Antiepileptic Drugs, 5th Edition



Mechanisms of Action

Charles P. Taylor PhD

Director, Department of CNS Pharmacology, Pfizer Global Research and Development, Ann Arbor, Michigan

Gabapentin (Neurontin or l-[aminomethyl]cyclohexaneacetic acid) is a novel amino acid derived by addition of a cyclohexyl group to the chemical backbone of γ-aminobutyric acid (GABA) (74), the major rapid inhibitory neurotransmitter of mammalian brain (Figure 28.1A). X-ray crystallography and molecular modeling studies also indicate a structural similarity between gabapentin and L-leucine (Figure 28.1B). Gabapentin is an effective and safe anticonvulsant in extensive and well-controlled clinical trials (3,50). As a chemical derivative of GABA, gabapentin was expected originally to mimic the actions of GABA at inhibitory synaptic receptors in the brain and to mimic the action of baclofen or to prevent seizures by increasing rapid chloride-dependent inhibition. However, with one controversial exception, subsequent studies have shown that gabapentin is inactive at GABAA and GABAB receptors. In fact, many studies in vivo and in vitro have failed to pinpoint the mechanism of anticonvulsant action of gabapentin clearly.

Gabapentin interacts with a specific high-affinity binding site in mammalian brain membranes that is an auxiliary protein subunit of voltage-gated calcium channels. In addition, gabapentin reduces the release of several neurotransmitters in a manner that may be related to this high-affinity binding site. This chapter reviews the pharmacology of gabapentin.

The cellular mechanisms of anticonvulsant drugs are the subject of several reviews (70,91,96), and the pharmacology of gabapentin has been summarized in the literature (24,92). In addition to gabapentin, agents that have shown significant efficacy for treatment of refractory partial seizures in controlled clinical trials include sodium channel modulators (phenytoin, carbamazepine, zonisamide, and lamotrigine) and drugs that interact with GABA receptors or GABA metabolism or uptake (phenobarbital, vigabatrin, and the GABA-uptake blocker tiagabine; perhaps also valproic acid and felbamate). Of these clinically proven anticonvulsants, only gabapentin and vigabatrin are amino acids, a feature suggesting that the mechanisms of both drugs may relate to the neurotransmitter or metabolic amino acids of the brain.


Anticonvulsant Actions in Animal Models

To date, no published reports have shown that gabapentin alters seizurelike discharges in vitro, although one report indicates that gabapentin has an additive effect with vigabatrin to alter seizurelike discharges in rat hippocampal slices (46). Another unpublished study showed that gabapentin application, at 100 µmol/L for 60 minutes, did not inhibit stimulus-induced burst firing in hippocampal slices, a model of seizures in isolated hippocampal tissue (W. Wilson et al., unpublished data, methods of ref. 79). However, in vivo, gabapentin reduces the behavioral seizure score, and at higher dosages, it shortens the local afterdischarge duration in rats kindled by repeated electrical stimulation of the limbic system, a model of partial seizures (44,57,104). Furthermore, gabapentin treatment reduces the duration and development of maximal dentate activation (108), a model of epileptic seizures in anesthetized rats (85). In addition, systemic administration of gabapentin to rodents prevents seizures from a variety of different electrical and chemical stimuli (Table 28.1). Tonic extensor seizures from many different chemical agents are prevented, whereas threshold clonic seizures from several agents are not. In contrast to the prototype agents phenytoin and carbamazepine (39), clonic seizures from pentylenetetrazol are prevented by gabapentin. Gabapentin is active at low doses and thus prevents seizures from maximal electroshock in rats after doses equivalent to or lower than those of phenytoin or carbamazepine. With an intravenous dose of gabapentin (15 mg/kg) that prevents seizures from maximal electroshock in practically all rats, plasma gabapentin concentrations are approximately 5.0 µg/mL (30 µmol/L) at the time of peak anticonvulsant action (102).



FIGURE 28.1. A: Chemical structures of gabapentin in comparison with the inhibitory neurotransmitter γ-aminobutyric acid (GABA) and the enantiomers of 3-isobutyl GABA (pregabalin and R(-)-3-isobutyl GABA. Gabapentin and pregabalin cross the blood-brain barrier in animals and prevent seizures when these agents are given systemically, whereas GABA does not. Although the two-dimensional structures of gabapentin and GABA are similar, three-dimensional models show differences in the spacing of amino and carboxyl functional groups. B: Computer-generated molecular model of gabapentin superimposed on a similar model for L-leucine. This stereo pair may be viewed either with a stereo viewer or by slightly crossing the eyes to superimpose the two images. The carboxyl and amino moieties of the two molecules match closely in three-dimensional space, as do the lipophilic hydrocarbon moieties. These results suggest that gabapentin may be expected to mimic the actions of L-leucine and similar branched-chain neutral amino acids at a variety of biologic sites. The computer-modeled structure agrees with three-dimensional crystal structure of gabapentin (crystal coordinates deposited with the Cambridge structural database, Cambridge CB2 1E2, United Kingdom). (B, from E. Lunney, Pfizer Global R & D, unpublished data and Ibers JA. Gabapentin and gabapentin monohydrate. Acta Crystallographica 2001;57:641-643, with permission.



Convulsant Agent

Dose Route

ED50 (mgkg)


Tonic extensor seizures (maximal electroshock)




Tonic extensor seizures (maximal electroshock)

i.v., p.o.

7.2, 9.1


Tonic extensor seizure (thiosemicarbazide)

i.v., p.o.

6.3, 5.0a


Tonic extensor seizure (3-mercaptopropionic acid)




Tonic extensor seizure (strychnine)




Tonic extensor seizure (pentylenetetrazole)




Tonic extensor seizure (N-methyl-D-aspartate)




Tonic extensor seizure (kainate)




Threshold clonic seizure (pentylenetetrazole)




Threshold clonic seizure (bicuculline)




Threshold clonic seizure (picrotoxin)




Threshold clonic seizure (strychnine)




Behavioral seizure score (hippocampal kindled rats)



DBA/2J mouse

Tonic extensor seizure (audiogenic)



Wistar rat (strain)

Absence seizures (electroencephalogram)


Not effective (25-100 mg/kg)

Gerbil (genetic strain)

Tonic extensor seizure



Baboon (genetic strain)

Photogenic myoclonus


Not effective (1.0-240 mg/kg)

ED50, median effective dose; i.p., intraperitoneal; i.v., intravenous; p.o., oral.

a From Bartoszyk et al. (1); others are unpublished data.

b Lowest effective dose.

c At this dose, seizures were significantly delayed but not prevented.



Gabapentin is active in several genetic models of seizures. However, unlike the anticonvulsants valproate, ethosuximide, and the benzodiazepines (53), gabapentin fails to prevent spike-and-wave events in the electroencephalogram of rats with genetic absence seizures. Gabapentin did not alter generalized seizure activity in lethargic mice (a second genetic model of absence) (34). In comparison with other antiepileptic drugs, the spectrum of anticonvulsant activity with gabapentin suggests efficacy for treatment of partial seizures and generalized tonic-clonic seizures but not for generalized absence. The profile of activity of gabapentin in animal models of seizure activity suggests that gabapentin may have a different pharmacologic mechanism than that of other antiepileptic drugs.

Another novel GABA derivative, S-(+)-3-isobutyl-GABA or pregabalin (94), has shown a profile of anticonvulsant activity very similar to that of gabapentin. The two enantiomers (optical isomers) of pregabalin differ greatly in their anticonvulsant potency and thus offer a tool with which to study various mechanisms that may be involved in the anticonvulsant action of pregabalin. The enantiomers of pregabalin are stereoselective for their inhibition of [3H]-gabapentin binding, a finding suggesting that anticonvulsant activity may relate to potency for displacing [3H]-gabapentin binding (94) (see also the later discussion of binding).


FIGURE 28.2. The anticonvulsant activity of gabapentin reaches a maximum approximately 3 hours after intravenous administration (electroshock response, closed circles). However, gabapentin concentrations in blood plasma (open triangles) and in brain interstitial space (open squares) peak soon after dosing and are declining at the time of maximal anticonvulsant action (concentrations may be converted to micromolar by multiplying by 5.8). These results indicate that the anticonvulsant mechanism of gabapentin requires significant time to be expressed and suggest that biochemical changes must take place before anticonvulsant effects are seen. (From Welty DF, Wang Y, Busch JA, et al. Pharmacokinetics and pharmacodynamics of CI-1008 (pregabalin) and gabapentin in rats using maximal electroshock. Epilepsia 1997;38:35-36, with permission.)

