Basic NeuroscienceIn vivo models of cortical acquired epilepsy
Introduction
The term epilepsy is used to define over 40 different types of neurological pathologies resulting from different etiologies. They usually are all characterized by the occurrence of unprovoked recurrent seizures, which consists in a period of abnormal (paroxysmal) brain electrical activity. Epilepsy is either genetically determined or acquired (secondary). The causes of acquired epilepsies are multiple (stroke, cortical trauma, brain tumor, infections, etc.) and the recurring seizures are typically primarily focal (Timofeev et al., 2014). The common feature of acquired epilepsies is neuronal death leading to deafferentation. Brain activities could be roughly described using simple words such as activity, silence, excitation, and inhibition. During normal brain activities there is a balance between activity and silence and also between excitation and inhibition. If for any reason the balance is shifted, some homeostatic plasticity occurs to reestablish the balance. Deafferentation increases overall silence in the involved neuronal network. This triggers up-regulation of neuronal excitability. If deafferentation is large it would lead to uncontrolled up-regulation of neuronal excitability and epilepsy. A period between initial insult and development of epilepsy is called epileptogenesis and its exact mechanisms remain largely unknown, therefore, there is no effective treatment of epileptogenesis. Over the past several years, our studies aim to understand the network mechanisms of epileptogenesis in order to develop effective treatment of epileptogenesis and therefore, to prevent the development of epilepsy.
Neuronal synchronization is achieved via different mechanisms: chemical synaptic transmission, electrical synaptic transmission, ephaptic and non-specific interactions such as alterations in the extracellular ionic balance (reviewed in Timofeev et al., 2012). Chemical transmission occurs when an action potential is fired by the presynaptic neuron, invading the nerve terminal, allowing calcium to enter in the synaptic terminal, which will trigger the release of synaptic vesicle containing neurotransmitter, and finally causing depolarization or hyperpolarization of the postsynaptic neuron (Eccles, 1964, Katz and Miledi, 1968).
Electrical synapses use a spike-independent mechanism of communication between neurons connected via gap junctions allowing a direct flux of ions between them (Carlen et al., 2000, Dermietzel and Spray, 1993, Perez Velazquez and Carlen, 2000). When cells are connected by gap junctions, any change in the membrane potential in one cell will trigger a current flow in the other one leading to corresponding changes. In the neocortex, GABA-releasing interneurons (Galarreta and Hestrin, 1999, Gibson et al., 1999) and glial cells (Mugnaini, 1986) are interconnected by gap junctions. However as gap junctions have relatively high resistances, they act as low pass filters (Galarreta and Hestrin, 2001), therefore they are best situated to transmit to coupled neurons low frequency oscillations, but not fast processes, like action potentials.
Ephaptic interactions refer to influences produced by local electric fields. The extracellular currents produced by the activity of neurons constituting the local field potential might directly influence the electrical properties of surrounding neurons (Frohlich and McCormick, 2010, Jefferys, 1995). Although these influences are relatively weak, they might affect the neuronal excitability by exerting a global influence and may provide a significant impact when the network is already quite synchronized such as during slow-wave sleep or during a seizure.
Normal neuronal activity leads to the opening of different ionic channels and produce short-term changes in the extracellular ionic composition. The most affected ion concentrations are for those that have a lower extracellular concentration such as calcium and potassium. For example, during a normal slow oscillation, the dominant type of activity observed during slow-wave sleep and under some anesthetics (see Section 2), the extracellular calcium concentration varies from 1.2 mM during silent states to 1.0 mM during active states (Crochet et al., 2005, Massimini and Amzica, 2001), this can strongly impact the chemical synaptic transmission (Crochet et al., 2005). On the contrary, low calcium concentration promotes the opening of hemichannels (Thimm et al., 2005), which increases the electrical coupling. The changes in the extracellular milieu are even more dramatic during seizures as extracellular potassium concentration can reach up to 7–18 mM (Amzica et al., 2002, Moody et al., 1974, Prince et al., 1973) and the extracellular calcium concentration can decrease to as low as 0.4–0.7 mM (Amzica et al., 2002, Heinemann et al., 1977, Pumain and Heinemann, 1981, Pumain et al., 1983). These changes dramatically reduce the synaptic excitability and almost abolish any synaptic response (Seigneur and Timofeev, 2011). With the same levels of neuronal activity, the extent of changes in extracellular ionic concentrations should depend on extracellular space. The brain extracellular space during wake is smaller than during sleep or anesthesia (Xie et al., 2013) therefore, synchronous neuronal activity during wake should produce major changes in local ionic concentration.
