DSC1, a Drosophila channel with sequence similarity to the voltage-gated sodium channel (NaV), was identified over 20 years ago. This channel was suspected to function as a non-specific cation channel with the ability to facilitate the permeation of calcium ions (Ca2+). A honeybee channel homologous to DSC1 was recently cloned and shown to exhibit strict selectivity for Ca2+, while excluding sodium ions (Na+), thus defining a new family of Ca2+ channels, known as CaV4. In this study, we characterize CaV4, showing that it exhibits an unprecedented type of inactivation, which depends on both an IFM motif and on the permeating divalent cation, like NaV and CaV1 channels, respectively. CaV4 displays a specific pharmacology with an unusual response to the alkaloid veratrine. It also possesses an inactivation mechanism that uses the same structural domains as NaV but permeates Ca2+ ions instead. This distinctive feature may provide valuable insights into how voltage- and calcium-dependent modulation of voltage-gated Ca2+ and Na+ channels occur under conditions involving local changes in intracellular calcium concentrations. Our study underscores the unique profile of CaV4 and defines this channel as a novel class of voltage-gated Ca2+ channels.

DSC1 (Drosophila sodium channel 1, or NaV2) is a voltage-gated ion channel initially identified in the fruit fly Drosophila in 1987 based on its sequence homology with vertebrate sodium channels (Ramaswami and Tanouye, 1989; Salkoff et al., 1987). DSC1 expression is predominantly observed in embryonic and adult neurons (Castella et al., 2001; Hong and Ganetzky, 1994), and tissue-specific isoforms exist in the German cockroach (where this channel is called BSC1 [Liu et al., 2001]). However, its expression is largely absent in non-neuronal tissues. In Drosophila, DSC1 is also found in olfactory organs such as antenna segments or maxillary palps, where it plays a role in the processing of olfactory information (Kulkarni et al., 2002). These expression features are also found in honeybees, with RT-PCR signals positive in the brain, muscle, antenna, and ganglion but absent in the gut (Gosselin-Badaroudine et al., 2016). Additionally, it has been suggested to be involved in the response to various stresses or insecticides, including DTT and pyrethroids, in both adult and larval stages (Zhang et al., 2013; Rinkevich et al., 2015; Chen et al., 2018; Dong et al., 2014). However, it is important to note that the direct action of these insecticides on the DSC1 channel itself has not yet been experimentally tested.

DSC1 belongs to the family of 24 transmembrane (TM) helices channels, which are organized into four homologous domains, each containing six TM α helices. Similar to the other channels from this family, such as voltage-gated Na+ and Ca2+ channels, NaV and CaV, respectively (Chen-Izu et al., 2015; Catterall, 2000), these domains are split into two subdomains: a voltage-sensor subdomain, comprising the first four (S1–S4) α helices, with the S4 being the voltage-sensitive element with four to seven positive charges distributed along the helices, and a pore subdomain made of the S5 and S6 helices (that form the pore walls) and a reentrant, the loop connecting these two helices (that makes the channel ionic selectivity filter [SF]). In all these channels, extra- and intracellular loops connecting the helices and the domains play a crucial role in toxin specificity or intracellular channel regulation, among other functions.

Meticulous inspection of the DSC1 amino acids sequences from fly, cockroach, and honeybee (called CaV4, [Gosselin-Badaroudine et al., 2016]) revealed that, although being more similar to the NaV than to the CaV channels (40–45% and 20–25% homology when compared to mammalian or insect NaV or CaV channels, respectively), the DSC1 pore, and more precisely the channel SF, was clearly different from both NaV and CaV channels. The selectivity filter is composed of positively or negatively charged amino acids located in each of the four pore domains that form two rings called EEEE and DCS (divalent cation selectivity) loci (Cens et al., 2007; Cibulsky and Sather, 2000; Sather and McCleskey, 2003; Heinemann et al., 1992; Yang et al., 1993; Tang et al., 2014; Hille, 1992a). The EEEE locus plays a major role in the selection between monovalent and divalent cations and contains the DEKA sequence in NaV and the EEEE sequence in CaV (see Table 1 for the sequences of the loci in NaV, CaV, and DSC1/CaV4 channels). The EEEE and DCS loci in DSC1/CaV4 are different from NaV, CaV, DSC1, or BSC1 channels, and contain the sequence of amino acids DEEA and DEED, respectively (see Fig. 1), suggesting a particular ionic selectivity. Moreover, the loop between domains III and IV (LIII–IV), which plays a critical role in NaV fast inactivation through a motif of three specific amino acids (IFM in human NaV, and MFM in fruit fly or honeybee NaV [Gosselin-badaroudine et al., 2015]), is partly conserved in DSC1 with the homologous sequence MFL in Drosophila and in most other arthropods (Cui et al., 2012), including honeybees (Gosselin-Badaroudine et al., 2016). This observation suggests that DSC1 may represent an evolutionary intermediate stage between CaV and NaV channels (Dudev and Lim, 2014b). It is conceivable that DSC1 retains the selectivity of a CaV channel while already incorporating the inactivation mechanism typical of a NaV channel. These characteristics delineate a novel voltage-gated calcium channel (VGCC) family, which differs in structure, biophysical, and pharmacological aspects from the three previously recognized subfamilies of VGCC namely CaV1, CaV2, and CaV3 (Gosselin-Badaroudine et al., 2016; Dong et al., 2015). Consequently, Gosselin-Badaroudine et al. [2016] coined the term CaV4 for the honeybee DSC1 ortholog (Gosselin-Badaroudine et al., 2016), a nomenclature that has been subsequently adopted.

Although the ionic selectivity of mammalian CaV and NaV channels has been extensively explored through biophysical, molecular, and structural studies (Neumaier et al., 2015; Sather et al., 1994; Sather and McCleskey, 2003 for review), studies focused on the basal metazoan channels, and more precisely, on DSC1, or its cockroach, honeybee, or sea anemone orthologs (Zhou et al, 2004; Dudev and Lim, 2014b; Moran et al, 2015; Gosselin-Badaroudine et al, 2016), are relatively limited (Gosselin-Badaroudine et al., 2016; Dudev and Lim, 2014a; Gur Barzilai et al., 2012). These studies have suggested that various arrangements of the EEEE locus (EEEE, DEKA, DKEA, and DEEA) in the SF can give rise to different relative Na+/Ca2+/Ba2+/K+ permeabilities. However, their specific selectivity profiles and inactivation properties under different experimental conditions received limited analysis (Gosselin-Badaroudine et al., 2016).

The first biophysical study of the DSC1 channel has been carried out in Xenopus oocytes with the DSC1 cockroach ortholog of the Drosophila channel (BSC1 [Zhou et al, 2004]). However, due to the existence of endogenous Ca2+-activated Cl current in Xenopus oocytes, the recordings were mostly conducted using a high Ba2+ concentration (50 mM), and in the absence of external Cl, restricting the contaminating chloride conductance to an inward current (outward flux of Cl). On the basis of tail current analysis, BSC1 was described as a cation-permeable channel exhibiting slow activation, inactivation, and deactivation kinetics, and displaying a permeability to divalent cations (Ca2+ and Ba2+) and, to a lesser extent, to monovalent ions (Na+). A recent work expressing CaV4 (the honeybee DSC1 ortholog) extends this characterization and suggests that CaV4 was not only able to permeate Ca2+ but was also impermeable to Na+ (Gosselin-Badaroudine et al., 2016), Na+ permeability was only restored by the DEEA→DEKA mutation in the SF. However, analysis of CaV4 inactivation has never been performed yet, although differences between Ba2+ and Ca2+ currents kinetics have been clearly identified.

Therefore, in the present study, we use the intra-oocyte online-BAPTA-injection procedure, which allows us to efficiently inhibit the Ca2+-activated Cl current, and therefore record Ca2+ and Ba2+ currents without any contamination. BAPTA injection spares the Ca2+-dependent inactivation, at least on CaV1.2 and CaV1.3 Ca2+ channels, which relies on a very local increase in Ca2+ concentration in the vicinity of the intracellular mouth of the channel.

We demonstrate that DSC1/CaV4 is a high voltage–activated Ca2+ channel with a high affinity for divalent cations, an inactivation mechanism driven by the MFL motif in the loop IIII–IV and subjected to an unusual Ca2+-dependency. CaV4 is insensitive to both NaV and CaV regulators, with the exception of diltiazem and veratrine. While diltiazem reduces Ba2+ current amplitude and slows inactivation, veratrine, an alkaloid known for slowing NaV channel inactivation and deactivation, exerts the opposite effect on CaV4. This unique behavior therefore truly defines CaV4 as a novel class of VGCC.

cDNA, cRNA, and Xenopus oocyte preparation

Cloning of Am-CaV4 channel in the pPol_Not1 vector, an expression vector containing the T7 promoter, the Xenopus laevis β-globin 5′- and -3′-untranslated region, the Xenopus β-globin 3′-untranslated regions, and polyA and polyC tracts, was previously done (Gosselin-Badaroudine et al., 2016). The constructs were linearized with Not1, and T7 RNA polymerase was used to synthesize RNA using mMESSAGE mMACHINE T7 kits (Ambion). The RnCaV1.2, RnCaV2.1, DSC1, BSC1, RnCaVα2δ1, and Rn-CaVβ2a channel subunits had the following Genbank accession nos. M67515.1, M64373, AAK01090.1, ABF70206, NP_037051, and NP_446303.1, respectively. Sequence alignment was performed with Vector NTI Manager (In Vitrogen). The CaV2.1 chimera, harboring the CaV1.2 C-end (starting 10 amino acids after the end of the S6 segment of domain IV, AAAAC mutant) was previously described (Cens et al., 2006). Oocyte preparation and injection were also performed (Rousset et al., 2017; Cens et al., 2013, 2015) and carried out in strict accordance with the recommendations and relevant guidelines of our institution. Surgery was performed under anesthesia and efforts were made to minimize suffering. The experimental protocols were approved by the Direction Départementale des Services Vétérinaires (authorization N° C34.16). Briefly, oocytes were surgically removed from anesthetized (using-MS 222, Sigma-Aldrich ref A5040, at 0.2 %) female Xenopus and dissociated with collagenase 1A (ref C9891; Sigma-Aldrich) at 1 mg/ml in a low Ca2+ solution (in mM: NaCl, 82; KCl, 2; MgCl2, 1; HEPES, 5; pH = 7.2 with NaOH) for 1.5 h. Batches of ∼30 oocytes were then isolated and individually pressure-injected (50–300 ms at 1 bar) with ∼30–50 nl of RNA (or cDNA) at a concentration of 1 µg/μl. Injected oocytes were kept at 19°C in NDS solution (in mM: NaCl, 96; KCl, 2; CaCl2, 1.8; MgCl2, 1; HEPES, 5; Na-pyruvate, 2.5; gentamycin, 0.05; pH = 7.2 with NaOH) for 1–3 days before recordings.

