Store depletion induces Gαq-mediated PLCβ1 activity to stimulate TRPC1 channels in vascular smooth muscle cells

Depletion of sarcoplasmic reticulum (SR) Ca2+ stores activates store-operated channels (SOCs) composed of canonical transient receptor potential (TRPC) 1 proteins in vascular smooth muscle cells (VSMCs), which contribute to important cellular functions. We have previously shown that PKC is obligatory for activation of TRPC1 SOCs in VSMCs, and the present study investigates if the classic phosphoinositol signaling pathway involving Gαq-mediated PLC activity is responsible for driving PKC-dependent channel gating. The G-protein inhibitor GDP-β-S, anti-Gαq antibodies, the PLC inhibitor U73122, and the PKC inhibitor GF109203X all inhibited activation of TRPC1 SOCs, and U73122 and GF109203X also reduced store-operated PKC-dependent phosphorylation of TRPC1 proteins. Three distinct SR Ca2+ store-depleting agents, 1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid acetoxymethyl ester, cyclopiazonic acid, and N,N,N′,N′-tetrakis(2-pyridylmethyl)ethane-1,2-diamineed, induced translocations of the fluorescent biosensor GFP-PLCδ1-PH from the cell membrane to the cytosol, which were inhibited by U73122. Knockdown of PLCβ1 with small hairpin RNA reduced both store-operated PLC activity and stimulation of TRPC1 SOCs. Immunoprecipitation studies and proximity ligation assays revealed that store depletion induced interactions between TRPC1 and Gαq, and TRPC1 and PLCβ1. We propose a novel activation mechanism for TRPC1 SOCs in VSMCs, in which store depletion induces formation of TRPC1-Gαq-PLCβ1 complexes that lead to PKC stimulation and channel gating.—Shi, J., Miralles, F., Birnbaumer, L., Large, W. A., Albert, A. P. Store depletion induces Gαq-mediated PLCβ1 activity to stimulate TRPC1 channels in vascular smooth muscle cells.

Plasma membrane store-operated channels (SOCs) are physiologically induced by extracellular agents, which stimulate the classic phosphoinositol signaling pathway composed of Gaq-coupled receptors, PLC activation, phosphatidyinositol 4,5-bisphosphate (PIP 2 ) hydrolysis, and generation of inositol 1,4,5-trisphosphate (IP 3 )a n dd i a c y l g l y e r c o l( D A G ) that leads to IP 3 -mediated depletion of endoplasmic/ sarcoplasmic reticulum (SR) Ca 2+ stores. In vascular smooth muscle cells (VSMCs), SOCs have been proposed to mediate Ca 2+ entry pathways, which regulate cellular functions such as contraction, proliferation, and migration that are linked to regulation of vascular tone, and the development of hypertension and atherosclerosis (1)(2)(3). Consequently, understanding molecular mechanisms involved in gating SOCs is an important objective in vascular physiology.
It is now firmly established that the archetypal storeoperated current I crac , which is characterized by high Ca 2+ permeability, pronounced inward rectification, and a unitary conductance in order of fS, is formed by Orai1 channel proteins (4)(5)(6)(7). Moreover, it is recognized that Ca 2+ store depletion induces oligomerization and translocation of the endoplasmic/SR Ca 2+ sensor STIM1 to the plasma membrane where it induces Orai1 channel opening (4)(5)(6)(7). It is also apparent that many cell types express SOCs, which have much lower Ca 2+ permeabilities, relatively linear current-voltage (I/V) relationships, and larger unitary conductances compared to Orai1-mediated I crac . These SOCs are proposed to be mediated by the canonical transient receptor potential (TRPC) family of Ca 2+ -permeable nonselective cation channel proteins (TRPC1-C7) (8,9), with TRPC1, TRPC3, and TRPC4 subtypes particularly implicated in composing SOCs. Because TRPC subunits form (continued on next page) (120 mg/kg), and mice were killed using cervical dislocation according to the UK Animals Scientific Procedures Act of 1986. Portal veins or second-order mesenteric arteries were dissected free and cleaned of fat, connective tissue, and endothelium in physiologic salt solution containing 126 mM NaCl, 6 mM KCl, 10 mM glucose, 11 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 1.2 mM MgCl 2 , and 1.5 mM CaCl 2 (pH adjusted to 7.2 using 10 M NaOH). Vessels were enzymatically dispersed into single VSMCs as previously described (19,21).

