"A combined hardware and software platform for ambulatory seizure onset detection is presented. The hardware is developed around commercial off-the-shelf components, featuring ADS1299 analog front ends for electroencephalography from Texas Instruments and a Broadcom ARM11 microcontroller for algorithm execution. The onset detection algorithm is a patient-specific support vector machine algorithm. It outperforms a state-of-the-art detector on a reference data set, with 100% sensitivity, 3.4 second average onset detection latency, and on average 1 false positive per 24 hours. The more comprehensive European Epilepsy Database is then evaluated, which highlights several real-world challenges for seizure onset detection, resulting in reduced average sensitivity of 93.5%, 5 second average onset detection latency, and 85.5% specificity. Algorithm enhancements to improve this reduced performance are proposed."

Epilepsy -- the condition of recurrent, unprovoked seizures -- manifests in brain voltage activity with characteristic spatio-temporal patterns. One of the patterns typically observed during a seizure is a traveling wave. To characterize these waves, we analyze high-density local field potential (LFP) data recorded in vivo from human cortex during a seizure from three patients. We show that traveling wave patterns emerge in the LFP with consistent quantitative features. Using a mean-field approach we model the neuronal population activity observed in the LFP and obtain explicit traveling wave solutions for this model. We then employ the LFP data to constrain the model and obtain parameter configurations that support traveling wave solutions with features consistent with the observed LFP waves. In particular, our model formulation is able to capture the "reverberation" of the activity following the traveling wave that was found in the clinical data. We obtain biologically reasonable parameter estimates for two important features: the timescales of the model and the extent of the connectivity. In this way, we link the observed LFP waves during seizure to proposed biological mechanisms. We also study the linear stability of the traveling wave solutions by constructing an Evans function. We find for some parameters the existence of two waves: one wave is slow and narrow whereas the other wave is fast and wide. Moreover, the fast and wide wave has speed and width consistent with the observed LFP waves. We numerically analyze the Evans function to determine stability (instability) of the fast (slow) wave.

GRIN2A and GRIN2B encode the GluN2A and GluN2B subunits of the NMDA receptor, a subtype of ionotropic glutamate receptor that displays voltage-dependent block by Mg2+ and a high permeability to Ca2+. These receptors play important roles in synaptogenesis, synaptic transmission and synaptic plasticity, as well as contributing to neuronal loss and dysfunction in several neurological disorders. Recently, individuals with a range of childhood onset epilepsies, intellectual disability and other neurodevelopmental abnormalities have been found to carry heterozygous gene-disrupting or protein-altering point mutations in GRIN2A and GRIN2B. This thesis addresses the hypothesis that these point mutations cause key functional disturbances to NMDA receptor properties that contribute to neurodevelopmental disorders. To test this hypothesis, a group of related mutations were selected for functional assessment in heterologous systems: four missense mutations affecting residues in or near the subunit pore regions, all of which are associated with epilepsy and intellectual disability. To model the impact of gene disrupting mutations in GRIN2A, a preliminary analysis of the functional consequences of GluN2A haploinsufficiency in a genetically modified rat was also performed. Three of the four missense mutations were found to be associated with profound alterations in fundamental NMDA receptor properties: compared to wild type, GluN2AN615K was found to reduce Mg2+ block, GluN2BN615I and GluN2BV618G to cause potentiation by Mg2+, and GluN2AN615K and GluN2BN615I showed reduced conductance. GluN2AR586K was not found to influence the parameters assessed. When GluN2AN615K was expressed alongside wild type subunits in the same NMDA receptor, it was found to have a dominant negative effect. Finally, I established successful gene targeting in a new rat Grin2A knock-out model, and observed that heterozygous neurons had lower GluN2A protein expression and current density, making a good model to study human epilepsies associated with loss of a GRIN2A allele. This thesis provides evidence that three missense mutations in GRIN2A and GRIN2B affect physiologically important properties of the NMDA receptor, and that GluN2A haploinsufficiency influences important neural properties in vitro. This is consistent with these mutations causing disease and highlights these and related mutations as potential therapeutic targets in the future.