Wild-type zebrafish Kv3.3a, iR4H, or a 1:1 mixture of wild type and iR4H RNAs were expressed in Xenopus oocytes for voltage clamp analysis. (A) iR4H is inactive when expressed alone and suppresses current amplitude when co-expressed with the wild type subunit. Currents were evoked by pulsing from a holding potential of –90 mV to +60 mV for 50 ms. Representative traces are shown for wild type alone, a 1:1 mixture of wild type and iR4H, or iR4H alone. (B) Average current amplitudes measured at +60 mV are shown after expression of wild type alone, a 1:1 mixture of wild type and iR4H, or iR4H alone. Values are provided as mean ± SEM. Note that the level of suppression at a 1:1 ratio, with 32.1% of the wild-type current amplitude remaining, is consistent with the prediction of the binomial distribution (31.2% wild-type current amplitude remaining) for the hypothesis that a channel formed from one mutant and three wild type subunits is active and expressed on the cell surface, whereas additional mutant subunits in the tetramer abolish activity (Minassian et al., 2012; Mock et al., 2010). (C) Incorporation of an iR4H mutant subunit results in a dominant shift in the voltage dependence of activation in the hyperpolarized direction. Normalized conductance for wild type expressed alone (black squares) or for wild type and iR4H expressed at a 1:1 ratio (red circles) has been plotted as a function of voltage. Data are provided as mean ± SEM. If error bars are not visible, they are smaller than the size of the symbol. Data were fitted with a Boltzmann function to estimate the midpoint voltage, V_0.5, and the slope factor. Values of V_0.5: wild type alone (n = 14), 13 ± 1 mV; 1:1 mixture (n = 14), 7 ± 1 mV. Values of slope factor: wild type alone (n = 14) 11 ± 1; 1:1 mixture (n = 14) 7 ± 0.3.

Wild-type zebrafish Kv3.3a, iR4H, or a 1:1 mixture of wild type and iR4H RNAs were expressed in Xenopus oocytes for voltage clamp analysis. (A) iR4H is inactive when expressed alone and suppresses current amplitude when co-expressed with the wild type subunit. Currents were evoked by pulsing from a holding potential of –90 mV to +60 mV for 50 ms. Representative traces are shown for wild type alone, a 1:1 mixture of wild type and iR4H, or iR4H alone. (B) Average current amplitudes measured at +60 mV are shown after expression of wild type alone, a 1:1 mixture of wild type and iR4H, or iR4H alone. Values are provided as mean ± SEM. Note that the level of suppression at a 1:1 ratio, with 32.1% of the wild-type current amplitude remaining, is consistent with the prediction of the binomial distribution (31.2% wild-type current amplitude remaining) for the hypothesis that a channel formed from one mutant and three wild type subunits is active and expressed on the cell surface, whereas additional mutant subunits in the tetramer abolish activity (Minassian et al., 2012; Mock et al., 2010). (C) Incorporation of an iR4H mutant subunit results in a dominant shift in the voltage dependence of activation in the hyperpolarized direction. Normalized conductance for wild type expressed alone (black squares) or for wild type and iR4H expressed at a 1:1 ratio (red circles) has been plotted as a function of voltage. Data are provided as mean ± SEM. If error bars are not visible, they are smaller than the size of the symbol. Data were fitted with a Boltzmann function to estimate the midpoint voltage, V_0.5, and the slope factor. Values of V_0.5: wild type alone (n = 14), 13 ± 1 mV; 1:1 mixture (n = 14), 7 ± 1 mV. Values of slope factor: wild type alone (n = 14) 11 ± 1; 1:1 mixture (n = 14) 7 ± 0.3.

(A) Histograms show the percentage of cells with the indicated average firing frequencies in 2 Hz bins at different times postfertilization. Black bars, control cells. Red bars, iR4H-expressing cells. Data are the same as shown in Figure 1C. No recordings were made from control cells at the four dpf time point. (B) The fold-change in average firing frequency in iR4H-expressing cells compared to control cells is shown as a function of time postfertilization time. Data are provided as mean ± SEM.

