Abstracts

For many years the mechanisms underlying the generation of seizures in humans has eluded clear definition. Surprisingly, decades of recordings from human resected neocortical tissue slices have failed to distinguish epileptic from non-epileptic tissue at the intracellular level. Therefore, no meaningful correlations or inferences have been drawn between information collected at the single cell level and the clinical level, such as clinical seizure control or electrographic clinical seizures. The purpose of this study is to suggest that correlations may be found between intracellular electrical abnormalities and more macro measures of clinical epilepsy.
Human cortical tissue was removed from 2 patients with intractable epilepsy presumed secondary to viral encephalitis. The tissue was obtained using a protocol approved by the IRB. The sites selected for the slice studies were determined by electrocorticography, source localization and clinical relevance. Upon resection, the tissue was placed into artificial CSF (aCSF in mM: 118 NaCl, 3 KCl, 1.5 CaCl[sub]2[/sub], 1 Mg Cl[sub]2[/sub], 25 NaHCO[sub]3[/sub], 1 NaH[sub]2[/sub]PO[sub]4[/sub] and 30 D-glucose, pH of 7.4) bubbled with carbogen (95% O[sub]2[/sub] and 5 % CO[sub]2[/sub]). Slices (500 [mu]m) were sectioned perpendicular to the gyri of the seizure focus such that the 6-layered cortex could be identified. The slices were immediately transferred to aCSF bubbled with carbogen at room temperature. Experiments were conducted in a recording chamber at 29[deg] C. After 30 minutes the [K⁺][sub]o[/sub] concentration of the aCSF was raised from 3 to 5 mM. Population recordings were obtained with suction electrodes positioned onto the surface of the cortical layers. Intra- and extracellular recordings were simultaneously obtained from human cortical slices.
Seizure-like activity in the tissue slices was induced by application of 5-15 [mu]M NMDA and/or 20 [mu]M bicuculline. In tissue from one patient, application of lamotrigine (160 [mu]M) reduced the population burst amplitude (n=2 slices) and desynchronized the unit response from the population response. However, the unit still demonstrated bursting activity. In contrast, phenytoin had no observable effect. Similarly, phenytoin had no clinical effect, whereas lamotrigine was highly effective. In a second case, application of anticonvulsants to resected tissue had no effect and correspondingly, did not improve the patient clinically.
This study provides encouragement that it may be possible to correlate the behavior of single patched cells with macro measures of clinical epilepsy, such as seizure frequency and EEG changes. If this proves true for an even larger group of patients, this gives support for the use of human tissue slice preparations for the study of pediatric neocortical epilepsies.
[Supported by: PACE (CJM)
Falk Foundation (CJM, WvD, KEH)
PEW Fellowship (FP)
NIH F32HL67659 (AKT)
NIH HL60120 (JMR)]