Electrophysiology
Whole-cell and single-channel cation currents were made with an AXOpatch 200B amplifier (Axon Instruments, Union City, CA, USA) at room temperature (20-23°C) as described previously (21). Whole-cell currents were filtered at 1 kHz (23 dB, low-pass 8-pole Bessel filter, Frequency Devices model LP02; Scensys, Aylesbury, United Kingdom) and sampled at 5 kHz (Digidata 1322A and pCLAMP 9.0 software; M o l e c u l a rD e v i c e s ,S u n n y v a l e ,C A ,U S A ) .W h o l e -c e l lI/V relationships were obtained by applying 750 ms duration voltage ramps from +100 to 2150 mV every 30 s from a holding potential of 0 mV. Single-channel currents were filtered between 0.1 and 0.5 kHz and acquired at 1-5 kHz. Single-channel I/V relationships were evaluated by manually altering the holding potential of 280 mV between 2120 and +120 mV. For singlechannel analysis, single-channel current amplitudes were calculated from idealized traces of $60 s in duration using the 50% threshold method and analyzed using pCLAMP 9.0 software. Events lasting for ,6.664 ms [23 rise time for a 100 Hz (23 dB) low-pass filter] were excluded from analysis to maximize the number of channel openings reaching their full current amplitude. Open probability was used as a measure of channel activity and was calculated automatically by pCLAMP 9. Single-channel current amplitude histograms were plotted from the event data of the idealized traces with a 0.01 pA bin width. Amplitude histograms were fitted using gaussian curves with peak values corresponding to channel open levels.
Mean channel amplitudes at different membrane potentials were plotted, and I/V relationships were fitted by linear regression with the gradient determining conductance values. Figures were prepared using MicroCal Origin 6.0 software (MicroCal Software, Northampton, MA, USA), in which inward single-channel openings are shown as downward deflections.

Knockdown of PLCb1
We used a lentiviral-mediated delivery of pLKO.

IP 3 ELISA
Cells or tissues were quickly lysed or homogenized on ice. IP 3 production determinations were performed with a rabbit IP 3 ELISA kit (BlueGene Biotech, Shanghai, China) following the manufacturer's instructions. The data were reported as picograms of IP 3 per milligrams of total cell lysate protein.

Immunoprecipitation and Western blot
Freshly isolated vessel segments or primary cultured cells were lysed by RIPA buffer and then transferred to a microcentrifuge tube (VWR, Lutterworth, United Kingdom). Total cell lysate protein was extracted and immunoprecipitated using antibodies raised against targeted proteins with an EMD Millipore Catch and Release Kit (EMD Millipore, Billerica, MA, USA) followed by 1-dimensional protein gel electrophoresis (15-20 mg total protein per lane). Separated proteins were transferred onto PVDF membranes and then membranes were incubated with the primary antibodies overnight at 4°C. Visualization was performed with a horseradish peroxidase-conjugated secondary antibody (80 ng/ml) and ECL reagents (Pierce Biotechnology, Inc., Rockford, IL, USA) for 1 min and exposure to photographic films. Band intensities were calculated using Image Studio software (Li-Cor Biosciences, Cambridge, United Kingdom) andthenwerenormalizedtocontrol bands. Data shown represent findings from $3 different animals.

Immunocytochemistry
Freshly isolated VSMCs were fixed with 4% paraformaldehyde (Sigma-Aldrich) for 10 min, washed with PBS, and permeabilized with PBS containing 0.1% Triton X-100 for 20 min at room temperature. Cells were incubated with PBS containing 1% bovine serum albumin for 1 h at room temperature and then were incubated with primary antibodies in PBS containing 1% bovine serum albumin overnight at 4°C. In control experiments, cells were incubated without the primary antibody. The cells were washed and incubated with secondary antibodies conjugated to a fluorescent probe. Unbound secondary antibodies were removed by washing with PBS, and nuclei were labeled with DAPI mounting medium (Sigma-Aldrich). Cells were imaged using a Zeiss LSM 510 laser-scanning confocal microscope. The excitation beam was produced by an argon (488 nm) or helium/neon laser (543 and 633 nm) and delivered to the specimen via aZeiss Apochromat 363 oil-immersion objective (numerical aperture, 1.4). Emitted fluorescence was captured using LSM 510 software (release 3.2; Carl Zeiss). Two-dimensional images cut horizontally through approximately the middle of the cells were captured (1024 3 1024 pixels). Raw confocal imaging data were processed and analyzed using Zeiss LSM 510 software. Final images were produced using PowerPoint (Microsoft XP).
In cell-attached patch experiments, the membrane potential was set to 0 mV by perfusing cells in a KCl external solution containing 126 mM KCl, 1.5 mM CaCl 2 , 10 mM HEPES, and 11 mM glucose (pH adjusted to 7.2 with 10 M KOH). A total of 5 mm nicardipine was included to prevent smooth muscle cell contraction by blocking Ca 2+ entry through voltage-dependent Ca 2+ channels.

