Mar. Drugs 2006, 4, 37-69
Marine Drugs
ISSN 1660-3397
© 2006 by MDPI
www.mdpi.org/marinedrugs
Special Issue on “Marine Drugs and Ion Channels” Edited by Hugo Arias
Review
Marine Toxins Targeting Ion Channels
Hugo R. Arias
Department of Pharmaceutical Sciences, College of Pharmacy, Western University of Health Sciences,
Pomona, California 91766, USA
Tel. (909) 469-5424, Fax (909) 469-5600, E-mail: harias@westernu.edu
Received: 10 January 2006 / Accepted: 27 March 2006 / Published: 6 April 2006
Abstract: This introductory minireview points out the importance of ion channels for cell
communication. The basic concepts on the structure and function of ion channels triggered
by membrane voltage changes, the so-called voltage-gated ion channels (VGICs), as well as
those activated by neurotransmitters, the so-called ligand-gated ion channel (LGICs), are
introduced. Among the most important VGIC superfamiles, we can name the voltage-gated
Na+ (NaV), Ca2+ (CaV), and K+ (KV) channels. Among the most important LGIC super
families, we can include the Cys-loop or nicotinicoid, the glutamate-activated (GluR), and
the ATP-activated (P2XnR) receptor superfamilies. Ion channels are transmembrane
proteins that allow the passage of different ions in a specific or unspecific manner. For
instance, the activation of NaV, CaV, or KV channels opens a pore that is specific for Na+,
Ca2+, or K+, respectively. On the other hand, the activation of certain LGICs such as
nicotinic acetylcholine receptors, GluRs, and P2XnRs allows the passage of cations (e.g.,
Na+, K+, and/or Ca2+), whereas the activation of other LGICs such as type A γ-butyric acid
and glycine receptors allows the passage of anions (e.g., Cl− and/or HCO3−). In this regard,
the activation of NaV and CaV as well as ligand-gated cation channels produce membrane
depolarization, which finally leads to stimulatory effects in the cell, whereas the activation
of KV as well as ligand-gated anion channels induce membrane hyperpolarization that
finally leads to inhibitory effects in the cell. The importance of these ion channel
superfamilies is emphasized by considering their physiological functions throughout the
body as well as their pathophysiological implicance in several neuronal diseases. In this
regard, natural molecules, and especially marine toxins, can be potentially used as
modulators (e.g., inhibitors or prolongers) of ion channel functions to treat or to alleviate a
specific ion channel-linked disease (e.g., channelopaties).
Mar. Drugs 2006, 4
38
Keywords: voltage-gated ion channels, ligand-gated ion channels, natural toxins, marine
toxins.
1. Introduction
The communication between individual neurons is performed at a morphologically and functionally
highly specialized region named the synapse (reviewed in [111]). The name synapse was coined from
the Greek syn, meaning “together”, and haptein, meaning “to clasp”. The synapse is structurally
shaped by the presynaptic and the postsynaptic membrane both separated by the synaptic cleft. Figure
1 shows a simple scheme of the synapse. A very interesting review on the mechanisms of vertebrate
synaptogenesis has been recently published [104].
Figure 1. Diagram showing the most important components of the synapse, including the pre(axon terminal) and the post-synaptic cells (dendritic spine) and the synaptic cleft (taken from
www.en.wikipedia.org/wiki/Synapse). CaV channels involved in neurotransmitter release,
vesicles for neurotransmitter storage, and neurotransmitter re-uptake transporters are depicted
in the presynaptic cell. Neurotransmitters are released from the presynaptic membrane to the
synaptic cleft. Neurotransmitters reach the postsynaptic membrane by diffusion, and then
activate neurotransmitter receptors. Neurotransmitter transporters are also shown in the
presynaptic membrane.
An important concept on the synaptic mechanism is that the transference of information between
cells is achieved neither by a direct contact nor electrically but chemically. The chemo-electrical
communication mechanism starts with the elicited action potential on the nerve ending which activates
voltage-gated Na+ (NaV) channels (reviewed in [81]). The opening of voltage-gated Na+ channels
depolarizes the presynaptic membrane. The depolarization process activates both voltage-gated Ca2+
(CaV) and K+ (KV) channels. On one hand, the opening of CaV channels enhances Ca2+ permeation in
Mar. Drugs 2006, 4
39
the presynaptic membrane (reviewed in [84]). The raised intracellular Ca2+ concentration triggers a
series of lipid membrane-vesicle fusion processes (reviewed in [91,101]) that finally produce
neurotransmitter release. More specifically, the increase in intracellular Ca2+ activates calmodulin, and
subsequently protein kinase C, which in turn phosphorylates other proteins that are involved in the
process of vesicle mobilization, docking, and priming (reviewed in [104, 111]). On the other hand, the
activation of KV channels provokes membrane hyperpolarization, stopping finally, the presynaptic
action potential and thus, neurotransmitter release. Some ligand-gated ion channels (LGICs) with high
permeability to Ca2+ (e.g., N-methyl-D-aspartate-subtype glutamate receptor and the α7 nicotinic
acetylcholine receptor subtype) can also induce these intracellular processes (reviewed in [77]). A
specialized region of the presynaptic membrane called “active zone” fuses with the neurotransmittercontaining vesicles by means of different fusion proteins. Among these proteins we can name the
synaptophysin-synaptobrevin complex, a hallmark of synaptic vesicle maturation (e.g., see [15]). The
vesicles contain several additional fusion proteins located within its lipid membrane bilayer. After
several decades of study, the fusion process is being understood (reviewed in [91,101,111]). Most
steps after vesicle-membrane fusion are actually rationalized (reviewed in [29,39]). For example, it is
known that after the vesicle content is released into the synaptic cleft the time course of
neurotransmitter clearance is between 0.1 and 2.0 ms. The neurotransmitter molecules diffuse through
the synaptic space in less than 0.2 ms reaching concentrations of 1-5 mM. Nevertheless, these
parameters vary between synapses on a single neuron as well as among synapses from different
neurons.
Neurotransmitter release can be modulated by different molecular entities: (1) extracellular and
intracellular messengers (reviewed in [42]), and/or (2) presynaptic receptors (reviewed in [49,59,74]).
For instance, M2-type muscarinic receptors inhibit the release of the neurotransmitter acetylcholine
(ACh) in the presynaptic membrane from the neuromuscular junction. The modulation of
neurotransmitter release may lead to short- or long-term adaptations of the synapse to altered
environmental signals (reviewed in [59,76]). Finally, the chemical information is converted into
electrical currents on the postsynaptic membrane.
The idea of chemical transmission started in the first decade of 1900 (e.g., see [61]), however, it
took about 50 more years to be finally established. The molecules bearing chemical information were
called neurotransmitters. Up to date, hundreds of neurotransmitters and neuropeptides have been
identified as responsible for chemical signaling. In addition, it has been found that a single neuron can
synthesize and liberate several of these substances at the same time. Different neurotransmitters
operate at different parts of the nervous system and have different effects. Some promote the
transmission of impulses while others inhibit it. At least three mechanisms modulating the
neurotransmitter concentration in the synaptic cleft are well known: (a) enzymatic inactivation. For
instance, the neurotransmitter ACh is catabolized to choline and acetate by neuronal
acetylcholinesterases; (b) presynaptic membrane re-uptake by specific neurotransmitter transporters.
Figure 1 shows neurotransmitter transporters. This mechanism is used for the re-uptake of monoamine
neurotransmitters such as norepinephrine, epinephrine, dopamine, and serotonin (5-hydroxytryptamine;
5-HT); and (c) soluble neurotransmitter-binding proteins. Several proteins homologous to the soluble
ACh-binding protein (AChBPs) secreted by Lymnaea stagnalis glial cells [90] have been characterized.
For instance, AChBPs from mollusks such as Aplysia californica [47] and Bulinus truncatus [24]. The
Mar. Drugs 2006, 4
40
secretion of these proteins is modulated by the neurotransmitter itself [90] (reviewed in [89]).
Functionally, they act as neurotransmitter carriers and they may buffer the neurotransmitter
concentration in the synapse. Another potential function is that they may transport neurotransmitter
molecules to extrasynaptic sites, allowing the interaction with, for instance, glial cells or other adjacent
neurons. Similar binding proteins for other neurotransmitters might also exist in the central nervous
system (CNS) from vertebrates.
The postsynaptic membrane is highly specialized in the recognition and binding of
neurotransmitters by means of lipid-embedded protein receptors. On the basis of their elicited
physiological response two main receptor classes have been distinguished: ionotropic receptors and
metabotropic receptors. The binding of one kind of neurotransmitter to its specific metabotropic
receptor produces a slow biological response. The mechanism by which the activated receptor
produces its final effect on the cell is mediated not directly but indirectly by the enzymatic interaction
with one of the members of the G-protein family. An example of metabotropic receptor is the
adrenergic receptor. On the other hand, the specific binding of a ligand to its ionotropic receptor
induces a fast opening of the ion channel intrinsically coupled to the receptor. Figure 1 shows the
location of postsynaptic ionotropic receptors. Interestingly, perisynaptic or extracellular ionotropic
receptors have also been found (reviewed in [35,49,59]). The function of these receptors is not totally
clear, but some perisynaptic receptor types may produce separate and sometimes opposite effects as its
postsynaptic counterpart (reviewed [39,59]).
2. Ion channels
Ion flux across excitable cells mediated by activation of ion channels is a signaling mechanism that
it is extensively used by different organisms from the three major superkindoms of life. Ion channels
are responsible for generating and orchestrasting the electrical signals throughout the body. These
signals ultimately modulate the thinking process in the brain, the heartbeat in the heart, and the muscle
contraction in the muscle, to name only a few important physiological functions. Using methods such
as molecular biology and electrophysiology, researchers have cloned, expressed, and characterized
genes encoding many of these ion channels.
Ion channels may be gated by different signals, including membrane voltage changes, mechanical
forces, and chemicals. Ion channels that are triggered by specific endogenous ligands (e.g.,
neurotransmitters, neuropeptides, etc.) are called LGICs. LGICs are genetically, structurally, and
functionally different from voltage-gated ion channels (VGICs). One of the main distinctions is that
VGICs are gated not by endogenous ligands but by membrane voltage changes (reviewed in
[17,81,95]).
Several ion channels from different organs and tissues have been implicated in the pathophysiology
of distinct diseases (reviewed in [2]). For instance, mutations in the epithelial chloride channel , CFTR
(the cystic fibrosis transmembrane regulator), are responsible for cystic fibrosis, mutations in the renal
chloride channel are responsible for hypercalciuric nephrolithiasis (Dent’s disease), truncation in the
ATP-sensitive pancreatic potassium channel is involved in familial persistent hyperinsulinemic
hypoglycemia from the infancy, mutations in the epithelial sodium channel are related with Liddle’s
syndrome (produces hereditary hypertension and pseudoaldosteronism), mutations in the skeletal-
Mar. Drugs 2006, 4
41
muscle chloride channel are involved with different myotonias (e.g., Becker’s or Thomsen’s myotonia),
mutations in the ryanodine calcium channel are related with the central core storage disease and with
malignant hyperthermia. The list of the most important diseases related with VGICs or LGICs are
provided in the following sections.
In addition, ion channels are the targets of several marine toxins. One of the most important toxins
from the point of view of their diversity and specificity are the snail toxins obtained from different
Conus species. In this regard, Layer and McIntosh review, in this special issue, the most important
structural details of these conotoxins as well as their therapeutical potential for the treatment of
different diseases.
3. Voltage-Gated Ion Channels
Voltage-gated ion channels are complex proteins that are embedded in the lipid membrane of the cell.
