Abstracts

Rationale: The therapy of focal neocortical epilepsy remains unsatisfactory. Systemic antiepileptic drugs fail to control seizures in a large fraction of patients and surgical resection produces adequate seizure control in only about 60% of patients. We have explored the feasibility of using a small ultraviolet light emitting diode (UV LED) to release low concentrations of GABA from a new caged GABA, 4-[[(2H-benzopyran-2-one-7-amino-4-methoxy)carbonyl]amino]butanoic acid (BC204) Methods: We used standard techniques to prepare 600 μm cortical brain slices, which were perfused with artificial cerebrospinal fluid (ACSF) in a chamber with a glass coverslip forming the bottom. A UV LED package was placed immediately below the coverslip allowing unfocused light to directly illuminate the slice from below. A computer or pulse generator TTL pulse controlled light activation. In our initial experiments we illuminated the slice during perfusion with control ACSF and monitored the temperature with a small thermocouple as a function of time and UV LED current. We then examined the effect of 4.5 sec UV illumination (100 mA) during perfusion with ACSF lacking magnesium and containing 100 μM 4-aminopyridine (0 Mg/4-AP ACSF) in the absence and presence of BC204. We compared the baseline standard deviation (SD), number of spikes, and number of paroxysmal bursts in the 4.5 sec epoch before illumination and the 4.5 sec epoch during illumination. Results: When we varied current through the UV LED, the highest temperature increases observed during a 30 sec observation period were 0.09 ± 0.02 and 0.32 ± 0.02°C, at 50 and 400 mA, respectively. Our subsequent BC204 uncaging experiments then utilized 100 mA of current for 5 sec. When the slice was perfused with 0 Mg/4-AP ACSF lacking BC204, the UV LED had no effect on baseline SD, spike number, or bursts (p>0.6 for all 3 situations). In the presence of 30 μM BC204, baseline SD was reduced from 0.016 ± 0.01 to 0.009 ± .007 mV; burst number decreased from 1.8 ± 1.3 to 0.4 ± 0.5; and spike number diminished from 21.3 ± 14.3 to 4.2 ± 4.1 (p<0.01 for all 3 parameters). When the BC204 concentration was reduced to 10 μM, baseline SD; burst number; and spikes were also reduced (p<0.01 for all groups). Additional analysis showed that BC204 did not affect paroxysmal activity in the absence of UV light. BC204 concentrations as high as 100 μM failed to damage cultured rat neurons. Conclusions: Our results indicate that a relatively low concentration of the caged compound BC204 can release sufficient GABA after illumination with an LED to suppress paroxysmal activity in a 600 μm thick slice of brain tissue. Although it is difficult to extrapolate from these results to focal human epilepsy, our robust results in a very extreme seizure model make us optimistic that it will be possible to configure UV LEDs close to an epileptic focus exposed to BC204 or a related compound. Closed loop activation of the UV LED could uncage sufficient GABA to suppress focal seizures without having to permanently destroy the cortical seizure focus.