Current Topics in Medicinal Chemistry 2005, 5, 15-30 15

Voltage Gated ion Channels: Targets for Anticonvulsant Drugs

Adam C. Errington1,3, Thomas Stöhr2 and George Lees1,3,*

1
Department of Pharmacology and Toxicology, Otago School of Medical Sciences, University of Otago, PO Box 913,
Dunedin, New Zealand, 2Schwarz Biosciences GmBH, Alfred Nobel Strasse 10, 40789 Monheim am Rhein, Germany,
3
Sunderland Pharmacy School, University of Sunderland, Chester Road Campus, Sunderland SR1 3SD, UK

Abstract: Epilepsy is one of the most prevalent neurological syndromes in the world today. Epilepsy describes a group of
brain disorders whose symptoms and causes are diverse and complicated, but all share a common behavioural
manifestation: the seizure. Seizures result from the abnormal discharge of groups of neurons within the brain, usually
within a focal point, that can result in the recruitment of large brain regions into epileptiform activity. Although the range
of explanations for the development of seizures can be as varied as genetic composition to acute head trauma, the net
result is often similar. The excitability of neurons is governed by the input they receive from their neighbours and the
intrinsic excitability of the neuron. In this review we focus on elements that are crucial to determining the intrinsic
excitability of neurons in the CNS, the voltage gated ion channels (VGICs).
VGICs as well as being important for physiological function are critical in producing hyperexcitability such as that
associated with seizure discharges. Many drugs routinely used in the clinical setting, as well as several novel experimental
drugs, have shown interactions with VGICs that underpin, at least in part, their anticonvulsant action. We review the
physiological roles of voltage gated ion channels that are selective for sodium, potassium and calcium conductance and
attempt to highlight their role in the pathology of epilepsy. This is supplemented by the mechanisms of drug action at
these important anticonvulsant targets for classical and clinically relevant compounds (e.g. phenytoin, ethosuximide) as
well as some important second generation drugs (e.g. gabapentin, levetiracetam) and novel experimental agents (e.g.
retigabine, losigamone, safinamide). We also briefly discuss the urgent need for new drugs in this arena and the potential
of combinatorial methods and recombinant screening to identify leads.
Key Words: Voltage gated ion channels; epilepsy; anticonvulsants

INTRODUCTION the electrogenic properties intrinsic to each neuron. These

Epilepsy is one of the most prevalent neurological
disorders with current estimates approximating between 0.5
– 2% of the global population being affected. The use of the
term epilepsy is to some extent misleading as it actually
refers to a collection of neurophysiological disorders that are
diverse in terms of both their etiology and symptomology.
The typical behavioural manifestation common to all forms
of epilepsy is the aberrant synchronized discharge of
neuronal populations termed the ‘seizure’. Seizures result
from phasic changes in the firing properties of groups of
neurons, usually within a discrete focal point, to an
intermittent high-frequency burst-firing mode. Induction of a
population of neurons into a pattern of burst-firing resulting
in synchronized disruptive seizure discharges is determined
by changes in excitability of neurons that are essentially
governed by two major factors. The first determinant of
neuronal excitability in any given neuron, is the balance of
synaptic input from neighboring cells within the network in
the form of chemical transmission (through ligand gated ion
channels and G-protein coupled metabotropic receptors) and
aplastic but extremely rapid electrical transmission (through
gap juctions). Secondly, and perhaps most importantly are
*Address correspondence to this author at Chair & Head of Department of
Pharmacology & Toxicology, Otago School of Medical Sciences, University
of Otago, PO Box 913, Dunedin, New Zealand; Tel: 64 3 4797265; Fax: 64
3 4799777; E-mail: george.lees@stonebow.otago.ac.uk
1568-0266/05 $50.00+.00
are determined to a great extent by the complement and
localization of functional voltage gated ion channels
expressed in the cell membrane [1]. It is intuitive, therefore,
that these determinants of physiological excitability would
be crucially involved in the pathological hyperexcitability
occurring during the seizure discharges characteristic of the
epilepsies. Indeed both excitatory and inhibitory neurotrans-
mission as well as voltage-gated channels (that can be
modulated selectively by natural or synthetic toxins), have
been shown to be crucial for seizure like activity in
acute chemoconvulsant models in rodent brain slices and in
vivo [2].
It is not surprising, therefore, that many of the drugs used
in the treatment of seizures and epilepsy target the channels
and receptors that are vital for the initiation and spread of
seizure activity. In this review we focus specifically upon the
voltage gated ion channels, discussing their function both
physiologically and pathologically and evaluate their
relevance as anticonvulsant drug targets by outlining how
they can be manipulated for therapeutic benefit.
ELECTROPHYSIOLOGICAL PROPERTIES AND
MOLECULAR BIOLOGY OF VOLTAGE GATED
SODIUM CHANNELS (VGSCS)
The action potential is the cellular mechanism by which
excitable cells within the nervous system communicate with
one another. Action potentials propagating in large
© 2005 Bentham Science Publishers Ltd.
16 Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1
pyramidal cells of the cortical surface of the brain constitute
the unitary events in an EEG. At the cellular level, using the
newly developed voltage clamp technique in the 1950s
Hodgkin and Huxley showed that a transient voltage
activated sodium current (INa) was responsible for the rapid
rising phase of this crucial intracellular communication
process. Although the techniques required to elucidate the
molecular correlate for this current were not available in the
1950s these seminal studies identified three essential features
that have come to characterize the voltage gated sodium
channel (VGSC) family, namely: voltage dependent
activation, rapid inactivation and selective ion conductance
[3-5]. Following this series of elegant and pioneering
experiments a large number of detailed electrophysiological
and molecular studies have elucidated much about the
structure and function of the channel(s) responsible for INa.
Molecular biology has determined the major structural
component of the channel to be a protein of approximately
260 kDa, called the α-subunit, which has four repeated
(named I-IV) structural motifs consisting of six α-helical
transmembrane spanning domains separated by intra and
extracellular loops [6]. These four repeated domains fold
together to form a central pore and it is their structural
components that determine selectivity and conductance of
the ion channel. Although α subunits alone can form
functional channels co-expression of accessory β (33 kDa)
subunits is most commonly required to confer the correct
gating properties found in native channels. Channels formed
purely from α subunits exhibit slower inactivation and
positive shifts in the voltage dependence of activation
compared to native neurons when expressed in Xenopus
oocytes. To date three β (β1−3) accessory subunits
have been cloned from rat and human and their effects
upon channel gating have been extensively characterised
[6-8].
Based upon sequence homology, several different sodium
channel α subunits, recently named by IUPHAR as NaV1.1-
Table 1. Voltage Gated ion Channel Genes Bearing Mutations which have been Linked to Inherited Epilepsy.
Voltage-gated ion channel family Gene name
Potassium channels KCNQ2
KCNQ3
KCNQ3
Sodium channels SCN1A
SCN1A
SCN2A
SCN2A
SCN1B
Calcium channels CACNA1A
CACNA1H
BFNC: Benign familial neonatal convulsions
BFNIC: Benign familial neonatal/infantile convulsions
CAE: Childhood absence epilepsy
GEFS+: Generalized epilepsy with febrile seizures plus
IGE: Idiopathic generalised epilepsy
JME: Juvenile myoclonic epilepsy
SMEI: Severe myoclonic epilepsy of infancy
Lees et al.
1.9 [9], have been characterised. cDNA clones have been
isolated from rat brain cDNA libraries and show that the
major central nervous system (CNS) channels are encoded
for by four genes entitled Scn1a, Scn2a, Scn3a and Scn8a [9-
11]. The corresponding protein products of these genes (or
splice variants of them) form the major brain sodium channel
α-subunits classified as NaV1.1-1.3 which are blocked by the
puffer fish ‘fugu’ poison (Tetrodotoxin, TTX). Several
mutations of the Scn1a and the Scn2a genes have been
identified in individuals suffering from monogenetic epilepsy
syndromes such as generalized epilepsy with febrile seizures
plus and severe myoclonic epilepsy of infancy (table 1). The
gene Scn8a encodes the primarily CNS Nav 1.6 α-subunit
(formerly SkM2) and a further CNS channel resistant to
TTX, Scn5a (Nav1.5, formerly NaCh6), has also been
described. Crystal structures for these large homo-oligomeric
pores remain elusive but the fine structure of the channel
complex has been determined by means of cryo-electron
microscopy and image reconstruction analysis. This reveals
water filled pores (thought to contain the gating charges)
deep within the membrane plane and multiple intracellular
vestibules which constitute a likely pathway for the permeant
ions [12].
Functionally, VGSCs can be simply thought to exist in
three conformational states and it is the transition between
these states that allows selective and temporally regulated
ionic conductance. At the resting potential of most CNS
neurons, VGSCs can be found in a closed non-conducting
state commonly referred to as the ‘resting’ state (Fig. 1C). In
response to change in membrane potential, such as that
provided by excitatory synaptic input [13], the channels
undergo a physical conformational change. Activation occurs
within a fraction of a millisecond and as predicted by
Hodgkin & Huxley [14] is due to movement of gating
charges within the membrane electric field. These so called
‘gating currents’ have been measured electrophysiologically
[15,16] and attributed to sequences of positively charged
Syndrome Reference
BFNC [113,119]
BFNC [119,120]
JME [121]
GEFS+ [122-124]
SMEI [125,126]
GEFS+ [127]
BFNIC [128]
GEFS+ [129]
IGE [130]
CAE [131]
Voltage Gated Ion Channels: Targets for Anticonvulsant Drugs
Fig. (1). (A) Structures of anticonvulsant drugs that act upon VGSCs. (B) Diagram of the subunit structure of VGSC α and β subunits
showing the pore forming region, the voltage sensor and inactivation gate. (C) Gating conformations of the voltage gated subunit.
arginine and lysine residues in the S4 regions of the α
subunit [6] (Fig. 1B). The movement of the S4 region
perpendicular to the plane of the membrane results in
opening of the activation gate and allows sodium flux. For
the VGSC it was estimated that the movement of
approximately six charges across the entire membrane would
be required for gating [7]. Neutralization of the charged
residues in the S4 region markedly alters the voltage
dependence of activation [17]. Upon opening of the
activation gate, the channel pore (formed by a re-entrant loop
between transmembrane segments V and VI) selectively
conducts sodium ions along their electrochemical gradient
from the extracellular space to the cell interior thus
depolarizing further the neuronal transmembrane potential.
However, within a few milliseconds of opening of the
activation gate, sustained depolarization results in
termination of the sodium conductance by a process known
as inactivation. Inactivation usually (but not always, as
discussed below) occurs rapidly, attenuating the majority of
the sodium current and for this reason is often referred to as
‘fast’ inactivation. The process of fast inactivation has been
strongly linked to the short intracellular loop between
domains III and IV (the so-called III-IV linker) [18,19] (Fig.
1B) which is also known as the inactivation gate.
Enzymatically cutting this loop [17], deleting amino acids in
the linker [18], or directing antibodies against it [19], all
produce a marked slowing or disruption of the inactivation
process. It is hypothesised that the intracellular portion of the
protein may bind to a so called ‘inactivation gate receptor’
on the cytoplasmic side of the IVS6 domain using a structural
motif comprising a hydrophobic group of isoleucine,
phenylalanine and methionine (IFM) [20,21] (Fig. 1C).
Entry of sodium channels into the fast inactivated-state is
highly voltage dependent and becomes significantly more
pronounced upon depolarization. The ability of channels to
recover from the fast inactivated state is dependent upon
Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1 17
membrane potential and time and is a mechanism that
ensures adequate time for recovery before reopening of the
channel [22].
