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

J. Biol. Chem., Vol. 277, Issue 37, 34143-34149, September 13, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/37/34143    most recent M204477200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Vanhauwe, J. F. Articles by Hamm, H. E. Search for Related Content PubMed PubMed Citation Articles by Vanhauwe, J. F. Articles by Hamm, H. E. Social Bookmarking                      What's this?

Thrombin Receptors Activate G_o Proteins in Endothelial Cells to Regulate Intracellular Calcium and Cell Shape Changes^*

Jurgen F. Vanhauwe^§, Tarita O. Thomas^§, Richard D. Minshall^¶, Chinnaswamy Tiruppathi^¶, Anli Li, Annette Gilchrist, Eun-ja Yoon, Asrar B. Malik^¶, and Heidi E. Hamm^¶**

From the  Institute for Neuroscience and Department of Molecular Pharmacology and Biological Chemistry, Northwestern University, Chicago, Illinois 60611, the  Department of Pharmacology, Vanderbilt University, Nashville, Tennessee 37232, and the ^¶ Department of Pharmacology, University of Illinois at Chicago, Chicago, Illinois 60612

Received for publication, May 7, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Thrombin receptors couple to G_i/o, G_q, and G_12/13 proteins to regulate a variety of signal transduction pathways that underlie the physiological role of endothelial cells in wound healing or inflammation. Whereas the involvement of G_i, G_q, G₁₂, or G₁₃ proteins in thrombin signaling has been investigated extensively, the role of G_o proteins has largely been ignored. To determine whether G_o proteins could contribute to thrombin-mediated signaling in endothelial cells, we have developed minigenes that encode an 11-amino acid C-terminal peptide of G_o1 proteins. Previously, we have shown that use of the C-terminal minigenes can specifically block receptor activation of G protein families (1). In this study, we demonstrate that G_o proteins are present in human microvascular endothelial cells (HMECs). Moreover, we show that thrombin receptors can stimulate [³⁵S]guanosine-5'-O-(3-thio)triphosphate binding to G_o proteins when co-expressed in Sf9 membranes. The potential coupling of thrombin receptors to G_o proteins was substantiated by transfection of the G_o1 minigene into HMECs, which led to a blockade of thrombin-stimulated release of [Ca²⁺]_i from intracellular stores. Transfection of the -adrenergic kinase C terminus blocked the [Ca²⁺]_i response to the same extent as with G_o1 minigene peptide, suggesting that this G_o-mediated [Ca²⁺]_i transient was caused by G stimulation of PLC. Transfection of a G_i1/2 minigene had no effect on thrombin-stimulated [Ca²⁺]_i signaling in HMEC, suggesting that G derived from G_o but not G_i could activate PLC. The involvement of G_o proteins on events downstream from calcium signaling was further evidenced by investigating the effect of G_o1 minigenes on thrombin-stimulated stress fiber formation and endothelial barrier permeability. Both of these effects were sensitive to pertussis toxin treatment and could be blocked by transfection of G_o1 minigenes but not G_i1/2 minigenes. We conclude that the G_o proteins play a role in thrombin signaling distinct from G_i1/2 proteins, which are mediated through their G subunits and involve coupling to calcium signaling and cytoskeletal rearrangements.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Thrombin is a multifunctional serine protease that catalyzes conversion of fibrinogen to fibrin, a process that is crucial in blood coagulation (1). In addition, thrombin plays a central role in a variety of biological functions such as platelet aggregation, mitogenesis of fibroblasts, monocytic cell chemotaxis, and endothelial cell monolayer permeability (2-4). Many of its actions, including the regulation of biochemical, transcriptional, and functional responses in endothelial cells occur through activation of protease-activated receptors (PARs),¹ which belong to the superfamily of G protein-coupled receptors.

Four PARs have been cloned so far, but only PAR1, PAR3, and PAR4 can be activated by thrombin (5). The activation and signal transduction pathways of PAR1, the prototype of the PAR family, have been studied in great detail. Thrombin cleaves the N-terminal extracellular domain of PAR1 at a specific site, which unmasks a new N terminus that then serves as a tethered agonist ligand and activates the receptor by binding intramolecularly to the body of the receptor (6). Cleaved, i.e. irreversibly activated, PAR1 can couple to members of the G_i/o, G_q, and G_12/13 protein families and regulate a variety of intracellular effectors (1).

Although the role of G_o proteins has been generally believed to be confined to the brain and heart, several reports indicate that G_o proteins may serve to regulate various intracellular pathways in non-neuronal cell lines (7, 8). In addition, several new effectors have been identified that are specifically or differentially regulated by G_o proteins (versus G_i proteins) (9-11). Pertussis toxin (PTX)-mediated ADP-ribosylation of the Cys³⁵¹ residue in the C terminus of G_i and G_o proteins disables their interaction with receptors and thus prevents receptor-mediated activation of these G proteins. Many effects of thrombin are mediated through pertussis toxin-sensitive G proteins (12-18). This method, however, does not distinguish between G_i and G_o proteins, and the importance of the latter subtype could be inadequately appreciated (5).

To dissect out the contribution of G_i1/2 and G_o1 proteins in thrombin-regulated signaling pathways in HMECs, we have designed a dominant negative strategy based on minigene vectors that encode the C-terminal 11-amino acid sequence from G. Previously, we have shown that these minigenes are quite specific in such a way that G_q-based minigenes blocked only thrombin activation of G_q protein-mediated pathways (phosphatidylinositol bisphosphate hydrolysis and intracellular calcium increase) but not G_i1/2 or G_12/13 protein-mediated pathways (1, 19). The specificity of G C-terminal peptides has been shown dramatically by Gilchrist et al. (20), where one or two amino acid substitutions inhibited the ability of peptides to block receptor-mediated activation of signaling pathways.

