RhoA Interaction with Inositol 1,4,5-Trisphosphate Receptor and Transient Receptor Potential Channel-1 Regulates Ca2+ Entry

We tested the hypothesis that RhoA, a monomeric GTP-binding protein, induces association of inositol trisphosphate receptor (IP3R) with transient receptor potential channel (TRPC1), and thereby activates store depletion-induced Ca2+ entry in endothelial cells. We showed that RhoA upon activation with thrombin associated with both IP3R and TRPC1. Thrombin also induced translocation of a complex consisting of Rho, IP3R, and TRPC1 to the plasma membrane. IP3R and TRPC1 translocation and association required Rho activation because the response was not seen in C3 transferase (C3)-treated cells. Rho function inhibition using Rho dominant-negative mutant or C3 dampened Ca2+ entry regardless of whether Ca2+ stores were emptied by thrombin, thapsigargin, or inositol trisphosphate. Rho-induced association of IP3R with TRPC1 was dependent on actin filament polymerization because latrunculin (which inhibits actin polymerization) prevented both the association and Ca2+ entry. We also showed that thrombin produced a sustained Rho-dependent increase in cytosolic Ca2+ concentration [Ca2+]i in endothelial cells overexpressing TRPC1. We further showed that Rho-activated Ca2+ entry via TRPC1 is important in the mechanism of the thrombin-induced increase in endothelial permeability. In summary, Rho activation signals interaction of IP3R with TRPC1 at the plasma membrane of endothelial cells, and triggers Ca2+ entry following store depletion and the resultant increase in endothelial permeability.

The increase in cytosolic calcium concentration ([Ca 2ϩ ] i ) 1 activated by depletion of Ca 2ϩ stores is required for signaling multiple processes in non-excitable cells (1). The increase in [Ca 2ϩ ] i regulates responses ranging from tension development through activation of actin and myosin motors to modulation of cell-cell and cell-extracellular matrix adhesive forces (2)(3)(4). For example in endothelial cells, increase in [Ca 2ϩ ] i triggered by depletion of Ca 2ϩ stores is essential for the increase in endothelial permeability induced by thrombin (5,6). [Ca 2ϩ ] i increases rapidly after the release of Ca 2ϩ stores in the endoplasmic reticulum (ER). This is followed by a more sustained response secondary to Ca 2ϩ entry through plasmalemmal channels (1,(7)(8)(9)(10)(11). The initial release of cytosolic Ca 2ϩ occurs following the heterotrimeric G protein-coupled receptor-mediated generation of inositol 1,4,5-trisphosphate (IP 3 ), which in turn binds to inositol 1,4,5-trisphosphate receptor (IP 3 R) on ER, and signals Ca 2ϩ release from ER stores. Depletion of ER Ca 2ϩ induces activation of store-operated channels (SOC), resulting in Ca 2ϩ entry and replenishment of ER stores (8,(11)(12)(13). Molecular cloning and functional expression studies showed that the Drosophila transient receptor potential channel (TRPC) family of proteins are prominent plasmalemmal Ca 2ϩ channels in non-excitable cells (10,(12)(13)(14). These channels are activated in response to stimulation of G proteincoupled receptors (1,10,12,13,15,16). TRPC-1, -2, -4, and -5 are likely candidates for endogenous SOCs because they are activated by Ca 2ϩ store depletion (1,12,14,17). TRPC1 activates Ca 2ϩ entry upon store depletion in a variety of cell types including endothelial cells (12, 18 -22). Despite their involvement in the mechanism of Ca 2ϩ entry, the signals by which store depletion activate TRPC1-induced Ca 2ϩ entry are unclear.
It is believed that interaction of IP 3 R with TRPC is required for activation of store depletion-induced Ca 2ϩ entry (1,7,8,13,16,(23)(24)(25). Interaction may be in the form of chemical or conformation coupling. Both models help to explain activation of Ca 2ϩ entry through TRP channels (1,7,8,13,16). In the chemical-coupling model, Ca 2ϩ store depletion induces release of diffusible messenger(s) from ER that activate SOC (26 -29); however, identity of these mediator(s) is not known. In the conformational coupling model, store depletion causes a IP 3 R conformational change that enables it to interact with TRPC, thereby resulting in channel opening (7,8,13,16). If a change in IP 3 R conformation is the sole requirement for its coupling to TRPC and its activation, Ca 2ϩ entry through these channels should be essentially complete upon store depletion. However, patch clamp and whole cell fluorescence studies showed that complete activation of Ca 2ϩ entry through SOC was achieved several tens to hundreds seconds after Ca 2ϩ store depletion (1, 23, 30 -34). Furthermore, this model fails to explain Ca 2ϩ entry by TRPC1 as these channels have been shown to be localized within intracellular membranes (35). These inconsistencies suggest that coupling involves other events such as translocation and docking of IP 3 R and TRPC1 at the plasma membrane (1,23,24,34). Thus, it is possible that proteins capable of trafficking IP 3 R and TRPC1 to the membrane induce association of components of the complex and trigger Ca 2ϩ entry. The monomeric GTP-binding proteins regulating SOC activation (24, 36 -39) may be important in signaling the interaction of IP 3 R and TRPC1, and thus in activation of Ca 2ϩ entry.