The properties of rapid activation and inactivation of
VGSCs are essential determinants in neuronal firing
frequency. As such they are also crucial for signal
transmission in neuronal networks including recruitment of
oscillatory activity of putative importance for both cognition
and consciousness (i.e. γ frequency) [23]. However the
ability of sodium channels to rapidly cycle through conform-
ational states and facilitate high frequency neuronal firing is
also indispensible for the generation of synchronous bursting
activity that is characteristic of epilepsy. Indeed computer
models of normal and disruptive (epileptic) neuronal
network oscillations require the inclusion of the fast transient
sodium current (INa) carried by neuronal VGSCs [1,23].
As previously stated, sodium channel inactivation is
often thought of as a single process but it is increasingly
clear that there are at least two distinct kinetic classes of
inactivation (ultra-slow activation is discussed later). The
second type of sodium channel inactivation, termed slow, is
postulated as a cellular back-up mechanism that prevents INa
from becoming fully available rapidly. This type of
inactivation appears to be independent of the III-IV linker
(crucial for fast inactivation) and has recovery time constants
in the order of several hundreds of milliseconds to seconds
[22]. Recruitment of VGSCs into the slowly inactivated
conformation appears to proceed only under conditions of
prolonged or high frequency depolarization [24-26].
Furthermore, there is an apparent proportional relationship
between the time taken to recover from the slowly
inactivated state and the duration of depolarizing challenge
[27]. Type II and IIA sodium channels expressed in the
Xenopus oocyte expression system showed the time constant
of recovery from slow inactivation was intrinsically related
18 Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1
to the duration of previous depolarizations by a power law:
τ(t) = p•tD (where τ is the recovery constant for the channel
from slow inactivation, t is the duration of previous
depolarization, p is a constant kinetic setpoint and D a
positive scaling power). This type of inactivation process is
especially pertinent to epileptiform activity as the neuronal
networks involved are often exposed to periods of prolonged
high frequency neuronal activity [28-31] that may be
‘damped down’ by slow inactivation. It appears therefore
that pharmacological enhancement or stabilization of the
slowly inactivated conformation of VGSCs may be useful in
disrupting prolonged neuronal seizure discharges.
Although I Na is the principal sodium current responsible
for neuronal activity the existence of a second type of
sodium current, termed the persistent current (INaP), has been
confirmed in a number of neuronal cell types [32]. This
current, although only representing 1-3% of the total peak
current is thought to be crucial for determination of spike
threshold in neurons [32]. The mechanism behind these
persistent currents is not clearly defined but has been
variously attributed to the failure of a fraction of channels
(responsible for conducting INa) to inactivate or the ability to
switch to an ‘ultra-slow’ inactivating gating conformation
[32-35]. The voltage dependence of activation of INaP differs
from that of INa. Persistent Na+ currents are activated at
potentials below spike threshold (-60mV) and reach peak at
between –40 and –35mV with V50 for activation at around
–50mV [33,36-38]. INaP has been implicated in the fine
control of neuronal excitability based upon the sub-threshold
activation kinetics and lack of rapid inactivation. As
previously highlighted, synchronous neuronal network
activity (like that associated with seizures) is strongly
dependent upon the ability of neurons to fire bursts of action
potentials. INaP has been shown to be crucial in both
hippocampal CA1 neurons and layer V neocortical neurons
in the induction of burst firing modes. Evidence exists that
INaP is also thought to, at least in part, contribute to rhythmic
sub-threshold oscillations that can determine behaviour of
entire neuronal networks in a number of structures including
the neocortex and entorhinal cortex [23]. It appears that INaP
is acting as a mechanism for ‘fine-tuning’ neuronal
excitability. Selective modulation of this current may
represent an attractive target for future anticonvulsant drug
development. It is possible that a ligand selective for INaP
may be able reduce hyperexcitability without the drawbacks
associated with attenuating INa.
A new and exciting putative target for future
anticonvulsant drug intervention has come to light relatively
recently and is based upon a somewhat unexpected form of
plasticity displayed by VGSCs (see refs [7,39] for detailed
review). Biochemical experiments including two dimensional
phosphopeptide mapping have shown purified sodium
channel α subunits have five distinct phosphorylation sites
throughout the glycopeptide. Four of these sites are found on
serine residues (573, 610, 623 and 687) in the large
intracellular loop connecting domains I and II (LI-II) and a
further one is located of the domain III-IV linker which
constitutes the inactivation gate [40]. In intact synaptosomes
the Na + flux induced by veratridine is reduced by circa 26%
when activators of cAMP dependent protein kinase (PKA)
are concurrently applied due to rapid phosphorylation of α
Lees et al.
subunits. In functional studies activation of PKA resulted in
a 25-40% reduction in peak Na+ current mediated by NaV1.2
channels expressed in Xenopus oocytes [41] or CNaIIA-1
[42] (CHO cell derived stable line expressing NaV1.2) cells
without shifting voltage dependence of activation or
inactivation.
To complement this evidence it has also been shown that
protein kinase C (PKC) is able to phosphorylate sodium
channel α subunits. The amplitude of sodium current is
reduced and the rate of inactivation slowed in neuroblastoma
treated with fatty acids such as OAG (1-oleoyl-2-acetyl-sn-
glycerol) or DOG (1,2-dioctanoyl-rac-glycerol) that activate
PKC by mimicking the endogenous second messenger
molecule DAG. Slowed inactivation of sodium channels
produced by activation of PKC results from phosphorylation
of the site situated on the inactivation gate that does not
appear to be targeted by PKA. In Xenopus oocytes
transiently expressing rat brain Na+ channels the same
effects as described above can be achieved using phorbol
esters. In our laboratory we have shown that OAG can
reduce sodium currents in voltage clamped N1E-115
neuroblastoma and also inhibits sustained repetitive firing in
rat cultured cortical neurons (Fig. 2AB).
Furthermore, activation of G-protein coupled receptors,
including the muscarinic acetylcholine (mAChR) receptor
and the dopamine (D1) receptor that are coupled to the PKC
and PKA pathways, respectively, also results in reduction of
Na+ currents. The effects of the reduced Na+ current and
slowing of inactivation produced by activation of PKC are
evident in pyramidal neurons of the hippocampus. Activation
of mAChRs using carbachol converts intrinsically burst
firing CA1 neurons into a pattern of regular spiking [43,44].
This is achieved by a reduction in not only the transient Na+
current but also the persistent current crucial for determining
spike threshold. Similar effects can be seen in hippocampal,
striatal and cortical neurons where activation of the D1
receptor pathway reduces the number of spikes in an evoked
burst.
Neuromodulation of sodium channel function, particularly
somatic channels responsible for spike generation, may have
an important influence upon the firing patterns of many
neurons. The ability to activate PKC or PKA directly by
mimicking endogenous second messengers or indirectly via
GPCRs to produce reduction in both persistent and transient
Na+ currents may prove an exciting target for future AED
design. Many large drug companies now use high-
throughput combinatorial screens to highlight new leads and
kinase effectors are the focus of much attention in a variety
of therapeutic areas. To date, however, there are no known
clinical or experimental agents that exploit such pathways to
produce anticonvulsant effects.
DRUGS ACTING UPON VGSCs
A number of anticonvulsant molecules in the pharma-
copoeia today have primary modes of action that involve
modulation of neuronal VGSCs. The classical compound in
this class was discovered by Merritt and Putnam [45] for its
ability to inhibit maximal electroshock-induced seizures
(MES) in rodents. Since its discovery in 1938, phenytoin
[5,5-diphenylhydantoin, DPH] (Fig. 1A) and the pro-drug
Voltage Gated Ion Channels: Targets for Anticonvulsant Drugs
Fig. (2). (A) Sustained repetitive firing in current clamped cultured rat cortical neurons evoked by somatic current injection. The PKC
activator OAG (20µM) markedly reduces the frequency of action potentials within the burst. (B) Sodium currents from voltage clamped
mouse N1E-115 neuroblastoma cells are attenuated by application of 10µM OAG for five minutes. Cells were voltage clamped at –100mV
and test pulses were delivered to 0mV.
Primidone have become first line treatments for partial and
generalized tonic clonic seizures due to their ability to reduce
seizures without producing marked sedation [46]. Much is
now known about the mechanism(s) of action of DPH and it
has become clear that the biophysical characteristics of its
interaction with sodium channels are central to its efficacy as
an anticonvulsant.
Early evidence upon the mechanism of action of DPH
was obtained in experiments using supramaximal stimulation
of frog sciatic nerve. It was discovered that DPH produced
no effect upon an initial action potential triggered by
stimulation but blocked a second rebound spike [47]. This
appeared to suggest DPH would selectively inhibit high
frequency neuronal firing, a hypothesis that has been
confirmed in several studies since. Notably high frequency
evoked action potentials in mouse spinal cord neurons were
inhibited by DPH without impairment of spontaneous low
frequency action potentials [48].
Further and more extensive voltage clamp studies have
delineated much about the biophysical interaction of DPH
with sodium channels. Several studies using mouse
neuroblastoma have shown that DPH produces tonic block of
the transient sodium current INa and that this block is highly
voltage dependent [49,50]. For example DPH has been
reported to have IC50 = 120µM at –85mV, but IC50 = 10µM
at –60mV [51]. Indeed the block of sodium currents by DPH
in N18 mouse neuroblastoma could be reversed by up to
95% if cells were hyperpolarized to –120mV [49]. It was
predicted that because the block of sodium channels
followed the voltage dependence of inactivation of INa that
DPH inhibition was related to the inactivated state. Indeed
DPH does shift the voltage dependence of steady state
inactivation in the hyperpolarizing direction and thus reduces
the population of channels available for activation at any
Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1 19
given potential [49,50,52]. The binding of DPH to the fast
inactivated conformation of the sodium channels is relatively
slow and this, along with the strong use and voltage
dependence, explains how the molecule is able to selectively
inhibit high frequency epileptiform discharges whilst leaving
normal cellular activity unaffected.
In addition it has been reported that DPH is able to
reduce I NaP in dissociated cortical neurons [53], and that this
translates to a reduction in cellular excitability in layer V
neurons maintained in an intact slice (it is noteworthy that
DPH concentrations were lower than those required to
inhibit INa [54]). Single channel experiments revealed that
DPH preferentially impairs late channel openings rather than
those occurring immediately in response to a voltage step
which may explain the reduction of persistent currents. A
more recent study however disputes this claim showing that
low micromolar concentrations of DPH produced little or no
effect upon INaP [55].
Carbamazepine [5H-Dibenz[b,f]azepine-5-caboxamide,
CBZ] (Fig. 1A) is an iminostilbene derivative of tricyclic
antidepressants (structurally it is completely different to
DPH in that it does not contain the ureide moeity
traditionally thought to be important to anticonvulsant
molecules) [56]. However CBZ demonstrated shared
anticonvulsant properties in vivo with DPH in potently
inhibiting hindlimb extension in the MES seizure test whilst
proving relatively ineffective against pentylenetetrazol (PTZ)
induced epileptiform activity [57]. To complement this pre-
clinical data, since its therapeutic launch CBZ has been
successfully used in the control of generalised (tonic-clonic)
convulsive seizures and partial (focal motor or complex
partial) seizures, a niche common with DPH [58].
Like DPH early clues towards the mechanism of
action of CBZ were derived from its ability (or 10,11-
20 Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1 Lees et al.