These channels conduct ions at very high rates (~1 million ions per second) and are regulated by the
voltage across the membrane. The best known VGICs are NaV, KV, and CaV channels, as well as
voltage-gated Cl− channels. This classification corresponds to the type of ion that each channel allows
to pass. Subunits homologous to subunit α from the different VGICs form the structure of the ion pore.
Subunit α also bears the voltage sensor that allows the channel to detect and gate in response to
changes in the transmembrane voltage (reviewed in [17,95]). The opening of only one of these ion
channels allows the passage of about 10 million ions per second (reviewed in [2]). In this regard, every
time that a channel is open, a current of few picoamperes (pA) is generated (1 Ampere = 1
coulomb/sec = 6.24×1018 electrons moving through a surface in one second). Since these channels are
very efficient, there are only few thousand per cell of a given type. Consistent with the normal
electrochemical gradients across the cell membrane for these ions, the opening of NaV or CaV channels
induces membrane depolarization by allowing positive Na+ or Ca2+ ions flow into the cell. In contrast,
the opening of KV or voltage-gated Cl− channels induces membrane hyperpolarization (K+ exits from,
whereas Cl− enters, the cell, increasing the number of negative charges at the cytoplasmic surface of
the membrane). Additional subunits (e.g., α2, β1, β2, and γ) from these ion channels have accessory
functions. For instance, they modulate ion channel function, and interact with cytoskeleton proteins for
anchoring as well as with protein kinases for phosphorylation processes.
Given their physiological importance, VGICs are the targets for numerous small molecules and
toxins of natural origin. Malfunctioning of these VGICs is implicated in many important diseases, and
these ion channels are under intense scrutiny as potential targets for drugs for the treatment of different
diseases. In this regard, Messerli and Greenberg, in this volume, review the effects of Cnidarian toxins
(marine toxins) in VGICs.
3.1. The Voltage-Gated Na+ Channel Superfamily
Voltage-gated Na+ channels were purified from Electrophorus electricus electric organs in 1978 [3].
Since then, a good deal of information on the structure and function of different NaV channels has been
obtained. Mammalian NaV channels from brain are structurally formed by three different subunits, α,
β1, and β2, whereas channels formed by subunits α and β3 are present in dorsal root ganglia, and
those comprised by subunits α and β1 are found in skeletal muscle (reviewed in [36,110]). There are at
Mar. Drugs 2006, 4
42
least nine α subunits forming channels NaV1.1-NaV1.9. The Nomenclature Committee from the
International Union of Pharmacology has recently reviewed and accepted a new nomenclature for NaV
channel types [22,110].
In the NaV channel structure, the α subunit, which makes up the core structure, is formed by four
domains (I-IV), each containing six transmembrane segments (S1-S6) (reviewed in [17,110]). In
contrast, each β1 and β2 subunit has just one transmembrane segment. Figure 2 depicts the schematic
structure of the NaV channel. Interestingly, the ion-conducting pore of the channel is supported by the
S5 and S6 transmembrane segments from the α subunit. The outer part of the pore is lined by a reentrant loop (the P loop) that links the transmembrane segments S5 and S6, while the C-terminal ends
of the S6 segments line the cytoplasmic section of the pore. Certain amino acid residues in the Cterminal segment of the P loop are critical for determining the conductance and ion selectivity of the
Na+ channel. The activation gate is probably formed by the S6 segments at a point where they come
into close apposition in the cytoplasmic half of the pore. The conformational change from closed to
open is ultimately dependent on the motion of the positively charged, S4 segments, which act as
voltage sensors by responding to changes in the electric field within the membrane. The
transmembrane segment S4 is positively-charged and functions as a sensor for voltage changes across
the lipid membrane. The activation of this sensor induces complex conformational changes in the
protein that finally open the pore allowing the flux of Na+ into the cytoplasm of the cell. In vivo, Na+
influx produces a regenerative membrane depolarization in the millisecond time frame. The rapid
depolarization activates both KV channels, which contribute to the rapid repolarizing phase of the
action potential, and CaV channels. The activation of presynaptic CaV channels, in turn, induces
neurotransmitter release. Figure 3B shows a voltage-clamp recording of NaV channel activity in
response to a step depolarization of the membrane.
NaV channels undergo cyclic conformational changes among at least three functionally different
states. Figure 3A shows the three main conformational states. In the resting state, the ion channel is
nonconducting. Membrane voltage changes trigger ion channel opening in the millisecond time regime.
This conducting state is called the activated state. A few milliseconds after the initial activation of NaV
channels, an intrinsic process of inactivation takes place in which the short intracellular loop
connecting domains III and IV serves as the fast inactivation gate of the ion channel. This loop folds
into the intracellular mouth of the ion pore and occludes it, ending the process of membrane
depolarization as well as of neurotransmitter release. The hydrophobic residues isoleucinephenylalanine-methionine (IFM) are considered to serve as the inactivating domain. The inactivated
state is a closed state which is distinct from the resting state. Although the channel inactivation process
is mechanistically different from the desensitization found in LGICs, both processess produce similar
modulatory effects on these ion channels. Finally, the gating process can be modulated by different
protein kinases and phosphatases. Phosphorylation of the inactivation gate by protein kinase C slows
inactivation, whereas phosphorylation of the intracellular loop between domains I and II by either
protein kinase C or A reduces channel activation. Such modulations modulate patterns of electrical
activity in neurons and other excitable cells.
Mar. Drugs 2006, 4
43
Figure 2. Schematic representation of the structure of the NaV channel (modified from [36]).
(A) The α subunit consists of four homologous domains (I-IV), each containing six
transmembrane segments (S1-S6). Positively-charged S4 segments [S4(+)] form the voltage
sensor. Single segments and domains are connected by intracellular and extracellular loops.
The intracellular loop between domain I and II presents several phosphorylation sites (P).
Extracellular loops contain several glycosylation sites (Y). The isoleucine-phenylalaninemethionine (IFM) motif has a crucial role in the process of channel inactivation. The
inactivation gate is depicted in red and includes the IFM motif. (B) Transmembrane
arrangement of domains I-IV (DI-DIV) around the ion pore, seen from the cytoplasmic end.
The inactivation gate is depicted in red, close to the IFM motif [see (A)]. (C) Binding sites (1-5)
for different ligands [e.g., tetrodotoxin (TTX)] as well as for local anesthetics (LAs). The
selectivity filter is depicted in blue.
Different mutations in NaV channels are implicated in several diseases, the so-called
“channelopathies”, from the muscle, heart, or nervous system (reviewed in [2,41,110]). For example,
mutations in cardiac NaV1.5 channels are responsible for long-QT type 3 and Brugada syndromes
which produce cardiac arrhythmias, whereas mutations in the skeletal-muscle NaV1.4 channel are
implicated in “hyperkalemic periodic paralysis” (a muscle weakness disorder aggravated by high K+),
in Masseter-muscle rigidity (succinylcholine-induced), as well as in different kinds of myotonias (e.g.,
Mar. Drugs 2006, 4
44
paramyotonia congenital, a syndrome of cold- and exercise-induced muscle stiffness). In addition,
mutations in the neuronal NaV1.1 channel produce “generalized epilepsy with febrile seizures plus,
type II”, whereas mutations in the β1 subunit produce “generalized epilepsy with febrile seizures plus,
type I”. Finally, knockout mice lacking the NaV1.8 channel are viable but their response to pain is not
normal.
Figure 3. Functional properties of NaV channels (modified from [41]). (A) Schematic
representation of a NaV channel undergoing the major gating transitions (resting ↔ open ↔
inactivated). (B) Voltage-clamp recording of NaV channel activity in response to membrane
depolarization. Downward deflection of the current trace (red) corresponds to inward
movement of Na+.
NaV channels are targets for the action of natural as well as synthetic drugs (reviewed in [2,36,37]).
Among clinically important synthetic drugs we can name anticonvulsants (e.g., carbamazepine,
phenytoin, and valproic acid), local anesthetics (e.g., bupivacaine, cocaine, lidocaine, mepivacaine,
and tetracaine), as well as Class IA (e.g., disopyramide, procainamide, and quinidine), Class IB (e.g.,
lidocaine, mexiletine, phenytoin, and tocainide), and Class IC antiarrhythmics (e.g., encainide,
flecainide, and propafenone). Recent research efforts have focused on the development of analgesic
agents which selectively block dorsal root ganglion sodium channels that are involved in transmission
of pain sensation.
NaV channels present at least six binding sites for synthetic and natural molecules (Sites 1-6, as well
as the binding site for local anesthetics) that are located in different domains throughout the protein
(reviewed in [36]). Table 1 shows the most important natural molecules including marine toxins that
affect NaV channels. Natural drugs targeting NaV channels are chemically very diverse and include
Mar. Drugs 2006, 4
45
polypeptides, alkaloids, cyclic polyethers, esters, and heterocycles. For example, the reversible
blockers tetrodotoxin (from puffer fish and related species from the Tetraodontidae family),
maculotoxin (actually tetrodotoxin present in the blue-ringed octopus Hapalochlaena maculosa),
saxitoxin (from different algal species of the order Dinoflagellata such as Alexandrium tamarense,
Gymnodinium catenatum, and Pyrodinium bahamense, as well as from certain freshwater
cyanobacteria), and µ–conotoxins GIIIA, GIIIB, and GIIIC (polypeptides from the marine snail Conus
geographus) and PIIIA (from C. purpurascens) bind to Site 1. Certain amino acids from the pore
lining in the extracellular mouth of the ion channel are critical for the binding of these toxins. The
main pharmacological action of these toxins is to block current flow through the pore thus, decreasing
membrane depolarization and, as a consequence, also inhibiting neurotransmitter release. The toxic
effects induced by these toxins include diarrhea, vomiting, paralysis, paraesthesia, dyspnea,
hypotention, convulsions, mental impairments, cardiac arrhythmia, respiratory failure, and finally
death.
Local anesthetics also block the ion channel but its binding site is located in a different domain
from that for Site 1. Interestingly, local anesthetics pharmacologically behave as noncompetitive
antagonists (NCAs) of nicotinic acetylcholine receptors (AChRs) (reviewed in [9,12]). Batrachotoxin
(alkaloid from the skin of Colombian arrow poison frogs of the genus Phyllobates), veratridine
(alkaloid from plants of the Liliaceae family), aconitine (alkaloid from the plant Aconitum napellus),
and grayanotoxin (diterpenes from several Rhododendron species) bind to Site 2. The main action
elicited by these drugs is to produce persistent channel activation. In the heart, batrachotoxin,
veratridine, and grayanotoxin, but not aconitine, produce positive ionotropic effects (i.e., increased
contraction force). Class 1 scorpion α-toxins (from the genera Buthus, Androctonus, and Leiurus), δatracotoxins (from Australian funnel web spiders of the Atracinae family such as Atrax robustus which
produces robustoxin and Hadronyche versutus that produces versutoxin), µ-agatoxin-I and -IV (from
the web spider Agelenopsis aperta), sea anemone toxins (from the genera Anthopleura and Anemonia),
several toxins from the spider Phoneutria nigriventer, and δ–conotoxins such as PVIA (from C.
purpurascens) as well as TxVIA (also called King-Kong peptide TxIA), TxIB, and TxIIA (from C.
textile neovicarius) bind to Site 3. This site is located in the loop formed between the transmembrane
segments S3 and S4 from domain IV. The main pharmacological effects of these toxins are to slow
inactivation, and/or enhance or prolong activation. Persistent channel activation sustains membrane
depolarization, enhancing neurotransmitter release, leading to rigid paralysis and, finally, death.
Toxins that bind to Site 3 enhance the activity of molecules acting on Site 2. Scorpion β-toxins from
the genus Centuroides bind to Site 4. Class 2 Scorpion β-toxins shift the voltage-dependence of
activation to more negative membrane potentials without modifying the channel inactivation process,
whereas Class 3 Scorpion β-toxins alter the voltage-dependence of both activation and inactivation.
