Abstract
Federico Rossi L, Wykes RC, Kullmann DM, Carandini M. Nat Commun 2017;8:217.
Focal epilepsy involves excessive cortical activity that propagates both locally and distally. Does this propagation follow the same routes as normal cortical activity? We pharmacologically induced focal seizures in primary visual cortex (V1) of awake mice, and compared their propagation to the retinotopic organization of V1 and higher visual areas. We used simultaneous local field potential recordings and widefield imaging of a genetically encoded calcium indicator to measure prolonged seizures (ictal events) and brief interictal events. Both types of event are orders of magnitude larger than normal visual responses, and both start as standing waves: synchronous elevated activity in the V1 focus and in homotopic locations in higher areas, i.e. locations with matching retinotopic preference. Following this common beginning, however, seizures persist and propagate both locally and into homotopic distal regions, and eventually invade all of visual cortex and beyond. We conclude that seizure initiation resembles the initiation of interictal events, and seizure propagation respects the connectivity underlying normal visual processing.
Commentary
For over a century, clinicians and scientists alike have relied on the EEG as a readout of ictal and interictal brain activity. EEG can cover large areas of the brain with high temporal precision, but, as with many electrophysiological techniques, the spatial resolution is limited. Recent advances in indicators of neuronal activity, such as voltage and calcium sensors, now make it now possible to use fluorescence microscopy to image brain activity during ictal and interictal periods, at least in animal models. Although these methods, especially calcium imaging, sacrifice temporal precision, they can show activity in large, contiguous areas of the brain, while simultaneously maintaining sufficient spatial precision to resolve local circuits.
Focal epileptic seizures are initiated in small regions of tissue but typically expand to encompass large swaths of the brain. The extent of the spread is a major determinant of seizure severity in patients; thus, a better understanding of how and why seizure activity invades new territory can provide clues as to what processes are compromised in the epileptic brain and identify new therapeutic targets for interventions aimed at minimizing seizure severity. Previous studies have investigated seizure propagation and identified synaptic (1, 2), cellular (3), and electrical mechanisms (4) that facilitate propagation (5). Rossi et al., in a recent Nature Communications paper, used widefield calcium imaging to investigate a related but distinct issue: how seizures and interictal spikes develop and propagate with respect to the underlying functional organization of the brain region in which they occur.
To accomplish this, the authors thinned the skulls and implanted chronic cranial windows in mice that expressed the genetically encoded calcium indicators GCaMP6f or GCaMP3 in excitatory pyramidal neurons. To image the activity of these neurons, they used a tandem-lens epifluorescence macroscope, a tool capable of providing a field of view containing a significant fraction of the cortical surface (6). They then presented visual stimuli to five awake, head-fixed mice to map the retinotopic organization of their visual cortex. Because the visual cortex has a hierarchical organization—with neurons in V1 projecting axons to regions of higher visual cortices that represent the same portion of visual space (homotypic connectivity)—these retinotopic maps provide information about which groups of cells are synaptically connected. With these maps in hand, they then placed a pipette containing the GABAA receptor antagonist picrotoxin into V1 to induce interictal and ictal events. They imaged these events at speeds sufficient not only to capture the spatial growth of ictal events and interictal spikes, but also the faster oscillatory activity typical of seizures. The resulting spatiotemporal descriptions of these events in a well-characterized brain region yielded insight into the rules of propagation and an opportunity to compare event types.
This experimental design allowed a test of whether epileptic activity propagates along the same circuits as normal visually driven activity or whether it follows different rules, such as propagating only to adjacent areas or creating new pathological circuits. The major finding was that both ictal events and interictal spikes propagated to homotopic brain regions. As the picrotoxin took effect and spike and seizure discharges occurred around the injection pipette, activity increases were visible not only in adjacent areas, but also “jumped” to higher visual centers. These jumps were to regions that represented the same visual space as the injection site, arguing that seizures and spikes respect the organization of visual cortices and travel the same pathways as healthy responses. If we extrapolate, we might assume that similar behavior is exhibited by spontaneous focal epileptic activity in other brain regions. This could yield the exciting ability to predict the regions and circuits affected by seizures based on their point of origin. Conversely, when trying to localize a focus, knowledge of how seizure activity is distributed might constrain our hypothesis about where it originates.
A second major finding was that interictal spikes and the early stages of seizures look remarkably similar. As the title of the paper suggests, both can be described as “standing waves.” This means that in all regions where the event was observed, the temporal evolution of activity was the same. From this common beginning, however, seizures and interictal spikes showed significant divergence. Following an initial peak of activity, interictal spikes ceased while seizures continued and grew. This is an interesting finding because the relationship between interictal spikes and seizures remains mysterious (7, 8). While certainly not settling the issue, this observation is consistent with the view that interictal spikes and seizures are initiated by common mechanisms and arise from the same populations of cells. To speculate even further, it is also consistent with the idea that interictal spikes represent the events that trigger seizures, but only in a minority of the time. Of course, both of these events are the result of local drug application, so it may not be surprising that they initiate in the same population of cells.
Ongoing seizure activity was high amplitude and oscillatory in the 6 to 11 Hz range. As they progressed, seizures invaded new cortical territory, expanding outwards both from around the drug pipette and the homotopic foci in higher visual areas. Eventually, many seizures grew to encompass the entire visual cortex, and some expanded beyond its borders. Looking at phase delay across the cortical surface revealed that individual oscillations propagated outwards from foci, arriving at more distant regions with a short delay. Notably, the phase delay between the primary focus and those in higher visual areas was close to zero, indicating these areas oscillated nearly synchronously. This argues that not only are regions recruited into seizures along the pathways supporting healthy activity but that these synaptically coupled (yet distant) regions maintain their connectivity during the events.
An important fact to keep in mind is that the animals used in this study were not epileptic, so any alterations in connectivity or cellular physiology present in their brains, such as axon sprouting (9), are not reflected in these results. It is an open question whether seizure activity during spontaneous seizures in epileptic animals also follows homotypic patterns. Additionally, the calcium indicators in these experiments were restricted to excitatory neurons; thus, a desirable extension of this work would be to image the epileptiform activity with indicators in inhibitory neurons. Finally, while macroscopic imaging offers an expansive global view of the brain, it is superficial. How deeper brain regions are engaged in both event types studied here is a fascinating question that, admittedly, is not trivial to assess. All in all, the results presented here provide a satisfying and illuminating view of seizure and spike propagation in, relatively speaking, a well-characterized cortical region.