All animal work was performed at the University of Sheffield (United Kingdom), licensed by the Home Office under the Animals (Scientific Procedures) Act 1986 (PPL_PCC8E5E93) and approved by the University of Sheffield Ethical Review Committee (180626_Mar). For all in vitro work, mice were culled by cervical dislocation, which is a schedule 1 method.

Experiments were performed using Ca_V1.3 knockout mice (Ca_V1.3^–/–) on a C57BL/6N background and control mice (littermate heterozygous or C57BL/6N). Mice from both sexes were used and ranging from postnatal day 4 (P4) to P35. The semicircular canals with their ampullae and cochleae (apical turn) were dissected out from the inner ear as reported previously (Jeng et al., 2020a,2021; Spaiardi et al., 2020a; Carlton et al., 2021), using an extracellular solution composed of (in mM) 135 NaCl, 5.8 KCl, 1.3 CaCl₂, 0.9 MgCl₂, 0.7 NaH₂PO₄, 5.6 D-glucose, and 10 HEPES-NaOH. Sodium pyruvate (2 mM), amino acids, and vitamins were added from concentrates (Thermo Fisher Scientific, United Kingdom). The pH was adjusted to 7.5 (osmolality ∼308 mmol kg^–1). The dissected organs were fixed at the bottom of the recording chamber by a nylon-meshed silver ring and were continuously perfused with the above extracellular solution (0.5 ml/min) using a peristaltic pump (Masterflex L/S, Cole Palmer, United States). Hair cells were viewed using a upright microscopes (Olympus BX51; Leica DM-LFS) equipped with Nomarski Differential Interface Contrast (DIC) optics with a 60X or 64X water immersion objective and x15 eyepieces.

Voltage-clamp whole-cell experiments were performed at room temperature (18–22°C) for K⁺ current recordings, and near-body temperature (32–36°C) for Ca²⁺ currents recordings using an Optopatch amplifier (Cairn Research Ltd., United Kingdom) as previously described (Johnson and Marcotti, 2008; Jeng et al., 2020b; Spaiardi et al., 2020b). The patch pipettes were pulled to 2–3 MΩ tip resistance from soda glass capillaries (Hilgenberg, Germany) and coated with surf-wax (Mr. Zoggs SexWax, United States) to minimize the fast capacitance transient across the wall of the patch pipette. For K⁺ current recordings, the patch pipette filling solution contained (in mM): KCl 131, Na₂-Phosphocreatine 10, MgCl₂ 3, EGTA-KOH 1, Na₂ATP 5, and HEPES 5; pH adjusted to 7.28 with KOH (osmolality was 294 mmol kg^–1). For Ca²⁺ current recordings, the pipette intracellular solution contained (in mM): 106 Cs-glutamate, 20 CsCl, 3 MgCl₂, 1 EGTA-CsOH, 5 Na₂ATP, 0.3 Na₂GTP, 5 HEPES-CsOH, 10 Na₂-phosphocreatine, pH 7.3 with CsOH (294 mmol kg^–1). Data acquisition was controlled by pClamp software using a Digidata board (Molecular Devices, United States). Recordings were low-pass filtered at either 2.5 or 5 kHz (8-pole Bessel), sampled at 5, 10, or 100 kHz and stored on computer for off-line analysis using Clampfit (Molecular Devices, United States) and Origin (OriginLab, United States) software. Membrane potentials reported in the text and figures were corrected for the uncompensated residual series resistance (R_s) and the liquid junction potential (LJP), which was either −4 mV for the K⁺-based and −11 mV for the Cs⁺-glutamate-based intracellular solution, measured between electrode and bath solutions. For K⁺ current recordings from vestibular hair cells, R_s was calculated off-line from the capacitive artifact elicited by applying a voltage step from either −74 to −64 mV in type II and −124 to −44 mV in type I hair cells. The different voltage step was used in order to minimize artifact contamination by inward and outward rectifier voltage-gated channels in the two hair cell types (Spaiardi et al., 2017). Voltage clamp protocols are referred to a holding potential of −64 mV for the K⁺-based intracellular solution or −91 mV for the Cs⁺-based intracellular solution.

