- Email: [email protected]
Regulation of voltage-gated cardiac sodium current by epidermal growth factor receptor kinase in guinea pig ventricular myocytes
Journal of Molecular and Cellular Cardiology 42 (2007) 760 – 768 www.elsevier.com/locate/yjmcc
Original article
Regulation of voltage-gated cardiac sodium current by epidermal growth factor receptor kinase in guinea pig ventricular myocytes Hui Liu a , Hai-Ying Sun a , Chu-Pak Lau a,c , Gui-Rong Li a,b,c,⁎ a b c
Department of Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China Department of Physiology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China Research Centre of Heart, Brain, Hormone and Healthy Aging, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China Received 30 July 2006; received in revised form 15 October 2006; accepted 23 October 2006 Available online 22 December 2006
Abstract Voltage-gated cardiac fast sodium channel current (INa) plays a critical role in the initiation and propagation of the myocardial action potential, and regulation of cardiac INa by protein tyrosine kinases (PTKs) is not well documented, though it is known that ion channels are among the targets of PTKs. The present study was therefore designed to investigate whether/how cardiac INa was modulated by PTKs in guinea pig ventricular myocytes using whole-cell patch clamp and immunoprecipitation and Western blotting approaches. It was found that cardiac INa was enhanced by epidermal growth factor (EGF), and the effect was antagonized by the selective epidermal growth factor receptor (EGFR) kinase inhibitor tyrphostin AG556 while potentiated by orthovanadate (a protein tyrosine phosphatase (PTP) inhibitor). In addition, AG556 inhibited, while orthovanadate increased INa, and the inhibition of INa by AG556 was antagonized by orthovanadate. Immunoprecipitation and Western blotting analysis demonstrated that tyrosine phosphorylation level of cardiac sodium channels was enhanced by EGF or orthovanadate, and reduced by AG556. The AG556-induced reduction of phosphorylation level was significantly reversed by orthovanadate. Our results demonstrate the novel information that EGFR kinase enhances, and PTPs reduce native cardiac INa in guinea pig ventricular myocytes. © 2006 Elsevier Inc. All rights reserved. Keywords: Protein tyrosine kinases; Epidermal growth factor receptor; Protein tyrosine phosphatases; Voltage-gated sodium current; Ion channels; Guinea pig ventricular myocytes; Heart
1. Introduction Voltage-gated cardiac fast sodium channel current (INa) plays a critical role in the initiation and propagation of myocardial action potential, and is a primary target for the development of antiarrhythmic agents. It is generally believed that cardiac INa, like other types of ion channel currents, is regulated by protein phosphorylation [1,2]. It has been demonstrated that cardiac INa was modulated by the serine/threonine protein kinases, protein kinase A (PKA) and/or protein kinase C (PKC) [3–8].
⁎ Corresponding author. Department of Medicine, L8-01, Laboratory Block, FMB, The University of Hong Kong, 21 Sassoon Road, Pokfulam, Hong Kong SAR China. Tel.: +86 852 2819 9513; fax: +86 852 2855 9730. E-mail address: [email protected] (G.-R. Li). 0022-2828/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.yjmcc.2006.10.013
Protein tyrosine kinases (PTKs), including receptor PTKs (e.g. EGFR kinase, epidermal growth factor receptor kinase) and nonreceptor PTKs (e.g. Src-related kinases), are associated with long term cellular processes such as cell growth, differentiation, and oncogenesis [1,9]. Recent studies showed that protein phosphorylation of tyrosine residues modulated ion channels [1,10,11], including Ca2+, and several types of K+ channels [11–14], as well as volume-sensitive Cl− channels [15] in different types of cells. However, the regulation of cardiac INa by PTKs is not fully understood. Although several PTK inhibitors were recently shown to inhibit INa in rabbit ventricular myocytes [16], and a very recent study showed that cloned rat cardiac sodium channel NaV1.5 was modulated by Fyn, a Src-family tyrosine kinase [17], it is unknown whether cardiac INa would be modulated by EGFR kinase. The present study was therefore designed to use whole-cell patch voltage
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
clamp, and immunoprecipitation and Western blotting approaches to investigate whether and how cardiac INa would be regulated by EGFR kinase in guinea pig ventricular myocytes. We found that activation of EGFR kinase increased cardiac INa in ventricular myocytes from guinea pig heart. 2. Materials and methods 2.1. Myocyte preparation Guinea pigs of either gender (250–300 g) were sacrificed by cervical dislocation after anesthetization with pentobarbital (40 mg/kg, i.p.) in accordance with the guideline of Animal Care and Use Committee for Teaching and Research of University of Hong Kong. Ventricular myocytes from guinea pig hearts were enzymatically dissociated by the procedure described previously [18] and the isolated myocytes were kept in a high-K+ storage medium [18] at room temperature for 1 h before electrophysiological recording. 2.2. Solutions and chemicals INa was determined in the solution containing (mM) 5 NaCl, 135 CsCl (or choline Cl), 1 MgCl2, 10 glucose, 10 HEPES, 1.0 CaCl2, 1.0 CoCl (pH adjusted to 7.3 with CsOH). The pipette solution contained (mM) 5 NaCl, 130 CsCl, 1 MgCl2, 10 HEPES, 5 Cs-EGTA, 5 MgATP, 0.1 GTP, with pH adjusted to 7.2 with CsOH. 3-(4-Chlorophenyl)1-(1,1dimethylethyl)-1h-pyrazolo[3,4-d]pyrimidin-4-amine (PP2, Tocris, Bristol, UK) was prepared as 10 mM stock solution in dimethylsulfoxide (DMSO, Sigma-Aldrich, St Louis, MO). All other reagents were obtained from Sigma-Aldrich. Epidermal growth factor (EGF) was reconstituted using 10 mM acetic acid containing 0.1% BSA to 20 μg/ml stock solution. Tyrphostin AG556 (AG556) and AG1295 were prepared as 100 mM stock solution in DMSO. The stocks were divided into aliquots and stored at − 20 °C. Cs-orthovanadate (VO43− ) stock solution (0.5 mM) was made with distilled water, and pH of VO43− working solution was adjusted to 7.3–7.4 with HCl. 2.3. Electrophysiology INa was recorded as described previously [18–20] using the whole-cell patch clamp technique with an EPC-9 amplifier and Pulse software (HEKA Electronik, Lambrecht, Germany). Borosilicate glass (1.2-mm OD) pipettes were prepared with a Brown-Flaming puller (model P-97, Sutter Instrument Co, Novato, CA) to produce a tip with resistance of 0.6–0.9 MΩ when filled with the pipette solution. After a giga-ohm seal was obtained, the cell membrane was ruptured by gentle suction to establish the whole-cell configuration. Series resistance (Rs) and capacitance were electronically compensated. Care was taken to ensure that the voltage drop across Rs was < 5 mV. Data were collected after 20 min of rupture membrane. When experiments showed any change of Rs or inadequate voltage control, data were discarded. All the experiments were performed under
761
similar conditions (e.g. similar patch duration for each protocol) at room temperature (21–22 °C). 2.4. Immunoprecipitation and Western blot Ventricular cells were treated with EGF, VO43−, AG556, or AG556 plus VO43− for 20 min at room temperature, and centrifuged at 4 °C. The cell pellet was then lysed with a lysis buffer containing (mM) 25 Tris, 150 NaCl, 100 NaF, 1.0 EDTA, 1.0 VO43−, 1.0 phenylmethylsulfonyl fluoride, and 1% sodium deoxycholate, 0.1% SDS, 1% Triton X-100, 1 μg/ml leupeptin, and 1 μg/ml aprotenin. Protein quantification of lysates was made using a protein assay reader (Bio-Rad Laboratories, Hercules, CA), and diluted to equal concentrations. Proteins were immunoprecipitated overnight at 4 °C using 2 μg of antisodium channel antibody (III–IV loop, Upstate Biotechnology, Lake Placid, NY) and 50 μl of protein A/G beads (Upstate). Immunoprecipitated proteins bound to pelleted protein A/G beads were washed thoroughly with PBS solution, denatured in Laemmli sample buffer, separated using SDS-PAGE, and electroblotted onto nitrocellulose membranes. The immunoblots were probed with anti-phosphotyrosine antibody (1:1000, Upstate) overnight at 4 °C in a blocking solution containing 5% nonfat dry milk in TBS with 0.1% Tween 20, and subsequently treated with goat anti-mouse IgG-HRP antibody (1:5000, Santa Cruz Biotechnology, Santa Cruz, CA) for 1 h at room temperature. Blots were developed with enhanced chemiluminescence (ECL, Amersham Biosciences) and exposed on X-ray film (Fuji Photo Film GmbH, Düsseldorf, Germany). The blots were then stripped and reprobed with the anti-sodium channel antibody to determine total sodium channel protein levels. The film was scanned, imaged by a Bio-Imaging System (Syngene, Cambridge, UK), and analyzed via GeneTools software (Syngene). 2.5. Statistical analysis Nonlinear curve-fitting was performed using Pulsefit (HEKA) and Sigmaplot (SPSS, Chicago, IL). Paired and/or unpaired Student's two-tailed t-test were used as appropriate to evaluate the statistical significance of the differences between two group means, and ANOVA was used for multiple groups. Group data are expressed as means ± SE. Values of P <0.05 were considered statistically significant. 3. Results 3.1. Effect of EGF on INa Fig. 1A shows the time course of INa recorded in a representative cell with a 30-ms voltage step from − 130 to − 35 mV (inset) in the absence and presence of 100 ng/ml EGF. EGF gradually increased INa, accelerated activation and inactivation, and caused a cross over of current traces (Fig. 1B). The current traces were fitted to mono-exponential function with activation and inactivation time constants (τm and τh) in the absence and presence of 100 ng/ml EGF (Fig. 1B). The τm and
762
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
Fig. 1. Effect of EGF on INa. (A) Time course of INa recorded in a representative cell with a 30-ms step to − 35 from −130 mV (inset) in the absence and presence of 100 ng/ml EGF. (B) Raw data of INa (from panel A) were fitted to monoexponential function (superimposed with raw data) with activation and inactivation time constants (τm and τh) shown in the absence (control) and presence of 100 ng/ml EGF. EGF increased INa, accelerated activation and inactivation, and induced a cross over of current traces. (C) Voltage-dependent INa recorded in a representative cell with 30-ms voltage steps from − 130 mV to between − 70 and +5 (5-mV increment, 0.2 Hz) as shown in the inset in the absence (control) and presence of 100 ng/ml EGF (10 min). (D) Voltagedependence of inactivation (I/Imax, determined with the protocol shown in the inset) and activation (g/gmax) of INa were determined with voltage-dependent INa in the absence and presence of 100 ng/ml EGF. Curves were fitted to Boltzmann distribution: y = 1/{1 + exp[(Vm − V0.5)/S]}, where Vm is membrane potential, V0.5 is the midpoint of voltage, and S is the slope factor. V0.5 of INa inactivation: − 85.5 ± 1.3 mV in control, and − 87.1 ± 1.4 mV in EGF (n = 6, P = NS), and S: 5.3 ± 0.1 mV for control, and 5.1 ± 0.1 mV for EGF (P = NS vs. control). V0.5 of INa activation: −43.3 ± 1.9 mV for control, and − 44.6 ± 1.7 mV for EGF (n = 6, P = NS), and S: − 5.7 ± 0.2 mV for control; − 6.0 ± 0.3 mV for EGF (P = NS). (E) Recovery of INa from inactivation was determined with the protocol shown in the inset. The recovery curves were fitted to mono-exponential functions. No change in time constant was observed with 100 ng/ml EGF.
