Convulsant actions of the neurosteroid pregnenolone sulfate in mice.

Pregnenolone sulfate (PS) is an endogenous neurosteroid known to antagonize GABA(A) receptor-mediated inhibitory responses and potentiate NMDA receptor-mediated excitatory responses in vitro. To assess the actions of the steroid as a modulator of seizure susceptibility in vivo, PS (30-300 nmol) was administered intracerebroventricularly in mice. At doses of 50 to 150 nmol, PS elicited seizures characterized by head jerks, rearing and falling, severe forelimb and hindlimb clonus, opisthotonos and explosive running. The seizures increased in severity and frequency with time and eventually progressed to status epilepticus, tonic hindlimb extension and death. The doses producing convulsions in 50% (CD(50)) and 97% (CD(97)) of animals were 92 and 205 nmol, respectively. A subconvulsant dose of PS (50 nmol) significantly increased the convulsant potencies of systemically administered pentylenetetrazol (30-50 mg/kg) and NMDA (50-100 mg/kg). Systemically administered PS at doses as high as 100 mg/kg failed to induce seizures or alter the convulsant potencies of pentylenetetrazol and NMDA. Protection against PS (205 nmol)-induced seizures and lethality was conferred by the GABA(A) receptor positive allosteric modulators clonazepam and allopregnanolone, and by the NMDA receptor antagonists dizocilpine and (R)-CPP. The overall pharmacological profile suggests that the convulsant actions of PS are mediated predominantly via its effects on GABA(A) receptors, and also possibly by effects on NMDA receptors.

Induction of seizures by the potent K+ channel-blocking scorpion venom peptide toxins tityustoxin-K(alpha) and pandinustoxin-K(alpha).

The scorpion venom peptide toxins tityustoxin-K(alpha) (TsTx-K(alpha)) and pandinustoxin-K(alpha) (PiTx-K(alpha)) are novel, highly potent and selective blockers of voltage-activated K+ channels. PiTx-K(alpha) preferentially blocks rapidly inactivating (A-type) K+ channels whereas TsTx-K(alpha) is selective for slowly inactivating (delayed rectifier-type) channels. K+ channel blockers are known to induce seizures, but the specific K channel types that can serve as convulsant targets are not well defined. To address this issue, we examined for convulsant activity the K+ channel type-specific scorpion toxins and the selective K+ channel antagonists 4-aminopyridine (4-AP), an inhibitor of A-type voltage-activated K+ channels, and paxilline, a selective blocker of large conductance (maxi K) Ca(2+)-activated K+ channels. Intracerebroventricular injection of recombinant TsTx-K(alpha) and PiTx-K(alpha) in mice produced limbic and clonic-tonic seizures. The severity of the seizures increased during the 60-min period following injection, culminating in continuous clonic seizure activity (status epilepticus), tonic hindlimb extension, and eventually in death. The estimated doses producing limbic and clonic seizures in 50% of animals (CD50) for TsTx-K(alpha) and PiTx-K(alpha) were 9 and 33 ng, respectively. 4-AP produced seizure activity similar to the toxins (CD50, 76 ng) whereas paxilline failed to induce seizures at doses up to 13.5 microg. Carbamazepine protected fully against the toxin- and 4-AP-induced seizures whereas phenytoin had variable activity against the clonic component although it was protective against tonic hindlimb extension. The AMPA receptor antagonist GYKI 52466 also conferred full protection against toxin-induced seizures, but the NMDA receptor antagonists (R)-CPP and dizocilpine failed to affect limbic and clonic seizures, although they protected against hindlimb extension. We conclude that selective blockade of delayed rectifier- or A-type voltage-activated K+ channels can produce limbic, clonic and tonic seizures, whereas blockade of maxi K-type Ca(2+)-activated K+ channels does not. The convulsant effects may be related to enhanced glutamate release and, in the case of the limbic and clonic convulsions, activation of AMPA receptors.

Finasteride, a 5alpha-reductase inhibitor, blocks the anticonvulsant activity of progesterone in mice.