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / In last years, considerable efforts have been made to identify neuroactive and neuroprotective peptides derived from the venom of different natural species. In this work, were studied the activity of Dinoponera quadriceps native (DqV) and denatured (DqDV) venom on chemically induced seizures models in vivo and on in vitro cortical astrocytes viability. Male Swiss mice (28- 33g) were pretreated with DqV (0.1 or 0.5 mg/kg, e.v., n = 6-8), DqDV (0.5 or 2.0 mg/kg, i.p., n = 6-8) or DqDV (0.1 or 0.5 mg/kg, e.v., n = 6-8). 30 minutes after the intraperitoneal pretreatment or ten minutes after intravenous pretreatment with the venom was induced seizures in all animals by the administration of pentylenetetrazole (80 mg/kg) pilocarpine (400 mg/kg) or strychnine (3.0 mg/kg). In behavioral analysis, we recorded the time to the first seizure and to death and the percentage of survival. To determine the parameters of oxidative stress was dissected three brain areas (prefrontal cortex, hippocampus and striatum) of animals used in behavioral analysis, in order to determine the degree of lipid peroxidation, by measuring the levels of malondialdehyde (MDA), and the content of nitrite. In in vitro assay, cell viability of cortical astrocytes was determined after treatment with different concentrations of DqV, PTZ and DqV + PTZ. The data were analyzed by ANOVA followed by a Student Newman-Keuls (for in vivo tests) or Bonferroni (for in vitro experiments) as post hoc tests. It was observed that the DqV had effect only on the PTZ model, both in behavioral analysis as for the determination of the oxidative parameters. Pretreatment with DqV significantly reduced the time until the occurrence of first seizure (0.1 mg/kg: 77.83 Â 5.27 compared to the 101.0 Â 3.31 seconds in the control group; 0.5 mg/kg: 74.43 Â 3.94 compared to the 101.0 Â 3.31 seconds in the control group), while DqDV caused an increase in the percentage of survival when pretreated by i.p. (0.5 mg/kg: 25% of survival compared with 0% in the control group, 2.0 mg/kg: 62.5% of survival compared with 0% in control group) and e.v (0.5 mg/kg: 28.57% of survival compared with 0% in control group). routes. In relation to the oxidative stress parameters, both pretreatments with DqV and with DqDV caused increases of MDA levels in all three brain areas analyzed. The nitrite content also increased after pretreatment with DqV in the three areas of the brain and after pretreatment with DqDV via e.v., only in the hippocampus. About cell viability assays, were observed that DqV was not able to change this parameter. The PTZ reduced the cell viability of astrocytes in a dose-dependent way, with an IC 50 (cytotoxicity index to 50% of the cell population under study) corresponding to 33.12 mM. The combined treatment of DqV (100 Âg/mL) and PTZ (IC 50) also caused a reduction in cell viability. The results suggest that the DqV, probably, has both neurotoxic and neuroprotective components, and that the astrocytes should be the cells involved in the venomâs neurotoxicity. / Nos Ãltimos anos, esforÃos considerÃveis tÃm sido feitos no sentido de identificar peptÃdeos naturais neuroativos e neuroprotetores derivados do veneno de diferentes espÃcies animais. Nesse trabalho foi estudada a atividade do veneno de Dinoponera quadriceps nativo (vDq) e desnaturado (vdDq) em modelos animais de convulsÃo quimicamente induzidos in vivo e sobre a viabilidade de astrÃcitos corticais in vitro. Camundongos Swiss machos (28-33g) foram prÃ-tratados com o vDq (0,1 ou 0,5 mg/kg, e.v, n= 6-8), vdDq (0,5 ou 2,0 mg/kg, i.p., n= 6-8) ou com vdDq (0,1 ou 0,5 mg/kg, e.v., n= 6-8). Meia hora apÃs o prÃ-tratamento intraperitoneal ou dez minutos apÃs o prÃ-tratamento endovenoso com o veneno foi induzida a convulsÃo em todos os animais atravÃs da administraÃÃo de pentilenotetrazol (80 mg/kg), pilocarpina (400 mg/kg,) ou estricnina (3,0 