Istvan Tarnawa

Activity of 2,3-benzodiazepines at Native Rat and Recombinant Human Glutamate Receptors In Vitro: Stereospecificity and Selectivity Profiles

The activity and selectivity of the glutamate receptor antagonists belonging to the 2,3-benzodiazepine class of compounds have been examined at recombinant human non-NMDA glutamate receptors expressed in HEK293 cells and on native rat NMDA and non-NMDA receptors in vitro. The racemic 2,3-benzodiazepines GYKI52466, LY293606 (GYKI53405) and LY300168 (GYKI53655) inhibited AMPA (10 microM)-mediated responses in recombinant human GluR1 receptors expressed in HEK293 cells with approximate IC50 values of 18 microM, 24 microM and 6 microM, respectively and AMPA (10 microM) responses in recombinant human GluR4 expressing HEK293 cells with approximate IC50 values of 22 microM, 28 microM and 5 microM, respectively. GYKI 52466, LY293606 and LY300168 were non-competitive antagonists of AMPA receptor-mediated responses in acutely isolated rat cerebellar Purkinje neurons with approximate IC50 values of 10 microM, 8 microM and 1.5 microM, respectively. The activity of racemic compounds LY293606 and LY300168 was established to reside in the (-) isomer of each compound. At a concentration of 100 microM, GYKI52466, LY293606 and LY300168 produced < 30% inhibition of kainate-activated currents evoked in HEK293 cells expressing either human homomeric GluR5 or GluR6 receptors or heteromeric GluR6+KA2 kainate receptors. The activity of the 2,3-benzodiazepines at 100 microM was weak at kainate receptors, but was stereoselective. Similar levels of inhibition were observed for kainate-induced currents in dorsal root ganglion neurons. Intact tissue preparations were also used to examine the stereoselective actions of the 2,3-benzodiazepines. In the cortical wedge preparation, the active isomer of LY300168, LY303070, produced a non-competitive antagonism of AMPA-evoked depolarizations with smaller changes in depolarizations induced by kainate and no effect on NMDA-dependent depolarizations. LY303070 was also effective in preventing 30 microM AMPA-induced depolarizations in isolated spinal cord dorsal roots with an approximate IC50 value of 1 microM. Synaptic transmission in the hemisected spinal cord preparation was stereoselectively antagonized by the active isomers of LY300168 and LY293606. In summary, these results indicate that 2,3-benzodiazepines are potent, selective and stereospecific antagonists of the AMPA subtype of the non-NMDA glutamate receptor.

Electrophysiological studies with a 2,3-benzodiazepine muscle relaxant: GYKI 52466.

The effects of GYKI 52466, a new 2,3-benzodiazepine with muscle relaxant and anticonvulsant properties, were investigated and compared to those of midazolam in electrophysiological experiments. The effects of the drugs on the reflex potentials evoked by afferent nerve stimulation and recorded from the spinal roots in unanesthetized spinal cats were studied. GYKI 52466 exerted a strong inhibitory effect on the monosynaptic as well as the polysynaptic ventral root reflexes, while the dorsal root responses decreased slightly. In contrast, midazolam markedly enhanced the dorsal root responses, did not modify the monosynaptic reflex and partially inhibited the polysynaptic reflex. The spontaneous firing of cerebellar Purkinje cells was depressed by midazolam, but not by GYKI 52466. These results suggest strongly that, contrary to the classical 1,4-benzodiazepines, potentiation of the GABA-A receptor-mediated inhibition does not play a significant role in the pharmacological actions of GYKI 52466.

Blockers of Voltage-Gated Sodium Channels for the Treatment of Central Nervous System Diseases

Voltage gated sodium channels play important roles both in vital physiological functions and several pathological processes of the central nervous system. Epilepsy, chronic pain, neurodegenerative diseases, and spasticity are all characterized by an over-excited state of specific groups of central neurons that is accompanied by an abnormally increased activity of sodium channels. An efficient strategy of controlling such diseases is to use blockers that preferentially act on these over-excited cells. State dependently acting agents, such as phenytoin, or lamotrigine, leave normal physiological functions relatively intact, resulting in a favorable therapeutic window. Nine isoforms of the channel forming alpha subunit are known, which show distinct expression patterns in different tissues. Another possible way to decrease the chance of adverse effects is to develop agents selectively inhibiting the channel subtype involved in the pathomechanism of the disease to be treated. Many recent patents claim sodium channel blockers with improved characteristics regarding state dependency or subtype selectivity. Such agents may offer a breakthrough in the treatment of a variety of central nervous system diseases. This review focuses on the current trends in sodium channel research, surveying the traditional and newly emerging therapeutic fields, and the diverse medicinal chemistry of sodium channel blockers.

