Dual action of the Gαq-PLCβ-PI(4,5)P2 pathway on TRPC1/4 and TRPC1/5 heterotetramers

The transient receptor potential canonical (TRPC) 1 channel is widely distributed in mammalian cells and is involved in many physiological processes. TRPC1 is primarily considered a regulatory subunit that forms heterotetrameric channels with either TRPC4 or TRPC5 subunits. Here, we suggest that the regulation of TRPC1/4 and TRPC1/5 heterotetrameric channels by the Gαq-PLCβ pathway is self-limited and dynamically mediated by Gαq and PI(4,5)P2. We provide evidence indicating that Gαq protein directly interacts with either TRPC4 or TRPC5 of the heterotetrameric channels to permit activation. Simultaneously, Gαq-coupled PLCβ activation leads to the breakdown of PI(4,5)P2, which inhibits activity of TRPC1/4 and 1/5 channels.

The TRPC subfamily of the TRP channels consists of seven members. Among those, TRPC1, TRPC4, and TRPC5 channels are classified into a subgroup which can be activated by receptor stimulation 1 . TRPC4 and TRPC5 are expressed in selective tissues, including smooth muscle and neurons, and are non-selective cation channels which can be activated by specific G-protein alpha subunits 2,3 . TRPC1 is the first mammalian TRP channel to be cloned and is ubiquitously expressed in various tissues, but its membrane expression pattern and channel function remain controversial 4 .
Although homotetrameric TRPC1 is controversial about channel function, failure to regulate the expression or mutation of a TRPC1 channel is known to cause diseases. For example, the TRPC1 channel is upregulated in the squamous layer of the Darier's disease patient 5 . It has been observed that tumor necrosis factor alpha exposure increases TRPC1 expression without significantly altering the expression of other TRPC isoforms in human pulmonary artery endothelial cells 6 . It is suggested that the TRPC1 channel is important for adaptation to biomechanical stress and that TRPC1 dysregulation can induce maladaptive cardiac hypertrophy and failure 7 . In TRPC1 knockdown neurons, specific migratory characteristics such as distance covered, locomotion speed, and directionality were increased 8 . In hippocampal neurons from Trplc1/Trpc4/Trpc5-triple-knockout (Trpc1/4/5 −/− ) mice, action potential-triggered excitatory postsynaptic currents were significantly reduced 9 .
Although TRPC1 is closely related to the diseases, its electrophysiological function is not observed. Since TRPC1 channel did not have electrophysiological function, most studies were concentrated on homotetrameric TRPC4 and TRPC5 channels. Recently, the current-voltage (I/V) relationships of the TRPC1/4 and TRPC1/5 channels exhibit an outward rectification, in contrast to the double rectifying TRPC4 and TRPC5 channels 8,10-12 . Thus, the TRPC1 channel has been suggested to be a regulatory subunit rather than a channel itself 12 .
Most TRP channels are regulated by signaling downstream of the Gαβγ timer, which hydrolyzes PI(4,5)P 2 pattern and observed the interactions between TRPC4β/4β, TRPC5/5, TRPC1α/4β, TRPC1α/5, and TRPC4β/5 pairs at the plasma membrane with Förster resonance energy transfer (FRET) (Fig. 1A). HEK293 cells were cotransfected with CFP-or YFP-tagged TRPC1 (CFP-TRPC1α and YFP-TRPC1α), TRPC4 (TRPC4β-YFP and TRPC4β-CFP) and TRPC5 (CFP-TRPC5 and YFP-TRPC5). As negative control experiments, we confirmed that FRET efficiency of CFP-tagged TRPC channels with cytosolic YFP (empty-YFP) or YFP-tagged muscarinic receptor 3 (YFP-M3) were almost zero FRET efficiency. A co-immunoprecipitation (Co-IP) assay provided additional evidence for the formation of heterotetrameric TRPC1α/4β, TRPC1α/5, and TRPC4β/5 ( Supplementary  