Electrophysiological recording on Xenopus oocytes

For two-electrode voltage-clamp recordings of Xenopus oocytes, pipettes (GC150T10; Harward Electromedical Instruments) of 0.2–1 MΩ were filled with 3 M KCl. The basic recording solution (BANT10) contained (in mM) BaOH, 10; TEAOH, 20; NMDG, 50; CsOH, 2; and HEPES, 10; pH 7.2 with methanesulfonic acid. In the Ca2+-containing solution, 10 mM CaOH replaced 10 mM BaOH. The low divalent solution was prepared with (in mM) Na-acetate, 100; MgCl2, 2; HEPES, 5; and EGTA, 4; pH 7.2 with NaOH, in which the free Ca2+ concentration was evaluated to be <12.5 nM (WinMaxC, from Chris Patton, Stanford University, Pacific Grove, CA, USA; [email protected]). The amplifier (Geneclamp 500B, Axon Instruments) was connected to the bath using the virtual ground head stage and agar–KCl (3M) bridges. Junction potentials between the pipettes and the Bant10 solutions were nulled with the two pipettes in the bath before impalement. The holding potential was set to −80 or −100 mV, and depolarizing pulses of different amplitudes and durations were applied every 10 or 15 s, unless otherwise mentioned (see below). The leak and the capacitive transients due to the oocyte membrane were subtracted online using a P/5 procedure of the pClamp acquisition software (Axon Instruments, version 7.0, Molecular Devices). Current traces analysis was performed with Clampfit (Axon Instruments version 10, Molecular Devices).

Online BAPTA injection

After the impalement of the two electrodes and the establishment of the voltage clamp, a third microelectrode filled with 100 mM BAPTA and 5 mM HEPES (pH 7.2 with KOH) and connected to a home-made injector was impaled. One or two pressure pulses (2 bars, 100 ms) were given to inject 10–30 nl of the BAPTA solution and the evolution of the inhibition of the Ca2+-activated Cl current was followed on the tail current recorded after a voltage pulse to 0 mV (holding potential −100 mV, see Fig. 1 A). Chelation of Ca2+ was considered complete when the tail current almost disappeared (<10% of its initial value). Switching from a Ba2+ to a Ca2+-containing solution induced an almost instantaneous change in current inactivation (Fig. 1 B) without any modification of the tail current, indicating the lack of contamination by the Cl conductance.

Current analysis

Inactivation kinetics were quantified using three methods: calculation of R400 (see Fig. 1 B) or fitting the inactivation phase with decreasing mono- or biexponential equations for short (400) or long (2.5 s pulses), respectively, using Clampfit ver. 10. (Axon Instruments). The equation used for fitting the inactivation phases of the current was for a mono-exponential decay:
with I, the current amplitude; t, the time; A and τ the amplitude and time constant of the exponential, respectively; and C, the amplitude of the non-inactivating current.
For biexponential decay,
where I, the current amplitude; t, the time; A1, A2, and τ1, τ2, the amplitudes and time constants of the slow and fast exponential components, respectively; and C, the amplitude of the non-inactivating current.
The current–voltage curves were obtained by applying 400-ms long depolarizations of −70 to +50 mV (with 10 mV increment) at a frequency of 1/15 s from a holding potential of −80 or −100 mV. The normalized-current voltage curves were fitted using the equation:
where I and Imax are the peak-current amplitudes recorded during the voltage step V and at the peak of the current–voltage curve, respectively; G is the normalized macroscopic conductance; Erev, is the reversal potential; Eact is the half activation potential; and kact is the slope factor.
The normalized isochronal (2.5 s) inactivation curves (also called steady-state inactivation curves) were obtained by applying conditioning depolarizing pulses of 2.5 s in duration and of −80 to +40 mV amplitudes (by 10 mV increment) just before a 400 ms test pulse at +10 mV. The normalized current amplitudes measured during the test pulse were plotted against the voltage of the conditioning pulses. The curves were then fitted using the equation
where I and Imax are the peak-current amplitudes recorded during the voltage step at +10 V for conditioning voltage V and at −80 mV, respectively; R is the fraction of channels that does not inactivate; Ein is the half inactivation potential, and kin is the slope factor.
The reactivation curves were obtained by applying a first 400-ms long depolarization to 0 mV, followed by a return to the holding potential during various durations (t) and a second, 100-ms long, depolarization to 0 mV. The current amplitude recorded during the second depolarization (I2) was normalized according to the current amplitude recorded during the first depolarization (I1) and plotted against the duration t. The curve was then fitted using the following equation:
where R is the proportion of the current that was not inactivated during t, and τ is the time constant of reactivation.
For the cation concentration–conductance curves, the data points were fitted using a hyperbolic Langmuir isotherm function of the following form:
where I is the peak current of the current–voltage curves (using voltage ramps) recorded with a solution at different Ba concentrations; I2.5 is the peak current of the current–voltage curve recorded when [Ba] = 2.5 mM; Imax is the estimated maximum current amplitude; Kd is the concentration for half-maximum effect; and [Ba] is the extracellular Ba concentration.
Dose–response curves for Cd2+ or Ni2+ were fitted using the dose-response equation of the following form:
where I2 is the current amplitude measured in the presence of different concentrations of inhibitory divalent cation (Cd2+ or Ni2+); I1 is the current amplitude measured in the absence of inhibitory cation; R is the fraction of the current that is not affected by the inhibitory cations (close to 0 in these cases); pIC50 is the log of the inhibitory dose at 50% (IC50); X is the log of the inhibitory cation concentrations; and p is a slope factor.

All values were stored in Excel (Microsoft, Office 16), graphs, and their fits were done with Origin (Microcal software, ver 6.0). Statistical tests were performed using Sigmaplot 12.2, and final figures were assembled using Lotus Freelance Graphics 1997 edition. All averaged values are given as mean ± SEM (standard error of the mean). Student’s t test or Mann–Whitney rank sum test, when normality test failed, or ANOVA were used to assess the differences between mean values, with a statistical significance noted in the figures: * = P < 0.05, ** = P < 0.01, *** = P < 0.001.

Xenopus oocytes were injected with in vitro synthesized RNA encoding the Am-CaV4 channel isoform previously described (Gosselin-Badaroudine et al., 2016). Voltage-clamp recordings were made 1–3 days later using 10 mM Ba2+ as a charge carrier in the extracellular solution. Large inward currents can be recorded in these conditions (see Fig. 1 A), with amplitudes reaching several µA. Such currents induce, most of the time, the appearance of a contaminating Ca2+- or Ba2+-activated Cl current during both the voltage steps to 0 mV and during the repolarization to −100 mV (seen as a notch and a large tail current, respectively, in the current time-course; see trace in red in Fig. 1 A). This current cannot be efficiently blocked by prior injection of either EGTA, EDTA, or even BAPTA. In our hands, the only procedure that allows us to precisely inhibit this contaminating Cl current is to perform, at the beginning of the recording, an injection of 40–80 nl of a solution containing 100 mM BAPTA and 5 mM HEPES. In these conditions only, we can follow both the inhibition of the contaminating Cl current during both the pulse and the repolarization (Fig. 1). This inhibition required 12–15 min to be complete, as seen by the measurement of the tail current amplitude in the Ba2+ solution after BAPTA injection (trace in blue, Fig. 1 A). When the steady-state effect of BAPTA is reached, the perfusion of the Ca2+-containing solution does not cause any kind of contamination, during either the pulse or the repolarization (see Fig. 1 A). Online injection allows repetition of BAPTA injections if the [Ca]i is not properly kept at a low value.

In these conditions, Ba2+ and Ca2+ currents can be recorded without any contamination (Fig. 2 A), and current–voltage curves and isochronal inactivation curves can be obtained (Fig. 2, B and C). Notably, the potential for half-activation obtained in 10 mM Ba2+ (Eact = −11 ± 1 mV; n = 40; see Table 2 for activation and inactivation parameters) distinctly defines CaV4 as a high voltage–activated Ca2+ channel. The average current amplitude at +10 mV (−2.2 ± 0.3 µA, n = 47) 2 days after RNA injection is not affected by the co-expression of Apis mellifera of NaV (TEH1-4) or CaV (CaVβ) auxiliary subunits (not shown, but refer to Gosselin-Badaroudine et al., 2016). Perfusion of 10 mM Ca2+ instead of Ba2+ induces a shift of the current–voltage curves toward positive voltages by ∼10 mV (Eact varies from −11 to −2.5 mV, Table 2), as expected by a better screening of the surface charges by Ca2+, and decreases the peak current amplitude to 1.6 ± 0.3 µA (n = 30). However, Ca2+ has two unexpected effects on channel properties: (1) it does not shift the steady-state inactivation curves, as the curves in Ba2+ and Ca2+ superimposed almost perfectly (see Fig. 2 C and Table 1) and (2) it reduces significantly the speed of inactivation, producing an unusual Ca-dependent inactivation with slower kinetics in Ca2+ than in Ba2+ (R400 at 0 mV = 0.18 ± 0.03; n = 23 and 0.35 ± 0.03; n = 17, in Ba2+ and Ca2+, respectively; see Fig. 1 B for R400 calculation and also Fig. 2 A and Fig. 4 B). This slowing of inactivation occurs rapidly and is completely reversible (Fig. 1 C). While inactivation is slowed by Ca2+, changing Ba2+ for Ca2+ does not change the reactivation time course (Fig. 2 D) with the time constants needed to reactivate the channel being almost similar in Ba2+ and in Ca2+ (see Table 2 and Fig. 2 E). Interestingly, the potential for half-inactivation of CaV4 (Ein = −49 ± 0.6 mV, see Table 2) was more hyperpolarized than that of Ca2+ channels (CaV1.2 or CaV2.1), but more depolarized than that of NaV channels (NaV1.5), when recorded in similar conditions, Ein = −23, −38, and −80 mV, respectively (see Table 3).

