Design of a novel mouse model of Kcnc1 developmental and epileptic encephalopathy.

A. Design and structure of the conditional Kcnc1-A421V allele. Upon Cre-mediated recombination, the inserted WT coding sequence (CDS) flanked by LoxP sites is removed and the A421V variant inserted into exon 2 is expressed. B. Sequencing results indicate successful targeting of c.1262C>T to introduce the heterozygous A421V variant. C. PCR confirmation of two HET founders (Kcnc1-A421V/+) and two WT littermates. 167 bp, WT allele fragment; 207 bp, floxed allele fragment. D. Breeding strategy to generate control and experimental mice in which the Kcnc1-A421V variant is expressed globally and PV cells are fluorescently-labeled for targeted recording. E. Survival plot of WT (N=46; black) and Kcnc1-A421V/+ (N=33; green) mice. ***P<0.001 by log-rank Mantel-Cox curve comparison.

Impaired Cognitive Function in Kcnc1-A421V/+ Mice.

A. Example pathways for WT (black) and Kcnc1-A421V/+ (green) during the four-day acquisition phase of the Barnes maze test. Blue dot indicates the escape hole. B. Average escape latency during the acquisition phase for WT (black; N=15) and Kcnc1-A421V/+ (green; N=13) mice. C. Group data for time spent in target quadrant during probe trial of Barnes maze. D-F. Group data for WT (black; N=16) and Kcnc1-A421V/+ mice (green; N=15) during the Y-maze test for spatial working memory. D. Spontaneous alternation percentage. E. Total number of arm entries. G. Total distance traveled. Data are shown as mean ± SEM and were analyzed by Two-way repeated measures ANOVA with Tukey’s post-hoc test (B) and unpaired t-test (D-F). Significance is denoted as *P<0.05 or **P<0.01.

PV-INs from Kcnc1-A421V/+ mice exhibit attenuated voltage-gated potassium channel currents and impaired membrane Kv3.1 expression.

A. Representative image of a cell being recorded in the outside-out nucleated macropatch configuration. B-C. Example family of traces of voltage-gated K+ channel currents from a PV-IN from WT (B, black) and Kcnc1-A421V/+ (C, green) mice (P16-21). D. Average voltage-gated K+ channel current density for WT (n=13 macropatches, N=3 mice) and Kcnc1-A421V/+ (n=17, N=3 mice). E. Maximum K+ channel current density per PV-IN macropatch in WT and Kcnc1-A421V/+ mice. F. Averaged normalized plots of K+ conductance relative to voltage command indicating voltage-dependence of activation curves for WT and Kcnc1-A421V/+ mice. G. Average activation time constant for the voltage-gated K+ channel currents relative to voltage command potential in both WT and Kcnc1-A421V/+ mice. H. Representative images of individual cortical PV-INs from WT and Kcnc1-A421V/+ mice stained for Kv3.1 (green). Plasma membrane (PM), nucleus (nuc), and cytosol (cyt) are indicated in the top left panel. Note the markedly less pronounced Kv3.1 intensity in the plasma membrane in each of the Kcnc1-A421V/+ examples and more prominent cytosolic labeling (presumably corresponding to endoplasmic reticulum). I. Group quantification of ratio of membrane to cytosolic Kv3.1 for WT (n=49 cells, N=6 mice ) and Kcnc1-A421V/+ (n=48 cells, N=5 mice). Data are shown as mean ± SEM or individual data points and significance was determined using either Repeated-measures Two-way ANOVA or unpaired t-test where ***P<0.001.

Impaired PV-IN intrinsic Excitability in Juvenile and Adult Kcnc1-A421V/+ mice.