Progesterone is an effective anticonvulsant against pentylenetetrazol (PTZ) seizures. This action is hypothesized to require the metabolic conversion of progesterone to the gamma-aminobutyric acidA receptor potentiating neuroactive steroid allopregnanolone by 5alpha-reductase isoenzymes followed by 3alpha-hydroxy oxidoreduction. We evaluated this possibility using the competitive 5alpha-reductase inhibitor finasteride. Progesterone (50-200 mg/kg, i.p.) protected mice against PTZ-induced seizures in a dose-dependent manner (ED50, 94 mg/kg). Pretreatment with finasteride (50-300 mg/kg, i.p.) produced a dose-dependent (ED50, 146 mg/kg) reversal of the protective effects of progesterone (2 x ED50 dose = 188 mg/kg). In contrast, finasteride (up to 300 mg/kg) failed to affect the anticonvulsant activity of allopregnanolone (10-30 mg/kg, i.p.; ED50, 12 mg/kg). Finasteride (up to 300 mg/kg) did not block the protective effect of high doses of progesterone (250-350 mg/kg) on tonic hindlimb extension in the maximal electroshock seizure test (progesterone ED50, 235 mg/kg). The anticonvulsant activity of progesterone against PTZ-induced seizures can be blocked by 5alpha-reductase inhibition, providing strong evidence that the anticonvulsant effect of the steroid in this model is mediated by its active metabolite allopregnanolone.

Lack of anticonvulsant tolerance to the neuroactive steroid pregnanolone in mice.

GABA-potentiating neuroactive steroids such as pregnanolone have potent protective effects in the pentylenetetrazol seizure test. We sought to determine if tolerance develops to the anticonvulsant activity of pregnanolone with chronic administration. Mice were treated with two daily injections of a 2 x ED50 dose of pregnanolone (25 mg/kg, i.p.) for 7 days. On the day after the chronic treatment protocol, the dose-response relationship for protection in the pentylenetetrazol seizure test was obtained. The ED50 value after the chronic treatment protocol was not significantly different from that in naive mice (12 mg/kg), indicating that tolerance does not develop to the anticonvulsant activity of pregnanolone. In subsequent experiments, we extended the chronic treatment protocol to 14 days with three daily injections of pregnanolone (25 mg/kg, i.p.). Again, no tolerance was observed (ED50, 13 mg/kg). The anticonvulsant activity of pregnanolone was well correlated with plasma levels in both the naive and chronically (14 day) treated mice. The estimated plasma concentrations of pregnanolone representing threshold (10%) protection (125-150 ng/ml) and 50% protection (575-700 ng/ml) were similar in naive and chronically treated animals. In both chronically treated and naive animals, plasma levels of pregnanolone declined rapidly (t1/2, 16-19 min) and there was a corresponding reduction in the anticonvulsant activity. Our results with pregnanolone suggest that tolerance does not develop to the anticonvulsant activity of neuroactive steroids as it does with other GABA potentiating drugs such as benzodiazepines, supporting the potential clinical utility of neuroactive steroids in chronic seizure therapy.

Further studies on anti- and proconvulsant effects of inhibitors of nitric oxide synthase in rodents.

We confirmed that the effects of inhibitors of nitric oxide (NO) synthase, such as Nomega-nitro-L-arginine methyl ester and Nomega-nitro-L-arginine, differ depending on several experimental factors. Both compounds but not their less active enantiomers delayed picrotoxin-induced clonus in mice yet increased the incidence of clonus following low-dose picrotoxin. Nomega-nitro-L-arginine methyl ester significantly reduced the latencies of both myoclonus and clonus in older but not younger Sprague-Dawley rats receiving pentylenetetrazol s.c. By contrast, there was no significant change in the latencies for myoclonus and clonus in Wistar rats (older and younger). However, when pentylenetetrazol was administered i.p. rather than s.c., Nomega-nitro-L-arginine methyl ester dramatically increased latencies of convulsive indicators, including tonus, in both Sprague-Dawley and Wistar rats. Nomega-nitro-L-arginine methyl ester also delayed tonus but not myoclonus or clonus in mice, regardless of the systemic route of administration of pentylenetetrazol. Both Nomega-nitro-L-arginine methyl ester and NG-nitro-L-arginine increased the tonic CD50 of pentylenetetrazol in mice and Nomega-nitro-L-arginine methyl ester delayed 4-aminopyridine-induced tonus. However, Nomega-nitro-L-arginine methyl ester reduced the tonic CD50 of both picrotoxin and 4-aminopyridine in mice and failed to suppress tonus following maximal electroshock. Evidently, inhibitors of NO synthase are not universally effective antitonic drugs.

Ibogaine block of the NMDA receptor: in vitro and in vivo studies.