Previous works on Am-CaV4 have evidenced a particular selectivity profile with a high permeability ratio for Ca2+ over Na+ (Gosselin-Badaroudine et al., 2016). As expected, the relationship between the current reversal potential and the extracellular Ba2+ concentrations strictly obey the Nernst law for divalent cations with a slope of 29.5 mV (Fig. 3 A). Moreover, the concentration–conductance curve for Ba2+ gives a Kd of 1.9 ± 0.4 mM (Fig. 3 B), which is similar to the value found for the mammalian L type CaV1.2 channel: 1.9 ± 0.4 mM from single channel recordings (Guia et al., 2001) or 1.1 ± 0.1 mM in our conditions (not shown). When mixtures of different Ba2+/Ca2+ mole fractions were perfused, CaV4 displayed the usual anomalous mole fraction effect (Fig. 3 C for traces and 3 D for graph), consistent with what had been observed with mammalian CaV1.2 or CaV2.1 Ca2+ channels (Cens et al., 2007). Indeed, the current amplitude displays a marked minimum for low Ca2+ mole fractions (see the 2 mM Ca2+ condition marked with an arrow in Fig. 3 D, for example), suggesting that the Ca2+ binding sites within the channel have a higher affinity for Ca2+ than for Ba2+.

Perfusion of solutions containing either Ba2+, Sr2+, Ca2+, or Na+ allows depicting more precisely the permeation profile of the channel by calculating averaged current ratios: Ba/Ca = 1.92 ± 0.1 (n = 32) and Ba/Sr = 0.98 ± 0.07 (n = 5), with almost no Na+ current flowing through the channel (Fig. 3, E and F). By measuring the reversal potential measurement in different Ba2+ and Ca2+ conditions, the permeability ratio PBa/PCa can be calculated as 0.54, a value slightly larger than the PBa/PCa determined on the mammalian L-type Ca2+ channel (0.4 [Hille, 1992b]).

As evidenced in Figs. 1 and 2, the switch from the Ba2+-containing to the Ca2+-containing solution produces a marked slowing of the inactivation time course, better seen when the traces are normalized, as displayed in Fig. 4 A. This slowing is obtained in less than a second after exchanging the extracellular solution (Fig. 1 C), suggesting a direct impact of the cation on the channel protein itself, and not the involvement of secondary messengers (such as kinases, phosphatases, or PIP2 degradation or synthesis, for example). The slowing can be evaluated by measuring R400, the ratio of the current recorded 400 ms after the start of the depolarization over the current recorded at the peak amplitude (see Fig. 1 B), and R400 values were significantly smaller for Ba2+ than for Ca2+ currents (Fig. 4 B). However, interestingly, neither the current activation time-course nor the time-to-peak current was significantly different in Ba2+ or Ca2+ (Fig. 4 C). Analysis of the current inactivation kinetics during 2.5-s long depolarizations reveals a biexponential time course and shows that this slowing of inactivation occurs without any modification of the values of the two time constants, but rather results from a significant increase in the proportion of the slow time constant of inactivation (Fig. 4 D, left and right). Interestingly, both time constants exhibit almost no voltage-dependence.

This effect of Ca2+ on CaV4 channel inactivation is puzzling. NaV channel inactivation is mediated by the loop located between domains III and IV, which contains a specific sequence of three amino acids—IFM—crucial for rapid inactivation (Kontis et al., 1997; West et al., 1992). NaV channels also harbor potential binding sites for Ca2+ and/or calmodulin (CaM, a Ca2+-binding molecule) within the loop III–IV region and on their C-terminal tail that could be involved in Ca2+-dependent regulation of NaV gating (Johnson, 2020; Salvage et al., 2021), although defining this regulation has proven to be challenging. On the other hand, L-type Ca2+ channels exhibit a well-documented phenomenon known as Ca2+-dependent inactivation (CDI). This occurs when incoming Ca2+ ions bind to a CaM molecule attached to the L-type Ca channel via CaM-binding sites located on the C-terminal tail of the channel (preIQ and IQ domains) and produces a marked acceleration of the Ca2+ current inactivation (Peterson et al., 1999; Tadross et al., 2008; Budde et al., 2002; Ben-Johny and Yue, 2014; Ben-Johny et al., 2014). NaV and CaV channels also possess putative Ca2+-binding sequences (EF-hands) on their C-termini, playing a role in the CDI (at least for CaV1.2 channels), although their capacity for binding Ca2+ ions is debated. CaV4 channels do possess many of these sequences that are involved in NaV and CaV Ca2+ sensitivity. Putative CaM-binding sites are present on the loop III–IV and the C-terminal tail of the CaV4 channel sequence (see Fig. 5, A and B). An MLF sequence, homologous to the IFM sequence found in NaV channels, is also present on the CaV4 sequence, at a localization similar to the IFM in NaV channels, i.e., on the loop III–IV (see Fig. 5 A), and an EF-hand sequence can be found in their proximal C-termini. To investigate the impact of these regions on CaV4 inactivation, we opted to create three CaV4 mutants: one in which the MLF sequence was mutated to AAA (CaV4(6)), another one in which the C-terminal tail was replaced by that of CaV1.2 (CaV4(8)), and a third in which both of these mutations were combined (CaV4(10)). For a visual representation of these mutations, please refer to Fig. 6, B–D. The CaV4(6) mutant does display a complete loss of inactivation at all voltages (Fig. 6, B and E; and Tables 2 and 4 for values), while in the case of the CaV4(8) mutant, the steady-state inactivation parameters are slightly modified compared with those of CaV4 (Fig. 6 C and Tables 2 and 4), although the inactivation time course is slowed. The double mutant, with the modified IFM and C-terminus (CaV4(10)), behaves like the CaV4(6) mutant and shows a complete loss of inactivation. The fact that in these three mutants, the shape of the current–voltage curve (half-activation potentials, slopes of activation, and reversal potentials (see Fig. 6 F and Table 5) is similar to control values demonstrates that the global functioning of the channel (voltage-dependency of activation and channel selectivity) is preserved and that only inactivation is affected. These results strongly indicate that both sequences are central to the voltage-dependent inactivation of CaV4 and that the MLF sequence undeniably plays a role in CaV4 similar to that played by IFM sequence in NaV channels. Interestingly, with any mutants, the Ca2+-dependent slowing of inactivation observed with wild-type CaV4 vanished (as depicted in Fig. 7, A–C), and the R400 values are similar in Ba2+ and Ca2+ (Fig. 7 B). Analysis of the CaV4(8) inactivation time constants (the only one that displays a measurable inactivation) reveals that the values are very close in Ba2+ and in Ca2+ and close to the values obtained with the wild-type channel (Fig. 4 D and Fig. 7 C). The changes in inactivation for this mutant can be best explained by a change in the proportion of the slow time constant that is clearly larger than that of the wild-type channel at any potentials (∼0.4 versus ∼0.15, Fig. 4 D and Fig. 7 C). However, as opposed to the wild-type, changing Ba2+ for Ca2+ does not modify these values drastically, as seen in Fig. 4 C, explaining the fact that the R400 values (taken at 0 mV) were identical in these two conditions (Fig. 7 B).

To be sure that this cation-dependent inactivation (CatDI) was due to cation binding onto an intracellular target and not on a binding-site located extracellularly, we analyzed the Ca-dependency of the outward current. Outward currents in Ba2+ and Ca2+ solutions (probably carried by K+) were then recorded during depolarization to 0 or +70 mV. While the inward current recorded at 0 mV in Ca2+ displays the usual slowing of inactivation (when compared to Ba2+ conditions), the outward current recorded at +70 mV was similar in Ca2+ and Ba2+ (see Fig. 8 A for current traces and 8 B inset for normalized current traces). Statistical analysis of the inactivation time constant at both of these two potentials confirmed the Ca2+ does not affect current inactivation for positive potential (Fig. 8 B).

We then sought to investigate the role of calmodulin in CatDI of CaV4 by coexpressing either wild-type CaM or a mutant invalidated in the four Ca2+ binding-sites (CaM1234). The results are displayed in Fig. 9, A and C, and demonstrate that CaM does not play any role in CatDI, since neither CaM nor CaM1234 affect the Ca2+-induced changes in the R400 values.

Moreover, the fact that the mutant CaV4(8), which exchanges its C-terminus with that of CaV1.2, does not exhibit a CatDI, while the same substitution in CaV2.1 results in the appearance of the clear real CDI (AAAAC in Fig. 9, B and C) strongly implies that the mechanism of CatDI is distinct despite the involvement of sequences located at similar positions in the channels’ primary sequence. A control experiment of the effect of CaM1234 on L-type Ca2+ channel CDI is shown in Fig. 9 D, where CDI is strongly affected by the mutated CaM.

We concluded this characterization by pharmacological profiling of the CaV4. The dose–response inhibition curves with Cd2+ and Ni2+ show 10-fold greater sensitivity to Cd2+ (respective IC50 of 31.9 ± 2.4 and 433.1 ± 27.3 µM, Fig. 10 A), a specificity commonly observed for high-voltage-activated Ca2+ channels (Hille, 1992a). The use of a panel of insecticides (permethrin, allethrin, ivermectin, Picrotoxin, fipronil, chlorantraniliprole, and chlothianidin) of CaV3 (mibefradil, NCC-55-0396, TT-A2, amiloride), CaV1 modulators (nifedipine, PN200-110, Bay-K8644, verapamil, diltiazem), and of toxins and venom (SNX-482, atrachotoxin HV1a, Thomisus honustus and Synema globosum venoms, and PTx-II) was quite ineffective, except for diltiazem, which is responsible for significant inhibition of the peak current at 20 µM (49 ± 3%, n = 3, Fig. 10 B).

We also investigated the effect of several NaV channel modulators. Some of them, such as the pyrethroids (see Fig. 8 B) and TTX (see Gosselin-Badaroudine et al., 2016) were infective, but veratrine, an alkaloid mixture extracted from the plant Veratrum album, induced a small inhibition of the current amplitude (13% ± 0.3, n = 12, when perfused at 30 µM), and a well-resolved acceleration of current inactivation. Indeed, R400 decreases to 50% of its control value (R400 = 0.2 ± 0.03 [n = 12] to 0.10 ± 0.01 [n = 12] for control and veratrine, respectively, Mann–Whitney rank sum test P value = 0.004, see Figs. 10 and 11). At a stimulation frequency of 0.5 Hz, this acceleration develops with a time-constant of 8.2 s, a little faster than the time-constant required for the change in current amplitude (17 s, not shown). The acceleration of inactivation was almost fully reversible, while the effect on the current amplitude was not (Fig. 11 B). A similar effect was obtained in the presence of Ca2+ in the perfusion medium instead of Ba2+ (Fig. 12). During these short depolarizations, a condition necessary to carry out dose-response curves of veratrine, the inactivation time course can be better estimated by a fit with a single exponential. Increasing veratrine concentration decreases the CaV4 inactivation time constant in a dose-dependent manner (Fig. 11, A–C), and a similar effect was also obtained with the mutant channel CaV4(6). While the CaV4(6) inactivation time constant was too large to be effectively measured for low doses of veratrine, the perfusion of high doses increases the inactivation up to the values recorded for the wild type at similar doses (at 100 µM, see Fig. 11 C). Interestingly, outward current time constants are smaller than the inward current but do not vary between 40 and +110m mV (not shown). On outward currents, veratrine has a tendency to decrease the time constant of inactivation down to values close to that obtained for the inward current (59.5 ± 2.5 and 58.9 ± 5.7 ms for control and veratrine traces at +70 mV, respectively, see histogram Fig. 13), but this effect does not reach a significant level (Mann–Whitney rank sum test P value = 0.057). Similar effects were produced with veratridine, which is the major component of veratrine (not shown), and the effects were undoubtedly different from those produced by veratrine or veratridine on NaV channels, for which a marked slowing of current inactivation and deactivation were reported without noticeable effects on the peak current (Sutro, 1986; Sigel, 1987), and the effects that were also retrieved with the honeybee NaV channel expressed in Xenopus oocytes (not shown). Altogether, these effects seem to indicate that the veratrine-induced acceleration of inactivation is independent of the usual MFL-gated inactivation and involves other mechanisms. In the absence of other specific biophysical or pharmacological markers of CaV4, veratrine could be used as a pharmacological tool to reveal the presence of CaV4 in situ, in muscles or neurons for instance.

To achieve a more comprehensive understanding of cellular excitability in honeybees and their sensitivity to pesticides, our research group has been actively engaged in identifying and cloning ion channels expressed in larvae and adult bees. We have previously published preliminary characterizations of the ligand-gated RDL, glutamate and nicotinic receptors, and the voltage-gated CaV1, CaV2, CaV3, CaV4, and NaV1 channels (Cens et al., 2015; Gosselin-Badaroudine et al., 2017; Cens et al., 2013; Gosselin-Badaroudine et al., 2016; Rousset et al., 2017; Gosselin-badaroudine et al., 2015; Collet et al., 2016; Brunello et al., 2022). CaV4, a channel homologous to Drosophila DSCI, did not conform to the classical NaV channel archetype, and neither its biophysical nor pharmacological properties allowed us to classify this channel in any of the three already established CaV channel subfamilies (CaV1, CaV2, and CaV3). Consequently, we proposed the existence of a novel class of Ca2+ channel termed CaV4 (Gosselin-Badaroudine et al., 2016). In this study, we provided evidence demonstrating that the CaV4 channel exhibits distinctive characteristics with regard to cation permeability, voltage dependency, Ca2+-dependent inactivation mechanisms, and pharmacological properties.

CaV4 channel permeation and selectivity

The results depicted in Figs. 2 and 3 unequivocally demonstrate that CaV4 is a high-voltage activated Ca2+ channel that permeates both Ca2+ and Ba2+ and exhibits an anomalous mole fraction between Ba2+ and Ca2+, suggesting a higher selectivity for Ca2+. Moreover, the CaV4 saturation curves for divalent cations (Fig. 3) and the higher sensitivity to Cd2+ versus Ni2+ are typical of HVA Ca2+ channels (Guia et al., 2001). The shift of the current–voltage curve observed upon extracellular cations changes (Fig. 2) is also reminiscent of that observed in mammalian CaV1.2 or CaV2.1 Ca2+ channels, for example.

Na+ ions were unable to permeate CaV4 even in the presence of EGTA in the extracellular solution, consistent with our previous findings (Gosselin-Badaroudine et al., 2016). While channels harboring a Ca2+ or Ba2+ channel permeability with DEEA selectivity filters have been described in Cnidaria (Gur Barzilai et al., 2012), Drosophila, and cockroach (Zhou et al, 2004), these channels retained a permeability to Na+ in the absence of Ca2+ or Ba2+, which is not the case here.

The sequences of the SF at both loci (EEEE and DCS) identified in NaV and CaV channels (Heinemann et al., 1992; Cens et al., 2007; Neumaier et al., 2015) are unique in CaV4, with three negative charges at the EEEE locus, DEEA (numbered position 0) and four negative charges at the DCS locus (DEED; see Table 1 for the SF sequences of different CaV and NaV channels). The presence of these negative charges pointing toward the pore at the DCS locus (Wu et al., 2016) is known to affect the selectivity for monovalent cations in NaV channels (Heinemann et al., 1992) and lightly those of divalent cations in the CaV. Therefore, the fact that four negative charges are found in CaV4 but not in the DSC1, BSC1, or other NaV or CaV channels (Table 1) could explain the differences in Na permeability between these channels. The aspartate at position +1 (relative to the E in the SF of domain II) and the glutamate at position 0 of domain IV (conserved in all CaV channels [Abderemane-Ali et al., 2019; Shaya et al., 2014] but replaced by tryptophan and alanine, respectively, in CaV4) may also play a role in this particular selectivity. Substituting the (+1)-aspartate by non-charged amino acids in CaV1.2 did not alter the reversal potential in the presence of divalent cations (Abderemane-Ali et al., 2019) but affected the channel kinetics. Whether this aspartate impacts the permeation of monovalent cation has not been examined.

The observation that divalent cations can permeate, whereas smaller monovalent Na+ ions (respective size of 1.35 Å for Ca2+, 0.99 Å, for Ba2+ and 0.95 Å for Na+) cannot do so in their absence could also be attributed to the blockade of the inward monovalent current by extracellular Mg2+, present in the low Ca solution and not effectively chelated by EGTA.

Positions 0 (EEEE locus), +1 (aspartate), and/or +4 (DCS locus) are thus pertinent candidates for the strict selectivity of CaV4 to divalent cations. However, the definitive proof of their involvement in specific cation coordination in the CaV4 pore will require additional mutagenesis studies. In any case, the CaV4 selectivity pattern is a novel feature in the CaV family, and future structural studies on CaV4 may provide important new information on the role of these amino acids in the formation and properties of the cation binding sites and the Ca2+ coordination in CaV4 pore.

CaV4 channel inactivation

The sequence similarities between CaV4 and NaV channels extend to the loop connecting domains III and IV and, in particular, to the IFM sequence known to be crucial for NaV fast inactivation. Hydrophobic interactions between the IFM sequence and the S4–S5 linkers of DIII and DIV and part of the activation gate at the bottom of the DIV–S6 stabilize the pore in a closed-inactivated conformation (a mechanism formerly called hinged-lid [Catterall, 2013; Jiang et al., 2020; Liu et al., 2023]). The IFM sequence is replaced by MFM and MFL in honeybee NaV and CaV4 channels, respectively. CaV4 inactivation kinetics (100–1,000 ms) are slower than those of honeybee or mammalian NaV channels (2–10 ms [Fux et al., 2018]) and are not affected by the membrane potential. Moreover, CaV4 steady-state inactivation appears to be insensitive to surface charge density, as opposed to the voltage dependence of activation. Inactivation was also noticeably slower in Ca2+ compared with Ba2+. All of these features are specific to CaV4 and are not observed in other CaV or NaV channels (see Figs. 2, 3, and 4).

Interestingly, the MFL to AAA substitution (as the CaV4(6) mutant, and as shown in Fig. 6) completely abolishes fast inactivation, suggesting a potential shared mechanism with NaV channels (Zhou et al, 2004; Catterall, 2013). The observation that the inactivation kinetics are not voltage-dependent can be explained by a voltage-independent gating step that acts as the rate-limiting factor in the inactivation process. However, the insensitivity of the steady-state inactivation to changes in the extracellular divalent cation concentration is a more challenging puzzle to unravel.

It has been demonstrated that the voltage-dependent parameters of voltage-gated ion channels are responsive to an elevation of extracellular cation concentration (see Neumaier et al., 2015 for review). According to the Gouy-Chapman-Stern model, fixed negative charges present at the surface of the membrane and/or the channel create a local negative surface potential that is reduced by increasing extracellular divalent cations, resulting in a positive shift of the voltage-dependent parameters (Hille, 1992a; Gilbert and Ehrenstein, 1969; Neumaier et al., 2015). In this theoretical framework, the four voltage-sensing domains (VSD) of a channel should be affected, i.e., all the voltage-dependent parameters, including the voltage-dependence of activation and inactivation, should be sensitive to the changes in the surface potential. In NaV and CaV channels, this prediction is validated (see Table 3). In CaV4, both voltage dependency and kinetics of inactivation should be under the control of at least the IFM motif and the S4 helices of domain III and/or IV (Capes et al., 2013; Angsutararux et al., 2021). The observation that only inactivation is not influenced by this surface charges screening strongly suggests a highly localized effect on the S4 of domains III and/or IV (Capes et al., 2013; Angsutararux et al., 2021; Lewis and Raman, 2013). This effect is not observed with NaV or CaV channels (see Fig. 5) (Neumaier et al., 2015). Interestingly, the S4 helices of domains III and IV of CaV4 exhibit a high degree of conservation with those of NaV or CaV channels, featuring the same number of positive charges at similar positions.

Alternatively, alteration of either the interactions of the S4s with other channel helices during gating or modifications of the fixed charges at the surface of the channels both in terms of their quantity and arrangement may be questioned instead of focusing solely on S4 itself. Indeed, interactions involving voltage-sensor in domains III and IV, with residues surrounding pore sequence (S4–S5 linkers in domains III and IV and the III–IV loop, S3 helices negative charges), are known to modulate both voltage dependence and kinetics of activation, and inactivation in NaV and CaV channels (Fernández-Quintero et al., 2021; Lewis and Raman, 2013; Capes et al., 2013; Angsutararux et al., 2021; Hsu et al., 2017). These interactions could potentially be responsible for the fact that the potential for half-activation and inactivation of CaV4 are more hyperpolarized compared with other mammalian CaV channels such as CaV2.1 or CaV1.2, see Tables 1 and 2), and depolarized when compared with the NaV channels. However, whether they can modulate the sensing of a local electric field by the S4, or the properties of this field as referred to the existence of the surface potential, is not known and will require additional experimental work to be understood.

This insensitivity may also be related to modification of the electrostatic surface potential in the surrounding of these two voltage-sensors either by a modification in the number of the charges at their surface or a change in the affinity of these surface charges to divalent cations (Elinder et al., 1996; Madeja, 2000) producing either saturation at low concentration or insensitivity to binding. The number and position of the negative or positive charges in the S3–S4 segment of NaV1.5, CaV4, or CaV1.2 for example are different. However, the precise molecular structure of CaV4 is not known. Calculating the real surface potential locally, in the vicinity of these VSDs is therefore highly challenging and speculative. Finally, one cannot exclude the possibility of similar electrostatic surface potential to other channels partially, or totally, counterbalanced by a direct effect of the incoming Ca2+ ion on the VDI. In CaV1.2 channel, it has been demonstrated, for example, that mutation of a single aspartate residue in the selectivity filter of CaV1.2 (D707 at position +1, after the conserved E of domain II) can remove CDI without any other effect on the channel properties (Abderemane-Ali et al., 2019), demonstrating the role of intrapore cation-binding site on inactivation. Unfortunately, studies looking at the effect of the divalent cations on the role of these DIII–DIV voltage sensors are lacking. The molecular mechanism of CatDI will therefore require additional experiments to be fully explained.

CaV4 channel CatDI versus CaV1.2 CDI

Another surprising observation is the slowing of inactivation when Ca2+ is the charge carrier (called here CatDI). The CDI of CaV channels has been well described (Budde et al., 2002; Ben-Johny et al., 2014), and produces, on mammalian CaV1 or CaV2 channels, faster inactivation kinetics. In fact, in CaV1 or CaV2 channels, this Ca dependency appears to be more a Ca2+ calmodulin–dependent regulation of the classical voltage-dependent inactivation rather than a completely distinct mechanism (Cens et al., 1999; De Leon et al., 1995; Budde et al., 2002). Calmodulin bound to the CaV1.2 C-terminal tail constitutes the Ca-sensing element able to detect, very locally, incoming Ca2+ ions during channel opening and to trigger channel CDI. An EF-hand structure is also important for this process, although it is proposed not to be able to bind Ca2+ ions, but rather to play a role in CaM binding (Gardill et al., 2019). The displacement of CaM from one binding site to another on the channel may allow or accelerate a pre-existing voltage-dependent inactivation, thus producing the so-called CDI (Cens et al., 1999, 2006; De Leon et al., 1995; Budde et al., 2002). In NaV channels, while these Ca2+/CaM binding structures (IQ and EF-hand sequences) are also present on the C-terminal tail and the III–IV loop (Salvage et al., 2021), the Ca2+-dependent regulation of gating is translated by a decrease in the peak current and/or a shift in the inactivation curve, depending on the expressed NaV channels (Pitt and Lee, 2016; Salvage et al., 2021; Ben-Johny et al., 2014). In contrast, a clear effect on the inactivation kinetics has rarely been described, probably due to the lack of Ca2+ permeability of the NaV channel. These regulations have thus been analyzed by varying the bulk intracellular Ca2+ concentration, which also induced activation of intracellular regulatory pathways (CaM kinase, PKC…). This is not the case for CaV4, for which the slowing of inactivation by Ca2+ occurs in <2–5 s (see Fig. 1), which strongly suggests a direct effect of Ca2+ on CaV4. While this slowing clearly requires an inward flux of Ca2+, since it is not seen on outward current (Fig. 6), it does not have the hallmark of CDI, i.e., it is not proportional to the amplitude of the Ca2+ influx (see Fig. 4 D), and, as opposed to the classical CDI, CaM does not seem to be involved (Ben-Johny and Yue, 2014; Peterson et al., 1999). Loss of CatDI with the CaV4(8) chimera suggests, however, that the C-terminal tail, in one way or another, may be involved. We conclude that CatDI is clearly different from CDI. The aspartate at position +1 relative to SF in domains II (D707), crucial for CaV1.2 CDI (Abderemane-Ali et al., 2019), is conserved in all mammalian CaV channels but not in NaV or CaV4. The role of this substitution in the lack of CDI in CaV4, harboring CaM, CaM-binding sites, and the EF-hand motif in their sequence needs to be further studied. Interestingly, still in CaV1.2, mutation of the glutamate in the SF of domain III has the same effect as D707 and completely suppresses CDI. In CaV4, the glutamate at the SF of domain III is conserved, but not that of the SF at domain IV. Alteration of this Ca2+ binding site within the CaV4 SF could therefore drive the change not only in the cationic channel selectivity, as suggested earlier, but also in the Ca2+ dependence of the inactivation mechanism. Mutagenesis experiments are in progress to explore this possibility. All of these properties make CaV4 CatDI a novel mechanism regulating Ca2+ influx.

CaV4 channel pharmacology

CaV4 insensitivity to neonicotinoids and phenyl insecticides is not surprising. However, due to the amino acid sequence similarity with the NaV channel, a potential susceptibility to pyrethroids could be expected. This was not the case either, and a close examination of the amino acids reveals that key positions for knockdown resistance in Drosophila NaV channels (specifically in S3–S4 DII [M918], S6 DII [L1011], and S6 DIII [F1534]; Rinkevich et al., 2013) are occupied by amino acids that confer resistance (I1202, M1297, and C1830, respectively) in CaV4.

The only molecules that were active on the CaV4 channels were a CaV1 channel antagonist, diltiazem (which decreases current amplitude), and a NaV channel regulator, veratrine (which accelerates inactivation kinetics; see Figs. 7 and 8). The amino acids that have been shown to be implicated in diltiazem binding within the pore of the CaV1.1 channel (Zhao et al., 2019) and implicated in the effects on current amplitude (Tyr1365, Ala1369, and Ile1372 of S6DI [Hering et al., 1996]) are not conserved in CaV4, although the inhibition of the peak current is similar. In the CaV4 channel, the benzothiazepines binding site is therefore not completely conserved. We are currently working on pinpointing the channel’s susceptibility to diltiazem, and identifying the specific amino acids involved in these binding sites.

The effect of veratrine is unexpected: it leads to an acceleration of the inactivation kinetics on CaV4, which contrasts with its opposite effect recorded on NaV channels, where it slows the inactivation and deactivation kinetics. Interestingly, this acceleration is also observed with the mutant that lacks inactivation, (CaV4(6), Fig. 11), and both processes seem to converge toward similar values (Fig. 11 C). Veratridine, the major component of veratrine has similar effects on CaV4, but slows honeybee NaV channel inactivation and deactivation. On the mammalian NaV1.5 channel, two binding sites for veratridine have been proposed: (1) in the pore at the level of the SF, and (2) at the intracellular mouth of the pore, at the bottom of the S6 (Gulsevin et al., 2022). A mutagenesis analysis revealed that the most probable binding site is site 2, in which amino acids L409, E417, and I1466 would be the most important for the stabilization of the ligand. These amino acids are conserved at homologous positions in CaV4, and molecular docking of veratrine on an alpha-fold model of CaV4 shows poses of the molecule at this site (not shown), suggesting that the same or very close binding sites on NaV1.5 and CaV4 can produce opposite effects, as already seen for the CaV1.2 Ca2+ channel and dihydropyridines (Zhao et al., 2019). Amino acids at equivalent positions in domains I, II, III, and IV of NaV, CaV channels, or subunits of Kv channels have been shown to be central for the formation of the activation gate and for the development of fast and slow C-type inactivation (Shi and Soldatov, 2002; Chancey et al., 2007; Liu et al., 2023). The fact that the CaV4(6) mutant, which lacks inactivation, is also sensitive to veratrine with the same functional effects has led us to conclude that this acceleration of inactivation does not depend on the MFL motif but requires another molecular mechanism. One may speculate that upon binding at the bottom of one or multiple S6 segments, veratrine could induce a structural rearrangement of the pore and/or the SF and produce an acceleration of inactivation via a process similar to the C-type inactivation that is present in most voltage-gated ion channel types and has been attributed to a restructuration (either constriction or dilatation) of the selectivity filter inducing non-conductivity, and thus inactivation (Reddi et al., 2022; Cuello et al., 2010; Pavlov et al., 2005; Irie et al., 2010). In potassium channels, different mutations within the selectivity filter or at the internal mouth of the channel (specifically at the bottom of the S6 segment) have been demonstrated to impact the C-type inactivation (Li et al., 2021; Cuello et al., 2010; Tan et al., 2022; Reddi et al., 2022). Furthermore, although infrequent, there is precedent for the regulatory effects of certain drugs or ions on C-type inactivation (Armstrong and Hoshi, 2014; Chen et al., 2013). In this context, the fact that the inactivation time constant is independent of the voltage but dependent upon the direction of the permeating ion (see Fig. 10 B) could possibly be explained by the differential stabilization of the pore structure by divalent cation binding at the extracellular ion binding site during inward but not outward currents. Alternatively, an open-channel block can also be a possible mechanism. The single-channel signatures of these two processes might be different, and this eventuality is now being evaluated at the single-channel level. In any case, in the absence of other specific inhibitors, the use of veratrine may constitute a very useful tool to identify the expression of CaV4 in living insect tissues, but it may also possibly help, through structural studies, to shed new light on the C-type channel inactivation mechanism and regulation.

In conclusion, this study has revealed several distinctive properties of the CaV4 channel, including (1) its exclusive permeability to calcium, (2) insensitivity of the voltage-dependent inactivation to surface potential, (3) a specific type of CatDI, and (4) a pharmacological profile divergent from NaV and CaV channels. These unique features unequivocally define a novel type of voltage-gated Ca2+ channels specific to insects, establishing a new phylogenetic and functional connection between NaV and CaV. The identified characteristics demand further comprehensive analysis and are poised to offer invaluable insights into the functional molecular mechanisms not only of CaV4 but also of NaV and other CaV channels. It should be noted, however, that purified CaV4 channels from native insect tissues have not yet been obtained. Therefore, the biophysical and pharmacological properties of this channel in vivo may differ significantly from those described here if regulatory subunits exist, as described for other voltage-gated Ca, Na, or K channels.

The data are directly available from the corresponding author upon reasonable request by email.

Christopher J. Lingle served as editor.

A special thanks to Mecédès Charreton (INRAE), Sylvain Roque (Centre National de la Recherche Scientifique), and Karine Allenne (University of Montpellier) for their invaluable help in honeybee and Xenopus preparation and administrative work.

The authors would like to acknowledge the Centre National de la Recherche Scientifique (CNRS), the Institut National de la Santé et de la Recherche Médicale (INSERM), l’Institut national de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), and the Agence National pour la Recherche (ANR Synaptic-Bee: ANR-20-CE34-0017-01) for financial support. We acknowledge the support of the Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant (RGPIN-2020-06359) and NSERC Alliance International Catalyst grant (572132-2022) to Mohamed Chahine.

Author contributions: Investigation: A. Bertaud, T. Cens, M. Rousset, L. Soussi, A. Kadala, C. Collet, P. Gosselin-Badaroudine, J. Roussel, M. Chahine, and P. Charnet. Conceptualization: T. Cens, P. Bois, M. Chahine, and P. Charnet. Formal analysis: T. Cens, P. Gosselin-Badaroudine, and P. Charnet. Methodology: T. Cens, C. Ménard, A. Chavanieu, C. Collet, and P. Charnet. Writing - review & editing: T. Cens, C. Collet, P. Bois, J.-B. Thibaud, M. Vignes, M. Chahine, and P. Charnet. Data curation for molecular structure: A. Chavanieu, S. Esteran, M. Chahine, and P. Charnet. Validation: J.-B. Thibaud, M. Chahine, and P. Charnet. Resources: S. Dutertre and P. Charnet. Funding acquisition: M. Chahine and P. Charnet. Project administration: P. Charnet. Writing - original draft: P. Charnet.

Abderemane-Ali
,
F.
,
F.
Findeisen
,
N.D.
Rossen
, and
D.L.
Minor
Jr.
2019
.
A selectivity filter gate controls voltage-gated calcium channel calcium-dependent inactivation
.
Neuron
.
101
:
1134
1149.e3
.
Angsutararux
,
P.
,
P.W.
Kang
,
W.
Zhu
, and
J.R.
Silva
.
2021
.
Conformations of voltage-sensing domain III differentially define NaV channel closed- and open-state inactivation
.
J. Gen. Physiol.
153
:e202112891.
Armstrong
,
C.M.
, and
T.
Hoshi
.
2014
.
K⁺ channel gating: C-Type inactivation is enhanced by calcium or lanthanum outside
.
J. Gen. Physiol.
144
:
221
230
.
Ben-Johny
,
M.
,
P.S.
Yang
,
J.
Niu
,
W.
Yang
,
R.
Joshi-Mukherjee
, and
D.T.
Yue
.
2014
.
Conservation of Ca2+/calmodulin regulation across Na and Ca2+ channels
.
Cell
.
157
:
1657
1670
.
Ben-Johny
,
M.
, and
D.T.
Yue
.
2014
.
Calmodulin regulation (calmodulation) of voltage-gated calcium channels
.
J. Gen. Physiol.
143
:
679
692
.
Brunello
,
L.
,
C.
Ménard
,
M.
Rousset
,
M.
Vignes
,
P.
Charnet
, and
T.
Cens
.
2022
.
Different efficiency of auxiliary/chaperone proteins to promote the functional reconstitution of honeybee glutamate and acetylcholine receptors in Xenopus laevis oocytes
.
Insect Mol. Biol.
31
:
620
633
.
Budde
,
T.
,
S.
Meuth
, and
H.-C.
Pape
.
2002
.
Calcium-dependent inactivation of neuronal calcium channels
.
Nat. Rev. Neurosci.
3
:
873
883
.
Capes
,
D.L.
,
M.P.
Goldschen-Ohm
,
M.
Arcisio-Miranda
,
F.
Bezanilla
, and
B.
Chanda
.
2013
.
Domain IV voltage-sensor movement is both sufficient and rate limiting for fast inactivation in sodium channels
.
J. Gen. Physiol.
142
:
101
112
.
Castella
,
C.
,
M.
Amichot
,
J.B.
Bergé
, and
D.
Pauron
.
2001
.
DSC1 channels are expressed in both the central and the peripheral nervous system of adult Drosophila melanogaster
.
Invert. Neurosci.
4
:
85
94
.
Catterall
,
W.A.
2000
.
Structure and regulation of voltage-gated Ca2+ channels
.
Annu. Rev. Cell Dev. Biol.
16
:
521
555
.
Catterall
,
W.A.
2013
.
Voltage-gated sodium channels at 60: Structure, function, and pathophysiology
.
J. Physiol.
590
:
2577
2589
.
Cens
,
T.
,
S.
Restituito
,
S.
Galas
, and
P.
Charnet
.
1999
.
Voltage and calcium use the same molecular determinants to inactivate calcium channels
.
J. Biol. Chem.
274
:
5483
5490
.
Cens
,
T.
,
M.
Rousset
,
C.
Collet
,
M.
Charreton
,
L.
Garnery
,
Y.
Le Conte
,
M.
Chahine
,
J.C.
Sandoz
, and
P.
Charnet
.
2015
.
Molecular characterization and functional expression of the Apis mellifera voltage-dependent Ca2+ channels
.
Insect Biochem.
58
:
12
27
.
Cens
,
T.
,
M.
Rousset
,
C.
Collet
,
V.
Raymond
,
F.
Démares
,
A.
Quintavalle
,
M.
Bellis
,
Y.
Le Conte
,
M.
Chahine
, and
P.
Charnet
.
2013
.
Characterization of the first honeybee Ca²⁺ channel subunit reveals two novel species- and splicing-specific modes of regulation of channel inactivation
.
Pflugers Arch.
465
:
985
996
.
Cens
,
T.
,
M.
Rousset
,
A.
Kajava
, and
P.
Charnet
.
2007
.
Molecular determinant for specific Ca/Ba selectivity profiles of low and high threshold Ca2+ channels
.
J. Gen. Physiol.
130
:
415
425
.
Cens
,
T.
,
M.
Rousset
,
J.P.
Leyris
,
P.
Fesquet
, and
P.
Charnet
.
2006
.
Voltage- and calcium-dependent inactivation in high voltage-gated Ca(2+) channels
.
Prog. Biophys. Mol. Biol.
90
:
104
117
.
Chancey
,
J.H.
,
P.E.
Shockett
, and
J.P.
O’Reilly
.
2007
.
Relative resistance to slow inactivation of human cardiac Na+ channel hNav1.5 is reversed by lysine or glutamine substitution at V930 in D2-S6
.
Am. J. Physiol. Cell Physiol.
293
:
C1895
C1905
.
Chen-Izu
,
Y.
,
R.M.
Shaw
,
G.S.
Pitt
,
V.
Yarov-Yarovoy
,
J.T.
Sack
,
H.
Abriel
,
R.W.
Aldrich
,
L.
Belardinelli
,
M.B.
Cannell
,
W.A.
Catterall
, et al
.
2015
.
Na+ channel function, regulation, structure, trafficking and sequestration
.
J. Physiol.
593
:
1347
1360
.
Chen
,
H.
,
D.
Zhang
,
J.
Hua Ren
, and
S.
Ping Chao
.
2013
.
Effects of L-type calcium channel antagonists verapamil and diltiazem on fKv1.4ΔN Currents in Xenopus oocytes
.
Iran. J. Pharm. Res.
12
:
855
866
.
Chen
,
X.
,
Y.
Wang
,
W.
Wu
,
K.
Dong
, and
Z.
Hu
.
2018
.
DSC1 channel-dependent developmental regulation of pyrethroid susceptibility in Drosophila melanogaster
.
Pestic. Biochem. Physiol.
148
:
190
198
.
Cibulsky
,
S.M.
, and
W.A.
Sather
.
2000
.
The EEEE locus is the sole high-affinity Ca(2+) binding structure in the pore of a voltage-gated Ca(2+) channel: Block by Ca(2+) entering from the intracellular pore entrance
.
J. Gen. Physiol.
116
:
349
362
.
Collet
,
C.
,
A.
Kadala
,
B.
Vaissière
,
M.
Rousset
,
T.
Cens
,
Y.
Le Conte
,
M.
Chahine
,
J.B.
Thibaud
, and
P.
Charnet
.
2016
.
Differential action of pyrethroids on honey bee and bumble bee voltage-gated sodium channels
.
Biophys. J.
110
:
112a
.
Cuello
,
L.G.
,
V.
Jogini
,
D.M.
Cortes
, and
E.
Perozo
.
2010
.
Structural mechanism of C-type inactivation in K(+) channels
.
Nature
.
466
:
203
208
.
Cui
,
Y.-J.
,
L.-L.
Yu
,
H.-J.
Xu
,
K.
Dong
, and
C.-X.
Zhang
.
2012
.
Molecular characterization of DSC1 orthologs in invertebrate species
.
Insect Biochem. Mol. Biol.
42
:
353
359
.
Dong
,
K.
,
Y.
Du
,
F.
Rinkevich
,
Y.
Nomura
,
P.
Xu
,
L.
Wang
,
K.
Silver
, and
B.S.
Zhorov
.
2014
.
Molecular biology of insect sodium channels and pyrethroid resistance
.
Insect Biochem. Mol. Biol.
50
:
1
17
.
Dong
,
K.
,
Y.
Du
,
F.
Rinkevich
,
L.
Wang
, and
P.
Xu
.
2015
.
The Drosophila Sodium Channel 1 (DSC1): The founding member of a new family of voltage-gated cation channels
.
Pestic. Biochem. Physiol.
120
:
36
39
.
Dudev
,
T.
, and
C.
Lim
.
2014a
.
Ion selectivity strategies of sodium channel selectivity filters
.
Acc. Chem. Res.
47
:
3580
3587
.
Dudev
,
T.
, and
C.
Lim
.
2014b
.
Evolution of eukaryotic ion channels: Principles underlying the conversion of Ca²⁺-selective to Na⁺-selective channels
.
J. Am. Chem. Soc.
136
:
3553
3559
.
Elinder
,
F.
,
M.
Madeja
, and
P.
Arhem
.
1996
.
Surface charges of K channels. Effects of strontium on five cloned channels expressed in Xenopus oocytes
.
J. Gen. Physiol.
108
:
325
332
.
Fernández-Quintero
,
M.L.
,
Y.
El Ghaleb
,
P.
Tuluc
,
M.
Campiglio
,
K.R.
Liedl
, and
B.E.
Flucher
.
2021
.
Structural determinants of voltage-gating properties in calcium channels
.
Elife
.
10
:e64087.
Fux
,
J.E.
,
A.
Mehta
,
J.
Moffat
, and
J.D.
Spafford
.
2018
.
Eukaryotic voltage-gated sodium channels: On their origins, asymmetries, losses, diversification and adaptations
.
Front. Physiol.
9
:
1406
.
Gardill
,
B.R.
,
R.E.
Rivera-Acevedo
,
C.-C.C.
Tung
, and
F.
Van Petegem
.
2019
.
Crystal structures of Ca2+-calmodulin bound to NaV C-terminal regions suggest role for EF-hand domain in binding and inactivation
.
Proc. Natl. Acad. Sci. USA
.
116
:
10763
10772
.
Gilbert
,
D.L.
, and
G.
Ehrenstein
.
1969
.
Effect of divalent cations on potassium conductance of squid axons: Determination of surface charge
.
Biophys. J.
9
:
447
463
.
Gosselin-Badaroudine
,
P.
,
P.
Charnet
,
C.
Collet
, and
M.
Chahine
.
2017
.
Metaflumizone inhibits the honeybee NaV 1 channel by targeting recovery from slow inactivation
.
FEBS Lett.
591
:
3842
3849
.
Gosselin-Badaroudine
,
P.
,
A.
Moreau
,
L.
Delemotte
,
T.
Cens
,
C.
Collet
,
M.
Rousset
,
P.
Charnet
,
M.L.
Klein
, and
M.
Chahine
.
2015
.
Characterization of the honeybee AmNaV1 channel and tools to assess the toxicity of insecticides
.
Sci. Rep.
5
:
12475
.
Gosselin-Badaroudine
,
P.
,
A.
Moreau
,
L.
Simard
,
T.
Cens
,
M.
Rousset
,
C.
Collet
,
P.
Charnet
, and
M.
Chahine
.
2016
.
Biophysical characterization of the honeybee DSC1 orthologue reveals a novel voltage-dependent Ca2+ channel subfamily: CaV4
.
J. Gen. Physiol.
148
:
133
145
.
Guia
,
A.
,
M.D.
Stern
,
E.G.
Lakatta
, and
I.R.
Josephson
.
2001
.
Ion concentration-dependence of rat cardiac unitary L-type calcium channel conductance
.
Biophys. J.
80
:
2742
2750
.
Gulsevin
,
A.
,
A.M.
Glazer
,
T.
Shields
,
B.M.
Kroncke
,
D.M.
Roden
, and
J.
Meiler
.
2022
.
Veratridine can bind to a site at the mouth of the channel pore at human cardiac sodium channelchannel NaV1.5
.
Int. J. Mol. Sci.
23
:
2225
2238
.
Gur Barzilai
,
M.
,
A.M.
Reitzel
,
J.E.M.
Kraus
,
D.
Gordon
,
U.
Technau
,
M.
Gurevitz
, and
Y.
Moran
.
2012
.
Convergent evolution of sodium ion selectivity in metazoan neuronal signaling
.
Cell Rep.
2
:
242
248
.
Heinemann
,
S.H.
,
H.
Terlau
,
W.
Stühmer
,
K.
Imoto
,
S.
Numa
,
W.
Stühmer
,
K.
Imoto
, and
S.
Numa
.
1992
.
Calcium channel characteristics conferred on the sodium channel by single mutations
.
Nature
.
356
:
441
443
.
Hering
,
S.
,
S.
Aczél
,
M.
Grabner
,
F.
Döring
,
S.
Berjukow
,
J.
Mitterdorfer
,
M.J.
Sinnegger
,
J.
Striessnig
,
V.E.
Degtiar
,
Z.
Wang
, and
H.
Glossmann