To delineate specific functions for G_i and G_o proteins in the signaling of thrombin receptors, we have introduced these minigenes into HMECs. Our findings indicate that HMECs contain G_o proteins and that PAR1 has the potential to couple to G_i and G_o proteins when co-expressed in Sf9 cell membranes. In addition, we show that G_o1 minigenes block thrombin-stimulated release of [Ca²⁺]_i, whereas G_i1/2 minigenes do not. The involvement of G_o proteins, but not G_i proteins, was further established in pathways that are known to be downstream of calcium signaling, such as stress fiber formation and endothelial barrier permeability. Together our data demonstrate the importance of G_o proteins in the signaling of thrombin receptors in endothelial cells.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Reagents-- All of the cell culture reagents were purchased from Invitrogen. The parent pcDNA 3.1() vector was obtained from Invitrogen; pEGFP was from CLONTECH, retroviral Tet-inducible vectors pRevTRE2 and pREvTRE2-dEGFP were from CLONTECH (Palo Alto, CA). All of the restriction enzymes were procured from New England Biolabs (Beverly, MA). The highly purified -thrombin (~2000 units/mg) and PTX were obtained from Calbiochem. Alexa Fluor 568 phalloidin, DAPI, Oregon Green Bapta-1 acetoxymethylester, Pluronic F127, and the Prolong Antifade kit were purchased from Molecular Probes (Eugene, OR). Anti-G_o1/2 antibodies were from Dr. D. Manning (University of Pennsylvania, Philadelphia, PA). 3-Isobutyl-1-methylxanthine, forskolin, and isoproterenol were from Sigma. Electrodes for endothelial monolayer resistance measurements were obtained from Applied Biosciences (Troy, NY). Virions producing the rat G_i1, G_i2, G_i3, and G_o1 were obtained from Dr. S. Graber (West Virginia University, Morgantown, WV), whereas those for PAR1 were obtained from Dr. C. Chinni (University of Cambridge, Cambridge, UK). [³⁵S]GTPS was from PerkinElmer Life Sciences.

Preparation of Sf9 Membranes-- Sf9 cells were grown at 27 °C and at an ambient atmosphere in suspension in a shaking incubator and transfected as described before (21). Harvested Sf9 cells were washed with ice-cold 50 mM Tris-HCl buffer, pH 7.4, resuspended in hypotonic 10 mM Tris-HCl buffer, pH 7.4, and homogenized with 10 strokes of a Bio-Homogenizer (BioSpec Products, Inc.) at high speed. The homogenate was centrifuged at 30,000 × g for 20 min at 4 °C. The membrane pellet was resuspended in 50 mM Tris-HCl buffer, pH 7.4, containing 10% glycerol and stored in aliquots at 80 °C.

[³⁵S]GTPS Binding Experiments-- [³⁵S]GTPS binding experiments were performed as described previously (22). Briefly, 10 µg of Sf9 cell membrane protein was diluted in 50 mM Tris-HCl buffer, pH 7.4, containing 5 mM MgCl₂, 1 mM EGTA, 100 mM NaCl, 0.1 mM dithiothreitol, 10 µg/ml saponin, and 1 µM GDP and preincubated with TRAP for 15 min at room temperature in a volume of 125 µl in a 96-well plate. Then 25 µl of [³⁵S]GTPS diluted 1000-fold in assay buffer was added to the wells, and the assay mixtures were further incubated for 30 min at room temperature. The reactions were terminated by rapid filtration, after which the filters were washed four times with 200 µl of 50 mM Tris-HCl buffer, pH 7.4, containing 100 mM NaCl, 5 mM MgCl₂, and 1 mM EGTA. Filter-bound radioactivity was counted in a liquid scintillation spectrometer. Nonspecific [³⁵S]GTPS binding was measured in the presence of 100 µM GTPS and never exceeded 10% of basal binding. Basal [³⁵S]GTPS binding was estimated in the absence of TRAP.

Endothelial Cell Culture-- For our studies we used a human dermal microvascular endothelial cell line that was transformed using SV-40 (HMEC-1; obtained from Dr. E. Ades (Centers for Disease Control, Atlanta, GA). The cells were maintained in MCDB 131 medium supplemented with 5% fetal bovine serum, penicillin/streptomycin (5000 units/ml; 5000 µg/ml), hydrocortisone (500 µg/ml), epidermal growth factor (0.01 µg/ml), and L-glutamine (2 mM) in an atmosphere of 95% air, 5% CO₂ at 37 °C. The cells were seeded at 1 × 10⁵ cells/ml and subcultured after detachment with 0.05% trypsin,/0.5 mM EDTA. All of the studies utilized cell passages 18-26.

Plasmid Constructs-- cDNA minigene constructs were designed as described previously (19). The C terminus of the G protein-coupled receptor kinase 2 (ARK-ct) has been shown to be a potent and specific G inhibitor (23). The ARK-ct construct codes for residues 548-671 of the rat homolog ARK.

