Aminopyridines Potentiate Synaptic and Neuromuscular Transmission by Targeting the Voltage-activated Calcium Channel β Subunit*♦

Aminopyridines such as 4-aminopyridine (4-AP) are widely used as voltage-activated K+ (Kv) channel blockers and can improve neuromuscular function in patients with spinal cord injury, myasthenia gravis, or multiple sclerosis. Here, we present novel evidence that 4-AP and several of its analogs directly stimulate high voltage-activated Ca2+ channels (HVACCs) in acutely dissociated neurons. 4-AP, 4-(aminomethyl)pyridine, 4-(methylamino)pyridine, and 4-di(methylamino)pyridine profoundly increased HVACC, but not T-type, currents in dissociated neurons from the rat dorsal root ganglion, superior cervical ganglion, and hippocampus. The widely used Kv channel blockers, including tetraethylammonium, α-dendrotoxin, phrixotoxin-2, and BDS-I, did not mimic or alter the effect of 4-AP on HVACCs. In HEK293 cells expressing various combinations of N-type (Cav2.2) channel subunits, 4-AP potentiated Ca2+ currents primarily through the intracellular β3 subunit. In contrast, 4-AP had no effect on Cav3.2 channels expressed in HEK293 cells. Furthermore, blocking Kv channels did not mimic or change the potentiating effects of 4-AP on neurotransmitter release from sensory and motor nerve terminals. Thus, our findings challenge the conventional view that 4-AP facilitates synaptic and neuromuscular transmission by blocking Kv channels. Aminopyridines can directly target presynaptic HVACCs to potentiate neurotransmitter release independent of Kv channels.

. The classical view is that the beneficial effect of 4-AP results from blocking Kv channels, which leads to increases in the duration of action potentials, Ca 2ϩ influx, and neurotransmitter release (6 -8). This conception assumes that voltageactivated Ca 2ϩ channels (VACCs) are stimulated indirectly by increased excitability of cells after Kv channels are blocked by 4-AP. However, other Kv channel blockers such as tetraethylammonium (TEA) have limited effects in potentiating neurotransmitter release and improving neuromuscular function. Furthermore, 4-AP is effective in the treatment of patients with VACC antagonist overdose (9), Lambert-Eaton syndrome caused by impairment of presynaptic VACCs (10), or episodic ataxia type 2, a disorder caused by mutation of Cav2.1 (11). These observations raise the possibility that 4-AP may directly stimulate VACCs in addition to its effect on Kv channels.
Although VACCs are essential for synaptic and neuromuscular transmission, there is no evidence that 4-AP can directly stimulate VACCs in neurons. In this study, we investigated the direct effect of 4-AP on VACCs in dissociated neurons and its role in potentiating neurotransmitter release. We discovered that 4-AP and several of its analogs have a profound potentiating effect on high voltage-activated Ca 2ϩ channels (HVACCs) independent of Kv channels. The intracellular ␤ subunit is largely responsible for this potentiating effect. Furthermore, 4-AP potentiates neurotransmitter release from both sensory and motor nerve terminals independent of Kv channels. Our findings strongly suggest that aminopyridines facilitate synaptic and neuromuscular transmission primarily through direct stimulation of HVACCs. 4-AP and 3,4-DAP have been used as specific Kv channel blockers for Ͼ30 years, and their direct effects on HVACCs have long been overlooked. The interpretation of their effects on the intracellular Ca 2ϩ level and neurotransmitter release should be revised on the basis of this new information.

EXPERIMENTAL PROCEDURES
Isolation of Dorsal Root Ganglion, Superior Cervical Ganglion, and Hippocampal Neurons-Male Sprague-Dawley rats (5-6 weeks old; Harlan, Indianapolis, IN) were anesthetized with isoflurane and then rapidly decapitated. The thoracic and lumbar segments of the vertebral column were dissected. The dorsal root ganglions (DRGs) were quickly removed and transferred immediately into Dulbecco's modified Eagle's medium (DMEM; Invitrogen). The ganglion fragments were placed in a flask containing 5 ml of DMEM in which trypsin (type I, 0.5 mg/ml; Sigma) and collagenase D (1 mg/ml; Pfizer) had been dissolved. After incubation at 34°C in a shaking water bath for 30 min, soybean trypsin inhibitor (type II-s, 1.25 mg/ml; Sigma) was added to stop trypsin digestion. The cell suspension was subjected to centrifugation (500 rpm, 5 min) to remove the supernatant and replenished with DMEM. The cultured DRG neurons were replenished with DMEM containing 10% fetal bovine serum (Invitrogen) and penicillin/streptomycin/glutamine supplement (1%; Invitrogen). Neurons were then plated onto a 35-mm culture dish containing poly-L-lysine (50 g/ml)precoated coverslips and kept for at least 1 h before electrophysiological recordings (12).
