Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

In a number of neurological conditions ranging from acute insults such as strokes to chronic neurodegenerative disorders such as amyotropic lateral sclerosis (ALS), the excessive release of excitatory amino acid neurotransmitters may mediate neuronal injury or death (Lipton and Rosenberg, 1994). Glutamate is the principal excitatory neurotransmitter in the brain. Neurotoxicity due to glutamate appears to be predominantly mediated by excessive Ca²⁺ influx through glutamate-associated ion channels (Choi, 1987; Randall and Thayer, 1992; Lipton and Rosenberg, 1994).

Riluzole is a neuroprotective agent used in the treatment of the degenerative motor neuron disease ALS (Bensimen et al., 1994; Wokke, 1996). In addition to its effectiveness in treating ALS, riluzole may also be useful in the treatment of spinal cord injury and ischemia (Malgouris et al., 1989; Wokke, 1996). This drug also has sedative and anticonvulsant properties, whereas at higher concentrations, it acts as an anesthetic (Mantz et al., 1992; Wokke, 1996).

The molecular mechanism for the neuroprotective actions of riluzole is currently being investigated. It appears likely that the therapeutic actions of riluzole may stem from its ability to inhibit glutamate release from nerve terminals in the brain and spinal cord under pathological conditions (Martin et al., 1993; Lipton and Rosenberg, 1994; Wokke, 1996). Riluzole has been shown to inhibit the release of glutamate from presynaptic nerve terminals at several sites in the brain (Cheramy et al., 1992; Martin et al., 1993).

The specific cellular mechanism for the presynaptic inhibitory action is unknown. Riluzole inhibits voltage-gated Na⁺ channels in central nervous system neurons (Benoit and Escande, 1991; Mantz et al., 1992; Prakriya and Mennerick, 2000; Urbani and Belluzzi, 2000). Also, riluzole has recently been shown to activate two different varieties of cloned background K⁺ channels that set the resting potential in many cells, including neurons (Duprat et al., 2000). Activation of these nonvoltage-gated K⁺ channels could lead to membrane hyperpolarization, and a decrease in excitability and transmitter release.

Voltage-gated K⁺ channels are present in presynaptic terminals where they may regulate action potential, waveform, and transmitter release (Hille, 1992). Kv1.4 K⁺ channels are rapidly inactivating A-type channels that are present in axons and nerve terminals in many areas of the brain, including cells that secrete glutamate and other excitatory neurotransmitters (Storm, 1987; Sheng et al., 1992).

Bovine adrenal zona fasciculata (AZF) cells express bKv1.4-type K⁺ channels that are highly homologous to Kv1.4 channels in other cells, including cardiac myocytes and neurons (Mlinar and Enyeart, 1993; Wymore et al., 1996; Enyeart et al., 2000). Kv1.4 channels in all these cells display N-type inactivation, which involves the docking of an inactivation ball formed from N-terminal amino acids with a site in the pore (Hoshi et al., 1990; Ruppersberg et al., 1991).

We have studied the effects of riluzole on whole-cell bKv1.4 K⁺ currents in bovine AZF cells. In addition to inhibiting Kv1.4 K⁺ currents, riluzole dramatically and irreversibly slows inactivation of bKv1.4 K⁺ channels by a mechanism that may involve oxidation of the N-terminal inactivation domain.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Tissue culture media, antibiotics, fibronectin, and fetal bovine sera were obtained from Invitrogen (Carlsbad, CA). Coverslips were purchased from Bellco Glass, Inc. (Vineland, NJ). Enzymes, riluzole, reduced glutathione, MgATP, NaUTP, 5-adenylylimido-diphosphate (AMP-PNP, lithium salt), NaGTP, cAMP, 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA), and 4-amino-pyridine (4-AP) were obtained from Sigma Chemical Co. (St. Louis, MO).

Isolation and Culture of AZF Cells. Bovine adrenal glands were obtained from steers (age 1-3 years) within 30 min of slaughter at a local slaughterhouse. Fatty tissue was removed immediately and the glands were transported to the laboratory in ice-cold phosphate-buffered saline containing 0.2% dextrose. Isolated AZF cells were prepared as previously described (Enyeart et al., 1996). After isolation, cells were either resuspended in DMEM/F12 (1:1) with 10% fetal bovine serum, 100 U/ml penicillin, 0.1 mg/ml streptomycin, and the antioxidants 1 µM tocopherol, 20 nM selenite, and 100 µM ascorbic acid (DMEM/F12+) and plated for immediate use; or resuspended in fetal bovine serum/5% dimethyl sulfoxide, divided into 1-ml aliquots each containing about 2 × 10⁶ cells, and stored in liquid nitrogen for future use. Cells were plated in 35-mm dishes containing 9-mm² glass coverslips that had been treated with fibronectin (10 µg/ml) at 37°C for 30 min and then rinsed with warm, sterile PBS immediately before adding cells. Dishes were maintained in DMEM/F12+ at 37°C in a humidified atmosphere of 95% air and 5% CO₂.