Activation of monomeric Rho family GTP-binding proteins, Rho, Rac, and Cdc42, depends on the GTP/GDP exchange cycle (40 -42). These proteins can traffic from cytosol to plasma membrane on activation and they also regulate vesicle trafficking (40 -42). We and others have shown that thrombin rapidly induces activation of RhoA (but not Rac or Cdc42) in endothelial cells (43)(44)(45). 2 In the present study, we addressed the possibility that RhoA induces the interaction of IP 3 R and TRPC1 required for activation of store depletion-induced Ca 2ϩ entry. We observed that RhoA associated with IP 3 R and TRPC1 at the plasma membrane after thrombin stimulation of endothelial cells. Plasma membrane translocation of IP 3 R and TRPC1 and store depletion-induced Ca 2ϩ entry were dependent on Rho because inhibition of Rho activation prevented these responses. We also showed that Rho-activated Ca 2ϩ entry has an important functional consequence in mediating the thrombin-induced increase in transendothelial permeability.
Endothelial Cell Culture-HPAEC were cultured in endothelial growth medium supplemented with 10% FBS. HMEC were grown in endothelial basal medium MCDB-131 supplemented with 10% FBS. Cells were maintained at 37°C in a humidified atmosphere of 5% CO 2 and 95% air until they formed confluent monolayer. Cells from each primary flask were detached with 0.05% trypsin containing 0.02% EDTA, resuspended in fresh culture medium, and passaged as described below. In all experiments, HPAEC between passages 5 and 8 were used.
Cell Transfection-C3 transferase was introduced into endothelial monolayer by transfecting C3 using LipofectAMINE (46). After rinsing the monolayer with Opti-MEM I, cells were incubated with Opti-MEM I medium containing 3.5 g/ml LipofectAMINE for 45 min. C3 transferase (2-4 g/ml) was then added to this medium and cells were allowed to incubate for 6 -8 h after which media was replaced with 10% FBS, endothelial growth medium and cells were used for experiments the following day. The EGFP vector containing wild type RhoA was a generous gift from Dr. M. Philips (New York University School of Medicine, NY), pCMV5 vector and vector containing dominant-negative (N19dn) Rho mutant were provided by Dr. T. Kozasa (University of Illinois at Chicago, IL). Transfection was performed on 50 -70% confluent HPAEC grown on 4-well Lab-Tek chambers or 24-well plates using Superfect reagent following the supplier's protocol. The efficiency of transfection in HPAEC ranged from 10 to 20%. We also determined whether Rho regulates Ca 2ϩ entry by modulating the activity of TRPC1. We overexpressed TRPC1 by 3-fold in HMEC as in these cells a high efficiency of transfection can be obtained. Briefly, HMEC grown to 50% confluence were incubated with LipofectAMINE-TRPC1 cDNA or LipofectAMINE-pcDNA3.1 vector complexes for 4 h. LipofectAMINE-DNA complexes were made by incubating 4 g of LipofectAMINE with 0.5 g of plasmid DNA in 0.2 ml of Opti-MEM I for 45 min at 22°C. LipofectAMINE-DNA complexes were diluted with 0.8 ml of Opti-MEM before being added to HMEC, pre-washed 2 times with Opti-MEM I, for 4 h. To end the transfection procedure, 2 ml of MCDB 131-medium supplemented with 10% FBS was added to each well for up to 16 h after which they were fed again with fresh 10% FBS-MCDB medium and allowed to grow further till confluence (47). Cells overexpressing TRPC1 were then treated without or with C3 to determine the role of Rho in modulating TRPC1-induced Ca 2ϩ entry.
