Doi:10.1016/j.nurt.2007.01.013

Neurotherapeutics: The Journal of the American Society for Experimental NeuroTherapeutics Diana Conte Camerino, Domenico Tricarico, and Jean-François Desaphy Pharmacology Division, Department of Pharmacobiology, School of Pharmacy, University of Bari, Bari, Italy Summary: Because ion channels are involved in many cellular
tions have demonstrated that channel mutations can either in- processes, drugs acting on ion channels have long been used for crease or decrease affinity for the drug, modifying its potential the treatment of many diseases, especially those affecting elec- therapeutic effect. Together with the discovery of channel gene trically excitable tissues. The present review discusses the phar- polymorphisms that may affect drug pharmacodynamics, these macology of voltage-gated and neurotransmitter-gated ion findings highlight the need for pharmacogenetic research to channels involved in neurologic diseases, with emphasis on allow identification of drugs with more specific effects on neurologic channelopathies. With the discovery of ion chan- channel isoforms or mutants, to increase efficacy and reduce nelopathies, the therapeutic value of many basic drugs targeting side effects. With a greater understanding of channel genetics, ion channels has been confirmed. The understanding of the structure, and function, together with the identification of novel genotype–phenotype relationship has highlighted possible ac- primary and secondary channelopathies, the number of ion tion mechanisms of other empirically used drugs. Moreover, channel drugs for neurologic channelopathies will increase sub- other ion channels have been pinpointed as potential new stantially. Key Words: Voltage-gated, neurotransmitter-gated,
drug targets. With regards to therapy of channelopathies, ex-perimental investigations of the intimate drug– channel interac- ion channel, drug therapy, channelopathy, pharmacogenetics.
INTRODUCTION
Beyond their usefulness in the clinical setting,natural ion channel ligands, especially toxins with high Ion channels are involved in many, if not all, cellular binding affinity, have also largely contributed to the dis- functions and are altered in many pathological conditions covery of the various ion channels and the understanding either indirectly or directly, as in the channelopathies. Itis not surprising, therefore, that drugs targeting ion chan- of their structure and function long before their molecu- nels constitute important therapeutic interventions for a number of diseases. The use of ion channel modulators Historically, the role of ion channels was most obvious as drugs was operative long before their existence be- in the membrane of electrically excitable cells, such as came known. Ion channel function is modulated by many the neuron, the cardiac myocyte, and the skeletal muscle natural agents of the animal and plant kingdoms, which fiber. Consequently, a number of drugs able to modulate contribute to the dangerous effects of poisons or the cell excitability by acting on voltage-gated or neurotrans- beneficial effects of medicinal herbs. Once isolated, mitter-gated ion channels in these tissues have reached these lead compounds have served as the basis for the blockbuster status in the pharmaceutical industry, gener- synthesis of more specific ligands with fewer side ef- ating large profits. Examples are the antiepileptic drugs fects. For instance, cocaine extracted from coca leaves (AEDs), which include blockers of voltage-gated sodium entered clinical practice in the 1880s for its analgesic properties, but the occurrence of CNS and cardiovascular and, more recently, openers of potassium channels and toxicity led medicinal chemists to synthesize new deriv- antagonists of AMPA and NMDA glutamate receptors.
atives, thus giving rise to the pharmaceutical class of Today more than 400 genes are known that encode local anesthetics, which are selective blockers of sodium even more ion channel subunits due to alternative splic-ing, each subunit being likely the target of many phar-macological agents. Covering all drugs acting on ion Address correspondence and reprint requests to: Diana Conte Cam- channels is beyond the scope of this review, which will erino, Ph.D., Sezione di Farmacologia, Dipartimento FarmacoBio- instead focus mainly on drugs acting on ion channels logico, Facoltà di Farmacia, Università degli Studi di Bari, via Orabona4 – CAMPUS, I-70125, Bari, Italy. E-mail: conte@farmbiol.uniba.it.
involved in neurologic disorders and especially their use Vol. 4, 184 –198, April 2007 The American Society for Experimental NeuroTherapeutics, Inc.
in channelopathies. The sections that follow each detail channels participate to the repolarization of the postsyn- the pharmacology of an ion channel family. In addition, aptic action Kv3.1Ϫ/Ϫ mice show impaired a synopsis of drug information for the neurologic chan- motor skills and reduced muscle contraction force. Dou- ble Kv3.1/Kv3.3 knockout mice show ataxia, myoclo-nus, and other neurological abnormalities. Kv channelsare also involved in neurological symptoms observed in PHARMACOLOGY OF POTASSIUM
paraneoplastic neurological syndromes, which are re- CHANNELS
mote effects of cancer with an autoimmune responseagainst CNS and peripheral nervous In neuro- Kϩ channels are classified on the basis of the primary myotonia associated with limbic encephalitis and small amino acid sequence of the pore-containing unit (␣-sub- cell lung cancer cells (SCLC), function of Kv1.1/Kv1.2 unit) into three major families: 1) voltage-gated Kϩ channels is progressively lost because of an abnormally channels (Kv) containing six or seven transmembrane enhanced turnover and degradation of the proteins. The regions with a single pore, including also KCNQ, hERG, immune system is also modulated by the Kv1.3 channel, eag, and the Ca2ϩ-activated Kϩ channels; 2) inward which is expressed in many cells involved in immune rectifiers (Kir) containing only two transmembrane re- gions and a single pore; and 3) two-pore tandem Kϩ Kv channel blockers. The voltage-gated Kϩ channels
channels containing four transmembrane segments with have been investigated through the use of peptide toxins two pores. The pore subunits coassemble with auxiliary from animals and plants, such as dendrototoxins, kali- subunits, affecting their pharmacological responses and toxin, hongotoxin, margatoxin, and others that block the channel pore at picomolar to nanomolar concentrations Pharmacology of voltage-gated potassium
and serve as tools for the analysis of their structure– (Kv) channels
function relationships. These toxins block Kv1.1– 6 Following the cloning of the four Kϩ channel genes in channel Although Kv channels were the first Drosophila, several members of related voltage-gated Kϩ to be molecularly characterized, no selective blockers or channel (Kv) genes were identified in mammals and di- openers are currently available. Tetraethylammonium vided into eight gene families: KCNA (Kv1.1– 8), KCNB (TEA) and 4-aminopyridine (4-AP) are classic Kv chan- (Kv2.1–2), KCNC (Kv3.1– 4), KCND (Kv4.1–3), KCNF nel blockers, which can discriminate between various (Kv5.1), KCNG (Kv6.1– 4), KCNS (Kv9.1–3) and channel subtypes. The Kv1.1, Kv3.1– 4, and Kv7.2 chan- KCNV (Kv8.1–3). The Kv1– 4 families can form homo- nels are more sensitive to TEA. Kv1.1–5, Kv1.7 and or heteromeric channels with other subunits from within Kv3.1–2 are inhibited by micromolar concentrations of their own family or with the electrically silent families 4-AP, but millimolar concentrations are needed to block (Kv5, Kv6, Kv8, and Kv9). The ␤-subunits of Kv chan- nels influence channel properties, trafficking, and drug Other members, such as Kv2.1, Kv3.4, hERG, eag1, and KCNQ channels, are insensitive to 4-AP. Several Kv1.1–2 channels are involved in neuronal chan- 4-AP analogs have been tested against Kv channels, and nelopathies. Kv1.1 is expressed in many neurons, motor the order of potency as Kv inhibitors ranks as follows neurons, retina, and heart and skeletal muscle, whereas 3,4-DAP Ͼ 4-AP Ͼ 3-AP Ͼ These drugs cause Kv1.2 is expressed mainly in the cerebellum, hippocam- neuronal firing and release of neurotransmitters such as pus, and thalamus. These low-voltage activated channels, acetylcholine (ACh). Thus, 4-AP and 3,4-diaminopyri- located in the axons of neurons, do not affect the first dine (3,4-DAP) (EU/3/02/124) (25– 60 mg/day) are ef- action potential but increase the action potential thresh- fective in those conditions associated with loss or re- Kv1.1 and Kv1.2 can form heteromultimers impor- duced quantal release of neurotransmitters such as tant for repolarization of the presynapsis in neurons and episodic ataxias, myasthenia gravis (MG), Lambert– Eaton myasthenic syndromes (LEMS), and degenerative Loss-of-function mutations of KCNA1 are associated with episodic ataxia type 1 (EA1), which is characterized In episodic ataxias type 2 and 6 (EA2 and EA6) the by episodic failure of cerebellar excitation, while hyper- drugs enhance the excitability of spinocerebellar axis excitability of motor neurons is commonly observed. An that is compromised by gain-of-function mutations of the imbalance between inhibitory and excitatory input in the Ca(v)2.1 calcium channel ␣-subunit. They reduce attack cerebellum destabilizes motor controls during exercise frequency but not duration, suggesting different mecha- nisms for triggering and maintaining the These Other neuronal Kv channels are the Kv3.1–2 control- drugs may be also effective in migraine, which is often ling the release of GABA. These high-voltage activated Neurotherapeutics, Vol. 4, No. 2, 2007 TABLE 1. Drugs acting on ion channels used in the neurologic channelopathies
Clinically used drugs for neurological symptoms Symptomatic treatment independent on genetic origin Use-dependent block of muscle sodium channels ClC-1 channel activation through CA inhibition, others Direct activation of muscle BK channels; CA inhibition ACTZ may worsen paralysis in some patients Kϩ-sparing diuretics in patients worsened by ACTZ ␤ -adrenoceptor agonists: Na/K pump activation Inhibition of acetylcholinesterase enzyme Block of presynaptic Kv channels leading to ACh release Possible activation of the presynaptic BK ? Prevention of convulsions by Naϩ channel blockade Symptomatic treatment independent on genetic origin Positive allosteric GABA R modulator, other targets Na channel blockade, NMDA receptor antagonist Na channel blockade, KA/AMPA receptor antagonist Symptomatic treatment independent on genetic origin Blockade of T-type Ca2ϩ channels, other targets HVA calcium channel blockade, other targets reduced neuronal firing through CA inhibition Hyperekplexia
Familial erythermialgia
Use-dependent block of nerve sodium channels Paraneoplastic channelopathies
Immunosuppressive therapy is effective.
