Basic mechanisms of antiepileptic drugs and their pharmacokinetic/pharmacodynamic interactions: an update


Basic mechanisms of antiepileptic drugs and their pharmacokinetic/pharmacodynamic interactions: an update
W3adys3aw Lasoñ1,2, Monika Dudra-Jastrzêbska3,5, Konrad Rejdak4, Stanis3aw J. Czuczwar3,5
􏰀Department of Experimental Neuroendocrinology, Polish Academy of Sciences, Smêtna 12, PL 31-343 Kraków, Poland
Department of Drug Management, Institute of Public Health, Jagiellonian University, Medical College, Grzegórzecka 20, PL 31-351 Kraków, Poland
!Department of Pathophysiology, “Department of Neurology, Medical University of Lublin, Jaczewskiego 8, PL 20-090 Lublin, Poland

Department of Physiopathology, Institute of Agricultural Medicine, Jaczewskiego 2, PL 20-950 Lublin, Poland Correspondence: W3adys3aw Lasoñ, e-mail: lason@if-pan.krakow.pl

Abstract:
This article aims to summarize the current views of AED action and the promising new targets for the pharmacotherapy of epilepsy. In the first section of this paper, a neurobiological basis of epilepsy treatment and brief pharmacological characteristics of classical and new AEDs will be presented. In the second part, the results of experimental studies that have combined AEDs with similar or dif- ferent mechanisms of action will be discussed.
Key words:
antiepileptic drugs, epilepsy, drug interactions, seizures
Abbreviations: AED(s) – antiepileptic drug(s), AMPA – a-amino- 3-hydroxy-5-methyl-4-isoxazolepropionate, CBZ – carbamazepine, ESM – ethosuximide, FBM – felbamate, GABA – g-aminob- utyrate, GBP – gabapentin, LCM – lacosamide, LEV – leveti- racetam, LTG – lamotrigine, NMDA – N-methyl-D-aspartate, OXC – oxcarbazepine, PB – phenobarbital, PGB – pregabalin, PHT – phenytoin, RTG – retigabine, TGB – tiagabine, TPM – topi- ramate, VGB – vigabatrin, VPA – valproate, ZNS – zonisamide
Introduction
The beginning of a rational pharmacotherapy for epi- lepsy dates back to the second half of nineteenth cen-
tury when bromides were introduced as fairly effi- cient, but toxic, anticonvulsant agents. The first syn- thetic antiepileptic drug (AED) was phenobarbital (PB), which efficiently combated tonic seizures, in- hibited partial seizures to a lesser extent, and had no effect in the absence epilepsy. Employment of electro- convulsive shock by Putnam and Merritt for the screening of potential anticonvulsants in late 1930s [82] led to the discovery of diphenylhydantoin (phenytoin; PHT), which is a non-sedative still widely used as an AED [67, 91]. The maximal electroconvul- sive shock test was the most valuable model for the discovery of new AEDs because substances that pre- vented hind limb extension in rodents in this test usu-
2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘
271

ally showed high efficacy in the treatment of partial and tonic-clonic seizures in the clinic. Seizures in- duced by the GABA receptor antagonist, pentylenete- trazole, are a very useful screening test for the identi- fication of AEDs, particularly drugs that are effective in the treatment of myoclonic epilepsies. Despite their long history and the implementation of more ad- vanced pharmacological models, both of these tests are still broadly used in preclinical studies. The chemical structure of the majority of AEDs marketed before 1965 was based on PB and comprised some derivatives of hydantoin and imides of succinic acid [67, 91]. From 1965 to 1990, several new AEDs with quite distinct chemical formulas appeared on the pharmaceutical market, such as benzodiazepines, imi- nostilbene (carbamazepine; CBZ) and carboxylic acid with a branched aliphatic chain, known as valproic acid. In the 1990s, pharmaceutical companies launched a battery of new generation AEDs represented by phenyltriazine (lamotrigine; LTG), a cyclic analog of GABA (gabapentin; GBP), an aminosulfonic deriva- tive of monosaccharide (topiramate; TPM), a deriva- tive of nipecotic acid (tiagabine; TGB) and a deriva- tive of pyrrolidine (levetiracetam; LEV) [67, 119]. According to physician expectation, an ideal AED should totally inhibit seizures without producing any undesired effects. Regretfully, all known AEDs pos- sess limited clinical efficacy and frequently produce undesired symptoms, which range from mild distur- bances of central nervous system (CNS) functions to fatal cases of bone marrow damage or liver insuffi- ciency [67]. Therefore, the physician has to select the appropriate drug or combination of drugs that will provide optimal suppression of epileptic attacks in an individual patient accompanied with acceptable levels of side effects. It is assumed that total control of epi- leptic attacks can be achieved in approximately 50% of patients, and a significant improvement can be ob- served in another 25% of patients. The ultimate suc- cess of epilepsy treatment depends on the type of epi- leptic attack, etiology and many other factors. To minimize toxic effects, treatment with a single drug is preferred. However, if the drug is not effective despite its proper therapeutic blood level, substitution with another drug rather than the administration of both drugs together is suggested. On the other hand, poly- therapy may be required, especially when more than one kind of epileptic attacks is diagnosed in the same patient [67, 91]. The measurement of drug plasma
blood levels facilitates the optimization of the phar- macotherapy of epilepsy during the first period of dose adjustment, in cases of unsuccessful treatment, the appearance of toxic symptoms, and during poly- therapy. It should be kept in mind that the clinical effi- cacy of some drugs is not always strictly correlated with their blood levels, and therefore, the recom- mended drug concentrations should be regarded as a suggestion only. The final shape of the therapeutic procedure will depend on the clinical estimation of drug efficiency and its toxic effects [113, 121].
Neurobiological basis of AED action
The molecular targets for classic, new generation and potential AEDs are voltage-dependent sodium, cal- cium, potassium channels, h-channels, GABA) recep- tors, excitatory amino acid receptors, some enzymes and synaptic proteins [90, 92, 93]. The GABA mimetic effects and the blockade of voltage dependent sodium channels dominate the other mechanisms of AEDs [7, 13, 15, 68, 80, 90, 119]. GABA is the most important inhibitory transmitter in the CNS and even slight defi- ciencies in GABAergic transmission lead to hyper- excitability and pathological neuronal discharges [63].
The balance between synaptic excitation and inhi- bition depends mainly on the correct anatomical and functional organization of the neuronal net, which is composed of glutamatergic neurons and GABAergic interneurons [63, 119]. Although only 10–20% of neurons synthesize GABA in cortical structures of mammals, this amino acid efficiently controls the level of activity of all cortical neurons. The profi- ciency of synaptic inhibition in local neuronal loops depends on the divergence of innervation. The inner- vation of single GABAergic interneuron may inhibit several thousands of glutamatergic cells, or many GABA interneurons may convergence on the same glutamatergic cell. In contrast, GABA interneurons are stimulated by glutamatergic neurons and inhibited by other interneurons. GABA is the product of gluta- mate decarboxylation, and after its release to the syn- aptic cleft, it is taken up by glia and neuronal cells, where it is metabolized by GABA aminotransferase to succinic semialdehyde. Four transporters (GAT1-4) participate in GABA uptake. GABA exerts its phar- macological action via membrane GABA), GABA*
272 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘

and GABA+ receptors. GABA) and GABA+ recep- tors are ionotropic receptors that form channels per- meable to Cl􏰇􏰇 ions in the neuronal membrane, whereas GABA* receptors are metabotropic. GABA) recep- tors play a role in the formation of fast inhibitory postsynaptic potentials and, therefore, play a key role in controlling seizure phenomena. GABA) receptor agonists enhance chloride channel conductance to de- crease resistance and increase hyperpolarization of the neuronal membrane [29]. As a consequence, the stimulation of GABA) receptors usually increases seizure threshold in the presence of epilepsy, but not in the absence of epilepsy, and inhibits the spreading of seizure activity [15, 68, 119]. Conversely, GABA) receptor antagonists, such as bicuculline, picrotoxin or pentylenetetrazole, are some of the most potent seizure-inducing agents. Molecular cloning has re- vealed that the GABA) receptor forms a pentameric protein complex that is composed of two a subunits, two b subunits that contain the GABA binding site and one g or d subunit. The subunit composition (i.e., the receptor configuration) determines the affinity and pharmacodynamic effectiveness of GABA) receptor agonists, modulators and antagonists. An enhancement of GABAergic inhibitory transmission is responsible for the antiepileptic effects of drugs that directly bind and activate GABA) receptors or influence GABA synthesis, transport and metabolism [15, 26, 68, 119].
Glutamate is the main excitatory neurotransmitter in the CNS. It activates ionotropic receptors that are named after specific agonists, AMPA (a-amino-3 hydroxy-5-methyl-4-isoxazolepropionate), NMDA (N- methyl-D-aspartate) and kainate receptors, and meta- botropic receptors that act via G protein influence on various second messenger systems and ion channel activity. NMDA receptors consist of NR1 subunits combined with one or more NR2 (A–D) subunits, which form channels that are permeable to sodium and calcium ions. The activity of NMDA receptors is regulated via a strychnine-insensitive glycine-binding site and other modulatory sites, such as polyamine, Zn 􏰊, H􏰊. At resting membrane potentials, the pore of this receptor is blocked by magnesium ions, which are removed after membrane depolarization. The role of NMDA receptors in experimental epileptogenesis, neuroplasticity, seizures and excitotoxicity has been firmly established. Antagonists of NMDA receptors, such as dizocilpine or ketamine, inhibit seizures in- duced by pentylenetetrazole, pilocarpine, maximal electroshock or sensory stimulation. Furthermore,
Mechanisms of antiepileptic drug actions
9􏰋=@OI􏰋=M 􏰌=I􏰂􏰍 AJ =􏰃􏰎
they delay the development of amygdala kindling but have a weaker effect on fully developed seizures in this model. Unfortunately, both competitive and non- competitive NMDA receptor antagonists show serious undesired effects, such as psychomotor, memory and cognitive disturbances, and psychotomimetic-like ef- fects in experimental animals and in initial clinical studies. Moreover, these undesired symptoms are en- hanced in experimental models of epilepsy [2, 48, 49, 68]. Therefore, some allosteric modulators of NMDA receptors, especially modulators that interact with the strychnine-insensitive glycine-binding site and the polyamine-binding site, are more promising as poten- tial AEDs [39, 68]. Antagonists of the glycine binding site and the polyamine-binding site show affinity for NR1/NR2A subunits and NR1A/NR2B complexes, respectively. The partial agonist of the glycine bind- ing site, D-cycloserine, exerts anticonvulsant activity most likely via the desensitization of NMDA recep- tors. This compound also augments the seizure- suppressing effects of some AEDs and, in low doses, has a positive influence on memory processes. The beneficial effects in experimental models of seizures have been observed after the concomitant administra- tion of glycine- and polyamine-binding site antagonists [39]. Lacosamide (LCM), an antagonist of the glycine- binding site on NMDA receptors, has been registered as an AED in 2008 [5, 25, 33, 106]. Another example is felbamate, which shows inhibitory action on voltage-dependent sodium channels and antagonistic activity toward the glycine-binding site on NMDA re- ceptors [9, 68].
Glutamatergic AMPA receptors play the main role in the mediation of excitatory synaptic conductance in the central nervous system. AMPA receptor complexes consist of various combinations of four homologous GluR1-GluR4 subunits and function as cation channels that are permeable for Na􏰊 and K􏰊 ions and, in some configurations, also to calcium ions. Complexes that contain the GluR2 subunit show low permeability to calcium ions. In the presence of an agonist, the AMPA receptor undergoes desensitization. However, inhibi- tors of desensitization, such as cyclothiazide, are posi- tive allosteric modulators of this receptor [24]. Nega- tive allosteric modulators of AMPA receptors are 2,3- benzodiazepines. NMDA receptor-dependent calcium ion influx and the subsequent activation of protein ki- nases lead to the phosphorylation of AMPA receptor and to an increase in its activity. This process may be responsible for the pathological hyperactivity of gluta-
2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ 273