New Non Competitive AMPA Antagonists

New halogen atom substituted 2,3-benzodiazepine derivatives condensed with an azole ring on the seven membered part of the ring system of type 3 and 4 as well as 5 and 6 were synthesized. It was found that chloro-, dichloro- and bromo-substitutions in the benzene ring and additionally imidazole ring condensation on the diazepine ring can successfully substitute the methylenedioxy group in the well known molecules GYKI 52466 (1) and GYKI 53773 (2) and the 3-acetyl-4-methyl structural feature in 2, respectively, preserving the highly active AMPA antagonist characteristic of the original molecules. From the most active compounds (3b,i) 3b (GYKI 47261) was chosen for detailed investigations. 3b revealed an excellent, broad spectrum anticonvulsant activity against seizures evoked by electroshock and different chemoconvulsive agents indicating a possible antiepileptic efficacy. 3b was found to be highly active in a transient model of focal ischemia predictive of a therapeutic value in human stroke. 3b also reversed the dopamine depleting effect of MPTP and antagonized the oxotremorine induced tremor in mice indicating a potential antiparkinson activity.

NR2B containing NMDA receptor dependent windup of single spinal neurons

Windup, the frequency dependent build-up of spinal neuronal responses, is implicated in the development of central sensitization of nociceptive pathways. N-methyl-D-aspartate (NMDA) receptors have been shown to be involved in these processes but the role of various receptor subtypes at the spinal level is not fully understood. In our experiments, we compared the inhibitory effect of MK-801 (a nonselective NMDA receptor antagonist, 0.01-3 mg/kg i.v.) and CI-1041 (an NR2B subunit specific NMDA receptor antagonist, 0.3-10 mg/kg i.v.) on the formation of dorsal horn neuronal windup in spinalized rats, in vivo. Both types of antagonist blocked windup considerably at doses not affecting the normal synaptic transmission. These results are in agreement with the well-documented effectivity of NR2B subtype selective NMDA receptor antagonists in chronic pain models and give the first direct evidence that spinal mechanisms are involved in this effect.

Apparent antinociceptive and anti-inflammatory effects of GYKI 52466

GYKI 52466 (1-(4-aminophenyl)-4-methyl-7,8-methylenedioxy-5H-2,3-benzodiazepine) was examined in a battery of analgesia and anti-inflammatory tests in rats and mice, respectively. Its 3-N-acetyl (GYKI 53773) and 3-N-methylcarbamoyl (GYKI 53784) derivatives were also examined in some assays. These 2,3-benzodiazepines, known as prototypic non-competitive antagonists of AMPA receptors, showed a peculiar profile in some routinely used antinociceptive tests. They were found fairly potent in rat tail flick and mouse phenylquinone writhing assays but the dose-response curves were rather shallow as compared to that of morphine. Their action is stereoselective, i.e., the (+) isomers were found inactive, in agreement with the previous in vitro studies. Their antinociceptive effect could not be reversed by naloxone and the GYKI compounds did not potentiate significantly the morphine-induced analgesia. In the mouse hot plate assay the 2,3-benzodiazepines were active only in doses inducing visible motor incapacitation. In rats, GYKI 52466 weakly reduced the hypersensitivity accompanying acute carrageenan edema. However, it potently inhibited the hyperalgesia during Freund adjuvant-induced chronic arthritis. In the latter assay GYKI 52466 also attenuated the body weight loss without altering the paw edema. The present findings confirm reports in the literature which indicate AMPA receptors may contribute to certain forms of pathological hyperalgesia, e.g., to that detectable in inflamed tissues.

Protective effect of the antiepileptic drug candidate talampanel against AMPA-induced striatal neurotoxicity in neonatal rats.

2,3-Benzodiazepines represent a family of specific, noncompetitive AMPA receptor antagonists with anticonvulsant and neuroprotective properties. In this study, the antiexcitotoxic potency of the clinical antiepileptic drug candidate, talampanel (4 x 2 mg/kg), and that of two related 2,3-benzodiazepines, 5-(4-aminophenyl)-8-methyl-9H-1,3-dioxolo[4,5-h][2,3]-benzodiazepine (GYKI 52466) (4 x 10 mg/kg) and GYKI 53784 (4 x 2 mg/kg), was investigated in 7-day-old rats. The AMPA antagonists were applied in four consecutive i.p. injections at 1-h intervals, the first dosage was given shortly after the intrastriatal injection of (S)-alpha-amino-3-hydroxy-5,7-methylisoxazole-4-propionic acid (AMPA) (2.5 nmol). All tested compounds protected animals from brain damage induced by AMPA as assessed 5 days later by using a tissue volume determination method based on computer-aided serial section reconstruction. GYKI 53784 (56.1 +/- 5.0% protection) and talampanel (42.5 +/- 5.3% protection) were more potent neuroprotective agents than GYKI 52466 (21.8 +/- 2.8% protection). Furthermore, the three compounds attenuated the unilateral AMPA injection-induced turning behavior and seizure-like events.Our present findings are in agreement with those of other investigators who found talampanel neuroprotective in various in vivo experimental models. These data indicate that besides being a promising antiepileptic drug candidate talampanel may have a value in the pharmacotherapy of acute and chronic neurodegenerative diseases, including perinatal ischemia/hypoxia-induced brain injuries, as well.