For Ca²⁺ current recordings, the composition of the extracellular solution contained (in mM): NaCl 101, CaCl₂ 5, CsCl 5.8, MgCl₂ 0.9, HEPES 10, glucose 5.6, tetraethylammonium (TEA) 30, 4-aminopyridine (4AP) 15 (pH adjusted with HCl was 7.5; osmolality: 312 mOsm/kg). The higher Ca²⁺ concentration (5 mM) was used to better visualize the Ca²⁺ current in Ca_V1.3^–/– mice. Extracellular TEA, 4AP, and intracellular Cs⁺ (see above) were used to block the K⁺ channels (cochlear IHCs: Marcotti et al., 2003; Jeng et al., 2020b,2021; vestibular hair cells: Rennie and Correia, 1994; Biel et al., 2009). Moreover, the K⁺ channel blockers linopirdine (80 μM; Tocris, United Kingdom) was also used to block I_K,n in adult IHCs (Marcotti et al., 2003). In some experiments, CdCl₂ 0.1 mM was also added to the above extracellular solution to block the Ca²⁺ current (Hille, 2001) in vestibular hair cells. The amplitude of the Ca²⁺ current was measured by either subtracting the linear leakage current component, measured between −81 and −91 mV, or the current blocked by Cd²⁺ from the current recorded in the presence of TEA and 4AP (see above).

Statistical comparisons of means were made by Student’s two-tailed t-test or, for multiple comparisons, analysis of variance (one-way or two-way ANOVA) was applied. P < 0.05 was selected as the criterion for statistical significance. Mean values are quoted as means ± SD.

Type I hair cells of the ampullae sensory epithelium (the crista) exhibit a large low-voltage activated outward rectifying K⁺ currents, named I_K,L (Rennie and Correia, 1994; Rüsch and Eatock, 1996), which was recorded in both adult wild-type (Figure 1A) and Ca_V1.3^–/– mice (Figure 1B). Since I_K,L is almost completely activated at −60 mV, hyperpolarizing voltages from the holding potential of −64 mV produced deactivating tail currents (e.g., −104 and −124 mV: Figures 1A,B), while depolarizations elicit an instantaneous increase of the outward currents. The mean steady-state current-voltage (I-V) relationship for the total current recorded in type I hair cells was not significantly different between wild-type (P17-P22; n = 12) and Ca_V1.3^–/– mice (P19-P20; n = 6) (P = 0.9742, F = 0.4149, DFn = 15, two-way ANOVA, Figure 1C). The negligible steady-state current at −124 mV is consistent with I_K,L being fully deactivated at this potential, and with the absence of inward rectifier currents in mouse crista type I hair cells (Spaiardi et al., 2020a). Note that the outward tail currents in type I hair cells, which were recorded at −44 mV, were smaller following the voltage step to −24 mV than those at −44 mV in both wild-type (Figures 1A,D) and Ca_V1.3^–/– mice (Figures 1B,E,F). The presence of progressively smaller outward tail current following larger outward K⁺ currents (Figures 1D,E), which in some cases includes tail current reversal (Figure 1F), is consistent with intercellular (e.g., in the calyceal synaptic cleft) K⁺ accumulation inducing a shift in the K⁺ equilibrium potential (Lim et al., 2011; Contini et al., 2012, 2017, 2020; Spaiardi et al., 2020a). Progressive intercellular K⁺ accumulation during depolarization is also responsible for the “apparent” inactivation of the outward current, which is due to the progressive decrease of the driving force for K⁺ to exit the hair cells (e.g., Spaiardi et al., 2017).