Retroviral minigenes were constructed as follows. The cDNA encoding the last 11 amino acids of human G subunits (G_i1/2 and G_o) or the G_i1/2 C terminus in random order (G_iR) were synthesized (Great American Gene Company) with newly engineered 5'- and 3'-ends. The 5'-end contained a BamHI site followed by the ribosome-binding consensus sequence (5'-GCCGCCACC-3'), a methionine (ATG) for translation initiation, and a glycine (GGA) to protect the ribosome-binding site during translation and the nascent peptide against proteolytic degradation. A HindIII site was synthesized at the 3'-end immediately following the translational stop codon (TGA). The DNA was brought up in sterile double distilled H₂O (stock concentration, 100 µM). Complimentary DNA was annealed in 1× NE Buffer 3 (50 mM Tris-HCl, 10 mM MgCl₂, 100 mM NaCl, 1 mM dithiothreitol; New England Biolabs) at 85 °C for 10 min and then allowed to cool slowly to room temperature. The annealed cDNA were ligated for 1 h at room temperature into a Tet-inducible retroviral vector pRevTRE2 in murine Maloney tumor virus (CLONTECH) previously cut with BamHI and HindIII. Following ligation, the samples were heated to 65 °C for 5 min to deactivate the T4 DNA ligase. The ligation reaction (1 µl) was electroporated into 50 µl of competent ARI814 cells (Bio-Rad Escherichia coli Pulsar), and the cells were immediately placed into 1 ml of superoptimal catabolite medium (Invitrogen). After 1 h at 37 °C, 100 µl was spread on LB/ampicillin plates and incubated at 37 °C for 12-16 h. To verify that insert was present, several colonies were grown overnight in LB/ampicillin, and their plasmid DNA was purified (Qiagen SpinKit). The plasmid DNA was digested with NcoI (New England Biolabs, Inc.) for 1 h at 37 °C and run on a 1.5% (3:1) agarose gel. Vector alone produced one band (6.5 kb), whereas vector with insert resulted in two bands (5.1 and 1.4 kb). DNA with the correct pattern was sequenced (Northwestern University Biotechnology Center) to confirm the appropriate sequence.

For optimal results, the retroviral vectors were packaged using the pantropic GP-293 cell line (CLONTECH) with vesicular stomatitis virus glycoprotein, an envelope glycoprotein from the vesicular stomatitis virus. As a control, we used the enhanced GFP inserted into the parental vector (pRevTRE2-dEGFP; CLONTECH). Retroviral minigenes were generated by infecting the packaging cells GP-293 with pRevTRE2 minigenes and vesicular stomatitis virus glycoprotein using Effectene reagent according to instructions from the manufacturer (CLONTECH). 12-16 h later, the medium was replaced by 5 ml of fresh medium/10-cm dish, and the virus produced by the cells was collected 2-3 days post-transfection by filtration through 0.45-µm cellulose acetate filters and stored in aliquots at 80 °C. The virus titer reached 4 × 10⁶ plaque-forming units/ml.

Transfection and Infections-- For pcDNA-based minigenes, HMECs were transiently transfected with DNA (2 µg/100-mm plate or 500 ng/well for a 6-well plate) using Effectene transfection reagent (Qiagen). To monitor the efficiency of transfection, the cells were co-transfected with pEGFP, a plasmid vector containing enhanced green fluorescent protein to monitor stress fiber formation, or DsRED (CLONTECH), a plasmid vector containing red fluorescent protein for [Ca²⁺]_i imaging. After 3 h, the medium was changed, and fresh medium was added. After 48 h, the cells co-transfected with the fluorescent proteins were replated onto coverslips and analyzed using a fluorescent microscope to determine the efficiency of transfection. Adenylyl cyclase and HMEC monolayer permeability experiments were performed using cells infected with retroviral minigenes. HMECs were infected with retroviral minigene virus (2 × 10⁶ plaque-forming units/well for 6-well plate). The expression of the minigene peptide was induced 24 h after viral infection with 2 µg/ml doxycyclin, and the experiments were performed 24 h after induction. Use of retroviral minigenes led to ~100% transfection efficiency. This was confirmed using infection of the pRevTRE2-dEGFP vector, which exhibited expression of GFP in virtually every cell (data not shown).

Western Blot Analysis-- Endothelial cell lysates were resolved by SDS-PAGE on a 10-20% separating gel under reducing conditions. For immunoblotting analysis, the proteins were transferred to polyvinylidene difluoride membranes using standard semi-dry transfer method. The membranes were blocked with 5% dry milk in phosphate-buffered saline, 0.05% Tween 20 for 1 h at room temperature. The membranes were incubated with indicated primary antibody (diluted in blocking buffer) at 4 °C overnight. Following washes, the membranes were incubated at room temperature with peroxidase-labeled secondary antibodies and detected using luminol-based chemiluminescent detection system (LumiGLO, Kirkegaard and Perry Laboratories, Gaithersburg, MD).

cAMP Assay-- HMECs were seeded onto 6-well plates at 1 × 10⁵ cells/well 24 h before transfection. The cells were transfected with retroviral minigene constructs, and 24 h before the assay, the cells were seeded into a 24-well plate. Thereafter, the cells were washed once with serum-free medium containing 1 mM 3-isobutyl-1-methylxanthine, a phosphodiesterase inhibitor, and further incubated for 20 min in 500 µl of serum-free medium containing 1 mM 3-isobutyl-1-methylxanthine. After the preincubation, 50 µl of preincubation medium supplemented with forskolin (final concentration, 10 µM) was added to each well. To detect the inhibitory effect, 100 nM thrombin or 10 µM NECA was added along with forskolin. Basal cAMP accumulation was measured in the absence of forskolin and compounds. The reactions were terminated by the addition of 100 µl of 1 N HCLO₄. The samples were frozen and thawed, and 200 µl of KOH/K₃PO₄ (0.5 M, pH 13.5) was added to neutralize the samples (final pH, 7.4). After formation of the KClO₄ precipitate (30 min at 4 °C), the plates were centrifuged (10 min at 650 × g, 4 °C). The amount of cAMP in each well was determined with a commercial ¹²⁵I-labeled cAMP radioimmunoassay kit (Biomedical Technologies Inc., Stoughton, MA).

