TRPC6 channels stimulated by angiotensin II are inhibited by TRPC1/C5 channel activity through a Ca²⁺- and PKC-dependent mechanism in native vascular myocytes

Abstract

The present work investigated interactions between TRPC1/C5 and TRPC6 cation channel activities evoked by angiotensin II (Ang II) in native rabbit mesenteric artery vascular smooth muscle cells (VSMCs). In low intracellular Ca²⁺ buffering conditions (0.1 mm BAPTA), 1 nm and 10 nm Ang II activated both 2 pS TRPC1/C5 channels and 15–45 pS TRPC6 channels in the same outside-out patches. However, increasing Ang II to 100 nm abolished TRPC6 activity but further increased TRPC1/C5 channel activity. Comparison of individual patches revealed an inverse relationship between TRPC1/C5 and TRPC6 channel activity suggesting that TRPC1/C5 inhibits TRPC6 channel activity. Inclusion of anti-TRPC1 and anti-TRPC5 antibodies, raised against intracellular epitopes, in the patch pipette solution blocked TRPC1/C5 channel currents but potentiated by about six-fold TRPC6 channel activity evoked by 1–100 nm Ang II in outside-out patches. Bath application of T1E3, an anti-TRPC1 antibody raised against an extracellular epitope, also increased Ang II-evoked TRPC6 channel activity. With high intracellular Ca²⁺ buffering conditions (10 mm BAPTA), 10 nm Ang II-induced TRPC6 channel activity was increased by about five-fold compared to channel activity with low Ca²⁺ buffering. In addition, increasing intracellular Ca²⁺ levels ([Ca²⁺]_i) at the cytosolic surface inhibited 10 nm Ang II-evoked TRPC6 channel activity in inside-out patches. Moreover, in zero external Ca²⁺ (0 [Ca²⁺]_o) 100 nm Ang II induced TRPC6 channel activity in outside-out patches. Pre-treatment with the PKC inhibitor, chelerythrine, markedly increased TRPC6 channel activity evoked by 1–100 nm Ang II and blocked the inhibitory action of [Ca²⁺]_i on TRPC6 channel activity. Co-immunoprecipitation studies shows that Ang II increased phosphorylation of TRPC6 proteins which was inhibited by chelerythrine, 0 [Ca²⁺]_o and the anti-TRPC1 antibody T1E3. These results show that TRPC6 channels evoked by Ang II are inhibited by TRPC1/C5-mediated Ca²⁺ influx and stimulation of PKC, which phosphorylates TRPC6 subunits. These conclusions represent a novel interaction between two distinct vasoconstrictor-activated TRPC channels expressed in the same native VSMCs.

Introduction

Stimulation of plasmalemmal canonical transient receptor potential channels (TRPCs) produces an influx of Na⁺ and Ca²⁺ into cells. In vascular smooth muscle cells (VSMCs) this excitatory response has been associated with several physiological functions such as contraction and cell proliferation (Inoue et al. 2006; Abramowitz & Birnbaumer, 2009).

In VSMCs we have demonstrated that vasoconstrictor G-protein-coupled receptor (GPCR) agonists evoke two distinct classes of TRPC conductances in the same cell which differ in biophysical and molecular properties, and also in activation pathways. For example, in rabbit mesenteric artery VSMCs angiotensin II (Ang II) activates a 2 pS channel which appears to be composed of TRPC1 and TRPC5 subunits (Saleh et al. 2006, 2008). In addition Ang II evokes a conductance of higher unitary values (three sub-conductance levels of 15–45 pS) with TRPC6 channel characteristics (Saleh et al. 2006). Both conductances are linked to AT₁ receptors, phospholipase C (PLC) activation and the production of diacylglycerol (DAG) from hydrolysis of phosphatidylinositol-4,5-bisphosphate (PIP₂). However, subsequently there is divergence of the transduction mechanisms of these two ion channels. DAG stimulates the TRPC1/C5 conductance by a protein kinase C (PKC)-dependent mechanism whereas a PKC-independent action mediates the opening of TRPC6 channels by DAG (Saleh et al. 2006). Indeed, PKC stimulation inhibited TRPC6 channels in this preparation (Saleh et al. 2006). Noradrenaline and endothelin-1 (ET-1) also activated two separate TRPC conductances in, respectively, rabbit portal vein and coronary artery VSMCs although different TRPC isoforms and/or transduction pathways were involved (Inoue et al. 2001; Peppiatt-Wildman et al. 2007; Saleh et al. 2008; Ju et al. 2010).

A significant observation in mesenteric artery was that low Ang II concentrations (1 nm) stimulated TRPC6 but higher concentrations of Ang II (100 nm) only evoked TRPC1/C5 channels (Saleh et al. 2006). This study provided no mechanism for this anomalous concentration-dependent effect of Ang II but it is possible that there might be some interaction between these two separate TRPC conductances.

In the present work, we investigated whether TRPC1/C5 stimulation influences TRPC6 channel activity in mesenteric artery VSMCs. The results reveal a novel physiological mechanism where Ang II-mediated Ca²⁺ influx through TRPC1/C5 channels inhibits TRPC6 channels by a PKC-dependent mechanism.