Patch-Clamp Experiments. Patch-clamp recordings of K⁺ channel currents were made in the whole-cell configuration. The standard pipette solution consisted of 120 mM KCl, 2 mM MgCl₂, 1 mM CaCl₂, 20 mM HEPES, 11 mM BAPTA, 200 µM GTP, 5 mM MgATP, and 100 µM cAMP with pH buffered to 7.2 by using KOH. Deviations from the standard solution are described in the text. The external solution consisted of 140 mM NaCl, 5 mM KCl, 2 mM CaCl, 2 mM MgCl₂, 10 mM HEPES, and 5 mM glucose, buffered to pH 7.4 by using NaOH. All solutions were filtered through 0.22-µm cellulose acetate filters. Drugs were applied externally by bath perfusion controlled manually by a six-way rotary valve. Riluzole was dissolved in dimethyl sulfoxide at a concentration of 200 mM.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

In whole-cell recordings from AZF cells made with standard pipette solutions containing 5 mM MgATP, voltage steps to +20 mV from a holding potential of 80 mV typically activate two types of K⁺ currents. These include a voltage-gated rapidly inactivating bKv1.4 K⁺ channel and the noninactivating, ATP-dependent I_AC K⁺ current that develops over a period of minutes (Mlinar et al., 1993; Mlinar and Enyeart, 1993; Enyeart et al., 1997). I_AC expression is completely and selectively inhibited by the presence of 100 µM cAMP in the pipette solution, allowing bKv1.4 to be recorded in isolation (Enyeart et al., 1997). cAMP by itself has no effect on bKv1.4 amplitude or kinetics (Mlinar and Enyeart, 1993; Enyeart et al., 1998).

When AZF cells were superfused with riluzole at concentrations from 1 to 100 µM, two distinct effects were observed. The first of these was a concentration-dependent inhibition of bKv1.4 that was rapidly reversible upon washing with control saline (Fig. 1, A and B). Inhibition of bKv1.4 was half-maximal at a concentration of 70 µM (Fig. 1C).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Time- and concentration-dependent inhibition of bKv1.4 K+ current by riluzole. Whole-cell K+ currents were recorded from AZF cells in response to voltage steps applied at 30-s intervals from a holding potential of 80 mV to a test potential of +20 mV. After recording currents in control saline, cells were superfused with riluzole at concentrations ranging from 1 to 100 µM. A and B, current traces and associated time-dependent plot of bKv1.4 amplitude for cell superfused with increasing concentrations of riluzole as indicated. Numbers on traces correspond to currents recorded at times indicated on graph. C, inhibition curve, for riluzole constructed from experiments as in A and B. Fraction of unblocked bKv1.4 current is plotted against riluzole concentration. Data are fit with the equation I/Imax = 1/[1 + (B/IC50)X], where B is the riluzole concentration, IC50 is the concentration that reduces bKv1.4 by 50%, and X is the Hill slope. Data are normalized, mean values of the indicated number of determinations.

The rapid and complete reversal of riluzole-mediated inhibition of bKv1.4 upon washing with saline revealed that this drug also dramatically and irreversibly slowed the kinetics of bKv1.4 inactivation. In the experiment illustrated in Fig. 2A, superfusion of the cell with 100 µM riluzole reduced peak bKv1.4 current (trace 1) by 61% within 2 min (trace 2). Upon returning to control saline for 2 min, peak bKv1.4 current was restored to a value 37% greater than the control amplitude, whereas the inactivation time constant increased from 19.4 to 768 ms (trace 3).

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Riluzole slows bKv1.4 inactivation kinetics. A, whole-cell K+ currents were recorded as described in the legend of Fig. 1. After recording bKv1.4 currents in control saline (1), cell was superfused with riluzole (100 µM). When steady-state block was reached (2), the cell was again superfused with control saline for an additional 90 min (3-6). B, 4-AP inhibits slowly inactivating current. While recording K+ current as described above, the cell was sequentially superfused with control saline (1), 100 µM riluzole (2), control saline (3), and 4-AP (4) before finally returning to control saline (5). Numbers on current traces correspond to graph at right where peak bKv1.4 amplitudes are plotted against time.

