H. Straub

Comparison of Brain Extracellular Fluid, Brain Tissue, Cerebrospinal Fluid, and Serum Concentrations of Antiepileptic Drugs Measured Intraoperatively in Patients with Intractable Epilepsy

Summary:u2002 Purpose: The mechanisms of drug resistance in epilepsy are only incompletely understood. According to a current concept, overexpression of drug efflux transporters at the blood–brain barrier may reduce levels of antiepileptic drugs (AEDs) in epileptogenic brain tissue. Increased expression of drug efflux transporters such as P‐glycoprotein has been found in brain tissue surgically resected from patients with medically intractable epilepsy, but it is not known whether this leads to decreased extracellular (interstitial) AED concentrations in affected brain regions. This prompted us to measure concentrations of AEDs in the extracellular space of human neocortical tissue by using intraoperative microdialysis (IOMD) in those parts of the brain that had to be removed for therapeutic reasons. For comparison, AED levels were determined in brain tissue, subarachnoid CSF, and serum.

Melatonin reduces low-Mg2+ epileptiform activity in human temporal slices

Scizure susceptibility waxes and wanes in an apparently circadian manner in many epileptic patients. Fluctuations of melatonin concentration with highest levels during the night and lowest levels in the early morning could be involved in this phenomenon. Therefore, the action of melatonin on epileptic activity was tested. The experiments were carried out on human temporal neocortical slices cut from tissue resected for surgical treatment of epilepsy. Autoradiographic studies were performed on parallel slices with 100–120 pmol 2-[125I]iodomelatonin/l in the absence or presence of unlabelled melatonin. High-affinity binding sites of melatonin could be demonstrated in layers II–V of the temporal cortex. The binding was saturable, specific and occurred with low capacity. In electrophysiological studies, epileptiform field potentials were elicited by omission of Mg2+ from the superfusate and recorded from layers II–V. The frequency of occurrence of epileptiform field potentials was reduced to 0.5 of the initial value with application of melatonin (10 and 100 nmol/l) in each case. This effect was reversible upon washing. The findings favour the hypothesis that melatonin depresses epileptiform neuronal activity through specific neocortical receptors.

Differential sensitivity to induction of spreading depression by partial disinhibition in chronically epileptic human and rat as compared to native rat neocortical tissue

Spreading depression (SD) is characterized by a transient breakdown of neuronal function concomitant with a massive failure of ion homeostasis. It is a phenomenon that can be induced in neocortical tissue by raising excitability, e.g. injection of K(+), application of glutamatergic agonists, or blocking Na(+)/K(+) ATPase. Here we report a novel method of SD induction using minimal disinhibition with application of low concentrations (5 microM) of the GABA(A) receptor blocker bicuculline. This procedure-while subthreshold for epileptiform activity-readily induced spontaneous SDs in native rat neocortical slices, accompanied by typical depolarizations of neurons and glial cells. In contrast, in human neocortical preparations obtained from epilepsy surgery, in approximately 20% of the slices spontaneous epileptiform activity appeared with this bicuculline dosage without SDs. Raising the concentration of bicuculline to an epileptogenic dose (10 microM) in human tissue also resulted in the generation of epileptiform activity only. Likewise, in slices from pilocarpine-treated, chronically epileptic rats, bicuculline also only induced epileptiform activity without eliciting SDs. The experiments indicate that chronic epilepsy causes a differential sensitivity to partial GABA(A) receptor blockade with regard to induction of SD.

Simultaneous blockade of intracellular calcium increases and of neuronal epileptiform depolarizations by verapamil.

The specific L-type calcium channel blocker verapamil exerts an antiepileptic effect on neurons. This effect is assumed to depend on the blockade of transmembraneous calcium flux during epileptic discharges. In order to test this hypothesis, fura-dextran loaded snail neurons were rendered epileptic by pentylenetetrazole (40 mmol/l). The effect of verapamil (20 or 40 mumol/l) on free intracellular calcium ([Ca2+]i) transients was investigated by means of fluorescence ratio-imaging and simultaneous intracellular membrane potential recording. During epileptic depolarization [Ca2+]i increased especially in the outermost submembraneous areas of the neuron. [Ca2+]i reached peak values 6-22 s after the onset of epileptic depolarizations. Application of verapamil progressively shortened the epileptic depolarizations. This shortening of epileptic depolarizations developed along with a diminution of the submembraneous calcium signals down to noise level. The effect was found to be reversible. It is concluded that the antiepileptic effect of verapamil depends largely on its ability to block transmembraneous calcium flux.

