Agents that increase tyrosine phosphorylation activate a non-selective cation current in single rabbit portal vein smooth muscle cells

Abstract

1. The effects of agents that increase tyrosine phosphorylation were studied with whole-cell recording of membrane currents in single smooth muscle cells from the rabbit portal vein.

2. In K⁺-free conditions with the chloride equilibrium potential at about -50 mV, intracellular application via the patch pipette of 200 μM sodium orthovanadate (Na₃VO₄), which inhibits tyrosine phosphatases, activated a ‘noisy’ inward current at a holding potential of -50 mV.

3. Intracellular dialysis with 100 μM (pY)EEI, a peptide activator of the cytosolic tyrosine kinase pp60^c-src, and bath application of 5 μM insulin, which activates receptor-coupled tyrosine kinases, also evoked a ‘noisy’ inward current. The current-voltage relationships and the reversal potential (about +10 mV) of the Na₃VO₄-, pp60^c-src- and insulin-induced currents were similar to those of the noradrenaline-evoked non-selective cation current (I_cat).

4. The inward currents evoked by noradrenaline, Na₃VO₄, (pY)EEI and insulin were all greatly potentiated when the bathing calcium concentration was reduced from 1.5 mM to 50 μM.

5. The single channel conductance estimated from spectral density analysis of the whole-cell current was about 20 pS for noradrenaline, Na₃VO₄, (pY)EEI and insulin. Moreover for all agents the spectra were described by the sum of two Lorentzians with similar corner frequencies.

6. Noradrenaline-evoked I_cat was inhibited to a similar degree by the tyrosine kinase inhibitors genistein and tyrphostin 23 and their inactive analogues daidzein and tyrphostin A1, respectively.

7. In the presence of Na₃VO₄, application of noradrenaline evoked a cation current of similar peak amplitude to control I_cat although the rate of decay of I_cat was enhanced in the presence of Na₃VO₄.

8. This study shows that stimulation of both cytosolic and receptor-coupled tyrosine kinases evokes a non-selective cation current and the conductance is similar to that activated by noradrenaline.

It is well recognised that phosphorylation of tyrosine residues of cellular proteins by protein tyrosine kinases (PTKs) plays an important role in the proliferation of smooth muscle cells. However, it has become increasingly apparent that PTKs may also be involved in the contraction of smooth muscle. There are two major classes of PTK. The first group are receptor-coupled PTKs, which span the cell membrane and are stimulated at the extracellular domain by agents such as epidermal growth factor (EGF) and insulin. The second group of PTKs are located in the cytoplasm, although some, like pp60^c-src, can be bound to the cytoplasmic surface of the cell membrane. A notable feature is that the activity of PTKs is unusually high in smooth muscle. For example, the apparent activity of pp60^c-src is about 500 times greater in smooth muscle extracts than in either cardiac or skeletal muscle (Di Salvo et al. 1997).

There are several lines of evidence to indicate that tyrosine phosphorylation is involved in smooth muscle contraction, which are briefly summarised below (see reviews by Hollenberg, 1994; Di Salvo et al. 1997; and Hughes & Wijetunge, 1998). First, stimulants of receptor-coupled PTKs (e.g. EGF), which cause tyrosine phosphorylation, contract many vascular preparations by a direct action on smooth muscle cells and this contraction is blocked by inhibitors of PTKs such as genistein. Second, agents that inhibit protein tyrosine phosphatases (PTPs, which dephosphorylate tyrosine), e.g. sodium orthovanadate (Na₃VO₄), increase smooth muscle tyrosine phosphorylation and may produce contraction themselves and potentiate the contractions evoked by other stimulants. It has also been demonstrated that some classical vasoconstrictors (e.g. α₁-adrenoceptor stimulants) induce tyrosine phosphorylation and contraction and both of these effects are potentiated by Na₃VO₄ and inhibited by genistein (Jin et al. 1996). Therefore activation of PTKs may induce vascular smooth muscle contraction and these kinases may also be involved in the responses to well-known vasoconstrictor agents.

There has been little published on the regulation of ion channels in vascular smooth muscle by PTKs. Wijetunge & Hughes (1995) demonstrated that inclusion of the non-receptor PTK pp60^c-src in the patch pipette solution produced an increase in the amplitude of voltage-dependent calcium currents in rabbit ear artery cells and this effect was blocked by the PTK inhibitors genistein and tyrphostin 23. There have also been a few reports to suggest that inhibitors of PTK activity modulate K⁺ currents in vascular smooth muscle cells but it is not certain whether these effects can be attributed to an alteration of PTK activity or other effects of these agents (see Hughes & Wijetunge, 1998, for details).

