Search

Loop diuretics such as furosemide and bumetanide, which act by inhibiting the Na-K-2Cl cotransporter NKCC2 at the thick ascending limb of the loop of Henle, have been shown to exert anti-seizure effects. However, the exact mechanism of this effect is not known. For bumetanide, it has been suggested that inhibition of the NKCC isoform NKCC1 in the membrane of brain neurons may be involved; however, NKCC1 is expressed by virtually all cell types in the brain, which makes any specific targeting of neuronal NKCC1 by bumetanide impossible. In addition, bumetanide only poorly penetrates the brain. We have previously shown that loop diuretics azosemide and torasemide also potently inhibit NKCC1. In contrast to bumetanide and furosemide, azosemide and torasemide lack a carboxylic group, which should allow them to better penetrate through biomembranes by passive diffusion. Because of the urgent medical need to develop new treatments for neonatal seizures and their adverse outcome, we evaluated the effects of azosemide and torasemide, administered alone or in combination with phenobarbital or midazolam, in a rat model of birth asphyxia and neonatal seizures. Neither diuretic suppressed the seizures when administered alone but torasemide potentiated the anti-seizure effect of midazolam. Brain levels of torasemide were below those needed to inhibit NKCC1. In addition to suppressing seizures, the combination of torasemide and midazolam, but not midazolam alone, prevented the cognitive impairment of the post-asphyxial rats at 3 months after asphyxia. Furthermore, aberrant mossy fiber sprouting in the hippocampus was more effectively prevented by the combination. We assume that either an effect on NKCC1 at the blood-brain barrier and/or cells in the periphery or the NKCC2-mediated diuretic effect of torasemide are involved in the present findings. Our data suggest that torasemide may be a useful option for improving the treatment of neonatal seizures and their adverse outcome.

Objectives: The loop diuretic bumetanide has been proposed previously as an adjunct treatment for neonatal seizures because bumetanide is thought to potentiate the action of γ--aminobutyric acid (GABA)ergic drugs such as phenobarbital by preventing abnormal intracellular accumulation of chloride and the subsequent "GABA shift." However, a clinical trial in neonates failed to demonstrate such a synergistic effect of bumetanide, most likely because this drug only poorly penetrates into the brain. This prompted us to develop lipophilic prodrugs of bumetanide, such as the N,N-dimethylaminoethyl ester of bumetanide (DIMAEB), which rapidly enter the brain where they are hydrolyzed by esterases to the parent compound, as demonstrated previously by us in adult rodents. However, it is not known whether esterase activity in neonates is sufficient to hydrolyze ester prodrugs such as DIMAEB. Methods: In the present study, we examined whether esterases in neonatal serum of healthy term infants are capable of hydrolyzing DIMAEB to bumetanide and whether this activity is different from the serum of adults. Furthermore, to extrapolate the findings to brain tissue, we performed experiments with brain tissue and serum of neonatal and adult rats. Results: Serum from 1- to 2-day-old infants was capable of hydrolyzing DIMAEB to bumetanide at a rate similar to that of serum from adult individuals. Similarly, serum and brain tissue of neonatal rats rapidly hydrolyzed DIMAEB to bumetanide. Significance: These data provide a prerequisite for further evaluating the potential of bumetanide prodrugs as add-on therapy to phenobarbital and other antiseizure drugs as a new strategy for improving pharmacotherapy of neonatal seizures. © 2020 The Authors. Epilepsia published by Wiley Periodicals LLC on behalf of International League Against Epilepsy

Because of its potent inhibitory effect on the Na+-K+-2Cl- symporter isotype 1 (NKCC1) in brain neurons, bumetanide has been tested with varying results for treatment of seizures that potentially evolve as a consequence of abnormal NKCC1 activity. However, because of its physicochemical properties, bumetanide only poorly penetrates into the brain. We previously demonstrated that NKCC1 can be also inhibited by azosemide and torasemide, which lack the carboxyl group of bumetanide and thus should be better brain-permeable. Here we studied the brain distribution kinetics of azosemide and torasemide in comparison with bumetanide in mice and used pharmacokinetic-pharmacodynamic modelling to determine whether the drugs reach NKCC1-inhibitory brain concentrations. All three drugs hardly distributed into the brain, which seemed to be the result of probenecid-sensitive efflux transport at the blood-brain barrier. When fractions unbound in plasma and brain were determined by equilibrium dialysis, only about 6-17% of the brain drug concentration were freely available. With the systemic doses (10 mg/kg i.v.) used, free brain concentrations of bumetanide and torasemide were in the NKCC1-inhibitory concentration range, while levels of azosemide were slightly below this range. However, all three drugs exhibited free plasma levels that would be sufficient to block NKCC1 at the apical membrane of brain capillary endothelial cells. These data suggest that azosemide and torasemide are interesting alternatives to bumetanide for treatment of seizures involving abnormal NKCC1 functionality, particularly because of their longer duration of action and their lower diuretic potency, which is an advantage in patients with seizures.

