- Email: [email protected]
Modulation of voltage-dependent sodium channels by the δ-agonist SNC80 in acutely isolated rat hippocampal neurons
Neuropharmacology 47 (2004) 1102–1112 www.elsevier.com/locate/neuropharm
Modulation of voltage-dependent sodium channels by the d-agonist SNC80 in acutely isolated rat hippocampal neurons Christina Remy a, Stefan Remy b, Heinz Beck b, Dieter Swandulla a, Michael Hans a, a b
Institute of Physiology, University of Bonn, Wilhelmstr. 31, 53111 Bonn, Germany Department of Epileptology, University of Bonn Medical Center, Bonn, Germany
Received 20 November 2002; received in revised form 10 May 2004; accepted 18 June 2004
Abstract Following activation, voltage-gated Na+ currents (INa) inactivate on two different time scales: fast inactivation takes place on a time scale of milliseconds, while slow inactivation takes place on a time scale of seconds to minutes. Both fast and slow inactivation processes govern availability of Na+ channels. In this study, the effects of the d-opioid receptor agonist SNC80 on slow and fast inactivation of INa in rat hippocampal granule cells were analyzed in detail. Following application of SNC80, a block of the peak Na+ current amplitude (EC50: 50.6 lM, Hill coefficient: 0.518) was observed. Intriguingly, SNC80 (50 lM) also caused a selective effect on slow but not fast inactivation processes, with a notable increase in the fraction of Na+ channels undergoing slow inactivation during prolonged depolarization. In addition, recovery from slow inactivation was considerably slowed. At the same time, fast recovery processes were unaffected. The effects of SNC80 were not mimicked by the peptide d-receptor agonist DPDPE (10 lM), and were not inhibited by the opioid receptor antagonists naloxone (50–300 lM) or naltrindole (10 and 100 lM), indicating an opioid receptor independent modulation of Na+ channels. These data suggest that SNC80 not only affects d-opioid receptors, but also voltage-gated Na+ channels. SNC80 is to our knowledge hitherto the only substance that selectively influences slow but not fast inactivation processes and could provide an important tool in unraveling the mechanism underlying these distinct biophysical processes. # 2004 Elsevier Ltd. All rights reserved. Keywords: Sodium channel; SCN80; Slow inactivation; Opioids; Hippocampus
1. Introduction Voltage-dependent Na+ channels are ubiquitous membrane proteins in excitable cells. They are transmembrane proteins composed of one of several poreforming (a) subunits and accessory b-subunits. At resting membrane potential Na+ channels are closed, but are rapidly activated by depolarization, giving rise to Na+ inward currents (see for review Catterall, 1992). These Na+ currents decrease rapidly towards baseline levels as Na+ channels undergo inactivation during prolonged depolarization. Following inactivation, Na+ channels require repolarization in order to return to Corresponding author. Tel.: +49-228-287-2326; fax: +49-228-2872313. E-mail address: [email protected] (M. Hans).
0028-3908/$ - see front matter # 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropharm.2004.06.034
the resting state. Na+ channels in neuronal membranes are able to cycle through all these states on a millisecond time scale and are therefore able to mediate rapid events such as action potentials. The structural determinants of Na+ channels necessary for rapid inactivation processes have been extensively investigated (Stu¨hmer et al., 1989; West et al., 1992). In addition to fast inactivation processes, Na+ channels may also show slow inactivation and recovery from inactivation. These slow processes can modulate the availability of Na+ channels on a time scale of seconds to minutes and have been suggested to serve as molecular mechanism that preserves traces of previous activity (Toib et al., 1998). As the availability of Na+ channels crucially modulates action potential threshold, waveform, and the propensity to generate repetitive discharges, slow inactivation is also of considerable
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
interest. Interestingly, rate constants for slow recovery vary widely in the published literature (Ruben et al., 1992; Fleidervish and Gutnick, 1996; Fleidervish et al., 1996; Colbert et al., 1997). A large part of this variation may arise from the phenomenon that recovery rates depend on the duration of the depolarization used to inactivate Na+ channels, with more prolonged depolarization resulting in slower recovery rates (Toib et al., 1998; Ellerkmann et al., 2001). The rate constants of recovery are quantitatively related to the duration of prior depolarization by a power law relationship (Toib et al., 1998; Ellerkmann et al., 2001). There is good agreement that the structural correlate of fast Na+ channel inactivation is a conformational change of the interdomain linker III–IV (Stu¨hmer et al., 1989; West et al., 1992). In contrast, the mutations known to affect slow inactivation neither share a common localization nor point at the involvement of a particular region of the Na+ channel protein (Cummins and Sigworth, 1996; Alekov et al., 2001; Mitrovic et al., 2000; Hayward et al., 1997; MelamedFrank and Marom, 1999; Vilin and Ruben, 2001). Thus, the structural basis of slow inactivation and recovery may be complex and has not been fully resolved. Recovery from slow inactivation seems to be remarkably conserved across different types of neurons and species (Ellerkmann et al., 2001), whereas this is not the case for fast inactivation processes (Martina and Jonas, 1997). There are also pharmacological differences between fast recovery and slow recovery from inactivation. Anticonvulsants such as carbamazepine, lamotrigine and phenytoin strongly affect fast recovery from inactivation (Remy et al., 2003; Ragsdale and Avoli, 1998), but do not appear to interact with slow recovery processes. It would be of considerable interest to identify substances that do the converse, i.e. interact strongly with slow but not fast recovery processes. This may be particularly relevant since naturally occurring mutations in Na+ channel genes that affect slow inactivation processes are associated with several diseases such as hyperkaliemic periodic paralysis, myotonia, idiopathic ventricular fibrillation and long-QTsyndrome (for review, see Vilin and Ruben, 2001). In this study, we have investigated the effects of the putative selective d-opioid receptor agonist SNC80 (Bilsky et al., 1995; Calderon et al., 1997) on voltagedependent Na+ channels in hippocampal granule neurons. These effects did not require activation of opioid receptors, but were presumably due to a direct interaction with membrane components. We have examined both fast and slow inactivation processes in detail. Our results indicate that application of SNC80 inhibits INa amplitude and selectively prolongs the time course for recovery from slow inactivation without affecting other kinetic parameters.
1103
2. Methods 2.1. Preparation of dissociated dentate granule cells We prepared coronal slices (400 lm) from the hippocampus of male Wistar rats (40 5 day-old) with a vibratome (LeicaVT1000S) in a bath filled with ice-cold bicarbonate-buffered saline (ACSF) containing (in mM): NaCl 125, NaHCO3 25, KCl 3, NaH2PO4 1.25, MgCl2 1, CaCl2 2, and glucose 20 (pH was adjusted to 7.4 by gassing the solution with carbogen (95% O2, 5% CO2), osmolarity 300 mosmol l1). Prior to decapitation, rats were deeply anesthetized with ether. Slices were transferred to a storage chamber containing v ACSF at room temperature (21–23 C) and subsequently stored for up to 10 h. After an equilibration period of 60 min, the first slice was transferred to a tube with 5 ml solution containing (in mM): CH3SO3Na 145, KCl 3, MgCl2 1, CaCl2 0.5, HEPES 10, and glucose 15 (pH 7.4, adjusted with NaOH, osmolarity 310 mosmol l1). Pronase (protease type XIV, 2 mg/ml; Sigma) was added to the oxygenated medium. After an incubation period of 15 min at v 35 C, the slice was washed in the pronase-free solution of identical composition. The dentate gyrus was dissected and triturated with fire-polished Pasteur pipettes of decreasing apertures. The hilus was separated from the slice during trituration. The Petri dish containing the cell suspension was then mounted on the stage of an inverted microscope (Telaval, Zeiss, Jena, Germany). Isolated cells were allowed to settle for 10 min and subsequently superfused with an extracellular solution containing (in mM): CH3SO3Na 120; TEA 20; KCl 3; BaCl2 5; MgCl2 1; HEPES 10; 4-aminopyridine (4-AP) 4; CdCl2 0.03 and glucose 10 (pH 7.4, adjusted with NaOH; osmolarity 310 mosmol ll). Only those dissociated cells that showed a small round or ovoid soma (capacitance 6 2 pF) with a single process reminiscent of dentate granule cell (DGC) morphology in situ were included in this study. 2.2. Whole cell patch-clamp recording Patch pipettes (resistance 3 1 MX) fabricated from borosilicate glass capillaries (1.5 mm o.d., 1 mm i.d.; Science Products, Hofheim, Germany) on a Narishige P83 puller were filled with an intracellular solution containing (in mM): Cs-methanesulfonate 90; TEA 20; MgCl2 5; HEPES 10; BAPTA 5; CaCl2 0.5; adenosine50 -triphosphate (50 ATP-Na2) 10 and guanosine-50 -triphosphate (50 GTP-Na2) 0.5 (pH 7.4 with NaOH, 300 mosmol l1). Recordings were conducted in the tightseal whole cell configuration at room temperature using an Axopatch 200 A amplifier (Axon Instruments, Union City, CA, seal resistance >1 GX, average series resistance 4 2 MX). Series resistance was compen-
1104
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
sated between 70% and 90%, resulting in maximal residual voltage error below 1 mV (0:78 0:13 mV, n ¼ 5 for activation and 0:69 0:09 mV, n ¼ 8 for inactivation). Current signals were filtered at 5 kHz (3 dB, 4-pole low-pass Bessel filter) and sampled at least with 10 kHz (DigiData 1322A, Axon Instruments, Union City, CA). Linear leak and residual capacitance currents were digitally subtracted using a P/4 protocol (Bezanilla and Armstrong, 1977). INa properties were analyzed as described in detail elsewhere (Reckziegel et al., 1998). SNC80 was dissolved in 10% HCl and v stored at 20 C as 100 mM stock solution. Chemicals were obtained either from Tocris (SNC80, Naloxone hydrochloride) or Sigma, Germany (all others). 2.3. Voltage clamp protocols and fitting The voltage-dependence of activation and inactivation was determined using standard protocols. The conductance G(V) was calculated according to: GðV Þ ¼ Imax =ðV VNa Þ;
ð1Þ
where VNa is the reversal potential, V the command potential, and Imax the peak current amplitude. G(V) was then fitted by the following Boltzmann equation: GðV Þ ¼ Gmax =ð1 þ expððV50 V Þ=km Þ
ð2Þ
where Gmax is the maximum Na+ conductance, V50 is the voltage where G(V) is half of Gmax, and km indicates the slope of the relationship between channel inactivation and membrane voltage. Double pulse experiments were performed with a conditioning pulse at 10 mV followed by varying recovery periods from 1 to 1000 ms at either 70 or 80 mV and a subsequent testpulse at 10 mV for 10 ms. Time constants of recovery sfast and sintermediate for fast inactivation processes and sslow1 and sslow2 for slow inactivation processes were extracted from the equation: IðtÞ ¼ A0 þ A1 ð1 expð t=s1 ÞÞ þ A2 ð1 expð t=s2 ÞÞ
ð3Þ
where I(t) is the current amplitude at the time point t after onset of the voltage command and A is the amplitude contribution of the different recovery time constants. Slow recovery processes were investigated using a conditioning pulse to 10 mV of various durations (1, 3, 10, 30, 100, 300 s) followed by a 1 s recovery interval at 80 mV. This interval allowed for a virtually complete recovery of the channels from rapid inactivation and was followed by a series of brief (5 ms) testpulses to 10 mV at a frequency of 0.33 Hz aimed at monitoring the process of recovery from slow inactivation. When the current amplitude reached steady state values, the pulse protocol was terminated
and the current amplitude was normalized to these values (Imax). The relationship between the recovery time constants sslow and the duration of the preceding depolarization (t) was fitted by a power law function of the following equation: sðtÞ ¼ aðt=aÞb
ð4Þ
where a is the constant kinetic setpoint [s] and b the scaling power. Concentration–response curves were fitted with a Hill equation of the form: E cc ¼ y Emax k þ cc
ð5Þ
where c is the concentration of the used substance, Emax the maximal possible effect, k the concentration at which a half-maximal effect was obtained and E the effect evoked by the concentration c. c is the Hill coefficient. Statistical comparison was carried out with a twotailed Student’s t-test. All results are shown as the mean S:E:M: All experiments were conducted in accordance with the guidelines of the University of Bonn Animal Care Committee and were approved by the board of proper use of experimental animals of the state Nordrhein-Westfalen.