In the present study we demonstrate that activators of PTKs induce a non-selective cation current in rabbit portal vein smooth muscle cells. This conductance resembles the noradrenaline-evoked non-selective cation current (I_cat), which has been proposed to produce membrane depolarisation and contraction in vascular smooth muscle (Byrne & Large, 1988; Wang & Large, 1991; Helliwell & Large, 1996). Preliminary experiments from this study have appeared in abstract form (Albert et al. 1999).

METHODS

Cell isolation

New Zealand White rabbits (2-3 kg) were killed by an i.v. injection of sodium pentobarbitone (120 mg kg⁻¹) and the portal vein was removed into normal physiological salt solution (PSS). The tissue was dissected free of connective tissue and fat before being cut into strips and placed in ‘Ca²⁺-free’ PSS. The tissue was enzymatically dispersed in two sequential enzyme steps. First, the strips of tissue were incubated in ‘Ca²⁺-free’ PSS with 0.3 mg ml⁻¹ protease type XIV (Sigma) for 5 min and then washed in ‘Ca²⁺-free’ PSS. In the second step the strips were incubated with 1 mg ml⁻¹ collagenase type 1A (Sigma) in 100 μm Ca²⁺-PSS for 10 min and were then washed in 100 μm Ca²⁺-PSS. All enzyme and wash procedures were carried out at 37^oC. After the enzyme treatments, the strips were incubated in 100 μm Ca²⁺-PSS at room temperature (20-25^oC) for 10 min before the cells were released into the solution by gentle mechanical agitation of the strips of tissue using a wide-bore Pasteur pipette. The suspension of cells was then centrifuged (1000 r.p.m.) to form a loose pellet, which was resuspended in 0.75 mm Ca²⁺-PSS. The cells were then plated onto glass coverslips and stored at 4^oC before use (1-6 h).

The normal PSS contained (mm): NaCl 126, KCl 6, CaCl₂ 1.5, MgCl₂ 1.2, glucose 10 and Hepes 11, and the pH was adjusted to 7.2 with 10 m NaOH. ‘Ca²⁺-free’ PSS, 100 μm Ca²⁺-PSS and 0.75 mm Ca²⁺-PSS had the same composition except that either Ca²⁺ was omitted or 1.5 mm CaCl₂ was replaced by 100 μm CaCl₂ and 0.75 mm CaCl₂, respectively.

Electrophysiology

Cell membrane currents were recorded with a List L/M-PC patch clamp amplifier at room temperature using the standard whole-cell recording configuration. Patch pipettes were manufactured from borosilicate glass and had resistances between 4 and 8 MΩ when filled with the standard internal patch pipette solution. Series resistance was not compensated. The liquid junction potential between the pipette solution and the external solution was calculated by measuring the change in voltage when the external solution was replaced with pipette solution. The change in voltage was < 3 mV and was not compensated for in the final records. All experiments were carried out at a holding potential of -50 mV. The voltage protocol used to evaluate the current-voltage (I-V) characteristics of evoked currents involved stepping to -120 mV for 50 ms from the holding potential of -50 mV before a voltage ramp was applied (0.3 V s⁻¹) to +40 mV. To calculate the leak-subtracted I-V characteristics, ramps applied before current activation were subtracted from those applied at the peak amplitude of the evoked current. To avoid errors in calculating the I-V relationships due to the current fluctuations or ‘noise’ at least three ramps were used to obtain control and evoked currents. To further remove the variability of the current amplitudes the I-V relationships were measured at 20 mV intervals from -120 to +40 mV and normalised to the current amplitude evoked at -40 mV. The voltage ramps were generated and the data captured with a Pentium (P5-100) personal computer (Gateway, Ireland) using a CED 1401plus interface and software (Cambridge Electronic Design Ltd, Cambridge, UK). Data were filtered at 1 kHz and sampled at 5 kHz. Long term recordings were played back from the analog output of a CDATA digital tape-recorder (Cygnus Technology Inc., Delaware, PA, USA) using the 1401plus and CED sigavg software. Data were plotted for analysis and figure preparation using MicroCal Origin software (MicroCal Software Inc., USA). The activation rate of the noradrenaline-evoked I_cat was calculated as the time taken for the inward current to increase from 10 to 90% of the peak amplitude.

Noise analysis

For a comprehensive description of the procedures used for data preparation and for calculating the spectal density function of I_cat and estimating the single channel conductance of the non-selective cation channels, refer to Helliwell & Large (1998). In brief, the signal was first high-pass filtered (0.3 Hz, -3 db) before being low-pass filtered (800 Hz, -3 db), using 8-pole Butterworth filters (Barr and Stroud, UK). This band-pass signal was then amplified (x 10) prior to sampling at 2 kHz using the 1401plus and Spike2 software (Cambridge Electronic Design Ltd). Spectral density analysis was performed using Spike2 software between frequencies of 1 and 500 Hz on a 10 s section of data where I_cat remained constant. Leak-subtracted spectra were calculated by subtracting a 10 s section of the holding current before I_cat was evoked. The spectra were described and fitted by the sum of two Lorentzians with Origin software using a least squares method where each parameter was unconstrained.

