SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Appendix

We investigated synergism between inositol 1,4,5-trisphosphate (Ins(1,4,5)P3) and diacylglycerol (DAG) on TRPC6-like channel activity in rabbit portal vein myocytes using single channel recording and immunoprecipitation techniques. Ins(1,4,5)P3 at 10 μm increased 3-fold TRPC6-like activity induced by 10 μm 1-oleoyl-2-acetyl-sn-glycerol (OAG), a DAG analogue. Ins(1,4,5)P3 had no effect on OAG-induced TRPC6 activity in mesenteric artery myocytes. Anti-TRPC6 and anti-TRPC7 antibodies blocked channel activity in portal vein but only anti-TRPC6 inhibited activity in mesenteric artery. TRPC6 and TRPC7 proteins strongly associated in portal vein but only weakly associated in mesenteric artery tissue lysates. Therefore in portal vein the conductance consists of TRPC6/C7 subunits, while OAG activates a homomeric TRPC6 channel in mesenteric artery myocytes. Wortmannin at 20 μm reduced phosphatidylinositol 4,5-bisphosphate (PIP2) association with TRPC6 and TRPC7, and produced a 40-fold increase in OAG-induced TRPC6/C7 activity. Anti-PIP2 antibodies evoked TRPC6/C7 activity, which was blocked by U73122, a phospholipase C inhibitor. DiC8-PIP2, a water-soluble PIP2 analogue, inhibited OAG-induced TRPC6/C7 activity with an IC50 of 0.74 μm. Ins(1,4,5)P3 rescued OAG-induced TRPC6/C7 activity from inhibition by diC8-PIP2 in portal vein myocytes, and this was not prevented by the Ins(1,4,5)P3 receptor antagonist heparin. In contrast, Ins(1,4,5)P3 did not overcome diC8-PIP2-induced inhibition of TRPC6 activity in mesenteric artery myocytes. 2,3,6-Tri-O-butyryl-Ins(1,4,5)P3/AM (6-Ins(1,4,5)P3), a cell-permeant analogue of Ins(1,4,5)P3, at 10 μm increased TRPC6/C7 activity in portal vein and reduced association between TRPC7 and PIP2, but not TRPC6 and PIP2. In contrast, 10 μm OAG reduced association between TRPC6 and PIP2, but not between TRPC7 and PIP2. The present work provides the first evidence that Ins(1,4,5)P3 modulates native TRPC channel activity through removal of the inhibitory action of PIP2 from TRPC7 subunits.

Abbreviations 
DAG

diacylglycerol

Ins(1,4,5)P3

inositol 1,4,5-trisphosphate

IP

immunoprecipitation

MLCK

myosin light chain kinase

OAG

1-oleoyl-2-acetyl-sn-glycerol

PIP2

phosphatidylinositol 4,5-bisphosphate

PI-PLC

phosphoinositide-phospholipase C

PKC

protein kinase C

ROC

receptor-operated channel

SOC

store-operated channel

VDCC

voltage-dependent Ca2+ channel

WB

Western blotting

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Appendix

Canonical transient receptor potential (TRPC) channels are non-selective cation channels which when stimulated allow Na+ and Ca2+ ions to enter cells. TRPC channels are widely expressed in vascular smooth muscle cells where they have been linked to various physiological and pathological responses, notably contraction, cell growth and proliferation (Inoue et al. 2006; Abramowitz & Birnbaumer, 2009).

An important question concerns the activation mechanism(s) of TRPC channels where there is growing interest in the role of membrane phospholipids. It is generally accepted that TRPC channels are activated by G-protein-coupled receptors linked to phosphoinositide-phospholipase C (PI-PLC, Clapham, 2003; Albert & Large, 2006; Hardie, 2007; Abramowitz & Birnbaumer, 2009). The present study focuses on native members of the TRPC3/C6/C7 subfamily in freshly dispersed vascular myocytes. In a study on the action of noradrenaline in rabbit portal vein myocytes, it was shown that diacylglycerol (DAG), one of the products of PI-PLC-mediated hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2), stimulated a non-selective cation conductance (Helliwell & Large, 1997). Subsequently it was shown that TRPC6 was a component of this conductance (Inoue et al. 2001). However, it was also considered that other TRPC subunits might contribute to the ion channel (Inoue et al. 2001), and in light of our data we refer to this as a TRPC6-like channel. A significant observation was that DAG activated the TRPC6-like channel in portal vein myocytes by a protein kinase C (PKC)-independent mechanism (Helliwell & Large, 1997), which is considered to be a hallmark mechanism of heterologously expressed TRPC3/C6/C7 conductances (Hofmann et al. 1999; Okada et al. 1999; Trebak et al. 2003; Estacion et al. 2004; Shi et al. 2004), and this has been observed with diverse native TRPC conductances in several blood vessels (Albert et al. 2005, 2006; Peppiatt-Wildman et al. 2007).

The other product of PIP2 hydrolysis, inositol 1,4,5-trisphosphate (Ins(1,4,5)P3), also interacts with expressed TRPC3/C6/C7 channels but with varied effects (Trebak et al. 2003; Estacion et al. 2004; Shi et al. 2004; Vazquez et al. 2006). In rabbit portal vein myocytes it was shown that Ins(1,4,5)P3 increased open probability and the rate of activation of TRPC6-like channels induced by the DAG analogue 1-oleoyl-2-acetyl-sn-glycerol (OAG, Albert & Large, 2003). The potentiation by Ins(1,4,5)P3 of OAG-evoked TRPC6-like channel activity was observed in isolated membrane patches and was not inhibited by the classical Ins(1,4,5)P3 receptor antagonist heparin. Consequently it was concluded that this effect of Ins(1,4,5)P3 was not due to its interaction with the classical Ins(1,4,5)P3 receptor on the sarcoplasmic reticulum (SR) and may represent a novel effect on native TRPC channels (Albert & Large, 2003). In other experiments it was demonstrated that Ins(1,4,5)P3 constricts rat cerebral arteries by a direct interaction between Ins(1,4,5)P3 receptors and TRPC3 channels which was also independent of SR Ca2+ release (Xi et al. 2008). Therefore it appears that Ins(1,4,5)P3 interacts with vascular TRPC channels but it is not known how this interaction leads to increased channel activity.

More recently it has been shown that PIP2 itself has a direct effect on TRP channel function in addition to its role as a substrate for the PI-PLC-mediated generation of DAG and Ins(1,4,5)P3 (see Hardie, 2007; Nilius et al. 2008). In vascular smooth muscle, PIP2 activates several isoforms of native TRPC1 channels (Saleh et al. 2009a,b) but inhibits TRPC6 channel activity in rabbit mesenteric artery myocytes (Albert et al. 2008). In the latter study PIP2 was associated with TRPC6 proteins in unstimulated tissues and a striking observation was that in tissues in which PIP2 levels had been depleted TRPC6 channel activity evoked by OAG was increased greatly compared to control tissues. This finding together with the observation that OAG decreased co-association between PIP2 and TRPC6 proteins indicated that antagonism between DAG and PIP2 regulates TRPC6 channel activity (Albert et al. 2008 and see Large et al. 2009 for fuller description).