3. Results 3.1. SNC80 reduces INa amplitude We first examined the effects of SNC80 on the maximal INa amplitude. Maximal Na+ currents under control conditions were elicited by depolarizing pulses to 10 mV following a hyperpolarizing prepulse to remove fast inactivation (Fig. 1A, inset). These currents were reversibly blocked by bath application of 0.5 lM TTX (not shown, n ¼ 5). TTX-insensitive current components were not observed. Application of the putative selective d-agonist SNC80 potently inhibited INa in a dose-dependent manner (Fig. 1B). Following application of SNC80, the inhibitory effects occurred within a few seconds, saturated within 30 s and recovered following washout. The concentration-dependence of this effect could be fitted with a Hill curve shown superimposed on the data points in Fig. 1B (EC50: 50.6 lM, Hill coefficient: 0.518). 3.2. Effect of SNC80 on the voltage-dependent activation and fast inactivation of INa To determine whether the pronounced reduction in INa amplitude might be due to a change in voltagedependent activation or inactivation, we investigated
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
1105
1:6 mV, k ¼ 5:8 0:3, n ¼ 8). At lower concentration of SNC80 (1 lM), the shift in the voltage-dependence of inactivation was less pronounced (1.9 mV) but significantly different from control (Fig. 2C, left column, n ¼ 7, p < 0:05). 3.3. SNC80 does not alter recovery from fast inactivation Recovery from fast inactivation was analyzed at membrane potentials of 70 and 80 mV using a double pulse protocol (Fig. 3A, inset). The time course of recovery from fast inactivation was in both cases best fit with the sum of two exponential functions (Eq. (3)), with two time constants sfast and sintermediate. The recovery rates did not differ significantly at the two different holding potentials (Fig. 3B; 80 mV: sfast ¼ 8:9 1:2 ms, sintermediate ¼ 234 37 ms, n ¼ 11, 70 mV: sfast ¼ 8:2 0:8 ms, sintermediate ¼ 249 107 ms, n ¼ 6). In the presence of SNC80 (50 lM), the time course of recovery from fast inactivation was not significantly affected at both holding potentials (80 mV: sfast ¼ 9:1 0:7 ms, sintermediate ¼ 292 65 ms, n ¼ 11, 70 mV: sfast ¼ 8:3 0:7 ms, sintermediate ¼ 234 51 ms, n ¼ 6) Fig. 1. SNC80 inhibits Na+ currents in dentate granule cells in a dose-dependent manner. (A) Representative recording of a Na+ current elicited by a 5 ms testpulse to 10 mV from a 100 ms prepulse potential of 100 mV under control conditions (ACSF superfusion, solid line), during application of 1 lM (dashed line) and 100 lM SNC80 (dotted lines) and following washout. The voltage protocol is depicted in the inset. (B) Concentration–response relationship for the blockade of the Na+ current by SNC80. The percentage of the SNC80-mediated block of the Na+ current amplitude under ACSF superfusion is plotted versus the concentration of SNC80. Fitting the data points with a Hill equation (see Methods) yielded the following parameters: EC50 ¼ 50:6 lM, k ¼ 0:518. The Hill curve is shown superimposed on the data points (solid line).
these properties under control conditions and following application of 100 lM SNC80. The properties of voltage-dependent activation and inactivation were determined using standard voltage protocols (Fig. 2A, insets) and the conductance/amplitude–voltage relationship fitted with a Boltzmann equation shown superimposed on the data points in Fig. 2B. Application of 100 lM SNC80 (open triangles, Fig. 2B) did not significantly alter the voltage-dependence of activation (V1=2Act : 29:0 3:5 mV, k ¼ 6:3 0:7, n ¼ 5) compared to control conditions (closed triangles, Fig. 2B, V1=2Act : 30:3 1:7 mV, k ¼ 5:5 0:5, n ¼ 5). In contrast, a small but significant shift (4 mV) of the voltage-dependence of inactivation in a hyperpolarizing direction was observed in SNC80-treated cells (open triangles, Fig. 2B, V1=2Inact ¼ 58:1 2:3 mV, k ¼ 6:4 0:4, n ¼ 8, p < 0:01) compared to control conditions (closed triangles, Fig. 2B, V1=2Inact ¼ 54:1
3.4. SNC80 slows the time course of recovery from slow inactivation We then investigated the effects of SNC80 on slow recovery from inactivation. Recovery rates of INa have been previously demonstrated to be dependent on the duration of prior depolarization used to elicit inactivation (Toib et al., 1998; Ellerkmann et al., 2001). Accordingly, we have examined the rates of slow recovery from inactivation following depolarizing pulses of different durations ranging from 1 to 300 s (Fig. 4A, inset). We examined one population of cells under control conditions (Fig. 4A1) and compared it to a second population of cells preincubated with 50 lM SNC80 for 15 min (Fig. 4A2). After an interval (1 s) to allow virtually complete recovery from fast inactivation (see Fig. 3B), the slow recovery of INa was monitored as described previously using brief (5 ms) depolarizing pulses at 0.33 Hz . The time course of recovery from slow inactivation (Figs. 4A1 and A2) was best fitted by a biexponential equation with two time constants sslow1 and sslow2. In addition, both sslow1 and sslow2 were related to the duration of the conditioning pulse by a power law relation (Eq. (4)) (Toib et al., 1998; Ellerkmann et al., 2001). SNC80 (50 lM) dramatically slowed the time course of recovery from slow inactivation (open triangles Figs. 4B1–4), with significant effects both on sslow1 and sslow2 (Figs. 4B3 and 4, control: n ¼ 6 7, SNC80: n ¼ 6 10, p < 0:05, asterisks). We also analyzed the amplitudes associated with sslow1 and sslow2,
1106
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
Fig. 2. Effects of the d-agonist SNC80 on the activation and fast inactivation kinetics of the Na+ channels in dentate granule cells. (A) Families of representative recordings elicited by the voltage protocols shown in the insets. Voltage steps to different testpulse potentials from 60 to +60 mV from a 100 ms prepulse potential of 100 mV were used to characterize the activation behavior of the Na+ channel (upper panel). The voltage-dependent inactivation of the Na+ channels was analyzed by eliciting a Na+ current with a 15 ms testpulse to 10 mV after varying prepulse potentials from 100 to 20 mV (lower panel), (B) Activation and inactivation curves of the Na+ current under control conditions (closed triangles) and during application of 100 lM SNC80 (open triangles). The average values of normalized conductances were plotted versus either the prepulse (inactivation) or the testpulse potential (activation). For each individual recording, a Boltzmann function (Eq. (2)) was fitted to the data points. The solid lines superimposed on the data points show Boltzmann curves that were constructed with values for half-maximal activation (V1/2Act), inactivation (V1/2Inact) and the slope factor, (C) SNC80 significantly shifts V1/2Inact into the hyperpolarizing direction at concentrations of 1 lM (n ¼ 7, p < 0:05) and 100 lM (n ¼ 8, p < 0:01).
termed Aslow1 and Aslow2. The contribution of the fast component, Aslow1, was not altered by SNC80 application over a range of prepulse durations (Fig. 4B1). In contrast, Aslow2 exhibited a significant increase after application of SNC80 (50 lM) at all prepulse durations (Fig. 4B2 asterisks, p < 0:05). Thus, in addition to the slowing of the recovery rates, SNC80 selectively increases the fraction of Na+ channels that recover with sslow2. The relation of sslow1 and sslow2 to the duration of the conditioning pulse was then fitted with a power law function (Eq. (4)), shown superimposed on the data points (Figs. 4B3 and 4). The constant kinetic setpoint was considerably increased in the presence of SNC80 compared to control conditions (sslow1: from a ¼ 1:3 to 3.5 s; sslow2: a ¼ 31:3 41:1 s). The scaling power for sslow1 decreased from 0.19 under control conditions to 0.16 under drug application, whereas the scaling power for sslow2 increased from 0.07 to 0.17, respectively. These findings suggest that SNC80 apparently increases the time scale over which Na+ channels retain a trace of prior activity.