Solutions and drugs

The cells were perfused in a standard K⁺-free external solution containing (mm): NaCl 126, CaCl₂ 1.5 or 0.05, Hepes 11 and glucose 10, pH to 7.2 with NaOH. The standard internal patch pipette solution contained (mm): CsCl 18, caesium aspartate 108, MgCl₂ 1.2, Hepes 10, glucose 11, BAPTA 10 and CaCl₂ 1 (free internal calcium ([Ca²⁺]_i) approximately 14 nm, calculated using EQCAL software), pH to 7.2 with Tris. Under these conditions voltage-gated Ca²⁺ currents, K⁺ currents and calcium-activated conductances were abolished at the holding potential of -50 mV and I_cat could be recorded in isolation. Moreover, with the anion gradient the chloride equilibrium potential (E_Cl) was about -50 mV, i.e. the holding potential. Propranolol (1 μm) was added to all solutions containing noradrenaline to block β-adrenoceptor activation. In experiments evaluating I-V characteristics, 2-5 μm nicardipine, 100 μm niflumic acid and 100 μm DIDS were added to prevent contamination by voltage-dependent Ca²⁺ currents, Ca²⁺-activated Cl⁻ currents and volume-activated Cl⁻ currents. Drugs were from Sigma (UK) except (pY)EEI peptide, which was from Biomol (USA). Na₃VO₄ and (pY)EEI were dissolved in H₂O and stored as a 10 mm stock solution at -20°C and then diluted down to the final concentration on the day of experimentation. Bovine pancreas insulin was made up on the day of the experiment by preparing a 10 mm stock solution in dilute HCl which was then sonicated and subsequently diluted down to the final concentration immediately prior to use. The values are the mean of n cells ±s.e.m. Statistical analysis was carried out using Student's t test and the level of significance set at P < 0.05.

RESULTS

Current induced by the tyrosine phosphatase inhibitor sodium orthovanadate

Figure 1A illustrates a typical noradrenaline-activated I_cat in 1.5 mm extracellular calcium ([Ca²⁺]_o) at a holding potential of -50 mV. In this series of experiments bath application of 100 μm noradrenaline evoked an inward current with a 10-90% activation rate of 7 ± 2 pA s⁻¹(n = 11) and a peak amplitude of 24 ± 5 pA, which declined to 15 ± 2 pA after 120 s in the continued presence of the agonist. This was a non-selective cation current (I_cat), which has been described previously (Byrne & Large, 1988; Wang & Large, 1991).

Figure 1. Comparison of noradrenaline-evoked I_cat and the current activated by intracellular Na₃VO₄.

A, I_cat evoked by bath application of 100 μm noradrenaline, denoted by the filled bar. B, inward current evoked by inclusion of 200 μm Na₃VO₄ in the patch pipette. The asterisk indicates the time of achieving the whole-cell configuration. In A and B the holding potential was -50 mV. C, the mean I-V relationship of the noradrenaline-activated I_cat in 5 cells. D, the mean I-V relationship of the Na₃VO₄-activated inward current in 4 cells.

To increase tyrosine phosphorylation in the cell the competitive PTP inhibitor Na₃VO₄ (Huyer et al. 1997) was included in the standard patch pipette internal solution. Inclusion of 200 μm Na₃VO₄ in the patch pipette evoked a slowly developing inward current with a typically ‘noisy’ appearance in 1.5 mm[Ca²⁺]_o (Fig. 1b). The mean time from achieving the whole-cell configuration to the onset of the inward current was 107 ± 17 s (n = 18) and the inward current reached a mean peak amplitude of 6 ± 1 pA after 350 ± 41 s. In control cells, where Na₃VO₄ was not included in the patch pipette, a ‘noisy’ inward current was not activated (holding current at -50 mV was 22 ± 1.3 pA, n = 5) after bath perfusion with 1.5 mm[Ca²⁺]_o for 526 ± 67 s (n = 5).