In the present experiments we studied the potentiating effect of Ins(1,4,5)P3 on native TRPC channels, and show that in rabbit portal vein myocytes the effect of Ins(1,4,5)P3 is due to interaction with PIP2 on TRPC7 subunits of the TRPC6/C7 ion conductance.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Appendix

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). Portal vein was dissected free from fat and connective tissue in physiological salt solution containing (mm): NaCl (126), KCl (6), glucose (10), Hepes (11), MgCl2 (1.2), CaCl2 (1.5), with pH adjusted to 7.2 with 10 m NaOH. An incision was made along the longitudinal axis of the blood vessel and the exposed endothelium was gently removed using a cotton bud. Enzymatic digestion and smooth muscle cell isolation were subsequently carried out using methods previously described (Albert & Large, 2003).

Electrophysiology

Whole-cell and single cation channel currents were recorded in voltage-clamp mode using cell-attached and inside-out patch configurations with an AXOpatch 200B amplifier (Axon Instruments, USA) at room temperature (20–23°C). Patch pipettes were manufactured from borosilicate glass to produce pipettes with resistances of about 5 MΩ for whole-cell recording and 10 MΩ for single channel recording when filled with patch pipette solution. To reduce ‘line’ noise the recording chamber (vol. ca 150–200 μl) was perfused using two 20 ml syringes, one filled with external solution and the other used to drain the chamber, in a ‘push and pull’ technique. The external solution could be exchanged twice within 30 s.

Whole-cell currents and single channel activities were initially recorded onto digital audiotape (DAT) using a Sony PCM-R300 digital tape-recorder (Bio-Logic Science Instruments, Claix, France) at a bandwidth of 5 kHz and a sample rate of 48 kHz. For off-line analysis whole-cell current were filtered at 500 Hz (−3 db, low pass 8-pole Bessel filter, model LP02, Frequency Devices Inc., Ottawa, IL, USA) and acquired using a Digidata 1322A and pCLAMP 9.0 (Molecular Devices, Sunnyvale, CA, USA) at a sample rate of 2 kHz. Single channel currents were filtered at 0.5–1 kHz and acquired at a sampling rate of 5–10 kHz. Data were captured with a Dell Dimension 5150 personal computer.

To evaluate current–voltage (I–V) relationships of single channel currents, membrane potential was manually altered between −70 mV and +50 mV. Single channel current amplitudes were calculated from idealised traces of at least 30 s in duration using the 50% threshold method and analysed using pCLAMP 9.0 software with events lasting for <0.664–1.328 ms (×2 rise time for a 0.5–1 kHz, −3 db, low pass Bessel filter) being excluded. Single channel current amplitude histograms were plotted from this analysis, and these graphs were fitted with Gaussian curves with the peak of these curves determining the unitary amplitude of the single channel currents. Open probability (NPo) was used as a measure of channel activity and calculated automatically using pCLAMP 9. Figure preparation was carried out using Origin 6.0 (OriginLab Corp., Northampton, MA, USA) where inward single cation channel openings are shown as downward deflections.

In both rabbit portal vein and mesenteric artery myocytes we have shown that with the conditions used noradrenaline and angiotensin II activate only two cation conductances in both preparations (both TRPC channels, Albert et al. 2009). The first is termed a receptor-operated channel (ROC), which is composed of, at least, TRPC6 subunits and has subconductance states of 15–45 pS (Saleh et al. 2006 and present work). The second conductance consists of TRPC1 subunits and is referred to as a store-operated channel (SOC), and has a much lower conductance of about 2 pS (Albert & Large, 2002; Saleh et al. 2006). In the present work we investigated the former TRPC6-like conductances by setting the threshold level for ROC detection markedly above the full open levels for TRPC1 SOCs. In this manner only TRPC6-like channels were measured.

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. Tissues were defrosted and mechanically disrupted with an Ultraturrax homogeniser and further disrupted by sonication on ice for 2–3 h. Tissues were subsequently centrifuged at 12,000 g for 60 min at 4°C and the supernatant was discarded. The total cell lysate (TCL) was then collected by centrifugation at 12,000 g for 10 min in 10 mg ml−1 RIPA lysis buffer (Santa Cruz Biotechnology, Santa Cruz, CA, USA), supplemented with protease inhibitors. Protein content was quantified using the Bio-Rad protein dye reagent (Bradford method). TCL was retained on ice for subsequent experimental procedures. The immunoprecipitation protocol was carried out using the Millipore Catch and Release kit, where spin columns were loaded with 500 μg of TCL and 2–6 μg of antibody and immunoprecipitated for 1 h at room temperature.

Immunoprecipitated samples were eluted with Laemmli sample buffer and incubated at 60°C for 5 min. One-dimensional protein gel electrophoresis was performed in 4–12% Bis-Tris gels in a Novex mini-gel system (Invitrogen) with 10–20 μg of total protein loaded in each lane. Separated proteins were transferred onto PVDF membranes using iBlot apparatus (Invitrogen, UK). Western blotting was subsequently carried out on membranes which were incubated with the appropriate primary antibody at 4°C overnight. Primary antibodies were diluted in 1:200 for a 0.2 mg ml−1 stock of antibodies and 1:800 for a 0.8 mg ml−1 stock, and where possible alternative antibodies raised against different epitopes were used for immunoprecipitation and Western blot analysis. Following antibody removal membranes were washed for 2 h with milk/phosphate buffered saline with Tween-20 (PBST) and were subsequently incubated with horseradish peroxidase-conjugated secondary antibody diluted 1:10,000–50,000 (0.8 mg ml−1 stock concentration) in milk/PBST. Membranes were then washed 3 times for 15 min in phosphate buffered saline (PBS) + 0.1% Tween, followed by a final wash in PBS before being treated with electrochemiluminescence reagents (Pierce Biotechnology, Inc., Rockford, IL, USA) for 1 min and exposed to photographic films. Data shown represent experiments from at least 3 different animals.

Anti-TRPC and anti-PIP2 antibodies

Polyclonal TRPC1, TRPC4, TRPC5 and TRPC6 antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA) and polyclonal TRPC3 antibody was from Alomone Laboratories (Jerusalem, Israel). Polyclonal TRPC7 antibody was generated (Sigma Genosys, Cambridge, UK) against the sequence Glu-Lys-Phe-Gly-Lys-Asn-Leu-Asn-Lys-Asp-His-Leu-Arg-Val-Asn corresponding to positions 843–857 of human and mouse TRPC7 (accession nos: NP065122 and NP036165, respectively) and previously characterized by Goel et al. (2002). For discussion on the confirmation of selectivity and negligible cross-reactivity of these anti-TRPC antibodies see Methods in Saleh et al. (2008).

Mouse anti-PIP2 antibody generated against liposomes of human origin constituted with the phospholipid (Osborne et al. 2001) was purchased from Assay Designs (Ann Arbor, MI, USA) and Santa Cruz Biotechnology. It has a predicted molecular mass of 74 kDa according to manufacturer's specifications, and has been used in electrophysiological and immunoprecipitation experiments (see references in Saleh et al. 2008).