The lipid-soluble polyether marine toxins brevetoxins (produced by the dinoflagellate Karenia brevis)
and ciguatoxins (from several dinoflagellate species including Gambierdiscus toxicus) bind to Site 5.
The effects elicited by these NaV channel activators include repetitive neuronal firing, shifting the
voltage-dependence, and blocking the channel inactivation process. Finally, insecticides such as
pyrethroids (from the flowers of certain Chrysanthemum species) and dichlorodiphenyltrichloroethane
(DDT) bind to Site 6, a poorly defined binding domain. These compounds produce hyperexcitability
and paralysis preferentially in invertebrates rather than in mammals. Sites of action of fast inactivation
Mar. Drugs 2006, 4
46
inhibitors such as pronase, N-bromoacetamide and chloramine-T have not been positively identified as
well as for prolongers including DPI 201-106 and BDF9148 that are in experimental stage as
therapeutics for the treatment of congestive heart failure. Finally, the location of binding sites for other
natural toxins are still under debate. These include homobatrachotoxin (steroidal alkaloids from birds
of the genus Pitohui) which activates NaV channels, and phoneutriatoxins (from the Banana spider
Phoneutria nigriventer) as well as α- and β-pompilidotoxins (from the solitary wasps Anoplius
samariensis and Batozonellus maculifrons) which slow the NaV channel inactivation process.
Table 1. Binding site location for different toxins and drugs on the NaV channel.
Bindig Site
Toxin/Drug (origin)
Pharmacological Effect
1
Tetrodotoxin (marine)
Ion channel block
Saxitoxin (marine)
µ-Conotoxins GIIIA and GIIIB (marine)
2
Veratridine (plant)
Persistent activation
Batrachotoxin (animal)
Aconitine (plant)
Grayanotoxin (plant)
3
Scorpion α-toxins (animal)
Slow inactivation
Sea anemone toxins (marine)
Enhancement of persistent activation
δ-Atracotoxins (animal)
δ–Conotoxins PVIA,
TxIA, TxIB, and TxIIA (marine)
Scorpion β-toxins (animal)
4
Shift voltage-dependent activation to
more negative potentials
5
6
Local anesthetic
Brevetoxins (marine)
Shift voltage-dependent activation to
Ciguatoxins (marine)
more negative potentials
Pyrethroids (plant)
Slow
DDT (synthetic)
deactivation processes
Local anesthetics
Ion channel block (use-dependent)
activation,
inactivation
[synthetics in general; plant product (e.g.,
cocaine)]
Additional
binding
sites
Additional
sites
DPI 201-106 (synthetic)
Prolongation of action potential
BDF9148 (synthetic)
binding
Pronase (synthetic)
N-bromoacetamide (synthetic)
Chloramine-T (synthetic)
Disruption of fast inactivation
and
Mar. Drugs 2006, 4
47
In this special issue, French and collaborators review the general aspects of the structural and
functional properties of marine toxins that interact with NaV channels. In addition, Geffeny and Ruben
cover the information regarding the specific interaction of tetrodotoxin with NaV channels, and finally
Nicholson and Lewis present updated information regarding the pharmacological effects of
ciguatoxins on NaV channels.
3.2. The Voltage-Gated Ca2+ Channel Superfamily
CaV channels were first identified in crustacean muscle fibers by Fatt and Katz [40]. The use of
molecular biology has helped to identify subclasses of these channels in all types of excitable cells in
vertebrates and invertebrates, and even in plants. CaV channels can be classified depending on their
activity or site of activity. For instance, in neurons, CaV2.1 and CaV2.2 channels are located
presynaptically and are involved in neurotransmitter release. Almost all other types have been found in
cell bodies and dendrites. The currently accepted nomenclature for CaV channels is found in Catterall
et al. [23], and summarized here in Table 2.
CaV channels are structurally formed by several subunits including α1, α2, δ, β, and γ (reviewed in
[95,109]). The α1 subunit is formed by six transmembrane segments, and contains the ion channel, the
voltage sensor, and the gate. This structural arrangement is shown in Figure 4, and it is similar to NaV
channels (see Fig. 2). The β subunit is located intracellularly and it is involved in membrane
trafficking of α1 subunits. The α2 subunit is a highly glycosylated extracellular protein that is attached
to the membrane-spanning δ subunit by means of disulfide bonds. The α2 subunit provides structural
support, whereas the δ subunit modulates the voltage dependent activation and steady-state
inactivation processes. The γ subunit is a glycoprotein having four transmembrane segments.
The biophysical properties of these channels were studied by using voltage- and patch-clamp
methods. CaV channels are normally closed at resting membrane potentials, and open upon membrane
depolarization. The resultant current through the cell membrane can be characterized by a number of
properties including the membrane potential range over which the channel reaches its maximum open
probability as well as the current kinetics. The use of selective drugs and toxins has helped to identify
specific current components observed in an entire cell that correspond to a particular channel type.
There is accumulating evidence to support the view that CaV channels are implicated in several
diseases and channelopathies (reviewed in [2]). For instance, autoantibodies against P/Q-type CaV
channels are produced in the Lambert-Eaton syndrome, reducing the number of these channels in the
presynaptic cell and provoking myasthenic syndrome (i.e., muscle weakness). Mutations in
dihydropiridine-sensitive CaV channels, more specifically in the S4 region, are related with the muscle
disorder “hypokalemic period paralysis”. Other mutations have been implicated in diseases such as
spino-cerebellar ataxia (type 6), familial hemiplegic migraine, episodic ataxia (type 2), and X-linked
congenital night blindness among others.
Mar. Drugs 2006, 4
48
Table 2. Effect of several drugs on different CaV channel types.
Channel
Current
Tissue Localization
Drugs (origin)
Pharmacological effect
Skeletal muscle
Dihydropiridines (synthetic)
Contraction
Transverse tubules
Phenylalkylamines (synthetic)
Type
CaV1 Channels
CaV1.1
L
Benzothiazepines (synthetic)
CaV1.2
L
Cardiac myocytes
Dihydropiridines (synthetic)
Contraction
Endocrine cells
Phenylalkylamines (synthetic)
Hormone release
Neuronal cell bodies and
Benzothiazepines (synthetic)
Synaptic integration
Endocrine cells
Dihydropiridines (synthetic)
Hormone release
Neuronal cell bodies and
Phenylalkylamines (synthetic)
Synaptic integration
proximal dendrites
Benzothiazepines (synthetic)
Retina
Not established
Neurotransmitter release
Nerve terminals and
ω-Agatoxins IVA and IVB (animal)
Inhibits neurotransmitter
proximal dendrites
CaV1.3
CaV1.4
L
L
CaV2 Channels
CaV2.1
P/Q
dendrites
CaV2.2
CaV2.3
N
R
release
Nerve terminals and
ω-Conotoxins GVIA, MVIIC, and
Inhibits neurotransmitter
dendrites
MVIIA (marine)
release
Neuronal cell bodies and
SNX-482 (animal)
Inhibits repetitive firing
Neuronal cell bodies and
Kurtoxin (animal)
Inhibits repetitive firing
dendrites
Mibefradil (synthetic)
Inhibits pacemaker
Cardiac myocytes
Pimozide (synthetic)
Neuronal cell bodies and
Kurtoxin (animal)
Inhibits repetitive firing
dendrites
Mibefradil (synthetic)
Inhibits pacemaker
Cardiac myocytes
Pimozide (synthetic)
Neuronal cell bodies and
Kurtoxin (animal)
dendrites
Mibefradil (synthetic)
dendrites
CaV3 Channels
CaV3.1
CaV3.2
CaV3.3
T
T
T
Inhibits repetitive firing
Pimozide (synthetic)
L = long-lasting currents; P/Q =; R = ; T = transient currents; N = for their involvement in
norepinephrine.
Mar. Drugs 2006, 4
49
Figure 4. Schematic representation of the structure of the CaV channel (taken from
www.omedon.co.uk/ionchan/channels/calcium). The α1 subunit consists of four homologous
domains (I-IV; depicted in different colors), each containing six transmembrane segments (S1S6). This subunit contains the ion conducting pore, the voltage sensor, and the gate. The β
subunit is located intracellularly and it is involved in membrane trafficking of α1 subunits. The
α2 subunit is attached to the membrane-spanning δ subunit by means of disulfide bonds. The
α2 subunit provides structural support, whereas the δ subunit modulates the voltage-dependent
activation and steady-state inactivation processes. The γ subunit is a glycoprotein having four
transmembrane segments.
Different CaV channel families are targets for distinct therapeutically relevant drugs (reviewed in
[2,37,109]). Table 2 shows the pharmacological effect of several drugs including marine toxins on
different CaV channels. For instance, the CaV1 channel family (L-type currents) is the target for the
pharmacological action of different drugs used in the treatment of cardiovascular diseases including
phenylalkylamines, dihydropyridines, and benzothiazepines (class IV antiarrhythmics). Interestingly,
these classes of drugs bind to different sites. Phenylalkylamines (e.g., verapamil, devapamil,
gallopamil, and thiapamil) enter the pore from the cytoplasmic side and bind to a site located at
segment S5 from domain IV, similar to that for local anesthetics in the NaV channel, and occlude the
pore. Dihydropyridines (e.g., nifedipine, nicardipine, nitrendipine, nimodipine, nisoldipine, and
felodipine) inhibit the ion channel by an allosteric mechanism. This site is formed by segment S6 from
domains III and IV and segment S5 from domain III. Other dihydripyridine derivatives (e.g., Bay
K8644) activate CaV channels producing vasoconstriction and positive ionotropic effects.
Benzothiazepines (e.g., diltiazem) access the channel from the extracellular side and bind to a site
located close to the extracellular mouth. Snake toxins such as calcicludine and calciseptine (from
Dendroaspis angusticeps) as well as TaiCatoxin (from Oxyuranus scutellatus scutellatus) are also
Mar. Drugs 2006, 4
50
potent and selective blockers of L-type channels. The CaV2 channel family is specifically blocked by
spider and marine snail toxins. For example, ω-agatoxins IVA and IVB (from the funnel web spider
Agelenopsis asperta) as well as ω-conotoxins MVIIC and MVIIA (from C. magus) inhibit the gating
process in CaV2.1 channels (P/Q-type currents), whereas ω-conotoxins GVIA (from C. geographus)
blocks CaV2.2 channels (N-type currents). Since SNX-482 (from the tarantula Hysterocrates gigas)
specifically inhibits CaV2.3 channels, this toxin was crucial for the characterization of R-type currents.
A polyamine with analgesic properties obtained from the fishing spider Dolomedes okefinokensii
reversibly blocks L-and R-type CaV channels.
On the other hand, the activation gating on CaV3 channels (T-type currents) is inhibited by kurtoxin
(obtained from South African scorpions). Mibefradil and pimozide block CaV3 channels slightly more
strongly than either CaV1 or CaV2 channels. In addition, ω-grammotoxin SIA (obtained from the South
American rose tarantula Grammostola spatulata), HWTX-I (obtained from the Chinese bird spider),
taicatoxin (obtained from the Australian Taipan snake), and several spider toxins from Phoneutria
ngriventer, block different CaV channels as well.
Regarding CaV channel marine toxins, Schroeder and Lewis cover the effect of ω-conotoxins and
their therapeutical applications, whereas Zhang is reviewing different aspects of the interaction of SO3 conotoxins with N-type CaV channels.