At least two types of neuronal synchronization can be considered, local synchronization, which mediates generation of the local field potential and rely on all types of transmission, and long-range synchronization which is measured using distant recording electrodes and mainly relies on chemical synaptic transmission.
During normal brain activities, the central nervous system maintains the homeostasis between the excitation and the inhibition. Under anesthesia they are balanced (Haider et al., 2006, Shu et al., 2003), while during natural states of vigilance the inhibitory activities dominate (Haider et al., 2013, Rudolph et al., 2007). Epileptiform activities are the result of a shift in the balance of excitation and inhibition (Dichter and Ayala, 1987, Galarreta and Hestrin, 1998, Nelson and Turrigiano, 1998, Tasker and Dudek, 1991, Timofeev and Steriade, 2004). The most known procedure to induce epileptiform discharges consists in blocking inhibition by using GABAergic antagonist (bicuculline, penicillin, etc.), however a disbalance in the other direction: decreased excitation (or increased inhibition) as epileptogenic factor was much less investigated. The model of trauma-induced epilepsy that will be described in the main body of this review is a model in which the excitation is decreased as multiple glutamatergic thalamocortical fibers are severed by the undercut (Topolnik et al., 2003b). The undercut is a procedure in which the white matter is cut underneath a given cortical area using a custom-made knife (see Section 4.1). The model of ketamine–xylazine anesthesia leading to paroxysmal activities in cats could also be considered as a model in which the excitation is decreased, as ketamine is an NMDA receptor antagonist. The third model that will be described in this review is kindling which could result from a disbalance toward excitation through hebbian synaptic plasticity. These three models will be discussed in more details in the following sections.
In the case of genetically determined epilepsies, the altered genes are usually present in every cells of the body therefore increasing the probability of generating generalized seizures. In acquired epilepsies, seizures are primarily focal but may become secondarily generalized (Timofeev et al., 2014). Focal seizures are also sometimes called partial seizures and might be further subdivided into simple or complex partial seizures depending whether the consciousness is impaired or not, respectively. Generalized seizures include absence (also called petit-mal seizures), myoclonic, clonic, tonic, tonic-clonic (also called grand-mal seizures), and atonic seizures.
Neocortical epilepsy is generated cortically, is focal and is mainly nocturnal (Timofeev, 2011). These seizures are usually composed of spike-wave/polyspike-wave (SW/PSW) electrographic discharges at 1.0–2.5 Hz and fast runs (runs of fast EEG spikes) at 7–16 Hz, which develop without discontinuity from the slow oscillation (<1 Hz) (Boucetta et al., 2008, Steriade and Amzica, 1994, Steriade et al., 1998a, Steriade and Contreras, 1995, Timofeev, 2010, Timofeev et al., 2002, Timofeev et al., 2014).
Just prior to the onset of seizure, ripples which are waxing-and-waning very fast oscillations (80–200 Hz) can be observed at the focus (Grenier et al., 2001, Grenier et al., 2003). The use of a gap junction blocker such as halothane anesthetic gas prevents the occurrence of ripples and stops neocortical seizures, suggesting that electrical gap junctions play a critical role in the generation of these rhythms. Following the appearance of ripples, the SW/PSW complexes occur at 1–2.5 Hz and they are often accompanied by fast runs (7–16 Hz) (Timofeev and Steriade, 2004).
The cortically generated seizures that arise without any discontinuity from the slow oscillation are accompanied by the generation of large amplitude field potentials SW/PSW complexes, which suggests the presence of enhanced local synchrony (Timofeev and Steriade, 2004). The precise measurements during SW/PSW complexes suggested that long-range synchrony recorded on a wave-by-wave basis is rather loose (Boucetta et al., 2008, Derchansky et al., 2006, Meeren et al., 2002, Timofeev and Steriade, 2004). During the SW complexes, the correlation of field potentials varies between 0.3 and 0.8 and each cycle would propagate at a speed of 100 mm/s in vivo and 10 mm/s in vitro (McCormick and Contreras, 2001, Steriade, 2003).