Solutions and drugs

In whole-cell recording a K+-free external solution was used containing (mm): NaCl (126), CaCl2 (1.5), Hepes (10) and glucose (11), pH adjusted to 7.2 with 10 m NaOH. 5 μm nicardipine, 100 μm 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid (DIDS) and 100 μm niflumic acid were included to block voltage-dependent Ca2+ channels (VDCCs) and Ca2+-activated and swell-activated Cl conductances. In cell-attached patch experiments the membrane potential was set to approximately 0 mV by perfusing cells in a KCl external solution containing (mm): KCl (126), CaCl2 (1.5), Hepes (10) and glucose (11), pH adjusted to 7.2 with 10 m KOH. Nicardipine (5 μm) was also included to minimise smooth muscle cell contraction by blocking Ca2+ entry through voltage-dependent Ca2+ channels.

The patch pipette solution used for whole-cell recording was K+ free and contained (mm): CsCl (18), caesium aspartate (108), MgCl2 (1.2), Hepes (10), glucose (11), BAPTA (10), CaCl2 (4.8 mm, internal Ca2+ concentration approximately 100 nmol l−1 as calculated using EQCAL software, Biosoft, Cambridge, UK), Na2ATP (1), NaGTP (0.2), pH 7.2 with Tris. The bathing solution used in inside-out patch recording (intracellular solution) had the same composition as the pipette solution for whole-cell recording except that 1 mm BAPTA and 0.48 mm CaCl2 were used to provide an internal Ca2+ concentration of 100 nmol l−1.

The patch pipette solution used for both cell-attached and inside-out patch recording (extracellular solution) was K+ free and contained (mm): NaCl (126), CaCl2 (1.5), Hepes (10), glucose (11), TEA (10), 4-AP (5), iberiotoxin (0.0002), DIDS (0.1), niflumic acid (0.1) and nicardipine (0.005), pH adjusted to 7.2 with NaOH. Under these conditions VDCCs, voltage-gated K+ currents, swell-activated Cl currents and Ca2+-activated Cl and K+ conductances were abolished and non-selective cation currents were recorded in isolation.

All drugs were dissolved in distilled H2O or dimethyl sulphoxide (DMSO; 0.1%). DMSO alone had no effect on channel activity. In experiments using diC8-PIP2, the PLC inhibitor U71332 (2 μm) and PI-3-kinase inhibitor wortmannin (50 nm) were also included to prevent PIP2 acting as a substrate for DAG/Ins(1,4,5)P3 and PIP3 respectively. Values are the mean of n cells ±s.e.m. Statistical analysis was carried out using Student's t test for paired (comparing effects of agents on the same cell) or unpaired data (comparing effects of agents between cells) with the level of significance set at P < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Appendix

Ins(1,4,5)P3 potentiates OAG-induced TRPC6-like channel activity in portal vein but not in mesenteric artery myocytes

In the first series of experiments we studied the interaction between OAG and Ins(1,4,5)P3 on TRPC6-like channel activity in inside-out patches from portal vein myocytes. Bath application of 10 μm OAG activated TRPC6-like channel currents which had three mean subconductance states of 13 ± 2 pS, 27 ± 3 pS and 44 ± 4 pS (n= 6), which are similar to the conductance values of OAG-evoked TRPC6 channel currents previously described in mesenteric artery myocytes (Saleh et al. 2006). Figure 1Aa shows that 10 μm Ins(1,4,5)P3 markedly potentiated OAG-induced TRPC6-like channel activity in portal vein myocytes as previously described (Albert & Large, 2003), from a control mean open probability (NPo) value of 0.02 ± 0.01 to 0.11 ± 0.02 (n= 6, P < 0.05) in inside-out patches held at –50 mV. In addition, Fig. 1Ab shows that the amplitude histograms of channel currents evoked in the presence of OAG and OAG + Ins(1,4,5)P3 could both be fitted by the sum of four Gaussian curves, which represented one closed and three subconductances states. The Gaussian curves had similar peak values in the presence of OAG and OAG + Ins(1,4,5)P3, which indicates that Ins(1,4,5)P3 increased the activity of the same channels (Fig. 1Ab). Figure 1C shows the effect of different concentrations of Ins(1,4,5)P3 on mean relative NPo values of OAG-evoked TRPC6-like channel activity in portal vein with 50 μm Ins(1,4,5)P3 potentiating the effects of OAG by more than 10-fold.

image

Figure 1. Ins(1,4,5)P3 potentiates OAG-evoked TRPC6-like activity in portal vein but not TRPC6 activity in mesenteric artery myocytes Aa, bath application of 10 μm OAG activated TRPC6-like channel activity, which was increased following co-application of 10 μm Ins(1,4,5)P3 in an inside-out patch held at −50 mV from a portal vein myocyte. Ab, amplitude histograms of OAG-evoked TRPC6-like channel currents in the absence and presence of Ins(1,4,5)P3 shown in Aa. Both histograms could be fitted by the sum of four Gaussian curves denoting one closed state and three subconductance states with peak amplitudes of −0.68 pA, −1.36 pA and −2.15 pA. In contrast, B, co-application of 10 μm Ins(1,4,5)P3 did not increase OAG-induced TRPC6 channel activity in inside-out patches held at −50 mV from a mesenteric artery myocyte. C, the excitatory action of different concentrations of Ins(1,4,5)P3 on relative mean NPo of OAG-evoked TRPC6 channel activity in portal vein and its lack of effect in mesenteric artery myocytes. Each value is from at least 5 myocytes.

Download figure to PowerPoint

Previously we had shown that angiotensin II and OAG stimulated TRPC6 channels in rabbit mesenteric artery myocytes (Saleh et al. 2006). Therefore we studied the effects of Ins(1,4,5)P3 on OAG-evoked TRPC6 activity this preparation. In contrast to the data obtained in portal vein myocytes, Fig. 1B and C shows that 10 μm and 50 μm Ins(1,4,5)P3 had no effect on 10 μm OAG-induced TRPC6 activity in inside-out patches from mesenteric artery myocytes (mean NPo of control OAG responses was 0.03 ± 0.01, OAG + 10 μm Ins(1,4,5)P3: 0.02 ± 0.02, and OAG + 50 μm Ins(1,4,5)P3: 0.03 ± 0.01, n= 5). Bath application of 1–50 μm Ins(1,4,5)P3 did not stimulate channel activity when applied on its own to inside-out patches from either portal vein (n= 5) or mesenteric artery myocytes (n= 5, data not shown).

Evidence that TRPC6/C7 heteromeric channels compose OAG-induced cation channel activity in portal vein myocytes

The distinct excitatory effect of Ins(1,4,5)P3 on OAG-evoked TRPC6-like channel activity in portal vein but not in mesenteric artery myocytes (see Fig. 1) suggests that the molecular composition of these channels is different in these two preparations. We investigated the molecular properties of these channels by comparing the effects of anti-TRPC antibodies on channel activity in portal vein and mesenteric artery myocytes. For these studies we used the physiological neurotransmitter noradrenaline to evoke TRPC channel activity in cell-attached patches. Subsequently membrane patches were excised into the inside-out configuration in which channel activity was sustained at constant levels in control experiments. Anti-TRPC antibodies were then bath applied to the cytosolic surface of membrane patches.