The similar ‘noisy’ appearance of Na₃VO₄-evoked inward currents compared to noradrenaline-activated I_cat, under conditions that minimised potential contamination from other ion currents (see Methods), suggests that Na₃VO₄ may be activating the same non-selective cation channels. This was initially investigated by comparing the I-V characteristics of the noradrenaline-activated I_cat and the Na₃VO₄-evoked inward current using slow ramps (see Methods). Similar to I_cat described by Helliwell & Large (1996), the I-V relationships of noradrenaline-activated I_cat were linear between -40 mV and the reversal potential (V_rev) of about +10 mV but exhibited inward rectification at voltages positive to V_rev (Fig. 1C). At voltages negative to -40 mV the I-V relationships exhibited rectifying properties such that the relationships became S shaped. The I-V relationship of Na₃VO₄-evoked inward current also exhibited similar inward rectification and S-shaped characteristics (Fig. 1D). Moreover the V_rev for noradrenaline-induced I_cat (+10 ± 1 mV, n = 5) and the Na₃VO₄-evoked current (+12 ± 3 mV, n = 4) were similar. The ‘noisy’ appearance and the I-V characteristics of the Na₃VO₄-evoked inward current suggest that tyrosine phosphorylation may activate the same non-selective cation channels underlying the noradrenaline-activated I_cat.

Activators of cytosolic and receptor tyrosine kinases evoke inward currents

The above results indicate that tyrosine phosphorylation induces an inward current in vascular smooth muscle cells and therefore we investigated whether cytosolic and receptor-coupled PTK activators also induce an inward current. Thus, we investigated the potential role of the cytoplasmic PTK pp60^c-src in mediating the tyrosine phosphorylation-activated I_cat. To activate pp60^c-src we used the 11 amino acid residue peptide (pY)EEI, which contains the binding motif for the src homology domain (SH-2 domain) found within pp60^c-src. Binding of the SH-2 domain to this motif leads to the activation of pp60^c-src. Inclusion of 100 μm (pY)EEI in the patch pipette elicited a slowly developing inward current with a typically ‘noisy’ appearance (Fig. 2A). The time from achieving the whole-cell configuration to the onset of the inward current varied between 15 and 330 s with a mean onset time of 194 ± 59 s (n = 5) and the inward current reached a mean peak amplitude of 6 ± 2 pA after 365 ± 85 s.

Figure 2. Currents produced by (pY)EEI and insulin.

A, inclusion of 100 μm (pY)EEI peptide in the patch pipette activated a ‘noisy’ inward current 20 s after achieving the whole-cell configuration (denoted by the asterisk). B, bath application of 5 μm insulin (denoted by filled bar) activated an inward current with a peak amplitude of 10 pA. In A and B the holding potential was -50 mV. C, the mean I-V relationship of the (pY)EEI-evoked inward current from 3 cells. D, the mean I-V relationship of the insulin-evoked inward current from 4 cells.

To study the role of receptor-coupled PTKs we investigated the effect of stimulating the insulin tyrosine kinase receptor. Bath application of 5 μm insulin evoked a slowly developing inward current with a typically ‘noisy’ appearance (Fig. 2b). The lag period before the current developed varied between 5 and 400 s with a mean lag time of 164 ± 94 s (n = 5) and reached a mean peak amplitude of 7 ± 2 pA after 540 ± 152 s.

Figure 2C and D illustrates the I-V relationships of the currents activated by, respectively, (pY)EEI and insulin. It can be seen that the I-V relationships have similar S-shaped characteristics to that of noradrenaline. Moreover, V_rev was +10 ± 0.3 mV (n = 3) and +10 ± 1.3 mV (n = 4) for (pY)EEI and insulin, respectively, which are similar to that for noradrenaline.

These results show that stimulation of both pp60^c-src and the insulin receptor tyrosine kinase can evoke inward currents with characteristics similar to those of noradrenaline-evoked I_cat in rabbit portal vein smooth muscle cells.

The effect of extracellular Ca²⁺ on the Na₃VO₄-, (pY)EEI- and insulin-activated I_cat

The similar I-V characteristics of the noradrenaline- and Na₃VO₄-induced currents and also the ‘noisy’ appearance of the currents evoked by (pY)EEI and insulin suggest that the agents which increase tyrosine phosphorylation activate the same I_cat as noradrenaline. Consequently we carried out further experiments to confirm this notion.

Previous studies have shown that reducing [Ca²⁺]_o increases the peak amplitude of the noradrenaline-activated I_cat (Helliwell & Large, 1996). Therefore we compared the effect of reducing [Ca²⁺]_o from 1.5 mm to 50 μm on the tyrosine phosphorylation- and noradrenaline-activated I_cat. Figure 3A shows that changing the [Ca²⁺]_o from 1.5 mm to 50 μm potentiated the amplitude of the noradrenaline-activated I_cat from 26 to 196 pA. In nine cells, bath application of 100 μm noradrenaline activated a ‘noisy’ inward current with a mean peak amplitude of 10 ± 2 pA, which was increased to 82 ± 25 pA on changing from 1.5 mm to 50 μm[Ca²⁺]_o. This potentiation was statistically significant (P < 0.0001) compared to the leak inward holding current change of 10 ± 1 pA (n = 45) evoked when changing from 1.5 mm to 50 μm[Ca²⁺]_o in the absence of noradrenaline. Taking account of the leak changes, the effect on I_cat produced by lowering the [Ca²⁺]_o from 1.5 mm to 50 μm equated to a 6 (± 1)-fold potentiation.