τh of INa were 0.55 ± 0.04 and 2.13 ± 0.17 ms during control, and 0.44 ± 0.04 and 1.76 ± 0.15 ms after EGF application (n = 6, P < 0.01 vs. control). Fig. 1C displays the voltage-dependent INa increased by EGF in a typical experiment. The voltage dependence of inactivation (availability, I/Imax) of INa was determined with the protocol as shown in the inset (Fig. 1D), and inactivation curves were fitted to Boltzmann distribution to obtain voltage for half-inactivation (V0.5) and slope factor (S). The voltage dependence of conductance activation variable (g/gmax) of INa was determined from INa–V relationships measured from voltage-dependent INa (Fig. 1C)
for each cell as described previously [18], and also fitted to Boltzmann distribution. No change of voltage-dependence of INa was observed with EGF (Fig. 1D), and V0.5 of INa inactivation was − 85.5 ± 1.3 mV in control, and − 87.1 ± 1.4 mV in EGF (n = 6, P = NS). S was 5.3 ± 0.1 mV for control, and 5.1 ± 0.1 mV for EGF (P = NS vs. control). The V0.5 of INa activation was − 43.3 ± 1.9 mV for control, and − 44.6 ± 1.7 mV for EGF (n = 6, P = NS), and the S was − 5.7 ± 0.2 mV for control; −6.0 ± 0.3 mV for EGF (P = NS). Recovery of INa from inactivation was determined using a paired pulse protocol as shown in the inset (Fig. 1E), and recovery curves were fitted to mono-exponential functions. The recovery time constant (4.74 ± 0.33 ms in control) was not altered by 100 ng/ml EGF (5.47 ± 0.03 ms in EGF, n = 6, P = NS). The above results suggest that the enhancement of cardiac INa by EGF is probably mediated by the increased tyrosine phosphorylation of sodium channel by EGFR kinase. If this is true, the effect should be antagonized by an EGFR kinase inhibitor or potentiated by a protein tyrosine phosphatase (PTP) inhibitor. To test this, we firstly employed EGF, and then coapplied EGF and the highly selective EGFR kinase inhibitor AG556 or the PTP inhibitor VO43−. As Fig. 2 is shown, AG556 antagonizes the increase of INa by EGF, while VO43− potentiates the effect (Figs. 2A and B). EGF at 100 ng/ml enhanced INa (at − 35 mV, − 2.1 ± 0.3 nA) to 115.8 ± 3.2% (− 2.3 ± 0.4 nA) of control (n = 7, P < 0.05), and the effect was antagonized to 96.9 ± 1.7% (− 2.0 ± 0.3 nA) by combination of EGF with 5 μM AG556 (Fig. 2C), while co-application of EGF and 1 mM VO43− potentiated INa (− 2.0 ± 0.3 nA) to 126.7 ± 3.1% (− 2.6 ± 0.4 nA) of control (n = 6, P < 0.05 vs. EGF: − 2.2 ± 0.4 nA) (Fig. 2D). These results indicate that cardiac INa is likely augmented by an increase of tyrosine phosphorylation of cardiac sodium channels by activating EGFR kinase or by inhibiting PTPs, and decreased by a reduction of tyrosine phosphorylation of the channels by inhibiting EGFR kinase. 3.2. Effects of VO43− on INa To further confirm the effect of inhibiting PTPs on INa, VO43− was directly applied in a different set of experiments. Fig. 3A illustrates the time course of INa recorded in a representative cell in the absence and presence of VO43−. VO43− at 1 mM gradually increased INa. The effect reached a steady-state level within 3 min, and recovered on washout. VO43−, as EGF, accelerated activation and inactivation of INa, and induced a cross over of the current. The raw data of activation (m) and inactivation (h) were also fitted to a mono-exponential function. The τm and τh are shown in Fig. 3B in the absence (control) and presence of 1 mM VO43−. Mean values of τm and τh at − 35 mV were 0.53 ± 0.04 and 2.08 ± 0.16 ms in control and 0.41 ± 0.03 and 1.65 ± 0.12 ms in VO43− (n = 7, P < 0.01 vs. control). Voltage-dependent INa (Fig. 3C) was substantially enhanced by 1 mM VO43−, and the effect was reversed by washout. VO43− at 0.3, 1, and 3 mM increased INa (at − 35 mV, from − 2.2 ± 0.3 nA of control to − 2.3 ± 0.3, − 2.7 ± 0.4, and − 2.9 ± 0.4 nA, respectively, n = 7, P < 0.05 or P < 0.01 vs. control) in a concentration-dependent manner.
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
763
was employed. Fig. 4A illustrates the time-course of INa recorded in the absence and presence of 5 μM AG556 in a typical experiment. INa was gradually decreased to a steadystate level by AG556, and partially recovered upon washout. Original INa traces at corresponding time points are shown in right of the panel of Fig. 4A. With fitting activation and inactivation of INa, the τh (at − 35 mV) was found to be slightly increased by 5 μM AG556 (1.8 ± 0.1 ms in control, 2.1 ± 0.1 ms in AG556, n = 7, P < 0.05), while the τm was not affected (0.54 ± 0.1 ms in control, 0.53 ± 0.1 ms in AG556, P = NS). Fig. 4B displays voltage-dependent INa traces recorded in a representative cell with the protocol as shown in the inset of Fig. 1C. Voltage-dependent INa was substantially inhibited by 5 μM AG556, and the effect was partially reversed by washout for 10 min. INa was − 2.5 ± 2.0 nA (at − 35 mV) for control, − 1.4 ± 0.1 nA for 5 μM AG556 (n = 10, P < 0.01 vs. control), and −2.2 ± 0.2 nA after washout (P < 0.01 vs. AG556).
Fig. 2. EGF on INa and AG556 or VO3− 4 effect. (A) Time course of INa recorded in a representative cell with a 30-ms step to −35 from −130 mV (inset) during control, in the presence of 100 ng/ml EGF, and co-application of EGF and 5 μM AG556. Original currents at corresponding time points were shown in the right inset of the panel. (B) Time course of INa recorded in a representative cell during control, in the presence of 100 ng/ml EGF, and co-application of EGF and 1 mM VO3− 4 . Original currents at corresponding time points were shown in the right inset of the panel. (C) Histogram showing percentage values of INa during control, after application of 100 ng/ml EGF, and combination of EGF with 5 μM AG556 (n = 7, **P < 0.01 vs. control, ##P < 0.01 vs. EGF). (D) Mean percentage values of INa in control, after EGF, and EGF plus 1 mM VO3− 4 (n = 6, **<0.01 vs. control, ##P < 0.01 vs. EGF).
The effects of VO43− on voltage-dependent activation and inactivation, and recovery of INa were examined with the same procedure as described in Fig. 1D. No significant changes in the kinetics were observed with 1 mM VO43− (Figs. 3D and E). V0.5s of INa activation and inactivation were −45.5 ± 1.8 and −88.6 ± 0.8 mV in control, and −47.3 ± 3.0 mV and −90.5 ± 1.3 mV with VO43− (n= 8, P = NS). INa recovery time constant was 4.11 ± 0.45 ms in control, and 4.12± 0.27 ms with 1 mM VO43− (n= 8, P = NS). These results suggest that the decrease of tyrosine dephosphorylation of cardiac sodium channel by inhibiting PTPs with VO43−, as activating EGFR kinase with EGF, enhances current amplitude, and accelerates activation and inactivation of cardiac INa without affecting voltage-dependent and recovery kinetics. 3.3. Effects of AG556 on INa To verify the effect of directly inhibiting EGFR kinase on cardiac INa, the highly selective EGFR kinase inhibitor AG556
Fig. 3. Enhancement of INa by VO3− 4 . (A) Time course of INa recorded in a representative cell with a 30-ms step from − 130 mV to − 35 mV in the absence and presence of 1 mM VO3− 4 (OV). (B) Raw data of INa were fitted to monoexponential function with activation and inactivation time constants (τm and τh) 3− shown in the absence (control) and presence of 1 mM VO3− 4 (OV). VO4 , as EGF, enhanced INa, accelerated activation and inactivation and induced a cross over of the current. (C) Voltage-dependent INa recorded in a representative cell using the same protocol as shown in the inset of Fig. 1C in the absence and presence of 1 mM VO3− 4 . (D) Mean values of voltage-dependence of inactivation (I/Imax) and activation (g/gmax) of INa were determined in the absence and presence of 1 mM VO3− 4 . Curves were fitted to Boltzmann distribution. Voltagedependent inactivation and activation were not affected by VO3− 4 . (E) Recovery curves of INa were fitted to mono-exponential functions. No change in time constant was observed with 1 mM VO3− 4 .
764
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
conductance was not significantly altered by AG556 (− 46.8 ± 0.6 mV for control; − 48.5 ± 0.7 mV for AG556, n = 6, P = NS). Recovery of INa from inactivation was well fitted to monoexponential functions with time constant of 5.26 ± 0.14 ms (n = 7) in control (Fig. 4D). After administration of 5 μM AG556, recovery of INa was slowed, and the curve was only fitted to bi-exponential functions with rapid and slow time constants of 2.8 ± 0.4 and 219.3 ± 71 ms. The effect was reversed by washout (τ: 6.16 ± 0.28 ms, P = NS vs. control). These results suggest that the inhibition of tyrosine phosphorylation of cardiac sodium channels with the EGFR kinase inhibitor AG556 reduces INa, negatively shifts inactivation potential, and slows recovery time course of INa. 3.4. Influence of VO43− on AG556 effects
Fig. 4. Inhibition of INa by AG556. (A) Time course of INa recorded in a typical experiment with a 30-ms voltage step to − 35 from − 130 mV (inset) in the absence and presence of 5 μM AG556. Original current traces are shown in the right of the panel. (B) Voltage-dependent INa traces recorded in a representative myocyte with the same protocol as shown in the inset of Fig. 1C during control, in the presence of 5 μM AG556, and washout for 10 min. (C) Mean values of voltage-dependence of inactivation (I/Imax) and activation (g/gmax) of INa in the absence and presence of 5 μM AG556. Curves were fitted to Boltzmann distribution. Voltage-dependence of inactivation was negatively shifted by AG556. (D) Mean values of curves for recovery of INa from inactivation in the absence and presence of 5 μM AG556. Recovery curves were fitted to monoexponential functions during control, after drug washout (for 15 min); however, the curve was fitted to bi-exponential functions in the presence of 5 μM AG556.