epoxycarbamazepine) to inhibit, at clinically relevant
concentrations, repetitive firing in cultured mammalian CNS
neurons [59]. Further and more detailed voltage clamp
studies in mouse neuroblastoma cells [49,52] showed CBZ
produced tonic block of INa that was highly voltage
dependent; with approximately 40% of INa blocked by CBZ
(20µM) at a holding potential of –75mV, but virtually zero
block when held at –120mV [49]. These studies also showed
that, like DPH, the effects of CBZ were use dependent and
block was accumulated with repetitive stimulation (Fig. 3).
CBZ produces shifts in the voltage dependence of steady
state fast inactivation (in the hyperpolarzing direction) by
stabilizing the inactivated conformation of the sodium
channel and thus reduces the ability of the channel to revert
the resting state (Fig. 3). The time constant for recovery from
fast inactivation is also prolonged by CBZ.
Differences in the binding kinetics of DPH and CBZ to
the sodium channel have been described with the latter
binding 5 times faster than the former but with three fold
lower affinity [60]. This may give CBZ an advantage in the
treatment of seizures characterized by comparatively brief
depolarizing discharges and may explain differential
responses within certain patient sub groups to these agents.
Although it is still commonly used in the clinic, toxicological
drawbacks related to CBZ (including hepatic toxicity) have
led to the development of structural congeners that maintain
anticonvulsant efficacy whilst eliminating, to some extent,
the toxicity. Oxcarbazepine [OXC] (Fig. 1A) shares with
CBZ the dibenzazpine nucleus bearing the 5-carboxamide
substituent but is structurally different at the 10,11- position.
It appears that this substitution results in altered
biotransformation when compared to CBZ bypassing the
10,11-epoxide metabolite thought to be responsible for the

Fig. (3). Voltage-gated Na+ channels are important targets for a number of clinically important anticonvulsant drugs including
carbamazepine (100 µM throughout).
A. Voltage clamped currents showing tonic block exerted by CBZ at two different holding potentials (test pulse applied to elicit the peak
inward current at 0.5 Hz). As shown in the graph on the right, the proportion of the current blocked is increased with depolarization (note
that, over the time course of these experiments, the currents had a slight tendency to increase or “run-up” due to dephosphorylation of the
Na+ channel complex). B. The tonic block can be enhanced by activating the channels at higher frequencies across the physiological range
(up to 200Hz). The marked facilitation of block is clearly shown by progressively diminishing currents in the train (overlays are shown in
inset) evoked in this experiment at 10 Hz for 3s. C. Paired pulse protocols (top right) can be used to show that CBZ generates a large
hyperpolarizing shift in the steady-state inactivation profile for the Na+ current and (not shown) it retards recovery from the fast-inactivated
state.
CBZ, LTG and DPH exert a broadly similar effect (exhibiting subtle differences in kinetics, voltage-dependence or frequency dependence).
The concensus interpretation of these data are that the anticonvulsant drugs bind preferentially to (and stabilize) inactivated channel states.
This preferential interaction with deploarised, ectopic and rapidly firing channels (in diseased tissue) underpins the therapeutic ratio of these
clinically effective anticonvulsants.
All experiments illustrated above (Errington and Lees Unpublished) were conducted on NIE115 murine neuroblastoma (in which native
evoked Na+ currents are tetrodotoxin sensitive) . Data are presented as Mean with SEM represented by vertical error bars (.n=3-4).
Voltage Gated Ion Channels: Targets for Anticonvulsant Drugs
toxicity associated with the parent molecule [61]. Further
modification has led to (S)-(-)-(10)-10-acetoxy-10,11-
dihydro-5H-dibenz/b,f/azepine-5-carboxamide (BIA 2-093)
(Fig. 1A) which was found to be particularly effective in
preventing MES induced seizures with similar potency to
CBZ but with fewer cognitive and motor deficits. Like CBZ
and OXC, BIA 2-093 has been shown to inhibit VGSCs in a
frequency and voltage dependent manner in N1E-115 cells
confirming a shared mechanism of action with the former
[62]. Secondary to the state dependent block of VGSCs
CBZ, OXC and BIA 2-093 have all been shown to inhibit
veratrine induced neurotransmitter (glutamate, aspartate,
GABA, dopamine) release in rat striatal slices.
Lamotrigine [3,5-diamino-6-(2,3-dichloro-phenyl)-1,2,4-
triazine , LTG] (Fig. 1A) is the first of a new generation of
drugs developed in the 1980’s and 90’s to have sodium
channel inhibition as its primary mechanism of action.
Although between 60-80% of patients responded well to
existing agents some patients did not have seizures
effectively controlled. The observation that the folates
possessed convulsant activity in animals led to the screening
of putative antifolates as potential anticonvulsant candidates.
From these screens LTG emerged as the lead compound
showing similar pharmacological properties as CBZ and
DPH [63-65]. Early evidence implicating the VGSC as
the primary target for LTG was derived from neurochemical
studies in which glutamate and GABA release evoked by
veratridine (but not potassium) were suppressed in slices of
rat cortex [64]. This was later confirmed by whole cell patch
clamp [66] studies on cultures of rat cortical neurons in
which LTG reduced the incidence of synaptic events
significantly without affecting evoked postsynaptic
responses. It was also demonstrated that while LTG did not
attenuate single action potentials it did markedly reduce
burst firing [67-68]. Voltage clamp studies in N4TG1 mouse
neuroblastoma confirm that LTG exhibits similar biophysical
interactions with VGSCs compared to CBZ and DPH. LTG
tonically inhibits INa in a concentration dependent manner
from a holding potential of –80mV but like CBZ and DPH
the block is highly voltage dependent and can be reversed by
hyperpolarization. The voltage dependence of steady state
inactivation is also shifted in the hyperpolarizing direction to
a similar extent to CBZ and DPH [52]. It has been proposed
that LTG may be unique in acting on the slow inactivated
state of the channel, rather than classical fast inactivation
gating, but whether this truly represents the physiological
slow inactivated state or slow binding kinetics with the fast
inactivated state is unclear. Like DPH, LTG has been shown
to block INaP in rat neocortical neurons, but the degree of
block is less than 10% at a concentration of 10µM.
Interestingly two derivatives of LTG (sipatrigine, 202W92)
that are putative neuroprotective agents, are able to inhibit
INaP far more profoundly with 70 and 90% inhibition
respectively at 10µM [55].
There is striking and widely accepted evidence that all
(PHT, CBZ & LTG) these anticonvulsants and another group
of drugs that modulate the VGSCs, the local anesthetics
(LAs), do so by binding to a receptor site inside the
cytoplasmic vestibule of the ion conducting pore of the
channel. More specifically if either F1764 or Y1771 are
mutated to alanine in transmembrane segment IVS6 the
Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1 21
resultant sodium channels are far less susceptible to voltage
and frequency dependent block by LAs [69,70]. The
similarity between the blocking action of the LAs and the
anti-epileptic drugs (AEDs) suggested that this binding site
may be shared across both classes of molecules and indeed
the F1764A and Y1771A mutants showed greatly attenuated
block by DPH and LTG [11]. Furthermore Kuo [56] has
shown that applications of mixtures of LTG, DPH and CBZ
cannot produce double occupancy of the channel. He
concludes that despite dissimilarity in structure, these drugs
(and probably CBZ derivatives OXC and BIA 2-093) likely
share the same or very similar receptor site on the sodium
channel α subunit.
Despite the effectiveness of the aforementioned AEDs in
controlling seizures, in patients with generalized convulsive
seizures, some patient groups are still found to be refractory
to existing therapies. This has led to further development of
anticonvulsant compounds despite negative market forces
including long lead times and brief patent life of new
molecules. Some of these newer drugs, discovered by
screening large libraries of compounds in various animal
models, have been shown to act (to some degree) on VGSCs.
Topiramate [2,3:4,5-bis-O-(1-methyl-ethylidene)-36-D-
fructopyranose sulfamate, TPM] (Fig. 1A) is an
anticonvulsant drug which was licensed for clinical use in
many European countries in the late 1990’s. In the traditional
animal models (MES, PTZ) TPM showed a profile similar to
that of DPH [71] and it has therefore become clinically
established as an add-on therapy in simple or complex partial
seizures with or without secondary generalization. The
mechanism of action of TPM, although still unclear appears
to exhibit three facets, namely: blockade of kainate evoked
excitatory currents [72], enhancement of inhibition through
GABAergic chloride channels [73] and depression of sodium
currents [74]. TPM significantly reduced sustained repetitive
firing in cultured hippocampal pyramidal cells at
therapeutically relevant (10-30µM) concentrations and this
was later confirmed to be due to attenuation of INa. In
cerebellar granule cells [75] and dissociated neorcortical
neurons [76], TPM produced voltage dependent tonic block
of I Na and shifted the steady state inactivation curve to more
negative values in a fashion previously described for LTG,
CBZ, DPH, OXC and BIA 2-093. TPM has also been
reported to inhibit sustained sodium conductances (INaP) in
evoked in neocortical neurons in response to slowly
depolarizing membrane potential ramps at concentrations
lower than those required to inhibit INa. In this way TPM
mimics DPH and, furthermore, it is conceivable that the
inhibition of the persistent current may represent a major
contribution to the reduction in cellular excitability produced
by all these drugs.
The tetronic acid derivative Losigamone [(±)-5(R,S)-5-
(2- chlorophenyl)hydroxymethyl) - 4 – methoxy (5H0 -furanone,
LSG] (Fig. 1A) is an experimental anticonvulsant currently
undergoing phase II/III clinical trials in patients with partial
and secondary generalised seizures. LSG has been shown to
produce a reduction in spontaneous synaptic activity in
cultured hippocampal neurons and a marked suppression of
sustained repetitive firing evoked by somatic current
injection [77]. However the effects of LSG are apparently
different from many classical sodium channel-affecting
22 Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1
anticonvulsants previously described. In the majority of
cellular studies inhibition of SRF can be closely correlated to
impairment of INa. In whole cell recordings made from
hippocampal neurons in brain slice preparations LSG did not
inhibit INa at relevant concentrations. LSG did however
produce inhibition of INaP and it has been suggested that this
may prove to be crucial to its ability to decrease neuronal
hyperexcitability [78]. In this respect LSG may represent the
first compound to be used in epilepsy that inhibits INaP
without significant effect on transient sodium currents.
Another new anticonvulsant ligand which acts upon
voltage gated sodium channels is currently in phase III trials.
Safinamide [Propanamide, 2-[[[4-[(3-fluorophenyl)methoxy]
phenyl]methyl]amino]-, (2S)-monomethanesulfonate, PNU-
151774E, NW-1015, SAF] (Fig. 1A) produced inhibition of
sustained repetitive firing, inhibition of TTX sensitive
sodium currents and displayed higher affinity for the Na+
channel [3H] batrachotoxin receptor site (site 2) than
phenyoin or lamotrigine (EC50: 8.2µM) [79]. Although the
actions of SAF appear to be complex and multifaceted,
including inhibition of L and N-type calcium channels, the
compound nonetheless represents another molecule whose
anticonvulsant actions are conferred by interaction with
VGSCs.
Clearly the VGSC represents a major ion channel target
for both existing clinical and experimental anticonvulsant
molecules. It is apparent that despite varied chemical
structures many compounds can interact to produce state
dependent block of INa and that in some cases there is
overlapping reduction of INaP providing a dual mechanism
for reducing hyperexcitability.