matergic conductance and epileptoidal neuronal dis- charges. Phosphatases, including calcineurin, dephos- phorylate AMPA receptors and decrease their activity. Antagonists of AMPA and kainate receptors inhibit seizures in chemical (e.g., pentylenetetrazole, bicu- culline) and genetic models of seizures [38]. They also potentiate the anticonvulsant effects of classical AEDs, such as PB or valproate (VPA) [13]. These agents affect motor coordination and memory pro- cesses to a lesser extent than NMDA receptor antago- nists. Among the already registered AEDs that pos- sess the ability to block AMPA receptors, one should also mention barbiturates and TPM.
The non-competitive AMPA receptor antagonist, talampanel, shows a broad spectrum of anticonvulsant activity and is currently being tested in clinical trials [35, 36]. Kainate receptors are less abundant in the brain than AMPA receptors, but their broad distribu- tion suggests that they are present in a majority of neurons. Kainate receptors consist of GluR5-GluR7 and KA1-KA2 subunits, and their activity can be af- fected by the same ligands that bind to AMPA recep- tors, although with various preferences. GluR5 and GluR6 subunits form homomeric channels, but the other subunits are parts of heteromeric combinations with GluR5 and GluR6 subunits. The activation of postsynaptic kainate receptors leads to long-term neu- ronal depolarization and augmented intracellular Ca 􏰊 influx, which may play a role in changes in synaptic plasticity and in the seizurogenic and neurotoxic ef- fects of kainate. The potent epileptogenic and neurotoxic effects of kainate may be also connected to the presynap- tic inhibition of GABA release, because this agent de- creases GABA) and GABA* receptor-dependent inhibi- tory synaptic potentials in the CA1 hippocampal field. Presynaptic kainate receptors can be also involved during an enhancement of glutamate release, as the activation of these receptors increases calcium ion concentration in synaptosomes [39, 45, 68, 90]. The above-mentioned facts suggest that modulators of kai- nate receptor activity should not be overlooked as po- tential AEDs [13, 65, 90]. In addition to ionotropic excitatory amino acid receptors, there are eight meta- botropic glutamate receptor (mGluR) subtypes that are classified into three groups based on amino acid sequence similarity, agonist pharmacology and the signal transduction pathways to which they couple. The activation of group I mGluRs (i.e., mGlu􏰀 and mGlu#) stimulates inositol phosphate metabolism and the mobilization of intracellular Ca 􏰊. Agonists of
group II (i.e., mGlu and mGlu!) and group III (i.e., mGlu” and mGlu$&) mGluRs inhibit adenylyl cyclase. A fourth group of mGluRs that couple to phospholi- pase D has been also reported. Group I mGluRs are localized postsynaptically on both glutamatergic cor- tical and hippocampal neurons and on GABAergic in- terneurons. Their activation causes the phosphoryla- tion and inactivation of many types of potassium channels, which results in depolarization and neuronal hyperexcitability. Group II and III mGluRs are inhibi- tory presynaptic autoreceptors on glutamatergic neu- ronal endings or inhibitory presynaptic heterorecep- tors that are localized on some GABAergic neuronal terminals. Agonists of group I mGluRs evoke seizures in experimental animals and elongate the duration of both interictal and ictal discharges in hippocampal slices; their antagonists act in the opposite way. In- deed, some antagonists of mGlu􏰀 or mGlu# receptors and several agonists that act on group III mGluRs show anticonvulsant activity in animal models of gen- eral epilepsy or in the absence epilepsy. Therefore, these agents may be regarded as potential AEDs [40, 70, 90]. Furthermore, the agonist of group II mGluRs, LY354740, markedly enhances the antiepileptic activ- ity of diazepam [40]. After its release, the uptake of glutamate within the human brain is mediated by 5 subtypes of high-affinity, sodium-dependent excita- tory amino acid transporters (EEATs) that are local- ized in the cell membranes of astrocytes (EAAT1, EAAT2) and neurons (EAAT3-5). Excessive activity of the glutamatergic system plays an essential role in the pathomechanism of epileptic attacks of various etiologies [13, 68]. Attenuation of excitatory trans- mission to inhibit ictal neuronal activity can be achieved through a decrease in glutamate synthesis, the modula- tion of presynaptic receptors or the calcium-dependent release of the glutamate, an increase in glutamate uptake and a decrease of postsynaptic glutamate receptor activi- ties [13, 68]. Several AEDs inhibit glutamatergic system activity; however, these effects seem to play a minor role in their mechanisms of action. Voltage-gated sodium channels (VDSCs) are pro- teins that are comprised of four repeated domains of six transmembrane segments, which form sodium ion-selective pores. The brain contains subtypes I, II, IIa, III and VI of sodium channels, which are sensitive to tetrodotoxin, and their conductance ranges from 2.5 to 25 pS. VDSCs are responsible for the generation of action potentials and are the main targets for many AEDs, including PHT, CBZ and LTG [68, 69, 90, 91, 274 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀

117, 119]. These drugs block sodium channels during high frequency discharges, but in therapeutic concen- trations, they have no effect on physiological synaptic transmission. They show an ability to inhibit both re- petitive action potentials in the epileptic focus and the spreading of epileptic discharges. Voltage-dependent calcium channels (VDCCs) are divided into several subtypes, L, N, P/Q, T and R, according to their elec- trophysiological characteristics. VDCCs are formed of an a1 subunit, which is the sensor of the potential, and a2, b, d and g subunits which regulate the kinet- ics and amplitude of calcium currents. Blockade of N or P/Q channels inhibits the presynaptic release of ex- citatory amino acids. However, a potential role for these channels in AED action has not been elucidated. The low-voltage calcium channel, T, plays an essen- tial role in the mechanism of the thalamo-cortical os- cillatory activity and the generation of spike-wave discharges; it also plays a pathological role in the ab- sence epilepsy [90, 91, 119].
Potassium channels are classified into the inward rec- tifier (KE4) superfamily, which comprises the recep- tor-coupled ATP-sensitive and voltage-dependent chan- nels, and the shaker-related superfamily which includes Ca 􏰊-activated potassium channels. The muscarine- sensitive Kv7 (KCNQ) type potassium channels are responsible for a slowly activating current whose threshold is near resting potential and play an essen- tial role in neuronal repolarisation and hyperpolarisa- tion that follows paroxysmal depolarization shifts [3, 61, 120–123]. It has been well established that Kv7/ KCNQ/M potassium channels mediate the M-current which inhibits repetitive firing and burst generation. There are four neuronal Kv7 subunits (Kv7.2– Kv7.5) which serve as the target for some recently approved AED retigabine and currently being tested in early clinical phases ICA-105665 [3, 120, 121, 123]. Some of these channels, such as the muscarine-sensitive KCNQ2, are important regulators of neuronal excit- ability. An involvement of potassium channels in the mechanism of AEDs, with the exception of retiga- bine, remains largely unknown [119]. Mutations of membrane ion channels and receptors disturb bioelec- tric neuronal activity and are an important causative factor in the pathogenesis of epilepsy [1, 122]. In- deed, some genes whose mutations lead to distur- bances in brain development, metabolism, neurode- generation and abnormal neuronal activity are often associated with certain inherited types of epilepsies. Channelopathies due to the mutation of subunits of
Mechanisms of antiepileptic drug actions
9􏰋=@OI􏰋=M 􏰌=I􏰂􏰍 AJ =􏰃􏰎
voltage-dependent sodium, potassium, calcium and chloride neuronal channels deserve special attention [44]. A single amino acid mutation can significantly alter ion channel kinetic parameters and the affinity of its ligands. Thus substitution of lysine with methio- nine in the g2 subunit of the GABA) receptor de- creases inhibitory postsynaptic potentials. On the other hand, the substitution of alanine with asparagine in the a1 subunit lowers GABA) receptor sensitivity, elongates the time of its desensitization and shortens the opening time for chloride channels. Changes in the same locus can be reflected in the heterogenous phenotype, whereas, the same epileptic syndrome may be caused by various genetic defects. The well know genetically determined epilepsy syndromes in- clude benign familial neonatal or infantile convul- sions (BFNC or BFIC), autosomal dominant noctur- nal frontal lobe epilepsy (ADNFLE), absence epi- lepsy, myoclonic epilepsies, generalized idiopathic epilepsies and febrile seizures. Mutation in neuronal potassium channel subunits (KCNQ2 or KCNQ3) are associated with benign familial neonatal convulsions (BFNC), whereas mutation of voltage-gated sodium channels are the substrate for generalized epilepsy with febrile seizures plus [1]. Furthermore, some other ge- netic disorders e.g., cereidolipofuscinosis, galactosialo- sidosis or gangliosidosis GM1, have epileptic seizures as one of their symptoms [1, 102].
According to Sills [100], the mechanistic classes of current AEDs contain fast Na􏰊 channel blockers (e.g., PHT, CBZ, LTG, oxcarbazepine, rufinamide, and esli- carbazepine), slow Na􏰊 channel blockers (e.g., laco- samide), high-voltage Ca 􏰊 channel blockers (e.g., GBP, pregabalin), low-voltage Ca 􏰊 channel blockers (e.g., ethosuximide), GABA) receptor activators (e.g., PB, benzodiazepines, stiripentol), GABA transami- nase inhibitors (e.g., vigabatrin; VGB), GABA uptake inhibitors (e.g., TGB), SV2A ligands (e.g., LEV), and multiple mechanism drugs (e.g., VPA, felbamate, TPM, and zonisamide (ZNS)). In fact, the majority of AEDs show multiple mechanisms, although the inter- ference of a drug with one of the above-mentioned single targets (i.e., ion channel, membrane receptor, enzyme) is underlined. Therefore, one can propose other classifications of AEDs that are based on the primary, secondary and tertiary mechanisms of their action. The best-recognized primary mechanisms, such as the blockade of Na􏰊 or Ca 􏰊 channels and the modulation of the GABAergic system, are important for the prevention (through an enhancement of seizure
2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ 275

threshold) and suppression of ongoing abnormal neu- ronal activity. The secondary action may involve in- teractions with additional protein targets, such as car- bon anhydrase, or interference with endogenous anti- convulsant systems (e.g., adenosine, neurosteroids, neuropeptides, and antioxidant systems). These mecha- nisms may act in a synergistic or an additive way to po- tentiate the primary mechanisms of action. The tertiary mechanism of AEDs, including the long-term effects on neuroplastic processes (e.g., neurotrophin, cytoki- nes, hormone synthesis and release) and genetic and epigenetic effects, could be involved in the delay of morphological changes, neuroplasticity and epilepto- genesis (i.e., adaptive changes in the neuronal, endo- crine and immune systems), which thus changes the courses of evolution and involution of epilepsy [4, 48].
Although the primary action is the most useful and widely accepted basis for mechanistic classification and the designing of AEDs, the secondary and tertiary mechanisms should not be disregarded in the prepara- tion of the “personal” characteristics of a given AED. It remains an open question whether such a full neuro- chemical characteristic of an AED is more helpful in predicting the spectrum of its clinical efficacy and un- desirable effect profile than the simple, but “trun- cated”, classification system compromised solely of the primary mechanism of AED action. Furthermore, the thorough knowledge of the mechanisms of AED actions may provide a better rationale for the existing opinion that the combination of drugs with different mechanisms is preferable in the polytherapy of epi- lepsy. According to Gil-Nagel [31], the mode of ac- tion may predict the spectrum of efficacy and side- effect profile of AEDs to some extent. Therefore, se- lective sodium channel blockers (e.g., PHT, CBZ, OXC, ESL) are only effective in partial and secondary generalized tonic-clonic seizures, but drugs with a single GABAergic mechanism are likely to have a narrow spectrum of efficacy in partial seizures only. He also stressed that selective calcium channel block- ers (e.g., ESM, GBP, PGB) may not be effective in pri- mary generalized tonic-clonic seizures. Although AEDs with multiple mechanisms (e.g., TPM, LEV, LTG, VPA, ZNS) are likely to be broad spectrum, they usually have a single predominant mechanism at thera- peutic concentrations. The author concluded that knowledge of a mechanism of action may be useful for the selection of broad spectrum AEDs in unknown types of epilepsy and in guiding add-on and substitu- tion decisions.
In the first part of this paper, a brief pharmacological characterization of classic and new AEDs based on the dominant mechanism of action, along with their other neurochemical properties, will be presented. In the sec- ond part, the results of experimental studies of the com- bination of AEDs with similar or different mechanisms of action will be discussed.
AEDs that enhance GABAergic system activity
Barbiturates
PB was the first efficient synthetic AED. It possesses relatively low toxicity, and it is cheap and still widely used in clinical practice. Although the majority of barbi- turates prevent seizures, only some of them, including PB, can be used for the treatment of epilepsy because they show maximal seizure-suppressing activity in doses lower than doses that induce somnolence. PB is active, although not very selective, in almost all preclinical tests that are used for screening of potential AEDs. It inhibits tonic hind limb extension in the maximal electroconvul- sive shock test, pentylenetetrazole-induced clonic sei- zures and amygdala kindling [67, 91].
The mechanism of action of PB is most likely connected with an enhancement of synaptic inhibition via a positive modulatory effect of this drug on GABA) receptors.
The possible involvement of GABA) receptors in the antiepileptic properties of barbiturates was ini- tially controversial because the sleep-inducing pento- barbital had a stronger effect on GABA) receptor ac- tivity than the recognized AED, PB. Further studies have provided evidence that barbiturates potentiate pre- and postsynaptic activity by increasing the prob- ability and time of GABA) receptor-dependent chloride channel openings [116]. Intracellular recordings of mouse cortical and spinal cord neurons has shown that in therapeutic concentrations, PB enhances the cellular response to iontophoretically administered GABA. Furthermore, patch-clamp analysis of single ion channels provided evidence that PB elongated the time but had no effect on the frequency, of GABA) receptor-dependent neuronal discharges [116]. The electrophysiological data correlate well with results of radioreceptor assays that show that barbiturates bind to the ionophore part of the GABA) receptor complex
276 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘

to enhance GABA affinity to this receptor and inhibit its dissociation. Functional binding of barbiturates can occur even when the configuration of the GABA) receptor complex is composed of the same subunits, such as a or b (1–3). In supratherapeutic concentra- tions, PB restricts the sustained repetitive neuronal discharges, and this mechanism may be of signifi- cance in extinguishing status epilepticus with high doses of PB. In addition to the positive modulatory ef- fect on GABA) receptors, barbiturates can block VDCCs, intensify voltage-dependent potassium chan- nel activity and inhibit AMPA-receptor-related gluta- matergic transmission [20, 101, 119].
The main indications for barbiturates are general- ized tonic-clonic seizures, partial seizures and drug- resistant status epilepticus [67].
Mephobarbital
This is an N-methyl derivative of PB, which undergoes N-demethylation in the liver reticular endoplasmic sys- tem. The pharmacological activity of mephobarbital during long-term therapy results from an accumulation of PB. The indications are similar to PB [26].
Primidone
This is desoxybarbiturate, which is transformed to an amide of phenylethylmalonic acid and to PB. Primi- done shows similar efficacy to PB in the maximal electroconvulsive shock, but it is less active in the pentylenetetrazole test. Primidone is indicated in gen- eralized tonic-clonic seizures, partial seizures, psy- chomotor and myoclonic seizures [67, 91].
Benzodiazepines
Benzodiazepines are far more effective in inhibiting pentylenetetrazole-induced clonic seizures than sei- zures induced by the maximal electroconvulsive shock test. For example, clonazepam is an extremely potent anticonvulsant in the pentylenetetrazole test, but it is almost inactive in the maximal electroconvul- sive shock test. Benzodiazepines prevent seizure spreading in the kindling model and in the model of generalized seizures induced by electrical stimulation of the amygdala, but they do not suppress the abnor- mal neuronal discharges in the site of stimulation.
The antiepileptic activity of benzodiazepines are mainly connected with the positive allosteric modula-
Mechanisms of antiepileptic drug actions
9􏰋=@OI􏰋=M 􏰌=I􏰂􏰍 AJ =􏰃􏰎
tion of GABA) receptors and an intensification of synaptic inhibition [91, 116]. Electrophysiological measurements demonstrated that therapeutic concen- trations of benzodiazepines increase the frequency but not the conductance or the time of opening of GABA)-related chloride channels. Benzodiazepines enhance the effects of GABA but not the effects that are produced by other amino acids that increase the conductance of chloride channels, including glycine, b-alanine and taurine. The selective enhancement of the effect of GABA is reflected by the shift of the dose-response curve to the left without any alteration in the maximal effect. This effect means an increased sensitivity of neurons to GABA in the presence of the same number of accessible chloride channels [91]. Subunit composition is of vital importance for the binding and pharmacological effects of benzodi- azepines, and the co-expression of a, b and g subunits is required. It has been postulated that a (1–6) subunits participate in the modulation of GABA) receptor func- tion by benzodiazepines, and the g subunits (1–3) de- termine benzodiazepine binding. The d subunit is found in GABA) receptor complexes, which are in- sensitive to benzodiazepines. Repeated administration of benzodiazepines leads to the development of toler- ance via a decrease in the ability to enhance the GABA)-dependent chloride current and a reduction in the number of binding sites [101]. Besides the in- teraction with the GABA) receptor, benzodiazepines inhibit adenosine uptake and block voltage-dependent sodium and calcium channels. The latter mechanism is confirmed by the observation that in the high con- centrations that are used in the treatment of status epi- lepticus, diazepam and some other benzodiazepines inhibit the sustained high-frequency neuronal dis- charges in manner similar to PHT, CBZ and VPA.
Benzodiazepines are recommended for the treat- ment of pycnoleptic states of unconsciousness, myoclonic-astatic, propulsive attacks and in status epilepticus [67, 89].
Tiagabine (TGB)
TGB, which is a derivative of nipecotic acid, sup- presses seizures in the maximal electroconvulsive shock test and partial and tonic-clonic seizures in the model of amygdala kindling [18]. TGB is a potent and selective inhibitor of the GABA GAT-1 transporter, and it inhibits the uptake of this amino acid from the synaptic cleft to glia and neuronal cells [41]. In accor-
2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ 277

dance with this neurochemical mechanism, electro- physiological data indicate that TGB elongates the duration of inhibitory synaptic currents in CA1 hippo- campal neurons. TGB has no effect on biogenic amine uptake and does not bind to the majority of neuro- transmitter receptors, although it shows a slight affin- ity for histamine and benzodiazepine receptors.
This drug is indicated as an add-on therapy in drug-resistant partial and secondary generalized epi- leptic fits [6, 21, 78, 94].
Vigabatrin
This is analog of GABA (g-vinyl-GABA) and shows potent anticonvulsive effects in amygdala kindling and in models of audiogenic and chemical seizures evoked by pentylenetetrazole, picrotoxin and strych- nine [67, 91]. The effectiveness of VGB in the maxi- mal electroconvulsive shock test is controversial. A paradoxical pro-convulsive effect of VGB in the electroconvulsive test was also observed, but it might be due to the withdrawal of this drug.
VGB is an irreversible inhibitor of GABA transami- nase (GABA-T) and, therefore, evokes a long-term in- crease of GABA concentrations in brain tissue, an ef- fect that correlates well with the AED action. VGB in- hibits both neuronal and glial GABA-T; however, it shows a stronger affinity for the enzymes that are pres- ent in neurons [41]. Depolarizing stimuli, such as an extracellular increase in potassium ions during sei- zures, release the accumulated GABA from presynap- tic terminals to intensify GABAergic transmission. In addition to inhibiting GABA-T, VGB decreases gluta- mate and aspartate content in brain tissue [41].
This AED is recommended in the treatment of pro- pulsive fits of unconsciousness (monotherapy), partial and complex epilepsy in adults and partial epilepsy and in Lennox-Gastaut syndrome in children as an ad- junctive drug [21, 32, 117].
AEDs affecting the activity of cation channels (Na+, Ca2+, K+)
Phenytoin (PHT)
PHT (diphenylhydantoin) effectively inhibits all kinds of partial and generalized tonic-clonic seizures but fails to counteract generalized absences. The drug dis-
plays anticonvulsant activity without any prominent depression of the CNS. In the toxic range, there might be central excitation, and lethal doses are accompa- nied by decerebrate rigidity. The most characteristic activity of PHT is its ability to affect seizures induced by maximal electroshock in experimental animals. The characteristic tonic phase of seizure activity may be completely blocked, but the duration of clonic sei- zures may be extended [67, 91, 115]. AEDs that are effective against human generalized tonic-clonic con- vulsions share this pattern of activity. Furthermore, PHT does not exert any protective action against pentylenetetrazole-induced clonic seizures. In in vitro studies performed on mouse spinal cord neurons, this AED limits high frequency repetitive firing. This par- ticular effect correlates with a slower reactivation of VDSCs, and interestingly, the higher the frequency of channel openings are, the better the inhibitory effect of PHT on this parameter (i.e., use-dependence). It is noteworthy that the block of high frequency repetitive firing occurs within the therapeutic range of PHT in the spinal cord fluid, which is highly correlated with the free plasma concentration of this AED. This con- centration is responsible for the selective blockade of VDSCs, and no effects of PHT on glutamate or GABA neuronal activity are observed [91, 119]. With 5–10 fold higher concentrations, PHT may exert toxic effects due to a number of mechanisms, including a depression of basal neuronal activity or an enhance- ment of GABA-mediated events. Apart from its basic mechanism of action, PHT may lower the conduc- tance of calcium channels and limit calmodulin- dependent phosphorylations in the CNS [115, 119].
This AED is indicated in partial, generalized tonic-clonic, and psychomotor seizures [77].
Carbamazepine (CBZ)
CBZ, a dibenzoazepine derivative, shares many phar- macological effects with PHT, although there are also considerable differences. Apart from the antiepileptic activity, CBZ is effective in manic-depressive patients resistant to lithium carbonate. Similar to PHT, this AED limits the high frequency-sustained repetitive firing of cortical or spinal neurons that results from its preferential binding to the inactivated forms of VDSCs. This particular effect leads to slower transi- tions to the activated forms [51]. In the therapeutic concentration range, CBZ does not affect GABA- or glutamate-mediated neuronal events [34]. 10,11-
278 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘

Epoxycarbamazepine, a metabolite of CBZ, exerts an identical profile of activity, and this may indicate that the metabolite participates in the final effects of the parent drug [66]. In addition, CBZ reduces neuronal calcium influx. In contrast to PHT, CBZ is a ligand of A􏰀 and A adenosine receptors and displays some af- finity for benzodiazepine receptors. CBZ is indicated in the treatment of generalized tonic-clonic and sim- ple or complex partial seizures [85].
Oxcarbazepine (OXC)
This AED is a keto analog of CBZ and, undergoes a rapid metabolic transition to the active 10- monohydroxy derivative, which is subsequently glu- curonidated and eliminated via the kidneys. OXC possesses similar mechanisms of action to CBZ, but it is a weaker activator of microsomal (cytochrome P-450) enzymes [83]. However, OXC may significantly acti- vate CYP3A and reduce the effective concentration of oral contraceptives [84].
Its main indication is in the treatment of partial sei- zures [19].
Lamotrigine (LTG)
LTG [6-(2,3-Dichlorophenyl)-1,2,4-triazine-3,5-diami- ne] inhibits tonic seizures induced by maximal elec- troshock and partial and secondarily generalized con- vulsions in the kindling model. However, it is ineffec- tive against pentylenetetrazole-induced clonic seizures [17, 79]. Similarly to PHT and CBZ, LTG preferen- tially binds to the slowly inactivating conformation of VDSCs, which is encountered during prolonged depo- larization or epileptic discharges. This mechanism of action may be associated with the ability of LTG to control partial secondarily generalized seizures. LTG has a broader spectrum of anticonvulsant activity compared to PHT and CBZ, which may suggest the existence of other mechanisms of action. For exam- ple, LTG reduces veratridine-induced synaptic release of glutamate or aspartate at micromolar concentra- tions through the blockade of presynaptic sodium channels. This AED is an effective antagonist of AMPA glutamate receptors [12, 42, 55]. Considering the novel mechanism of LTG, it is emphasized that this drug can block dendritic action potentials and en- hance currents that are carried by h channels [57, 91]. The main indications for LTG include partial seizures
Mechanisms of antiepileptic drug actions
9􏰋=@OI􏰋=M 􏰌=I􏰂􏰍 AJ =􏰃􏰎
with or without secondarily generalized tonic-clonic seizures and Lennox-Gastaut syndrome [22, 27].
Topiramate (TPM)
TPM is a monosaccharide derivative with a broad spectrum of anticonvulsant activity [99, 118]. This drug is effective in a variety of animal models of epi- lepsy, including maximal electroshock, pentylenete- trazole-induced and kindled partial and secondarily generalized seizures [13]. TPM reduces voltage- dependent sodium currents in cerebellar granulae cells, and this action is similar to PHT regarding the inactivation of sodium channels. In addition, TPM ac- tivates a hyperpolarizing potassium current, enhances GABA-mediated actions, and diminishes glutamate- induced excitation via AMPA receptors. Interestingly, this antiepileptic interacts with the GABA) receptor complex at a site different from either barbiturate- or benzodiazepine-binding sites. TPM is also a weak in- hibitor of carbonic anhydrase [13, 91].
This AED is recommended against many types of epi- leptic seizures, including drug-resistant convulsions [67].
Ethosuximide (ESM)
ESM is an imide of succinic acid that is effective against pentylenetetrazole-induced clonic convulsions but not against tonic seizures induced by maximal electroshock. This drug is also not effective against kindled seizures. This pharmacological profile is in good correlation with the clinical activity of ESM against generalized ab- sences. Its mechanism of action is related to the reduc- tion of the low threshold T calcium current in the tha- lamic neurons. This current is apparently involved in the generation of rhythmic 3 Hz spike-wave discharges that are typical of absence seizures [12]. At therapeutic con- centrations, ESM inhibits T calcium currents in rat ven- tral basal or guinea pig thalamic neurons but does not af- fect the T calcium channel voltage-dependent inactiva- tion kinetics, repetitive neuronal firing, or the enhance- ment of GABA- mediated events [91].
This AED is effective against generalized absences [67].
Valproate (VPA)
Contrary to PHT and ESM, VPA [di-n-propylacetic acid] is an effective inhibitor of seizure activity in a variety of experimental animal models. VPA blocks the tonic hind limb extension in the maximal electro-
2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ 279

