Ca2+ dependent but PKC independent signalling mediates UTP induced contraction of rat mesenteric arteries.

Uridine triphosphate (UTP) can be released from damaged cells to cause vasoconstriction. Although UTP is known to act through P2Y receptors and PLC activation in vascular smooth muscle, the role of PKC in generating the response is somewhat unclear. Here we have used Tat-linked membrane permeable peptide inhibitors of PKC to assess the general role of PKC and also of specific isoforms of PKC in the UTP induced contraction of rat mesenteric artery. We examined the effect of PKC inhibition on UTP induced contraction, increased cytoplasmic Ca(2+) and reduction of K(+) currents and found that PKC inhibition caused a relatively small attenuation of contraction but had little effect on changes in cytoplasmic Ca(2+). UTP attenuation of both voltage-gated (Kv) and ATP-dependent (KATP) K(+) currents was abolished when intracellular Ca(2+) was decreased from 100 to 20 nM. PKC inhibition reduced slightly the ability of UTP to attenuate Kv currents but had no effect on KATP current inhibition. In conclusion, both UTP induced contraction of mesenteric artery and the inhibition of Kv and KATP currents of mesenteric artery smooth muscle cells by UTP are relatively independent of PKC activation; furthermore, the inhibition of both Kv and KATP currents requires intracellular Ca(2+).

Introduction

Contraction of arterial smooth muscle results from an increase in internal [Ca2+], occurring either by entry through Ca2+ permeable channels or by release from intracellular Ca2+ stores. Many vasoconstrictors stimulate Gq linked receptors activating phospholipase C leading to generation of DAG, IP3 and subsequent activation of PKC and Ca2+ release. In many instances intracellular Ca2+ is raised further through PKC induced reduction of K+ currents causing depolarization and subsequent activation of L-type Ca2+ channels (1). For example, we and others have shown that PKC is involved in generating the contraction of arterial smooth muscle in response to both Ang II and ET-1 (2–4). UTP, which is released in response to cellular injury (5), causes contraction of vascular smooth muscle by activating G-protein coupled P2Y receptors (6, 7); but in rat cerebral arteries UTP induced contraction was unaffected by the general PKC inhibitor bisindolylmaleimide and UTP inhibition of voltage-gated K+ channels was not dependent on PKC (8). However, in rat mesenteric artery smooth muscle cells UTP, Ang II and ET-1 all increase intracellular Ca2+ and induce the translocation of PKCα, PKCδ and PKCε (9). Interestingly, the contraction induced by Ang II works almost exclusively through PKCε while ET-1 works through PKCα (4), but whether specific PKC isoforms are involved in generating UTP induced contraction is not known.

In this study we have used myography, Ca2+ imaging and electrophysiological techniques to investigate the degree of PKC involvement in the signalling mechanisms of UTP induced contraction of rat mesenteric arteries. In our study we have used cell-permeable general and isoform specific peptide inhibitors of PKC and found that PKC activation has only a minimal role in generating UTP induced contraction. Our data also show that UTP induced attenuation of K+ currents is dependent on intracellular Ca2+.

Materials and Methods

Preparation of arterial smooth muscle cells

All experiments were carried out on adult male Wistar rats (200–300 g) killed by stunning and cervical dislocation. The care of animals was in accordance with the UK Animals (scientific procedures) Act 1986. Investigations carried out in this study conformed to the Guide for the Care and use of laboratory animals published by the US National Institutes of Health (NIH publications No. 85–23 revised 1996). The procedures used in this study were approved by the University of Leicester Animal Care and Use Committee.

The mesentery was cleaned of fat and connective tissue and mesenteric arteries were either stored in ice cold physiological saline for myography or were immediately subjected to enzymatic digestion to obtain isolated smooth muscle cells as described previously by Rainbow et al. (10). Briefly, the arteries were incubated at 35 °C for 31 min in a low Ca2+ solution comprising (mM): 137 NaCl, 5.4 KCl, 0.42 Na2HPO4, 0.44 NaH2PO4, 1 MgCl2, 0.1 CaCl2, 10 Hepes, 4 glucose and 6 mannitol, adjusted with NaOH to pH7.4 to which was added (mg ml−1): 0.9 albumin, 1.4 papain and 0.9 dithioerythritol. This initial digestion was followed by a further digestion for 12.5 min in the low Ca2+ solution containing (mg ml−1): 0.9 albumin, 1.4 collagenase type F and 0.9 hyaluronidase type I-S. Arteries were then washed in low Ca2+ solution containing 0.9 mg ml−1 albumin. Individual cells were isolated by gentle trituration in low Ca2+ solution and stored at 4 °C until used.