3.3. The Voltage-Gated K+ Channel Superfamily
The first KV channel sequence was determined from the Shaker mutant of Drosophila melanogaster
in 1987 [94]. Since then, a great effort has been done to elucidate the structural arrangement of this
channel family that recently culminated in the crystallographic structure of the mammalian Shaker KV
channel [71]. KV channels are more diverse and widespread than other ion channel families (reviewed
in [76]). In this regard, KV channels have been divided in nine families (e.g., KV1-KV9) (reviewed in
[95]). KV1-KV4 channels are also called Shaker, Shab, Shaw, and Shal, respectively. The current
nomenclature for these channels is found in Gutman et al. [46]. Regarding their assembly properties,
families KV1-KV4 can form homomeric or heteromeric channels, whereas families KV5-KV9 cannot
form homomeric channels but interact with members from the KV2 family to form heteromeric
channels. This is why subunits KV5-KV9 are called modulatory subunits.
Figure 5 depicts the most important structural features of a prototypic KV channel. KV channels are
composed of four separate α subunits, each containing six transmembrane segments (S1-S6) (reviewed
in [95]). The α subunits are assembled as a tetramer to form a central pore that is specific for K+ ions.
Four accessory β subunits (e.g., β1 and β2) may be complexed with the α subunits in a stoichiometry
α4β4 to form the functional KV channel. The α subunit determines the basic gating properties. More
specifically, the peptide chain (H5 or loop P) between segments S5 and S6 lines the water-filled
channel pore. This region is highly conserved in all VGICs. Mutations in this region alter the
permeation properties of the ion channel. As for NaV channels, KV channels have an S4 segment,
bearing a cluster of positively-charged residues (e.g., lysines and arginines), that serves as the voltage
sensor. Voltage-dependent fast inactivation (N-type) of the channel is mediated by a long chain of
amino acids with a ball in the amino terminal that swings into the mouth of the ion channel and blocks
the permeation pathway from the cytoplasmic side (i.e., the “ball and chain” model). Several residues
Mar. Drugs 2006, 4
51
in S6 are involved in the slow inactivation process (C-type). Figure 5 illustrates the basic “ball and
chain”model.
Figure 5. Schematic representation of the structure of the KV channel (modified from [2]). (A)
Each α subunit consists of six transmembrane segments (S1-S6). Positively-charged amino
acids (e.g., lysine and arginine) in the S4 segment form the voltage sensor. Key residues lining
the channel pore (H5 or P loop) are found between S5 and S6. The “ball and chain” structure at
the N-terminal participates in N-type “fast inactivation”, occluding the permeation pathway,
whereas certain residues at S6 are involved in the C-type “slow inactivation” process. Several
phosphorylation sites (P) are present in the C-terminal domain. (B) The proper KV channel is
assembled by the interaction of four α subunits. The pore region appears to have wide
intracellular and extracellular vestibules (~28-34 Å wide and 4-8 Å deep) that lead to a
constricted pore of 9-14 Å in diameter at its entrance, tapering to a diameter of 4-5 Å at a depth
of ~6 Å from the vestibule. K+ ions (small green balls) are passing through the ion filter from
the cytoplasmic to the extracellular side of the ion channel (efflux).
KV channel mutations are implicated in several important diseases (reviewed in [76]). Among them
we can name cardiac arrythmias (e.g., long-QT syndromes), congenital deafness, epilepsy, diabetes,
Mar. Drugs 2006, 4
52
misregulation of blood pressure, multiple sclerosis, Alzheimer's disease and other dementias, migraine,
pain, anxiety and bipolar disorder.
KV channels are also the target of several synthetic and natural drugs (reviewed in [2,37,54,86]).
Among synthetic drugs we can name antidiabetic drugs (e.g., glipizide, glyburide, and tolazamide),
antihypertensive drugs (e.g., diazoxide and minoxidil), and Class III antiarrhythmics (e.g., amiodarone,
clofillium, dofetilide, and sotalol). Among nonselective blockers we can include inorganic cations (e.g.,
Li+, Cs+, Ba+, Hg+, and Zn2+) and quaternary ammonium derivatives (e.g., tetramethy-, tetrahethyl-,
tetrabutyl-ammonium derivatives). Aminopyridines (e.g., 4-aminopyridine, and 3,4-diaminopyridine)
are more selective drugs in clinical trial for the treatment of long-standing spinal cord injury. These
drugs assist recovery of sensory and motor functions, enhance pulmonary functions, and diminish
muscle spasticity.
Regarding natural toxins, Scorpionidae toxins including charybdotoxin (obtained from Leiurus
quinquestriatus hebraeus), agitoxin-2 (obtained from Leiurus quinquestriatus hebraeus), margatoxin
(obtained from Centruroides margaritatus), noxiustoxin (obtained from Centruroides noxius),
anuroctoxin (obtained from Anuroctonus phaiodactylus), tityustoxin-Kα and -Kβ (obtained from the
Brazilian scorpion Tityus serrulatus), hongotoxin-1 (isolated from Centruroides limbatus), and
parabutoxin-3 (obtained from Parabuthus transvaalicus) selectively block KV channels. For instance,
parabutoxin 3 and hongotoxin-1 block several KV1 channels, whereas margatoxin, noxiustoxin, and
agitoxin-2 prefer the KV1.3 subtype. Agitoxin-2 interacts with the outer vestibule of the ion channel.
Several snake toxins such as α- and β-dendrodotoxin [obtained from green (Dendroaspis viridis) and
black mamba (Dendroaspis polylepis)] are KV channel blockers as well. Unlike scorpion toxins which
act by pore occlusion, spider toxins appear to bind to four equivalent sites at the surface of the KV
channel (one per subunit) close to the voltage sensor and thus, modify the gating process. Among them,
we can include phrixotoxin-1 (obtained from the Chilean fire tarantula Phrixotrichus auratus),
stromatoxin (obtained from Stromatopelma calceata), hanatoxin-1 and -2 (obtained from the Chilean
rose tarantula Grammostola spatulata), heteropodatoxin-2 (obtained from the huntsman spider
Heteropoda venatoria), as well as heteroscodratoxin-1 and -2 (obtained from Heteroscodra maculata).
These toxins inhibit KV2 and KV4 channel families. Other natural toxins include MCD peptide
(obtained from the honeybee Apis mellifera) which has high affinity for KV1.1, KV1.2, and KV1.6
channel subtypes, stichodactyla toxin (obtained from the sea anemone Stichodactyla helianthus),
phalloidine (alkaloid obtained from the death-cap mushroom Amanita phalloides), and correolide
derivatives (triterpenes purified from the root and bark of the Costa Rican tree Spachea correae) that
are selective blockers of KV1 channels and may be used as immunosuppressants by inhibiting the
lymphocyte KV1.3 channel. Among marine toxins we can name κ-conotoxin PVIIA (obtained from C.
purpurascens) which is the first identified marine toxin that blocks Drosophila KV channels. However,
no mammalian channel has been shown to be sensitive to the action of this conotoxin. Interestingly, a
new member from the I-conotoxin superfamily, conotoxin ViTx (obtained from C. virgo), inhibits the
vertebrate channels KV1.1 and KV1.3 but not KV1.2 [53], whereas conkunitzin-S1 (obtained from C.
striatus) and 6-bromo-2-mercaptotryptamine (obtained from the marine snail Calliostoma
canaliculatum) interact with the Shaker KV channel. Ciguatoxin-1 (obtained from the dinoflagellate
Gambierdiscus toxicus) is also a blocker from these ion channels. In general, these blockers prolong
Mar. Drugs 2006, 4
53
the duration of presynaptic action potentials in neurons, increasing Ca2+ influx by indirect CaV
activation, enhancing neurotransmitter release and subsequently, provoking strong convulsions.
4. Ligand-Gated Ion Channels
Among LGIC receptors, four different superfamilies have been described in detail, namely the Cysloop or nicotinicoid receptor superfamily, the ionotropic glutamate receptor (GluR) superfamily, the
ATP-activated receptor (P2XnR) superfamily (reviewed in [7,8,69]), and the transient receptor
potential-canonical channels (reviewed in [102]). These receptor superfamilies are considered different
from each other based upon genetical, structural, and functional properties. An updated data base for
the LGIC receptor superfamilies can be found in the following web page:
www.pasteur.fr/recherche/banques/LGIC/LGIC.html (see [64,65]).
4.1. Structural and Functional Similitude Among LGIC Superfamilies
Although the Cys-loop receptor, the GluR, and the P2XnR superfamily are genetically and
structurally different among them, all receptor members present at least some common features. Figure
3 shows the most important structural characteristics that are preserved throughout the different LGIC
receptor superfamilies. From the structural point of view, we can generalize that these LGIC receptor
superfamilies are formed by three main modules: a large extracellular portion, the transmembrane
portion, and a smaller cytoplasmic portion. The extracellular domain contains the neurotransmitter
binding sites and several consensus sequences involved in glycosylation. The transmembrane domain
has two faces, one internal face that is mainly related with the formation of the intrinsic ion channel
and another external face which interacts with the lipid membrane. The aqueous pore allows the
passage of different cations or anions across the cell membrane finally producing excitatory (for cation
flux) or inhibitory (for anion flux) cell responses. The cytoplasmic portion serves as a structural link to
cytoskeleton proteins for receptor anchorage, clustering, stabilization, and assembly (reviewed in [50]).
This domain is also related with receptor modulation (e.g., desensitization) by phosphorylationdephosphorylation processes (e.g., reviewed in [19,49,56,58,93]).
From the functional point of view, LGIC receptor superfamilies also have some basic aspects that
are common among them. LGIC superfamilies present a very simple repertoire of functional properties.
Each receptor recognizes its specific neurotransmitter, and upon binding, the intrinsically-coupled ion
channel is opened, augmenting in turn the possibility of ions to cross the lipid membrane. Thus, after
channel opening, a new ionic concentration is found at the aqueous solutions bathing the opposed
faces of the lipid bilayer of the cell. In particular, the cytoplasmic compartment will have a higher
+
content of Na or Ca2+ (for cation channels) or of Cl– or HCO3– (for anion channels). The increased
cation concentration in the cytoplasm produces membrane depolarization, whereas the increased anion
concentration in the cytoplasm induces membrane hyperpolarization.
Considering the high level of complexity of the neuronal network, the activation of certain cation
ion channels can result however in either neuronal inhibition or disinhibition (reviewed in [4]). The
opposite is also true, the activation of certain anionic channels may finally produce membrane
depolarization in immature neurons (e.g., see review by [39]).
Mar. Drugs 2006, 4
54
In general, these receptor superfamilies are functionally involved in higher order brain mechanisms
such as cognition, learning, memory, and behavior, as well as in peripheral functions including muscle
contraction, bone formation, cardiac and gastrointestinal modulation, embryonic development, sensory
perception, nociception, and pain processing (reviewed in [38,43,49,70,103]). The malfunctioning of
these receptors has been considered as the origin of several neurological disorders as well as drug
addiction (e.g., see [38,43,49,70,103]).
All these biologically relevant receptor properties are triggered by the binding of the specific
neurotransmitter. Upon binding, the receptor protein undergoes several conformational changes.
Several lines of experimental evidence suggest the existence of a high number of intermediate states
(reviewed in [8,26,30,82]). To make it simpler, however, we will assume that receptors exist in a
minimum of three interconvertible states. The diagram indicating the existence of at least three
receptor states is shown in Fig. 6. In the absence of the specific neurotransmitter, most receptors are in
the resting state (R). The diagram shows a generalized receptor with just one neurotransmitter binding
site. However, up to five binding sites may exist for pentameric receptors formed by just one subunit
subtype (e.g., homomeric receptors). The resting state is defined by the existence of a closed ion
channel. Nevertheless, the receptor is still activatable. In this regard, the receptor is activated (A) in the
presence of the neurotransmitter, and the ion channel is opened in a very fast time frame (R → A). The
time regime is different among receptor subtypes. Although ion channels open with higher probability
in the presence of agonists, thermodynamics fluctuations also allow ion channels to be open but with
much lower probability (e.g., see review by [39]). In the prolonged presence of the neurotransmitter,
the activated receptor is concomitantly commuted to a desensitized state (D) in a timescale that in
general is slower than the activation process (A → D). Neither a specific energy source nor an ionic
gradient is needed to induce the R → A → D conformational shift. The desensitized state is refractory
to the pharmacological action of the neurotransmitter or other agonists, and in this case the ion channel
remains closed. In addition, the desensitized state has a relatively higher affinity for the specific
neurotransmitter as well as for other agonists and some antagonists than that in the resting state.