3.5. Effects of SNC80 on entry into slow inactivation It was also apparent from the data shown in Figs. 4A1 and A2 that application of SNC80 significantly increases the fraction of INa undergoing slow inactivation (compare leftmost data points in Figs. 4A1 and A2). We have examined this phenomenon in more detail in Fig. 5. The fraction of INa undergoing slow inactivation was derived as the ratio of INa amplitude 1 s after the conditioning pulses (open circles, Fig. 5A1) to INa amplitude following complete recovery from slow inactivation (Fig. 5A2, asterisks). Assessing the time course of entry into slow inactivation (Fig. 5A2), it becomes apparent that SNC80 promotes entry into an inactivated state preferentially during the first seconds of the conditioning pulse. Application of SNC80 (open squares) significantly enhanced the fraction of INa undergoing slow inactivation at all prepulse durations tested (Fig. 5A2, control: n ¼ 6 7, SNC80: n ¼ 6 10, p < 0:05). Compared to control conditions, the fraction of available INa in the presence of SNC80 was reduced by 8.5%, 34.3%, 43.1%, 51.3%, 55.6% and
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
1107
using a 5 ms testpulse to 10 mV (see inset in Fig. 5B1). In our hands, slow inactivation was clearly dependent on the voltage of the conditioning pulse, resulting in a profound reduction of the available current with more depolarizing conditioning potentials (Fig. 5B2; n ¼ 6). Remarkably, preincubation of SNC80 (50 lM) significantly augments this voltage-dependent reduction of available current (n ¼ 9, Fig. 5B2, p < 0:05, asterisks), displaying the most prominent effects at more depolarized conditioning potentials. 3.6. Selectivity of SNC80 mediated effects
Fig. 3. Fast recovery from inactivation is not effected by the d-agonist SNC80. (A) Fast recovery from inactivation was studied using a double pulse protocol as depicted in the inset. Representative families of current traces during the recovery period are shown for control conditions (ACSF), after application of 50 lM SNC80 and after washout of SNC80 (wash), (B) The current amplitude during recovery from inactivation from membrane potentials of either 70 mV (closed squares: ACSF, open squares: 50 lM SNC80) or 80 mV (closed triangles: ACSF, open squares: 50 lM SNC80) was normalized to Imax at the end of each series and plotted versus the duration of the interpulse interval. The time course of recovery was determined by a biexponential fit to the data points of each individual recording (Eq. (3)) and the best fit to the data is represented by the solid lines.
76.5% for prepulses of 1, 3, 10, 30, 100 and 300 s, respectively. We next asked the question whether the effects of SNC80 on slow inactivation of INa were voltage-dependent. To test this idea, we applied 5 s conditioning pulses from 100 mV to different conditioning potentials ranging from 100 to +20 mV and monitored the fraction of INa recovered from slow inactivation after 1 s
The major effects of SNC80 were a block of the maximal INa conductance and a pronounced effect on slow inactivation processes. We have tested whether these two effects of the putative d-opioid agonist SNC80 were mediated by d-opioid receptors. As a first experiment, we tested whether a structurally distinct dopioid agonist mimics the effects of SNC80. In contrast to the pronounced effects of SNC80, incubation with the selective peptide agonist DPDPE (10 lM) did not alter slow inactivation induced by 10 and 100 s prepulses (Fig. 6A, n ¼ 4). Next, we tested whether the effects of SNC80 on slow inactivation processes are inhibited by opioid antagonists. Preincubation of granule cells with different concentrations of the unselective opioid antagonist naloxone (at least 15 min, 50–300 lM) did not alter the fraction of INa that had undergone slow inactivation after prepulses of 10 and 100 s duration (Fig. 6B, closed and open bars). As a control experiment, we tested whether the highest concentration of naloxone (300 lM) applied alone has effects on slow inactivation of Na+ channels, and found this not to be the case (Fig. 6B). Likewise, the more selective d-opioid antagonist naltrindole (10 and 100 lM) did not alter the effects of SNC80 on slow inactivation. Thus, the efficacy of SNC80 in modulating slow inactivation was maintained even in the presence of high concentrations of different d-opioid antagonists. Collectively, these results indicate that the effects of SNC80 on INa are not mediated by d-opioid receptors.
4. Discussion It is well established that d-opioid receptors modulate multiple types of ion channels via G-proteins, in particular inwardly rectifying K+ channels and voltagegated Ca++ channels (Christie et al., 2000). In this study, we have therefore investigated the effects of SNC80, a putative selective d-opioid receptor agonist, on voltage-gated Na+ channels in native hippocampal neurons. We have demonstrated that SNC80 selectively prolongs the time course of recovery from slow inacti-
1108
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
Fig. 4. The number of available Na+ channels after slow recovery from inactivation is strongly reduced after application of 50 lM SNC80. (A) Time course of Na+ channel recovery from slow inactivation recorded under control conditions (A1) and after drug application (A2). Inactivation was induced by prepulses to 10 mV ranging from 1 to 300 s. The current amplitude during recovery was normalized to Imax, yielding the fraction of available Na+ channels. The normalized values are plotted versus the recovery time for the indicated prepulse durations (n ¼ 6 10). Insets: Representative families of original traces during recovery from inactivation induced by a 100 s prepulse in control solution (A1) and after drug application (A2). Recovery from inactivation was fitted with a biexponential function (Eq. (3)) and best fits are shown as solid lines (A1–2), (B) The resulting time constants (sslow1 and sslow2, panels B3 and 4) and associated amplitudes (Aslow1 and Aslow2, panels B1 and 2) were plotted as function of the prepulse duration. The relationship between the prepulse duration (t) and the recovery time constants (sslow1, sslow2) was fitted to a power law function tðtÞ ¼ aðt=aÞb, where a is the constant kinetic setpoint and b the scaling power(Eq. (4), B3–4) and best fits are shown as solid lines.
vation without effects on fast inactivation processes. In addition, SNC80 reduced the maximal Na+ current amplitude. Surprisingly, we found that these powerful effects were not mediated by activation of d-opioid receptors. 4.1. Effects of SNC80 on slow but not fast inactivation The first interesting aspect of this study is the selectivity of the effects of SNC80 for slow but not fast inactivation processes. Firstly, SNC80 appeared to facilitate inactivation by prolonged depolarization. For
example, the fraction of INa available following inactivation by 300 s depolarizing prepulses was reduced by ~80% following application of SNC80 (see Fig. 5A). In addition, SNC80 potently prolonged the time course of recovery from slow inactivation induced by prolonged depolarization, without affecting recovery from fast inactivation induced by brief depolarizations. The finding further supports the idea that different channel domains might subserve fast and slow inactivation. Fast inactivation in Na+ channels occurs by a ‘‘balland-chain’’ or ‘‘hinged lid’’ mechanism, and is thought to be due to a conformational change of the
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
1109
Fig. 5. SNC80 considerably reduces the fraction of available Na+ channels. (A1) Representative traces of Na+ currents recorded 1 s after depolarizing conditioning pulses to 10 mV of 10, 100 or 300 s duration (open circles) and following complete recovery from slow inactivation (asterisks) in control conditions and in the presence of 50 lM SNC80. (A2) The fraction of Na+ channels available 1 s after a depolarizing conditioning pulse is plotted as function of the prepulse duration under control conditions (closed squares) and in the presence of 50 lM SNC80 (open squares). (B1) Representative traces of Na+ currents recorded 1 s after depolarizing conditioning pulses of 5 s duration to 100, 40 and +20 mV (open circles) and following complete recovery from slow inactivation (asterisks) under control conditions and in the presence of 50 lM SNC80 (holding potential: 100 mV). (B2) The fraction of Na+ channels available 1 s after a depolarizing conditioning pulse is plotted as function of the prepulse potential under control conditions (closed triangles) and in the presence of 50 lM SNC80 (open triangles). Note the strong decrease in availability of Na+ channels in the presence of SNC80 at more depolarized prepulse potentials.
‘‘inactivation particle’’, formed by three hydrophobic amino acids (Ile-Phe-Met) in the interdomain linker III–IV (Stu¨hmer et al., 1989; West et al., 1992). The structural basis of slow inactivation is less clear. A number of studies have demonstrated that mutations in the SCN4A and SCN5A genes, causing structural alterations in the transmembrane segments IV/S4, IV/ S6, II/S1, II/S5, and the loops between IV=S4 þ S5
and II=S5 þ S6 of the NaV1.4 and NaV1.5 a subunits, have strong effects on slow inactivation (Cummins and Sigworth, 1996; Alekov et al., 2001; Mitrovic et al., 2000; Hayward et al., 1997; Melamed-Frank and Marom, 1999; for review, see Vilin and Ruben, 2001). Some of these mutations have been suggested to underlie disease phenotypes such as hyperkalemic periodic paralysis and related myotonic disorders (for review,
1110
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
Fig. 6. SNC80-mediated inhibition of the Na+ current is not mediated by d-opioid receptors. (A) Effects of the peptide d-receptor agonist DPDPE (10 lM) on slow inactivation. The fraction of Na+ channels available after conditioning pulses of 10 s (closed bars) and 100 s (open bars) are depicted. DPDPE did not mimic the effects of SNC80, (B) Neither preincubation of naloxone at the concentrations indicated (50, 100 and 300 lM), nor preincubation with the d-selective antagonist naltrindole (10 and 100 lM) altered the SNC80-mediated reduction in the fraction of Na+ currents available after conditioning pulses of 10 s (closed bars) and 100 s (open bars). Naloxone alone (300 lM) did not alter slow inactivation. The number of experiments is indicated by the numbers in brackets.
