Fig. 3.

Dual effect of photo-depolarization of ChR2-expressing PV+ interneurons on the duration of seizures. a A closed-loop system was used to detect the onset of seizures and to vary the delay of optogenetic stimulation of ChR2-expressing PV+ neurons during pathological ECoG activity. FPGA field-programmable gate array. b Effects of photostimulation delivered at varying delays (increasing from top to bottom). Sample traces (left) illustrate pairs of consecutive electrographic seizures without and with photostimulation (blue rectangles indicate photostimulation). Intervening periods between seizures are omitted. The bar charts (middle and right) show a switch from light-induced seizure suppression with 0 and 0.5 s delays to seizure prolongation with >2 s delay (middle - normalised duration, right - absolute duration, n = 8, 7, 7 and 5 mice for 0, 0.5, 2 and >2 s delays respectively). In the scatter plots, each data point represents a pair of consecutive seizures, allowing a comparison of the duration of the discharge that was used to trigger the laser to that of the preceding control discharge. A linear regression line (blue) below the line of identity (dashed black) indicates that photo-depolarization reduced the seizure duration, whilst a line above the line of identity indicates prolongation of the seizure duration. c Normalised duration of ictal discharges during delayed photostimulation compared to randomized reshuffled data (*p < 0.05 paired t-test, n = 5; ### p < 0.001, unpaired t-test, n = 9, includes data from additional animals in randomized group). Error bars represent s.e.m. d Analysis of a subset of 0 s delay trials shown in b where the control seizure was longer than 2.5 s. The anti-seizure effect of immediate PV+ interneuron photostimulation persists, arguing against selection bias. e Summary of seizure durations in d. Statistics in b and e, paired t-test. *p < 0.05, **p < 0.01, ***p < 0.001; error bars represent s.e.m. Source data are provided as a Source Data file