Voltage-gated Na+ channels (VGSC) are transmembrane proteins that are essential for the initiation and propagation of action potentials in neuronal excitability. Because neurons express a mixture of Na+ channel isoforms and protein kinase C (PKC) isozymes, the nature of which channel is being regulated by which PKC isozyme is not known. We showed that DRG VGSC Nav1.7 (TTX-sensitive) and Nav1.8 (TTX-resistant), expressed in Xenopus oocytes were differentially regulated by protein kinase A (PKA) and PKC isozymes using the two-electrode voltage-clamp method. PKA activation resulted in a dose-dependent potentiation of Nav1.8 currents and an attenuation of Nav1.7 currents. PKA-induced increases (Nav1.8) and decreases (Nav1.7) in peak currents were not associated with shifts in voltage-dependent activation or inactivation. The PKA-mediated increase in Nav1.8 current amplitude was prevented by chloroquine, suggesting that cell trafficking may contribute to the changes in Nav1.8 current amplitudes. A dose-dependent decrease in Nav1.7 and Nav1.8 currents was observed with the PKC activators phorbol 12-myristate, 13-acetate (PMA) and phorbol 12,13-dibutyrate. PMA induced shifts in the steady-state activation of Nav1.7 and Nav1.8 channels by 6.5 and 14 mV, respectively, in the depolarizing direction. The role of individual PKC isozymes in the regulation of Nav1.7 and Nav1.8 was determined using PKC-isozyme-specific peptide activators and inhibitors. The decrease in the Nav1.8 peak current induced by PMA was prevented by a specific ϵPKC isozyme peptide antagonist, whereas the PMA effect on Nav1.7 was prevented by ϵPKC and βIIPKC peptide inhibitors. The data showed that Nav1.7 and Nav1.8 were differentially modulated by PKA and PKC. This is the first report demonstrating a functional role for ϵPKC and βIIPKC in the regulation of Nav1.7 and Nav1.8 Na+ channels. Identification of the particular PKC isozymes(s) that mediate the regulation of Na+ channels is essential for understanding the molecular mechanism involved in neuronal ion channel regulation in normal and pathological conditions.
Voltage-gated Na+ channels play a pivotal role in the initiation and generation of action potentials in neurons. These Na+ channels also contribute to the dynamic regulation of neuronal excitability and are good candidates for modulation by second messenger pathways. The functional channels are 260 kDa transmembrane proteins (α-subunit) that associate with one or more 32- to 36-kDa auxiliary β-subunits (β1-β3). Biochemical, peptide mapping, and mutational analysis of brain Na+ channels have shown that the cytoplasmic loop between domains DI and DII possesses several shared protein kinase A (PKA) and protein kinase C (PKC) phosphorylation sites (Cantrell et al. 2002; Murphy et al. 1993; Smith and Goldin 1996). The inactivation gate has a unique site for PKC phosphorylation (Qu et al. 1996; West et al. 1991).
PKA and PKC have an effect on native and cloned heterologously expressed cardiac (Murphy et al. 1996), skeletal muscle (Numann et al. 1994), and brain Na+ channels (Numann et al. 1991). However, little is known about the effects of phosphorylation on isolated Na+ channels expressed in sensory neurons (Fitzgerald et al. 1999). Other studies using in vivo models have reported increased kinase activity (PKA and/or PKC) in nociceptive C fibers during hyperalgesia (Dina et al. 2001; Igwe and Chronwall 2001; Martin et al. 1999).
Activation of PKA by forskolin or 8-bromo cAMP reduces peak current levels of brain Nav1.1 and Nav1.2 Na+ channels without significantly changing their steady-state properties. This effect can be repressed by PKA peptide inhibitors (Li et al. 1992; Smith and Goldin 1996, 1998). Interestingly, similar results have been observed in Xenopus oocytes and transfected Chinese Hamster Ovary (CHO) cells (Li et al. 1992; Smith and Goldin 1996, 1998). On the other hand, PKA activation by proinflammatory, hyperalgesic agents such as serotonin and prostaglandin E2, results in a dose-dependent increase in the amplitude of peak tetrodotoxin-resistant (TTX-R) currents in sensory neurons (England et al. 1996; Gold et al. 1996). The rise in current after PKA activation is accompanied by changes in gating properties and increased rates of inactivation (England et al. 1996; Gold et al. 1996, 1998). Similar effects are observed with cloned Nav1.5 and Nav1.8 (TTX-R) channels expressed in oocytes and COS-7 cell, respectively, when they are stimulated by PKA activators (Fitzgerald et al. 1999; Schreibmayer et al. 1994; Zhou et al. 2000). The potentiation of Nav1.5 currents can be attributed to an increased transport of channels to the cell surface because the potentiation can be prevented by the Golgi blocker chloroquine (Zhou et al. 2000). These findings illustrate two important points: first, PKA modulates the various Na+ channel isoforms (brain vs. cardiac and nociceptor Na+ channels), most likely by different mechanisms, and, second, the channels exhibit similar biochemical behavior in different expression systems, pointing to an interchangeability among amphibian (Xenopus oocytes), mammalian, and native cell expression systems.