see Lehmann-Horn and Jurkat-Rott, 1999). Site-directed mutagenesis studies have revealed that slow inactivation may result from a structural rearrangement of the outer channel pore (Xiong et al., 2003), but another study has failed to reveal effects of pore mutations on slow inactivation (Struyk and Cannon, 2002). Therefore, the structural basis of slow inactivation and recovery may be complex and requires further experimental evaluation. In any case, however, the determinants of fast and slow inactivation seem to be distinct from each other (Rudy, 1978; Cummins and Sigworth, 1996), a finding supported by the selective effects of SNC80 in this study. As described in previous studies (Marom, 1998; Ellerkmann et al., 2001), we observed an increase in the time constants of slow recovery with more prolonged conditioning depolarizations. Such an increase can be explained by assuming a model of Na+ channel inactivation with a number of interlinked slow inactivated states (I1–In): A I1 I2 I3 I4 . . . In where A is the available state. After membrane depolarization, Na+ channels undergo slow inactivation, with more and more channels entering remote slow
inactivated states during more prolonged conditioning depolarization. In terms of the model described above, a change in scaling power as seen for sslow2 might arise because entry into remote inactivated states is facilitated. Such a mechanism is consistent with the increased entry into slow inactivation that we observed (compare Fig. 4 panel A1 and B1, Fig. 5A). Alternatively, it is conceivable that SNC80 binds with a higher affinity to channels that have undergone slow inactivation. Regardless of the mechanism, the effects of SNC80 cause recovery from inactivation to be significantly slower for each given duration of the inactivating depolarization. In addition, the dramatic change in scaling power of sslow2 strongly augments the steepness of the relation between the duration of the conditioning depolarization and the slow recovery rate. This, in turn, implies that the prior history of membrane depolarization is encoded more efficiently in terms of the subsequent availability of Na+ channels (Toib et al., 1998; Ellerkmann et al., 2001). In addition to effects on slow inactivation processes, we also observed a potent reduction of Na+ current amplitude following application of SNC80. This Na+ current amplitude reduction appeared not to be due to
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
effects of SNC80 on fast voltage-dependent activation and inactivation behavior. Similar findings have also been described for the j-opioid agonist U50,488 (Zou et al., 2000). 4.2. Effects of SNC80 are not mediated by opioid receptors Interestingly, both the effects of SNC80 on slow inactivation processes and on Na+ current amplitude appeared not to require activation of opioid receptors. The potent and selective d-opioid receptor agonist DPDPE did not mimic the effects of SNC80. Conversely, in the presence of high concentrations of the opioid receptor antagonists naloxone and naltrindole, the effects of SNC80 on slow inactivation proved to be undiminished. Opiates and endogenous opioid peptides act through multiple receptors, classified as l-, d- and j-opioid receptors (Massotte and Kieffer, 1998; Kieffer, 1999; Kieffer and Gaveriaux-Ruff, 2002). Because of the strong analgesic action of opioid receptor activation (Hayes and Vogelsang, 1991), a large number of selective receptor agonists have been developed for clinical use. More recently, it has been appreciated that certain putative opioid receptor agonists also exert opioid receptor independent effects on membrane currents, in particular on Na+ currents. For instance, the j-opioid receptor agonist U50,488 was able to attenuate behavioral and primary afferent nerve responses to noxious colorectal distension in vivo. This effect could be mimicked by Na+ channel-blockers and in vitro experiments on colon sensory neurons show an inhibition of voltage-activated Na+ currents in an opioid receptor independent manner (Su et al., 2002). Similarly, an inhibitory effect of U50,488 on voltage-activated Na+ currents was described in hippocampal CA3 neurons. Again, the decrease of the Na+ current amplitude was not related to activation of opioid receptors (Zou et al., 2000). Likewise, a concentration-dependent reduction in peak transient Na+ current by the l-opioid receptor agonist morphine on cardiac myocytes was identified (Hung et al., 1998). In another study, effects of naltrindole, a highly specific d-opioid receptor antagonist, which has been shown to inhibit graft rejection in vivo and suppress allogenic lymphocyte reaction in vitro similarly to cyclosporin A, could not be blocked in dreceptor or triple opioid receptor knockout mice (Gaveriaux-Ruff et al., 2001). Taken together, these data suggest that putative selective opioid receptor agonists show highly specific effects on voltage-gated ion channels. In particular, the effects on slow inactivation may be relevant, given that a number of Na+ channel mutations that give rise to human diseases alter slow inactivation of these channels. In the future, it will be important to analyze such
1111
selective mechanisms of action of putative opioid receptor agonists, in particular, those intended for clinical use. In addition, the highly selective action of the opioid receptor agonist tested in this study on biophysical Na+ channel characteristics may help to extend our understanding of the biophysical mechanisms underlying slow and fast inactivation of these channels.
Acknowledgements This research was supported by the graduate program of the Deutsche Forschungsgemeinschaft ‘‘Pathogenese von Krankheiten des Nervensystems’’.
References Alekov, A.K., Peter, W., Mitrovic, N., Lehmann-Horn, F., Lerche, H., 2001. Two mutations in the IV/S4-S5 segment of the human skeletal muscle Na+ channel disrupt fast and enhance slow inactivation. Neurosci. Lett. 306, 173–176. Bezanilla, F., Armstrong, C.M., 1977. Inactivation of the sodium channel. I. Sodium current experiments. J. Gen. Physiol. 70, 549–566. Bilsky, E.J., Calderon, S.N., Wang, T., Bernstein, R.N., Davis, P., Hruby, V.J., McNutt, R.W., Rothman, R.B., Rice, K.C., Porreca, F., 1995. SNC 80, a selective, nonpeptidic and systemically active opioid delta agonist. J. Pharmacol. Exp. Ther. 273, 359–366. Calderon, S.N., Rice, K.C., Rothman, R.B., Porreca, F., FlippenAnderson, J.L., Kayakiri, H., Xu, H., Becketts, K., Smith, L.E., Bilsky, E.J., Davis, P., Horvath, R., 1997. Probes for narcotic receptor mediated phenomena. 23. Synthesis, opioid receptor binding, and bioassay of the highly selective delta agonist (+)-4[(alpha R)-alpha-((2S,5R)-4-allyl-2,5-dimethyl-1-piperazinyl)-3methoxybenzyl]-N,N-diethylbenzamide (SNC 80) and related novel nonpeptide delta opioid receptor ligands. J. Med. Chem. 40, 695–704. Catterall, W.A., 1992. Cellular and molecular biology of voltagegated sodium channels. Physiol. Rev. 72 (Suppl.), S15–S48. Christie, M.J., Connor, M., Vaughan, C.W., Ingram, S.L., Bagley, E.E., 2000. Cellular actions of opioids and other analgesics: implications for synergism in pain relief. Clin. Exp. Pharmacol. Physiol. 27, 520–523. Colbert, C.M., Magee, J.C., Hoffman, D.A., Johnston, D., 1997. Slow recovery from inactivation of Na+ channels underlies the activity-dependent attenuation of dendritic action potentials in hippocampal CA1 pyramidal neurons. J. Neurosci. 17, 6512–6521. Cummins, T.R., Sigworth, F.J., 1996. Impaired slow inactivation in mutant sodium channels. Biophys. J. 71, 227–236. Ellerkmann, R.K., Riazanski, V., Elger, C.E., Urban, B.W., Beck, H., 2001. Slow recovery from inactivation regulates the availability of voltage-dependent Na+ channels in hippocampal granule cells, hilar neurons and basket cells. J. Physiol. 532, 385– 397. Fleidervish, I.A., Gutnick, M.J., 1996. Kinetics of slow inactivation of persistent sodium current in layer V neurons of mouse neocortical slices. J. Neurophysiol. 76, 2125–2130. Fleidervish, I.A., Friedman, A., Gutnick, M.J., 1996. Slow inactivation of Na+ current and slow cumulative spike adaptation in
1112
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
mouse and guinea-pig neocortical neurones in slices J. Physiol. 493 (Pt. 1), 83–97. Gaveriaux-Ruff, C., Filliol, D., Simonin, F., Matthes, H.W., Kieffer, B.L., 2001. Immunosuppression by delta-opioid antagonist naltrindole: delta- and triple mu/delta/kappa-opioid receptor knockout mice reveal a nonopioid activity. J. Pharmacol. Exp. Ther. 298, 1193–1198. Hayes, S.R., Vogelsang, J., 1991. Opiate receptors and analgesia: an update. J. Post Anesth. Nurs. 6, 125–128. Hayward, L.J., Brown, R.H., Cannon, S.C., 1997. Slow inactivation differs among mutant Na+ channels associated with myotonia and periodic paralysis. Biophys. J. 72, 1204–1219. Hung, C.F., Tsai, C.H., Su, M.J., 1998. Opioid receptor independent effects of morphine on membrane currents in single cardiac myocytes. Br. J. Anaesth. 81, 925–931. Kieffer, B.L., 1999. Opioids: first lessons from knockout mice. Trends Pharmacol. Sci. 20, 19–26. Kieffer, B.L., Gaveriaux-Ruff, C., 2002. Exploring the opioid system by gene knockout. Prog. Neurobiol. 66, 285–306. Lehmann-Horn, F., Jurkat-Rott, K., 1999. Voltage-gated ion channels and hereditary disease. Physiol. Rev. 79, 1317–1372. Marom, S., 1998. Slow changes in the availability of voltage-gated ion channels: effects on the dynamics of excitable membranes. J. Membr. Biol. 161, 105–113. Martina, M., Jonas, P., 1997. Functional differences in Na+ channel gating between fast-spiking interneurones and principal neurones of rat hippocampus. J. Physiol. 505, 593–603. Massotte, D., Kieffer, B.L., 1998. A molecular basis for opiate action. Essays Biochem. 33, 65–77. Melamed-Frank, M., Marom, S., 1999. A global defect in scaling relationship between electrical activity and availability of muscle sodium channels in hyperkalemic periodic paralysis. Pflugers Arch. 438, 213–217. Mitrovic, N., George, Jr., A.L., Horn, R., 2000. Role of domain 4 in sodium channel slow inactivation. J. Gen. Physiol. 115, 707–718. Ragsdale, D.S., Avoli, M., 1998. Sodium channels as molecular targets for antiepileptic drugs. Brain Res. Brain Res. Rev. 26, 16–28. Reckziegel, G., Beck, H., Schramm, J., Elger, C.E., Urban, B.W., 1998. Electrophysiological characterization of Na+ currents in
acutely isolated human hippocampal dentate granule cells. J. Physiol. 509 (Pt. 1), 139–150. Remy, S., Urban, B.W., Elger, C.E., Beck, H., 2003. Anticonvulsant pharmacology of voltage-gated sodium channels in hippocampal neurons of control and chronically epileptic rats. Eur. J. Neurosci. 17, 2648–2658. Ruben, P.C., Starkus, J.G., Rayner, M.D., 1992. Steady-state availability of sodium channels. Interactions between activation and slow inactivation. Biophys. J. 61, 941–955. Rudy, B., 1978. Slow inactivation of the sodium conductance in squid giant axons. Pronase resistance. J. Physiol. 283, 1–21. Struyk, A.F., Cannon, S.C., 2002. Slow inactivation does not block the aqueous accessibility to the outer pore of voltage-gated Na+ channels. J. Gen. Physiol. 120, 509–516. Stu¨hmer, W., Conti, F., Suzuki, H., Wang, X., Noda, M., Yahagi, N., Kubo, H., Numa, S., 1989. Structural parts involved in activation and inactivation of the sodium channel. Nature 339, 597–603. Su, X., Joshi, S.K., Kardos, S., Gebhart, G.F., 2002. Sodium channel blocking actions of the kappa-opioid receptor agonist U50,488 contribute to its visceral antinociceptive effects. J. Neurophysiol. 87, 1271–1279. Toib, A., Lyakhov, V., Marom, S., 1998. Interaction between duration of activity and time course of recovery from slow inactivation in mammalian brain Na+ channels. J. Neurosci. 18, 1893–1903. Vilin, Y.V., Ruben, P.C., 2001. Slow inactivation in voltage-gated sodium channels: molecular substrates and contributions to channelopathies. Cell Biochem. Biophys. 35, 171–190. West, J.W., Patton, D.E., Scheuer, T., Wang, Y., Goldin, A.L., Catterall, W.A., 1992. A cluster of hydrophobic amino acid residues required for fast Na(+)- channel inactivation. Proc. Natl. Acad. Sci. USA 89, 10910–10914. Xiong, W., Li, R.A., Tian, Y., Tomaselli, G.F., 2003. Molecular motions of the outer ring of charge of the sodium channel: do they couple to slow inactivation? J. Gen. Physiol. 122, 323–332. Zou, B., Chen, Y., Wu, C., Zhou, P., 2000. Blockade of U50488H on sodium currents in acutely isolated mice hippocampal CA3 pyramidal neurons. Brain Res. 855, 132–136.