In contrast to most other anticonvulsants, the time of peak anticonvulsant action with gabapentin is delayed about 2 hours after intravenous administration, past the time of peak drug concentration in either the blood plasma or the brain interstitial space (102) (Figure 28.2). This delay suggests that anticonvulsant action may be caused by biochemical changes that require gabapentin to be present in brain for a substantial period. Alternatively, gabapentin may require prolonged binding to an extracellular receptor before its anticonvulsant action becomes significant. Existing data do not distinguish among these possibilities.

Other Pharmacologic Actions

Gabapentin is active in several animal models of spasticity (5). Spasticity was the initial therapeutic target of clinical studies with gabapentin, and subsequently, efficacy for treating spasticity has been demonstrated in clinical studies (13,26,66). In addition, studies have shown that gabapentin is active both in animal models of analgesia (1,21, 36,59,61,107) and in clinical trials for treatment of neuropathic pain (4,73). Similarly, gabapentin is active in animal models of anxiety (78), and it reduces anxiety scores in clinical studies (60). The cellular mechanisms of


gabapentin for reducing spasticity, anxiety, and pain are not well understood, but they may overlap with mechanisms that are relevant for treatment of epilepsy (see later). Finally, high doses of gabapentin reduce neuronal pathology both in vitro (72) and in an animal model of amyotrophic lateral sclerosis (28). However, the mechanism of neuroprotective effects of gabapentin may differ from those for epilepsy and the other indications mentioned earlier (101). Furthermore, the proposed neuroprotective mechanism, that is, inhibition of branched-chain amino acid aminotransferase (BCAA-T) (35,41), probably occurs only at high concentrations of gabapentin that are not reached with clinical treatments.

Gabapentin, at 300 mg/kg, administered intravenously to mice, causes a 30% reduction in spontaneous locomotor activity and a slight reduction in behavioral responsiveness, but mice are still able to walk normally and to respond to sensory stimulation; no sleeping, catalepsy, or changes in pinnal or corneal reflexes are observed. At higher doses, gabapentin causes ptosis, exophthalmos, ataxia, muscle relaxation, and further slowing of behavioral responses. These results indicate that gabapentin has a weaker tendency for causing dose-related behavioral side effects than prototype anticonvulsants such as phenytoin or carbamazepine, which cause unresponsiveness and depress respiration at doses of ≥ 100 mg/kg when these agents are given intravenously to mice or rats (C. P. Taylor and M. G. Vartanian, unpublished data).


Effects on GABA Receptors, GABA Turnover, and GABA Function

Despite previous findings that gabapentin is inactive as an agonist or antagonist at GABAB receptors, a recent study indicates that gabapentin is an agonist of one subtype of recombinant heterodimeric GABAB receptor, GABAB 1a/2b, a subtype thought to act mostly postsynaptically (55). Other recent results (39a,97a) indicate that gabapentin does not directly interact with GABAB receptor but may indirectly activate GABAA receptors by elevating extracellular GABA concentration. In addition, the actions of gabapentin on neurotransmitter release (D. Dooley et al., personal communication) and in an electrophysiologic model with anesthetized rats (84) are not reversed by antagonists of GABABreceptors. Therefore, it remains to be demonstrated whether a GABAB agonistlike action of gabapentin is relevant for anticonvulsant pharmacology.

Gabapentin increases the apparent rate of synthesis of GABA by up to twice control levels in several brain regions, a finding suggesting that altered GABA metabolism may underlie anticonvulsant action (43). However, the time course of changes in GABA synthesis induced by gabapentin does not correspond very well with the time course of anticonvulsant effects, so the relevance of this effect remains unclear. Regardless of the mechanism responsible, gabapentin treatment in human patients causes a rapid and sustained elevation of whole-tissue GABA concentration in brain (63,64), and these changes in GABA concentration may lead to changes in GABA synaptic function. However, the pool of GABA measured in such studies is largely confined anatomically within GABAergic neurons of brain, and only a very small fraction of total GABA is free in extracellular space, where it is able to interact with neuronal GABA receptors. One analysis suggests that increased cellular concentration of GABA in vitro (106) can cause a steady inhibitory GABAA-mediated current in cultured neurons. If such a mechanism also occurs in vivo, it would reduce brain excitability and thus could reduce or prevent seizure activity.

Gabapentin and pregabalin alter the inhibition of evoked potentials in hippocampus of anesthetized rats (86,108), a finding suggesting a change in GABA function in rat brain in vivo.This effect differs from that of baclofen (84), but the relevance and cellular mechanisms involved are not clear. Furthermore, gabapentin appears to reduce inhibition mediated by GABAergic interneurons in this in vivo model, and this is difficult to reconcile with an anticonvulsant effect. Despite this difficulty, these results suggest that gabapentin treatment alters the function of GABAergic inhibition in hippocampal neurons in vivo.

In vitro data with isolated optic nerve segments from neonatal rats suggest that gabapentin enhances the release of GABA from reversal of the GABA transporter (38). High-pressure liquid chromatography studies of optic nerve tissue (54) indicate that the effect of gabapentin in this system differs from that of vigabatrin, which increases GABA concentration in nerve segments (and brain tissues) by inhibiting GABA degradation. In this system, GABA can be released in a calcium-independent manner by application of the inhibitor and substrate of GABA transporters, nipecotic acid (56). The released GABA acts at GABAA receptors on axons and causes an easily measured depolarization. Preincubation of optic nerves with gabapentin (100 µmol/L for 1 hour) increases depolarizations from application of 1.0 mmol/L nipecotic acid. Nipecotic acid responses in optic nerve are not altered by acute application of gabapentin, and acute application causes no GABA agonistlike or antagonistlike effects. Furthermore, nipecotic acid responses both in control conditions and after enhancement by gabapentin are completely blocked by the GABAA antagonist bicuculline, a finding indicating that the effect is caused by changes in GABA release. Similar results with gabapentin have been obtained with field potential records and neuronal voltage-clamp records from the CA1 area of rat hippocampal slices (32,33,39a).

Previous studies have shown that increases in extracellular potassium or application of glutamate agonists release GABA from cultured brain cells both by calcium-dependent and calcium-independent mechanisms, the latter arising


from alterations in the equilibrium of GABA transport (22,65). Some of the anticonvulsant effects of gabapentin therefore may arise from increased nonsynaptic GABA release during seizure activity, which has previously been hypothesized to be altered in epileptic brain tissues (17,22).

Gabapentin's prevention of seizures in rodents from administration of GABA antagonists or GABA synthesis inhibitors and findings of increased brain GABA concentrations in humans suggest that gabapentin may enhance GABA activity in brain by a novel means, in comparison with benzodiazepines, barbiturates, or GABA uptake inhibitors (52).

Recent findings (103a) indicate that application of gabapentin or pregabalin to rat neuronal cultures for two hours increases the fraction of GAT1 GABA transportor protein located at cell membranes.

Effects on Neurotransmitter Release

Gabapentin slightly reduces the release of several monoamine neurotransmitters from mammalian brain tissue slices in vitro (68,75), and this action has been reexamined with both gabapentin and pregabalin (15). This more recent study shows that the actions of gabapentin are most pronounced when slices are stimulated for many seconds by modest concentrations of potassium to evoke release (rather than by electrical stimulation or high potassium concentrations). Under these conditions, the release of preloaded norepinephrine is reduced about 35% by 100 µmol/L gabapentin, and the half-maximal concentration is near 10 µmol/L (Figure 28.3). These studies also show that the effects of gabapentin and pregabalin are not additive, and so both compounds are likely to share a common cellular mechanism. The relevance of changes in monoamine release for anticonvulsant action is unclear, but changes in the release of other neurotransmitters could easily be relevant, and gabapentin also has been reported to reduce glutamate release from brain slices (16, 48). Although the molecular mechanism of reduced neurotransmitter release has not been clearly established, it is likely that changes in intracellular calcium are involved, because gabapentin also limits the increase in calcium fluorescence caused by potassium application to synaptosomes (21a,51, 97), and this could relate to binding of gabapentin to the α2δ auxiliary calcium channel protein (21). Furthermore, the actions of gabapentin on neurotransmitter release differ from those of the GABAB agonist baclofen (68). This has been confirmed by findings that a selective GABAB antagonist did not alter the effects of gabapentin on norepinephrine release in vitro (D. Dooley et al., personal communication). Despite all these effects of gabapentin on release of neurotransmitters in biochemical studies, the application of either gabapentin or pregabalin does not alter synaptic potentials in anesthetized rats evoked by single action potentials (86) and has only rather subtle inhibitory effects on synaptic currents in CA1 neurons (54a,97) or entorhinal neurons (21a) in tissue slices. The findings to date suggest that neurotransmitter release is modified by gabapentin only in a relatively subtle manner and particularly under conditions of prolonged depolarization or activation of cellular second messengers.