Transendothelial Resistance Measurement-The time course of endothelial cell retraction in real time, a measure of increased endothelial permeability, was determined as described (48). HPAE cells (200,000 cells) grown to confluence on gelatin-coated small gold electrode (4.9 ϫ 10 Ϫ4 cm 2 ) were left untreated or treated with C3, as reported previously (45). Cells were then stimulated with thrombin to measure changes in electrical resistance of endothelial monolayer in real time. The small electrode and larger counter electrode were connected to a phase-sensitive lock-in amplifier. A constant current of 1 A was supplied by a 1-V, 4000-Hz alternating current connected serially to 1 M⍀ resistor between the small electrode and the larger counter electrode. The voltage between the small electrode and large electrode was monitored by lock-in amplifier, stored, and processed on a computer. Data are presented as change in resistive (in-phase) portion of impedance normalized to its initial value at time.
[Ca 2ϩ ] i Measurement-Increase in [Ca 2ϩ ] i was measured using the Ca 2ϩ -sensitive fluorescent dyes fura-2AM or fluo3-AM. For loading cells with fura-2AM, cells grown on 25-mm coverslips were incubated with 3 M fura for 15 min at 37°C. Cells were then washed 2 times with HBSS and imaged using an Attoflor Ratio Vision digital fluorescence microscopy system (Atto Instruments, Rockville, MD) equipped with a Zeiss Axiovert S100 inverted microscope and a F-Fluar ϫ40, 1.3 NA oil immersion objective. Regions of interest in individual cells were marked and excited at 334 and 380 nm with emission at 520 nm. The 334/380 excitation ratio that increased as a function of [Ca 2ϩ ] i was captured at 5-s intervals.
For loading cells with fluo-3AM, cells were grown on 4-well Lab-Tek chambers and incubated with 3 M fluo-3AM for 15 min at 37°C. Cells were then washed 2ϫ with HBSS and viewed using a LSM510 Zeiss confocal microscope with 63X 1.2 NA water immersion objective. A series of time lapse confocal images were acquired at 12.5-s intervals following thrombin stimulation using 488 and 568 nm excitation laser lines for fluo3 and dsRed fluorescence, respectively. We used the Zeiss Multitrack imaging configuration to acquire these images independent of each other to avoid cross-talk between the two fluorescent indicators.
Patch Clamping of Endothelial Cells-Patch clamp in a whole cell configuration was performed on HMEC attached to a coverslip. Patch electrodes made from 1.5-mm borosilicate glass tubing without filament (Narishige, Japan) had a resistance typically between 3 and 6 M⍀ when filled. Cell membrane and pipette capacitative transients were subtracted from the records by the amplifier circuitry before sampling. Voltages were not compensated for liquid junction potentials. Membrane currents were measured with an EPC-7 amplifier in conjunction with pClamp 8.1 software and a Digidata 1322 A/D converter (Axon Instruments, Foster City, CA). The currents were filtered at 2 kHz (low-pass bessel filter) and sampled at an interval of 10 ms. Coverslips with cells were perfused at a rate of ϳ2 ml/min, whereas the bath solution was continuously removed by a vacuum line. Complete solution changes were achieved within 10 s. The standard extracellular solution contained (mM) 135 sodium glutamate, 1 MgCl 2 , 4 CaCl 2 , 10 glucose, and 10 HEPES, pH 7.4 (NaOH). The pipette was filled with (mM) 135 N-methyl-D-aspartic acid glutamate, 10 CsCl, 10 BAPTA, 1 MgCl 2 , 1 ATP, 10 HEPES, pH 7.2 (CsOH). After formation of the giga-ohm seal the patch was ruptured by negative suction and the whole cell patch configuration was achieved. SOC currents were then measured at a holding potential of Ϫ50 mV by depleting the store with 30 M 3-deoxy-3-fluoro-D-myoinositol-IP 3 (non-metabolizable IP 3 ) included in the pipette solution. All experiments were performed at room temperature.
Co-immunoprecipitation of Rho, IP 3 R, and TRPC1-HPAEC grown in 100-mm dishes were serum starved followed by quick washing in ice-cold phosphate-buffered saline. Cells were then lysed in buffer containing 50 mM Tris, pH 7.4, 150 mM NaCl, 0.25 mM EDTA, pH 8.0, 1% 2 P. Kouklis, unpublished observations. deoxycholic acid, 0.5% Nonidet P-40, 0.1% SDS, 1 mM NaF, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, and 2 g/ml each of leupeptin, aprotinin, and pepstatin A. The lysate was scraped and centrifuged at 4°C at 14,000 ϫ g for 10 min. Cell lysate containing an equal amount of protein was then incubated with rabbit IgG or incubated with anti-rabbit polyclonal Rho, IP 3 receptor, or TRPC1 Abs for 3-4 h followed by addition of protein A-agarose beads overnight at 4°C. Beads were collected by centrifugation, washed 3 times with lysis buffer without detergents after which proteins were eluted from the beads by boiling the samples suspended in Laemmli sample buffer. Each sample was then electrophoresed on 10 or 4 -20% linear gradient SDS-PAGE gels, transferred to nitrocellulose for West-ern blotting with IP 3 receptor, TRPC1, or Rho Abs. Specificity of the TRPC1 Ab as described recently (22) was confirmed by using peptide immunogen as a negative control.