A. Representative images taken at 10X (left) and 40X (right) magnification of a layer IV neocortical PV-IN recorded in the whole-cell configuration to characterize intrinsic excitability. B-C. Representative example traces for juvenile (P16-21) WT (B, black) and Kcnc1-A421V/+ (C, green) PV-INs generating APs at current injections of-100, 200, 300, and 400 pA. The inset shows an expanded view of APs generated in response to the 400-pA current injection in both genotypes. D. Average relationship between PV-IN AP frequency in response to a range of current injections for juvenile WT (n=20 cells, N=9 mice) and Kcnc1-A421V/+ (n=36 cells, N=12 mice). E. Representative overlaid examples of single APs and the corresponding phase plots for juvenile WT (black) and Kcnc1-A421V/+ (green) PV-INs. F. Representative images for a layer IV neocortical PV-IN from an adult mouse. G-H. Representative example traces displaying intrinsic excitability in adult (P32-42) WT and Kcnc1-A421V/+PV-INs. Inset show an expanded view of APs induced by the 400-pA current step. I. Average relationship between PV-IN AP frequency and current injection for adult WT (n=14 cells, N=3 mice) and Kcnc1-A421V/+ mice (n=17 cells, N=5 mice). J. Representative overlaid examples of single APs and the corresponding phase plots for adult WT (black) and Kcnc1-A421V/+ (green) PV-INs. Data are shown as mean ± SEM and significance (***P<0.001) was determined using repeated-measures Two-way ANOVA.

Membrane and Action Potential Properties of WT and Kcnc1-A421V/+ Neurons at P16-21.

Membrane and AP Properties of Adult (P32-42) WT and Kcnc1-A421V/+ PV-INs.

Unaltered potassium currents and physiological function of excitatory neurons in P16-21 Kcnc1-A421V/+ mice.

A. Representative image of a neocortical excitatory cell being recorded in the outside-out nucleated macropatch configuration. B-C. Example family of traces of voltage-gated K+ channel currents from a excitatory cell from WT (B, black) and Kcnc1-A421V/+ (C, green) mice. D. Average voltage-gated K+ channel current density for WT (n=8 macropatches, N=3 mice) and Kcnc1-A421V/+ (n=10, N=3 mice) relative to membrane potential. E. Peak voltage-gated K+ channel current density in WT and Kcnc1-A421V/+ mice. F. Normalized voltage-dependent activation curves for WT and Kcnc1-A421V/+ mice. G. Average voltage-dependent activation time constant for the voltage-gated K+ channel currents. H. Representative images taken at 10X (left) and 40X (right) magnification of a layer IV neocortical excitatory cell recorded in the whole-cell configuration to characterize intrinsic excitability. I-J. Representative example traces showing excitatory cell AP generation in WT (I, black) and Kcnc1-A421V/+ (J, green) in response to depolarizing current injections. K. Average relationship between excitatory cell AP frequency in response to a range of current injections for WT (n=23 cells, N=3 mice) and Kcnc1-A421V/+ (n=22, N=3 mice). Data are shown as mean ± SEM and all results failed to reach significance determined via Repeated-measures Two-way ANOVA or unpaired t-test.

Juvenile (P16-21) Kcnc1-A421V/+ mice exhibit normal PV-IN mediated inhibitory synaptic neurotransmission.

A. Representative images showing simultaneous whole-cell patch-clamp recordings of cortical PV-IN and nearby excitatory cell (left, 10x magnification; right 40x magnification). B. Example traces of a PV-IN and excitatory cell pair in which generation of APs in the PV-IN (bottom trace) is sufficient to induce clear unitary inhibitory postsynaptic potentials (uIPSPs) in the excitatory cell (arrowhead, top trace). The inset shows an example view of the individual uIPSPs corresponding to each AP in the PV-IN. C-D. Example traces of uIPSCs in both WT (C) and Kcnc1-A421V/+ pairs of synaptically-connected neurons. 10 APs were generated in the PV-IN (top trace) at 40 Hz and the uIPSCs are shown below. The black and green traces are the averages of numerous individual sweeps shown in gray. E. Probability of synaptic connection between PV-IN and excitatory cell in WT (n=21 64 pairs from N=7 mice) and Kcnc1-A421V/+ (n= 5 of 43 pairs from N=11 mice). F. Average failure probability relative to presynaptic stimulation frequency in WT and Kcnc1-A421V/+ neuron pairs. G-I. Average uIPSC magnitude for the first five APs in WT and Kcnc1-A421V/+ at 20 Hz (G), 40 Hz (H), and 80 Hz(I) stimulus frequencies. J-K. Paired-pulse ratios for both WT and Kcnc1-A421V/+ mice relative to stimulus frequency. Ratio of second uIPSC to the first is provided in J, while K displays the average ratio of the last uIPSC to the first. L. Average latency from peak of presynaptic AP to peak of the uIPSC in WT and Kcnc1-A421V/+ mice. Data are shown as mean ± SEM and all results failed to reach significance determined via repeated-measures Two-way ANOVA or unpaired t-test.

