Wondering About Wandering

Abstract

Original Article Citation:

Hippocampal Spreading Depolarization as a Driver of Postictal Ambulation

Mitlasóczki B, Gutiérrez Gómez A, Kamali M, Babushkina N, Baues M, Kück L, Haubrich AN, Tamiolakis T, Breuer A, Granak S, Schwering Sohnrey M, Gerhauser I, Baumgärtner W, Schwarz MK, Ewell L, Opitz T, Pitsch J, Musall S, Surges R, Mormann F, Beck H, Wenzel M. Sci Transl Med. 2025;17(816):eadv3260. doi:10.1126/scitranslmed.adv3260. PMID: 40961224

Original Article Abstract:

Postseizure (postictal) symptoms are regularly encountered in epilepsy and can be life threatening, yet their neurobiological underpinnings remain understudied. Using two-photon or widefield imaging, field potential and unit recordings, optogenetics, and basic behavioral assessment under healthy conditions or viral encephalitis, we studied seizures and postictal symptoms in mice. We show a propensity of the hippocampus for seizure-associated spreading depolarization (sSD). Through optogenetic stimulation, we provide evidence that induced isolated hippocampal spreading depolarization SD is sufficient to elicit postictal ambulation (PIA), whereas induced isolated seizure-like episodes are not. Furthermore, PIA occurred in the absence of SD progression to the neocortex. In addition, we analyzed Behnke-Fried depth-electrode recordings in four patients with focal epilepsy. Of 13 recorded seizures, we observed five slow shifts at seizure termination in the regionwise analysis that could reflect putative sSD. In support of our experiments in mice, we also found an increased vulnerability of the human temporomesial system (hippocampus and amygdala) for this phenomenon and longer recovery times of affected as compared with nonaffected brain regions. This work suggests sSD as a previously underrecognized pathoclinical entity underlying distinct postictal symptoms in epilepsy.

Commentary

In temporal lobe epilepsy, seizures can be immediately followed by abnormal behavior during the postictal period. For example, people can begin to walk aimlessly. This “post-ictal wandering” is a major clinical concern because anyone who begins to walk without a clear sense of direction is at a high risk of injury.

Mechanisms responsible for postictal wandering are largely unknown. In this article, ¹ Mitlasóczki et al hypothesized that a phenomenon called spreading depolarization (SD) causes postictal ambulation. This hypothesis is based on a large literature about SD, which occurs not only in epilepsy but also migraine, stroke, and traumatic brain injury. Originally described in 1944, ² SD begins with a massive depolarization of a population of neurons accompanied by intense neuronal firing, leading to a major disruption of ion gradients and blockade of action potentials. A large negative shift in extracellular potential occurs during the period of neuronal silencing, followed by subsequent restoration of neuronal function as ion pumps restore ion gradients. An important part of SD is that it “spreads,” meaning the depolarization and subsequent depression of activity travels slowly across gray matter. As more research has been conducted, it has become recognized that SDs are complex, with different subtypes, pro- and anti-seizure effects, and complex interactions with seizures.^3–5

Recent attention to SDs in epilepsy has increased because it has become clear that SDs are underestimated due to the common use of AC amplifiers to record EEG, preventing detection of large shifts of extracellular potential during SD. Thus, it is increasingly recognized that EEG suppression immediately after a seizure, a common phenomenon called postictal depression, may reflect SD. However, seizure activity can occur during SD, ⁴ making more research important.

