Further evaluation of the potential anxiolytic activity of imidazo[1,5-a][1,4]diazepin agents selective for α2/3-containing GABAA receptors.

Positive allosteric modulators of GABAA receptors transduce a host of beneficial effects including anxiolytic actions. We have recently shown that bioavailability and anxiolytic-like activity can be improved by eliminating the ester functionality in imidazo[1,5-a][1,4]diazepines. In the present series of experiments, we further substantiate the value of heterocyle replacement of the ester for potential treatment of anxiety. None of three esters was active in a Vogel conflict test in rats that detects anxiolytic drugs like diazepam. Compounds 7 and 8, ester bioisosters, were selective for alpha 2 and 3 over alpha 1-containing GABAA receptors but also had modest efficacy at GABAA alpha 5-containing receptors. Compound 7 was efficacious and potent in this anxiolytic-detecting assay without affecting non-punished responding. The efficacies of the esters and of compound 7 were predicted from their efficacies as anticonvulsants against the GABAA antagonist pentylenetetrazole (PTZ). In contrast, the related structural analog, compound 8, did not produce anxiolytic-like effects in rats despite anticonvulsant efficacy. These data thus support the following conclusions: 1) ancillary pharmacological actions of compound 8 might be responsible for its lack of anxiolytic-like efficacy despite its efficacy as an anticonvulsant 2) esters of imidazo[1,5-a][1,4]diazepines do not demonstrate anxiolytic-like effects in rats due to their low bioavailability and 3) replacement of the ester function with suitable heterocycles markedly improves bioavailability and engenders molecules with the opportunity to have potent and efficacious effects in vivo that correspond to human anxiolytic actions.

mGlu2/3 agonist-induced hyperthermia: an in vivo assay for detection of mGlu2/3 receptor antagonism and its relation to antidepressant-like efficacy in mice.

An assay to detect the on-target effects of mGlu2/3 receptor antagonists in vivo would be valuable in guiding dosing regimens for the exploration of biological effects of potential therapeutic import. Multiple approaches involving blockade of mGlu2/3 receptor agoinist-driven behavioral effects in mice and rats were investigated. Most of these methods failed to provide a useful method of detection of antagonists in vivo (e.g., locomotor activity). In contrast, the mGlu2/3 receptor agonist LY379268 produced dose-dependent increases in body temperature of mice. The hyperthermic effects of LY379268 was abolished in mGlu2 and in mGlu2/3 receptor null mice but not in mGlu3 null mice. Hyperthermia was not produced by an mGlu8 receptor agonist. Agonist-induced hyperthermia was prevented in a dose-dependent manner by structurally-distinct mGlu2/3 receptor antagonists. The blockade was stereo-specific. Moreover, this biological readout was responsive to both orthosteric and to negative allosteric modulators of mGlu2/3 receptors. Antagonism of agonist-induced hyperthermia predicted antidepressant-like efficacy in the mouse forced swim test. As with the hyperthermic response, the antidepressant-like effects of mGlu2/3 receptor antagonists were shown to be due to mGlu2 and not to mGlu3 or mGlu8 receptors through the use of receptor knock-out mice. The ability to rapidly assess on-target activity of mGlu2/3 receptor antagonists enables determination of parameters for setting efficacy doses in vivo. In turn, efficacy-related data in the preclinical laboratory can help to set expectations of therapeutic potential and dosing in humans.

Increased expression of neuronal Src and tyrosine phosphorylation of NMDA receptors in rat brain after systemic treatment with MK-801.

We have observed that systemic treatment with the uncompetitive N-methyl-D-aspartate (NMDA) receptor antagonist MK-801 increases Src expression and NMDA receptor phosphorylation in rat brain. A partial cDNA encoding rat neuronal Src was isolated and its sequence was used to design specific oligonucleotide probes. Systemically administered MK-801 (5 mg/kg for 4 h) increased by 28+/-4% mRNA expression of neuronal Src in the superficial layers of the parietal cortex. This effect was observed at doses as low as 0.2 mg/kg. A similar, although more modest, induction was observed 6 h after phencyclidine (15 mg/kg) administration, but not after high doses of memantine and ketamine. The MK-801-induced effect was not blocked by pretreatment with clozapine. Consistent with the increase in mRNA levels, cortical Src protein was increased to 186 +/- 24% of control 24 h after MK-801 treatment. Total cellular Src activity was also increased in parietal cortex homogenates 4 h after MK-801 (5 mg/kg). Moreover, MK-801 treatment (0.5 mg/kg and 5 mg/kg for 4 h) increased tyrosine phosphorylation, but not protein levels, of the NMDA receptor subunit NR2A. These results provide evidence for a contribution of Src and tyrosine phosphorylation of NMDA receptors in the pharmacological actions of uncompetitive NMDA receptor antagonists.

Preclinical evaluation of newly approved and potential antiepileptic drugs against cocaine-induced seizures.