In portal vein and mesenteric artery myocytes, noradrenaline activated cation channel currents in inside-out patches with three mean subconductances states of respectively 16 ± 2 pS, 31 ± 3 pS and 47 ± 3 pS (n= 5, portal vein) and 15 ± 3 pS, 29 ± 3 pS and 45 ± 4 pS (n= 5, mesenteric artery), which all had reversal potentials (Er) of ∼0 mV. These conductances values are similar to the conductances of OAG-, and Ang II-evoked currents in portal vein and mesenteric artery myocytes (see earlier and Saleh et al. 2006), and therefore show that noradrenaline and OAG activate the same channels.

Figure 2Aa, Ab and C shows that bath application of 1:200 dilution of anti-TRPC6 and anti-TRPC7 antibodies significantly inhibited noradrenaline-activated cation channel activity by over 85% in portal vein smooth muscle cells. Moreover pre-incubating anti-TRPC6 and anti-TRPC7 antibodies with their respective antigenic peptides prevented inhibition of noradrenaline-induced channel activity by these anti-TRPC antibodies (Fig. 2C). Figure 2C also shows that anti-TRPC antibodies raised against TRPC1, TRPC3, TRPC4 and TRPC5 had no effect on noradrenaline-induced channel activity in portal vein myocytes. Figure 2Ba, Bb and C show that bath application of 1:200 dilution of anti-TRPC6 antibodies but not anti-TRPC7 antibodies had an inhibitory effect on noradrenaline-induced TRPC6 activity in mesenteric artery myocytes. These results confirm earlier work showing that only anti-TRPC6 antibodies and not anti-TRPC1, anti-TRPC3 or anti-TRPC7 antibodies inhibited angiotensin-II-evoked ROC activity in mesenteric artery myocytes (Saleh et al. 2006).

image

Figure 2. Evidence for different TRPC subunit composition of cation channels in portal vein and mesenteric myocytes A, bath application of 1:200 dilution of anti-TRPC6 (a) and anti-TRPC7 antibodies (b) to the cytosolic surface of inside-out patches held at −50 mV from portal vein myocytes produced a pronounced inhibition of noradrenaline-evoked channel activity. B, bath application of 1:200 dilution of anti-TRPC6 (a) but not anti-TRPC7 antibodies (b) inhibited channel activity in inside-out patches held at −50 mV from mesenteric artery myocytes. Note that in A and B channel activity was initially evoked by 10 μm noradrenaline in cell-attached patches before excision into inside-out configuration (see Methods). The dotted line in Bb represents the closed level. C, mean data that illustrate anti-TRPC6 and anti-TRPC7 antibodies significantly inhibited channel activity in portal vein myocytes and that this inhibition was prevented following pre-incubation of the antibodies with their antigenic peptides (AgP). In contrast, anti-TRPC6 but not anti-TRPC7 antibodies inhibited channel activity in mesenteric artery myocytes. Each value represents at least n= 5. D, co-immunoprecipitation (IP) experiments showing that TRPC7 and TRPC6 are expressed in both portal vein (a) and mesenteric artery (b) tissue lysates. Moreover, there is strong co-association between TRPC7 and TRPC6 proteins in portal vein (a) but only a weak interaction between these proteins in mesenteric artery (b). Note that in portal vein and mesenteric artery, expression of TRPC7 and association of TRPC7 with TRPC6 are markedly reduced following pre-incubation with AgP during the IP step. The conditions for IP above the lanes in Da are the same for Db.

Download figure to PowerPoint

The differential sensitivity to anti-TRPC antibodies indicates that both TRPC6 and TRPC7 proteins are constituents of the conductance in portal vein but that the channels in mesenteric artery myocytes are composed of only TRPC6 subunits. This proposal was further investigated using co-immunoprecipitation studies. Figure 2Da and b shows that in respectively portal vein and mesenteric artery tissue lysate immunoprecipitation (IP) with anti-TRPC7 followed by Western blotting (WB) with anti-TRPC7 antibodies produced predicted bands of ∼90 kDa. In addition, Fig. 2Da and b shows that when IP was carried out with anti-TRPC6 and WB with anti-TRPC7 antibodies there were also predicted bands of ∼90 kDa. Moreover Fig. 2Da and b shows that when IP and WB were carried out using anti-TRPC6 antibodies there were predicted bands of ∼110 kDa in portal vein and mesenteric artery tissue lysates. These data indicate that TRPC7 proteins are expressed both in portal vein and mesenteric artery, as well as TRPC6 subunits as previously described (Inoue et al. 2001; Saleh et al. 2006). However, Fig. 2Da and b illustrates that there is a marked association between TRPC6 and TRPC7 proteins in the portal vein but there is much weaker association between these subunits in mesenteric artery.

These results provide evidence that noradrenaline-induced TRPC6-like channels in portal vein are composed of TRPC6 and TRPC7 subunits in a heteromeric structure. Consequently, in the rest of the present work we term these TRPC6/C7 channels. However, these data indicate that the ion channel in mesenteric artery is a homomeric TRPC6 conductance.

Endogenous and exogenous PIP2 have a marked inhibitory action on TRPC6/C7 channel activity in native rabbit portal vein myocytes

Previously we demonstrated in mesenteric artery myocytes that PIP2 inhibited the excitatory action of OAG on TRPC6 channel activity (Albert et al. 2008; Large et al. 2009). In the present study we investigated whether the potentiating effect of Ins(1,4,5)P3 on OAG-evoked TRPC6/C7 channels in portal vein was due to an interaction with PIP2. Therefore it was initially necessary to explore whether PIP2 also has an inhibitory action on TRPC6/C7 channels in portal vein myocytes.

In control portal vein myocytes, bath application of 10 μm OAG activated TRPC6/C7 channel activity in cell-attached patches at −50 mV with a mean open probability (NPo) of 0.05 ± 0.03 (n= 9, Fig. 3Aa and Bb). In cells pre-treated with 20 μm wortmannin for 30 min to deplete tissue PIP2 levels, OAG-evoked TRPC6/C7 channel activity was increased about 40-fold (NPo of 2.04 ± 0.37, n= 10, P < 0.01, Fig. 3Ab and Bb). Moreover, Fig. 3Aa and b shows that pre-treatment of cells with 20 μm wortmannin markedly decreased the latency to onset, defined as time taken from application of 10 μm OAG to initial channel activation, from 134 ± 17 s (n= 9, latency range 30–204 s) in control cells to 16 ± 4 s (P < 0.05, n= 10, latency range 3–40 s). It should be noted that application of 10 μm or 50 μm Ins(1,4,5)P3 on their own had no effect on channel activity in portal vein myocytes pre-treated with 20 μm wortmannin (n= 5, data not shown).