Adult (P32-42) Kcnc1-A421V/+ mice exhibit altered PV-IN mediated synaptic neurotransmission.

A. Example presynaptic cortical PV-IN and post-synaptic excitatory neuron. Arrowhead and inset display the uIPSPs induced in the postsynaptic cell when the PV-IN generates APs. B-C. Example traces of uIPSCs in both adult WT (B) and Kcnc1-A421V/+ (C) pairs of synaptically-connected neurons. 10 APs were generated in the PV-IN at 40 Hz and the evoked uIPSCs are displayed in the trace below where the black and green traces are the averages of numerous individual sweeps shown in gray. D. Connection probability between WT (14 of 36, N=8 mice) and Kcnc1-A421V/+ (13 of 36, N=8 mice) pairs PV-INs and nearby excitatory cells. E. Average frequency-dependent rate of failure for the first five APs in adult WT and Kcnc1-A421V/+ neuron pairs. F-H. Average uIPSC magnitude of the first five APs for adult WT and Kcnc1-A421V/+ at 20 Hz (F), 40 Hz (G), and 80 Hz (H). I-J. Paired-pulse ratios for WT and Kcnc1-A421V/+ neuron pairs (uIPSC2/uIPSC1 provided in I, uIPSClast/uIPSCfirst provided in J) relative to stimulus frequency. K. Average latency from AP peak to onset of the uIPSC in WT and Kcnc1-A421V/+ mice. Data are shown as mean ± SEM or individual values and significance (*P<0.05, **P<0.01) was determined using unpaired t-test or Repeated-measures Two-way ANOVA.

In vivo 2P calcium imaging reveals paroxysmal hypersynchronous discharges and altered neuronal excitability in Kcnc1-A421V/+ mice.

A. A. Experimental setup for in vivo 2P calcium imaging with representative calcium transients from cells expressing AAV-hSyn-GCaMP8m and mean dF/F0 of the whole field of view (FOV) aligned to locomotion speed. B. Representative 2P field of view during a hypersynchronous discharge in a Kcnc1-A421V/+ mouse. C. Mean dF/F of the field of view (top), calcium transients of individual somata (middle), and locomotion speed (bottom) during a paroxysmal discharge in the Kcnc1-A421V/+ mouse relative to typical baseline activity shown in a WT mouse. Note that, in contrast to epochs of low-amplitude synchronization in WT associated with transition from quiet wakefulness to locomotion, there is no locomotion during the larger-amplitude hypersynchronous discharges identified in Kcnc1-A421V/+ mice. D. Frequency of paroxysmal discharges in each mouse (N=5 WT, N=7 Kcnc1-A421V/+ mice). E. Experimental design for in vivo 2P calcium imaging of somata positive (PV+) and negative (PV-) for parvalbumin. F. Example calcium transients of PV+ (bottom) and PV-(top) somata aligned to locomotion speed in a WT (left) and Kcnc1-A421V mouse (right). G-H. Transients per minute during quiet rest in (G) PV-(WT, N=n=885 cells, 4 mice, mean = 1.17; Kcnc1-A421V, n=1041 cells, N=3 mice, mean=1.63) and (H) PV+ cells (WT, N=4 mice, n=110 cells, mean=0.94; Kcnc1-A421V, N=3 mice, n=100 cells, mean=1.50). I-J. Mean peak height in (I) PV-(WT, mean=0.46; Kcnc1-A421V, mean=0.41) and (J) PV+ cells (WT, mean=0.48; Kcnc1-A421V, mean=0.40). Data points are shaded by mouse identity. Statistical comparisons were performed using mixed-effects modeling.

Kcnc1-A421V/+ mice exhibit spontaneous seizures and seizure-induced death.