Methods

Cell isolation

New Zealand White rabbits (2–3 kg) were killed using i.v. sodium pentobarbitone (120 mg kg⁻¹, in accordance with the UK Animals (Scientific Procedures) Act, 1986). Experimental methods were carried out as specified by St George's animal welfare committee and according to the policies of The Journal of Physiology (Drummond, 2009). Single cells were prepared from rabbit mesenteric arteries (1st to 5th order) in physiological salt solution containing (mm): NaCl (126), KCl (6), glucose (10), Hepes (11), MgCl₂ (1.2), CaCl₂ (1.5) and pH to 7.2 with 10 m NaOH. Enzymatic digestion and VSMC isolation were subsequently carried out using methods previously described (Saleh et al. 2006).

Electrophysiology

Whole-cell currents and single channel currents using inside-out and outside-out patch configurations were recorded with an Axopatch 200B amplifier (Axon Instruments, USA) at room temperature (20–23°C) as previously described (Saleh et al. 2006).

To evaluate current–voltage (I–V) relationships of whole-cell recordings and single channel currents, voltage ramps were applied from −100 to +70 mV (0.5 V s⁻¹) from a holding potential of −50 mV and membrane potential was manually altered between −70 mV and +50 mV, respectively. Whole-cell currents were filtered at 500 Hz (−3 db, low pass 8-pole Bessel filter, Frequency Devices, model LP02, Scensys Ltd, Aylesbury, UK) and acquired using a Digidata 1322A and pCLAMP v. 9.0 software at a sample rate of 2 kHz. Records were filtered between 0.1–0.5 kHz and sampled at 1–5 kHz for analysis of TRPC1/C5 channel currents and filtered at 0.5–1 kHz and sampled between 5–10 kHz for analysis of TRPC6 channel currents.

Analysis of TRPC1/C5 and TRPC6 channel activity

Ang II evoked TRPC1/C5 and TRPC6 channel currents in the same patch and in this study we measured open probability (NP_o) values and unitary amplitudes of both conductances. This was possible due to the different conductance values of these two channels, with TRPC1/C5 having a conductance of about 2 pS and TRPC6 having multiple sub-conductance states between 15 and 45 pS (Saleh et al. 2006). Consequently, these two channels have very different channel current amplitudes when recorded at the same membrane potential (see Fig. 1Ab).

Figure 1. Concentration-dependent effect of Ang II on TRPC1/C5 and TRPC6 channel activity in outside-out patches.

Aa, bath application of 1 nm Ang II activated two distinct cation channels at −70 mV. One channel had a small unitary amplitude of about −0.2 pA (labelled TRPC1/C5, see Ab) and the other channel had briefer openings to three much larger amplitudes of −1 to −3 pA (labelled TRPC6, Ab). Inward cation channel openings are shown as downward current deflections. Continous lines represent closed levels and dashed lines represent open levels. Aa and c, with 100 nm Ang II, TRPC6 channel openings were abolished but the activity of TRPC1/C5 channels was increased. Ba, amplitude histogram of channel currents recorded in (Ab) using a 0.05 pA bin width was fitted with four Gaussian curves representing one closed and the three open levels of TRPC6 of about −1 pA, −2 pA and −3 pA. Bb, amplitude histogram of channel currents recorded in (Ab), with a bin width of 0.01 pA to increase resolution, was fitted with four Gaussians representing one closed state, two open levels of TRPC1/C5 with a unitary amplitude of about −0.2 pA and also the lowest sub-conductance amplitude of TRPC6 at about −1 pA. Note that the number of observations are the same between 0 pA and −1.6 pA in Ba and b, but as the bin width used in Bb provides greater resolution than Ba the number of observations per peak amplitude was smaller. Bc, amplitude histogram of channel currents induced by 100 nm Ang II, with a bin width of 0.01 pA, shows that only TRPC1/C5 channel openings were present. C, concentration-dependent effect of Ang II on TRPC1/C5 and TRPC6 channel activity. Da, relationship between 10 nm Ang II-evoked TRPC1/C5 NP_o and TRPC6 NP_o from the same outside-out patches showing an inverse relationship in which an increase in TRPC1/C5 channel activity is related to a reduction in TRPC6 channel activity. Db, semi-log plot of data shown in Da illustrating a significant inverse correlation between TRPC6 and TRPC1/C5 NP_o values.

Single channel current amplitudes of TRPC1/C5 and TRPC6 were calculated from idealised traces of at least 30 s in duration using the 50% threshold method and analysed using pCLAMP v. 9.0 software. TRPC1/C5 events lasting for <1.32–6.64 ms and TRPC6 events lasting for <0.64–1.32 ms (×2 rise time of filtering used, see above) were excluded, to maximise the number of channel openings reaching their full current amplitude. Figure 1Ab shows that at −70 mV the unitary channel current amplitude of TRPC1/C5 was about −0.2 pA, and thus a 50% threshold level was generated at about −0.1 pA by pCLAMP v. 9.0 software. In contrast, the channel current amplitude of the lowest sub-conductance level of TRPC6 was about −1 pA, which produced a 50% threshold level of about −0.5 pA that is approximately five-fold greater threshold value than the value for openings of TRPC1/C5 channels.

NP_o was used as a measure of channel activity and was calculated automatically by pCLAMP 9. To calculate NP_o values for TRPC1/C5 and TRPC6 channel activity activated by Ang II in the same patches, two idealised traces were created. One trace was produced using 50% threshold levels for TRPC1/C5 channel activity and the other trace was created using 50% threshold levels for TRPC6 channel activity. Thus, TRPC1/C5 channel openings were excluded from traces measuring TRPC6 NP_o due to the five-fold greater threshold values used (see above). Distinction between TRPC1/C5 and TRPC6 channel NP_o values was also assisted by the low activity of these two channel subtypes induced by 1 and 10 nm Ang II (Fig. 1C), which reduced the possibility of TRPC1/C5 channel activity contaminating TRPC6 channel activity.