shock test and displays a protective activity against kindled seizures and pentylenetetrazole-induced clonic convulsions at doses that are lower than toxic concen- trations [67]. Its mechanism of action is still a matter of controversy. At a therapeutic concentration, VPA reduces high frequency repetitive firing in mouse cor- tical or spinal neurons [47]. This particular activity re- sembles PHT or CBZ to a considerable degree and seems to be related to the longer period of time that the sodium channel needs to recover from the inacti- vation state. However, VPA does not modify neuronal responses to iontophoretically applied GABA. This AED is responsible for the moderate inhibition of T calcium channel activity. An additional mechanism is one that is involved in its effect on GABA metabolism. Brain GABA concentrations are elevated by VPA and in in vitro studies, VPA stimulates the GABA synthe- sizing enzyme, glutamate decarboxylase, and inhibits GABA metabolizing enzymes, GABA transaminase, succinic semialdehyde dehydrogenase, and aldehyde reductase.
The drug also reduces the activity of voltage- dependent T-type calcium channels. There has been no proof for a correlation between the increase in brain GABA and the anticonvulsant activity of VPA [67, 91].
This AED may be used for the therapy of absences, myoclonic, partial, and tonic-clonic seizures [67].
Retigabine
RTG [N-(2-amino-4-(4-fluorobenzylamino)-phenyl) car- bamic acid ethyl ester]. This drug shows anticonvul- sant properties in a variety of animal models, includ- ing maximal electroconvulsive shock, amygdala kin- dling, pentylenetetrazole-, picrotoxin-, and NMDA- induced seizures and in genetic models of epilepsy [3, 17]. The anticonvulsant action of this drug results mainly from the opening of voltage-gated KCNQ2/3 potassium channels, which are abundantly expressed in the brain, are associated with seizures and appear to control neuronal excitability via the M-current [89]. It has been found that RTG activates neuronal KCNQ- type potassium channels by inducing a large hyperpo- larizing shift of steady-state activation. Importantly, re- tigabine activates neuronal Kv7.2-Kv7.5 (KCNQ2- KCNQ5) potassium channel subunits but not on car-
diac Kv 7.1 (KCNQ1) channel [122–124]. Molecular studies demonstrated that RTG binds to hydrophobic pocket formed upon channel opening between the cyto- plasmic parts S5 and S6 involving Trp236 and the chan- nel’s gate, which account for the marked shift in voltage-dependent activation [122, 123].
In higher concentrations, RTG potentiates GABA- induced currents in rat cortical neurons. This drug also inhibits 4-aminopyridine-induced glutamate re- lease and the de novo synthesis of GABA in the hip- pocampus [71]. This drug has been recently registered as adjunctive treatment of partial onset seizures with or without secondary generalization in adults aged 18 years and above with epilepsy [10, 72, 79].
Zonisamide (ZNS)
ZNS is an aromatic sulfonamide derivative that shows anticonvulsant activity in the maximal electroshock test and the kindling model of epilepsy [43]. This drug effectively inhibits partial or secondarily gener- alized convulsions, but it is ineffective against minimal pentylenetetrazole-induced seizure activity, which is a model of myoclonic seizures [56, 72]. ZNS is an in- hibitor of T calcium channels and, similar to PHT and CBZ, it reduces high frequency repetitive firing in spinal cord neurons, which is probably dependent on the longer inactivation time of VDSCs [95, 107].
Its main recommendation is drug-resistant partial seizures in the form of an adjuvant drug [97, 107, 125].
Antagonists of receptors for excitatory amino acids
Felbamate (FBM)
FBM is a dicarbamate that is active against both maximal electroshock and pentylenetetrazole-induced convulsions [108]. In vitro studies have documented that FBM is an NMDA receptor antagonist at thera- peutic concentrations, and moreover, it potentiates GABA-mediated events in hippocampal neurons. A simultaneous reduction and enhancement of excita- tory and inhibitory neurotransmission, respectively, may play a considerable role in FBM’s broad spec- trum of anticonvulsive activity. Due to a risk of aplas- tic anemia, this AED has limited recommendations
280 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘

and may be used practically as an adjuvant in Lennox-Gastaut syndrome [76, 77].
Lacosamide (LCM)
LCM, (R)-2-acetamido-N-benzyl-3-methoxypropion- amide, demonstrates high anticonvulsant activity in a broad range of animal models of partial onset and pharmacoresistant seizures, generalized tonic-clonic seizures and status epilepticus [104]. LCM prevents psychomotor seizures in the 6 Hz model and sound- induced seizures in an audiogenic model. The drug is also effective in the maximal electroshock test and the amygdala and hippocampal kindling models. How- ever, the effectiveness of LCM was less pronounced against clonic seizures induced by pentylenetetrazole, bicuculline or picrotoxin in rodents [5]. The anticon- vulsant mechanism in humans is not fully understood. LCM neither influences voltage-activated potassium channels nor VDCCs. The drug selectively enhances sodium channel slow inactivation with no influence on fast inactivation [5].
According to clinical trials that have evaluated the efficacy and safety of LCM, the drug is a safe and an effective agent for the adjunctive treatment of refrac- tory partial-onset seizure [25]. LCM is still being evaluated in clinical trials as a treatment for chronic refractory neuropathy and pain in diabetic neuropathy. LCM is also considered as a therapy for migraine pro- phylaxis and fibromyalgia syndrome [33, 67].
Talampanel
Talampanel [7-acetyl-5-(4-amino-phenyl)-8,9dihydro-8- methyl-7H-1,3-dioxolo(4,5H)-2,3-benzodiazepine] is a new allosteric inhibitor of the AMPA receptor with a weak ability to inhibit kainate receptors [35]. It has no activity at NMDA receptors. TLP treatment sup- presses seizures in the maximal electroshock and pen- tylenetetrazole tests in rodents but is only weakly ac- tive in animal models of absence epilepsy [35]. Ac- cording to the results from a phase 2 trial of TLP for adults with recurrent gliomas, the treatment was well tolerated but had no significant activity as a single agent [36]. This drug is still being examined in pre- clinical and clinical trials as a treatment for patients with amyotrophic lateral sclerosis, adults with partial seizures or advanced Parkinson’s disease [125].
Mechanisms of antiepileptic drug actions
9􏰋=@OI􏰋=M 􏰌=I􏰂􏰍 AJ =􏰃􏰎
AEDs sharing other or unknown mechanism of action
Acetazolamide
Acetazolamide is a prototype inhibitor of carbonic an- hydrase and shows potent anticonvulsant activity against maximal electroshock-induced convulsions and, at higher dosages, against convulsions produced by the GABA) receptor antagonists, pentylenetetra- zole and bicuculline. As an uncompetitive anhydrase inhibitor in central glial cells, acetazolamide is re- sponsible for the elevation of carbon dioxide in the extraneuronal space and in neurons themselves. How- ever, how this event is involved in the inhibition of seizure activity is still a matter of dispute [67, 91].
The apparent disadvantage of this AED is the quick development of tolerance to its anticonvulsant action. Acetazolamide may be transiently effective against tonic-clonic, partial, and absence seizures [86].
Sulthiame
Sulthiame is an inhibitor of carbonic anhydrase [62, 91, 111, 112]. It is recommended in epileptic seizures of focal origin with or without secondary generaliza- tion, especially benign partial epilepsies in childhood, such as rolandic epilepsy [67].
Gabapentin (GBP)
In rodents, GBP [1-(aminomethyl)cyclohexaneacetic acid] inhibits the tonic hind limb extension produced by maximal electroshock and pentylenetetrazole-induced clonic seizures. This profile of activity is similar to VPA [13]. The mechanism of action of GBP is not fully un- derstood. In addition to being a cyclic analog of GABA, this AED does not bind to GABA) receptors. However, GBP may promote the release of an extravesicular pool of this amino acid neurotransmitter. In addition, GBP can increase the synthesis of brain GABA probably via an enhancement of glutamate decarboxylase activity. This AED also reduces the activity of the aminotransfe- rase, BCAA, which results in a reduced synthesis of glu- tamate. GBP and its structurally related drug – prega- balin – bind to the proteins of cortical neurons with an amino acid sequence that is identical to the a d subunits of the voltage-operated calcium L-type channel [100]. However, it has no effect on the activity of L-, T-, or N-
2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ 281

type calcium channels that are located in spinal dorsal root neurons [91, 110].
This AED is recommended against partial seizures with/or without secondary generalization. GBP can also be helpful in the treatment of chronic pain (e.g., neuropathic), migraine and bipolar disorder [67].
Levetiracetam (LEV)
LEV [(S)-2-(2-oxopyrrolidin-1-yl)butanamide] possesses a unique pharmacological profile of activity. It counter- acts partial and secondarily generalized tonic-clonic convulsions in the kindling model, but it is ineffective against maximal electroshock- or pentylenetetrazole- induced seizures [67, 91, 119]. Its mechanism of action has not been clarified. There is no convincing evidence pointing to the interaction of LEV with VDSCs or glu- tamate receptors. However, it reverses the effects of negative allosteric modulators of GABA) receptors, b-carboline or zinc ions [88]. An involvement of GABA-mediated events in the anticonvulsant activity of this AED is uncertain. LEV binds selectively to the synaptic vesicle protein, SV2A which is located in syn- aptic vesicles of the rat brain [88]. This protein is in- volved in the exocytosis of neurotransmitters, particu- larly glutamate. There is some evidence that SV2A is involved in the mediation of the in vivo anticonvulsant activity of LEV [119]. Clinical use of LEV is mainly restricted to drug-resistant partial seizures, mainly in the form of an add-on therapy.
A summary of the mechanisms of action of AEDs is listed in Table 1.
Pregabaline (PGB)
PGB [(S)-(+)-3-isobutyl-GABA] inhibits maximal elec- troshock and GABA A receptor antagonists-induced threshold clonic seizures in rodents. It is also effective against audiogenic seizures in DBA/2 mice and sup- press hippocampal kindling in rats. The mechanism of PGB action has not been fully elucidated. This drug binds to a b type 1 and 2 subunits of voltage-gated calcium channels, decreases calcium inward currents, and lowers glutamate brain concentration. PGB does not interact with GABA receptors, uptake or metabo- lism. PGB is recommended against refractory partial seizures as an add-on therapy [51].
Interactions concerning AEDs Pharmacokinetic interactions
Pharmacokinetics concerns problems that are related to drug absorption, distribution, metabolism and elimination from the body [28]. Consequently, there might be interactions that occur at these levels, for ex- ample, at the metabolic level. If there are pharmaco- logically active metabolites of a drug, then the exist- ing interactions may involve the parent drug and/or these metabolites [for review see 83].
A good example of an interaction at the level of drug absorption is the oral use of activated charcoal, which significantly reduces the absorption of various drugs, including AEDs [83].
The distribution of AEDs is closely associated to the degrees of their binding to blood albumins, and any interactions are likely when at least 90% of an AED is protein bound. PHT, TGB, and VPA fulfill this condition [91]. It is noteworthy that VPA can dis- place PHT from the protein bound form, which results in an increase in its free form that is responsible for the therapeutic effect. Compensative mechanisms are usually activated, and the enhanced metabolism of PHT is responsible for the eventual reduction of its to- tal blood concentration. Transiently, the free plasma concentration of PHT may even be elevated due to the inhibitory effect of VPA on PHT metabolism [83]. The estimation of total blood PHT concentration may be misleading because its reduction may be accompa- nied by toxicity that results from the transient increase in free plasma concentration [84]. A VPA-like effect towards PHT may be exerted by phenylbutazone or tolbutamide. AEDs, which are not protein bound to a considerable degree, are usually devoid of the inter- actions mentioned above. These AEDs include ESM, GBP, PGB, and VGB [85].
Because of different metabolic pathways, meta- bolic interactions can occur at the level of cytochrome P-450 (CYP) or UDP-glucuronosyltransferase (UGT) enzymes. The first pathway is involved in the metabo- lism of CBZ, FBM, PB, PHT, TGB, TPM and ZNS. The second pathway metabolizes LTG and VPA [87]. Among a variety of liver CYP isoenzymes for the me- tabolism of AEDs, CYP3A4, CYP2C9, and CYP2C19 are mainly responsible [96]. It is remarkable that CBZ, PB, PHT, and primidone are activators of CYP isoenzymes, but VPA is a CYP inhibitor. Conse-
282 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘

Tab. 1. Mechanisms of action of AEDs
Mechanisms of antiepileptic drug actions
9􏰋=@OI􏰋=M 􏰌=I􏰂􏰍 AJ =􏰃􏰎
AED
Benzodiazepines Carbamazepine Ethosuximide Phenobarbital Phenytoin Valproate Felbamate Gabapentin Lamotrigine Levetiracetam Oxcarbazepine Tiagabine Topiramate Vigabatrin Zonisamide Bivaracetam Carabersat Carisbamate Eslicarbazepine Ganaxolone Lacosamide Pregabalin Retiagabine Stiripentol Talampanel
Increase
in GABA level
+

  • +
    +
  • + +
  • +
    Affinity to GABA) receptor
    +
    +
    +
    +
    +
    Blockade of sodium channels
    +
  • +
  • + + + +
    Blockade of calcium channels
  • (L-type) + (T-type)
  • (T-type) (L-type)
  • (N-, P/Q-type)
  • (N-, P/Q-, R-, T-type) + (N-type)
  • (N-, P-type)
    +(L-type)
  • (N-, P-, T-type)
  • (N-, P/Q-type)
    Inhibition of glutamate excitation
    +
  • +
  • +
    Activation of potassium channels
    Other
    +
  • + + +
  • +
    +
  • + + +
  • +
    +
  • +
  • – Action experimentally proven. Controversial mechanisms were not included
  • Kv7.2-7.5
    quently, activators may accelerate their own metabo- lism or the metabolism of other drugs [83]. For exam- ple, CBZ enhances the metabolic breakdown of olan- zapine and warfarin [83]. Although newer AEDs are generally devoid of diffuse stimulatory effects on a series of CYP isoenzymes, they can enhance the ac- tivity of a particular CYP isoform. For example, OXC activates CYP3A isoenzymes, which are involved in the metabolism of oral contraceptive drugs [30, 84] or dihydropyridine calcium channel antagonists. TPM is a weak activator of the CYP3A4 isoenzyme, but at daily doses exceeding 200 mg, it may reduce the ef- fectiveness of oral contraception [84]. Due to the in-
    hibitory effect of VPA on CYP isoenzymes, this par- ticular AED may significantly elevate the plasma con- centrations of PB or PHT [74, 75, 103]. In addition to the inhibitory influence on CYP or UGT (see below), VPA is also an inhibitor of epoxide hydrolase, which is the enzyme that plays a role in the metabolism of CBZ. Eventually, VPA may cause an increase in the concentration of carbamazepine-10,11-epoxide, which may manifest in the form of an enhanced toxicity with plasma CBZ concentrations in the therapeutic range [83].
    Among the many UGT isoenzymes, UGT1A4 is associated with the glucuronidation of LTG, and UGT1A3 is associated with the glucuronidation of
    +
    +
    2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ 283 VPA [28, 46]. Another isoenzyme, UGT2B7, may also be involved in the metabolism of VPA [38]. Be- cause VPA is an UGT inhibitor, it may affect the me- tabolism of LTG. In contrast, OXC is an UGT activa- tor, and it accelerates the metabolism of LTG [83].
    Interactions involving AEDs at the level of renal excretion are rare. Nevertheless, drugs that alkalize urine can reduce the tubular reabsorption of PB, which may be of therapeutic significance in the case of a PB overdose [85]. FBM, GBP, LEV, TPM and VGB undergo renal elimination, but any participation of the active tubular transport in this process has not yet been confirmed. In fact, active tubular transport is required for the pharmacokinetic interactions to occur. However, the possibility of the interactions of AEDs with other drugs that are excreted in the same way cannot be excluded [83, 84].
    Pharmacokinetic interactions among AEDs in clinical conditions may concern either classical AEDs or both, classical and newer AEDs. Considering the first type of interactions, one cannot always predict the final outcome based on the influence of AEDs on metabolizing enzymes. The plasma concentration of PB may be elevated not only by VPA but also by PHT. Furthermore, PB administered in low doses may re- duce the plasma level of PHT, but when it is given in the higher dose range, it exerts an opposite effect [83]. Both PHT and PB lead to a reduction in the plasma concentration of CBZ [83], but as mentioned above, VPA elevates the plasma concentration of CBZ- 10,11-epoxide. Generally, CBZ, PB and PHT cause a decrease in the plasma concentration of VPA [83]. Regarding the second type of interactions, there is evidence that VGB may significantly reduce the plasma levels of PHT and CYP activators, including LTG, TGB, TPM and ZNS. In contrast, VPA produces an increase in the concentration of LTG [83]. A third type of interaction could also be found, an interaction between newer AEDs themselves. These interactions are rarer, mainly because of the better pharmacokinetic profiles of these AEDs. However, OXC retained the feature of CBZ to stimulate CYP enzymes, although to a lesser degree, and this mechanism may lower the plasma concentrations of LTG and TPM [83].
    Certainly, the drugs that are prescribed in epileptic patients for indications other than epilepsy can also affect the concentrations of AEDs in plasma [60, 73, 84]. There are many possibilities for these interac- tions; therefore, only the most important ones that cause substantial modifications of the AED concen-
    trations are mentioned below. The competition of macrolides, erythromycin, clarithromycin, and olean- domycin with CBZ for the isoenzyme CYP3A4 may cause a several-fold increase in the plasma concentra- tions of this AED. Moreover, significant elevations in CBZ concentration are observed in combinations with danazol, fluconazole, diltiazem, cimetidine, ticlopid- ine, and isoniazid [60]. Isoniazid can also induce a considerable rise in the plasma concentration of ethosuximide. Diltiazem, cimetidine, ticlopidine, or isoniazid cause a rise in the plasma concentration of PHT, which may precipitate toxicity in some patients. Aspirin may produce toxic effects in patients on VPA; there might be a substantial increase in the plasma concentration of this AED, which is displaced by as- pirin from plasma albumin. Also, aspirin inhibits the metabolic degradation of VPA. A significant rise in the plasma level of VPA may be a consequence of isoniazid co-administration. Other drugs, such as ke- toconazole, naproxen, ibuprofen, and mefenamic acid, produce a similar effect on the plasma concen- tration of VPA, but the clinical significance of these interactions is unknown [60]. Interestingly, there is limited data on the interactions between PB and other drugs. One important interaction is between this AED and activated charcoal, which eventually results in a strong inhibition of the absorption of PB from the intestines. This interaction is of therapeutic signifi- cance for the management of a PB overdose. The pharmacokinetic parameters of the newer AEDs can be hardly influenced by other drugs [60, 83].
    Pharmacodynamic interactions
    In contrast to pharmacokinetic interactions, pharma- codynamic interactions occur without any significant changes in the plasma or brain concentrations of the combined drugs. In clinical testing, the basic pharma- cokinetic parameters are estimated in the plasma or serum. In contrast, drug brain concentrations may be taken into consideration in experimental studies. This is quite remarkable because changes in the plasma concentrations of AEDs are not necessarily followed by respective changes in brain drug levels. Pharmaco- dynamic interactions assume that the net activity of the two combined AEDs results from the summation of their individual receptor or non-receptor effects. To evaluate the nature of these interactions, a basic con- vulsive test may be applied, for example, in the maxi- mal electroshock test in rodents (a model of human
    284 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ clonic-tonic and partial seizures, to a certain degree), the pentylenetetrazole test (a model of myoclonic convulsions), and the amygdala-kindled convulsions in rats (a model of partial seizures) [49]. The available data have been generally derived from the maximal electroshock test with the use of isobolographic analysis. Fewer results have been obtained in the pen- tylenetetrazole test, and only isolated data have been available from amygdala-kindled seizures in rats. The anticonvulsant activity of AEDs is usually displayed as ED#􏰅 values (in mg/kg) that are the median doses necessary to protect 50% of animals against seizures. A synergy (or hyper additive synergy) is defined when the actual anticonvulsant effect of two combined AEDs is greater than the theoretical sum of their indi- vidual effects. If the experimentally denoted anticon- vulsant effect is not significantly different from the theoretical sum, then an addition (or additive synergy) occurs. Finally, when the experimental effect is lower than the theoretical one, then an antagonism is evi- dent. Apart from the interactions of the anticonvulsant activity, interactions that are based on neurotoxicity need to be performed to fully characterize the nature of interactions among AEDs. Certainly, the best inter- actions from a preclinical point of view are character- ized by anticonvulsant synergy and neurotoxic antago- nism [16]. The interaction index, which is a helpful pa- rameter that characterizes the nature of interactions, may be calculated by dividing a given, experimentally obtained ED#􏰅 (or TD#􏰅) and the respective theoretical ED#􏰅 value. As a matter of fact, the interaction index is separately calculated for the anticonvulsant activity and neurotoxic effects. The benefit index, which char- acterizes both the anticonvulsant activity and neuro- toxicity, may be calculated by dividing the respective protective indices, the experimental and theoretical. Usually when the benefit index exceeds the value of 1.3, the combination of two AEDs is promising from a preclinical point of view. However, isobolography can precisely characterize the final outcome of an in- teraction between two AEDs for various fixed drug- dose ratio combinations in terms of synergy, addition or antagonism. Details of how to perform an isobolo- graphic analysis may be found in a series of papers [8, 14, 50, 52, 109, 114].
    Interactions among classical AEDs
    If not stated otherwise, the results listed below were obtained in the maximal electroshock test in mice.
    Mechanisms of antiepileptic drug actions
    9􏰋=@OI􏰋=M 􏰌=I􏰂􏰍 AJ =􏰃􏰎
    A combination of PB with PHT exerts an additive an- ticonvulsant effect, but the neurotoxic effects were antagonistic. However, the protective index of PHT alone was higher than for the combined treatment [14]. However, there are also data that point to the an- ticonvulsant synergy of this particular combination. Because there was no neurotoxicity measured in this study, any calculation of the protective index is im- possible. When CBZ was co-administered with PB, anticonvulsant and neurotoxic addition was evident. Identical interactions occurred for the combinations of VPA with CBZ or PB for the anticonvulsant activ- ity. Respective antagonism or additivity was ob- served, which makes the combination of VPA + CBZ superior from the preclinical point of view. A com- bined treatment of VPA with PHT produces synergy in the convulsive test and neurotoxic additivity. The protective index of this combination was considerably better than the protective index for these AEDs when given alone. In the pentylenetetrazole test in mice, very similar results were found for the combined treatment of VPA with ESM [14].
    Interactions between classical and newer AEDs
    Generally, the results presented below have been ob- tained in the maximal electroshock test in mice. Drug ratios were calculated based on the ED#􏰅s of AEDs in combinations. The ratio of 1:1 means that AEDs were put into a mixture in equal fractions of their ED#􏰅s. Ratios of 3:1 or 1:3 point to 75% of the ED#􏰅 of drug Aand25%ofED#􏰅ofdrugBor25%oftheED#􏰅of drug A and 75% of the ED#􏰅 of drug B, respectively.