The effects of verapamil and flunarizine on epileptiform activity induced by bicuculline and low Mg2+ in neocortical tissue of epileptic and primary non-epileptic patients

In human neocortical slices the specific L-type calcium channel blocker verapamil had been shown to be antiepileptic in the low Mg(2+)-model of epilepsy. The present investigation demonstrated: (1) verapamil exerted also an antiepileptic effect on epileptiform field potentials (EFP) induced by the GABAA-antagonist bicuculline. (2) The unspecific calcium channel modulator flunarizine, which in contrast to verapamil penetrates the blood-brain barrier, depressed EFP in the low Mg(2+)-model and in the bicuculline model. (3) There was no significant difference in the antiepileptic efficacy of verapamil and flunarizine in epileptic (epilepsy surgery) and primary non-epileptic (tumor surgery) neocortical slices.

Changes of extracellular calcium concentration induced by application of excitatory amino acids in the human neocortex in vitro

The influence of the glutamate subreceptor agonists N-methyl-D-aspartate (NMDA) and alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) on cortical field potentials and on changes in extracellular free calcium concentration ([Ca2+]o) was tested on human neocortical slices (eleven from nine different patients). The tissue used was a small portion of that which is normally removed for the treatment of a brain tumor. [Ca2+]o and field potentials were measured by Ca(2+)-selective microelectrodes. Local pressure-microejection of NMDA (100 mumol/l)- and AMPA (1 mmol/l)-induced negative field potentials with maximal amplitudes of 0.9 +/- 0.1 mV (11 slices, mean +/- S.E.M.) and 1.0 +/- 0.1 mV (nine slices), respectively. The negative field potentials induced by NMDA were accompanied by monophasic decreases of [Ca2+]o (0.8 +/- 0.1 mmol/l, nine slices). AMPA elicited no (three slices) or only minor decreases of [Ca2+]o (0.2 +/- 0.1 mmol/l, five slices). The responses to the glutamate subreceptor agonists NMDA and AMPA were reversibly depressed by adding their specific antagonists DL-2-amino-5-phosphonovalerate (APV, 100 mumol/l, six slices) and 6-cyano-7-nitroquinoxalin-2,3-dion (CNQX, 5 mumol/l, four slices), respectively. The results correspond to findings in animal experiments and are consistent with the interpretation that in the human neocortex the Ca2+ permeability of channels gated by NMDA is higher than those gated by AMPA.

Picrotoxin-induced epileptic activity in hippocampal and neocortical slices (guinea pig): suppression by organic calcium channel blockers

Epileptic activity induced by the GABAA receptor antagonist bicuculline is known to be blocked by organic calcium antagonists. To further analyse the mechanism underlying convulsant activity induced by substances reducing GABA-mediated synaptic transmission, the effect of organic calcium channel blockers on epileptic activity induced by the GABAA channel blocker picrotoxin in hippocampal and neocortical slices of guinea pigs were investigated. Verapamil and flunarizine suppressed paroxysmal depolarization shifts (PDS) of single neurons and accompanying epileptic field potentials (EFP). As a measure of drug action the repetition rate of epileptic events were used. The depression down to 10% the initial value (90% depression) is indicated. In the hippocampus verapamil suppressed PDS/EFP within 70 +/- 16 min (40 mumol/l) and within 39 +/- 5 min (60 mumol/l). This suppression was reversible with washout of verapamil. Flunarizine irreversibly blocked EFP/PDS within 108 +/- 14 min (18 mumol/l). In the neocortex verapamil reversibly suppressed EFP within 146 +/- 6 min (40 mumol/l) and 127 +/- 26 min (60 mumol/l). Flunarizine irreversibly blocked EFP within 181 +/- 30 min (3 mumol/l) and 109 +/- 13 min (18 mumol/l). The results suggest that voltage dependent calcium channels are essentially involved in picrotoxin-induced epileptic activity.

