20. Stahl, S. J., Wingfield, P. T., Kaufman, J. D., Pannell, L. K., Cioce, V., Sakata, J., Taylor, W. G., Rubin, J. S., Bottaro, D. P. (1997) Functional and biophysical characterization of recombinant human hepatocyte growth factor isoforms produced in Escherichia coli. Biochem. J. 326,763-772
  21. Garcia, J. G. N., Davis, H. W., Patterson, C. E. (1995) Regulation of endothelial cell gap formation and barrier dysfunction. Role of myosin light chain phosphorylation. J. Cell. Physiol. 163,510-522[CrossRef][Medline]
  22. Liu, F., Verin, A. D., Wang, P., Day, R., Wersto, R. P., Chrest, F. J., English, D. K., Garcia, J. G. N. (2001) Differential regulation of sphingosine 1-phosphate- and VEGF-induced endothelial cell chemotaxis. Am. J. Respir. Cell Mol. Biol. 24,711-719[Abstract/Free Full Text]
  23. Garcia, J. G. N., Schaphorst, K. L., Patterson, C. E., Shi, S., Verin, A. D., Natarajan, V. (2000) Diperoxovanadate alters endothelial cell focal contacts and increases permeability: role of tyrosine phosphorylation. J. Cell. Physiol. 89,2333-2343
  24. Giaiver, I., Keese, C. R. (1993) A morphological biosensor for mammalian cells. Nature (London) 366,591-592[CrossRef][Medline]
  25. Tiruppathi, C., Malik, A. B., Del Vecchio, P. J., Keese, C. R., Giaever, I. (1992) Electrical method for detection of endothelial cell shape change in real time: assessment of endothelial barrier function. Proc. Natl. Acad. Sci. USA 89,7919-7923[Abstract/Free Full Text]
  26. Borbiev, T., Verin, A. D., Shi, S., Liu, F., Garcia, J. G. N. (2001) Regulation of endothelial cell barrier function by calcium/calmodulin-dependent protein kinase II. Am. J. Physiol. 280,L983-L990[Abstract/Free Full Text]
  27. Hartmann, G., Naldini, L., Weidner, K. M., Sachs, M., Vigna, E., Comoglio, P. M., Birchmeier, W. (1992) A functional domain in the heavy chain of scatter factor/hepatocyte growth factor binds the c-Met receptor and induces cell dissociation but not mitogenesis. Proc. Natl. Acad. Sci. USA 89,11574-11578[Abstract/Free Full Text]
  28. Guerin, C., Luddy, C., Abounader, R., Lal, B., Laterra, J. (2000) Glioma inhibition by HGF/NK2, an antagonist of scatter factor/hepatocyte growth factor. Biochem. Biophys. Res. Commun. 273,287-293[CrossRef][Medline]
  29. Chan, A. M., Rubin, J. S., Bottaro, D. P., Hirschfield, D. W., Chedid, M., Aaronson, S. A. (1991) Identification of a competitive HGF antagonist encoded by an alternative transcript. Science 254,1382-1385[Abstract/Free Full Text]
  30. Day, R. M., Cioce, V., Breckenridge, D., Castagnino, P., Bottaro, D. P. (1999) Differential signaling by alternative HGF isoforms through c-Met: activation of both MAP kianse and PI 3-kinase pathways is insufficient for mitogenesis. Oncogene 18,3399-3406[CrossRef][Medline]
  31. Verin, A. D., Liu, F., Bogatcheva, N., Borbiev, T., Hershenson, M. B., Wang, P., Garcia, J. G. N. (2000) Role of Ras-dependent ERK activation in phorbol ester-induced endothelial cell barrier dysfunction. Am. J. Physiol. 279,L360-L370
  32. Liang, C. C., Chen, H. C. (2001) Sustained activation of extracellular signal-regulated kinase stimulated by hepatocyte growth factor leads to integrin alpha 2 expression that is involved in cell scattering. J. Biol. Chem. 276,21146-21152[Abstract/Free Full Text]
  33. Harrington, E. O., Loffler, J., Nelson, P. R., Kent, K. C., Simons, M., Ware, J. A. (1997) Enhancement of migration by protein kinase C{alpha} and inhibition of proliferation and cell cycle proression by protein kinase Cdelta in capillary endothelial cells. J. Biol. Chem. 272,7390-7397[Abstract/Free Full Text]
  34. Schaphorst, K. L., Pavalko, F., Patterson, C. E., Garcia, J. G. N. (1997) Thrombin-mediated barrier dysfunction in endothelium. Role of adhesion protein phosphorylation. Am. J. Respir. Cell Mol. Biol. 17,441-455
  35. Machide, M., Kamitori, K., Nakamura, Y., Kohsaka, S. (1998) Selective activation of phospholipase C {gamma}1 and protein kinase C subspecies in intracellular signaling by hepatocyte growth factor/scatter factor in primary cultured rat neocortical cells. J. Neurochem. 71,592-602[Medline]
  36. Dejana, E., Valiron, O., Navarro, P., Lampugnani, M. G. (1997) Intercellular junctions in the endothelium and the control of vascular permeability. Ann. N. Y. Acad. Sci. 811,36-43[Free Full Text]
  37. Hinck, L., Nelson, W. J., Papkoff, J. (1994) Wnt-1 modulates cell-cell adhesion in mammalian cells by stabilizing beta-catenin binding to the cell adhesion protein cadherin. J. Cell Biol. 124,729-741[Abstract/Free Full Text]
  38. Papkoff, J., Aikawa, M. (1998) Wnt-1 and HGF regulate GSK3ß activity and ß-catenin signaling in mammary epithelial cells. Biochem. Biophys. Res. Commun. 247,851-858[CrossRef][Medline]
  39. Papkoff, J. (1997) Regulation of complexed and free catenin pools by distinct mechanisms: differential effects of Wnt-1 and v-Src. J. Biol. Chem. 272,4536-4543[Abstract/Free Full Text]
  40. Cross, D. A., Alessi, D. R., Cohen, P., Andjelkovich, M., Hemmings, B. A. (1995) Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature (London) 378,785-789[CrossRef]<