A. Representative example trace of the EEG collected from an adult Kcnc1-A421V/+ mouse during a nonfatal seizure. After the seizure-related spike-wave discharges, there are large spikes that are associated with diffuse whole-body jerks. B. Representative generalized tonic-clonic seizure resulting in seizure-induced sudden death in a Kcnc1-A421V/+ mouse. C. Raster plot indicating nonfatal seizures (blue bars), seizure-induced sudden death (red bars), interictal runs of spikes (green bars) without clear behavior manifestation, and periods of repetitive myoclonic seizures (yellow shading) for each mouse examined (N=12). Recordings in WT (N=4) control mice did not show epileptic seizures or runs of spikes.

Pvalb-TdTomato reporter effectively labels PV-INs in WT and Kcnc1-A421V/+ mice.

A. Representative immunohistochemistry images for parvalbumin in WT (top row) and Kcnc1-A421V/+ (bottom row) mice (P21-33) showing a high degree of overlap between the tdTomato reporter and parvalbumin expression. The asterisk indicates rare cells that are tdTomato+, but parvalbumin-. The arrowhead indicates cells that are parvalbumin+ but tdTomato-. Scale bar, 100mum. B. Counts of PV cells per unit area (mm2) are not different between WT and Kcnc1-A421V/+ mice (N=3 mice/genotype). C. Sensitivity rate (probability of a cell being tdTomato+ if it is parvalbumin+; N=6 mice/genotype). D. False-positive probability (proportion of all tdTomato+ cells that are parvalbumin negative; N=6 mice/genotype).

Early postnatal development of Kcnc1-A421V/+ mice.

A. Representative example image showing littermate WT and Kcnc1-A421V/+ mice at postnatal day 21. B. Average body weights for WT (N=10; black) and Kcnc1-A421V/+ (N=11; green) at postnatal days 7, 14, and 21. C. Average brain weights for WT and Kcnc1-A421V/+ mice at postnatal day 7 (WT, N = 2; Kcnc1-A421V/+, N =6), 14 (WT, N=3; Kcnc1-A421V/+, N=3), and 21 (WT, N=4; Kcnc1-A421V/+, N=6). Note that for both B and C, error bars depicting SEM are present, but do not appear beyond the symbols. D. Onset of developmental and motor benchmarks for WT (N=7) and Kcnc1-A421V/+ (N=10) mice. E. Onset of startle reflex and other typical behavioral/motor milestones in WT (N=7) and Kcnc1-A421V/+ (N=10) mice. ***P<0.001 by Mixed-Effects Analysis followed by Sidak’s Multiple comparisons post-hoc test.

Loss of potassium current density in A421V-expressing HEK cells.

A-C. Example Kv3.1 currents in HEK cells expressing WT (A; black), a 50:50 mixture of WT and A421V (B; blue), and A421V (C; green) Kv3.1 subunits. Voltage command protocol shown below WT example. D. Average current density for WT (black), WT + A421V (blue), and A421V (green). Note that the A421V variant leads to a profound loss of function. E. Average normalized conductance relative to membrane potential for WT (black), WT + A421V (blue), and A421V (green). G-V curves for the A421V variant should be interpreted with caution as the recorded currents are exceedingly small and cannot be easily differentiated from the small endogenous delayed rectifier potassium currents known to be present in HEK cells.

Intrinsic physiology of neocortical layer II-IV PV-INs remains impaired when resting membrane potential is not normalized.

(A-B) Representative example traces for juvenile (P16-21) WT (A, black) and Kcnc1-A421V/+ (B, green) PV-INs generating APs at current injections of -100, 200, 300, and 400 pA from their resting membrane potential without DC bias current (as in Figure 3). C. Average frequency-current relationship for WT (n=20 cells, N=9 mice) and Kcnc1-A421V/+ (n=32 cells, N=12 mice) PV-INs with uncorrected resting membrane potential. Data are shown as mean ± SEM and significance (***P<0.001) was determined using repeatedmeasures Two-way ANOVA.

Subtle abnormalities in neocortical layer V PV-INs from juvenile (P16-21) Kcnc1-A421V/+.