Functional regulation of Na+ channels by PKC has been shown in cloned channels expressed in Xenopus oocytes and stable mammalian cell lines and in channels in cultured native neurons (DRG and brain). Activation of PKC by phorbol esters [phorbol 12-myristate, 13-acetate (PMA) and phorbol 12,13-dibutyrate (PDBu)] causes a marked decrease in the peak current of a number of cloned Na+ channel isoforms (Nav1.2, Nav1.4, and Nav1.5) in oocytes (Bendahhou et al. 1995; Murray et al. 1997; Schreibmayer et al. 1991). Similar decreases in macroscopic currents are also observed when CHO, Chinese hamster lung 1610, and neuroblastoma cells are treated with PKC activators (Godoy and Cukierman 1994; Numann et al. 1991; Qu et al. 1994). These results imply that phosphorylation by PKC inhibits the function of Na+ channels. The inhibition of channel activity has been associated with a slowed time constant of inactivation (Nav1.2 in CHO cells, oocytes, and neuroblastoma cells) and negative shifts in steady-state inactivation (Nav1.2 in neuroblastomas and Nav1.5 in Chinese hamster lung 1610 cells) (Godoy and Cukierman 1994; Qu et al. 1994).
However, different results have been observed when similar PKC activators are used on isolated rat DRG sensory neurons (Gold et al. 1996, 1998). In this case, the PKC activators PMA and PDBu cause a dose-dependent increase in the amplitude of TTX-R Na+ currents. The rise in amplitude is attributed to increased conductance, a slight hyperpolarized shift in steady-state availability, and increased rates of activation and inactivation (Gold et al. 1998). Because DRG neurons express at least two types of TTX-R channel (Nav1.8 and Nav1.9), the specific type of Na+ channel that is modulated is unknown (Amaya et al. 2000; Sangameswaran et al. 1997).
The characterization of the modulatory role of individual PKC isozymes has been largely limited by the lack of selective isozyme activators and inhibitors. The goal of this study was to characterize the electrophysiological effects of PKA and PKC activators on the Nav1.7 and Nav1.8 channels of sensory neurons and to determine the regulatory role of PKC isozymes. We report that both PKA and PKC (ϵPKC and βIIPKC isozymes) inhibited Nav1.7 currents. The potentiation of peak currents by Nav1.8 is in agreement with the data reported for the same channel in COS-7 cells (Fitzgerald et al. 1999). On the other hand, the inhibition of currents by PKC (ϵPKC) differs from previous studies conducted using native sensory neurons (Gold et al. 1998).
Complementary DNA (cDNA) coding for rat dorsal root ganglion (DRG) Nav1.7/pCDNA3 was obtained from Gail Mandel (Department of Neurobiology, State University of New York), while Nav1.8/pSP64T was cloned as previously described (Vijayaragavan et al. 2001). cRNA was prepared using the T7 (Nav1.7/pCDNA3) or SP6 (Nav1.8/pSP64T) mMessage mMachine kit (Ambion, TX).
Expression and electrophysiology in Xenopus oocytes
Xenopus oocytes were used because we observed disparities in the levels of current expression between Nav1.7 and Nav1.8 channels in mammalian cells (tsA201 and CHO cells). Xenopus laevis females were anesthetized with 1.5 mg/ml tricaine (Sigma, Oakville, ON, Canada), and two or three ovarian lobes were surgically removed. Follicular cells surrounding the oocytes were removed by incubation at 22°C for 2.5 h in calcium-free oocyte medium (OR2) composed of (in mM) 82.5 NaCl, 2.5 KCl, 1 MgCl2, and 5 HEPES. The pH was adjusted to 7.6 with 1 N NaOH containing 2 mg/ml collagenase (Sigma). The oocytes were washed with calcium-free medium supplemented with 50% Leibovitz's L-15 medium (Gibco Life Technologies, Burlington, ON, Canada) containing 15 mM HEPES, 5 mM l-glutamine, and 10 mg/ml gentamycin (OR3). The pH was adjusted to 7.6 with 1 N NaOH. The oocytes were stored in OR3 medium until further use. Stage VI-V oocytes were selected and microinjected with a 50-nl mixture of equal concentrations of α-subunit coding for the Nav1.7 or Nav1.8 channels and β1-subunit cRNA. The oocytes were stored at 18°C and used for experiments 3–5 days postinjection, depending on the level of expression of each channel type.
The whole cell sodium current from cRNA-injected oocytes was measured using a two-microelectrode voltage clamp. The oocytes were impaled with <2 MΩ electrodes containing 3 M KCl and were voltage-clamped with an OC-725 oocyte clamp (Warner Instruments, Hamden, CT). The Ringer's bath solution contained (in mM) 90 NaCl, 2 KCl, 1.8 CaCl2, 2 MgCl2, and 5 HEPES (pH 7.6). Currents were filtered at 1.5 kHz with an 8-pole Bessel filter and sampled at 10 kHz. Data were acquired and analyzed using pCLAMP software v7 (Axon Instruments). The dose-dependent curves and the PKC activator concentrations eliciting half-maximal inhibition (EC50) were determined by least-square fitting the dose response curves to the following function where I is the current in the presence, I0 is the current in the absence of the PKC activators. The n value was set to 1 for experiments conducted with PMA, PDBu for Nav1.7 and for PMA on Nav1.8 and Yo was set to 0. However for PDBu on Nav1.8, the n value was set at 0.77 and Yo was set 0.32. The voltage dependence of activation was determined by eliciting depolarizing pulses from a holding potential of –100 mV to potentials ranging from –80 to +60 mV in 5-mV increments. Current activation curves of the channels were plotted using the following Boltzmann equation where the GNa (conductance) values for each clamped oocyte were determined by dividing the peak sodium current by the driving force (Vm – ENa). The reversal potential (ENa) for the oocytes expressing the channels was estimated by extrapolating the linear ascending segment between 0 to +20 mV for Nav1.7 and between +20 to +40 mV for Nav1.8 of an I-V curve to the zero current level. V is equal to the test voltage, V1/2 is the voltage at which the channels are half-maximal activated, and kv is the slope factor. Conductance versus voltage data were fitted with a two-state Boltzmann equation.