Figure 3. Effect of [Ca²⁺]_o on the noradrenaline-, Na₃VO₄-, (pY)EEI- and insulin-activated I_cat.

Inward currents were activated by extracellular 100 μm noradrenaline (A), intracellular 200 μm Na₃VO₄(B), intracellular 100 μm (pY)EEI (C) and extracellular 5 μm insulin (D). Currents were initially evoked in 1.5 mm[Ca²⁺]_o which was then lowered to 50 μm[Ca²⁺]_o during activation of I_cat as denoted above the records. In A and D noradrenaline and insulin, respectively, were bath applied as indicated by the horizontal bar. In B and C the asterisk indicates the time of achieving the whole-cell configuration. All the experiments were carried out at a holding potential of -50 mV.

Figure 3B illustrates that changing [Ca²⁺]_o from 1.5 mm to 50 μm also potentiated the peak amplitude of a Na₃VO₄-evoked inward current from 12 to 38 pA. In 14 cells, the mean peak amplitude of the Na₃VO₄-evoked inward current was significantly increased from 6 ± 1 to 22 ± 3 pA (P < 0.001). Taking leak changes into account this equates to a 3 (± 0.2)-fold potentiation.

Changing [Ca²⁺]_o from 1.5 mm to 50 μm also increased the peak amplitude of a (pY)EEI-evoked inward current from 4 to 22 pA (Fig. 3c). In five cells, the mean peak amplitude of the (pY)EEI-activated inward current was significantly potentiated from 6 ± 2 to 24 ± 4 pA (P < 0.001), which equates to a 3 (± 0.4)-fold increase.

In Fig. 3D, the peak amplitude of an insulin-evoked inward current is shown to increase from 7 to 30 pA on changing [Ca²⁺]_o from 1.5 mm to 50 μm. In five cells the mean peak amplitude of the insulin-activated inward current was significantly potentiated from 7 ± 2 to 25 ± 5 pA (P < 0.01), equating to a 2 (± 0.3)-fold increase in amplitude.

Noise analysis of noradrenaline-activated I_cat and the tyrosine phosphorylation-activated inward currents

This study shows that agents which increase tyrosine phosphorylation by inhibiting PTPs or by stimulating pp60^c-src and the insulin tyrosine kinase receptor activate an inward current with a similar ‘noisy’ appearance, I-V characteristics and sensitivity to changes in [Ca²⁺]_o to the noradrenaline-activated I_cat. To investigate further whether tyrosine phosphorylation activates the same non-selective cation channels as those underlying noradrenaline-evoked I_cat, we used noise analysis to compare the kinetic behaviour and single channel conductance of the marcoscopic whole-cell currents in 1.5 mm[Ca²⁺]_o.

Figure 4A illustrates a typical spectral density function of a noradrenaline-activated I_cat, which is described by the sum of two Lorentzians and thus suggests the presence of three resolvable states (Helliwell & Large, 1998). In this example the corner frequencies of the lower (f_c(1)) and the higher (f_c(2)) Lorentzian components were 11 and 89 Hz, respectively. The mean values of f_c(1) and f_c(2) are shown as time constants τ₁ and τ₂ in Table 1. The relative contribution of the two Lorentzians within the total variance of the signal is quantified as a ratio of the zero frequency asymptote, G₂(0), of the higher frequency Lorentzian component divided by the zero frequency asymptote, G₁(0), of the lower frequency Lorentzian component. The G(0) ratio for the noradrenaline-activated I_cat was 0.06 ± 0.01 (Table 1), reflecting a predominant contribution from the lower frequency Lorentzian component. The single channel conductance (γ) of the cation channels underlying the noradrenaline-activated I_cat was estimated from the spectral density functions to be 19 ± 2 pS (Table 1).

Figure 4. Spectral density functions of the noradrenaline-, Na₃VO₄-, (pY)EEI- and insulin-activated I_cat.

A, a typical example of the spectral density function of noradrenaline-activated I_cat in 1.5 mm[Ca²⁺]_o. The spectra could be described by two Lorentzians with corner frequencies of 11 and 89 Hz. The f_c(1) and f_c(2) corner frequencies are represented as the time constants τ₁ and τ₂, respectively. The estimated single channel conductance (γ) of the noradrenaline-activated I_cat was 23 pS. B-D, examples of the spectral density function of the Na₃VO₄-, (pY)EEI- and insulin-evoked inward currents, respectively, showing similar noise characteristics to those of the noradrenaline-activated I_cat. The estimated single channel conductance of the Na₃VO₄-, (pY)EEI- and insulin-activated inward currents was 24, 23 and 19 pS, respectively.