The pronounced slowing of inactivation kinetics was all the more remarkable because it persisted, even in the presumed absence of riluzole. In the experiment illustrated in Fig. 2A, superfusion of the cell with control saline for 2 min fully reversed bKv1.4 inhibition. However, the slowed inactivation was only slightly changed after 30, 60, and 90 min of continuous washing (Fig. 2A, traces 4, 5, and 6, and graph). After 90 min of washing, the inactivation time constant was 645 ms.

The increase in peak bKv1.4 amplitude, which also persisted with prolonged washing, is directly attributable to the slowed inactivation. Because the kinetics of bKv1.4 activation and inactivation overlap, selective slowing of inactivation kinetics increases the maximum number of bKv1.4 channels that are simultaneously open in response to a depolarization (Mlinar and Enyeart, 1993; Enyeart et al., 1998).

The dual effects of riluzole observed in this study, including its inhibitory actions and slowing of inactivation kinetics, were due to the exclusive interaction of this drug with bKv1.4 K⁺ channels. The rapidly inactivating bKv1.4 K⁺ current is inhibited by 4-AP with an IC₅₀ of 630 µM (Mlinar and Enyeart, 1993). Accordingly, the slowly inactivating K⁺ current present after exposure to riluzole and subsequent wash was inhibited almost completely by 3 mM 4-AP, indicating that it is a modified bKv1.4 current (Fig. 2B). The noninactivating I_AC K⁺ current, whose expression was suppressed by cAMP in these experiments, is less sensitive to 4-AP (Gomora and Enyeart, 1999).

Concentration-Dependent Slowing of bKv1.4 Inactivation Kinetics by Riluzole. Riluzole slowed bKv1.4 inactivation at concentrations similar to those that inhibited this current. In each of the four experiments illustrated in Fig. 3A, bKv1.4 currents were recorded in control saline (trace 1) before superfusing each cell with riluzole at a single concentration that ranged from 1 to 100 µM, as indicated. When steady-state inhibition of bKv1.4 was achieved (trace 2), the cell was again superfused with control saline until inhibition was fully reversed and a stable current was obtained (trace 3). Overall, at a concentration of 10 µM, riluzole significantly increased the inactivation time constant from its original value of 28.9 ± 3.9 (n = 6) to 75.4 ± 11.2 ms (n = 6) (Fig. 3B). At concentrations of 50 and 100 µM, riluzole further increased inactivation time constants to 254.2 ± 33.1 (n = 8) and 623.2 ± 47.6 ms (n = 13), respectively (Fig. 3B). At still higher concentrations, this drug appeared to be toxic, and high resistance seals could not be maintained.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Concentration-dependent slowing of bKv1.4 inactivation by riluzole. bKv1.4 currents were recorded as described in the legend of Fig. 1. After recording K+ current in control saline, cells were superfused with riluzole at concentrations ranging from 1 to 100 µM. After steady-state block was achieved, cells were superfused with control saline until a new steady-state current was reached. The inactivating component of this current was then fit with a single exponential to determine i. A, K+ current traces from four cells exposed to riluzole at concentrations of 1, 10, 50, or 100 µM as indicated. Numbers correspond to currents recorded in control saline (1), in the presence of riluzole (2), and after wash (3). B, inactivation time constants derived from experiments as in A are plotted against riluzole concentration for the indicated number of determinations. Values are mean ± S.E.M.

Riluzole Does Not Change Activation Kinetics. Gating kinetics of bKv1.4 channels, including activation and inactivation, have been described in detail (Mlinar and Enyeart, 1993; Enyeart et al., 1998). The time-dependent bKv1.4 current activated by a voltage step can be fit by the following equation: I = I_max[1  exp(T/_a)]⁴ [exp(T/_i)], where _a is the voltage-dependent activation time constant, and _i is the voltage-independent time constant of inactivation.

