Uni-Logo
You are here: Home Seizure Prediction Workshop 2007 Abstracts Animal Models of Epilepsy
Document Actions

Animal Models of Epilepsy

5.1) Dynamics of Spike-Wave Discharges in Young Adult and Aged Fischer 344 Rats

Sandeep P. Nair (1), Deng-Shan Shiau (2), Peter Jukkola (1), J. Chris Sackellares (2), Kevin M. Kelly (1,3)
(1) Department of Neurology, Allegheny Singer Research Institute, Allegheny General Hospital, Pittsburgh, PA, USA; (2) Optima Neuroscience Inc., Gainesville, FL, USA; (3) Drexel University College of Medicine
Several studies with animal models of aging, including studies in our laboratory (Kelly et al, 2001), have shown an age-related increase in the incidence and duration of 7-9 Hz generalized spike-wave discharges (SWDs; absence seizures) as the animal ages. In an effort to elucidate the mechanisms underlying aging-related changes in the expression of SWDs, we studied the dynamical electroencephalographic (EEG) properties associated with spontaneously occurring SWDs in young (4 month) and aged (20 month) Fischer 344 rats. The short term maximum Lyapunov exponent (STLmax), a measure of chaoticity, and the pattern match regularity statistic (PMRS), a measure of signal complexity based on the likelihood of signal pattern similarity, were utilized to extract a dynamical profile of the EEG signal. A statistical comparison of preictal dynamical values (2 min before a SWD) showed no significant difference between the two age groups in either STLmax (p=0.18) or PMRS (p=0.19). However, the same statistical test performed on postictal dynamical values (2 min after a SWD) revealed a significant difference between the two groups in both STLmax (p=0.009) and PMRS values (p= 0.01). A comparison of the difference between average preictal and postictal dynamical values suggested that brain “resetting” to its normal interictal state was more effective in the 4 month cohort compared to the 20 month cohort by both STLmax (p=0.007) and PMRS (p=0.008) values. These preliminary results suggest that brain recovery following SWDs was more sustained in young adult animals compared to aged animals. Supported by a Targeted Research Initiative for Seniors grant from the Epilepsy Foundation to S Nair.

Download the poster as PDF file

5.2) Insights into Epileptiform Activity using Phase Resetting Curves

J.L. Perez Velazquez (1), L. Garcia Dominguez (1), S. Lo, R. F. Galán (2), R. Guevara Erra (1)
(1) Hospital for Sick Children and University of Toronto; (2) Department of Biology, Carnegie Mellon University and Center for the Neural basis of Cognition
Oscillatory coordinated cellular activity is a major characteristic of brain function. In this study, we focus on the characterization of the dynamics of epileptiform activity, based on seizures that manifest themselves with very periodic activity, termed absence seizures. Taking advantage of this long-lasting periodic activity, our approach consists in obtaining experimentally the phase response curves (PRC), which describe the alteration of the phase due to an input at each point of the cycle, and incorporating these into models of coupled oscillators. We use a rat model of absence seizures that results from injection with gamma-hydroxybutyric acid (GHB). As a result, rhythmic synchronized spike-and-wave (SWD) discharges occur in the neocortex and thalamus. Intracerebral recordings are obtained from the cortex and thalamus. PRCs were obtained by stimulating either the thalamus or the cortex, and evaluating the alteration of the oscillation. The electrical stimuli used were the minimal that did not alter profoundly the oscillation. In addition, larger stimulations were tested for their ability to halt the SWD. Only brief stopping (desynchronization) of the SWD was observed in some cases (55%) at large stimulation intensities, phenomenon for which no specific phase of the perturbation was noted. The experimentally obtained PRCs, for the cortex and thalamus, were approximated by polynomials. Incorporating these functions into a Kuramoto-like system of two coupled differential equations representing the time evolution of the phases, we study the phase preferences of the stationary states and their stability, and the results from the model are compared with the experimental recordings.