Table 1.

Characteristics of the noradrenaline-activated I_cat and the Na₃VO₄-, pYEEI- and insulin-evoked inward currents obtained from the spectral density function in 1.5 mM [Ca²⁺]₀

	τ1	τ2	G2(0)/G1(0)	γ	n
	(ms)	(ms)		(pS)
NA-activated Icat	15 ± 1.0	1.6 ± 0.3	0.06 ± 0.01	19 ± 2.0	6
Na3VO4-evoked current	15 ± 0.8	2.0 ± 0.2	0.06 ± 0.01	23 ± 2.0	7
(pY)EEI-evoked current	14 ± 0.9	1.9 ± 0.3	0.05 ± 0.02	22 ± 2.5	8
Insulin-evoked current	14 ± 0.5	1.5 ± 0.1	0.04 ± 0.01	21 ± 1.8	4

Time constants τ₁ and τ₂ were calculated from the respective corner frequencies, f_c(1) and f_c(2), according to the equation τ=½2πf_c. NA, noradrenaline.

Spectral density function analysis of the Na₃VO₄-, (pY)EEI- and insulin-activated inward currents in 1.5 mm[Ca²⁺]_o produced similar results to those obtained with the noradrenaline-activated I_cat (Fig. 4B-D and Table 1). The lower Lorentzian time constants (τ₁) were approximately 15 ms, the higher Lorentzian time constants (τ₂) were between 1.5 and 2 ms and in all three spectra the lower frequency Lorentzian component had the greater contribution to the total variance (Table 1). In addition, the single channel conductances underlying the tyrosine phosphorylation-activated inward currents were approximately 20 pS (Table 1).

Ion substitution experiments

In order to demonstrate that the currents evoked by the PTK activators were non-selective cation currents it would be desirable to measure V_rev after replacing external Na⁺ ions with impermeable cations. In four experiments we replaced half of the external NaCl with TrisCl (i.e. 63 mm NaCl + 63 mm TrisCl) and in these conditions the currents induced by 200 μm Na₃VO₄ included in the patch pipette solution were reduced by 93 ± 6%(n = 4). Consequently it was not possible to obtain an accurate estimate of V_rev in external TrisCl solution. Nevertheless the reduction of the current produced by the impermeant cation Tris is consistent with the current activated by Na₃VO₄ being a non-selective cation current.

Effect of inhibitors of tyrosine kinases on I_cat

In the next series of experiments we investigated the effects of PTK inhibitors on I_cat. Figure 5A shows a control noradrenaline-evoked I_cat and in Fig. 5B it can be seen that bath application of 75 μm genistein produced a marked and reversible inhibition of I_cat. The concentration-effect relationship of genistein is illustrated in Fig. 5D where it can be seen that the concentration required to reduce I_cat by 50% (IC₅₀) was 30 μm genistein. However, it was apparent that the analogue of genistein daidzein, which is ineffective against PTKs, also caused inhibition of I_cat (Fig. 5c). Thus 75 μm daidzein reduced I_cat by 91 ± 4%(n = 5) compared to 95 ± 3%(n = 5) for 75 μm genistein. The PTK inhibitor tyrphostin 23 (50 μm) also reduced I_cat by 70 ± 4%(n = 4). However, its inactive analogue tyrphostin A1 (50 μm) also reduced I_cat by 67 ± 7%(n = 4). Therefore it appears that these pharmacological inhibitors of PTKs have non-selective effects on I_cat. The rapid onset of action suggests that these compounds might, for example, act by direct blockade of the cation channel.

Figure 5. Effects of genistein and daidzein on I_cat.

A, control I_cat recorded by bath application of 100 μm noradrenaline. B and C illustrate the inhibition of I_cat by 75 μm genistein and 75 μm daidzein, respectively, applied during the sustained phase of the noradrenaline-evoked I_cat. In A-C the holding potential was -50 mV. D illustrates the concentration-dependent inhibitory effect of genistein where the current amplitude in the presence of genistein was normalised to the current amplitude immediately prior to application of the antagonist. The relative current was plotted against various concentrations of genistein on a logarithmic scale. The data could be described by a logistic equation of the following form:

where x denotes the concentration of genistein and n the Hill coefficent. The calculated IC₅₀ value was 30 μm and the Hill coefficient was 2. Each data point represents the mean of 4-5 cells.

