Petra Bloms-Funke

The novel antiepileptic drug levetiracetam (ucb L059) induces alterations in GABA metabolism and turnover in discrete areas of rat brain and reduces neuronal activity in substantia nigra pars reticulata

Levetiracetam ((S)-alpha-2-oxo-pyrrolidine acetamide, ucb L059) is a novel anticonvulsant drug presently in clinical development. Its mechanism of action is unknown although a recently novel specific binding site for [3H]levetiracetam, unique to brain, may be involved. This binding site has yet been characterized, but some evidence suggested a possibly indirect interaction with the GABA system. We therefore examined levetiracetams effects on GABA metabolism and turnover in several rat brain regions after systemic administration of anticonvulsant doses. Furthermore, in order to study functional effects of levetiracetam on a well defined system of GABAergic neurons in a brain region that has been critically involved in anticonvulsant drug action, we examined levetiracetams action on spontaneous firing of substantia nigra pars reticulata (SNR) neurons in anesthetized rats. Although levetiracetam did not alter the activity of the GABA synthesizing and degrading enzymes glutamic acid decarboxylase (GAD) and GABA aminotransferase (GABA-T) in vitro, systemic administration induced significant alterations in these enzymes in several brain regions, indicating that these enzyme alterations were no direct drug effects but a consequence of postsynaptic changes in either GABAergic or other neurotransmitter-related systems. In the striatum, levetriacetam, 170 mg/kg i.p., induced a significant increase in GABA-T activity while GAD activity markedly decreased. When GABA turnover was estimated after inhibition of GABA-T by aminooxyacetic acid (AOAA), treatment with levetiracetam (given 15 min prior to injection of AOAA) significantly reduced GABA turnover in the striatum. Since the substantia nigra pars reticulata (SNR) receives a strong GABAergic input from the striatum, we examined if the alterations in GABA metabolism and turnover in the striatum led to functional alterations in neuronal activity in the SNR by recording single unit activity of SNR neurons after i.p. injection of levetiracetam. While injection of vehicle did not affect SNR neuronal activity, a significant decrease in spontaneous neuronal firing was recorded after levetiracetam. Since a substantial body of evidence suggests that the SNR is a critical site at which decrease of neuronal firing results in protection against various seizure types, the suppressive effect of levetiracetam on SNR activity may contribute to the anticonvulsant action of this drug.

The anticonvulsant gabapentin decreases firing rates of substantia nigra pars reticulata neurons

Gabapentin is a novel anti-epileptic drug which enhances GABA (gamma-aminobutyric acid) turnover in certain brain regions, including substantia nigra. However, the functional consequences of GABA turnover increases in response to gabapentin and their potential involvement in the anticonvulsant action of this drug are not known. In the present study, we examined the effects of gabapentin on the extracellular, single unit activity of nondopaminergic (presumably GABAergic) neurons of the substantia nigra pars reticulata in rats. During the recordings, the animals were infused with the narcotic opioid analgesic fentanyl, associated with a skeletal muscle relaxant and artificial ventilation. The spontaneous firing of substantia nigra pars reticulata neurons was determined up to about 2 h after i.v. or i.p. administration of gabapentin at doses of 15-30 mg/kg. After both routes of administration, gabapentin markedly reduced neuronal firing when administered at a dose of 20-30 mg/kg, while 15 mg/kg were ineffective in this regard. The suppressive effect of gabapentin was rapid in onset (2 min after i.v. and about 20 min after i.p. injection), reached peak values of about 70% below predrug baseline after about 45-60 min, and remained at this level for at least 2 h. Vehicle administration had no effect on substantia nigra pars reticulata neurons. The ability of gabapentin to alter substantia nigra pars reticulata firing does correlate with its known ability to increase nigral GABA turnover. Since a substantial body of evidence suggests that the substantia nigra pars reticulata is a critical site at which decrease of neuronal firing by potentiation of GABAergic influences results in protection against various seizure types, the suppressive effect of gabapentin on substantia nigra pars reticulata activity may contribute to the anticonvulsant action of this drug.

Extracellular single-unit recordings of piriform cortex neurons in rats: influence of different types of anesthesia and characterization of neurons by pharmacological manipulation of serotonin receptors.