mg/kg). Na anÃlise comportamental, foram registrados os tempos para ocorrÃncia da primeira convulsÃo e morte e o percentual de sobrevida. Para determinaÃÃo dos parÃmetros de stress oxidativo foram utilizadas trÃs Ãreas cerebrais (cÃrtex prÃ-frontal, hipocampo e corpo estriado) de animais utilizados na anÃlise comportamental, a fim de se determinar o grau de peroxidaÃÃo lipÃdica, pela mensuraÃÃo dos nÃveis de malondialdeÃdo (MDA), e o conteÃdo de nitrito. No ensaio in vitro, foi determinada a viabilidade celular de astrÃcitos corticais apÃs o tratamento com diferentes concentraÃÃes de vDq, PTZ e vDq + PTZ. Os dados foram analisados por ANOVA e Student-Newman-Keuls como pÃs-teste, para os ensaios in vivo, e ANOVA seguido pelo pÃs-teste de Bonferroni, para as experimentaÃÃes in vitro. Foi observado que o vDq apresentou efeito apenas no modelo de PTZ, tanto na anÃlise comportamental quanto na determinaÃÃo dos parÃmetros oxidativos. O prÃ-tratamento com vDq reduziu significativamente o tempo para ocorrÃncia da primeira convulsÃo (0,1 mg/kg: 77,83 Â 5,27 comparado com 101,0 Â 3,31 segundos no grupo controle; 0,5 mg/kg: 74,43 Â 3,94 comparado com 101,0 Â 3,31 segundos no grupo controle), enquanto que o vdDq causou aumento do percentual de sobrevida quando prÃ-tratado por via i.p. (0,5 mg/kg: 25% de sobrevida comparado com 0% no grupo controle; 2,0 mg/kg: 62,5% de sobrevida comparado com 0% no grupo controle) e e.v (0,5 mg/kg: 28,57% de sobrevida comparado com 0% no grupo controle). Em relaÃÃo aos parÃmetros de estresse oxidativo, tanto o prÃ-tratamento com o vDq quanto com o vdDq causaram aumentos dos nÃveis de MDA nas trÃs Ãreas cerebrais analisadas. O conteÃdo de nitrito tambÃm sofreu elevaÃÃo apÃs prÃ-tratamento com vDq, nas trÃs Ãreas cerebrais, e apÃs o prÃ-tratamento com vdDq, via e.v., apenas no hipocampo. Quanto aos ensaios de viabilidade celular, foi observado que o vDq, isoladamente, nÃo foi capaz de alterar esse parÃmetro. O PTZ causou reduÃÃo da viabilidade de astrÃcitos de modo dose-dependente e com uma IC 50 (Ãndice de citotoxicidade para 50% da populaÃÃo celular em estudo) correspondente a 33,12 mM. O tratamento combinado entre vDq (100 Âg/mL) e PTZ (IC 50) tambÃm causou reduÃÃo da viabilidade das cÃlulas. Os resultados sugerem que o vDq possivelmente apresenta ambos componentes neurotÃxicos e neuroprotetores, e os astrÃcitos devem ser uma das cÃlulas envolvidas na neurotoxicidade do veneno.

Studies have shown that some canines have the ability to predict seizures in people with epilepsy, and that canines can be trained to recognize changes in humans before an epileptic seizure and make these predictions. It is not known with any certainty to what the canines are alerting. However, canines’ exceptional sense of smell and their ability to discriminate human scent is well established. Therefore, it is possible that the canines could be responding to an olfactory cue, such as the release of some volatile organic compounds (VOCs) prior to the onset of a seizure. Individuals release a wide array of VOCs, both odorous and non-odorous, from their bodies. The odorous VOCs collectively make up human scent and a number of these VOCs have been identified as biomarkers of different diseases. Evidence suggests that canines can perceive these biomarkers, leading to early detection of underlying physical ailments before individuals are aware of their own symptoms. The main purpose of this study was to use headspace solid phase microextraction (HS-SPME) with gas chromatography-mass spectrometry (GC-MS) to analyze hand odor, saliva and breath samples from epileptic with and without seizure activity to determine if the human scent profiles resulting from a seizure event differs from the scent profiles in the absence of seizure activity. the HS-SPME-GC-MS method was also used to analyze and compare hand odor, saliva and breath samples of healthy individuals and epilepsy patients to determine if the profiles can be differentiated. Comparison of the VOCs in each specimen from healthy individuals and epileptic patients revealed compounds that could be used as potential biomarkers to differentiate between healthy and epileptic individuals. Comparison of the VOCs in each specimen from epileptic patients with and without seizure activity revealed compounds that could be used as potential biomarkers for epileptic seizures. Finally, canine trials were used to verify that these compounds are indeed biomarkers.