Modulation of voltage-dependent sodium channels by the d-agonist SNC80 in acutely isolated rat hippocampal neurons Christina Remy a, Stefan Remy b, Heinz Beck b, Dieter Swandulla a, Michael Hans a, a b
Institute of Physiology, University of Bonn, Wilhelmstr. 31, 53111 Bonn, Germany Department of Epileptology, University of Bonn Medical Center, Bonn, Germany
Received 20 November 2002; received in revised form 10 May 2004; accepted 18 June 2004
Abstract Following activation, voltage-gated Na+ currents (INa) inactivate on two different time scales: fast inactivation takes place on a time scale of milliseconds, while slow inactivation takes place on a time scale of seconds to minutes. Both fast and slow inactivation processes govern availability of Na+ channels. In this study, the effects of the d-opioid receptor agonist SNC80 on slow and fast inactivation of INa in rat hippocampal granule cells were analyzed in detail. Following application of SNC80, a block of the peak Na+ current amplitude (EC50: 50.6 lM, Hill coefficient: 0.518) was observed. Intriguingly, SNC80 (50 lM) also caused a selective effect on slow but not fast inactivation processes, with a notable increase in the fraction of Na+ channels undergoing slow inactivation during prolonged depolarization. In addition, recovery from slow inactivation was considerably slowed. At the same time, fast recovery processes were unaffected. The effects of SNC80 were not mimicked by the peptide d-receptor agonist DPDPE (10 lM), and were not inhibited by the opioid receptor antagonists naloxone (50–300 lM) or naltrindole (10 and 100 lM), indicating an opioid receptor independent modulation of Na+ channels. These data suggest that SNC80 not only affects d-opioid receptors, but also voltage-gated Na+ channels. SNC80 is to our knowledge hitherto the only substance that selectively influences slow but not fast inactivation processes and could provide an important tool in unraveling the mechanism underlying these distinct biophysical processes. # 2004 Elsevier Ltd. All rights reserved. Keywords: Sodium channel; SCN80; Slow inactivation; Opioids; Hippocampus
1. Introduction Voltage-dependent Na+ channels are ubiquitous membrane proteins in excitable cells. They are transmembrane proteins composed of one of several poreforming (a) subunits and accessory b-subunits. At resting membrane potential Na+ channels are closed, but are rapidly activated by depolarization, giving rise to Na+ inward currents (see for review Catterall, 1992). These Na+ currents decrease rapidly towards baseline levels as Na+ channels undergo inactivation during prolonged depolarization. Following inactivation, Na+ channels require repolarization in order to return to Corresponding author. Tel.: +49-228-287-2326; fax: +49-228-2872313. E-mail address: [email protected] (M. Hans).
0028-3908/$ - see front matter # 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropharm.2004.06.034
the resting state. Na+ channels in neuronal membranes are able to cycle through all these states on a millisecond time scale and are therefore able to mediate rapid events such as action potentials. The structural determinants of Na+ channels necessary for rapid inactivation processes have been extensively investigated (Stu¨hmer et al., 1989; West et al., 1992). In addition to fast inactivation processes, Na+ channels may also show slow inactivation and recovery from inactivation. These slow processes can modulate the availability of Na+ channels on a time scale of seconds to minutes and have been suggested to serve as molecular mechanism that preserves traces of previous activity (Toib et al., 1998). As the availability of Na+ channels crucially modulates action potential threshold, waveform, and the propensity to generate repetitive discharges, slow inactivation is also of considerable
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
interest. Interestingly, rate constants for slow recovery vary widely in the published literature (Ruben et al., 1992; Fleidervish and Gutnick, 1996; Fleidervish et al., 1996; Colbert et al., 1997). A large part of this variation may arise from the phenomenon that recovery rates depend on the duration of the depolarization used to inactivate Na+ channels, with more prolonged depolarization resulting in slower recovery rates (Toib et al., 1998; Ellerkmann et al., 2001). The rate constants of recovery are quantitatively related to the duration of prior depolarization by a power law relationship (Toib et al., 1998; Ellerkmann et al., 2001). There is good agreement that the structural correlate of fast Na+ channel inactivation is a conformational change of the interdomain linker III–IV (Stu¨hmer et al., 1989; West et al., 1992). In contrast, the mutations known to affect slow inactivation neither share a common localization nor point at the involvement of a particular region of the Na+ channel protein (Cummins and Sigworth, 1996; Alekov et al., 2001; Mitrovic et al., 2000; Hayward et al., 1997; MelamedFrank and Marom, 1999; Vilin and Ruben, 2001). Thus, the structural basis of slow inactivation and recovery may be complex and has not been fully resolved. Recovery from slow inactivation seems to be remarkably conserved across different types of neurons and species (Ellerkmann et al., 2001), whereas this is not the case for fast inactivation processes (Martina and Jonas, 1997). There are also pharmacological differences between fast recovery and slow recovery from inactivation. Anticonvulsants such as carbamazepine, lamotrigine and phenytoin strongly affect fast recovery from inactivation (Remy et al., 2003; Ragsdale and Avoli, 1998), but do not appear to interact with slow recovery processes. It would be of considerable interest to identify substances that do the converse, i.e. interact strongly with slow but not fast recovery processes. This may be particularly relevant since naturally occurring mutations in Na+ channel genes that affect slow inactivation processes are associated with several diseases such as hyperkaliemic periodic paralysis, myotonia, idiopathic ventricular fibrillation and long-QTsyndrome (for review, see Vilin and Ruben, 2001). In this study, we have investigated the effects of the putative selective d-opioid receptor agonist SNC80 (Bilsky et al., 1995; Calderon et al., 1997) on voltagedependent Na+ channels in hippocampal granule neurons. These effects did not require activation of opioid receptors, but were presumably due to a direct interaction with membrane components. We have examined both fast and slow inactivation processes in detail. Our results indicate that application of SNC80 inhibits INa amplitude and selectively prolongs the time course for recovery from slow inactivation without affecting other kinetic parameters.
1103
2. Methods 2.1. Preparation of dissociated dentate granule cells We prepared coronal slices (400 lm) from the hippocampus of male Wistar rats (40 5 day-old) with a vibratome (LeicaVT1000S) in a bath filled with ice-cold bicarbonate-buffered saline (ACSF) containing (in mM): NaCl 125, NaHCO3 25, KCl 3, NaH2PO4 1.25, MgCl2 1, CaCl2 2, and glucose 20 (pH was adjusted to 7.4 by gassing the solution with carbogen (95% O2, 5% CO2), osmolarity 300 mosmol l1). Prior to decapitation, rats were deeply anesthetized with ether. Slices were transferred to a storage chamber containing v ACSF at room temperature (21–23 C) and subsequently stored for up to 10 h. After an equilibration period of 60 min, the first slice was transferred to a tube with 5 ml solution containing (in mM): CH3SO3Na 145, KCl 3, MgCl2 1, CaCl2 0.5, HEPES 10, and glucose 15 (pH 7.4, adjusted with NaOH, osmolarity 310 mosmol l1). Pronase (protease type XIV, 2 mg/ml; Sigma) was added to the oxygenated medium. After an incubation period of 15 min at v 35 C, the slice was washed in the pronase-free solution of identical composition. The dentate gyrus was dissected and triturated with fire-polished Pasteur pipettes of decreasing apertures. The hilus was separated from the slice during trituration. The Petri dish containing the cell suspension was then mounted on the stage of an inverted microscope (Telaval, Zeiss, Jena, Germany). Isolated cells were allowed to settle for 10 min and subsequently superfused with an extracellular solution containing (in mM): CH3SO3Na 120; TEA 20; KCl 3; BaCl2 5; MgCl2 1; HEPES 10; 4-aminopyridine (4-AP) 4; CdCl2 0.03 and glucose 10 (pH 7.4, adjusted with NaOH; osmolarity 310 mosmol ll). Only those dissociated cells that showed a small round or ovoid soma (capacitance 6 2 pF) with a single process reminiscent of dentate granule cell (DGC) morphology in situ were included in this study. 2.2. Whole cell patch-clamp recording Patch pipettes (resistance 3 1 MX) fabricated from borosilicate glass capillaries (1.5 mm o.d., 1 mm i.d.; Science Products, Hofheim, Germany) on a Narishige P83 puller were filled with an intracellular solution containing (in mM): Cs-methanesulfonate 90; TEA 20; MgCl2 5; HEPES 10; BAPTA 5; CaCl2 0.5; adenosine50 -triphosphate (50 ATP-Na2) 10 and guanosine-50 -triphosphate (50 GTP-Na2) 0.5 (pH 7.4 with NaOH, 300 mosmol l1). Recordings were conducted in the tightseal whole cell configuration at room temperature using an Axopatch 200 A amplifier (Axon Instruments, Union City, CA, seal resistance >1 GX, average series resistance 4 2 MX). Series resistance was compen-
1104
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
sated between 70% and 90%, resulting in maximal residual voltage error below 1 mV (0:78 0:13 mV, n ¼ 5 for activation and 0:69 0:09 mV, n ¼ 8 for inactivation). Current signals were filtered at 5 kHz (3 dB, 4-pole low-pass Bessel filter) and sampled at least with 10 kHz (DigiData 1322A, Axon Instruments, Union City, CA). Linear leak and residual capacitance currents were digitally subtracted using a P/4 protocol (Bezanilla and Armstrong, 1977). INa properties were analyzed as described in detail elsewhere (Reckziegel et al., 1998). SNC80 was dissolved in 10% HCl and v stored at 20 C as 100 mM stock solution. Chemicals were obtained either from Tocris (SNC80, Naloxone hydrochloride) or Sigma, Germany (all others). 2.3. Voltage clamp protocols and fitting The voltage-dependence of activation and inactivation was determined using standard protocols. The conductance G(V) was calculated according to: GðV Þ ¼ Imax =ðV VNa Þ;
ð1Þ
where VNa is the reversal potential, V the command potential, and Imax the peak current amplitude. G(V) was then fitted by the following Boltzmann equation: GðV Þ ¼ Gmax =ð1 þ expððV50 V Þ=km Þ
ð2Þ
where Gmax is the maximum Na+ conductance, V50 is the voltage where G(V) is half of Gmax, and km indicates the slope of the relationship between channel inactivation and membrane voltage. Double pulse experiments were performed with a conditioning pulse at 10 mV followed by varying recovery periods from 1 to 1000 ms at either 70 or 80 mV and a subsequent testpulse at 10 mV for 10 ms. Time constants of recovery sfast and sintermediate for fast inactivation processes and sslow1 and sslow2 for slow inactivation processes were extracted from the equation: IðtÞ ¼ A0 þ A1 ð1 expð t=s1 ÞÞ þ A2 ð1 expð t=s2 ÞÞ
ð3Þ
where I(t) is the current amplitude at the time point t after onset of the voltage command and A is the amplitude contribution of the different recovery time constants. Slow recovery processes were investigated using a conditioning pulse to 10 mV of various durations (1, 3, 10, 30, 100, 300 s) followed by a 1 s recovery interval at 80 mV. This interval allowed for a virtually complete recovery of the channels from rapid inactivation and was followed by a series of brief (5 ms) testpulses to 10 mV at a frequency of 0.33 Hz aimed at monitoring the process of recovery from slow inactivation. When the current amplitude reached steady state values, the pulse protocol was terminated
and the current amplitude was normalized to these values (Imax). The relationship between the recovery time constants sslow and the duration of the preceding depolarization (t) was fitted by a power law function of the following equation: sðtÞ ¼ aðt=aÞb
ð4Þ
where a is the constant kinetic setpoint [s] and b the scaling power. Concentration–response curves were fitted with a Hill equation of the form: E cc ¼ y Emax k þ cc
ð5Þ
where c is the concentration of the used substance, Emax the maximal possible effect, k the concentration at which a half-maximal effect was obtained and E the effect evoked by the concentration c. c is the Hill coefficient. Statistical comparison was carried out with a twotailed Student’s t-test. All results are shown as the mean S:E:M: All experiments were conducted in accordance with the guidelines of the University of Bonn Animal Care Committee and were approved by the board of proper use of experimental animals of the state Nordrhein-Westfalen.