The voltage dependence of inactivation was determined by eliciting 500-ms conditioning pulses to voltages between –110 and +30 mV in 5-mV increments followed by a standard test pulse to either –10 mV (Nav1.7+β1) or +10 mV (Nav1.8+β1). Test currents were normalized and plotted versus the conditioning voltage. Inactivation curves were fitted to the following Boltzmann relation
The capacitances of the oocytes were measured by calculating the area under the capacitance transient using pClampfit 8.0. Changes in Xenopus oocyte membrane surface areas were monitored by following changes in membrane capacitance (Cm). While real-time changes in Cm cannot be accurately measured using the two-microelectrode voltage-clamp technique, the steady-state Cm can be recorded (Zhou et al. 2000). To measure Cm, the transient capacitative current was recorded by applying a square voltage step to –65 mV from a holding potential of –100 mV, where no membrane currents were activated. The integration of this transient current generated the charge (Q) transferred by applying the voltage step (V). The Cm was then calculated by dividing Q by V.
Drug and peptide preparation
All reagent grade chemicals were obtained from Sigma. Stock solutions (50 mM) of the PKC activators PMA and PDBu were prepared in 100% DMSO and then diluted to produce 50 mM working stocks, which were further diluted to the appropriate concentrations immediately before use. A 250 mM stock solution of the PKA activator forskolin was prepared in 100% DMSO. Stock solutions of the PKC inhibitor calphostin C (5 mM) and the PKA inhibitors chloroquine (500 mM) and H-89 (10 mM) were also prepared in 100% DMSO and diluted to the appropriate concentrations immediately before use. The maximum DMSO concentration in the Ringer solution was 0.02%. Stock solutions were stored at –20°C in aliquots. PKC isozyme-specific agonist and antagonist peptides were obtained from Dr. Daria Mochly-Rosen (Stanford University, CA). All peptides used were >90% pure. Stock peptides were stored as small aliquots at –80°C and thawed slowly on ice before use.
Three methods were used to apply the agents— bath superfusion: PKA and PKC activators were applied continuously during the time course of the experiment using a bath superfusion system; preincubation: oocytes were incubated in the PKA and PKC inhibitor solutions for 30–60 min at 18°C before adding the respective kinase activator; and intracellular application: 50 nl of the PKC peptide antagonist (50 μM) or agonist (0.1–0.5 μM) were microinjected into the vegetal pole of the oocytes, which were allowed to recover for 1 h (antagonist) or used immediately (agonist).
To account for potential nonspecific effects by DMSO, separate sets of experiments were performed. Oocytes were superfused with Ringer solution containing 0.02% DMSO (the amount in the 50 μM forskolin bath solution). No rise in Nav1.8 current amplitudes was observed in experiments conducted for 30 min with DMSO alone (data not shown).
Data in results are expressed as means ± SE. The currents of paired groups of oocytes injected with Nav1.7+β1 or Nav1.8+β1 treated with Ringer (control) were directly compared with oocytes superfused with the PKA/PKC or PKA/PKC activators and co-injected with the specific PKC peptide antagonist/agonist. Statistical analysis were performed using a repeated-measure ANOVA with the Dunnett's ANOVA test. The homogeneity of the correlation between repeated measures was tested with the Brown and Forsythe test. The results were considered significant if P values were under 0.05. The data were analyzed using the SAS statistical package program (SAS Institute, Cary, NC).
Native DRG neurons express a mixture of TTX-sensitive (Nav1.1, Nav1.2, Nav1.3, and Nav1.7) and TTX-resistant (Nav1.8 and Nav1.9) Na+ channel isoforms. To determine how the Nav1.7 and Nav1.8 isoforms are modulated by the activation of PKA and PKC, they were expressed in Xenopus oocytes in combination with the Na+ channel β1-subunit because previous studies showed that the auxiliary subunit modulates the gating, kinetics and expression of these channels (Vijayaragavan et al. 2001). This expression system was chosen because Nav1.8 is poorly expressed in mammalian cells (see methods).