Statistical analysis
This was performed using paired (comparing the effects of agents on the same cell) or unpaired (comparing the effects of agents between cells) Student's t tests with the level of significance set at a value of P , 0.05.

Activation of TRPC1 channels involves Gaq, PLC, and PKC activities
In our initial experiments, we confirmed that SOCs recorded in the present study are mediated by TRPC1 subunits using anti-TRPC1 antibodies as blocking agents. Figure 1A shows that passive depletion of internal Ca 2+ stores following cell dialysis with a patch pipette solution containing 20 mM BAPTA and no added Ca 2+ evoked whole-cell cation currents with relative linear I/V relationships and E rev of ;+20 mV, which are similar properties to store-operated TRPC1 currents previously described in VSMCs (22). In addition, Fig. 1B and Supplemental Fig. S1B show that bath application of 50 mM BAPTA-AM, a cell-permeable Ca 2+ chelator, activated single-channel activity in cell-attached patches with a unitary conductance of ;2p S ;a g a i n ,t h e s e properties are similar to those previously shown for single TRPC1 SOCs in VSMCs (15,17,18,20,22). Complementary to these findings, Fig. 1A shows that bath application of 1 mg/ml 21 of TIE3, an extracellularacting anti-TRPC1 antibody (14,26), inhibited mean peak whole-cell current densities from 24.21 6 0.63 pA/pF to 1.34 6 0.22 pA/pF (n =6 )a t280 mV. Figure 1B also shows that BAPTA-AM-evoked channel activity, maintained following excision of cell-attached patches into the inside-out configuration, was inhibited by bath application of 1:200 dilution of an intracellularacting anti-TRPC1 antibody to the cytosolic surface of patches, with mean open probability values reduced from 0.64 6 0.06 to 0.16 6 0.03 (n =7)at280 mV.
It is well known that Gaq-mediated PLC activity and production of DAG lead to PKC stimulation, and we have shown that Gaq-coupled receptor agonists and DAG analogs evoke PKC-dependent activation of TRPC1 channels in VSMCs (15,21,25). We therefore examined if Gaq and PLC activities are also required for activation of TRPC1 channels by store depletion in freshly isolated rabbit portal vein VSMCs using well-characterized pharmacologic inhibitors of G proteins, PLC, and PKC on store-operated whole-cell and single-channel TRPC1 currents.
Our previous data indicate that PKC-dependent phosphorylation of TRPC1 proteins is pivotal for activation of TRPC1 SOCs (20,25), and therefore, we studied if PLC activity is involved in this pathway. Immunoprecipitation of freshly isolated rabbit portal vein vessel lysates with a mixture of anti-phosphorylated serine and anti-phosphorylated threonine antibodies followed by Western blotting with an anti-TRPC1 antibody revealed that TRPC1 proteins displayed a low level of constitutive phosphorylation, which was inhibited by pretreatment with 2 mM U73122 or 3 mM GF109203X ( Fig. 2A, B, left panel). Moreover, pretreatment of vessels with 10 mMCPA (Fig.2A,middlepanel)or50mM BAPTA-AM ( Fig. 2A, right panel) for 10 min increased phosphorylation of TRPC1 proteins by ;2-fold, which were reduced by coapplication of 2 mM U73122 or 3 mM GF209203X (Fig. 2B). In control experiments, pretreatment of vessels with BAPTA-AM, CPA, U73122, or GF109203X did not alter TRPC1 expression levels (Supplemental Fig. S2A).
These findings provide pharmacologic evidence that store depletion is coupled to Gaq-mediated PLC activity and that this pathway induces PKC-dependent phosphorylation of TRPC1 proteins, which is important for stimulation of TRPC1 channels.

PLCb1 mediates TRPC1 SOCs in VSMCs
Previous studies have stated that PLCb1, a PLC isoform, is involved in activation of TRPC channels (27)(28)(29), and therefore, we investigated if PLCb1 contributes to PLCmediated stimulation of TRPC1 SOCs in VSMCs. Western blot studies showed that PLCb1 protein is expressed in primary cultured rabbit portal vein VSMCs and that PLCb1 shRNAs reduced PLCb1 expression by ;75% compared to scrambled shRNA sequences (Fig. 3A). In control experiments, PLCb1 knockdown did not alter TRPC1, Gaq, and b-actin expression levels (Supplemental Fig. S2B). It should also be noted that primary cultured VSMCs expressed SOCs with similar single-channel properties and activation mechanisms as TRPC1 SOCs present in freshly dispersed VSMCs (Supplemental Fig. S1B). Moreover, primary cultured VSMCs maintained in 1% fetal calf serum for 3-7dd i splayed a contractile phenotype (Supplemental Fig. S1C).