    Combinations of GBP in a variety of fixed drug ra- tios with many AEDs (e.g., CBZ, PB, PHT, VPA) ex- erted anticonvulsant synergy, and only in one case, a combination with PB, a pharmacokinetic factor con- tributed to the final effect [8]. A very good profile of interaction has been observed for LTG + VPA (1:1) with anticonvulsant synergy and neurotoxic antago- nism being evident. Although the anticonvulsant syn- ergy occurred for a combination of LTG with PB (at 1:1 and 1:3), the same kind of interaction was also noted for neurotoxicity (at the ratio of 1:1). This result indicates that the net result of this interaction is nega- tive at the ratio of 1:1. An even worse outcome has been found for the combined treatment of LTG and CBZ at ratios of 1:1 and 3:1. In the ratio of 1:3, only a tendency toward antagonism was observed. In all three ratios, additive neurotoxic activity with a trend
    2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ 285 towards synergy was evident [59]. According to Shank et al. [99], various drug ratios combinations of TPM with PB or CBZ showed anticonvulsant syn- ergy, and an additivity of TPM with PHT was shown. Unfortunately, the neurotoxic effects of these combi- nations were not evaluated. Therefore, the full pre- clinical profile of the combinations of TPM with clas- sical AEDs is not available. Few data can be found on interaction of RTG with other AEDs. RTG enhanced protective effects of CBZ, diazepam, FBM, LTG, PHT, PB and VPA against audiogenic seizures in DBA/2 mice without affecting pharmacokinetic pa- rameters of these drugs [23]. Furthermore, the isobo- lographic analysis showed a clear-cut synergy for the combination of retigabine with valproate and no ef- fect of these drug combinations on motor coordina- tion, long term memory or muscular strength was ob- served [59].
    In the pentylenetetrazole test in mice, a combina- tion of VGB with PB displayed a promising preclini- cal profile with the anticonvulsive synergy accompa- nied by moderate toxicity. However, this interaction was not free from the pharmacokinetic mechanism because there was an increase in the brain PB concen- tration [58]. There are examples that some AEDs in combinations undergo many pharmacokinetic interac- tions. Stiripentol, when co-administered with ESM, PB or VPA in different fixed-dose ratios produces ad- ditivity in the pentylenetetrazole test in mice [56]. However, brain concentrations of PB and ESM were elevated, but the concentration of VPA was reduced.
    In addition, PB diminished and VPA increased the brain level of stiripentol. Regarding neurotoxicity, all combinations of stiripentol with the classical AEDs were also additive. These examples clearly indicate how important it is to verify the pharmacokinetic ef- fect on the pharmacodynamic effects.
    Interactions among newer AEDs
    As in previous interactions, the co-administration of newer AEDs was mainly performed using the maxi- mal electroshock test in mice [51]. In a variety of dose ratios (1:3, 1:1, 3:1), LEV exerts synergy when com- bined with TPM or OXC. Especially, these combina- tions were associated with positive neurotoxic pro- files [126]. Also, a combination of OXC with GBP seems preclinically attractive because the anticonvul- sant synergy was accompanied by no neurotoxicity. A very good preclinical profile may be ascribed to the combined treatment of TPM with LTG, especially at the fixed dose ratio of 1:1. At this ratio, the anticon- vulsant synergy was associated with neurotoxic an- tagonism [54]. The anticonvulsant synergy was also evident for a combination of GBP with TGB in all three fixed dose ratios of 1:3, 1:1 and 3:1 with no neu- rotoxicity for the 1:3 and 3:1 ratios [55]. When GBP was combined with LTG, an anticonvulsant synergy with no neurotoxicity was shown. However, brain GBP concentration was significantly elevated. It is remark- able that a clear-cut antagonism has been observed for the combined treatment of OXC with LTG regarding the
    Tab. 2. Effects of combinations of some antiepileptics (AEDs) evaluated experimentally with isobolography in mice
    Drug A
    Drug B LTG OXC TGB TPM
    CBZ Ant)@@ GBP S* LEV Add􏰅 OXC Ant5O􏰉 TGB Add􏰆- TPM S)􏰉 VPA S)􏰉
    Add)@@ Add􏰆- S􏰆- S􏰅 S)@@ S)@@
    S􏰅 NE S􏰅
    – Add) S)@@ Add)􏰉 – Add􏰆- S)@@ Add􏰆- – Add)@@ S􏰆- NE
    VGB VPA NE NE
    S􏰅 S)@@ NE Add􏰅
    NE Add)@@ S)@@ Add􏰆-
    NE NE Add􏰅 –
    Ant – Antagonism; S – synergy; Add – additivity; * – the increased level of GBP in brain has been observed; 􏰅 – no neurotoxicity observed for an- tiepileptics at the fixed dose ratio of 1:1, recorded in the chimney test or passive avoidance task; )@@ – additive neurotoxicity in the chimney test calculated by isobolography; )􏰉 – antagonistic neurotoxicity; 5O􏰉 – synergistic neurotoxicity; CBZ – carbamazepine; GBP – gabapentin; LEV – levetiracetam; LTG – lamotrigine; – – no possibility of combination; 􏰆- – neurotoxicity not evaluated; NE – not evaluated by isobolography; OXC – oxcarbazepine; 5O􏰉 – synergistic neurotoxic effects; TGB – tiagabine; TPM – topiramate; VGB – vigabatrin; VPA – valproate
    286 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ anticonvulsant activity. This preclinically negative re- sult was paralleled by synergistic neurotoxicity. In the threshold electroshock test in mice, combinations of VGB with TGB at 1:3, 1:1 and 3:1 fixed-dose ratios were additive with no neurotoxicity [52]. In the same test, a combination of VGB with GBP exerted anti- convulsant synergy at the dose ratio of 1:3, but the dose ratios of 1:1 and 3:1 were additive. Neurotoxic- ity was not found for any of these dose ratios [57].
    The results of the combined treatments with AEDs are listed in Table 2.
    Are there correlations between preclinical and clinical data on the interactions among AEDs?
    Due to intensive preclinical research and clinical tri- als, there are now approximately 30 available AEDs. It is obvious that the pathophysiology of an epileptic seizure is similar in experimental animals and hu- mans. Also, the mechanisms utilized by AEDs to in- hibit convulsions in experimental and clinical epilep- tology are identical, although there might be differ- ences in pharmacokinetic profiles. Taking this into consideration, one may conclude that the experimen- tal studies on AED combinations can have quite a sig- nificant predictive value. A number of AED combina- tions have been presented [103–105] with positive clinical outcomes (the results of preclinical studies are shown in parentheses): PHT + PB (addition/synergy), VPA + CBZ (addition), PHT + CBZ (addition), CBZ + TPM (synergy), CBZ + GBP (synergy), CBZ + LTG (antagonism), VPA + LTG (synergy), TPM + LTG (synergy). As can be seen from the above data, the only discrepancy concerns the combined treatment of CBZ + LTG, which was scored as positive in clinical conditions in contrast to the experimental evaluation that pointed to an evident antagonism. Can this dis- crepancy be interpreted in terms of the low predictive value of preclinical data on AED combinations? Pre- clinical research on this issue assumes addition as a simple summation of the partial protective effects of AEDs. Consequently, if a partial protective effect of drug A amounts to 35% in non-seizing mice and that of drug B to 45%, then a combined treatment of these drugs should result in a total protection within 80% if there is an additive interaction. The final protective
    Mechanisms of antiepileptic drug actions
    9􏰋=@OI􏰋=M 􏰌=I􏰂􏰍 AJ =􏰃􏰎
    effect of 60%, although exceeding the partial effects of drugs A and B (a positive finding in clinical condi- tions), is regarded as an antagonism from the preclini- cal point of view. This might be a reason for some po- tential discrepancies in the interpretation of experi- mental and clinical data. The clinical evaluation of this particular AED combination (CBZ + LTG) is not homogenous. The existing clinical evidence also points to a better outcome of LTG monotherapy vs. the combination of CBZ + LTG and to an increase in seizure frequency in 18% of patients who take this combination [14, 74, 83].
    Is the mechanism of action of AEDs relevant to clinical practice?
    Individually tailored treatment for a given patient, which is adequate for particular seizure syndrome or demographic characteristics, is an attractive concept in the modern management of epilepsy. However, it should be emphasized that the mechanism of action is only a supplementary factor in the process of selec- tion of proper AED for an individual patient. Many other factors, such as efficacy, tolerability and safety, are the main criteria for the selection of AEDs, al- though other properties, such as pharmacokinetics and ease of use, are also important [121]. There are practi- cal implications of a detailed knowledge of the mode of action for a particular drug. Accumulating clinical experience suggests that drugs with complex mecha- nisms of action that influence different ion channels or neurotransmitter systems, seem to display broader clinical efficacy for the suppression of both focal and generalized seizures. In contrast, drugs that act through a more specific mechanism are effective in narrower clinical spectrums. Another interesting issue is whether the etiology of epilepsy impacts the drug effect. There is only limited clinical evidence that a particular pathomechanism of a given epileptic syn- drome might preferentially respond to an AED with a specific mechanism of action. This concept is sup- ported by studies investigating the use of VGB, a GABA mimetic drug, in the treatment of infantile spasms. A meta-analysis of 10 clinical studies demon- strated that of the 313 patients without tuberous scle- rosis complex, 170 (54%) had a complete cessation of their infantile spasms, but 73 of the 77 patients with
    2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ 287 tuberous sclerosis complex (95%) had a complete ces- sation of their seizures. It was concluded that VGB should be considered as a first-line monotherapy for the treatment of infantile spasms in infants with either a confirmed diagnosis of tuberous sclerosis or those at high risk, i.e., infants with a first-degree relative with tuberous sclerosis complex [32]. The above findings give some hope that further investigation of the pathomechanism of selected epileptic syndromes may pave the way for individually directed drug selections that act not only on seizure suppression but also on epileptogenesis.
    Conclusions
    The management of resistant epileptic patients with AED combinations is based on the pharmacodynamic interactions between AEDs, and only combinations that show the best preclinical profiles need to be con- sidered. Pharmacodynamics may be complicated by the pharmacokinetic interactions of both the AEDs that are used in the combined treatment and other drugs that are prescribed for other than epilepsy rea- sons. A failure of the combined treatment with AEDs may result from the above-mentioned pharmacoki- netic factors. Also, the pharmacodynamic interactions of AEDs with non-AEDs cannot be underestimated. Both the experimental and clinical data indicate that methylxanthines (e.g., theophylline, pentoxifylline) and caffeine significantly reduce the protective effi- cacy of AEDs [11]. Patients receiving methylxanthine medications or heavy coffee drinkers may experience no or considerably reduced benefits from combined AED treatment even though these combinations are rationally chosen. Although the choice of specific drug is still mainly based on clinician’s experience, the better understanding of pathogenesis of epilepsy and mechanisms of the AED action should allow a ra- tional approach to be applied [37].
