A Staring Spell

Image Courtesy of Ava Hoffman.

It comes without warning: all motion halts and activity stills. Moments later, the world returns between blinks, all memory of the lost time gone. These staring spells are the hallmarks of absence seizures, which are brief episodes of unresponsiveness and loss of consciousness. Epilepsy, a neurological disorder that affects nearly seventy million people worldwide, is characterized by the occurrence of recurring seizures due to abnormal electrical activity in the brain. 

Absence epilepsy primarily presents in children, comprising ten percent of childhood seizures. They can occur up to several hundred times a day and prevent normal engagement in school and social interactions. Knowledge about absence seizures has evolved considerably with the help of neuroimaging techniques and computational methods. However, there are still many questions on the mechanisms by which absence seizures occur that Hal Blumenfeld, Yale School of Medicine professor of neurology and director of the Yale Clinical Neuroscience Imaging Center, and his lab seek to answer. “For years, we’ve worked on trying to understand the basic cellular mechanisms of what goes wrong during loss of consciousness in absence seizures because that’s been a puzzle,” Blumenfeld said. 

The Puzzle of Absence Seizures

 It is now understood that absence seizures are caused by abnormal rhythmic activity in the corticothalamic network, an interconnected circuit in the brain that regulates attention and cognitive processing. But some of the unexplained phenomena that Blumenfeld and his lab encountered were discrepancies in brain activity during absence seizures between children and animal models. 

Common techniques used to map seizures include functional magnetic resonance imaging (fMRI), a technique that maps the flow of oxygenated blood in the brain utilizing its different magnetic properties from deoxygenated blood, as well as electroencephalograms (EEGs), which measure electrical activity generated by neurons in the brain. During absence seizures in children, the cerebral cortex typically shows a decrease in blood-oxygen-level-dependent (BOLD) signal on fMRIs and a repetitive spike-wave discharge (SWD) pattern on EEGs. SWDs are the defining electrographic characteristic of absence seizures. However, previous experiments in animal models showed confusing results: studies instead observed an increase in cerebral fMRI signal which did not resemble the decreases seen in children. 

There were also major discrepancies in behavioral response between children and animal models during absence seizures. “Absence seizures interrupt an individual’s ability to respond normally to the environment, whether it’s something that’s simple and repetitive, like tapping on a button, or more challenging like responding to a specific stimulus,” Blumenfeld said. However, previous attempts to characterize such changes in animals failed because behavioral activities suppressed or interrupted seizures. “The problem is that nobody had ever tested absence seizures in an animal model where animals were in a state where the seizures wouldn’t be interrupted… the tasks that were used were very exciting for the animals,” Blumenfeld said. The pursuit to understand these discrepancies drove a five-year-long project led by Cian McCafferty, then a postdoctoral student in Blumenfeld’s lab and current lecturer and principal investigator in the Department of Anatomy and Neuroscience at University College Cork. 

Validating an Animal Model

In their study published in Nature Communications, McCafferty and colleagues used a common model for absence epilepsy: genetic absence epilepsy rats of Strasbourg (GAERS). “As a rat model, the behavior can be more easily interrogated than a mouse model. They are less prone to impulsive or hyper-aggressive behavior [than mice],” McCafferty said. 

Previously, fMRI scans of absence seizures had only been done on anesthetized animals, as the cold and loud fMRI machine generates a distressing environment. In this study, researchers were able to habituate the rats to the machine and record fMRI without anesthesia. “[McCafferty] would wrap them up like a little child, almost like swaddling a baby, to make them very comfortable,” Blumenfeld said. “And he would train them until they’re very calm and habituate them. So they’re at the point that they would be able to not move without any drugs or anesthesia and sit still for long enough to do an MRI scan… just like children do.” Unlike measurements from anesthetized rodents, a decrease in blood flow accompanied absence seizures in these rats, just like in children. The researchers then determined that previously reported increases in blood flow were due to anesthesia rather than a characteristic of the seizure itself. 

Besides a new method for fMRI scans, this study also developed two behavioral tests that did not disrupt absence seizures. In one task, rats were trained to respond about once a minute to eighty decibel (dB) sound signals, around the volume of a noisy restaurant, by licking a sensor to receive a sugary water reward. However, this intensity kept the rats aroused and inhibited seizures. So, researchers lowered the intensity to forty-five dB, or average room noise, every few minutes, or upon detection of an SWD (signaling seizure onset)—whichever came sooner. This change allowed for seizures to occur and interrupt behavior. On average, rats responded to 88.2 percent of all sound signals before seizures but only 0.4 percent during seizures. 

In the other task, researchers trained rats to spontaneously lick at a spout by giving sugar rewards at random intervals. The average rate of licking decreased during SWDs, indicating that activity was interrupted. Licking recovered within a few seconds after SWDs ended. However, five percent of all seizures were “spared,” meaning that rats demonstrated at least one lick during SWDs. No rats responded to sound signals during a seizure as an auditory response was a more demanding task. These behavioral changes are consistent with observations in humans. “Just like children, the rats had a decrease of fMRI activity in their cortex and these changes in behavior. We finally had a model that we could trust,” Blumenfeld said. 

Measuring Neuronal Activity

Once the model was validated, the group turned to investigating the underlying causes of absence seizures on the cellular level. In EEGs, they found for the first time an overall decrease in neuronal firing both at the surface and deep in the brain, which Blumenfeld hypothesized was most likely responsible for the loss of consciousness. They also discovered a decrease in neuronal activity a few seconds before SWDs started and a transiently higher activity at seizure initiation before the overall decrease again. In addition, they found that neuronal patterns were more rhythmic during seizures. During normal function, neurons encode information in varied firing signals. The increase in rhythmicity and loss of irregular firing, then, indicates that important signaling is lost. 

After discovering these changes in neuronal firing, the researchers went on to characterize individual neurons and discovered four different patterns of neuronal firing that contribute to the overall physiology of an absence seizure SWD. The majority of neurons decreased in firing, contributing to the overall lower activity. However, there are groups of neurons that show increased firing, some that have no change in firing, and some that display a transient increase in firing just before the seizure begins. “We think that the different neurons might be playing different roles in the seizure initiation and maintenance—in particular, the group of neurons with abnormal transient increase in firing might be critical for triggering the onset of the seizure. And identifying those can be very exciting to try to prevent the seizures from getting started,” Blumenfeld said. 

This study also reported systematic neuronal and behavioral changes forty to eighty seconds prior to seizure initiation, consistent with the directions of changes at seizure onset. Still, McCafferty remains cautious with these findings. “These changes are quite preliminary. One thing that would be interesting to see is whether those trajectories of behavior and EEG happen at other times when it doesn’t lead to a seizure,” McCafferty said. 

Toward Targeted Therapeutics

McCafferty is interested in one day using these findings to predict and inhibit seizures. While there is still a long way to go from identifying a trend to establishing reliable predictive power, he believes that pre-onset changes may inform an algorithm to predict seizures in children with absence epilepsy. “Other people have suggested things that happen in a shorter period before the seizure starts that could lead to the seizure,” McCafferty said. There is robust evidence that sensory stimuli can prevent seizures and, in some cases, even stop them at an early stage. It may be possible to devise portable devices that detect neuronal changes and prevent an anticipated seizure or restore partial functionality after seizure onset. 

At the same time, Blumenfeld’s lab is working to determine the different neuronal cell types, including their genetic identities and how different groups connect to one another. These efforts will help develop targeted therapeutics for absence seizures. “Prior to relatively recently, it looked like things were going wrong in the whole brain all at the same time [during an absence seizure], so it’s really hard to figure out how to fix that. But if it turns out that there are only some neurons that you need to target to fix, that could facilitate the development of more targeted therapies,” McCafferty said. Maybe one day, equipped with targeted therapies and devices to stop seizures, scientists can eradicate these staring spells.