Advertisement

A CACNA1C variant associated with cardiac arrhythmias provides mechanistic insights in the calmodulation of L-type Ca2+ channels

  • Author Footnotes
    ‡ These authors contributed equally to this work.
    Juan Zhao
    Footnotes
    ‡ These authors contributed equally to this work.
    Affiliations
    Centre de recherche de l’Institut de Cardiologie de Montréal, Université de Montréal, Montréal, Québec, Canada
    Search for articles by this author
  • Author Footnotes
    ‡ These authors contributed equally to this work.
    Emilie Segura
    Footnotes
    ‡ These authors contributed equally to this work.
    Affiliations
    Centre de recherche de l’Institut de Cardiologie de Montréal, Université de Montréal, Montréal, Québec, Canada

    Département de Pharmacologie et Physiologie, Faculté de Médecine, Montréal, Québec, Canada
    Search for articles by this author
  • Mireille Marsolais
    Affiliations
    Centre de recherche de l’Institut de Cardiologie de Montréal, Université de Montréal, Montréal, Québec, Canada

    Département de Pharmacologie et Physiologie, Faculté de Médecine, Montréal, Québec, Canada
    Search for articles by this author
  • Lucie Parent
    Correspondence
    For correspondence: Lucie Parent
    Affiliations
    Centre de recherche de l’Institut de Cardiologie de Montréal, Université de Montréal, Montréal, Québec, Canada

    Département de Pharmacologie et Physiologie, Faculté de Médecine, Montréal, Québec, Canada
    Search for articles by this author
  • Author Footnotes
    ‡ These authors contributed equally to this work.
Open AccessPublished:October 20, 2022DOI:https://doi.org/10.1016/j.jbc.2022.102632
      We recently reported the identification of a de novo single nucleotide variant in exon 9 of CACNA1C associated with prolonged repolarization interval. Recombinant expression of the glycine to arginine variant at position 419 produced a gain in the function of the L-type CaV1.2 channel with increased peak current density and activation gating but without significant decrease in the inactivation kinetics. We herein reveal that these properties are replicated by overexpressing calmodulin (CaM) with CaV1.2 WT and are reversed by exposure to the CaM antagonist W-13. Phosphomimetic (T79D or S81D), but not phosphoresistant (T79A or S81A), CaM surrogates reproduced the impact of CaM WT on the function of CaV1.2 WT. The increased channel activity of CaV1.2 WT following overexpression of CaM was found to arise in part from enhanced cell surface expression. In contrast, the properties of the variant remained unaffected by any of these treatments. CaV1.2 substituted with the α-helix breaking proline residue were more reluctant to open than CaV1.2 WT but were upregulated by phosphomimetic CaM surrogates. Our results indicate that (1) CaM and its phosphomimetic analogs promote a gain in the function of CaV1.2 and (2) the structural properties of the first intracellular linker of CaV1.2 contribute to its CaM-induced modulation. We conclude that the CACNA1C clinical variant mimics the increased activity associated with the upregulation of CaV1.2 by Ca2+–CaM, thus maintaining a majority of channels in a constitutively active mode that could ultimately promote ventricular arrhythmias.

      Keywords

      Abbreviations:

      CaM (calmodulin), CDI (Ca2+-dependent inactivation), cDNA (complementary DNA), CK2 (casein kinase II), CMV (cytomegalovirus), HEK (human embryonic kidney cell line), LQTS (long-QT syndrome), TBB (4,5,6,7-tetrabromobenzotriazole), TS (Timothy syndrome)
      Cardiac contraction during the systole is handled by the influx of Ca2+ into cardiomyocytes in response to depolarization during phase 2 of the cardiac action potential (
      • Grant A.O.
      Cardiac ion channels.
      ). Voltage-gated L-type calcium channel CaV1.2 are expressed in the T-tubules such that localized Ca2+ entry triggers a sustained and more global Ca2+ release by the sarcoplasmic reticulum in the dyadic cleft (
      • Wang S.Q.
      • Song L.S.
      • Lakatta E.G.
      • Cheng H.
      Ca2+ signalling between single L-type Ca2+ channels and ryanodine receptors in heart cells.
      ). Cardiac L-type CaV1.2 channels are heteromultimeric protein complexes formed by the pore-forming CaVα1C subunit bound to the extracellular CaVα2δ1 auxiliary subunits (
      • Wu J.
      • Yan Z.
      • Li Z.
      • Qian X.
      • Lu S.
      • Dong M.
      • et al.
      Structure of the voltage-gated calcium channel Cav1.1 at 3.6 Å resolution.
      ,
      • Segura E.
      • Bourdin B.
      • Tétreault M.P.
      • Briot J.
      • Allen B.G.
      • Mayer G.
      • et al.
      Proteolytic cleavage of the hydrophobic domain in the Ca(V)α2δ1 subunit improves assembly and activity of cardiac Ca(V)1.2 channels.
      ) and to the cytoplasmic CaVβ (
      • Colecraft H.M.
      • Alseikhan B.
      • Takahashi S.X.
      • Chaudhuri D.
      • Mittman S.
      • Yegnasubramanian V.
      • et al.
      Novel functional properties of Ca(2+) channel beta subunits revealed by their expression in adult rat heart cells.
      ) that binds with nanomolar affinity to the first intracellular linker (
      • Van Petegem F.
      • Duderstadt K.E.
      • Clark K.A.
      • Wang M.
      • Minor Jr., D.L.
      Alanine-scanning mutagenesis defines a conserved energetic hotspot in the CaValpha1 AID-CaVbeta interaction site that is critical for channel modulation.
      ). The CaVα1 subunit is formed by a single polypeptide chain of 24 transmembrane helices grouped into four structural homologous domains (domains I, II, II, and IV) (Fig. 1). Although not a specific auxiliary subunit, calmodulin (CaM) contributes to Ca2+-dependent facilitation and Ca2+-dependent inactivation (CDI) of CaV1.2 (
      • Adams P.J.
      • Ben-Johny M.
      • Dick I.E.
      • Inoue T.
      • Yue D.T.
      Apocalmodulin itself promotes ion channel opening and Ca(2+) regulation.
      ,
      • Zühlke R.D.
      • Pitt G.S.
      • Deisseroth K.
      • Tsien R.W.
      • Reuter H.
      Calmodulin supports both inactivation and facilitation of L-type calcium channels.
      ,
      • Zühlke R.D.
      • Pitt G.S.
      • Tsien R.W.
      • Reuter H.
      Ca2+-sensitive inactivation and facilitation of L-type Ca2+ channels both depend on specific amino acid residues in a consensus calmodulin-binding motif in the(alpha)1C subunit.
      ) through binding to the isoleucine–glutamine motif in the C-terminal tail of CaVα1C (
      • Pate P.
      • Mochca-Morales J.
      • Wu Y.
      • Zhang J.Z.
      • Rodney G.G.
      • Serysheva I.I.
      • et al.
      Determinants for calmodulin binding on voltage-dependent Ca2+ channels.
      ,
      • Romanin C.
      • Gamsjaeger R.
      • Kahr H.
      • Schaufler D.
      • Carlson O.
      • Abernethy D.R.
      • et al.
      Ca(2+) sensors of L-type Ca(2+) channel.
      ,
      • Pitt G.S.
      • Zühlke R.D.
      • Hudmon A.
      • Schulman H.
      • Reuter H.
      • Tsien R.W.
      Molecular basis of calmodulin tethering and Ca2+-dependent inactivation of L-type Ca2+ channels.
      ,
      • Mouton J.
      • Feltz A.
      • Maulet Y.
      Interactions of calmodulin with two peptides derived from the c-terminal cytoplasmic domain of the Ca(v)1.2 Ca2+ channel provide evidence for a molecular switch involved in Ca2+-induced inactivation.
      ).
      Figure thumbnail gr1
      Figure 1The CaV1.2 variant is located in the intracellular linker before the binding site for the CaVβ subunit. The LQTS-related CaVα1C mutation G449 is located before the α-interacting domain (AID). A, the cryo-EM 3D structure of the rabbit CaV1.1 oligomeric complex at 3.6 Å for CaVα1S and at 3.9 Å for CaVβ (Protein Data Bank code: 5GJV). CaVα1C and CaVα1S share 81% homology in their primary protein sequence. L-type calcium channels share similar structure, being composed of the pore-forming subunit CaVα1 in red and CaVβ in blue and an intracellular subunit bound to CaVα1 through the intracellular helix linking domains I and II of CaVα1 (shown in dark green). The first transmembrane domain of CaVα1 (DI) is shown in yellow. The human LQTS-related CaVα1C G419R variant