[Ca²⁺]_i Response-- In single cell fluorescence measurements, the DsRED (CLONTECH) fluorescence reporter gene was used to confirm the transfected cells. HMECs were transfected with pcDNA-G_i1/2, pcDNA-G_o1, or pcDNA-G_iR minigene DNA and with or without ARK-ct DNA along with DsRED. After 48 h, the cells were transferred to coverslips at a low confluency in a 24-well plate and allowed to adhere for at least 2 h. The medium was aspirated, and each coverslip was incubated at 37 °C for 30 min in 0.5 ml of loading buffer (20 mM Hepes, pH 7.4, 130 mM NaCl, 5 mM KCl, 2 mM CaCl₂, 1 mM MgSO₄, 0.83 mM Na₂HPO₄, 0.17 mM NaH₂PO₄, 1 mg/ml bovine serum albumin, 25 mM mannose) containing 0.1% (v/v) Pluronic F127 and 10 µM Oregon Green Bapta-1 acetoxymethyl ester. The cells were washed twice with and incubated in Ca²⁺ buffer (10 mM Hepes, pH 7.4, 140 mM NaCl, 5 mM KCl, 0.5 mM CaCl₂, 0.55 mM MgCl₂). The coverslips were placed in the chamber that was mounted on the stage of an upright microscope. The experiment was performed at room temperature. The transfected cells were identified using a green filter by observing DsRED fluorescence. The basal conditions were established for 40 s before addition of thrombin (~70 nM). Recordings (exposure time) were made every 10 s and continued for 170 s after stimulation with thrombin. The images were quantified using the NIH Image Program.

Immunofluorescent Microscopy-- As a marker for transfected cells, the pEGFP plasmid containing the gene for enhanced green fluorescent protein was transiently co-transfected together with minigene constructs as described above. HMECs were grown on gelatin-coated coverslips, serum-starved for 24 h, washed with HBSS, and fixed with 4% paraformaldehyde. The coverslips were washed three times for 5 min in 100 mM glycine in HBSS to quench and remove the fixative followed by three washes for 10 min in HBSS. The cells were permeabilized with 0.1% Triton X-100 and washed three times for 10 min in HBSS. Thereafter, the cells were incubated for 90 min at room temperature with 200 nM Alexa Fluor 568-phallodin to visualize polymerized F-actin. The coverslips were washed three times for 10 min in HBSS and labeled with 1 µg/ml DAPI for 30 min to visualize the nucleus. The coverslips were finally washed three times for 10 min in HBSS and mounted on a drop of ProLong Antifade mounting medium (Molecular Probes). The cells were observed with a Zeiss 510 laser scanning confocal microscope (New York, NY) using 364-, 488-, and 568-nm excitation laser lines to detect DAPI (BP 385-470 nm emission), fluorescein isothiocyanate/Alexa 488 (BP505-550 emission), and rhodamine/Alexa 568 fluorescence (LP585 emission) with the ×63 1.4 NA water immersion objective. The acquired images were later assembled using Adobe Photoshop, MS PowerPoint, and Macromedia Freehand image processing software.

Transendothelial Electrical Resistance Assay-- Endothelial cell retraction measured in real time in response to thrombin was measured as described before (24). HMECs were infected with retroviral minigene constructs and seeded on gelatin-coated gold electrodes (4.9 × 10⁴ cm²) and grown to confluence. The small and larger counter electrodes were connected to a phase-sensitive lock-in amplifier. A constant current of 1 µA was applied by a 1 V, 4000 Hz AC signal connected serially to 1 M resistor between the small and large counter electrodes. The voltage between the small electrode and the large counter electrode was monitored by a lock-in amplifier, stored, and processed by a personal computer. The same computer controlled the output of the amplifier and switched the measurement to different electrodes in the course of the experiment. Prior to the experiments, the monolayers were washed two times with serum-free medium and incubated for 2 h in 1% serum-supplemented medium.

Data Analysis-- The data were analyzed using GraphPad Prism 2.01 (GraphPad Software, San Diego, CA). Statistical comparisons were made using a two-tailed Student's t test. The experimental values were considered significant at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Coupling of PAR1 to G_o1 Proteins in Sf9 Membranes-- Thrombin is known to couple to multiple G proteins including G_i, G_q, and G_12/13 proteins (25). Activation of the thrombin receptor results in initiating an array of signal transduction pathways such as phosphatidylinositol hydrolysis, mobilization of Ca²⁺ stores, induction of stress fiber formation, and activation of mitogen-activated protein kinase (1, 5). Some signaling pathways regulated by thrombin have been shown to be sensitive to PTX, e.g. intracellular calcium release, activation of Na⁺/H⁺ exchanger, arachidonic acid release, induction of PAR1 gene expression, von Willebrand Factor release, and endothelial relaxation (12-18). PTX abolishes the interaction between receptors and all members of the G_i subfamily (except G_z) through ADP-ribosylation of the Cys³⁵¹ residue in the C terminus of the G subunit. The effect of PTX has often been attributed to G_i proteins, because G_o proteins are generally believed to play a role in the brain, where it constitutes about 1% of the total protein content (26).