Single channel current amplitude histograms were plotted from event data created from idealised traces. To resolve unitary TRPC6 current amplitudes, amplitude histograms were created using a bin width of 0.05 pA, and to provide greater resolution for analysing smaller TRPC1/C5 current amplitudes a bin width of 0.01 pA was used. Histograms were fitted with Gaussian curves, with the peak of these curves determining the unitary amplitude of the single channel currents.

Figure preparation was carried out using MicroCal Origin software 6.0 (MicroCal Software Inc., MA, USA) where inward single cation channel openings are shown as downward deflections.

Immunoprecipitation and Western blotting

Dissected tissues were flash frozen and stored in 10 mm Tris-HCl (pH 7.4) at −80°C for subsequent use. Briefly (see Ju et al. 2010), proteins from total cell lysate were extracted and then immunoprecipitated using anti-TRPC6 antibodies (expected molecular mass of about 120 kDa) with a Millipore (MA, USA) Catch and Release kit followed by one-dimensional protein gel electrophoresis (10–20 μg of total protein per lane). Separated proteins were transferred onto PVDF membranes, and then membranes were incubated with a primary antibody raised against phosphorylated serine/threonine residues (1 μg ml⁻¹) at 4°C overnight. Visualisation was carried out using a horseradish peroxidase-conjugated secondary antibody (80 ng ml⁻¹) and enhanced chemiluminescence (ECL) reagents (Thermofisher Scientific, MA, USA) for 1 min and exposure to photographic films. Quantitative analysis of unstimulated and Ang II-evoked phosphorlyated TRPC6 proteins was carried out using densitometry, measured relative to control bands using Adobe Photoshop software (San Jose, CA, USA). Data shown represent n values of at least three separate experiments.

Solutions and drugs

The external solution used in whole-cell recording and outside-out patches and in the patch pipette solution used in inside-out patch experiments contained (mm): NaCl (126), CaCl₂ (1.5), Hepes (10) and glucose (11), pH to 7.2 with 10 m NaOH. Nicardipine (5 μm), DIDS (100 μm) and niflumic acid (100 μm) were included to block voltage-dependent Ca²⁺ channels (VDCCs) and Ca²⁺-activated and swell-activated Cl⁻ conductances. In 0 mm[Ca²⁺]_o external solution, 1.5 mm CaCl₂ was omitted and 1 mm BAPTA was added (<10 nm free Ca²⁺).

The patch pipette solution used for whole-cell recording and outside-out patches, and the external (intracellular) solution used in inside-out patches was K⁺-free and contained (mm): CsCl (18), caesium aspartate (108), MgCl₂ (1.2), Hepes (10), glucose (11), Na₂ATP (1), NaGTP (0.2), pH 7.2 with Tris. Intracellular free Ca²⁺ concentration was set at 100 nm using different concentrations of CaCl₂ and BAPTA (4.8 mm CaCl₂+ 10 mm BAPTA or 0.048 mm CaCl₂+ 0.1 mm BAPTA as calculated using EQCAL software).

Polyclonal TRPC1, TRPC4, TRPC5, TRPC6 and TRPC7 antibodies were generated (Genscript, NJ, USA) against peptide sequences from putative intracellular regions which were previously characterized by Goel et al. (2002). T1E3, an anti-TRPC1 antibody raised against a peptide sequence from a putative extracellular region of TRPC1 (Xu & Beech, 2001), was also generated (Genscript). Anti-phosphorylated serine and threonine and anti-β-actin antibodies were purchased from Santa Cruz (USA). All drugs were purchased from Calbiochem (UK), Sigma (UK) or Tocris (UK) and agents were dissolved in distilled H₂O or DMSO (0.1%). DMSO alone had no effect on channel activity. Data values are mean of n cells ±s.e.m. Statistical analysis was carried out using paired (comparing effects of agents on the same cell) or unpaired (comparing effects of agents between cells) Student’s t test with the level of significance set at P < 0.05.

Results

Concentration-dependent effect of Ang II on TRPC1/C5 and TRPC6 channel activity in outside-out patches from rabbit mesenteric artery VSMCs

Previously we demonstrated using cell-attached recording in rabbit mesenteric artery VSMCs that 1 nm Ang II evoked TRPC6 channels but that 100 nm only stimulated TRPC1/C5 channels (Saleh et al. 2006). However, in that study the analysis did not show whether 1 nm Ang II also evoked TRPC1/C5 channels or provide an explanation for the concentration-dependent action of Ang II on channel activity. In order to investigate a possible interaction between TRPC6 and TRPC1/C5 channels, the optimal approach was to use isolated patches rather than cell-attached recording. Therefore in the first experiments we investigated the effect of various Ang II concentrations on both TRPC1/C5 and TRPC6 channel activity in outside-out patches. In these studies we used low Ca²⁺ buffering patch pipette solution (0.1 mm BAPTA + 100 nm free Ca²⁺) to mimic physiological conditions.

Figure 1Aa, b and B show that bath application of 1 nm Ang II to an outside-out patch held at −70 mV evoked channel activity that was composed of two distinct channel subtypes. One conductance had a small unitary amplitude channel (∼−0.2 pA, Fig. 1Ab and Bb) and the other conductance had larger amplitude levels (∼−1 to −3 pA, Fig. 1Ab and Ba.

