The tail current induced by Ca²⁺ channel activation in cultured hippocampal pyramidal neurons is a Cl⁻ current. (A) Depolarization (at room temperature for A-C) to 0 mV for 200 ms results in Ca²⁺ current followed by a tail current that reverses at E_Cl. (B) The reversal potential follows E_Cl: black, 23 mM [Cl⁻]_in(E _Cl= −46.8 mV), red, 10 mM [Cl⁻]_in(E _Cl= −67.8 mV), blue, 5 mM [Cl⁻]_in(E _Cl= −85.3 mV). (C) The average reversal potential is −45.1 ± 0.9 mV at 23 mM [Cl⁻]_in (n = 14), −67.2 ± 1.1 mV at 10 mM [Cl⁻]_in (n = 5), and −83.5 ± 1.4 mV at 5 mM [Cl⁻]_in (n = 6). (D) Depolarizing neurons (in acute slices at 35°C with 2.5 mM Ca²⁺) to 0 mV for 1 ms results in Ca²⁺ current followed by a tail current that reverses at E_Cl as in the case of cultured neurons: −45 ± 0.5 mV (n = 25 for 10 recordings from cultured neurons and 15 from acute slices) (E_Cl= −46.8 mV). The tail (left) is sensitive to the CaCC blocker NFA (300 μM) (middle). The dashed line in A and D indicates zero current. See Supplemental Figure 1A for time course plot.

Ca²⁺-activated Cl⁻ current in hippocampal pyramidal neurons is sensitive to two different CaCC blockers and is important for controlling action potential duration. (A) CaCC is significantly reduced to 28 ± 5.1% of control by 100 μM NFA (top left, n = 9, p < 0.001) and to 18 ± 5.9% of control by 100 μM NPPB (top right, n = 10, p < 0.001). IC50 is 33 ± 2.3 μM for NFA (bottom left, n = 5), and 26.2 ± 1.36 μM for NPPB (bottom right, sequentially increasing blocker concentration at room temperature, n = 5). The membrane voltage was held at −70 mV, depolarized to 0 mV to induce Ca²⁺ influx and repolarized back to −90 mV to elicit Ca²⁺-activated Cl⁻ current. * p < 0.01. See Supplemental Figure 1D and 1E for time course plots. (B) Reducing CaCC with 100 μM NFA (red) causes the spike width to increase by ~30% compared to action potentials prior to NFA application (control, black) and after washout of NFA (blue) (top panel). A single action potential is elicited by injecting 2-ms current barely reaching threshold for neurons in acute slices at 35°C in B and C. Bottom panel shows the dose response curve for NFA, which was applied at increasing concentrations sequentially (n = 5). The dose for half maximal spike widening (measured at 1/3 total spike height) is 55.08 ± 5.72 μM NFA; the maximal spike widening corresponds to an increase of the spike width by ~65%. * p < 0.05. See Supplemental Figure 1F for time course plot and experimental details and Supplemental Figure 4 for CaMKII inhibitor control. (C) Apply 100 μM NPPB caused the spike width to increase by ~30% (control, black; NPPB, red; washout, blue) (top panel). Bottom panel shows the dose response curve for NPPB (n = 5). The dose for half maximal spike widening (measured at 1/3 total spike height) is 48.07 ± 10.22 μM; the maximal spike widening corresponds to an increase of the spike width by ~70%. * p < 0.05. See Supplemental Figure 1G for time course plot and experimental details.