    Acknowledgment:
    This manuscript has been inspired by the GlaxoSmithKline-supported Conference “AED therapy: does MoA matter?” in London, June 2010.
    References:
  1. Armijo JA, Shushtarian M, Valdizan EM, Cuadrado A, de las Cuevas I, Adin J: Ion channels and epilepsy. Curr Pharm Des, 2005, 11, 1975–2003.
  2. Barnes GN, Slevin JT: Ionotropic glutamate receptor biol- ogy: effect on synaptic connectivity and function in neuro- logical disease. Curr Med Chem, 2003, 10, 2059–2072.
  3. Barrese V, Miceli F, Soldovieri MV, Ambrosino P, Iamotti FA, Cilio MR, Taglialatela M: Neuronal potassium chan- nel openers in the management of epilepsy: role and po- tential of retigabine. Clin Pharmacol, 2010, 2, 225–236.
  4. Basta-Kaim A, Budziszewska B, Leœkiewicz M, Kubera M, Jag3a G, Nowak W, Czuczwar SJ, Lasoñ W: Effects of new antiepileptic drugs and progabide on the mitogen-induced proliferative activity of mouse spleno- cytes. Pharmacol Rep, 2008, 60, 925–932.
  5. Beyreuther BK, Freitag J, Heers C, Krebsfänger N, Scharfenecker U, Stöhr T: Lacosamide: a review of pre- clinical properties. CNS Drug Rev, 2007, 13, 21–42.
  6. Bialer M: New antiepileptic drugs that are second gen- eration to existing antiepileptic drugs. Expert Opin In- vestig Drugs, 2006, 15, 637–647.
  7. Bialer M, Johannessen SI, Kupferberg HJ, Levy RH, Perucca E, Tomson T: Progress report on new antiepilep- tic drugs: a summary of the Eigth Eilat Conference (EILAT VIII). Epilepsy Res, 2007, 73, 1–52.
  8. Borowicz KK, Œwi1der M, £uszczki J, Czuczwar SJ: Effect of gabapentin on the anticonvulsant activity of antiepileptic drugs against maximal electroshock in mice – an isobolographic analysis. Epilepsia, 2002, 43, 956–963.
  9. Brodie MJ: Felbamate: a new antiepileptic drug. Lancet, 1993, 341, 1445–1446.
  10. Brodie MJ, Lerche H, Gil-Nagel A, Elger C, Hall S, Shin P, Nohria V, Mansbach H; RESTORE 2 Study Group: Efficacy and safety of adjunctive ezogabine (retigabine) in refractory partial epilepsy. Neurology, 2010, 75, 1817–1824.
  11. Chroœciñska-Krawczyk M, Rastnaraj N, Patsalos PN, Czuczwar SJ: Effect of caffeine on the anticonvulsant effects of oxcarbazepine, lamotrigine and tiagabine in a mouse model of generalized tonic-clonic seizures. Pharmacol Rep, 2009, 61, 819–826.
  12. Coulter DA, Huguenard JR, Prince DA: Calcium cur- rents in rat thalamocortical relay neurones: kinetic prop- erties of the transient, low-threshold current. J Physiol, 1989, 414, 587–604.
  13. Czapiñski P, B3aszczyk B, Czuczwar SJ: Mechanisms of action of antiepileptic drugs. Curr Top Med Chem, 2005, 5, 3–14.
  14. Czuczwar SJ, Kap3añski J, Œwiderska-Dziewit G, Ger- gont A, Kroczka S, Kacinski M: Pharmacodynamic in- teractions between antiepileptic drugs: preclinical data based on isobolography. Expert Opin Drug Metab Toxi- col, 2009, 5, 1–6.
  15. Czuczwar SJ, Patsalos PN: The new generation of GABA enhancers. CNS Drugs, 2001, 15, 339–350.
  16. Czuczwar SJ, Borowicz KK: Polytherapy in epilepsy:
    the experimental evidence. Epilepsy Res, 2002, 52, 15–23.
    288 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘
  17. Czuczwar P, Wojtak A, Cioczek-Czuczwar A, Parada- Turska J, Maciejewski R, Czuczwar SJ: Retiga- bine:the newer potential antiepileptic drug. Pharmacol Rep, 2010, 62, 211–219.
  18. Dalby NO, Nielsen EB: Comparison of the preclinical anticonvulsant profiles of tiagabine, lamotrigine, gaba- pentin and vigabatrin. Epilepsy Res, 1997, 28, 63–72.
  19. Dam M, Ekberg R, Løyning Y, Waltimo O, Jakobsen K: A double-blind study comparing oxcarbazepine and car- bamazepine in patients with newly diagnosed, previously untreated epilepsy. Epilepsy Res, 1989, 3, 70–76.
  20. Davies JA: Mechanisms of action of antiepileptic drugs. Seizure, 1995, 4, 267–271.
  21. Deckers CLP, Czuczwar SJ, Hekster YA, Keyser A, Kubova H, Meinardi H, Patsalos PN et al.: Selection of antiepileptic drug polytherapy based on mechanism of action: The evidence reviewed. Epilepsia, 2000, 41, 1364–1374.
  22. De Romanis F, Sopranzi N: Lamotrigine in the therapy of resistant epilepsy. Clin Ter, 1999, 150, 279–282.
  23. De Sarro G, Di Paola ED, Conte G, Pasculli MP, De Sarro A: Influence of retigabine on the anticonvulsant activity of some antiepileptic drugs against audiogenic seizures in DBA/2 mice. Naunyn Schmiedebergs Arch Pharmacol, 2001, 363, 330–336
  24. De Sarro G, Gitto R, Russo E, Ibbadu GF, Barreca ML, De Luca L, Chimirri A: AMPA receptor antagonists as potential anticonvulsant drugs. Curr Top Med Chem, 2005, 5, 31–42.
  25. Doty P, Rudd GD, Stoehr T, Thomas D: Lacosamide. Neurotherapeutics, 2007, 4, 145–148.
  26. Dichter MA, Brodie MJ: New antiepileptic drugs. N Engl J Med, 1996, 334, 1583–1590.
  27. Dickins M, Chen C: Lamotrigine. Chemistry, biotrasfor- mation and pharmacokinetics. In: Antiepileptic Drugs. Eds. Levy RH, Mattson RH, Meldrum BS, Perucca E, Lippincott, Williams & Wilkins, Philadelphia, 2002, 370–379.
  28. Eadie MJ: Formation of active metabolites of anticonvul- sant drugs. A review of their pharmacokinetic and thera- peutic significance. Clin Pharmacokinet, 1991, 21, 27–41.
  29. Enz R, Cutting GR: Molecular composition of GABAC receptors. Vision Res, 1998, 38, 1431–1441.
  30. Fattore C, Cippola G, Gatti G, Limido GL, Sturm Y, Bernsconi C, Perucca E: Induction of ethinylestradiol and levonorgestrel metabolism by oxcarbazepine in healthy women. Epilepsia, 1999, 40, 783–787.
  31. Gil-Nagel A: Does mode of action predict spectrum of efficacy and side-effect profile? In: Antiepileptic Drug Therapy: Does Mechanism of Action Matter? Radisson Edwardian Heathrow Hotel, London, UK, Monday 7 June 2010.
  32. Hancock E, Osborne JP: Vigabatrin in the treatment of infantile spasms in tuberous sclerosis: literature review. J Child Neurol, 1999, 14, 71–74.
  33. Harris JA, Murphy JA: Lacosamide and epilepsy. CNS Neurosci Ther, 2010, 15.
  34. Hough CJ, Irwin RP, Gao XM, Rogawski MA, Chuang DM: Carbamazepine inhibition of N-methyl-D-aspartate- evoked calcium influx in rat cerebellar granule cells.
    J Pharmacol Exp Ther, 1996, 276, 143–149.
    Mechanisms of antiepileptic drug actions
    9􏰋=@OI􏰋=M 􏰌=I􏰂􏰍 AJ =􏰃􏰎
  35. Howes JF, Bell C: Talampanel. Neurotherapeutics, 2007, 4, 126–129.
  36. Iwamoto FM, Kreisl TN, Kim L, Duic JP, Butman JA, Albert PS, Fine HA: Phase 2 trial of talampanel, a gluta- mate receptor inhibitor, for adults with recurrent malig- nant gliomas. Cancer, 2010, 116, 1776–1782.
  37. Jacobs MP, Fischbach GD, Davis MR, Dichter MA, Dingle- dine R, Lowenstein DH, Morrell MJ et al.: Future directions for epilepsy research. Neurology, 2001, 57, 1536–1542.
  38. Jakus R, Graf M, Ando RD, Balogh B, Gacsaly I, Levay G, Kantor S, Bagdy G: Effect of two noncompetitive AMPA receptor antagonists GYKI 52466 and GYKI 53405 on vigilance, behavior and spike-wave discharges in a genetic rat model of absence epilepsy. Brain Res, 2004, 1008, 236–244.
  39. Kalia LV, Kalia SK, Salter MW: NMDA receptors in clinical neurology: excitatory times ahead. Lancet Neu- rol, 2008, 7, 742–755
  40. K3odziñska A, Bijak M, Chojnacka-Wójcik E, Kroczka B, Swiader M, Czuczwar SJ, Pilc A: Roles of group II metabotropic glutamate receptors in modulation of sei- zure activity. Naunyn Schmiedebergs Arch Pharmacol, 2000, 361, 283–288.
  41. Leach JP, Sills GJ, Majid A, Butler E, Carswell A, Thompson GG, Brodie MJ: Effects of tiagabine and vi- gabatrin on GABA uptake into primary cultures of rat cortical astrocytes. Seizure, 1996, 5, 229–234.
  42. Lee CY, Fu WM, Chen CC, Su MJ, Liou HH: Lamotrigine inhibits postsynaptic AMPA receptor and glutamate re- lease in the dentate gyrus. Epilepsia, 2008, 49, 888–897.
  43. Leppik IE: Zonisamide: chemistry, mechanism of action, and pharmacokinetics. Seizure, 2004, 13, Suppl 1, S5-9, discussion S10.
  44. Lerche H, Weber YG, Jurkatt-Rott K, Lehmann-Horn F: Ion channel defects in idiopathic epilepsies. Curr Pharm Des, 2005, 11, 2737–2752.
  45. Leœkiewicz M, Lasoñ W: The neurochemical mecha- nisms of temporal lobe epilepsy: an update (Polish). Przegl Lek, 2007, 64, 960–964.
  46. Liston HL, Markovitz JS, Devane L: Drug glucuronida- tion in clinical psychopharmacology. J Clin Psychophar- macol, 2001, 21, 500–515.
  47. Löscher W: Basic pharmacology of valproate: a review after 35 years of clinical use for the treatment of epi- lepsy. CNS Drugs, 2002, 16, 669–694.
  48. Löscher W, Köhling R: Functional, metabolic, and syn- aptic changes after seizures as potential targets for antie- pileptic therapy. Epilepsy Behav, 2010, 19, 105–113.
  49. Löscher W, Schmidt D: Which animal models should be used in the search for new antiepileptic drugs? A pro- posal based on experimental and clinical considerations. Epilepsy Res, 1988, 2, 145–181.
  50. £uszczki JJ: Isobolographic analysis of interaction be- tween drugs with nonparallel dose-response relationship curves: a practical application. Naunyn-Schmiedeberg’s Arch Pharmacol, 2007, 375, 105–114.
  51. £uszczki JJ: Third-generation antiepileptic drugs: mechanisms of action, pharmacokinetics and interac- tions. Pharmacol Rep, 2009, 61, 197–216.
  52. £uszczki JJ, Czuczwar SJ: Isobolographic characteriza- tion of interactions between vigabatrin and tiagabine in
    2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ 289 two experimental models of epilepsy. Prog Neuropsy-
    chopharmacol Biol Psychiatry, 2007, 30, 529–538.
  53. £uszczki JJ, Czuczwar SJ: Preclinical profile of combina-
    tions of some second-generation antiepileptic drugs: an
    isobolographic analysis. Epilepsia, 2004, 45, 895–907.
  54. £uszczki JJ, Czuczwar M, Kiœ J, Pasztelan M, Krysa J,
    Czuczwar SJ: Interactions of lamotrigine with topiramate and first generation antiepileptic drugs in the maximal electroshock test in mice: an isobolographic analysis. Epilepsia, 2003, 44, 1003–1013.
  55. £uszczki JJ, Kolacz A, Wojda E, Czuczwar M, Przesmycki K, Czuczwar SJ: Synergistic interaction of gabapentin with tiagabine in the hot-plate test in mice: an isobolographic analysis. Pharmacol Rep, 2009, 61, 459–467.
  56. £uszczki JJ, Ratnaraj N, Patsalos P, Czuczwar SJ: Char- acterization of the anticonvulsant, behavioral, and phar- macokinetic interaction profiles of stiripentol in combi- nation with clonazepam, ethosuximide, phenobarbital, and valproate using isobolographic analysis. Epilepsia, 2006, 47, 1841–1854.
  57. £uszczki JJ, Ratnaraj N, Patsalos PN, Czuczwar SJ: Isobolographic and behavioral characterizations of inter- actions between vigabatrin and gabapentin in two experi- mental models of epilepsy. Eur J Pharmacol, 2008, 595, 13–21.
  58. £uszczki JJ, Wójcik-Æwik3a J, Andres MM, Czuczwar SJ: Pharmacological and behavioral characteristics of interactions between vigabatrin and conventional antiepi- leptic drugs in pentylenetetrazole-induced seizures in mice: an isobolographic analysis. Neuropsychopharma- cology, 2005, 30, 958–973.
  59. £uszczki JJ, Wu JZ, Raszewski G, Czuczwar SJ: Isobo- lographic characterization of interactions of retigabine with carbamazepine, lamotrigine, and valproate in the mouse maximal electroshock-induced seizure model. Naunyn Schmiedebergs Arch Pharmacol, 2009, 379, 163–179.