Solutions and chemicals

For myography experiments the physiological bathing solution comprised (mM): 135 NaCl, 6 KCl, 1.2 MgCl2, 1.8 CaCl2, 4 glucose, 6 mannitol, 10 HEPES, adjusted to pH 7.4. For the 60 mM K+ solution, KCl was increased to 60 and NaCl decreased to 81 mM, other solution changes were made by adding or omitting compounds as necessary. Whole-cell recording was done with pipette solution comprising (mM): 110 KCl, 30 KOH, 1 MgCl2, 1 or 3.9 CaCl2, 1.0 Na2ATP, 0.1 ADP, 0.5 GTP, 10 EGTA, and 10 HEPES adjusted to pH 7.2. The free [Ca2+] calculated with the program Maxchelator (http://www.stanford.edu/~cpatton/maxc.html) were 20 and 100 nM for 1 and 3.9 mM CaCl2 respectively. The 6 mM K+ extracellular solution comprised (mM): 135 NaCl, 6 KCl, 1 MgCl2, 0.1 CaCl2, 10 HEPES, 4 glucose and 6 mannitol; the 140 mM K+ solution contained (mM): 140 KCl and no NaCl. Both solutions were adjusted to pH 7.4. All chemicals were obtained from Sigma-Aldrich. The general and isoform-specific PKC inhibitor peptides were linked to a HIV derived Tat peptide, C-YGRKKRRQRRR, using a disulphide link between N-terminal cysteine residues. This makes the peptides membrane permeable and the disulphide link is cleaved by the reducing environment of the cell. The peptide inhibitors used were: PKC20–28 (N-Myr-FARKGALRQ); PKCα (SLNPQWNET); PKCβI (KLFIMN); PKCβII (QEVIRN); PKCδ (SFNSYELGSL) and PKCε (EAVSLKPT). Peptides were synthesized by Pepceuticals Ltd. and linking the inhibitory peptides to the Tat peptide was done by Dr R. Norman of the Department of Cardiovascular Sciences, University of Leicester.

Myography

To record contractile force, 2–4 mm segments of third order mesenteric arteries were mounted in a Myo-interface model 500A myograph (JP Trading, Denmark). In addition to the compounds being tested, all bathing solutions contained 20 µM L-NAME to eliminate endogenous nitric oxide synthesis and were added to the bath directly, which was maintained at 37 °C.

Ca2+ imaging

Intracellular Ca2+ was measured in isolated mesenteric arterial smooth muscle cells cultured for 3 days. Cells were pre-loaded with Fura-2-AM (5 μM) for 30 min. Fluorescent signals were measured every 3 seconds following alternate excitation with 340 and 380 nm light using a monochromator (deltaRAM, Photon Technology, UK) and 45 second pulses of UTP were applied every 3 min to stimulate Ca2+ release. Emitted light >520 nm was recorded with a Roper Scientific CCD97 camera and values are expressed as the ratio of light emitted following excitations at 340 and 380 nm respectively. All measurements were made at 30–32 °C.

Electrophysiology

Whole-cell Kv and KATP currents were recorded from single smooth muscle cells using the patch clamp technique. Currents were filtered at 2 kHz (-3 dB) and recorded with an Axopatch 200A amplifier (Molecular Devices) and digitized at 10 kHz. Patch pipettes were made from thick-walled borosilicate glass (Clark Electromedical, Pangbourne, Berks, UK) using a pp-83 vertical puller (Narishige, Tokyo, Japan). Electrode resistances before sealing were 3–5 MΩ and after sealing were >1 GΩ . All experiments were done at 30 °C, maintained using a Dagan HW-30 temperature controller.

Analysis

Results are expressed as means ± S.E.M. Intergroup differences were tested by analysis of variance followed by Bonferroni's post hoc test, P<0.05 was considered statistically significant.