CaCC shortens the duration of action potentials recorded in CA3 pyramidal neurons, but does not affect transmitter release from their axon terminals upon stimulation of Schaffer collaterals (acute slices, 35°C). (A–B)Reducing CaCC in CA3 neurons in acute hippocampal slices with 100 μM NFA (red) slows the action potential repolarization compared to action potentials before applying NFA (control, black) and after washout of NFA (blue) (A). (B) NFA increases CA3 spike duration. Control, black, 0.99 ± 0.04 ms; NFA, red, 1.41 ± 0.036 ms; n = 10, * p < 0.01. See Supplemental Figure 1H for time course plot and experimental details, Supplemental Figure 2 and 4 for controls of the NFA effect. (C) In the presence of 100 μM APV, the AMPA-fEPSP (black) induced by 10 nerve stimuli at 10 Hz was not altered by 100 μM NFA (red) (n = 7, p = 0.8). Blue, 20 μM CNQX applied at the end of the experiment. The 6^th response is shown in each case. See Supplemental Figure 1R for time course plot. (D) At the holding potential of −65 mV (E_Cl = −64.4 mV with 10 mM [Cl⁻]_in), the NMDA-EPSC (black) was not altered by 100 μM NFA (red) (n = 5, p = 0.2). Blue, 100 μM APV applied at the end of the experiment (recorded in the presence of CNQX). See Supplemental Figure 1S for time course plot. (E–H) When CaCC contribution is maximized by increasing the driving force for Cl⁻(E _Cl ~ 0 mV with 130 mM [Cl⁻]_in, holding potential at −65 mV) and removing Mg²⁺ block of NMDA-Rs by using a zero-Mg²⁺/4 mM Ca²⁺ external solution, 100 μM NFA (red) reduces NMDA-EPSC (black) by 28 ± 4.3% (n = 10, p < 0.05) (E). Blue, 100 μM APV applied at the end of the experiment. See Supplemental Figure 1T for time course plot. (F) Chelating internal Ca²⁺ by including 10 mM BAPTA in a 130 mM [Cl⁻]_in pipette solution eliminates the effect of 100 μM NFA on NMDA-EPSC (black, 20 μM CNQX; red, 100 μM NFA + 20 μM CNQX; blue, 100 μM APV + 20 μM CNQX; n = 10, p = 0.13). See Supplemental Figure 1U for time course plot. (G) With 10 mM EGTA in a 130 mM [Cl⁻]_in pipette solution, 100 μM NFA (red) reduces NMDA-EPSC (black) by 32 ± 9% (n = 5, p < 0.01). Blue, 100 μM APV applied at the end of the experiment. See Supplemental Figure 1V for time course plot. (H) CaCC contributes to NMDA-EPSC when there is maximal driving force for Cl⁻ (Figure 5E, n = 10, p < 0.05) and when there is Ca²⁺ influx through NMDA-Rs (Figure 5G, n = 5, p < 0.01). Unlike BAPTA that quickly chelates Ca²⁺ (Figure 5F, n = 10), the slow Ca²⁺ chelator EGTA cannot eliminate CaCC modulation of NMDA-EPSC (Figure 5G, n = 5). (I) AMPA-EPSPs recorded from CA1 neurons (black) by stimulating axons of CA3 neurons are not affected by the NFA block of CaCCs (red). These traces are superimposed with the AMPA-EPSP recorded after washout of NFA (blue). See Supplemental Figure 1W for time course plot. (J) Blocking CaCC with NFA had no effect on pharmacologically isolated AMPA-EPSPs regardless of stimulation intensities or AMPA-EPSP amplitudes (n = 11, p > 0.05). See Supplemental Figure 5 for NFA effect on pharmacologically isolated NMDA-fEPSP.

The tail current recorded at room temperature from cultured hippocampal pyramidal neurons is a Ca²⁺-activated Cl⁻ current. (A) 100 μM CdCl₂ eliminates tail current (left) by blocking Ca²⁺ current (right top, red). The Cl⁻ tail current requires Ca²⁺ for activation (black) (right bottom)(n = 10, p < 0.01). See Supplemental Figure 1B for time course plot. (B) 10 mM BAPTA in the pipette solution left Ca²⁺ current intact (right top, red) but eliminated tail current (left) (right bottom)(red, n = 12, p < 0.01).(C) With external 2.5 mM Ca²⁺ replaced by 2.5 mM Ba²⁺, Ba²⁺ ions that carry the inward current (right top, red) cannot activate the Cl⁻ tail current (right bottom)(n = 13, p < 0.01). For A–C, scale bar for right top panel is 100 ms, 200 pA, and arrow points to the 150–200 ms time window where the average tail current measurement is taken. See Supplemental Figure 1C for time course plot. (D) Tail current diminishes in amplitude as prepulse voltage approaches E_Ca (> + 100 mV) (n = 52). (E) Tail current increases with increasing duration of prepulse to 0 mV (n = 22, p < 0.01) but not + 100 mV (n = 22, p = 0.9).