During the spike component, cortical neurons are depolarized and fire action potentials whereas during the wave component cortical neurons are hyperpolarized and silent (Timofeev, 2010). Just prior to and during the initial segment of a seizure, neocortical neurons firing increases and then it decreases toward the end of the seizure (Bazhenov et al., 2004). Any intense neuronal firing will lead to an increase in the extracellular potassium concentration and a decrease in the extracellular calcium concentration (Heinemann et al., 1977), which will change neuronal excitability (Boucetta et al., 2013, Seigneur and Timofeev, 2011) and would be an important factor in the generation of paroxysmal activities (Somjen, 2002). For example, during the spike component, fast-spiking presumed inhibitory interneurons fire high-frequency action potential trains and due to the high extracellular potassium concentration during seizure (Heinemann et al., 1977), the reversal potential of chloride-dependent IPSPs becomes depolarized (Cohen et al., 2002, Payne et al., 2003, Timofeev et al., 2002).
The strong depolarization will lead to an activation of high-threshold intrinsic currents such as the high-threshold calcium and the persistent sodium currents which will further contribute to the neuronal depolarization (Timofeev, 2010, Timofeev et al., 2004, Timofeev and Steriade, 2004). However, the increased neuronal activity will lead to an increase in the intracellular calcium and sodium concentrations, which in turn will activate both sodium- and calcium-activated potassium currents that will hyperpolarize neurons and contribute to the wave component of spike-wave complexes (Timofeev, 2010, Timofeev et al., 2004, Timofeev and Steriade, 2004). This hyperpolarization then contribute to the activation of hyperpolarization-activated depolarizing current (Ih) in a subset of neurons that will lead them to depolarize up to the firing threshold and initiate the next paroxysmal cycle (Timofeev, 2010, Timofeev et al., 2004, Timofeev and Steriade, 2004). Due to the dramatic decrease in the extracellular calcium concentration during paroxysmal activities (Heinemann et al., 1977, Pumain et al., 1983), chemical synaptic transmissions are strongly impaired but electrical synaptic transmission via gap-junction increases contributing to the local network synchronization (Jefferys, 1995).
Fast runs are also referred to runs of fast EEG spikes that occur at 7–16 Hz and accompany most of neocortical SW/PSW complexes. They last about 5 s on average but the range of their duration can vary from 1 to 30 s (Boucetta et al., 2008). The onset and offset of fast runs occur almost simultaneously in different recording sites, however during fast runs the different recording sites (even closely located) behave in a quasi-independent manner with low or even asynchronous patterns of coherence between them. Four different types of synchronization could be observed during fast runs, (1) synchronous and in phase, this pattern was observed in approximately 20% of cases; (2) synchronous with phase shift, this pattern was the most commonly observed (about 50% of cases); (3) patchy consisting in repeated phase/phase shift transitions, this pattern was observed in about 10% of cases; (4) non-synchronous (occurring in about 20% of cases), with slightly different frequencies in different recording sites or without the oscillatory activity in one of the recorded sites (Boucetta et al., 2008). Runs of fast EEG spikes likely originate in neocortex. This conclusion is based on two facts: (a) thalamocortical neurons do not show oscillatory activities during cortical fast runs and (b) fast runs can be easily triggered in isolated neocortical slabs (Timofeev et al., 1998). The firing of fast-spiking neurons (presumed-interneurons) is strongly reduced when not completely abolished during fast runs suggesting that inhibitory interneurons do not play any significant role in the generation of fast runs (Boucetta et al., 2008, Timofeev et al., 2002, Timofeev and Steriade, 2004). However, bursting neurons, both intrinsically bursting and fast-rhythmic bursting neurons fire more action potentials than any other type of neurons during fast runs suggesting they might play an important role in the generation of these EEG spikes (Boucetta et al., 2008). The intrinsically bursting neurons appear more suited to play a major role in the generation of these fast runs as upon depolarization, they generate burst of action potentials recurring at 5–15 Hz (Agmon and Connors, 1989, Connors and Gutnick, 1990) similar to the frequency of fast runs (7–16 Hz), while fast-rhythmic bursting neurons display bursts recurring at 20–50 Hz, but mainly at 30–40 Hz (Gray and McCormick, 1996, Steriade et al., 1998b). Intrinsically bursting neurons were also shown to be depolarized and to fire earlier than other neuronal types during fast runs (Boucetta et al., 2008).