Ibogaine is an hallucinogenic indole alkaloid claimed to have anti-addictive properties. Although its mechanism of action is unknown, binding studies have indicated that the drug may interact with N-methyl-D-aspartate (NMDA) receptors. We further investigated the nature of the interaction between ibogaine and NMDA receptors in voltage clamp and binding studies, and sought to confirm that the drug has NMDA receptor blocking activity in vivo. In whole-cell recordings from cultured rat hippocampal neurons, ibogaine caused a slow, concentration-dependent block of NMDA-induced currents (IC50, 3.1 microM at -60 mV). In contrast, ibogaine failed to affect either kainate- or gamma-aminobutyric acid-evoked currents. The blockade of NMDA currents was use- and voltage-dependent, and the long lasting ibogaine block could be occluded by co-application of Mg2+. Ibogaine also inhibited equilibrium [3H]dizocilpine binding to NMDA receptors in rat forebrain membranes (IC50, 3.2 microM). We conclude that ibogaine is an open channel NMDA receptor antagonist. Administration of ibogaine to mice resulted in complete protection in the maximal electroshock test (ED50, 31 mg/kg, i.p.) and partial protection against NMDA-induced lethality, confirming that ibogaine can block NMDA receptors in vivo.

Neuroactive steroids protect against pilocarpine- and kainic acid-induced limbic seizures and status epilepticus in mice.

Several structurally related metabolites of progesterone (3 alpha-hydroxy pregnane-20-ones) and deoxycorticosterone (3 alpha-hydroxy pregnane-21-diol-20-ones) and their 3 beta-epimers were evaluated for protective activity against pilocarpine-, kainic acid- and N-methyl-D-aspartate (NMDA)-induced seizures in mice. Steroids with the 3-hydroxy group in the alpha-position and 5-H in the alpha- or beta-configurations were highly effective in protecting against pilocarpine (416 mg/kg, s.c.)-induced limbic motor seizures and status epilepticus (ED50 values, 7.0-18.7 mg/kg, i.p.). The corresponding epimers with the 3-hydroxy group in the beta-position were also effective but less potent (ED50 values, 33.8-63.5, i.p.). Although the neuroactive steroids were considerably less potent than the benzodiazepine clonazepam in protecting against pilocarpine seizures, steroids with the 5 alpha,3 alpha-configuration had comparable or higher protective index values (TD50 for motor impairment divided by ED50 for seizure protection) than clonazepam, indicating that some neuroactive steroids may have lower relative toxicity. Steroids with the 5 alpha,3 alpha- or 5 beta,3 alpha-configurations also produced a dose-dependent delay in the onset of limbic seizures induced by kainic acid (32 mg/kg, s.c.), but did not completely protect against the seizures. However, when a second dose of the steroid was administered 1 hr after the first dose, complete protection from the kainic acid-induced limbic seizures and status epilepticus was obtained. The steroids also caused a dose-dependent delay in NMDA (257 mg/kg, s.c.)-induced lethality, but did not completely protect against NMDA seizures or lethality. We conclude that neuroactive steroids are highly effective in protecting against pilocarpine- and kainic acid-induced seizures and status epilepticus in mice, and may be of utility in the treatment of some forms of status epilepticus in humans.

Anticonvulsant efficacy of ADCI (5-aminocarbonyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine) after acute and chronic dosing in mice.

ADCI (5-aminocarbonyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5, 10-imine), a low-affinity uncompetitive N-methyl-D-aspartate (NMDA) antagonist, is a broad-spectrum anticonvulsant with a favorable side-effect profile. In the present study, we sought to determine if tolerance develops to the anticonvulsant activity of ADCI, using the maximal electroshock (MES) test to assess seizure protection. Mice were treated with three daily injections of a 2 x ED50 dose for MES protection (18 mg/kg, intraperitoneally, i.p.) or vehicle for 7 or 14 days. On the day after the chronic treatment protocol, all animals received a challenge dose of ADCI (18 mg/kg) and 15 min later were evaluated in the MES test. In control animals, 83-94% of animals were protected and the ADCI plasma levels immediately after the MES test were 5.5-9.7 micrograms/ml. In treated animals, 29 and 0% of animals were protected at 7 and 14 days, respectively, and the ADCI plasma levels were 77 and 52% of the control values. [3H]Dizocilpine binding to brain NMDA receptors was unaltered by the chronic drug treatment. In subsequent experiments, we determined that 14-day chronically treated animals could be completely protected by increased doses of ADCI (ED50 28.9 mg/kg). In both naive and chronically treated animals receiving a challenge dose of ADCI, plasma drug levels decreased in two phases, the first with a time constant of approximately 55 min and the second with a much slower rate. The estimated plasma concentrations of ADCI reflecting threshold (3-5 micrograms/ml) and 50% protection (5-7.5 micrograms/mg) were similar in naive and chronic animals.(ABSTRACT TRUNCATED AT 250 WORDS)

Anticonvulsant activity of neurosteroids: correlation with gamma-aminobutyric acid-evoked chloride current potentiation.