image

Figure 3. PIP2 has an inhibitory action on OAG-evoked TRPC6/C7 channel activity in portal vein myocytes Aa, bath application of 10 μm OAG induced TRPC6/C7 channel activity in inside-out patches held at −50 mV, which was markedly increased following pre-treatment of myocytes with 20 μm wortmannin for 30 min (Ab). Ba, amplitude histogram of OAG-evoked TRPC6/C7 channel currents at −50 mV in the absence (filled bars) and after pre-treatment with (open bars) 20 μm wortmannin showing that the same channels with three subconductance states were activated in both conditions. Bb, mean NPo values of OAG-evoked TRPC6/C7 channel activity in control cells and following pre-treatment of cells with 1 μm and 20 μm wortmannin. Bc, IP blots showing PIP2 associates with TRPC6 and TRPC7 at rest in portal vein tissue lysates, and that this interaction is greatly reduced following pre-treatment with 20 μm wortmannin for 30 min. Pre-treatment of cells with 1 μm wortmannin for 30 min had no effect on associations between TRPC6 and PIP2. C, in a myocyte pre-treated with 20 μm wortmannin for 30 min, bath application of 10 μm OAG activated TRPC6/C7 channel activity in an inside-out patch held at −50 mV which was almost completely abolished by co-application of 10 μm diC8-PIP2. D, concentration–response curve of diC8-PIP2 on TRPC6/C7 channel activity induced by 10 μm OAG showing that diC8-PIP2 inhibited responses by 50% at a concentration of 0.74 μm (IC50). Each point is from at least 5 myocytes.

Download figure to PowerPoint

Figure 3Ba shows that histograms of OAG-activated cation channel current amplitudes from a control cell and a myocyte pre-treated with wortmannin could both be fitted by the sum of four Gaussian curves with similar peak values, indicating that the channels in these two patches contained one closed level and three similar subconductance states. Thus OAG-evoked channel currents in control and in wortmannin-treated cells had similar subconductance values of ∼13 pS, 27 pS and 42 pS, which all had an Er of ∼0 mV, i.e. OAG evokes the same conductance in the absence and presence of 20 μm wortmannin. Figure 3Ba also illustrates an interesting observation that OAG-induced channel activity opened most frequently to the lower 13 pS conductance level in control cells compared to the full 42 pS conductance level following pre-treatment with wortmannin. These results suggest that endogenous PIP2 inhibits TRPC6/C7 channel activation by exogenous OAG.

Wortmannin also inhibits PI-3-kinase and myosin light chain kinase (MLCK) with IC50 values of 5 nm and 200 nm respectively (Nakanishi et al. 1992; Arcaro & Wymann, 1993). Therefore the potentiating action of 20 μm wortmannin on OAG-evoked TRPC6/C7 could possibly be caused by an action of this agent on either of these two kinases. We compared the effects of 1 μm wortmannin, which will block both PI-3-kinase and MLCK but not PI-4- or PI-5-kinases, with 20 μm wortmannin on OAG-evoked channel activity and PIP2 levels in portal vein tissue lysates. Figure 3Bb shows that pre-treatment of myocytes with 1 μm wortmannin for 30 min had no effect on OAG-induced TRPC6/C7 channel activity. In addition, Fig. 3Bc illustrates IP experiments showing that PIP2 associates with TRPC6 at rest, and that this interaction was not altered by 1 μm wortmannin but markedly reduced following pre-treatment with 20 μm wortmannin. Moreover, Fig. 3Bc also shows that PIP2 interacts with TRPC7 at rest, and this association is also reduced following pre-treatment with 20 μm wortmannin. These data suggest that the effects of 20 μm wortmannin are due to an action on PI-4- and PI-5-kinases and not PI-3-kinase or MLCK.

Pre-treatment of portal vein myocytes with 20 μm wortmannin for 30 min also significantly increased the mean amplitude of whole-cell cation conductances induced by 10 μm OAG at −50 mV from a control value of −9 ± 2 pA (n= 5) to −31 ± 4 pA (n= 6, P < 0.05). These results further show that endogenous PIP2 has a pronounced inhibitory action on activation of TRPC6/C7 channel activity.

To quantify the effect of exogenous PIP2 on TRPC6/C7 channel activity we studied the effect of diC8-PIP2, a water-soluble analogue of this phospholipid, on OAG-evoked TRPC6/C7 channel activity in inside-out patches from portal vein myocytes. In these experiments we removed endogenous PIP2 by pre-treating myocytes with 20 μm wortmannin for 30 min.

Figure 3C and D shows that TRPC6/C7 channel activity induced by 10 μm OAG was almost completely abolished by co-application of 10 μm diC8-PIP2 with a concentration of 0.74 μm producing 50% inhibition (IC50). These results show that exogenous PIP2 also has a marked inhibitory action on TRPC6/C7 channel activity in portal vein.

Anti-PIP2 antibodies activate TRPC6/C7 through a U73122-sensitive mechanism

To further investigate the influence of an inhibitory action of endogenous PIP2 on TRPC6/C7 channel activity, we studied the effect of applying anti-PIP2 antibodies, which have been shown to act as inhibitors of actions of PIP2 (Suh & Hille, 2005), to excised inside-out patches of portal vein myocytes.

Figure 4Aa and b illustrates that bath application of 1:200 dilution of anti-PIP2 antibodies to quiescent inside-out patches at −50 mV activated cation channel activity with a mean NPo of 1.27 ± 0.38 (n= 6), which declined slowly to a mean NPo of 0.08 ± 0.04 (n= 6) after 20.5 ± 2.7 min (P < 0.05, n= 6). These anti-PIP2 antibody-induced cation channel currents were composed of three subconductance states of 14 pS, 28 pS and 43 pS with an Er of ∼0 mV, which are similar to those values of OAG-evoked channel currents described above, and indicate that anti-PIP2 antibodies also activate TRPC6/C7 channels. Figure 4Aa and b also shows that following the run-down of anti-PIP2 antibody-evoked TRPC6/C7 channel activity after 20–30 min, co-application of 10 μm OAG stimulated channel activity, which shows that anti-PIP2 antibodies do not directly block the ion channels.

image

Figure 4. Anti-PIP2 antibodies stimulate TRPC6/C7 channel activity in inside-out patches from portal vein Aa, bath application of 1:200 dilution of anti-PIP2 antibodies activated TRPC6/C7 channel activity, which declined to baseline levels after about 20 min in the continued presence of the anti-PIP2 antibodies, but channel currents were re-activated following co-application of 10 μm OAG. Ab, mean NPo values of TRPC6/C7 activated by anti-PIP2 antibodies, after run-down, and following co-application of OAG. Ba, co-application of 2 μm U73122 completely abolished anti-PIP2 antibody-evoked TRPC6/C7 channel activity, but it was restored following co-application of 10 μm OAG. Bb, mean NPo values of TRPC6/C7 activated by anti-PIP2 antibodies, after application of U73122 and then co-application of OAG. Note that the inactive analogue of U73122, U73443 had no effect on channel activity induced by anti-PIP2 antibodies.