STRUCTURE, FUNCTION AND ANTICONVULSANT
PHARMACOLOGY OF VOLTAGE GATED
CALCIUM CHANNELS
The voltage gated calcium channels (VGCCs) are critical
for the proper functioning of excitable cells within the
Fig. (4). (A) Structures of anticonvulsant drugs acting at VGCCs. (B) Diagram of the subunit structure of the VGCC α subunit and the
accessory β, α2 and δ subunits.
Lees et al.
central nervous system. Calcium influx into the cellular
interior occurring in response to channel gating is essential
for the translation of transient electrical signals into
biochemical events, both acutely, as in neurotransmitter
release, and more chronically, as in regulating gene
expression and even cell-death. VGCCs were initially
described by Paul Fatt and Bernard Katz in crustacean
muscle but have subsequently been shown to exist in an
array of mammalian cells types including, crucially, CNS
neurons. VGCCs were first classified according to their
pharmacological and biophysical characteristic into P/Q, N,
L, R (high voltage activated, HVA) and T (low voltage
activated, LVA) types [80]. HVA channels have been
extensively characterised biochemically revealing that
VGCCs are complex heteromeric structures. VGCCs have
been demonstrated to be formed by at least three main
subunits namely α1, α2δ and β, as well as, in some cases a γ
subunit. The main component of the VGCC is the
approximately 2000 amino acid 190-250kDa α subunit. The
VGCC α subunit is structurally similar to VGSC α subunits,
with four repeated domains each being produced by six
transmembrane spanning segments. As with its sodium and
potassium conducting counterparts it is predicted that the
calcium channel voltage sensor is the S4 membrane spanning
segment and the pore is formed by a re-entrant loop between
the S5 and S6 segments (Fig. 4B). Ten VGCC α subunits
have thus far been identified through homology screening
and this has led to reclassification of the different Ca2+
channel types. The cloned subunits are know divided into
three functionally related families based upon their
biophysical and pharmacological characteristics [81].
Cav1.1-1.4 subunits form the L-type channels whilst Cav2.1,
Cav2.2 and Ca v2.3 form the P/Q, N and R types respectively.
The LVA T-type channels include an α subunit from the
Cav3.1-3.3 types [80].
The first of the calcium currents to be described was the
HVA L-type (because the current is long lasting, particularly
Voltage Gated Ion Channels: Targets for Anticonvulsant Drugs
when Ba2+ is used as the permeant ion [82]) current in
smooth and skeletal muscle. The L-type channels are
characterized by slow voltage dependent inactivation, high
voltage of activation, modulation by cAMP dependent
phosphorylation pathways and inhibition by dihydro-
pyridines [83]. The T-type LVA calcium channels that were
discovered in cerebellar purkinje neurons [84] but
characterised in detail in dorsal root ganglion neurons [82].
The T-type designation is based upon the rapid inactivation
of this type of channel resulting in transient currents.
Furthermore the T-type channels are activated at much more
negative membrane potentials, have small single channel
conductances and are insensitive to block by
dihydropyridines.
A further type of Ca2+ current was revealed based upon
kinetic and pharmacological divergence from the existing L
and T-type currents. Nowycky et al. showed using whole
cell and single-channel recording from dorsal root ganglion
neurons a calcium current that activated at more negative
potentials than did the L-type but more positive than the T-
type [82]. The new current, which became known as the N-
type, also showed more rapid inactivation than L but slower
inactivation than the transient type. However it is the
sensitivity to block by the snail peptide ω-conotoxin GVIA
that is the crucial distinguishing factor for identification of
N-type channels [85]. The remaining P, Q and R-type
calcium channels were identified in a range of neuronal types
based upon sensitivity to other peptide toxins. P-type
currents are highly sensitive to block by the spider toxin ω-
agatoxin IVA and were first recorded in purkinje neurons
whilst the Q-type were first found in cerebellar granule
neurons [86] and display much lower affinity for the toxin.
Finally the R-type were found also in cerebellar granule
neurons and take their name from the fact they are resistant
to the subunit specific organic and peptide Ca2+ channel
blockers [86].
As described above, VGCCs are formed by a pore
forming α1 subunit linked with several accessory subunits.
As with the VGSC these accessory subunits are not required
for the formation of functional ion conducting pores but are
essential for conferring correct gating kinetics and cell
surface expression of the ion channel. The VGCC β subunits
(β1-4) are 53-65kDa proteins found within the intracellular
compartment which induce large changes to the voltage
dependence of activation and the rate of inactivation.
VGCCs also include an α2δ (α2δ 1-4) subunit which are 123-
129kDa dimers that are linked to the α1 subunit by a
dispulphide link. The α2 component is a membrane spanning
protein whilst the δ component is found extracellularly (Fig.
4B). The α2δ subunit is also thought to modulate channel
gating kinetics and regulate levels of expression but to a
lesser degree than the β subunit [80]. Furthermore several
25-36kDa γ (γ1-5) subunits have been identified but the
expression and functional status of this subunit is still
relatively unknown.
There is considerable evidence available (reviewed
recently by Jones [87]) concerning the role of VGCC in
epileptogenesis although there are few available AEDs that
target the VGCC to exert their anticonvulsant effects. In
particular genetic mutations in human calcium channel genes
(table 1) or those of several mouse species have been
Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1 23
associated with epileptic phenotypes. For example the
stargazer mouse has shown that mutations of γ subunits
result in loss of γ2 subunits without compensatory up
regulation of other γ subunit types. It has been suggested by
some authors that this may contribute to the pathogenesis in
this mutant mouse strain that results in spike wave seizures
characteristic of absence epilepsy [87]. This is still
somewhat controversial however as some groups implicate
the mutant γ subunit analogue (stargazin protein) an
important role in molecular trafficking and cell surface
expression of AMPA receptors [88].
The LVA T-type VGCC has been implicated in playing a
significant role in controlling rhythmic discharges in
thalamocortical neurons that display epileptiform spike wave
discharges associated with absence seizures. 2-Ethyl-2-
methylsuccinimide [Ethosuximide, ESM] (Fig. 4A) is an
AED that has an unusual clinical profile. ESM shows
efficacy against absence seizures but not against partial or
generalized tonic-clonic seizures. This has been shown to be
due to the ability of ESM to partially inhibit T-type calcium
channels at therapeutically relevant concentrations and hence
reduce the hyperexcitability of thalamocortical neurons that
is associated specifically with absence seizures [89].
Furthermore it has also been demonstrated that methyl-
phenyl suximide (the active metabolite of the related
compound methsuximide) but not its inactive analog also
produces inhibition of T-type currents in thalamic neurons
[90]. To complement these findings a recent study has shown
that both of these agents are able to inhibit cloned human T-
type VGCCs expressed in HEK293 cells whereas non-
anticonvulsant congeners were ineffective [91].
A recent and interesting development in the field of
VGCC anticonvulsant pharmacology are the novel drugs 1-
(aminomethyl) cyclohexanacetic acid [Gabapentin, GBP]
and S(+)-3-isobutyl-gamma-amniobutyric acid (Pregabalin,
PGL) (Fig. 4A). GBP has proven to be effective against a
wide range of seizure models in animals and clinically useful
for the treatment of partial seizures. It has been approved as
an add-on therapy in the U.K. since 1992 and the U.S. since
1993. GBP was designed to be a CNS penetrating GABA
analog but was later found to have no effect at the GABA
receptors, enzymes or transporters.
A putative binding site for GBP was first identified on
the α2δ subunit of HVA VGCCs using [3H]-gabapentin on
purified porcine brain [92]. The α2 and δ components were
found to both be required for binding of GBP although the
disulphide link between the two did not necessarily need to
be intact. Binding of GBP to these subunits also appears to
be subunit specific with higher affinity binding to α2δ-1 than
α2δ-2 and no binding at α2δ-3 and α2δ-4. This binding
evidence is supported by electrophysiological data in a
number of different preparations that show inhibition of
calcium currents. Several studies report inhibition of HVA
calcium channels in dissociated rat brain neurons and dorsal
root ganglion cells [93,94], reduction of Ca2+ influx into
synaptosomes through P/Q-type channels and decreased
glutamate release [95]. Recent experiments by Bayer and
colleagues have shown that GBT selectively blocks P/Q-type
Ca2+ channels in dorsal horn neurons (Fig. 5). The authors
suggest that this may result from interaction of the GBP
binding α2δ-1 or α2δ-2 subunits with the α1A pore forming
24 Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1 Lees et al.

Fig. (5). (A) Postsynaptic responses to exogenously applied glutamate (1mM) and glycine (1mM) to whole cell patch clamped dorsal root
neurons. Gabapentin (1µM) does not significantly effect the amplitude or duration of either the evoked excitatory or inhibitory currents. (B)
AMPA-EPSCs and GLY-IPSCs evoked by extracellular stimulation are significantly attenuated by gabapentin (1µM) when N-type channels
are blocked but not when P/Q-type channels are blocked. This demonstrates gabapentin is able to preferentially block P/Q-type calcium
channels in sensory neurons (from ref. [95]).

subunit of the P/Q-type channels [96]. Although other
potential mechanisms have been proposed for GBP including
enhancement of the GABA synthesizing enzyme, glutamic
acid decarboxylase (GAD), it is now generally accepted that
actions of GBP and that of its follow-up compound PGL are
at least in part mediated by interaction with α2δ−1/2 subunits
of VGCCs. The clinically important AEDs LTG and TPM
have also shown some activity against VGCCs although
modulation of VGSC is widely accepted as their primary
mode of action. Patch clamp experiments have shown that
TPM is able to produce inhibition of L-type VGCCs in
dentate granule cells [97] whilst in rat cortical neurons LTG
produces concentration dependent reduction in N and P but
not L-type currents [98].
STRUCTURE, FUNCTION AND ANTICONVULSANT
PHARMACOLOGY OF VOLTAGE GATED
POTASSIUM CHANNELS
Potassium channels that are gated by changes in
membrane potential represent a functionally and structurally
diverse family of ion channels that are crucial for many key
cellular processes including setting resting membrane
potential, determining firing threshold, action potential
repolarization [14] and determination of postsynaptic
excitability. They represent one of the greatest unexploited
targets for future AED development: only a few
experimental compounds evidently target these channels as
their primary mode of action. Several families of mammalian
K+ channels exist, including those gated by a change in
membrane potential (Kv); those which favor the movement
of K+ ions into the cell rather than outward (Kir); and the
recently discovered twin pore K channels (KCNK, TWIK,
TRAAK and related channels) that appear to generate
neuronal resting membrane potential. The largest of these
“superfamilies”, and also the first to be discovered, are the
channels that give rise to IKv (and IKA) currents. The Kv
channel sub-units are the best characterised of the K+
channels in terms of both structure and function. The first of
the K v channels to be cloned were those from the Drosophila
fruit fly voltage gated K+ channel genes Shaker, Shal, Shab
and Shaw but since, a further 29 related members of the Kv
family have been discovered. These have been divided into
eight gene families each having several members {KCNA
(Kv1.1-1.8); KCNB (Kv2.1-2.3); KCNC (Kv3.1-3.4); KCND
(Kv4.1-4.3); KCNF (Kv5.1); KCNG (Kv6.1-6.4); KCNS
(Kv9.1-9.3); KCNV (Kv8.1-8.3)} and can form homo or
heteromeric channels (Kv1, Kv2, Kv3 and Kv4) with other
subunits within their own family.
As previously discussed, Kv channels, are typically
formed by tetrameric assemblies of Kv α−subunits [99] each
of which having six transmembrane spanning domains (S1-
S6) much like a single sodium channel repeated motif (Fig.