Results

UTP induced contraction depends on Ca2+ entry

Under control conditions (6 mM K+ and 1.8 mM Ca2+) we found that repeated 5 min applications of UTP (100 µM) with 10 min wash intervals gave consistent contractions for at least 8 applications. This enabled comparisons of the contractions before and after pharmacological interventions without desensitization of UTP responses being a major factor; thus for all myography experiments described we used 100 µM UTP. Fig. 1A shows contractile responses to 60 mM K+ and to UTP in the absence and presence of diltiazem (50 µM), an L-type Ca2+ channel blocker. As expected, the response to 60 mM K+ was completely abolished by diltiazem, indicating that the depolarization induced by 60 mM K+ activated L-type channels causing Ca2+ entry and thereby contraction. Approximately 50% of the contraction induced by UTP remained in the presence of diltiazem (Fig. 1A & C), suggesting that UTP application triggers additional mechanisms for raising intracellular Ca2+. UTP has been shown to activate P2Y receptors which are Gq-linked (6, 7), and in line with this, pre-treatment of arterial segments with the PLC inhibitor U73122 (1 µM) virtually abolished the UTP induced contractions to 4.6 ± 1.9% of control values (n=10, P<0.05).

Fig. 1.

Contractile responses of rat mesenteric artery. (A) Example trace showing contractile responses to the application of 60 mM K+ or 100 μM UTP in the absence or presence of 50 µM diltiazem as indicated. (B) Example trace showing the effect of U73122 on 100 μM UTP and 60 mM K+ induced contractions. (C) Mean contractile responses, normalized to their respective control amplitudes, to 100 μM UTP (n=15) or 60 mM K+ (n=15) in the presence of 50 μM diltiazem (n=15), or to 100 μM UTP in the presence of 1 μM U71322 (n=10). (*P<0.05, two-way ANOVA.)

UTP induces Ca2+ release

Since blocking Ca2+ entry through L-type Ca2+ channels only partially reduced UTP induced contraction we examined changes in Ca2+ induced fluorescence of Fura-2 by UTP in the presence of 0.1 and 1.3 mM external Ca2+. As can be seen in Fig. 2A & 2B, in cells bathed with 1.3 mM Ca2+ UTP induced a concentration dependent increase in Fura-2 fluorescence ratio which corresponds to an increase in cytoplasmic Ca2+. In the presence of 0.1 mM Ca2+ there was still a substantial increase in fluorescence ratio (Fig. 2C & 2D). These data suggests that UTP causes Ca2+ release from intracellular stores, as Ca2+ entry in 0.1 mM external Ca2+ would be less than 8% of that in 1.3 mM Ca2+ assuming a similar open probability of Ca2+ permeable ion channels present on the cell membrane.

Fig. 2.

UTP increases intracellular Ca2+ concentration. (A) Overlay of the Fura-2 ratios recorded from 3 cells (different colours) in response to 45 second pulses of UTP at the concentrations (in µM) indicated in the presence of 1.3 mM external Ca2+. (B) Concentration response plot of the change in Fura-2 ratio. Data points are the mean ± s.e.m. of between 65 and 185 cells measured from 3 different preparations, all in 1.3 mM external Ca2+. (C) Overlay of the Fura-2 ratios recorded from 3 cells in response to 45 second pulses of UTP (100 µM) in the presence of 0.1 or 1.3 mM external Ca2+ as indicated. (D) Mean changes in Fura-2 ratio in response to 100 μM UTP measured in cells bathed in 0.1 mM Ca2+, 1.3 mM Ca2+ or 1.3 mM Ca2+ following pre-treatment for 15 minutes with Tat-PKC20-28-IP (100 nM) as indicated. (*P<0.05, one-way ANOVA.)