5.3) The Generation of Epileptic Seizures Requires Interaction of Sclerotic and Intact Networks - A Study in a Mouse Model for Temporal Lobe Epilepsy

Ute Häussler (1,2,3), Ralph Meier (1,2), Antoine Depaulis (3), Ad Aertsen (1,2), Ulrich Egert (1,2)
(1) Bernstein Center for Computational Neurosience Freiburg, Hansastrasse 9a, 79104 Freiburg, Germany; (2) Neurobiology and Biophysics, Institute for Biology III, Schänzlestrasse 1, 79104 Freiburg, Germany; (3) INSERM U836-Inserm Universite Joseph Fourier-CEA, Rue de la Piscine 2280, 38400 St. Martin d’Hères, France
Network structures and dynamics initiating epileptic seizures in mesial Temporal Lobe Epilepsy (MTLE) are still not fully understood. MTLE is accompanied by severe changes of the hippocampal histology, especially cell loss in CA1 and the hilus, granule cell dispersion and mossy fiber sprouting. Excision of those sclerotic areas is necessary to stop epileptic seizures and the development of less invasive therapy options requires the knowledge which brain areas participate in seizure generation. To determine, whether seizures are initiated by the sclerotic hippocampal areas or if a larger network participates in those processes, we used a model for MTLE in mice. A single unilateral injection of kainate into the dorsal hippocampus induced histological changes comparable to hippocampal sclerosis. Recordings of epileptiform events (EE) in-vivo indicated that hypersynchronous spiking involved not only the sclerotic areas of the injected hippocampus but also the temporal hippocampus, although histologically unchanged. To investigate whether initiation of EEs occurred in those sclerotic areas, we recorded slices from this region on multielectrode arrays. Surprisingly, it was impossible to induce EEs there. In contrast, in slices from the temporal hippocampus without obvious histological damage we could induce EEs (bicuculline) with the same rate of recurrence as in controls. Analysis of the coherence between MEA electrodes revealed, however, that slices from epileptic mice showed a changed activity structure within the dentate gyrus. Although apparently structurally intact, the network dynamics in these slices thus differed. The network necessary for EE initiation therefore likely consists of subnetworks with various degrees of degeneration.

5.4) Evolution of Correlations and High Frequency Components in Multi-Channel EEG Recordings from Rat Kindling and Kainate Models of Temporal Lobe Epilepsy.

Alberto Capurro (1,3), Ad Aertsen (1,2), Joacir Cordeiro (3), Ralph Meier (1,2), Monika Haeffner (3), Andreas Schulze-Bonhage (1,3)
(1) Bernstein Center for Computational Neuroscience Freiburg, Albert-Ludwigs-University; (2) Neurobiology and Biophysics, Institute of Biology III, Albert-Ludwigs-University; (3) Center for Epilepsy, Dept. Neurosurgery, University Medical Center Freiburg
We assessed multi-channel correlations and high frequency (HF, 100-500 Hz) content in EEG recordings from two animal models of temporal lobe epilepsy, hippocampal kindling and intra-hippocampal kainate injection. We found that in the kindling model, HF discharges (ripples, 100-400 Hz) developed over subsequent days and were not limited to the kindling site, but also extended to the contralateral hippocampus with high correlation. In contrast, the HF discharges of the kainate model (fast ripples, 400-500 Hz) occurred only in the lesioned hippocampus. During the afterdischarge (AD) of the kindling model, a behavioral seizure (Racine 5) was observed after 18-25 days of stimulation in all rats. The HF power was always high before and during the initial phase of the seizure, but similar ripples were present in the AD many days before seizures appeared (seizures occurred during the primary AD in most cases). The correlation between all pairs of channels increased during the AD from the first kindling day onwards. We did not find a clear difference when behavioral seizures first appeared. During the AD, sudden inversions of the sign of the correlation occurred between pairs of contacts placed in the right hippocampus as well as between pairs placed in both hippocampii. Overall, our data support a key role of HF discharges and enhanced correlations during the epileptogenic process in both, the kindling and kainate models of temporal lobe epilepsy.