The concentration–response relationship for the inhibition of INa (at − 35 mV) by AG556 (0.1–30 μM) was evaluated, and fitted to the Hill equation: E = Emax/[1 + (IC50/C)b], where E is the percentage for inhibiting INa at concentration C, Emax is the maximum inhibition, IC50 is the concentration for halfmaximum action, and b is the Hill coefficient. The IC50 of AG556 for inhibiting INa was 7.6 μM (close to that for EGFR tyrosine kinase inhibition in other system [21,22]), and Emax was 61.4%. Fig. 4C illustrates the curves for voltage dependence of inactivation and activation conductance of INa evaluated with the procedure as described in Fig. 1D. AG556 negatively shifted the V0.5 of inactivation (availability) by 7.8 mV, from − 89.9 ± 1.5 mV in control to − 97.7 ± 1.1 mV in AG556, (n = 7, P < 0.01), and the effect was partially reversed by 3.2 mV (to − 94.5 ± 1.6 mV) by washout. However, the V0.5 of activation
To study whether the AG556 effects on INa would be antagonized by the PTP inhibitor VO43−, we pretreated cells with VO43− and then AG556 was co-applied. Fig. 5A displays the time course of INa recorded in a typical experiment at − 35 mV. A small reduction of INa was observed with 5 μM AG556 in the presence of 1 mM VO43−; however, an additional inhibition was noted as VO43− was removed. Original current traces at corresponding time points are shown in right of the panel. Fig. 5B summarizes the group data for the percent changes of INa with AG556, and VO43− plus AG556. AG556 (5 μM) alone inhibited INa by 33.4 ± 2.2% (from − 2.3 ± 0.4 nA of control to − 1.5 ± 0.2 nA, n = 16, P < 0.01), and the effect was reduced to 8.8 ± 2.3% (− 2.2 ± 0.3 nA, n = 6, P < 0.01 vs. AG556 alone) in cells pretreated with 1 mM VO43− (−2.4 ± 0.3 nA). These results indicate that reduction of cardiac INa by AG556 is substantially antagonized by the PTP inhibitor VO43−. In addition, pretreatment with 1 mM VO43− substantially antagonized the negative shift of inactivation potential of INa by AG556 (Fig. 5C). The inactivation V0.5 (− 86.4 ± 1.2 mV) was not significantly changed by 5 μM AG556 (− 89.7 ± 1.0 mV, n = 6, P = NS) in cells pretreated with 1 mM VO43−. Moreover, the slowed recovery of INa from inactivation by AG556 (Fig. 4D) was also antagonized by the pretreatment with 1 mM VO43− (Fig. 5D). The recovery was fitted to monoexponential functions (τ: 5.2 ± 0.15 and 4.1 ± 0.16 ms before and after AG556, n = 6, P = NS). These results further suggest that tyrosine phosphorylation by EGFR kinase contributes to the modulation of cardiac INa via the interaction of EGFR kinase and PTPs in guinea pig ventricular myocytes. 3.5. Other PTK inhibitors on INa Tyrphostin AG1295, a selective inhibitor of platelet-derived growth factor receptor (PDGFR) kinase, was employed to study the effect of inhibiting PDGFR kinase on INa. AG1295 at 100 μM (100 times higher than IC50s for inhibiting PDGFR tyrosine kinase [23]) produced no significant inhibition of INa (− 2.31 ± 0.22 nA for control, − 2.27 ± 0.23 nA for AG1295, n = 4, P = NS). We also used PP2, a membrane permeable Srcfamily PTK inhibitor, to examine the effect of inhibiting Srcrelated kinases on native cardiac INa. PP2 at 5 μM (1000 times
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
765
higher than IC50s for inhibiting Src-family tyrosine kinase [24]) did not affect INa in five cells (− 2.51 ± 0.24 nA in control, − 2.46 ± 0.23 nA in PP2, P = NS). Voltage-dependent activation and inactivation, and recovery of INa were not influenced by the application AG1295 or PP2 (data not shown). 3.6. Tyrosine phosphorylation level of cardiac sodium channels Tyrosine phosphorylation level of cardiac sodium channels was studied using immunoprecipitation and Western blotting analysis. Fig. 6A illustrates tyrosine phosphorylation level of sodium channels during control and in the presence of EGF, VO43−, AG556, and VO43− plus AG556 in a representative experiment. EGF (100 ng/ml) or VO43− (1 mM) increased tyrosine phosphorylation level of cardiac sodium channels; however, AG556 at 30 μM (a high concentration that caused a substantial reduction of tyrosine phosphorylation level of cardiac sodium channels), decreased the phosphorylation level, and the effect was significantly antagonized by pretreatment with 1 mM VO43− . Fig. 6B summarizes the quantitative tyrosine phosphorylation levels of sodium channels. The
Fig. 6. Tyrosine phosphorylation levels of cardiac sodium channels in the absence 3− (control) and presence of 100 ng/ml EGF, 1 mM VO3− 4 , 30 μM AG556, or VO4 (OV) plus AG556. (A) Upper panel: tyrosine phosphorylation levels of cardiac sodium channels obtained by immunoprecipitation and Western blot. Lower panel: total sodium channel protein levels obtained by stripping and reprobing the membrane with the anti-sodium channel antibody. (B) Histogram summarizes the mean values of the relative levels of tyrosine-phosphorylated sodium channel protein analyzed by the densitometry. The amount of protein from the immunoprecipitation (as in panel A) was relative to those from the Western blots. The relative values of treatment groups were then normalized by that of solvent control. n = 4, *P < 0.05, **P < 0.01 vs. control; #P < 0.01 vs. AG556.
tyrosine phosphorylation level was increased by EGF (23.8 ± 5.4%, n = 4, P < 0.01 vs. control) or VO43− (20.2 ± 2.1%, n = 4, P < 0.01 vs. control), decreased by AG556 (54.7 ± 7.4%, P < 0.01), and reduced by 15.3 ± 8.5% with AG556 in cells pretreated with 1 mM VO43− (P < 0.01 vs. AG556 alone). These results indicate that tyrosine phosphorylation level of sodium channel protein is increased by EGF or VO43−, and decreased by the EGFR kinase inhibitor AG556. 4. Discussion 4.1. Novel findings of the present study
Fig. 5. VO3− 4 effect on AG556. (A) Time course of INa recorded in a typical experiment with a 30-ms step to − 35 from − 130 mV (inset) under conditions of 3− the pretreatment with 1 mM VO3− 4 (OV), VO4 plus 5 μM AG556, and washout of VO3− 4 . Current traces at corresponding time points are shown in the right of the panel. (B) Histogram illustrates the mean values of INa at − 35 mV during control, in the presence of 5 μM AG556 (n = 8, **P < 0.01 vs. control), VO3− 4 plus AG556 (n = 6, **P < 0.01 vs. control; ##P < 0.01 vs. AG556). (C) Voltage-dependence of 3− inactivation (I/Imax) of INa in the presence of 1 mM VO3− 4 , and VO4 plus 5 μM AG556 (n = 6), and curves were fitted to Boltzmann distribution. (D) Recovery curves of INa were fitted to monoexponential functions. With pretreatment of 1 mM VO3− 4 , no change in time constant was observed with 5 μM AG556.
In the present study, we demonstrated that EGF enhanced voltage-gated cardiac INa in guinea pig ventricular myocytes, and accelerated activation and inactivation of the current (Fig. 1). The augmentation of INa by EGF was antagonized by the selective EGFR kinase inhibitor AG556, and potentiated by the PTPs inhibitor VO43− (Fig. 2). VO43−, like EGF, enhanced INa and accelerated both activation and inactivation of the current (Fig. 3). In addition, AG556 decreased INa, negatively shifted inactivation potential, and slowed recovery of INa from inactivation (Fig. 4). These effects were antagonized by VO43− (Fig. 5). Immunoprecipitation and Western blotting analysis demonstrated that tyrosine phosphorylation level of cardiac sodium channels was increased by EGF or VO43−, and reduced
766
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
by AG556. The decreased phosphorylation level by AG556 was significantly reversed by VO43− (Fig. 6). These results suggest that native cardiac INa is regulated by EGFR kinase and/or PTPs. 4.2. Regulation of cardiac INa by PKC and PKA The voltage-gated cardiac sodium channel of the heart was down-regulated by PKC [6,25]. However, modulation of cardiac INa by cAMP/PKA is inconsistent. Earlier reports showed that INa was inhibited by β-adrenoceptor stimulation and cAMP in rat [4,26], and guinea pig [5] cardiac myocytes, while later studies demonstrated an enhancement of INa by cAMP/PKA in rabbit ventricular myocytes [3], and cloned human cardiac sodium channels expressed in Xenopus laevis oocytes [27]. Ono and colleagues [28] demonstrated that voltage-dependent activation and inactivation of INa were negatively shifted by cAMP in guinea pig and canine cardiac myocytes, and cAMP inclusion in pipette solution could result in either increase or decrease of cardiac INa depending on the selected holding and test potentials. The complex effects of PKA on cardiac sodium channel function suggest that cAMPdependent PKA may phosphorylate multiple serine/threonine sites that cause different functional effects [29]. 4.3. Comparison with previous reports regarding INa regulation by PTKs It is well documented that PTKs regulate the long term cellular processes (e.g. cell growth, differentiation, oncogenesis, etc.) [1,9], and also participate in the modulation of cellular plasma membrane ion channels [11], including Ca2+ channels and several types of K+ channels [11–14], as well as volumesensitive Cl− channels [15] in different types of cells. The regulation of ion channels by PTKs are mostly based on the effects of soluble PTK inhibitors that are relatively specific for certain kinase families (e.g., AG556, PP1, and PP2), and supportive evidence would be from an opposite effect and/or antagonizing effect (against PTK inhibitors) resulting from inhibition of PTPs [11,14,15]. However, limited information is available in the literature regarding the regulation of cardiac INa by PTKs. An earlier report described that cardiac INa was inhibited by several PTK inhibitors (including the broad spectrum PTK-inhibitor genistein, the EGFR kinase inhibitor AG957, and Src-related PTK inhibitor PP2) in rabbit ventricular cells [16]. This study suggested a clue of possible modulation of cardiac INa by PTKs, the concentrations of PTK inhibitors were much higher than those of inhibiting PTKs in other systems [16]; however, PTP inhibitor was not used to confirm the effects observed with different PTK inhibitors, and the possible PTK-independent effects [11] on cardiac INa would not be excluded. In the present study, we did not find that PP2 at 5 μM (1000 times higher than IC50s for inhibiting Src-family tyrosine kinase [24]) had any effect on INa in guinea pig ventricular myocytes. Nevertheless, we found that EGF and the PTP inhibitor VO43− (Figs. 1 and 3) increased cardiac INa, and accelerated time-dependent activation
and inactivation without affecting voltage-dependence and recovery of the current. VO43− antagonized the current reduction by the EGFR kinase inhibitor AG556 (Fig. 5). The simplest explanation for the actions of EGF, VO43−, AG556, and their interaction is that phosphorylation and/or dephosphorylation of a critical tyrosine residue on sodium channel or a signaling molecule directly or indirectly augments or decreases sodium channel activity. In this model (Fig. 7), activation of EGFR tyrosine kinase by EGF increases sodium channel activity by enhancing the tyrosine phosphorylation of the channel, whereas inhibition of EGFR kinase by AG556 decreases the channel activity by reducing the tyrosine phosphorylation. In addition, inhibition of PTP (with VO43−) increases INa by allowing its unopposed phosphorylation. This may also explains why pretreatment with VO43− precluded the effect of AG556. Once tyrosine becomes phosphorylated during VO43− pretreatment, blocking EGFR kinase will have little effect because there is no efficient means of dephosphorylating the residue. This notion is supported by the results from immunoprecipitation and Western blotting analysis, which provide a direct evidence that tyrosine phosphorylation level of cardiac sodium channels is enhanced by EGF or VO43−, and decreased by AG556 (Fig. 6). These results are consistent with the reports that the coordinated action of PTKs and PTPs is required for protein tyrosine phosphorylation [30,31]. In the heart, Malhotra and colleagues demonstrated with immunohistochemical technique that tyrosine-phosphorylated and nonphosphorylated sodium channel β1-subunits were differentially localized in cardiac myocytes [32]. A recent report showed that human NaV1.5 channel was modulated by the co-transfection of PTPH1 in HEK 293 cells [33]. In the neuron, Ratcliffe and colleagues reported that voltage-gated sodium channels were associated with receptor protein tyrosine phosphatase beta [34]. Moreover, it was reported that L-type Ca2+ channels in GH3
Fig. 7. Schematic model for regulation of cardiac INa channel by EGFR tyrosine kinase and PTPs. Tyrosine phosphorylation sites on cardiac sodium channel or signaling molecules are postulated. The tyrosine of the sodium channel is phosphorylated by EGFR kinase and dephosphorylated by PTPs. Phosphorylation of tyrosine by one or more steps (dashed line) favors INa channel opening, whereas dephosphorylation favors closure. AG556 inhibited EGFR tyrosine kinase reduced the phosphorylation, while VO3− 4 inhibited PTPs and decreased dephosphorylation of the channel. In addition, EGFR kinase and PTPs may also affect gene expression in cardiac cells.
