HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 48, 45838-45846, November 29, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/48/45838    most recent M209213200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Dormond, O. Articles by Rüegg, C. Search for Related Content PubMed PubMed Citation Articles by Dormond, O. Articles by Rüegg, C. Social Bookmarking                      What's this?

Prostaglandin E₂ Promotes Integrin _V3-dependent Endothelial Cell Adhesion, Rac-activation, and Spreading through cAMP/PKA-dependent Signaling^*

Olivier Dormond, Manuela Bezzi, Agnese Mariotti, and Curzio Rüegg^§

From the Centre Pluridisciplinaire d'Oncologie (CePO), University of Lausanne Medical School, CH-1011 Lausanne, Switzerland

Received for publication, September 9, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We have recently reported that the inhibition of endothelial cell COX-2 by non-steroidal anti-inflammatory drugs suppresses _V3- (but not ₅₁-) dependent Rac activation, endothelial cell spreading, migration, and angiogenesis (Dormond, O., Foletti, A., Paroz, C., and Ruegg, C. (2001) Nat. Med. 7, 1041-1047). Here we investigated the role of the COX-2 metabolites PGE₂ and TXA2 in regulating human umbilical vein endothelial cell (HUVEC) adhesion and spreading. We report that PGE₂ accelerated _V3-mediated HUVEC adhesion and promoted Rac activation and cell spreading, whereas the TXA2 agonist U46619 retarded adhesion and inhibited spreading. We show that the cAMP level and the cAMP-regulated protein kinase A (PKA) activity are critical mediators of these PGE₂ effects. _V3-mediated adhesion induced a transient COX-2-dependent rise in cAMP levels, whereas the cell-permeable cAMP analogue 8-brcAMP accelerated adhesion, promoted Rac activation, and cell spreading in the presence of the COX-2 inhibitor NS-398. Pharmacological inhibition of PKA completely blocked _V3-mediated adhesion. A constitutively active Rac mutant (L61Rac) rescued _V3-dependent spreading in the presence of NS398 or U46691, but did not accelerate adhesion, whereas a dominant negative Rac mutant (N17Rac) suppressed spreading without affecting adhesion. ₅₁-mediated HUVEC adhesion, Rac activation, and spreading were not affected by PGE₂, U46691, 8-brcAMP, or the inhibition of PKA. In conclusion, these results demonstrate that PGE₂ accelerates _V3-mediated endothelial cell adhesion through cAMP-dependent PKA activation and induces _V3-dependent spreading via cAMP- and PKA-dependent Rac activation and may contribute to the further understanding of the regulation of vascular integrins _V3 by COX-2/PGE₂ during tumor angiogenesis and inflammation.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Tumor angiogenesis, i.e. the formation of new blood vessels in response to angiogenic stimuli, promotes tumor progression by stimulating tumor cell survival, tumor invasion, and metastasis formation (1). Many molecules involved in mediating or regulating angiogenesis have been identified (2). They include growth factors (i.e. vascular endothelial growth factors, VEGF)¹ and their cell surface receptors, matrix-degrading enzymes (e.g. matrix metalloproteinases), vascular remodeling ligands, and receptors (i.e. angiopoietins and Tie receptors) and adhesion receptors of the integrin and cadherin families. Integrins are the main receptors for extracellular matrix proteins and consist of two non-covalently associated  and  subunits (3). Integrin ligand binding affinity and adhesion-promoting activity are regulated by intracellular events ("inside out" signaling) (4). Upon ligand binding, integrins rapidly cluster and recruit structural (e.g. -actinin, talin, vinculin) and signaling (e.g. focal adhesion kinase, paxillin, c-Src) proteins to form characteristic structures called focal contacts or focal adhesions (5). Integrins and focal adhesions propagate tensional forces between the extracellular matrix and the cytoskeleton necessary to stabilize cell adhesion and initiate signaling events essential to cell survival, proliferation and differentiation ("outside in" signaling) (6). Integrin _V3 is highly expressed in angiogenic endothelial cells but not, or to a much lower extent, in quiescent endothelial cells (7-9). Several studies have demonstrated that _V3 antagonists effectively inhibit angiogenesis, including tumor angiogenesis. An anti-_V3 function-blocking mAb or an antagonistic RGD-based cyclic peptide suppressed cornea vascularization (10), retinal neovascularization (11), and tumor angiogenesis (7, 13). Tumstatin, an endogenous degradation fragment of collagen IV, suppresses tumor angiogenesis by interacting with _V3 and inhibiting protein synthesis in endothelial cells (14). Furthermore, disruption of tumor vessels by high doses of tumor necrosis factor and interferon  is associated with the inhibition of the _V3 function in endothelial cells (15).

Non-steroidal anti-inflammatory drugs (NSAIDs) are widely used therapeutics for the treatment of pain and inflammation. NSAIDs act by inhibiting cyclooxygenase (COX) activity and the synthesis of prostaglandins and thromboxans (16). There are two known COX isoforms: 1) COX-1, which is ubiquitously expressed and contributes to tissue homeostasis, and 2) COX-2, which is expressed in activated leukocytes and cancer cells and promotes inflammation and cancer progression (17). Prolonged intake of NSAIDs, including COX-2 inhibitors, significantly decreases the risk of developing colon cancer and suppresses the progression of pre-malignant lesions (polyps) (18-20). Moreover, NSAIDs suppress the progression of established experimental tumors in mice (21, 22). Recent reports indicate that the anti-tumor activity of NSAID involves the inhibition of tumor angiogenesis (23, 24). COX-2 inhibitors decrease VEGF production in fibroblasts and tumor cells and prevent VEGF-induced MAPK activation in endothelial cells (23, 25), block _V3-mediated endothelial cell spreading and migration in vitro, and suppress fibroblast growth factor 2-induced angiogenesis in vivo (26). This latter effect was not associated with any detectable changes in either integrin cell surface expression or integrin affinity. We identified suppression of _V3-dependent activation of the small GTPases Cdc42 and Rac as the mechanism by which NSAIDs suppress spreading. Exogenous administration of PGE₂ or PGI₂, but not TXA2, rescued this NSAID effect, thus demonstrating that prostaglandins are critical regulators of _V3-mediated endothelial cell spreading and migration.