Noradrenaline-activated I_cat in the presence of Na₃VO₄

Our results suggest that inhibition of PTPs activates a non-selective cation current with similar characteristics to those of noradrenaline-evoked I_cat. Therefore we investigated the effect of Na₃VO₄ on the response to noradrenaline. In these experiments 200 μm Na₃VO₄ was applied in the patch pipette solution and after 5 min a small cation current was activated (Fig. 6A). Then 100 μm noradrenaline was applied to the cells and the amplitude and time course of the noradrenaline-evoked I_cat were measured. The mean peak amplitude of the noradrenaline-activated I_cat was not significantly changed, being 81 ± 24 pA (n = 7) and 94 ± 21 pA (n = 6), respectively, in the absence and presence of 200 μm Na₃VO₄ in the patch pipette solution. However, in the presence of 200 μm Na₃VO₄ the rate of decay of the noradrenaline-evoked I_cat was significantly increased (Fig. 6a). Thus after 240 s I_cat had declined to 0.24 ± 0.01 (n = 4) of the peak current in control conditions and to 0.02 ± 0.01 (n = 4) in the presence of Na₃VO₄ in the pipette solution (Fig. 6b).

Figure 6. Effect of Na₃VO₄ on the noradrenaline-evoked I_cat.

A, typical noradrenaline-evoked I_cat in the presence and absence of 200 μm Na₃VO₄ recorded from the same population of cells. Noradrenaline-evoked I_cat was activated in 50 μm[Ca²⁺]_o to increase the peak current amplitude. The current traces were normalised to the peak amplitude to illustrate the changes in the time course, since the two currents had similar peak amplitudes in the presence (210 pA) and absence (180 pA) of Na₃VO₄ in the patch pipette. The holding potential was -50 mV. Note that the peak amplitude and ‘noisy’ appearance of the Na₃VO₄-evoked I_cat are significantly increased on reducing [Ca²⁺]_o from 1.5 mm to 50 μm. B illustrates the time course of the decay of noradrenaline-evoked I_cat in the absence (▪) and presence (□) of 200 μm Na₃VO₄. The current was normalised to the peak amplitude (1.0). Asterisks represent statistically significant differences from control values: *P < 0.05, **P < 0.01 and ***P < 0.001.

DISCUSSION

The present experiments show that agents which cause an increase in tyrosine phosphorylation induced an inward current in rabbit portal vein smooth muscle cells. Biochemical measurements were not carried out in our cells but nevertheless it seems likely that the conductance change occurred as a result of tyrosine phosphorylation because we used agents that increase tyrosine phosphorylation by three distinct mechanisms: sodium orthovanadate increases tyrosine phosphorylation in smooth muscle by inhibiting PTPs (Di Salvo et al. 1997); (pY)EEI increases the activity of pp60^c-src severalfold (Hughes & Wijetunge, 1998); and insulin is a well-known activator of receptor-coupled PTKs. The data with PTK inhibitors were inconclusive as the inactive analogues were as potent at blocking I_cat as the PTK inhibitors. It is suggested that these compounds might inhibit the cation channel directly, i.e. downstream of the PTK, although it is possible that the inactive compounds may act as inhibitors of undefined PTKs. Nevertheless the data with the PTK activators provide strong evidence that tyrosine phosphorylation stimulates the opening of non-selective cation channels in rabbit portal vein smooth muscle cells. It is worth reiterating that this conductance mechanism is activated by both receptor-coupled and cytosolic PTKs.

Our data are in direct contrast to the results obtained in cultured porcine coronary smooth muscle where it was shown that genistein, an inhibitor of PTKs, activated a large conductance (140 pS) non-selective cation channel (Minami et al. 1994). However, in guinea-pig ileum it has been proposed that tyrosine phosphorylation may be involved in the stimulation of the muscarinic receptor-operated non-selective cation channel (Inoue et al. 1994). It is possible that PTKs may have a general role in the activation or modulation of non-selective cation channels in smooth muscle.