The superior cervical ganglion (SCG) neurons were removed and placed in cold (4°C) DMEM. The ganglion fragments were then placed in a flask containing 5 ml of DMEM in which trypsin and collagenase D had been dissolved. After incubation at 34°C in a shaking water bath for 30 min, the flask was shaken vigorously by hand to release the neuronal somata from the fragments. Soybean trypsin inhibitor was then added to stop trypsin digestion. The cell suspension was subjected to centrifugation (500 rpm, 5 min) to remove the supernatant and replenished with DMEM.
To dissociate hippocampal neurons, neonatal rats (2-3 days old) were deeply anesthetized with isoflurane and then decapitated. The whole brain was removed and placed immediately in ice-cold DMEM. The hippocampus was identified and dissected under a microscope, and neurons were isolated by an enzymatic protocol. The tissues were incubated in DMEM with 0.25% trypsin for 15 min at 37°C, and neurons were mechanically dissociated using a glass pipette. The neurons were then plated (1.5 ϫ 10 6 cell/ml) on poly-L-lysine-precoated coverslips. The isolated neurons were grown in DMEM containing 10% fetal bovine serum and penicillin/streptomycin/glutamine supplement for 24 h. The neuron basal medium (2% B-27; Invitrogen) was replaced every 3-4 days.
Cell Culture and Transfection-Human embryonic kidney HEK293 cells were grown in DMEM (containing 10% fetal bovine serum but no antibiotics) to 80% confluence and maintained at 37°C in a humidified incubator with 5% CO 2 . The cells were transiently transfected with N-type Ca 2ϩ channel subunits, including ␣ 1 B (from rat SCG), ␣ 2 ␦ (from rat SCG), and ␤ 3 (from rat brain), or Cav3.2␣ cDNA (from rat brain), all in pcDNA3.1 using Lipofectamine 2000 (Invitrogen). When expression of more than one subunit was induced, the multiplasmids were cotransfected at a 1:1 ratio. Fifteen hours after transfection, the cells were dissociated with 0.05% trypsin. Cells were then plated onto a 35-mm culture dish containing poly-Llysine-precoated coverslips and kept for another 18 h in an incubator at 30°C before electrophysiological recordings.
Electrophysiological Recordings-Electrodes with a resistance of ϳ2 megaohms were pulled from glass capillaries using a micropipette puller and fire-polished. Neurons were recorded in the whole-cell configuration using an EPC-10 amplifier (HEKA Instruments, Lambrecht, Germany). After whole-cell configuration was established, the cell membrane capacitance and series resistance were electronically compensated. All experiments were performed at room temperature (ϳ25°C). Signals were filtered at 1 kHz, digitized at 10 kHz, and acquired using the Pulse program (HEKA Instruments). The whole-cell Ca 2ϩ current, carried by barium (I Ba ), was recorded using an extracellular solution consisting of 140 mM N-methyl-D-glucamine, 2 mM MgCl 2 , 3 mM BaCl 2 , 10 mM glucose, and 10 mM HEPES (pH 7.4 adjusted with HCl; osmolarity of 320 mosM). In some recordings, N-methyl-D-glucamine was replaced with the same concentration of TEA. The pipette internal solution contained 120 mM CsCl, 1 mM MgCl 2 , 10 mM HEPES, 10 mM EGTA, 4 mM MgATP, and 0.3 mM NaGTP (pH 7.2 adjusted with CsOH; osmolarity of 300 mosM). I Ba was elicited by a series of command potentials from Ϫ70 to 50 mV for 150 ms in 10-mV steps (5-s intervals) from a holding potential of Ϫ90 mV (12). T-type I Ba was recorded by depolarizing the cells from Ϫ90 to Ϫ45 mV for 150 ms (12). The steady-state inactivation of VACCs was measured by depolarizing cells to a series of prepulse potentials from Ϫ100 to 20 mV for 2500 ms in 10-mV steps (15-s intervals), followed by a command potential to 0 for 150 ms.