Measurement of Rho Activity-pGEX-2T containing Rhotekin-Rho binding domain was provided by Dr. M. A. Schwartz (Scripps Research Institute, La Jolla, CA). Bacterial expressed GST-Rhotekin Rho binding domain protein (GST-RBD) was purified from isopropyl-1-thio-␤-D-galactopyranozide (1 mM)-induced DH5␣ cells previously transformed with the appropriate plasmid as described (49). Confluent HPAE cells grown in 100-mm dishes were stimulated for the indicated times with 50 nM thrombin. Cells were then quickly washed with ice-cold Trisbuffered saline and lysed in buffer (50 mM Tris, pH 7.4, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 500 mM NaCl, 10 mM MgCl 2 , 10 g/ml each of aprotinin and leupeptin, and 1 mM phenylmethylsulfonyl fluoride). Cell lysates were clarified by centrifugation at 14,000 ϫ g at 4°C for 2 min and equal volumes of cell lysates were incubated with GST-RBD beads (15 g) at 4°C for 1 h. The beads were washed three times with wash buffer (50 mM Tris, pH 7.4, 1% Triton X-100, 150 mM NaCl, 10 mM MgCl 2 , 10 g/ml each of aprotinin and leupeptin, and 0.1 mM phenylmethylsulfonyl fluoride), and bound Rho was eluted by boiling each sample in Laemmli sample buffer. Eluted samples from beads and total cell lysate were then electrophoresed on 12.5% SDS-PAGE gels, transferred to nitrocellulose, blocked with 5% nonfat milk, and analyzed by Western blotting using a polyclonal anti-RhoA, anti-IP 3 R, or anti-TRPC1 Abs. In addition, cell lysate from each sample was Western blotted with anti-RhoA Ab to confirm equal protein loading in each lane.
Immunofluorescence Studies-Cells were fixed for 15 min with 2% paraformaldehyde in HBSS containing 10 mM HEPES buffer, pH 7.4, at room temperature. Cells were thoroughly rinsed 4 times with HBSS and then permeabilized with 0.1% Triton X-100 for 3 min. After rinsing, cells were incubated with 1% ovalbumin in HBSS followed by incubation with anti-IP 3 R or anti-TRPC1 Abs. Cells were then washed 3 times with HBSS and incubated with Alexa-labeled secondary Abs plus DAPI to stain nuclei. Specific IP 3 R or TRPC1 immunostaining was determined by staining EGFP-Rho expressing cells with control IgG or with TRPC1 Ab preabsorbed with an equivalent amount (w/w) to its immunizing peptide. For staining actin filaments, cells were incubated with Alexa-labeled phalloidin plus DAPI. Cells were then washed 3 times with HBSS, mounted with Prolong Antifade mounting medium, and viewed with a 63 ϫ 1.2 NA objective using a Zeiss LSM 510 confocal microscope. To avoid cross-talk between the fluorescent dyes, separate images were acquired with the Multitrack imaging configuration in which each laser line (364, 488, and 568 nm) is line scanned independently and thereby recorded quasi-simultaneously.
Statistical Analysis-Comparison between experimental groups were made using paired t test using SigmaPlot software. Differences in mean values were considered significant at p Ͻ 0.05.

Activation of Rho Signals Store Depletion-induced Ca 2ϩ Entry by Regulating TRPC1 Function-We inhibited endogenous
Rho function either by overexpressing the dominant-negative mutant of RhoA (19Ndn-Rho) or by ADP ribosylating Rho with C3 transferase to determine the role of Rho in store depletionmediated Ca 2ϩ entry. In addition, we determined Rho activation and transendothelial electrical resistance changes (a measure of endothelial permeability in real-time) (48) to address the role of Rho-regulated Ca 2ϩ entry in mediating a thrombin-induced increase in endothelial permeability.