PKC inhibitor peptides are relatively ineffective at inhibiting UTP induced responses

We have shown that the vasoconstrictors AngII, ET-1 and UTP all mobilize PKCα, PKCδ and PKCε in mesenteric smooth muscle cells (9). We investigated the extent of various PKC isoform involvement in UTP induced contractions by using Tat-linked membrane permeable peptide inhibitors targeting particular PKC isoforms. Previous experiments in our lab show that ET-1 induced contraction was markedly inhibited by application of Tat-PKCα-IP (4) and we therefore used this as a control to confirm the efficiency of the Tat-linked inhibitors in our current investigation. Inhibition of the ET-1 induced contraction by Tat-PKCα-IP is shown in Fig. 3A. Because ET-1 induces a prolonged contraction we quantified the effect of Tat-PKCα-IP by normalizing responses to the contractions induced by 60 mM K+ (Fig. 3B). It is evident that Tat-PKCα-IP inhibited the ET-1 induced contraction significantly and to a similar extent as in our previous work (4). UTP responses could be compared directly (normalized to control UTP induced contractions) and as Fig. 3C shows, in contrast to ET-1, UTP induced contractions were unaffected by Tat-PKCα-IP. None of the PKC isoform-specific peptide blockers affected UTP induced contractions, only the non-specific peptide blocker Tat-PKC20–28-IP produced a small but significant inhibition of UTP induced contractions (decreased to 82.8 ± 6.8%, P<0.05, n=7). We found that the UTP induced increase in Fura-2 ratio (with 1.3 mM Ca2+) was also reduced slightly, but not significantly, by pre-treatment with Tat-PKC20–28-IP (Fig. 2D).

Fig. 3.

Effect of PKC inhibitory peptides on ET-1 and UTP induced contractions. (A) Overlaid traces showing contractile responses of two arterial segments to 60 mM K+, 100 μM UTP and 3 nM ET-1. The segment corresponding to the light blue trace was exposed to Tat-PKCα-IP (100 nM) for the duration indicated; for clarity the traces have been scaled to their respective initial responses to 60 mM K+. (B) Mean contractions, normalized to those obtained with 60 K+, to 3 nM ET-1 in the absence and presence of 100 nM Tat-PKCα-IP. (C) Mean UTP (100 μM) induced contractions, normalized to control, in the presence of the Tat-linked peptide PKC inhibitors indicated (100 nM, except Tat-PKCδ-IP at 50 nM). (*P<0.05, two-way ANOVA or t-test as appropriate.)

UTP reduces Kv current in a Ca2+-dependent manner

Whole-cell Kv currents were induced following depolarizing voltage pulses to potentials more positive than about -40 mV from a holding potential of -65 mV. To minimize contamination by BK channel activity, penitrem A (100 nM), a specific BK channel blocker, was included in the external solution (11). Kv current density (normalized to cell capacitance) was quite variable in these cells; for this reason we have plotted currents normalized to control current at +60 mV to enable comparison between cells. The average cell capacitance of mesenteric arterial smooth muscle cells was 13.9 ± 0.7 pF and the average current recorded was 290 ± 52 pA at +60 mV. Kv current densities measured at +60 mV under control conditions (100 nM penitrem A) were 18.7 ± 2.7 (n=9) and 21.9 ± 4.8 (n=10) pA pF−1 where the pipette solution contained 20 and 100 nM free Ca2+ respectively; these values are not statistically different (P=0.57, two-tailed t-test). UTP has been shown to inhibit Kv currents in rat cerebral arteries in a Rho kinase dependent manner (8). We found UTP (100 µM) inhibited Kv currents of rat mesenteric artery smooth muscle by a similar amount to that reported by Luykenaar et al. (8), provided the pipette solution contained 100 nM free Ca2+ (Fig. 4C & D). Under these conditions Kv current inhibition by UTP was more prevalent at positive potentials (see Fig. 4D); UTP reduced the Kv current recorded at +60 mV to 59.8 ± 7.3% of the control amplitude (P<0.05, n=9). However, in recordings made from cells where the pipette solution contained only 20 nM free Ca2+, UTP did not inhibit Kv currents (see Fig. 4A & B).

Fig. 4.

Inhibition of Kv current by UTP requires intracellular Ca2+. (A) Examples of whole-cell currents under control conditions and 9 minutes after applying UTP (100 μM). These currents were recorded with 20 nM free Ca2+ in the patch pipette. (B) Mean I-V plots, normalized to control current at +60 mV, under the conditions of part A (n=4). (C) Examples of whole-cell currents under control conditions and 9 minutes after applying UTP (100 μM) in a cell where the pipette contained 100 nM free Ca2+. (D) Mean I-V plots, normalized to control current at +60 mV, under the conditions of part C (n=9). (E) Examples of whole-cell currents under control conditions and 9 minutes after applying UTP (100 μM). These currents were recorded from a cell pre-treated for over 10 minutes with Tat-PKC20-28-IP (100 nM) and with 100 nM free Ca2+ in the patch pipette. (F) Mean I-V plots, normalized to control current at +60 mV, under the conditions of part E (n=7).