Different receptor subtypes have distinct desensitized rates and each receptor may present various
desensitized states. For instance, the AChR α7 subtype presents a fast desensitization rate, whereas the
α2β2 subtype practically does not desensitize. In general, the desensitization process is a reversible
process: after agonist washout, the receptor can be activated again (D → R). The receptor can recover
from the desensitized state by means of at least two pathways: a ligand-bound receptor can reopen, or
the neurotransmitter dissociates first and then the receptor can return to the resting state.
Desensitization is a classical example of allosteric protein behavior that considers the previous
existence of discrete conformational states. Physiologically, desensitization might be a major
determinant of synaptic efficacy by controlling the number of available receptors (reviewed in [82]).
5. The Cys-Loop or Nicotinicoid Receptor Superfamily
This superfamily is called “Cys-loop receptor superfamily” because all receptor members conserve
a 15-residue-spaced disulfide loop between amino acids 128 and 142 (corresponding to the AChR α1
subunit) located in the extracellular domain of the receptor. Figure 7 shows the most important
structural characteristics of the Cys-loop receptor superfamily, and its distinction with the ionotropic
Mar. Drugs 2006, 4
55
GluR and P2XnR superfamilies. More recently, the term “nicotinicoid receptor superfamily” has been
imprinted. Examples of this kind of receptors are the members of four families of ion channel-coupled
receptors which are gated by specific neurotransmitters, including type A and C GABA receptors
(GABAAR and GABACR), glycine receptors (GlyRs), type 3 5-HT receptors (5-HT3Rs), and both
neuronaland
muscle-type
nicotinic
receptors
(AChRs)
(reviewed
in
[7,8,13,14,30,32,55,63,67,69,75,78]). The hirerarchy for the different LGIC receptor superfamilies is
summarized in Table 3.
Neurotransmitters
Activated (Open)
ions
Resting (Closed)
Desensitized (Closed)
Figure 6. Diagram showing the general dynamics of multiple conformational states on LGIC
receptor superfamilies. In the absence of neurotransmitter molecules (shown here as red
triangles), the receptor is in the resting state. The receptor is shown here bearing just two
neurotransmitter binding sites, but there is evidence indicating that up to five binding sites may
exist. In the resting state, the ion channel is closed but the receptor is still activatable. After
neurotransmitter binding, the receptor is activated and the ion channel is opened, allowing the
passage of different ions across the lipid membrane. In general, the agonist affinity for the
activated receptor is relatively lower (shown here as a loosed interaction between the
neurotransmitter and its binding site) than that in the desensitized state (shown here as a tight
interaction between the neurotransmitter and its binding site). The transition from the resting to
the activated state is relatively fast, but the time regime is different among receptor subtypes.
In the prolonged presence of the neurotransmitter, the receptor becomes refractive to the
pharmacological action of the neurotransmitter molecule and thus, its ion channel is closed.
Different receptor subtypes have distinct desensitized rates and various desensitized states. In
general, the desensitization process is a reversible process: after agonist washout, the receptor
can be activated again. The receptor can recover from the desensitized state by means of at
least two pathways: a ligand-bound receptor can reopen, or the agonist dissociates first and
then the receptor can return to the resting state.
Mar. Drugs 2006, 4
56
Table 3: Hierarchy of ligand-gated ion channel receptor superfamilies.
Superfamily of Cys-Loop or Nicotinicoid Receptors
Family of AChRs
Family of 5-HT3Rs
Subfamily of epithelial receptors
Family of GABA receptors
Family of GlyRs
Subfamily of GABAARs
Subfamily of
GlyRs
Subfamily of GABACRs
Subfamily of neuronal
α-bungarotoxin-sensitive
receptors
Subfamily of muscle receptor
Subfamily of GABA-induced
cation channel receptors
(invertebrates)
Subfamily of heteromeric
neuronal receptors
Subfamily of heteromeric
protostomian receptors
Subfamily of ACh-induced anion
channel receptors
(invertebrates)
Superfamily of Ionotropic Glutamate Receptors
Family of excitatory GluRs
Family of inhibitory
GluRs
Subfamily of NMDAR subunits
Subfamily of Glu-induced
anion channel receptors
(invertebrates)
Subfamily of AMPAR subunits
Subfamily of KAR subunits
Subfamily of kainate binding
proteins
Subfamily of delta subunits
Superfamily of Inotropic ATP Receptors
Family of P2XnRs
Mar. Drugs 2006, 4
Figure 7. Schematic representation of the structural organization of the Cys-loop receptor,
GluR, and P2XnR superfamilies. (A) Schematic illustration of the primary sequence of several
subunits from members of the Cys-loop receptor superfamily, including α (α1-α10) an non-α
(β1-β4, γ, ε, and δ) subunits from the AChR, the 5-HT3AL subunit from the 5-HT3R, the α1
subunit from the GABAAR, and the α1 subunit from the GlyR. M1-M4, transmembrane
domains; C⎯C, Cys-Cys bridge found in the Cys-loop ion-channel superfamily (homologous
to Cys128 and Cys148 of the α1 AChR subunit); CC, Cys-Cys pair found in the α subunits from
both muscle- and neuronal-type AChRs (corresponding to Cys192-Cys193 from the α1 AChR
subunit); Υ, oligosaccharide groups. (B) Diagram of the tertiary and quaternary organization of
the LGIC receptor superfamilies. AChR subunits are formed by a long N-terminal hydrophilic
extracellular region; four highly hydrophobic domains (M1-M4). The intrinsic ion channel is
composed by five M2 segments, one from each subunit. Moreover, M1-M2 and M2-M3 are
connected by minor hydrophilic stretches; and a major hydrophilic segment (M3-M4 loop)
facing the cytoplasm. Additionally, the M4 domain orientates the C-terminus to the synaptic
57
Mar. Drugs 2006, 4
58
side of the membrane. On the contrary, GluR subunits present three transmembrane domains
(Ml, M3, and M4) and a reentrant loop (P), as well as a cytoplasmic orientation for the COOHterminus; while the P2XnR subunits have only two transmembrane domains (Ml and M2) and
both N- and C-terminus are cytoplasm oriented. (C) Schematic representation of the oligomeric
organization for LGIC receptor superfamilies. The hypothetical pentameric AChR is formed by
two α subunits and three non-α chains. The two ligand binding sites (L) are located at the
interfaces of one α subunit and one non-α chain. Regarding homomeric AChRs (e.g. α7, α8,
and α9), up to five ligand binding sites may putatively exist. On the contrary, GluRs present
four ligand binding sites, each one located in a lobe formation from each subunit, whereas
P2XnRs only have three ligand binding sites, each one located on one subunit.
Apart from the fact that all receptors from this superfamily have a 15-residue Cys-loop, they also
share some structural and functional features that make them a true superfamily of receptors. Among
the most important structural and functional features, we can name:
(1) High homology in amino acid [63,66] and pair base [78] sequence. Homology of ~30% is
observed among different subunits, whereas ~70% homology is obtained considering just one
subunit from different species. The genetic evidence also suggests that all members have evolved
from the same ancestor protein [63,66,67,78].
(2) A pentameric quaternary structure. This structural characteristic has been confirmed by
electron microscopy of tubular crystals of native membranes containing high density of AChRs
prepared from Torpedo electric organs. The enduring effort from Unwin’s laboratory has recently
rendered a receptor three-dimensional image with a resolution of 4 Å [99].
(3) A large N-terminal domain formed by ~220 residues located extracellularly, bearing the
neurotransmitter binding sites. This funnel-shaped domain has an outer diameter of 70-80 Å and
protrudes ~60 Å above the lipid membrane surface. This domain had been previously modeled
mainly as antiparallel β-sheet secondary structures arranged perpendicular to the membrane
surface [66]. This model was recently confirmed by determining the crystallographic structure of
the AChBP from L. stagnalis [20], which resembles the extracellular portion of the AChR, in
particular the α7 homomeric receptor, by 24%.
(4) An interfacial location for the neurotransmitter binding sites. The comparison with the
different AChBP structures, which also bind AChR agonists (e.g., ACh, CCh, nicotine, etc.) and
competitive antagonists [e.g., α-bungarotoxin (α-BTx), d-tubocurarine, etc.], have confirmed that
ligand sites are located between two subunits [18,20,24,25,27,48,87,106] (reviewed in [89]).
Additional comparisons with other members of the superfamily including GABAARs
[28,33,52,96], GlyRs [62], and 5-HT3Rs [83,84] also support such interfacial location.
(5) An agonist binding site comprised by several loops (A-F) (reviewed in [7]), which have been
re-named as “binding segments A-F” because some segments are actually in β-sheet rather than a
loop conformation (reviewed in [69]). Binding segments A-C form the principal component of the
agonist binding domain (which is located in the α subunit from AChRs and homologous subunits
from other receptors), whereas binding segments D-F form the secondary component (which is
located in the non-α subunit from AChRs and homologous subunits from other receptors). The
Mar. Drugs 2006, 4
59
evidence that supports this structural feature was provided by photoaffinity labeling and sitedirected mutagenesis experiments (reviewed in [7]), and more lately by cysteine- [92] and lysinescanning mutagenesis [88], as well as by comparative studies using the recently found AChBP
structures as templates [see paragraph (4)].
(6) Four transmembrane segments (M1-M4) per subunit (twenty transmembrane segments in total),
where the M2 transmembrane segment from each subunit (five M2 segments in total) forms the
wall of the ion channel. The ion channel is the main domain for the action of noncompetitive
antagonists from endogenous as well as exogenous origin (reviewed in [5,6,9,10,12-14]). The
transmembrane domain (including the phospholipid headgroup region) measure about 40 Å.
Earlier hydropathy analysis suggested that the AChR have four transmembrane segments and that
each segment is mainly in an α-helix secondary structure. This result has been confirmed by
electron microscopy aided by computational methods [99,100], photoaffinity labeling (reviewed in
[13,14]), substituted cysteine accessibility method (reviewed in [21,51], and references therein),
tryptophan-scanning mutagenesis, as well as by circular dichroism and nuclear magnetic
resonance spectroscopy (reviewed in [8,21]). However, these ion channels contain significant
amounts of non-helical structures as well. For instance, β-sheet secondary structures have been
observed in the external portion of M1 and other segments of M3.
(7) A “hydrophobic girdle”, the so-called “kink”, in the middle of the M2 segment that is formed
by residues that are conserved through all superfamily member subunits. This hydrophobic girdle
constitutes the “gate” that is occluding ion flux in the resting (closed) state. In the activated (open)
state, the rotation of one or two of the M2 segments would disrupt the hydrophobic interactions
with adjacent helices allowing the passage of ions through the channel. In this regard,
anticlockwise rotations in the extracellular domain as well as in the M2 domain were observed in
refined electron microscopy images when the AChR was activated [100].
(8) A dynamic mechanism, called “ion channel gating”, where agonist binding provokes a
conformational change in the protein that finally leads the opening of the intrinsically-associated
ion channel that is located far apart from the proper ligand binding domain. This modular
mechanism has been studied in detail in the AChR by Auerbach’s laboratory. The main
conclusion from these studies is that the chemical signal triggered by ACh binding is locally
transmitted to the extracellular loop formed between M2 and M3 segments, the M2-M3 linker
(reviewed in [21,30,31]), as a “conformational wave” that finally reaches the ion channel [45]
(reviewed in [44]).
