BY RICHARD ROBINSON
Mossy cells of the hippocampus are partially lost in temporal lobe epilepsy (TLE), and that loss is thought to contribute to seizure activity. But exactly how they contribute — whether the remaining cells are overall excitatory or inhibitory within the complex hippocampal network —has been unclear, and so designing treatment strategies to compensate for mossy cell loss has been hampered.
Now, a new study in the February 16 issue of Science reveals that increasing mossy cell firing can reduce electrographic seizure duration and reduce the likelihood they will generalize to become convulsive seizures.
The results indicate that the remaining mossy cells continue to inhibit aberrant electrical activity in the hippocampus. Further, the study shows that a decrease in mossy cell activity can produce cognitive deficits in otherwise normal mice, suggesting that loss of mossy cells may be the source of similar deficits in people with TLE.
"It has been a longstanding question in temporal lobe epilepsy whether it is the loss of mossy cells themselves, or an overall change in the network that allows seizures to develop," said the first author of the new study, Anh Bui, PhD, currently a medical student at University of California, Irvine, and formerly in the laboratory of Ivan Soltesz, PhD, James R. Doty professor of neurosurgery and neurosciences at Stanford University School of Medicine. "Our study suggests that the loss of mossy cells is sufficient to allow seizures to generalize."
Dr. Bui and colleagues weren't the first to attempt to solve the puzzle of the mossy cell in TLE, but previous work has been hindered by the inability to target mossy cells specifically. To overcome that obstacle, the team turned to optogenetics, the technique of using light to selectively turn on or turn off neuronal firing.
Mice were produced to express one of two light-sensitive rhodopsin proteins only in their mossy cells. Both proteins were ion channels, one of which decreased membrane polarization and thus promoted firing, while the other increased polarization and inhibited firing. The proteins were activated when light was delivered to the hippocampus via an implanted optical fiber, allowing the mossy cells to be switched on or off. The mice received unilateral injections to the hippocampus of the excitotoxin kainic acid, a standard way to induce an experimental form of TLE, and seizure-recording electrodes were implanted as well.
To test how activity of the remaining mossy cells affected electrographic, non-convulsive seizure activity, Dr. Bui recorded seizure onset, then randomly delivered either no light (as a control), or light sufficient to trigger the ion channels. In mice bearing the inhibitory ion channel, electrographic seizure dynamics following light administration were no different from the control. But in those bearing the excitatory ion channel, light reduced the duration of the seizures.
"Stimulation of mossy cells controls seizures rather than inhibition," Dr. Bui said, "presumably because they are overall inhibitory in the network," with synapses reaching multiple cell types, including inhibitory interneurons. "When you excite the mossy cells, that leads to more inhibition of the network, so that is able to stop the seizures."
As in humans with TLE, a small proportion of electrographic seizures evolve into convulsive seizures. To test the role of mossy cells in that evolution, Dr. Bui recorded seizure electrographic activity continuously for up to six months, "enough time to collect sufficient data for statistical analysis," she said. They used light to inhibit mossy cells. She found that inhibition of mossy cells during an electrographic seizure increased the likelihood of the seizure generalizing to convulsion but had no effect on the duration of the resulting convulsive seizure. Conversely, exciting mossy cells during the electrographic stage reduced the number of seizures that generalized, but again had no effect on duration of those that did.
"What we think is going on is that decreased mossy cell activity is allowing seizures to propagate and to generalize," but without affecting the dynamics of the generalized seizure itself.
Finally, Dr. Bui and colleagues asked whether the loss of mossy cell activity might also contribute to the cognitive deficits that are commonly seen in TLE. To do this, they examined the ability to learn in mice without kainate-induced TLE but with an inhibitory rhodopsin in place. They found that when the mice experienced a light-induced inhibition of their mossy cells, they were less able to encode new spatial memories. The effect seemed to be highly selective, as inhibition did not interfere with recall of previously encoded spatial memories, or with learning or recall of object recognition. The next step for the team is to test whether exciting the remaining mossy cells in epileptic mice might restore cognitive function.
The study was funded by the National Institute of Neurological Disorders and Stroke.
This study nicely tests two competing hypotheses — aberrant excitation versus incomplete inhibition — of residual mossy cell function in TLE, said Kazutoshi Nakazawa, MD, PhD, associate professor of psychiatry and neurobiology at the University of Alabama School of Medicine in Birmingham. Previous studies, including Dr. Nakazawa's own work, have suggested an overall inhibitory role for mossy cells, and the new study "confirms that convincingly in the TLE model," he said.
The study also highlights the importance of mossy cells during acquisition of spatial information. "It remains unknown whether, in an epilepsy model, cognitive function can be rescued by manipulating mossy cells," he said. "That is the next important question to be addressed."
"The idea that mossy cells contribute to temporal lobe epilepsy has been hypothesized for over 25 years, but it has been very hard to explore without the selectivity used in this study," said Helen Scharfman, PhD, professor in the departments of child and adolescent psychiatry, neuroscience and physiology, and psychiatry at NYU Langone Medical Center. Mossy cells are known to have synapses on both inhibitory and excitatory neurons, "so it has been hard to predict the effect on the network of their loss. This study, because of its ability to manipulate mossy cells so selectively, really moves the field forward," suggesting that mossy cells normally inhibit the brain.
The study also provides a proof of principle that "direct manipulation of mossy cells may be therapeutic," although there are many steps before that possibility could be realized, she said, and there may be more practical ways to control seizures for those currently untreated by anti-seizure drugs.
"This study also suggests the mossy cells may contribute to cognitive impairment. It's important to think not only about controlling seizures in temporal lobe epilepsy, but also improving the quality-of-life issues that people with epilepsy experience," including cognitive deficits, she said.
Neither the study author nor the independent experts had disclosures relevant to this study.
LINK UP FOR MORE INFORMATION:
Bui AD, Nguyen TM, Limouse C, et al. Dentate gyrus mossy cells control spontaneous convulsive seizures and spatial memory. Science 2018; 359(6377):787-790.
Scharfman HE. The enigmatic mossy cell of the dentate gyrus. Nat Rev Neurosci 2016; 17(9): 562–575.