Summary Ethanol is commonly used as a solvent in injectable formulations of poorly water-soluble drugs. The concentrations of ethanol in such formulations are generally considered reasonably safe. It is long known that ethanol can potentiate central effects of sedatives and tranquillizers, particularly the benzodiazepines, most likely as a result of a synergistic interaction at the GABA A receptor. However, whether this occurs at the low systemic doses of ethanol resulting from its use as solvent in parenteral formulations of benzodiazepines is not known. In the present study we evaluated whether a commercial ethanol-containing aqueous solution of diazepam exerts more potent anti-seizure effects than an aqueous solution of diazepam hydrochloride or an aqueous emulsion of this drug in the intrahippocampal kainate model of temporal lobe epilepsy in mice. Spontaneous epileptic seizures in this model are known to be resistant to major antiepileptic drugs. Administration of the ethanol-containing formulation of diazepam caused an almost complete suppression of seizures. This was not seen when the same dose (5mg/kg) of diazepam was administered as aqueous solution or emulsion, although all three diazepam formulations resulted in similar drug and metabolite concentrations in plasma. Our data demonstrate that ethanol-containing solutions of diazepam are superior to block difficult-to-treat seizures to other formulations of diazepam. To our knowledge, this has not been demonstrated before and, if this finding can be translated to humans, may have important consequences for emergency treatment of acute seizures, series of seizures, and initial treatment of status epilepticus in patients.

Recent pharmacological studies have shown antidystonic effects of NMDA and non-NMDA receptor antagonists in an inbred line of Syrian hamsters (dt(sz)) with primary generalized dystonia, i.e. a neurological syndrome of sustained muscle contractions which occurs in the absence of any pathomorphological alterations. This prompted us to examine the levels of kynurenic acid (KYNA), the endogenous broad spectrum antagonist of the excitatory amino acid receptors. The concentrations of KYNA were determined by HPLC in forebrain, cerebellum, brainstem and plasma in dystonic hamsters and age-matched non-dystonic controls. Dystonia in mutant hamsters is transient and disappears completely at the age of 70 days. In order to examine if neurochemical changes are associated with dystonia, KYNA was determined at the age of maximum severity (30 days) and after remission (70 days). The levels of KYNA were significantly increased in forebrain, cerebellum and brainstem (37-130 percent) in dystonic hamsters at the age of maximum severity of dystonia (30 days of life) compared to both a genetically related non-dystonic inbred line and a non-related outbred line of hamsters. The increase of KYNA in brain regions was accompanied by enhanced plasma levels. However, there was no correlation between brain and plasma levels. Since the changes in KYNA levels disappeared in parallel with dystonia (70 days), the present data provide further evidence that abnormal activity of excitatory amino acids may be pathogenetically involved in dystonia in mutant hamsters. With regard to the recent finding of antidystonic effects of glutamate receptor antagonists the increased levels of kynurenic acid may be interpreted as a counteracting process to an overactivity of the glutamatergic system.