AMPA receptor antagonists, GYKI 52466 and NBQX, do not block the induction of long-term potentiation at therapeutically relevant concentrations

The involvement of alpha-amino-3-hydroxy-5-methylizoxazole-4-propionic acid (AMPA) receptors in induction of long-term potentiation (LTP) was examined in rat hippocampal slice preparation. Using conventional extracellular recording, excitatory postsynaptic potentials (EPSPs) and population action potentials (PSs), evoked by low-frequency stimulation of the Schaffer collateral-commissural fibres, were recorded in the CA1 region. The effects of a competitive AMPA receptor antagonist, 6-nitro-7-sulfamoylbenzo(f)quinoxaline-2, 3-dione (NBQX), and that of a non-competitive blocker, 1-(4-aminophenyl)-4-methyl-7,8-methylendioxy-5H-2,3-benzodiazepine (GYKI 52466) have been examined. 0.25-0.5 microM of NBQX and 20-40 microM of GYKI 52466 did not suppress the induction of LTP. LTP was attenuated only at the highest concentrations tested (1 microM NBQX or 80 microM GYKI 52466). These in vitro concentrations, however, exceed the brain levels needed for in vivo anticonvulsant action. Furthermore, even at the highest concentrations both compounds suppressed only the expression but not the induction of LTP. Namely after their washout LTP reappeared. Thus, at pharmacologically relevant concentrations these AMPA receptor antagonists apparently do not suppress LTP, a cellular mechanism underlying memory formation. These experiments suggest that in clinical practice AMPA receptor blockade may have some advantage over N-methyl-D-aspartate receptor antagonism, which is accompanied by severe memory impairment.

NR2B receptors are involved in the mediation of spinal segmental reflex potentials but not in the cumulative motoneuronal depolarization in vitro

Windup, the frequency dependent build-up of spinal neuronal responses is an electrophysiological model of the development of the central sensitization in the chronic pain states. NR2B subunit containing NMDA-type glutamate receptors are implicated in the windup of dorsal horn neurons, while their role at the motoneuronal level is controversial. The cumulative motoneuronal depolarization in hemisected rat spinal cord preparation is an in vitro model of windup. The role of NR2B receptors in this process, and in the mediation of dorsal root stimulation evoked ventral root reflex potentials was elucidated. Three selective NR2B antagonists; CP-101,606; CI-1041 and Co-101244 (1 microM) were used. They had only weak, but statistically significant inhibitory effect on the early part of ventral root response, and did not influence the cumulative depolarization. On the contrary, non-selective NMDA antagonist APV (40 microM) decreased both responses markedly. We conclude that the pharmacological sensitivities of windup at the sensory and motor levels are different. NR2B containing NMDA receptors have major role in the mediation of the windup of dorsal horn neurons, but their contribution to this phenomenon at the motor level is negligible.

Functional characterization of sodium channel blockers by membrane potential measurements in cerebellar neurons: prediction of compound preference for the open/inactivated state.

Voltage-gated sodium channel (VGSC) blockers are widely used in the therapy, but most currently available blockers have suboptimal profile. However, discovery of new drug candidates has been hampered by the lack of appropriate in vitro assays. We established a fluorometric, plate reader-based membrane potential assay for testing the inhibitory potency of various VGSC blocking drugs, using primary cultures of cerebellar neurons, and veratridine, as activator of VGSCs. Since inhibition was strongly dependent on the depolarizing effect of veratridine, the EC(80) value of veratridine was determined on each experimental day, and this concentration was used for drug testing. This strict control on agonist effect seems to improve the reliability of the dose-inhibition measurements with antagonists. Veratridine responses could be completely inhibited by tetrodotoxin (TTX, IC(50)=17 nM), consistent with the exclusive expression of TTX-sensitive VGSCs. A variety of compounds known to block sodium channels inhibited veratridine-induced membrane depolarization concentration-dependently. Furthermore, inhibitory potencies of drugs strongly depended on whether their administration preceded or followed veratridine application. Potency of lamotrigine, carbamazepine, phenytoin and lidocaine was approximately 10-fold higher when applied after a steady-state depolarization had been achieved by a supramaximal veratridine dose, compared with those from a different protocol, where cells were preincubated with the antagonists prior to veratridine application. On the contrary, there was only relatively small difference between the IC(50) values of GBR 12909 obtained from the two different protocols (0.51 microM versus 1.23 microM). In contrast with most sodium channel blockers, this compound lacks binding preference to inactivated channels. We suggest that comparison of the results obtained with a particular blocker in the pretreatment and post-treatment schedules may be suitable for drawing conclusions regarding the state-dependency of its action. Thus, relevant information can be obtained about the potential therapeutic utility of different drugs by applying non-electrophysiological methods.