PKA inhibited Nav1.7 and increased Nav1.8 currents
PKA activators and inhibitors were used to investigate the effects of PKA on the functional properties of cloned rat Nav1.7 and Nav1.8 voltage-gated Na+ channels. Xenopus oocytes expressing Nav1.7 or Nav1.8 were superfused with Ringer solution for 3–5 min until the inward Na+ currents were stable. Forskolin (0.1–50 μM) was added after this initial period. Forskolin caused a bell-shaped dose-dependent decrease in Nav1.7 currents, where 10 μM (6.9 ± 4.0% inhibition, n = 5) and 50 μM (6.7 ± 2.0% inhibition, n = 7) forskolin caused similar levels of peak current inhibition as 0.1 μM (9.8 ± 2.2% inhibition, n = 4) forskolin (Fig. 1A). The threshold concentration for the forskolin-induced decrease in Nav1.7 was 100 nM (9.8 ± 2.2%, n = 5), whereas the greatest decrease in peak currents occurred with 1 μM forskolin. Representative whole cell current traces before and after the treatment with 1 μM forskolin are shown in Fig. 1, B and C, respectively. A wash-out with Ringer solution was observed but was not complete (data not shown). Forskolin (1 μM) caused a 23.6 ± 3.8% (n = 6) decrease in peak current that was not associated with any significant changes (P > 0.05) in the time constant of the current decay (Table 1) or shift in the current-voltage relationship (Fig. 1D). While 1 μM forskolin reduced the peak current amplitude, it did not alter the steady-state activation or inactivation (P > 0.05) of the channels (Table 1, Fig. 2A). The decrease in the amplitude of Nav1.7 currents was abolished by preincubating the oocytes with 1 μM H-89, a specific inhibitor of PKA (Hidaka et al. 1984). The 27.1% inhibition (Fig. 2B, □) of peak currents with 1 μM forskolin was prevented by 1 μM H-89 (Fig. 2B, ▵), where a 4.7 ± 5.6% increase in peak currents was observed within 30 min (n = 5). This increase was not significant and was inconsistent because in some experiments the steady state was reached after >30 min. Simultaneous measurements of electrical capacitance, an indication of cell membrane surface area, changed very little (Fig. 2C). The decrease in peak current was thus not a result of a change in membrane surface area. To control for nonspecific effects of forskolin, oocytes expressing Nav1.7 were injected with 0.03 units of the PKA catalytic subunit. The PKA catalytic subunit caused a 76.0 ± 6.1% (n = 7) inhibition of Nav1.7 current. In a parallel set of control experiments, oocytes injected with PBS exhibited a 21.6 ± 2.6% (n = 5) decrease in Nav1.7 current. All experiments were conducted under Ringer superfusion
On the other hand, forskolin potentiated the current amplitudes of oocytes expressing Nav1.8 channels. A dose-dependent increase in current was observed. The highest concentration of forskolin tested (50 μM) produced a 57.0 ± 7.5% (n = 5) rise in current (Fig. 3A). Little wash-out with Ringer solution was observed for Nav1.8 channels treated with 50 μM forskolin (data not shown). Figure 3, B and C, shows whole cell current traces of representative experiments before and after, respectively, exposure to 50 μM forskolin. The increase in peak current was not associated with a change (P > 0.05) in the time constant of inactivation (Table 1) or a shift in the current/voltage relationship (Fig. 3D). The potentiation of currents was also not associated with shifts in steady-state activation or inactivation (P > 0.05) of the channel (Table 1, Fig. 4A).
The forskolin-induced slow, unsaturable increase in current was probably not solely due to the direct phosphorylation of the channel. Direct phosphorylation usually occurs rapidly and in a saturable manner as observed with Nav1.7 and voltage-gated K+ channels KV1.5 (Kwak et al. 1999). PKA regulates the cell transport machinery (Muniz et al. 1997). To determine whether the activation of PKA promotes trafficking of Nav1.8 channels, oocytes were pretreated with 100 μM chloroquine before adding 50 μM forskolin. Chloroquine is a monovalent carboxylic ionophore that disrupts vesicular trafficking from the Golgi complex to the plasma membrane (Tartakoff 1983). Chloroquine preincubation (30 min) almost completely inhibited the forskolin-induced rise in peak Nav1.8 currents (103.9 ± 8.4%, n = 5, □, Fig. 4B). Chloroquine alone did not have an effects on the channels (data not shown). A significant 6.5 ± 3.6% (P < 0.05, n = 4) increase in the capacitance of the oocytes was observed after a 30-min exposure to 50 μM forskolin (Fig. 4C). The increase in capacitance was also inhibited by the pretreatment with 100 μM chloroquine (data not shown). The chloroquine inhibition of forskolin-induced rise in Nav1.8 currents and capacitance could be due to an increase in the surface area of the oocytes cell membrane as a result of an increase in the number of inserted channels.
PKC-induced modulation of Nav1.7 and Nav1.8
To determine the impact of PKC on oocytes expressing Nav1.7, we tested the effect of the PKC activators PMA and PDBu on the properties of the channels. A dose-dependent decrease in Nav1.7 current was observed with both PMA and PDBu (Fig. 5A). The concentration of the PKC activator eliciting half-maximal inhibition was determined by least-square fits to the function described in methods. The fitting resulted in an EC50 value of 9.73 ± 0.5 and 35.4 ± 0.2 nM for PMA and PDBu respectively. The mean maximal inhibition of 500 nM PMA was 83.4 ± 4.9% (n = 6) and 84.9 ± 1.4% (n = 5) for 100 nM PDBu. At –10mV, PMA (10 nM) inhibited Nav1.7 by 54.7 ± 1.72% (n = 5) and PDBu (20 nM) inhibited Nav1.7 by 39.9 ± 4.63% (n = 4). Representative whole cell traces before and after the addition of 10 nM PMA are shown in Fig. 5, B and C.