Recording of Excitatory Postsynaptic Currents in the Spinal Cord-Rats were anesthetized with 2-3% isoflurane, and the lumbar segment of the spinal cord was removed through laminectomy. The spinal tissue was immediately placed in icecold sucrose artificial cerebrospinal fluid presaturated with 95% O 2 and 5% CO 2 . The sucrose artificial cerebrospinal fluid contained 234 mM sucrose, 3.6 mM KCl, 1.2 mM MgCl 2 , 2.5 mM CaCl 2 , 1.2 mM NaH 2 PO 4 , 12.0 mM glucose, and 25.0 mM NaHCO 3 . The tissue was then placed in a shallow groove formed in a gelatin block and glued onto the stage of a vibratome. Transverse spinal cord slices (400 m) were cut in the ice-cold sucrose artificial cerebrospinal fluid and preincubated in Krebs solution oxygenated with 95% O 2 and 5% CO 2 at 34°C for at least 1 h before they were transferred to the recording chamber. The Krebs solution contained 117.0 mM NaCl, 3.6 mM KCl, 1.2 mM MgCl 2 , 2.5 mM CaCl 2 , 1.2 mM NaH 2 PO 4 , 11.0 mM glucose, and 25.0 mM NaHCO 3 .
Excitatory postsynaptic currents (EPSCs) were recorded using the whole-cell voltage-clamp method, as we described previously (14,15). The slice was perfused continuously with Krebs solution (5.0 ml/min) at 34°C maintained by an in-line solution heater and a temperature controller. Lamina II neurons in the spinal slice were identified using a fixed-stage microscope with differential interference contrast/infrared illumination. EPSCs were recorded at a holding potential of Ϫ60 mV with an electrode (impedance of 5-8 megaohms) filled with the following internal solution: 135.0 mM gluconate, 5.0 mM TEA, 2.0 mM MgCl 2 , 0.5 mM CaCl 2 , 5.0 mM HEPES, 5.0 mM EGTA, 5.0 mM MgATP, 0.5 mM NaGTP, and 10 mM QX314. The solution was adjusted to pH 7.2-7.4 with 1 M KOH (osmolarity of 290 -300 mosM). EPSCs were evoked by electrical stimulation (0.2 ms, 0.3-0.6 mA, and 0.2 Hz) through a bipolar tungsten electrode placed on the dorsal root zone. Monosynaptic EPSCs were identified on the basis of the constant latency of evoked EPSCs and the absence of conduction failure of evoked EPSCs in response to a 20-Hz electrical stimulation.
Recording of End-plate Potentials-To assess the effect of 4-AP on neuromuscular transmission, we recorded end-plate potentials (EPPs) using a phrenic nerve-diaphragm preparation (16). Rats were anesthetized with isoflurane, and the diaphragm and the attached phrenic nerve were removed rapidly and pinned in a Sylgard-lined 35-mm Petri dish. The phrenic nervediaphragm was superfused with oxygenated Ringer's solution containing 116 mM NaCl, 5 mM KCl, 2 mM CaCl 2 , 1 mM MgSO 4 , 1 mM NaH 2 PO 4 , 23 mM NaHCO 3 , and 11 mM glucose (pH adjusted to 7.2-7.3), and the solution was continuously gassed with 95% O 2 and 5% CO 2 . Muscle contraction was selectively blocked with 2.3 M -conotoxin GIIIB (Bachem, King of Prus-sia, PA), which preferentially blocks muscle-specific voltage-activated Na ϩ channels (17). Intracellular sharp-electrode recording was performed using glass microelectrodes (10 -15 megaohms, filled with 3 mM KCl) at 25°C. Microelectrodes were lowered slowly using a manipulator until EPPs of muscle fibers were recorded. EPPs were evoked with supramaximal stimuli applied to the phrenic nerve via a suction electrode. EPP signals were processed by a Multiclamp 700B amplifier (Molecular Devices, Foster City, CA). The areas under the curve of evoked EPPs were integrated and analyzed before and during drug application.
