Drug-Resistant Epilepsy: Could It Be Astrocyte Disfunction?

by Moheb Costandi

July 27, 2012

Epilepsy is a chronic and debilitating neurological condition that affects around 50 million people worldwide. Anti-convulsant drugs can control seizures for only about two-third of people with the disorder, and little progress has been made developing new therapies for these drug-resistant cases. 

Traditionally, epilepsy researchers have focused on the role that neurons play in the condition. It is widely accepted that seizures arise from abnormal electrical discharges in groups of overactive neurons; anti-epileptic drugs dampen down this hyperexcitability.

This ‘neurocentric’ approach was challenged during a symposium at the FENS Forum on Neuroscience in Barcelona earlier this month, in which researchers described the role played by astrocytes in the condition.

Astrocytes are one of several types of non-neuronal brain cells called glia. Glial cells had largely been dismissed as little more than support cells that nourish and protect neurons. In the past decade, however, it has become clear that they do much more, and astrocytes in particular have emerged as key players in every aspect of brain function.

Detlev Boison of Legacy Research in Portland, Oregon, discussed the role of adenosine, a ubiquitous cellular metabolite that also functions as an inhibitory neurotransmitter, acting as a "native" anti-convulsant. In the adult brain, adenosine kinase (ADK), the enzyme that catalyses the synthesis of adenosine, is expressed almost exclusively by astrocytes.

In 2008, Boison and his colleagues reported that transgenic mice overexpressing ADK are prone to seizures and exhibit some of the cognitive impairments seen in epilepsy, leading Boison to propose the ADK hypothesis of epileptogenesis.

“ADK is an unusual enzyme because it is inhibited by adenosine,” says Boison. “Once adenosine levels reach a certain threshold, ADK shuts down. Brain injury causes a surge of ATP from dying cells, which is rapidly degraded into adenosine. This leads to an adenosine deficiency, which can be a direct cause of epileptic seizures.”

More recently, Boison’s group reported that tuning down ADK expression in astrocytes almost completely abolishes seizures in the mouse model of epilepsy, suggesting that the enzyme could be a target for gene therapy.

Also during the symposium, Ole Petter Ottersen of the University of Oslo discussed the role of the balance of potassium and water in cells could affect how susceptibile a person may be to seizures. Astrocytes express aquaporin-4 (AQP4), a membrane channel that regulates the flow of water via protuberances called endfeet, which come into close contact with brain capillaries. AQP4 also interacts with the Kir4.1 potassium channel to regulate the clearance and reuptake of potassium.

Early studies showed deleting the AQP4 gene in mice increases the severity of seizures and slows potassium homeostasis. Subsequently, Ottersen and his colleagues published evidence that human temporal lobe epilepsy involves a loss of AQP4 expression in astrocyte endfeet at capillaries.

“It’s now very clear that there is perturbed expression of both astroglial AQP4 and Kir4.1 in medial temporal lobe epilepsy,” says Ottersen. “These are possible targets for therapy in the future, but we still don’t know the exact mechanism by which AQP4 affects potassium homeostasis.”

Christian Steinhäuser, of the University of Bonn noted that temporal lobe epilepsy involves dramatic changes in astrocyte structures and relationships. In the hippocampus, a medial temporal lobe structure that is often the source of seizures, individual astrocytes make contact with up to 140,000 synapses, and are therefore perfectly positioned to orchestrate neuronal signaling [See: One Man’s Continuing Contribution to the Science of Memory].

Steinhäuser and his colleagues have found that astrocyte communication is disrupted in epilepsy. Like neurons, astrocytes form functional networks and are connected via gap junctions, or electrical synapses, consisting of channels that link adjacent cells and enable electrical signals to travel near-instantaneously. 

Steinhäuser’s group has shown that the electrical coupling between astrocytes is disrupted in tissue samples from people who have temporal lobe epilepsy. Animal studies confirm a role for gap junctions in epilepsy: Mice lacking the Connexin36 gene, which encodes a gap junction protein, have increased sensitivity to drug-induced seizures. Expression of other connexin genes also has been found to be altered in human epilepsies.

“Electrical uncoupling is one of the earliest events [following experimentally-induced seizures in mice],” says Steinhäuser, “but this process appears to be reversible during an early time window, opening up the possibility of pharmacological intervention.”

Finally, Eliana Scemes, of Albert Einstein College of Medicine in New York, summarized research into pannexins, a family of proteins that was discovered about 10 years ago. Pannexins are related to gap junction proteins, but in the vertebrate brain they primarily help transport ions and large molecules across cell boundaries. These proteins are expressed by both neurons an astrocytes, and release ATP in response to high levels of potassium.

ATP can activate neurons by binding to P2X pannexin receptors; this signaling is disrupted in epilepsy. Scemes and her colleagues hypothesized that pannexins may contribute to seizures by releasing ATP. To test this, they induced seizures in wild-type (normal mice) and in genetically engineered animals lacking the pannexin1 gene.

They found that the genetically engineered animals were less susceptible to seizures, and that blocking pannexin1 channels reduced the severity of seizures in the normal mice. A further set of experiments showed that pannexin1 channels are activated during epilepsy-like activity in slices of brain tissue from the normal mice, and that both the size and duration of abnormal electrical discharges were significantly reduced in slices from the mutant animals.   

More recently, the researchers reported that ATP signaling is deficient in astrocytes isolated from mice lacking the Pannexin1 gene. Based on their findings, Scemas and her colleagues propose that pannexin1 and P2X receptors act together to contribute to epileptic seizures. They theorize that pannexin1 releases ATP, which then activates nearby neurons by binding to P2X receptors.

“There is now compelling evidence that astrocytes modulate synaptic signaling between neurons,” says Steinhäuser, who chaired the symposium, “and “we are all convinced that astrocytes are very promising new targets in the search for anti-epileptic drugs.