Inhibition of acetylcholinesterase enzyme Abbreviations: LA: local anesthetics; nAChR: nicotinic acetylcholine receptor; GABA R: ␥-aminobutyric acid receptor type A; GlyR: glycine receptor; HVA: high voltage activated; LEMS: Lambert-Eaton myasthenic syndrome; SMEI: severe myoclonic epilepsy of infancy; BFNC: Begnin familial neonatal convulsions (BFNC); GEFSϩ: Generalized epilepsy with febrile seizures;CA: carbonic anhydrase.
The treatment of MG is generally based on the use of Pharmacology of KCNQ, hERG, and eag1 channels
anticholinesterases, such as pyridostigmine, which work Other genes were found to encode different voltage- also in Quinidine and 3,4-DAP are also effective dependent Kϩ channels, which are KCNQ1–5 (Kv7.1–5) in MG and LEMS, because they improve the release of ACh that is disrupted by the autoantibodies directed against motor nerve terminals. Inhibition of Kv1.3 by Kv KCNQ2–5 channels are expressed in neurons, being re- blockers may also explain their efficacy in LEMS and sponsible for the M-currents inhibited by ACh through MG. Indeed, immunosuppressive therapy is successful in the muscarinic receptors involved in the ACh-dependent the treatment of the neuronal paraneoplastic channelopa- postsynaptic depolarization. Combinations of KCNQ3 thies, but not in other paraneoplastic neurological with KCNQ2, KCNQ4, or KCNQ5 give rise to various heteromeric channels that underlie these currents. Muta- The Kv blockers are not effective in neuromyotonia, tions in KCNQ2 or KCNQ3 result in a loss of function however, which is characterized by peripheral nerve hy- and an increase in cellular excitability, leading to benign perexcitability. Neuromyotonia responds to the AEDs familial neonatal convulsion. Mutations in KCNQ4 cause carbamazepine, phenytoin, and pregabalin. Other nonse- lective Kv channel blockers include linopirdine, a KCNQ Because of their important physiological functions, channel blocker that evokes quantal ACh release in the KCNQ channels are potential drug targets. KCNQ1 and CNS; the antiarrhythmic drugs flecainide and bupiva- hERG are not involved in neuronal channelopathies, but caine (100 –250 ␮mol/L); verapamil, nifedipine, nicardi- have a physiopathological role in the Indeed,drugs blocking these channels can precipitate long QT pine, diltiazem (20 –200 ␮mol/L), all of them well- intervals, with potentially fatal effects. Eag1 is expressed known vasodilating and antiarrhythmic drugs; and in several brain areas and in skeletal muscle. KCNH1 is riluzole (100 –200 ␮mol/L), a neuroprotective agent used involved in the cell cycle and in cell proliferation. An in treating amyotrophic lateral sclerosis.
abnormal expression is found in 70% of human cervical The blocking of Kv channels by drugs, as observed in and breast carcinomas, so it is an oncogenic marker. No anorexigens and dopamine agonists, may cause vasocon- striction with an increased risk of pulmonary hyperten- KCNQ channel blockers. Interest in KCNQ channels
sion and valvular heart disease. A reduced expression of was heightened by the observation that compounds de- Kv1.5 and Kv2.1 is associated with chronic pulmonary veloped as cognition enhancers, such as linopirdine and XE-991, are blockers of KCNQ channels. Blocking of Kv channel openers. No drugs are available that open
M-currents underlies the enhancement of transmitter re- either wild-type Kv1.1 or the Kv1.1 mutants responsible lease by these Linopirdine increases ACh release for EA1. The current therapy of EA1 is based on the use in rat brain and improves performance in animal models of acetazolamide (ACTZ) (trade name: Diamox; 250 – of learning and memory. Although clinical data with 500 mg/day), a carbonic anhydrase inhibitor that relieves linopirdine were inconclusive, several analogs such as the kinesigenic attacks and corrects dystonia in humans.
XE-991 and DMP-543 were developed as orally active It is possible that ACTZ is effective in EA1 indirectly by ACh-releasing agents with potential in Alzheimer’s dis- ameliorating the neurotransmission in the cerebellum ease. Heteromers derived from the cardiac KCNQ1/ and spinal cord. ACTZ does not interact with Kv chan- minK are 14- to 18-fold less sensitive to XE-991 block- nels in heterologous expression systems, suggesting that ade compared with either KCNQ1 alone or neuronal its effect in EA1 is mediated by other factors. One pos- KCNQ2/3 combination, demonstrating selectivity of sibility includes a direct interaction of ACTZ with neu- these compounds for neurotransmitter release over car- ronal BK channels that modulate neurotransmitter re- diac function. The compounds are 10 to 20 times more lease in the CNS, thereby compensating for the lack of potent in releasing ACh from the hippocampus, with presynaptic Kv in EA1. ACTZ can indeed open skeletal improved half-life and brain–plasma distribution com- muscle BK It is noteworthy that the lack of pared with linopirdine. Compounds inhibiting M-cur- BK channels produces cerebellar ataxia along with Pur- rents selectively are potentially useful for treating cog- nitive deficits in neurodegenerative diseases and are Promising effects in EA1 can also be observed with likely to be forthcoming. These compounds are also KCNQ openers, such as retigabine (currently used as an promising drugs in the treatment of EA2, EA6, and anticonvulsant) and flupirtine (used as an analgesic), both of which are also effective in paroxysmal dystonia KCNQ channel openers. These are the newest anti-
mutant Convulsions in EA1 are controlled by epileptic agents. The loss-of-function mutations in carbamazepine and phenytoin (see discussion of sodium KCNQ channels associated with benign familial neonatal convulsions support the use of this class of drugs as Neurotherapeutics, Vol. 4, No. 2, 2007 antiepileptic Retigabine, the desaza-analog of Pharmacology of calcium-activated potassium
flupirtine (approved in Europe for general nociceptive ) channels
pain), was originally identified as an anticonvulsant.