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
767
cells and in guinea pig ventricular myocytes was also regulated by the interplay of PTKs and PTPs [15,35]. Nevertheless, the opposite response of INa to PTPs and PTKs were observed in other systems. In neuronal cells, the inhibition of protein tyrosine phosphatase-β associated brain neuronal sodium channel with pervanadate suppressed INa [34]. In PC12 cell line, several receptor tyrosine kinase activators (e.g. EGF) produced an inhibition of INa [36]. Whether these diverse results are related to the subtle variation of sodium channels, different species, tissue/cell types, and/or experimental conditions remain to be studied.
autoimmune myocarditis [38]. These studies suggest that suppression of EGFR kinase is likely cardioprotective; however, whether this effect is related to the INa inhibition remains to be studied. In summary, the present study provides the novel information that cardiac native INa is regulated by EGFR kinase and PTPs. EGFR kinase positively, while PTPs negatively modulates cardiac INa. Electrical excitability of the heart would be controlled by the interplay of EGFR kinase and PTPs.
4.4. Limitations and potential significance of the present study
The study was supported by Sun Chieh Yeh Heart Foundation of Hong Kong. The authors thank Professor TakMing Wong in the Department of Physiology for his substantial support.
One of limitations was that cardiac INa was recorded at room temperature (to ensure the well-controlled voltage clamp). This would underestimate the effects of the enzymes (i.e. kinases and phosphatrases) on the regulation of the channel. In addition, the present study reported the regulation of voltage-gated cardiac sodium current by EGFR kinase in guinea pig ventricular myocytes; however, molecular mechanisms were not clarified. For instance, it is unknown how the increased phosphorylation would enhance the activation and inactivation of cardiac sodium channel, which remains to be studied with a computer modeling of the sodium channel in the future. Another limitation was that we could not well explain why our results from native cardiac myocytes were not consistent with those observed in cloned NaV1.5 channel expressed in HEK 293 cells, in which Fyn, a Src-family tyrosine kinase, regulated cloned cardiac NaV1.5 channels by positively shifting inactivation potential without affecting the current amplitude [17], and co-expression of NaV1.5 with PTPH1 negatively shifted inactivation potential [33]. One of possibilities for the discrepancy is likely related to the various responses of sodium channel to different kinases in native cardiac cells (EGFR kinase) and expressed HEK 293 cells (Src-family kinases). An opposite regulative effect of ICl.vol by EGFR kinase and Src-family kinases was recently reported in human atrial myocytes [15]. In addition, the different response may also be related to that the native cardiac cells have intact sodium channel with both α- and β-subunits in the present study, while HEK 293 cells expressed only cloned NaV1.5 αsubunit was used in the recent reports [17,33]. However, these possibilities should be investigated in detail by making comparative study in native cardiac cells and HEK 293 cells expressing both NaV1.5 and sodium channel β-subunits in the future. The results from the present observation demonstrated the strong evidence that cardiac INa in native guinea pig ventricular myocytes was modulated by the coordination of EGFR kinase and PTPs. Activation of EGFR kinase increased, while activation of PTPs decreased cardiac INa channels, suggesting that EGFR kinase and PTPs could regulate cardiac electrical excitability. It is interesting to note that the inhibition of EGFR kinase by AG556 reduced ischemic zones, improved heart performance in experimental myocardial ischemia of a rat model [37], and ameliorated progression of experimental
Acknowledgments
References [1] Levitan IB. Modulation of ion channels by protein phosphorylation and dephosphorylation. Annu Rev Physiol 1994;56:193–212. [2] Marx S. Ion channel macromolecular complexes in the heart. J Mol Cell Cardiol 2003;35:37–44. [3] Matsuda JJ, Lee H, Shibata EF. Enhancement of rabbit cardiac sodium channels by beta-adrenergic stimulation. Circ Res 1992;70:199–207. [4] Schubert B, Vandongen AM, Kirsch GE, Brown AM. Inhibition of cardiac Na+ currents by isoproterenol. Am J Physiol 1990;258:H977–82. [5] Ono K, Kiyosue T, Arita M. Isoproterenol, DBcAMP, and forskolin inhibit cardiac sodium current. Am J Physiol 1989;256:C1131–7. [6] Qu Y, Rogers J, Tanada T, Scheuer T, Catterall WA. Modulation of cardiac Na+ channels expressed in a mammalian cell line and in ventricular myocytes by protein kinase C. Proc Natl Acad Sci U S A 1994;91: 3289–93. [7] Lu T, Lee HC, Kabat JA, Shibata EF. Modulation of rat cardiac sodium channel by the stimulatory G protein alpha subunit. J Physiol 1999;518 (Pt 2):371–84. [8] Ahmmed GU, Xu Y, Hong DP, Zhang Z, Eiserich J, Chiamvimonvat N. Nitric oxide modulates cardiac Na(+) channel via protein kinase A and protein kinase G. Circ Res 2001;89:1005–13. [9] Hunter T. Signaling—2000 and beyond. Cell 2000;100:113–27. [10] Siegelbaum SA. Channel regulation. Ion channel control by tyrosine phosphorylation. Curr Biol 1994;4:242–5. [11] Davis MJ, Wu X, Nurkiewicz TR, Kawasaki J, Gui P, Hill MA, et al. Regulation of ion channels by protein tyrosine phosphorylation. Am J Physiol: Heart Circ Physiol 2001;281:H1835–62. [12] Gao Z, Lau CP, Wong TM, Li GR. Protein tyrosine kinase-dependent modulation of voltage-dependent potassium channels by genistein in rat cardiac ventricular myocytes. Cell Signal 2004;16:333–41. [13] Tiran Z, Peretz A, Attali B, Elson A. Phosphorylation-dependent regulation of Kv2.1 channel activity at tyrosine 124 by Src and by protein–tyrosine phosphatase epsilon. J Biol Chem 2003;278:17509–14. [14] Ogura T, Shuba LM, McDonald TF. L-type Ca2+ current in guinea pig ventricular myocytes treated with modulators of tyrosine phosphorylation. Am J Physiol 1999;276:H1724–33. [15] Du XL, Gao Z, Lau CP, Chiu SW, Tse HF, Baumgarten CM, et al. Differential effects of tyrosine kinase inhibitors on volume-sensitive chloride current in human atrial myocytes: evidence for dual regulation by Src and EGFR kinases. J Gen Physiol 2004;123:427–39. [16] Wang Y, Wagner MB, Kumar R, Cheng J, Joyner RW. Inhibition of fast sodium current in rabbit ventricular myocytes by protein tyrosine kinase inhibitors. Pflugers Arch 2003;446:485–91. [17] Ahern CA, Zhang JF, Wookalis MJ, Horn R. Modulation of the cardiac sodium channel NaV1.5 by Fyn, a Src family tyrosine kinase. Circ Res 2005;96:991–8.
768
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
[18] Li GR, Lau CP, Shrier A. Heterogeneity of sodium current in atrial vs. epicardial ventricular myocytes of adult guinea pig hearts. J Mol Cell Cardiol 2002;34:1185–94. [19] Pu J, Balser JR, Boyden PA. Lidocaine action on Na+ currents in ventricular myocytes from the epicardial border zone of the infarcted heart. Circ Res 1998;83:431–40. [20] Hanck DA, Sheets MF. Time-dependent changes in kinetics of Na+ current in single canine cardiac Purkinje cells. Am J Physiol 1992;262:H1197–207. [21] Gazit A, Osherov N, Posner I, Yaish P, Poradosu E, Gilon C, et al. Tyrphostins. 2. Heterocyclic and alpha-substituted benzylidenemalononitrile tyrphostins as potent inhibitors of EGF receptor and ErbB2/neu tyrosine kinases. J Med Chem 1991;34:1896–907. [22] Levitzki A, Gazit A, Osherov N, Posner I, Gilon C. Inhibition of protein– tyrosine kinases by tyrphostins. Methods Enzymol 1991;201:347–61. [23] Kovalenko M, Gazit A, Bohmer A, Rorsman C, Ronnstrand L, Heldin CH, et al. Selective platelet-derived growth factor receptor kinase blockers reverse sis-transformation. Cancer Res 1994;54:6106–14. [24] Hanke JH, Gardner JP, Dow RL, Changelian PS, Brissette WH, Weringer EJ, et al. Discovery of a novel, potent, and Src familyselective tyrosine kinase inhibitor. Study of Lck- and FynT-dependent T cell activation. J Biol Chem 1996;271:695–701. [25] Xiao GQ, Qu Y, Sun ZQ, Mochly-Rosen D, Boutjdir M. Evidence for functional role of epsilonPKC isozyme in the regulation of cardiac Na(+) channels. Am J Physiol: Cell Physiol 2001;281:C1477–86. [26] Herzig JW, Kohlhardt M. Na+ channel blockade by cyclic AMP and other 6-aminopurines in neonatal rat heart. J Membr Biol 1991;119:163–70. [27] Frohnwieser B, Chen LQ, Schreibmayer W, Kallen RG. Modulation of the human cardiac sodium channel alpha-subunit by cAMP-dependent protein kinase and the responsible sequence domain. J Physiol (London) 1997;498:309–18. [28] Ono K, Fozzard HA, Hanck DA. Mechanism of cAMP-dependent modulation of cardiac sodium channel current kinetics. Circ Res 1993; 72:807–15.
[29] Murphy BJ, Rogers J, Perdichizzi AP, Colvin AA, Catterall WA. cAMPdependent phosphorylation of two sites in the alpha subunit of the cardiac sodium channel. J Biol Chem 1996;271:28837–43. [30] Sun H, Tonks NK. The coordinated action of protein tyrosine phosphatases and kinases in cell signaling. Trends Biochem Sci 1994;19:480–5. [31] Ostman A, Bohmer FD. Regulation of receptor tyrosine kinase signaling by protein tyrosine phosphatases. Trends Cell Biol 2001;11:258–66. [32] Malhotra JD, Thyagarajan V, Chen C, Isom LL. Tyrosine-phosphorylated and nonphosphorylated sodium channel beta1 subunits are differentially localized in cardiac myocytes. J Biol Chem 2004;279: 40748–54. [33] Jespersen T, Gavillet B, van Bemmelen MX, Cordonier S, Thomas MA, Staub O. Cardiac sodium channel Na(v)1.5 interacts with and is regulated by the protein tyrosine phosphatase PTPH1. Biochem Biophys Res Commun 2006;348:1455–62. [34] Ratcliffe CF, Qu Y, McCormick KA, Tibbs VC, Dixon JE, Scheuer T, et al. A sodium channel signaling complex: modulation by associated receptor protein tyrosine phosphatase beta. Nat Neurosci 2000; 3:437–44. [35] Cataldi M, Taglialatela M, Guerriero S, Amoroso S, Lombardi G, di Renzo G, et al. Protein-tyrosine kinases activate while protein–tyrosine phosphatases inhibit L-type calcium channel activity in pituitary GH3 cells. J Biol Chem 1996;271:9441–6. [36] Hilborn MD, Vaillancourt RR, Rane SG. Growth factor receptor tyrosine kinases acutely regulate neuronal sodium channels through the src signaling pathway. J Neurosci 1998;18:590–600. [37] George J, Biner S, Keren P, Barshack I, Goldberg I, Sherez J, et al. Tyrphostin AG-556 reduces myocardial infarct size and improves cardiac performance in the rat. Exp Mol Pathol 2003;74:314–8. [38] George J, Barshack I, Goldberg I, Keren P, Gazit A, Levitzki A, et al. The effect of early and late treatment with the tyrphostin AG-556 on the progression of experimental autoimmune myocarditis. Exp Mol Pathol 2004;76:234–41.