TMEM16B is expressed in hippocampal pyramidal neurons and is important for the control of action potential duration and the generation of CaCC. (A) RT-PCR reveals TMEM16B (top) but not TMEM16A (middle) expression in neurons 8, 14, and 20 days in vitro (DIV). β-actin is included as loading control (bottom). (B) Quantitative RT-PCR revealed that 16B-shRNAs #2 (left) and #5 (right) reduced the expression of TMEM16B mRNA to 54 ± 1.0% (n = 3) and 37 ± 5.5% (n = 3) of scrambled shRNA control respectively. * p < 0.05. See Supplemental Figure 3C for additional control for knockdown of TMEM16B expression by 16B-shRNA #2 and #5. (C) CaCC tail current from cultured hippocampal pyramidal neurons at room temperature is significantly reduced (by 67.5%) by TMEM16B shRNA knockdown (red), but not by scrambled RNA control (black). Inset (top left): TMEM16B knockdown does not affect the Ca²⁺ current. (D) Ca²⁺-activated Cl⁻ current is reduced by 60 ± 4.5% and 61 ± 5% by TMEM16B shRNA #2 and #5, respectively (scrambled, n = 17, shRNA #2, n = 18, p < 0.01, shRNA #5, n = 12, p < 0.01, 12 days after infection). (E) Left, TMEM16B shRNA #2 knockdown broadens spike waveform (red) as compared to scrambled RNA control (black) (12 days after infection). Right, summary of the effect of TMEM16B-shRNA #2 and #5 on action potential duration (resting potential ~65 mV). (F) In neurons recorded (with ~140 mM Cl⁻ on both sides) at 35°C 10 days after infection with lentivirus for scrambled shRNA (top panels), raising intracellular Ca²⁺ from 0 (left) to 0.5 μM Ca²⁺(right) gives rise to CaCC that is blocked by 300 μM NFA. This CaCC current was greatly reduced in neurons infected with shRNA #5 (middle panels). Current-voltage curves (bottom panels) plotted for current recorded by applying voltage steps from −80 to +80 mV (10 mV intervals) from holding potential of 0 mV (E_Cl = 0) without intracellular Ca²⁺ (left, bottom) and with 0.5 μM intracellular Ca²⁺ (right, bottom), show that 16B-shRNA #2 and #5 reduced CaCC in hippocampal neurons by ~80% at positive voltages. *p < 0.01. (G) TMEM16B mRNA expression in mouse dentate granule cells, hilar interneurons, CA1 pyramidal neurons (top) and CA3 pyramidal neurons (bottom) as revealed by in situ hybridization. (H) TMEM16B protein is detected in cultured hippocampal neurons at 13, 15, and 17 DIV by western blotting. * indicates a non-specific band. Bottom panel is the neuron-specific β-tubulin as a loading control. See Supplemental Figure 3B for antibody specificity controls.

CaCC activation by Ca²⁺ influx through NMDA-R provides a brake to excitatory synaptic responses, temporal summation, and EPSP-spike coupling (acute slices, 35°C). (A) 100 μM NFA (red) increases the amplitude of large (6 mV) EPSP (right) to 147 ± 2.9% of control, n = 10, p < 0.05) (Supplemental 1X) without affecting small (left, n = 10, p > 0.05) excitatory synaptic potentials. (B) Reducing CaCC with 100 μM NFA amplifies the larger EPSPs under physiological conditions (left, n = 10, p < 0.05), but dampens the larger EPSPs in the presence of 130 mM [Cl⁻]_in (middle, n = 9, p < 0.05). 100 μM NFA has no effect on EPSP amplitude when BAPTA is included with 10 mM [Cl⁻]_in to chelate Ca²⁺ (right, n = 16, p > 0.05). The input-output relations summarize averages of ten recordings in each condition. * p < 0.05. See Supplemental Figure 1X–Z for time course plots, Table 1 and Supplemental Information for experimental details. (C) A burst of three nerve stimuli at 40 Hz results in EPSPs that summate and grow progressively. Reducing CaCC with 100 μM NFA (red) enhances the summation efficacy (the ratio of third EPSP over first EPSP) while leaving the first EPSP (EPSPs ≤ 2 mV) unchanged (black, control prior to NFA application). Blue, 20 μM CNQX and 100 μM APV application at the end of the experiment. (D) Reducing CaCC with 100 μM NFA enhances summation and increases the 3^rd/1^st EPSP ratio under physiological conditions (left, n =8, p < 0.01), but has the opposite effect in 130 mM [Cl⁻]_in (middle, n =9, p < 0.01). 100 μM NFA has no effect on EPSP summation when BAPTA is included with 10 mM [Cl⁻]_in to chelate Ca²⁺ (right, n =5 p > 0.05). See Supplemental Figure 1A′-C′ for time course plots, Table 1 and Supplemental Information for experimental details. (E) A burst of five nerve stimuli at 40 Hz generates EPSPs that summate to reach the threshold after (red) but not before (black) application of 100 μM NFA. Blue, 20 μM CNQX and 100 μM APV application at the end of the experiment. Five superimposed recordings are shown. (F) Reducing CaCC with 100 μM NFA enhances EPSP-spike coupling under physiological conditions (left, n = 7, p < 0.01), but dampens EPSP-spike coupling in 130 mM [Cl⁻]_in (middle, n = 8, p < 0.01). 100 μM NFA has no effect on EPSP-spike coupling when BAPTA is included with 10 mM [Cl⁻]_in to chelate Ca²⁺ (right, n = 5, p > 0.05). See Supplemental Figure 1D′–F′ for time course plots and Supplemental Figure 6 for NFA effect on pharmacologically isolated NMDA-EPSP spike coupling.