Seizures and status epilepticus are among the neurological complications of cocaine overdose in humans. The aim of the present study was to evaluate the protective effectiveness and therapeutic index (separation between anticonvulsive and side effect profiles) of 14 newly approved and potential antiepileptic drugs using a murine model of acute cocaine toxicity and the inverted-screen test for behavioral side effect testing. Cocaine (75 mg/kg i.p.) produces clonic seizures (approximately 90% of mice), and conventional antiepileptic drugs have been reported to be either ineffective or only effective at doses producing significant sedative/ataxic effects. Clobazam, flunarizine, lamotrigine, topiramate, and zonisamide were ineffective against seizures up to doses producing significant motor impairment. In contrast, felbamate, gabapentin, loreclezole, losigamone, progabide, remacemide, stiripentol, tiagabine, and vigabatrin produced dose-dependent protection against cocaine-induced convulsions with varied separations between their anticonvulsant and side effect profiles: the protective index values (toxic TD(50)/anticonvulsive ED(50)) ranged from 1.26 (felbamate) to 7.67 (loreclezole), and gabapentin had the highest (protective index >152). Thus, several drugs were identified with greater protective efficacy and reduced motor impairment compared with classic antiepileptic drugs. Based on the proposed mechanism of action of these new anticonvulsants, it is noteworthy that 1) drugs that enhance gamma-aminobutyric acid-mediated neuronal inhibition in a manner distinct from barbiturates and benzodiazepines offer the best protective/behavioral side effect profiles, and 2) functional antagonists of Na(+) and Ca(2+) channels are generally ineffective. Overall, this study provides the first description of the effectiveness of new antiepileptic drugs against experimentally induced cocaine seizures and points to several drugs that deserve clinical scrutiny for this indication.

Evaluation of ADCI against convulsant and locomotor stimulant effects of cocaine: comparison with the structural analogs dizocilpine and carbamazepine.

Both the antiepileptic, carbamazepine, and the N-methyl-D-aspartate receptor antagonist, dizocilpine, have shown preclinical efficacy against behavioral and toxic effects of cocaine. Nonetheless, side effects or toxicity of these compounds either alone or in conjunction with cocaine are problematic. 5-Aminocarbonyl-10,11-dihydro-5h-dibenzo[a,d]cyclohepten-5,1 0-imine (ADCI), a molecular hybrid of these compounds, has been shown to have a broad anticonvulsant profile with a good protective index (behavioral TD50/anticonvulsant ED50). In male Swiss Webster mice, ADCI and dizocilpine produced dose-dependent protection against the convulsant effects of cocaine that were insensitive to carbamazepine. However, in contrast to dizocilpine, ADCI did not produce behavioral impairment on the inverted screen test demonstrating a protective index of greater than 15; the protective index for dizocilpine was 1.2. All three compounds attenuated the locomotor stimulant effects of cocaine without significantly decreasing locomotor activity on their own, although the cocaine antagonism was not always dose dependent. Only dizocilpine increased spontaneous locomotor activity when given alone and augmented the locomotor stimulant effects of cocaine. The results confirm the novel anticonvulsant activity of ADCI and its lack of phencyclidine-like behavioral side effects. The data also suggest a modest blocking action of these compounds against the locomotor stimulatory effects of cocaine.

Tolerance to behavioral effects of physostigmine under interval schedules of positive or negative reinforcement.

The present experiments examined whether the rate and type of events maintaining responding help determine physostigmine's behavioral effects. The first two experiments examined the acute and chronic effects of physostigmine, respectively, on lever pressing of rats under variable-interval schedules of food presentation. The third examined the chronic effects of physostigmine on lever pressing under random-interval schedules of shock avoidance. Three different variable intervals (18, 56, and 180 s) and two different random intervals (20 and 60 s) were studied, each associated with a distinctive stimulus. Baseline rates of responding were directly related to the scheduled rate of food delivery or shock avoidance. Acute administration of 0.154-1.233 mumol/kg (0.1-0.8 mg/kg) physostigmine sulfate produced monotonic decreases in overall response rate under all schedules of food presentation. Acute effects (per cent of control response rate) did not differ systematically under the various interval values. Large doses (i.e., 0.4 or 0.8 mg/kg) suppressed the rate of food delivery as well. When initially administered, 0.967 mumol/kg (0.4 mg/kg) physostigmine salicylate also suppressed avoidance response rates and per cent shocks avoided. Tolerance developed to the effects of this dose of physostigmine salicylate on pellet or shock-avoidance frequency more rapidly than to effects on overall response rate. Tolerance to the latter developed only very gradually and could in the case of shock-avoidance response rates be considered partial at best. Tolerance was not affected by the scheduled rate of food or shock presentation. Blood acetylcholinesterase levels showed no recovery during chronic physostigmine. Tolerance is discussed in terms of the reinforcement-loss hypothesis.