Data Analysis-The Ca 2ϩ and K ϩ current data were analyzed using the PulseFit software program (HEKA Instruments). Whole-cell Ca 2ϩ current-voltage (I-V) relationships for individual neurons were constructed by calculating the mean peak inward current at each test potential. Conductance-voltage (G-V) curves were calculated by dividing the current at each potential by the driving force (V Ϫ V r ), where V is the test potential and V r is the reversal potential extrapolated from the I-V curve. The normalized conductance (G/G max ) for G-V relationships and inactivation curves were fit with the Boltzmann function: where G min is the minimal conductance of VACCs, G max is the maximal conductance, V 0.5 is the voltage for 50% activation or inactivation of VACCs, and k is a voltage-dependent slope factor. The percent augmentations of total I Ba and subtypes of I Ba were calculated as the ratio of 4-AP-augmented I Ba to the total peak I Ba or VACC subtype currents during control, respectively. The effects of 4-AP on EPPs and EPSCs were analyzed using Clampfit (Axon Instruments). Statistical data are presented as means Ϯ S.E. All comparisons between means were tested for significance using Student's paired or unpaired t test or oneway analysis of variance unless indicated otherwise. p Ͻ 0.05 was considered to be statistically significant.

4-AP Potentiates VACC Currents in Dissociated Neurons-
The whole-cell VACC currents in DRG neurons were elicited by a series of depolarizing pulses (from Ϫ70 to 50 mV for 150 ms in 5-mV increments) from a holding potential of Ϫ90 mV. Bath application of 4-AP (5 mM) caused a large increase in I Ba in all the neurons tested (Fig. 1A). The effect of 4-AP was rapid and completely washed out in 3 min. To further characterize the effect of 4-AP on I Ba , we examined the effect of 4-AP on the steady-state activation and inactivation kinetics of I Ba . Although 5 mM 4-AP did not significantly alter the current-voltage relationship, it significantly shifted the steady-state activation to the left (n ϭ 16) (Fig. 1B). In nine additional neurons, the steadystate inactivation of I Ba was examined through a series of prepulse potentials (Ϫ90 to 10 mV for 500 ms), followed by depolarization of the cell to 0 mV for 150 ms. Following application of 5 mM 4-AP, the voltage-dependent steady-state inactivation of I Ba was shifted significantly to the left (i.e. more negative potentials) (Fig. 1C). Because 4-AP is a well known Kv blocker, we also compared the concentration-effect relationship of 4-AP on I Ba and Kv channel currents in DRG neurons. At the same concentrations (1-5 mM) that blocked Kv channels, 4-AP also significantly potentiated I Ba (Fig. 2). 4-AP had a 50% maximal effect on Kv channels and VACCs at the same concentration (ϳ2.5 mM). When HVACCs were blocked with 100 M Cd 2ϩ , subsequent application of 4-AP failed to increase I Ba (n ϭ 6) (Fig. 3A).
To determine whether the potentiating effect of 4-AP on VACCs is tissue-specific, we examined the effect of 4-AP on I Ba in dissociated SCG and hippocampal neurons. As in DRG neurons, 5 mM 4-AP substantially increased the amplitude of I Ba in hippocampal neurons (n ϭ 9) (Fig. 1D). However, the potentiation of I Ba by 5 mM 4-AP in SCG neurons was relatively small (n ϭ 17) (Fig. 1D) compared with that in DRG and hippocampal neurons.