We showed using GST-Rhotekin fusion protein, which specifically binds to activated Rho, a 2-3-fold increase in Rho activity following thrombin stimulation of HPAEC (Fig. 1A). To determine the effects of N19dn-Rho expression on thrombininduced Ca 2ϩ transients, dsRed plasmid cDNA (transfection marker) was co-expressed with empty vector or vector containing dn-Rho plasmid in HPAEC. These cells were then loaded with flou3 to measure the Ca 2ϩ transients in real time in the transfected cells (dsRed-positive) by confocal microscopy (Fig.  1B). In cells expressing the empty vector (outlined) or untransfected cells in the same field, we observed the characteristic Ca 2ϩ transient following thrombin stimulation as indicated by increase in flou3 fluorescence intensity after challenge (Fig.  1B). However, cells expressing dnRho (outlined) exhibited smaller and short-lived Ca 2ϩ transients (Fig. 1B). These findings indicate that RhoA is important in regulating the increase in [Ca 2ϩ ] i in response to thrombin exposure of HPAEC.
To determine whether Rho is responsible for activation of SOC, we used thapsigargin and IP 3 because these agents induce store depletion-mediated Ca 2ϩ entry independent of ligand-receptor-G protein-coupled processes (12). Thapsigargininduced Ca 2ϩ entry was measured in fura 2-loaded control or C3-pretreated HPAEC incubated in Ca 2ϩ -free HBSS (Fig. 2). Inhibition of Rho markedly suppressed Ca 2ϩ entry following the depletion of ER stores with thapsigargin without affecting the Ca 2ϩ release response (Fig. 2, A and B). Fig. 3 shows the IP 3 -activated store-operated current in control and C3-pretreated endothelial cells. As evident from the figure, IP 3 activated store-operated current in control cells with a mean value FIG. 5. Rho-mediated Ca 2؉ entry regulates thrombin-induced endothelial retraction. HPAEC grown to confluence on gold electrodes were left untreated or were treated with 10 g/ml C3 transferase in 10% endothelial growth medium for 16 h. Cells were then stimulated with 50 nM thrombin to measure the time course of transendothelial electrical resistance changes (a measure of loss of endothelial barrier function). As results from the experiments were similar, data from a representative experiment are shown (n ϭ 3). ␣-T, ␣-thrombin. of 1.8 Ϯ 0.05 pA, which was inactivated by perfusion of lanthanum, a blocker of the store-operated Ca 2ϩ channels (Fig. 3A) (50). This value of mean current density is within the range reported in endothelial cells (12). However, inhibition of Rho by C3 transferase reduced the IP 3 -induced Ca 2ϩ current (Fig. 3, A  and B). Thus, these observations show that Rho regulates Ca 2ϩ entry by modulating the function of store-operated Ca 2ϩ channels.
As TRPC1 is a primary store-operated Ca 2ϩ channel in human endothelial cells (12,20), we determined the role of Rho on endogenous TRPC1 function in HMEC in which TRPC1 expression was increased 3-fold. 3 Cells expressing control vector or TRPC1 were pretreated without or with C3 transferase after which they were loaded with fura-2AM to measure changes in cytosolic Ca 2ϩ following thrombin stimulation (Fig. 4). We observed a sustained increase in [Ca 2ϩ ] i in response to thrombin in TRPC1-overexpressing cells that was markedly reduced after inhibition of Rho by C3 transferase (Fig. 4). Furthermore, C3 transferase prevented a thrombin-induced decrease in transendothelial electrical resistance, indicating that Rho mediates the decrease in endothelial barrier function by regulating [Ca 2ϩ ] i (Fig. 5).
Rho Interacts with IP 3 R and TRPC1-Because the above results show Rho regulation of TRPC1-mediated Ca 2ϩ entry, we addressed the possibility that Rho activates TRPC1 function by inducing its coupling with IP 3 R. We first tested whether Rho interacts with IP 3 R and TRPC1 by immunoprecipitating whole cell lysates with rabbit anti-Rho Ab. In addition, whole cell lysate immunoprecipitated with rabbit anti-IgG was also included to rule out nonspecific interactions. Fig. 6A shows the results of Western blots of whole cell lysates immunoprecipitated with rabbit anti-Rho Ab or control IgG. We observed that IP 3 R and TRPC1 were present in cell lysate immunoprecipi-tated using anti-Rho Ab, whereas these proteins were not detected in lysates immunoprecipitated using control IgG. Conversely, Rho was also present in lysates immunoprecipitated using IP 3 R or TRPC1 antibody (Fig. 6B).
We then used the GST-Rhotekin pull-down assay in which active Rho is selectively bound by a GST fusion protein containing the Rho-binding region of Rhotekin to address whether IP 3 R and TRPC1 interact with Rho following thrombin stimulation. Lysates from unstimulated or stimulated cells were incubated with GST-Rhotekin fusion protein after which they were subjected to Western blotting with anti-IP 3 R, anti-TRPC1, or anti-Rho Abs. The resulting blot showed that thrombin induced the association of IP 3 R and TRPC1 with activated Rho (Fig. 6C); however, as compared with TRPC1, the association of IP 3 R with Rho was increased within 1 min after challenge (Fig. 6C). These findings show that thrombin activated the association of Rho with IP 3 R and TRPC1 in a time-dependent manner consistent with its role in regulating Ca 2ϩ signaling.