6C). The S4 transmembrane spanning segment provides the
voltage sensing capacity [100] whilst the pore of the
complete channel is formed by the re-entrant loop between
S5-S6 (P-loop). The P-loop contains the structural motif
GYG which forms the basis of the selectivity filter [101] and
makes the channel selective for K+ ion conductance.
MacKinnon’s group has made a huge contribution to our
knowledge of the fine molecular structure of voltage gated
Voltage Gated Ion Channels: Targets for Anticonvulsant Drugs
Fig. (6). (A) Structures of anticonvulsant drugs acting at VGKCs. (B) Diagram showing the structure of potassium channel α subunits
including the voltage gated (Kv), KCNQ and the twin pore KCNK type channels. (C) Differences in subunit composition of complete ion
channels. Na+ and Ca2+ channels are homomers comprising a single α subunit whilst potassium channels are formed by heteromeric
assembly of four smaller α subunits.
K+ channels through X-ray crystallographic studies of
related bacterial channels [102]
Due to the diversity of K+ channels brought about by
many gene families, splice variation and heteromeric
assembly (see subunit structures in Fig 6C) the range of
physiological properties and roles of these channels are
extremely diverse. The Kv channels can however be
classified into three broad groups based upon their voltage
dependence of activation and the extent to which they
inactivate. At resting membrane potential Kv channels reside
in a closed state and are activated in response to
depolarization whereupon they allow K+ ions to flow from
the cellular interior resulting in hyperpolarization of
membrane potential. The first functional class of K v channels
are the so-called low voltage activated channels (LVA, e.g.
Kv1.1, KCNA). These are non-inactivating channels that
produce sustained outward potassium currents in response to
relatively low changes in membrane potential away from rest
(~ -60mV). LVA channels activate and deactivate slowly and
as such are often known as delayed rectifier channels. The
slow kinetic features associated with LVA channels means
they are important in determining spike frequency adaptation
but have little effect upon single spikes or the duration or
rise time of the first spike within a burst. The second major
subtype of the delayed rectifier family, are the high voltage
activated channels (HVA, e.g. Kv3, KCNC). HVA activate
and deactivate more rapidly than LVA channels but
activation only occurs in response to large depolarization,
such as that brought about by an action potential. HVA
channels are therefore mostly responsible for the repolariz-
ation phase of the action potential and allowing high
frequency neuronal firing. This is typified by expression of
Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1 25
Kv3 channels in cortical interneurons where it has been
correlated to the short duration action potentials that allow
very high firing frequencies. The significant role the delayed
rectifier channels play in controlling cellular excitability and
burst firing frequency makes them an ideal therapeutic target
that may be exploited for anticonvulsant benefit.
Levetiracetam [(S)-α-ethyl-2-oxo-pyrrolidine acetamide,
LEV] (Fig 6A) is a pyrrolidine derivative that is a well-
tolerated adjunctive therapy for refractory partial seizures in
adults. Despite its clinical success the mechanism of action
of LEV has remained elusive with GABAergic facilitation,
inhibition of Na+ /low voltage activated Ca2+ current and
inhibition of excitatory ionotropic conductances being ruled
out as potential candidates [103,104]. Complementing this
lack of activity at established molecular target sites for
anticonvulsants in vitro, is the notable observation that LEV
differs from existing anticonvulsants in showing lack of
activity against MES and PTZ induced seizures in vivo.
Recently however Madeja and colleagues [105] have
postulated that the anticonvulsant properties of LEV may be
substantially contributed to by a reduction in the
conductance of delayed rectifier K+ channels. The authors of
this study demonstrated using acutely isolated rat and guinea
pig CA1 neurones, and the whole cell patch clamp
technique, that a 26% reduction in the delayed rectifier K+
current is produced by 100µM LEV. This inhibition was
demonstrated to be sufficient to produce marked impairment
of repetitive action potential firing. This study implicates
LEV as the first compound to have its primary mode of
action as a blocker of delayed rectifier channels although no
distinction is made between the LVA and HVA types.
26 Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1
The final major functional type of Kv channels are those
channels which are capable of inactivation and give rise to
the so called transient outward K+ current or A currents.
These channels activate more rapidly than the delayed
rectifier type but also rapidly and spontaneously deactivate
owing to a ‘ball and chain’ or slower C-type inactivation
mechanism depending upon the subunit composition of the
channel. There is no evidence to date that any existing
clinical or experimental drugs exploit modulation of A-
currents as their primary mode of action. However it has
been reported that LTG and CBZ are able to potentiate
outward KA conductance at concentrations similar to those at
which they exert sodium channel inhibition [106,107]. 4-AP
models owing to the occlusion of a primary target by the
convulsant toxin.
The M-current (IM) is a slowly activating, non-
inactivating and slowly deactivating current that was first
described in frog sympathetic ganglia by Brown and Adams
in 1980 [108]. The name is derived from the ability of
muscarinic receptor stimulation to inhibit the K+ current and
this has been shown to be present in a number of CNS
neurons [109]. IM has an important role in determining
neuronal excitability as it represents the only sustained
inhibitory current in the range of potentials around spike
threshold. The slow activation and deactivation kinetics of I M
mean that it plays a significant part in controlling resting
membrane potential, responsiveness to synaptic input and as
a braking mechanism on repetitive firing. The molecular
identity of the M-current was unknown until it was
discovered that currents induced by heteromers of
KCNQ2/KCNQ3 potassium channel subunits shared similar
pharmacological and biophysical properties to native IM
[110]. The KCNQ genes encode a growing family of K+
channel products that are structurally related to the Kv
channel α subunits and like the classical Kv channels are
gated by changes in membrane potential. They have six
transmembrane spanning domains, a single pore forming
loop that forms the selectivity filter and sequences of
charged amino acids in the S4 domain that probably
represent the voltage sensor (Fig. 6BC). Although it has not
been shown directly, KCNQ subunits, of which there are five
known types (KCNQ1, KCNQ2, KCNQ3, KCNQ4 and
KCNQ5), likely form homo or heterotetramers to produce
functional channels [111]. The significance of these channels
in the control of cellular excitability is demonstrated by the
observation that mutations of the KCNQ 2/3 channels result
in the rare condition of benign familial neonatal convulsions
(BFNC, table 1) [112,113]. BNFC mutations in either
KNCQ2 or KNCQ3 may only produce a reduction of
between 20-30% [114] in the current amplitude of the
expressed heteromeric channels but this is often sufficient to
induce tonic-clonic seizures in the neonate.
Owing to the clear link between, an albeit rare idiopathic
form of epilepsy, KCNQ2/3 mutations and the correlation
between these channel subunits and the M-current (that is so
crucial for determining cellular excitability), it would appear
that K+ channel opening represents an ideal therapeutic
target for novel anticonvulsant ligands. Indeed an
experimental anticonvulsant ligand already undergoing
Phase II clinical trials has been shown to act, at least in part,
through opening KCNQ2/KCNQ3 potassium channels.
Lees et al.
Retigabine [(D-23129, N-(2-amino-4-(4-fluorobenzylamino)-
phenyl) carbamic acid ethyl ester, RTG] (Fig 6A) is
structurally unrelated to any currently marketed anticon-
vulsant drug and has been shown to exert anticonvulsant
effects in broad range of seizure models. It was first found
that RTG activated a K+ conductance in marginally
depolarized NG108-15 neuronal cells at concentrations
approximating to therapeutic plasma levels [115]. Although
it was unknown at the time of this study it was later observed
that the pharmacological profile of the RTG induced current
was consistent with that of the M-current. It has since been
confirmed in Chinese hamster ovary (CHO) cells transfected
with either KCNQ2 (homomer) or KCNQ2/KCNQ3
(heteromer) that RTG is indeed that first molecule in a new
class of KCNQ K+ channel opening drugs [116] (Fig. 7). The
direct interaction of Retigabine with the KCNQ channels that
are responsible for BFNC may be able to overcome the small
loss of function underlying this rare epileptic condition and
illustrates the importance of molecular biology and genetics
to the development of therapeutic agents for the treatment of
idiopathic syndromes.
A further important family of potassium channels has
been recognized only within the last eight to ten years. Leak
currents (resting or background conductances) have been
described in physiology for many years but only in 1996 was
it discovered that these currents were carried by discrete
molecular entities within the cell membrane. The first of
these ion channel structures was cloned from the
neuromuscular junction of Drosophila and became known as
KCNK0. Later in the same year the first mammalian gene for
this type of channel subunit, KCNK1, was also discovered
although this was found to be physiologically non-
functional. Although the differences in structural and
functional properties associated with KCNK channels are too
extensive to cover fully in this review (reviewed in detail by
Goldstein et al. [117]) there is some interesting pharma-
cology related to these channels that deserves attention.
Briefly the KCNK family of K+ channels subunits differ in
that they have two pore forming or P domains and four
(rather than six) transmembrane spanning segments (Fig.
6B). They are open at rest and are subject to modulation by a
number of physiologically pertinent agents including cyclic
nucleotides, noradrenaline, serotonin and fatty acids. This
type of ligand modulation at KCNK channels is capable of
influencing channel gating and conductance properties and
as such is predicted to contribute to control of membrane
potential and cellular excitability.
The neuroprotective agent sipatrigine (STG), which is
chemically related to LTG, has been shown to potently
inhibit at least two of the two pore domain channels (that are
known to be sensitive to arachidonic acid), namely KCNK2
(also known as TREK-1) and KCNK4 (also known as
TRAAK) when they are expressed in HEK293 cells [118]. If
one assumes that the KCNK channels are the main targets
for the neuroprotective actions of STG it may be expected
that inhibition of the leak conductance would result in
depolarization and increased excitability/excitotoxicity.
However, if KCNK2/4 were preferentially expressed on
GABAergic interneurones, reduced net excitability may be
produced by an increase in polysynaptic inhibitory tone.
Alternatively chronic depolarization resulting from the
Voltage Gated Ion Channels: Targets for Anticonvulsant Drugs
Fig. (7). The anticonvulsant agent retigabine is mechanistically novel. It produces (A) a concentration dependent increase in potassium
conductance through KNCQ2/KNCQ3 heteromers expressed in CHO cells by directly opening the channel. (B) Retigabine also shifts the
voltage dependence of activation to more negative potentials in a reversible fashion. This results in the cellular potential being taken further
from the threshold potential for spike firing (from ref. [115]).
decreased leak conductance may cause reduction in excitatory
neurotransmitter release by depolarization mediated block of
glutamatergic neurons. Although the anticonvulsant
compound LTG produces only <10% block of this current at
10µM [118] it is important to consider the other mechanisms
by which the anticonvulsant is known to act. A small
reduction in the KCNK2/4 current produced by LTG
application may, on its own, produce only insignificant
changes in excitability but the resulting depolarization would
undoubtedly recruit more inactivation of both sodium
channel and calcium channels. It is known that LTG binds
potently to the inactivated conformation of the sodium
channel (as above) to reduce electrogenesis and also reduces
excitatory neurotransmitter release that likely results from
reduction in availability of calcium channels in the
presynaptic terminal. In these terms, what looks like a
paradoxical excitant or pro-convulsant surrogate effect of the
drug may represent a synergistic pathway to promote its
potency in blocking epileptiform activity. This type of
structure-dependent, non-selective cross talk with other
channels may explain the diverse pharmacological profiles of
DPH, CBZ and LTG in the living brain (despite their
commonality at the VGSCs).
Clearly the functional diversity amongst the K+ channels
make them prospective targets for future AED development.