FIGURE 28.3. Gabapentin reduces the release of tritiated norepinephrine from rat neocortical brain slices in vitro (15). Tritium overflow (from [3H]norepinephrine that was preloaded into rat neocortex tissue) was evoked by a 120-second application of elevated potassium solution (25 mmol/L). Fractions (5 mL) were collected from superfusate samples (1 mL/min) and were analyzed for tritium. Each data point represents the mean from six or more replicate experiments. Significant differences from the control group were seen with 10 µmol/L and greater of gabapentin (single asterisk denotes p < 0.05; double asterisks denote p < 0.01 by ANOVA). Error bars denote standard error of the mean. Gabapentin had a maximal inhibition of approximately 40% and an IC50 for this effect of approximately 9 µmol/L. (From Dooley DJ, Donovan CM, Pugsley TA. Stumulus-dependent modulation of [3H]norepinephrine release from rat neocortical slices by gabapentin and pregabalin. J Pharmacol Exp Ther 2000; 296:1086-1098, with permission.)

Effects on Cytosolic Enzymes of Brain Tissue

Gabapentin has been studied for effects on several enzymes that metabolize amino acids in rat brain tissues (25) (Figure 28.4). Several effects have been found, but they only occur with concentrations of gabapentin greater than those relevant for human therapy. Specifically, the activity of glutamic acid decarboxylase (GAD), BCAA-T, GABA transaminase (GABA-T), glutamate dehydrogenase (GDH), glutamine synthase, glutaminase, alanine aminotransferase, aspartate aminotransferase and γ-glutamyl transferase have been studied. These results are summarized in Table 28.2.

GAD is the primary synthetic enzyme for GABA, and it has been proposed that anticonvulsant actions of valproic acid may result from increased GAD activity (45). Studies with gabapentin and pregabalin show that they both enhance GAD activity in protein extracts partially purified from porcine brain (93). It is possible that the increased GABA turnover observed in rat brain after administration of gabapentin (43) results from activation of GAD. However, in vitro results need to be confirmed with purified isoenzymes of GAD. The high concentrations (and doses) of gabapentin


needed to enhance GAD activity suggest that this mechanism may not be relevant for anticonvulsant activity in humans.


FIGURE 28.4. Simplified schematic diagram showing pathways of amino acid metabolism in the brain. In vitro, gabapentin interacts with branched-chain amino acid aminotransferase (BCAA-T), glutamic acid decarboxylase (GAD), glutamate dehydrogenase (GDH), and γ-aminobutyric acid transaminase (GABA-T), but not with glutaminase or glutamine synthase. It is unclear which of these interactions (if any) are necessary for anticonvulsant actions.

BCAA-T of rat brain is the primary enzyme that metabolizes cytosolic L-leucine, L-isoleucine, or L-valine together with α-ketoglutarate to form glutamate; the other product of this reaction, α-ketoglutarate, is formed in glia and is exported to neurons (35,41). In a partially purified in vitro enzyme preparation, gabapentin inhibits BCAA-T activity (25). Gabapentin is a competitive inhibitor of BCAA-T metabolism of L-leucine, L-isoleucine, and L-valine, with antagonist affinity constant (Ki) values of 300 to 1,300 µmol/L for gabapentin and Michaelis constant (Km) values of 500 to 1,000 µmol/L for L-leucine, L-isoleucine, or L-valine (A. Goldlust et al., personal communication). In contrast, pregabalin is a weak noncompetitive inhibitor of BCAA-T, with Ki values about 20-fold higher. Relatively high concentrations of gabapentin decrease the de novo synthesis of glutamate in isolated retinal tissues (41) by this mechanism and this effect may also relate to changes in glutamine metabolism (6,41). BCAA-T inhibition by gabapentin may account for neuroprotective


action of gabapentin (101), both in a transgenic model of amyotrophic lateral sclerosis (28) and in an in vitro model (72). However, it seems unlikely that BCAA-T inhibition accounts for the anticonvulsant action of gabapentin because of the high doses and concentrations required. In any case, activity at this enzyme suggests a biologic similarity between gabapentin and endogenous amino acids such as L-leucine (Figure 28.1B).




Action (at 10.0 mmol/L)

Inhibition Type

Ki (Apparent)



Threefold enhancement





Sixfold enhancement



Glutamate decarboxylase (93)


78% Enhancement (at 2.5 mmol/L)



Branched-chain amino acid aminotransferase


64% Inhibition


0.8 mmol/L

GABA aminotransferase (rat brain)


32% Inhibition

Mixed (competitive and noncompetitive)

23 mmol/L

Glutamine synthase


No effect




No effect


Alanine aminotransferase (rat brain)


No effect


Aspartate aminotransferase (rat brain)


No effect


γ-glutamyltransferase (rat brain, substrate activity)


<50% of glutamine


Ki, antagonist affinity constant; GDH, glutamate dehydrogenase; GABA, γ-aminobutyric acid.

From refs. 25 and 93, with permission.

GABA-T is the primary degradative enzyme for the neurotransmitter GABA and converts it to succinic semialdehyde and glutamate. This enzyme is the primary target of the irreversible enzyme inhibitor and anticonvulsant agent vigabatrin (γ-vinyl GABA, (40,42)). Gabapentin inhibits GABA-T activity, but in contrast to vigabatrin, inhibition is very weak and completely reversible, with both competitive and noncompetitive components (apparent Ki near 25 mmol/L). The two enantiomers of pregabalin also inhibit GABA-T with Kivalues of 6.6 and 64 mmol/L for pregabalin and the R enantiomer, respectively (A. Goldlust et al., personal communication). The concentrations of soluble GABA in the whole brain are in the range of 1,000 to 2,000 µmol/L (19) and are even higher within GABAergic neurons. In comparison, concentrations of gabapentin in brain tissues are probably too low to be significant on GABA-T. In addition, the spectrum of pharmacologic actions of gabapentin in animals is different from that of vigabatrin (58), a finding suggesting that GABA-T inhibition may not be significant with gabapentin.

GDH is found mostly in the mitochondria of rat brain neurons. It is responsible for the degradation of cytosolic glutamate under certain conditions and also for synthesis of glutamate under other conditions. L-leucine and its chemical analog 2-aminobicyclo [2.2.1] heptane-2-carboxylic acid (BCH) enhance the action of GDH when added at millimolar concentrations, and presumably as a result, they increase synaptosomal ammonia concentrations (18). Likewise, 10 mmol/L of gabapentin and pregabalin significantly increase GDH activity from rat brain (11,25). However, this effect occurs only at millimolar concentration and therefore is unlikely to be relevant for clinical treatment.

Effects on Neurotransmitter Receptors and Ion Channels

Radioligand Binding Studies

Gabapentin has been tested in a wide range of standardized radioligand binding assays and electrophysiologic responses in vitro and is remarkably silent in most tests (14). Gabapentin failed to displace radioligand binding at concentrations up to 100 µmol/L in standardized assays for GABAA, GABAB (including the subtypes GABAB1a/2 and GABAB1b/2) (39a,97a), benzodiazepine, glutamate, N-methyl-D-aspartic acid (NMDA), quisqualate, kainate, glycine, MK-801, and strychnine-insensitive glycine receptors. In addition, gabapentin was inactive at the following: A1 and A2 adenosine receptors; α1, α2, and β-adrenergic receptors; D1 and D2 dopamine receptors; histamine H1 receptors; S1 and S2 serotonin receptors; M1, M2, and nicotinic acetylcholine receptors; δ, κ, and µ opiate receptors; leukotriene B4, D4, and thromboxane A2 receptors; phorbol ester dibutyrate receptors; and binding sites on calcium channels labeled by nitrendipine and diltiazem and on sodium channels labeled by batrachotoxinin. These negative results suggest that gabapentin has a novel mechanism of action in comparison with many other drugs active in the central nervous system.