Figure 1Ba shows that the amplitude histogram of 1 nm Ang II-evoked channel currents illustrated in Fig. 1Ab created using a 0.05 pA bin width could be fitted by the sum of four Gaussian curves, which represented one closed level and three conductance levels with amplitudes of about −1 pA, −2 pA and −3 pA, and corresponded to TRPC6 sub-conductances states of 15 pS, 27 pS and 45 pS (data not shown, for discussion of criteria used to define sub-conductance states see Albert et al. 2003; Saleh et al. 2006). In contrast, Fig. 1Bb shows that same events created using a 0.01 pA bin width, to provide greater resolution (see Methods), revealed peak amplitudes of the smaller channel currents, which could be fitted by the sum of three Gaussian curves representing one closed and two open levels with unitary amplitudes of about −0.2 pA that corresponded to the TRPC1/C5 unitary conductance of 2 pS (data not shown, Saleh et al. 2006, 2008). The lowest TRPC6 sub-conductance level (−1.01 pA) is also seen in the histogram.

Figure 1Aa, c and C show that increasing the concentration of Ang II from 1 nm to 100 nm increased TRPC1/C5 channel activity but abolished TRPC6 channel activity. Figure 1C illustrates the concentration-dependent effect of Ang II on TRPC1/C5 and TRPC6 mean NP_o values, and shows that 1 nm and 10 nm Ang II activated both channel subtypes to the same extent and that 100 nm Ang II further increased TRPC1/C5 NP_o but completely inhibited TRPC6 channel activity in all preparations.

It is possible that TRPC6 channel activity may contaminate TRPC1/C5 NP_o values in patches containing both channel subtypes and therefore we measured Ang II-induced TRPC1/C5 channel activity in the presence of anti-TRPC6 antibodies. Inclusion of anti-TRPC6 antibodies (1:200), raised against an intracellular epitope, in the patch pipette solution blocked 1 nm and 10 nm Ang II-induced TRPC6 channel activity in outside-out patches (data not shown, see Saleh et al. 2006). In the presence of anti-TRPC6 antibodies, the mean NP_o value of TRPC1/C5 channel activity induced by 1 nm and 10 nm Ang II was respectively 0.07 ± 0.01 (n= 8) and 0.17 ± 0.02 (n= 8), which is virtually identical to the mean NP_o of TRPC1/C5 channel activity of 0.05 ± 0.02 (1 nm Ang II, n= 6) and 0.14 ± 0.03 (10 nm Ang II, n= 6) in control patches (Fig. 1C). These results indicate that Ang II-induced TRPC6 channel activity does not significantly contaminate TRPC1/C5 channel NP_o.

Even though Fig. 1C indicates that 1 nm and 10 nm Ang II activated TRPC1/C5 and TRPC6 channel activity with similar mean NP_o values, it was apparent that there was considerable variability between the activities of these two channel subtypes in individual patches. Figure 1Da and b show this variability in a graph of NP_o values of TRPC1/C5 versus TRPC6 channel activities evoked by 10 nm Ang II in the same outside-out patches. The graphs shown in Fig. 1D illustrate a significant inverse relationship between TRPC1/C5 and TRPC6 NP_o values, with increased TRPC1/C5 channel activity correlated to lower levels of TRPC6 channel activation. An inverse relationship between TRPC1/C5 and TRPC6 activities was also found with 1 nm Ang II (data not shown).

These data suggest that increased TRPC1/C5 channel activity leads to TRPC6 channel inhibition.

Inhibition of TRPC1/C5 channel activity potentiates Ang II-evoked TRPC6 channel activity

To investigate an interaction between TRPC1/C5 and TRPC6 channel activities, we studied the effects of blocking TRPC1/C5 channel currents with anti-TRPC1 and anti-TRPC5 antibodies on TRPC6 NP_o. Low Ca²⁺ buffering solution containing 0.1 mm BAPTA Ca²⁺ buffering and 100 nm free Ca²⁺ was used, and Ang II was bath applied 5 min after obtaining outside-out patches to allow anti-TRPC antibodies to dialyse throughout the isolated patch.

Figures 2Aa, b and C and 5B show that inclusion of anti-TRPC1 (1:200 dilution) in the pipette solution significantly increased the mean NP_o value of TRPC6 channel activity induced by 1 nm, 10 nm and 100 nm Ang II by about six-fold at −70 mV. In addition, Fig. 2Aa, c and C illustrate that inclusion of anti-TRPC5 antibodies (1:200 dilution) in the pipette solution also significantly potentiated mean NP_o values of Ang II-evoked TRPC6 channel activity by six-fold. In the presence of anti-TRPC1 or anti-TRPC5 antibodies, 2 pS TRPC1/TRPC5 channel activity was not observed (compare high resolution records in Fig. 2Aa and b). I–V relationships of channel activity potentiated by anti-TRPC1 and anti-TRPC5 antibodies had three sub-conductances states between 15 and 45 pS with a reversal potential of 0 mV (data not shown), i.e. they are TRPC6 channels. Pre-incubation of anti-TRPC1 and anti-TRPC5 antibodies with their corresponding antigenic peptide (Fig. 2C), and also inclusion of anti-TRPC4 (1:200) and anti-TRPC7 (1:200) antibodies in the pipette solution, had no effect on Ang II-evoked TRPC6 channel activity (data not shown).

Figure 2. Inhibition of TRPC1/C5 channel activity produces an increase in TRPC6 channel activity in outside-out patches.