The production of selective modulators of the various K+
channel types may serve to further relieve the symptoms of
those patients with refractory or idiopathic epilepsies whose
conditions are poorly controlled by the existing compounds.
CONCLUSION
To sum up we wish to stress that voltage-gated ion
channels have been crucial in the treatment of epilepsy (often
their mode of action emerged after serendipitous discovery
of their clinical promise). One by one the large multi-
national pharmaceutical companies are withdrawing from
biorational drug discovery in the epilepsy field (because of
the prohibitive development costs and the long lead times).
This is clearly detrimental to the significant number of
Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1 27
refractory patients: there is a surprisingly large incidence of
sudden death in epilepsy patients (SUDEP) and a clear
clinical need for new drugs. The market analysts appear to be
ignoring completely the obvious promise that such drugs
hold in the massive pharmaceutical market of neuropathic
pain. These chronic pain syndromes are notoriously difficult
to treat but licensed anticonvulsants (even those with overt
side effects in the form of sedation, impairment of hepatic
function and serious blood dyscrasias are often useful in the
treatment of chronic pain). Gabapentin has become the gold-
standard in the treatment of conditions like post-herpetic and
trigeminal neuralgia. The NIH in the USA continue to fund
drug discovery research in epilepsy and it seems likely that
novel drugs in the future will exploit the superfamilies of
ion-selective pores reviewed here.
REFERENCES
[1] Traub, R. D.; Borck, C.; Colling, S. B.; Jefferys, J. G. On the
structure of ictal events in vitro. Epilepsia 1996, 37, 879-891.
[2] Jefferys, J. G. R. Models and Mechanisms of Experimental
Epilepsies. Epilepsia 2003, 44, 44-50.
[3] Hodgkin, A. L.; Huxley, A. F. The dual effect of membrane
potential on sodium conductance in the giant axon of Loligo. J.
Physiol. (London) 1952, 116, 497-506.
[4] Hodgkin, A. L.; Huxley, A. F. The components of membrane
conductance in the giant axon of Loligo. J. Physiol (London) 1952,
116, 473-496.
[5] Hodgkin, A. L.; Huxley, A. F. Currents carried by sodium and
potassium ions through the membrane of the giant axon of Loligo.
J. Phsiol. (London) 1952, 116, 473-506.
[6] Catterall, W. A. Cellular and molecular biology of voltage-gated
Na+ channels. Physiol. Rev. 1992, 72, S15-S48.
[7] Catterall, W. A. Molecular properties of brain sodium channels: an
important target for anticonvulsant drugs. Adv. Neurol. 1999, 79,
441-456.
[8] Qu, Y.; Curtis, R.; Lawson, D.; Gilbride, K.; Ge, P.; DiStefano, P.
S.; Silos-Santiago, I.; Catterall, W. A.; Scheuer, T. Differential
modulation of sodium channel gating and persistant sodium
currents by the β1, β2 and β3 subunits. Mol. Cell. Neurosci. 2001,
18, 570-580.
[9] Goldin, A. L.; Barchi, R. L.; Caldwell, J. H.; Hofmann, F.; Howe,
J. R.; Hunter, J. C.; Kallen, R. G.; Mandel, G.; Meisler, M. H.;
Netter, Y. B.; Noda, M.; Tamkun, M. M.; Waxman, S. G.; Wood, J.
N.; Catterall, W. A. Nomenclature of voltage-gated sodium
channels. Neuron 2000, 28, 365-368.
28 Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1
[10] Noda, M.; Ikeda, T.; Kayano, T.; Suzuki, H.; Takeshima, M.;
Kurasaki, H.; Takahashi, S.; Numa, S. Existence of distinct sodium
channel messenger RNAs in rat brain. Nature 1986, 320, 121-127.
[11] Ragsdale, D. S.; Avoli, M. Sodium channels as molecular targets
for antiepileptic drugs. Brain Res. Rev. 1998, 26, 16-28.
[12] Catterall, W. A. A 3D view of sodium channels. Nature 2001, 409,
988-991.
[13] Hille, B. Ion channels of excitable membranes; Sinauer Assoc.:
Sunderland, MA., 1992.
[14] Hodgkin, A. L.; Huxley, A. F. A quantative description of
membrane current and its application to conduction and excitation
in nerve. J Physiol (Lond) 1952, 117, 500-544.
[15] Armstrong, C. M.; Bezanilla, F. Charge movement associated with
the opening and closing of the activation gates of Na channels. J.
Gen. Physiol. 1974, 63, 533-552.
[16] Armstrong, C. M.; Bezanilla, F. Currents related to the movement
of the gating particles of the sodium channels. Nature 1973, 242,
459-461.
[17] Stümer, W.; Conti, F.; Suzuki, X.; Wang, M.; Noda, M.; Yahadi,
N.; Kubo, H.; Numa, S. Structural parts involved in activation and
inactivation of the voltage gated sodium channel. Nature 1989,
339, 597-603.
[18] Patton, D. E.; West, J. W.; Catterall, W. A.; Goldin, A. L. Amino
acid residues required for fast Na+ channel inactivation: charge
neutralizations and deletions in the III-IV linker. Proc. Nat. Acad.
Sci. U. S. A. 1992, 89, 10905-10909.
[19] Vassilev, P. M.; Scheuer, T.; Catterall, W. A. Identification of an
intracellular peptide segment involved in sodium channel
inactivation. Science 1988, 241, 1658-1661.
[20] West, J. W.; Patton, D. E.; Scheuer, T.; Wang, Y.; Goldin, A. L.;
Catterall, W. A. A cluster of hydrophobic amino acid residues
required for fast Na+ channel inactivation. Proc. Nat. Acad. Sci.
U. S. A. 1992, 89, 10910-10914.
[21] McPhee, J. C.; Ragsdale, D. S.; Scheuer, T.; Catterall, W. A. A
critical role for transmembrane segment IVS6 of the sodium
channel α subunit in fast inactivation. J. Biol. Chem. 1995, 270,
12025-12034.
[22] Goldin, A. L. Mechanisms of sodium channel inactivation. Curr.
Opin. Neurobiol. 2003, 13, 284-290.
[23] Traub, R. D.; Jefferys, G. R.; Whittington, M. A. Fast Oscillations
in Cortical Circuits.; The MIT Press: Cambridge, MA, 1999.
[24] Featherstone, I. A.; Richmond, J. E.; Ruben, P. C. Interaction
between fast and slow inactivation in Skm1 sodium channels.
Biophys. J. 1996, 71, 3098-3109.
[25] Richmond, J. E.; Featherstone, I. A.; Hartmann, H. A.; Ruben, P.
C. Slow inactivation in human cardiac sodium channels. Biophys.
J. 1998, 74, 2945-2952.
[26] Fleidervish, I. A.; Friedman, A.; Gutnick, M. J. Slow inactivation
of Na+ current and slow cumulative spike adaptation in mouse and
guinea pig neocortical neurones in slices. J. Physiol. 1996, 493, 83-
97.
[27] Toib, A.; Lyakhov, V.; Marom, S. Interaction between duration of
activity and time course of recovery from slow inactivation in
mammalian brain Na+ channels. J. Neurosci. 1998, 18, 1893-1903.
[28] Avoli, M.; Barbarosie, M.; Lucke, A.; Nagao, T.; Lopantsev, V.;
Kohling, R. Synchronous GABA-mediated potentials and
epileptiform discharges in the rat limbic system in vitro. J.
Neurosci. 1996, 16, 3912-3924.
[29] Lopantsev, V.; Avoli, M. Participation of GABAA-mediated
inhibition in ictal-like discharges in the rat entorhinal cortex. J.
Neurophysiol. 1998, 79, 352-360.
[30] Avoli, M.; D'Antuono, M.; Louvel, J.; Kohling, R.; Biagini, G.;
Pumain, R.; D'Arcangelo, G.; Tancredi, V. Network and
pharmacological mechanisms leading to epileptiform
synchronization in the limbic system in vitro. Prog. Neurobiol.
2002, 68, 167-207.
[31] Dougalis, A.; Lees, G.; Ganellin, C. R. The sleep lipid oleamide
may represent and endogenous anticonvulsant: an in vitro
comparative study in the 4-aminopyridine rat brain slice model.
Neuropharmacology 2003.
[32] Taylor, C. P. Na+ currents that fail to inactivate. Trends Neurosci.
1993, 16, 455-460.
[33] Alzheimer, C.; Schwindt, P. C.; Crill, W. E. Modal gating of Na+
channels as a mechanism of peristant Na+ current in pyramidal
neurons from rat and cat sensorimotor coretx. J. Neurosci. 1993,
13, 660-673.
Lees et al.
[34] Brown, A. M.; Schwindt, P. C.; Crill, W. E. Different voltage
dependence of transient and persistant Na+ currents is compatible
with modal gating hypothesis for sodium channels. J.
Neurophysiol. 1994, 71, 2562-2565.
[35] Huguenard, J. R.; Hamill, O. P.; Prince, A. Developmental changes
in Na + conductance in rat cortical neurons: appearance of a slowly
inactivating component. J. Neurophysiol. 1988, 59, 778-795.
[36] Crill, W. E. Persistant sodium current in mammalian central
nervous system. Annu. Rev. Physiol. 1996, 58, 349-362.
[37] Stafstrom, C. E.; Schwindt, P. C.; Chubb, M. C.; Crill, W. E.
Properties of persistant sodium conductance and calcium
conductance of layer V neurons from cat sensorimotor coretx in
vitro. J. Neurophysiol. 1985, 53, 152-170.
[38] Stafstrom, C. E.; Schwindt, P. C.; Crill, W. E. Negative slope
conductance due to a persistant subthreshold sodium current in cat
neocortical neurons in vitro. Brain Res. 1982, 236, 221-226.
[39] Cantrell, A. R.; Catterall, W. A. Neuromodulation of Na+ channels:
an unexpected form of cellular plasticity. Nat. Rev. Neurosci. 2001,
2, 397-407.
[40] Murphy, B. J.; Rossie, S.; DeJongh, K. S.; Catterall, W. A.
Identification of sites of selective phosphorylation and
dephosphorylation of the rat brain sodium channel α subunit by
cAMP dependent protein kinase and phosphoprotein phosphatases.
J. Biol. Chem. 1993, 268, 27355-27362.
[41] Smith, R. D. Goldin, A. L. Phosphorylation of brain sodium
channels in the I-II linker modulates function in Xenopus oocytes.
J. Neurosci. 1996, 16, 1965-1974.
[42] Li, M.; West, J. W.; Lai, Y.; Scheuer, T.; Catterall, W. A.
Functional modulation of brain sodium channels by cAMP-
dependent phosphorylation. Neuron 1992, 8, 1151-1159.
[43] Alroy, G.; Su, H.; Yaari, Y. Protein kinase C mediates muscarinic
block of intrinsic bursting in rat hippocampal neurons. J. Physiol.
1999, 518, 71-79.
[44] Azouz, R.; Jensen, M. S.; Yaari, Y. Muscarinic modulation of of
intrinsic burst firingi in rat hippocampal neurons. Eur. J. Neurosci.
1994, 6, 961-966.
[45] Merritt, H. H.; Putnam, T. J. Sodium diphenylhydantoinate in the
treatment of convulsive disorders. J. Am. Med. Assoc. 1938, 111,
1068-1073.
[46] Brodie, M. J.; Dichter, M. A. Antiepileptic drugs. N. Engl. J. Med.
1996, 334, 168-175.
[47] Toman, J. E. P. The neuropharmacology of antiepileptics.