To determine whether PAR1, the most characterized thrombin receptor, can potentially couple to G_o proteins, we expressed this receptor in Sf9 cells along with G_o1, G₁ and G₂, and measured TRAP-stimulated [³⁵S]GTPS binding to membranes prepared from these cells. As a negative control, we co-expressed PAR1, G₁, and G₂ (but not G_o) in Sf9 cells. Fig 1 shows that TRAP stimulated [³⁵S]GTPS binding up to 700 ± 50% above the basal level in a concentration-dependent manner in Sf9 membranes co-expressing PAR1 and G_o112 proteins. In Sf9 membranes co-expressing only PAR1 and G₁₂, TRAP stimulated [³⁵S]GTPS binding to a significantly lower level (230 ± 10% above basal level) (p < 0.05). In addition, we found that TRAP could stimulate [³⁵S]GTPS binding to membranes prepared from Sf9 cells co-expressing PAR1 and G_i112, G_i212, or G_i312 heterotrimers. The level of stimulation in the latter membranes was apparently lower than in membranes co-expressing the G_o112 heterotrimer. Because we did not determine the level of PAR1 expression (because of the lack of commercially available radioligands), we could not conclude whether the lower stimulation level by G_i12 reflected a lower coupling efficiency or a lower expression of receptor or G protein.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   [35S]GTPS binding to Sf9 membranes. membranes prepared from Sf9 cells co-expressing PAR1 and Go12 (filled squares) or G12 (open squares) were preincubated with the indicated concentrations of TRAP for 15 min and further incubated for 30 min after addition of [35S]GTPS at room temperature. Bound [35S]GTPS retained after filtration and four washes was counted. The results are expressed as percentages of stimulation over basal (100 × (cpm stimulated  cpm basal)/cpm basal)). Basal [35S]GTPS was measured in the absence of TRAP and is indicated on the left scale. The results are the means ± S.E. of three experiments performed in duplicate.

Presence of G_o Proteins in HMECs-- Because we demonstrated coupling of PAR1 to G_o1 proteins when co-expressed in Sf9 cells, we next investigated the presence of G_o proteins in HMECs. HMEC lysates were subjected to SDS-PAGE and blotted onto polyvinylidene difluoride membranes. The presence of G_o proteins was detected with an antibody that recognizes both G_o1 and G_o2 proteins, but not G_i proteins (8). Fig. 2 shows that HMECs contain G_o proteins. This unexpected finding led us to further investigate the coupling of G_o proteins to thrombin receptors in HMECs.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Western blot analysis shows presence of Go proteins in HMEC lysates. Lysates from HMECs were subjected to SDS-PAGE and transferred to polyvinylidene difluoride membranes using a standard semi-dry transfer method. The membranes were probed for G protein expression using a specific Go antibody. Lanes 1-5 show serial 10-fold dilutions of purified Go (range 0.01-100 ng/lane) starting with the highest concentration. Lane 6 is another loading of 100 ng of purified Go protein, and lane 7 represents 50 µg of HMEC lysate.

Inhibition of Adenylyl Cyclase-- To determine whether G_i or G_o proteins are involved in thrombin-mediated inhibition of adenylyl cyclase, HMECs were infected with pRevTRE2-G_i1/2 or pRevTRE2-G_o1 minigenes. We could not demonstrate thrombin-mediated inhibition of adenylyl cyclase after either forskolin or isoproterenol stimulation of HMEC. This might be attributable to a different source of HMECs than used previously (27). We tested several different conditions including preincubation with thrombin before challenging the cells with forskolin and thrombin: leaving out thrombin in either the preincubation or incubation step (to eliminate receptor desensitization), elimination of 3-isobutyl-1-methylxanthine (to reduce protein kinase A effects), different concentrations of forskolin, or variation of the incubation or preincubation times. Under none of these experimental conditions did we demonstrate inhibition of forskolin- or isoproterenol-stimulated cAMP formation by thrombin (data not shown). Indeed, thrombin-mediated inhibition of adenylyl cyclase has only been reported in specific endothelial cell lines (28-31).

To verify whether our experimental conditions were sufficient to measure inhibition of adenylyl cyclase activity, we measured it in a CHO cell line that stably expressed A₃ adenosine receptors, which couple preferentially to G_i/o proteins. Our results show that stimulation of A₃ adenosine receptors inhibited adenylyl cyclase in CHO cells (Fig. 3). When these CHO cells were infected with retroviral minigene viruses, and the expression of the 11-amino acid C-terminal peptide of G_i1/2 or G_o1 proteins was induced 24 h before the experiment, this inhibitory action of A₃ adenosine receptors on cAMP formation was blocked almost completely (Fig. 3). Infection of a control virus (pRevTRE2-G_iR) that encodes a peptide based on the C-terminal sequence of G_i1/2 in random order did not have an effect on the inhibition of cAMP formation through A₃ adenosine receptors. This blockade of inhibition was observed in CHO cells infected with pRevTRE2-G_i1/2 or pRevTRE2-G_o1 minigene viruses, indicating that both proteins can couple to A₃ adenosine receptors to inhibit adenylyl cyclase.