Here, we extend these observations by reporting that prostaglandin E₂ accelerated _V3-mediated endothelial cell adhesion through the cAMP-dependent activation of protein kinase A (PKA) and induced spreading via cAMP- and PKA-dependent activation of Rac. In contrast, ₅₁-mediated endothelial cell adhesion was not regulated by PGE_2, intracellular cAMP levels, or PKA activity.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Reagents and Antibodies-- Bovine gelatin, bovine plasma fibronectin, human plasma vitronectin, leupeptin, aprotinin, and phenylmethylsulfonyl fluoride (PMSF) were purchased from Sigma. Butaprost, PGE₁ alcohol, and sulprostone were purchased from Cayman Chemical (Ann Arbor, MI). NS-398, PGE₂, 8-brcAMP, forskolin, and H-89 were obtained from Biomol (Plymouth Meeting, PA). U46619 was from Calbiochem. The anti-Rac mAb clone 102 was from BD Biosciences. The DNA molecular weight marker was from Roche Molecular Biochemicals.

Cell Culture and Electroporation-- Human umbilical vein endothelial cells (HUVEC) were prepared and cultured as previously described (15) except for the use of M199 (Invitrogen) as a basal medium. For electroporation, subconfluent HUVEC were collected and incubated on ice for 5 min with 25 µg of L61Rac- or N17Rac-encoding plasmids or empty plasmid and 5 µg of pEGFP-C1 plasmid (Clontech) in M199 medium without fetal calf serum and electroporated with a Gene Pulser (Bio-Rad). HUVEC were resuspended in complete medium and cultured for 48 h before use in the experiments. Electroporation efficiency (routinely ~80%) was assessed by the analysis of enhanced green fluorescent protein fluorescence by flow cytometry.

Cell Adhesion Assay and Spreading Assays-- Maxisorp II Nunc enzyme-linked immunosorbent assay plates (Nunc, Roskilde, Denmark) were coated with fibronectin (5 µg/ml), gelatin (0.5%), or vitronectin (0.5 µg/ml) in PBS overnight at 4 °C, and assays were done as described previously (15). Briefly, HUVEC were resuspended in a serum-free M199 medium and plated at 3 × 10⁴ cells/well and incubated at 37 °C. At given times, unattached cells were removed by rinsing the wells with warm PBS. Attached cells were fixed in 2% paraformaldehyde (Fluka Chemie, Buchs, Switzerland), stained with 0.5% crystal violet (Sigma), and quantified by an optical density reading at 620 nm (Packard Spectra Count). Results are given as optical density values and represent the mean of duplicate wells ± S.D. of specific adhesion (equal to the adhesion on an extracellular matrix protein minus the adhesion on bovine serum albumin). If not stated otherwise, pharmacological agents were added at the time of plating and used at the following concentrations: NS-398, 100 µM; PGE₂, 100 ng/ml; 8-brcAMP, 1 mM; H-89, 5 µM; U46619, 50 µM; butaprost, 20 µM; PGE₁ alcohol, 10 ng/ml; sulprostone, 25 µM; and forskolin, 10 µM. For spreading determination, the percentage of spread cells was counted in three representative high power fields at different times after plating. Non-spread cells were defined as small round cells with little or no membrane protrusions, whereas spread cells were defined as large cells with extensive visible lamellipodia (26). Results represent the percentage of spread cells in three high power fields ± S.D.

GTPase Assays-- HUVEC were plated on gelatin (0.5%)- or fibronectin (5 µg/ml)-coated wells (Evergreen Scientific, Los Angeles, CA) in M199 medium with 1% fetal calf serum. After 30 min (or as indicated), cells were washed once with ice-cold PBS and immediately lysed in buffer containing 1% Nonidet P-40, 50 mM Tris, pH 7.4, 150 mM NaCl, 10% glycerol, 1 mM PMSF, 1 µg/ml aprotinin, and 1 µg/ml leupeptin. Cleared extracts were mixed with 20 µg of GST·PAK in the presence of glutathione-agarose beads (Sigma). After 1 h incubation at 4 °C, beads were pelleted by centrifugation and washed three times in lysis buffer, and the proteins were eluted in SDS-PAGE sample buffer and analyzed by Western blotting using a monoclonal antibody to Rac. Total Rac was determined in cell lysate. The ECL system was used for detection (Amersham Biosciences).

Determination of Intracellular cAMP Concentrations-- HUVEC were plated on fibronectin or gelatin-coated wells (Evergreen Scientific) for 45 min. HUVEC were washed twice in PBS, and cellular cAMP was extracted with 0.1 M HCl and quantified using an enzyme immunoassay (Biomol) according to the manufacturer's protocol. Values were normalized to the protein concentration using a Bio-Rad protein assay and expressed as fmol/µg protein.

Protein Kinase A Assay-- HUVEC were plated in a serum-free medium in gelatin or fibronectin-coated wells (Evergreen Scientific). At the indicated time, cells were washed once in cold PBS and lysed with cold hypotonic extraction buffer (20 mM Tris, pH 7.5, 5 mM EDTA, 1 mM PMSF, 10 µg/ml aprotinin). PKA activity was determined by the incorporation of phosphate in Leu-Arg-Arg-Ser-Leu-Gly (Kemptide) using the non radioactive Peptag system (Promega, Madison, WI). PKA activity was normalized to protein concentration and expressed as picomoles of incorporated phosphate per minute per microgram of protein.