Electroencephalogr. Clin. Neurophysiol. 1949, 1, 33-44.
[48] McLean, M. J.; McDonald, R. L. Multiple actions of phenytoin on
mouse spinal cord neurons in cell culture. J. Pharmacol. Exp. Ther.
1986, 237, 779-789.
[49] Willow, M.; Gonoi, T.; Catterall, W. A. Voltage clamp analysis of
the inhibitory actions of diphenylhydantoin and carbamazepine on
voltage-sensitive sodium channels in neuroblastoma cells. Mol.
Pharmacol. 1985, 27, 549-558.
[50] Matsuki, N.; Quandt, F. N.; Ten Eick, R. E.; Yeh, J. Z.
Characterization of the block of sodium channels by phenytoin in
mouse neuroblastoma cells. J. Pharmacol. Exp. Ther. 1984, 228,
523-530.
[51] Taylor, C. P.; Meldrum, B. S. Na+ channels as targets for
neuroprotective drugs. Trends Pharmacol. Sci. 1995, 16, 309-315.
[52] Lang, D. G.; Wang, C. M.; Cooper, B. R. Lamotrigine, Phenytoin
and Carbamazepine interactions on the sodium current present in
N4TG1 mouse neuroblastoma cells. J. Pharmacol. Exp. Ther.
1993, 266, 829-834.
[53] Chao, T. I.; Alzheimer, C. Effects of phenytoin on the persistent
Na+ current of mammalian CNS neurones. Neuroreport 1995, 6,
1778-1780.
[54] Lampl, I.; Schwindt, P.; Crill, W. Reduction of Cortical Pyramidal
Neuron Excitability by the Action of Phenytoin on Persistent Na+
Current. J. Pharmacol. Exp. Ther. 1998, 284, 228-237.
[55] Spadoni, F.; Hainsworth, A. H.; Mercuri, N. B.; Caputi, L.;
Martella, G.; Lavaroni, F.; Bernardi, G.; Stefani, A. Lamotrigine
derivatives and riluzole inhibit I NaP in cortical neurons. Neuroreport
2002, 13, 1167-1170.
[56] Kuo, C. C. A Common Anticonvulsant Binding Site for Phenytoin,
Carbamazepine, and Lamotrigine in Neuronal Na+ Channels. Mol.
Pharmacol. 1998, 54, 712-721.
[57] Krall, R. L.; Penry, J. K.; White, B. G.; Kupferberg, H. J.;
Swinyard, E. A. Anti-epileptic drug development. II.
Anticonvulsant drug screening. Epilepsia 1978, 19, 404-428.
Voltage Gated Ion Channels: Targets for Anticonvulsant Drugs
[58] Meldrum, B. Do preclinical seizure models preselect certain
adverse effects of antiepileptic drugs. Epilepsy Res. 2002, 50, 33-
40.
[59] McLean, M. J.; McDonald, R. L. Carbamazepine and 10,11-
epoxycarbamazepine produce use and voltage dependent limitation
of rapidly firing action potentials in mouse central neurons in cell
culture. J. Pharmacol. Exp. Ther. 1986, 238, 727-738.
[60] Kuo, C. C.; Bean, B. P.; Chen, L. L.; Chen, R. C. Carbamzepine
inhibition of neuronal Na+ currents: quantitative distinction from
phenytoin and possible clinical implications. Mol. Pharmacol.
1977, 51, 1077-1083.
[61] Tecoma, E. S. Oxcarbazepine. Epilepsia 1999, 40, S37-S46.
[62] Bonifácio, M. J.; Sheridan, R. D.; Parada, A.; Cunha, R. A.;
Patmore, L.; Soares-da-Silva, P. Interaction of the novel
anticonvulsant, BIA 2-093, with voltage gated sodium channels:
comparison with carbamazepine. Epilepsia 2001, 42, 600-608.
[63] Leach, M. J.; Baxter, M. G.; Critchley, M. A. E. Neurochemical
and behavioural aspects of Lamotrigine. Epilepsia 1991, 32, S4-S8.
[64] Leach, M. J.; Marden, C. M.; Miller, A. A. Pharmacological studies
on Lamotrigine, a novel potential antiepileptic drug. II.
Neurochemical studies on the mechanism of action. Epilepsia
1986, 27, 490-497.
[65] Miller, A. A.; Wheatly, P.; Sawyer, D. A.; Baxter, M. G.; Roth, B.
Pharmacological studies on lamotrigine, a novel potential
antiepiletic drug: I. Anticonvulsant profile in mice and rats.
Epilepsia 1986, 27, 483-489.
[66] Hamill, O. P.; Marty, A.; Neher, E.; Sakmann, B.; Sigworth, F. J.
Improved patch clamp techniques for high-resolution current
recordings from cells and cell free memebrane patches. Pflug.
Arch. 1981, 391, 85-100.
[67] Lees, G.; Leach, M. J. Studies on the mechanism of action of the
novel anticonvulsant lamotrigine (Lamictal) using primary
neurological cultures from rat cortex. Brain Research 1993, 612,
190-199.
[68] Wang, C. M.; Lang, D. G.; Cooper, B. R. Lamotrigine effects on
ion channels in neuronal cells. Epilepsia 1993, 34, 117-118.
[69] Ragsdale, D. S.; McPhee, J. C.; Scheuer, T.; Catterall, W. A.
Molecular determinants of state-dependent block of Na channels by
local anesthetics. Science 1994, 265, 1724-1728.
[70] Ragsdale, D. S.; McPhee, J. C.; Scheuer, T.; Catterall, W. A.
Common molecular determinants of local anesthetic,
antiarrhythmic and anticonvulsant block of voltage gated Na
channels. Proc. Nat. Acad. Sci. U. S. A. 1996, 93, 9270-9275.
[71] Shank, R. P.; Gardocki, J. F.; Vaught, J. L.; Davis, C. B.;
Schupsky, J. J.; Raffa, R. B.; Dodgson, S. J.; Nortey, S. O.;
Maryanoff, B. E. Topiramate: preclinical evaluation of a
structurally novel anticonvulsant. Epilepsia 1994, 35, 450-460.
[72] Gibbs, J. W. III; Sombati, S.; DeLorenzo, R. J.; Coulter, D. A.
Cellular actions of topiramate: blockade of kainate-evoked inward
currents in cultured hippocampal neurons. Epilepsia 2000, 41, S10-
S16.
[73] Brown, S. D.; Wolf, H. H.; Swinyard, E. A.; Twyman, R. E.;
White, H. S. The novel anticonvulsant topiramate enhances
GABA-mediated chloride flux. Epilepsia 1993, 34, 123-124.
[74] DeLorenzo, R. J.; Sombati, S.; Coulter, D. A. Effects of topiramate
on sustained repetitive firing and spontaneous recurrent seizure
discharges in cultured hippocampal neurons. Epilepsia 2000, 41,
S40-S44.
[75] Zona, C.; Ciotti, M. T.; Avoli, M. Topiramate attenuates voltage-
gated sodium currents in rat cerebellar granule cells. Neurosci. Lett.
1997, 231, 123-126.
[76] Taverna, S.; Sancini, G.; Mantegazza, M.; Franceschetti, S.;
Avanzini, G. Inhibition of transient and persistent Na+ current
fractions by the new anticonvulsant topiramate. J. Pharmacol. Exp.
Ther. 1999, 288, 960-968.
[77] Draguhn, A.; Jungclaus, M.; Sokolowa, S.; Heinemann, U.
Losigamone decreases spontaneous synaptic activity in cultured
hippocampal neurons. Eur. J. Pharmacol. 1997, 325, 245-251.
[78] Gebhardt, C.; Breustedt, J. M.; Noldner, M.; Chatterjee, S. S.;
Heinemann, U. The antiepileptic drug losigamone decreases the
persistent Na+ current in rat hippocampal neurons. Brain Res.
2001, 920, 27-31.
[79] Salvati, P.; Maj, R.; Caccia, C.; Cervini, M. A.; Fornaretto, M. G.;
Lamberti, E.; Pevarello, P.; Skeen, G. A.; White, H. S.; Wolf, H.H.;
Faravelli, L. Biochemical and electrophysiological studies on the
Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1 29
mechanism of action of PNU-151774E, a novel antiepileptic
compound. J. Pharmacol. Exp. Ther. 1999, 288, 1151-1159.
[80] Catterall, W. A. Structure and regulation of voltage-gated Ca2+
channels. Annu. Rev. Cell. Dev. Biol. 2000, 16, 521-555.
[81] Ertel, E.; Campbell, K. P.; Harpold, M. M.; Hofmann, F.; Mori, Y.
Nomenclature of voltage-gated calcium channels. Neuron 2000, 25,
533-535.
[82] Nowycky, M. C.; Fox, A. P.; Tsien, R. W. Three types of neuronal
calcium channel with different calcium agonist sensitivity. Nature
1985, 316, 440-443.
[83] Reuter, H. Calcium channel modulation by neurotransmitters,
enzymes and drugs. Nature 1983, 301, 569-574.
[84] Llinás, R. Yarom, Y. Electrophysiology of mammalian inferior
olive neurones in vitro. Different types of voltage dependent ionic
conductances. J. Physiol. 1981, 315, 569-584.
[85] McCleskey, E. W.; Fox, A. P.; Feldman, D. H.; Cruz, L. J.;
Olivera, B. M.; Tsien, R. W.; Yoshikami, D. ω-Conotoxin: direct
and persistant blockade of specific calcium channels in nuerons but
not muscle. Proc. Nat. Acad. Sci. U. S. A. 1987, 84, 4327-4331.
[86] Randall, A.; Tsien, R. W. Pharmacological dissection of multiple
types of Ca2+ channel currents in rat cerebellar granule neurons. J.
Neurosci. 1995, 15, 2995-3012.
[87] Jones, O. T. Ca2+ channels and epilepsy. Eur. J.Pharmacol. 2002,
447, 211-225.
[88] Chen, L.; El Husseini, A.; Tomita, S.; Bredt, D. S.; Nicoll, R. A.
Stargazin Differentially Controls the Trafficking of {alpha}-
Amino-3-hydroxyl-5-methyl-4-isoxazolepropionate and Kainate
Receptors. Mol. Pharmacol. 2003, 64, 703-706.
[89] Coulter, D. A.; Huguenard, J. R.; Prince, D. A. Specific petit mal
anticonvulsants reduce calcium currents in thalamic neurons.
Neurosci. Lett. 1989, 98, 74-78.
[90] Coulter, D. A.; Huguenard, J. R.; Prince, D. A. Differential effects
of petit mal anticonvulsants and convulsants on thalamic neurons:
calcium current reduction. Br. J. Pharmacol. 1990, 100, 800-806.
[91] Gomora, J. C.; Daud, A. N.; Weiergraber, M.; Perez-Reyes, E.
Block of cloned human T-type calcium channels by succinimide
antiepileptic drugs. Mol. Pharmacol. 2001, 60, 1121-1132.
[92] Gee, N. S.; Brown, J. P.; Dissanayake, V. U.; Offord, J.; Thurlow,
R.; Woodruff, G. N. The novel anticonvulsant drug, gabapentin
(neurontin), binds to the alpha2delta subunit of a calcium channel.
J. Biol. Chem. 1996, 271, 768-776.
[93] Sutton, K. G.; Martin, D. J.; Pinnock, R. D.; Lee, K.; Scott, R. H.
Gabapentin inhibits high-threshold calcium channel currents in
cultured rat dorsal root ganglion neurones. Br. J. Pharmacol. 2002,
135, 257-265.
