www.ScienceTranslationalMedicine.org 21 November 2012 Vol 4 Issue 161 161fs40 1 FOCUS Epilepsy is a devastating disease afecting 1% of the world’s population. 30% of epilep- sy patients are resistant to available drugs, a percentage that has not changed since the 1950s. his illustrates how little progress has been made in understanding the pathogenic mechanisms of this disease and in devel- oping new treatments. Given this, and the fact that epilepsy patients whose seizures are controlled by anti-epileptic drugs oten sufer from detrimental side efects, there is a dire need for new therapeutic approaches for treating epilepsy (1). Drug development for epilepsy faces two main challenges: inding drugs that abolish seizure activity without deleterious side efects, and preventing the emergence of epilepsy whether inherited or acquired. Inherited epilepsies originate from gene mutations that usually result in reorgani- zation of neuronal networks in the brain during development. Acquired epilepsies may occur ater insults to the brain such as trauma or meningitis that lead to a complex reorganization of brain neuronal networks in a process called epileptogenesis (Fig. 1). Eventually this reorganization results in spontaneous seizures, sometimes ater a sei- zure-free latent period that may have lasted for decades. Preventing epileptogenesis and controlling seizures may involve difer- ent molecular targets and hence diferent therapeutic strategies. he ideal therapeutic strategy would target only the brain regions responsible for seizure genesis and would prevent seizures without interfering with physiological function. None of the current anti-epileptic drugs fulill these criteria. Reporting in this issue of Science Trans- lational Medicine, Wykes et al. (2) take a diferent tack using gene therapy instead of drugs. Working in a rat model of focal neocortical epilepsy, they use gene therapy to deliver two diferent therapeutic genes to the rat brain. hey show that expression of the therapeutic genes by a subpopulation of excitatory principal (pyramidal) neurons in the epileptogenic region decreases the excit- ability of these neurons, abrogating epileptic activity and preventing epileptogenesis. he authors created an epileptogenic re- gion in rat brain by injection of tetanus tox- in into the rat motor cortex. In this model, the rats display epileptic activity and motor seizures similar to epilepsia partialis con- tinua observed in human patients. he au- thors used two diferent gene therapy strate- gies. First, using an optogenetics approach, they injected a lentiviral vector carrying a gene encoding the light-activated chloride ion channel halorhodopsin into the rat brain along with tetanus toxin to induce fo- cal epilepsy (Fig. 1). he investigators used a miniaturized wireless system to record EEG activity around the clock and a newly designed algorithm that automatically de- tects and classiies epileptic activity. A week later, the researchers shone laser light in 20 s pulses (delivered by an optic iber) into the epileptogenic region of the rat brain to acti- vate halorhodopsin expressed by excitatory principal neurons (Fig. 1). Once activated, the halorhodopsin channels allowed chlo- ride ions to low into the neurons resulting in membrane hyperpolarization, a decrease in neuronal excitability, and a decrease in epileptic activity as detected by EEG. Con- trol rats not exposed to light showed no ab- rogation of epileptic activity because halor- hodopsin remained inactive. his optogenetics strategy has several advantages: It can target one small brain re- gion, can be modulated by switching light on and of, and does not interfere with normal function because only a few neurons are tar- geted (the number of neurons is determined by the volume of tissue accessed by the viral construct). But there are a number of open questions including how long halorhodop- sin delivered by gene therapy will be ex- pressed by excitatory neurons and whether long-term expression of this foreign protein could have deleterious consequences. Ide- ally, to be able to apply optogenetics to treat epilepsy, one would need to use a closed loop system that is able to detect seizure on- set and then trigger optogenetic silencing of the excitatory neurons in the epileptic focus. A proof-of-concept that closed loop sys- tems could be used to control epilepsy has been provided recently by Buzsaki and col- leagues (3) and Huguenard and co-workers (4). In these studies, electrical stimulation and light-activated halorhodopsin, respec- tively, were used to abolish epileptic activ- ity in rodents with absence-like seizures. Absence-like seizures are characterized by short-lasting behavioral arrest and can be controlled efectively with existing anti- epileptic drugs. Such closed loop systems should now be validated in animal models of non-absence seizures that are refractory to drug treatment such as the rat model of focal neocortical epilepsy used by Wykes and co-workers. In their second strategy, which did not involve optogenetics, Wykes et al. injected a lentiviral vector carrying the gene encoding a native potassium ion channel, Kv1.1, into the rat brain (Fig. 1). Expression of Kv1.1 by neurons led to an outward low of potassium ions resulting in membrane hyperpolariza- tion and reduced neuronal excitability. he principal diference between this approach and the optogenetics strategy is that over- expression of the Kv1.1 channel produced a constitutive decrease in neuronal excit- ability without the need for activation by an external light source. When the authors injected the viral vector carrying the Kv1.1 gene concomitantly with tetatus toxin into the rat brain, they were able to prevent epi- leptogenesis as shown by EEG. Even more exciting, when the researchers injected the virus carrying the Kv1.1 gene into the epi- leptogenic region of rats in which epilepsy had been established one week earlier with tetanus toxin, there was a steady reduction in epileptic activity. he reduction in epilep- tic activity continued over several weeks and ultimately was abolished, but there were no motor or behavioral deicits. he gene therapy approach of Wykes et al. has several advantages over current anti- epileptic drugs. Such drugs have major limi- tations because they act nonspeciically in the brain resulting in side efects such as cog- nitive deicits. In contrast, gene therapy acts speciically delivering the therapeutic gene to a small number of neurons in a localized brain region (Fig. 1). A further challenge with using anti-epileptic drugs for treating epilepsy is that neurons in epileptogenic regions oten express multi-drug resistant EPILEPSY Treating Epilepsy with a Light Potassium Diet Christophe Bernard E-mail: christophe.bernard@univ-amu.fr Aix Marseille Université, INS, and Inserm, UMRS 1106, Marseille Cedex 05, France. Gene therapy in a rat model of focal neocortical epilepsy decreases neuron excitability, preventing epileptogenesis and abrogating epileptic activity (Wykes et al.).