Reverse Transcription Polymerase Chain Reaction-- Total RNA was prepared from HUVEC using the RNeasy system (Qiagen, Basel, Switzerland). 2 µg of total RNA were reversed transcribed (Superscript II, Invitrogen), and cDNA was subjected to PCR amplifications using primer pairs specific for E-prostanoid (EP) subtypes EP₁, EP₂, EP₃, and EP₄ cDNAs as described by Sheng et al. (27). PCR primer sequences were as follows: EP₁F, 5'-ACCGACCTGGCGGGCCACCTGA-3'; EP₁R, 5'-CGCTGAGCGTGTTGCACACCAG-3'; EP₂F, 5'-TCCAATGACTCCCAGTCTGAGG-3'; EP₂R, 5'-TGCATAGATGACAGGCAGCACG-3'; EP₃F, 5'-GATCACCATGCTGCTCACTG-3'; EP₃R, 5'-AGTTATGCGAAGAGCTAGTCC-3'; EP₄F: 5'-GGGCTGGCTGTCACCGACCTG-3'; and EP₄R, 5'-GGTGCGGCGCATGAACTGGCG-3' (Microsynth, Balgach, Switzerland). Amplification conditions were 40 cycles of 30 s at 94 °C, 1 min at 62 °C and 1 min at 72 °C.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

PGE₂ Accelerates _V3-mediated Cell Spreading and Adhesion-- We have previously reported that NSAIDs suppressed _V3-dependent endothelial cell spreading and migration and that this effect was prevented by exogenous addition of prostaglandins (i.e. PGE₂ or PGI₂). These results identified COX-2 and derived prostaglandins as critical regulators of vascular integrin _V3 function (26). To investigate the role of prostaglandins in the regulation of _V3 function in greater detail, we first characterized the effects of PGE₂ and TXA2, two prostanoids implicated in the modulation of angiogenesis (28, 29), on the integrin-mediated adhesion and spreading kinetics of human umbilical vein endothelial cells. HUVEC attach on gelatin and vitronectin through integrin _V3, whereas they predominantly use integrin ₅₁ to attach to fibronectin (15). On vitronectin and gelatin, half-maximal and maximal adhesion and spreading were observed at 30-45 and 60 min after plating, respectively (Figs. 1A and 2, top and middle panels). On fibronectin, adhesion and spreading proceeded with a faster kinetics: half-maximal and maximal levels were observed at 15 and 30 min after plating, respectively (Figs. 1A and 2, bottom panels). The addition of PGE₂ during adhesion on vitronectin or gelatin accelerated HUVEC cell attachment in a dose-dependent manner to reach a kinetics similar to adhesion on fibronectin (Fig. 1A, top and middle panels). Sixty minutes after plating, maximal adhesion was observed on both substrates regardless of the presence or absence of PGE₂. The acceleration of HUVEC adhesion on gelatin and vitronectin induced by PGE₂ was paralleled by an acceleration in the spreading kinetics (Fig. 2). HUVEC spreading on fibronectin was not further accelerated by PGE₂. The TXA2 analogue U46619 caused a mild but consistent retardation of HUVEC adhesion and the inhibition of spreading on vitronectin and gelatin, resulting in a ~30% suppression of cell adhesion and an 80% inhibition of spreading at 60 min at a dose of 50 µM (Figs. 1B and 2). U46619 had no effect on HUVEC adhesion or spreading to fibronectin (Figs. 1B and 2).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   PGE2 accelerates and U46691 delays V3-mediated cell adhesion. HUVEC were plated on vitronectin, gelatin or fibronectin in the absence or presence of PGE2 (A) or U46619 (B) at the indicated concentrations. Attached cells were fixed at the indicated times and revealed by crystal violet staining. Results are given as optical density (O.D.) values and represent the mean of duplicate determination ± S.D. (n = 3).

View larger version (12K): [in this window] [in a new window]   Fig. 2.   PGE2 promotes and U46691 suppresses V3-mediated cell spreading. HUVEC were plated on vitronectin, gelatin, or fibronectin in the absence or presence of PGE2 (100 ng/ml) or U46619 (50 µM), as indicated. Attached cells were fixed at 15, 30, 45, and 60 min after plating and stained by crystal violet. Results represent the percentage of spread cells over total adherent cells in three high power fields ± S.D. (n = 2).

From these results we concluded that PGE₂ accelerated the initial _V3-dependent HUVEC adhesion and spreading kinetics on gelatin and vitronectin, whereas U46619 delayed adhesion and suppressed spreading. ₅₁-mediated HUVEC adhesion and spreading on fibronectin were not affected.

HUVEC Express Functional PGE₂ receptors EP₂ and EP₄-- PGE₂ binds to and activates the following four different E-prostanoid receptor subtypes: EP₁, EP₂, EP₃, and EP₄ (30). To determine which EP receptor was involved in mediating the PGE₂ effect, we first determined the EP receptor subtype mRNA expression by reverse transcription PCR. Amplification products for EP₂ and EP₄, but not EP₁ and EP₃, were obtained (Fig. 3A). To demonstrate the functionality of EP₂ and EP₄ receptors, we performed adhesion assays in the absence or presence of butaprost, a selective EP₂ receptor agonist, and PGE₁ alcohol, a selective EP₄ receptor agonist. Both agonists accelerated HUVEC adhesion to gelatin with similar kinetics as observed for PGE₂ (Fig. 3B), although they had no effect on HUVEC adhesion to fibronectin (data not shown). The EP₃ agonist sulprostone had no effect (data not shown). Taken together, these results identify EP₂ and EP₄ prostanoid receptors as functional PGE₂ receptors on HUVEC.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   HUVEC express functional PGE2 receptors EP2 and EP4. A, expression of EP1, EP2, EP3, and EP4 receptor mRNA in HUVEC was analyzed by reverse transcription PCR. The amplified products were visualized by agarose gel electrophoresis and ethidium bromide staining. M, DNA molecular weight marker. B, HUVEC were plated on gelatin in the absence or presence of 20 µM butaprost, an EP2 agonist, 10 ng/ml PGE1 alcohol, an EP4 agonist, and 100 ng/ml PGE2. Attached cells were fixed at 0, 15, 30, 45, and 60 min after plating and revealed by crystal violet staining. Results are given as optical density values (O.D.) and represent the mean of duplicate determination ± S.D. (n = 3).