Effects of Riluzole Are Independent of Test Potential. The effects of riluzole on bKv1.4 current amplitude and inactivation kinetics were similar over a wide range of test potentials. In the experiment illustrated in Fig. 4, current-voltage relationships were obtained in control saline, after steady-state inhibition by 100 µM riluzole, and again after washing with saline. Initially, in control saline, bKv1.4 appeared as a rapidly inactivating current that increased in amplitude with increasingly positive depolarizations (Fig. 4, A and B). When the cell was exposed to riluzole (100 µM), bKv1.4 current amplitude was reduced by 62 to 73% at test potentials ranging from 30 to +40 mV. The slowing of bKv1.4 inactivation kinetics was also obvious even in the presence of riluzole, where current amplitudes were much reduced (Fig. 4A).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Voltage-independent actions of riluzole. Current-voltage relationships were obtained for bKv1.4 in control saline, after steady-state block by riluzole (100 µM) and again after washing in saline. Voltage steps were applied at 30-s intervals from a holding potential of 80 mV to test potentials between 40 and +40 mV in 10-mV increments. A, current records at test potentials between 40 and +40 mV in control saline (left), 100 µM riluzole (middle), and after wash (right). B, current-voltage relationships: peak current amplitudes taken from the experiment illustrated in A are plotted against test potential. C, inactivation time constants: control current traces and traces following treatment with riluzole in A were fit with single exponentials. Inactivation time constants (i) were plotted against test potential.

Molecular Mechanism of Riluzole. The molecular mechanism by which riluzole acts to irreversibly slow bKv1.4 inactivation kinetics is unknown. The inactivation kinetics of some A-type K⁺ channels is regulated through phosphorylation by kinases, including cAMP-dependent protein kinase (A-kinase) and protein kinase C (Covarrubias et al., 1994; Drain et al., 1994). Riluzole has been reported to inhibit background K⁺ channels through an A-kinase-dependent mechanism (Duprat et al., 2000). However, it is unlikely that riluzole-mediated slowing of inactivation kinetics is mediated through A-kinase, because pipette solutions containing cAMP at a concentration that fully activates this enzyme did not slow inactivation kinetics. Furthermore, it is unlikely that any other common protein kinase is involved because riluzole-induced slowing of inactivation was observed even when ATP in the pipette solution was replaced with nucleotides such as UTP or AMP-PNP, which are poor substrates for these enzymes.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Nonhydrolyzable nucleotides do not affect riluzole actions. Whole-cell recordings were made with pipette solutions containing ATP (1 or 5 mM), UTP (1 mM), or AMP-PNP (1 mM). bKv1.4 currents were activated by voltage steps to +20 mV applied at 30-s intervals from a holding potential of 80 mV. Cells were superfused with riluzole until a steady-state block was achieved and then washed with control saline. A, current traces in control saline (1), after riluzole (2), and after wash (3) with pipettes containing ATP, UTP, or AMP-PNP as indicated. B, i: the inactivating component of bKv1.4 from experiments as in Fig. 1 was fit with single exponentials to determine i. Columns illustrate i with different nucleotides in the absence () or presence () of riluzole. Values are mean ± S.E.M.

Riluzole Slows Inactivation through Oxidation of bKv1.4 K⁺ Channels. The inactivation of cloned Kv1.4 K⁺ channels from rat brain is slowed or removed by oxidation of a cysteine located on the N-terminal ball domain. Rapid inactivation kinetics is restored by application of reducing agents to the intracellular side of the channel (Ruppersberg et al., 1991). To determine whether riluzole mediates slowing of bKv1.4 inactivation kinetics in AZF cells by a mechanism involving oxidation, the antioxidant reduced glutathione was added to the patch pipette solution.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Glutathione prevents slowing of inactivation by riluzole. bKv1.4 currents were recorded with pipettes containing standard solution or the same solution supplemented with reduced glutathione (3 mM). bKv1.4 currents were activated by voltage steps to +20 mV applied at 30-s intervals from a holding potential of 80 mV. A, bKv1.4 current traces in control saline (1), after steady-state block by 100 µM riluzole (2), and after reversal of block with control saline (3). B, current traces from experiments as in A were fit with single exponentials to determine inactivation time constants measured in control saline before and after exposure to riluzole. Values are mean ± S.E.M. for the indicated number of determinations in the presence or absence of glutathione, as indicated.

Riluzole-Dependent Slowing of bKv1.4 Inactivation Is Eliminated by Membrane Depolarization. Our results strongly suggest that riluzole-mediated slowing of bKv1.4 inactivation occurs through oxidation of a cysteine that is probably located on the inactivating ball domain. If so, bKv1.4 channels might be rendered resistant to riluzole when the ball domain docks with its site in the mouth of the channel pore. To test this hypothesis, we studied the effect of riluzole on bKv1.4 inactivation kinetics after exposing cells to this drug at holding potentials of either 40 or 80 mV. At a holding potential of 80 mV, bKv1.4 channels are almost entirely in the closed conformation, whereas at 40 mV, these channels are inactivated (Mlinar and Enyeart, 1993; Enyeart et al., 1998).