4-AP Increases N-and L-type, but Not T-type, I Ba -To determine whether 4-AP affects the N-and L-type VACCs, the specific Ca 2ϩ channel blockers nimodipine (5 M, L-type), -conotoxin GVIA (2 M, N-type), -agatoxin IVA (100 nM, P/Q-type), and -conotoxin MVIIC (500 nM, N-and P/Q-type) were selectively combined to define Land N-type Ca 2ϩ channels (18,19). In this protocol, the effect of 5 mM 4-AP was initially tested without blockers to determine the augmentation of total I Ba . After the initial effect of 4-AP was washed out, a series of grouped blockers was used to isolate defined subtypes of HVACC currents before the effect of 4-AP was re-examined. 4-AP significantly increased N-type I Ba (isolated using nimodipine and -agatoxin IVA; n ϭ 6) and L-type I Ba (isolated by co-application of -conotoxin GVIA, -conotoxin MVIIC, and -agatoxin IVA; n ϭ 6) (Fig. 3, B and C). However, 4-AP did not significantly affect T-type I Ba (Fig. 3D). Blocking Voltage-activated K ϩ Channels Does Not Mimic or Alter the Effect of 4-AP on I Ba -Because 4-AP is a well known A-type Kv channel blocker, we determined whether the effect of 4-AP on HVACCs is mediated by its effect on Kv channels in DRG neurons. When I Ba was recorded in the N-methyl-D-glucamine external solution, bath application of 140 mM TEA, a blocker selectively acting on slow transient outward K ϩ currents (20), alone had no significant effect on I Ba . Even in the presence of TEA, 5 mM 4-AP produced a large potentiation of I Ba (n ϭ 6) (Fig. 4A). 4-AP blocks many Kv subtypes that constitute A-type Kv currents (21)(22)(23). In DRG neurons, A-type Kv currents are mediated primarily by Kv1.4, Kv3.4, and Kv4.3 (13,24). Thus, to test whether blocking Kv channels mimics or alters 4-AP-induced potentiation of I Ba in DRG neurons, we applied 100 nM ␣-dendrotoxin, a specific Kv1 blocker (25,26); 1 M phrixotoxin-2, a specific Kv4.2/Kv4.3 blocker (27); or 1 M BDS-I (blood-depressing substance-I), a specific Kv3.4 blocker (28). After testing and confirming the initial 4-AP effect on I Ba , these Kv channel blockers were then applied individually to the  same DRG neurons. None of these A-type Kv blockers alone had any significant effect on I Ba . The potentiating effect of 4-AP on I Ba was not altered by ␣-dendrotoxin (n ϭ 8), phrixotoxin-2 (n ϭ 9), or BDS-I (n ϭ 9) in any of the DRG neurons tested (Fig.  4, B and C).
It has been reported that 4-AP may increase Ca 2ϩ currents through calmodulin-dependent protein kinase II in cardiomyocytes (29). However, the calmodulin-dependent protein kinase II inhibitor KN-93 used in that study is also a broad-spectrum Kv channel blocker (30). Pretreatment with 2 M of KN-93 significantly decreased the base-line I Ba in DRG neurons, but it did not significantly alter the potentiating effect of 4-AP on the amplitude of I Ba (n ϭ 6, p Ͼ 0.05) (Fig. 4D). When 10 M KN-93 was included in the pipette solution, moreover, it also failed to alter the effect of 4-AP on I Ba (n ϭ 7).

Potentiation of HVACCs by Other Aminopyridines and 4-AP Analogs-
We next determined the structurefunction relationship of the effects of aminopyridines on HVACCs. The concentration-response effects of 4-AP and other aminopyridines on I Ba in DRG neurons were compared. Different concentrations of aminopyridines were applied in a random order. The cells were depolarized to Ϫ10 mV for 150 ms from a holding potential of Ϫ90 mV. Although 2-aminopyridine, 3-aminopyridine, and 3,4-DAP all significantly increased the amplitude of I Ba , 4-AP wasmostefficaciousinthepotentiating I Ba (Fig. 5, A and B).

4-AP Increases Synaptic Transmission in the Spinal Cord Independent of Kv Channels-
To determine whether the Kv channels are involved in the potentiating effect of 4-AP on glutamate release from sensory nerve terminals, we examined the effect of 4-AP on glutamatergic EPSCs in the spinal cord evoked from the dorsal root. At concentrations between 0.1 and 2 mM, 4-AP caused concentration-dependent increases in the amplitude of monosynaptic EPSCs (Fig. 7A). To block the Kv channels, we used two broad-spectrum A-type Kv channel blockers, ␣-dendrotoxin (100 nM) for Kv1.1, Kv1.2, Kv1.4, and Kv1.6 (21) and CP339818 (3 M) for Kv4.2 and Kv4.3 (32). In DRG neurons, 100 nM ␣-dendrotoxin and 3 M CP339818 diminished Kv channel currents. Bath application of ␣-dendrotoxin and CP339818 to the spinal cord slice only slightly increased the amplitude of EPSCs. However, ␣-dendrotoxin and CP339818 failed to significantly attenuate the potentiating effect of 0.5 mM 4-AP on EPSCs in all of the spinal dorsal neurons examined (Fig. 7B). Blocking the glutamate ␣-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors with 20 M 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) completely abolished EPSCs at the end of the experiment.