3. Results 3.1. SNC80 reduces INa amplitude We first examined the effects of SNC80 on the maximal INa amplitude. Maximal Na+ currents under control conditions were elicited by depolarizing pulses to 10 mV following a hyperpolarizing prepulse to remove fast inactivation (Fig. 1A, inset). These currents were reversibly blocked by bath application of 0.5 lM TTX (not shown, n ¼ 5). TTX-insensitive current components were not observed. Application of the putative selective d-agonist SNC80 potently inhibited INa in a dose-dependent manner (Fig. 1B). Following application of SNC80, the inhibitory effects occurred within a few seconds, saturated within 30 s and recovered following washout. The concentration-dependence of this effect could be fitted with a Hill curve shown superimposed on the data points in Fig. 1B (EC50: 50.6 lM, Hill coefficient: 0.518). 3.2. Effect of SNC80 on the voltage-dependent activation and fast inactivation of INa To determine whether the pronounced reduction in INa amplitude might be due to a change in voltagedependent activation or inactivation, we investigated
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
1105
1:6 mV, k ¼ 5:8 0:3, n ¼ 8). At lower concentration of SNC80 (1 lM), the shift in the voltage-dependence of inactivation was less pronounced (1.9 mV) but significantly different from control (Fig. 2C, left column, n ¼ 7, p < 0:05). 3.3. SNC80 does not alter recovery from fast inactivation Recovery from fast inactivation was analyzed at membrane potentials of 70 and 80 mV using a double pulse protocol (Fig. 3A, inset). The time course of recovery from fast inactivation was in both cases best fit with the sum of two exponential functions (Eq. (3)), with two time constants sfast and sintermediate. The recovery rates did not differ significantly at the two different holding potentials (Fig. 3B; 80 mV: sfast ¼ 8:9 1:2 ms, sintermediate ¼ 234 37 ms, n ¼ 11, 70 mV: sfast ¼ 8:2 0:8 ms, sintermediate ¼ 249 107 ms, n ¼ 6). In the presence of SNC80 (50 lM), the time course of recovery from fast inactivation was not significantly affected at both holding potentials (80 mV: sfast ¼ 9:1 0:7 ms, sintermediate ¼ 292 65 ms, n ¼ 11, 70 mV: sfast ¼ 8:3 0:7 ms, sintermediate ¼ 234 51 ms, n ¼ 6) Fig. 1. SNC80 inhibits Na+ currents in dentate granule cells in a dose-dependent manner. (A) Representative recording of a Na+ current elicited by a 5 ms testpulse to 10 mV from a 100 ms prepulse potential of 100 mV under control conditions (ACSF superfusion, solid line), during application of 1 lM (dashed line) and 100 lM SNC80 (dotted lines) and following washout. The voltage protocol is depicted in the inset. (B) Concentration–response relationship for the blockade of the Na+ current by SNC80. The percentage of the SNC80-mediated block of the Na+ current amplitude under ACSF superfusion is plotted versus the concentration of SNC80. Fitting the data points with a Hill equation (see Methods) yielded the following parameters: EC50 ¼ 50:6 lM, k ¼ 0:518. The Hill curve is shown superimposed on the data points (solid line).
these properties under control conditions and following application of 100 lM SNC80. The properties of voltage-dependent activation and inactivation were determined using standard voltage protocols (Fig. 2A, insets) and the conductance/amplitude–voltage relationship fitted with a Boltzmann equation shown superimposed on the data points in Fig. 2B. Application of 100 lM SNC80 (open triangles, Fig. 2B) did not significantly alter the voltage-dependence of activation (V1=2Act : 29:0 3:5 mV, k ¼ 6:3 0:7, n ¼ 5) compared to control conditions (closed triangles, Fig. 2B, V1=2Act : 30:3 1:7 mV, k ¼ 5:5 0:5, n ¼ 5). In contrast, a small but significant shift (4 mV) of the voltage-dependence of inactivation in a hyperpolarizing direction was observed in SNC80-treated cells (open triangles, Fig. 2B, V1=2Inact ¼ 58:1 2:3 mV, k ¼ 6:4 0:4, n ¼ 8, p < 0:01) compared to control conditions (closed triangles, Fig. 2B, V1=2Inact ¼ 54:1
3.4. SNC80 slows the time course of recovery from slow inactivation We then investigated the effects of SNC80 on slow recovery from inactivation. Recovery rates of INa have been previously demonstrated to be dependent on the duration of prior depolarization used to elicit inactivation (Toib et al., 1998; Ellerkmann et al., 2001). Accordingly, we have examined the rates of slow recovery from inactivation following depolarizing pulses of different durations ranging from 1 to 300 s (Fig. 4A, inset). We examined one population of cells under control conditions (Fig. 4A1) and compared it to a second population of cells preincubated with 50 lM SNC80 for 15 min (Fig. 4A2). After an interval (1 s) to allow virtually complete recovery from fast inactivation (see Fig. 3B), the slow recovery of INa was monitored as described previously using brief (5 ms) depolarizing pulses at 0.33 Hz . The time course of recovery from slow inactivation (Figs. 4A1 and A2) was best fitted by a biexponential equation with two time constants sslow1 and sslow2. In addition, both sslow1 and sslow2 were related to the duration of the conditioning pulse by a power law relation (Eq. (4)) (Toib et al., 1998; Ellerkmann et al., 2001). SNC80 (50 lM) dramatically slowed the time course of recovery from slow inactivation (open triangles Figs. 4B1–4), with significant effects both on sslow1 and sslow2 (Figs. 4B3 and 4, control: n ¼ 6 7, SNC80: n ¼ 6 10, p < 0:05, asterisks). We also analyzed the amplitudes associated with sslow1 and sslow2,
1106
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
Fig. 2. Effects of the d-agonist SNC80 on the activation and fast inactivation kinetics of the Na+ channels in dentate granule cells. (A) Families of representative recordings elicited by the voltage protocols shown in the insets. Voltage steps to different testpulse potentials from 60 to +60 mV from a 100 ms prepulse potential of 100 mV were used to characterize the activation behavior of the Na+ channel (upper panel). The voltage-dependent inactivation of the Na+ channels was analyzed by eliciting a Na+ current with a 15 ms testpulse to 10 mV after varying prepulse potentials from 100 to 20 mV (lower panel), (B) Activation and inactivation curves of the Na+ current under control conditions (closed triangles) and during application of 100 lM SNC80 (open triangles). The average values of normalized conductances were plotted versus either the prepulse (inactivation) or the testpulse potential (activation). For each individual recording, a Boltzmann function (Eq. (2)) was fitted to the data points. The solid lines superimposed on the data points show Boltzmann curves that were constructed with values for half-maximal activation (V1/2Act), inactivation (V1/2Inact) and the slope factor, (C) SNC80 significantly shifts V1/2Inact into the hyperpolarizing direction at concentrations of 1 lM (n ¼ 7, p < 0:05) and 100 lM (n ¼ 8, p < 0:01).
termed Aslow1 and Aslow2. The contribution of the fast component, Aslow1, was not altered by SNC80 application over a range of prepulse durations (Fig. 4B1). In contrast, Aslow2 exhibited a significant increase after application of SNC80 (50 lM) at all prepulse durations (Fig. 4B2 asterisks, p < 0:05). Thus, in addition to the slowing of the recovery rates, SNC80 selectively increases the fraction of Na+ channels that recover with sslow2. The relation of sslow1 and sslow2 to the duration of the conditioning pulse was then fitted with a power law function (Eq. (4)), shown superimposed on the data points (Figs. 4B3 and 4). The constant kinetic setpoint was considerably increased in the presence of SNC80 compared to control conditions (sslow1: from a ¼ 1:3 to 3.5 s; sslow2: a ¼ 31:3 41:1 s). The scaling power for sslow1 decreased from 0.19 under control conditions to 0.16 under drug application, whereas the scaling power for sslow2 increased from 0.07 to 0.17, respectively. These findings suggest that SNC80 apparently increases the time scale over which Na+ channels retain a trace of prior activity.