Three subfamilies of Ca2ϩ-activated Kϩ channels can Retigabine is effective in various epilepsy models and be distinguished. The first is the large conductance chan- was shown to activate M-currents in various types of nels K 1.1 encoded by KCMNA1 (slo1) gene (BK) and cultured neurons, suggesting KCNQ2/3 opening as a new the K 4.1–2 and K 5.1 channels, which are, however, mode of action for anticonvulsant drugs. The M-current less Ca2ϩ sensitive but activated by Naϩ and OHϪ is a slowly activating current whose threshold is near the BK channels are widely expressed and are in- resting potential. Retigabine shifts the activation of volved in hypertension, coronary artery spasm, urinary KCNQ M-current to more hyperpolarized membrane po- incontinence, stroke, psychoses, and several neurological tentials, and also slows deactivation and accelerates ac- disorders including epilepsy and schizophrenia. The sec- tivation, reducing neuronal firing. Retigabine acts on all ond subfamily is the intermediate conductance K 3.1 neuronal KCNQ subunits, but not on the cardiac encoded by the KCNN4 gene (IK), expressed in thymo- KCNQ1. A single 236-tryptophan residue within the S5 cytes, where it plays a role in immunostimulation and segment of KCNQ2 is critical for the effects of retigab- erythrocytes. This channel is also expressed in several ine on gating and binding, and it is believed to be a part cancer cell lines being involved in cell proliferation. The of a hydrophobic pocket when the channel opens. Reti- third of the subfamilies of calcium-activated potassium gabine enhances GABA-activated ClϪ current at concen- channels are the small conductance channels K 2.1–3 trations higher than those affecting KCNQ channels. The encoded by the KCNN1–3 genes (alias hSK1–3), found ina variety of cells including sympathetic neurons, intesti- effect on GABA receptors occurs independently of the nal smooth muscle, bladder smooth muscle, hepatocytes, benzodiazepine site. It is quite possible that this action and brown adipocytes. In excitable cells, the SK chan- may contribute, along with the effect on KCNQ chan- nels are responsible for the slow after-hyperpolarization nels, to the anticonvulsant activity of retigabine. It seems that follows action potential. Calmodulin is associated also likely that the effects on GABA may contribute to with the SK ␣-subunit, and is necessary for Ca2ϩ binding the dose-limiting CNS side effects that have occurred in clinical trials, including somnolence, dizziness, ataxia, BK channel openers. These drugs stabilize the cell by
confusion, speech disorder, vertigo, tremor, amnesia and increasing efflux of Kϩ ions when the intracellular Ca2ϩ ions rise, leading to hyperpolarization and thus decrease Analogs of retigabine with more favorable pharmaco- logical profile are under investigation, including the appealing as a therapeutic target than ATP-sensitive Kϩ benzanilide derivative ICA-27243 which is a more se- lective KCNQ opener and does not affect GABA channels in the heart. Furthermore, extensive Kϩ efflux and a late channel opening (at approximately 0 mV) Phase III clinical trial in patients with partial seizures circumvent the adverse cardiac effects (hypotension, re- with or without secondary generalization who were re- flex tachycardia, and coronary artery steal) associated fractory to available therapies. Indications of efficacy were obtained in open-label trials with 35% to 44% of The different subunit compositions of BK channels in responders (Ն50% reduction in seizure frequency). In a various tissues open the possibility of finding tissue- multicenter double-blind clinical trial, there was a dose- selective BK openers. In fact, the skeletal muscle BK is dependent reduction in seizure frequency with a maxi- composed of the ␣-subunit alone, the vascular BK is composed of ␣ ϩ ␤1, and the neuronal types of ␣ ϩ ␤4 More recently, the anti-inflammatory drugs meclofe- or ␣ ϩ ␤3. These channels also show different responses namate and diclofenac (10 – 40 ␮mol/L) were shown to to modulators. Peptide toxins can discriminate between be agonists of KCNQ2/3 channels. The effects of these peripheral BK formed by ␣-subunit alone, which is sen- drugs are KCNQ2/3 selective and synergic to retigabine, sitive to charybdotoxin (ChTX), and neuronal BK chan- suggesting different sites of interaction. These drugs show nels formed by ␣ ϩ ␤4 subunits, which is resistant to anticonvulsant activity in vivo in maximal electroshock sei- ChTX but sensitive to iberiotoxin (IbTX). Slotoxin, from zure (MES) tests. This may also contribute to their effects in a scorpion venom, selectively blocks ␣-subunit of mam- migraine, neuropathic pain, and The antistroke malian BK channels (K ϭ 1.5 nmol/L) and can distin- agent BMS-204352 (trade name: Maxipost) is also a mod- guish among ␣, ␣ ϩ ␤1, and ␣ ϩ ␤4 more efficiently ulator of all neuronal KCNQ The effect is stereoselective, because the S-enantiomer is an agonist The BK openers comprise a large series of synthetic and the R-enantiomer is an antagonist; both enantiomers benzimidazolone derivatives such as NS004 and NS1619, biaryl amines, biarylureas, pyridyl amines, 3-aryloxin- Neurotherapeutics, Vol. 4, No. 2, 2007 doles, benzopyrans, dihydropyridines, and natural mod- the KCMNA1 gene encoding the slo1 BK ␣-subunit ulators such as dihydrosoyasaponin-1 (DHS-1) and fla- are linked to generalized epilepsy with dyskinesia. Ab- vonoids. Both NS004 and NS1619 are known as normal function of the BK channels present in the pre- ␣-subunit-selective BK openers. NS1619 is the only synaptic terminals affects the release of inhibitory neu- compound without any effects on other ion channels. The rotransmitters. The Ca2ϩ influx into the presynaptic 3-fluoro aryloxindole analog BMS-204352 is neuropro- terminal via voltage-gated Ca2ϩ channels initiates neu- tective and reduces infarct size in two rat stroke models.
rotransmitter release and activates presynaptic BK chan- BMS-204352 has no effect on heart rate and mean arte- nels, which in turn terminates neurotransmitter release.
rial pressure in conscious dogs. In hippocampal slices it Neuronal BK channels composed of a ␤3-subunit variant (␤3b-V4) fail to terminate action potential and hence L). BMS-204352 was well tolerated in phase I and II contribute to neuronal excitability observed in idiopathic clinical trials, but failed to show efficacy against placebo epilepsy. Currently, no selective blockers are available as an antistroke agent in phase III clinical trials. Re- cently, BMS-204352 showed dose-related anxiolytic ef- In this respect, the antiepileptic effects of carbonic ficacy due to activation of KCNQ2–5 channels, and the anhydrase inhibitors including ACTZ, zonisamide, and likely that this drug would be of benefit in other disease sulthiame appear to be related to their ability to lower intracellular pH through inhibition of neuronal CA en- Other than benzimidazolone derivatives, a wide struc- zymes with reduction of the neuronal Lowering tural diversity of drugs showing BK activation properties of intracellular pH is known to inhibit several ion chan- has emerged. We have found that some carbonic anhy- nels, including neuronal BK channels. Other unselective drase inhibitors increase Kϩ currents in membrane BK blockers are verapamil and gallopamil, which pro- patches isolated from muscle fibers by interacting with duce a flickering block of vascular BK channels. The anesthetic ketamine inhibits BK channels by an indirect order of potency as BK channel openers is ACTZ Ͼ mechanism. The antifungal clotrimazole inhibits BK ac- bendroflumethiazide Ͼ ethoxzolamide Ͼ dichlorphena- tivity, thereby increasing hormonal secretion and neuro- mide (DCP) (trade name: Daranide). Their action as BK channel openers is not correlated with the inhibition of IK channel blockers and openers. IK channel block-
the carbonic anhydrase enzymes. ACTZ and other car- ers may be of therapeutic interest for immunosuppressive bonic anhydrase inhibitors are effective in preventing the therapy, through modulation of thymocytes and erythro- insulin-induced paralysis and in restoring the serum Kϩ cytes Nonselective IK blockers such as clo- levels in Kϩ-depleted rats, thus explaining their efficacy trimazole have shown antiproliferative effects on lym- phocytes and cancer cell lines. Clinical evaluation of and DCP are indeed the first-line drugs in PP to reduce attack frequency and restore serum Kϩ levels. ACTZ is IK channel openers may be beneficial in hypertension, effective in hypoPP type 1, but efficacy in hypoPP type cystic fibrosis, and peripheral vascular disease. Although 2 is unclear. We have identified a BK channel in slow (1-EBIO) and clinically used benzoxazoles are de- Various drugs such as niflumic, flufenamic, and mefe- scribed as pharmacological activators of the IK namic acids, as well as 17-␤ estradiol, activate BK chan- nels in a nonselective manner. Channel activation by17-␤ estradiol could contribute to its nongenomic effect SK channel blockers and openers. SK channel
on the vasculature (acute vasorelaxation). Tamoxifen, an blockers have been suggested for the treatment of estrogen receptor antagonist with mixed estrogenic prop- myotonic muscular dystrophy, in which an abnormal erties, activates BK channels at therapeutic concentra- activation of this channel has been found. They are also tions and blocks other ion channels such as Kv channels; proposed in the treatment of gastrointestinal dysmotility, this may explain the tamoxifen-induced QT prolongation memory disorders, epilepsy, narcolepsy, and alcohol in- toxication. Three classes of SKCa blockers are known: the neuroprotection by BK openers may be the activation peptide toxins such as apamin and leiurotoxin I (scylla- of the mitochondrial BK channels, which couple the toxin), bis-quinolinium blockers, and neuromuscular intracellular Ca2ϩ levels to the electrical activity of the SK channel openers may be important in diseases BK channel blockers. Blockers of BK channels may
involving loss of synaptic plasticity, including age-re- have a role in those conditions associated with abnormal lated loss of memory and learning in Alzheimer’s Neurotherapeutics, Vol. 4, No. 2, 2007 Pharmacology of inward rectifier potassium
noxidil and pinacidil activate the vascular Kir6.1/SUR2B (Kir) channels
complex. However, the first-generation KCO-K Since the initial cloning of the first inward rectifiers Kir1.1 limited use, in that their lack of tissue selectivity con- (ROMK1) and Kir2.1 (IRK1), new members of this family tributes to side effects such as hypotension, tachycardia, have been identified, including the G protein-coupled Kir3 and the ATP-sensitive These channels play important Second-generation KCOs synthesized include cy- roles in many organs including brain, heart, kidney, endocrine clobutenediones (WAY151616), 2H-1,4-benzoxazine de- cells, ear, and retina. Seven Kir subfamilies are known: rivatives, dihydropyridine-related structures (ZM244085), Kir1.1 (KCNJ1), Kir2.1–4 (KCNJ2,4,12,14), Kir3.1–4 and tertiary carbinols (ZD6169), all showing enhanced (KCNJ6,5,9,3), Kir4.1–2 (KCNJ10,15), Kir5.1 (KCNJ16), potency and tissue selectivity. However, these com- Kir6.1–2 (KCNJ8,11), and Kir7.1 (KCNJ13). Kir2.1 is ex- pounds fail to show any advantage over other antihyper- pressed in heart, skeletal muscle, and several brain areas, coas- tensive and antiasthmatic drugs in Phase III clinical sembling with other Kir2 to form functional channels. ATP opens Kir2 channels, possibly through phosphorylation by effects have been also investigated in neu- PKA or PIP2. Loss-of-function mutations of Kir2.1 are linked romuscular disorders. Cromakalim repolarizes muscle fi- to Andersen’s syndrome, which is a periodic paralysis associ- bers in hypoPP patients as well as in Kϩ-depleted rats. A ated with arrhythmias, dysmorphisms, Kϩ homeostasis anom- few experiments also show that this drug may, in vitro, suppress the abnormal myofiber hyperexcitability of channels are octameric complexes of Kir6.1–2 myotonic patients. Furthermore, pinacidil is effective in and the sulfonylurea receptor subunits (SUR1, SUR2A, reducing the attack frequency and in restoring muscle and SUR2B) with 1:1 stoichiometry. These channels are strength in Diazoxide was also found to ame- metabolically regulated and couple energy status of the liorate the weakness and paralysis in human hypoPP, involved in several physiopathological processes involv- These drugs hyperpolarize the sarcolemma when the ing glucose metabolism and heart contractility dysfunc- ATP/ADP ratio increases, reducing Ca2ϩ influx and elec- tion. In skeletal muscle, a reduced activity of K trical activity. This has a beneficial effect on cellular nels in human hypoPP patients carrying the R528H energy saving. An additional mechanism has been pro- mutation of dihydropyridine receptor was found. Simi- posed for diazoxide, which is also considered as a mito- larly, in Kϩ-depleted rats a reduced expression or activ- opener. It causes swelling, stimulation of respira- ity of the Kir6.2/SUR2A subunits was observed in fast- tion, inhibition of the MPTP pore, and Ca2ϩ overload twitch muscles, suggesting a contribution of this channel reduction in mitochondria, thereby contributing to neu- to the hypoPP phenotype. Other Kir channels play a role in epithelial transport of Kϩ ions such as the ROMK1, not skeletal muscle selective. New benzopyran deriva- which is associated with the renal Bartter’s syndrome tives, such as 2H-1,4-benzoxazine analogs, have there- characterized by hypokalemia. Kir3.1 (GIRK1) forms fore been synthesized and tested on the muscular KATP heteromeric channels with other members of the family, which are activated by G protein (G␤␥ subunit) and by channels in the nanomolar concentration range—show- ing however, a peculiar bell-shaped, dose–response channel openers. Knowledge of the tissue-se-
curve that may limit their use in vivo.