Download the poster as PDF file

5.5) Changes in Inter-Hippocampal Coherence Precede Epileptiform Activity in Mice with Induced Epilepsy

Ralph Meier (1), Ute Häussler (1), Ad Aertsen (1,2), Antoine Depaulis (3), Ulrich Egert (1)
(1) Bernstein Center for Computational Neuroscience (Freiburg, D); (2) Faculty of Biology, Albert-Ludwigs-University (Freiburg, D); (3) Inserm U836 Université Joseph Fourier-CEA (Grenoble, F)
The Mesial Temporal Lobe Epilepsy (MTLE) syndrome is among the most prevalent forms of focal epilepsies, however, network structures and understanding of the dynamics involved in the generation of seizures are still elusive. Thus, there is an urgent need to investigate on the time scale of processes initiating epileptic seizures, allowing for more detailed examination of seizure initialization mechanisms. We addressed these questions using the in vivo intrahippocampal kainate model for induced MTLE in mice. These mice show histological changes comparable to human hippocampal sclerosis in the injected hippocampus. We recorded recurrent epileptiform activity (EA) in the injected and in the contralateral, intact hippocampus. We investigated on changes in inter-hippocampal coherence preceding the onset of epileptiform events to determine ongoing seizure generation processes on a timescale suitable for acute intervention. We found, that inter-hippocampal coherence decreased significantly in high frequency bands (> 80 Hz) up to 12 seconds before the onset of EA. This indicates an early decoupling of the ipsilateral hippocampus from the contralateral, intact hippocampus during the seizure initiation phase. Additionally, this time scale limits the possible range for cellular and mechanisms leading to increased synchronicity in the network, ulitmately initiating the seizure. Acknowledgements: This work was supported by the German Federal Ministry of Education and Research (BMBF grant 01GQ0420 to BCCN Freiburg), INSERM, Fondation pour la Recherche Medicale, Fondation de l’Avenir and DAAD (U.H.).

5.6) A Markov Source Model of Seizure Progression

Sridhar Sunderam (1), Nick Chernyy (1), Jonathan Mason (1), Steven L. Weinstein (2), Steven J. Schiff (3,1), Bruce J. Gluckman (1,3)
(1) Engineering Science and Mechanics, Pennsylvania State University, University Park, PA; (2) Neurology, Children's National Medical Center, Washington, DC; (3) Department of Neurosurgery, Pennsylvania State University, University Park, PA
Seizures can have onset, middle and terminal stages with distinctive dynamics, but are usually treated as monolithic events. We are employing the rodent tetanus toxin model of temporal lobe epilepsy to test-bed closed-loop seizure control with low frequency electrical field modulation. In this model, seizures have been characterized (Finnerty and Jefferys, J. Neurophys 2000) to have five distinct stages characterized by the frequency of field postsynaptic potential (FPSP) "spikes". The fourth stage (~9-16Hz) is strongly correlated with secondary generalization (rearing and myoclonus). Seizures start in 2-3 days, and achieve a maximal seizure rate of ~30/day in a week. We have implemented an automated quantitative model of seizure evolution and dynamics with a hidden Markov model (HMM). A HMM is a stochastic model in which the observed time series reflects transitions between discrete underlying states. A HMM comprises "hidden" states with fixed probabilities, state transition probabilities, and state-dependent measurement distributions. With only the number of states prespecified, a HMM was trained on the FPSP spike frequency time series (1/4s bins) derived from sampled data segments that include baseline and seizures. When this trained HMM was then used to determine the most likely state sequence of other seizures, it identified a seizure progression through discrete stages consistent with published descriptions. Notably, a model-identified state with 9-16Hz discharges often preceded clonic behavior. Because seizure patterns vary over time and between animals, such a model and analysis tool will be useful for comparison of different treatment protocols. (Support: NIH R01EB001507, K02MH01493 and R01MH50006.)
« November 2017 »
November
MoTuWeThFrSaSu
12345
6789101112
13141516171819
20212223242526
27282930
Personal tools