Based on the evidence provided by low resolution electron microscopy, a concerted
anticlockwise rotation of all extracellular domains has been hypothetized as a plausible
mechanism for gating [98]. Additional evidence has been shown in other receptor members as
well. For instance, electrostatic interactions were found between a Lys located in the M2-M3
linker and acidic residues located in the Cys-loop and in the loop 2 from the GABAAR [52].
Homologous residues in the GlyR [1] and the 5-HT3R (reviewed in [72]) were also found to be
critical in the gating process.
(9) A desensitization process, where the prolonged exposure of the receptor to agonists as well as
some competitive and noncompetitive antagonists produces a conformational change in the
protein that finally leads to a refractory state where the agonist does not open the ion channel
Mar. Drugs 2006, 4
60
(reviewed in [82]). This conformational state is depicted in Fig. 6. Qualitatively, the
desensitization process is similar among the members of the Cys-loop receptor superfamily.
Desensitization might be very important in the prevention of citotoxicity, in the process of
nicotine addiction, or in other pathological states (reviewed in [82]). For instance, a Ser248 to Phe
mutation in the pore domain of the α4 subunit results in faster desensitization of α4β2 AChRs
causing a type of human cortical epilepsy [107].
(10) A small cytoplasmic domain formed by the loop between M3 and M4 of variable length
(from 80 to 265 amino acids). This region contains the greatest sequence divergence among the
members of this receptor superfamily. This region is very important for receptor modulation by
phosphorylation-deposphorylation processes (reviewed in [21,49,93]). For instance, tyrosine
kinases have been found to reduce nicotine-induced currents, whereas phosphotyrosine
phosphatases have been found to enhance nicotine-induced currents in chromaffin cells [105].
Receptor phosphorylation by serine/threonine and/or tyrosine kinases also increases the
desensitization rate in several members of this LGIC superfamily (reviewed in [49,93]).
The anchorage of certain receptors to the cytoskeleton is also mediated by the interaction of
non-receptor proteins with the cytoplasmic domain. Among them we can name rapsyn for AChRs
(reviewed in [108]) and the tubulin-binding element gephyrin for both GABAARs and GlyRs
(reviewed in [21,57,73]). Protein-protein contacts may affect receptor assembly, trafficking,
clustering, targeting, modulation, and turnover, among other processes.
Finally, the cytoplasmic domain is considered an additional determinant for ion channel
conductance (reviewed in [80]). Using different 5-HT3R chimaeras, a novel determinant of singlechannel conductance was located to a putative amphipathic helix (the ‘HA stretch’), apart from the
M2 domain. Key residues may be components of narrow openings within the inner vestibule of
the channel, which contribute to the permeation pathway.
(11) And finally, an extracellular location for the very short C-terminal. This small area is
involved, at least in certain neuronal-type AChRs (e.g., α4β2 and α4β4), in the binding of
neurosteroids [34,79] (reviewed in [5,6,9,10,13,14]). Neurosteroids are molecules synthesized in
the CNS that modulate some of the functions of this receptor superfamily. For instance, 17βestradiol may enhance the activity of α4β2 and α4β4 neuronal AChRs [34,79] (reviewed in
[5,6,9,10,13,14,49,60]). GABAAR function is also enhanced by different neurosteroids (reviewed
in [14,16]), which presents therapeutical opportunities for the treatment of several mood disorders.
Opposite to that, neurosteroids other than 17β-estradiol (e.g., progesterone, hydrocortisone, etc.)
inhibit noncompetitively several AChRs by interacting with the lipid-protein interface (reviewed
in [5,6,9,10,13,14,49]).
Since one of the topics covered in this special issue includes the action of different marine drugs on
distinct AChRs, we will give more detailed information on the structure and function of the AChR
family.
5.1. The Nicotinic Acetylcholine Receptor Family
The AChR family can be subdivided in two main receptor types (see Table 4): (a) the muscle-type
AChR that is formed by a combination of four different subunits in a stoichiometry that depends on the
Mar. Drugs 2006, 4
61
developmental stage of the muscle. The muscle in the embryonic stage has receptors (as well as the
Torpedo or Electrophorus receptor) with the stoichiometry α12β1γδ, whereas muscle cells in the adult
stage has receptors with the stoichiometry α12β1εδ. Both receptors are arranged in an anticlockwise
manner as αγ/εαδβ; and (b) neuronal-type AChRs which can be formed by the combination of several
α (α2-α10) and β (β2-β4) subunits. Functional, active homomeric AChRs are formed only by subunits
α7, α8, or α9, whereas heteromeric AChRs can be shaped by a combination of two, three, or even four
different subunits (reviewed in [7,75]).
From the functional point of view, the activation of AChRs provokes an excitatory response on the
cell because these receptors are connected to a cation-selective channel that allows the passage of
cations such as Na+, K+, or Ca2+. The activation of these ion channel receptors increases cation
permeability which subsequently produces membrane depolarization. Depolarization of the membrane
provokes specific physiological responses on each involved cell. For instance, if the muscle membrane
depolarization is large enough, an action potential is elicited. This action potential propagates from the
neuromuscular junction all over the muscle fiber. During the propagating action potential the release of
+
Ca2 ions from intracellular stores in the muscle cell (e.g., ER) is stimulated. The final response in the
muscle is the contraction of its myofibrils. On the other hand, in neurons, the excitatory signal
provided by the activation of the cation channel is summed with other signals (including inhibitory
ones provided by the activation of anionic channels) and subsequently re-directed to another neuron or
to an endocrine gland cell as a nerve action potential.
The AChR family is implicated in several important physiological functions (reviewed in [70]). For
instance, AChRs may directly or indirectly act on cognition, learning, memory and attention,
neurodegeneration, pain perception, reward, anxiety and depression control, body temperature,
auditory function development, control of voluntary motion, respiration, sleep and wakefulness,
protective responses to stress (e.g., hypoxia), cardiovascular and gastrointestinal tract, as well as on
muscle contraction and tone. From the pathophysiological point of view, AChRs are involved in
different neurological disorders including Alzheimer’s and Parkinson’s disease, schizophrenia, drug
addiction, Tourette’s syndrome, myasthenia gravis and myasthenic syndromes, depression, anxiety,
ulcerative colitis, attention-deficit hyperactivity disorder, and neuroinflammation [97], as well as on
different types of epilepsies, including nocturnal lobe, juvenile myoclonic, and catamenial epilepsy
(reviewed in [49,68,70]).
Several members from the AChR family are targets for the pharmacological action of different
natural toxins (reviewed in [7,11]). The list of alkaloids that act as agonists includes (-)nicotine
(obtained from the leaves of the tobacco plant Nicotiana tabacum), (+)anatoxin-a (obtained from the
blue-green algae Anabena flos-aquae), (-)cytisine (obtained from the plant Thermopsis alterniflora),
(±)epibatidine (obtained from the skin of the Ecuadorian frog Epipedobates tricolor), lobeline
(obtained from the plant Lobelia inflata), and anabaseine (obtained from Nemertine marine worm
species and from venom glands of certain ants). Among competitive antagonists we can name
alkaloids such as d-tubocurarine (obtained from the bush Chondrodendron tomentosum),
methyllycaconitine (obtained from the seeds of the plant Delphinium brownii), and dihydro-βerythroidine (obtained from different Erythrina plant species), snake polypeptides such as α- and κbungarotoxins [obtained from different Elapidae (e.g., the many-banded krait Bungarus multicinctus)
and Hydrophiidae (sea snakes) species] and cobrotoxins (e.g., from Naja naja siamensis), as well as
Mar. Drugs 2006, 4
62
marine toxins such as lophotoxins (obtained from gorgonian corals such as Lophogorgia and
Pseudopterogorgia), neosurugatoxin (obtained from the Japanese ivory shell Babylonia japonica), as
well as α- and αA-conotoxins (obtained from several Conus species). Finally, among noncompetitive
antagonists we can quote the alkaloid ibogaine (obtained from the shrub Tabernanthe iboga), and the
psi-conotoxin PIIIE (obtained from C. purpurascens). For more details on conotoxins see the
minireview by Layer and McIntosh in this special issue.
Although some toxins bind to different AChRs in a nonspecific manner, several toxins present very
high affinity and high specificity for certain receptor types, pointing out the importance for potential
therapeutic uses. In this regard, Kem and colleagues review the activity of benzylidene-anabaseine
derivatives as partial specific agonists of the α7 AChR as well as the noncompetitive mechanism for
these compounds in muscle-type AChRs. In addition, Wonnacott and Gallagher cover the structurefunction relationship for anatoxin-a.
Since this special issue includes marine drugs interacting neither with 5-HT3Rs, GABAARs,
GABACRs, and GlyRs (members from the Cys-loop receptor superfamily), nor with members from the
GluR or P2XnR superfamily, we will not cover these receptors any further.
6. Conclusions
Since the decade of 1900, where the concept of neurotransmission was first defined, until now, a great
deal of information on the synapse structure and function has been accumulated with the help of
different approaches from different research areas such as molecular and cell biology, biochemistry,
electrophysiology, biophysics, and genetically-altered animals. One of the most important entities for
synapse functioning, ion channels, is the focus of this special issue. Since the structural
characterization of AChRs, the archetype of the LGIC receptor superfamilies, until today, additional
LGIC as well as VGIC superfamilies have been extensively studied. In this regard, our actual
knowledge on the structure and function of different marine drugs acting on LGIC as well as on VGIC
superfamilies will pave the way to new therapies for the treatment of distinct neurological diseases.
Acknowledgements
This review was supported by Western University Health Sciences Intramural Grants (to HRA). I
thank to Dr. Robert French for critical reading of the manuscript and valuable suggestions.
References
1.
2.
3.
Absalom, N. L.; Lewis, T. M.; Kaplan, W.; Pierce, K. D.; Schofield, P. R. Role of charged
residues in coupling ligand binding and channel activation in the extracellular domain of the
glycine receptor. J. Biol. Chem. 2003, 278, 50151-50157.
Ackerman, M. J.; Clapham, D. E. Ion channels − Basic science and clinical disease. N. Engl. J.
Med. 1997, 336, 1575-1586.
Agnew, W. S.; Levinson, S. R.; Brabson, J. S.; Raftery, M. A. Purification of the tetrodotoxinbinding component associated with the voltage-sensitive sodium channel from Electrophorus
electricus electroplax membranes. Proc. Natl. Acad. Sci. USA 1978, 75, 2602-2610.
Mar. Drugs 2006, 4
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
63
Alkondon, M.; Albuquerque, E. X. The nicotinic acetylcholine receptor subtypes and their
function in the hippocampus and cerebral cortex. Prog. Brain Res. 2004, 145, 109-120.
Arias, H. R. Noncompetitive inhibition of nicotinic acetylcholine receptors by endogenous
molecules. J. Neurosci. Res. 1998, 52, 369-379.
Arias, H. R. Binding sites for exogenous and endogenous non-competitive inhibitors of the
nicotinic acetylcholine receptor. Biochim. Biophys. Acta Rev. Biomembr. 1998, 1376, 173-220.
Arias, H. R. Localization of agonist and competitive antagonist binding sites on nicotinic
acetylcholine receptors. Neurochem. Int. 2000, 36, 595-645.
Arias, H. R. In Biological and Biophysical Aspects of Ligand-Gated Ion Channel Receptor
Superfamilies; Arias, H. R., Ed.; Research Signpost: India, 2006, Chapter 1, p. 1-25.
