Evaluation of the degree of injury in CCI models. A Experimental procedure. Mice were subjected to different degrees injury with CCI models. Behavior test and tissue evaluation were performed at 1, 3, 7, 14, 21 days and 3 months post- traumatic brain injury (TBI). Skull electrodes were implanted at 11 weeks after brain damage. Long-term video-electroencephalogram (v-EEG) was monitored for one week, then followed by pentylenetetrazol (PTZ) injection. B Summary of behavior test in CCI models compared to sham groups at different time points. C Bar graph showing the percent mortality in CCI models compared to sham groups within 2 weeks. D Representative cresyl violet stained brain slices from three cortical regions in CCI models. E Summary of tissue loss in CCI models. (*p < 0.05, ***p < 0.001 vs. sham group; #p < 0.05, ###p < 0.001 vs. mild group, Error bars indicate standard error of the mean).

Characteristics of spontaneous seizures and pentylenetetrazol-induced seizure susceptibility in CCI models. A Locations of craniectomy window (black dotted circle), cortical injury (red circle), and placement of skull electrodes. B Long-term v-EEG monitor for 7 days on the TBI models. The epileptiform discharges of spontaneous seizure were showed in the right. C Evaluation of spontaneous seizures in the TBI models with different degree injury. It showed the percentage of spontaneous seizures(left), number of seizures per day (middle), and duration of seizures in TBI models. D Representative traces of ictal epileptiform discharges after pentylenetetrazol (PTZ) injection on the TBI models. *indicated interictal discharges. The red arrow indicated the onset of epileptiform discharges of generalized convulsive seizures. N = 4-6 for each group. E Quantification of latencies to myoclonic jerk. N = 4-6 for each group. F Quantification of latencies to generalized convulsive seizures. N = 4-6 for each group. G Quantification of seizure severity scores. N = 4-6 for each group. (*p < 0.05, ***p < 0.001. Error bars indicate standard error of the mean).

Increased KNa1.1 channel expression after TBI. A The expression of KNa1.1 protein at the different time points after TBI. N = 5-7 for each group. B On the day 14 after TBI, immunofluorescence staining showed enhanced fluorescence intensity and increased number of co-localized KNa1.1 channel on neurons around lesions (white arrowhead). N=3 for each group. C The expression of KNa1.1 channel on the lesion area on the neocortices from patients with PTE and patients with mesial temporal lobe epilepsy. N=3 for each group. D Immunofluorescence showed increased and more co-localized KNa1.1 channel on neurons in the patient with PTE compared with the patient with mesial temporal lobe epilepsy. (*p < 0.05, **p < 0.01, ***p < 0.01. Error bars indicate standard error of the mean).

Increased KNa1.1 preferentially in GABAergic inhibitory interneurons on Dpi-14d mice. A The expression of VGLUT1 and GAD67 protein after TBI. B At Dpi-14d, immunofluorescence staining showed the distribution of KNa1.1 channel and VGLUT1on neurons. The white arrowheads indicated the neurons with co-localized KNa1.1 channel and VGLUT1. The white arrows indicated the neurons labelled with VGLUT1 but without KNa1.1 channel. C At Dpi-14d, immunofluorescence staining showed the distribution of KNa1.1 channel and GAD67 on neurons. The white arrowheads indicated the neurons with co-localized KNa1.1 channel and GAD67. (*p < 0.05. Error bars indicate standard error of the mean).

Altered electrophysiological characteristics of perilesional TBI neurons. A Macroscopic EEG recording showed interictal and ictal discharges in TBI mice. Neurons in layer II/III of perilesional cortex were selected for patch clamp recording. B Representative traces and action potential (AP) firing of PNs from sham and TBI groups with -100, 100, +300 pA current injection. C Representative traces and AP firing of INs from sham and TBI groups with -100, 0, +300 pA current injection. D Summary of afterhyperpolarization and AP frequency in PNs and INs. PNs: N = 18, 21 for sham and TBI groups. INs: N = 6 for sham and TBI groups. E Representative traces of spontaneous excitatory postsynaptic currents (sEPSC) and spontaneous inhibitory postsynaptic currents (sIPSC) of PNs (Black, sham group; red, TBI group). N = 18, 21 for sham and TBI groups. F Quantitative summary of sEPSC and sIPSC frequency and excitation/inhibition ratio of PNs. N = 18, 21 for sham and TBI groups. (*p < 0.05, ***p < 0.001. Error bars indicate standard error of the mean).

Evaluation of homozygous kcnt1-/- mice and age-matched wild type controls. A Diagram of KNa1.1 protein in the kcnt1−/− mice showed corresponding mistranslated site in its pore-formation domain. B PCR verification of kcnt1 exon 3-11 deletion. The predicted amplicons for WT and the exon 3-11 deleted mutant were 489 bp and 606 bp, respectively. The corresponding sample from the WT, kcnt1+/-, and kcnt1−/−. C Western blots showed KNa1.1 expression from the WT, WT-TBI (for a positive control), and kcnt1−/− mice. D Representative image showed similar gross morphology of brains dissected from 10 to 12-week-old male WT, kcnt1−/− mice. N = 7 in WT and kcnt1−/− groups, respectively. Scar bar represented the length of 1cm. E Evaluation of motor task in kcnt1−/− and WT mice (WT, black; kcnt1−/−, blue). N = 40, 28 in WT and kcnt1−/− groups, respectively. F Representative diagram showed cultured primary cortical neuron from kcnt1−/− mice. dV/dt is plotted as function of membrane voltage for the action potentials (APs) (WT, black; kcnt1−/−, blue). Horizontal dotted lines correspond to the dV/dt value that is 10% of peak dV/dt for a given cell. G Comparisons of AP threshold and rheobase of cultured primary cortical neuron from WT and kcnt1−/−mice. N= 17, 11 in WT and kcnt1−/− groups, respectively. H -40, 0, 60, 100, and 200 pA current injections (0.6 s) were used to elicit firing in WT and kcnt1−/− (blue) neurons from a holding potential of −70 mV. Comparisons of AP frequency and resting membrane potential of cultured primary cortical neuron from kcnt1−/− and WT mice. N= 17, 11 in WT and kcnt1−/− groups, respectively. (*p < 0.05. Error bars indicate standard error of the mean).

Decreased seizure susceptibility in homozygous kcnt1-/- mice after TBI. A Representative MRI image showed both TBI and kcnt1−/− -TBI mice appeared lesions in the cortex after TBI. B The motor skill learning of WT mice and kcnt1-/- mice after TBI were improved (accelerating rotarod) over successive trials. N = 11,17, 16 for Sham, TBI, and kcnt1−/−-TBI groups, respectively. C The latency to seizure, including generalized and partial seizure, and distribution of Racine scores after injection of PTZ. N = 11,9,7,12 for Sham, TBI, kcnt1−/−-sham, and kcnt1−/−-TBI groups, respectively. (*p < 0.05, **p < 0.01, vs. sham group; #p < 0.05, ##p < 0.01 vs. TBI group, Error bars indicate standard error of the mean).