Original article
Regulation of voltage-gated cardiac sodium current by epidermal growth factor receptor kinase in guinea pig ventricular myocytes Hui Liu a , Hai-Ying Sun a , Chu-Pak Lau a,c , Gui-Rong Li a,b,c,⁎ a b c
Department of Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China Department of Physiology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China Research Centre of Heart, Brain, Hormone and Healthy Aging, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong SAR, China Received 30 July 2006; received in revised form 15 October 2006; accepted 23 October 2006 Available online 22 December 2006
Abstract Voltage-gated cardiac fast sodium channel current (INa) plays a critical role in the initiation and propagation of the myocardial action potential, and regulation of cardiac INa by protein tyrosine kinases (PTKs) is not well documented, though it is known that ion channels are among the targets of PTKs. The present study was therefore designed to investigate whether/how cardiac INa was modulated by PTKs in guinea pig ventricular myocytes using whole-cell patch clamp and immunoprecipitation and Western blotting approaches. It was found that cardiac INa was enhanced by epidermal growth factor (EGF), and the effect was antagonized by the selective epidermal growth factor receptor (EGFR) kinase inhibitor tyrphostin AG556 while potentiated by orthovanadate (a protein tyrosine phosphatase (PTP) inhibitor). In addition, AG556 inhibited, while orthovanadate increased INa, and the inhibition of INa by AG556 was antagonized by orthovanadate. Immunoprecipitation and Western blotting analysis demonstrated that tyrosine phosphorylation level of cardiac sodium channels was enhanced by EGF or orthovanadate, and reduced by AG556. The AG556-induced reduction of phosphorylation level was significantly reversed by orthovanadate. Our results demonstrate the novel information that EGFR kinase enhances, and PTPs reduce native cardiac INa in guinea pig ventricular myocytes. © 2006 Elsevier Inc. All rights reserved. Keywords: Protein tyrosine kinases; Epidermal growth factor receptor; Protein tyrosine phosphatases; Voltage-gated sodium current; Ion channels; Guinea pig ventricular myocytes; Heart
1. Introduction Voltage-gated cardiac fast sodium channel current (INa) plays a critical role in the initiation and propagation of myocardial action potential, and is a primary target for the development of antiarrhythmic agents. It is generally believed that cardiac INa, like other types of ion channel currents, is regulated by protein phosphorylation [1,2]. It has been demonstrated that cardiac INa was modulated by the serine/threonine protein kinases, protein kinase A (PKA) and/or protein kinase C (PKC) [3–8].
⁎ Corresponding author. Department of Medicine, L8-01, Laboratory Block, FMB, The University of Hong Kong, 21 Sassoon Road, Pokfulam, Hong Kong SAR China. Tel.: +86 852 2819 9513; fax: +86 852 2855 9730. E-mail address: [email protected] (G.-R. Li). 0022-2828/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.yjmcc.2006.10.013
Protein tyrosine kinases (PTKs), including receptor PTKs (e.g. EGFR kinase, epidermal growth factor receptor kinase) and nonreceptor PTKs (e.g. Src-related kinases), are associated with long term cellular processes such as cell growth, differentiation, and oncogenesis [1,9]. Recent studies showed that protein phosphorylation of tyrosine residues modulated ion channels [1,10,11], including Ca2+, and several types of K+ channels [11–14], as well as volume-sensitive Cl− channels [15] in different types of cells. However, the regulation of cardiac INa by PTKs is not fully understood. Although several PTK inhibitors were recently shown to inhibit INa in rabbit ventricular myocytes [16], and a very recent study showed that cloned rat cardiac sodium channel NaV1.5 was modulated by Fyn, a Src-family tyrosine kinase [17], it is unknown whether cardiac INa would be modulated by EGFR kinase. The present study was therefore designed to use whole-cell patch voltage
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
clamp, and immunoprecipitation and Western blotting approaches to investigate whether and how cardiac INa would be regulated by EGFR kinase in guinea pig ventricular myocytes. We found that activation of EGFR kinase increased cardiac INa in ventricular myocytes from guinea pig heart. 2. Materials and methods 2.1. Myocyte preparation Guinea pigs of either gender (250–300 g) were sacrificed by cervical dislocation after anesthetization with pentobarbital (40 mg/kg, i.p.) in accordance with the guideline of Animal Care and Use Committee for Teaching and Research of University of Hong Kong. Ventricular myocytes from guinea pig hearts were enzymatically dissociated by the procedure described previously [18] and the isolated myocytes were kept in a high-K+ storage medium [18] at room temperature for 1 h before electrophysiological recording. 2.2. Solutions and chemicals INa was determined in the solution containing (mM) 5 NaCl, 135 CsCl (or choline Cl), 1 MgCl2, 10 glucose, 10 HEPES, 1.0 CaCl2, 1.0 CoCl (pH adjusted to 7.3 with CsOH). The pipette solution contained (mM) 5 NaCl, 130 CsCl, 1 MgCl2, 10 HEPES, 5 Cs-EGTA, 5 MgATP, 0.1 GTP, with pH adjusted to 7.2 with CsOH. 3-(4-Chlorophenyl)1-(1,1dimethylethyl)-1h-pyrazolo[3,4-d]pyrimidin-4-amine (PP2, Tocris, Bristol, UK) was prepared as 10 mM stock solution in dimethylsulfoxide (DMSO, Sigma-Aldrich, St Louis, MO). All other reagents were obtained from Sigma-Aldrich. Epidermal growth factor (EGF) was reconstituted using 10 mM acetic acid containing 0.1% BSA to 20 μg/ml stock solution. Tyrphostin AG556 (AG556) and AG1295 were prepared as 100 mM stock solution in DMSO. The stocks were divided into aliquots and stored at − 20 °C. Cs-orthovanadate (VO43− ) stock solution (0.5 mM) was made with distilled water, and pH of VO43− working solution was adjusted to 7.3–7.4 with HCl. 2.3. Electrophysiology INa was recorded as described previously [18–20] using the whole-cell patch clamp technique with an EPC-9 amplifier and Pulse software (HEKA Electronik, Lambrecht, Germany). Borosilicate glass (1.2-mm OD) pipettes were prepared with a Brown-Flaming puller (model P-97, Sutter Instrument Co, Novato, CA) to produce a tip with resistance of 0.6–0.9 MΩ when filled with the pipette solution. After a giga-ohm seal was obtained, the cell membrane was ruptured by gentle suction to establish the whole-cell configuration. Series resistance (Rs) and capacitance were electronically compensated. Care was taken to ensure that the voltage drop across Rs was < 5 mV. Data were collected after 20 min of rupture membrane. When experiments showed any change of Rs or inadequate voltage control, data were discarded. All the experiments were performed under
761
similar conditions (e.g. similar patch duration for each protocol) at room temperature (21–22 °C). 2.4. Immunoprecipitation and Western blot Ventricular cells were treated with EGF, VO43−, AG556, or AG556 plus VO43− for 20 min at room temperature, and centrifuged at 4 °C. The cell pellet was then lysed with a lysis buffer containing (mM) 25 Tris, 150 NaCl, 100 NaF, 1.0 EDTA, 1.0 VO43−, 1.0 phenylmethylsulfonyl fluoride, and 1% sodium deoxycholate, 0.1% SDS, 1% Triton X-100, 1 μg/ml leupeptin, and 1 μg/ml aprotenin. Protein quantification of lysates was made using a protein assay reader (Bio-Rad Laboratories, Hercules, CA), and diluted to equal concentrations. Proteins were immunoprecipitated overnight at 4 °C using 2 μg of antisodium channel antibody (III–IV loop, Upstate Biotechnology, Lake Placid, NY) and 50 μl of protein A/G beads (Upstate). Immunoprecipitated proteins bound to pelleted protein A/G beads were washed thoroughly with PBS solution, denatured in Laemmli sample buffer, separated using SDS-PAGE, and electroblotted onto nitrocellulose membranes. The immunoblots were probed with anti-phosphotyrosine antibody (1:1000, Upstate) overnight at 4 °C in a blocking solution containing 5% nonfat dry milk in TBS with 0.1% Tween 20, and subsequently treated with goat anti-mouse IgG-HRP antibody (1:5000, Santa Cruz Biotechnology, Santa Cruz, CA) for 1 h at room temperature. Blots were developed with enhanced chemiluminescence (ECL, Amersham Biosciences) and exposed on X-ray film (Fuji Photo Film GmbH, Düsseldorf, Germany). The blots were then stripped and reprobed with the anti-sodium channel antibody to determine total sodium channel protein levels. The film was scanned, imaged by a Bio-Imaging System (Syngene, Cambridge, UK), and analyzed via GeneTools software (Syngene). 2.5. Statistical analysis Nonlinear curve-fitting was performed using Pulsefit (HEKA) and Sigmaplot (SPSS, Chicago, IL). Paired and/or unpaired Student's two-tailed t-test were used as appropriate to evaluate the statistical significance of the differences between two group means, and ANOVA was used for multiple groups. Group data are expressed as means ± SE. Values of P <0.05 were considered statistically significant. 3. Results 3.1. Effect of EGF on INa Fig. 1A shows the time course of INa recorded in a representative cell with a 30-ms voltage step from − 130 to − 35 mV (inset) in the absence and presence of 100 ng/ml EGF. EGF gradually increased INa, accelerated activation and inactivation, and caused a cross over of current traces (Fig. 1B). The current traces were fitted to mono-exponential function with activation and inactivation time constants (τm and τh) in the absence and presence of 100 ng/ml EGF (Fig. 1B). The τm and
762
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
Fig. 1. Effect of EGF on INa. (A) Time course of INa recorded in a representative cell with a 30-ms step to − 35 from −130 mV (inset) in the absence and presence of 100 ng/ml EGF. (B) Raw data of INa (from panel A) were fitted to monoexponential function (superimposed with raw data) with activation and inactivation time constants (τm and τh) shown in the absence (control) and presence of 100 ng/ml EGF. EGF increased INa, accelerated activation and inactivation, and induced a cross over of current traces. (C) Voltage-dependent INa recorded in a representative cell with 30-ms voltage steps from − 130 mV to between − 70 and +5 (5-mV increment, 0.2 Hz) as shown in the inset in the absence (control) and presence of 100 ng/ml EGF (10 min). (D) Voltagedependence of inactivation (I/Imax, determined with the protocol shown in the inset) and activation (g/gmax) of INa were determined with voltage-dependent INa in the absence and presence of 100 ng/ml EGF. Curves were fitted to Boltzmann distribution: y = 1/{1 + exp[(Vm − V0.5)/S]}, where Vm is membrane potential, V0.5 is the midpoint of voltage, and S is the slope factor. V0.5 of INa inactivation: − 85.5 ± 1.3 mV in control, and − 87.1 ± 1.4 mV in EGF (n = 6, P = NS), and S: 5.3 ± 0.1 mV for control, and 5.1 ± 0.1 mV for EGF (P = NS vs. control). V0.5 of INa activation: −43.3 ± 1.9 mV for control, and − 44.6 ± 1.7 mV for EGF (n = 6, P = NS), and S: − 5.7 ± 0.2 mV for control; − 6.0 ± 0.3 mV for EGF (P = NS). (E) Recovery of INa from inactivation was determined with the protocol shown in the inset. The recovery curves were fitted to mono-exponential functions. No change in time constant was observed with 100 ng/ml EGF.