lective expression of various SUR subunits (SUR1, Experiments are ongoing with new compounds to cir- cumvent this problem. Recent findings that skeletal mus- made possible the search for tissue-selective cle, cardiac, and vascular tissues express SUR1 and SUR2B subunits, which are the high-affinity binding site ceptors for antagonists of the pancreatic channels, such for sulfonylureas, have toxicological For as sulfonylureas (glibenclamide and tolbutamide) and example, the interaction of sulfonylureas with sarcolem- glinides developed as antidiabetic drugs, and for the Kϩ channels enhances insulin sensitivity, contrib- uting to the sulfonylurea-dependent hypoglycemia. Con- antihypertensive, and antihypoglycemic drugs. First- cyanoguanidines (pinacidil), thioformamides (aprikalim), Kir channel openers. The pharmacology of Kir chan-
thiadiazines (diazoxide), and pyridyl nitrates (nicorandil, minoxidil). Diazoxide activates the pancreatic Kir6.2/ Andersen’s syndrome is symptomatic and aimed mainly SUR1 channel. The main cardiac and skeletal muscle at correcting the ventricular fibrillation and arrhythmias.
Kir6.2/SUR2A complex is activated by cromakalim, and Kϩ-sparing diuretics such as triamterene and spironolac- pinacidil, and cromakalim, diazoxide, nicorandil, mi- tone are effective in this disorder, possibly by improving Neurotherapeutics, Vol. 4, No. 2, 2007 Kϩ homeostasis. ACTZ may also be effective in Anders- Ca 1.2 is the cardiac and smooth muscle isoform in- en’s syndrome, through a mechanism as yet unknown.
volved in EC coupling. It is found also in pancreatic ␤ ACTZ may improve Kϩ homeostasis, as observed with cells, where it is involved in insulin secretion, and in the the Kϩ sparing diuretics. Alternatively, the intracellular somatodendritic area of neurons. Ca 1.3– 4 are the sen- acidification due to CA inhibition by ACTZ may indi- sory isoforms localized in photoreceptors and cochlear cells, where they control neurotransmitter release, and in the action potential. Other drugs possibly effective in this endocrine cells, neurons, and to lesser extent in cardiac cells, where they control cardiac rhythm. Ca 2.1–3 chan- nicorandil and pinacidil, which may correct the lack of nels are linked to EA2, EA6, and migraine, disorders inward rectifier Kϩ currents in cardiomyocytes and ac- characterized by gain-of-function mutations of these tion potential prolongation. It is unlikely that the drug channels in the cerebellum and spinal cord with neuronal effects in Andersen’s syndrome pass through activation atrophy and degeneration. Ca 2.2 is a neuronal channel of BK channels, which are not expressed in the heart.
involved in pain sensation and inflammation. Ca 3 chan- and Kir channel blockers. K
nels are expressed in somatic and dendritic areas of blockers, such as sulfonylureas and glinides, are not in neurons in the hippocampus, hypothalamus, thalamus, use in neuronal channelopathies. Kir channels are cerebellum, and cortex and are responsible for the T- blocked by Ba2ϩ and Csϩ ions and polyamines, which currents. Abnormal openings of these channels are the basis for the low-frequency discharge (3 Hz) of the hy-pothalamic nucleus associated with absence epilepsy.
Pharmacology of two-pore potassium channels
The two-pore potassium ion channels (K2P) are re- Ca 1.1–1.4 channel antagonists
sponsible for the background Kϩ conductance in the Ca subunits have similar pharmacology and are tar- cells at Fifteen mammalian genes belong to the gets of phenylalkylamines, dihydropyridines, and benzo- KCNK family encoding the K2P channels, including thiazepines, and are used as antihypertensive and antiar- TASK1–3, TREK1–2, TRAAK, TWIK1–2, TALK1–2, and others. These are controlled by several stimuli, including Ca2ϩ antagonists in the treatment of hypoPP patients was oxygen tension, pH, lipids, mechanical stretch, neuro- transmitters, and G protein-coupled receptors. These Ca 2.1–3 channel antagonists
channels are also the targets for volatile and local anes- Ca 2.1–3 channels are insensitive to classical Ca2ϩ thetics. TASK is involved in chemoreception; its inhibi- antagonists, but are specifically blocked by high-affinity tion by extracellular protons or hypoxia depolarizes the peptide toxins. Selective antagonists would be of benefit cells and starts the firing of respiratory motor neurons in EA2– 6 and migraine disorders. Acetazolamide is the with increased frequency in the respiratory reflexes.
first-line treatment in EA2 (which is thus named aceta- TREK channels are expressed in neurons involved in zolamide-responsive ataxia) and can also help EA6 pa- thermoregulation. These are modulated by lipids and tients. The ACTZ effects in EA2– 6 appear to be medi- fatty acids. Volatile anesthetics open TREK channels, ated by mechanisms not involving the P/Q type Ca2ϩ whereas local anesthetics block these channels. The fact It is likely that opening of presynaptic BK that the neuroprotective agent riluzole is an activator of channels by ACTZ would buffer the P/Q mutant-depen- TREK raises the question as to whether it can be a drug dent increase of presynaptic Ca2ϩ ion levels, thus con- trolling neurotransmitter release. In addition, inhibitionof the R-type Ca2ϩ channel by ACTZ would reduce PHARMACOLOGY OF VOLTAGE-GATED
CALCIUM (CA ) CHANNELS
channels have strong anti-inflammatory and analgesic effects comparable or superior to opiates. This channel is Ten different genes encode different ␣-subunits com- indeed the target of cannabinoids, opioids, neuropeptide posing the voltage-gated Ca2ϩ Ca 1.1– 4 Y, and substance P. Intrathecal administration of zi- (␣ , ␣ , ␣ , and ␣ ) mediate L-type Ca2ϩcurrents; conotide (trade name: Prialt), a synthetic analog of the Ca 2.1–3 (␣ , ␣ , ␣ ) mediate P/Q-type, N-type, and ␻-conotoxin MVIIA, is effective in patients not respon- R-type Ca2ϩcurrents, respectively; and Ca 3.1–3 (␣ , sive to opiates. Nonselective blockers of these channels ␣ , ␣ ) mediate T-type currents. The ␣-subunits coas- are mibefradil, piperazines, gabapentin, and volatile an- semble with ␣2, ␤, ␦, and ␥ subunits to form functional Ca 1.1 is the skeletal muscle isoform localized in the Ca 3 channel antagonists
T-tubules participating in EC coupling. Loss-of-function Ca 3 channel antagonists are not related to drugs tar- mutations are associated with hypoPP type 1 in humans.