Blockade of astrocyte metabolism causes delayed excitation as revealed by voltage-sensitive dyes in mouse brainstem slices

Abstract.Fluoroacetate is known to block cell metabolism and to change potassium conductances selectively in astrocytes. In a functional neuronal network with ongoing activity, we investigated the effects of such a blockade of the astrocytic metabolism by fluoroacetate on neuronal signal propagation. Transverse 400-µm slices were prepared from the caudal medulla of mice of postnatal day 3–8, which contained the hypoglossal nucleus receiving excitatory synaptic input from the ventral respiratory group. Propagation of excitation within this network was measured by optical imaging using the voltage-sensitive dye RH 795. A 464-element photodiode array allowed fast recordings of voltage changes within a small population of cells. The spatial and temporal resolution was advanced to 32xa0µm and 1.27xa0ms, respectively. Changes of cellular membrane potential levels were expressed as relative changes of fluorescence (ΔI/I). Stimulus-evoked excitation of neurons propagating from the ventral respiratory group to the hypoglossal nucleus peaked after 7.2±0.6xa0ms (n=6). The latency of this early excitatory response is consistent with the time course of stimulus-evoked EPSPs in whole-cell recordings. Mean changes of fluorescence in the hypoglossal nucleus were −2.1±0.5×10−3 (ΔI/I). After incubation in 1xa0mM fluoroacetate, the early depolarization was reduced to 69.1±9.8% of control (n=6, p=0.034). Additionally, fluoroacetate induced a delayed excitatory response, such that fluorescence intensity did not return to baseline within 1xa0s. Propagation velocity and spatial distribution of the voltage signal were not affected by fluoroacetate. Our results suggest that blockade of astrocyte metabolism impairs fast synaptic transmission and induces a delayed excitation, probably resulting from the combination of reduced repolarization of neurons and persistent depolarization of astrocytes.

Anoxic terminal negative DC-shift in human neocortical slices in vitro

In animal models, the hallmark of a hypoxic condition is a strong negative shift of the DC potential (anoxic terminal negativity, ATN). This DC-shift is interpreted to be primarily due to a breakdown of the membrane potential of neurons. Such massive neuronal depolarizations have not been reported for all human neocortical neurons in vitro even during prolonged hypoxic periods. This poses the question whether ATN develop also in human neocortical slices made hypoxic. ATN could be observed when human brain slice preparations (n = 15, 13 patients) were subjected to periods of hypoxia (10 to 120 min). These ATN were usually monophasic and appeared with a latency of 16 +/- 4 min (mean +/- S.E.M.). Separating the ATN according to their slopes of rise, steep (> 10 mV/min) and flat (< 10 mV/min) ATN could be distinguished. Steep and flat ATN may be regarded as two different entities of reactions since steep ATN had also greater amplitudes and slopes of decay as compared a flat ATN. With repetitive hypoxias, the latency of both the steep and flat ATN was reduced for the following hypoxic episodes. During hypoxic DC-shifts, evoked potentials were suppressed. With the 1st through 4th hypoxia, they recovered fully within 30 min after reoxygenation when hypoxia was terminated at the plateau of ATN; with extension of hypoxia, recovery was only partial. From the 5th hypoxia onwards, recovery usually did not take place or was not complete.

Strychnine-induced epileptiform activity in hippocampal and neocortical slice preparations: suppression by the organic calcium antagonists verapamil and flunarizine.

Alongside GABA, glycine is the major inhibitory transmitter in the central nervous system. Application of the glycine receptor blocker strychnine is known to evoke epileptiform phenomena. The present paper addresses the question whether postsynaptic calcium currents through L-type channels contribute to strychnine-induced epileptiform field potentials (EFP). To test for this, the antiepileptic effect of the specific L-type calcium channel blocker, verapamil, in hippocampal and neocortical slices was investigated. In parallel with this, the antiepileptic efficacy of the unspecific calcium channel modulator, flunarizine, was tested with respect to pharmacotherapy of epilepsies. In both preparations, the L-type calcium channel blocker, verapamil, was able to suppress EFP. In neocortical slices, EFP were blocked in all experiments, whereas in hippocampal slices, in 3 out of 11 experiments, no complete suppression occurred. Flunarizine acted in a similar way. It is concluded that L-type calcium channels are involved in strychnine-induced epilepsy, but to a greater extent in the neocortex than in the hippocampus.