3.5. Effects of SNC80 on entry into slow inactivation It was also apparent from the data shown in Figs. 4A1 and A2 that application of SNC80 significantly increases the fraction of INa undergoing slow inactivation (compare leftmost data points in Figs. 4A1 and A2). We have examined this phenomenon in more detail in Fig. 5. The fraction of INa undergoing slow inactivation was derived as the ratio of INa amplitude 1 s after the conditioning pulses (open circles, Fig. 5A1) to INa amplitude following complete recovery from slow inactivation (Fig. 5A2, asterisks). Assessing the time course of entry into slow inactivation (Fig. 5A2), it becomes apparent that SNC80 promotes entry into an inactivated state preferentially during the first seconds of the conditioning pulse. Application of SNC80 (open squares) significantly enhanced the fraction of INa undergoing slow inactivation at all prepulse durations tested (Fig. 5A2, control: n ¼ 6 7, SNC80: n ¼ 6 10, p < 0:05). Compared to control conditions, the fraction of available INa in the presence of SNC80 was reduced by 8.5%, 34.3%, 43.1%, 51.3%, 55.6% and
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
1107
using a 5 ms testpulse to 10 mV (see inset in Fig. 5B1). In our hands, slow inactivation was clearly dependent on the voltage of the conditioning pulse, resulting in a profound reduction of the available current with more depolarizing conditioning potentials (Fig. 5B2; n ¼ 6). Remarkably, preincubation of SNC80 (50 lM) significantly augments this voltage-dependent reduction of available current (n ¼ 9, Fig. 5B2, p < 0:05, asterisks), displaying the most prominent effects at more depolarized conditioning potentials. 3.6. Selectivity of SNC80 mediated effects
Fig. 3. Fast recovery from inactivation is not effected by the d-agonist SNC80. (A) Fast recovery from inactivation was studied using a double pulse protocol as depicted in the inset. Representative families of current traces during the recovery period are shown for control conditions (ACSF), after application of 50 lM SNC80 and after washout of SNC80 (wash), (B) The current amplitude during recovery from inactivation from membrane potentials of either 70 mV (closed squares: ACSF, open squares: 50 lM SNC80) or 80 mV (closed triangles: ACSF, open squares: 50 lM SNC80) was normalized to Imax at the end of each series and plotted versus the duration of the interpulse interval. The time course of recovery was determined by a biexponential fit to the data points of each individual recording (Eq. (3)) and the best fit to the data is represented by the solid lines.
76.5% for prepulses of 1, 3, 10, 30, 100 and 300 s, respectively. We next asked the question whether the effects of SNC80 on slow inactivation of INa were voltage-dependent. To test this idea, we applied 5 s conditioning pulses from 100 mV to different conditioning potentials ranging from 100 to +20 mV and monitored the fraction of INa recovered from slow inactivation after 1 s
The major effects of SNC80 were a block of the maximal INa conductance and a pronounced effect on slow inactivation processes. We have tested whether these two effects of the putative d-opioid agonist SNC80 were mediated by d-opioid receptors. As a first experiment, we tested whether a structurally distinct dopioid agonist mimics the effects of SNC80. In contrast to the pronounced effects of SNC80, incubation with the selective peptide agonist DPDPE (10 lM) did not alter slow inactivation induced by 10 and 100 s prepulses (Fig. 6A, n ¼ 4). Next, we tested whether the effects of SNC80 on slow inactivation processes are inhibited by opioid antagonists. Preincubation of granule cells with different concentrations of the unselective opioid antagonist naloxone (at least 15 min, 50–300 lM) did not alter the fraction of INa that had undergone slow inactivation after prepulses of 10 and 100 s duration (Fig. 6B, closed and open bars). As a control experiment, we tested whether the highest concentration of naloxone (300 lM) applied alone has effects on slow inactivation of Na+ channels, and found this not to be the case (Fig. 6B). Likewise, the more selective d-opioid antagonist naltrindole (10 and 100 lM) did not alter the effects of SNC80 on slow inactivation. Thus, the efficacy of SNC80 in modulating slow inactivation was maintained even in the presence of high concentrations of different d-opioid antagonists. Collectively, these results indicate that the effects of SNC80 on INa are not mediated by d-opioid receptors.
4. Discussion It is well established that d-opioid receptors modulate multiple types of ion channels via G-proteins, in particular inwardly rectifying K+ channels and voltagegated Ca++ channels (Christie et al., 2000). In this study, we have therefore investigated the effects of SNC80, a putative selective d-opioid receptor agonist, on voltage-gated Na+ channels in native hippocampal neurons. We have demonstrated that SNC80 selectively prolongs the time course of recovery from slow inacti-
1108
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
Fig. 4. The number of available Na+ channels after slow recovery from inactivation is strongly reduced after application of 50 lM SNC80. (A) Time course of Na+ channel recovery from slow inactivation recorded under control conditions (A1) and after drug application (A2). Inactivation was induced by prepulses to 10 mV ranging from 1 to 300 s. The current amplitude during recovery was normalized to Imax, yielding the fraction of available Na+ channels. The normalized values are plotted versus the recovery time for the indicated prepulse durations (n ¼ 6 10). Insets: Representative families of original traces during recovery from inactivation induced by a 100 s prepulse in control solution (A1) and after drug application (A2). Recovery from inactivation was fitted with a biexponential function (Eq. (3)) and best fits are shown as solid lines (A1–2), (B) The resulting time constants (sslow1 and sslow2, panels B3 and 4) and associated amplitudes (Aslow1 and Aslow2, panels B1 and 2) were plotted as function of the prepulse duration. The relationship between the prepulse duration (t) and the recovery time constants (sslow1, sslow2) was fitted to a power law function tðtÞ ¼ aðt=aÞb, where a is the constant kinetic setpoint and b the scaling power(Eq. (4), B3–4) and best fits are shown as solid lines.
vation without effects on fast inactivation processes. In addition, SNC80 reduced the maximal Na+ current amplitude. Surprisingly, we found that these powerful effects were not mediated by activation of d-opioid receptors. 4.1. Effects of SNC80 on slow but not fast inactivation The first interesting aspect of this study is the selectivity of the effects of SNC80 for slow but not fast inactivation processes. Firstly, SNC80 appeared to facilitate inactivation by prolonged depolarization. For
example, the fraction of INa available following inactivation by 300 s depolarizing prepulses was reduced by ~80% following application of SNC80 (see Fig. 5A). In addition, SNC80 potently prolonged the time course of recovery from slow inactivation induced by prolonged depolarization, without affecting recovery from fast inactivation induced by brief depolarizations. The finding further supports the idea that different channel domains might subserve fast and slow inactivation. Fast inactivation in Na+ channels occurs by a ‘‘balland-chain’’ or ‘‘hinged lid’’ mechanism, and is thought to be due to a conformational change of the
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
1109
Fig. 5. SNC80 considerably reduces the fraction of available Na+ channels. (A1) Representative traces of Na+ currents recorded 1 s after depolarizing conditioning pulses to 10 mV of 10, 100 or 300 s duration (open circles) and following complete recovery from slow inactivation (asterisks) in control conditions and in the presence of 50 lM SNC80. (A2) The fraction of Na+ channels available 1 s after a depolarizing conditioning pulse is plotted as function of the prepulse duration under control conditions (closed squares) and in the presence of 50 lM SNC80 (open squares). (B1) Representative traces of Na+ currents recorded 1 s after depolarizing conditioning pulses of 5 s duration to 100, 40 and +20 mV (open circles) and following complete recovery from slow inactivation (asterisks) under control conditions and in the presence of 50 lM SNC80 (holding potential: 100 mV). (B2) The fraction of Na+ channels available 1 s after a depolarizing conditioning pulse is plotted as function of the prepulse potential under control conditions (closed triangles) and in the presence of 50 lM SNC80 (open triangles). Note the strong decrease in availability of Na+ channels in the presence of SNC80 at more depolarized prepulse potentials.
‘‘inactivation particle’’, formed by three hydrophobic amino acids (Ile-Phe-Met) in the interdomain linker III–IV (Stu¨hmer et al., 1989; West et al., 1992). The structural basis of slow inactivation is less clear. A number of studies have demonstrated that mutations in the SCN4A and SCN5A genes, causing structural alterations in the transmembrane segments IV/S4, IV/ S6, II/S1, II/S5, and the loops between IV=S4 þ S5
and II=S5 þ S6 of the NaV1.4 and NaV1.5 a subunits, have strong effects on slow inactivation (Cummins and Sigworth, 1996; Alekov et al., 2001; Mitrovic et al., 2000; Hayward et al., 1997; Melamed-Frank and Marom, 1999; for review, see Vilin and Ruben, 2001). Some of these mutations have been suggested to underlie disease phenotypes such as hyperkalemic periodic paralysis and related myotonic disorders (for review,
1110
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
Fig. 6. SNC80-mediated inhibition of the Na+ current is not mediated by d-opioid receptors. (A) Effects of the peptide d-receptor agonist DPDPE (10 lM) on slow inactivation. The fraction of Na+ channels available after conditioning pulses of 10 s (closed bars) and 100 s (open bars) are depicted. DPDPE did not mimic the effects of SNC80, (B) Neither preincubation of naloxone at the concentrations indicated (50, 100 and 300 lM), nor preincubation with the d-selective antagonist naltrindole (10 and 100 lM) altered the SNC80-mediated reduction in the fraction of Na+ currents available after conditioning pulses of 10 s (closed bars) and 100 s (open bars). Naloxone alone (300 lM) did not alter slow inactivation. The number of experiments is indicated by the numbers in brackets.