View larger version (43K): [in this window] [in a new window]   Fig. 3.   Effect of retroviral GiR, Gi, and Go viruses on NECA-mediated inhibition of adenylyl cyclase in CHO cells stably expressing A3 adenosine receptors. CHO cells infected with retroviral minigene virus or treated with PTX were preincubated with serum-free medium for 20 min and then further incubated for 20 min with serum-free medium alone (basal) or supplemented with forskolin (10 µM) or a mixture of forskolin (10 µM) and NECA (30 µM). The cells were lysed, and cAMP formation was determined using a commercial radioimmunoassay kit. The results are expressed as percentages of the forskolin-stimulated cAMP level. The bars represent the means ± S.E. of three experiments performed in duplicate. The legends under the bars represent either the retroviral minigene virus or treatment with PTX (100 ng/ml for 18 h). A shows cAMP levels as a percent of forskolin in unstimulated (open bars) and forskolin-stimulated (black bars) cells, as well as cells treated with forskolin and NECA (gray bars). The asterisks indicate that the cAMP levels in forskolin- and NECA-treated cells were significantly lower than in the forskolin-treated cells (paired Student's t test; p < 0.05). B shows the cAMP levels of cells treated with NECA and thrombin as a percentage of the forskolin level. Forskolin levels were set at 100% and are represented by a dashed line. The asterisks indicate that the cAMP levels were significantly higher than in cells infected with control virus (pRevTRE2-GiR) (paired Student's t test; p < 0.05).

Stimulation of Intracellular Calcium Release-- Previously, we have shown that thrombin-mediated stimulation of [Ca²⁺]_i was blocked in HMECs transfected with pcDNA-G_q, but not pcDNA-G_i1/2 minigene vectors, which eliminated a role for G_i proteins in this signaling event of thrombin receptors (1). In this study, we loaded HMECs transfected with pcDNA-G_i1/2, pcDNA-G_o1, or pcDNA-G_iR minigene vectors with Oregon Green Bapta-1 for 30 min and measured the thrombin-induced [Ca²⁺]_i in single cells. Fig 4 shows that HMECs transfected with pcDNA-G_o1 minigene vector showed a marked reduction in their calcium response to thrombin, but HMECs transfected with pcDNA-G_i1/2 or pcDNA-G_iR minigene vectors were unaffected. This means that G_o proteins, but not G_i proteins, play a role in the thrombin-stimulated [Ca²⁺]_i increase.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Go minigene peptides inhibit thrombin-stimulated [Ca2+]i release. HMECs were transfected with pcDNA-GiR, pcDNA-Gi1/2, or pcDNA-Go1 minigene vectors. After 48 h, the cells were loaded with Oregon Green Bapta-1 acetoxymethyl ester at 37 °C for 30 min. The basal conditions were established before the addition of thrombin (~70 nM) in Ca2+ buffer. A, the arrow indicates the addition of thrombin. Recordings were made every 10 s and continued for 170 s after stimulation with thrombin. The images were quantitated using NIH Image. The data represent the mean fold increases in fluorescence ± S.E. recorded in 13-21 cells, measured in at least three experiments and two to four coverslips per experiment. B, data from the time point at 90 s, represented as the mean stimulated fluorescence over basal fluorescence, calculated as: (Fstimulated  Fbasal)/Fbasal (where F = fluorescence signal). Open, black, and gray bars represent cells transfected with pcDNA-GiR, pcDNA-Gi1/2, and pcDNA-Go1, respectively. The asterisk indicates that the signal was significantly lower than in control cells transfected with pcDNA-GiR (paired Student's t test; p < 0.05).

To further delineate the mechanism whereby G_o proteins participate in thrombin-mediated calcium signaling pathways, we used a scavenger of G, the -adrenergic receptor kinase C-terminal vector (23), which would allow us to differentiate the role of the G versus G subunits of G_o. We co-transfected ARK-ct DNA along with minigene DNA and measured [Ca²⁺]_i (Fig. 5). ARK-ct expression in G_iR minigene transfected cells led to the same amount of inhibition of [Ca²⁺]_i as in G_o minigene transfected cells, suggesting that G_o is involved in activation of PLC. As expected, the addition of the ARK-ct with the G_q minigene together had no further effect on [Ca²⁺]_i, which is through the G subunit of G_q. Thus, G_o, but not G_i, activation by PAR1 in HMEC cells leads to G activation of PLC. Because G_i can couple to PAR1 when overexpressed in Sf9 cells, these data provide evidence of a cell type-specific coupling of PAR1 to PLC.

View larger version (53K): [in this window] [in a new window]   Fig. 5.   Go regulates [Ca2+]i through its G subunit. The cells were transfected with GiR, Gq, and Go minigene DNA with or without ARK-ct DNA. Thrombin-stimulated [Ca2+]i measurements were performed as described for Fig. 4. The data are represented as percentages of inhibition compared with the control condition in which HMEC cells were transfected with the GiR minigene, which is set at 100% and taken from the peak signal at 90 s. The data represent the mean fold increase in fluorescence ± S.E. recorded in 14-26 cells. Each experimental calcium signal was significantly lower than GiR transfected cells (paired Student's t test; p < 0.001).

Stimulation of Stress Fiber Formation-- The unexpected finding that G_o proteins are involved in the calcium response to thrombin in HMECs led us to further investigate the importance of G_o proteins in downstream signaling events. It has already been shown that thrombin-mediated F-actin stress fiber formation can in part be regulated by [Ca²⁺]_i (32). Therefore, HMECs were co-transfected with pcDNA-GFP and pcDNA-G_i1/2 or pcDNA-G_o1 minigene vectors, and thrombin-induced stress fiber formation was monitored 48 h after transfection. The cells were serum-starved for 24 h prior to thrombin stimulation, permeabilized, and stained for F-actin with Alexa 568-phalloidin, and their nuclei were stained with DAPI. Transfected cells were distinguished from untransfected cells by detecting GFP fluorescence. Thrombin-stimulated stress fiber formation is blocked almost completely in G_o1 minigene transfected cells. In contrast, the presence of the G_i1/2 minigene did not change the level of thrombin-stimulated stress fiber formation as compared with cells transfected with the G_iR minigene (Fig. 6, section II).