The endogenous neuroinhibitory and neuroprotective excitatory amino acid receptor antagonist kynurenic acid has been hypothetically linked to the pathogenesis of epilepsy and several other brain disorders. In the present study, alterations in kynurenic acid levels were examined in the kainate model of temporal lobe epilepsy. Kainate was systemically injected in rats at a dose (10 mg/kg s.c.) which induces a characteristic behavioural syndrome with stereotypies and focal (limbic) and generalized seizures, eventually progressing into severe status epilepticus. Kynurenic acid was determined 3 h after kainate injection in various brain regions (olfactory bulb, frontal cortex, piriform cortex, amygdala, hippocampus, nucleus accumbens, caudate/putamen, thalamus, superior and inferior colliculus, pons and medulla, and cerebellar cortex) and in plasma, using a sensitive high-performance liquid chromatographic method. When data were analysed irrespective of individual seizure severity, significant increases in kynurenic acid were determined in all brain regions examined except the hippocampus, nucleus accumbens and pons/medulla. The most marked (200-500%) increases above controls were seen in the piriform cortex, amygdala, and cerebellar cortex. Furthermore, a significant kynurenic acid increase of about 200% above control was determined in plasma. When kynurenic acid levels were determined in subgroups of rats with different behavioural alterations in response to kainate, the most marked kynurenic acid increases were seen in subgroups with status epilepticus. Rats which only developed mild (focal) seizures or stereotyped behaviours (wet dog shakes) also exhibited significantly increased kynurenic acid levels, thus indicating that the increase in kynurenic acid in response to kainate was not solely due to sustained convulsive seizure activity. Whereas it was previously proposed that kynurenic acid is involved only in later stages of seizure disorders, the present data demonstrate that marked increases in central and peripheral kynurenic acid levels occur early after the onset of neuroexcitation, at least in the kainate model.

The NMDA/glycine receptor partial agonist, D-cycloserine, has recently been reported to exert anticonvulsant effects in different seizure models in mice and rats. In view of the high doses (> 100 mg/kg) needed for these effects, actions other than those mediated by the glycine site might be involved. In this respect, inhibition of pyridoxal phosphate-dependent enzymes involved in amino acid metabolism might play a role. In the present experiments, D-cycloserine was administered at an anticonvulsant dose (320 mg/kg) to mice and rats and levels of 11 amino acids, including several neurotransmitters, were determined in brain cortex and plasma at different times after administration. In addition, the concentration of D-cycloserine was determined in plasma and brain. Compared to peak concentrations of D-cycloserine in plasma, only about 20% of D-cycloserine appeared in the brain. The only marked alteration in brain amino acids was an increase in alanine levels, while amino acids acting as neurotransmitters were hardly altered. The data indicate that the anticonvulsant action of D-cycloserine is not secondary to changes in levels of amino acid neurotransmitters.

The technique of estimating gamma-aminobutyric acid (GABA) turnover by inhibiting its major degrading enzyme GABA-T (4-aminobutyrate:2-oxoglutarate aminotransferase; EC 2.6.1.19) and measuring GABA accumulation has been used repeatedly, but, at least in rats, its usefulness has been limited by several difficulties, including marked differences in the degree of GABA-T inhibition in different brain regions after systemic injection of GABA-T inhibitors. In an attempt to improve this type of approach for measuring GABA turnover, the time course of GABA-T inhibition and accumulation of GABA in 12 regions of rat brain has been studied after systemic administration of aminooxyacetic acid (AOAA), injected at various doses and with different routes of administration. A total and rapidly occurring inhibition of GABA-T in all regions was obtained with intraperitoneal injection of 100 mg/kg AOAA, whereas after lower doses, marked regional differences in the degree of GABA-T inhibition were found, thus leading to underestimation of GABA synthesis rates, e.g., in substantia nigra. The activity of the GABA-synthesizing enzyme GAD (L-glutamate-1-decarboxylase; EC 4.1.1.15) was not reduced significantly at any time after intraperitoneal injection of AOAA, except for a small decrease in olfactory bulbs. Even the highest dose of AOAA tested (100 mg/kg) was not associated with toxicity in rats, but induced motor impairment, which was obviously related to the marked GABA accumulation found with this dose. The increase in GABA concentrations induced with intraperitoneal injection of 100 mg/kg AOAA was rapid in onset, allowing one to estimate GABA turnover rates from the initial rate of GABA accumulation, i.e., during the first 30 min after AOAA injection. GABA turnover rates thus determined were correlated in a highly significant fashion with the GAD activities determined in brain regions, with highest turnover rates measured in substantia nigra, hypothalamus, olfactory bulb, and tectum. Pretreatment of rats with diazepam, 5 mg/kg i.p., 5-30 min prior to AOAA, reduced the AOAA-induced GABA accumulation in all 12 regions examined, most probably as a result of potentiation of postsynaptic GABA function. The data indicate that AOAA is a valuable tool for regional GABA turnover studies in rats, provided the GABA-T inhibitor is administered in sufficiently high doses to obtain complete inhibition of GABA degradation.