_V3-mediated Adhesion Results in a COX-2-dependent Increase in cAMP Levels-- PGE₂ and TXA₂ exert many of their biological effects through G-protein-dependent adenylcyclase (AC) (31). To test whether the PGE₂ and U46619 effects on _V3-mediated HUVEC adhesion involved modulation of cAMP levels, we measured cAMP concentrations in HUVEC in response to adhesion to gelatin or fibronectin. Adhesion on gelatin caused a transient 2-fold rise in cAMP levels (Fig. 4A, left panel), whereas adhesion on fibronectin caused a transient, 4-fold rise that was more prolonged compared with the one on gelatin (Fig. 4A, right panel). Peak cAMP concentrations were reached 15 min after plating on both substrates and returned to pre-adhesion levels within 30 min on gelatin and within 60 min on fibronectin. Adhesion in the presence of U46619 completely prevented the rise in cAMP level on gelatin, but only marginally suppressed the cAMP increase on fibronectin (Fig. 4A). HUVEC adhesion on gelatin or fibronectin in the presence of exogenous PGE₂ resulted in identical maximal cAMP levels and kinetics (Fig. 4A). The increase in cAMP levels observed in response to HUVEC adhesion to gelatin was completely abolished by NS-398, and this effect was reversed by the addition of exogenous PGE₂, whereas the rise in cAMP induced by adhesion to fibronectin was insensitive to NS-398 treatment (Fig. 4B). To collect direct evidence for a role of cAMP in accelerating _V3-dependent adhesion and promoting spreading, we plated HUVEC on gelatin in the absence or presence of the cell-permeable cAMP analog 8-brcAMP and NS-398 or U46619. 8-brcAMP accelerated HUVEC adhesion, and this effect was insensitive to NS-398 or U46619 (Fig. 4C). NS-398 slightly delayed the initial kinetics of attachment and, after 1 h of incubation, control and NS-398-treated HUVEC attached to identical extents. The addition of 8-brcAMP induced strong HUVEC spreading in the presence of NS-398 or U46619 (Fig. 4D). Similar results were obtained by elevating cAMP levels with forskolin (data not shown).

View larger version (35K): [in this window] [in a new window]   Fig. 4.   Cell adhesion induces a transient cAMP rise and 8-brcAMP promotes adhesion and spreading. HUVEC were plated on gelatin or fibronectin in the absence or presence of 100 ng/ml PGE2 or 50 µM U46619 (A) or 100 µM NS-398, 100 ng/ml PGE2, or a combination thereof, as indicated (B). HUVEC were lysed before (t = 0) and at 15, 30, 45, and 60 min after plating, and the cAMP concentration in the lysates was determined using an enzyme immunoassay. Results are given as fmol/µg protein, and values represent the mean of duplicate determinations ± S.D. (n = 3). C, HUVEC were plated on gelatin in the absence or presence of 100 µM NS-398, 1 mM 8-brcAMP, 50 µM U46619, 1 mM 8-brcAMP, or combinations thereof, as indicated. Attached cells were fixed at the indicated times and revealed by crystal violet staining. Results are given as optical density (O.D.) values and represent the mean of duplicate determinations ± S.D. (n = 3). D, HUVEC plated under the same conditions as described for panel C were fixed, stained, and photographed 60 min after plating (n = 3).

Taken together, these data demonstrate that _V3-mediated HUVEC adhesion results in a transient and COX-2-dependent increase in cAMP levels, whereas ₅₁-mediated adhesion induces a robust and COX-2 independent cAMP rise. Exogenous addition of 8-brcAMP also accelerates _V3-dependent HUVEC adhesion in the presence of a COX-2 inhibitor.

Rac Mediates cAMP-induced Spreading but Not Adhesion-- The small GTP-binding protein Rac is a critical regulator of cell spreading (32). COX-2 activity and prostaglandin production are essential for _V3-dependent Rac activation in HUVEC (26). In light of the ability of 8-brcAMP to promote cell spreading in the presence of NS-398, we asked the question of whether 8-brcAMP was able to reverse the inhibition of _V3-mediated Rac activation caused by NS-398. First, we analyzed the activation kinetics of Rac in HUVEC plated on gelatin and fibronectin using a PAK pull-down assay. HUVEC adhesion on gelatin and fibronectin resulted in rapid activation of Rac with peak activities at 15 and 30 min, respectively (Fig. 5A). Next, we measured Rac activity in HUVEC at 30 min after plating on gelatin in the absence or presence of NS-398, 8-brcAMP, or a combination thereof. The addition of NS-398 during adhesion completely inhibited Rac activation, and this effect was nearly completely reversed by the concomitant addition of 8-brcAMP (Fig. 5B). The addition of 8-brcAMP during adhesion resulted in a slight increase in Rac activity compared with adhesion on gelatin alone.

View larger version (34K): [in this window] [in a new window]   Fig. 5.   8-brcAMP promotes Rac activation and Rac-dependent V3-mediated spreading. A, HUVEC were plated on gelatin or fibronectin and, after 13, 30, and 45 min, cells were lysed and total and active Rac determined. B, HUVEC were plated on gelatin in the absence or presence of 100 µM NS-398, 1 mM 8-brcAMP, and a combination thereof. After 30 min, cells were lysed, and total and active Rac determined. (n = 2). The bars under the blots give the active Rac/total Rac ratio determined from the scans of the blots. C, control HUVEC and HUVEC overexpressing constitutive active (L61Rac) or dominant negative (N17Rac) mutants were plated on gelatin in the presence of 8-brcAMP (1 mM) or U46691 (50 µM) as indicated. Attached cells were fixed at the indicated times and revealed by crystal violet staining. Results are given as optical density (O.D.) values and represent the mean of duplicate determinations ± S.D. (n = 2). D, control HUVEC and HUVEC overexpressing L61Rac or N17Rac mutants were plated on gelatin in the presence of NS-398 (100 µM), U46691 (50 µM), and 8-brcAMP (1 mM) as indicated. Attached cells were fixed 60 min after plating. Results represent the percentage of spread cells over total adherent cells in three high power fields ± S.D. (n = 2).