[94] Stefani, A.; Spadoni, F.; Bernardi, G. Gabapentin inhibits calcium
currents in isolated rat brain neurons. Neuropharmacology 1998,
37, 83-91.
[95] Fink, K.; Dooley, D. J.; Meder, W. P.; Suman-Chauhan, N.; Duffy,
S.; Clussman, H.; Göthert, M. Inhibition of neuronal Ca2+ influx by
gabapentin and pregabalin in the human neocortex.
Neuropharmacology 2001, 42, 229-236.
[96] Bayer, K.; Ahmadi, S.; Zeilhofer, H. U. Gabapentin may inhibit
synaptic transmission in the mouse spinal cord dorsal horn through
a preferential block of P/Q-type Ca2+ channels.
Neuropharmacology 2004, 46, 743-749.
[97] Zhang, X.; Velumian, A. A.; Jones, O. T.; Carlen, P. L. Modulation
of high-voltage activated calcium channels in dentate granule cells
by topiramate. Epilepsia 2000, 41, S52-S60.
[98] Stefani, A.; Spadoni, F.; Siniscalchi, A.; Benardi, G. Lamotrigine
inhibits Ca2+ currents in cortical neurons: functional implications.
Eur. J. Pharmacol. 1996, 307, 113-116.
[99] MacKinnon, R. Determination of the subunit stoichiometry of a
voltage-activated potassium channel. Nature 1991, 350, 232-235.
[100] Bezanilla, F. The voltage sensor in voltage-dependent ion channels.
Physiol. Rev. 2000, 80, 555-592.
[101] Yellen, G. The voltage gated potassium channels and their
relatives. Nature 2002, 419, 35-42.
[102] MacKinnon, R. Potassium channels. FEBS Lett. 2003, 555, 62-65.
[103] Zona, C.; Marchetti, C.; Margineau, D. G. The novel antiepileptic
drug candidate levetiracetam does not modify the biophysical
porperties of the Na+ channel in rat cortical neurons in culture.
Society for neuroscience abstracts 1999, 25, 1868.
[104] Zona, C.; Niespodziany, I.; Marchetti, C.; Klitgaard, H.; Bernardi,
G.; Margineanu, D. G. Levetiracetam does not modulate neuronal
30 Current Topics in Medicinal Chemistry, 2005, Vol. 5, No. 1
voltage-gated Na+and T-type Ca2+currents. Seizure 2001, 10, 279-
286.
[105] Madeja, M.; Georg Margineanu, D.; Gorji, A.; Siep, E.; Boerrigter,
P.; Klitgaard, H.; Speckmann, E. J. Reduction of voltage-operated
potassium currents by levetiracetam: a novel antiepileptic
mechanism of action? Neuropharmacology 2003, 45, 661-671.
[106] Zona, C.; Tancredi, V.; Longone, P.; D'Arcangelo, G.; D'Antuono,
M.; Manfredi, M.; Avoli, M. Neocortical potassium currents are
enhanced by the antiepileptic drug lamotrigine. Epilepsia 2002, 43,
685-690.
[107] Zona, C.; Tancredi, V.; Palma, E.; Pirrone, G. C.; Avoli, M.
Potassium currents in rat cortical neurons in culture are enhanced
by the antiepileptic drug carbamazepine. Can. J. Physiol.
Pharmacol. 1990, 68, 545-547.
[108] Brown, D. A.; Adams, P. R. Muscarinic supression of a novel
voltage-sensitive K+ current in a vertebrate neuron. Nature 1980,
283, 673-676.
[109] Halliwell, J. V.; Adams, P. R. Voltage clamp analysis of
muscarinic excitation in hippocampal neurons. Brain Res. 1982,
250, 71-92.
[110] Wang, H. S.; Pan, Z.; Shi, W.; Brown, B. S.; Wymore, R. S.;
Cohen, I. S.; Dixon, J. E.; McKinnon, D. KNCQ2 and KNCQ3
potassium channels subunits: molecular correlates of the M-
channel. Science 1998, 282, 1890-1893.
[111] Robbins, J. KCNQ potassium channels: physiology,
pathophysiology, and pharmacology. Pharmacol. Ther. 2001, 90,
1-19.
[112] Charlier, C.; Singh, N. A.; Ryan, S. G.; Lewis, T. B.; Reus, B. E.;
Leach, R. J.; Leppert, M. A pore mutation in an idiopathic epilepsy
family. Nat. Genet. 1998, 18, 53-55.
[113] Biervert, C.; Schroeder, B. C.; Kubisch, C.; Berkovic, S. F.;
Propping, P.; Jentsch, T. J.; Steinlein, O. K. A potassium channel
mutation in neonatal human epilepsy. Science 1998, 279, 403-406.
[114] Schroeder, B. C.; Kubisch, C.; Stein, V.; Jensch, T. J. Moderate
loss of function of cyclic-AMP-modulated KNCQ2/KNCQ3 K+
channels causes epilepsy. Nature 1998, 396, 687-690.
[115] Rundfeldt, C. The new anticonvulsant retigabine (D-23129) acts as
an opener of K+ channels in neuronal cells. Eur. J. Pharmacol.
1997, 336, 243-249.
[116] Rundfeldt, C.; Netzer, R. The novel anticonvulsant retigabine
activates M-currents in Chinese hamster ovary-cells tranfected with
human KCNQ2/3 subunits. Neurosci. Lett. 2000, 282, 73-76.
[117] Goldstein, S. A. N.; Bockenhauer, D.; O'Kelly, I.; Zilberberg, N.
Potassium leak channels and the KNCK family of two-P-domain
subunits. Nat. Rev. Neurosci. 2001, 2, 175-184.
[118] Meadows, H. J.; Chapman, C. G.; Duckworth, D. M.; Kelsell, R.
E.; Murdock, P. R.; Nasir, S.; Rennie, G.; Randall, A. D. The
neuroprotective agent sipatrigine (BW619C89) potently inhibits the
human tandem pore-domain K+ channels TREK-1 and TRAAK.
Brain Res. 2001, 892, 94-101.
[119] Singh, N. A.; Westenskow, P.; Charlier, C.; Pappas, C.; Leslie, J.;
Dillon, J.; Anderson, V. E.; Sanguinetti, M. C.; Leppert, M. F.
KCNQ2 and KCNQ3 potassium channel genes in benign familial
neonatal convulssions: expansion of the functional mutation
spectrum. Brain 2003, 126, 2726-2737.
[120] Hirose, S.; Zenri, F.; Akiyoshi, H.; Fukama, G.; Iwata, H.; Inoue,
T.; Yonetani, M.; Tsutsumi, M.; Muranaka, H.; Kurokawa, T.;
Hanai, T.; Wada, K.; Kaneko, S.; Mitsudome, A. A novel mutation
Lees et al.
of KCNQ3 (c.925T-->C) in a Japanese family with benign familial
neonatal convulsions. Ann. Neurol. 2000, 47, 822-826.
[121] Vijai, J.; Kapoor, A.; Ravishankar, H. M.; Cherian, P. J.; Girija, A.
S.; Rajendran, B.; Rangan, G.; Jayalakshmi, S.; Mohandas, S.;
Radhakrishnan et, a. Genetic association analysis of KCNQ3 and
juvenile myoclonic epilepsy in a South Indian population. Human
Genetics 2003, 113, 461-463.
[122] Wallace, R. H.; Scheffer, I. E.; Barnett, S.; Richards, M.; Dibbens,
L.; Desai, R. R.; Lerman-Sagie, T.; Lev, D.; Mazarib, A.; Brand,
N.; Ben-Zeev, B.; Goikhman, I.; Singh, R.; Kremmidiotis, G.;
Gardner, A.; Sutherland, G. R.; George, A. L. J.; Mulley, J. C.;
Berkovic, S. F. Neuonal sodium-channel alpha1-subunit mutations
in generalized epilepsy with febrile seizures plus. Am. J. Hum.
Genet. 2001, 68, 859-865.
[123] Spampanato, J.; Escayg, A.; Meisler, M. H.; Goldin, A. L.
Functional effects of two voltage gated sodium channel mutations
that cause generalized epilepsy with febrile seizures plus type 2. J.
Neurosci. 2001, 21, 7481-7490.
[124] Ito, M.; Nagafuji, H.; Okazawa, H.; Yamakawa, K.; Sugawara, T.;
Mazaki-Miyazaki, E.; Hirose, S.; Fukama, G.; Mitsudome, A.;
Wada, K.; Kaneko, S. Autosomal dominant epilepsy with febrile
seizures plus with missense mutations of the (Na+)-channel alpha 1
subunit gene, SCN1A. Epilepsy Res. 2002, 48, 15-23.
[125] Ceulemans, B. P.; Claes, L. R.; Lagae, L. G. Clinical correlations
of mutations in the SCN1A gene: from febrile seizures to severe
myoclonic epilepsy in infancy. Pediatr. Neurol. 2004, 30, 236-243.
[126] Sugawara, T.; Tsurubuchi, Y.; Fujiwara, T.; Mazaki-Miyazaki, E.;
Nagata, K.; Montal, M.; Inoue, T.; Yamakawa, K. Nav1.1 channels
with mutations of severe myoclonic epilepsy in infancy display
attenuated currents. Epilepsy Res. 2003, 54, 201-207.
[127] Sugawara, T.; Tsurubuchi, Y.; Agarwala, K. L.; Ito, M.; Fukuma,
G.; Mazaki-Miyazaki, E.; Nagafuji, H.; Noda, M.; Imoto, K.;
Wada, K.; Mitsudome, A.; Kaneko, S.; Montal, M.; Nagata, K.;
Hirose, S.; Yamakawa, K. A missense mutation of the Na+ channel
alpha II subunit gene Na(v)1.2 in a patient with febrile and afebrile
seizures causes channel dysfunction. Proc. Nat. Acad. Sci. U. S. A.
2001, 98, 6384-6389.
[128] Berkovic, S. F.; Heron, S. E.; Giordano, L.; Marini, C.; Guerrini,
R.; Kaplan, R. E.; Gambardella, A.; Steinlein, O. K.; Grinton, B.
E.; Dean, J.T.; Bordo, L. Hodgson, B.L.; Yamamoto, T.; Mulley,
J.C.; Zara, F.; Scheffer, I.E. Benign familial neonatal-infantile
seizures: characterization of a new sodium channelopathy. Ann.
Neurol. 2004, 55, 550-557.
[129] Wallace, R. H.; Scheffer, I. E.; Parasivam, G.; Barnett, S.; Wallace,
G. B.; Sutherland, G. R.; Berkovic, S. F.; Mulley, J. C. Generalized
epilepsy with febrile seizures plus: mutation of the sodium channel
subunit SCN1B. Neurology 2002, 58, 1426-1429.
[130] Chioza, B.; Osei-Lah, A.; Nashef, L.; Suarez-Merino, B.; Wilkie,
H.; Sham, P.; Knight, J.; Asherson, P.; Makoff, A. J. Haplotype and
linkage disequilibrium analysis to characterise a region in the
calcium channel gene CACNA1A associated with idiopathic
generalised epilepsy. Eur. J. Hum. Genet.: EJHG 2002, 10, 857-
864.
[131] Khosravani, H.; Altier, C.; Simms, B.; Hamming, K. S.; Snutch, T.
P.; Mezeyova, J.; McRory, J. E.; Zamponi, G. W. Gating effects of
mutations in the Cav3.2 T-type calcium channel associated with
childhood absence epilepsy. J. Biol. Chem. 2004, 279, 9681-9684.