Studies show that Rho-GTPases are translocated to PM upon activation (51)(52)(53). Because the above results from the GSTpull down assay indicate that activated Rho associates with IP 3 R and TRPC1, we used confocal imaging to address whether Rho mediates the recruitment of IP 3 R and TRPC1 at the PM. Thrombin-induced co-localization of IP 3 R and TRPC1 in cells expressing GFP-Rho is shown in Fig. 7. In an unstimulated cell (Fig. 7C), GFP-Rho was homogenously distributed in the cytosol with little staining on the membrane. Furthermore, GFP-Rho was co-localized with IP 3 R and TRPC1 near the perinuclear area (Fig. 7, C and D, yellow). Upon thrombin stimulation, we observed some of the Rho translocated to the PM. IP 3 R was also present at the PM following thrombin stimulation. Marked increase in TRPC1 staining at the PM was observed following a 3-min thrombin stimulation. Interestingly, Rho was co-localized with IP 3 R and TRPC1 at the membrane (Fig. 7, C and D). We also observed the appearance of condensed structures con-FIG. 6. Interactions of IP 3 R, TRPC1, and Rho. A and B, total cell lysates immunoprecipitated with rabbit IgG, rabbit anti-Rho, anti-IP 3 R, or anti-TRPC1 Abs were separated by SDS-PAGE, transferred to nitrocellulose membrane, and Western blotted using anti-Rho, anti-IP 3 R, or anti-TRPC1 Ab to determine their association. Experiments were repeated at least two times. C, cells stimulated without or with thrombin were lysed and the cell lysates were incubated with GST-Rhotekin fusion protein to pull-down activated Rho. Lysates were then electrophoresed, transferred to nitrocellulose, and Western blotted with anti-RhoA, anti-IP 3 R, or anti-TRPC1 Abs. Representative Western blots from multiple experiments (n ϭ 3) shows the association of IP 3 R as well as TRPC1 with Rho following thrombin stimulation. Activated Rho is shown by the amount of RBD-bound Rho, whereas Rho-cytosol represents the amount of Rho in whole cell lysates indicating an equal amount of protein loaded in each lane. RBD, Rhotekin-Rho binding domain; ␣-T, ␣-thrombin.
taining GFP-Rho, IP 3 R, and TRPC1 within the cytosol after thrombin challenge (Fig. 7, C and D). This pattern of IP 3 R and TRPC1 staining and co-localization with Rho was specific be-cause it was not observed in cells stained with control IgG or TRPC1 Ab preabsorbed to its immunogenic peptide. However, in the absence of Rho activation in C3 transferase-treated cells, FIG. 7. Rho-dependent translocation of IP 3 R and TRPC1 to the plasma membrane. Confocal images showing co-localization of Rho with IP 3 R (C and E) or TRPC1 (D and F) in cells pretreated without (C and D) or with C3 (E and F). HPAEC were transiently transfected with EGFP-Rho (green) fixed following thrombin stimulation, and then stained with anti-IP 3 R or TRPC1 Abs (red) plus DAPI (blue) to visualize the co-localization of IP 3 R and TRPC1 with Rho. Note that the pattern of IP 3 R and TRPC1 immunostaining is specific because it was not observed in cells stained with control IgG (A) or with TRPC1 Ab pre-absorbed to its immunizing peptide (B). This experiment was repeated at least 2 times; results are from a representative experiment. ␣-T, ␣-thrombin; bar, 10 or 20 m. Arrows represents the translocation (red or green) and co-localization (merge) of Rho, IP 3 R, and TRPC1 at the plasma membrane, whereas the arrowhead shows the puncta containing GFP-Rho, IP 3 R, and TRPC1 within the cytosol after stimulation. thrombin failed to induce the translocation of IP 3 R and TRPC1 to the membrane (Fig. 7, E and F). These imaging data along with biochemical results demonstrate that Rho signals recruitment of IP 3 R and TRPC1 to the plasma membrane.