In VSMCs expressing scrambled shRNA, passive store depletion activated whole-cell TRPC1 currents, which were inhibited by bath application of 2 mM U73122 (Fig. 3B). In contrast, treatment of VSMCs with PLCb1 shRNAs greatly reduced the development of store-operated whole-cell TRPC1 currents at all membrane potentials tested (Fig. 3B,  C). Furthermore, PLCb1 knockdown reduced 50 mM BAPTA-AM-evoked and 10 mMC P A -e v o k e ds i n g l e TRPC1 channel activities by .70% (Fig. 3D and Supplemental Fig. S3A). In contrast, bath application of 1 mM phorbol 12,13-dibutyrate, a direct PKC activator, Figure 1. G-protein, PLC, and PKC activities mediate TRPC1 SOCs. A) Representative recording shows development of a storeoperated whole-cell cation current following break-in into the whole-cell configuration (w.c.), which was inhibited by bath application of the external-acting TRPC1 antibody TIE3. Vertical deflections represent currents evoked by voltage ramps from +100 to 2150 mV (750 ms duration) every 30 s from a holding potential of 0 mV. B) Representative trace shows that BAPTA-AMevoked single cation channel activity in cell-attached patches held at 280 mV was maintained following patch excision into the inside-out configuration (i/o) and inhibited by bath application of an internal-acting TRPC1. C ) Trace shows that development of a store-operated whole-cell TRPC1 current was prevented by inclusion of GDP-b-S in the patch pipette solution. D, E ) Storeoperated whole-cell TRPC1 currents were inhibited by bath applications of U731222 (D) or GF109203X (E). F ) Mean data show the inhibitory effects of GDP-b-S, U73122, and GF109203X on store-operated whole-cell TRPC1 current densities at 280 mV (each data set is n = 6). ***P , 0.001. G, H ) Original recording traces show that BAPTA-AM-evoked single cation channel activity in cell-attached patches held at 280 mV was inhibited by GDP-b-S or a mixture of anti-Gaq and anti-Ga11 antibodies to the cytosolic surface of inside-out patches (G), whereas a mixture of anti-Gai1/2 and anti-Gai3 antibodies had no effect (H). I ) Mean data show inhibitory actions of GDP-b-S, anti-Gaq/11 antibodies, U73122, and GF109203X on BAPTA-AM-evoked TRPC1 channel activity (each data set is n = 6). **P , 0.01; *** P , 0.001.
to PLCb1 knockdown VSMCs readily induced single TRPC1 channel activity, which indicates that PLCb1isinvolvedin stimulation of TRPC1 SOCs upstream from PKC activity (Fig. 3D). These results provide clear evidence that PLCb1 playsamajorroleinactivationofTRPC1SOCsinVSMCs.

Store-depleted PLC activity is mediated by PLCb1 isoform
Our results suggest that store depletion stimulates PLC activity mediated by PLCb1. However, there is no previous evidence for store-operated PLC activity in VSMCs, and so we investigated this idea in more detail. Stimulation of PLC activity induces PIP 2 hydrolysis at the plasma membrane to generate DAG and IP 3 , with the latter molecule diffusing into the cytosol. To monitor store-operated PLC activity in VSMCs, we transfected primary cultured VSMCs with GFP-PLCd1-PH, a fluorescent biosensor with a high affinity for PIP 2 and IP 3 (30)(31)(32)(33), and measured signal changes (in relative fluorescent units) at the plasma membrane [fluorescent intensity in membrane (Fm)] and within the cytosol [fluorescent intensity in cytosol (Fc)]. To provide a comprehensive analysis on whether store depletion induces PLC activity, we studied the effect of BAPTA-AM, CPA, and N,N,N9,N9-tetrakis(2-pyridylmethyl)ethane-1,2diamineed (TPEN), a cell-permeable low-affinity Ca 2+ chelator that selectively lowers Ca 2+ levels within SR Ca 2+ stores, on GFP-PLCd1-PH signals.