Fig. S1). When coexpressed with TRPC4β, TRPC1α was successfully targeted to the plasma membrane, which was not the case when it was expressed alone 32 . TRPC1α affected the properties of both TRPC4β and TRPC5 currents. The I/V relationships of the heterotetrameric TRPC channels are substantially different from the doubly rectifying I/V shapes of homotetrameric TRPC4β or TRPC5 channels (Fig. 1C,E). The stimulation of M3 receptors with 100 μM carbachol (CCh) elicited outwardly rectifying currents in HEK293 cells expressing TRPC1α/4β (Fig. 1D) and TRPC1α/5 (Fig. 1F). A receptor-stimulated current was not observed when TRPC1α was expressed alone (Fig. 1B). These results suggest that TRPC1 contributes to the formation of a distinct pore in a complex with TRPC4β or TRPC5, although TRPC1 channel cannot function alone. To exclude the possibility of heterotetrameric TRPC1α/4β or TRPC1α/5 channels mixed together with homotetrameric TRPC4β or TRPC5 channels in our coexpression system, we only recorded the currents from those cells in which the fluorescence intensity of YFP-TRPC1α was at least three times greater than that of TRPC4β-CFP or CFP-TRPC5 at the plasma membrane ( Supplementary Fig. S2). When the fluorescence ratio of YFP-tagged TRPC1α to CFP-tagged TRPC4β or TRPC5 was at least 3, the I/V shapes of outwardly rectifying heterotetrameric TRPC1/4 and TRPC1/5 currents were recorded. However, when the fluorescence ratio of YFP to CFP was less than 3, double rectified or mixed currents were observed. Consequently, we only recorded whole-cell currents in cells with a fluorescence ratio of at least 3 and excluded the contribution of homotetrameric currents. Therefore, our experiments were limited to heterotetrameric TRPC1α/4β and TRPC1α/5 under these conditions. Inactivation of heterotetrameric channels after Gα q -PLC pathway activation. Following GPCR stimulation, TRPC1α/4β and TRPC1α/5 channels show transient activation (Fig. 1D,F,G) 10 . Single channel recording of TRPC1α/5 heteromeric channels showed a transient increase in open probability by CCh stimulation ( Supplementary Fig. S3). Additionally, 100 μM histamine transiently activated currents with an outwardly rectifying I/V curve in the renal cancer A498 cells 33,34 , which express TRPC1 and TRPC4 (Fig. 1H,I). These results suggest that receptor agonists, such as histamine and acetylcholine, transiently activate heterotetrameric TRPC1/4 in native cells, as well as in HEK cells expressing TRPC1α/4β. Therefore, we first examined the molecular mechanism of channel inactivation after GPCR stimulation.
As an alternative method to deal with PI(4,5)P 2 hydrolysis by PLCβ, we used pharmacological tools to activate PLCβ. As well known, the application of PLC activator m-3M3FBS (100 μM) induced PI(4,5)P 2 hydrolysis and diacylglycerol (DAG) production as muscarinic stimulation, whereas there was no response to its inactive analog o-3M3FBS ( Supplementary Fig. S5). To monitor DAG kinetics, we used YFP tagged DAG sensor (YFP-C1AC1A) 37 . After pretreatment of m-3M3FBS and o-3M3FBS, 100 nM EA was used to stimulate the TRPC1α/4β and TRPC1α/5 channels. With m-3M3FBS pretreatment to activate PLCβ, TRPC1α/4β and TRPC1α/5 currents were not activated by EA stimulation (Fig. 2H,I). However, during pretreatment with m-3M3FBS, the currents were slightly reduced. Unlike with m-3M3FBS, with o-3M3FBS pretreatment, TRPC1α/4β and TRPC1α/5 currents were increased by EA stimulation. Therefore, we suggest that the PLC downstream pathway is involved in the inactivation of heterotetrameric TRPC1α/4β and TRPC1α/5 channels but that the upstream pathway of PLCβ is not responsible for the inactivation of heterotetrameric channels.