Section snippets
The slow oscillation (<1 Hz)
In humans, sleep occupies about a third of their life and is composed of rapid-eye-movements (REM) sleep (also called paradoxical sleep) and of non-REM sleep. The non-REM sleep is then subdivided into three stages and the stage 3 (S3) is often referred to as slow-wave sleep. The hallmark of this sleep stage is the presence of large amplitude slow waves in the field potential (Blake and Gerard, 1937) and in cortical neurons (Steriade et al., 1993a, Steriade et al., 1993b, Steriade et al., 2001,
The ketamine–xylazine anesthesia model
Ketamine–xylazine anesthesia is often used as a model to study slow-wave sleep as it reproduces the main characteristics on the natural slow oscillation (Chauvette et al., 2007, Chauvette et al., 2011, Chauvette et al., 2012, Contreras and Steriade, 1995, Sharma et al., 2010), however some differences exist such as an increased synchrony and correlation between different cortical regions, an increased rhythmicity, and an increased amplitude of the slow oscillation compared to slow-wave sleep (
The trauma-induced epilepsy model
Traumatic brain injury (TBI) is a major risk factor for the development of epilepsy (Annegers et al., 1998, Feeney and Walker, 1979, Temkin et al., 1995). Immediate seizures (less than 24 h after TBI) and early seizures (less than a week after TBI) are risk factors for the development of later epilepsy (Temkin, 2003). A study revealed that adult patients with moderate to severe head injury had a risk of about 86% to have recurrent seizure two years after TBI (Haltiner et al., 1997). However, a
The kindling model
Kindling is the relatively permanent increased propensity to convulsions that follows repeated electrical or chemical stimulation of certain areas of the brain even at intensities that are initially too low to produce any behavioral response or convulsions. Once established, the changes are so persistent that even the administration of a weak stimulus will elicit a seizure (Goddard, 1967, Goddard et al., 1969, Kandratavicius et al., 2014). Neocortical kindling is characterized by a higher
Conclusions
We have described different approaches used to model neocortical seizures and epileptogenesis. ketamine–xylazine anesthesia in cats leads to the occurrence of seizure in a large number of animals. These seizures can be focal or generalized and are usually composed of SW/PSW components, often accompanied with fast runs of EEG/LFP spikes. During these seizures, the extracellular milieu is modified by the increased neuronal activity in such a way that the chemical synaptic transmission is
Acknowledgments
In recent years, our work was supported by CIHR (MOP-325213, MOP-324941, MOP-37862 and MOP-67175), NIH-NINDS (1R01-NS060870 and 1R01-NS059740), NSERC (Grant 298475), and FRQ-S, FRQ-NT. We are thankful to Sergiu Ftomov for his excellent technical assistance.
References (172)
- et al.
Repetitive burst-firing neurons in the deep layers of mouse somatosensory cortex
Neurosci Lett
(1989) - et al.
Extracellular Ca2+ fluctuations in vivo affect afterhyperpolarization potential and modify firing patterns of neocortical neurons
Exp Neurol
(2013) - et al.
The role of gap junctions in seizures
Brain Res Brain Res Rev
(2000) - et al.
Sleep oscillations in the thalamocortical system induce long-term neuronal plasticity
Neuron
(2012) - et al.
Long-term risk of epilepsy after traumatic brain injury in children and young adults: a population-based cohort study
Lancet
(2009) - et al.
Intrinsic firing patterns of diverse neocortical neurons
Trends Neurosci
(1990) - et al.
Bidirectional multisite seizure propagation in the intact isolated hippocampus: the multifocality of the seizure focus
Neurobiol Dis
(2006) - et al.
Gap junctions in the brain: where, what type, how many and why
Trends Neurosci
(1993) - et al.
Endogenous electric fields may guide neocortical network activity
Neuron
(2010) - et al.