Download figure to PowerPoint

Figure 4Ba and b shows that co-application of the PI-PLC inhibitor 2 μm U73122, <3 min after stimulation of TRPC6/C7 channel activity by anti-PIP2 antibodies and hence before any significant decline in activity, produced a pronounced inhibition of channel activity by 94 ± 3% (n= 5, P < 0.05). Figure 4Ba and b also shows that U73122-induced inhibition of anti-PIP2 antibody-evoked TRPC6/C7 channel activity was overcome by co-application of 10 μm OAG, which rules out the possibility that U73122 was acting as a direct channel blocker. Figure 4Bb illustrates that the structurally similar but inactive analogue of U73122, 2 μm U73343, had no effect on anti-PIP2 antibody-induced TRPC6/C7 channel activity (n= 4), providing further evidence that U73122 is not having a non-selective inhibitory action on these channels.

These data support the hypothesis that endogenous PIP2 has a marked inhibitory action on TRPC6/C7 channel activity in portal vein myocytes. Moreover simply reducing the level of PIP2 within a patch leads to channel activation via the constitutively active PI-PLC-dependent pathway, which is likely to involve generation of DAG.

Ins(1,4,5)P3 rescues TRPC6/C7 channel activity from inhibition by PIP2 in portal vein but not TRPC6 activity in mesenteric artery

The present work and previous data indicate that PIP2 has a pronounced inhibitory action on TRPC6/C7 channels in portal vein (see Fig. 3) and TRPC6 channels in mesenteric artery myocytes (Albert et al. 2008). Consequently in the next series of experiments we investigated interactions between Ins(1,4,5)P3 and exogenous PIP2 on OAG-induced TRPC6/C7 and TRPC6 channel activity in respectively portal vein and mesenteric artery smooth muscle cells. In these experiments cells were pre-treated with 20 μm wortmannin for 30 min to deplete tissue PIP2 levels.

Figure 5A illustrates in inside-out patches from portal vein myocytes that inhibition of 10 μm OAG-evoked TRPC6/C7 channel activity by 10 μm diC8-PIP2 was partially rescued by co-application of 10 μm Ins(1,4,5)P3. In six myocytes, 10 μm Ins(1,4,5)P3 significantly rescued the PIP2-mediated inhibition of OAG-induced TRPC6/C7 channel activity by 54 ± 6% (P < 0.01, Fig. 5A and F).

image

Figure 5. Ins(1,4,5)P3 rescues OAG-induced TRPC6/TRPC7 channel activity from inhibition by PIP2 in inside-out patches from portal vein myocytes A, in myocytes pre-treated with 20 μm wortmannin for 30 min, 10 μm Ins(1,4,5)P3 rescued 10 μm OAG-evoked TRPC6/C7 channel activity from inhibition by 10 μm diC8-PIP2 in portal vein myocytes. B, in contrast, 10 μm and 50 μm Ins(1,4,5)P3 did not overcome PIP2-induced inhibition of OAG-induced TRPC6 channel activity in a mesenteric artery myocyte. C, heparin did not prevent Ins(1,4,5)P3 from rescuing OAG-evoked TRPC6/C7 channel activity from inhibition by diC8-PIP2. D and E show respectively that the non-hydrolysable analogue of Ins(1,4,5)P3, 10 μm 3-F-Ins(1,4,5)P3, and the metabolite of Ins(1,4,5)P3, 10 μm Ins(1,4)P2, both rescued OAG-evoked TRPC6/C7 channel activity from inhibition by diC8-PIP2. F, mean data showing action of Ins(1,4,5)P3, 3-F-Ins(1,4,5)P3 and Ins(1,4)P2 on diC8-PIP2-mediated inhibition of OAG-evoked TRPC6/C7 channel activity. Each value is the mean from at least 5 cells.

Download figure to PowerPoint

In contrast Fig. 5B shows that 10 μm Ins(1,4,5)P3 did not rescue OAG-induced TRPC6 channel activity from inhibition by PIP2 in inside-out patches from mesenteric artery myocytes, with a relative mean NPo of TRPC6 channel activity induced by 10 μm OAG in the presence of 10 μm diC8-PIP2 of 0.03 ± 0.02 compared to 10 μm diC8-PIP2 plus 10 μm Ins(1,4,5)P3 of 0.03 ± 0.01 (n= 7, P > 0.05). In addition, 50 μm Ins(1,4,5)P3 also did not overcome PIP2-mediated inhibition of TRPC6 channel activity in mesenteric artery myocytes (n= 5, P > 0.05, Fig. 5B).

We investigated whether this rescuing action of Ins(1,4,5)P3 on PIP2-induced inhibition of OAG-evoked TRPC6/C7 channel activity was mediated via classical Ins(1,4,5)P3 receptors by testing the action of heparin, an Ins(1,4,5)P3 receptor antagonist, on these responses. Figure 5C and F illustrates that the reversal of PIP2-induced inhibition of OAG-evoked TRPC6/C7 activity by Ins(1,4,5)P3 was not prevented by pre-treatment with 1 mg ml−1 heparin (n= 6).

We also tested the effect of 3-F-Ins(1,4,5)P3, a non-hydrolysable analogue of Ins(1,4,5)P3, and Fig. 5D and F shows that this compound also partly rescued channel activity subsequent to inhibition by PIP2. A major metabolite of Ins(1,4,5)P3, inositol-1,4-bisphosphate (Ins(1,4)P2), also produced a significant reversal of PIP2-induced inhibition of OAG-induced TRPC6/C7 channel activity (Fig. 5E and F).

These data show that both Ins(1,4,5)P3 and Ins(1,4)P2 interact with PIP2 in regulating TRPC6/C7 channel activity in portal vein myocytes which are not mediated by the classical Ins(1,4,5)P3 receptor. In contrast, in mesenteric artery myocytes there is no interaction between Ins(1,4,5)P3 and PIP2 on TRPC6 channel activity.

Ins(1,4,5)P3 reduces association of PIP2 with native TRPC7 but not TRPC6 proteins

An important proposal from the present work is that the potentiating effect of Ins(1,4,5)P3 on channel activity is produced through an interaction with TRPC7 subunits. This hypothesis would explain why the potentiating effect of Ins(1,4,5)P3 occurs in portal vein where channels are composed of TRPC6/C7 subunits but not in mesenteric artery where channels are composed of only TRPC6 subunits. To further investigate interactions between Ins(1,4,5)P3 and PIP2 on TRPC6 and TRPC7 proteins in portal vein, we studied the association between PIP2 and TRPC6 proteins, and between PIP2 and TRPC7 proteins following pre-treatment with OAG and the cell-permeant analogue of Ins(1,4,5)P3, 2,3,6-tri-O-butyryl-Ins(1,4,5)P3/AM (6-Ins(1,4,5)P3, Xi et al. 2008).