Next, we postulated that decreasing the PI(4,5)P 2 concentration would reduce the TRPC1α/4β and TRPC1α/5 currents. To confirm the effect of PI(4,5)P 2 on the suppression of channel activity, we added PI(4,5)P 2 directly to the internal solution of the pipette and tested whether added PI(4,5)P 2 reverses the inactivation caused by the active mutant Gα q (Q209L). In HEK293 cells coexpressing heterotetrameric channels and Gα q (Q209L), the current increased for approximately 50 seconds after rupture and then decreased again ( Supplementary Fig. S6A,C and Fig. 2G) when a pipette containing 20 μM of PI(4,5)P 2 was used. The decreased current was not reactivated by EA. This phenomenon occurs because the added PI(4,5)P 2 temporarily restores the activity of the channels, but the expressed Gα q (Q209L) protein hydrolyzes PI(4,5)P 2 . Thus, we used a metabolically stable form of PI(4,5)P 2 , diC8 ms-PIP2, to stably supply PI(4,5)P 2 , even in the presence of Gα q (Q209L). When 20 μM diC8 ms-PI(4,5)P 2 Similarly, activation of the M3 receptor depleted PI(4,5)P 2 in HEK293 cells expressing M3 receptor (Fig. 2J,K). CCh stimulation also led to a decrease of PI(4,5)P 2 and a subsequent increase of DAG even in HEK Heterotetrameric (E) TRPC1α/4β and (F) TRPC1α/5 channels coexpressed with Gα q (WT) or Gα q (Q209L) were stimulated by 100 nM EA at +100 mV. (G) When PI(4,5)P 2 or ms-PI(4,5)P 2 was added or not to the pipette solution, delta currents (EA stimulated current-basal current) of cells expressing heterotetrameric channels and G alpha protein were quantified at 100 mV. Cells expressing (H) TRPC1α/4β or (I) TRPC1α/5 were pretreated with 100 μM m-3M3FBS or o-3M3FBS and stimulated with EA. (J) For cells expressing M3 and YFP-PH and stimulated with 100 μM CCh, the YFP-PH translocation curve was obtained by fitting the Norm data to a singleexponential decay function (black solid curve). The area enclosed by the dashed box is enlarged in the right panel. (K) The line scan shows the YFP-PH intensity along the white dashed line. All data are present as the mean ± SEM. n.s., not significant, *p < 0.05, **p < 0.01, ***p < 0.001. The scale bar represents 10 μm.
Scientific REPORTS | (2018) 8:12117 | DOI:10.1038/s41598-018-30625-0 cells expressing TRPC1α/4β or TRPC1α/5 ( Supplementary Fig. S7A,B). We compared the time-course between PI(4,5)P 2 and DAG with the halftime for PI(4,5)P 2 and DAG. The halftime constant (Τ 1/2 ) for decreased PI(4,5)P 2 level at the plasma membrane was smaller than that for the increased level of DAG at the plasma membrane. The Τ 1/2 for the current inactivation of the heterotetrameric channel was the slowest among three half times. The Τ 1/2 order of current >DAG >PIP2 was observed in all the concentrations ( Supplementary Fig. S7C,D). Furthermore, as the concentration of CCh was higher, Τ 1/2 was the faster. Based on these findings, we thought that PI(4,5)P 2 depletion might be a candidate for the inactivation of TRPC1α/4β and TRPC1α/5 currents. PI(4,5)P 2 is essential for maintaining TRPC1α/4β and TRPC1α/5 activation. To further establish the possible role of PI(4,5)P 2 depletion during channel inactivation, we used two independent methods to deplete PI(4,5)P 2 using two types of phosphatases, rapamycin inducible using Inp54p and VSP, while recording the channel activity. With these methods, we depleted PI(4,5)P 2 only without generating IP 3 or DAG.
