A microdialysis study on the mechanism of action of gabapentin

Gabapentin inhibits γ-amino butyric acid release in the locus coeruleus but not in the spinal dorsal horn after peripheral nerve injury in rats

BACKGROUND

Gabapentin reduces acute postoperative and chronic neuropathic pain, but its sites and mechanisms of action are unclear. Based on previous electrophysiologic studies, the authors tested whether gabapentin reduced γ-amino butyric acid (GABA) release in the locus coeruleus (LC), a major site of descending inhibition, rather than in the spinal cord.

METHODS

Male Sprague-Dawley rats with or without L5-L6 spinal nerve ligation (SNL) were used. Immunostaining for glutamic acid decarboxylase and GABA release in synaptosomes and microdialysates were examined in the LC and spinal dorsal horn.

RESULTS

Basal GABA release and expression of glutamic acid decarboxylase increased in the LC but decreased in the spinal dorsal horn after SNL. In microdialysates from the LC, intravenously administered gabapentin decreased extracellular GABA concentration in normal and SNL rats. In synaptosomes prepared from the LC, gabapentin and other α2δ ligands inhibited KCl-evoked GABA release in normal and SNL rats. In microdialysates from the spinal dorsal horn, intravenous gabapentin did not alter GABA concentrations in normal rats but slightly increased them in SNL rats. In synaptosomes from the spinal dorsal horn, neither gabapentin nor other α2δ ligands affected KCl-evoked GABA release in normal and SNL rats.

DISCUSSION

These results suggest that peripheral nerve injury induces plasticity of GABAergic neurons differently in the LC and spinal dorsal horn and that gabapentin reduces presynaptic GABA release in the LC but not in the spinal dorsal horn. The current study supports the idea that gabapentin activates descending noradrenergic inhibition via disinhibition of LC neurons.

A comparison of the pharmacokinetics and pharmacodynamics of pregabalin and gabapentin

Pregabalin and gabapentin share a similar mechanism of action, inhibiting calcium influx and subsequent release of excitatory neurotransmitters; however, the compounds differ in their pharmacokinetic and pharmacodynamic characteristics. Gabapentin is absorbed slowly after oral administration, with maximum plasma concentrations attained within 3-4 hours. Orally administered gabapentin exhibits saturable absorption--a nonlinear (zero-order) process--making its pharmacokinetics less predictable. Plasma concentrations of gabapentin do not increase proportionally with increasing dose. In contrast, orally administered pregabalin is absorbed more rapidly, with maximum plasma concentrations attained within 1 hour. Absorption is linear (first order), with plasma concentrations increasing proportionately with increasing dose. The absolute bioavailability of gabapentin drops from 60% to 33% as the dosage increases from 900 to 3600 mg/day, while the absolute bioavailability of pregabalin remains at > or = 90% irrespective of the dosage. Both drugs can be given without regard to meals. Neither drug binds to plasma proteins. Neither drug is metabolized by nor inhibits hepatic enzymes that are responsible for the metabolism of other drugs. Both drugs are excreted renally, with elimination half-lives of approximately 6 hours. Pregabalin and gabapentin both show dose-response relationships in the treatment of postherpetic neuralgia and partial seizures. For neuropathic pain, a pregabalin dosage of 450 mg/day appears to reduce pain comparably to the predicted maximum effect of gabapentin. As an antiepileptic, pregabalin may be more effective than gabapentin, on the basis of the magnitude of the reduction in the seizure frequency. In conclusion, pregabalin appears to have some distinct pharmacokinetic advantages over gabapentin that may translate into an improved pharmacodynamic effect.

Effects of gabapentin on cocaine self-administration, cocaine-triggered relapse and cocaine-enhanced nucleus accumbens dopamine in rats

Gabapentin is a gamma-aminobutyric acid (GABA) analogue, with GABAmimetic pharmacological properties. Gabapentin is used for the treatment of seizures, anxiety and neuropathic pain. It has been proposed that gabapentin may be useful in the treatment of cocaine dependence. However, clinical trials with gabapentin have shown conflicting results, while preclinical studies are sparse. In the present study, we investigated the effects of gabapentin on intravenous cocaine self-administration and cocaine-triggered reinstatement of drug-seeking behavior, as well as on cocaine-enhanced dopamine (DA) in the nucleus accumbens (NAc). We found that gabapentin (25-200 mg/kg, i.p., 30 min or 2 h prior to cocaine) failed to inhibit intravenous cocaine (0.5 mg/kg/infusion) self-administration under a fixed-ratio reinforcement schedule or cocaine-triggered reinstatement of cocaine-seeking behavior. In vivo microdialysis showed that the same doses of gabapentin produced a modest increase (approximately 50%, p<0.05) in extracellular NAc GABA levels, but failed to alter either basal or cocaine-enhanced NAc DA. These data suggest that gabapentin is a weak GABA-mimic drug. At the doses tested, it has no effect in the addiction-related animal behavioral models here tested. This is in striking contrast to positive findings in the same animal models shown by another GABAmimetic--gamma-vinyl GABA (see companion piece to present article).