Certain neurosteroids rapidly alter the excitability of neurons, in part by potentiating gamma-aminobutyric acid (GABA)-evoked chloride currents, and, like other GABA potentiating drugs, may have anticonvulsant activity. We compared the abilities of a series of isomeric metabolites of progesterone and deoxycorticosterone (3-hydroxy pregnane-20-ones and 3-hydroxy pregnane-21-ol-20-ones) to enhance GABA-evoked chloride currents in cultured hippocampal neurons with their abilities to protect against pentylenetetrazol (PTZ)-induced seizures in mice. Metabolites with 3-hydroxy in the alpha-position and 5-H in the alpha- or beta-configuration were highly effective at potentiating GABA-evoked chloride current and also showed potent anticonvulsant activity in the PTZ seizure test. The corresponding metabolites with hydroxyl groups in the 3 beta-position were considerably less potent in enhancing GABA responses and were inactive in the PTZ test. All of the neurosteroids failed to protect against tonic hindlimb extension in the maximal electroshock seizure test. 5 alpha-Pregnane-3 alpha,11 beta,21-triol-20-one, a corticosterone metabolite reported to block voltage-dependent Ca++ channels, was inactive in either of the anticonvulsant tests. At higher doses, neurosteroids effective in the PTZ test also produced motor impairment. Relative motor toxicity was lower (higher protective index) for compounds with the 5 alpha-configuration than for their corresponding 5 beta-epimers. The anticonvulsant profile of the neurosteroids resembled that of the benzodiazepine clonazepam. Although the anticonvulsant steroids had greater in vitro potencies than clonazepam, they were less potent in vivo, and they had lower protective indices.(ABSTRACT TRUNCATED AT 250 WORDS)

Stimulation of the slow calcium current in bullfrog skeletal muscle fibers by cAMP and cGMP.

The effects of adenosine 3',5'-cyclic monophosphate (cAMP) and guanosine 3',5'-cyclic monophosphate (cGMP) on slow calcium currents (ICa) were investigated using the Vaseline-gap voltage-clamp technique in bullfrog skeletal muscle cut fibers. Both cAMP and cGMP induced a pronounced increase in the amplitude of ICa when applied to the cut ends of fibers. Both cyclic nucleotides also decreased time to peak current at all membrane potentials. The current-voltage relationship was shifted toward more negative potentials by cAMP as well as cGMP. The potentiating effects of cAMP and cGMP on ICa were additive. 8-Bromo analogues of both nucleotides had similar effects on ICa. The beta-adrenergic agonist isoproterenol, applied extracellularly, also produced an increase in the amplitude of ICa and produced a leftward shift in the current-voltage relationship. These results suggest that both cAMP and cGMP modulate calcium slow channels in bullfrog skeletal muscle fibers, causing stimulation of the ICa. The effect of cyclic nucleotides on ICa in bullfrog skeletal muscle contrasts with that in mammalian cardiac muscle, in which the same nucleotides produce opposite effects on the slow ICa, i.e., in cardiac muscle cAMP stimulates, and cGMP inhibits, the slow ICa.

Blockade of K+ contractures in skeletal muscle by opioid drugs: a nonstereospecific effect.

The effect of several opioid drugs was tested on the K+ contractures in frog's skeletal muscle. These contractures are produced by the entrance of extracellular Ca2+ ions via the voltage-dependent, slow Ca2+ channels located in the T tubules. Morphine and other opioid agonists in concentrations ranging from 10(-10) to 10(-5) M inhibited K+ contractures. The stereoisomers, dextrorphan and levorphanol, were found to have identical potency in inhibiting high K+ contractures, suggesting that this was a nonstereospecific blockade of voltage-dependent calcium channels by the opioid drugs despite the low effective drug concentrations. In agreement with this conclusion it was found that the inhibition of K+ contractures by the opioids was not antagonized by naloxone. It also was observed using a sucrose gap apparatus that these opioid drugs in concentrations used to block the high K+ contractures did not reduce the K+-induced membrane depolarization. Raising the bathing solution Ca2+ concentration from 1.08 to 5 mM produced a reversal of the opioid-induced block of K+ contractures. Finally it was shown that while opioids completely blocked K+ contractures, they did not produce any effect on caffeine contractures showing that opioids do not deplete intracellular Ca2+ stores or inhibit the release of Ca2+ from intracellular sarcoplasmic reticulum stores. It was concluded that several opioid drugs in very low concentrations block K+ contractures in frog's skeletal muscle by a nonstereospecific block of voltage-dependent slow calcium channels.