The first used expression of danio rerio voltage-sensing phosphatase (DrVSP), a membrane-resident voltage-controllable phosphoinositide phosphatase that dephosphorylates PI(4,5)P 2 to phosphatidylinositol 4-phosphate (PI(4)P) (Fig. 3A), to reduce the endogenous level of PI(4,5)P 2 38,39 . For this, we used the following three time ramp pulse (3TRP) protocol: the current was recorded at a 500-ms ramp from +100 to −100 mV with a holding potential of −60 mV; a brief step to +100 mV for 2 seconds to activate the phosphatase was followed by a 2 nd ramp pulse; the current was recorded by a 3 rd ramp pulse after ~6 seconds of recovery (Fig. 3B). Measurement of PI(4,5)P 2 with CFP-PH showed PI(4,5)P 2 depletion by the +100 mV pulse and recovery after ~6 seconds (Fig. 3C,D). In control experiments, only VSP transfected HEK293 cell currents (Fig. 3F) and the EA-activated TRPC1α/4β and TRPC1α/5 currents were the same at the 1 st , 2 nd , and 3 rd ramps in the absence of DrVSP (Fig. 3G,I). However, in the presence of DrVSP, depletion of PI(4,5)P 2 reduced the TRPC1α/4β and TRPC1α/5 at 2 nd ramp current, respectively, relative to their original values (Fig. 3E,H,J). Addition of ms-PI(4,5) P 2 to pipette solution prevented VSP induced currents depletion (Fig. 3E).
PI(4,5)P 2 was also depleted using the FRB/FKBP rapamycin-inducible system. Supplementary Fig. 8A shows a schematic of the efficient depletion of PI(4,5)P 2 by this system and the translocation of Inp54p-CFP from the cytosol to the membrane and of YFP-PH from the membrane to the cytosol. Treatment with 20 nM rapamycin recruited Inp54p to the plasma membrane, resulting in the time-dependent depletion of PI(4,5)P 2 ( Supplementary Fig. S8B,C). The depletion of PI(4,5)P 2 with Inp54p reduced the EA-induced TRPC1α/4β and TRPC1α/5 currents (Fig. 3K,L). In a further experiment, the currents and movements of Inp54p and YFP-PH were simultaneously recorded. Interestingly, the tendency of EA-induced TRPC1α/4β and TRPC1α/5 currents to decrease was similar to the YFP-PH decrease in the cell membrane ( Supplementary Fig. S8D,E). However, in the absence of the key protein, rapamycin stimulation did not decrease the TRPC1α/4β and TRPC1α/5 currents (Supplementary Fig. S8F,G). When ms-PI(4,5) was added to the pipette solution as CCh stimulation, rapamycin-induced Inp54p did not decrease the currents in heterotetrameric TRPC channels. We analyzed the kinetics of the PI(4,5)P 2 decrease and the current decrease. Fitting the PI(4,5)P 2 depletion time course to a sigmoidal curve revealed that both the PI(4,5)P 2 depletion and the reduction in current had similar time courses, with Τ 1/2 values for YFP-PH, TRPC1α/4β, and TRPC1α/5 of 62.2 ± 6.8 s, n = 7; 67.2 ± 10.9 s, n = 6; and 57.0 ± 16.1 s, n = 6, respectively (Fig. 3M). These results suggest that PI(4,5)P 2 depletion without IP 3 or DAG production induced the current inhibition of heterotetrameric channels.
To establish the relevance of PI(4,5)P 2 hydrolysis for the inactivation observed during physiological GPCR stimulation, instead of the activator EA, we used the PLC inhibitor U73122, which completely inhibits the hydrolysis of PI(4,5)P 2 during receptor stimulation (Supplementary Fig. S9A and B). Pretreating cells with U73122 gradually increased the currents and prevented time-dependent TRPC1α/4β and TRPC1α/5 current inactivation (Fig. 3N,O). In the presence of U73122, CCh still slightly reduced the current initially; however, importantly, CCh persistently increased the ratio (CCh stimulated current/ basal current) to 470% (TRPC1α/4β, n = 8) and 598% (TRPC1α/5, n = 7) (Fig. 3P). However, transient activation was observed during pretreatment with U73433, an inactive analog of U73122 ( Supplementary Fig. S9C,D). Together, these results strongly suggest that PI(4,5)P 2 regulates these channels activity and PI(4,5)P 2 reduction decrease current of TRPC1α/4β and TRPC1α/5 by GPCR stimulation. Activated Gα q directly binds and activates TRPC1α/4β and TRPC1α/5. CCh stimulation activated TRPC1α/4β and TRPC1α/5 and continually increased the currents in the presence of a PLC inhibitor ( Fig. 3N-P), and PLC activation did not induce an increase in heterotetrameric current (Fig. 2H,I). Because activation downstream of PLCβ is independent of channel activation, we postulated that Gα q activates TRPC1α/4β and TRPC1α/5. To understand the activation mechanism of the heterotetramers, we employed the FRB/FKBP rapamycin-inducible system and Gα q (Q209L, L254 A), which reportedly lacks the ability to activate PLCβ 40 . The induction of currents by rapamycin was not observed in the absence of key molecules (Supplementary Fig. S10). As shown in Fig. 4A, RFP-FKBP-Gα q (Q209L, L254A) was translocated to the plasma membrane by 20 nM rapamycin treatment, and the mutant did not hydrolyze PI(4,5)P 2 . Importantly, the same treatment activated TRPC1α/4β and TRPC1α/5 heterotetramers (Fig. 4B,C,F), increasing the possibility that Gα q interacted with and regulated the channels. In addition, the movements of PI(4,5)P 2 and Gα q were recorded simultaneously while recording currents from TRPC1α/4β and TRPC1α/5 (Fig. 4D,E). After normalization, TRPC1α/4β and TRPC1α/5 currents indeed showed similar dynamics to that of the normal translocation of RFP-FKBP-Gα q (Q209L, L254A) to the plasma membrane.