No changes in the time constants of inactivation were observed at peak voltage (–10 mV) when the Nav1.7 channels were treated with either 10 nM PMA or 20 nM PDBu (P > 0.05, Table 1). However, significant shifts (P < 0.05) in steady-state activation were observed with both PKC activators. PMA (10 nM) induced a 6.5-mV depolarizing shift (Table 1, Fig. 5D) while PDBu (20 nM) induced a 9-mV depolarized shift (Table 1, Fig. 6A). No shift in steady-state inactivation (P > 0.05) was observed with either PKC activator (Table 1). Figure 6B shows that the inhibition of the peak Nav1.7 current by 20 nM PDBu was prevented in oocytes preincubated with 100 nM calphostin C, a general PKC inhibitor (14.5 ± 8.6% inhibition in current (n = 5) compared with a 54.4 ± 2.4% decrease (n = 4) with the 20 nM PDBu control). A slight but significant decrease in oocyte capacitance was observed during the time course of the experiments in the presence of 20 nM PDBu (n = 3, Fig. 6C).
We also observed a dose-dependent decrease in the peak current amplitude with Nav1.8. Based on the least-square fits, 50% inhibition was obtained with 1.38 ± 0.6 nM PMA and 1.82 ± 0.3 nM PDBu (Fig. 7A). A mean maximal inhibition of 77.8 ± 3.1% was observed with 10 nM PMA (n = 5) and 99.9 ± 0.02% with 100 nM PDBu (n = 7). A 2 nM concentration of the PKC activators was employed for subsequent experiments. Representative whole cell current traces before and after exposure to 2 nM PDBu are shown in Figs. 7, B and C, respectively. No changes in the time constant of the current (P > 0.05) were observed at the peak voltage (+10 mV) with PMA (Fig. 7D) and PDBu (Table 1). Also, no shift in voltage-dependent inactivation (P > 0.05) was observed with either PKC activator (Table 1, Fig. 8A). However, the steady-state activation of Nav1.8 was shifted (P < 0.05) by 14 and 12 mV in the depolarized direction for PMA (Fig. 7D) and PDBu, respectively (Fig. 8A). The PDBu-induced decrease in peak Nav1.8 current was completely abolished by preincubating the oocytes with 100 nM calphostin C (Fig. 8B). PDBu (2 nM) reduced the current by 1.9 ± 3.4% in the presence of calphostin C (○, n = 4) versus 66.7 ± 0.6% (squares, n = 4) in the absence of calphostin C. Calphostin C (100 nM) thus prevented the PMA-induced decrease in Nav1.8 current by 97.2%. No change in capacitance was observed during the time course of the 2 nM PDBu treatment (n = 3, Fig. 8C).
Role of individual PKC-isozymes in the modulation of Nav1.7 and Nav1.8
Previous studies using Xenopus oocytes demonstrated the existence of several PKC isozymes, including α-, βI-, βII-, γ-, δ-, ϵ-, and ζPKC (Johnson and Capco 1997). Two PKC isozyme subfamilies, the conventional cPKC isozymes α-, βI-, βII-, and γPKC that contain the Ca2+ binding domain (C2-containing) and the novel nPKC (δ-, θ-, ϵ-, and ηPKC) or C2-less isozymes, are stimulated by phorbol esters. To determine which PKC isozymes are involved in the inhibition of Nav1.7 and Nav1.8 currents, we used specific peptide activators (ϵV1.7-PKC agonist) and inhibitors (αV5.3-, βIIV5.3-, ϵV1.2-, and γC2.4-PKC antagonist). The peptides were preinjected into the oocytes expressing the channels prior to adding the PMA (see methods). Figure 9A shows that the slow, time-dependent decrease in peak Nav1.7 current on exposure of the oocytes to 10 nM PMA was prevented when they were preinjected solely with selected peptide-isozyme-specific PKC antagonists. The βIIV5.3-PKC and ϵV1.2-PKC antagonists decreased the PMA inhibition of Nav1.7 to 25.9 ± 1.8% (P = 0.02, n = 4) and 28.5 ± 2.9% (P = 0.02, n = 5), respectively, compared with PMA alone (65.7 ± 10.8% inhibition, n = 3).
However, PMA inhibition of Nav1.7 alone (65.7 ± 10.8% inhibition, n = 3) was similar to that observed with PMA in the presence of other PKC isozyme-specific peptide antagonists such as γC2.4 (60.2 ± 16.3% inhibition, n = 3), αV5.3 (56.5 ± 9.7% inhibition, n = 3), and the scramble negative control ϵV1.2-PKC peptide (56.5 ± 9.7% inhibition, n = 3).