PKC inhibition reduces but does not abolish UTP modulation of Kv currents

To assess the involvement of PKC in the UTP induced modulation of Kv currents a comparison of the inhibition was made in the absence and presence of Tat-PKC20–28-IP, a non-isoform specific inhibitor of PKC. As shown in Fig. 4E & F, UTP inhibition persisted in the presence of Tat-PKC20–28-IP, though to a lesser extent than in its absence. Again the inhibition of Kv currents by UTP in the presence of this PKC inhibitor remained more pronounced at positive potentials and was similar in extent to UTP inhibition of Kv currents of rat cerebral arteries in the presence of the non-peptide PKC inhibitors calphostin C and bisindolylmaleimide (8).

KATP channels are strongly inhibited by UTP in a Ca2+ dependent manner

To measure KATP currents the membrane potential was held at -60 mV and external K+ was raised from 6 to 140 mM, setting EK at 0 mV and resulting in inward KATP currents. At this membrane potential Kv and BK currents were virtually absent. Pinacidil (10 µM) was used to activate KATP channels further and their identity was confirmed by applying the KATP channel blocker glibenclamide (10 µM). Application of UTP (100 µM) following activation by pinacidil caused a marked reduction in KATP current when the pipette solution contained 100 nM free Ca2+, but, similar to the case with Kv currents, UTP inhibition was much less and did not reach statistical significance when the pipette contained only 20 nM free Ca2+ (Fig. 5).

Fig. 5.

Inhibition of KATP currents by UTP is not dependent on PKC activation but does require intracellular Ca2+. (A and B) Representative KATP current traces obtained at -60 mV in symmetrical 140 mM K+ following the application of pinacidil (10 μM), UTP (100 μM) and glibenclamide (10 μM) as indicated in the presence of 20 nM (A) or 100 nM (B) free Ca2+ in the patch pipette. The arrow indicates the change from 6 to 140 mM external K+. (C) Mean KATP current, normalized to that in the presence of pinacidil, under the conditions indicated. Note that blocking PKC by pre-treatment for over 10 minutes with Tat-PKC20-28-IP (100 nM) had little effect on the inhibition of the current by UTP. (*P<0.05, two-way ANOVA, n=8).

PKC inhibition has little effect on UTP modulation of KATP currents

To establish the extent of PKC involvement in the UTP induced inhibition of KATP currents cells were pre-exposed to 100 nM Tat-PKC20–28-IP for at least 10 min before recording. Whole-cell recordings were established with 100 nM Ca2+ in the pipette and a comparison of the UTP inhibition of pinacidil activated KATP currents between control cells and pre-treated cells was made. No significant difference between the inhibition of KATP current by UTP in control cells and in cells pre-treated with Tat-PKC20–28-IP (n=8) was observed (Fig. 5C).

Discussion

In the presence of a functional epithelium, endothelial P2Y1 receptor activation results in mesenteric arteriole dilation (12, 13). However, if the endothelial layer is damaged or, as in our experiments, nitric oxide synthesis is supressed, UTP causes a potent contraction of rat mesenteric artery segments. This is partly dependent on Ca2+ entry through L-type Ca2+ channels, as indicated by the ability of diltiazem to reduce the contraction to about 50% of control values (see Fig. 1C). UTP induced contraction of rat cerebral arteries was also sensitive to diltiazem but to a somewhat lesser degree (8). UTP is known to activate Gq linked P2Y1, P2Y2, P2Y4 and P2Y6 receptors on vascular smooth muscle which generate PLC dependent Ca2+ signals (6, 7, 14). In our Ca2+ measurements with 0.1 mM external Ca2+, which would support only a small Ca2+ influx, there was still a substantial increase in intracellular Ca2+ in response to UTP which is consistent with Ca2+ release from intracellular stores. Sanchez-Fernandez et al. (15) show that UTP causes Ca2+ mobilization from intracellular stores of culture bovine aortic cells which persisted in the presence of L-type channel blockers or following removal of extracellular Ca2+. Sugihara et al. (16) reported a dual action of UTP on arterial smooth muscle with contributions from both P2X and P2Y receptor signalling. These authors show that Ca2+ entry through L-type channels mediate a phasic contraction while Ca2+ release from endoplasmic reticulum caused tonic contraction of rat aortic rings (16).