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Effect of holding potential (HP) on riluzole-mediated slowing of inactivation. After recording bKv1.4 in control saline, the HP was either maintained at 80 mV or switched to 40 mV. The cell was then sequentially superfused for 5 min with riluzole (100 µM) followed by saline. bKv1.4 current was then recorded again in response to a voltage step to +20 mV from a HP of 80 mV. A, current traces show bKv1.4 before and after riluzole exposure at HPs of 40 mV (top traces) or 80 mV (bottom traces) as indicated. B, i values obtained from experiments as in A are plotted as mean ± S.E.M. for the indicated number of determinations for cells exposed to riluzole at either 40 or 80 mV.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, two separate actions of riluzole on bKv1.4 K⁺ currents in bovine AZF cells were described. These include a concentration-dependent reduction in current amplitude and a pronounced slowing of inactivation that persists after prolonged washing. The distinctive irreversible slowing of bKv1.4 inactivation may result from riluzole-mediated oxidation of a cysteine in the N-terminal inactivation ball. Oxidation of a voltage-gated K⁺ channel suggests an entirely new target and molecular mechanism of action for this agent. Regardless of the mechanism, the marked slowing of inactivation kinetics could mediate inhibition of glutamate release from nerve terminals, an action that could contribute to the neuroprotective action of this drug.

Molecular Mechanism of Riluzole. Although riluzole-induced inhibition of bKv1.4 current and slowing of inactivation kinetics occur over a similar range of concentrations, the marked difference in reversibility with washing and sensitivity to antioxidants indicate that two separate receptors and/or mechanisms are involved.

Oxidation of N-Terminal Cysteine by Riluzole. Inactivation of rat brain Kv1.4 K⁺ channels is markedly slowed by oxidation of a cysteine residue located at position 13 in the N-terminal ball domain or channel inactivation gate (Ruppersberg et al., 1991). The bKv1.4 K⁺ channel of AZF cells is highly homologous to the rat brain channel and contains a cysteine residue at position 13 in the N-terminal ball domain (Enyeart et al., 2000). The slowing of bKv1.4 inactivation by riluzole and the near complete inhibition of this effect by reduced glutathione and DTT suggest that oxidation of the N-terminal cysteine may be the causative mechanism.

View larger version (46K): [in this window] [in a new window]   Fig. 8.   Model for voltage-dependent oxidation of Kv1.4 K+ channels by riluzole. At a holding potential (HP) of 40 mV (top), Kv1.4 channel is inactivated and N-terminal cysteine is protected from oxidation by riluzole. At HP of 80 mV (bottom), N-terminal cysteine of closed channel is oxidized by riluzole and forms disulfide bond with neighboring cysteine.

Slowed Inactivation Linked to Increase in bKv1.4 Amplitude. The increase in bKv1.4 current amplitude observed after riluzole washout is directly attributable to the slowed inactivation kinetics. In a simplified gating scheme shown below, bKv1.4 channels shuttle from closed to open and finally to the inactivated state during the course of a depolarizing voltage step. Transitions to the open and inactivated conformations are determined by a voltage-dependent activation rate constant (K_a), and a voltage-independent inactivation rate constant (K_i) (Mlinar and Enyeart, 1993; Enyeart et al., 1998):
According to this scheme, riluzole reduces K_i but has no effect on K_a. Consequently, a greater number of bKv1.4 channels collect in the open state during a depolarization, producing a larger peak bKv1.4 current.

Mechanism for Inhibition of Glutamate Release. The cellular mechanism responsible for the neuroprotective effects of riluzole is not certain. In this regard, excessive activation of glutamate receptors may mediate neuronal injury or death in various neuropathological conditions, both acute and chronic (Choi, 1987; Randall and Thayer, 1992; Lipton and Rosenberg, 1994). It has been proposed that the beneficial effects of riluzole stem from its ability to limit this excitotoxicity by inhibiting the release of glutamate from presynaptic terminals (Malgouris et al., 1989). Riluzole reduces the release of glutamate from the caudate nucleus and from cultured cerebellar granule and hippocampal cells (Cheramy et al., 1992; Martin et al., 1993; Prakriya and Mennerick, 2000).