To selectively activate ϵ-PKC, 0.5 μM of the ϵV1.7-PKC peptide agonist was injected into oocytes expressing Nav1.7 and peak inward Na+ currents were measured in Ringer solution. The ϵV1–7 peptide is derived from the regulatory V1-region of ϵPKC and selectively activates the translocation of ϵPKC (Dorn et al. 1999). Figure 9B shows the effect of this ϵPKC agonist (pseudo ϵV1–7 RACK2) on Nav1.7 currents. In these experiments, oocytes expressing the channels were injected with pseudo ϵV1–7 RACK2, and currents were immediately recorded in Ringer solution alone. Pseudo ϵV1–7 RACK2 inhibited peak currents by 64.9% (n = 3, P > 0.05) Fig. 9B summarizes the percentage inhibition of Nav1.7 currents by PMA (control) and the effect of various PKC isozyme-specific peptide antagonists and agonists. These results demonstrate the involvement of ϵPKC and βIIPKC but not γPKC and αPKC in PMA-induced inhibition of Nav1.7.
Similar experiments were also conducted with oocytes expressing Nav1.8 channels. PMA (2 nM) induced a slow time-dependent decrease in Nav1.8 peak currents as it did with Nav1.7. However, only the ϵV1.2-PKC antagonist prevented this inhibition (Fig. 9C). Figure 9D summarizes the percentage inhibitions by the various specific PKC peptide antagonists and agonists. Unlike Nav1.7, only the ϵV1.2-PKC antagonist prevented the PMA-induced decrease in Nav1.8 current (29.2 ± 3.1% inhibition, P = 0.006, n = 7). In the presence of other isozyme-specific peptide antagonists such as βIIV5.3-PKC (61.2 ± 8.4% inhibition, n = 4), γC2.4 (45.5 ± 3.3% inhibition, n = 4, P ≥ 0.05), αV5.3 (53.5 ± 4.8% inhibition, n = 4), and the ϵV1.2 scramble negative control (44.2 ± 3.2% inhibition, n = 4, P ≥ 0.05), PMA showed similar or statistically insignificant differences in current inhibition compared with control uninjected oocytes expressing Nav1.8 (60.5 ± 9.8% inhibition, n = 4). A lower concentration of pseudo ϵV1–7 RACK2 (0.1 μM) than that needed for Nav1.7 (0.5 μM) was required to produce a time-dependent decrease in currents in oocytes expressing Nav1.8 channels (Fig. 9C). Pseudo ϵV1–7 RACK2 inhibited peak currents by 53.8% (n = 5, P > 0.05), which is comparable to the control (PMA alone) and supports the notion that activation of ϵPKC leads to a decrease in Nav1.8 currents.
The present study shows that the cloned voltage-gated Na+ channels, Nav1.7 and Nav1.8, from the sensory nervous system are regulated in different ways by PKA and PKC. Nav1.7 is regulated by both ϵPKC and βIIPKC, whereas Nav1.8 is regulated by ϵPKC. This is the first report demonstrating a functional role for PKC isozymes in the regulation of Nav1.7 and Nav1.8 neuronal Na+ channels.
Modulation by PKA
The activation of PKA reduced Nav1.7 peak currents by 24% but did not alter the steady-state activation, inactivation, or time constant of fast inactivation. The data suggest that PKA inhibited currents without altering the voltage sensitivity or kinetics of Nav1.7 gating. The effects of forskolin were abolished by the specific PKA inhibitor H-89, indicating that PKA activation plays an essential role in current inhibition. A similar PKA-dependent inhibition of currents, with no change in the voltage dependence of activation or inactivation or kinetics, has been reported for Nav1.2 channels (Li et al. 1992). However, in the study by Li et al. (1992), the reduction in current amplitudes resulted from a decrease in the probability of channels opening in response to depolarization. Phosphorylation of several sites on the DI-DII linker domain (Ser573, Ser610, Ser623, and Ser687) is required for the reduction in Nav1.2 current (Cantrell et al. 2002; Murphy et al. 1993; Rossie and Catterall 1987). These consensus phosphorylation sites are conserved in Nav1.7, suggesting that a similar mechanism may account for the reduction in current induced by low concentrations of forskolin (1 μM).
An unexpected finding was that at high concentrations (5–50 μM), forskolin failed to alter Nav1.7 currents. The relative specific effects of low concentrations of forskolin argue against a nonspecific direct effect on the channel and suggest that cAMP may have a biphasic effect on Nav1.7. One possibility is that at high concentrations, elevated cAMP levels may activate protein phosphatases such as phosphatase 2A and calcineurin (Feschenko et al. 2002; Fimia and Sassone-Corsi 2001). These phosphatases are known to dephosphorylate PKA-specific phosphorylation sites on Nav1.2 channels (Chen et al. 1995; Murphy et al. 1993) and hence may act by suppressing the PKA-dependent phosphorylation of Nav1.7 when high concentrations of forskolin are used.
Alternatively, the biphasic effect of forskolin could be explained by a differential modulation of PKA phosphorylation. Biphasic modulation of Nav1.2 has been observed when one (Ser573) or more of the consensus PKA sites on the DI-DII linker is eliminated (Smith and Goldin 2000). In addition, Smith and Goldin (2000) also observed that the length of the DI-DII linker upstream from the conserved PKA phosphorylation sites is an important determinant of Nav1.2 current potentiation. PKA phosphorylation sites on the DI-DII interdomain of Nav1.7 are highly conserved. However, the region of the DI-DII linker immediately upstream from Ser573, a key PKA phosphorylation site, is 25 amino acids longer in Nav1.7 than in Nav1.2 or Nav1.8. In addition to its role in PKA-dependent phosphorylation, the DI-DII region also contributes to the binding of synaptotagamin, a synaptic protein that may regulate PKA-induced potentiation of Na+ currents (Sampo et al. 2000). The consensus phosphorylation sites, the 25-aminoacid insertion and the binding of regulatory proteins may contribute to the differential effects of high and low forskolin concentrations on Nav1.7 channels.
