Abstract
Dossi E, Blauwblomme T, Moulard J, Chever O, Vasile F, Guinard E, Le Bert
M, Couillin I, Pallud J, Capelle L, Huberfeld G, Rouach N Sci
Transl Med 2018;10:eaar3796 Epilepsies are characterized by recurrent seizures, which disrupt normal
brain function. Alterations in neuronal excitability and
excitation–inhibition balance have been shown to promote seizure
generation, yet molecular determinants of such alterations remain to be
identified. Pannexin channels are nonselective, large-pore channels
mediating extracellular exchange of neuroactive molecules. Recent data
suggest that these channels are activated under pathological conditions
and regulate neuronal excitability. However, whether pannexin channels
sustain or counteract chronic epilepsy in human patients remains
unknown. We studied the impact of pannexin-1 channel activation in
postoperative human tissue samples from patients with epilepsy
displaying epileptic activity ex vivo. These samples were obtained from
surgical resection of epileptogenic zones in patients suffering from
lesional or drug-resistant epilepsy. We found that pannexin-1 channel
activation promoted seizure generation and maintenance through adenosine
triphosphate signaling via purinergic 2 receptors. Pharmacological
inhibition of pannexin-1 channels with probenecid or
mefloquine—two medications currently used for treating gout and
malaria, respectively—blocked ictal discharges in human cortical
brain tissue slices. Genetic deletion of pannexin-1 channels in mice had
anticonvulsant effects when the mice were exposed to kainic acid, a
model of temporal lobe epilepsy. Our data suggest a proepileptic role of
pannexin-1 channels in chronic epilepsy in human patients and that
pannexin-1 channel inhibition might represent an alternative therapeutic
strategy for treating lesional and drug-resistant epilepsies.
Commentary
Pannexins (Panx) are a relatively new family of integral membrane proteins that were discovered in 2000 (1), cloned in 2004 (2), and recently implicated in animal models of epilepsy (3, 4). Owing to their recent discovery, the precise physiological function of these proteins remains unclear, but one member of the pannexin family, Panx1, is expressed ubiquitously in mammals, including in neurons and astrocytes (2). To date, interest in Panx1 has focused on its enormous pore that allows small intracellular molecules (such as ATP) to efflux from the cell (5). In fact, the pore of Panx1 is large enough that, when assaying Panx1 activity, many researchers use Panx1-dependent uptake of fluorescent dyes. Although closed at rest, many stimuli have been identified that lead to opening of the Panx1 pore, including depolarization of the membrane and elevated intracellular calcium levels (6).
While a physiological role for Panx1 expression remains to be found, this protein has been implicated in several distinct pathological states including ischemia and epilepsy (2). However, the reports linking Panx1 to epilepsy have been difficult to interpret as distinct animal models have produced conflicting results. In a rodent pilocarpine model of seizures, one group noted that Panx1 functioned to inhibit seizure initiation (4) while others have noted a role for Panx1 in promoting hippocampal excitability (3).
In a recent report, Dossi and colleagues made strides to determine the precise role of Panx1 in human epilepsy. Rather than use an animal model, they conducted electrophysiological recordings on neurosurgical specimens from epilepsy patients undergoing resections. They were particularly interested in bursts of epileptiform activity that were occurring spontaneously in epileptic tissue (but not in the healthy margins used as a control). By employing a series of Panx1 permeable compounds, they demonstrated that areas of epileptic tissue had active neuronal Panx1 while the healthy tissue did not. Intriguingly, compounds that blocked Panx1 also limited the epileptiform activity without affecting the activity of surrounding areas. One of the most exciting aspects of their report was their use of two FDA approved medications to antagonize Panx1 activity: mefloquine and probenecid (discussed in greater detail, following). Application of either of these compounds to the human epilepsy tissue reduced the incidence of epileptiform activity. To confirm this result, the authors also successfully used both compounds to reduce seizure incidence in a mouse model of temporal lobe epilepsy.
While the precise mechanism linking Panx1 activity back to epilepsy remains hazy, Dossi et al. began to sketch out how Panx1 may play a role. Consistent with prior reports (2), the authors noted that the epileptiform activity that opens the Panx1 channels led to the release of ATP. However, they found that antagonizing Panx1 prevented this increase in extracellular ATP—of particular interest given that rising extracellular ATP is a key mechanism in promoting pathologic excitatory neurotransmission. Next, Dossi et al. observed that blocking P2 receptors (which bind extracellular ATP and ultimately facilitate Panx1 opening [4]) had the same effect as blocking the Panx1 receptors themselves in limiting epileptiform activity. In short, they found a positive feedback loop whereby epileptiform activity causes Panx1 to release ATP, which then binds to P2 receptors and recruits more Panx1 channels to release ATP, theoretically producing spiraling levels of excitatory activity.
While the use of neurosurgical specimens allows investigators the rare opportunity to study human disease in situ, the approach also carries significant limitations in terms of generalizability. Many of the experiments performed by Dossi et al. utilize only a few distinct patient samples, relying on sectioning a single tissue into multiple slices to generate additional data points. While this is understandable because of the rarity of such tissues, it certainly creates some doubt about the generalizability of their findings. A second point of concern is the heterogeneity of the dataset examined. Tissue was utilized from patients with primary brain tumors (grade I–IV gliomas), malformations of cortical development, and dysembryoplastic neuroepithelial tumors. This wide range of pathology certainly highlights the possibility that Panx1 may be involved as a common mechanism for many etiologies of epilepsy but further complicates the interpretation of Panx1 in any given epilepsy subtype. To the credit of Dossi et al., they confront this heterogeneity head-on, providing explicit descriptions of the samples used in every experiment. That said, the possibility remains that the aggregated results do not reflect the situation of any given sample.
As with all preclinical investigations, it is premature to conclude that either mefloquine or probenecid would be appropriate for use as antiepileptic therapies. Clearly, use of either agent may ultimately have significant therapeutic potential, but neither agent has been FDA approved for this purpose. In practice, adding an agent such as probenecid may only exacerbate the already complicated pharmacology of antiepileptic agents. Probenecid antagonizes glucuronidation and ultimately increases the steady state concentration of a range of drugs (7). If used clinically, probenecid may prove to be the antithesis of levetiracetam, an agent that has found widespread use because it rarely interferes with drug metabolism. The identification of mefloquine as an antiepileptic is also likely to raise eyebrows. As with most antimalarial agents, mefloquine has a known propensity for neuropsychiatric complications (i.e., depression, insomnia, and agitation) at prophylactic doses (8). Apart from potential side effect concerns, it is tempting to speculate that adverse effects of mefloquine use may result from antagonism of the physiological role of Panx1 in the central nervous system.
Despite the caveats outlined above, Dossi et al. provide several reasons to think that Panx1 is a good antiepileptic target, even if mefloquine and probenecid are not the ideal agents for this purpose. In their human tissue experiments, they noted that Panx1 was inactive in healthy control tissue, suggesting that Panx1 may have “use-dependent” features that restrict Panx1 activity to epileptogenic tissue. Moreover, application of Panx1 inhibitors to human control tissue produced no detectable changes in neuronal activity, indicating that under physiological conditions, Panx1 may not contribute to neural activity. Along this line, genetic knock-out mice (in which Panx1 is absent) are fertile, grow to the expected size, and have lifespans identical to wild-type littermates (9), consistent with a role for Panx1 predominantly in disease states. The elegant demonstration of these properties for Panx1 by Dossi and colleagues merits excitement for a potential new class of therapeutics. Future work will hopefully begin bridging the gap to confirm their role in epilepsy and bring these agents into clinical use as antiepileptics. Synergy between Panx1 antagonists and existing antiepileptics could allow for lower doses of either agent that ultimately circumvents concerns with their use. Further work will certainly reveal whether neurologists will be able to capitalize on the promise of Panx1 as a therapeutic target.