Data are the same as shown in Figure 1C and Figure 2B.

Histograms show the percentage of cells with the indicated average firing frequencies in 2 Hz bins at different times postfertilization. Black bars, control cells. Red bars, aR3H-expressing cells. Data are the same as shown in Figure 2B.

Histograms show the percentage of cells with the indicated average firing frequencies in 2 Hz bins at different times postfertilization. Black bars, control cells. Red bars, aR3H-expressing cells. Data are the same as shown in Figure 2B.

(A) and B) Representative complex spikes recorded in control (A) and aR3H-expressing (B) Purkinje cells are shown (red dots). (C) Graph shows mean complex spike frequency ± SEM at 5.25 dpf in control Purkinje cells and cells expressing aR3H. Mean frequencies were 0.40 ± 0.05 Hz in control cells (n = 20) and 0.27 ± 0.06 Hz in aR3H cells (n = 3). Values do not differ significantly by unpaired t-test (p=0.32).

(A) A representative trace recorded at 5.25 dpf in a Purkinje cell expressing exoWT is shown. (B) Average frequency of simple spikes at 5.25 dpf was calculated from individual 10 s traces for control Purkinje cells and cells expressing exoWT. Each symbol represents a trace. Control data obtained at 5.25 dpf are the same as shown in Figure 1C and are repeated here for comparison to exoWT. For control cells, the mean firing frequency ± SEM was 7.7 ± 0.5 Hz (n = 117). For exo-WT-expressing cells, the mean firing frequency ± SEM was 4.8 ± 0.7 Hz (n = 16). (C) Average frequency of complex spiking is shown for control and exoWT-expressing Purkinje cells at 5.25 dpf. Values, provided as mean ± SEM, were 0.4 ± 0.05 Hz in control cells and 0.3 ± 0.1 Hz in exo-WT expressing cells. Control data obtained at 5.25 dpf are the same as shown in Figure 2—figure supplement 2C and are repeated here for comparison to exoWT.

At 5 dpf, live zebrafish were adapted to a LED light, which was turned off at time 0 s (Hsieh et al., 2014). (A) Representative recordings from a control Purkinje cell (left) and a Purkinje cell expressing aR3H (right) before and after turning off the LED are shown. Time 0 s is indicated by dashed vertical lines. (B) Raster plots show firing of simple spikes (black) and complex spikes (red) before and after turning off the LED at time 0 s (dashed vertical lines) for control cells (left) and aR3H-expressing cells (right). Each row represents a different trial. Data were obtained from 8 control cells in 5 animals and 8 aR3H-expressing cells in 4 animals. (C) The average frequency of tonic firing per 100 ms interval was normalized to the average frequency calculated over the 10 s period before turning the LED off at time 0 s (dashed vertical lines). The fold-change in frequency from all trials was averaged and plotted versus time for control (left) and aR3H-expressing (right) Purkinje cells (blue lines; n = 8 control cells, n = 8 aR3H-expressing cells). The gray shaded areas represent the SEM. The solid red horizontal line indicates the original firing frequency before the imposition of sudden darkness.

Live images of a Purkinje cell expressing exogenous wild type Kv3.3 were acquired during cerebellar development in vivo using a laser scanning confocal microscope at different times post-fertilization as indicated. Maximum intensity projections of stacks of 1 μm optical sections are shown. Scale bar: 10 μm.

Live images of a Purkinje cell expressing exogenous wild type Kv3.3 were acquired during cerebellar development in vivo using a laser scanning confocal microscope at different times post-fertilization as indicated. Maximum intensity projections of stacks of 1 μm optical sections are shown. Scale bar: 10 μm.

(A) The cartoon illustrates the stereotyped morphology of CaP motor neurons, which have a large cell body located dorsally in the spinal cord and an axon that grows into the ventral myotome where it branches and forms synapses on fast twitch muscle fibers (Myers et al., 1986; Westerfield et al., 1986). (B) EGFP (left column), or iR4H (middle column) or aR3H (right column) fusion proteins with EGFP (green) were co-expressed in motor neurons with a presynaptic marker, synaptophysin-mCherry (Syn-mCherry, red). Live images of CaP motor neurons were obtained at 48 hpf using a laser scanning confocal microscope. Representative maximum intensity projections are shown. Top row, EGFP fluorescence; middle row, mCherry fluorescence; bottom row, merged images. Images were traced for three-dimensional reconstruction and synapses were counted using Imaris software. (C–H) Panels show mean values ± SEM obtained from traced images for (C) synapse number; (D) distal branch number, where distal indicates below the midline (dashed line in cartoon shown in part A); (E) distal branch length; (F) length of the main axon shaft; (G) proximal branch number, where proximal indicates at or above the midline (dashed line shown in cartoon in part A); and (H) proximal branch length. Statistical significance was assessed by one-way ANOVA Kruskal-Wallis test, followed by Dunn’s multiple comparison test. *, p<0.05; **, p<0.01. EGFP, n = 12 neurons; iR4H, n = 33 neurons; aR3H, n = 19 neurons.

(A) The cartoon illustrates the stereotyped morphology of CaP motor neurons, which have a large cell body located dorsally in the spinal cord and an axon that grows into the ventral myotome where it branches and forms synapses on fast twitch muscle fibers (Myers et al., 1986; Westerfield et al., 1986). (B) EGFP (left column), or iR4H (middle column) or aR3H (right column) fusion proteins with EGFP (green) were co-expressed in motor neurons with a presynaptic marker, synaptophysin-mCherry (Syn-mCherry, red). Live images of CaP motor neurons were obtained at 48 hpf using a laser scanning confocal microscope. Representative maximum intensity projections are shown. Top row, EGFP fluorescence; middle row, mCherry fluorescence; bottom row, merged images. Images were traced for three-dimensional reconstruction and synapses were counted using Imaris software. (C–H) Panels show mean values ± SEM obtained from traced images for (C) synapse number; (D) distal branch number, where distal indicates below the midline (dashed line in cartoon shown in part A); (E) distal branch length; (F) length of the main axon shaft; (G) proximal branch number, where proximal indicates at or above the midline (dashed line shown in cartoon in part A); and (H) proximal branch length. Statistical significance was assessed by one-way ANOVA Kruskal-Wallis test, followed by Dunn’s multiple comparison test. *, p<0.05; **, p<0.01. EGFP, n = 12 neurons; iR4H, n = 33 neurons; aR3H, n = 19 neurons.

Representative confocal image stacks of CaP neurons expressing EGFP, or EGFP fusion proteins of iR4H or aR3H were traced using Imaris software. Projected traces are shown from sagittal and transverse perspectives. Red dots indicate presynaptic specializations labeled by a co-expressed synaptophysin-mCherry fusion protein.

(A) A live image of a Purkinje cell expressing iR4H and membrane-bound tdTomato (mTomato) is shown at 4 dfp. Scale bar = 5 μm. (B) At 5 dpf, the zebrafish was incubated in acridine orange and the cell was re-imaged. Left, mTomato; middle, acridine orange; right, merged image. In apoptotic cells, acridine orange appears as green puncta, which corresponds to the dye binding to condensed, fragmented chromatin (arrows) (Atale et al., 2014). Less intense green staining is present in nearby healthy cells that are not expressing iR4H and do not contain condensed, fragmented chromatin (Atale et al., 2014). Scale bar: 5 μm.

(A) A live image of a Purkinje cell expressing iR4H and membrane-bound tdTomato (mTomato) is shown at 4 dfp. Scale bar = 5 μm. (B) At 5 dpf, the zebrafish was incubated in acridine orange and the cell was re-imaged. Left, mTomato; middle, acridine orange; right, merged image. In apoptotic cells, acridine orange appears as green puncta, which corresponds to the dye binding to condensed, fragmented chromatin (arrows) (Atale et al., 2014). Less intense green staining is present in nearby healthy cells that are not expressing iR4H and do not contain condensed, fragmented chromatin (Atale et al., 2014). Scale bar: 5 μm.