A permanent change in brain function resulting from daily electrical stimulation
Exp Neurol
(1969)
Long-term synchronized electrophysiological and behavioral wireless monitoring of freely moving animals
J Neurosci Methods
Risk of seizure recurrence after the first late posttraumatic seizure
Arch Phys Med Rehabil
Functional characterization of central alpha-adrenoceptors by yohimbine diastereomers
Eur J Pharmacol
Cellular mechanisms of the slow (<1 Hz) oscillation in thalamocortical neurons in vitro
Neuron
Dose-dependent proconvulsant and anticonvulsant actions of the alpha 2 adrenergic agonist, xylazine, on kindled seizures in the rat
Pharmacol Biochem Behav
Intractable epilepsy and craniocerebral trauma: analysis of 163 patients with blunt and penetrating head injuries sustained in war
Injury
Changes in long-range connectivity and neuronal reorganization in partial cortical deafferentation model of epileptogenesis
Neuroscience
Neurotransmitter actions in the thalamus and cerebral cortex and their role in neuromodulation of thalamocortical activity
Prog Neurobiol
The kindling model of epilepsy: a review
Prog Neurobiol
The multiple effects of ketamine on electroencephalographic activity and behavior in WAG/Rij rats
Pharmacol Biochem Behav
Extracellular potassium activity during epileptogenesis
Exp Neurol
Inactivity produces increases in neurotransmitter release and synapse size
Neuron
Cortical and thalamic components of neocortical kindling-induced epileptogenesis in behaving cats
Exp Neurol
State-dependent slow outlasting activities following neocortical kindling in cats
Exp Neurol
Seizure activity occurring in two dogs after S-ketamine-induction
Schweiz Arch Tierheilkd
Polyneuronal innervation of spiny stellate neurons in cat visual cortex
J Comp Neurol
Spatial buffering during slow and paroxysmal sleep oscillations in cortical networks of glial cells in vivo
J Neurosci
A population-based study of seizures after traumatic brain injuries
N Engl J Med
Early and late posttraumatic seizures in traumatic brain injury rehabilitation patients: brain injury factors causing late seizures and influence of seizures on long-term outcome
Epilepsia
Cellular and homeostatic network mechanisms of posttraumatic epilepsy, Département d’anatomie et physiologie
Neocortical post-traumatic epileptogenesis is associated with loss of GABAergic neurons
J Neurotrauma
Synaptic strength modulation following cortical trauma: a role in epileptogenesis
J Neurosci
Facilitatory effects of intermittent photic stimulation on visual cortical kindling
Epilepsia
Long-term changes in neocortical activity after chemical kindling with systemic pentylenetetrazole: an in vitro study
J Neurophysiol
Potassium model for slow (2–3 Hz) in vivo neocortical paroxysmal oscillations
J Neurophysiol
Layer-specific excitatory circuits differentially control recurrent network dynamics in the neocortex
Nat Neurosci
Brain potentials during sleep
Am J Physiol
Focal generation of paroxysmal fast runs during electrographic seizures
Epilepsia
Synchronous neuronal activity is a signal for axonal sprouting after cortical lesions in the adult
J Neurosci
Practice parameter: antiepileptic drug prophylaxis in severe traumatic brain injury: report of the Quality Standards Subcommittee of the American Academy of Neurology
Neurology
Properties of slow oscillation during slow-wave sleep and anesthesia in cats
J Neurosci
Local origin and long-range synchrony of active state in neocortex during slow oscillation
Origin of active states in local neocortical networks during slow sleep oscillation
Cereb Cortex
Interneuron-mediated inhibition synchronizes neuronal activity during slow oscillation
J Physiol
Generalized seizure activity in an adult rhesus macaque (Macaca mulatta) during ketamine anesthesia and urodynamic studies
Comp Med
Traumatic brain injury: risks of epilepsy and implications for medicolegal assessment
Epilepsia
On the origin of interictal activity in human temporal lobe epilepsy in vitro
Science
Cellular basis of EEG slow rhythms: a study of dynamic corticothalamic relationships
J Neurosci
Modulation of synaptic transmission in neocortex by network activities
Eur J Neurosci
The slow (<1 Hz) rhythm of non-REM sleep: a dialogue between three cardinal oscillators
Nat Neurosci
Cited by (26)
Repeated applications of high potassium elicit long-term changes in a motor circuit from the crab, Cancer borealis
2022, iScienceCitation Excerpt :Maintaining the appropriate ionic composition of the extracellular milieu is critical for normal physiological function, and the potassium gradient is particularly important for the maintenance of resting membrane potential and normal activity levels. It is therefore unsurprising that altered potassium concentrations occur in a wide array of conditions including heart disease, kidney failure, thermal stress, tissue damage, epilepsy, traumatic brain injury, and stroke (Arnold et al., 2014; Baylor and Nicholls, 1969; Chauvette et al., 2016; Katayama et al., 1990; Morrison et al., 2011; Rodgers et al., 2007; Seeburg and Sheng, 2008). In addition to these pathological states, altered extracellular potassium levels are routinely used by researchers as a physiologically relevant depolarizing stimulus to increase neuronal activity or as a proxy for excitatory inputs (Ballerini et al., 1999; Ruangkittisakul et al., 2011; Rybak et al., 2014; Sharma et al., 2015).