View larger version (32K): [in this window] [in a new window]   Fig. 6.   Go minigene vector or treatment with PTX block thrombin-stimulated stress fiber formation. HMECs were co-transfected with pEGFP and minigene vectors, and after 24 h the cells were serum-starved. After another 24 h, cells were stimulated with 10 nM thrombin for 10 min, fixed, and stained with 200 nM Alexa Fluor 568-phalloidin and 1 µg/ml DAPI. The experiments were performed three times and viewed in a field of at least 100 transfected cells. Section I, HMEC transfected with pcDNA-Go1 minigene vector: A shows actin stress fiber staining with Alexa Fluor 568-phalloidin; B shows nuclear staining with DAPI; C shows GFP fluorescence to identify transfected cells; D shows an overlay of these images. Section II, Gi1/2 minigene transfected HMECs. A shows actin stress fiber staining with Alexa Fluor 568- phalloidin; B is an overlay of actin stress fiber and nuclear staining and GFP fluorescence. Section III, thrombin-induced stress fiber formation in untreated (A and B) and PTX treated (C and D) HMECs. A and C show unstimulated cells, whereas B and D show thrombin-stimulated cells (10 nM for 10 min).

If G_o is required for thrombin-induced stress fiber formation, then pertussis toxin, which ADP-ribosylates G_o as well as G_i  subunit, should inhibit stress fiber formation. Treatment of HMECs with pertussis toxin (100 ng/ml) resulted in a strong blockade of thrombin-induced stress fiber formation (Fig. 6, section III). Together, these data clearly show the involvement of G_o proteins, but not G_i proteins, in F-actin stress fiber formation.

Transendothelial Electrical Resistance Assay-- Thrombin-induced stress fiber formation may lead to cellular contraction and disruption of the endothelial monolayer. This leads to an increase in permeability of the endothelial monolayer and may have physiological significance in inflammation and wound healing. To check the involvement of G_o proteins in endothelial permeability, we measured transendothelial electrical resistance on microelectrodes. Confluent HMEC monolayers were seeded onto electrodes and challenged with thrombin. When thrombin induces cell rounding through modification of the cellular cytoskeleton, a transient decrease in resistance can be measured. Fig. 7 shows the thrombin-induced decrease in electrical resistance on electrodes covered with cells infected with pRevTRE2 or pRevTRE2-G_iR virus expressing the G_i random sequence in all of the cells. Similarly a decrease in permeability was observed when HMECs were infected with pRevTRE2-G_i1/2 minigene virus. However, thrombin treatment in HMECs infected with pRevTRE2-G_o1 minigene virus or pretreated with PTX (100 ng/ml) resulted in a dramatic blockade of the thrombin-induced transendothelial electrical resistance change compared with either G_iR or G_i1/2 minigene transfected HMECs. Hence, G_o proteins are the major G_i/o family proteins involved in the thrombin-induced endothelial barrier dysfunction.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   HMEC infected with pRevTRE2-Go minigene viruses block thrombin-induced endothelial barrier dysfunction. The cells were infected with minigene viruses and plated on gold electrodes. The production of the C-terminal peptides was induced using 2 µg/ml doxycyclin. After serum deprivation, the cells were stimulated with 25 nM thrombin, and changes in transendothelial electrical resistance were monitored in real time. A representative from at least three experiments is shown. Statistical analysis was performed with the peak values. The peak values from PTX-treated HMECs were significantly smaller than peak values from cells infected with pRevTRE2-Gi1/2 minigene virus. A similar reduction was observed in cells infected with pRevTRE2-Go1 virus. The cells infected with pRevTRE2-GiR showed a smaller peak value than control, but our results indicate that Go proteins may play a prominent role in disruption of the endothelial monolayer (paired Student's t test; p < 0.05).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The role of G_o proteins has been primarily established in the brain, where it represents about 1% of the total protein content (26). Although the role of G_o proteins had been somewhat obscure, several reports now indicate distinct roles and effectors for these G subunits, which share about 80% amino acid identity with G_i proteins. Novel effectors or regulators such as Rap1GAP, VMAT2, RGS14, or Pcp2 have been described that G_o specifically or differentially targets compared with G_i proteins (9-11, 33, 34).

PTX-mediated ADP-ribosylation of G_i or G_o proteins inhibits their interaction with G protein-coupled receptors, and several thrombin-mediated effects have been shown to be sensitive to PTX (12, 13, 17), such as stimulation of PAR1 gene expression (18), NO release (15), endothelial relaxation (14), Ca²⁺ influx, and release of tissue plasminogen activator and Von Willebrand factor (16). To determine whether G_o proteins serve a specific role in the signaling pathways of thrombin receptors in HMECs, we investigated whether PAR1 has the potential to activate G_o proteins in vitro. TRAP stimulation of PAR1 induced a 7-fold increase in [³⁵S]GTPS binding to Sf9 membranes co-expressing PAR1 and the G_o112 heterotrimer. This stimulation of [³⁵S]GTPS binding was significantly higher than in membranes expressing PAR1 along with the G₁₂ dimer. We also found in similar experiments that TRAP stimulates [³⁵S]GTPS to Sf9 membranes co-expressing PAR1 and G_i1, G_i2, or G_i3 heterotrimers (data not shown). Because no radioligands are commercially available to determine the level of receptors, we could not distinguish whether PAR1 couples preferentially to a specific G protein subtype. Hence, we conclude that PAR1 has the ability to couple to and activate G_o proteins, as well as G_i proteins.