Different from the type I hair cells, the macroscopic current recorded from adult type II hair cells is not dominated by I_K,L (Figures 1G,H), but instead express several other currents that have been previously described in great details (Meredith and Rennie, 2016). This include a transient (I_K,A) and a delayed rectifying (I_K,v) outward K⁺ current, an inward rectifying K⁺ current (I_K,1), and the mixed inward rectifying Na⁺/K⁺ current (I_h). All these currents were evident from the inward and outward current profile recorded from type II hair cells of Ca_V1.3^–/– mice (Figure 1H). The mean steady-state I-V for the total current recorded in type II hair cells was also not significantly different between wild-type (P17-P22; n = 4) and Ca_V1.3^–/– mice (P18-P19; n = 4) (P = 0.7010, F = 0.7756, DFn = 15, two-way ANOVA, Figure 1I).

The above results demonstrate that the absence of the Ca_V1.3 Ca²⁺ channel subunit does not impair the normal developmental acquisition of voltage-dependent K⁺ currents in vestibular type I and II hair cells in the crista. This is different from the cochlea, where the normal expression of the K⁺ currents characteristic of adult inner hair cells (IHCs), which are the fast activating BK current I_K,f and negatively activating delayed rectifier I_K,n (Kros et al., 1998; Marcotti et al., 2003; Oliver et al., 2003), was either prevented (apical-coil) or reduced (basal-coil) in Ca_V1.3^–/– mice (Brandt et al., 2003; Jeng et al., 2020a).

Although type I and type II hair cells are morphologically distinct, this is not always visible when working with the intact organ. Therefore, the correct identification of type I hair cells relies on the presence of I_K,L (Figure 1). Recordings were performed using intracellular Cs⁺, which is known to block I_h (Biel et al., 2009) and the outward rectifying K⁺ currents (Bao et al., 2003) in type II hair cells while having little effect on I_K,L (Griguer et al., 1993; Rüsch and Eatock, 1996; Chen and Eatock, 2000; Rennie and Correia, 2000). An additional consideration affecting the correct identification of I_K,L is its state of deactivation at the holding potential used for the recordings, which is generally set at −90 mV. Figures 2A,C shows representative current responses recorded from a type I hair cell with a largely deactivated I_K,L, as indicated by the small inward current at −91 mV, and the small instantaneous current upon depolarizing and hyperpolarizing voltage steps. In the example shown in Figures 2B,C, however, I_K,L is largely activated at −91 mV and exhibits large instantaneous currents. Although this variability has previously been reported (Hurley et al., 2006; Spaiardi et al., 2017), we found that the reversal potential for the macroscopic current was significantly more hyperpolarised in type I hair cells showing a largely activated I_K,L at −91 mV (Figure 2C). While a deactivated I_K,L at the holding potential was primarily recorded in early postnatal hair cells (n = 15), more mature cells tend to exhibit a largely activated current (n = 8) (Figure 2D). Since the reversal potential of the macroscopic K⁺ current in type I hair cells is primarily determined by the ions flowing through the K,L channels, it should reach values near the reversal potential of the mixed Cs⁺/K⁺ current (called a “biionic potential”; see Hille, 2001) through K,L channels in the presence of a large I_K,L and intracellular Cs⁺ (Spaiardi et al., 2020a). Given the reported permeability ratio of Cs⁺ to K⁺ of about 0.31 (Rüsch and Eatock, 1996; Spaiardi et al., 2020a), the reversal potential should be close to −40 mV. The finding that the reversal potential in older hair cells (Figure 2C) is close to −60 mV could indicate that the relative permeability of K,L channels to Cs⁺ is likely to be larger than that previously estimated, at least in more mature type I hair cells. However, the presence of a residual calyx might also produce a shift of the mixed Cs⁺/K⁺ equilibrium potential toward more negative voltages during inward currents (Spaiardi et al., 2017, 2020a). This would imply a tighter attachment of the calyx to more mature type I hair cells.

The above results are consistent with I_K,L increasing in amplitude and activating at more hyperpolarized potentials during post-natal development, as also previously shown in rat type I hair cells (Hurley et al., 2006). Different K⁺ channels subunits have also been found to be expressed by rodent type I hair cells during postnatal development which may account for the above changes (Hurley et al., 2006; Spitzmaul et al., 2013).