Next, we asked the question of whether Rac activation was involved in mediating both the acceleration of _V3-dependent HUVEC adhesion and HUVEC spreading in response to 8-brcAMP. To address these questions we electroporated HUVEC with an expression vector encoding for a constitutive active (L61Rac) or a dominant negative form (N17Rac) of Rac (26) and then tested the adhesive and spreading properties of these cells. L61Rac did not accelerate HUVEC adhesion to gelatin, and N17Rac did not delay it. Also, N17 Rac did not prevent the acceleration of HUVEC adhesion induced by 8-brcAMP, and L61Rac did not prevent the adhesion delay caused by U46691 (Fig. 5C). In contrast, L61Rac fully reversed the inhibition of HUVEC spreading caused by NS-398 and U46691, whereas N17Rac suppressed HUVEC spreading, and this effect was not reversed by 8-brcAMP (Fig. 5D).

From these results we concluded that V3-dependent HUVEC spreading in response to cAMP elevation requires Rac activation, whereas cAMP-induced acceleration of _V3-dependent HUVEC adhesion does not.

_V3-dependent HUVEC Adhesion and Rac Activation Depend on PKA-- Protein kinase A was reported to regulate _V3-dependent angiogenesis (33) and Rac-dependent migration of carcinoma cells (34). Because elevated cAMP levels promote PKA activation, we next tested whether HUVEC adhesion to gelatin or fibronectin promoted PKA activation. A basal PKA activity was observed at the time of plating, and adhesion on gelatin induced a transient increase in PKA activity (Fig. 6A). NS-398 strongly suppressed adhesion-induced PKA activation but did not inhibit basal PKA activity, whereas the pharmacological PKA inhibitor H-89 fully suppressed both basal and adhesion-induced PKA activity (Fig. 6A). PGE₂ induced a robust PKA activation even in the presence of NS-398 (Fig. 6A). HUVEC adhesion to fibronectin induced a strong increase in PKA activity, which was completely insensitive to NS-398 (Fig. 6B). To test whether PKA activity was required for _V3- and ₅₁-dependent HUVEC adhesion and spreading, we plated cells on gelatin and fibronectin in the absence or presence of H-89. H-89 strongly suppressed HUVEC adhesion to gelatin, whereas it had no effect on HUVEC adhesion on fibronectin (Fig. 6, C and D). Furthermore, the addition of 8-brcAMP did not reverse H-89-induced inhibition of cell adhesion on gelatin (Fig. 5C). In PAK pull-down assays we investigated the requirement of PKA in _V3- and ₅₁-dependent Rac activation. H-89 completely suppressed _V3-dependent Rac activation, whereas it did not affect ₅₁-mediated Rac activation (Fig. 6E). The addition of PGE₂ or 8-brcAMP did not rescue inhibition of _V3-dependent Rac activation caused by H-89 (Fig. 6E). We next tested whether constitutively active Rac could rescue the suppression of _V3-dependent HUVEC adhesion caused by H-89. The result of this experiment demonstrated that expression of L61Rac was not sufficient to rescue suppression of _V3-dependent HUVEC adhesion caused by H-89. HUVEC plated on fibronectin in the presence of H-89 spread normally (Fig. 6G), consistent with the full activation of Rac in the presence of H-89 observed during HUVEC adhesion to fibronectin (Fig. 6E, right panel).

View larger version (21K): [in this window] [in a new window]   Fig. 6.   V3-dependent HUVEC adhesion requires PKA activation. A and B, HUVEC were plated on gelatin (A) or fibronectin (B) in the presence of 100 µM NS-398 ± 100 ng/ml PGE2 or 5 µM H-89 as indicated, and PKA activity was measured in the cell lysate at the indicated times. Results are given as incorporated picomoles of PO4 per minute per microgram of protein, and values represent the mean of duplicate determinations ± S.D. (n = 2). C and D, HUVEC were plated on gelatin (C) or fibronectin (D) in the presence of the PKA inhibitor H-89 (5 µM), 1 mM 8-brcAMP, or a combination thereof, as indicated. Attached cells were fixed at the indicated times and revealed by crystal violet staining. Results are given as optical density (O.D.) values and represent the mean of duplicate determination ± S.D. (n = 3). E, HUVEC were plated on gelatin (left panel) or on fibronectin (right panel) in the absence or presence of 100 nM PGE2, 1 mM 8-brcAMP, and 5 µM H-89 as indicated. After 30 min, cells were lysed, and active and total Rac were determined. (n = 2). The bars under the blots give the active Rac/total Rac ratio determined from the scans of the blots. F, control HUVEC and HUVEC expressing L61Rac were plated on gelatin in the absence or presence of 5 µM H-89. Attached cells were fixed at the indicated times and revealed by crystal violet staining. Results are given as optical density (O.D.) values and represent the mean of duplicate determination ± S.D. (n = 3). G, HUVEC plated on fibronectin in the absence or presence of the PKA inhibitor H-89 (5 µM) were scored for spreading. Results are given as the percentage of spreading and represent the mean of triplicate determination ± S.D. (n = 3).