Aa, control response showing that 10 nm Ang II evokes both TRPC1/C5 and TRPC6 channel activity (see inset) at −70 mV. Ab, following inclusion of anti-TRPC1 or anti-TRPC5 (Ac) antibodies at 1:200 dilutions in the patch pipette, 10 nm Ang II induced much greater TRPC6 channel activity than in control conditions. The inset in Ab shows that anti-TRPC1 antibodies blocked Ang II-induced TRPC1/C5 channel activity. B, co-application of the extracellular anti-TRPC1 antibody T1E3 at 1:100 dilution potentiated 10 nm Ang-evoked TRPC6 channel activity. C, mean data showing that anti-TRPC1, anti-TRPC5 and T1E3 antibodies significantly increased 10 nm Ang II-evoked TRPC6 channel activity. The potentiating actions of these antibodies were prevented by pre-treatment with their antigenic peptides. Each value is from at least n= 6, *P < 0.05.

Figure 5. TRPC6 is inhibited by TRPC1/C5 channel activity via a Ca²⁺ and PKC-dependent mechanism.

Aa, control response to 10 nm Ang II and b, pre-treatment with 3 μm chelerythrine produced a pronounced increase in 10 nm Ang II-evoked TRPC6 channel activity in outside-out patches at −70 mV. B, mean data showing the effect of chelerythrine and anti-TRPC1 antibodies on the concentration–response curve of Ang II-evoked TRPC6 channel activity. C, immunoprecipitation (IP) of total tissue cell lysate protein with anti-TRPC6 antibodies followed by Western blotting (WB) with anti-phosphorylated Ser/Thr antibodies showing that TRPC6 proteins are phosphorylated at rest and increased by 10 nm Ang II. Ang II-evoked phosphorylation of TRPC6 proteins is prevented in a 0 [Ca²⁺]_o conditions, b by pre-treatment with T1E3 (1:100) and c pre-treatment with 3 μm chelerythrine. Cb, Inhibitory action of T1E3 was prevented by pre-incubation with its antigenic peptide (1:50). Lower panels in Ca, b and c and d and e show that the experimental conditions did not alter total β-actin or TRPC6 protein levels. D, mean data of the inhibitory action of 0 [Ca²⁺]_o, T1E3 (1:100) and 3 μm chelerythrine on intensity of Ang II-evoked phosphorylated TRPC6 proteins, measured relative to control bands. Each value represents at least n= 3, *P < 0.05, n.s. not significant.

Figure 2B and C illustrate that bath application of T1E3 at 1:100 dilution, an anti-TRPC1 antibody raised against a putative extracellular epitope (Xu & Beech, 2001), also potentiated mean NP_o values of TRPC6 channel activity activated by 10 nm Ang II by six-fold in outside-out patches. This effect was not observed when T1E3 was pre-incubated with its antigenic peptide (Fig. 2C).

In a further series of experiments, we used a different protocol in which channel activity was first evoked by Ang II in the cell-attached configuration and then subsequently the patch was excised into the inside-out patch conformation and antibodies were applied to the cytosolic surface of the membrane, i.e. channels were activated prior to exposure to antibodies. Figure 3A and B show that bath application of anti-TRPC1 or anti-TRPC5 antibodies (1:200 dilution) significantly augmented mean NP_o of Ang II-evoked TRPC6 channel activity in inside-out patches by about five-fold. Pre-incubation of these anti-TRPC antibodies with their antigenic peptides prevented the potentiation of Ang II-induced TRPC6 channel activation, and anti-TRPC4 and anti-TRPC7 antibodies also had no effect on Ang II-evoked TRPC6 channel activity (Fig. 3B).

Figure 3. Inhibition of TRPC1/C5 channel activity augmented TRPC6 channel activity in inside-out patch and whole-cell recording configurations.

A, bath application of a anti-TRPC1 or b anti-TRPC5 antibodies to the cytosolic surface of inside-out patches at −70 mV potentiated TRPC6 channel activity which was initially activated by 10 nm Ang II in cell-attached patches (see text). B, mean data showing anti-TRPC1 and anti-TRPC5 antibodies significantly potentiated TRPC6 channel activity, which was prevented by pre-treatment with their antigenic peptides. Note that anti-TRPC4 and anti-TRPC7 antibodies had no effect on Ang II-evoked TRPC6 channel activity. Each value was from at least n= 6, P < 0.05. C, bath application of 10 nm Ang II evoked a whole-cell cation current, at a holding potential of −50 mV, which was increased in amplitude after co-application of T1E3 at 1:100 dilution. Vertical deflections represent applied voltage ramps between −100 mV and +70 mV (see Methods) with individual current responses to these ramps shown below the long-term recording. D, mean I–V relationships of 10 nm Ang II-evoked whole-cell cation currents showing that T1E3 increased mean current amplitudes at all membrane potentials and that the action of T1E3 was prevented by pre-treatment with its antigenic peptide.

Figure 3C and D illustrate that bath application of the extracellular anti-TRPC1 antibody T1E3 at 1:100 dilution significantly increased the mean amplitude of 10 nm Ang II-evoked whole-cell cation currents by about five-fold at −50 mV. Pre-incubation of T1E3 with its antigenic peptide prevented the increase in Ang II-evoked whole-cell conductance by T1E3 (Fig. 3D).

These results indicate that inhibiting TRPC1/C5 channel activity potentiates TRPC6 channel stimulation and in the next series of experiments we investigated the mechanism underlying this interaction.