With the present conditions of K⁺-free solutions and the E_Cl set at about -50 mV (the holding potential), the inward current, with a V_rev of about +10 mV, evoked by the activators of tyrosine phosphorylation is likely to result from the opening of non-selective cation channels. Indeed the characteristics of the tyrosine phosphorylation-evoked current and noradrenaline-induced I_cat were similar in terms of current-voltage relationship, reversal potential, sensitivity to external Ca²⁺ ions, and single channel conductance and kinetic parameters estimated by spectral density analysis. It is worth noting that the single channel conductance estimated from fluctuation analysis in the present paper with noradrenaline (about 20 pS) is similar to the single channel conductance obtained in isolated patches with phenylephrine (23 pS; Inoue & Kuriyama, 1993) and preliminary data in isolated patches in our laboratory (authors' unpublished data). It was not possible to obtain V_rev measurements when external Na⁺ ions were replaced by Tris because the amplitude of the current was reduced by over 90%. Nevertheless the observation that the current was decreased when Na⁺ was replaced by the relatively impermeant cation Tris is in accordance with the idea that the inward current is carried by Na⁺ ions, i.e. it is a non-selective cation current. It is well established that I_cat stimulated by noradrenaline is a non-selective cation conductance (Byrne & Large, 1988; Wang & Large, 1991) and therefore it can be clearly stated that noradrenaline and tyrosine phosphorylation activate a similar conductance. However, there were some differences between the noradrenaline- and the tyrosine phosphorylation-evoked I_cat. The amplitude and rate of rise of the noradrenaline-evoked I_cat were greater than the values of tyrosine phosphorylation-induced I_cat. These discrepancies may be due to differences in the efficacy of the transduction mechanisms linking α₁-adrenoceptors and tyrosine phosphorylation to the cation channel. Also, the slowness of the response to Na₃VO₄ and the peptide (pY)EEI may reflect to some extent the time for intracellular dialysis through the patch pipette. This may reflect that noradrenaline can activate the cation channel by a more direct route than PTK activators, e.g. direct G-protein activation of the channel. Also, tyrosine phosphorylation-induced I_cat was potentiated to a lesser extent by reducing the external Ca²⁺ concentration than was seen with noradrenaline. There is no obvious explanation for this observation and further work is needed to provide an answer.

It is not clear how tyrosine phosphorylation opens the non-selective cation channel but it is possible that pp60^c-src could have direct effects on the channel or may regulate the activity of other effector mechanisms such as other kinases. Previously we have shown that noradrenaline activates I_cat via a G-protein coupled to phospholipase C (PLC) and the resulting 1,2-diacyl-sn-glycerol (DAG) has a central role in the activation of the cation channel by a protein kinase C-independent mechanism (Helliwell & Large, 1997). Moreover, recently we proposed that myosin light chain kinase (MLCK), or an isoform of MLCK, may also be involved in the transduction mechanism linking the α₁-adrenoceptor to the opening of the non-selective cation channels (Aromoloran et al. 2000). The cascade we proposed was as follows: α₁-adrenoceptor - G-protein - PLC activation - DAG - MLCK - opening of cation channel.

It is possible that PTKs may be involved in this model but the results with the PTK inhibitors did not provide evidence supporting this proposal as the inactive analogues also inhibited I_cat. It is also possible that the PTK pathway represents an alternative route for channel activation and non-selective cation channels may be opened by several biochemical cascades that converge on the ion channel in vascular smooth muscle. It is becoming increasingly evident that phosphorylation is involved in the activation of I_cat (see also Aromolaran et al. 2000) and it is tempting to speculate that phosphorylation of the channel protein itself leads to channel opening, as occurs, for example, with the cystic fibrosis transmembrane regulator channel. In this case there may be more than one phosphorylation site on the cation channel molecule, e.g. one as a site substrate for PTKs and another site as a target for MLCK. Further experiments are needed to confirm the role of phosphorylation in cation channel opening.

A few experiments were carried out to investigate the interaction between tyrosine phosphorylation-induced I_cat and noradrenaline-evoked I_cat. In this study, after a small current had been activated by intracellular application of Na₃VO₄, the peak amplitude of the noradrenaline-evoked I_cat was similar to the control value. This may mean that Na₃VO₄ and noradrenaline activate different populations of the same type of cation channel. However, additive currents may not have been observed because the concentration of noradrenaline may have been supramaximal. With Na₃VO₄ the rate of decay of the current was increased, suggesting an interaction between the noradrenaline-evoked pathway and tyrosine phosphorylation. This interaction may be at the level of the receptor, the transduction mechanism or the channel protein itself. The present work does not reveal the underlying mechanisms and further work is required using single channel recording to provide a better understanding of the interplay between tyrosine phosphorylation and the α₁-adrenoceptor cascade.

The present work provides an electrophysiological mechanism underlying the contraction of vascular smooth muscle produced by tyrosine phosphorylation described in the Introduction. Stimulation of I_cat will produce membrane depolarisation and the consequent opening of voltage-dependent calcium channels to increase intracellular calcium concentration and thus evoke contraction. In addition it has been shown that this non-selective cation conductance is permeable to divalent cations (Byrne & Large, 1988; Wang & Large, 1991; Helliwell & Large, 1996) and therefore influx of Ca²⁺ through this channel may produce contraction directly, independently of voltage-dependent calcium channels.

In conclusion this study shows that both receptor-coupled and cytosolic PTKs evoke a non-selective cation current in rabbit portal vein smooth muscle cells and that this conductance has similar properties to those of noradrenaline-induced I_cat. Moreover this cation current can be activated by receptor and intracellular mechanisms that are separate from G-protein-linked membrane receptors and provides evidence for an important physiological role of tyrosine phosphorylation in regulating vascular smooth muscle tone.