From these experiments, we concluded that PKA activity mediates basal and PGE₂/8-brcAMP-stimulated _V3-dependent HUVEC adhesion and Rac activation and that active Rac does not reconstitute cell adhesion in the absence of PKA activity. In contrast, PKA is not required for ₅₁-dependent cell adhesion and spreading.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

COX-2 expression in tumor cells and the tumor microenvironment promotes tumor progression, and this effect involves the induction of tumor angiogenesis. Mice lacking COX-2 have deficient VEGF expression, reduced tumor angiogenesis, and suppressed tumor growth (35). COX-2 inhibitors decrease VEGF production in fibroblasts and tumor cells and prevent mitogen-activated protein kinase activation in response to VEGF (23, 25, 35). Pharmacological inhibition of COX-2 in endothelial cells blocks V3-mediated Rac activation, resulting in reduced cell spreading and migration in vitro and suppressed angiogenesis in vivo (26). Two of the major COX-2-derived prostanoids, PGE₂ and TXA2, have been shown to promote angiogenesis (28, 29), but the mechanisms involved are only partially characterized. In this study we have investigated the effect of PGE₂ and TXA2 on _V3- and ₅₁-mediated HUVEC adhesion and spreading. Here we report the following. First, PGE₂ accelerated HUVEC adhesion, induced Rac-activation, and stimulated Rac-dependent spreading mediated by integrin _V3, whereas the TXA2 agonist U46691 delayed adhesion and inhibited spreading mediated by _V3. PGE₂ signaled to HUVEC through EP receptors 2 and 4. Second, _V3-mediated HUVEC adhesion resulted in a COX-2/PGE₂-dependent transient rise in cAMP concentration and activation of the cAMP-dependent PKA. Third, _V3-mediated HUVEC adhesion required PKA but not Rac activity, whereas _V3-mediated spreading required both PKA and Rac activities. Fourth, integrin ₅₁-dependent HUVEC adhesion and Rac activation and spreading were not regulated by COX-2/PGE₂ or TXA2 and did not depend on PKA activity. Taken together, these observations demonstrate that _V3-dependent HUVEC adhesion and Rac-dependent spreading are positively regulated by COX-2-derived PGE₂ through cAMP/PKA-dependent signaling, whereas ₅₁-mediated adhesion and spreading occur independently of this pathway.

There is increasing evidence indicating that cAMP levels can be regulated by integrin ligation and, on the other hand, that cAMP contributes to the modulation of integrin function. For example, activating anti-₁ integrin antibodies induces MCF-7 breast carcinoma cell migration by stimulating a rise in intracellular cAMP concentration (36). The mechanism by which integrin ligation leads to an increase in cAMP level is not completely known. Our results suggest the existence of two distinct and integrin-specific mechanisms. The first one involves _V3-mediated and COX-2-dependent production of PGE₂, the stimulation of EP₂/EP₄ receptors, and the activation of AC. This pathway is consistent with the requirement of COX-2 activity to induce a cAMP rise in response to _V3 ligation (Fig. 4A), cell spreading (Fig. 4D), and migration (26) with the classical signaling pathway of prostaglandins (20). The stimulation of COX-2-dependent PGE₂ production upon _V3-dependent adhesion could result from the _V3-mediated activation of phospholipase A2 (PLA2) and the production of arachidonic acid, the substrate of COX. In this respect, it has been recently reported that _V3 ligation induces membrane translocation and activation of PLA2 with the subsequent release of arachidonic acid in bovine pulmonary artery endothelial cells (37). PLA2 appears to be stimulated by integrin ligation in several cell types, and in some cases its activation has been linked to the production of arachidonic acid, the activation of PKC, and cell spreading (38, 39). The second mechanism involves ₅₁-mediated, COX-2/PGE₂-independent activation of AC. This is supported by the observation that NS-398 does not inhibit the increase in cAMP concentration observed in HUVEC plated on fibronectin (Fig. 4B). ₁ integrin ligation with the RGD cell binding sequence of fibronectin or with ₁-activating antibodies followed by mechanical stress has been reported to cause a rapid increase in intracellular cAMP levels and PKA activity in endothelial cells (40). G-protein  subunit inhibitors suppressed this effect, suggesting that integrin ligation and mechanical stress may stimulate AC through the activation of integrin-coupled heterotrimeric G-proteins (40).

cAMP modulates integrin-dependent cell adhesion and migration, but, depending on the cell type and context, it can exert stimulatory or inhibitory effects. For example, an increase in cAMP levels was shown to promote adhesion of immature thymocytes to fibronectin (41), to inhibit adhesion of mature T lymphocytes to fibronectin (42), and to suppresses leukocyte adhesion and migration in response to chemoattractants (43). Our findings indicate that, in endothelial cells, an elevation of cAMP accelerates _V3-mediated adhesion and promotes Rac-dependent spreading via activation of PKA. Recently, cAMP-dependent PKA activation was reported to induce ₁-integrin mediated MDA-435 breast carcinoma cell migration in response to growth factors through the activation of Rac and the inhibition of RhoA (34). There is also emerging evidence that PKA can positively and negatively regulate integrin-mediated cell adhesion and migration of carcinoma and sarcoma-derived cell lines, endothelial cells, and neutrophils (33, 34, 44-46). The precise mechanisms by which PKA regulates integrin-dependent events and Rac activity in particular remain, however, largely unknown.

Taken together, we propose the following working model for the regulation of _V3-mediated endothelial cell adhesion and spreading by PGE₂. Upon _V3-dependent adhesion there is a transient rise in cAMP levels and PKA activity dependent on COX-2-mediated production of PGE₂ and on signaling through EP2/EP4 receptors. Increased PKA activity accelerates _V3-dependent adhesion through a Rac-independent mechanism and stimulates _V3-dependent spreading through a Rac-dependent mechanism. In contrast, the rise in cAMP levels and PKA activity observed after ₅₁-mediated adhesion does not depend on COX-2 or PGE₂ and is not required for ₅₁-dependent HUVEC adhesion and spreading.