Gabapentin had little or no effect in several electrophysiologic tests with intracellular voltage records from cultured mouse spinal cord neurons in vitro (69) or in rat hippocampal slices in vitro (29). Furthermore, gabapentin did not alter responses to the application of GABAA or GABAB agonists in rat hippocampal slices or cultured neocortical neurons (39a,97a). Unlike GABAA or GABAB agonists or glutamate agonists, application of gabapentin (up to 500 µmol/L) did not alter resting neuronal membrane potential (69). Gabapentin had no effect on inhibitory responses of spinal cord neurons to GABA or glycine (69).

NMDA-Dependent Responses

Different studies have reported mixed results with gabapentin on responses dependent on NMDA receptors. When gabapentin was added to solutions bathing rat neocortical wedges (methods of ref. 30), it caused no responses and did not alter responses to the application of glutamate agonists NMDA, kainic acid, D,L-α-amino-3-hydroxy-5-methylisoxazole propionic acid (AMPA), or the NMDA antagonists 3-(2-carboxypiperazine-4-yl)propyl-1 -phosphonic acid (CPP) or 7-chlorokynurenic acid with or without added glycine (L. Robichaud and P. Boxer, personal communication). Furthermore, the application of 100 µmol/L gabapentin did not change single-channel NMDA responses in outside-out patches with respect to the frequency of openings, mean open duration, burst duration, or mean number of openings per burst (69). In contrast to these negative results, another study (27) indicates that gabapentin enhances NMDA-induced currents in acutely isolated dorsal horn neurons, but only when the catalytic subunit of protein kinase C is applied intracellularly or when neurons are taken from acutely inflamed rats. Furthermore, the gabapentin effect reported by Gu and Huang (27) is accounted for by shifting the concentration-response curve so a lower concentration of the required coagonist, glycine, is required. Therefore, gabapentin is not expected to modulate NMDA receptors in saturating concentrations or very low concentrations of glycine. Enhancement of NMDA responses by gabapentin does not reconcile easily with an anticonvulsant effect. However, Gu and Huang (27) speculate that they recorded primarily from GABAergic


interneurons, which may have increased excitability in response to gabapentin. These results, like those from neurotransmitter release studies (see earlier), suggest that gabapentin action in cellular systems may require altered states of cytosolic protein phosphorylation or sustained depolarization before they can be observed. Furthermore, the results of Gu and Huang raise the possibility that the effects of gabapentin may vary widely in different types of neurons (e.g., GABA neurons versus glutamate neurons).

Voltage-Gated Sodium Channels

Acute application of gabapentin had no effect on sustained repetitive firing of sodium-dependent action potentials of mouse spinal cord neurons or voltage-clamped calcium currents in dorsal root ganglion neurons (69). However, a different report indicates that gabapentin reduces sustained repetitive action potentials in cultured spinal cord cells when it is applied for longer periods (99). The modulation of action potentials by sustained gabapentin application probably occurs by a different mechanism from that of phenytoin, carbamazepine, or lamotrigine (67), because gabapentin and pregabalin do not interact with batrachotoxinin binding (105) even at high concentrations (unpublished data). Voltage-clamped sodium currents in a Chinese hamster ovary (CHO) cell line expressing sodium channel α subunits are not altered even by prolonged application of gabapentin (90), and sodium currents in acutely isolated neocortex neurons are not altered by gabapentin (81). Furthermore, gabapentin does not alter veratridineinduced cell death in cultured neocortical neurons (D. M. Rock, personal communication), whereas phenytoin, carbamazepine, and lamotrigine counteract veratridine. Therefore, the modulation of repetitive action potentials by gabapentin (99) appears to arise from a different molecular site of action than those of phenytoinlike anticonvulsants and needs further study.

Voltage-Gated Calcium Channels

The actions of gabapentin on voltage-clamped calcium currents also have been controversial. A study with acutely isolated rat neurons reported a small decrease in calcium currents (80), and this has been confirmed more recently by the same group (81). However, another published study with acutely isolated human neurons reported no effect of gabapentin (76), and a more recent study has not shown changes in calcium currents with gabapentin applied to neuronal pyramidal cell bodies (97). This latter report suggested that calcium influx in synaptosomes (isolated presynaptic endings) was indeed sensitive to gabapentin, and this has been shown independently using fluorescent tracer methods (21a,51), a finding suggesting that presynaptic calcium channels may be more sensitive to gabapentin than channels on cell bodies. Another a study reported inhibition of calcium currents by gabapentin in cultured rat sensory neurons (89), but inhibition was only consistent and pronounced when neurons were previously exposed to an analog of cyclic adenosine monophosphate that activates protein kinase A. It is tempting to speculate that reduced calcium channel action is involved with the anticonvulsant action of gabapentin, particularly in light of high-affinity binding to the α2δ protein that interacts with calcium channels. However, it appears from the work published so far that gabapentin does not block calcium channels directly and completely (as do dihydropyridines or conotoxins). Furthermore, the reduced calcium channel function with gabapentin may not occur under all conditions, but may instead require the presence of presynaptic proteins (as in synaptosomes) or activation of specific cytosolic kinases. Additional work in this area is still needed to confirm the existing results with gabapentin and to clarify the conditions that are necessary for calcium channel modulation to occur.


Naturally occurring amino acids (and also gabapentin) are dually ionized at neutral pH. Because of this, they are not very soluble in lipids and have very little permeability to cell membranes by diffusion. To facilitate the transport of various amino acids across cell membranes, several families of specialized, membrane-bound proteins have evolved (12). Because gabapentin is an analog of a naturally occurring amino acid, it may be carried by one or more of the endogenous transport systems that are present in a variety of tissues. Initial studies of transport with gabapentin focused on transporters for glutamate and GABA that have specialized and highly selective transport mechanisms. Gabapentin and pregabalin at concentrations up to 1 mmol/L do not alter the uptake of [3H]-GABA or [3H]-D-aspartate into synaptosomes, cultured astrocytes, or CHO tumor cells. These results indicate that gabapentin is not a substrate or an inhibitor of either GABA or glutamate transporters (87).

System L Transport

Studies in humans and animals (82) and in vitro (83) show that gabapentin is transported across gut membranes into the bloodstream by a saturable mechanism similar to the large, neutral amino acid carrier (system L) of gut tissues. The transport of [3H]-gabapentin and [3H]-L-phenylalanine are mutually inhibitory, a finding indicating a common transport mechanism. These results and others published in the same report (82) indicate that one mechanism of gabapentin transit from gut to bloodstream is through a sodium-independent system L-like amino acid transporter.

The system L transporter and several other amino acid transporters have been characterized by radiotracer techniques


in CHO tumor cells (87). System L actually represents a family of related sodium-independent transporters in various tissues that are relatively selective for neutral BCAAs and aromatic amino acids (12). Experiments with system L in CHO cells extend and confirm the notion that gabapentin competes with leucine for transport by system L.

The transport of [3H]-gabapentin into CHO cells with different concentrations of substrate is shown in Figure 28.5A. Transport is mostly saturable, with Michaelis-Menten kinetics (Kmvalue of 35 µmol/L and Vmax of 4.4 nmol/min/mg). These values of Km and Vmax are similar to those reported elsewhere for L-leucine. Uptake of gabapentin was inhibited more than 90% by L-leucine, L-valine, L-isoleucine, L-phenylalanine, and the synthetic leucine analog BCH (2-aminobicyclo[2.2.1] heptane-2-carboxylic acid). The synthetic system A-specific amino acid analog MeAIB (methylaminoisobutyric acid) did not significantly inhibit gabapentin uptake. Thus, gabapentin behaves like a substrate of system L in CHO cells. A Dixon plot of the data with gabapentin inhibition of [3H]-L-leucine uptake (Figure 28.5B) gives a Ki value similar to that for unlabeled L-leucine, consistent with competitive inhibition by a single transporter. Finally, both gabapentin and L-leucine facilitate the efflux of the other compound when they are added at the extracellular side of the membrane (this is calledtransstimulation of efflux and is a measure of heteroexchange). The heteroexchange of L-leucine and gabapentin confirms that a single transporter (system L) carries both amino acids.