A rat model of somatosensory-evoked reflex seizures induced by peripheral stimulation
2019, Epilepsy ResearchCitation Excerpt :All our animals displayed similar, brief electrographic seizures that were followed by a refractory period, normal response, or another seizure. The neocortex is one of the brain structures involved in several types of acquired epilepsy, including TIS (Chauvette et al., 2016; Timofeev et al., 2014). Cortical seizures are characterized by spike– and polyspike– wave electrographic after-discharges, as well as fast oscillations that are followed by the formation of larger amplitude spike– and polyspike– wave field potential complexes (Timofeev et al., 2014).
Ionic and synaptic mechanisms of seizure generation and epileptogenesis
2019, Neurobiology of DiseaseSusceptibility to hippocampal kindling seizures is increased in aging C57 black mice
2017, IBRO ReportsCitation Excerpt :These include a loss of subgroups of hippocampal GABAergic interneurons (Shetty and Turner, 1998; Cadacio et al., 2003; Vela et al., 2003; Shi et al., 2004; Stanley and Shetty, 2004; Potier et al., 2006; Kuruba et al., 2011; Smith et al., 2000; Stanley et al., 2012; Spiegel et al., 2013) and an increase in hyperactive or hyperexcitable responses of hippocampal CA3 neurons in aging/aged animals (Vreugdenhil and Toescu, 2005; Wilson et al., 2005; Patrylo et al., 2007; Kanak et al., 2011; Lu et al., 2011; El-Hayek et al., 2013; Spiegel et al., 2013; Moradi-Chameh et al., 2014; Simkin et al., 2015; Villanueva-Castillo et al., 2017). In light of these findings and classic kindling as a widely used model of temporal lobe epilepsy (see reviews by Morimoto et al., 2004; Bertram, 2007; Sharma et al., 2007; Coppola and Moshé, 2012; Gilby and O'Brien, 2013; Chauvette et al., 2016; Gorter et al., 2016; Löscher, 2017), we explored whether susceptibility to hippocampal CA3 kindling seizures is different between young and aging C57 black mice. Male C57 black mice (C57BL/6N) were obtained from Charles River Laboratory (Senneville, Quebec, Canada).
Structural alterations in fast-spiking GABAergic interneurons in a model of posttraumatic neocortical epileptogenesis
2017, Neurobiology of DiseaseCitation Excerpt :We analyzed FS interneurons of neocortical layer V in the UC model to determine whether there were abnormalities in their axons and presynaptic terminals that could result in defects in GABAergic transmission. Interictal discharges originate in layer V of the UC (Prince and Tseng, 1993; Hoffman et al., 1994) and alterations in FS interneurons in this lamina could contribute to hyperexcitability and epileptiform bursts in in vitro cortical slices (Prince and Tseng, 1993; Hoffman et al., 1994) and seizures in vivo (Chauvette et al., 2016; Ping and Jin, 2016a). We identified FS cells in whole cell recordings from their electrophysiological phenotype (Xiang et al., 1998) and, retrospectively, appearance following biocytin labeling.
New-Onset Refractory Status Epilepticus Associated With the Use of Synthetic Cannabinoids
2017, PsychosomaticsCitation Excerpt :As described in the case, he had a history of episodes of decreased level of consciousness noted by family, which could have been epileptic episodes associated with pre-existing epilepsy. Second, he could have had initial seizures triggered by substance abuse (or withdrawal), which over time evolved into a refractory state via brain circuitry changes akin to a kindling model.21 Although these 2 models may explain his initial hospital presentation, it is still unknown why he suffered a progressive course of increasingly intractable seizures, despite not having any detectable SC metabolites in his urine.