In contrast, forskolin-induced activation of PKA increased Nav1.8 currents by 57%. This potentiation occurred over a relatively slow time course (5–10 min) and inhibited by chloroquine, an inhibitor of intracellular protein trafficking. Our data suggest that the majority of the increase in the Nav1.8 currents resulted from the incorporation of additional channels into the plasma membrane. PKA stimulation is known to increase the intracellular trafficking of several membrane-bound proteins (glucose transporter GLUT4, chloride channel CFTR, Kv1.1 K+ channel, and Nav1.5 Na+ channel) (Holman and Kasuga 1997; Levin et al. 1995; Takahashi et al. 1996; Zhou et al. 2000). In the axons of Aplysia bag cells, PKA activation enhances the rate of organelle transport along the microtubule track by two- to threefold (Azhderian et al. 1994). A similar mechanism may account for the observed increase in Nav1.8 currents. Cyclic AMP signaling may modulate an accessory protein that preferentially associates with certain Na+ channel isoforms and targets them for transport to the membrane. Differences in trafficking may explain why PKA activation potentiates the currents of some Na+ channel isoforms (Nav1.5 and Nav1.8, TTX resistant) but not others (Nav1.2 and Nav1.7, TTX sensitive).
Modulation by PKC
Phorbol esters such as PMA or PDBu activate a wide range of PKC isozymes that belong to the conventional cPKC (α-, βI-, βII-, and γPKC) and novel nPKC families (δ-, ϵ-, η-, and θPKC) and that differ in their sensitivity to Ca2+. In the present study, both PKC activators caused a dose-dependent decrease in the current amplitudes of Nav1.7 and Nav1.8 channels, which was prevented by a pretreatment with the general PKC inhibitor calphostin C. Nav1.8 channels appeared to be more sensitive to PKC regulation and required lower concentrations (Fig. 7A) than Nav1.7 (Fig. 5A) to reduce the currents. PDBu-induced reductions in Nav1.7 and Nav1.8 currents were not due to a time-dependent internalization of the plasma membrane, as previously reported for oocytes treated with PMA, because the capacitance remained constant during the time course of the experiment (Vasilets et al. 1990). The PDBu effect was accompanied by a depolarizing shift in the steady-state activation of Nav1.7 (Fig. 6A) and Nav1.8 (Fig. 8A) channels, but this could only partially account for the reduction in currents. Functional changes in both channels suggest that at least part of the effect of PKC might be mediated by phosphorylation of the channels. The effects of PKC activators on these Na+ channel isoforms need to be studied at the single channel level.
The unexpected finding that PKC caused an inhibition of Nav1.8 current is in marked contrast to other reports of the potentiation of TTX-R currents in DRG neurons by PMA, PDBu, and PGE2 (Gold et al. 1996, 1998). Decreases in current amplitudes after PKC activation have, however, been observed with cardiac Nav1.5 (ϵPKC isozyme), brain Nav1.2, and neuroblastoma Na+ channels (Godoy and Cukierman 1994; Murray et al. 1997; Numann et al. 1991; Xiao et al. 2001). Single-channel analyses show that PKC phosphorylation slows the inactivation and thus increases the time Nav1.2 channels remain open during prolonged depolarization (Numann et al. 1991). Several PKC phosphorylation sites have recently been implicated in the inhibition of Nav1.2 currents, including a serine residue on the DIII-DIV inactivation loop (Ser1506) and two sites on the DI-DII interdomain (Ser554 and Ser573) (Cantrell et al. 2002; Numann et al. 1991). Similar sites may also be required for the PKC-mediated reduction in peak currents for Nav1.7 and Nav1.8 channels because the sites are all highly conserved.
Several members of the PKC isozyme family (α-, βII-, ϵ-, and γPKC) have been implicated in the modulation of pain responses (Aley et al. 2000; Igwe and Chronwall 2001; Martin et al. 1999). Xenopus oocytes contain several PKC isozymes, including α-, βI-, βII-, γ-, δ-, ϵ-, and ζPKC (Johnson and Capco 1997). Studies of the role of individual PKC isozymes in the regulation of ion channels have been largely limited by the lack of specific isozyme activators and inhibitors. To identify the specific isozyme(s) required for Nav1.7 and Nav1.8 channel current inhibition, specific PKC-isozyme peptide inhibitors were preinjected into oocytes expressing Nav1.7 or Nav1.8 channels prior to adding PMA. In the present study, only the PKC ϵ and βII isozymes of PKC were involved in the inhibition of Nav1.7 peak currents by PMA as shown by the fact that specific peptide inhibitors of the isozymes, βIIV5.3-PKC and ϵV1.2-PKC, respectively, were able to prevent the PMA-induced inhibition. In the presence of the βIIV5.3-PKC and ϵV1.2-PKC antagonists, PMA inhibition was reduced to 26 and 29%, respectively. The other PKC isozyme peptide inhibitors tested, αV5.3 and γC2.4, were unable to prevent PMA inhibition, inhibiting the currents by 57 and 60% respectively after a challenge by 10 nM PMA. In addition, the specific ϵPKC peptide activator (ϵV1.7 PKC antagonist, 65% inhibition) emulated the effect of PMA (66% reduction) on Nav1.7 peak currents, implying that phosphorylation by ϵPKC is sufficient to inhibit currents.