Inhibition of TRPC6 channel activity is mediated by Ca²⁺ influx through TRPC1/C5 channels

Native TRPC1/C5 channels in neurones and VSMCs and expressed TRPC1/C5 channels in cell lines are thought to have a significant permeability to Ca²⁺ with a permeability ratio of Ca²⁺ to Na⁺ (P_Ca/P_Na) of about 50 (Strubing et al. 2001; Alfonso et al. 2008; Saleh et al. 2008), whereas TRPC6 channels are thought to have limited Ca²⁺ permeability (Estacion et al. 2006). Moreover, changes in intracellular Ca²⁺ concentration ([Ca²⁺]_i) are proposed to have both excitatory and inhibitory actions on TRPC6 channel activity (Shi et al. 2004). We examined the possibility that influx of Ca²⁺ through TRPC1/C5 channels inhibits TRPC6 activity. In the first experiments we compared the effects of different [Ca²⁺]_i buffering by BAPTA on TRPC6 channel stimulation by Ang II in outside-out patches.

Figure 4Aa and b show that increasing intracellular BAPTA concentration from 0.1 mm to 10 mm in the pipette solution greatly increased Ang II-evoked TRPC6 mean NP_o values from, respectively, 0.14 ± 0.03 (n= 6) to 0.64 ± 0.08 (n= 10, P < 0.001). With 10 mm BAPTA, 10 nm Ang II-evoked TRPC1/C5 channel currents were observed (Fig. 4Ab, inset i) with a mean NP_o value 0.22 ± 0.08 (n= 10), which was similar to the mean NP_o value of 0.14 ± 0.05 mm (n= 6) obtained with 0.1 mm BAPTA (Figs 1B and 4Aa). With 10 mm BAPTA, bath application of T1E3 (1:100) blocked TRPC1/C5 (Fig. 4Ab, inset ii) but had no further effect on TRPC6 channel activity with mean NP_o values of 0.70 ± 0.09 (n= 6, Fig. 4Ab).

Figure 4. TRPC1/C5-mediated Ca²⁺ influx inhibits TRPC6 channel activity.

Aa, control response to 10 nm Ang II with 0.1 mm BAPTA in the pipette solution in outside-out patches at −70 mV. Ab, response with 10 mm BAPTA in the pipette solution to 10 nm Ang II. TRPC6 channel activity was increased with 10 mm BAPTA but TRPC1/C5 channel activity was similar with 0.1 mm and 10 mm BAPTA. Addition of T1E3 (1:100) blocked TRPC1/C5 channel activity but produced no further increase in TRPC6 channel activity. Ba, increasing [Ca²⁺]_i from 50 nm to 200 nm inhibited TRPC6 channel activity in an inside-out patch at −50 mV, which was previously induced by 10 nm Ang II in a cell-attached patch. Bb, pre-treatment with 3 μm chelerythrine for 5 min prevented the inhibition of Ang II-evoked TRPC6 channel activity by increased [Ca²⁺]_i. C, concentration–response curve of the inhibitory effect of [Ca²⁺]_i on Ang II-evoked TRPC6 channel activity in inside-out patches at −50 mV, with 113 nm[Ca²⁺]_i producing 50% inhibition. Note that chelerythrine and removal of ATP from the intracellular solution prevented the inhibitory effect of [Ca²⁺]_i on Ang II-evoked TRPC6 channel activity. Each point is from at least n= 6, *P < 0.05. D, in 0 [Ca²⁺]_o conditions, bath application of 100 nm Ang II activated both TRPC1/C5 and TRPC6 channel activity. It should be noted that in 1.5 mm[Ca²⁺]_o 100 nm Ang II did not activate TRPC6 channel activity (see Figs 1B and 5B).

We further tested the effect of [Ca²⁺]_i on TRPC6 channel activity in inside-out patches, which had been previously evoked by 10 nm Ang II in cell-attached patches (see above). Figure 4Ba and C show that increasing free Ca²⁺ levels at the cytosolic surface of inside-out patches reduced Ang II-induced TRPC6 channel activity, with a [Ca²⁺]_i producing 50% inhibition (IC₅₀) of 113 nm.

Our hypothesis was that Ca²⁺ influx through TRPC1/C5 channels inhibited TRPC6 channel activity, and therefore we tested the effect of removing Ca²⁺ from the external bathing solution (0 [Ca²⁺]_o+ 1 mm BAPTA, <10 nm free Ca²⁺, see Methods) on Ang II-evoked TRPC6 channel activity in outside-out patches. In 1.5 mm[Ca²⁺]_o, 100 nm Ang II only stimulated TRPC1/C5 channel activity (see Fig. 1A and B) but Fig. 4D shows that in the presence of 0 [Ca²⁺]_o bath application of 100 nm Ang II activated both TRPC1/C5 and TRPC6 channels with similar mean NP_o values of 0.28 ± 0.04 (n= 9) and 0.32 ± 0.03 (n= 9), respectively. As previously shown, in 0 [Ca²⁺]_o Ang II-evoked TRPC1/C5 and TRPC6 channel activity had, respectively, unitary conductances of about 7 pS and about 13 pS (data not shown but see Inoue et al. 2001; Albert & Large, 2002).

TRPC1/C5-induced inhibition of TRPC6 channel activity is mediated by Ca²⁺-dependent PKC phosphorylation of TRPC6 proteins

The present results indicate that during stimulation by Ang II, influx of Ca²⁺ through TRPC1/C5 channels leads to inhibition of TRPC6 channel activity. It is established that PKC has a profound inhibitory action on expressed TRPC6 channel activity (Shi et al. 2004), and that Ca²⁺-dependent PKC isoforms are expressed in VSMCs (Salamanca & Khalil, 2005). Therefore, we studied the effect of PKC inhibitors on Ang II-induced TRPC6 channel activity and also on Ang II-evoked phosphorylation of TRPC6 proteins.