In unstimulated cells, GFP-PLCd1-PH signals were predominantly found located at the plasma membrane and had a mean Fm:Fc ratio of ;7, which reflects the predominant cellular location of PIP 2 and also suggests that there is limited cytosolic IP 3 in these conditions (Fig. 4). Bath application of 50 mMBAPTA-AM,10mM CPA, or 1 mM TPEN for 10 min induced translocation of GFP-PLCd1-PH signals from the plasma membrane tothecytosol,whichrelatestoareductioninmeanFm: Fc ratio of ;80% (Fig. 4). These signal changes are likely to represent PLC-mediated PIP 2 hydrolysis at the plasma membrane and subsequent generation of cytosolic IP 3 (30)(31)(32)(33). In support of these ideas, coapplication of 2 mM U73122 reversed BAPTA-AM-, CPA-, and TPEN-induced translocations of GFP-PLCd1-PH signals (Fig. 4). Figure 5 shows that PLCb1 knockdown in VSMCs prevented translocation of GFP-PLCd1-PH signals by 50 mM BAPTA-AM, whereas in the presence of scrambled shRNAs, BAPTA-AM induced similar effects on GFP-PLCd1-PH signals as in Fig. 4 (data not shown). In comparison, stimulation of endogenously expressed a1 Gaq-coupled adrenoreceptors by bath application of 10 mM noradrenaline induced translocation of GFP-PLCd1-PH signals from the plasma membrane to the cytosol in the presence of PLCb1 shRNAs (Fig. 5). This indicates that other PLC isoforms, apart from PLCb1, are likely to have a dominant role in mediating PLC activity induced by this concentration of noradrenaline. These results cannot exclude the possibility that noradrenalineevoked PLCb1 activity produces a small but irresolvable contribution to overall evoked PLC activity, which is involved in mediating stimulation of TRPC1 channels (Fig. 8A, C ). These findings with noradrenaline also show that knockout of PLCb1d o e sn o th a v ea general inhibitory effect on PLC activity, indicating that PLCb1 shRNA is selective. Similar effects on GFP-PLCd1-PH signals in the presence of PLCb1 shRNAs were observed using 10 mMC P Aa n d1m MT P E N (Supplemental Fig. S4).
We also investigated store-depletion-evoked PLC activity by measuring IP 3 production using an ELISA. In primary cultured VSMCs, 10 mM noradrenaline induced an 8-fold increase in IP 3 levels, which was prevented by pretreatment of 2 mM U73122 (Supplemental Fig. S3B). In comparison, 50 mM BAPTA-AM evoked over a 4-fold increase in IP 3 , which was also inhibited by 2 mMU73122 (Supplemental Fig. S3B).
Taken together, our findings provide strong evidence that store depletion induces PLCb1 activity in VSMCs, which provides further support that Gaq-evoked PLC activity and PKC stimulation are important for activation of TRPC1 SOCs.

Store depletion induces interactions between TRPC1, Gaq and PLCb1
For store depletion to induce Gaq-evoked PLC activity and activate TRPC1 SOCs, it would seem appropriate that these molecules interact with one another, and therefore, we investigated these interactions using 2 techniques: coimmunoprecipitation, and proximity ligation assay. Immunoprecipitation with anti-TRPC1 antibodies followed by immunoblotting with either anti-GaqorantiPLCb1antibodies failed to show any interactions between these molecules in unstimulated primary cultured cell lysates (Fig. 6A). However, pretreatment of VSMCs with 50 mM BAPTA-AM for 10 min induced interactions between TRPC1 and Gaq, and between TRPC1 and PLCb1 (Fig.6A). Similar results were also obtained following pretreatment of freshly isolated vessel segments with 10 mM CPA (Supplemental Fig. S2C). As expected, transduction of VSMCs with Figure 3. Activation of TRPC1 SOCs is mediated by PLCb1. A) Western blots and mean data confirm that 2 different PLCb1 shRNA sequences (shRNA1 and shRNA2) reduced PLCb1 expression (n = 3 primary cell culture preparations). **P , 0.01. B) Representative traces show that peak amplitude of store-operated whole-cell TRPC1 currents was greatly reduced following transduction of cells with PLCb1 shRNA1 compared to scrambled shRNA sequences. In the presence of scrambled shRNA, storeoperated whole-cell currents were inhibited by U73122. C ) Mean I/V relationships show that PLCb1 knockdown with shRNA1 and shRNA2 reduced store-operated TRPC1 currents (n = 6). D) Representative recordings and mean data show that BAPTA-AMevoked TRPC1 SOC activities were reduced by both PLCb1 shRNA1 and shRNA2 sequences compared to scrambled shRNA, but this did not affect channel activation by phorbol 12,13-dibutyrate (PDBu) (n = 7). ***P , 0.001. PLCb1 shRNAs significantly decreased BAPTA-AMinduced associations between TRPC1 and PLCb1; however, PLCb1 knockdown did not affect the interaction between TRPC1 and Gaq (Fig. 6). Proximity ligation assays showed no apparent signals between TRPC1 and Gaq, and TRPC1 and PLCb1inrestingcells (Fig. 7A), whereas pretreatment of cells with 50 mM BAPTA-AM for 10 min induced robust fluorescent signals (red) at the plasma membrane, which denoted interactions between TRPC1 and Gaq, and TRPC1 and PLCb1 (Fig. 7A, B). These BAPTA-AM-evoked TRPC1-PLCb1 signals were greatly reduced following transduction of VSMCs with PLCb1 shRNAs, whereas BAPTA-AM-induced interactions between TRPC1 and Gaq remained unchanged (Fig. 7B, C,  D). These findings clearly indicate that store depletion induces formation of TRPC1-Gaq-PLCb1 complexes at the plasma membrane.