  60. Majkowski J, Patsalos PN: Interaction between antiepi- leptic and non-epileptic drugs. In: Antiepileptic Drugs. Combination Therapy and Interactions. Eds. Majkowski J, Bourgeois B, Patsalos P, Mattson R, Cambridge Uni- versity Press, Cambridge, 2005, 139–177.
  61. Maljevic S., Wuttke TV, Seebohm G, Lerce H: Kv7 channelopathies. Eur J Physiol, 2010, 460, 277–288.
  62. Masuda Y, Karasawa T, Shiraishi Y, Hori M, Yoshida K, Shimizu M: 3-Sulfamoylmethyl-1,2-benzisoxazole, a new type of anticonvulsant drug. Pharmacological pro- file. Arzneimittelforschung, 1980, 30, 477–483.
  63. Mathews GC: The dual roles of GABA in seizures and epilepsy generate more excitement. Epilepsy Curr, 2007, 7, 28–30.
  64. Mattson RH: Efficacy and adverse effects of established and new antiepileptic drugs. Epilepsia, 1995, 36, Suppl 2, S13–S26.
  65. Matute C: Therapeutic Potential of kainate receptors. CNS Neurosci Ther, 2010, 6.
  66. McLean MJ, Macdonald RL: Carbamazepine and 10,11- epoxycarbamazepine produce use- and voltage- dependent limitation of rapidly firing action potentials of mouse central neurons in cell culture. J Pharmacol Exp Ther, 1986, 238, 727–738.
  67. McNamara JO: Pharmacotherapy of the epilepsies In: The Pharmacological Basis of Therapeutics. Eds. Brun- ton LL, Lazo JS, Parker KL, Goodman and Gilman’s 11th edition, McGraw-Hill Medical Publishing Division, New York, 2006, 501–526.
  68. Meldrum BS, Rogawski MA: Molecular targets for an- tiepileptic drug development. Neurotherapeutics, 2007, 4, 18–61.
  69. Miller AA, Wheatley P, Sawyer DA, Baxter MG,
    Roth B: Pharmacological studies on lamotrigine, a novel potential antiepileptic drug: I. Anticonvulsant profile in mice and rats. Epilepsia, 1986, 27, 483–489.
  70. Moldrich RX, Chapman AG, De Sarro G, Meldrum BS: Glutamate metabotropic receptors as targets for drug ther- apy in epilepsy. Eur J Pharmacol, 2003, 22, 476, 3–16.
  71. Mora G, Tapia R: Effects of retigabine on the neurode- generation and extracellular glutamate changes induced by 4-aminopyridine in rat hippocampus in vivo. Neuro- chem Res, 2005, 30, 1557–1565.
  72. Nasreddine W, Beydoun A, Atweh S, Abou-Khalil B: Emerging drugs for partial onset seizures. Expert Opin Emerging Drugs, 2010, 15, 415–431.
  73. Parker AC, Pritchard B, Preston T, Choonara I: Induction of CYP1A2 activity by carbamazepine in children using the caffeine breath test. Br J Clin Pharmacol, 1998, 45, 176–178.
  74. Patsalos PN: Pharmacokinetic principles and mecha- nisms of drug interactions. In: Antiepileptic Drugs. Combination Therapy and Interactions. Eds. Majkowski J, Bourgeois B, Patsalos P, Mattson R, Cambridge Uni- versity Press, Cambridge, 2005, 47–56.
  75. Patsalos PN, Lascelles PT: Effect of sodium valproate on plasma protein binding of diphenylhydantoin. J Neurol Neurosurg Psychiat, 1977, 40, 570–574.
  76. Pellock JM, Faught E, Leppik IE, Shinnar S, Zupanc ML: Felbamate: consensus of current clinical experience. Epilepsy Res, 2006, 71, 89–101.
  77. Pennell PB, Ogaily MS, Macdonald RL: Aplastic anemia in a patient receiving felbamate for complex partial sei- zures. Neurology, 1995, 45, 456–460.
  78. Perucca E: The clinical pharmacology and therapeutic use of the new antiepileptic drugs. Fund Clin Pharmacol, 2001, 15, 405–407.
  79. Plosker GL, Scott LJ: Retigabine: in partial seizures. CNS Drugs, 2006, 20, 601–608, discussion 609–610.
  80. Poolos NP, Migliore M, Johnston D: Pharmacological upregulation of h-channels reduces the excitability of py- ramidal neuron dendrites. Nat Neurosci, 2002, 5, 767–774.
  81. Porter RJ, Partiot A, Sachdeo R, Nohria V, Alves WM: 205 Study Group: Randomized, multicenter, dose- ranging trial of retigabine for partial-onset seizures. Neurology, 2007, 68, 1197–1204.
  82. Putnam TJ, Merritt HH: Experimental determination of the anticonvulsant properties of some phenyl derivatives. Science, 1937, 85, 525–526.
  83. Rambeck B, May TW: Interactions between antiepileptic drugs. In: Antiepileptic Drugs. Combination Therapy and Interactions. Eds. Majkowski J, Bourgeois B, Patsalos P, Mattson R, Cambridge University Press, Cambridge, 2005, 111–138.
    290 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘
  84. Reddy DS: Clinical pharmacokinetic interactions be- tween antiepileptic drugs and hormonal contraceptives. Expert Rev Clin Pharmacol, 2010, 3, 183–192.
  85. Reinikainen KJ, Keränen T, Halonen T, Komulainen H, Riekkinen PJ: Comparison of oxcarbazepine and carba- mazepine: A double-blind study. Epilepsy Res, 1987, 1, 284–289.
  86. Reiss WG, Oles KS: Acetazolamide in the treatment of seizures. Ann Pharmacother, 1996, 30, 514–519.
  87. Rendic S, Di Carlo FJ: Human cytochrome P450 en- zymes: a status report summarizing their reactions, sub- strates, inducers, and inhibitors. Drug Metab Rev, 1997, 29, 413–580.
  88. Rigo JM, Hans G, Nguyen L, Rocher V, Belachew S: The anti-epileptic drug levetiracetam reverses the inhibi- tion by negative allosteric modulators of neuronal GABA- and glycine-gated currents. Br J Pharmacol, 2002, 136, 659–672.
  89. Riss J, Cloyd J, Gates J, Collins S: Benzodiazepines in epilepsy: pharmacology and pharmacokinetics. Acta Neurol Scand, 2008, 118, 69–86.
  90. Rogawski M: Diverse mechanisms of antiepileptic drugs in the development pipeline. Epilepsy Res, 2006, 69, 273–294.
  91. Rogawski MA, Porter RJ: Antiepileptic drugs: Pharma- cological mechanisms and clinical efficacy with consid- eration of promising developmental stage compounds. Pharmacol Rev, 1990, 42, 223–286.
  92. Rostock A, Tober C, Rundfeldt C, Bartsch R, Engel J, Polymeropoulos EE, Kutscher B et al.: D-23129: a new anticonvulsant with a broad spectrum activity in animal models of epileptic seizures. Epilepsy Res, 1996, 23, 211–223.
  93. Sabers A, Gram L: Newer anticonvulsants: Comparative review of drug interactions adverse effects. Drugs, 2000, 60, 23–33.
  94. Schachter SC: A review of the antiepileptic drug tiaga- bine. Clin Neuropharmacol, 1999, 22, 312–317.
  95. Schauf CL: Zonisamide enhances slow sodium inactiva- tion in Myxicola. Brain Res, 1987, 9, 413, 185–188.
  96. Scheyer RD: Valproic acid: drug interactions. In: Antie-
    pileptic Drugs. Ed. Levy RH, Mattson RH, Meldrum BS, Perucca E, Lippincott, Williams & Wilkins, Philadelphia, 2002, 801–807.
  97. Schulze-Bonhage A: Zonisamide in the treatment of epi- lepsy. Expert Opin Pharmacother, 2010, 11, 115–126.
  98. Schumacher TB, Beck H, Steinhäuser C, Schramm J,
    Elger CE: Effects of phenytoin, carbamazepine, and ga- bapentin on calcium channels in hippocampal granule cells from patients with temporal lobe epilepsy. Epilep- sia, 1998, 39, 355–363.
  99. Shank RP, Gardocki JF, Vaught JL, Davis CB, Schupsky JJ, Raffa RB, Dodgson SJ et al.: Topiramate: preclinical evaluation of structurally novel anticonvulsant. Epilep- sia, 1994, 35, 450–460.
  100. Sills GJ: Combining mechanisms: how can experimental studies inform clinical practice? The Conference Antie- pileptic drug therapy: does mechanism of action matter? Radisson Edwardian Heathrow Hotel, London, UK, Monday 7 June 2010.
    Mechanisms of antiepileptic drug actions
    9􏰋=@OI􏰋=M 􏰌=I􏰂􏰍 AJ =􏰃􏰎
  101. Sills GJ: The mechanisms of action of gabapentin and pregabalin. Curr Opin Pharmacol, 2006, 6, 108–113.
  102. Singh R, Gardner RJ, Crossland KM, Sheffer IE, Berko- vic SF: Chromosomal abnormalities and epilepsy: a re- view for clinicians and gene hunters. Epilepsia, 2002, 43, 127–140.
  103. Spina E, Perucca E, Levy R: Predictability of metabolic antiepileptic drug interactions. In: Antiepileptic Drugs. Combination Therapy and Interactions. Eds. Majkowski J, Bourgeois B, Patsalos P, Mattson R, Cambridge Uni- versity Press, Cambridge, 2005, 57–92.
  104. Stafstrom CE: Mechanisms of action of antiepileptic drugs: the search for synergy. Curr Opin Neurol, 2010, 23, 157–163.
  105. Stephen LJ, Brodie MJ: Seizure freedom with more than one antiepileptic drug. Seizure, 2002, 11, 349–351.
  106. Stöhr T, Kupferberg HJ, Stables JP, Choi D, Harris RH,
    Kohn H, Walton N, White HS: Lacosamide, a novel anti-convulsant drug, shows efficacy with a wide safety margin in rodent models for epilepsy. Epilepsy Res, 2007, 74, 147–154.
  107. Suzuki S, Kawakami K, Nishimura S, Watanabe Y, Yagi K, Seino M, Miyamoto K: Zonisamide blocks T-type cal- cium channel in cultured neurons of rat cerebral cortex. Epilepsy Res, 1992, 12, 21–27.
  108. Swinyard EA, Sofia RD, Kupferberg HJ: Comparative anticonvulsant activity and neurotoxicity of felbamate and four prototype antiepileptic drugs in mice and rats. Epilepsia, 1986, 27, 27–34.
  109. Tallarida RJ: An overview of drug combination analysis with isobolograms. J Pharmacol Exp Ther, 2007, 113, 197–209.
  110. Taylor CP, Gee NS, Su TZ, Kocsis JD, Welty DF, Brown JP, Dooley DJ et al.: A summary of mechanistic hypothe- sis of gabapentin pharmacology. Epilepsy Res, 1998, 29, 233–249.
  111. Temperini C, Innocenti A, Mastrolorenzo A, Scozzafava A, Supuran CT: Carbonic anhydrase inhibitors. Interac- tion of the antiepileptic drug sulthiame with twelve mammalian isoforms: kinetic and X-ray crystallographic studies. Bioorg Med Chem Lett, 2007, 17, 4866–4872.
  112. Thiry A, Dogné JM, Supuran CT, Masereel B: Anticon- vulsant sulfonamides/sulfamates/sulfamides with carbonic anhydrase inhibitory activity: drug design and mechanism of action. Curr Pharm Des, 2008, 14, 661–671.
  113. Tomson T: Drug selection for the newly diagnosed pa- tient: when is a new generation antiepileptic drug indi- cated? J Neurol, 2004, 251, 1043–1049.
  114. Trojnar MP, Kimber-Trojnar ̄, Trojnar MK, Czuczwar SJ: A review of experimental data on efficacy of antiepileptic drug combinations in seizure models – an isobolographic analysis. Epileptologia, 2006, 14, 301–311.
  115. Tunnicliff G: Basis of the antiseizure action of pheny- toin. Gen Pharmacol, 1996, 27, 1091–1097.
  116. Twyman RE, Rogers CJ, Macdonald RL: Differential regulation of gamma-aminobutyric acid receptor chan- nels by diazepam and phenobarbital. Ann Neurol, 1989, 25, 213–220.
  117. Walker MC, Sander JW: New antiepileptic drugs. Expert Opin Invest Drugs, 1999, 10, 1497–1510.
    2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘ 291
  118. Wauquier A, Zhou S: Topiramate: a potent anticonvul- sant in the amygdala-kindled rat. Epilepsy Res, 1996, 24, 73–77.
  119. White HS, Smith MD, Wilcox KS: Mechanism of action of antiepileptic drugs. Int Rev Neurobiol, 2007, 81, 85–110.
  120. Wickenden AD: Potassium channels as anti-epileptic drug targets. Neuropharmacology, 2002, 43, 1055–1060.
  121. Willmore LJ: Clinical pharmacology of new antiepileptic drugs. Neurology, 2000, 55, 17–24.
  122. Wolff H, Castle NA, Pardo LA: Voltage-gated potassium channels as therapeutic drug targets. Nat Rev Drug Dis- cov, 2009, 8, 982–1001.
  123. Wuttke TV, Lerche H: Novel anticonvulsant drugs target- ing voltage-dependent ion channels. Expert Opin Inves- tig Drugs, 2006, 15, 1167–1177.
  124. Wuttke TV, Seebohm G, Bail S, Maljevic S, Lerche H: The new anticonvulsant retigabine favors voltage- dependent opening of the Kv7.2 (KCNQ2) channel by binding to its activation gate. Mol Pharmacol, 2005, 67, 1009–1017.
  125. http://www.clinicaltrials.gov
  126. Zaccara G, Gangemi PF, Bendoni L, Menge GP,
    Schwabe S, Monza GC: Influence of single and repeated doses of oxcarbazepine on the pharmacokinetic profile of felodipine. Ther Drug Monit, 1993, 15, 39–42.
    Received: February 24, 2011; in the revised form: March 14, 2011; accepted: March 18, 2011.
    292 2D=H􏰁=?􏰂􏰃􏰂CE?=􏰃 4AF􏰂HJI􏰄 􏰅􏰀􏰀􏰄 $!􏰄 %􏰀` ‘

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: