Epileptogenesis: More Than Just the Latent Period

Commentary

Evolution of Network Synchronization During Early Epileptogenesis Parallels Synaptic Circuit Alterations.

Lillis KP, Wang Z, Mail M, Zhao GQ, Berdichevsky Y, Bacskai B, Staley KJ. J Neurosci 2015;35:9920–9934.

In secondary epilepsy, a seizure-prone neural network evolves during the latent period between brain injury and the onset of spontaneous seizures. The nature of the evolution is largely unknown, and even its completeness at the onset of seizures has recently been challenged by measures of gradually decreasing intervals between subsequent seizures. Sequential calcium imaging of neuronal activity, in the pyramidal cell layer of mouse hippocampal in vitro preparations, during early post-traumatic epileptogenesis demonstrated rapid increases in the fraction of neurons that participate in interictal activity. This was followed by more gradual increases in the rate at which individual neurons join each developing seizure, the pairwise correlation of neuronal activities as a function of the distance separating the pair, and network-wide measures of functional connectivity. These data support the continued evolution of synaptic connectivity in epileptic networks beyond the latent period: early seizures occur when recurrent excitatory pathways are largely polysynaptic, while ongoing synaptic remodeling after the onset of epilepsy enhances intranetwork connectivity as well as the onset and spread of seizure activity.

Epileptogenesis is the process through which neuronal networks are altered resulting in the generation of spontaneous, chronic seizures. Epileptogenesis is thought to involve three stages: 1) the initial insult or precipitating event, 2) the latent period, and 3) the chronic epilepsy phase. It has been suggested that during the latent phase, the process of acquired epileptogenesis is completed and culminates in seizure generation (Figure 1) (1). However, the current study suggests that epileptogenesis may be progressive with continuing changes in the neuronal network extending into the chronic epilepsy period (Figure 1).

FIGURE 1.

A model of progressive epileptogenesis. Epileptogenesis is thought to involve three stages: 1) the initial precipitating insult, including traumatic brain injury, stroke, brain infections, or prolonged seizures; 2) the latent period, which is a seizure-free period that can last weeks to months and involves cellular and molecular changes, which alter the excitability of the neuronal network; and 3) the chronic epilepsy phase in which animals exhibit spontaneous, recurrent seizures. The latent period has been defined as period between the initial insult and the onset of the first spontaneous seizure. However, recent evidence suggests that epileptogenesis may be progressive with continuing changes in the neuronal network extending into the period of chronic epilepsy. This diagram was modeled after a diagram of epileptogenesis from Rakhade and Jensen (10).

In support of the idea that epileptogenesis continues into the period of chronic epilepsy is evidence for the progression of epilepsy. Over 100 years ago, British neurologist, Sir William Gowers, coined the phrase “seizures beget seizures,” which is still widely used today to indicate the progressive nature of epilepsy. The observation that seizures beget seizures was largely based on untreated patients. Today, the progression of epilepsy is more difficult to measure given the introduction of more effective anticonvulsant medications and, thus, the inability to monitor the natural progression of epilepsy in the absence of treatment. In addition, we also now appreciate the heterogeneity of the epilepsies in which there are many different types of epilepsies with differences in the underlying etiology and types of seizure presentation. While it is apparent that in some patients epilepsy is not a self-perpetuating disorder, given that seizure frequency and severity remain stable and can even regress or remit (for review, see Blume [2]); it is also clear that some patients exhibit epilepsy progression, evident by increasing seizure frequency, progressive hippocampal damage, and the development of pharmacoresistance.

Several theories have been proposed regarding the mechanisms contributing to epilepsy progression, including progressive primary etiologies, such as expanding lesions and progressive neurodegeneration, development of pharmacologic resistance, and seizure-induced plasticity/kindling (for review, see Sutula [3]). However, we still know very little about the mechanisms underlying epilepsy progression. Understanding the mechanisms contributing to the progression of epilepsy will likely require studies into the emergent properties leading to seizure activity, including circuit-level investigation into network synchronization. The use of voltage-sensitive dyes and calcium imaging enables the activity of large populations of neurons within a network to be studied. These methods have been employed to examine changes in network dynamics associated with simultaneously recorded epileptiform activity (4, 5) (for review, see Goldberg and Coulter [1] and Coulter et al. [6]). The development of in vivo imaging techniques coupled with calcium imaging allows for the activity of neurons to be measured in an intact network during abnormal spontaneous electrographic activity associated with epilepsy (7). However, these studies are typically limited to acute seizure activity rather than the evolution of epileptogenesis.