In control experiments, neither BAPTA-AM nor CPA altered expression levels of TRPC1, Gaq, PLCb1, or PLCg1, and neither one induced interactions between TRPC1 and PLCg1 (Supplemental Fig. S2C). These negative results with PLCg1 indicate that interactions between TRPC1 and PLCb1 are selective.

Noradrenaline-evoked TRPC1 activity requires PLCb1
Our above results clearly demonstrate that agents that deplete internal Ca 2+ stores induce TRPC1 channel activity through a PLCb1-mediated pathway. In our final experiments, we investigated whether a similar role for PLCb1is involved in mediating TRPC1 channel activity evoked by the physiologic agonist and vasoconstrictor noradrenaline. Figure 8A, C shows that bath application of noradrenaline (1 nM to 100 mM) activated 2 pS cation channel activity in a concentration-dependent manner in cell-attached patches held at 280 mV from VSMCs expressing scrambled shRNA. The properties of these channels are similar to TRPC1 channel activity previously recorded using storedepleting (see above) and vasoconstrictor agents (15,21,34,35). In the presence of PLCb1 shRNA, noradrenaline-induced TRPC1 channel activity was greatly reduced (Fig. 8B, C). Interestingly, knockdown of PLCb1 seemed to preferentially inhibit TRPC1 channel activity evoked by higher concentrations of noradrenaline (1-100 mM), whereas levels of channel activity evoked by lower concentrations were maintained (Fig. 8B, C).
These results strongly suggest that PLCb1 has an important role in mediating TRPC1 channel activity induced by an endogenous agonist, which indicates the likely physiologic relevance of the proposed store-operated PLCb1-mediated pathway in stimulating TRPC1 channels.

DISCUSSION
The present work reveals for the first time that the classic phosphoinositol signaling pathway composed of Gaqmediated PLCb1 activity is stimulated by depletion of Ca 2+ levels within SR Ca 2+ stores in VSMCs. Moreover, storeoperated Gaq-PLCb1 activities coupled to PKC stimulation result in opening of TRPC1 SOCs. These results are likely to have widespread importance because phosphoinositol signaling and TRPC1 channels are ubiquitously expressed among cell types.
There is considerable evidence that SOCs are composed of a heteromeric TRPC1/C5 molecular template in contractile VSMCs; SOCs are absent in TRPC1 2/2 VSMCs, reduced and increased by knockdown and overexpression of TRPC1 proteins, respectively, and inhibited by anti-TRPC1 and anti-TRPC5 antibodies (3,(14)(15)(16)(17)(18)(19)(20)(21)(22). In addition, TRPC1 and TRPC5 proteins colocalize with one another (21). These studies have provided considerable evidence that TRPC1 is the essential subunit that confers channel gating by store depletion, and therefore, these heteromeric TRPC1/C5 templates in VSMCs are often termed TRPC1 SOCs (22). The present work shows that wellestablished store-depletion agents with distinct mechanisms of action (e.g., high intracellular BAPTA, BAPTA-AM, and CPA) activated whole-cell conductances with a relatively linear I/V relationship and an E rev of ;+20 mV, and also single-channel currents with a unitary conductance of ;2 pS in freshly isolated and primary cultured VSMCs, which exhibit contractile phenotypes. Importantly, these studies did not observe SOCs in VSMCs, which had properties that resembled Orai1-mediated I crac such as pronounced inward rectification and very positive E rev .Furthermore,our findings confirm that store-operated whole-cell and singlechannel currents were inhibited by anti-TRPC1 antibodies. These results provide strong evidence that TRPC1 SOCs, and not Orai1-mediated I crac , are recorded in the present study.
Store-operated conductances with a linear I/V relationship and an E rev of ;0 mV were present in freshly isolated cerebral VSMCs from TRPC1 2/2 mice (36), and inhibited by Orai1 small interfering RNA in contractile primary cultured mouse aorta VSMCs (37). There is currently no explanation why these 2 studies differ from the substantial number of studies, which indicate that SOCs in contractile VSMCs are mediated by TRPC1 channels. Recent studies suggesting that Orai1-mediated I crac is expressed in long-term cultured VSMCs with synthetic or proliferative phenotypes may provide explanations (38,39).