Figure 4Bb and C show that pre-treatment with the PKC inhibitor chelerythrine completely prevented intracellular Ca²⁺-dependent inhibition of 10 nm Ang II-evoked TRPC6 channel activity in inside-out patches. In addition, Fig. 5A and B also show that pre-treatment with 3 μm chelerythrine greatly increased TRPC6 channel activity induced by 1 nm, 10 nm and 100 nm Ang II in outside-out patches. Ang II (10 nm)-evoked channel activity initially activated in the presence of 3 μm chelerythrine in cell-attached patches was greatly inhibited by 92 ± 1% (n= 6) by anti-TRPC6 antibodies (1:200) following patch excision into the inside-out configuration. It should be noted that 3 μm chelerythrine blocked stimulation of TRPC1/C5 channel activity by Ang II in both inside-out and outside-out patches as previously described for cell-attached patches from mesenteric artery VSMCs (Saleh et al. 2006, 2008), and therefore TRPC1/C5 channel activity does not contaminate TRPC6 channel activity induced in the presence of chelerthyrine. Figure 4C shows that removal of ATP from the intracellular solution prevented the inhibitory effect of 200 nm Ca²⁺ on 10 nm Ang II-induced TRPC6 channel activity, which indicates that a phosphorylation process mediates the inhibitory action of [Ca²⁺]_i on TRPC6 channel activity.

Figure 5Ca, b and c show Western blots using anti-phosphorylated serine/threonine antibodies following immunoprecipiation with anti-TRPC6 antibodies, which illustrate that TRPC6 proteins (shown as a single band at a molecular weight of about 120 kDa) contain phosphorylated serine/threonine residues in un-stimulated conditions. In addition, Fig. 5Ca, b, c and D also show that the levels of phosphorylated serine/threonine residues in TRPC6 proteins was increased by 10 nm Ang II. Moreover, Fig. 5Ca, b, c and D show that Ang II-evoked serine/threonine phosphorylation of TRPC6 proteins is reduced in the presence of 0 [Ca²⁺]_o, by pre-treatment with the anti-TRPC1 antibody T1E3 (1:100) and pre-treatment with 3 μm chelerythrine, respectively. Figure 5Cb and D illustrate that the inhibitory action of T1E3 on Ang II-induced phosphorylation of TRPC6 proteins was prevented by pre-incubation of T1E3 with its antigenic peptide. The lower panels of Fig. 5Ca, b, c and d) show control experiments which illustrate that the experimental conditions did not have any effect on total β-actin or TRPC6 protein levels.

These results show that Ang II evokes PKC-dependent phosphorylation of TRPC6 proteins which requires influx of Ca²⁺ through TRPC1/C5 channels.

Discussion

The present data show that stimulation of rabbit mesenteric artery VSMCs by physiological concentrations of Ang II evokes an influx of Ca²⁺ through native TRPC1/C5 channels which inhibits a separate Ang II-induced TRPC6 conductance. The reduction in TRPC6 channel activity is produced by Ca²⁺-dependent PKC-mediated phosphorylation of TRPC6 subunits. These results represent a novel interaction between two separate native TRPC channels in the same cell that will have an important effect on physiological responses such as vasoconstriction and cell proliferation.

Concentration effect of Ang II on TRPC1/C5 and TRPC6 channel activity

With 1–10 nm Ang II there was a concentration-dependent increase in both TRPC1/C5 and TRPC6 channel activity. However, 100 nm Ang II produced a further rise in TRPC1/C5 NP_o value but abolished TRPC6 channel activity. Furthermore, comparison of different patches stimulated with 10 nm Ang II revealed an inverse relationship between TRPC1/C5 and TRPC6 channel activity. Together these data suggest that increasing TRPC1/C5 channel stimulation inhibits TRPC6 channel activity.

Using different configurations of membrane patch recording it was shown with all concentrations of Ang II (1–100 nm) that blocking TRPC1/C5 channel currents with intracellular anti-TRPC1 and anti-TRPC5 antibodies markedly enhanced TRPC6 NP_o whilst inhibiting TRPC1/C5 channel activity. The extracellular anti-TRPC1 antibody T1E3 produced similar data and the effects of the antibodies were completely prevented by pre-incubation with their respective antigenic peptides. The anti-TRPC antibodies did not evoke channel activity in the absence of Ang II. Overall these data support the proposal that stimulation of TRPC1/C5 channels limits TRPC6 channel activity.

Role of Ca²⁺ ions and PKC in inhibiting TRPC6 activity

There are several observations implicating a role of Ca²⁺ in TRPC1/C5 decreasing TRPC6 activation. First, increasing intracellular BAPTA from 0.1 to 10 mm produced an increase in Ang II-evoked TRPC6 NP_o similar to that produced by anti-TRPC1 and anti-TRPC5 antibodies. However, with 10 mm BAPTA Ang II-evoked TRPC1/C5 channel currents were observed and the NP_o was similar to that observed with 0.1 mm BAPTA. Therefore increasing intracellular BAPTA concentrations to increase Ca²⁺ buffering did not alter Ang II-evoked TRPC1/C5 NP_o but TRPC6 stimulation was greatly increased. When T1E3 was applied extracellularly with 10 mm BAPTA in the pipette solution, the Ang II-evoked TRPC1/C5 conductance was inhibited but there was no further increase in TRPC6 activity. Consequently, the inhibition of TRPC6 is not simply due to TRPC1/C5 channels opening and closing but due to an increase of [Ca²⁺] at the cytoplasmic surface of the membrane.