Studies investigating the evolution of epileptiform activity have largely relied on in vivo cortical and depth electrode recordings or acute and chronic in vitro field potential recordings. The limitations of these types of studies are that the activity measured is limited to the field around the electrodes, which lacks the spatial specificity to examine the dynamics of the entire network. Multi-electrode recording both in vivo and in vitro have been used to overcome these spatial limitations. However, it still remains difficult to monitor the activity of individual neurons throughout the evolution of epileptogenesis. The highlighted study by Lillis et al. employs the organotypic slice model (8) and calcium imaging to examine the evolution of network synchronization during epileptogenesis with single-cell resolution. In this study, the authors demonstrate a progressive recruitment of neurons throughout epileptogenesis. Synchronization within the local hippocampal network is assessed by measuring synchronous calcium signaling in a large number of CA1 pyramidal neurons in relation to distance between the neurons. This analysis demonstrated that early during epileptogenesis, neuronal synchrony is only observed in neurons in close proximity. In contrast, later throughout epileptogenesis, a high level of synchronization is apparent nearly independent of distance. The dynamics of the network also evolves, becoming more efficient at generating seizures with a more rapid seizure onset developing later during epileptogenesis. Using dual whole-cell patch clamp recordings, the authors also demonstrate an increase in the correlation coefficient of spontaneous excitatory postsynaptic currents (sEPSCs) and spontaneous inhibitory postsynaptic currents (sIPSCs) between principal neurons in the hippocampus associated with the progression of epileptogenesis. The authors demonstrate a continuous increase in the correlation coefficient as epileptogenesis progressed, suggesting a progressive synchronization of the network. Examining the functional connectivity of the network, by measuring correlations in the activity of neurons within the network, the authors demonstrate that during epileptogenesis the network transitions from low connectivity to robust connectivity between a large number of principal neurons during epileptiform activity. These data demonstrate that epileptogenesis continues into the chronic phase, which alters the functional connectivity of the network, recruiting a larger population of neurons, making the generation of epileptiform activity more efficient.

These findings force us to reevaluate the epileptogenic process and the definition of the latent period. It is apparent that there are numerous cellular and molecular changes that occur during the latent period (reviewed by Rakhade and Jensen [9]), which is clearly important for the transition from a seizure-free state to the generation of spontaneous, recurrent seizures. However, based on the findings in the highlighted study, the idea that the latent period culminates and ends with seizure onset may not be entirely accurate. Rather, the epileptogenic process may continue into the chronic epilepsy phase with dynamic changes in the neuronal network contributing to the progression of epilepsy. These data suggest that epileptogenesis continues to progress even after seizure onset (Figure 1). Insight into the mechanisms of continuing epileptogenesis would have significant clinical relevance since it would indicate that treatment even after the presentation of the first seizure would be beneficial in halting some aspects of epileptogenesis.

It should be noted that it remains to be determined whether the findings in this study translate to a more physiologically relevant model. It is fair to say that the use of organotypic slices as a model of “early posttraumatic epileptogenesis” is controversial. The organotypic slice has been characterized as a model of epilepsy (8) by the same laboratory that conducted the highlighted study. However, many researchers are reluctant to accept this model given that it is difficult to model a network disorder in the slice. Further, there are numerous changes that occur during the acute slice preparation, which can impact network function, including changes in pH, temperature, oxygen and glucose levels, exposure to cellular contents from damaged or dead cells, excitotoxicity, and cell death. Organotypic slices undergo synaptic remodeling and axonal sprouting, which is seen in experimental models of epilepsy; however, it is unclear whether the remodeling that occurs in the organotypic slice preparation adequately models the changes that occur in epilepsy. Despite these limitations, it appears that the basic foundation of the epileptogenic process is present in the organotypic slice model since these slices develop epileptiform activity, and this model may prove to be extremely useful in screening anti-epileptogenic compounds, which is prohibitively time consuming and expensive using other models.