is similar to the rabbit CaVα1C G449R and is equivalent to Gly358 in CaVα1S. Image was produced by Discovery Studio 2020 (BIOVIA Pipeline Pilot 2020). B, cartoon of the corresponding secondary structure for the CaVα1C pore-forming subunit of the L-type CaV1.2 channel showing the four homologous domains (domains I to IV) with the N and C termini located into the cytoplasm. The CaVβ subunit–binding site on the CaVα1C subunit is referred to the “α-interacting domain” or AID. The AID is located within 20 residues of the sixth transmembrane segment in domain I (IS6). The primary sequence for the AID motif is shown below the primary sequence for the short region extending from the end of S6 to the beginning of the AID. The relative position of three glycine variants reported in the Timothy syndrome (G402S, G406R, and G419R) is fully conserved across species and presented in red with the numbering in the rabbit clone used for this study. LQTS, long-QT syndrome.
      First clinically described in 1957 (
      • Jervell A.
      • Lange-Nielsen F.
      Congenital deaf-mutism, functional heart disease with prolongation of the Q-T interval and sudden death.
      ), the long-QT syndrome (LQTS) is a major cause of sudden death in healthy infants and young adults (
      • Liberthson R.R.
      Sudden death from cardiac causes in children and young adults.
      ,
      • Chugh S.S.
      • Reinier K.
      • Teodorescu C.
      • Evanado A.
      • Kehr E.
      • Al Samara M.
      • et al.
      Epidemiology of sudden cardiac death: clinical and research implications.
      ,
      • Schwartz P.J.
      • Stramba-Badiale M.
      • Crotti L.
      • Pedrazzini M.
      • Besana A.
      • Bosi G.
      • et al.
      Prevalence of the congenital long-QT syndrome.
      ). Congenital LQTS in the absence of structural defects (
      • Tester D.J.
      • Ackerman M.J.
      Postmortem long QT syndrome genetic testing for sudden unexplained death in the young.
      ) is often the result of inherited or de novo genetic mutations in the DNA of a variety of ion channels (
      • Tester D.J.
      • Will M.L.
      • Haglund C.M.
      • Ackerman M.J.
      Compendium of cardiac channel mutations in 541 consecutive unrelated patients referred for long QT syndrome genetic testing.
      ). Gain-of-function mutations within the CACNA1C gene, coding for CaVα1C, are associated with the LQTS type 8 also referred to as Timothy syndrome (TS) (
      • Splawski I.
      • Timothy K.W.
      • Sharpe L.M.
      • Decher N.
      • Kumar P.
      • Bloise R.
      • et al.
      Ca(V)1.2 calcium channel dysfunction causes a multisystem disorder including arrhythmia and autism.
      ,
      • Splawski I.
      • Timothy K.W.
      • Decher N.
      • Kumar P.
      • Sachse F.B.
      • Beggs A.H.
      • et al.
      Severe arrhythmia disorder caused by cardiac L-type calcium channel mutations.
      ,
      • Gillis J.
      • Burashnikov E.
      • Antzelevitch C.
      • Blaser S.
      • Gross G.
      • Turner L.
      • et al.
      Long QT, syndactyly, joint contractures, stroke and novel CACNA1C mutation: Expanding the spectrum of Timothy syndrome.
      ). Many TS variants were identified in a short region adjoining the sixth transmembrane segment of the CaVα1C protein (Fig. 1). The canonical TS1 variant Gly406Arg results from a de novo CACNA1C mutation in exon 8A (
      • Splawski I.
      • Timothy K.W.
      • Sharpe L.M.
      • Decher N.
      • Kumar P.
      • Bloise R.
      • et al.
      Ca(V)1.2 calcium channel dysfunction causes a multisystem disorder including arrhythmia and autism.
      ). An atypical form of TS type 2 is associated with the Gly402Ser and Gly402Arg variants in the alternatively spliced exon 8 (
      • Splawski I.
      • Timothy K.W.
      • Decher N.
      • Kumar P.
      • Sachse F.B.
      • Beggs A.H.
      • et al.
      Severe arrhythmia disorder caused by cardiac L-type calcium channel mutations.
      ). More recent de novo mutations have highlighted the importance of this region such as Glu407Gly/Ala (
      • Po C.
      • Zordan R.
      • Vecchi M.
      • Cerutti A.
      • Sartori S.
      • Trevisson E.
      • et al.
      Photosensitive epilepsy and long QT: expanding Timothy syndrome phenotype.
      ,
      • Colson C.
      • Mittre H.
      • Busson A.
      • Leenhardt A.
      • Denjoy I.
      • Fressard V.
      • et al.
      Unusual clinical description of adult with Timothy syndrome, carrier of a new heterozygote mutation of CACNA1C.
      ) and Arg518Cys/His (
      • Boczek N.J.
      • Ye D.
      • Jin F.
      • Tester D.J.
      • Huseby A.
      • Bos J.M.
      • et al.
      Identification and functional characterization of a novel CACNA1C-mediated cardiac disorder characterized by prolonged QT intervals with hypertrophic cardiomyopathy, congenital heart defects, and sudden cardiac death.