FIGURE 28.5. Tritiated gabapentin is transported into Chinese hamster ovary cells in vitro by system L. A: The initial rate of transport is saturable, with a Km or half-maximum rate of transport at about 35 µmol/L. B: Inhibition of L-leucine transport by gabapentin is competitive with a Ki value of about 25 µmol/L, very similar to the Km values for gabapentin transport (compare with A) and L-leucine transport (not shown). These data and others (see text) indicate that gabapentin and several other amino acids (L-leucine, L-isoleucine, L-valine, L-phenylalanine, and others) share a common transport mechanism, namely system L. (From T.-Z. Su and D. Oxender, Parke-Davis Research, unpublished data; also from Su TZ, Lunney E, Campbell G, et al. Transport of gabapentin, a gamma-amino acid drug, by system I alpha-amino acid transporters: a comparative study in astrocytes, synaptosomes, and CHO cells. J Neurochem 1995;64:2125-2131, with permission.)

System L transport also occurs in neuronal membranes and in astrocytes (6). In these systems, influx of [3H]-gabapentin was more than 90% independent of sodium and was reciprocally inhibited by L-leucine, BCH, and unlabeled gabapentin. Thus, rat brain membranes have facilitated transport for gabapentin, a finding that may explain the accumulation of gabapentin in whole brain tissues (mostly representing cytosol) at higher concentrations than in the brain interstitial space (102).

Gabapentin is rapidly permeable to the blood-brain barrier (98,102) and is transported into brain tissues by a saturable mechanism accounted for by system L (47). In spite of the clear evidence that gabapentin is a transported by system L in various tissues, this appears not to account for anticonvulsant actions, because nonmetabolized substrates of system L transport such as BCH do not share the anticonvulsant properties of gabapentin in animal models.


Studies with [3H]-gabapentin reveal a specific binding site in brain (88). An elegant series of biochemical and molecular studies shows conclusively that the gabapentin binding site is identical with the α2δ auxiliary protein subunit of voltage-gated calcium channels (23). A Scatchard analysis of specific binding activity shows a single site with apparent affinity constant or Kd of 38±2.8 nmol/L and density of binding sites or Bmax of 4.6±0.4 pmol/mg protein (Figure 28.6A). [3H]-Gabapentin binding is not altered by a wide variety of neuroactive substances including glutamate, GABA, NMDA, glycine, other anticonvulsants, or common neurotransmitters. In contrast, [3H]-gabapentin binding is competitively blocked by low micromolar or high nanomolar concentrations of L-leucine, L-isoleucine, L-valine, and L-phenylalanine (7,95), but the significance of these findings remain unclear. However, the results with endogenous amino acids suggest that in brain tissues in vivo, the [3H]-gabapentin binding site is saturated with endogenous BCAAs such as L-leucine.




FIGURE 28.6. A: [3H]-Gabapentin binding to rat neocortex synaptic plasma membranes is saturable. Specific binding, defined by binding displaced by 100 µmol/L of (R,S)3-isobutyl γ-aminobutyric acid (GABA), was determined at increasing concentration of [3H]-gabapentin. Analysis of the Scatchard plot (inset) yielded a 50% saturation (Kd) value of 38 nmol/L and a maximum density of binding sites (Bmax) of 4.6 pmol/mg protein. B: The autoradiographic localization of [3H]-gabapentin binding sites in a coronal section of rat brain tissue is shown. Highest densities of binding are seen in neocortex layers I and II (1, 2) and dendritic regions of hippocampus (CA1, CA2, and CA3 subfields and dentate gyrus or dg). The molecular layer of cerebellum (mol) also has a high density of binding, whereas the granule layer of cerebellum (gr), lateral septum (Is), and white matter (w)have low densities. [A, From Suman-Chauhan et al. (88); B, from Hill et al. (31), with permission.]

α2δ Protein has been cloned and functionally expressed at the cell membrane in a number of recombinant cell systems. α2δ Function is necessary for the high-level expression and function of the main ion-conducting calcium channel protein, α1. The function of α2δ proteins has been reviewed previously (20). Furthermore, four distinct α2δ-like proteins have been cloned and expressed in recombinant systems. Each of the four types function together with other calcium channel subunits (37). However, [3H]-gabapentin binds significantly only to the α2δ1 and α2δ2 isoforms, and not to α2δ3 or α2δ4 (49). The α2δ1 isoform is expressed widely in various tissues (e.g., brain, striate and cardiac muscle, glands), whereas expression of type 2 is somewhat more restricted to the central nervous system. It is not known which subtypes contribute most to the pharmacology of gabapentin for prevention of seizures. Mutagenesis studies of the α2δ1 protein have shown that binding of [3H]-gabapentin occurs to the extracellular portion of the protein, which binds gabapentin in the absence of the transmembrane domain (8), and requires several individual identified amino acid residues (100).

Autoradiographic studies (31) show that specific binding of gabapentin is highest in the superficial layers of neocortex and dendritic layers of hippocampus, with low levels of binding in white matter and brainstem (Figure 28.6B). Thus, gabapentin binding is densest in areas rich in synapses. [3H]-Gabapentin binding is displaced by unlabeled gabapentin and by several structural analogs of gabapentin including pregabalin and additional gabapentin and pregabalin analogs (10,88). The two enantiomers of pregabalin have different potencies for binding at the gabapentin site, and the same stereospecificity for these enantiomers is seen in anticonvulsant studies with whole animals (94). Together, these findings indicate that the anticonvulsant actions of gabapentin and chemically similar compounds relate to binding at the [3H]-gabapentin site.


Gabapentin has significant anticonvulsant action in well-controlled clinical studies. The profile of anticonvulsant activity in a variety of animal models suggests that gabapentin differs pharmacologically from other anticonvulsants. Several in vitro actions of gabapentin are summarized in Table 28.3, along with information on concentration dependence that helps to assess the relevance for anticonvulsant or behavioral actions. Although numerous pharmacologic actions have been reported in vitro, it is not yet clear which actions are the most relevant for preventing seizures. The access of gabapentin from the gut to the brain depends on system L amino acid transport. Gabapentin is a potent ligand for displacement of [3H]-gabapentin binding


to the α2δ protein, and this binding activity appears to correlate with anticonvulsant activity (9,10).


Test System

Primary Observation

Concentration of Gabapentin

Direct Action?


[3H]-Gabapentin binding to α28 protein

Specific binding inhibited by gabapentin and derivatives, L-leucine

IC50 = 0.8 µmol/L

Yes, also occurs with binding to solubilized or recombinant protein


Action potential firing (cultured mouse spinal neurons)

Reduced action potential firing (prolonged treatment)

IC50 = 20 µmol/L for application times 60 min

Probably indirect, [voltage-clamped Na+ currents not blocked (81)]


Monoamine neurotransmitter release (striatal or neocortical brain slices)

10%-40% Decrease in release of [3H]-norepinephrine, [3H]-dopamine, and [3H]-serotonin

10-100 µmol/L (IC50 ~10 µmol)

Direct action on Ca2+ channels (?) or vesicle release machinery (?)

15,68, 75

Ca2+ fluorescence and [3H]-norepinephrine release in synaptosomes

Reduced release and fluorescence by 15%-30%

~10 µmol/L

Direct action on Ca2+ channels (?)

21a,51, 97

[3H]-Glutamate release in trigeminal nucleus slices

Reduced in tissue pretreated with substance P

~10 µmol/L

May be a direct action on Ca2+channels or vesicle release machinery


Ca2+ channel function (voltage clamp) with cultured or acutely isolated neurons

~15%-50% decrease in current

1-100 µmol/L

Direct effect? may require activation of protein kinase A

2, 80,81, 89

Reduced synaptic currents in voltage-clamped neuronal tissue slices

Up to 50% decrease in synaptic currents

10-100 µmol/L

Unknown, presynaptic (?)

54a,62, 7797

Enhanced reversal of GABA transport (rat optic nerve and hippocampal slices)

Increase in electrophysiologic response to 1 mmol/L nipecotic acid or increased GABAAcurrent

100 µmol/L


32,33, 38

Enhanced GABA transporter activity in cultured neurons

100% increase in initial rate of GABA transport

EC50 approx. 20 µmol/L



Cell death from blockade of glutamate uptake (rat spinal cord cultures)

Reduced cell death of motor neurons

100 µmol/L

Blockade of BCAA-T and reduced glutamate synthesis (?)