τh of INa were 0.55 ± 0.04 and 2.13 ± 0.17 ms during control, and 0.44 ± 0.04 and 1.76 ± 0.15 ms after EGF application (n = 6, P < 0.01 vs. control). Fig. 1C displays the voltage-dependent INa increased by EGF in a typical experiment. The voltage dependence of inactivation (availability, I/Imax) of INa was determined with the protocol as shown in the inset (Fig. 1D), and inactivation curves were fitted to Boltzmann distribution to obtain voltage for half-inactivation (V0.5) and slope factor (S). The voltage dependence of conductance activation variable (g/gmax) of INa was determined from INa–V relationships measured from voltage-dependent INa (Fig. 1C)
for each cell as described previously [18], and also fitted to Boltzmann distribution. No change of voltage-dependence of INa was observed with EGF (Fig. 1D), and V0.5 of INa inactivation was − 85.5 ± 1.3 mV in control, and − 87.1 ± 1.4 mV in EGF (n = 6, P = NS). S was 5.3 ± 0.1 mV for control, and 5.1 ± 0.1 mV for EGF (P = NS vs. control). The V0.5 of INa activation was − 43.3 ± 1.9 mV for control, and − 44.6 ± 1.7 mV for EGF (n = 6, P = NS), and the S was − 5.7 ± 0.2 mV for control; −6.0 ± 0.3 mV for EGF (P = NS). Recovery of INa from inactivation was determined using a paired pulse protocol as shown in the inset (Fig. 1E), and recovery curves were fitted to mono-exponential functions. The recovery time constant (4.74 ± 0.33 ms in control) was not altered by 100 ng/ml EGF (5.47 ± 0.03 ms in EGF, n = 6, P = NS). The above results suggest that the enhancement of cardiac INa by EGF is probably mediated by the increased tyrosine phosphorylation of sodium channel by EGFR kinase. If this is true, the effect should be antagonized by an EGFR kinase inhibitor or potentiated by a protein tyrosine phosphatase (PTP) inhibitor. To test this, we firstly employed EGF, and then coapplied EGF and the highly selective EGFR kinase inhibitor AG556 or the PTP inhibitor VO43−. As Fig. 2 is shown, AG556 antagonizes the increase of INa by EGF, while VO43− potentiates the effect (Figs. 2A and B). EGF at 100 ng/ml enhanced INa (at − 35 mV, − 2.1 ± 0.3 nA) to 115.8 ± 3.2% (− 2.3 ± 0.4 nA) of control (n = 7, P < 0.05), and the effect was antagonized to 96.9 ± 1.7% (− 2.0 ± 0.3 nA) by combination of EGF with 5 μM AG556 (Fig. 2C), while co-application of EGF and 1 mM VO43− potentiated INa (− 2.0 ± 0.3 nA) to 126.7 ± 3.1% (− 2.6 ± 0.4 nA) of control (n = 6, P < 0.05 vs. EGF: − 2.2 ± 0.4 nA) (Fig. 2D). These results indicate that cardiac INa is likely augmented by an increase of tyrosine phosphorylation of cardiac sodium channels by activating EGFR kinase or by inhibiting PTPs, and decreased by a reduction of tyrosine phosphorylation of the channels by inhibiting EGFR kinase. 3.2. Effects of VO43− on INa To further confirm the effect of inhibiting PTPs on INa, VO43− was directly applied in a different set of experiments. Fig. 3A illustrates the time course of INa recorded in a representative cell in the absence and presence of VO43−. VO43− at 1 mM gradually increased INa. The effect reached a steady-state level within 3 min, and recovered on washout. VO43−, as EGF, accelerated activation and inactivation of INa, and induced a cross over of the current. The raw data of activation (m) and inactivation (h) were also fitted to a mono-exponential function. The τm and τh are shown in Fig. 3B in the absence (control) and presence of 1 mM VO43−. Mean values of τm and τh at − 35 mV were 0.53 ± 0.04 and 2.08 ± 0.16 ms in control and 0.41 ± 0.03 and 1.65 ± 0.12 ms in VO43− (n = 7, P < 0.01 vs. control). Voltage-dependent INa (Fig. 3C) was substantially enhanced by 1 mM VO43−, and the effect was reversed by washout. VO43− at 0.3, 1, and 3 mM increased INa (at − 35 mV, from − 2.2 ± 0.3 nA of control to − 2.3 ± 0.3, − 2.7 ± 0.4, and − 2.9 ± 0.4 nA, respectively, n = 7, P < 0.05 or P < 0.01 vs. control) in a concentration-dependent manner.
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
763
was employed. Fig. 4A illustrates the time-course of INa recorded in the absence and presence of 5 μM AG556 in a typical experiment. INa was gradually decreased to a steadystate level by AG556, and partially recovered upon washout. Original INa traces at corresponding time points are shown in right of the panel of Fig. 4A. With fitting activation and inactivation of INa, the τh (at − 35 mV) was found to be slightly increased by 5 μM AG556 (1.8 ± 0.1 ms in control, 2.1 ± 0.1 ms in AG556, n = 7, P < 0.05), while the τm was not affected (0.54 ± 0.1 ms in control, 0.53 ± 0.1 ms in AG556, P = NS). Fig. 4B displays voltage-dependent INa traces recorded in a representative cell with the protocol as shown in the inset of Fig. 1C. Voltage-dependent INa was substantially inhibited by 5 μM AG556, and the effect was partially reversed by washout for 10 min. INa was − 2.5 ± 2.0 nA (at − 35 mV) for control, − 1.4 ± 0.1 nA for 5 μM AG556 (n = 10, P < 0.01 vs. control), and −2.2 ± 0.2 nA after washout (P < 0.01 vs. AG556).
Fig. 2. EGF on INa and AG556 or VO3− 4 effect. (A) Time course of INa recorded in a representative cell with a 30-ms step to −35 from −130 mV (inset) during control, in the presence of 100 ng/ml EGF, and co-application of EGF and 5 μM AG556. Original currents at corresponding time points were shown in the right inset of the panel. (B) Time course of INa recorded in a representative cell during control, in the presence of 100 ng/ml EGF, and co-application of EGF and 1 mM VO3− 4 . Original currents at corresponding time points were shown in the right inset of the panel. (C) Histogram showing percentage values of INa during control, after application of 100 ng/ml EGF, and combination of EGF with 5 μM AG556 (n = 7, **P < 0.01 vs. control, ##P < 0.01 vs. EGF). (D) Mean percentage values of INa in control, after EGF, and EGF plus 1 mM VO3− 4 (n = 6, **<0.01 vs. control, ##P < 0.01 vs. EGF).
The effects of VO43− on voltage-dependent activation and inactivation, and recovery of INa were examined with the same procedure as described in Fig. 1D. No significant changes in the kinetics were observed with 1 mM VO43− (Figs. 3D and E). V0.5s of INa activation and inactivation were −45.5 ± 1.8 and −88.6 ± 0.8 mV in control, and −47.3 ± 3.0 mV and −90.5 ± 1.3 mV with VO43− (n= 8, P = NS). INa recovery time constant was 4.11 ± 0.45 ms in control, and 4.12± 0.27 ms with 1 mM VO43− (n= 8, P = NS). These results suggest that the decrease of tyrosine dephosphorylation of cardiac sodium channel by inhibiting PTPs with VO43−, as activating EGFR kinase with EGF, enhances current amplitude, and accelerates activation and inactivation of cardiac INa without affecting voltage-dependent and recovery kinetics. 3.3. Effects of AG556 on INa To verify the effect of directly inhibiting EGFR kinase on cardiac INa, the highly selective EGFR kinase inhibitor AG556
Fig. 3. Enhancement of INa by VO3− 4 . (A) Time course of INa recorded in a representative cell with a 30-ms step from − 130 mV to − 35 mV in the absence and presence of 1 mM VO3− 4 (OV). (B) Raw data of INa were fitted to monoexponential function with activation and inactivation time constants (τm and τh) 3− shown in the absence (control) and presence of 1 mM VO3− 4 (OV). VO4 , as EGF, enhanced INa, accelerated activation and inactivation and induced a cross over of the current. (C) Voltage-dependent INa recorded in a representative cell using the same protocol as shown in the inset of Fig. 1C in the absence and presence of 1 mM VO3− 4 . (D) Mean values of voltage-dependence of inactivation (I/Imax) and activation (g/gmax) of INa were determined in the absence and presence of 1 mM VO3− 4 . Curves were fitted to Boltzmann distribution. Voltagedependent inactivation and activation were not affected by VO3− 4 . (E) Recovery curves of INa were fitted to mono-exponential functions. No change in time constant was observed with 1 mM VO3− 4 .
764
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
conductance was not significantly altered by AG556 (− 46.8 ± 0.6 mV for control; − 48.5 ± 0.7 mV for AG556, n = 6, P = NS). Recovery of INa from inactivation was well fitted to monoexponential functions with time constant of 5.26 ± 0.14 ms (n = 7) in control (Fig. 4D). After administration of 5 μM AG556, recovery of INa was slowed, and the curve was only fitted to bi-exponential functions with rapid and slow time constants of 2.8 ± 0.4 and 219.3 ± 71 ms. The effect was reversed by washout (τ: 6.16 ± 0.28 ms, P = NS vs. control). These results suggest that the inhibition of tyrosine phosphorylation of cardiac sodium channels with the EGFR kinase inhibitor AG556 reduces INa, negatively shifts inactivation potential, and slows recovery time course of INa. 3.4. Influence of VO43− on AG556 effects
Fig. 4. Inhibition of INa by AG556. (A) Time course of INa recorded in a typical experiment with a 30-ms voltage step to − 35 from − 130 mV (inset) in the absence and presence of 5 μM AG556. Original current traces are shown in the right of the panel. (B) Voltage-dependent INa traces recorded in a representative myocyte with the same protocol as shown in the inset of Fig. 1C during control, in the presence of 5 μM AG556, and washout for 10 min. (C) Mean values of voltage-dependence of inactivation (I/Imax) and activation (g/gmax) of INa in the absence and presence of 5 μM AG556. Curves were fitted to Boltzmann distribution. Voltage-dependence of inactivation was negatively shifted by AG556. (D) Mean values of curves for recovery of INa from inactivation in the absence and presence of 5 μM AG556. Recovery curves were fitted to monoexponential functions during control, after drug washout (for 15 min); however, the curve was fitted to bi-exponential functions in the presence of 5 μM AG556.