4-AP Augments Neuromuscular Transmission Independent of Kv
Channels-We investigated whether the potentiating effect of 4-AP on acetylcholine release from motor nerve terminals involves Kv channels. At 0.5 mM, 4-AP substantially prolonged the duration of cholinergic EPPs with a small increase in their amplitude (n ϭ 10) (Fig.  8A). ␣-Dendrotoxin (100 nM) and CP339818 (3 M) were applied to determine the contribution of Kv channels to the effect of 4-AP on EPPs. ␣-Dendrotoxin and CP339818 alone induced only a small increase in the peak amplitude, but not the duration, of EPPs. Even when applied in the presence of ␣-dendrotoxin and CP339818, 0.5 mM 4-AP still profoundly increased the duration and amplitude of EPPs (n ϭ 11). The potentiating effect of 4-AP on the size of EPPs, quantified using the area under the curve, was not significantly altered by ␣-dendrotoxin and CP339818 (Fig. 8B). Bath application of the nicotinic acetylcholine receptor antagonist D-tubocurarine (5 M) blocked EPPs at the end of the experiments.

DISCUSSION
HVACCs at the presynaptic terminals are essential for synaptic and neuromuscular transmission. The potentiating effects of aminopyridines on neurotransmitter release are believed to result from blocking Kv channels, which subsequently leads to prolongation of the action potential, followed by activation of HVACCs. However, findings from our study directly challenge this conventional view by demonstrating that 4-AP and many of its analogs directly stimulate HVACCs in dissociated neurons. We found that 4-AP, at the same concentration that blocked Kv channels, significantly increased the amplitude of I Ba in neurons isolated from the DRG, SCG, and hippocampus. We also found that 4-AP significantly augmented N-and L-type, but not T-type, VACCs. 4-AP-induced augmentation of I Ba is likely mediated by HVACCs because 4-AP failed to produce any effect on I Ba when HVACCs were blocked with Cd 2ϩ . Under our recording conditions, K ϩ was removed from both internal and external solutions. Thus, the potentiating effect of 4-AP on HVACCs is not associated with K ϩ ions permeating through Kv channels. Furthermore, TEA, a Kv channel blocker acting mainly on slow transient outward K ϩ currents (20), did not mimic or attenuate the potentiating effect of 4-AP on HVACC currents. We found that three highly specific A-type Kv channel blockers, ␣-dendrotoxin (Kv1), BDS-I (Kv3.4), and phrixotoxin-2 (Kv4.2 and Kv4.3), all failed to mimic or alter the effect of 4-AP on HVACCs. Collectively, our data provide strong evidence that 4-AP potentiates HVACCs independent of Kv channels in neurons.
By determining the structurefunction relationship of aminopyridines on I Ba , we found that 4-AP (the amine at position 4 of the pyridine ring) is the aminopyridine that is most effective in stimulating HVACCs. Although 3,4-DAP shows a greater potency than 4-AP in blocking Kv channels (33), 4-AP seems to be superior to 3,4-DAP in the treatment of motor dysfunction in patients with multiple sclerosis (34). Consistent with this clinical observation, we found that 4-AP produced a greater effect than 3,4-DAP on HVACCs. By testing the effects of 4-AP analogs on I Ba , we demonstrated that 4-(aminomethyl)pyridine, 4-(methylamino)pyridine, and 4-di(methylamino)pyridine potentiated HVACCs with a similar or greater potency than 4-AP. Therefore, our findings suggest that 4-AP can be derivatized to yield compounds more effective in stimulating HVACCs. 4-AP analogs such as 4-di(methylamino)pyridine may be more efficacious than 4-AP in the treatment of neuromuscular dysfunctions.