geting others Ca2ϩ Mibefradil is fairly selec- Neurotherapeutics, Vol. 4, No. 2, 2007 tive for T-type versus L-type Ca2ϩ currents, but was anesthetics, because they were first developed as alter- withdrawn from the market for its low pharmacotoxico- natives to cocaine to obtain anesthesia and logical profile. The peptide kurtoxin inhibits activation Their cardioprotective effect was discovered later, lead- gating of Ca 3.1 and Ca 3.2 Other nonselec- ing to the birth of class I antiarrhythmics. Cardiac action tive blockers are penfluridol, pimozide, amiloride, and of Naϩ channel blockers is discussed elsewhere in ex- phenytoin. More specific and high-affinity blockers of cellent Most clinical LAs have a tertiary Cav3 channels would be useful for therapy. These chan- amine associated with a hydrophobic aromatic ring nels are the main target of ethosuximide, widely used through an amide or ester link. Experimental data sug- against absence seizures. Other AEDs with multiple gest that the two ends of the drugs may interact with modes of action, such as zonisamide (trade name: Zoni- channel pore-lining amino acids through hydrophobic or gram) and valproate, can inhibit T-type Ca2ϩ ␲-cation Although antiepileptic Naϩchannel blockers constitute a more heterogeneous chem-ical class, it is thought that their molecular receptor PHARMACOLOGY OF VOLTAGE-GATED
matches, at least partially, that of the SODIUM CHANNELS
Because the LA binding site lies within the Naϩ chan- The human genome contains nine genes encoding the nel pore, both lipophilicity and pKa of the drugs are main ␣-subunit of voltage-gated Naϩ channels (SCN1A, important determinants of channel block. At physiolog- SCN2A, and the rest) and at least four genes encoding ical pH, LAs equilibrate between a liposoluble neutral auxiliary ␤-subunits (SCN1B to SCN4B), the expression form, which may reach or leave the receptor across the of which is tissue-specific. Before the breakthroughs plasma membrane lipid phase even if the channel is made with molecular genetics, the complexity of the Naϩ closed, and a protonated form that needs channel opening channel repertoire was suggested by their different re- to enter the pore and inhibit Naϩ currents in a use- sponse to natural toxins. An initial distinction was made dependent In addition, according to the mod- on the basis on their sensitivity to tetrodotoxin (TTX), a ulated receptor hypothesis, the pore changes conforma- paralytic poison from some pufferfish. Naϩ channels tion during channel activity, thereby modifying the drug were therefore divided into TTX-sensitive, which are binding affinity as a function of channel state: the bind- expressed in the differentiated skeletal muscle and in ing affinity of LA is far greater for open and inactivated central and peripheral neurons, and TTX-resistant, in- sodium channels than for closed channels. Consequently, cluding the cardiac isoform and some subtypes expressed it is still a matter of debate whether the greater drug in peripheral nerves. Further distinction was possible affinity for a specific Naϩ channel isoform with respect with the use of ␮-conotoxins isolated from marine cone to others depends on subtle differences in the receptor snails, which have a much higher affinity for skeletal site or differences in channel gating that secondarily muscle Naϩ channels than for neuronal channels.
affect receptor accessibility, or both.
At least eight distinct neurotoxin binding sites have The state-dependent affinity of drugs for Naϩ channels been identified within the Naϩ channel protein, with has two important implications for therapy. First, be- different effects on ion permeation and gating resulting cause the transition of the channel from the low-affinity in either inhibition or enhancement of Naϩ currents.
closed state to the high-affinity open or inactivated state These toxins provide important tools for binding assays depends on membrane voltage, the drugs will blocks and definition of the channel structure–activity relation- Naϩ channels in a voltage-dependent manner. Thus, the ship. For instance, the activation of muscle Naϩ channels channel blocking will be greater in neurons than in skel- by the sea anemone toxin ATX II is used as a model for etal muscle fibers or in an ischemic area than in healthy the biophysical and pharmacological characterization of tissue. Second, channel block will increase with the fre- quency of stimulation, because the channel will spend ins have not found any therapeutic application, whereas more time in the open or inactivated states and the drug synthetic Naϩ channel ligands have provided us with will have less time to dissociate from the channel be- local anesthetics (LA), antiarrhythmics, and AEDs.
tween two action potentials. This frequency-dependent(or use-dependent) block is fundamental for the clinical Sodium channel blockers
use of LA drugs, allowing their selective action on hy- Because Naϩ channels are responsible for the upstroke perexcited tissues while preserving the normally func- and propagation of action potential in most excitable cells, drugs blocking Naϩ channels find application in a Use-dependent block is the basis for the use of LA in large spectrum of membrane hyperexcitability disorders, the treatment of myotonic syndromes, which are charac- including cardiac arrhythmias, epilepsies, myotonias, terized by high-frequency action potential discharges in and chronic Many of these drugs are chemically skeletal Currently, the preferred drug against related, and are grouped under the generic term local myotonia is mexiletine, which is marketed as an antiar- Neurotherapeutics, Vol. 4, No. 2, 2007 rhythmic drug. More anecdotally, other Naϩ channel zures in EA1, although Naϩ channel blockers may cause blockers used against myotonia include tocainide, pro- ataxia. Valproate, in combination with other AEDs, may cainamide, disopyramide, phenytoin, flecainide, and car- be valuable in severe myoclonic epilepsy of infancy bamazepine. There is, however, no controlled clinical (SMEI), but lamotrigine may exacerbate seizures. This trial available to confirm the efficacy and tolerability of knowledge is based on physician experience, and molec- ular genetic approaches have as yet had little impact on Mexiletine is administered in doses of 150 to 200 mg SMEI Because SCN1A mutations may pro- two to three times a day and is generally well tolerated.
duce a loss of function of the Naϩ it might Cardiac function and drug serum concentration should be appear reasonable to avoid treatment with Naϩ channel carefully monitored, to reduce the risk of CNS and blockers in SMEI patients before a clear genotype–phe- cardiac toxicity. The fact that mexiletine is useful in many patients suffering from either ClϪ channel or Naϩ As for Naϩ channel myotonias, there is also a possi- channel myotonia, or even from myotonic dystrophy, bility that Naϩ channel mutations in epilepsy may mod- suggests that mexiletine treatment is symptomatic. The ify the AED effect. Recently, an epileptic mutation in the drug, by blocking Naϩ channels, counteracts the hyper- auxiliary ␤ -subunit (mutation C121W in SCN1B) was excitability induced by the different genetic defects. In shown to modify the voltage-dependence of a neuronal the case of Naϩ channel myotonias, however, the muta- Naϩ channel (SCN3A), consequently reducing the sensi- tions themselves can modify the sensitivity of the chan- tivity of the mutant channel to Whether this finding is relevant for therapy remains to be verified. On from altered intrinsic drug affinity or from mutation- the other hand, a recent study suggests that a common induced altered gating that secondarily alters drug effect.
SCN1A polymorphism may influence the clinical use of Indeed, we proposed that the voltage-dependence of phenytoin and carbamazepine, thereby supporting the channel availability may be considered as a general in- Before its use in epilepsy, carbamazepine was origi- The few mutations that shift this voltage-dependency nally approved in the United States for treatment of toward more negative potentials increase mexiletine sen- trigeminal neuralgia, a chronic pain syndrome. Other sitivity, and the heterozygous patients may respond well AEDs, such as phenytoin and lamotrigine, and some LA, to therapy, because of the preferential block of mutants such as lidocaine and mexiletine, were also found to be with respect to wild-type channels. On the other hand, functional in the treatment of neuropathic pain in small mutations that shift voltage-dependency toward more cohorts of patients. It is believed that abnormal Naϩ positive potentials reduce drug sensitivity; in this case, channel excitability of injured axons and their respective the mexiletine effect may be attributable to wild-type dorsal root ganglia neurons is a primary event in neuro- channel block, and patients may benefit from a drug pathic pain. Recent findings strengthen this view, be- having a more specific action on the mutant channel. On cause a gain of function of the Nav1.7 channel expressed the way to defining a pharmacogenetic strategy to better in nociceptors was shown to cause familial erythermal- address treatment in individual myotonic patients, we gia, an inherited pain syndrome linked to SCN9A muta- showed that flecainide may exhibit greater benefit in tions and responsive to lidocaine and mexiletine treat- those patients carrying mutations with a positively and an increased expression of the Nav1.3 channel was also found along the pain pathway after In isolated neurons, the inhibition of high-frequency spinal cord In addition, the up-regulation of action potential discharges by Naϩ channel blockers in- TTX-resistant Naϩ channels induced by prostaglandins cluding phenytoin, carbamazepine, oxcarbazepine, lam- in nociceptors was recently pinpointed as the causal otrigine, and valproate is obtained with drug concentra- mechanism for the development of hyperalgesia in in- tions similar to those able to produce an anticonvulsive flammatory Although such a conclusion was re- effect in humans. These drugs are widely used in partial epilepsies and in symptomatic generalized tonic-clonic Naϩ channel involvement in the various types of pain, seizures. Lamotrigine and valproate are also efficient the development of isoform-specific Naϩ channel block- against absence seizures, most probably because they ers with greater efficiency and fewer side effects may have multiple targets over Naϩ channels. Also, the glu- represent a stimulating area in analgesia.