The concentration–response relationship for the inhibition of INa (at − 35 mV) by AG556 (0.1–30 μM) was evaluated, and fitted to the Hill equation: E = Emax/[1 + (IC50/C)b], where E is the percentage for inhibiting INa at concentration C, Emax is the maximum inhibition, IC50 is the concentration for halfmaximum action, and b is the Hill coefficient. The IC50 of AG556 for inhibiting INa was 7.6 μM (close to that for EGFR tyrosine kinase inhibition in other system [21,22]), and Emax was 61.4%. Fig. 4C illustrates the curves for voltage dependence of inactivation and activation conductance of INa evaluated with the procedure as described in Fig. 1D. AG556 negatively shifted the V0.5 of inactivation (availability) by 7.8 mV, from − 89.9 ± 1.5 mV in control to − 97.7 ± 1.1 mV in AG556, (n = 7, P < 0.01), and the effect was partially reversed by 3.2 mV (to − 94.5 ± 1.6 mV) by washout. However, the V0.5 of activation
To study whether the AG556 effects on INa would be antagonized by the PTP inhibitor VO43−, we pretreated cells with VO43− and then AG556 was co-applied. Fig. 5A displays the time course of INa recorded in a typical experiment at − 35 mV. A small reduction of INa was observed with 5 μM AG556 in the presence of 1 mM VO43−; however, an additional inhibition was noted as VO43− was removed. Original current traces at corresponding time points are shown in right of the panel. Fig. 5B summarizes the group data for the percent changes of INa with AG556, and VO43− plus AG556. AG556 (5 μM) alone inhibited INa by 33.4 ± 2.2% (from − 2.3 ± 0.4 nA of control to − 1.5 ± 0.2 nA, n = 16, P < 0.01), and the effect was reduced to 8.8 ± 2.3% (− 2.2 ± 0.3 nA, n = 6, P < 0.01 vs. AG556 alone) in cells pretreated with 1 mM VO43− (−2.4 ± 0.3 nA). These results indicate that reduction of cardiac INa by AG556 is substantially antagonized by the PTP inhibitor VO43−. In addition, pretreatment with 1 mM VO43− substantially antagonized the negative shift of inactivation potential of INa by AG556 (Fig. 5C). The inactivation V0.5 (− 86.4 ± 1.2 mV) was not significantly changed by 5 μM AG556 (− 89.7 ± 1.0 mV, n = 6, P = NS) in cells pretreated with 1 mM VO43−. Moreover, the slowed recovery of INa from inactivation by AG556 (Fig. 4D) was also antagonized by the pretreatment with 1 mM VO43− (Fig. 5D). The recovery was fitted to monoexponential functions (τ: 5.2 ± 0.15 and 4.1 ± 0.16 ms before and after AG556, n = 6, P = NS). These results further suggest that tyrosine phosphorylation by EGFR kinase contributes to the modulation of cardiac INa via the interaction of EGFR kinase and PTPs in guinea pig ventricular myocytes. 3.5. Other PTK inhibitors on INa Tyrphostin AG1295, a selective inhibitor of platelet-derived growth factor receptor (PDGFR) kinase, was employed to study the effect of inhibiting PDGFR kinase on INa. AG1295 at 100 μM (100 times higher than IC50s for inhibiting PDGFR tyrosine kinase [23]) produced no significant inhibition of INa (− 2.31 ± 0.22 nA for control, − 2.27 ± 0.23 nA for AG1295, n = 4, P = NS). We also used PP2, a membrane permeable Srcfamily PTK inhibitor, to examine the effect of inhibiting Srcrelated kinases on native cardiac INa. PP2 at 5 μM (1000 times
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
765
higher than IC50s for inhibiting Src-family tyrosine kinase [24]) did not affect INa in five cells (− 2.51 ± 0.24 nA in control, − 2.46 ± 0.23 nA in PP2, P = NS). Voltage-dependent activation and inactivation, and recovery of INa were not influenced by the application AG1295 or PP2 (data not shown). 3.6. Tyrosine phosphorylation level of cardiac sodium channels Tyrosine phosphorylation level of cardiac sodium channels was studied using immunoprecipitation and Western blotting analysis. Fig. 6A illustrates tyrosine phosphorylation level of sodium channels during control and in the presence of EGF, VO43−, AG556, and VO43− plus AG556 in a representative experiment. EGF (100 ng/ml) or VO43− (1 mM) increased tyrosine phosphorylation level of cardiac sodium channels; however, AG556 at 30 μM (a high concentration that caused a substantial reduction of tyrosine phosphorylation level of cardiac sodium channels), decreased the phosphorylation level, and the effect was significantly antagonized by pretreatment with 1 mM VO43− . Fig. 6B summarizes the quantitative tyrosine phosphorylation levels of sodium channels. The
Fig. 6. Tyrosine phosphorylation levels of cardiac sodium channels in the absence 3− (control) and presence of 100 ng/ml EGF, 1 mM VO3− 4 , 30 μM AG556, or VO4 (OV) plus AG556. (A) Upper panel: tyrosine phosphorylation levels of cardiac sodium channels obtained by immunoprecipitation and Western blot. Lower panel: total sodium channel protein levels obtained by stripping and reprobing the membrane with the anti-sodium channel antibody. (B) Histogram summarizes the mean values of the relative levels of tyrosine-phosphorylated sodium channel protein analyzed by the densitometry. The amount of protein from the immunoprecipitation (as in panel A) was relative to those from the Western blots. The relative values of treatment groups were then normalized by that of solvent control. n = 4, *P < 0.05, **P < 0.01 vs. control; #P < 0.01 vs. AG556.
tyrosine phosphorylation level was increased by EGF (23.8 ± 5.4%, n = 4, P < 0.01 vs. control) or VO43− (20.2 ± 2.1%, n = 4, P < 0.01 vs. control), decreased by AG556 (54.7 ± 7.4%, P < 0.01), and reduced by 15.3 ± 8.5% with AG556 in cells pretreated with 1 mM VO43− (P < 0.01 vs. AG556 alone). These results indicate that tyrosine phosphorylation level of sodium channel protein is increased by EGF or VO43−, and decreased by the EGFR kinase inhibitor AG556. 4. Discussion 4.1. Novel findings of the present study
Fig. 5. VO3− 4 effect on AG556. (A) Time course of INa recorded in a typical experiment with a 30-ms step to − 35 from − 130 mV (inset) under conditions of 3− the pretreatment with 1 mM VO3− 4 (OV), VO4 plus 5 μM AG556, and washout of VO3− 4 . Current traces at corresponding time points are shown in the right of the panel. (B) Histogram illustrates the mean values of INa at − 35 mV during control, in the presence of 5 μM AG556 (n = 8, **P < 0.01 vs. control), VO3− 4 plus AG556 (n = 6, **P < 0.01 vs. control; ##P < 0.01 vs. AG556). (C) Voltage-dependence of 3− inactivation (I/Imax) of INa in the presence of 1 mM VO3− 4 , and VO4 plus 5 μM AG556 (n = 6), and curves were fitted to Boltzmann distribution. (D) Recovery curves of INa were fitted to monoexponential functions. With pretreatment of 1 mM VO3− 4 , no change in time constant was observed with 5 μM AG556.
In the present study, we demonstrated that EGF enhanced voltage-gated cardiac INa in guinea pig ventricular myocytes, and accelerated activation and inactivation of the current (Fig. 1). The augmentation of INa by EGF was antagonized by the selective EGFR kinase inhibitor AG556, and potentiated by the PTPs inhibitor VO43− (Fig. 2). VO43−, like EGF, enhanced INa and accelerated both activation and inactivation of the current (Fig. 3). In addition, AG556 decreased INa, negatively shifted inactivation potential, and slowed recovery of INa from inactivation (Fig. 4). These effects were antagonized by VO43− (Fig. 5). Immunoprecipitation and Western blotting analysis demonstrated that tyrosine phosphorylation level of cardiac sodium channels was increased by EGF or VO43−, and reduced
766
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
by AG556. The decreased phosphorylation level by AG556 was significantly reversed by VO43− (Fig. 6). These results suggest that native cardiac INa is regulated by EGFR kinase and/or PTPs. 4.2. Regulation of cardiac INa by PKC and PKA The voltage-gated cardiac sodium channel of the heart was down-regulated by PKC [6,25]. However, modulation of cardiac INa by cAMP/PKA is inconsistent. Earlier reports showed that INa was inhibited by β-adrenoceptor stimulation and cAMP in rat [4,26], and guinea pig [5] cardiac myocytes, while later studies demonstrated an enhancement of INa by cAMP/PKA in rabbit ventricular myocytes [3], and cloned human cardiac sodium channels expressed in Xenopus laevis oocytes [27]. Ono and colleagues [28] demonstrated that voltage-dependent activation and inactivation of INa were negatively shifted by cAMP in guinea pig and canine cardiac myocytes, and cAMP inclusion in pipette solution could result in either increase or decrease of cardiac INa depending on the selected holding and test potentials. The complex effects of PKA on cardiac sodium channel function suggest that cAMPdependent PKA may phosphorylate multiple serine/threonine sites that cause different functional effects [29]. 4.3. Comparison with previous reports regarding INa regulation by PTKs It is well documented that PTKs regulate the long term cellular processes (e.g. cell growth, differentiation, oncogenesis, etc.) [1,9], and also participate in the modulation of cellular plasma membrane ion channels [11], including Ca2+ channels and several types of K+ channels [11–14], as well as volumesensitive Cl− channels [15] in different types of cells. The regulation of ion channels by PTKs are mostly based on the effects of soluble PTK inhibitors that are relatively specific for certain kinase families (e.g., AG556, PP1, and PP2), and supportive evidence would be from an opposite effect and/or antagonizing effect (against PTK inhibitors) resulting from inhibition of PTPs [11,14,15]. However, limited information is available in the literature regarding the regulation of cardiac INa by PTKs. An earlier report described that cardiac INa was inhibited by several PTK inhibitors (including the broad spectrum PTK-inhibitor genistein, the EGFR kinase inhibitor AG957, and Src-related PTK inhibitor PP2) in rabbit ventricular cells [16]. This study suggested a clue of possible modulation of cardiac INa by PTKs, the concentrations of PTK inhibitors were much higher than those of inhibiting PTKs in other systems [16]; however, PTP inhibitor was not used to confirm the effects observed with different PTK inhibitors, and the possible PTK-independent effects [11] on cardiac INa would not be excluded. In the present study, we did not find that PP2 at 5 μM (1000 times higher than IC50s for inhibiting Src-family tyrosine kinase [24]) had any effect on INa in guinea pig ventricular myocytes. Nevertheless, we found that EGF and the PTP inhibitor VO43− (Figs. 1 and 3) increased cardiac INa, and accelerated time-dependent activation
and inactivation without affecting voltage-dependence and recovery of the current. VO43− antagonized the current reduction by the EGFR kinase inhibitor AG556 (Fig. 5). The simplest explanation for the actions of EGF, VO43−, AG556, and their interaction is that phosphorylation and/or dephosphorylation of a critical tyrosine residue on sodium channel or a signaling molecule directly or indirectly augments or decreases sodium channel activity. In this model (Fig. 7), activation of EGFR tyrosine kinase by EGF increases sodium channel activity by enhancing the tyrosine phosphorylation of the channel, whereas inhibition of EGFR kinase by AG556 decreases the channel activity by reducing the tyrosine phosphorylation. In addition, inhibition of PTP (with VO43−) increases INa by allowing its unopposed phosphorylation. This may also explains why pretreatment with VO43− precluded the effect of AG556. Once tyrosine becomes phosphorylated during VO43− pretreatment, blocking EGFR kinase will have little effect because there is no efficient means of dephosphorylating the residue. This notion is supported by the results from immunoprecipitation and Western blotting analysis, which provide a direct evidence that tyrosine phosphorylation level of cardiac sodium channels is enhanced by EGF or VO43−, and decreased by AG556 (Fig. 6). These results are consistent with the reports that the coordinated action of PTKs and PTPs is required for protein tyrosine phosphorylation [30,31]. In the heart, Malhotra and colleagues demonstrated with immunohistochemical technique that tyrosine-phosphorylated and nonphosphorylated sodium channel β1-subunits were differentially localized in cardiac myocytes [32]. A recent report showed that human NaV1.5 channel was modulated by the co-transfection of PTPH1 in HEK 293 cells [33]. In the neuron, Ratcliffe and colleagues reported that voltage-gated sodium channels were associated with receptor protein tyrosine phosphatase beta [34]. Moreover, it was reported that L-type Ca2+ channels in GH3
Fig. 7. Schematic model for regulation of cardiac INa channel by EGFR tyrosine kinase and PTPs. Tyrosine phosphorylation sites on cardiac sodium channel or signaling molecules are postulated. The tyrosine of the sodium channel is phosphorylated by EGFR kinase and dephosphorylated by PTPs. Phosphorylation of tyrosine by one or more steps (dashed line) favors INa channel opening, whereas dephosphorylation favors closure. AG556 inhibited EGFR tyrosine kinase reduced the phosphorylation, while VO3− 4 inhibited PTPs and decreased dephosphorylation of the channel. In addition, EGFR kinase and PTPs may also affect gene expression in cardiac cells.
