Different susceptibility of acutely injured and chronically epileptic brain tissue to induction of cortical spreading depolarization and seizures

Epileptic seizures and cortical spreading depolarization (CSD) are two patterns of electrophysiological activity that can be studied in the healthy and in the injured brain. They can be triggered by the same pathologies, such as stroke, traumatic brain injury and fever, with disruption of the blood brain barrier as a common hallmark. The latter is associated with entry of serum components into the brain interstitial space, such as albumin and immune factors, and a shift of extracellular ion concentrations towards serum values. While CSD can acutely deteriorate patients’ outcome, hyperexcitability of the tissue is potentially indicative for the development of chronic seizures. Chronic epilepsy is usually preceded by an initial silent period, during which reorganization processes take place. In my thesis, I have examined the relationship between CSD, acute and chronic tissue excitability and blood brain barrier disruption, using electrophysiological recordings from acute temporo-hippocampal slices and immunocytochemistry stains of the temporo-hippocampal formation. The main findings of the work are that CSD and epileptic seizures can coexist in both acute brain injury and chronic epilepsy. However chronic epileptic tissue is more resistant to agents, which have proven to be potent inducers of epileptic seizures and/or CSDs in the healthy brain. Moreover, agents, which are harmless or even play a physiological function in the healthy brain, can become ictogenic in the chronically epileptic brain, as is the case with the neurotransmitter acetylcholine. These results provide evidence that hyperexcitability in acute brain injury and chronic epilepsy are driven by different mechanisms and therefore distinct treatment strategies should be considered. 5 3. Introduction and Aims Epileptic seizures are often observed after acute brain injury such as stroke (Camilo and Goldstein, 2004; Jungehulsing et al., 2013), subarachnoid hemorrhage (Dreier et al., 2012), traumatic brain injury (TBI)(Bolkvadze and Pitkänen, 2012), and fever in childhood (Dubé et al., 2007; McClelland et al., 2011) and might indicate risk for progression to chronic epilepsy. Usually, the initial event is followed by a latent phase, during which reorganization processes take place and eventually lead to the development of recurrent seizures. Reorganization might include cell loss, fiber sprouting and changes in expression of ion channels and receptors, associated with increased excitability, such as in voltage-gated ion channels (Bernard et al., 2004; Ellerkmann et al., 2003), and in the glutamatergic, GABAergic (Ferando and Mody, 2012; Maglóczky and Freund, 2005; Tolner et al., 2007) and cholinergic system (Friedman et al., 2007; Zimmerman et al., 2008). Not all patients develop epilepsy after an initial injury. In the case of stroke, up to 8.2% of patients develop chronic epilepsy (Jungehulsing et al., 2013) and in the case of TBI even up to 16 % (Annegers et al., 1996). The probability increases with the severity of the injury and is much higher in patients with late (more than 2 weeks) post-stroke (65-90%) or post-traumatic seizures (up to 86%), compared to patients with early seizures which usually develop within 24 hours (17-35%) (Haltiner AM, Temkin NR, Dikmen SS,Camilo & Goldstein, 2004). Therefore, according to the definition of the International League Against Epilepsy (ILAE) one late seizure per se can be diagnosed as epileptic disease (Jungehulsing et al., 2013) because the condition of an “enduring alteration of the brain that increases the likelihood of future seizures” (Fisher et al., 2005) is provided. Identifying the patients with acute brain injury who are prone to develop late seizures is a major therapeutic concern. So far, the pathophysiology of the injury, the severity of the lesion and EEG, have only limited prediction value. The discovery and validation of reliable biomarkers is therefore necessary. Lesions that are accompanied by prolonged periods of blood-brain barrier (BBB) disruption might facilitate epileptogenesis (Seiffert et al., 2004; Shlosberg et al., 2010; van Vliet et al., 2007). Under these conditions serum components enter the interstitial space and electrolyte levels are altered in a way that provokes hyperexcitability. For example, lowered calcium and increased potassium concentrations have proconvulsant effects. On the other hand, the serum protein albumin probably activates astrocytes associated with downregulation of Kir channels (Ivens et al., 2007; Tomkins et al., 2007), neuronal depolarization and reduced buffering of glutamate and potassium. Another electrophysiological phenomenon, which is commonly observed in acute brain injury and can facilitate lesion progression is cortical spreading depolarization (CSD) (Dohmen et al., 2008; Dreier et al., 2006; Oliveira-Ferreira et al., 2010). CSD is characterized by a breakdown of the electrochemical gradients across the neuronal membranes, which leads to massive glutamate 6 release, prolonged rise of intracellular calcium and massive depolarization of neurons and glial cells (Dreier, 2011; Somjen, 2001). The sustained depolarization of the neurons beyond the threshold for action potential generation prevents the sodium channels from recovering from inactivation. Generation of organized network activity is therefore blocked as is clinically evident in migraine aura (Hadjikhani et al., 2001). However, prolonged CSDs enhance hyperexcitability in the long run, probably due to excitatory plasticity processes (Berger et al., 2008; Ghadiri et al., 2012). I therefore became interested whether a causal relation exists between BBB disruption and CSD generation as a potential biomarker for development of chronic epilepsy. Furthermore, because prolonged seizures can cause rise in potassium and focal hypoxia, two potent agents provoking CSD, I tested whether CSDs co-occur with seizures in acutely compromised as well as in chronic epileptic tissue and can contribute to disease progression or even trigger a seizure. Finally, another crucial question is how a seizure is triggered in the chronically epileptic brain. An epileptic seizure is defined by the ILAE as ”a transient occurrence of signs and/or symptoms due to abnormal excessive or synchronous neuronal activity in the brain”(Fisher et al., 2005) and is therefore likely to be triggered by a transient change in the brain state (Quilichini and Bernard, 2012). Such transitions in brain states are typically the switch from wakefulness to slow wave sleep and REM sleep. On electrophysiological level this would be the switch between theta and gamma oscillations which occur in the hippocampus during exploratory behavior and REM sleep and fast sharp wave ripple oscillations, which can be detected during rest and slow wave sleep and are implemented in memory consolidation (Buzsáki, 1996; Buzsáki et al., 1992). These transitions are believed to be at least partially controlled by acetylcholine, as it can induce gamma and theta oscillations (Fano et al., 2011; Teitelbaum et al., 1975) and can block sharp-wave ripples (Norimoto et al., 2012; Behrens et al., in preparation). Acetylcholine levels are high during exploratory behavior, during REM-sleep (Jasper and Tessier, 1971) and upon arousal (Klinkenberg et al., 2011). Indeed cholinergic activation can induce seizure-like events in the chronically epileptic entorhinal cortex (Zimmerman et al., 2008). I therefore became interested in the mechanisms involved in cholinergic ictogenesis. I compared generation of seizures and CSDs in naïve rat tissue, in acutely injured brain as a result of photothrombotic stroke, as well as in chronic epileptic tissue from pilocarpine-treated rats and resectates from drug-resistant epileptic patients. The slices were examined electrophysiologically as to generation of spontaneous seizure-like events and CSDs, as well as to pharmacologically-induced epileptiform activity by cholinergic agonists, including blocker of the Kv7 blockers, which imitate the effect of acetylcholine on the muscarinic M1 receptor. The comparison was further complemented with histochemical analysis to study tissue morphology and changes in expression of Kv7 channels.