tamate antagonist AEDs, felbamate and topiramate, The tricyclic antidepressants are known to alleviate might block Naϩ channels. There is, however, no clear various pain syndromes, and Naϩ channel blockade by picture as to whether these drugs may be of benefit to these drugs was proposed as the key mechanism for pain patients with idiopathic epilepsy originating from a chan- relief. For instance, the receptor for the tricyclic antide- pressant amitriptyline overlaps with the LA receptor Phenytoin and carbamazepine are used to control sei- in Naϩ channels, and Naϩ channel blockade by amitrip- Neurotherapeutics, Vol. 4, No. 2, 2007 tyline conserves the hallmarks of LA It is Sodium channel activators
noteworthy that tricyclic antidepressants such as imipra- The therapeutic use of Naϩ channel openers may seem mine, clomipramine, and amitriptyline have improved inadvisable, because mutations resulting in a slight Naϩ channel gain of function can induce membrane overex- Sodium channel blockers are also increasingly studied citability in humans. Loss of function or expression of Naϩ channels has, however, been found to be associated nel blockers are in clinical trials for the acute treatment with some sodium channelopathies leading to periodic of cerebral stroke. Naϩ ions and channels may have an paralysis, cardiac arrhythmia, and epilepsy. In these dis- important role in the cascade of events leading to cell orders, drugs that can promote Naϩ currents may be damage after global or focal ischemia in the brain, and useful to restore normal excitability of affected tissue, many Naϩ channel blockers display considerable neuro- but no such drug is currently available for therapy.
protective effects in models of ischemia. In amyotrophic One promising possibility is the use of pharmacolog- lateral sclerosis, the only drug available today to reduce ical chaperones, which are low molecular weight com- motor neuron damage is riluzole. Although this drug may pounds that bind selectively to intracellularly retained have many molecular targets, it is thought that neuropro- proteins and promote their proper folding and membrane tection arises primarily from Naϩ channel blockade.
targeting. A number of genetic defects associated with Other than the classical Naϩ channel blockers, a num- epilepsy and cardiac arrhythmia impair Naϩ channel ber of drugs acting primarily on other targets, as well as trafficking, resulting in haploinsufficiency. It has been some dietary compounds, have been shown to block Naϩ shown that Naϩ channel ligands such as mexiletine can channels at relevant clinical concentrations. Recent ex- rescue mutant channels to the surface How- amples include the antidepressant selective serotonin re- ever, a number of issues must be addressed before ad- uptake inhibitor, fluoxetine, and the red grape polyphe- vancing to clinical trials in patients suffering from so- dium channelopathies. For instance, drugs may be able to example are drugs acting on ␤ -adrenergic receptors, achieve channel rescue without blocking its activity and which are able to block Naϩ channels in a manner re- promote normal gating behavior of the rescued mutant dundant to Some ␤ -adrenoceptor agonists have channels. Alternatively, such agents could increase the been used successfully to prevent attacks in hyperkale- expression of wild-type channels, which may counteract mic periodic paralysis patients, most probably by coun- the haploinsufficiency due to retention of the allelic teracting muscle membrane depolarization through acti- channel mutant (Desaphy and Conte Camerino, unpub- vation of the electrogenic Naϩ/Kϩ pump. Not all the ␤ -agonists may be used in hyperPP patients, however, because the Naϩ channel block exerted by some of them PHARMACOLOGY OF VOLTAGE-GATED
(including clenbuterol) may accentuate the paralysis.
CHLORIDE CHANNELS
Rather, we proposed the use of clenbuterol in patientssuffering from myotonic dystrophy, because of the pos- The voltage-gated chloride channels of the ClC family sibility of combining its well-known anabolic action with are encoded by nine different genes in mammals the antimyotonic activity through Naϩ channel use-de- (CLCN1, CLCN2, and the rest). This family may serve as a paradigm for using human hereditary diseases and Empiric pharmacology has thus supplied us with a mouse models to facilitate the elucidation of the cellular number of widely used drugs acting primarily as Naϩ roles of channels, whose function and even existence channel blockers. Current drugs have, however, a rela- were still mysterious some 15 years ago. Myotonia con- tively low selectivity profile, acting equally on the vari- genita (MC) was the first human disease proven to be ous Naϩ channel subtypes. At present, the sole guarantee caused by an ion channel defect, thus leading to the for safe use is provided by the voltage and use-dependent discovery of the CLCN1 gene encoding the ClC-1 chan- characteristics of Naϩ channel blockade. In the last de- nel responsible for the high ClϪ conductance of skeletal cade, the discovery of sodium channelopathies has tre- mendously increased our knowledge of Naϩ channel with MC mutations, thereby exacerbating muscle excit- involvement in pathological processes. In parallel, the ability and leading to abnormal myotonic runs of action elucidation of the intramolecular drug binding site pro- vides an opportunity to understand the complex drug– Human mutations in other CLCN genes have been channel interactions. These milestones promise the rapid linked to idiopathic epilepsy, kidney diseases with or development of pharmacogenetics, where old and new drugs which selectively affect specific Naϩ channel sub- channels are expressed in the plasma membranes and in types or mutants will be matched to specific diseases or intracellular organelles of many tissues, where they per- form a variety of functions including the regulation of Neurotherapeutics, Vol. 4, No. 2, 2007 cellular excitability, cell volume, transepithelial trans- perekplexia in humans, and GABA R mutations have been linked to idiopathic generalized epilepsy and An-gelman’s syndrome, a disorder characterized by severe Chloride channel blockers
mental retardation, convulsions, and delayed motor de- Much of the information regarding ClC channel phar- macology has been obtained by studying skeletal muscle The GlyR has a very modest pharmacological profile, ClϪ conductance, leading to the identification of a series and no therapeutic ligands are currently The of blockers—which are, however, poorly selective, loss of GlyR function due to mutations suggests, how- blocking many ClϪ channels and transporters in various ever, that selective GlyR agonists may find therapeutic Thus, more work is needed to develop ClϪ application in diseases characterized by a lack of motor channel blockers with enough affinity and specificity for Compared with GlyR, the drugs acting on GABA R Chloride channel openers
encompass a broad therapeutic field, including anxiety, To date, all mutations found in CLCN genes produce a epilepsy, mood disorders, sleep disorders, schizophrenia, loss of function of the encoded ClC channel. The drugs cognitive disorders, and general Drug af- of interest for chloride channelopathies should therefore finity depends on receptor subunit composition, thereby be able to increase ClϪ currents, a goal not yet offering a chance to develop GABA R subtype-specific The only exception is a bicyclic fatty acid, drugs. For instance, the ␤3 subunit of GABA R may be called lubiprostone, that selectively activates ClC-2 important for promoting sleep, because a mutation in this The U.S. Food and Drug Administration has subunit was found in a patient with chronic insomnia and recently approved lubiprostone (trade name: Amitiza) for ␤3 subunit knockout mice are insensitive to oleamide, an the treatment of idiopathic chronic constipation. Identi- endogenous fatty acid sleep promoter. Thus, finding fication of specific ClC channel openers would be useful drugs acting preferentially on ␤3 subunit-containing for counteracting hyperexcitability in myotonia and ep- GABA R may be useful in the treatment of insomnia.
Positive allosteric GABA R modulators, such as phe- Another way to enhance ClϪ currents is to indirectly nobarbital and the benzodiazepines (diazepam, clonaz- increase ClC channel expression or function by stimu- epam, lorazepam, and the like), are widely used in the lating intracellular biochemical pathways. For instance, treatment of partial and generalized epilepsies. Mutations muscle ClϪ conductance increases in vivo and in vitro found in ␣1 and ␥2 subunits of GABA R associated with after application of insulin-like growth factor-1 or tau- epilepsy result in diminished synaptic inhibition, provid- Chronic treatment with taurine may improve ing an explanation for the increased propensity to con- More recently, ACTZ, used as an alternative vulsions. Benzodiazepines may thus be particularly in- drug in myotonia, was shown to increase ClC-1 channel dicated in patients carrying GABA R mutations. Some currents in a mammalian cell line, most probably through mutations, however, may also impair GABA R expres- sion or benzodiazepine responsiveness, which theoreti- ACTZ was able to improve function of a myotonia con- cally may hamper their use in these patients. Clearly, genital ClϪ channel Thus, the indirect activa- more studies are needed to allow a pharmacogenetic tion of wild-type and mutant ClC-1 channels by ACTZ might contribute to its therapeutic effects in muscle dis-orders.