Although PAR1 is able to couple to G_i in Sf9 cells overexpressing either G_i1, G_i2, or G_i3, in HMECs, there was no inhibition of adenylyl cyclase by thrombin under a variety of conditions. In a CHO cell line that stably expressed A₃ adenosine receptors, a 30% inhibition of forskolin-stimulated cAMP formation could be observed, and this effect was blocked by infection of either pRevTRE2-G_i1/2 or pRevTRE2-G_o1 minigene virus. Taken together, these findings suggest that thrombin receptors do not use G_i or G_o proteins to inhibit adenylyl cyclase in HMECs. Interestingly, PAR1 does couple to G_i to regulate PAR-1 gene expression in HMEC cells (18), endothelial albumin endocytosis (35), or endothelial proliferation.²

Next, we showed that G_o proteins are expressed in HMECs. This finding further indicates that G_o proteins may play a role in the signaling of thrombin receptors in HMECs. Because calcium signaling has been shown to be sensitive to PTX in several cell lines, we wanted to verify whether transfection of minigene vectors encoding the C-terminal sequence of G_i1/2 or G_o1 proteins could block thrombin-stimulated intracellular calcium mobilization. Previously, we have shown that the G_q minigene vectors, but not the G_i1/2 minigene vectors can block thrombin-stimulated phosphatidylinositol hydrolysis and intracellular calcium mobilization (1). In addition, we demonstrated that a minigene vector encoding the G_q C-terminal peptide in which the last two residues were mutated could not block these effects. These findings demonstrated that G C-terminally based minigenes can specifically block G protein families. We showed in HMEC that G_o minigenes caused a significant reduction in [Ca²⁺]_i response to thrombin compared with G_i1/2 or G_iR minigenes. These results indicate that G_o, but not G_i, serves to couple thrombin receptors to calcium signaling. This Ca²⁺ transient is mediated by G activation of PLC, because a similar inhibition was induced by transfection with a scavenger of G signaling, the ARK-ct peptide, and their effects were not additive. Thus, there is a cell-specific lack of coupling of G_i to PLC.

To further demonstrate the importance of G_o proteins in signal transduction pathways downstream from calcium events, we measured thrombin-stimulated stress fiber formation in HMECs transfected with minigene vectors or pretreated with PTX. Stress fiber formation has been shown to depend on a number of signaling events, including intracellular calcium release and/or activation of Ca²⁺-dependent protein kinase C (32, 36). Thrombin-induced stress fiber formation was blocked in PTX-pretreated cells as well as in cells transfected with pcDNA-G_o1 minigene vector but not in cells transfected with pcDNA-G_i1/2 minigene vector, indicating the involvement of G_o proteins, but not G_i proteins, in this signaling event.

Formation of stress fibers often leads to cell rounding, which, in the case of endothelial cells, disrupts the endothelial monolayer. Hence, we measured endothelial monolayer permeability, a physiological response downstream of stress fiber formation. Thrombin induced a decrease in transendothelial electrical resistance, and this effect was blocked in cells pretreated with PTX. The cells infected with the pRevTRE2-G_o1 minigene virus also exhibited a muted thrombin response. The cells infected with the control virus pRevTRE2-G_iR had a normal response to thrombin, whereas the cells infected with pRevTRE2-G_i1/2 demonstrated a slightly reduced response. Hence, we showed that G_o proteins play an important role in the signaling of thrombin receptors distinct from G_i proteins, and we confirmed its involvement in different levels of the signaling cascade from calcium mobilization to stress fiber formation and endothelial barrier dysfunction.

In conclusion, we have demonstrated a novel role for G_o proteins in the signal transduction of thrombin receptors in HMECs in which they regulate calcium signaling and cytoskeletal rearrangements. In addition, we have shown that the minigene approach can be used to dissect out the effects of pertussis toxin and determine whether they are mediated by G_i or G_o proteins.

	FOOTNOTES

^* This work was supported by Grant HL60678-01A1 from the National Institutes of Health (to H. E. H.).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.

^§ These authors contributed equally to this work.

^** To whom correspondence should be addressed: Dept. of Pharmacology, Vanderbilt University Medical Center, 442 Robinson Research Bldg., Nashville, TN 37232. E-mail: heidi.hamm@vanderbilt.edu.

Published, JBC Papers in Press, May 30, 2002, DOI 10.1074/jbc.M204477200

² J. F. Vanhauwe, T. O. Thomas, R. D. Minshall, C. Tiruppathi, A. Li, A. Gilchrist, E.-J. Yoon, A. B. Malik, and H. E. Hamm, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: PAR, protease-activated receptor; GTPS, guanosine-5'-O-(3-thio)triphosphate; ARK-ct, -adrenergic kinase C terminus; CHO, Chinese hamster ovary; DAPI, 4',6-diamidino-2-phenylindole dihydrochloride; GFP, green fluorescent protein; HBSS, Hank's buffered salt solution; HMEC, human microvascular endothelial cell; NECA, N-ethyl-5'-carbamoyladenosine; PTX, pertussis toxin; Sf9, Spodoptera frugiperda 9; TRAP, Thrombin Receptor Activating Peptide.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles: