Ca_v1.4α1 Subunits Can Form Slowly Inactivating Dihydropyridine-Sensitive L-Type Ca²⁺ Channels Lacking Ca²⁺-Dependent Inactivation

Abstract

The neuronal L-type calcium channels (LTCCs) Ca_v1.2α1 and Ca_v1.3α1 are functionally distinct. Ca_v1.3α1 activates at lower voltages and inactivates more slowly than Ca_v1.2α1, making it suitable to support sustained L-type Ca²⁺ inward currents (I_Ca,L) and serve in pacemaker functions. We compared the biophysical and pharmacological properties of human retinal Ca_v1.4α1 using the whole-cell patch-clamp technique after heterologous expression in tsA-201 cells with other L-type α1 subunits. Ca_v1.4α1-mediated inward Ba²⁺ currents (I_Ba) required the coexpression of α2δ1 and β3 or β2a subunits and were detected in a lower proportion of transfected cells than Ca_v1.3α1. I_Ba activated at more negative voltages (5% activation threshold; -39mV; 15 mm Ba²⁺) than Ca_v1.2α1 and slightly more positive than Ca_v1.3α1. Voltage-dependent inactivation of I_Ba was slower than for Ca_v1.2α1 and Ca_v1.3α1(∼50% inactivation after 5 sec; α2δ1 + β3 coexpression). Inactivation was not increased with Ca²⁺ as the charge carrier, indicating the absence of Ca²⁺-dependent inactivation. Ca_v1.4α1 exhibited voltage-dependent, G-protein-independent facilitation by strong depolarizing pulses. The dihydropyridine (DHP)-antagonist isradipine blocked Ca_v1.4α1 with ∼15-fold lower sensitivity than Ca_v1.2α1 and in a voltage-dependent manner. Strong stimulation by the DHP BayK 8644 was found despite the substitution of an otherwise L-type channel-specific tyrosine residue in position 1414 (repeat IVS6) by a phenylalanine. Ca_v1.4α1 + α2δ1 + β channel complexes can form LTCCs with intermediate DHP antagonist sensitivity lacking Ca²⁺-dependent inactivation. Their biophysical properties should enable them to contribute to sustained I_Ca,L at negative potentials, such as required for tonic neurotransmitter release in sensory cells and plateau potentials in spiking neurons.

Introduction

Many cellular functions are controlled by a depolarization-induced influx of Ca²⁺ ions from the extracellular space through voltage-gated Ca²⁺ channels. In neurons, Ca²⁺ influx through presynaptic N-, P/Q-, and, to a limited extent, R-type Ca²⁺ channels is tightly coupled to neurotransmitter release from nerve terminals (stimulus-secretion coupling) (Catterall, 2000). L-type Ca²⁺ channels (LTCCs) are primarily targeted to dendrites and the cell soma (Catterall, 2000) and are responsible for Ca²⁺ signals, which activate signaling pathways controlling gene transcription (Graef et al., 1999).

We (Koschak et al., 2001), and others (for review, see Lipscombe, 2002), have recently found that Ca_v1.2α1 and Ca_v1.3α1 LTCCs possess distinct functional properties that thus allow them to serve distinct neuronal functions. Ca_v1.3α1 channels activate at more negative voltages, inactivate slower during depolarizing pulses, and exhibit lower dihydropyridine (DHP) antagonist sensitivity than Ca_v1.2α1. Their biophysical properties make them highly suitable to mediate tonic neurotransmitter release in sensory cells (such as in cochlear inner hair cells) (Platzer et al., 2000) to support plateau potentials in spiking neurons (Carlin et al., 2000; Alaburda et al., 2002; Morisset and Nagy, 2000) and contribute to diastolic depolarization and pacemaking in the sinoatrial node (Mangoni et al., 2003) (for review, see Lipscombe, 2002). In the mammalian retina, Ca_v1.3α1 subunits are expressed in photoreceptor nerve terminals and selected bipolar cell synapses (Morgans et al., 1998; Taylor and Morgans, 1998; Morgans, 1999). I_Ca,L in photoreceptors and bipolar cells shares most features of heterologously expressed Ca_v1.3α1 currents. It was therefore proposed that this channel underlies retinal I_Ca,L (Wilkinson and Barnes, 1996).

Recently, Ca_v1.4α1 subunits were discovered as a putative neuronal LTCC subunit (Bech-Hansen et al., 1998; Strom et al., 1998). Ca_v1.4α1 is expressed predominantly in the retina but also in other neurons such as dorsal root ganglia (Murakami et al., 2001). In the mammalian retina, its expression pattern resembles Ca_v1.3α1. Ca_v1.4α1 immunoreactivity has also been localized in the synapses of the outer and inner plexiform layer as well as on photoreceptor cell bodies (Firth et al., 2001; Morgans, 2001; Morgans et al., 2001; Ball et al., 2002; Berntson et al., 2003). Its physiological relevance for normal retinal function is evident from Ca_v1.4 α1 mutations causing incomplete X-linked congenital stationary night blindness (iCSNB2) in humans (Bech-Hansen et al., 1998; Strom et al., 1998). Ca_v1.4α1 could therefore also represent a synaptically localized Ca²⁺ channel in the retina. However, this interpretation is complicated by the fact that Ca_v1.4α1 is the only cloned mammalian Ca²⁺ channel α1 subunit that has not yet been functionally characterized. Its characterization in isolated neurons is hampered by the lack of Ca_v1.4α1-deficient mouse models. These would allow the identification of neurons, such as photoreceptors, with predominantly Ca_v1.4α1-mediated currents and isolate them from the residual Ca²⁺ current components. In the absence of such models, the biophysical and pharmacological characterization of recombinant Ca_v1.4α1 channels would allow us to address many open questions: can Ca_v1.4α1 form DHP-sensitive LTCCs and I_Ca,L as described in retinal neurons? Do its biophysical properties resemble Ca_v1.2α1 or rather the lower voltage-activated Ca_v1.3α1? Do Ca_v1.4α1 currents exhibit Ca²⁺-dependent inactivation? Ca_v1.4α1 subunits lack a tyrosine in transmembrane segment IVS6, which was found previously to be part of the DHP-binding pocket of other LTCCs (Peterson et al., 1996; Striessnig et al., 1998). This raises the question of whether Ca_v1.4α1 exhibits the typical DHP sensitivity by which retinal I_Ca,L has been defined.

Here we describe for the first time the successful functional expression of a human retinal Ca_v1.4α1 in mammalian cells. We show that Ca_v1.4α1 channels share many of the properties of Ca_v1.3α1, including intermediate DHP sensitivity, but lack Ca²⁺-dependent inactivation under identical experimental conditions.

Materials and Methods

Cloning of human Ca_V1.4α1 subunits. The Ca_v1.4α1 cDNA (Strom et al., 1998) (GenBank AJ224874; open reading frame length, 5898 bp) was cloned from five subfragments (F1–F5) using different native or artificial restriction enzyme (RE) sites [nucleotide numbers (nt) are given in parentheses; asterisks indicate artificial RE sites introduced by PCR]: F1,SalI^*-BamHI (nt, -5–812), F2, BamHI-SphI (nt, 812–1993), F3 SphI-ClaI (nt, 1993–3255), F4, ClaI-EcoRI (nt, 3255–4349), F5, EcoRI-XbaI^* (nt, 4349–5907). Fragments were generated by reverse transcriptase (RT)-PCR using proofreading pfu DNA polymerase (Stratagene, La Jolla, CA). First strand cDNA as a PCR template was synthesized from 1–1.5 μg of human retinal poly A⁺ RNA (Clontech, Cambridge, UK) with the Ready-To-Go T-primed first-strand reaction kit (Amersham Biosciences, Arlington Heights, IL). PCR fragments were subcloned into vectors pBluescript SK+ (Stratagene) or pSport-1 (Invitrogen, San Diego, CA). Sequence integrity of the subclones was determined by DNA sequencing (MWG Biotech, Ebersberg, Germany). The construction of the complete Ca_v1.4α1 was performed as follows: fragment F1 + 2 was generated by ligating the BamHI-SphI fragment (F2) into the corresponding RE sites of pSport-1-containing fragment F1. Fragment F4 + 5 was generated by ligating the ClaI-EcoRI fragment (F4) into the corresponding RE sites of pBluescript SK+-containing fragment F5. These steps were followed by a three fragment ligation of the SalI^*–SphI fragment (F1 + 2) and the SphI-ClaI fragment (F3) into the SalI and ClaI sites of the F4 + 5-containing pBluescript SK+. For subsequent expression studies, the Ca_v1.4α1 construct was either inserted into plasmid pGFP⁺ (Grabner et al., 1998; Koschak et al., 2001) (yielding Ca_v1.4α1 with GFP fused to its N terminus GFP-Ca_v1.4α1) or into the corresponding vector pGFP^-, which lacks the GFP sequence.

Transient expression of LTCCs in tsA-201 cells. tsA-201 cells were maintained at 37°C and 5% CO₂ in DMEM—Coon's F12 medium (Invitrogen) supplemented with 10% (v/v) FCS (Sebak, Aidenbach, Germany), 2 mml-glutamine, and 100 U/ml of penicillin streptomycin. For transient Ca²⁺ channel expression, cells were plated onto 10 cm tissue culture dishes 12 hr before transfection with Ca²⁺ phosphate precipitation using standard protocols. Human Ca_v1.4α1, human Ca_v1.3α1 (Koschak et al., 2001), human Ca_v2.1α1 (Wappl et al., 2002), or rabbit Ca_v1.2α1-a (Mikami et al., 1989) subunits were expressed together with α2δ1 (Ellis et al., 1988), rat β3 subunits (Castellano et al., 1993), or rat β2a (Perez-Reyes et al., 1992). Transfection protocols for Ca_v2.1α1, Ca_v1.2α1, and Ca_v1.3α1 subunits were as described previously (Koschak et al., 2001). Ca_v1.4 α1-transfected cells were incubated at 30°C and 5% CO₂ 6 – 8 hr after transfection for 2 – 3 d before recording. One day before recording, cells were transferred to 3 cm culture dishes containing glass coverslips for drug application experiments. Transfected cells were visualized as GFPCa_v1.4α1 or by cotransfected GFP fluorescence.

Membrane preparation and immunoblotting with affinity-purified sequence-directed antibodies. Immunoblotting was performed as described previously (Safayhi et al., 1997; Platzer et al., 2000) using a generic anti-α1 sequence-directed antibody (anti-α1_1382-1400; raised against residues 1382–1400 of Ca_v1.1α1) (Safayhi et al., 1997). Membranes from tsA-201 cells transfected with 3 μg of α1, 2 μg of β, 2.5 μg of α2δ1 subunit cDNA, and 2.5 μg of pUC18 carrier DNA in a 10 cm culture dish were prepared as described previously (Huber et al., 2000).

Electrophysiological recordings. Whole-cell patch-clamp experiments were performed at room temperature (Axopatch 200B amplifier; Axon Instruments, Foster City, CA) and linked to a personal computer equipped with pClamp version 7.0. Currents were recorded at sampling rates of 5 or 25 kHz and low-pass filtered at 2 or 5 kHz with a Digidata 1322A analog-to-digital board (Axon Instruments). Borosilicate glass pipettes were pulled using a Sutter P-97 (Linton Instruments, Palgrave, UK), microelectrode puller and fire polished, showing typical resistances of 2–3 MΩ when filled with internal solution. Capacitance compensation and series resistance compensations of 60% were used. The solutions for whole-cell measurements were as follows (in mm): (internal solution) 135 CsCl, 10 Cs-EGTA, and 1 MgCl₂, adjusted to pH 7.4, with CsOH; (recording solution) 15 BaCl₂ or 15 CaCl₂, 10 HEPES, 150 Choline-Cl, and 1 MgCl₂, adjusted to pH 7.4, with CsOH. The holding potential (HP) was -80 mV, unless stated otherwise. The presence of ATP in the pipette solutions did not affect run down of heterologously expressed L-type channels (see below) and was therefore omitted. All voltages were corrected for a liquid junction potential of -9 mV for Ba²⁺ and -8 mV for Ca²⁺-containing solutions. Leak and capacitative currents were measured using hyperpolarizing pulses. Raw currents were corrected for linear leak currents. The voltage dependence of activation was determined from current—voltage (I—V) curves obtained by step depolarizations from the holding potential to various test potentials. I—V curves were fitted according to the following:

1

where V_rev is the extrapolated reversal potential of I_Ba, V is the membrane potential, I is the peak current, G_max is the maximum conductance of the cell, V_{0.5, act} is the voltage for half-maximal activation, and k_act is the slope factor of the Boltzmann term. The time course of current activation was fitted to the following exponential functions:

2

where I(t) is the current at time t after the depolarization, A₀ the steady state current amplitude with the respective time constant of activation, τ₀, and C the remaining steady state current or to the following:

3

yielding time constants for a fast (τ_fast) and a slow (τ_slow) component.

Effects of DHPs were monitored continuously using 0.1 Hz depolarizing pulses (40 msec) to V_max. DHPs were dissolved in the recording solution from a 10 mm stock solution in dimethyl sulfoxide and perfused through a microcapillary onto cells using a gravity driven perfusion system. Only cells exhibiting stable currents (run down <5% during the first 60 sec) were used for analysis of DHP effects. The DHPs isradipine and BayK 8644 (kindly provided by Novartis, Basel, Switzerland, and Bayer, Wuppertal, Germany) were used as their racemic mixtures.

Activation of G-proteins was achieved by intracellular perfusion with guanosine 5′-[γ-thio]triphoshate (GTPγS; Sigma, St. Louis, MO) for >3 min under whole-cell conditions. The degree of voltage-dependent current facilitation was determined as the ratio (facilitation ratio) of absolute peak current amplitudes before [-PP (prepulse)] and after (+PP) a conditioning prepulse (5–200 msec to voltages between 80 and 140 mV).

Statistics. Data were analyzed using Clampfit 8.0 (Axon Instruments) and Origin 5.0 (Microcal Software, Northampton, MA). All data are presented as mean ± SE for the indicated number of experiments. Statistical significance was determined by unpaired student's t test except when stated otherwise (Kruskal—Wallis test followed by Dunn's multiple comparison procedure, or one-way ANOVA followed by Bonferroni test as indicated).

Results

Although the DNA sequences of the human and mouse Ca_v1.4α1 subunits are known (Bech-Hansen et al., 1998; Strom et al., 1998; Naylor et al., 2000), their successful functional expression has not been reported so far. We therefore constructed a full-length Ca_v1.4α1 cDNA derived from human retina for functional expression in tsA-201 cells. The Ca_v1.4α1 cDNA contains exons 1, 2, and 9a (Strom et al., 1998; Boycott et al., 2001).

We first confirmed the efficient expression of full-length Ca_v1.4α1 subunits (calculated molecular mass, 220 kDa) on the protein level by immunoblot analysis of transfected tsA-201 cell membranes (Fig. 1). As expected for the full-length form of Ca_v1.4α1, the immunostained band comigrated with the prestained myosin molecular mass standard (217 kDa), slightly faster than Ca_v1.3α1 (calculated molecular mass, 242.5 kDa) and Ca_v1.2α1 (Fig. 1). Its expression density was slightly lower than that of Ca_v1.2α1 and Ca_v1.3α1 (Fig. 1) (n = 4).

Figure 1.

Heterologous expression of different LTCC subunits in tsA-201 cells. Cells were transfected with Ca_v1.2α1, Ca_v1.3α1, or Ca_v1.4α1 together with β3 and α2δ1 subunit cDNA as described in Materials and Methods. Expression of α1 subunit proteins was analyzed in immunoblots of membranes prepared from lyzed cells after separation on 8% SDS-PAGE gels (10 μg of membrane protein per lane) using a generic anti-α1 sequence directed antibody (anti-CP_1382–1400). No α1 immunoreactivity was present in mock-transfected cells used as a control. One of four experiments yielding similar results is shown.

Next, we investigated whether the heterologously expressed Ca_v1.4α1 subunits can also form functional channels after expression in tsA-201 cells using the whole-cell patch-clamp technique. Using a standard transfection protocol and cotransfection with α2δ1 and β3 subunits, significant I_Ba was measurable during depolarization from an HP of -90 mV for both Ca_v1.4α1 and the GFP-Ca_v1.4α1 fusion protein (Fig. 2). Compared with Ca_v1.3α1, the expression efficiency was lower. Only 50 of 227 (22%) Ca_v1.4α1 + β3 + α2δ1 and 19 of 71 (28%) GFP-Ca_v1.4α1 + β3 + α2δ1 transfected (i.e., GFP-positive) cells yielded I_Ba (15 mm Ba²⁺ as charge carrier), exceeding endogenous currents (Fig. 2, legend). In contrast, 66% of GFP-positive cells transfected with Ca_v1.3α1 + β3 + α2δ1 (and >90% with Ca_v1.2α1 + β3 + α2δ1; data not shown) expressed L-type currents. When Ca_v1.4α1 was coexpressed with β3 subunits in the absence of α2δ1, current densities did not exceed those of endogenous currents measured in untransfected tsA-201 cells (Fig. 2A) (p > 0.05). This shows that Ca_v1.4α1 can associate with α2δ subunits. Because β subunits exert modulatory effects on Ca_v1.4α1-meditated currents (see below), the smallest functional complex is Ca_v1.4α1 + β + α₂δ. Coexpression of Ca_v1.4α1 + α2δ1 with β2a subunits, which are important for normal retinal function (Ball et al., 2002), also yielded significant I_Ba above endogenous currents (p < 0.01) (Fig. 2A).

Figure 2.

Biophysical properties of I_Ba and I_Ca through Ca_v1.4α1 subunits. Ca_v1.4α1 subunits were expressed together with β3 + α2δ1 (A—C) or β2a + α2δ1 (A), as described in Materials and Methods, using 15 mm Ba²⁺ (A—C) or 15 m_M Ca²⁺ (D) as charge carrier. A, Expression density was determined by depolarizing pulses to V_max.I_Ba was measured 48 – 82 hr after transfection for the indicated number of cells. Small currents measured in untransfected cells were attributable to endogenous non-L-type currents that were <1.72 pA/pF. Asterisks indicate statistically significant difference to untransfected cells (p < 0.01; Kruskal—Wallis test, followed by Dunn's multiple comparison test). B, Superimposed currents were activated by depolarizing Ca_v1.4α1-transfected cells during 50 msec pulses from an HP of -90 mV to between -60 and 40 mV in 10 mV steps. C, Normalized I—V curves for Ca_v1.4α1 coexpressed with β3 and α2δ1 subunits using 15 mm Ba²⁺ (black squares) or Ca²⁺ (black circles) as charge carriers. Biophysical parameters are given in Table 1. D, Protocol as in B but with 15 mm Ca²⁺ in the bath solution.

We analyzed the biophysical properties of Ca_v1.4α1-mediated currents in comparison with Ca_v1.2α1 and Ca_v1.3α1, which, like Ca_v1.4α1, are also expressed in sensory cells including the retina (Morgans et al., 1998; Taylor and Morgans, 1998; Morgans, 1999; Berntson et al., 2003). Ca_v1.4α1 I_Ba typically activated at more negative voltages (-38.8 ± 0.7 mV; n = 36; β3 + α2δ1 coexpression) than Ca_v1.2α1 but slightly more positive than Ca_v1.3α1 (p < 0.001) (Table 1) (Koschak et al., 2001). Representative currents activated by 50 msec step depolarizations to different test potentials (HP, -90 mV) are illustrated in Figure 2B. The I—V relationship (V_0.5,act, V_max) was shifted to more positive potentials with 15 mm Ca²⁺ as the charge carrier (p < 0.01) (Table 1, Fig. 2C).

Table 1.

Biophysical properties of Ba²⁺and Ca²⁺currents through heterologously expressed Ca_v1.4α1 subunits

							+ BayK 8644
Subunits coexpressed with α2δ1	Charge carrier (mM)	V0.5,act (mV)	Kact (mV)	Vmax (mV)	Activation threshold (mV)	Current density (pA/pF)	V0.5,act (mV)	Kact (mV)
Cav1.4α1 + β3	15Ba2+	−8.9 ± 0.8	−8.0 ± 0.4	4.2 ± 0.7	−38.8 ± 0.7	15.5 ± 2.0	−21.5 ± 1.3xxx	−5.7 ± 0.3xx
		n = 36	n = 36	n = 36	n = 36	n = 36	n = 9	n = 9
GFP-Cav1.4α1 + β3	15Ba2+	−11.0 ± 1.4	−9.1 ± 0.4	2.7 ± 1.1	−42.0 ± 1.1	6.0 ± 1.8	−24.9 ± 1.5xxx	−4.7 ± 0.5xxx
		n = 18	n = 18	n = 18	n = 18	n = 18	n = 8	n = 8
Cav1.4α1 + β2a	15Ba2+	−10.7 ± 1.4	−9.5 ± 0.3	2.6 ± 1.3	−45.6 ± 1.2**	4.6 ± 1.3
		n = 6	n = 6	n = 6	n = 6	n = 6	ND	ND
Cav1.3α1 + β3	15Ba2	−16.8 ± 0.7***	−9.2 ± 0.2	−1.7 ± 0.7***	−44.5 ± 0.5***	22.4 ± 4.2
		n = 52	n = 52	n = 52	n = 52	n = 52	ND	ND
Cav1.2α1 + β3	15Ba2	−3.0 ± 1.1**	−8.0 ± 0.2	13.4 ± 1.9***	−31.5 ± 0.5***	20.4 ± 5.6
		n = 20	n = 20	n = 20	n = 20	n = 20	ND	ND
Cav1.4α1 + β3	15Ca2+	0.6 ± 2.2**	−9.4 ± 0.5	15.1 ± 2.2***	−33.6 ± 1.8	12.7 ± 3.7
		n = 6	n = 6	n = 6	n = 6	n = 6	ND	ND
Cav1.3α1 + β3	15Ca2	−7.8 ± 1.1	−9.4 ± 0.4	7.1 ± 1.6*	−37.3 ± 0.9	13.5 ± 2.2
		n = 13	n = 13	n = 13	n = 13	n = 13	ND	ND
Cav1.2α1 + β3	15Ca2	9.2 ± 1.5+	−10.5 ± 0.3***	21.4 ± 1.2	−27.6 ± 1.3+	10.4 ± 2.1
		n = 22	n = 22	n = 22	n = 22	n = 22	ND	ND

GFP-Ca_v1.4α1 and Ca_v1.4α1 Ca²⁺ channels were coexpressed with rat β3 or rat β2a and rabbit α2δ1 subunits in tsA-201 cells, and the biophysical properties were determined using 15 mm Ba²⁺ or Ca²⁺ as a charge carrier. V_0.5,act, K_act, and V_max were obtained by fitting the data as described in Materials and Methods. The activation threshold was determined as the test potential at which 5% of the maximal current was activated. Data are given as means ± SE; ND, not determined. Statistical differences for V_0.5,act, K_act, V_max, and activation thresholds (calculated by one-way ANOVA, followed by Bonferroni test) are given in comparison with Ca_v1.4α1 + β3 + α2δ1 with Ba²⁺ (^*p<0.05; ^**p<0.01; ^***p<0.001) or Ca²⁺ (⁺p<0.05; ⁺⁺p<0.01; ⁺⁺⁺p<0.001) as charge carrier. For cells used for the calculation of biophysical parameters, current densities (no BayK 8644 present) are also given. For statistical comparisons of current densities, see Figure 2A. Statistically significant differences for BayK 8644 effects (^xxp<0.01; ^xxxp<0.001; Student's t tests) are also indicated.

The time course of activation determined during 50 msec depolarizations to V_max revealed similar activation time constants for I_Ba through GFP-tagged or non-GFP-tagged Ca_v1.4α1 subunits. When coexpressed with β3 + α2δ1, activation could be described by a monoexponential time course in the majority of cells (Ca_v1.4α1; 0.43 ± 0.08 msec; 9 of 14 cells). In the remaining cells, a biexponential onset of activation was measured (τ_fast = 0.46 ± 0.08 msec; τ_slow = 16.4 ± 3.9 msec; relative contribution of slow component, 6.03 ± 1.93%). Coexpression of β2a subunits primarily resulted in activation by a biexponential time course (three of four cells; τ_fast = 0.70 ± 0.13 msec; τ_slow = 6.6 ± 0.1 msec; relative contribution of slow component, 3.9 ± 1.1%). Therefore, with respect to the activation properties, I_Ba through Ca_v1.4α1 Ca²⁺ channels closely resembled Ca_v1.3α1 currents, which also activate with faster time courses and at lower voltages than Ca_v1.2 (Koschak et al., 2001; Scholze et al., 2001; Xu and Lipscombe, 2001). No changes in the (monoexponential) activation time course were detected for Ca_v1.3α1 currents in the presence of β2a subunits (monophasic activation; α2δ1 + β3, τ_act = 0.77 ± 0.06; n = 24; α2δ1 + β2a, τ_act = 0.73 ± 0.1; n = 4; p > 0.05).

One property that distinguishes Ca_v1.3α1 from Ca_v1.2α1 currents is its slower inactivation during prolonged depolarizations (Koschak et al., 2001). The experiments in Figure 3 illustrate that inactivation of Ca_v1.4α1 was even slower than for Ca_v1.3α1 (Fig. 3). Only 50.2 ± 2.9% (n = 17) of I_Ba inactivated after 5 sec of depolarization to V_max (Fig. 3A,B). After 10 sec, 84.2 ± 6.4% (n = 4) of Ca_v1.3α1 but only 68.1 ± 2.7% (n = 17) of Ca_v1.4α1I_Ba inactivated (p < 0.05) (Fig. 3A,B). Substitution of β2a for β3 subunits also significantly slowed inactivation (Fig. 3E,F). This also demonstrates that β subunits participate in fine tuning the Ca_v1.4α1 channel complex.

Figure 3.

Inactivation properties of Ca_v1.4α1 Ca²⁺ channels. A, I_Ba (black traces) through Ca_v1.3α1 and Ca_v1.4α1 subunits coexpressed with β3 and α2δ1 subunits were elicited by 10 sec depolarizing pulses from an HP of -90 mV to V_max. Representative current traces for Ca_v1.3α1 (n = 4) and Ca_v1.4α1 (n = 17) channels are shown. Traces were normalized to the peak current amplitudes. For the experiments shown, inactivation measured during 5 and 10 sec depolarizing pulses was as follows: Ca_v1.4α1, 41 and 56%; Ca_v1.3α1, 89 and 97%. A representative trace for current through Ca_v1.4α1 recorded with 15 mm Ca²⁺ as charge carrier is illustrated in gray superimposed on the I_Ba trace indicated in black. B, Percent current inactivation measured after 0.25, 5, and 10 sec depolarizations to V_max in Ca_v1.4α1-transfected cells using either 15 mm Ba²⁺ or 15 m_M Ca²⁺ as the charge carrier. Inactivation of currents during pulses was not significantly different for Ba²⁺ (black bars) and Ca²⁺ (gray bars) (n = 7; p > 0.05). C, D, Inactivation for Ca_v1.2α1 (C) and Ca_v1.3α1 (D) during 2 sec depolarizing pulses toV_max with 15 mm Ba²⁺ (black trace) or 15 m_M Ca²⁺ (gray trace) as charge carriers. For Ca_v1.3α1, a variable noninactivating I_Ca component was found (9 – 40%; n = 4), whereas remaining Ca_v1.2α1 currents were always <3.5% (n = 7). E, Inactivation of I_Ba through Ca_v1.4α1 cotransfected with β3 (black) or β2a (gray) and α2δ1. Currents were normalized to peak I_Ba. Currents were elicited by depolarization from an HP of -90 mV to V_max. F, Percentage of inactivation of I_Ba through Ca_v1.4α1 cotransfected with β3 (black; n = 17) or β2a (gray;n = 7), and α2δ1 was determined after 5 and 10 sec during a depolarization from an HP of -90 mV toV_max. Currents were normalized to peakI_Ba. Inactivation with β2a coexpression was 27.2±4.6% (after 5 sec) and 44.6 ± 6.4% (after 10 sec; n=7), respectively. Asterisks indicate a statistically significant difference to β2a coexpression (p < 0.01).

In addition to voltage, Ca²⁺ is also an important determinant of LTCC inactivation. Figure 3, C and D, illustrates that not only inactivation of Ca_v1.2α1 (for review, see Budde et al., 2002) but also of heterologously expressed Ca_v1.3α1 occurred in a Ca²⁺-dependent manner (percentage of current inactivation after 250 msec; Ca_v1.2α1, I_Ba, 60.7 ± 8%; n = 4; I_Ca, 84.5 ± 3.1%; n = 6;p < 0.05; Ca_v1.3α1, I_Ba, 37.5 ± 2.9%; n = 13; I_Ca, 68.8 ± 4.7%;n = 12; p < 0.001). In contrast, under the same experimental conditions (10 mm EGTA in the pipette solution), Ca_v1.4α1 did not exhibit accelerated inactivation with Ca²⁺ as charge carrier throughout its slow inactivation time course (Fig. 3A,B). As a consequence, the majority of I_Ca through Ca_v1.2α1 and Ca_v1.3α1, but hardly any Ca_v1.4α1 current, inactivated during 200 – 400 msec.

LTCCs are defined by their high sensitivity to DHP antagonists and their activation by DHP Ca²⁺ channel activators (Peterson et al., 1996), such as BayK 8644. In Ca_v1.4α1 subunits, a IVS6 tyrosine (position 1414 in the human Ca_v1.4α1 sequence) (Fig. 8), previously shown to contribute to the formation of the binding pocket (Peterson et al., 1996), is replaced by a phenylalanine. The corresponding mutation in Ca_v1.1α1 subunits reduces DHP antagonist sensitivity ∼3- to 5-fold (Peterson et al., 1996). Its role for agonist action has not yet been studied. Therefore, we tested the DHP sensitivity of heterologously expressed Ca_v1.4α1 channels (+ β3 + α2δ1). At -90 mV HP, the DHP antagonist isradipine (1 μm) blocked 82.7 ± 2.9% (n = 7) of I_Ba elicited by 0.1 Hz depolarizing pulses to V_max (Fig. 4A). The same concentration completely inhibited Ca_v1.3α1 currents under identical experimental conditions, as reported previously (Koschak et al., 2001) (Fig. 4B). Current inhibition by 300 nm concentrations was also slightly less pronounced for Ca_v1.4α1, compared with Ca_v1.3α1 (Koschak et al., 2001) (Fig. 4B). Changing the HP from -90 to -50 mV dramatically increased isradipine sensitivity of Ca_v1.4α1 (Fig. 4B, open triangle), indicating a voltage-dependent mechanism of DHP block, which is also typical for both Ca_v1.2α1 and Ca_v1.3α1 (Welling et al., 1997; Koschak et al., 2001). Activation of I_Ba through Ca_v1.4α1 by the Ca²⁺ channel activator BayK 8644 occurred in an LTCC-typical manner (Fig. 5). Perfusion of Ca_v1.4α1-transfected cells (yielding significant I_Ba already in the absence of drug) with 5 μm BayK 8644 resulted in a robust (9.6 ± 1.9-fold; n = 7) increase of the maximal I_Ba (Fig. 5A), similar to Ca_v1.3α1 stimulation (Koschak et al., 2001). Furthermore, BayK 8644 produced a typical ∼10 mV hyperpolarizing shift of the I—V curve (Table 1, Fig. 5B). Interestingly, in some GFP-positive cells (four of four tested) with no significant current under basal conditions, the presence of Ca_v1.4α1 currents was unmasked by application of the Ca²⁺ channel activator BayK 8644 (5 μm; 10.5 ± 1.3 pA/pF; n = 4). This suggested that an even >10-fold stimulation of Ca_v1.4α1 currents occurred in some cells. Interestingly, a similarly strong BayK 8644 dependence of L-type current components has also been described in retinal cone bipolar cells (Pan, 2000). Our data demonstrate that phenylalanine in position 1414 of Ca_v1.4α1 still supports full agonist sensitivity. Therefore, a tyrosine in this position is unlikely to be required for DHP agonist action in LTCCs.

Figure 8.

Sequence alignment of L-type Ca_v1.4α1 with Ca_v1.2α1 and Ca_v1.3α1. An amino acid exchange within the proposed DHP-binding domain of LTCCs is indicated by an arrow. A tyrosine residue conserved in all L-type α1 subunits is replaced by phenylalanine in position 1414 of Ca_v1.4α1. The sequence alignment also illustrates amino acid differences between Ca_v1.4α1 and Ca_v1.2α1 or Ca_v1.3α1, which might explain the differences in Ca²⁺-dependent inactivation described under our experimental conditions. Sequence stretches previously identified as critical determinants for calmodulin-binding and Ca²⁺-dependent inactivation (de Leon et al., 1995; Zuhlke et al., 1999; Mouton et al., 2001; Pitt et al., 2001) are indicated.

Figure 4.

DHP antagonist sensitivity of Ca_v1.4α1 Ca²⁺ channels. Ca_v1.4α1 was coexpressed with β3 and α2δ1 subunits in tsA-201 cells and recorded in bath solution containing 15 mm Ba²⁺ as charge carrier. A, I_Ba was elicited from depolarizations to V_max (filled circles) in the absence or presence (gray bar) of the DHP antagonist isradipine. For the experiment shown,I_Ba block by 1 μm isradipine was 80%. Corresponding peak current traces elicited by 40 msec depolarizations in the absence (control) and presence of isradipine are shown in the inset. B, Concentration-dependent inhibition was measured from a holding potential of -90 mV for Ca_v1.4α1 (black triangles) during superfusion of the cell with bath solution containing the indicated concentrations of isradipine. The dose—response relationship was compared with Ca_v1.3α1 (squares) (Koschak et al., 2001). Pronounced voltage dependence of isradipine block of I_Ba through Ca_v1.4α1 subunits was observed by changing the HP from -90 to -50 mV (open triangle). Asterisks indicate statistically significant difference (p < 0.05; data are means ± SE).

Figure 5.

DHP agonist sensitivity of Ca_v1.4α1 Ca²⁺ channels. For all experiments, Ca_v1.4α1 subunits were coexpressed with β3 and α2δ1 subunits in tsA-201 cells. Charge carrier was 15 mm Ba²⁺. One representative experiment (of 9) is shown. A, Stimulation of I_Ba through Ca_v1.4α1 Ca²⁺ channels by the Ca²⁺ channel activator BayK 8644 (BayK). I_Ba was elicited by depolarizations to V_max before and after application of BayK 8644-containing solution (5 μm). Maximal I_Ba is plotted against time. The inset shows representative traces in the absence (control) and presence of BayK 8644. B, Current—voltage relationship for Ca_v1.4α1 in the absence (black circles) and presence of 5 μm BayK 8644 (BayK; gray circles). V_{0.5, act} was -8 and -21.1 mV for the control and BayK 8644-modulated current, respectively. One representative experiment (of 7) is shown.

We also tested the modulation of Ca_v1.4α1 currents by G-protein activation and/or strong depolarizing pulses. Figure 6A shows the effect of 200 msec depolarizing prepulses to 80 mV on I_Ba through Ca_v1.4α1, elicited by a subsequent test pulse toV_max. Prepulses facilitated I_Ba in 59% (10 of 17) of Ca_v1.4α1 + β3+ α2δ1-transfected cells. The absence of facilitation in some cells has also been reported previously for Ca_v1.2α1 after expression in mammalian cells (Kamp et al., 2000). In cells showing facilitation, the extent of facilitation was slightly (but not significantly) smaller for Ca_v1.4α1 (facilitation ratio, 1.13 ± 0.02) than for Ca_v1.2 α1 (1.27 ± 0.07) and Ca_v1.3 α1 (1.26 ± 0.09) (Fig. 6B) when measured under identical experimental conditions. As expected (Bourinet et al., 1994), no facilitation was observed for the P/Q-type channel Ca_v2.1α1 subunit (Fig. 7B,C, lower panel) (n = 7) under these experimental conditions. The application of 200 msec prepulses not only increased the peak current but also accelerated the activation and inactivation kinetics of the facilitated Ca_v1.4α1 current (Fig. 6A). Whereas control I_Ba hardly inactivated (>97% of current remained at the end of the 400 msec test pulse) (Fig. 6A), slightly accelerated inactivation was detected for the facilitated current (92.8 ± 0.8% residual current;p < 0.001). Similar kinetic changes have also been observed for other facilitated L-type currents (Dai et al., 1999; Kamp et al., 2000). Coexpression with β2a + α2δ1 subunits also supported facilitation, which was observed in 85% of the experiments, and the average facilitation ratio was similar, as measured for β3 +α2δ1 (1.18 ± 0.03; n = 6). Similar facilitation was observed with Ca²⁺ as a charge carrier. Prepulses facilitated I_Ca in five of five Ca_v1.4α1 + β3 + α2δ1-transfected cells (facilitation ratio, 1.16 ± 0.06).

Figure 6.

Voltage-dependent facilitation of Ca_v1.4α1 subunits. A, Schematic representation of the voltage protocol used to elicit facilitation. Channel activity was recorded in tsA-201 cells transfected with Ca_v1.4α1, β3, and α₂δ1 subunits in 15 mm Ba²⁺ solution. Test pulses (TPs) of 400 msec were applied with or without a 200 msec PP. A representative current trace (facilitation ratio, 1.11) is shown. Note the differences in activation and inactivation kinetics time courses. B, Comparison of voltage-dependent facilitation of different LTCCs. Facilitation of Ca_v1.2α1, Ca_v1.3α1, and Ca_v1.4α1 I_Ba is expressed as the amplitude ratio of currents recorded with a prepulse (+PP) over the respective control currents without prepulse (-PP). A similar extent of facilitation was observed for all LTCC subtypes tested (p > 0.05). Facilitation was observed in seven of seven Ca_v1.2α1-transfected cells, eight of 20 Ca_v1.3α1-transfected cells, and 10 of 17 Ca_v1.4α1 (+ β3 + α2δ1)-transfected cells.

Figure 7.

Depolarization-induced facilitation of Ca_v1.4α1 subunits. A, Pulse protocol used to determine G-protein modulation of Ca_v1.4α1- and Ca_v2.1α1-mediated I_Ba. Facilitation was measured during 50 msec test pulses (TP) to V_max either with or without a 5 msec PP to 140 mV. B, Facilitation ratios for Ca_v1.4α1 and Ca_v2.1α1 Ca²⁺ channels (coexpressed with β3 and α2δ1) in the absence (control; gray bars) and presence of 300 μm intracellular GTPγS (black bars) using the pulse protocol described in A. Recordings were started after ≥3 min of dialysis with GTPγS. The following facilitation ratios were obtained in the absence and presence of GTPγS, respectively:Ca_v1.4α1, 1.1±0.02 (n=11 of 17 cells), 1.12±0.02 (n=7 of 14 cells); Ca_v2.1α1, 0.99 ± 0.01 (n = 7 of 7 cells), 1.3 ± 0.07 (n = 10 of 10 cells). Asterisk indicates statistically significant difference (p < 0.01). C, Representative current traces for the experiments described in B.

We found no evidence for a G-protein dependence of this prepulse facilitation of Ca_v1.4α1 currents. When the nonhydrolyzable GTP analog GTPγS was included in the pipette solution to activate expressed G-proteins in tsA-201 cells (Herlitze et al., 1997; Meza and Adams, 1998), facilitation induced by 5 msec depolarizing prepulses to 140 mV remained unaffected (Fig. 7). To prove that G-protein activation is feasible under our experimental conditions, the modulation of Ca_v2.1α1 channels was determined using the same experimental protocol. At least 3 min after establishing the whole-cell configuration, prepulse application caused the typical relief of G-protein modulation of Ca_v2.1α1 currents characterized by faster activation and the reduction of peak current amplitude (Fig. 7B,C). All Ca_v2.1α1 +β3 + α2δ1-expressing cells (seven of seven) dialyzed with GTPγS exhibited this typical G-protein-mediated inhibition (Fig. 7B,C), consistent with previous studies (Bourinet et al., 1996; Zhang et al., 1996; Herlitze et al., 1997; Meza and Adams, 1998; Canti et al., 1999). These experiments clearly demonstrated that voltage-dependent G-protein modulation can be measured under our experimental conditions, but that facilitation of Ca_v1.4α1 is unlikely to result from relief of G-protein-induced channel inhibition.

Discussion

Biophysical properties of Ca_v1.4α1

Here we report the first successful functional characterization of Ca²⁺ currents through Ca_v1.4α1 subunits. Experiments were performed under the same conditions previously used to compare the biophysical and pharmacological properties of Ca_v1.3α1 and Ca_v1.2α1 LTCCs using the whole-cell patch-clamp technique. Ca_v1.4α1 currents resemble more closely Ca_v1.3α1 than Ca_v1.2α1. Like Ca_v1.3α1, Ca_v1.4α1 activated more rapidly and at more negative voltages than heterologously expressed Ca_v1.2α1 and also inactivated more slowly. Ca_v1.4α1 exhibited no Ca²⁺-induced inactivation, which was found to exist in Ca_v1.3α1 (Xu and Lipscombe, 2001; our observations) and is well studied for Ca_v1.2α1 (for review, see Budde et al., 2002). Therefore, the slower inactivation of Ca_v1.4α1 became especially prominent when Ca²⁺ was the permeating cation (Fig. 3).

Ca²⁺-dependent inactivation is mediated through Ca²⁺ and calmodulin interaction with the C-terminal tail and is a typical property of LTCCs (for review, see Budde et al., 2002). However, the absence of Ca²⁺-dependent inactivation is not a specific property of Ca_v1.4α1. A putative neuronal Ca_v1.2α1 C-terminal splice variant, Ca_v1.2α1₈₆ (α1C_,86), also lacks Ca²⁺-induced inactivation (Soldatov et al., 1997). In this splice variant, 80 amino acid residues of the C-terminal tail are replaced by 81 essentially nonidentical amino acid residues. This eliminates important motifs essential for calmodulin-mediated Ca²⁺-dependent inactivation and also causes a profound acceleration of voltage-dependent inactivation of I_Ba. In contrast, inspection of the Ca_v1.4α1 sequence (Fig. 8) revealed that the regions recently identified as determinants for Ca²⁺-dependent inactivation in Ca_v1.2α1 (Qin et al., 1999; Zuhlke et al., 1999; Pate et al., 2000; Peterson et al., 2000; Romanin et al., 2000; Mouton et al., 2001; Pitt et al., 2001) are highly conserved in this subunit. Only a few amino acid differences, as compared with Ca_v1.2α1 and Ca_v1.3α1, exist in the F-helix of the EF hand (Peterson et al., 2000), peptide A (Pitt et al., 2001), the CB peptide (Pate et al., 2000), and peptide C (Pitt et al., 2001). Therefore, Ca_v1.4α1 subunits represent an ideal model to further study the role of these amino acid changes for the molecular mechanisms of Ca²⁺-dependent inactivation. Additional studies must also address the question whether Ca_v1.4α1 undergoes N-lobe calmodulin-mediated calcium-dependent inactivation revealed at low intracellular Ca²⁺ buffering (Liang et al., 2003).

DHP-sensitivity of Ca_v1.4α1

As for Ca_v1.3α1, the apparent DHP antagonist sensitivity of Ca_v1.4α1 was significantly lower (∼15-fold) than for Ca_v1.2α1 at negative holding potentials. This intermediate DHP antagonist sensitivity of Ca_v1.4α1 and Ca_v1.3α1 is in good accordance with data obtained on L-type currents in retinal cells, in which relatively high concentrations of DHPs are required to block I_Ca,L (Wilkinson and Barnes, 1996; Protti and Llano, 1998; Taylor and Morgans, 1998). Ca_v1.4α1 current inhibition by DHP antagonist was highly voltage dependent. At a more positive membrane potential, 100 nm isradipine inhibited >80% of I_Ba (Fig. 4B), indistinguishable from the block of Ca_v1.3α1 under the same experimental conditions (Koschak et al., 2001). Therefore, the low apparent affinity of Ca_v1.4α1 in comparison with Ca_v1.2α1 is likely to be attributable to differences in the voltage-dependent interaction of the DHP antagonist, as recently demonstrated also for Ca_v1.3α1 (Koschak et al., 2001). We could exploit a “natural mutation,” an amino acid exchange from tyrosine to phenylalanine in position 1414 (Fig. 8), of the DHP-binding domain to investigate the role of this residue for agonist action. In Ca_v1.1α1 (data not shown), Ca_v1.2α1, and Ca_v1.3α1, a tyrosine is found in this position (Tyr 1463; α1C-a numbering) (Striessnig et al., 1998). Mutation of this residue to phenylalanine was found to decrease DHP antagonist-binding affinity at least in Ca_v1.1α1 (Peterson et al., 1996). We show that the tyrosine to phenylalanine exchange in Ca_v1.4α1 does not cause a major change in DHP antagonist sensitivity, as compared with Ca_v1.3α1. The role of this tyrosine for DHP agonist sensitivity has not been investigated thus far. Because we found a robust stimulation of Ca_v1.4α1-mediated currents by BayK 8644, we can clearly demonstrate that the tyrosine hydroxyl is not required for BayK 8644 stimulation of LTCCs.

Functional implications

Our data provide a first answer to the important question whether Ca_v1.4α1 LTCCs can contribute to the L-type currents in retinal photoreceptors and bipolar cells, which are tightly coupled to neurosecretion. In photoreceptors (Wilkinson and Barnes, 1996; Taylor and Morgans, 1998; Kourennyi and Barnes, 2000; Stella et al., 2002) and bipolar cells (von Gersdorff and Matthews, 1996; Protti and Llano, 1998) of different species, DHP-sensitive I_Ca was found to posses properties not typically found for L-type currents in cardiac myocytes or neurons. These were described as faster activation, slower inactivation, negative activation thresholds, and intermediate DHP sensitivity. Such properties were described previously both in the current study and by others (for review, see Lipscombe, 2002) for Ca_v1.3α1. Ca_v1.3α1 is expressed in photoreceptor terminals in the outer plexiform layer (OPL) and, most likely, also bipolar cells synapses in the inner plexiform layer (IPL) (Morgans et al., 1998; Taylor and Morgans, 1998; Morgans, 1999). Therefore, these channels could account for the retinal I_Ca,L in these cells. However, we can now demonstrate that Ca_v1.4α1 can also mediate currents with similar properties. Because Ca_v1.4α1 is also expressed in the synapses of the OPL and IPL (Firth et al., 2001; Morgans, 2001; Morgans et al., 2001; Ball et al., 2002; Berntson et al., 2003), it may also participate in the formation of photoreceptor and bipolar cell I_Ca,L (Berntson et al., 2003). In humans, Ca_v1.4α1 mutations cause iCSNB2. Most of these mutations result in truncated subunits and should cause a complete loss of function. The most promising animal model to directly quantitate the contribution of Ca_v1.4α1 to retinal I_Ca,L are Ca_v1.4α1-deficient mice. In these animals, one should be able to correlate visual defects with the relative contribution of Ca_v1.4α1 to retinal I_Ca,L. Ca_v1.3α1-deficient mice do not seem to represent such a useful model because they do not exhibit electroretinogram changes (Platzer et al., 2000) (M.W. Seeliger, E. Schmid, J. Platzer, and J. Striessnig, unpublished observations).

On the basis of their biophysical characteristics and intermediate DHP sensitivity, Ca_v1.3α1 and Ca_v1.4α1 might be classified as a functional LTCC subgroup. Because of its lower activation threshold, Ca_v1.3α1 has indeed been shown to serve in an essential role for cardiac pacemaking in the sinoatrial node, which cannot be substituted by Ca_v1.2α1 expressed in the same cells (Zhang et al., 2002; Mangoni et al., 2003). Faster activation, lower activation thresholds, and slower inactivation make Ca_v1.3α1 and Ca_v1.4α1 also suited for certain neuronal functions. First, they can support neurotransmitter release from nonspiking neurons and sensory cells such as photoreceptors (Ca_v1.3α1 and Ca_v1.4α1) and cochlear inner hair cells (Ca_v1.3α1). In the latter, >90% of the current is carried by Ca_v1.3α1 (Platzer et al., 2000). In the darkness, photoreceptors are continuously depolarized by cGMP-gated channels to approximately -30 to -40 mV. During illumination, they hyperpolarize by approximately -20 to -30 mV (Witkovsky et al., 1997). To rapidly adjust tonic release to changes in illumination (i.e., changes in membrane potential),I_Ca,L should be rapidly gated, activated over a relatively negative voltage range, and slowly inactivated at depolarized potentials. These criteria are fulfilled by Ca_v1.3α1 and Ca_v1.4α1 but not Ca_v1.2α1. Second, sustained I_Ca,L, activating at negative voltages, are suitable to support plateau potentials in neurons elicited, for instance, by weak depolarizations to voltages just above the resting potential of a neuron. For example, such plateau potentials occur in motoneurons (Carlin et al., 2000; Alaburda et al., 2002) and second order pain neurons (Morisset and Nagy, 2000), in which they modulate motoneuron responses and pain processing, respectively. Such current components are believed to be mediated by Ca_v1.3α1 (Carlin et al., 2000; Alaburda et al., 2002), but systematic analysis of Ca_v1.4α1 expression (e.g., in motoneurons) has not yet been performed. Similarly, in bipolar cell nerve terminals, such low voltage-activated L-type Ca²⁺ currents also seem to account for the interesting finding that specific retinal bipolar cells, which are generally considered nonspiking cells, can respond to light-induced tonic depolarization by photoreceptors with Ca²⁺ action potentials and regenerative responses from a plateau potential (Burrone and Lagnado, 1997; Protti et al., 2000), mechanisms that are suitable to amplify small photoreceptor signals. Additional studies will have to determine the inactivation kinetics of Ca_v1.3α1 and Ca_v1.4α1 on even longer time scales than those shown here. Note that in mouse cochlea inner hair cells, Ca_v1.3α1-mediated currents inactivate to a much smaller extent than after heterologous expression (Zidanic and Fuchs, 1995; Kollmar et al., 1997; Platzer et al., 2000). It will be important to reveal the molecular substrate of this difference, which may not be only attributable to alternative splicing of α1 subunits (Koschak et al., 2001; Safa et al., 2001; Xu and Lipscombe, 2001).

Our work demonstrates for the first time that Ca_v1.4α1 subunits form functional LTCCs that can contribute to such currents. Therefore, this study paves the way for the analysis of Ca_v1.4α1 mutations responsible for iCSNB2. A detailed genotype—phenotype analysis of this disease will now be possible.