Abstract not available

Lacosamide (LCM) is a new antiepileptic drug with a previously unknown mode of action. Using electrophysiological recording techniques in a range of in vitro preparations I have determined a mechanism of action of the new drug. In a 4-aminopyridine model of tonic-clonic seizures in rat visual cortex in vitro, LCM stereoselectively reduced maximal frequency and duration of tonic activity with EC[50�s] of 71 and 41 [mu]M respectively. LCM (100 [mu]M) significantly reduced excitability in whole cell patch clamped neurons producing non-selective reduction in the incidence of excitatory/inhibitory postsynaptic currents (EPSCs; LCM: 46.1 � 15.5 %, P <0.01, n = 4, IPSCs; LCM: 24.9 � 9.6 %, P <0.01, n = 4) and block of spontaneous action potentials (EC₅₀ 61 [mu]M). The inhibitory effects of LCM did not result from changes in passive membrane properties (including resting membrane potential or input resistance) as assessed by application of voltage ramps between -70 to +20 mV. LCM did not mimic the effects of diazepam as an allosteric modulator of GABA[A] receptor currents, nor did it inhibit evoked excitatory currents mediated by AMPA or NMDA receptors. Unlike phenytoin (DPH), carbamazepine (CBZ) or lamotrigine (LTG) that blocked sustained action potential firing evoked by brief depolarising steps (750 ms) or ramps (-70 to 20 mV, 90 mV.sec⁻�), LCM could weakly reduce the frequency of action potentials evoked by brief depolarisation suggesting a potential interaction with VGSCs. In accordance with this, the effect of LCM upon neurotransmission was negated in the presence of tetrodotoxin (200 nM, TTX). The frequency of miniature EPSCs was not altered by the drug (100 [mu]M). These results discounted some crucial potential anticonvulsant targets for LCM but implied a potential interaction with electrogenic VGSCs. When SRF duration was prolonged (10 s) LCM produced significant (P <0.01, n = 4-10, EC₅₀: 48 [mu]M) inhibition, but not within the first second of the burst EC₅₀: 640 [mu]M). Evoked TTX sensitive sodium currents in N1E-115 neuroblastoma cells were significantly reduced by LCM, CBZ, LTG and DPH when V[h]: -60 mV. Hyperpolarizing pulses (500 ms) to -100 mV could reverse block by CBZ, LTG and DPH but not LCM. The V₅₀ for steady state fast inactivation was more hyperpolarized by CBZ (-79.45 � 2.64 mV, n = 5, P < 0.001), LTG (-72.30 � 1.70 mV, n = 6, P <0.05) and DPH (-77.17 � 2.32 mV, n = 6, P <0.05) but not by LCM (-65.02 � 1.75 mV, n = 6, CONTROL: -65.84 � 0.86 mV). In contrast to CBZ, LCM did not slow recovery from fast inactivation or produce frequency dependent facilitation of block of a 3 s, 10 Hz pulse train. LCM (100 [mu]M) did produce a (V₅₀: CONTROL ~64 mV, LCM -57.47 � 4.53 mV, P <0.001, n = 4-8) hyperpolarizing shift in the voltage dependence of slow sodium channel inactivation and promoted channel entry into the slow inactivated state (P <0.001, n = 6) but did not alter the rate of recovery. I therefore conclude that LCM produces inhibition of epileptiform cellular activity, at least in part, via enhancement of voltage gated sodium channel slow inactivation and represents a molecule possessing a unique anticonvulsant mechanism of action.

This thesis applies several standard nonlinear quantifiers to EEG analysis to examine both human primary generalised epilepsy (PGE) and rat models of human epilepsy. We analysed rat EEG, and then used the analysed data, in parallel with an impedance recording, to better understand the events during experiments. Next, the nonlinear analysis of EEG was used to attempt to model the behaviour of the impedance data. This modeling did not yield a useful predictive tool, so we recommend the continued recording of impedance data as a means of augmenting EEG recordings. The analyses were also applied to human data, and showed differences between the PGE and control groups in apparently normal EEG. We then attempted to use these differences to detect the presence of PGE in an unclassified subject – a diagnostic tool. This was done using a feed-forward neural network. We found that the inter-group differences were exploitable and facilitated the diagnosis of PGE in previously unknown subjects. The extent to which this is useful as a diagnostic tool should be assessed by further trials. Finally, the analyses were used to examine data from a paralysed human subject, in an attempt to identify the mental task being performed by that subject. This was not successful, suggesting that the same analyses that were useful in discriminating between PGE and control were not useful in detecting the mental state of the subject. It was also apparent that the presence of EMG (in an unparalysed state) assisted task-classification.

Carbonic anhydases (CAs) are ubiquitous enzymes present in human under 15 different isozymes. Each active one catalyzes the hydration reaction of carbon dioxide into bicarbonate anion and proton. Some isozymes contribute to basic physiological processes like among other respiration and acid-base homeostasis while other isozymes are involved in pathologies such as epilepsy (CA VII) and cancer (CA IX). Convulsions observed during epileptic seizures are partly attributed to carbonic anhydrase VII which play a role in neuronal excitation phenomenon. Carbonic anhydrase IX (CA IX) is overexpressed in most cancer tissues and is absent from their normal counterparts. It can acidify the extratumoral medium leading to metastatic behavior. To improve our knowledge on the role of these isozymes, the design of selective CA VII and IX inhibitors is of a great interest. Otherwise, such compounds can potentially be developed as antiepileptic or anticancer agents. A molecular modeling study which combines a direct (homology modelling, docking) and an indirect (pharmacophore, virtual screening) approach of drug design was conducted to create novel and selective inhibitors. In parallel, original indanesulfonamides were designed, synthesized and their inhibitory potencies against the CAs were determined. Docking studies of some derivatives allowed to rationalize the enzymatic data. Then, we evaluated the effect of some indanesulfonamides on a model of cancer cells. The study of the anticonvulsant activity was performed on an in vivo model. Finally, during this work other series of potentially CA inhibitors were also evaluated for their CA inhibitory activities and for one of them for its anticonvulsant effect.

During fetal development, the hippocampal structures are folded forming the hippocampal sulcus which penetrates into the temporal lobe and then the entity rotates.  During this process, the hippocampal sulcus will be closed and the inverted hippocampus takes a rounded form. After complete inversion, the hippocampus has an oval form in a plane perpendicular to its long axis. If this process has not been completed the hippocampus remains the rounded form. That condition is called incomplete hippocampal inversion (IHI). The aims of this study was to evaluate the frequency of IHI in non-epileptic and epileptic children and adults and to explore the development of the hippocampal region by studying premature neonates and fetuses. Magnetic resonance (MR) images of 201 epilepsy patients and 150 non-epileptic subjects were evaluated without knowing clinical data. IHI was found in 19 % in seizure free controls (20 left-sided and 8 bilateral). 30% of the 201 epilepsy patients had IHI (40 left-sided, 4 right-sided, 16 bilateral). The difference was statistically significant (p&lt;0.02). 25% of the temporal lobe epilepsy patients had IHI. The frequency was not significantly higher than in controls. There is no causality between temporal lobe epilepsy and IHI. 44% of the Rolandic epilepsy patients and 57% of the cryptogenic generalized epilepsy patients had IHI. IHI can be a sign of possible disturbed cerebral development in other parts of the brain. Cranial ultrasound examinations of 160 premature children were analyzed. The age at examination was 23-24 GW in 24 children, 25-28 GW in 72 children, and 29-36 GW in 64 children. IHI was found in 50%, 25% and 14%, respectively. The frequency difference between the children &lt; 25 GW and &gt; 25 GW was statistically significant (p&lt; 0.001). From 25 GW onwards, the frequency and laterality of IHI is similar to that in the adult population. MRIs of 63 fetuses without intracranial pathology were reviewed independently by two radiologists. Three MRIs were performed post mortem at gestation week (GW) 17-18 and 60 in utero at GW 19-35. The hippocampal sulcus was open, bi- or unilaterally, in 35 fetuses at GW 17-32. The oldest of them was at GW 32.  The sulcus was closed at GW 21 at the earliest, unilaterally, and always from GW 33 onwards bilaterally. In 26/63 fetuses (41%), the hippocampal development was asymmetric and in 23 fetuses, the right side had developed faster.