Rho Induced Association of IP 3 R with TRPC1 Requires Actin Polymerization-Because our studies show that Rho regulates TRPC1 and IP 3 R co-localization at the plasma membrane and Ca 2ϩ entry through TRPC1 activation, we addressed the possibility that Rho induces association of IP 3 R with TRPC1. We also determined whether actin polymerization (another Rhodependent event) participates in the Ca 2ϩ entry response. We treated HPAEC with C3 transferase and latrunculin-A (which inhibits actin polymerization by binding with actin monomers (54)) to address the contribution of Rho and actin polymerization, respectively, in mediating association of TRPC1 and IP 3 R. Lysates of control cells and cells treated with C3 transferase or latrunculin were immunoprecipitated using anti-IP 3 R antibody, and Western blotted using anti-TRPC1 or IP 3 R Abs. In addition, GFP-Rho-transfected, or cells pretreated without or with C3 transferase or latrunculin were fixed to assess actin filament polymerization. Stimulation of HPAEC with thrombin increased association of IP 3 R with TRPC1; however, inhibition of Rho and actin polymerization prevented the association (Fig.  8A). Inhibition of Rho markedly reduced the amount of actin stress fibers in thrombin-stimulated cells (Fig. 8B), whereas thrombin induced the formation of polymerized actin stress fibers in control and GFP-Rho-transfected cells (Fig. 8, B and  C). In GFP-Rho-transfected cells, these fibers co-localized with Rho in the perinuclear area and plasma membrane (Fig. 8C). Latrunculin pretreatment prevented the formation of actin stress fibers induced by thrombin; in some cells, actin was condensed in foci at the cell periphery (Fig. 8B). In addition, FIG. 9. Rho-activated interaction of IP 3 R with TRPC1 triggers Ca 2؉ entry. Upon activation Rho associates with IP 3 R and TRPC1, and the complex is translocated to the plasma membrane. Rho couples IP 3 R to TRPC1 in an actin filament polymerization-dependent manner, thereby linking Ca 2ϩ store emptying to Ca 2ϩ entry.
FIG. 8. Rho requires actin polymerization to couple IP 3 R with TRPC1 and activate Ca 2؉ entry. A, control HPAEC, HPAEC transfected with C3 transferase, or HPAEC pretreated with 250 nM latrunculin for 30 min were stimulated with thrombin for 2 min. Total cell lysate was immunoprecipitated with anti-IP 3 R Ab, separated by SDS-PAGE, transferred to nitrocellulose membrane, and Western blotted using anti-IP 3 R or anti-TRPC1 Ab. This experiment was repeated at least 2 times; Western blot shows the result from a representative experiment. B and C, HPAEC treated without or with 250 nM latrunculin for 15 min or C3 transferase pretreated cells were fixed following thrombin stimulation and then stained with Alexa-phalloidin (green) plus DAPI (blue) to label filamentous actin and nuclei (B). In addition, GFP-Rho expressing cells were left unstimulated or stimulated with thrombin after which they were fixed and stained for determining actin stress fiber formation as described above (C). This experiment was repeated at least 2 times; results are from a representative experiment. D and E, control cells or cells pretreated with 250 nM latrunculin-A for 15 min were loaded with fura2-AM in the presence or absence of latrunculin for 15 min to measure cytosolic Ca 2ϩ in the presence of 1.3 mM external Ca 2ϩ (D) or nominally Ca 2ϩ -free medium (E). As results from the experiments were similar, data from a representative experiment are shown (n ϭ 3). ␣-T, ␣-thrombin; bar, 20 m; Ϫ, absence; ϩ, presence. inhibition of actin polymerization by latrunculin prevented Ca 2ϩ entry while it had no effect on thrombin-induced Ca 2ϩ release from ER (Fig. 8, D and E). Taken together, these results demonstrate the important role of Rho-dependent actin polymerization in the mediating IP 3 R and TRPC1 coupling. DISCUSSION RhoA regulates discrete cellular events involving actin cytoskeletal dynamics, MLC phosphorylation, and vesicle trafficking (2,40,41). In the present study, we investigated the possible role of Rho in regulating Ca 2ϩ entry in endothelial cells. We interrogated a model in which Rho associates with IP 3 R and TRPC1, the predominant SOC in human endothelial cells (12,20), and thus signals interaction of these two components of the complex required for Ca 2ϩ entry. As RhoA activation has a central role in the mechanism of thrombin-induced increase in endothelial permeability subsequent to activation of MLC phosphorylation and actin filament re-arrangement (43-45), we also addressed the possible relevance of Rhoactivated Ca 2ϩ entry in signaling increased endothelial permeability.