With isolated patches it is unlikely that significant amounts of sarcoplasmic reticulum are attached to the membrane and therefore the effective increase in [Ca²⁺]_i is likely to be caused by influx of Ca²⁺. In agreement with this hypothesis, removal of extracellular Ca²⁺ permitted stimulation of TRPC6 channels by 100 nm Ang II, which was not observed in 1.5 mm[Ca²⁺]_o. Activation of TRPC1/C5 channels by 100 nm Ang II was not altered in 0 [Ca²⁺]_o. Therefore influx of Ca²⁺ ions through TRPC1/C5 channels inhibits TRPC6 by an action on the inner surface of the cell membrane.

It was demonstrated that TRPC6 NP_o was decreased by increasing [Ca²⁺]_i and that this inhibitory effect of Ca²⁺ was blocked by the PKC inhibitor chelerythrine. In the absence of ATP in the intracellular solution, increasing [Ca²⁺]_i no longer reduced TRPC6 NP_o indicating that a phosphorylation process mediated the inhibitory action of [Ca²⁺]_i. Chelerythrine also markedly potentiated Ang II-evoked stimulation of TRPC6 but blocked TRPC1/C5 channel activity (see also Saleh et al. 2006). In co-immunoprecipitation studies we showed that Ang II increased phosphorylation of TRPC6 subunits and that this effect was reduced by chelerythrine, 0 [Ca²⁺]_o and by T1E3. These data are consistent with a model in which influx of Ca²⁺ ions through TRPC1/C5 stimulates PKC which phosphorylates TRPC6 subunits to inhibit channel opening. In expression systems, PKC activation produces potent inhibition of whole-cell TRPC6 currents and TRPC6-mediated Ca²⁺ influx (Shi et al. 2004). Moreover, PKC has also been shown to phosphorylate serine 768 and serine 714 residues in TRPC6A and TRPC6B isoforms, respectively (Kim & Saffen, 2005). The present and previous studies do not elucidate how PKC-dependent phosphorylation of TRPC6 subunits leads to channel inhibition, but it may involve PIP₂. Previously it has been shown that PKC-mediated phosphorylation of TRPC1 facilitates activation of several TRPC1 isoforms by PIP₂ and phosphatidylinositol-3,4,5-trisphosphate (PIP₃, Saleh et al. 2009a,b;). If PKC stimulation also increases PIP₂ binding to TRPC6 proteins, this will cause inhibition of channel activity due to the inhibitory action of PIP₂ on TRPC6 ion channels (Albert et al. 2008; Ju et al. 2010).

Chelerythrine produced a larger potentiation of Ang II-evoked TRPC6 NP_o than anti-TRPC1 or anti-TRPC5 antibodies which reduced TRPC1/C5 channel activity by >90%. Therefore TRPC6 activity is not only inhibited by TRPC1/C5-mediated Ca²⁺ entry and PKC activation, but also by a separate mechanism involving PKC stimulation, probably due to DAG generated by PLC coupled to the AT₁ receptor.

Physiological implications of interaction between TRPC1/C5 and TRPC6 channels

TRPC1/C5 is a Ca²⁺-permeable non-selective cation channel and its primary function is to cause cell excitation leading to contraction and proliferation of vascular smooth muscle cells (Abramowitz & Birnbaumer, 2009). It is proposed that the inhibitory effect of TRPC1/C5 channels described here is a secondary role to limit excessive cell excitation by TRPC6 channels which have a much larger unitary conductance than TRPC1/C5 channels. This effect was observed with all Ang II concentrations studied indicating the physiological importance of this interaction between TRPC1/C5 and TRPC6 channels.

Previously we have demonstrated that other vascular preparations also possess two TRPC conductances, one consisting of a core TRPC1/C5 element and a second consisting of either TRPC3, TRPC6 or TRPC7 subunits either as homomeric or heteromeric conductances (Albert & Large, 2006; Saleh et al. 2006, 2008; Peppiatt-Wildman et al. 2007; Ju et al. 2010). If expression of two TRPC channel subtypes that interact with each other occurs in other preparations, interpretation of data in studies where anti-TRPC1 antibodies or molecular interventions are used to block TRPC1 channels might produce misleading functional results. In this respect the extracellular TRPC1 antibody T1E3 decreased ET-1-induced contraction in rat tail artery and cultured cerebral arteries but had no effect in fresh cerebral arteries or the basilar artery (Bergdahl et al. 2003, 2005). In rat mesenteric arteries, T1E3 increased contractions induced by ET-1, phenylephrine and the thromboxane mimetic U76619 (Kwan et al. 2009). These apparent conflicting data may be explained if TRPC1/C5 channels can cause not only direct cellular excitation but also produce inhibition of other excitatory TRPC conductances.

Conclusions

The present results show a novel physiological interaction in which during Ang II-evoked excitation of mesenteric arteries Ca²⁺ influx through TRPC1/C5 channels reduces TRPC6 channel activity via Ca²⁺-dependent and PKC-mediated phosphorylation of TRPC6 subunits.

Glossary

Abbreviations

AgP

antigenic peptide

Ang II

angiotensin II

IP

immunoprecipitation

PKC

protein kinase C

TRPC

canonical transient receptor potential channel

VSMC

vascular smooth muscle cell

Author contributions

J.S., M.J. and S.N.S. carried out the data collection and analysis. A.P.A. and W.A.L. were involved in conception, design and interpretation of data, and also wrote the manuscript. All authors approved the final version of the manuscript.