NMDA response (voltage clamp with isolated spinal cord neurons)

25% Increase in NMDA current (shift in glycine dependence)

100 µmol/L

Probably indirect, requires protein kinase C catalytic subunit

27, 69

Glutamic acid decarboxylase enzyme activity

45% Enhanced Vmax, no change in KM (58)

1 mmol/L

Apparently; occurs in partially purified enzyme in vitro


BCAA-T enzyme activity

Competitive inhibition

Ki = 0.7 mmol/L, calculated inhibition in vivo ~10%

Apparently; occurs in partially purified enzyme in vitro

25,35, 41

GABA transaminase enzyme activity

Mixed-type competitive and noncompetitive inhibition

Ki = 25 mmol/L, calculated inhibition in vivo <1%

Apparently; occurs in partially purified enzyme in vitro


GABA, γ-aminobutyric acid; BCAA-T, branched-chain amino acid aminotransferase; NMDA, N-methyl-D-aspartate; Vmax, rate constant of enzyme activity; KM, affinity constant of enyme activity; Ki, affinity constant of inhibition.

Gabapentin has a pronounced three-dimensional structural similarity to L-leucine (Figure 28.1), and gabapentin mimics the action of L-leucine at several sites, such as [3H]-gabapentin binding, system L transport, BCAA-T enzyme, and GDH enzyme. In contrast to L-leucine, gabapentin does not appear to be metabolized by enzymes of brain cytosol. It seems unlikely that L-leucine-like actions are necessary for the anticonvulsant effects of gabapentin, because other L-leucine analogs such as BCH do not share the anticonvulsant activity of gabapentin.

The findings of Kocsis et al. (38,54,32,33) suggest that gabapentin increases the nonsynaptic release of GABA from neuronal tissues, and the human studies of Petroff et al. (63, 64) indicate that gabapentin increases brain GABA concentrations. It is tempting to speculate that changes in GABA transporter function or GABA cellular concentration may be involved in the molecular basis of these effects. Numerous studies have addressed effects of gabapentin on sodium channels,


calcium channels, and NMDA receptors. To date, these have not offered a single consistent view of gabapentin action. However, it is hoped that additional studies soon will give a more complete picture of the cellular and molecular basis of the anticonvulsant action of gabapentin. Such studies may also give insight into basic mechanisms involved in the disease of epilepsy and the onset of seizures.


I wish to thank many colleagues and investigators who have studied gabapentin including Mark Vartanian, Barbra Stewart, Ti-Zhi Su, Arie Goldlust, Dale Oxender, David Rock, Greg Campbell, Mark Weber, Sean Donevan, David Hill, Jason Brown, Nick Gee, David Dooley, Susan Hutson, Kay LaNone, Kathy Sutton, Roland Jones, Sean Donevan, Jim Offord, Michael Quick, Jeff Kocsis, Leonard Meltzer, Beth Lunney, Sandy McKnight, Devin Welty, Janet Stringer, Ognen Petroff, Jeff Kocsis, and George Richerson.


  1. Abdi S, Lee DH, Chung JM. The anti-allodynic effects of amitriptyline, gabapentin, and lidocaine in a rat model of neuropathic pain. Anesth Analg1998;87:1360-1366.
  2. Alden KJ, Garcia J. Differential effect of gabapentin on neuronal and muscle calcium currents. J Pharmacol Exp Ther2001;297:727-735.
  3. Andrews J, Chadwick D, Bates D. Gabapentin in partial epilepsy. Lancet1990;335:1114-1117.
  4. Backonja M, Beydoun A, Edwards KR, et al. Gabapentin for the symptomatic treatment of painful neuropathy in patients with diabetes mellitus. JAMA1998;280:1831-1836.
  5. Bartoszyk GD, Meyerson N, Reimann W, et al. Gabapentin. In: Meldrum BS, Porter RJ, eds. Curent Problems in Epilepsy, Vol 4, London, John Libbey, P., 1986; 147-163.
  6. Brookes N. Interaction between the glutamine cycle and the uptake of large neutral amino acids in astrocytes. J Neurochem1993;60:1923-1928.
  7. Brown JP, Dissanayake VU, Briggs AR, et al. Isolation of the [3H]gabapentin-binding protein/alpha 2 delta Ca2+channel subunit from porcine brain: development of a radioligand binding assay for alpha 2 delta subunits using [3H]leucine. Anal Biochem 1998;255:236-243.
  8. Brown JP, Gee NS. Cloning and deletion mutagenesis of the alpha2 delta calcium channel subunit from porcine cerebral cortex: expression of a soluble form of the protein that retains [3H]gabapentin binding activity. J Biol Chem1998;273:25458-25465.
  9. Bryans JS, Davies N, Gee NS, et al. Identification of novel ligands for the gabapentin binding site on the alpha2delta subunit of a calcium channel and their evaluation as anticonvulsant agents. J Med Chem1998;41:1838-1845.
  10. Bryans JS, Wustrow DJ. 3-substituted GABA analogs with central nervous system activity: a review. Med Res Rev1999;19:149-177.
  11. Cho SW, Cho EH, Choi SY. Activation of two types of brain glutamate dehydrogenase isoproteins by gabapentin. FEBS Lett1998;426:196-200.
  12. Christensen HN. Role of amino acid transport and countertransport in nutrition and metabolism. Pharmacol Rev1990;70:43-77.
  13. Cutter NC, Scott DD, Johnson JC, et al. Gabapentin effect on spasticity in multiple sclerosis: a placebo-controlled, randomized trial. Arch Phys Med Rehabil2000;81:164-169.
  14. Diener HC, Hacke W, Hennerici M, et al. Lubeluzole in acute ischemic stroke: a double-blind, placebo-controlled phase II trial. Stroke1996;27:76-81.
  15. Dooley DJ, Donovan CM, Pugsley TA. Stumulus-dependent modulation of [3H]norepinephrine release from rat neocortical slices by gabapentin and pregabalin. J Pharmacol Exp Ther2000;296:1086-1098.
  16. Dooley DJ, Mieske CA, Borosky SA. Inhibition of K(+)-evoked glutamate release from rat neocortical and hippocampal slices by gabapentin. Neurosci Lett2000;280:107-110.
  17. During M, Ryder KM, Spencer DD. Hippocampal GABA transporter function in temporal-lobe epilepsy. Nature1995;376:174-177.
  18. Ericinska M, Nelson D. Activation of glutamate dehydrogenase by leucine and its nonmetalizable analogue in rat brain synaptosomes. J Neurochem1990;54:1335-1343.
  19. Ericinska M, Nelson D, Wilson DF, et al. Neurotransmitter amino acids in the CNS. I. Regional changes in amino acid levels in rat brain during ischemia and reperfusion. Brain Res1990;304:9-22.
  20. Felix R. Voltage-dependent Ca2+ channel alpha2delta auxiliary subunit: structure, function and regulation. Receptors Channels1999;6:351-362.
  21. Field MJ, Oles RJ, Lewis AS, et al. Gabapentin (Neurontin) and S-(+)-3-isobutyl GABA represent a novel class of selective antihyperalgesic agents. Br J Pharmacol1997;121:1513-1522.

21a. Fink K, dolley DJ, Meder WP, et al. Inhibition of neuronal Ca2+ influs by gabapentin and pregabalin in the human cortex. Neuropharmacol 2002;42:229-236.

  1. Gaspary HL, Wang W, Richerson GB. Carrier-mediated GABA release activates GABA receptors on hippocampal neurons. J Neurophysiol1998;80:270-281.
  2. Gee NS, Brown JP, Dissanayake VU, et al. The novel anticonvulsant drug, gabapentin (Neurontin), binds to the alpha2delta subunit of a calcium channel. J Biol Chem1996;271:5768-5776.
  3. Goa KL, Sorkin EM. Gabapentin: a review of its pharmacological properties and clinical potential in epilepsy. Drugs1993;46:409-427.
  4. Goldlust A, Su TZ, Welty DF, et al. Effects of anticonvulsant drug gabapentin on the enzymes in metabolic pathways of glutamate and GABA. Epilepsy Res1995;22:1-11.
  5. Gruenthal M, Mueller M, Olson WL, et al. Gabapentin for the treatment of spasticity in patients with spinal cord injury. Spinal Cord1997;35:686-689.
  6. Gu Y, Huang L-YM. Gabapentin actions on NMDA channels are PKC dependent. Mol Pharmacol2001.
  7. Gurney ME, Cutting FB, Zhai P, et al. Benefit of vitamin E, riluzole, and gabapentin in a transgenic model of familial amyotrophic lateral sclerosis. Ann Neurol1996;39:147-157.
  8. Haas HL, Wieser HG. Gabapentin: action on hippocampal slices of the rat and effects in human epilepticss. Paper presented at the Northern European Epilepsy Meeting, York, UK, 1986.
  9. Harrison NL, Simmonds MA. Quantitative studies on some antagonists of N-methyl-D-aspartate in slices of rat cerebral cortex. Br J Pharmacol1985;84:381-391.
  10. Hill DR, Suman CN, Woodruff GN. Localization of [3H]gabapentin to a novel site in rat brain: autoradiographic studies. Eur J Pharmacol1993;244:303-309.
  11. Honmou O, Kocsis JD, Richerson GB. Gabapentin potentiates the conductance increase induced by nipecotic acid in CA1 pyramidal neurons in vitro. Epilepsy Res1995;20:193-202.