Using pull-down assays, we showed that thrombin induced the association of IP 3 R and TRPC1 with activated Rho in a time frame corresponding to store depletion-induced Ca 2ϩ entry. We also showed that thrombin stimulation of endothelial cells induced translocation of Rho, and subsequent co-localization of Rho with TRPC1 and IP 3 R at the plasmalemma. We showed that plasmalemmal distribution of TRPC1 and IP 3 R and colocalization of these components with Rho were specific in that these responses were not observed in cells stained with control IgG or with TRPC1 Ab preabsorbed to its immunizing peptide. Previous findings that IP 3 R can exist as an integral plasma membrane protein (54 -56) lends credence to our observations. The present results also demonstrated that appearance of IP 3 R and TRPC1 at the membrane were required for the activated form of Rho because thrombin failed to induce PM translocation of these components in endothelial cells pretreated with C3. Inhibition of Rho function with either dnRho or C3 transferase significantly reduced Ca 2ϩ entry as well as SOC current in single cell recordings. This was the result of Rho activation of TRPC1 because overexpression of TRPC1 in endothelial cells produced a sustained C3 transferase-sensitive increase in [Ca 2ϩ ] i in response to thrombin exposure. We observed that thrombin-induced Ca 2ϩ release from ER stores was reduced by C3 in TRPC1-overexpressing cells. The present studies showing that IP 3 R-induced Ca 2ϩ release requires Rho activation are consistent with the findings that IP 3 R receptor activation occurs through a Rho-dependent pathway (57). Taken together, our results demonstrate an important role of Rho in inducing interaction of the key proteins required for activation of SOCinduced Ca 2ϩ entry. In addition, we observed that inhibition of Rho activation prevented increased endothelial permeability occurring secondary to the rise in [Ca 2ϩ ] i (5,6). Thus, Ca 2ϩ entry by this mechanism is important in increasing endothelial permeability.
Our results demonstrate that Rho activation is required for both thrombin and thapsigargin-or IP 3 -induced Ca 2ϩ entry. Studies have shown that thrombin binding to protease activated receptor-1 activates heterotrimeric G proteins, G 12 /G 13 and G q , which in turn induce Rho activation (58 -60). However, the mechanism of thapsigargin activation of Rho in the absence of the receptor-coupled pathway is not known. We observed only modest Rho activation induced by thapsigargin (data not shown); thus, it is possible that Rho can activate SOC independent of the mode of Ca 2ϩ store emptying (i.e. an agonist in the case of thrombin versus ER calcium pump blockade in case of thapsigargin).
Although evidence from several laboratories showed that IP 3 R coupling with TRPC can trigger Ca 2ϩ entry, the mechanisms responsible for this critical association event are not clear (23,25,34,(61)(62)(63)(64). Models involving chemical and conformation coupling have been suggested (1, 11, 14 -16, 65). Although there is evidence in favor of chemical coupling, the identity of the messengers is unknown (26 -29). There is also inconclusive evidence supporting the concept that coupling of IP 3 R with TRP channels is mediated through a IP 3 R conformation change. Additionally, as TRPC1 has been shown to be localized in intracellular membrane it is difficult to explain the activation of TRPC1-induced Ca 2ϩ entry based on the above proposals (35). Our results are important in this regard because they help to explain Rho-activated interaction of IP 3 R with TRPC1 and subsequent activation of Ca 2ϩ entry. RhoA may signal localization of IP 3 R with TRP channels by targeting these components of the Ca 2ϩ entry machinery to the PM. Furthermore, Rho-dependent plasmalemmal recruitment of these channels provides an explanation for the observed delay in complete activation of SOC following Ca 2ϩ store depletion (1, 23, 30 -34). Our finding of Rho-dependent PM recruitment of IP 3 R and TRPC1, which promotes activation of Ca 2ϩ entry following store emptying, is in accord with suggestions that SOC activation involves trafficking and docking of the channels in the plasmalemma (1, 24, 34, 36 -39).
Spatial rearrangement of actin filaments promoted the association of IP 3 R with TRPC1 and the Ca 2ϩ entry (62)(63)(64). Because RhoA regulates actin polymerization, we surmised that Rho-induced association of IP 3 R and TRPC1 is dependent on actin polymerization. We showed that thrombin failed to induce association of IP 3 R with TRPC1 in the absence of actin polymerization, and there was also no Ca 2ϩ entry. Thus, these results point to an important role of actin polymerization in organizing the PM association of IP 3 R and TRPC1.
The present results are consistent with the Rho-activated coupling model in which Rho, by signaling trafficking of IP 3 R and TRPC1 to PM, promotes the interaction of these components of the complex (Fig. 9). Membrane insertion of this complex thereby triggers Ca 2ϩ entry through TRPC1 after Ca 2ϩ store depletion. Rho-induced actin polymerization also participates in this process by maintaining stable interaction of IP 3 R and TRPC1 channels at the PM. Rho activation of Ca 2ϩ entry by this mechanism is functionally important in endothelial cells because it is a key determinant of increased endothelial permeability.