To investigate the interactions between Gα q and the heterotetrameric channels, we performed FRET measurements under receptor stimulation and co-IP experiments. The FRET efficiencies of TRPC1α/4β and TRPC1α/5 with Gα q were initially 2.3 ± 1.7% and 5.4 ± 1.5% and increased to 8.8 ± 1.6% (n = 7) and 15.9 ± 2.2% (n = 7), respectively, after CCh stimulation (Fig. 4G). Co-IP analysis revealed that neither TRPC1α/4β nor TRPC1α/5 interacted with wild-type Gα q but that both interacted with the activated Gα q mutant (Fig. 4H,I). HEK293 cells expressing TRP1/4β-Flag (or TRPC1/5-Flag) and YFP-tagged Gα q (WT) or Gα q (Q209L) were lysed, and the lysates were immunoprecipitated by either a GFP antibody or a Flag antibody, and coprecipitation of TRPC4β (Fig. 4H) and TRPC5 (Fig. 4I) with Gα q was detected with the indicated antibodies (top blots). Because receptor stimulation increased the membrane expression of the homotetrameric TRPC5 channel and increased the current 41 , we investigated whether CCh stimulation increased the cell membrane expression of the heterotetrameric TRPC1α/4β and TRPC1α/5 channels. When changes in the cell membrane expression were observed using a total internal reflection fluorescence (TIRF) microscope, the level of TRPC1α/4β expression did not significantly change with CCh stimulation (Supplementary Fig. S11A). However, in the case of TRPC1α/5, the expression level increased starting at 1 minute after CCh stimulation. In fact, because the transient activity of the current due to the CCh stimulation was a relatively rapid reaction, occurring within 30 seconds, it is difficult to understand how the increase in cell membrane expression caused the current increase. In addition, EA stimulation did not affect changes in the cell membrane expression of heteromeric channels (Supplementary Fig. S11B). In the surface biotinylation experiment, TRPC1α/4β did not react with CCh stimulation, but TRPC1α/5 showed an increase in surface expression 3 minutes after CCh stimulation ( Supplementary Fig. S11C and D). Collectively, these results demonstrated that 1) activated Gα q could activate TRPC1α/4β and TRPC1α/5 through direct binding to these channels and 2) PI(4,5)P 2 depletion reduced the current of heterotetrameric channels.

Discussion
We suggest here that the self-limiting regulation of TRPC1α/4β and TRPC1α/5 channels occurs through the Gα q -PLCβ pathway due to the following findings: (1) The Gα q subunit binds directly to TRPC1α/4β and TRPC1α/5 and activates these channels. (2) A decrease in membrane PI(4,5)P 2 , caused by PLCβ activation, inhibits channel activity. The transient activation of heterotetrameric channels due to Gα q -PLC stimulation was observed not only in an overexpression system but also under general physiological conditions. Even when we activated TRPC1α/4β and TRPC1α/5 channels by EA, PI(4,5)P 2 depletion with VSP or Inp54p inhibited the channels activity. PI(4,5)P 2 was essential for the activation of TRPC1α/4β and TRPC1α/5 channels.
Stimulation of the renal cancer cell line A498 with EA results in an I/V curve with an outwardly rectifying shape for TRPC1/4, in contrast to the double rectifying shape observed for TRPC4α. We also recorded transient activation currents consistent with the I/V shape of TRPC1/4 after histamine stimulation in A498 cells (Fig. 1H). In endothelial cells, the nitrosylation of native TRPC1/5 heterotetrameric channels after G protein-coupled ATP receptor stimulation elicited entry of calcium 42 . Although the mechanism was different, the endogenous TRPC1/5 heterotetrameric channels in secretory fibroblast-like synoviocytes were activated by reduced thioredoxin 43 . In addition, stimulation of Q7 cells, a striatal cell line obtained from wild-type mice, showed an outwardly rectifying I/V shape for TRPC1/5, whereas in Q111 cells, a Huntington's disease cell line obtained from mutant HTT knock-in mice, a doubly rectifying I/V shape was observed for TRPC5 because the expression of TRPC1 was suppressed 44 . Thus, studies on heterotetrameric TRPC1/4 or TRPC1/5 channels rather than homotetrameric TRPC4 or TRPC5 channels might have more pathophysiological relevance 8,43 .
Although TRPC1 does not form a functional homotetrameric ion channel, it is thought that the reason an abnormality in TRPC1 can cause disease is that the TRPC1 channel forms a heterotetrameric channel with TRPC4 and TRPC5. TRPC1 acts as a regulatory subunit that reduces the inward current of the homotetrameric TRPC4 and TRPC5 channels at negative potentials and increases the outward current at positive potentials, thus reducing cellular excitability. Single channel recording of the heterotetrameric channel showed reduced unitary conductance in TRPC1/4 and TRPC1/5 channels. Also, TRPC1 reduced the calcium permeability of TRPC4 or TRPC5 when forming heterotetramers 8,45 . Low levels of TRPC1 increased the formation of homotetrameric TRPC5, a highly Ca 2+ -permeable channel, and stimulated Ca 2+ -dependent apoptosis in Huntington's disease cells 44 .
TRPC channels have been introduced as subunits of receptor-operated channels 46,47 or store-operated channels 48,49 , although the precise mechanism by which the channel operates remains controversial. However, perfusion and pretreatment with 20 μM U73122, which is a PLC inhibitor. (P) Summary of the reduction in the inactivation ratio for the TRPC1α/4β and TRPC1α/5 currents caused by U73122 and U73433 pretreatment. All data are presented as the mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001; the scale bar represents 10 μm.
Interestingly, PI(4,5)P 2 has been shown to be essential for TRPC4 activation 13 . Recently, it has been reported that PLCδ1, rather than PLCβ or PLCγ, is essential for the Gα i -mediated activity of the TRPC4 channel 16 . In fact, PLCδ1 is thought to play a central role in the regulation of the TRPC4 channel by PI(4,5)P 2 and Ca 2+ . The concentrations of calcium, calmodulin, and PKC regulate TRPC4 and TRPC5 16,28,[57][58][59] . Thus, studies on the complex interactions among PI(4,5)P 2 , PLCδ, PKC and calcium and their effects on TRPC1/4 and TRPC1/5 channels are needed.
It has been reported that the TRPC1 channel produces the TRPC1/3 heterotetramer, together with TRPC3, TRPC4, and TRPC5 8,60 . Therefore, it is necessary to study the mechanism of TRPC1/3 heterotetrameric channel activation and why no change in the I/V curve shape is observed, unlike with TRPC1/4 or TRPC1/5 heterotetramers, although TRPC1 decreases calcium permeability. Finally, the phenomenon that Gα q -PLC stimulation slowly increases the cell membrane expression of the TRPC1/5 channel (Supplementary Fig. S10A) also requires future study.