Posttraining administration of pentylenetetrazol dissociates gabapentin effects on memory consolidation from that on memory retrieval process in mice

Gabapentin (GBP), an anticonvulsant drug, 10 mg/kg, i.p., but not 100 mg/kg, i.p., enhanced retention of an inhibitory avoidance task when given 20 min after training, as indicated by retention performance 48 h later. The immediate post-training administration of pentylenetetrazol (PTZ, 45 mg/kg, i.p.) impaired retention performance. The amnesic effects of the convulsant drug PTZ were not influenced by GBP at any level of doses. However, GBP 100 mg/kg, but not 10 mg/kg, delayed the latency to first clonic body seizures and decreased the duration of convulsion induced by PTZ. The enhancing effect of GBP on retention was not prevented by the opiate receptor antagonist, naltrexone (0.01 mg/kg, i.p.), which completely prevented the impairment of retention caused by PTZ. Further, naltrexone did not modify the convulsions induced by PTZ. In mice pretreated with naltrexone and that received PTZ, the administration of GBP again, enhanced retention performance during the retention test. Since previous results indicate that the amnesic action of PTZ are due to an effect on memory retrieval, the present results provide additional pharmacological evidence suggesting that GBP influenced memory consolidation and not memory retrieval of an inhibitory avoidance task in mice.

Reduction by gabapentin of K+-evoked release of [3H]-glutamate from the caudal trigeminal nucleus of the streptozotocin-treated rat

Recently, we showed that gabapentin can inhibit a facilitatory effect of substance P (SP) on K(+)-evoked glutamate release in rat trigeminal slices (Maneuf et al., 2001), and we have now examined the effect of gabapentin on glutamate release in the trigeminal slice from the streptozotocin (STZ)-treated rat. 1. At 4 weeks following STZ treatment (50 mg kg(-1) i.p.), blood glucose was increased in the majority of cases, compared to the control level. All the treated animals showed a significant degree (P<0.001) of tactile allodynia (assessed using von Frey filaments) that did not appear to correlate with blood glucose levels. 2. In this study, we demonstrated that, after STZ treatment, 30 microM gabapentin reduced K(+)-evoked release of [(3)H]-glutamate in either normal (11 mM) or high (30 mM) glucose conditions by 24 and 22%, respectively. In the normal rat, gabapentin (up to 100 microM) is ordinarily unable to affect release of glutamate from the trigeminal slice. 3. The uptake of glutamate in Sp5C punches from streptozotocin-treated rats was reduced under normal glucose conditions (41.7% of control), whereas high glucose restored uptake to normal (84.7% of control). 4. The addition of 1 microm substance P potentiated the evoked release of glutamate in both normal (40% increase) and high glucose (28%), and this was blocked by gabapentin (30 microM) in both conditions. It is interesting to speculate that this ability of gabapentin to reduce the release of glutamate in the trigeminal nucleus after streptozotocin treatment may be of relevance to the antihyperalgesic-allodynic actions of the drug.

Gabapentin-- actions on adult superficial dorsal horn neurons

Despite identification of GABA(B) receptors with gb1a-gb2 composition and the alpha2delta calcium channel subunit as putative molecular targets for gabapentin (GBP), its cellular mechanism of action has remained elusive. Therefore, we have used an in vitro spinal cord slice preparation to study the effects of GBP on lamina II neurons. The frequency and amplitude of spontaneous EPSCs and IPSCs were unaffected by GBP, suggesting presynaptic neurotransmitter release is not regulated. Direct modulation of postsynaptic membrane excitability is also unlikely since the level of holding current required to maintain neurons at -70, 0 and +45 mV was unaffected by GBP. Effects on excitatory and inhibitory synaptic transmission were variable across the population. Primary afferent-evoked fast glutamatergic EPSCs were unaffected by GBP, while evoked NMDA receptor-mediated EPSCs and IPSCs were variably affected. In contrast, GBP enhanced responses to bath applied NMDA in 71% of neurons. Thus, in adult rat dorsal horn, synaptic and extrasynaptic NMDA receptors may be differentially regulated by GBP perhaps due to differences in subunit composition.