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
767
cells and in guinea pig ventricular myocytes was also regulated by the interplay of PTKs and PTPs [15,35]. Nevertheless, the opposite response of INa to PTPs and PTKs were observed in other systems. In neuronal cells, the inhibition of protein tyrosine phosphatase-β associated brain neuronal sodium channel with pervanadate suppressed INa [34]. In PC12 cell line, several receptor tyrosine kinase activators (e.g. EGF) produced an inhibition of INa [36]. Whether these diverse results are related to the subtle variation of sodium channels, different species, tissue/cell types, and/or experimental conditions remain to be studied.
autoimmune myocarditis [38]. These studies suggest that suppression of EGFR kinase is likely cardioprotective; however, whether this effect is related to the INa inhibition remains to be studied. In summary, the present study provides the novel information that cardiac native INa is regulated by EGFR kinase and PTPs. EGFR kinase positively, while PTPs negatively modulates cardiac INa. Electrical excitability of the heart would be controlled by the interplay of EGFR kinase and PTPs.
4.4. Limitations and potential significance of the present study
The study was supported by Sun Chieh Yeh Heart Foundation of Hong Kong. The authors thank Professor TakMing Wong in the Department of Physiology for his substantial support.
One of limitations was that cardiac INa was recorded at room temperature (to ensure the well-controlled voltage clamp). This would underestimate the effects of the enzymes (i.e. kinases and phosphatrases) on the regulation of the channel. In addition, the present study reported the regulation of voltage-gated cardiac sodium current by EGFR kinase in guinea pig ventricular myocytes; however, molecular mechanisms were not clarified. For instance, it is unknown how the increased phosphorylation would enhance the activation and inactivation of cardiac sodium channel, which remains to be studied with a computer modeling of the sodium channel in the future. Another limitation was that we could not well explain why our results from native cardiac myocytes were not consistent with those observed in cloned NaV1.5 channel expressed in HEK 293 cells, in which Fyn, a Src-family tyrosine kinase, regulated cloned cardiac NaV1.5 channels by positively shifting inactivation potential without affecting the current amplitude [17], and co-expression of NaV1.5 with PTPH1 negatively shifted inactivation potential [33]. One of possibilities for the discrepancy is likely related to the various responses of sodium channel to different kinases in native cardiac cells (EGFR kinase) and expressed HEK 293 cells (Src-family kinases). An opposite regulative effect of ICl.vol by EGFR kinase and Src-family kinases was recently reported in human atrial myocytes [15]. In addition, the different response may also be related to that the native cardiac cells have intact sodium channel with both α- and β-subunits in the present study, while HEK 293 cells expressed only cloned NaV1.5 αsubunit was used in the recent reports [17,33]. However, these possibilities should be investigated in detail by making comparative study in native cardiac cells and HEK 293 cells expressing both NaV1.5 and sodium channel β-subunits in the future. The results from the present observation demonstrated the strong evidence that cardiac INa in native guinea pig ventricular myocytes was modulated by the coordination of EGFR kinase and PTPs. Activation of EGFR kinase increased, while activation of PTPs decreased cardiac INa channels, suggesting that EGFR kinase and PTPs could regulate cardiac electrical excitability. It is interesting to note that the inhibition of EGFR kinase by AG556 reduced ischemic zones, improved heart performance in experimental myocardial ischemia of a rat model [37], and ameliorated progression of experimental
Acknowledgments
References [1] Levitan IB. Modulation of ion channels by protein phosphorylation and dephosphorylation. Annu Rev Physiol 1994;56:193–212. [2] Marx S. Ion channel macromolecular complexes in the heart. J Mol Cell Cardiol 2003;35:37–44. [3] Matsuda JJ, Lee H, Shibata EF. Enhancement of rabbit cardiac sodium channels by beta-adrenergic stimulation. Circ Res 1992;70:199–207. [4] Schubert B, Vandongen AM, Kirsch GE, Brown AM. Inhibition of cardiac Na+ currents by isoproterenol. Am J Physiol 1990;258:H977–82. [5] Ono K, Kiyosue T, Arita M. Isoproterenol, DBcAMP, and forskolin inhibit cardiac sodium current. Am J Physiol 1989;256:C1131–7. [6] Qu Y, Rogers J, Tanada T, Scheuer T, Catterall WA. Modulation of cardiac Na+ channels expressed in a mammalian cell line and in ventricular myocytes by protein kinase C. Proc Natl Acad Sci U S A 1994;91: 3289–93. [7] Lu T, Lee HC, Kabat JA, Shibata EF. Modulation of rat cardiac sodium channel by the stimulatory G protein alpha subunit. J Physiol 1999;518 (Pt 2):371–84. [8] Ahmmed GU, Xu Y, Hong DP, Zhang Z, Eiserich J, Chiamvimonvat N. Nitric oxide modulates cardiac Na(+) channel via protein kinase A and protein kinase G. Circ Res 2001;89:1005–13. [9] Hunter T. Signaling—2000 and beyond. Cell 2000;100:113–27. [10] Siegelbaum SA. Channel regulation. Ion channel control by tyrosine phosphorylation. Curr Biol 1994;4:242–5. [11] Davis MJ, Wu X, Nurkiewicz TR, Kawasaki J, Gui P, Hill MA, et al. Regulation of ion channels by protein tyrosine phosphorylation. Am J Physiol: Heart Circ Physiol 2001;281:H1835–62. [12] Gao Z, Lau CP, Wong TM, Li GR. Protein tyrosine kinase-dependent modulation of voltage-dependent potassium channels by genistein in rat cardiac ventricular myocytes. Cell Signal 2004;16:333–41. [13] Tiran Z, Peretz A, Attali B, Elson A. Phosphorylation-dependent regulation of Kv2.1 channel activity at tyrosine 124 by Src and by protein–tyrosine phosphatase epsilon. J Biol Chem 2003;278:17509–14. [14] Ogura T, Shuba LM, McDonald TF. L-type Ca2+ current in guinea pig ventricular myocytes treated with modulators of tyrosine phosphorylation. Am J Physiol 1999;276:H1724–33. [15] Du XL, Gao Z, Lau CP, Chiu SW, Tse HF, Baumgarten CM, et al. Differential effects of tyrosine kinase inhibitors on volume-sensitive chloride current in human atrial myocytes: evidence for dual regulation by Src and EGFR kinases. J Gen Physiol 2004;123:427–39. [16] Wang Y, Wagner MB, Kumar R, Cheng J, Joyner RW. Inhibition of fast sodium current in rabbit ventricular myocytes by protein tyrosine kinase inhibitors. Pflugers Arch 2003;446:485–91. [17] Ahern CA, Zhang JF, Wookalis MJ, Horn R. Modulation of the cardiac sodium channel NaV1.5 by Fyn, a Src family tyrosine kinase. Circ Res 2005;96:991–8.
768
H. Liu et al. / Journal of Molecular and Cellular Cardiology 42 (2007) 760–768
[18] Li GR, Lau CP, Shrier A. Heterogeneity of sodium current in atrial vs. epicardial ventricular myocytes of adult guinea pig hearts. J Mol Cell Cardiol 2002;34:1185–94. [19] Pu J, Balser JR, Boyden PA. Lidocaine action on Na+ currents in ventricular myocytes from the epicardial border zone of the infarcted heart. Circ Res 1998;83:431–40. [20] Hanck DA, Sheets MF. Time-dependent changes in kinetics of Na+ current in single canine cardiac Purkinje cells. Am J Physiol 1992;262:H1197–207. [21] Gazit A, Osherov N, Posner I, Yaish P, Poradosu E, Gilon C, et al. Tyrphostins. 2. Heterocyclic and alpha-substituted benzylidenemalononitrile tyrphostins as potent inhibitors of EGF receptor and ErbB2/neu tyrosine kinases. J Med Chem 1991;34:1896–907. [22] Levitzki A, Gazit A, Osherov N, Posner I, Gilon C. Inhibition of protein– tyrosine kinases by tyrphostins. Methods Enzymol 1991;201:347–61. [23] Kovalenko M, Gazit A, Bohmer A, Rorsman C, Ronnstrand L, Heldin CH, et al. Selective platelet-derived growth factor receptor kinase blockers reverse sis-transformation. Cancer Res 1994;54:6106–14. [24] Hanke JH, Gardner JP, Dow RL, Changelian PS, Brissette WH, Weringer EJ, et al. Discovery of a novel, potent, and Src familyselective tyrosine kinase inhibitor. Study of Lck- and FynT-dependent T cell activation. J Biol Chem 1996;271:695–701. [25] Xiao GQ, Qu Y, Sun ZQ, Mochly-Rosen D, Boutjdir M. Evidence for functional role of epsilonPKC isozyme in the regulation of cardiac Na(+) channels. Am J Physiol: Cell Physiol 2001;281:C1477–86. [26] Herzig JW, Kohlhardt M. Na+ channel blockade by cyclic AMP and other 6-aminopurines in neonatal rat heart. J Membr Biol 1991;119:163–70. [27] Frohnwieser B, Chen LQ, Schreibmayer W, Kallen RG. Modulation of the human cardiac sodium channel alpha-subunit by cAMP-dependent protein kinase and the responsible sequence domain. J Physiol (London) 1997;498:309–18. [28] Ono K, Fozzard HA, Hanck DA. Mechanism of cAMP-dependent modulation of cardiac sodium channel current kinetics. Circ Res 1993; 72:807–15.
[29] Murphy BJ, Rogers J, Perdichizzi AP, Colvin AA, Catterall WA. cAMPdependent phosphorylation of two sites in the alpha subunit of the cardiac sodium channel. J Biol Chem 1996;271:28837–43. [30] Sun H, Tonks NK. The coordinated action of protein tyrosine phosphatases and kinases in cell signaling. Trends Biochem Sci 1994;19:480–5. [31] Ostman A, Bohmer FD. Regulation of receptor tyrosine kinase signaling by protein tyrosine phosphatases. Trends Cell Biol 2001;11:258–66. [32] Malhotra JD, Thyagarajan V, Chen C, Isom LL. Tyrosine-phosphorylated and nonphosphorylated sodium channel beta1 subunits are differentially localized in cardiac myocytes. J Biol Chem 2004;279: 40748–54. [33] Jespersen T, Gavillet B, van Bemmelen MX, Cordonier S, Thomas MA, Staub O. Cardiac sodium channel Na(v)1.5 interacts with and is regulated by the protein tyrosine phosphatase PTPH1. Biochem Biophys Res Commun 2006;348:1455–62. [34] Ratcliffe CF, Qu Y, McCormick KA, Tibbs VC, Dixon JE, Scheuer T, et al. A sodium channel signaling complex: modulation by associated receptor protein tyrosine phosphatase beta. Nat Neurosci 2000; 3:437–44. [35] Cataldi M, Taglialatela M, Guerriero S, Amoroso S, Lombardi G, di Renzo G, et al. Protein-tyrosine kinases activate while protein–tyrosine phosphatases inhibit L-type calcium channel activity in pituitary GH3 cells. J Biol Chem 1996;271:9441–6. [36] Hilborn MD, Vaillancourt RR, Rane SG. Growth factor receptor tyrosine kinases acutely regulate neuronal sodium channels through the src signaling pathway. J Neurosci 1998;18:590–600. [37] George J, Biner S, Keren P, Barshack I, Goldberg I, Sherez J, et al. Tyrphostin AG-556 reduces myocardial infarct size and improves cardiac performance in the rat. Exp Mol Pathol 2003;74:314–8. [38] George J, Barshack I, Goldberg I, Keren P, Gazit A, Levitzki A, et al. The effect of early and late treatment with the tyrphostin AG-556 on the progression of experimental autoimmune myocarditis. Exp Mol Pathol 2004;76:234–41.