In electrophysiological experiments we tested whether 6-Ins(1,4,5)P3, when applied extracellularly, could mimic the potentiating effects of Ins(1,4,5)P3 on inside-out patches as described earlier (see Fig. 1Aa). Figure 6Aa illustrates an experiment with cell-attached patch recording in a portal vein smooth muscle cell in which channel activity was initially induced with bath-applied 10 μm OAG. Subsequent addition of 10 μm 6-Ins(1,4,5)P3 to the bathing solution produced a marked increase in channel activity. The mean NPo values in OAG and OAG + 6-Ins(1,4,5)P3 were respectively 0.02 ± 0.01 and 0.08 ± 0.04 (n= 6, P < 0.05). These results with 6-Ins(1,4,5)P3 are similar to the potentiating effect of Ins(1,4,5)P3 on OAG-evoked TRPC6/C7 channel activity in inside-out patches from portal vein (see Fig. 1Aa). Moreover Fig. 6Ab shows that the amplitude histograms of the channel currents activated by both OAG and OAG + 6-Ins(1,4,5)P3 could be fitted by the sum of four Gaussian curves with similar peak values. These results indicate that OAG and OAG + 6-Ins(1,4,5)P3 both activate the same channels (compare with Fig. 1Ab).

image

Figure 6. 6-Ins(1,4,5)P3 potentiates OAG-induced TRPC6/C7 activity and reduces association between TRPC7 and PIP2 in portal vein myocytes Aa, bath application of 10 μm OAG activated TRPC6/C7 channel activity in a cell-attached patch held at −50 mV from a portal vein myocyte which was increased following co-application of the cell-permeant Ins(1,4,5)P3 analogue, 10 μm 6-Ins(1,4,5)P3. Ab, amplitude histograms showing that TRPC6/C7 channel currents activated by 10 μm OAG in the absence and presence of 10 μm 6-Ins(1,4,5)P3 had three subconductance states of −0.66 pA, −1.33 pA and −1.95 pA. B, co-immunoprecipitation blots showing that pre-treatment with 10 μm OAG for 30 min reduced association between TRPC6 proteins and PIP2 (a) but not between TRPC7 proteins and PIP2 (b) in portal vein tissue lysates. C, in contrast pre-treatment with 10 μm 6-Ins(1,4,5)P3 reduced associations between TRPC7 proteins and PIP2 (b) but not TRPC6 proteins and PIP2 (a) in portal vein tissue lysates. D, co-immunoprecipiation blots showing that pre-treatment of tissue lysates from mesenteric artery for 30 min with 10 μm OAG reduced associations between TRPC6 proteins and PIP2 (a), but these interactions were unaffected by 6-Ins(1,4,5)P3 (b).

Download figure to PowerPoint

Figure 6Ba shows that IP with anti-TRPC6 followed by WB with anti-PIP2 antibodies produced a predicted band at ∼75 kDa (see Methods) in portal vein tissue lysates that was substantially reduced following pre-treatment with 10 μm OAG for 30 min, whereas Fig. 6Bb shows that association between TRPC7 and PIP2 was unaffected by pre-treatment with OAG. In contrast, Fig. 6Ca and Cb shows that IP with anti-TRPC6 antibodies and then WB with anti-PIP2 antibodies produced ∼75 kDa bands that were unaffected by pre-treatment with 10 μm 6-Ins(1,4,5)P3 for 30 min, whereas interactions between TRPC7 and PIP2 were substantially reduced by 6-Ins(1,4,5)P3. In control experiments pre-treatment for 30 min with 20 μm wortmannin, 10 μm OAG or 10 μm 6-Ins(1,4,5)P3 had no effect on association between TRPC6 and TRPC7 proteins in portal vein (data not shown).

For comparison, we studied the effect of OAG and 10 μm 6-Ins(1,4,5)P3 on associations between TRPC6 proteins and PIP2 in mesenteric artery tissue lysates. Figure 6Da shows that pre-treatment for 30 min with 10 μm OAG markedly reduced association between TRPC6 and PIP2, whereas Fig. 6Db shows that pre-treatment with 10 μm 6-Ins(1,4,5)P3 had no effect on this interaction.

These results further support our proposal that the potentiating effect of Ins(1,4,5)P3 on TRPC6/C7 channel activity in portal vein is mediated by Ins(1,4,5)P3 disrupting association between TRPC7 proteins and PIP2 and that Ins(1,4,5)P3 does not interact with TRPC6 subunits.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Appendix

The novel finding of the present study is that in rabbit portal vein myocytes Ins(1,4,5)P3 potentiates OAG-evoked TRPC6/C7 channel activity by decreasing the inhibitory effect of PIP2 on the ion conductance. Moreover it appears that Ins(1,4,5)P3 interacts with PIP2 on TRPC7 and not TRPC6 subunits of the ion channel.

Comparison of TRPC6-like channels in portal vein and mesenteric artery myocytes

In rabbit portal vein smooth muscle cells Ins(1,4,5)P3 applied to the cytoplasmic surface of the membrane greatly potentiated channel activity by OAG. In contrast in mesenteric artery myocytes Ins(1,4,5)P3 had no synergistic interaction with OAG on channel activity. This was a surprising result since it was suggested that TRPC6 was a component of the cation conductance in portal vein (Inoue et al. 2001) and mesenteric artery myocytes (Saleh et al. 2006). We examined the effects of the full range of anti-TRPC antibodies against the noradrenaline-evoked responses and found that both anti-TRPC6 and anti-TRPC7 antibodies inhibited channel activity in portal vein myocytes. However only anti-TRPC6, and not anti-TRPC7, antibody reduced channel activity in mesenteric artery myocytes (also see Saleh et al. 2006). Co-immunoprecipitation studies revealed that both TRPC6 and TRPC7 proteins were expressed in portal vein and mesenteric artery. However a strong signal indicating co-association between TRPC6 and TRPC7 subunits was observed in portal vein but only weak association in mesenteric artery myocytes. Overall the data suggest the native conductance consists of a TRPC6/C7 heteromeric structure in portal vein and of a TRPC6 homomer in mesenteric artery myocytes. In expression systems it has been shown that Ins(1,4,5)P3 increased TRPC7 channel activity but not TRPC6 activity (Shi et al. 2004; Estacion et al. 2004), which supports the involvement of TRPC7 subunits in the portal vein conductance. Alternatively it is possible that heterotetrameric TRPC6/C7 conductances occur in both preparations but with different stoichiometries. For example, in portal vein where there is a greater evidence for a role of TRPC7 there may be a TRPC6:TRPC7 subunit ratio of 2:2 but 3:1 in mesenteric artery myocytes. In this situation the presence of a minimum of two TRPC7 subunits in channel structure may be necessary to confer sensitivity to Ins(1,4,5)P3. The present experiments do not unequivocally point to either of these two explanations although we favour the existence of heterotetrameric TRPC6/C7 and homomeric TRPC6 conductances in respectively portal vein and mesenteric artery myocytes. Previously it was reported that a heteromeric TRPC6/C7 ion channel was activated by vasopressin in the immortalised A7r5 cell line (Maruyama et al. 2006) but this is the first suggestion that TRPC6/C7 is a native ionic channel in freshly dispersed myocytes.

Inhibitory action of PIP2 on TRPC6/C7 channels in portal vein myocytes

The present data show that both endogenous and exogenous PIP2 inhibits TRPC6/C7 channel activity in rabbit portal vein myocytes. Qualitatively the inhibitory effect of exogenous PIP2 is similar to that observed in mesenteric artery myocytes although the PIP2 IC50 value of 0.74 μm in portal vein (present data) is about 10-fold lower than the value (∼7.6 μm) found in mesenteric artery myocytes (Albert et al. 2008). This may result from the different molecular compositions of the channels in the two preparations (see above).