Mitlasóczki et al provide evidence of several novel findings. ¹ First, they show that seizures with a subsequent SD are associated with an increase in postictal ambulation in mice. This finding was made possible by the clever use of multiple techniques. First, the investigators used a mouse model with focal seizures that is well known, ⁶ but unique because convulsive behavior is rare, allowing convulsions not to interfere with the experiments. The model uses an injection of Theiler's murine encephalomyelitis virus (TMEV) into one dorsal CA1 hippocampus to initiate focal seizures in the subsequent days and weeks. The investigators employed awake, head-fixed mice which could be recorded by electrodes inserted into the dorsal surface of the brain while the mouse walked on a treadmill. Part of the skull was removed so electrodes could be placed into the area of interest, dorsal CA1. The recordings were made of the local field potentials (LFPs) and calcium fluctuations to monitor seizures and SD. Calcium imaging was made possible using transgenic mice with the calcium indicator GCaMP6s expressed in cells with the Thy1 promoter (neurons). In addition, the investigators used a wide field of view to simultaneously examine CA1 and cortex. The recordings showed that when a seizure occurred there was a DC shift in CA1 consistent with SD, as well as a sudden and large rise in calcium associated with SD. Furthermore, there was an increase in ambulation shortly after the seizure and SD. However, SD was not significant in cortex.

These data suggested that hippocampal seizures associated with SD might cause postictal ambulation. To test that hypothesis they induced SD in a more controlled manner using normal mice after injection of an adeno-associated virus to express the excitatory opsin channelrhodopsin (ChR2) in CA1. Optogenetic stimulation was able to induce seizure activity and large calcium transients in CA1 neurons, similar to seizures associated with SD. Then the investigators varied stimulation parameters and compared SD with or with very little seizure activity. They found that ambulation was greatest when SD occurred with little seizure activity, especially if SD was bilateral. These data supported the hypothesis that SD led to ambulation, not the seizure.

Another interesting aspect of the study was the observation that SDs in hippocampus were associated with increased postictal ambulation rather than cortical SDs. That led the authors to ask if temporomesial structures were most vulnerable to SD. Although cortical SD is common in other contexts, ⁷ postictal ambulation often occurs in TLE, ⁸ suggesting importance of the temporal lobe. In four patients with focal drug-resistant epilepsy they analyzed multiple areas of the temporal lobe and indeed found that there was susceptibility of temporal regions to seizures with slow shifts relative to other areas they sampled. Although extratemporal regions were not possible to sample, the data suggested that temporomesial regions like hippocampus have a susceptibility to SD and postictal ambulation.

This study was a technical tour de force, and greatly advanced our understanding, but as with all such studies it provoked more questions. One is whether the results from the TMEV model, where seizures occurred during the acute phase post-infection, would apply to chronic epilepsy. Similarly, the seizures in the TMEV model were focal and those seizures that secondarily generalize are relevant to TLE and may be very different. There also was difficulty initiating SD without any seizure activity in normal animals using ChR2, which the authors note. In addition, seizures may have occurred in areas they did not record from, since activating CA1 will lead to glutamate release from CA1 axons distant to the recording site. Still their approaches were unique, clever and provide convincing evidence that hippocampal seizures associated with SD lead to ambulation.

The next question is how to explain the findings. For example, how would hippocampus initiate ambulation? Clues come from the 1960's when it was suggested that the hippocampus initiates walking, coincident with rhythmic slow activity. ⁹ Later studies showed that movements of the monkey were preceded by hippocampal LFPs and multi-unit activity, ¹⁰ again suggesting hippocampal activity initiated movement. However, the proof that the hippocampus initiates walking remains unclear.

Another question is how SD initiates walking if it is associated with profound depression of neural activity. Perhaps the transition from the seizure to SD is critical, or the recovery from SD, because at these times there would be the potential for hippocampal projections to activate other brain areas involved in walking. Consistent with this idea, seizure activity prior to SD in the Scna1-deficient mouse model of Dravet syndrome was associated with motor hyperactivity. ¹¹ Also, during recovery from SD, there is potentiation of hippocampal circuits. ¹² Clearly more experiments will be necessary to clarify underlying mechanisms, and they are very important to understand so that all types of abnormal behavior after seizures can be better controlled. Postictal ambulation is only one type of abnormality, with violent behavior another potentially dangerous example. Until more is known, we will continue wondering about postictal wandering.