A similar modulation by two different PKC isozymes (βPKC and ϵPKC) has been observed with L-type Ca2+ channels in cardiac cells (Hu et al. 2000; Zhang et al. 1997). Regulation of L-type Ca2+ channels by one or more PKC isozymes depends on the location of the channel. Similarly, native Nav1.7 channels may be modulated by various PKC isozymes, depending on their co-localization in the neuron. Xiao et al. (2003) reported that βII- and ϵPKC upregulate cloned human cardiac delayed slow rectifier K+ channels expressed in Xenopus oocytes.
Interestingly, only the ϵV1.2-PKC antagonist (29% PMA inhibition) and ϵV1.7-PKC agonist (54% inhibition) were able to prevent and emulate PMA-induced inhibition (60% inhibition) of Nav1.8 peak currents respectively. None of the other specific PKC isozyme peptide antagonists tested affected the reduction of currents by PMA.
Given the inherent limitations of the transient Xenopus oocyte heterologous expression system, additional experiments will have to be carried out in an appropriate mammalian expression system that is able to express both channels. However, Xenopus oocytes are indispensable and have been extensively used to study many regulatory aspects of ion channels and receptors by the signal transduction pathway. These include PKA/PKC mediated translocation studies of the glucose transporter (GLUT4) (Holman and Kasuga 1997), chloride channels (CFTR) (Takahashi et al. 1996), and cardiac delayed slow rectifier K+ channels (Xiao et al. 2003) as well as phosphorylation studies of Na+ channels (Nav1.1–1.2 and Nav1.5) in oocytes (Bendahhou et al. 1995; Schreibmayer et al. 1991; Smith and Goldin 1996). The results we obtained with oocytes are consistent with those reported for mammalian and native cells and thus indicate a high degree of interchangeability between the two expression systems (Ameen et al. 1999; Godoy and Cukierman 1994; Li et al. 1992; Mora et al. 1995; Numann et al. 1991). Furthermore, several mammalian PKC isozymes have also been identified in Xenopus oocytes (Johnson and Capco 1997).
Convergent modulation and physiological significance
Synergistic PKA and PKC (ϵ- and βIIPKC isozyme) regulation of ion channels may be particularly important in injured neurons that are subjected to a wide spectrum of inflammatory signals. Downregulation of Nav1.7 currents by PKA and PKC should increase the voltage thresholds required for action potential (AP) firing and a stronger depolarization should be needed to elicit a response. Thus the neurons that predominantly express Nav1.7 should exhibit a lower frequency of AP generation, which would protect the neurons against excitotoxicity caused by prolonged hyperexcitability.
In the present study, we observed a nonsynergistic modulation of Nav1.8 by PKC and PKA. Inhibition of Nav1.8 peak currents was mediated by the ϵPKC isozyme and potentiation was mediated by PKA. Modification of Nav1.8 by the ϵPKC isozyme via phosphorylation could result in short-term changes to nociceptive fibers. The results of Cesare et al. (1999), who reported that ϵPKC translocation and activation after acute bradykinin (BK) exposure (5 s) is transient, whereas prolonged exposure (5 min) causes a downregulation of the isozyme in small neurons, provides support for this idea. However, PKA-induced export of channels to the membrane may result in long-term changes in C-fibers. Potentiation of Nav1.8 peak currents is attributable to the increased transport of channels to the cell membrane. An increased number of active Nav1.8 channels on the cell surface as well as a redistribution of the channels as observed with chronic constriction nerve injury (CCI) or inflammatory induced hyperalgesia, may act in concert to increase in sensory nerve excitability (Coward et al. 2000; Gold et al. 2003; Khasar et al. 1998; Novakovic et al. 1998).
It is important to remember that sensory neurons also express a multitude of other ion channels and receptors, such as K+ and Ca2+ channels and GABA and vanilloid receptors, which are potential targets of modulation by PKC and PKA and play a key role not only in sensory excitability but also in transmitter release to amplify sensory signals (Chow et al. 1999; Premkumar and Ahern 2000; Sculptoreanu and De Groat 2003). For instance, an article by Zhang et al. (2001) reported that inhibiting K+ currents (Kv1.1, Kv1.2, Kv1.5) by PKC activation may play an important role in sensory neuron excitability. The ultimate response of sensory neurons to neurotransmitters that stimulate the PKC and/or PKA pathway depends on the global interplay of all channels, receptors, and their regulator proteins.
We thank Dr. Y. Okamura for valuable comments on the manuscript.
This study was supported by grants from the Heart and Stroke Foundation of Québec, the Canadian Institutes of Health Research (MOP-49502). Dr. M. Chahine is an Edwards Senior Investigator (Joseph C. Edwards Foundation).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 by the American Physiological Society