To further identify the subunit(s) involved in the potentiating effect of 4-AP on HVACCs, we determined the effect of 4-AP on N-type I Ba reconstituted by expressing ␣ 1 B, ␤ 3 , and ␣ 2 ␦ subunits in HEK293 cells. We found that 4-AP similarly produced a profound effect on the current amplitude of N-type, but not low voltage-activated T-type (Cav3.2), VACCs. The major pharmacological and functional properties of the N-type HVACCs are determined by the pore-forming ␣ 1 B subunit (35,36). In our study, 4-AP had only a small effect on I Ba when ␣ 1 B was expressed alone or with the ␣ 2 ␦ subunit. Strikingly, 4-AP profoundly potentiated I Ba when the ␤ 3 subunit was coexpressed with ␣ 1 B or ␣ 1 B plus ␣ 2 ␦. Therefore, the intracellular ␤ 3 subunit is the most important site of the HVACCpotentiating activity of 4-AP. The function of T-type VACCs does not require any ␤ subunit, which explains the lack of effect of 4-AP on T-type I Ba in dissociated neurons and HEK293 cells. The ␤ subunit of HVACCs affects the channel function mainly by increasing the current amplitude and by hyperpolarizing the voltage dependence of activation. The basis for these effects of the ␤ subunit is likely an increased number of channels expressed on the plasma membrane (37). We observed that 4-AP induced a hyperpolarizing shift in steady-state activation and inactivation, suggesting that HVACCs may recover more readily from inactivation and can be opened more effectively by the depolarization in the presence of 4-AP. Hence, the effects of 4-AP on the amplitude and activation and inactivation kinetics of I Ba in DRG neurons and HEK293 cells consistently suggest an important role for the ␤ subunit in 4-AP-induced stimulation of HVACCs.
Four ␤ subunits of HVACCs have been identified: ␤ 1 , ␤ 2 , ␤ 3 , and ␤ 4 . The ␤ 1 subunit appears to be the only ␤ subunit expressed in skeletal muscle (38). In contrast, only the ␤ 2 subunit is present in rat heart (39). The ␤ 3 subunit is strongly expressed in smooth muscle and brain, and the ␤ 4 subunit is the predominant subunit in the cerebellum (39). The presence of the ␤ 3 subunit in sensory neurons is suggested by the finding that knock-out of the ␤ 3 subunit in mice impairs sensory processing (40). We found that 4-AP had a profound effect on HVACCs in DRG and hippocampus neurons. On the other hand, 4-AP only slightly increased I Ba in SCG neurons, possibly because of the limited expression of the ␤ 3 subunit in the sympathetic neurons. Consistent with our findings, the side effects of 4-AP are generally limited to the gastrointestinal tract and central nervous system with little effect on the autonomic nervous system in humans (41,42). Thus, the magnitude of the potentiating effect of 4-AP on HVACCs varies among different populations of neurons, likely because of heterogeneous distribution of the ␤ 3 subunit. Further studies are warranted to determine the potential roles of other ␤ subunits in the effect of 4-AP on different subtypes of HVACCs.
Because the effects of 4-AP on HVACCs occur at mM concentrations, it could be argued that this mechanism of action may not be pertinent to its effect on neurotransmitter release and neuromuscular function that emerges at M concentrations. Notably, the blocking effect of 4-AP on Kv channels also requires mM concentrations, as we found in dissociated neurons. To determine whether Kv channels play a role in the effect of 4-AP on neurotransmitter release at the nerve terminals, we recorded glutamatergic EPSCs evoked from sensory nerve terminals in the spinal cord and cholinergic EPPs elicited by stimulation of the phrenic (motor) nerve. In both preparations, although blocking Kv channels slightly increased the amplitude  of EPSCs and EPPs, it did not mimic or alter the large potentiating effects of 4-AP on EPSCs and EPPs. These findings provide further evidence that 4-AP can promote neurotransmitter release at the presynaptic terminals independent of its effect on Kv channels. Although it is not clear why 4-AP stimulates the nerve terminals at a much lower concentration than is required to potentiate HVACCs at the soma, it is possible that there is a strong expression of the ␤ 3 subunit and/or a very close association between HVACCs and vesicles at the nerve terminals (43,44).
In summary, our study provides novel evidence that 4-AP and its analogs can directly stimulate HVACCs in neurons independent of Kv channels. Furthermore, the intracellular ␤ subunit is the critical site where 4-AP acts to increase the function of HVACCs. We also demonstrated that 4-AP enhances synaptic and neuromuscular transmission independent of Kv channels. This new information is essential to our understanding of the molecular mechanism of the therapeutic actions of aminopyridines. Identification of the ␤ 3 subunit as the target for the effect of 4-AP on HVACCs has important therapeutic implications because this subunit could be targeted for further development of more efficacious drugs to treat neuromuscular dysfunction. Our findings directly challenge the long-held view that aminopyridines such as 4-AP and 3,4-DAP produce their effects on neurotransmitter release and neuromuscular function through blocking Kv channels. 4-AP is still being widely used as a specific Kv channel blocker to increase intracellular Ca 2ϩ and neurotransmitter release. These 4-AP actions should be reinterpreted on the basis of this new information.