.
1996
.
Transfer of high sensitivity for benzothiazepines from L-type to class A (BI) calcium channels
.
J. Biol. Chem.
271
:
24471
24475
.
Hille
,
B.
1992a
.
Ion channels of Excitable Membranes
. Third edition.
Oxford University Press
,
Oxford, UK
.
Hille
,
B.
1992b
.
Ionic channels of Excitable Membranes
. Second edition.
Oxford University Press
,
Oxford, UK
.
Hong
,
C.S.
, and
B.
Ganetzky
.
1994
.
Spatial and temporal expression patterns of two sodium channel genes in Drosophila
.
J. Neurosci.
14
:
5160
5169
.
Hsu
,
E.J.
,
W.
Zhu
,
A.R.
Schubert
,
T.
Voelker
,
Z.
Varga
, and
J.R.
Silva
.
2017
.
Regulation of Na+ channel inactivation by the DIII and DIV voltage-sensing domains
.
J. Gen. Physiol.
149
:
389
403
.
Irie
,
K.
,
K.
Kitagawa
,
H.
Nagura
,
T.
Imai
,
T.
Shimomura
, and
Y.
Fujiyoshi
.
2010
.
Comparative study of the gating motif and C-type inactivation in prokaryotic voltage-gated sodium channels
.
J. Biol. Chem.
285
:
3685
3694
.
Jiang
,
D.
,
H.
Shi
,
L.
Tonggu
,
T.M.
Gamal El-Din
,
M.J.
Lenaeus
,
Y.
Zhao
,
C.
Yoshioka
,
N.
Zheng
, and
W.A.
Catterall
.
2020
.
Structure of the cardiac sodium channel
.
Cell
.
180
:
122
134.e10
.
Johnson
,
C.N.
2020
.
Calcium modulation of cardiac sodium channels
.
J. Physiol.
598
:
2835
2846
.
Kontis
,
K.J.
,
A.
Rounaghi
, and
A.L.
Goldin
.
1997
.
Sodium channel activation gating is affected by substitutions of voltage sensor positive charges in all four domains
.
J. Gen. Physiol.
110
:
391
401
.
Kulkarni
,
N.H.
,
A.H.
Yamamoto
,
K.O.
Robinson
,
T.F.C.
Mackay
, and
R.R.H.
Anholt
.
2002
.
The DSC1 channel, encoded by the smi60E locus, contributes to odor-guided behavior in Drosophila melanogaster
.
Genetics
.
161
:
1507
1516
.
De Leon
,
M.
,
Y.
Wang
,
L.
Jones
,
E.
Perez-Reyes
,
X.
Wei
,
T.W.
Soong
,
T.P.
Snutch
, and
D.T.
Yue
.
1995
.
Essential Ca2+-binding motif for Ca2+-sensitive inactivation of L-type Ca2+ channels
.
Science
.
270
:
1502
1506
.
Lewis
,
A.H.
, and
I.M.
Raman
.
2013
.
Interactions among DIV voltage-sensor movement, fast inactivation, and resurgent Na current induced by the NaVβ4 open-channel blocking peptide
.
J. Gen. Physiol.
142
:
191
206
.
Li
,
J.
,
R.
Shen
,
B.
Reddy
,
E.
Perozo
, and
B.
Roux
.
2021
.
Mechanism of C-type inactivation in the hERG potassium channel
.
Sci. Adv.
7
:
1
8
.
Liu
,
Y.
,
C.A.Z.
Bassetto
Jr.
,
B.I.
Pinto
, and
F.
Bezanilla
.
2023
.
A mechanistic reinterpretation of fast inactivation in voltage-gated Na+ channels
.
Nat. Commun.
14
:
5072
.
Liu
,
Z.
,
I.
Chung
, and
K.
Dong
.
2001
.
Alternative splicing of the BSC1 gene generates tissue-specific isoforms in the German cockroach
.
Insect Biochem. Mol. Biol.
31
:
703
713
.
Madeja
,
M.
2000
.
Extracellular surface charges in voltage-gated ion channels
.
News Physiol. Sci.
15
:
15
19
.
Moran
,
Y.
,
M.G.
Barzilai
,
B.J.
Liebeskind
, and
H.H.
Zakon
.
2015
.
Evolution of voltage-gated ion channels at the emergence of Metazoa
.
J. Exp. Biol.
218
:
515
525
.
Neumaier
,
F.
,
M.
Dibué-Adjei
,
J.
Hescheler
, and
T.
Schneider
.
2015
.
Voltage-gated calcium channels: Determinants of channel function and modulation by inorganic cations
.
Prog. Neurobiol.
129
:
1
36
.
Pavlov
,
E.
,
C.
Bladen
,
R.
Winkfein
,
C.
Diao
,
P.
Dhaliwal
, and
R.J.
French
.
2005
.
The pore, not cytoplasmic domains, underlies inactivation in a prokaryotic sodium channel
.
Biophys. J.
89
:
232
242
.
Peterson
,
B.Z.
,
C.D.
DeMaria
,
J.P.
Adelman
, and
D.T.
Yue
.
1999
.
Calmodulin is the Ca2+ sensor for Ca2+ -dependent inactivation of L-type calcium channels
.
Neuron
.
22
:
549
558
.
Pitt
,
G.S.
, and
S.Y.
Lee
.
2016
.
Current view on regulation of voltage-gated sodium channels by calcium and auxiliary proteins
.
Protein Sci.
25
:
1573
1584
.
Ramaswami
,
M.
, and
M.A.
Tanouye
.
1989
.
Two sodium-channel genes in Drosophila: Implications for channel diversity
.
Proc. Natl. Acad. Sci. USA
.
86
:
2079
2082
.
Reddi
,
R.
,
K.
Matulef
,
E.A.
Riederer
,
M.R.
Whorton
, and
F.I.
Valiyaveetil
.
2022
.
Structural basis for C-type inactivation in a Shaker family voltage-gated K+ channel
.
Sci. Adv.
8
:eabm8804.
Rinkevich
,
F.D.
,
Y.
Du
, and
K.
Dong
.
2013
.
Diversity and convergence of sodium channel mutations involved in resistance to pyrethroids
.
Pestic. Biochem. Physiol.
106
:
93
100
.
Rinkevich
,
F.D.
,
Y.
Du
,
J.
Tolinski
,
A.
Ueda
,
C.F.
Wu
,
B.S.
Zhorov
, and
K.
Dong
.
2015
.
Distinct roles of the DmNav and DSC1 channels in the action of DDT and pyrethroids
.
Neurotoxicology
.
47
:
99
106
.
Rousset
,
M.
,
C.
Collet
,
T.
Cens
,
F.
Bastin
,
V.
Raymond
,
I.
Massou
,
C.
Menard
,
J.B.
Thibaud
,
M.
Charreton
,
M.
Vignes
, et al
.
2017
.
Honeybee locomotion is impaired by Am-CaV3 low voltage-activated Ca2+ channel antagonist
.
Sci. Rep.
7
:
41782
.
Salkoff
,
L.
,
A.
Butler
,
A.
Wei
,
N.
Scavarda
,
K.
Giffen
,
C.
Ifune
,
R.
Goodman
, and
G.
Mandel
.
1987
.
Genomic organization and deduced amino acid sequence of a putative sodium channel gene in Drosophila
.
Science
.
237
:
744
749
.
Salvage
,
S.C.
,
Z.F.
Habib
,
H.R.
Matthews
,
A.P.
Jackson
, and
C.L.H.
Huang
.
2021
.
Ca2+-dependent modulation of voltage-gated myocyte sodium channels
.
Biochem. Soc. Trans.
49
:
1941
1961
.
Sather
,
W.A.
, and
E.W.
McCleskey
.
2003
.
Permeation and selectivity in calcium channels
.
Annu. Rev. Physiol.
65
:
133
159
.
Sather
,
W.A.
,
J.
Yang
, and
R.W.
Tsien
.
1994
.
Structural basis of ion channel permeation and selectivity
.
Curr. Opin. Neurobiol.
4
:
313
323
.
Shaya
,
D.
,
F.
Findeisen
,
F.
Abderemane-Ali
,
C.
Arrigoni
,
S.
Wong
,
S.R.
Nurva
,
G.
Loussouarn
, and
D.L.
Minor
Jr.
2014
.
Structure of a prokaryotic sodium channel pore reveals essential gating elements and an outer ion binding site common to eukaryotic channels
.
J. Mol. Biol.
426
:
467
483
.
Shi
,
C.
, and
N.M.
Soldatov
.
2002
.
Molecular determinants of voltage-dependent slow inactivation of the Ca2+ channel
.
J. Biol. Chem.
277
:
6813
6821
.
Sigel
,
E.
1987
.
Properties of single sodium channels translated by Xenopus oocytes after injection with messenger ribonucleic acid
.
J. Physiol.
386
:
73
90
.
Sutro
,
J.B.
1986
.
Kinetics of veratridine action on Na channels of skeletal muscle
.
J. Gen. Physiol.
87
:
1
24
.
Tadross
,
M.R.
,
I.E.
Dick
, and
D.T.
Yue
.
2008
.
Mechanism of local and global Ca2+ sensing by calmodulin in complex with a Ca2+ channel
.
Cell
.
133
:
1228
1240
.
Tan
,
X.F.
,
C.
Bae
,
R.
Stix
,
A.I.
Fernández-Mariño
,
K.
Huffer
,
T.H.
Chang
,
J.
Jiang
,
J.D.
Faraldo-Gómez
, and
K.J.
Swartz
.
2022
.
Structure of the Shaker Kv channel and mechanism of slow C-type inactivation
.
Sci. Adv.
8
:eabm7814.
Tang
,
L.
,
T.M.
Gamal El-Din
,
J.
Payandeh
,
G.Q.
Martinez
,
T.M.
Heard
,
T.
Scheuer
,
N.
Zheng
, and
W.A.
Catterall
.
2014
.
Structural basis for Ca2+ selectivity of a voltage-gated calcium channel
.
Nature
.
505
:
56
61
.
West
,
J.W.
,
D.E.
Patton
,
T.
Scheuer
,
Y.
Wang
,
A.L.
Goldin
, and
W.A.
Catterall
.
1992
.
A cluster of hydrophobic amino acid residues required for fast Na(+)-channel inactivation
.
Proc. Natl. Acad. Sci. USA
.
89
:
10910
10914
.
Wu
,
J.
,
Z.
Yan
,
Z.
Li
,
X.
Qian
,
S.
Lu
,
M.
Dong
,
Q.
Zhou
, and
N.
Yan
.
2016
.
Structure of the voltage-gated calcium channel Ca(v)1.1 at 3.6 Å resolution
.
Nature
.
537
:
191
196
.
Yang
,
J.
,
P.T.
Ellinor
,
W.A.
Sather
,
J.F.
Zhang
, and
R.W.
Tsien
.
1993
.
Molecular determinants of Ca2+ selectivity and ion permeation in L-type Ca2+ channels
.
Nature
.
366
:
158
161
.
Zhang
,
T.
,
Z.
Wang
,
L.
Wang
,
N.
Luo
,
L.
Jiang
,
Z.
Liu
,
C.-F.
Wu
, and
K.
Dong
.
2013
.
Role of the DSC1 channel in regulating neuronal excitability in Drosophila melanogaster: Extending nervous system stability under stress
.
PLoS Genet.
9
:e1003327.
Zhao
,
Y.
,
G.
Huang
,
J.
Wu
,
Q.
Wu
,
S.
Gao
,
Z.
Yan
,
J.
Lei
, and
N.
Yan
.
2019
.
Molecular basis for ligand modulation of a mammalian voltage-gated Ca2+ channel
.
Cell
.
177
:
1495
1506.e12
.
Zhou
,
W.
,
I.
Chung
,
Z.
Liu
,
A.L.
Goldin
, and
K.
Dong
.
2004
.
A voltage-gated calcium-selective channel encoded by a sodium channel-like gene
.
Neuron
.
42
:
101
112
.

Author notes

*

A. Bertaud and T. Cens contributed equally to this paper.

Disclosures: The authors declare no competing interests exist.

This article is distributed under the terms of an Attribution–Noncommercial–Share Alike–No Mirror Sites license for the first six months after the publication date (see http://www.rupress.org/terms/). After six months it is available under a Creative Commons License (Attribution–Noncommercial–Share Alike 4.0 International license, as described at https://creativecommons.org/licenses/by-nc-sa/4.0/).