This latter observation suggests that the ligation of integrin ₅₁ may generate a cAMP rise and PKA activation, which can then promote _V3-dependent endothelial cell adhesion and spreading/migration. Such a "cross-talk" is consistent with a recent report demonstrating that ligation of integrin ₅₁ stimulated _V3-dependent endothelial cell migration and angiogenesis (33). Optimal migration and angiogenesis is likely to require intermediate cAMP levels and PKA activity, whereas subthreshold or excessive PKA activity would result in static adhesion. The regulation of cAMP levels may occur through modulation of cAMP generation by AC or degradation by phosphodiesterase. Indeed, growth factor-stimulated carcinoma cell migration was shown to require both cAMP-dependent PKA activity and phosphodiesterase-mediated cAMP degradation (12, 34).

In conclusion, we have demonstrated that a PGE₂-mediated rise in cAMP promotes _V3-mediated endothelial cell adhesion through the activation of PKA and induces spreading via PKA-dependent Rac activation. These results may contribute to the further understanding of the regulation of vascular integrins _V3 by COX-2/PGE₂ during tumor angiogenesis and inflammation.

	ACKNOWLEDGEMENTS

The authors thank Dr. F. J. Lejeune for continuous support, Dr. J. Collard, A. Hall and S. Klein for providing reagents, and R. Driscoll for critical reading of the manuscript.

	FOOTNOTES

^* This work was supported by grants from the Swiss National Science Foundation (31-52946.97 and 31-63752.00), the Leenaards Foundation, and the Banque Centonale Vaudoise (BCV) Foundation.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Recipient of an M.D./Ph.D. fellowship (31-51279.97) from the Swiss National Science Foundation.

^§ Recipient of a SCORE-A award (32-41611.94) from the Swiss National Science Grant. To whom all correspondence should be addressed: Laboratory of the CePO, c/o Swiss Institute for Experimental Cancer Research (ISREC), 155 Chemin des Boveresses, CH-1066 Epalinges, Switzerland. Tel.: 41-21-692-5853; Fax: 41-21-692-5872; E-mail: curzio.ruegg@isrec.unil.ch.

Published, JBC Papers in Press, September 16, 2002, DOI 10.1074/jbc.M209213200

	ABBREVIATIONS

The abbreviations used are: VEGF, vascular endothelial growth factor; AC, adenylcyclase; COX, cyclooxygenase; EP, E-prostanoid; HUVEC, human umbilical vein endothelial cells; NSAIDs, non-steroidal anti-inflammatory drugs; PAK, p21-activated kinase; PBS, phosphate-buffered saline; PGE, prostaglandin E; PGI, prostaglandin I; PKA, protein kinase A; PLA2, phospholipase A 2; PMSF, phenylmethylsulfonyl fluoride; TXA2, thromboxane A2.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				A. A. Paulucci-Holthauzen, L. A. Vergara, L. J. Bellot, D. Canton, J. D. Scott, and K. L. O'Connor Spatial Distribution of Protein Kinase A Activity during Cell Migration Is Mediated by A-kinase Anchoring Protein AKAP Lbc J. Biol. Chem., February 27, 2009; 284(9): 5956 - 5967. [Abstract] [Full Text] [PDF]

				F. Finetti, R. Solito, L. Morbidelli, A. Giachetti, M. Ziche, and S. Donnini Prostaglandin E2 Regulates Angiogenesis via Activation of Fibroblast Growth Factor Receptor-1 J. Biol. Chem., January 25, 2008; 283(4): 2139 - 2146. [Abstract] [Full Text] [PDF]

				X. Shen, T. Koga, B.-C. Park, N. SundarRaj, and B. Y. J. T. Yue Rho GTPase and cAMP/Protein Kinase A Signaling Mediates Myocilin-induced Alterations in Cultured Human Trabecular Meshwork Cells J. Biol. Chem., January 4, 2008; 283(1): 603 - 612. [Abstract] [Full Text] [PDF]

				S. J. Netherton, J. A. Sutton, L. S. Wilson, R. L. Carter, and D. H. Maurice Both Protein Kinase A and Exchange Protein Activated by cAMP Coordinate Adhesion of Human Vascular Endothelial Cells Circ. Res., October 12, 2007; 101(8): 768 - 776. [Abstract] [Full Text] [PDF]

				A. C. Azim, H. Cao, X. Gao, M. Joo, A. B. Malik, R. B. van Breemen, R. T. Sadikot, G. Park, and J. W. Christman Regulation of cyclooxygenase-2 expression by small GTPase Rac2 in bone marrow macrophages Am J Physiol Lung Cell Mol Physiol, September 1, 2007; 293(3): L668 - L673. [Abstract] [Full Text] [PDF]

				M. J. Lorenowicz, M. Fernandez-Borja, and P. L. Hordijk cAMP Signaling in Leukocyte Transendothelial Migration Arterioscler. Thromb. Vasc. Biol., May 1, 2007; 27(5): 1014 - 1022. [Abstract] [Full Text] [PDF]

				Y.-W. E. Chang, J. W. Marlin, T. W. Chance, and R. Jakobi RhoA Mediates Cyclooxygenase-2 Signaling to Disrupt the Formation of Adherens Junctions and Increase Cell Motility Cancer Res., December 15, 2006; 66(24): 11700 - 11708. [Abstract] [Full Text] [PDF]

				W. A. Sands, H. D. Woolson, G. R. Milne, C. Rutherford, and T. M. Palmer Exchange Protein Activated by Cyclic AMP (Epac)-Mediated Induction of Suppressor of Cytokine Signaling 3 (SOCS-3) in Vascular Endothelial Cells. Mol. Cell. Biol., September 1, 2006; 26(17): 6333 - 6346. [Abstract] [Full Text] [PDF]

				A. Alfranca, M. A. Iniguez, M. Fresno, and J. M. Redondo Prostanoid signal transduction and gene expression in the endothelium: Role in cardiovascular diseases Cardiovasc Res, June 1, 2006; 70(3): 446 - 456. [Abstract] [Full Text] [PDF]