  1. Honmou O, Oyelese AA, Kocsis JD. The anticonvulsant gabapentin enhances promoted release of GABA in hippocampus: a field potential analysis. Brain Res1995;692:273-277.
  2. Hosford DA, Wang Y. Utility of the lethargic (lh/lh) mouse model of absence seizures in predicting the effects of lamotrigine, vigabatrin, tiagabine, gabapentin, and topiramate against human absence seizures. Epilepsia1997;38:408-414.
  3. Hutson SM, Berkich D, Drown P, et al. Role of branched-chain aminotransferase isoenzymes and gabapentin in neurotransmitter metabolism. J Neurochem1998;71:863-874.
  4. Jun JH, Yaksh TL. The effect of intrathecal gabapentin and 3-isobutyl gamma-aminobutyric acid on the hyperalgesia observed after thermal injury in the rat. Anesth Analg1998;86:348-354.
  5. Klugbauer N, Lacinova L, Marais E, et al. Molecular diversity of the dalcium channel alpha-2-delta subunit. J Neurosci1999;19:684-691.
  6. Kocsis JD, Honmou O. Gabapentin increases GABA-induced depolarization in rat neonatal optic nerve. Neurosci Lett1994;169:181-184.
  7. Krall RL, Penry JK, White BG, et al. Antiepileptic drug development. II. Anticonvulsant drug screening. Epilepsia1978;19:409-428.

39a. Lanneau C, Green A, Hirst WD, et al. Gabapentin is not a GABAB receptor agonist. Neuropharmacol 2001;41:965-975.

  1. Larsson OM, et al. Differential effect of gamma-vinyl GABA and valproate on GABA-transaminase from cultured neurons and astrocytes. Neuropharmacology1986;25:617-625.
  2. Lieth E, LaNoue KF, Berkich D, et al. Nitrogen shuttling between neurons and glial cells during glutamate synthesis. J Neurochem2001;76:1712-1723.
  3. Lippert B, Metcalf BW, Jung MJ, et al. 4-Amino-hex-5-enoic acid, a selective catalytic inhibitor of 4-aminobutyric-acid aminotransferase in mammalian brain. Eur J Biochem1977;74:441-445.
  4. Loscher W, Honack D, Taylor CP. Gabapentin increases aminooxyacetic acid-induced GABA accumulation in several regions of rat brain. Neurosci Lett1991;128:150-154.
  5. Loscher W, Reissmuller E, Ebert U. Anticonvulsant efficacy of gabapentin and levetiracetam in phenytoin-resistant kindled rats. Epilepsy Res2000;40:63-77.
  6. Löscher W. Valproate enhances GABA turnover in the substantia nigra. Brain Res1989;501:198-203.
  7. Luecke A, Musshoff U, Koehling R, et al. Gabapentin potentiation of the antiepileptic efficacy of vigabatrin in an in vitro model of epilepsy. Br J Pharmacol1998;124:370-376.
  8. Luer MS, Hamani C, Dujovny M, et al. Saturable transport of gabapentin at the blood-brain barrier. Neurol Res1999;21:559-562.
  9. Maneuf YP, McKnight AT. Gabapentin inhibits substance P-and calcitonin gene-related peptide-facilitated K+-evoked release of [3H]-glutamate from rat caudal trigeminal nucleus slices. Soc Neurosci Abstr2000;26:1931-1931.
  10. Marais E, Klugbauer N, Hofmann F. Calcium channel alpha-2-delta subunits: structure and gabapentin binding. Mol Pharmacol2001;59:1243-1248.
  11. McLean MJ, Ramsey RE, Leppik I, et al. Gabapentin as add-on therapy in refractory partial epilepsy: a double-blind, placebo-controlled, parallel-group study. Neurology1993;43:2292-2298.
  12. Meder WP, Dooley DJ. Modulation of K+-induced synaptosomal calcium influx by gabapentin. Brain Res2000;875:157-159.
  13. Meldrum BS. GABAergic mechanisms in the pathogenesis and treatment of epilepsy. Br J Clin Pharmacol1989;27:3S-11S.
  14. Michelitti G, Vergnes M, Marescaux C, et al. Antiepileptic drug evaluation in a new animal model: spontaneous petit mal epilepsy in the rat. Arzneimittelforshung1985;35:483-485.
  15. Mirchandani GR, Agulian S, Abi-Saab W, et al. Vigabatrin increases total GABA levels in rat optic nerve while gabapentin and pregabalin do not. Soc Neurosci Abstr1999;25:1868-1868.

54a. Misner DL, Kansagara AG, Bonhaus DW. Effects of gabapentin on hippocampal CA1 neurons. Soc Neurosci Abstr 2001;27:711.5.

  1. Ng GYK, Bertrand S, Sullivan R, et al. Gamma-aminobutyric acid type B receptors with specific heterodimer composition and postsynaptic actions in hippocampal neurons are targets of anticonvulsant gabapentin action. Mol Pharmacol2001;59:144-152.
  2. Ochi S, Lim JY, Rand MN, et al. Transient presence of GABA in astrocytes of the developing optic nerve. Glia1993;9:188-198.
  3. Otsuki K, Morimoto K, Sato K, et al. Effects of lamotrigine and conventional antiepileptic drugs on amygdala- and hippocampal-kindled seizures in rats. Epilepsy Res1998;31:101-112.
  4. Palfreyman MG, Schechter PJ, Buckett WR, et al. The pharmacology of GABA-transaminase inhibitiors. Biochem Pharmacol1981;30:817-824.
  5. Pan H-L, Chen S-R, Eisenach JC. Gabapentin suppresses ectopic nerve discharges and reverses allodynia in neuropathic rats. J Pharmacol Exp Ther1998;288:1026-1030.
  6. Pande AC, Davidson JR, Jefferson JW, et al. Treatment of social phobia with gabapentin: a placebo-controlled study. J Clin Psychopharmacol1999;19:341-348.
  7. Partridge BJ, Chaplan SR, Sakamoto E, et al. Characterization of the effects of gabapentin and 3-isobutyl-gamma-aminobutyric acid on substance P-induced thermal hyperalgesia. Anesthesiology1998;88:196-205.
  8. Patel MK, Gonzalez MI, Bramwell S, et al. Gabapentin inhibits excitatory synaptic transmission in the hyperalgesic spinal cord. Br J Pharmacol2000;130:1731-1734.
  9. Petroff OA, Hyder F, Rothman DL, et al. Effects of gabapentin on brain GABA, homocarnosine, and pyrrolidinone in epilepsy patients. Epilepsia2000;41:675-680.
  10. Petroff OA, Rothman DL, Behar KL, et al. The effect of gabapentin on brain gamma-aminobutyric acid in patients with epilepsy. Ann Neurol1996;39:95-99.
  11. Pin JP, Bockaert J. Two distinct mechanisms, differentially affected by excitatory amino acids, trigger GABA release from fetal mouse striatal neurons in primary culture. J Neurosci1989;9:648-656.
  12. Priebe MM, Sherwood AM, Graves DE, et al. Effectiveness of gabapentin in controlling spasticity: a quantitative study. Spinal Cord1997;35:171-175.
  13. Ragsdale DS, Scheuer T, Catterall WA. Frequency and voltage-dependent inhibition of type IIA Na+channels, expressed in a mammalian cell line, by local anesthetic, antiarrhythmic, and anticonvulsant drugs. Mol Pharmacol 1991;40:756-765.
  14. Reimann W. Inhibition by GABA, baclofen and gabapentin of dopamine release from rabbit caudate nucleus: are there common or different sites of action? Eur J Pharmacol1983;94:341-344.
  15. Rock DM, Kelly KM, Macdonald RL. Gabapentin actions on ligand- and voltage-gated responses in cultured rodent neurons. Epilepsy Res1993;16:89-98.
  16. Rogawski MA, Porter RJ. Antiepileptic drugs: pharmacological mechanisms and clinical efficacy with consideration of promising developmental stage compounds. Pharmacol Rev1990;42:223-286.
  17. Rothstein JD, Kuncl RW. Neuroprotective strategies in a model of chronic glutamate-mediated motor neuron toxicity. J Neurochem1995;65:643-651.
  18. Rothstein JD. Therapeutic horizons of amyotrophic lateral sclerosis. Curr Opinions Neurobiol1996;6:679-687.