(A-B) Representative example traces showing spiking in layer V neocortical PV-INs from WT (A, black) and Kcnc1-A421V/+ (B, green) mice in response to depolarizing current injections. (C) Average PV-IN spiking frequency relative to current injection. A significant interaction effect between genotype and current injection was observed (***P<0.001) by two-way ANOVA. Data are shown as mean ± SEM and significance of post-hoc comparisons (*P<0.05) by repeated-measures Two-way ANOVA followed by Sidak multiple comparisons test.

A. Representative images taken at 10X (left) and 40X (right) magnification of a PV-positive cell in the reticular thalamus recorded in the whole-cell configuration to characterize intrinsic excitability. B. Representative example traces of APs generated in response to current injections of varying magnitudes. C. Average hyperpolarization-induced rebound APs in WT (n=16 cells, N=4 mice) and Kcnc1-A421V/+ (n=19 cells, N=5 mice) in response to various current injections from 0 to -100 pA. D. Frequency-current relationship shows impaired intrinsic excitability in the RT PV cells relative to depolarizing current injections (0-400 pA). Data are shown as mean ± SEM and significance is denoted as *P<0.05 or ***P<0.001 by Repeated-measures Two-way ANOVA.

Intrinsic excitability is unchanged in excitatory cells from P32-42 Kcnc1-A421V/+ mice.

(A) Representative images taken at 10X (left) and 40X (right) magnification of a recorded layer IV excitatory cell. (B-C) Representative example traces of evoked APs in response to depolarizing current injections in cortical layer IV excitatory neurons from WT (B) and Kcnc1-A421V/+ (C) mice. (D) Average frequency-current relationship for WT (black; n=12, N=4) and Kcnc1-A421V/+ (n=12, N=4) mice. Data are shown as mean ± SEM and all results failed to reach significance determined via repeated-measures Two-way ANOVA.

Excitatory neuron to PV-IN unitary excitatory synaptic neurotransmission is unaltered in juvenile (P16-21) Kcnc1-A421V/+ mice.

Excitatory neuron to PV-IN excitatory synaptic neurotransmission is unaltered in Kcnc1-A421V/+ mice. A. Example traces of simultaneous recordings of a synaptically-connected PV-IN and excitatory cell neuron pair in which generation of APs in the excitatory cell leads to uEPSPs in the PV-IN. The inset shows an expanded view of the uEPSPs in the PV-IN generated from each AP in the excitatory cell. B-C. Representative example traces of uEPSCs generated in WT (B) and Kcnc1-A421V/+ pairs of excitatory cells (top trace) and nearby PV-INs (bottom trace). In response to a 20-Hz train of five APs generated in the excitatory cells, uIPSCs are recorded in the PV-INs (shown in gray) with the average of numerous sweeps shown for WT (black) or for Kcnc1-A421V/+ (green). D. Connection probability between the excitatory cell and the PV-IN for WT (n=9 of 64 pairs from N=7 mice) and Kcnc1-A421V/+ (n=6 of 43 pairs from N=5 mice). E. Average failure rate for WT and Kcnc1-A421V/+ neuron pairs relative to presynaptic stimulation frequency. F. Average uEPSC magnitude in response to a train of five APs generated at 20 Hz in WT and Kcnc1-A421V/+ neuron pairs. G. Paired-pulse ratio of the second uEPSC to the first uEPSC in WT and Kcnc1-A421V/+ mice. H. Average synaptic latency between peak of AP to peak of uEPSC in both WT and Kcnc1-A421V/+ mice.

A-B. Transients per minute during quiet rest in only active (A) PV-(WT, n=847 cells, N=4 mice; Kcnc1-A421V, n=901 cells, N=3 mice) and (B) PV+ cells (WT, n=65 cells; Kcnc1-A421V, n=60 cells). C-D. Transients per minute during running in (C) PV-(WT, n=885 cells, N=4 mice; Kcnc1-A421V, n=1041 cells, N=3 mice,) and (D) PV+ cells (WT, n=110 cells, N=4 mice; Kcnc1-A421V, n=100 cells, N=3 mice). E-F. Mean transient height in (E) PV- and (F) PV+ cells. Error bars indicate standard error of the mean, and values next to data points are the mean by genotype. Data points are shaded by mouse identity. Statistical comparisons were performed using mixedeffects modeling. G-H. Boxplot of percent of active (G) PV- and (H) PV+ cells for each mouse.