Arias, H. R.; Bhumireddy, P. Anesthetics as chemical tools to study the structure and function of
nicotinic acetylcholine receptors. Curr. Protein Pept. Sci. 2005, 6, 451-472.
Arias, H. R.; Bhumireddy, P.; Bouzat, C. Molecular mechanisms and binding site locations for
noncompetitive antagonists of nicotinic acetylcholine receptors. Int. J. Biochem. Cell Biol. 2006,
in press.
Arias, H. R.; Blanton, M. P. α-Conotoxins. Int. J. Biochem. Cell Biol. 2000, 32, 1017-1028.
Arias, H. R.; Blanton, M. P. Molecular and physicochemical aspects of local anesthetics acting on
nicotinic acetylcholine receptor-containing membranes. Mini Rev. Med. Chem. 2002, 2, 385-410.
Arias, H. R.; Kem, W. R.; Trudell, J. R.; Blanton, M. P. Unique general anesthetic binding sites
within distinct conformational states of the nicotinic acetylcholine receptor. Int. Rev. Neurobiol.
2002, 54, 1-50.
Arias, H. R.; Machu, T. K.; Trudell, J. R.; Blanton, M. P. In Recent Research Developments in
Biophysical Chemistry, Condat, C. A. and Baruzzi, A., Eds.; Research Signpost: India, 2002, p.
123-154.
Becher, A.; Drenckhahn, A.; Pahner, I.; Margittai, M.; Jahn, R.; Ahnert-Hilger, G. The
synaptophysin-synaptobrevin complex is developmentally upregulated in cultivated neurons but is
absent in neuroendocrine cells. J. Neurosci. 1999, 19, 1922-1931.
Belelli, D.; Lambert, J. J. Neurosteroids: endogenous regulators of the GABAA receptor. Nat. Rev.
Neurosci. 2005, 6, 565-575.
Bezanilla, F. Voltage-gated ion channels. IEEE Trans. Nanobiosci. 2005, 4, 34-48.
Bourne, Y.; Talley, T. T.; Hansen, S. B.; Taylor, P.; Marchot, P. Crystal structure of a CbtxAChBP complex reveals essential interactions between snake α-neurotoxins and nicotinic
receptors. EMBO J. 2005, 24, 1512-1522.
Brandon, N.; Jovanovic, J.; Moss, S. Multiple roles of protein kinases in the modulation of γaminobutyric acidA receptor function and cell surface expression. Pharmacol. Ther. 2002, 94,
113-122.
Brejc, K.; van Dijk, W. J.; Klassen, R.V.; Schuumans, M.; van Der Oost, J.; Smit, A. B.; Sixma, T.
K. Crystal structure of an ACh-binding protein reveals the ligand-binding domain of nicotinic
receptors. Nature 2001, 403, 269-276.
Cascio, M. Structure and function of the glycine receptor and related nicotinicoid receptors. J.
Biol. Chem. 2004, 279, 19383-19386.
Mar. Drugs 2006, 4
64
22. Catterall, W. A.; Goldin, A. L.; Waxman, S. G. International Union of Pharmacology.
International Union of Pharmacology. XXXIX. Compendium of voltage-gated ion channels:
sodium channels. Pharmacol. Rev. 2003, 55, 575-578.
23. Catterall, W. A.; Striessnig, J.; Snutch, T. P.; Perez-Reyes, E. International Union of
Pharmacology. International Union of Pharmacology. XL. Compendium of voltage-gated ion
channels: calcium channels. Pharmacol. Rev. 2003, 55, 579-581.
24. Celie, P. H. N.; Klaassen, R.V.; van Rossum-Fikkert, S. E.; van Elk, R.; van Nierop, P.; Smit, A. B.;
Sixma, T. K. Crystal structure of acetylcholine-binding protein from Bulinus truncatus reveals the
conserved structural scaffold and sites of variation in nicotinic acetylcholine receptors. J. Biol.
Chem. 2005, 280, 26457-26466.
25. Celie, P. H. N.; van Rossum-Fikkert, S. E.; van Dijk, W. J.; Brejc, K.; Smit, A. B.; Sixma, T. K.
Nicotine and carbamylcholine binding to nicotinic acetylcholine receptors as studied in AChBP
crystal structures. Neuron 2004, 41, 907.
26. Changeux, J-P.; Edelstein, S. J. Allosteric receptors after 30 years. Neuron 1998, 21, 959-980.
27. Chou, K-C. Insights from modelling the 3D structure of the extracellular domain of α7 nicotinic
acetylcholine receptor. Biochem. Biophys. Res. Commun. 2004, 319, 433-438.
28. Chou, K-C. Modelling extracellular domains of GABA-A receptors: subtypes 1, 2, 3, and 5.
Biochem. Biophys. Res. Commun. 2004, 316, 636-642.
29. Clement, J. G. Toxicology and pharmacology of bispyridium oximes-insight into the mechanism of
action vs Soman poisoning in vivo. Fundam. Appl. Toxicol. 1981, 1, 193-202.
30. Colquhoun, D.; Sivilotti, L. G. Function and structure in glycine receptors and some of their
relatives. Trends Neurosci. 2003, 27, 337-344.
31. Condat, C. A.; Lamberti, P. W.; Arias, H. R. In Biological and Biophysical Aspects of LigandGated Ion Channel Receptor Superfamilies; Arias, H. R., Ed.; Research Signpost: India, 2006,
Chapter 4, p. .
32. Connolly, C. N.; Wafford, K. A. The Cys-loop superfamily of ligand-gated ion channels: the
impact of receptor structure on function. Biochem. Soc. Trans. 2004, 32, 529-534.
33. Cromer, B. A.; Morton, C. J.; Parker, M. W. Anxiety over GABAA receptor structure relieved by
AChBP. Trends Biochem. Sci. 2002, 27, 280-287.
34. Curtis, L.; Buisson, B.; Bertrand, S.; Bertrand, D. Potentiation of human α4β2 neuronal nicotinic
acetylcholine receptor by estradiol. Mol. Pharmacol. 2002, 61, 127-135.
35. Dahan, M.; Levi, S.; Luccardini, C.; Rostaing, P.; Riveau, B.; Triller, A. Diffusion dynamics of
glycine receptors revealed by single-quantum dot tracking. Science 2003, 302, 442-445.
36. Denac, H.; Mevissen, M.; Scholtysik, G. Structure, function and pharmacology of voltage-gated
sodium channels. Naunyn-Schmiedeberg’s Arch. Pharmacol. 2000, 362, 453-479.
37. Escoubas, P.; Diochot, S.; Corzo, G. Structure and pharmacology of spider venos neurotoxins.
Biochimie 2000, 82, 893-907.
38. Farber, L.; Haus, U.; Spath, M.; Drechsler, S. Physiology and pathophysiology of the 5-HT3
receptor. Scand. J. Rheumatol. Suppl. 2004, 119, 2-8.
39. Farrant, M.; Nusser, Z. Variations on an inhibitory theme: phasic and tonic activation of GABAA
receptors. Nat. Rev. Neurosci. 2005, 6, 215-229.
Mar. Drugs 2006, 4
65
40. Fatt, P.; Katz, B. The electrical properties of crustacean muscle fibers. J. Physiol. 1953, 120, 171204.
41. George Jr., A. L. Inherited disorders of voltage-gated sodium channels. J. Clin. Invest. 2005, 115,
1990-1999.
42. Ghijsen, W. E.; Leenders, A. G. M. Differential signaling in presynaptic neurotransmitter release.
Cell Mol. Life Sci. 2005, 62, 937-954.
43. Gotti, C.; Clementi, F. Neuronal nicotinic receptors: from structure to pathology. Progr. Neurobiol.
2004, 74, 363-396.
44. Grosman, C. Linear free-energy relationships and the dynamics of gating in the acetylcholine
receptor channel. A Φ-value analysis of an allosteric transition at the single-molecule level. J.
Biol. Phys. 2002, 28, 267-277.
45. Grosman, C.; Zhou, M.; Auerbach, A. Mapping the conformational wave of acetylcholine receptor
channel gating. Nature 2000, 403, 773-776.
46. Gutman, G. A.; Chandy, K. G.; Adelman, J. P.; Aiyar, J.; Bayliss, D. A.; Clapham, D. E.;
Covarriubias, M.; Desir, G. V.; Furuichi, K.; Ganetzky, B.; Garcia, M. L.; Grissmer, S.; Jan, L. Y.;
Karschin, A.; Kim, D.; Kuperschmidt, S.; Kurachi, Y.; Lazdunski, M.; Lesage, F.; Lester, H. A.;
McKinnon, D.; Nichols, C. G.; O'Kelly, I.; Robbins, J.; Robertson, G. A.; Rudy, B.; Sanguinetti,
M.; Seino, S.; Stuehmer, W.; Tamkun, M. M.; Vandenberg, C. A.; Wei, A.; Wulff, H.; Wymore, R.
S. International Union of Pharmacology. International Union of Pharmacology. XLI.
Compendium of voltage-gated ion channels: potassium channels. Pharmacol. Rev. 2003, 55, 583586.
47. Hansen, S. B.; Talley, T. T.; Radić, Z.; Taylor P. Structural and ligand recognition characteristics
of an acetylcholine-binding protein from Aplysia californica. J. Biol. Chem. 2004, 279, 2419724202.
48. Harel, M.; Kasher, R.; Nicolas, A.; Guss, J. M.; Balass, M.; Fridkin, M.; Smit, A.B.; Brejc, K.;
Sixma, T. K.; Katchalski-Katzir, E.; Sussman, J. L.; Fuchs, S. The binding site of acetylcholine
receptor as visualized in the X-Ray structure of a complex between α-bungarotoxin and a
mimotope peptide. Neuron 2001, 32, 265.
49. Hogg, R. C.; Raggenbass, M.; Bertrand, D. Nicotinic acetylcholine receptors: from structure to
brain function. Rev. Physiol. Biochem. Pharmacol. 2003, 147, 1-46.
50. Huh, K. H.; Fuhrer, C. Clustering of nicotinic acetylcholine receptors: from the neuromuscular
junction to interneuronal synapses. Mol. Neurobiol. 2002, 25, 79-112.
51. Karlin, A. Emerging structure of the nicotinic acetylcholine receptors. Nat. Rev. Neurosci. 2002, 3,
102-114.
52. Kash, T. L.; Trudell, J. R.; Harrison, N. L. Structural elements involved in activation of the
gamma-aminobutyric acid type A (GABAA) receptor. Biochem. Soc. Trans. 2004, 32, 540-546.
53. Kauferstein, S.; Huys, I.; Lamthanh, H.; Stöcklin, R.; Sotto, F.; Menez, A.; Tytgat, J.; Mebs, D. A
novel conotoxin inhibiting vertebrate voltage-sensitive potassium channels. Toxicon 2003, 42, 4352.
54. Kažić, T.; Gojković-Bukarica, L. Ion channels and drug development. Focus on potassium
channels and their modulators. Facta Universitatis 1999, 6, 23-30.
Mar. Drugs 2006, 4
66
55. Keramidas, A.; Moorhouse, A. J.; Schofield, P. R.; Barry, P. H. Ligand-gated ion channels:
mechanisms underlying ion selectivity. Progr. Biophys. Mol. Biol. 2004, 86, 161-204.
56. Kittler, J. T.; Moss, S. J. Modulation of GABAA receptor activity by phosphorylation and receptor
trafficking: implications for the efficacy of synaptic inhibition. Curr. Opin. Neurobiol. 2003, 13,
341-347.
57. Kneussel, M., Betz, H. Receptors, gephyrin and gephyrin-associated proteins: novel insights into
the assembly of inhibitory postsynaptic membrane specializations. J. Physiol. 2000, 525, 1-9.