We have shown previously that UTP causes mobilization of PKCα, δ and ε in these cells (9). However, examination of the PKC dependence revealed that both contraction and the increased Ca2+ in response to UTP were relatively insensitive to Tat-linked membrane permeable peptide inhibitors of PKC (see Fig. 2D & 3C). The general PKC inhibitor peptide, Tat-PKC20–28-IP, was the only one to cause a small but significant decrease (17%) in the contraction; isoform specific inhibitors were without effect, although the ET-1 contraction was strongly inhibited by Tat-PKCα-IP as we have reported previously (4).

Many vasoconstrictors reduce smooth muscle K+ currents causing depolarization and activation of L-type Ca2+ channels which increases contraction. UTP has been shown to inhibit Kv channels of rat cerebral arteries (8) and KATP channels in rat coronary arteries (17). We also found that UTP reduced Kv currents and KATP currents of rat mesenteric arteries when the pipette contained 100 nM free Ca2+ (Fig. 4C & D). Of note, however, was that lowering pipette [Ca2+] from 100 to 20 nM abolished UTP inhibition of both currents (Fig. 4A & B). This is unlikely to result from BK channel inhibition at the higher Ca2+ level as these experiments were done in the presence of the BK channel blocker penitrem A. Furthermore, no difference in current density between cells recorded with 20 or 100 nM Ca2+ in the pipette was observed, indicating that BK current was absent under our recording conditions. In rat cerebral artery UTP still reduced Kv currents in the presence of the PKC inhibitor bisindolylmaleimide, but the reduction was not to the same degree as in its absence (8); we found a similar effect with the more specific peptide PKC inhibitor Tat-PKC20–28-IP on UTP reduction of mesenteric artery smooth muscle Kv currents (Fig. 4E & F). The inhibition of mesenteric artery KATP currents by UTP was considerable, with 100 μM UTP leading to an 85% reduction in KATP current; as was the case with Kv currents, lowering pipette free [Ca2+] to 20 nM virtually abolished the effect of UTP. The inhibition of KATP currents by UTP persisted in cells pre-treated with Tat-PKC20–28-IP (Fig. 5). It is known that activation of PLC is enhanced by Ca2+ (18), and recently Jones et al. (19) have shown that Ca2+ entry through P2X receptors can enhance ADP responses acting through P2Y receptors in platelets, possibly by a mechanism that involves enhanced PLC activation. Although the mechanism whereby intracellular Ca2+ appears necessary for UTP signalling to K+ channels in our experiments is unclear, a reduced PLC activation in experiments with low (20 nM) intracellular Ca2+ is certainly plausible.

We have shown that DiC8, an analogue of DAG, is an effective blocker of Kv and to a lesser extent KATP currents in these cells (11). This raises the possibility that DAG produced following PLC activation may contribute to the UTP induced reduction of both KV and KATP currents; it should be noted, however, that another DAG analogue (OAG), did not inhibit KV currents (11). Furthermore, Luykenaar et al. (8) have shown that in rat cerebral artery smooth muscle cells the Rho kinase inhibitor Y27632 abolished the inhibition of KV currents by UTP.

In conclusion, the contraction of rat mesenteric arteries by UTP is rather insensitive to peptide inhibitors of various PKC isoforms. The contraction depends partly on Ca2+ entry through L-type Ca2+ channels, but Ca2+ release is also important. In addition to initiating contraction, it appears that intracellular Ca2+ is also necessary for UTP signalling, as the reduction in both Kv and KATP currents are virtually abolished if free [Ca2+] is lowered to 20 nM. This raises the interesting possibility that UTP induced vasospasm has a positive feedback component, where raised intracellular Ca2+ increase the ability of UTP to inhibit K+ currents, thereby leading to increased depolarization and further Ca2+ entry.

Conflict of interest

The authors declare that they have no conflict of interest.