				B. Hoang, L. Zhu, Y. Shi, P. Frost, H. Yan, S. Sharma, S. Sharma, L. Goodglick, S. Dubinett, and A. Lichtenstein Oncogenic RAS mutations in myeloma cells selectively induce cox-2 expression, which participates in enhanced adhesion to fibronectin and chemoresistance Blood, June 1, 2006; 107(11): 4484 - 4490. [Abstract] [Full Text] [PDF]

				D. Wang, H. Wang, J. Brown, T. Daikoku, W. Ning, Q. Shi, A. Richmond, R. Strieter, S. K. Dey, and R. N. DuBois CXCL1 induced by prostaglandin E2 promotes angiogenesis in colorectal cancer J. Exp. Med., April 17, 2006; 203(4): 941 - 951. [Abstract] [Full Text] [PDF]

				N. Fortemaison, S. Blancquaert, J. E. Dumont, C. Maenhaut, K. Aktories, P. P. Roger, and S. Dremier Differential Involvement of the Actin Cytoskeleton in Differentiation and Mitogenesis of Thyroid Cells: Inactivation of Rho Proteins Contributes to Cyclic Adenosine Monophosphate-Dependent Gene Expression but Prevents Mitogenesis Endocrinology, December 1, 2005; 146(12): 5485 - 5495. [Abstract] [Full Text] [PDF]

				U. K. Misra and S. V. Pizzo Coordinate Regulation of Forskolin-induced Cellular Proliferation in Macrophages by Protein Kinase A/cAMP-response Element-binding Protein (CREB) and Epac1-Rap1 Signaling: EFFECTS OF SILENCING CREB GENE EXPRESSION ON Akt ACTIVATION J. Biol. Chem., November 18, 2005; 280(46): 38276 - 38289. [Abstract] [Full Text] [PDF]

				A. K. Howe, L. C. Baldor, and B. P. Hogan Spatial regulation of the cAMP-dependent protein kinase during chemotactic cell migration PNAS, October 4, 2005; 102(40): 14320 - 14325. [Abstract] [Full Text] [PDF]

				R. Mayoral, A. Fernandez-Martinez, L. Bosca, and P. Martin-Sanz Prostaglandin E2 promotes migration and adhesion in hepatocellular carcinoma cells Carcinogenesis, April 1, 2005; 26(4): 753 - 761. [Abstract] [Full Text] [PDF]

				J. Symowicz, B. P. Adley, M. M.M. Woo, N. Auersperg, L. G. Hudson, and M. S. Stack Cyclooxygenase-2 Functions as a Downstream Mediator of Lysophosphatidic Acid to Promote Aggressive Behavior in Ovarian Carcinoma Cells Cancer Res., March 15, 2005; 65(6): 2234 - 2242. [Abstract] [Full Text] [PDF]

				D. Wang, F. G. Buchanan, H. Wang, S. K. Dey, and R. N. DuBois Prostaglandin E2 Enhances Intestinal Adenoma Growth via Activation of the Ras-Mitogen-Activated Protein Kinase Cascade Cancer Res., March 1, 2005; 65(5): 1822 - 1829. [Abstract] [Full Text] [PDF]

				J. Zaric and C. Ruegg Integrin-mediated Adhesion and Soluble Ligand Binding Stabilize COX-2 Protein Levels in Endothelial Cells by Inducing Expression and Preventing Degradation J. Biol. Chem., January 14, 2005; 280(2): 1077 - 1085. [Abstract] [Full Text] [PDF]

				H. Jiang, A. S. Weyrich, G. A. Zimmerman, and T. M. McIntyre Endothelial Cell Confluence Regulates Cyclooxygenase-2 and Prostaglandin E2 Production That Modulate Motility J. Biol. Chem., December 31, 2004; 279(53): 55905 - 55913. [Abstract] [Full Text] [PDF]

				J. M. Enserink, L. S. Price, T. Methi, M. Mahic, A. Sonnenberg, J. L. Bos, and K. Tasken The cAMP-Epac-Rap1 Pathway Regulates Cell Spreading and Cell Adhesion to Laminin-5 through the {alpha}3{beta}1 Integrin but Not the {alpha}6{beta}4 Integrin J. Biol. Chem., October 22, 2004; 279(43): 44889 - 44896. [Abstract] [Full Text] [PDF]

				F. Yang and D. Bleich Transcriptional Regulation of Cyclooxygenase-2 Gene in Pancreatic {beta}-Cells J. Biol. Chem., August 20, 2004; 279(34): 35403 - 35411. [Abstract] [Full Text] [PDF]

				G.A. Stouffer and S.S. Smyth Effects of Thrombin on Interactions Between {beta}3-Integrins and Extracellular Matrix in Platelets and Vascular Cells Arterioscler. Thromb. Vasc. Biol., November 1, 2003; 23(11): 1971 - 1978. [Abstract] [Full Text] [PDF]

				C. Sanchez-Moreno, M. P. Cano, B. de Ancos, L. Plaza, B. Olmedilla, F. Granado, and A. Martin High-Pressurized Orange Juice Consumption Affects Plasma Vitamin C, Antioxidative Status and Inflammatory Markers in Healthy Humans J. Nutr., July 1, 2003; 133(7): 2204 - 2209. [Abstract] [Full Text] [PDF]

				S. M. Ellerbroek, K. Wennerberg, and K. Burridge Serine Phosphorylation Negatively Regulates RhoA in Vivo J. Biol. Chem., May 23, 2003; 278(21): 19023 - 19031. [Abstract] [Full Text] [PDF]

				N. Hatae, A. Kita, S. Tanaka, Y. Sugimoto, and A. Ichikawa Induction of Adherent Activity in Mastocytoma P-815 Cells by the Cooperation of Two Prostaglandin E2 Receptor Subtypes, EP3 and EP4 J. Biol. Chem., May 9, 2003; 278(20): 17977 - 17981. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/48/45838    most recent M209213200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Dormond, O. Articles by Rüegg, C. Search for Related Content PubMed PubMed Citation Articles by Dormond, O. Articles by Rüegg, C. Social Bookmarking                      What's this?