PHARMACOLOGY OF NEUROTRANSMITTER-
GATED CATIONIC CHANNELS
PHARMACOLOGY OF NEUROTRANSMITTER-
Neurotransmitter-gated cationic channels include nic- GATED CHLORIDE CHANNELS
otinic ACh receptors, glutamate receptors, and serotonin- The ␥-aminobutyric acid and glycine receptors ergic receptors. The nicotinic receptor channels are ex- (GABA R and GlyR) are the major inhibitory neuro- pressed in the neuromuscular plaque, the noradrenergic transmitter-gated receptors in the Both recep- and cholinergic ganglions, and the CNS. Several nico- tors are pentameric, formed from subunits produced by tinic receptor subtypes are known to show different phar- different genes or splice variants, or both. After neuro- macological responses to drugs and toxins. Although transmitter binding, the ingress of ClϪ ions within the loss-of-function mutations of nicotinic receptors are in- cell hyperpolarizes the postsynaptic membrane, resulting volved in MG and epilepsy, no drugs are known that are in neurotransmission inhibition. Both receptors can be capable of restoring normal channel function. The drugs gated also by taurine. Dysfunction of GlyR and and toxins targeting nicotinic receptors are instead used GABA R have been implicated in various channelopa- as ganglioplegic and muscle relaxants in anesthesia.
thies. GlyR mutations induce spasticity in mice and hy- The serotoninergic receptor channel (5-HT type) is Neurotherapeutics, Vol. 4, No. 2, 2007 expressed in the peripheral nerve and area postrema, pharmacogenetic approach aimed at defining the best where it plays a role in antinociception and antiemetic responses. Drugs targeting this channel, such as ondan- Today, the gold standard technology for studying ion setron and tropisetron, are in use as antiemetics. Gluta- channel pharmacology remains the patch– clamp tech- mate receptor channels are instead important targets for nique. The excellent signal and time resolutions of neuroprotective and antiepileptic drugs, and thereby are patch– clamp recordings allow the fine dissection of in- used in the neuronal channelopathies. Drugs acting on timate drug– channel interactions. The downside is that glutamate receptors reduce the excitotoxicity in the CNS, this is a laborious and time-consuming procedure, which being effective in acute hypoxic–ischemic brain injury limits its application as a screening platform. With the and in chronic neurodegenerative diseases such as Alz- increasing awareness of the importance of ion channels heimer’s, Parkinson’s, Huntington’s, and amyotrophic in drug discovery, owing in part to the understanding of lateral sclerosis. Glutamate antagonists are also effective channelopathies, high-throughput screening technologies in neuropathic pain, dementia, and melanoma, as well as are being developed in the pharmaceutical industry. On in neuroprotection. More recently, the use of glutamate the other hand, application of genomics and proteomics agonists in schizophrenia has been proposed.
to ion channels (i.e., channelomics) offers a tremendousopportunity to underscore the role of ion channels inmany diseases and their contribution to the success or CONCLUSION
It is thus likely that with a greater understanding of In this review, we sought to describe the pharmacol- channel genetics, structure, and function, together with ogy of ion channels involved in neurologic disorders the identification of novel primary and secondary chan- with a special regard to neurologic channelopathies.
nelopathies, ion channel drugs will continue to see a Many other channels have been excluded from the re- substantial growth the neurologic pharmaceutical sector.
view because their function or pharmacology are notnoticeably related to neurologic diseases (e.g., the cystic Acknowledgments: This work on ion channelopathies
fibrosis transmembrane conductance regulator chloride was supported by grants from Telethon-Italy (Grant no.
channel CFTR). Other channels have not been discussed here because their pharmacology, although very promis- RBNE01XMP4 and PRIN-2005059597), and the Italian ing, is quite distant from therapeutic applications (e.g., the transient receptor potential channel TRP). Alsonon-neurologic channelopathies were not covered in de-tail, such as the cardiac arrhythmias (e.g., long QT and REFERENCES
Brugada syndromes), kidney diseases (e.g., Dent’s dis- 1. Ruetsch YA, Böni T, Borgeat A. From cocaine to ropivacaine: the ease, Bartter’s syndrome), and neonatal diabetes. It history of local anesthetic drugs. Curr Top Med Chem 2001;1:175– should be mentioned, however, that the reader will find 2. Gutman GA, Chandy KG, Grissmer S, et al. International Union of excellent descriptions in the literature of these chan- Pharmacology. LIII. Nomenclature and molecular relationships of nelopathies and their treatment, which may serve as a voltage-gated potassium channels. Pharmacol Rev 2005;57:473– paradigm for possible therapeutic approaches in the 3. Dodson PD, Forsythe ID. Presynaptic Kϩ channels: electrifying regulators of synaptic terminal excitability. Trends Neurosci 2004; Recent advances in genetic medicine, especially in ion 4. Lehmann-Horn F, Lerche H, Jurkat-Rott K. Skeletal muscle chan- nelopathies: myotonias, periodic paralyses and malignant hyper- knowledge of the physiopathological processes leading thermia. In: Stålberg E, editor. Clinical neurophysiology of disor- to neurologic disorders. In many cases, this knowledge ders of muscle and neuromuscular junction, including fatigue.
has confirmed the validity of empirically used drugs Handbook of Clinical Neurophysiology 2. Boston: Elsevier Sci-ence; 2003:457– 483.
targeting ion channels, such as the sodium channel 5. Vedeler CA, Antoine JC, Giometto B, et al. Management of para- blockers and the GABA R agonists in epilepsy. As in neoplastic neurological syndromes: report of an EFNS Task Force.
the case of ACTZ, a novel mechanism of action of an 6. Coghlan MJ, Carroll WA, Gopalakrishnan M. Recent develop- empiric drugs is being uncovered because of our under- ments in the biology and medicinal chemistry of potassium channel standing of the drug-responsive channelopathies. More- modulators: update from a decade of progress. J Med Chem 2001; over, other ion channels have been pinpointed as new 7. Munoz-Caro C, Nino A. The nature of the receptor site for the candidate targets for drug therapy, such as the KCNQ reversible Kϩ channel blocking by aminopyridines. Biophys Chem potassium channels with retigabine currently in a phase III clinical trial evaluation for partial epilepsies. Specif- 8. Weisz CJ, Raike RS, Soria-Jasso LE, Hess EJ. Potassium channel blockers inhibit the triggers of attacks in the calcium channel ically in the channelopathies, the fine dissection of the mouse mutant tottering. J Neurosci 2005;25:4141– 4145.
biophysical effects of mutations may help to pursue a 9. Skeie GO, Apostolski S, Evoli A, et al. Guidelines for the treat- Neurotherapeutics, Vol. 4, No. 2, 2007 ment of autoimmune neuromuscular transmission disorders. Eur 31. McNaughton NCL, Davies CH, Randall A. Inhibition of ␣ channels by carbonic anhydrase inhibitors. J Pharmacol Sci 2004; 10. Michelakis E. Anorectic drugs and vascular disease: the role of voltage-gated Kϩ channels. Vascul Pharmacol 2002;38:51–59.
32. Cannon SC, Corey DP. Loss of Naϩ channel inactivation by anem- 11. Tricarico D, Barbieri M, Mele A, Carbonara G, Conte Camerino D.
one toxin (ATX II) mimics the myotonic state in hyperkalaemic Carbonic anhydrase inhibitors are specific openers of skeletal mus- periodic paralysis. J Physiol 1993;466:501–520.
cle BK channel of Kϩ-deficient rats. FASEB J 2004;18:760 –761.
33. Desaphy JF, Conte Camerino D, Franchini C, Lentini G, Tortorella 12. Rogawski MA. Diverse mechanisms of antiepileptic drugs in the V, De Luca A. Increased hindrance on the chiral carbon atom of development pipeline. Epilepsy Res 2006;69:273–294.
mexiletine enhances the block of rat skeletal muscle Naϩ channels 13. Ashcroft FM. From molecule to malady. Nature 2006;440:440 – in a model of myotonia induced by ATX. Br J Pharmacol 1999; 14. Peretz A, Degani N, Nachman R. Meclofenamic acid and diclofe- 34. Clare JJ, Tate SN, Nobbs M, Romanos MA. Voltage-gated sodium nac, novel templates of KCNQ2/Q3 potassium channel openers, channels as therapeutic targets. Drug Discov Today 2000;5:506 – depress cortical neuron activity and exhibit anticonvulsant proper- ties. Mol Pharmacol 2005;67:1053–1066.
35. Glaaser IW, Clancy CE. Cardiac Naϩ channels as therapeutic 15. Korsgaard MPG, Hartz BP, Brown WD, Ahring PK, Strøbæk D, targets for antiarrhythmic agents. Handb Exp Pharmacol 2006;171: Mirza NR. Anxiolytic effects of Maxipost (BMS-204352) and retigabine via activation of neuronal Kv7 channels. J Pharmacol 36. Yarov-Yarovoy V, McPhee JC, Idsvoog D, Pate C, Scheuer T, Catterall WA. Role of amino acid residues in transmembrane seg- 16. Wei AD, Gutman GA, Aldrich R, Chandy KG, Grissmer S, Wulff ments IS6 and IIS6 of the Naϩ channel ␣ subunit in voltage- H. International Union of Pharmacology. LII. Nomenclature and dependent gating and drug block. J Biol Chem 2002;277:35393– molecular relationships of calcium-activated potassium channels.
37. Hille B. Ion channels of excitable membranes. 3rd ed. Sunderland, 17. Ghatta S, Nimmagadda D, Xu X, O’Rourke ST. Large-conduc- tance, calcium-activated potassium channels: structural and func- 38. Heatwole C, Moxley RT III. The nondystrophic myotonias. Neu- tional implications. Pharmacol Ther 2006;110:103–116.