  1. Rowbotham MC, Harden N, Stacey B, et al. Gabapentin for the treatment of postherpetic neuralgia: a randomized controlled trial. JAMA1998;280:1837-1842.
  2. Satzinger G. Antiepileptics from gamma-aminobutyric acid. Arzneimittelforschung1994;44:261-266.
  3. Schlicker E, Reimann W, Gothert M. Gabapentin decreases monoamine release without affecting acetylcholine release in the brain. Arzheimittelforschung1985;35:1347-1349.
  4. Schumacher TB, Beck H, Steinhauser C, et al. Effects of gabapentin, phenytoin and carbamazepine on calcium currents in hippocampal granule cells from patients with temporal lobe epilepsy. Epilepsia1997.
  5. Shimoyama M, Shimoyama N, Hori Y. Gabapentin affects glutamatergic excitatory neurotransmission in the rat dorsal horn. Pain2000;85:405-414.
  6. Singh L, Field MJ, Ferris P, et al. The antiepileptic agent gabapentin (Neurontin) possesses anxiolytic-like and antinociceptive actions that are reversed by D-serine.Psychopharmacology (Berl)1996;127:1-10.
  7. Stasheff SF, Bragdon AC, Wilson WA. Induction of epileptiform activity in hippocampal slices by trains of electrical stimuli. Brain Res1985;344:296-301.
  8. Stefani A, Spadoni F, Bernardi G. Gabapentin inhibits calcium currents in isolated rat brain neurons. Neuropharmacology1998;37:83-91.
  9. Stefani A, Spadoni F, Giacomini P, et al. The effects of gabapentin on different ligand- and voltage-gated currents in isolated cortical neurons. Epilepsy Res2001;43:239-248.
  10. Stewart BH, Kugler AR, Thompson PR, et al. A saturable transport mechanism in the intestinal absorption of gabapentin is the underlying cause of the lack of proportionality between increasing dose and drug levels in plasma. Pharm Res1993;10:276-281.
  11. Stewart BH, Reyner EL, Lu RH. Mechanism of gabapentin (Neurontin) transport across monolayers of human colon adenocarcinoma cells (CACO-2). Amino Acids1993;5:204-204.
  12. Stringer JL, Lorenzo N. The reduction in paired-pulse inhibition in the rat hippocampus by gabapentin is independent of GABA(B) receptor activation. Epilepsy Res1999;33:93-97.
  13. Stringer JL, Lothman EW. Maximal dentate activation: a tool to screen compounds for activity against limbic seizures. Epilepsy Res1990;5:169-176.
  14. Stringer JL, Taylor CP. The effects of gabapentin in the rat hippocampus are mimicked by two structural analogs, but not by nimodipine. Epilepsy Res2000;41:155-162.
  15. Su TZ, Lunney E, Campbell G, et al. Transport of gabapentin, a gamma-amino acid drug, by system 1 alpha-amino acid transporters: a comparative study in astrocytes, synaptosomes, and CHO cells. J Neurochem1995;64:2125-2131.
  16. Suman CN, Webdale L, Hill DR, et al. Characterisation of [3H]gabapentin binding to a novel site in rat brain: homogenate binding studies. Eur J Pharmacol1993;244:293-301.
  17. Sutton KG, Scott RH, Lee K, et al. Gabapentin inhibits high threshold calcium channel currents in cultured dorsal root ganglion neurones. Soc Neurosci Abstr2000;26:234.4.
  18. Taylor CP. The anticonvulsant lamotrigine blocks sodium currents from cloned alpha-subunits of rat brain Na+channels in a voltage- dependent manner but gabapentin does not. Soc Neurosci Abstr 1993;23:1631.
  19. Taylor CP. Mechanisms of new antiepileptic drugs,3rd ed. Delgado-Escuelo AU, Wilson WA, Oslen RW, et al, eds. Philadelphia; Lippincott Williams & Wilkins, 1999:1011-1026.
  20. Taylor CP, Gee NS, Su TZ, et al. A summary of mechanistic hypotheses of gabapentin pharmacology. Epilepsy Res1998;29:233-249.
  21. Taylor CP, Vartanian MG, Andruszkiewicz R, et al. 3-alkyl GABA and 3-alkylglutamic acid analogues: two new classes of anticonvulsant agents. Epilepsy Res1992;11:103-110.
  22. Taylor CP, Vartanian MG, Yuen PW, et al. Potent and stereospecific anticonvulsant activity of 3-isobutyl GABA relates to in vitro binding at a novel site labeled by tritiated gabapentin. Epilepsy Res1993;14:11-15.
  23. Thurlow RJ, Hill DR, Woodruff GN. Comparison of the uptake of [3H]-gabapentin with the uptake of L-[3H]-leucine into rat brain synaptosomes. Br J Pharmacol1996;118:449-456.
  24. Upton N. Mechanisms of action of new antiepileptic drugs: rational design and serendipitous findings. Trends Pharmacol Sci1994;15:456-463.
  25. vanHooft JA, Dougherty D, Endeman D, et al. Gabapentin inhibits presynaptic but not postsynaptic voltage-operated calcium channels in rat neocortex and hippocampus. Soc Neurosci Abstr2000;26:662.7.

97a. Vartanian MG, Donovan DM, Weber ML, et al. Gabapentin does not interact with the GABAB receptor. Soc Neurosci Abstr 2001;27:603.3.

  1. Vollmer KO, vonHodenberg A, Koelle EU. Pharmacokinetics and metabolism of gabapentin in rat, dog and man. Arzneimittelforschung1986;36:830-839.
  2. Wamil AW, McLean MJ. Limitation by gabapentin of high frequency action potential firing by mouse central neurons in cell culture. Epilepsy Res1994;17:1-11.
  3. Wang M, Offord J, Oxender DL, et al. Structural requirement of the calcium-channel subunit alpha2delta for gabapentin binding. Biochem J1999;342:313-320.
  4. Welty DF, Schielke GP, Rothstein JD. Potential treatment of amyotrophic lateral sclerosis by the anticonvulsant gabapentin: a hypothesis. Ann Pharmacother1995;29:1164-1167.
  5. Welty DF, Schielke GP, Vartanian MG, et al. Gabapentin anticonvulsant action in rats: disequilibrium with peak drug concentrations in plasma and brain microdialysate.Epilepsy Res1993;16:175-181.
  6. Welty DF, Wang Y, Busch JA, et al. Pharmacokinetics and pharmacodynamics of CI-1008 (pregabalin) and gabapentin in rats using maximal electroshock. Epilepsia1997;35-36.

103a. Whitworth TL, Quick MW. Upregulation of γ-aminobutyric acid transporter expression: role of alkylated γ-aminobutyric acid derivatives. Bioch Soc Trans 2001;29:736-741.

  1. Williamson J, Lothman EW, Taylor CP, et al. Comparison of S-(+)-3-isobutyl GABA and gabapentin against kindled hippocampal seizures. Epilepsia1997;38:29-29.
  2. Willow M, Catterall WA. Inhibition of the binding of [3H]batrachotoxinin A 20-α-benzoate to sodium channels by the anticonvulsant drugs diphenylhydantoin and carbamazepine. Mol Pharmacol1982;22:627-635.
  3. Wu Y, Wang W, Richerson GB. GABA transaminase inhibition induces spontaneous and enhances depolarization-evoked GABA efflux through reversal of the GABA transporter.J Neurosci2001;21:2630-2639.
  4. Xiao W-H, Bennett GJ. Gabapentin has an antinocioceptive effect mediated via a spinal site of action in a rat model of painful peripheral neuropathy. Analgesia1996;2:267-273.
  5. Xiong Z-Q, Stringer JL. Effects of felbamate, gabapentin and lamotrigine on seizure parameters and excitability in the rat hippocampus. Epilepsy Res1997;27:187-194.