- In vivo electrophysiology. In order to study these electrical phenomena we will use an in vivo chronic epilepsy model. Chronic models include methods, in which epilepsy or epilepsy-like conditions are induced by electrical or chemical stimulation in previously healthy (non-epileptic) animals, mostly rats. In the most widely used models recurrent spontaneous seizures develop after a self-sustained status epilepticus (SSSE, 6-20 hours), which is elicited either by sustained electrical stimulation of the hippocampus, amygdala, etc or by the systemic or local administration of the excitotoxic glutamate analogue kainate or the cholinergic (muscarinic) agonist pilocarpine. These post-status epilepsy models are characterized by the fact that neuropathological changes reminiscent of mesiotemporal sclerosis (encountered in many patients with TLE) and recurrent spontaneous seizures develop after the status. Typically, the spontaneous seizures first occur after a latency period following the SSSE of about 3-4 weeks. These models also present major problems: high mortality caused by the status epilepticus, controversies regarding the human relevance of a epilepsy developing on a seriously damaged brain, high frequency of seizures (several/day) that means a sever form of epilepsy, rarely encountered in humans and last but not least these methods produce important discomfort to the animals. Based on these data we decided to use a much modern approach called kindling, a model in which there is a progressive increase in electrographic and behavioral seizure activity resulting from repeated application of short electrical stimuli to limbic brain regions such as amygdala or hippocampus, without inducing status epilepticus. The model reproduces the clinical phenomenology of complex partial seizures and secondary generalized seizures, reproducing almost all aspects of human epileptogenesis. The model is superior for the evolutive understanding of epileptogenesis and has the advantage of establishing limited and localized damage and producing seizures with a very focal onset, a pattern that may be more similar to the human condition (Dichter, 2006).Initially, seizures are stimulation induced, later the seizure severity and duration progressively increases and spontaneous seizures develop, reproducing the most typical evolutive aspects of human epilepsy.
- Imunolabelling. As interneurons are critical elements in controlling network synchrony we are interested in determining what changes may occur in different interneuron populations during epileptogenesis. We will establish whether interneuronal death and/or GABA-A receptor expression-changes are relevant to seizure genesis and determine the link between morphological/molecular changes and the electrical activity, which may lead to epileptogenesis (HFO and ISp). The morphological and immunohistochemical analysis will be performed not only at the end of the experiments ("complete epileptogenesis") but also when detecting important electrophysiological changes (modification of HFO, appearance of ISp, appearance of spontaneous seizures). Cell death will be studied with the Gallyas silver impregnation technique. Using standard immunohistochemistry slice preparation distinct populations of interneurons will be identified by immunostaining against two calcium binding proteins, calbindin and parvalbumin (PV), the neuropeptide somatostatine (SOM) and the alfa1 subunit of GABA-A receptors (alfa1-GABA-A-R). These markers were chosen so that besides positive identification of interneurons the differentiation of the most important interneuronal subpopulations (basket, bistratified, axo-axonic cells) will be possible. Calbindin immunoreactivity is observed in interneurons that make synaptic connections with a variety of cells, including smooth non-pyramidal neurons [glutamic acid decarboxylase (GAD) positive], and pyramidal and spiny non-pyramidal cells in the hippocampus and/or cortex. PV immunoreactive interneurons, in contrast, primarily innervate principal cell soma and their axon initial segment. Especially in the CA1 area SOM is expressed in bistratified cells co-expressing PV. The alfa1-GABA-A-R is expressed both in PV expressing principal cells and interneurons in the hippocampus with a strong immunoreactivity of cell bodies and dendrites which probably represents extrasynaptic receptors contributing to the strong tonic inhibition of hippocampal neurons.
- RT-QPCR. From the slices several limbic structures will be dissected and used separately for RNA extraction and consecutive RT-QPCR.
- Computer modelling. The results of the electrophysiological measurements will be implemented in computational models in order to improve the methods used, to reduce the number of animals needed for the experiments and for a better understanding of epileptogenic processes. This method was learned and extensively used by the members of the Department of Physiology during several research visits to the Anatomical Neuropharmacology Unit, Medical Research Council, Oxford University, UK, led by Prof. Dr. Peter Somogyi) and also during a bilateral project agreement with the Theoretical Neurobiology Unit, University of Antwerp, Belgium, led by Prof. Dr. Erik De Schutter. The method is readily implementad at the Department. For the simulations the NEURON program developed by M. Hines and T. Carnavale will be used, which is a flexible program created for implementing biologically realistic models of neurons generating chemical and electrical signals and of neural networks. The processing of information in the brain is a result of the propagation and interaction of electrical and chemical signals in and between neurons. This involves non-linear mechanisms that extend on a large scale of spatial and temporal magnitudes. The equations that describe the neural mechanisms usually don't have analytical solutions. Non-linearities and spatio-temporal complexities are specific for biological systems, so the use of qualitative and quantitative models that don't consider these characteristics is substantially limited. We will use multicompartmental numeric models that avoid this obstacle. Within the proiect we will create models of interneurons that have been recently characterized morphologically and electrophysiologically (basket, bistratified, OLM <oriens - lacunosum-moleculare>, axo-axonic cells). We will try to realize interneuronal connections that best fit the anatomical situation. Then we will determine the possibilities of simplification of these models, without significantly changing the electrical behaviour of the cells. This simplification allows a lot shorter simulation time and subsequently the built of a large neural network, a more realistic model of the structure and activity of the hippocampus. In this stage we will build neural networks that consider the morphological/molecular results obtained in the previous phase. The quantitative and qualitative elements necessary for the model to reproduce the high frequency oscillations recorded in vivo will be determined. We will build several models that reproduce the different stages of epileptogenesis which will be compared to determine the underlying pathophysiological mechanisms.
| 