58. Koles, L.; Wirkner, K.; Illes, P. Modulation of ionotropic glutamate receptor channels. Neurochem.
Res. 2001, 26, 925-932.
59. Kullmann, D. M.; Ruiz, A.; Rusakov, D. M.; Scott, R.; Semyanov, A.; Walker, M. C. Presynaptic,
extrasynaptic and axonal GABAA receptors in the CNS: where and why? Prog. Biophys. Mol.
Biol. 2005, 87, 33-46.
60. Lambert, J. J.; Belelli, D.; Harney, S. C.; Peters, J. A.; Frenguelli, B. G. Modulation of native and
recombinant GABAA receptors by endogenous and synthetic neuroactive steroids. Brain Res.
Brain Res. Rev. 2001, 37, 68-80.
61. Langley, J. N. On the reaction of cells and of nerve endings to certain poisons, chiefly as regards
the reaction of striated muscle to nicotine and to curare. J. Physiol. (London) 1905, 33, 374-413.
62. Laube, B.; Maksay, G.; Schemm, R.; Betz, H. Modulation of glycine receptor function: a novel
approach for therapeutic intervention at inhibitory synapses? Trends Pharmacol. Sci. 2002, 23,
519-527.
63. Le Novére, N.; Changeux, J-P. Molecular evolution of the nicotinic acetylcholine receptor: an
example of multigene family in excitable cells. J. Mol. Evol. 1995, 40, 155-172.
64. Le Novére, N.; Changeux, J-P. LGICdb: the ligand-gated ion channel database. Nucleic Acids Res.
2001, 29, 294-295.
65. Le Novére, N.; Changeux, J-P. The ligand gated ion channel database: an example of a sequence
database in neuroscience. Philos. Trans. R. Soc. Lond. B Biol. Sci. 2001, 356, 1121-1130.
66. Le Novére, N.; Corringer, P. J.; Changeux, J-P. Improved secondary structure predictions for a
nicotinic receptor subunit: incorporation of solvent accessibility and experimental data into a twodimensional representation. Biophys. J. 1999, 76, 2329-2345.
67. Le Novére, N.; Corringer, P. J.; Changeux, J-P. The diversity of subunit composition in nAChRs:
evolutionary origins, physiologic and pharmacologic consequences. J. Neurobiol. 2002, 53, 447456.
68. Leonard, S.; Bertrand, D. Neuronal nicotinic receptors: from structure to function. Nicotine
Tobacco Res. 2001, 3, 203-223.
69. Lester, H. A.; Dibas, M. I.; Dahan, D. S.; Leite, J. F.; Dougherty, D. A. Cys-loop receptors: new
twists and turns. Trends Neurosci. 2004, 27, 329-336.
70. Lloyd, G. K.; Williams, M. Neuronal nicotinic acetylcholine receptors as novel drug targets. J.
Pharmacol. Exp. Ther. 2000, 292, 461-467.
71. Long, S. B.; Campbell, E. B.; Mackinnon, R. Crystal structure of a mammalian voltage-dependent
Shaker family K+ channel. Science 2005, 309, 897-903.
72. Lummis, S. C. R. The transmembrane domain of the 5-HT3 receptor: its role in selectivity and
gating. Biochem. Soc. Trans. 2004, 32, 535-539.
Mar. Drugs 2006, 4
67
73. Lynch, J. W. Molecular structure and function of the glycine receptor chloride channel. Physiol.
Rev. 2004, 84, 1051-1095.
74. MacDermott, A. B.; Role, L. W.; Siegelbaum, S. A. Presynaptic ionotropic receptors and the
control of transmitter release. Annu. Rev. Neurosci. 1999, 22, 443-485.
75. Millar, N. S. Assembly and subunit diversity of nicotinic acetylcholine receptors. Biochem. Soc.
Trans. 2003, 31, 869-874.
76. Miller, C. An overview of the potassium channel family. Genome Biol. 2000, 4, 0004.1-0004.5.
77. Miller, R. J. Presynaptic receptors. Annu. Rev. Pharmacol. Toxicol. 1998, 38, 201-227.
78. Ortells, M. O.; Lunt, G. G. Evolutionary history of the ligand-gated ion-channel superfamily of
receptors. Trends Neurosci. 1995, 18, 121-127.
79. Paradiso, K.; Zhang, J.; Steinbach, J. H. The C terminus of the human nicotinic α4β2 receptor
forms a binding site required for potentiation by an estrogenic steroid. J. Neurosci. 2001, 21,
6561-6568.
80. Peters, J. A.; Kelley, S. P.; Dunlop, J. I.; Kirkness, E. F.; Hales, T. G.; Lambert, J. J. The 5hydroxytryptamine type 3 (5-HT3) receptor reveals a novel determinant of single-channel
conductance. Biochem. Soc. Trans. 2004, 32, 547-552.
81. Pichon, Y.; Prime, L.; Benquet, P.; Tiaho, F. Some aspects of the physiological role of ion
channels in the nervous system. Eur. Biophys. J. 2004, 33, 211-226.
82. Quick, M. W.; Lester, R. A. J. Desensitization of neuronal nicotinic receptors. J. Neurobiol. 2002,
53, 457-478.
83. Reeves, D. C.; Lummis, S. C. R. The molecular basis of the structure and function of the 5-HT3
receptor: a model ligand-gated ion channel. Mol. Membr. Biol. 2002, 19, 11-26.
84. Reeves, D. C.; Sayed, M. F. R.; Chau, P. L.; Price, K. L.; Lummis, S. C. R. Prediction of 5-HT3
receptor agonist-binding residues using homology modeling. Biophys. J. 2003, 84, 2338-2344.
85. Reid, C. A.; Bekkers, J. M.; Clements, J. D. Presynaptic Ca2+ channels: a functional patchwork
Trends Neurosci. 2003, 26, 683-687.
86. Roux, B. Ion conduction and selectivity in K+ channels. Annu. Rev. Biophys. Biomol. Struct. 2005,
34, 153-171.
87. Schapira, M.; Abagyan, R.; Totrov, M. Structural model of nicotinic acetylcholine receptor
isotypes bound to acetylcholine and nicotine. BMC Struct. Biol. 2002, 2, 1-8.
88. Sine, S. M.; Wang, H. L.; Bren, N. Lysine scanning mutagenesis delineates structural model of the
nicotinic receptor ligand binding domain. J Biol Chem. 2002, 277, 29210-29223.
89. Sixma, T. K.; Smit, A.B. Acetylcholine binding protein (AChBP): a secreted glial protein that
provides a high-resolution model for the extracellular domain of pentameric ligand-gated ion
channels. Annu. Rev. Biophys. Biomol. Struct. 2003, 32, 311-334.
90. Smit, A. B.; Syed, N. I.; Schaap, D.; van Minnen, J.; Klumperman, J.; Kits, K. S.; Lodder, H.; van
der Schors, R. C.; van Elk, R.; Sorgedrager, B.; Brejc, K.; Sixma, T. K.; Geraerts, W. P. A. gliaderived acetylcholine-binding protein that modulates synaptic transmission. Nature 2001, 411,
261-268.
91. Stevens, C. F. Presynaptic function. Curr. Opin. Neurobiol. 2004, 14, 341-345.
Mar. Drugs 2006, 4
68
92. Sullivan, D.; Chiara, D. C.; Cohen, J. B. Mapping the agonist binding site of the nicotinic
acetylcholine receptor by cysteine scanning mutagenesis: antagonist footprint and secondary
structure prediction. Mol. Pharmacol. 2002, 61, 463-472.
93. Swope, S. L.; Moss, S. J.; Raymond, L. A.; Huganir, R. L. Regulation of ligand-gated ion channels
by protein phosphorylation. Adv. Second Messenger Phosphoprot. Res. 1999, 33, 49-78.
94. Tempel, T. M.; Papazian, D. M.; Schwarz, T. L.; Jan, Y. N.; Jan, L.Y. Sequence of a probable
potassium channel component encoded at Shaker locus in Drosophila. Science 1987, 237, 770775.
95. Trimmer, J. S.; Rhodes, K. J. Localization of voltage-gated ion channels in mammalian brain. Annu.
Rev. Physiol. 2004, 66, 477-519.
96. Trudell, J. Unique assignment of inter-subunit association in GABAA α1β3γ2 receptors determined
by molecular modeling. Biochim. Biophys. Acta 2002, 1565, 91-96.
97. Tuppo, E. E.; Arias, H. R. The role of inflammation in Alzheimer's disease. Int. J. Biochem. Cell
Biol, 2005, 37, 289-305.
98. Unwin, N. The Croonian Lecture 2000. Nicotinic acetylcholine receptor and the structural basis of
fast synaptic transmission. Philos. Trans. R. Soc. London Ser. B 2000, 355, 1813-1829.
99. Unwin, N. Refined structure of the nicotinic acetylcholine receptor at 4 Å resolution. J. Mol. Biol.
2005, 346, 967-989.
100. Unwin, N.; Miyazawa, A.; Li, J.; Fujiyoshi, Y. Activation of the nicotinic acetylcholine receptor
involves a switch in conformation of the α subunits. J. Mol. Biol. 2002, 319, 1165-1176.
101. Van der Kloot, W.; Molgó, J. Quantal acetylcholine release at the vertebrate neuromuscular
junction. Physiol. Rev. 1994, 74, 899-991.
102. Vazquez, G.; Wedel, B. J.; Aziz, O.; Trebak, M.; Putney, Jr., J. W. The mammalian TRPC cation
channels. Biochim. Biohphys. Acta 2004, 1742, 21-36.
103. Vial, C.; Roberts, J.A.; Evans, R.J. Molecular properties of ATP-gated P2X receptor ion channels.
Trends Pharmacol. Sci. 2004, 25, 487-493.
104. Waites, C. L.; Craig, A. M.; Garner, C. C. Mechanisms of vertebrate synaptogenesis. Annu. Rev.
Neurosci. 2005, 28, 251-274.
105. Wang, K.; Hackett, J. T.; Cox, M. E.; van Hoek, M.; Lindstrom, J. M.; Parsons, S. J. Regulation
of the neuronal nicotinic acetylcholine receptor by SRC family tyrosine kinases. J. Biol. Chem.
2004, 279, 8779-8786.
106. Wei, D.; Sirois, S.; Du, Q-S.; Arias, H.R.; Chou, K-C. Theoretical studies of Alzheimer’s disease
drug candidate 3-[(2,4-dimethoxy)benzylidene]-anabaseine (GTS-21) and its derivatives. Biochem.
Biophys. Res. Commun. 2005, 338, 1059-1064.
107. Weiland, S., Witzemann, V., Villarroel, A., Propping, P., Steinlein, O. An amino acid exchange in
the second transmembrane segment of a neuronal nicotinic receptor causes partial epilepsy by
altering its desensitization kinetics. FEBS Lett. 1996, 398, 91-96.
108. Willmann, R.; Fuhrer, C. Neuromuscular synaptogenesis: clustering of acetylcholine receptors
revisited. Cell Mol. Life Sci. 2002, 59, 1296-1316.
109. Yamakage, M.; Namiki, A. Calcium channels - basic aspects of their structure, function and gene
encoding; anesthetic action on the channels - a review. Can. J. Anesth. 2002, 49, 151-164.
Mar. Drugs 2006, 4
69
110. Yu, F. H., Catterall, W. A. Overview of the voltage-gated sodium channel family. Genome Biol.
2003, 4, 207.
111. Ziv, N. E.; Garner, C. C. Cellular and molecular mechanisms of presynaptic assembly. Nat. Rev.
Neurosci. 2004, 5, 385-399.
Samples Availability: Available from the authors.
© 2006 by MDPI (http://www.mdpi.org). Reproduction is permitted for noncommercial purposes.