The inhibitory role of endogenous PIP2 in portal vein myocytes was revealed in tissues in which PIP2 levels had been significantly depleted by pre-treatment with wortmannin. In these PIP2-depleted preparations OAG-evoked TRPC6/C7 channel activity was approximately 40-fold greater than in control cells. In addition application of anti-PIP2 antibodies produced marked activation of channel activity. Channel activation by anti-PIP2 antibody appeared to depend on the generation of DAG since application of the PI-PLC inhibitor U73122 produced a rapid and almost complete inhibition of anti-PIP2-induced TRPC6/C7 channel activity. Application of OAG at this time produced marked stimulation of channel activity which was much greater than control values (see Fig. 3Bb), presumably due to removal of PIP2 by the antibody. In other experiments after prolonged (20–30 min) application of anti-PIP2 antibodies, TRPC6/C7 channel activity eventually declined to very low levels at which time OAG again produced marked channel activation, greater than control values (see Fig. 3Bb). These experiments demonstrate not only the competition between DAG and endogenous PIP2 but also the dual roles of PIP2 as, first, a precursor of DAG (and Ins(1,4,5)P3) and, secondly, a direct inhibitor of TRPC6/C7 channels. These data also highlight the interesting property of the transduction pathway that there must be constitutive hydrolysis of PIP2 by PI-PLC to generate DAG. Presumably under control situations PIP2 hydrolysis does not produce sufficient levels of DAG to remove the inhibitory action of PIP2 and cause channel activation. This latter statement is supported by the observation that the DAG lipase inhibitor RHC80267 causes TRPC6/C7 (portal vein) and TRPC6 (mesenteric artery) channel stimulation when applied to unstimulated cells (Helliwell & Large, 1997; Inoue et al. 2001; Saleh et al. 2006).

These data on TRPC6/C7 in portal vein are similar to native TRPC6 channels in mesenteric artery myocytes and indicate physiological antagonism between PIP2 and DAG on channel activation. However the data contrast with the facilitatory responses of PIP2 on expressed TRPC6 channels (Kwon et al. 2007; Lemonnier et al. 2008) which has been discussed previously (see Albert et al. 2008).

Interaction between Ins(1,4,5)P3 and PIP2 on native TRPC6/C7 channels

In portal vein myocytes in which exogenous PIP2 inhibited OAG-evoked TRPC6/C7 channel activity, subsequent addition of Ins(1,4,5)P3 rescued channel activity. This excitatory effect of Ins(1,4,5)P3 was mimicked by the stable Ins(1,4,5)P3 analogue, 3-F-Ins(1,4,5)P3, and the metabolite Ins(1,4)P2. This antagonism indicates that the facilitatory effect of Ins(1,4,5)P3 is due to counteracting the inhibitory effect of PIP2 on TRPC6/C7 channel activity. Furthermore, in co-immunoprecipitation experiments it was demonstrated that both TRPC6 and TRPC7 proteins co-associate with PIP2 in un-stimulated portal vein and mesenteric artery myocytes. Previously it has been shown using biochemical assays in expression systems that several phosphoinositides including PIP2 and Ins(1,4,5)P3 bind directly to both TRPC6 and TRPC7 proteins (Kwon et al. 2007). In our experiments the membrane permeant analogue 6-Ins(1,4,5)P3 reduced association of PIP2 with TRPC7 but not TRPC6 proteins. These data suggest that the facilitatory effects of Ins(1,4,5)P3 are due to decreasing association of PIP2 to the TRPC7 subunit of the TRPC6/C7 conductance in portal vein myocytes. Also the lack of a synergistic action between Ins(1,4,5)P3 and DAG on TRPC6 channel activity in mesenteric artery myocytes is explained by the fact that Ins(1,4,5)P3 does not reduce association of PIP2 with TRPC6. In contrast the DAG analogue OAG reduces co-association between PIP2 and TRPC6 subunits but not between PIP2 and TRPC7 proteins.

Comparison of the effects of Ins(1,4,5)P3 on different TRPC channels

Previously it was shown that Ins(1,4,5)P3 stimulates expressed human TRPC3 channels (Kiselyov et al. 1998) although contradictory data have also been presented (Trebak et al. 2003). In addition it was proposed that an interaction between Ins(1,4,5)P3 and TRPC3 underlies the cation current evoked by brain-derived nerve growth factor in neonatal rat pontine neurones (Li et al. 1999). In cerebral arteries it was shown that Ins(1,4,5)P3 causes TRPC3 channel activation and vasoconstriction (Xi et al. 2008). A notable difference between the above data with the present report is that Ins(1,4,5)P3 receptor antagonists inhibited Ins(1,4,5)P3 activation of TRPC3 channels (Kiselyov et al. 1998; Li et al. 1999; Xi et al. 2008). Moreover it was also shown that stimulation of TRPC7 channels by OAG in DT40 B lymphocytes requires Ins(1,4,5)P3 receptors (Vazquez et al. 2006). In contrast, the present results demonstrate that 1 mg ml−1 heparin does not inhibit the potentiation of TRPC6/C7 channels by Ins(1,4,5)P3 in portal vein myocytes. Therefore there is no evidence that the Ins(1,4,5)P3 receptor is involved in the latter mechanism. In addition, Ins(1,4,5)P3 also potentiates the activity of another distinct channel composed of TRPC1/C5/C7 subunits in portal vein through an Ins(1,4,5)P3 receptor-independent mechanism (Liu et al. 2005; Saleh et al. 2008). However, in coronary artery myocytes Ins(1,4,5)P3 increases the activity of TRPC3/C7 channel activity, which was blocked by 1 mg ml−1 heparin, which may suggest that TRPC3 subunits confer Ins(1,4,5)P3 receptor-mediated mechanisms (Peppiatt-Wildman et al. 2007). It is apparent that Ins(1,4,5)P3 interacts with both expressed and native TRPC channels, but there are variations in the role of Ins(1,4,5)P3 receptors. It is possible that Ins(1,4,5)P3 has multiple actions on TRPC channels. The effect of Ins(1,4,5)P3 on TRPC channels via an Ins(1,4,5)P3 receptor-independent mechanism appears to be a novel and potentially important effect of this signalling molecule.

Conclusions

The present work demonstrates that Ins(1,4,5)P3 potentiates native TRPC6/C7 channel activity by decreasing the inhibitory effect of PIP2 on TRPC7 subunits independently of Ins(1,4,5)P3 receptors. In physiological conditions in portal vein myocytes, noradrenaline stimulates α1-adrenoceptors to produce hydrolysis of PIP2 by PI-PLC, which generates DAG and Ins(1,4,5)P3. Reduction of PIP2 binding to the ion channel is caused by hydrolysis and by Ins(1,4,5)P3, which facilitates rapid and optimal activation of TRPC6/C7 channels by DAG.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Appendix

Appendix

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Appendix

Author contributions

M.J. carried out the majority of the experimental work and figure preparation. J.S. contributed some electrophysiological recordings and S.N.S. carried out immunoprecipitation and Western blotting. A.P.A. and W.A.L. were involved in conception, design, interpretation of data, and drafting of manuscript. All authors approved the final manuscript.

Acknowledgements

This work was funded by The Wellcome Trust and the British Heart Foundation.