We have previously reported that PKC-dependent phosphorylation of TRPC1 proteins is obligatory for gating of TRPC1 SOCs in VSMCs (15,17,18,(22)(23)(24)(25). In addition, PKCa-dependent phosphorylation of TRPC1 hasalsobeenreportedtoregulatestore-operatedCa 2+ entry in endothelial cells (40). However, it is not understood how store depletion is coupled to PKC stimulation, and therefore, this current study explored the possibility that store-operated Gaq-mediated PLC activity coupled to PKC is involved in opening of TRPC1 channels in VSMCs. Stimulation of store-operated whole-cell and single TRPC1 channel activities was prevented by G-protein, PLC, and PKC inhibitors. In addition, anti-Gaq/11 antibodies, but not by anti-Gai1-3 antibodies, inhibited store-operated single TRPC1 channel activity, which implicates a role for Gaq/11 subunits in evoking TRPC1 SOCs. Moreover, PLC and PKC inhibitors reduced store-operated increases in phosphorylation of TRPC1 proteins. PLC and PKC inhibitors also reduced constitutive phosphorylation levels of TRPC1 proteins. Basal PKC phosphorylation may explain why TRPC1 SOCs are activated by agents such as PIP 2 , calmodulin, and MANS peptide in inside-out patches, which are unlikely to contain functional SR Ca 2+ stores to drive Gaq-mediated PLC and PKC activities that are obligatory for channel gating (20,25,41). In future studies, it will be important to identify which PKC isoform is involved in gating TRPC1 channels and reveal which serine/threonine residues reported to be located at the putative pore region and N and C termini are involved (42).
To our knowledge, this is the firsttimethatdepletion of Ca 2+ within SR Ca 2+ stores has been proposed to induce Gaq-mediated PLC activity. Our data clearly show that the well-established store-depletion agents BAPTA-AM,CPA,andTPENallinducedtranslocationofGFP-PLCd1-PH signals from the plasma membrane to the cytosol, which corresponds to stimulation of PLC activity, PIP 2 hydrolysis, and production of IP 3 . In addition, store-operated changes in cellular distribution of GFP-PLCd1-PH signals were reduced by a PLC inhibitor. GFP-PLCd1-PH has previously been used to investigate changesinPIP 2 and IP 3 levels induced by stimulation of Gaq-coupled receptors and associated PLC-mediated signaling because it has a higher affinity for IP 3 over PIP 2 (30)(31)(32)(33). In contrast, other agents that have much greater selectively for PIP 2 over IP 3 are useful for measuring changes in PIP 2 levels regardless of PLC activity, such as Tubby (30)(31)(32)(33). A rise in [Ca 2+ ] i may trigger PLC activity (33); however, this is unlikely to stimulate PLC activity in the present study because BAPTA-AM and TPEN, which reduce or have little effect on [Ca 2+ ] i ,respectively, induced translocation of GFP-PLCd1-PH signals. Because BAPTA-AM, CPA, and TPEN deplete Figure 6. Store depletion evoked associations between TRPC1, Gaq and PLCb1. A) Representative Western blots show that in unstimulated primary cultured rabbit portal vein VSMCs, TRPC1 did not associate with Gaq or PLCb1. BAPTA-AM induced associations between TRPC1 and Gaq, and TRPC1 and PLCb1, which were reduced by transduction of cells with either PLCb1 shRNA1 or shRNA2 sequences. Primary cultured rabbit portal vein cell lysates initially immunoprecipitated (IP) with anti-TRPC1 antibodies were then Western blotted (WB) with anti-Gaq or anti-PLCb antibodies. B) Mean data for relative band intensities of BAPTA-AM-evoked interactions between TRPC1 and Gaq or PLCb1(n = 3, different primary cell culture preparations). Scram., scrambled. *P , 0.05. SR Ca 2+ stores by such distinct actions, it is unlikely that the similar effects of these agents on GFP-PLCd1-PH signals represent nonselective actions.
Transduction of primary cultured VSMCs with 2 distinct PLCb1 shRNAs produced significant reductions in both store-operated whole-cell and single TRPC1 channel activities and also prevented store-operated translocation of GFP-PLCd1-PH signals from the plasma membrane to the cytosol. These findings clearly show that the PLCb1i s oform significantly contributes to store-operated PLC activity in VSMCs. PLCb1 has also been linked to activation of TRPC channels in neurons (27)(28)(29). Previous studies have proposed that PLCg1 has an important role in activation of TRPC1/C4-mediated SOCs in keratinocytes (43) and I crac -like currents in hepatocytes (44). It is thought that PLCg1 enzymatic activity is not involved in activation of these channels; instead, PLCg1 may act as a scaffold protein via its SH-2 domain (43,44). These ideas are similar to those put forward for a role of PLCg1 in agonist-induced Ca 2+ entry (45). U73122 has also been shown to inhibit endogenous I crac -like currents and store-operated Ca 2+ entry in RBL-2H3 cells (46). These studies further emphasize the novelty of the present work, that store-operated PLCb1 enzymatic activity regulates TRPC1 SOCs.