18. Tricarico D, Mele A, Conte Camerino D. Carbonic anhydrase 39. Trip J, Drost G, van Engelen BG, Faber CG. Drug treatment for inhibitors ameliorate the symptoms of hypokalaemic periodic pa- myotonia. Cochrane Database Syst Rev 2006;1:CD004762.
ralysis in rats by opening the muscular Ca2ϩ-activated-Kϩ chan- 40. Weckbecker K, Wurz A, Mohammadi B, et al. Different effects of nels. Neuromuscul Disord 2006;16:39 – 45.
mexiletine on two mutant sodium channels causing paramyotonia 19. Tricarico D, Mele A, Conte Camerino D. Phenotype-dependent congenita and hyperkalemic periodic paralysis. Neuromuscul Dis- functional and pharmacological properties of BK channels in skel- etal muscle: effects of microgravity. Neurobiol Dis 2005;20:296 – 41. Desaphy J-F, De Luca A, Tortorella P, De Vito D, George AL Jr, Conte Camerino D. Gating of myotonic Na channel mutants de- 20. Szewczyk A, Skalska J, Glab M, et al. Mitochondrial potassium fines the response to mexiletine and a potent derivative. Neurology channels: from pharmacology to function. Biochim Biophys Acta 42. Takahashi MP, Cannon SC. Mexiletine block of disease-associated 21. Leniger T, Wiemann M, Bingmann D, Widman G, Hufnagel A, mutations in S6 segments of the human skeletal muscle Naϩ chan- Bonnet U. Carbonic anhydrase inhibitor sulthiame reduces intra- cellular pH and epileptiform activity of hippocampal CA3 neu- 43. Desaphy J-F, De Luca A, Didonna MP, George AL Jr, Conte rones. Epilepsia 2002;43:469 – 474.
Camerino D. Different flecainide sensitivity of hNav1.4 channels 22. Kubo Y, Adelman JP, Clapham DE, et al. International Union of and myotonic mutants explained by state-dependent block.
Pharmacology. LIV. Nomenclature and structure–function rela- tionships of inwardly rectifying potassium channels. Pharmacol 44. Berkovic SF. Influence of molecular genetic advances on therapy for the idiopathic epilepsies. Neurotherapeutics 2007 (in press).
channel openers: structure–activity relation- 45. Heron SE, Scheffer IE, Berkovic SF, et al. Channelopathies in ships and therapeutic potentials. Med Res Rev 2004;24:213–266.
idiopathic epilepsy. Neurotherapeutics 2007;4:295–304.
24. Ligtenberg JJ, Van Haeften TW, Van Der Kolk LE, et al. Normal 46. Lucas PT, Meadows LS, Nicholls J, Ragsdale DS. An epilepsy insulin release during sustained hyperglycaemia in hypokalaemic mutation in the ␤1 subunit of the voltage-gated sodium channel periodic paralysis: role of the potassium channel opener pinacidil results in reduced channel sensitivity to phenytoin. Epilepsy Res in impaired muscle strength. Clin Sci (Lond) 1996;91:583–589.
25. Wu L, Shen F, Lin L, Zhang X, Bruce IC, Xia Q. The neuropro- 47. Tate SK, Depondt C, Sisodiya SM, et al. Genetic predictors of the tection conferred by activating the mitochondrial ATP-sensitive maximum doses patients receive during clinical use of the anti- Kϩ channel is mediated by inhibiting the mitochondrial perme- epileptic drugs carbamazepine and phenytoin. Proc Natl Acad Sci ability transition pore. Neurosci Lett 2006;402:184 –189.
26. Tricarico D, Barbieri M, Laghezza A, Tortorella P, Loiodice F, 48. Legroux-Crespel E, Sassolas B, Guillet G, Kupfer I, Dupre D, Conte Camerino D. Dualistic actions of cromakalim and new po- Misery L. Treatment of familial erythermalgia with the association tent 2H-1,4-benzoxazine derivatives on the native skeletal muscle of lidocaine and mexiletine [In French]. Ann Dermatol Venereol channel. Br J Pharmacol 2003;139:255–262.
27. Tricarico D, Mele A, Lundquist AL, Desai RR, George AL Jr, 49. Waxman SG, Dib-Hajj S. Erythermalgia: molecular basis for an Conte Camerino D. Hybrid assemblies of ATP-sensitive Kϩ chan- inherited pain syndrome. Trends Mol Med 2005;11:555–562.
nels determine their muscle-type-dependent biophysical and phar- 50. Waxman SG, Hains BC. Fire and phantoms after spinal cord macological properties. Proc Natl Acad Sci U S A 2006;103:1118 – injury: Naϩ channels and central pain. Trends Neurosci 2006;29: 28. Lesage F. Pharmacology of neuronal background potassium chan- 51. Lai J, Porreca F, Hunter JC, Gold MS. Voltage-gated sodium nels. Neuropharmacology 2003;44:1–7.
channels and hyperalgesia. Annu Rev Pharmacol Toxicol 2004;44: 29. Catteral WA, Perez-Reyes E, Snutch TP, Striessnig J. International Union of Pharmacology. XLVIII. Nomenclature and structure– 52. Nassar MA, Baker MD, Levato A, et al. Nerve injury induces function relationships of voltage-gated calcium channels. Pharma- robust allodynia and ectopic discharges in Nav1.3 null mutant 30. Spacey SD, Hildebrand ME, Materek LA, Bird TD, Snutch TP.
53. Sindrup SH, Jensen TS. Are sodium channel blockers useless in Functional implications of a novel EA2 mutation in the P/Q-type peripheral neuropathic pain? [Editorial]. Pain 2006 Oct. 17 [Epub calcium channel. Ann Neurol 2004;56:213–220.
Neurotherapeutics, Vol. 4, No. 2, 2007 54. Wang GK, Russell C, Wang SY. State-dependent block of voltage- editor. Chloride movements across cellular membranes. Advances gated Naϩ channels by amitriptyline via the local anesthetic re- in Molecular and Cell Biology 38. Amsterdam: Elsevier; ceptor and its implication for neuropathic pain. Pain 2004;110: 63. Cuppoletti J, Malinowska DH, Tewari KP, et al. SPI-0211 acti- 55. Meola G, Sansone V. Treatment in myotonia and periodic paral- vates T84 cell chloride transport and recombinant human ClC-2 ysis. Rev Neurol (Paris) 2004;160:S55–S69.
chloride currents. Am J Physiol Cell Physiol 2004;287:C1173– 56. Hemmings HC Jr. Neuroprotection by Naϩ channel blockade.
J Neurosurg Anesthesiol 2004;16:100 –101.
64. De Luca A, Pierno S, Liantonio A, Camerino C, Conte Camerino 57. Lenkey N, Karoly R, Kiss JP, Szasz BK, Vizi ES, Mike A. The D. Phosphorylation and IGF-1-mediated dephosphorylation path- mechanism of activity-dependent sodium channel inhibition by the ways control the activity and the pharmacological properties of antidepressants fluoxetine and desipramine. Mol Pharmacol 2006; skeletal muscle chloride channels. Br J Pharmacol 1998;125:477– 58. Wallace CH, Baczko I, Jones L, Fercho M, Light PE. Inhibition of 65. Conte Camerino D, Tricarico D, Pierno S, et al. Taurine and cardiac voltage-gated sodium channels by grape polyphenols. Br J skeletal muscle disorders. Neurochem Res 2004;29:135–142.
59. Desaphy JF, Pierno S, De Luca A, Didonna P, Conte Camerino D.
66. Eguchi H, Tsujino A, Kaibara M, et al. Acetazolamide acts directly Different ability of clenbuterol and salbutamol to block sodium on the human skeletal muscle chloride channel. Muscle Nerve channels predicts their therapeutic use in muscle excitability dis- orders. Mol Pharmacol 2003;63:659 – 670.
67. Desaphy J-F, Rolland J-F, Valente EM, LoMonaco M, Conte 60. Bezzina CR, Tan HL. Pharmacological rescue of mutant ion chan- Camerino D. Functional alteration of ClC-1 channel mutants as- nels. Cardiovasc Res 2002;55:229 –232.
sociated with transient weakness in myotonia congenita [Abstract 61. Jentsch TJ, Poet M, Fuhrmann JC, Zdebik AA. Physiological 1291-Pos]. Biophys J 2007 [Epub in advance of print].
functions of CLC ClϪ channels gleaned from human genetic dis- 68. Bowery NG, Smart TG. GABA and glycine as neurotransmitters: ease and mouse models. Annu Rev Physiol 2005;67:779 – 807.
a brief history. Br J Pharmacol 2006;147:S109 –S119.
62. Pusch M, Liantonio A, De Luca A, Conte Camerino D. Pharma- 69. Cascio M. Modulating inhibitory ligand-gated ion channels. AAPS cology of CLC chloride channels and transporters. In: Pusch M, Neurotherapeutics, Vol. 4, No. 2, 2007

Source: http://www.trx2.jp/documents/Ion%20Channel%20Pharmacology.pdf