see Lehmann-Horn and Jurkat-Rott, 1999). Site-directed mutagenesis studies have revealed that slow inactivation may result from a structural rearrangement of the outer channel pore (Xiong et al., 2003), but another study has failed to reveal effects of pore mutations on slow inactivation (Struyk and Cannon, 2002). Therefore, the structural basis of slow inactivation and recovery may be complex and requires further experimental evaluation. In any case, however, the determinants of fast and slow inactivation seem to be distinct from each other (Rudy, 1978; Cummins and Sigworth, 1996), a finding supported by the selective effects of SNC80 in this study. As described in previous studies (Marom, 1998; Ellerkmann et al., 2001), we observed an increase in the time constants of slow recovery with more prolonged conditioning depolarizations. Such an increase can be explained by assuming a model of Na+ channel inactivation with a number of interlinked slow inactivated states (I1–In): A I1 I2 I3 I4 . . . In where A is the available state. After membrane depolarization, Na+ channels undergo slow inactivation, with more and more channels entering remote slow
inactivated states during more prolonged conditioning depolarization. In terms of the model described above, a change in scaling power as seen for sslow2 might arise because entry into remote inactivated states is facilitated. Such a mechanism is consistent with the increased entry into slow inactivation that we observed (compare Fig. 4 panel A1 and B1, Fig. 5A). Alternatively, it is conceivable that SNC80 binds with a higher affinity to channels that have undergone slow inactivation. Regardless of the mechanism, the effects of SNC80 cause recovery from inactivation to be significantly slower for each given duration of the inactivating depolarization. In addition, the dramatic change in scaling power of sslow2 strongly augments the steepness of the relation between the duration of the conditioning depolarization and the slow recovery rate. This, in turn, implies that the prior history of membrane depolarization is encoded more efficiently in terms of the subsequent availability of Na+ channels (Toib et al., 1998; Ellerkmann et al., 2001). In addition to effects on slow inactivation processes, we also observed a potent reduction of Na+ current amplitude following application of SNC80. This Na+ current amplitude reduction appeared not to be due to
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
effects of SNC80 on fast voltage-dependent activation and inactivation behavior. Similar findings have also been described for the j-opioid agonist U50,488 (Zou et al., 2000). 4.2. Effects of SNC80 are not mediated by opioid receptors Interestingly, both the effects of SNC80 on slow inactivation processes and on Na+ current amplitude appeared not to require activation of opioid receptors. The potent and selective d-opioid receptor agonist DPDPE did not mimic the effects of SNC80. Conversely, in the presence of high concentrations of the opioid receptor antagonists naloxone and naltrindole, the effects of SNC80 on slow inactivation proved to be undiminished. Opiates and endogenous opioid peptides act through multiple receptors, classified as l-, d- and j-opioid receptors (Massotte and Kieffer, 1998; Kieffer, 1999; Kieffer and Gaveriaux-Ruff, 2002). Because of the strong analgesic action of opioid receptor activation (Hayes and Vogelsang, 1991), a large number of selective receptor agonists have been developed for clinical use. More recently, it has been appreciated that certain putative opioid receptor agonists also exert opioid receptor independent effects on membrane currents, in particular on Na+ currents. For instance, the j-opioid receptor agonist U50,488 was able to attenuate behavioral and primary afferent nerve responses to noxious colorectal distension in vivo. This effect could be mimicked by Na+ channel-blockers and in vitro experiments on colon sensory neurons show an inhibition of voltage-activated Na+ currents in an opioid receptor independent manner (Su et al., 2002). Similarly, an inhibitory effect of U50,488 on voltage-activated Na+ currents was described in hippocampal CA3 neurons. Again, the decrease of the Na+ current amplitude was not related to activation of opioid receptors (Zou et al., 2000). Likewise, a concentration-dependent reduction in peak transient Na+ current by the l-opioid receptor agonist morphine on cardiac myocytes was identified (Hung et al., 1998). In another study, effects of naltrindole, a highly specific d-opioid receptor antagonist, which has been shown to inhibit graft rejection in vivo and suppress allogenic lymphocyte reaction in vitro similarly to cyclosporin A, could not be blocked in dreceptor or triple opioid receptor knockout mice (Gaveriaux-Ruff et al., 2001). Taken together, these data suggest that putative selective opioid receptor agonists show highly specific effects on voltage-gated ion channels. In particular, the effects on slow inactivation may be relevant, given that a number of Na+ channel mutations that give rise to human diseases alter slow inactivation of these channels. In the future, it will be important to analyze such
1111
selective mechanisms of action of putative opioid receptor agonists, in particular, those intended for clinical use. In addition, the highly selective action of the opioid receptor agonist tested in this study on biophysical Na+ channel characteristics may help to extend our understanding of the biophysical mechanisms underlying slow and fast inactivation of these channels.
Acknowledgements This research was supported by the graduate program of the Deutsche Forschungsgemeinschaft ‘‘Pathogenese von Krankheiten des Nervensystems’’.
References Alekov, A.K., Peter, W., Mitrovic, N., Lehmann-Horn, F., Lerche, H., 2001. Two mutations in the IV/S4-S5 segment of the human skeletal muscle Na+ channel disrupt fast and enhance slow inactivation. Neurosci. Lett. 306, 173–176. Bezanilla, F., Armstrong, C.M., 1977. Inactivation of the sodium channel. I. Sodium current experiments. J. Gen. Physiol. 70, 549–566. Bilsky, E.J., Calderon, S.N., Wang, T., Bernstein, R.N., Davis, P., Hruby, V.J., McNutt, R.W., Rothman, R.B., Rice, K.C., Porreca, F., 1995. SNC 80, a selective, nonpeptidic and systemically active opioid delta agonist. J. Pharmacol. Exp. Ther. 273, 359–366. Calderon, S.N., Rice, K.C., Rothman, R.B., Porreca, F., FlippenAnderson, J.L., Kayakiri, H., Xu, H., Becketts, K., Smith, L.E., Bilsky, E.J., Davis, P., Horvath, R., 1997. Probes for narcotic receptor mediated phenomena. 23. Synthesis, opioid receptor binding, and bioassay of the highly selective delta agonist (+)-4[(alpha R)-alpha-((2S,5R)-4-allyl-2,5-dimethyl-1-piperazinyl)-3methoxybenzyl]-N,N-diethylbenzamide (SNC 80) and related novel nonpeptide delta opioid receptor ligands. J. Med. Chem. 40, 695–704. Catterall, W.A., 1992. Cellular and molecular biology of voltagegated sodium channels. Physiol. Rev. 72 (Suppl.), S15–S48. Christie, M.J., Connor, M., Vaughan, C.W., Ingram, S.L., Bagley, E.E., 2000. Cellular actions of opioids and other analgesics: implications for synergism in pain relief. Clin. Exp. Pharmacol. Physiol. 27, 520–523. Colbert, C.M., Magee, J.C., Hoffman, D.A., Johnston, D., 1997. Slow recovery from inactivation of Na+ channels underlies the activity-dependent attenuation of dendritic action potentials in hippocampal CA1 pyramidal neurons. J. Neurosci. 17, 6512–6521. Cummins, T.R., Sigworth, F.J., 1996. Impaired slow inactivation in mutant sodium channels. Biophys. J. 71, 227–236. Ellerkmann, R.K., Riazanski, V., Elger, C.E., Urban, B.W., Beck, H., 2001. Slow recovery from inactivation regulates the availability of voltage-dependent Na+ channels in hippocampal granule cells, hilar neurons and basket cells. J. Physiol. 532, 385– 397. Fleidervish, I.A., Gutnick, M.J., 1996. Kinetics of slow inactivation of persistent sodium current in layer V neurons of mouse neocortical slices. J. Neurophysiol. 76, 2125–2130. Fleidervish, I.A., Friedman, A., Gutnick, M.J., 1996. Slow inactivation of Na+ current and slow cumulative spike adaptation in
1112
C. Remy et al. / Neuropharmacology 47 (2004) 1102–1112
mouse and guinea-pig neocortical neurones in slices J. Physiol. 493 (Pt. 1), 83–97. Gaveriaux-Ruff, C., Filliol, D., Simonin, F., Matthes, H.W., Kieffer, B.L., 2001. Immunosuppression by delta-opioid antagonist naltrindole: delta- and triple mu/delta/kappa-opioid receptor knockout mice reveal a nonopioid activity. J. Pharmacol. Exp. Ther. 298, 1193–1198. Hayes, S.R., Vogelsang, J., 1991. Opiate receptors and analgesia: an update. J. Post Anesth. Nurs. 6, 125–128. Hayward, L.J., Brown, R.H., Cannon, S.C., 1997. Slow inactivation differs among mutant Na+ channels associated with myotonia and periodic paralysis. Biophys. J. 72, 1204–1219. Hung, C.F., Tsai, C.H., Su, M.J., 1998. Opioid receptor independent effects of morphine on membrane currents in single cardiac myocytes. Br. J. Anaesth. 81, 925–931. Kieffer, B.L., 1999. Opioids: first lessons from knockout mice. Trends Pharmacol. Sci. 20, 19–26. Kieffer, B.L., Gaveriaux-Ruff, C., 2002. Exploring the opioid system by gene knockout. Prog. Neurobiol. 66, 285–306. Lehmann-Horn, F., Jurkat-Rott, K., 1999. Voltage-gated ion channels and hereditary disease. Physiol. Rev. 79, 1317–1372. Marom, S., 1998. Slow changes in the availability of voltage-gated ion channels: effects on the dynamics of excitable membranes. J. Membr. Biol. 161, 105–113. Martina, M., Jonas, P., 1997. Functional differences in Na+ channel gating between fast-spiking interneurones and principal neurones of rat hippocampus. J. Physiol. 505, 593–603. Massotte, D., Kieffer, B.L., 1998. A molecular basis for opiate action. Essays Biochem. 33, 65–77. Melamed-Frank, M., Marom, S., 1999. A global defect in scaling relationship between electrical activity and availability of muscle sodium channels in hyperkalemic periodic paralysis. Pflugers Arch. 438, 213–217. Mitrovic, N., George, Jr., A.L., Horn, R., 2000. Role of domain 4 in sodium channel slow inactivation. J. Gen. Physiol. 115, 707–718. Ragsdale, D.S., Avoli, M., 1998. Sodium channels as molecular targets for antiepileptic drugs. Brain Res. Brain Res. Rev. 26, 16–28. Reckziegel, G., Beck, H., Schramm, J., Elger, C.E., Urban, B.W., 1998. Electrophysiological characterization of Na+ currents in
acutely isolated human hippocampal dentate granule cells. J. Physiol. 509 (Pt. 1), 139–150. Remy, S., Urban, B.W., Elger, C.E., Beck, H., 2003. Anticonvulsant pharmacology of voltage-gated sodium channels in hippocampal neurons of control and chronically epileptic rats. Eur. J. Neurosci. 17, 2648–2658. Ruben, P.C., Starkus, J.G., Rayner, M.D., 1992. Steady-state availability of sodium channels. Interactions between activation and slow inactivation. Biophys. J. 61, 941–955. Rudy, B., 1978. Slow inactivation of the sodium conductance in squid giant axons. Pronase resistance. J. Physiol. 283, 1–21. Struyk, A.F., Cannon, S.C., 2002. Slow inactivation does not block the aqueous accessibility to the outer pore of voltage-gated Na+ channels. J. Gen. Physiol. 120, 509–516. Stu¨hmer, W., Conti, F., Suzuki, H., Wang, X., Noda, M., Yahagi, N., Kubo, H., Numa, S., 1989. Structural parts involved in activation and inactivation of the sodium channel. Nature 339, 597–603. Su, X., Joshi, S.K., Kardos, S., Gebhart, G.F., 2002. Sodium channel blocking actions of the kappa-opioid receptor agonist U50,488 contribute to its visceral antinociceptive effects. J. Neurophysiol. 87, 1271–1279. Toib, A., Lyakhov, V., Marom, S., 1998. Interaction between duration of activity and time course of recovery from slow inactivation in mammalian brain Na+ channels. J. Neurosci. 18, 1893–1903. Vilin, Y.V., Ruben, P.C., 2001. Slow inactivation in voltage-gated sodium channels: molecular substrates and contributions to channelopathies. Cell Biochem. Biophys. 35, 171–190. West, J.W., Patton, D.E., Scheuer, T., Wang, Y., Goldin, A.L., Catterall, W.A., 1992. A cluster of hydrophobic amino acid residues required for fast Na(+)- channel inactivation. Proc. Natl. Acad. Sci. USA 89, 10910–10914. Xiong, W., Li, R.A., Tian, Y., Tomaselli, G.F., 2003. Molecular motions of the outer ring of charge of the sodium channel: do they couple to slow inactivation? J. Gen. Physiol. 122, 323–332. Zou, B., Chen, Y., Wu, C., Zhou, P., 2000. Blockade of U50488H on sodium currents in acutely isolated mice hippocampal CA3 pyramidal neurons. Brain Res. 855, 132–136.