Both coimmunoprecipitation studies and proximity ligation assays showed that store depletion induced interactions between TRPC1, Gaqa n dP L C b1. Proximity ligation assays also identified that these interactions occurred at the plasma membrane and that they are likely to occur within 40 nm of each other (47).
PLCb1 knockdown did not affect the associations between TRPC1 and Gaq, which suggests that these 2 interactions may occur as separate events during the formation of TRPC1-Gaq-PLCb1 signaling complexes. Store depletion did not induce interactions between TRPC1 and PLCg1, which suggests selective associations between TRPC1 and PLCb1.
Our findings clearly show that TRPC1 channel activity induced by the endogenous Gaq-coupled receptor agonist and vasoconstrictor noradrenaline was prevented by knockdown of PLCb1. This suggests that PLCb1-mediated TRPC1 channel activation is likely to be physiologically important. Interestingly, reduction of noradrenalineevoked TRPC1 channel activity by knockdown of PLCb1 was most pronounced at concentrations of noradrenaline between 1 and 100 mM, which may suggest that these concentrations of noradrenaline are coupled to store depletion.
Taken together, our results indicate that in contractile VSMCs, depletion of Ca 2+ within SR Ca 2+ stores forms TRPC1-Gaq-PLCb1 signaling complexes, which leads to increased PLCb1 activity, production of DAG, and stimulation of PKC that induces TRPC1 channel gating (Fig. 8D). What is not yet understood is how store depletion induces formation of these complexes. A potential molecular candidate is STIM1, which is proposed to be involved in activation of overexpressed and endogenous TRPC1 channels through electrostatic and protein-protein interactions between STIM1 and TRPC1, including TRPC1 SOCs in VSMCs (8-13, 19, 38, 39, 48). In future experiments, it may be revealing to investigate if STIM1 and these STIM1 interaction sites mediate interactions with Gaqa n dP L C b1, and also examine whether interactions between STIM1 and TRPC1 lead to dissociation of G proteins into Gaqand Gbg subunits. It is increasingly apparent that STIM1 has diverse cellular partners, including ion channels such as Orai1 (4-7), TRPC channels (8,12,13) and voltagegated Ca 2+ channels (49,50), SR and plasma membrane Ca 2+ -ATPases (51,52), and adenylate cyclases (53). It will be intriguing to investigate if STIM1 coupled to Gaq-mediated PLC activity makes an important addition to this list, and also if Orai proteins have a role in these mechanisms.
We recently proposed an activation model of Gaqcoupled receptor-mediated TRPC1 channels in which interactions between TRPC1, MARCKS, PKC activity and PIP 2 are obligatory partners in channel gating (25). Gaq receptor-mediated phosphorylation of TRPC1 by PKC induced dissociation of the PIP 2 -binding protein MARCKS from TRPC1 and also caused MARCKS to release PIP 2 , which then acted as a gating ligand (25). This finding suggested that MARCKS behaves as a reversible PIP 2 buffer, providing a discrete pool of PIP 2 for channel gating, which is protected from PLC-mediated PIP 2 hydrolysis. It will be interesting to examine if store-operated TRPC1 channel activation by PKC involves similar roles for MARCKS and PIP 2 .
In conclusion, the present work proposes a novel gating pathway of TRPC1 channels; store depletion induces formation of TRPC1-Gaq-PLCb1 complexes at the plasma membrane, which evoke Gaq-mediated PLC activity, PKC stimulation, and channel gating. Interestingly, this pathway will also generate IP 3 , which introduces the intriguing Figure 8. Proposed signal pathway coupling store depletion to activation of TRPC1 channels. A, B) Traces show that bath application of noradrenaline evoked TRPC1 channel activity in a concentration-dependent manner in cell-attached patches held at 280 mV, which were greatly reduced in VSMCs expressing PLCb1 shRNA (B) compared to scrambled shRNA (A). C ) Mean data show the inhibitory effect of PLCb1 shRNA on noradrenaline-induced TRPC1 channel activity (n = 7). **P , 0.01; ***P , 0.001. D) Proposed activation model of TRPC1 channels in VSMCs. In the closed state, TRPC1 does not interact with Gaq and PLCb1. Following Ca 2+ store depletion, TRPC1 forms complexes with Gaq and PLCb1 to cause PIP 2 hydrolysis and formation of DAG, which stimulates PKC activity, phosphorylation of TRPC1 subunits, and channel opening. possibility that TRPC1-mediated Ca 2+ entry, store refilling, and IP 3 -mediated store depletion produce discrete localized Ca 2+ signals that selectively trigger cellular functions such as contraction, secretion, and gene regulation.