Following the positive identification of the patched hair cell as type I, a small inward current (Figure 3A, upper panel) became visible when superfusing the cells with an extracellular solution containing the K⁺ channel blockers TEA and 4AP, in the presence of 5 mM Ca²⁺ (see section “Materials and Methods”). In a subset of type I hair cells (n = 8), we found that the addition of 0.1 mM Cd²⁺ to the extracellular solution fully blocked the inward current, confirming its identity as I_Ca (Figure 3A, lower panel). The isolated I_Ca was obtained by subtracting the current recorded in the presence of TEA, 4AP, and Cd²⁺ to that obtained in the absence of Cd²⁺ (Figure 3B). Under this experimental condition, the inward current activated at about −61 mV and peaked near −21 mV (Figures 3B,D). Comparable results were obtained when I_Ca was isolated by performing the leakage-subtraction procedure (see section “Materials and Methods”) to the currents recorded in TEA and 4AP (Figures 3C,D).

In Ca_V1.3^–/– mice, the inward current was barely visible in type I hair cells when the extracellular solution contained TEA, 4AP, and 5 mM Ca²⁺ (Figure 3E). Following the isolation of I_Ca with Cd²⁺ (Figure 3F), or after leakage subtraction (Figure 3G), it became more easily detectable. The inward current peaked near −11 mV (Figure 3H). In type I hair cells, the mean peak I_Ca after Cd²⁺-subtraction was 121 pA in wild-type and 19 pA in Ca_V1.3^–/– mice (15.7% of the wilt-type value).

For type II hair cells, the isolation of the Ca²⁺ current was performed as shown in Figure 3, using either the isolated Cd²⁺-procedure (Figures 4A,D) or after leakage subtraction (Figures 4B,E) in both wild-type and Ca_V1.3^–/– mice, respectively. In WT type II hair cells the amplitude of the Cd²⁺-sensitive current (I_Ca: Figures 4A,C) was not significantly different to that obtained by leakage-subtraction (Figures 4B,C: P < 0.9336, F = 0.3916, DFn = 9, two-way ANOVA), or to the Cd²⁺-sensitive current measured in wild-type type I hair cells (Figures 3B,D: P < 0.8457, two-way ANOVA). The size of the Cd²⁺-sensitive I_Ca recorded in type II hair cells of Ca_V1.3^–/– mice (Figures 4D,F) was very similar to that measured in type I hair cells (Figures 3F,H, P < 0.7677, F = 0.6319, DFn = 9, two-way ANOVA). In type II hair cells, the mean peak I_Ca after Cd²⁺-subtraction was 142 pA in wild-type and 27 pA in Ca_V1.3^–/– mice (19.0% of the wild-type value).

We tested whether the size of I_Ca changed with age in hair cells from wild-type and Ca_V1.3^–/– mice. The peak I_Ca amplitude, which was detected between −21 and −10 mV, was plotted as a function of postnatal day (Figure 5). The size of I_Ca in both type I and type II hair cells appears to be comparable between early and late postnatal ages in both wild-type and Ca_V1.3^–/– mice.

We investigated I_Ca in cochlear IHCs by performing whole-cells recordings at 32–36 °C and with the same solutions used for the above vestibular hair cell recordings (intracellular Cs⁺, extracellular TEA and 4AP, and 5 mM Ca²⁺ (Figure 6). About 10% of the total I_Ca in apical IHCs has been shown to be carried by Ca²⁺ channels other than Ca_V1.3 (Platzer et al., 2000; Michna et al., 2003). Indeed, we found that the size of I_Ca, measured after leakage-subtraction, was largely reduced in IHCs from Ca_V1.3^–/– mice compared to littermate controls at both immature (P < 0.0001, F = 63.70, DFn = 9) and adult ages (Figure 6G, P < 0.0001, F = 20.33, DFn = 9, between the same voltage range used for the vestibular hair cells: −61 to +29 mV, two-way ANOVA). The nature of the residual Ca² channels in immature (∼16% of the wild-type value) and adult (∼12%) IHCs is unknown, but previous suggestions included Ca_V1.4 (Brandt et al., 2003) or, in immature avian auditory hair cells, Ca_V3.1 T-type subunits (Levic and Dulon, 2012).