In epilepsy research, there is a growing interest in the role of the piriform cortex (PC) in the development and maintenance of limbic kindling and other types of limbic epileptogenesis leading to complex partial seizures. Neurophysiological studies on PC or amygdala‐PC slice preparations from kindled rats showed that kindling of the amygdala induces long‐lasting changes in synaptic efficacy in the ipsilateral PC, including spontaneous discharges and enhanced susceptibility of PC neurons to evoked burst responses. These long‐lasting electrophysiological changes in the PC during kindling appear to be due, at least in part, to impaired function of gamma‐aminobutyric acid (GABA)ergic interneurons. The aim of the present study was to develop an anesthetic protocol allowing electrophysiological single‐unit recordings from inhibitory, presumably GABAergic PC interneurons in vivo. In addition to recording of spontaneously active PC neurons, microiontophoretic application of glutamate was used to activate silent neurons. Anesthesia of rats with ketamine/xylazine was not suited for single‐unit recordings in the PC because of marked cardiovascular depression. Anesthesia with chloral hydrate allowed recording of spontaneous or glutamate‐driven single‐unit activity in ≈40% of all animals. A similar percentage was obtained when recordings were done with the narcotic opioid fentanyl (plus gallamine), after all surgical preparations were performed under anesthesia with repeated administration of the barbiturate methohexital. To avoid brain accumulation of methohexital by repeated applications, we modified the anesthetic protocol in that methohexital was only injected once for initiation of surgical anesthesia, followed by the short‐acting anesthetic propofol which does not accumulate upon repeated application. Again, after surgical preparation, electrophysiological recordings were done under fentanyl (plus gallamine). By this procedure, spontaneous or glutamate‐driven single‐unit activity could be measured in all rats in either layer II or III of the PC. Based on shape and frequency of action potentials, two types of neurons were recorded. The predominant type was similar in its firing characteristics to GABAergic neurons in other brain regions, was mainly located in layer III, and could be suppressed by the serotonin2A receptor antagonist MDL 100,907, suggesting that this type of PC neuron represents inhibitory, putative GABAergic interneurons. This new in vivo preparation may be useful for evaluation of PC neurons in kindled rats. J. Neurosci. Res. 55:608–619, 1999. 

Modulation of spinal nociception by GluR5 kainate receptor ligands in acute and hyperalgesic states and the role of gabaergic mechanisms

GluR5 receptors modulate spinal nociception, however, their role in nociceptive hypersensitivity remains unclear. Using behavioural and electrophysiological approaches, we have investigated several GluR5 ligands in acute and hyperalgesic states. Furthermore, as the GABAergic system plays a role in GluR5 mediated effects in the brain, we also analysed the interaction between GluR5 agonists and GABA(A) antagonists in the spinal cord. In young rats in vivo, the GluR5 selective agonist ATPA was antinociceptive and antihyperalgesic in a model of inflammatory hyperalgesia (ED(50) approximately 4.6 and approximately 5.2 mg/kg, respectively), whereas the GluR5/GluR6 agonist SYM2081 was only antihyperalgesic. ATPA, but not SYM2081, was also able to inhibit nociceptive motoneurone responses in anaesthetised adult rats after intrathecal administration. In hemisected spinal cords in vitro, SYM2081 was inactive, whereas ATPA and another GluR5 agonist, (S)-5-iodowillardiine, inhibited nociceptive reflexes (EC(50) 1.1+/-0.4 micro M and 0.36+/-0.05 micro M, respectively). Both GluR5 agonists also inhibited motoneurone responses to repetitive dorsal root stimulation and their cumulative depolarisation, a correlate of wind-up. The GABA(A) antagonists bicuculline (10 micro M) and SR95531 (1 micro M) enhanced polysynaptic responses to single stimuli but abolished the cumulative depolarisation. Both bicuculline and SR95531 significantly attenuated the inhibition of nociceptive responses by 1 micro M ATPA (by approximately 50%). We conclude that selective GluR5 kainate receptor activation inhibits spinal nociception and its sensitisation caused by ongoing peripheral nociceptive drive. GABA(A) receptors are involved in tonic inhibition of segmental responses, but contribute to their sensitisation by repetitive primary afferent stimulation. Furthermore, there is a cross-talk between the two systems, presumably due to GluR5-mediated activation of GABAergic inhibitory interneurones in the spinal cord.