Materials and Methods
Cell culture and transient transfection. cDNA clones and human embryonic kidney (HEK293) cells (ATCC, Manassas, VA) were maintained according to the supplier's recommendations. HEK293 cells were incubated in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% heat-inactivated FBS, penicillin (100 U/ml), and streptomycin (100 μg/ml) at 37 °C in a 5% CO 2 humidified incubator. Cells were seeded in a confocal dish for recording images or a 12-well plate for whole-cell patch clamp recordings. The following day, Electrophysiology. Cells were transferred onto a solution chamber on the stage of an inverted microscope (IX70, Olympus, Japan). Whole cell configuration was used to measure TRPC channel currents in HEK cells as described previously 2,44 . Cells were left for 10-15 min to attach to coverslips. Whole cell currents were recorded using an Axopatch 200B amplifier (Axon Instruments). Patch pipettes were made from borosilicate glass and had resistances of 3-5 MΩ when filled with normal intracellular solutions. The normal Tyrode (NT) contained 135 mM NaCl, 5 mM KCl, 2 mM CaCl 2 , 1 mM MgCl 2 , 10 mM glucose, and 10 mM HEPES with a pH that was adjusted to 7.4 using NaOH. The internal solution contained 140 mM CsCl, 10 mM HEPES, 0.2 mM Tris-guanosine 5′-triphospate, 0.5 mM EGTA, and 3 mM Mg-adenosine 5′-triphosphate with a pH that was adjusted to 7.3 with CsOH. A voltage ramp pulse from +100 mV to −100 mV was applied for 500 ms at a −60 mV holding potential. Experiments were performed at room temperature (18-22 °C). The recording chamber was continuously perfused at a flow rate of 1-2 ml/min. , where E is the intrinsic FRET efficiency when fluorophore-tagged molecules are associated with each other, A b is the fraction of YFP-tagged molecules that are associated with CFP-tagged molecules, and the bracketed term is the ratio of YFP and CFP molar extinction coefficients scaled for the FRET cube excitation filter 63 . We determined this ratio to be 0.094 based on maximal extinction coefficients for ECFP and EYFP 64 and excitation spectra. TIRF imaging. To conduct the plasma membrane-translocation assay of the channel using TIRF imaging, HEK293 cells were plated at 35-mm coverslip bottom dish. Cell were transfected with M3, TRPC1α and YFP tagged TRPC4β or TRPC5 using FuGENE 6 (Promega) transfection reagent according to manufacturer's protocol. Imaging was performed at room temperature with a 100x oil objective on the stage of an IX81 TIRF microscope (Olympus, Tokyo, Japan). Fluorescence images of YFP were taken using a diode laser at 488 nm.

Microscopic image acquisition and FRET measurements.
Western blotting analyses, Co-IP, and surface biotinylation. For Western blotting, cells were seeded in 6-well plates. On the next day, 0.5-2 μg/well of cDNA was transfected into cells using the transfection reagent Lipofectamine 2000 (Invitrogen, U.S.A.) according to the manufacturer's protocol. After transfection for 24 h, the cells were harvested as follows. Lysates were prepared in lysis buffer (0.5% Triton X-100, 50 Tris-Cl, 150 NaCl, 1 EDTA, pH 7.5, [in mM]) by being passed through a 26-gauge needle seven to 10 times after sonication. Lysates were centrifuged at 13,000 × g for 10 min at 4 °C, and the protein concentration in the supernatants was determined. The proteins extracted in sample buffer were loaded onto 8% Tris-glycine SDS-PAGE gels and then subsequently transferred onto a PVDF membrane. The proteins were probed with GFP (Invitrogen), Flag (Sigma), or β-Actin (GeneTex, U.S.A) antibodies for GFP-tagged, Flag-tagged, or housekeeping proteins as indicated.
Surface biotinylation. PBS-washed cells were incubated in 0.5 mg/ml sulfo-NHS-LC-biotin (Pierce, U.S.A.) in PBS for 30 min on ice. Afterward, unreacted biotin was quenched by the addition of 100 mM glycine in PBS. The cells were then processed as described above to prepare cell extracts. Forty microliters of a 1:1 slurry of immobilized avidin beads (Pierce, U.S.A.) were added to 300 μl of the cell lysate containing 500 μg protein. After incubation for 1 h at room temperature, the beads were washed three times with 0.5% Triton-X-100 in PBS, and proteins were extracted in sample buffer. Collected proteins were then analyzed by Western blot.