      ). These missense variants are causing a gain of function in the CaV1.2 channel as a result of slower inactivation kinetics that promote larger Ca2+ influx for the same depolarizing pulse (
      • Han D.
      • Xue X.
      • Yan Y.
      • Li G.
      Highlight article: dysfunctional Cav1.2 channel in Timothy syndrome, from cell to bedside.
      ). Nonetheless, functional outcomes of other TS mutations included marked loss of current density, a gain-of-function shift in activation, and increased window current (
      • Hennessey J.A.
      • Boczek N.J.
      • Jiang Y.H.
      • Miller J.D.
      • Patrick W.
      • Pfeiffer R.
      • et al.
      A CACNA1C variant associated with reduced voltage-dependent inactivation, increased CaV1.2 channel window current, and arrhythmogenesis.
      ). We have recently identified in the first intracellular region of CaVα1C a missense variant, Gly419Arg, from a patient with prolonged QT interval (≈500 ms), syndactyly, left ventricular noncompaction, and slight delay in neurodevelopment (
      • Kelu Bisabu K.
      • Zhao J.
      • Mokrane A.E.
      • Segura É.
      • Marsolais M.
      • Grondin S.
      • et al.
      Novel gain-of-function variant in CACNA1C associated with Timothy syndrome, multiple accessory pathways, and noncompaction cardiomyopathy.
      ). Unlike other TS variants located close to the high-affinity binding domain of CaVβ, CaV1.2 Gly419Arg exhibited a gain-of-function shift in the activation gating and no decrease in the channel current decay (
      • Kelu Bisabu K.
      • Zhao J.
      • Mokrane A.E.
      • Segura É.
      • Marsolais M.
      • Grondin S.
      • et al.
      Novel gain-of-function variant in CACNA1C associated with Timothy syndrome, multiple accessory pathways, and noncompaction cardiomyopathy.
      ).
      Herein, we explored the regulation of the long QTS variant Gly419Arg (G449R in the rabbit clone numbering). Glycine residues, inserted between the sixth transmembrane segment and the high-affinity binding site for CaVβ, have been shown to confer higher flexibility to this region (
      • Findeisen F.
      • Minor Jr., D.L.
      Disruption of the IS6-AID linker affects voltage-gated calcium channel inactivation and facilitation.
      ,
      • Papa A.
      • Kushner J.
      • Hennessey J.A.
      • Katchman A.N.
      • Zakharov S.I.
      • Chen B.X.
      • et al.
      Adrenergic Ca(V)1.2 activation via Rad phosphorylation converges at α(1C) I-II loop.
      ,
      • Almagor L.
      • Chomsky-Hecht O.
      • Ben-Mocha A.
      • Hendin-Barak D.
      • Dascal N.
      • Hirsch J.A.
      The role of a voltage-dependent Ca2+ channel intracellular linker: A structure-function analysis.
      ), leading to reduced basal L-type CaV channel activity in cardiomyocytes (
      • Papa A.
      • Kushner J.
      • Hennessey J.A.
      • Katchman A.N.
      • Zakharov S.I.
      • Chen B.X.
      • et al.
      Adrenergic Ca(V)1.2 activation via Rad phosphorylation converges at α(1C) I-II loop.
      ). The reverse proposition, removing or substituting glycine residues in this locus, decreased the linker flexibility (
      • Almagor L.
      • Chomsky-Hecht O.
      • Ben-Mocha A.
      • Hendin-Barak D.
      • Dascal N.
      • Hirsch J.A.
      The role of a voltage-dependent Ca2+ channel intracellular linker: A structure-function analysis.
      ). Herein, we present evidence that the novel variant, whereby a conserved glycine is substituted by a larger arginine residue, promotes stronger activity (peak current density and activation gating) at physiological voltages akin to a constitutively hyperactive channel. This hyperactive mode was reconstituted in the WT channel by coexpression with CaM WT or pseudophosphorylated surrogates CaM T79D or CaM S81D and was abolished by the CaM antagonist W-13. In contrast, the functional parameters of the clinical glycine to arginine variant remained remarkably insensitive to these treatments. Substitution with the α-helix breaker proline residue yielded opposite results with channels more reluctant to open at physiological voltages but more likely to respond to the modulation by CaM. Altogether, the functional characterization of the glycine to arginine variant provides mechanistic insight on the regulation of CaV1.2 by CaM (sometimes referred to as calmodulation) and specifically the role played by the I–II linker as relaying the signal to the channel activation gate.

      Results

      Glycine substitution stimulates activation gating and peak current density of CaV1.2

      It is well known that gain-of-function mutations G402S and G406R (Fig. 1) decelerate inactivation kinetics (
      • Splawski I.
      • Timothy K.W.
      • Sharpe L.M.
      • Decher N.
      • Kumar P.
      • Bloise R.
      • et al.
      Ca(V)1.2 calcium channel dysfunction causes a multisystem disorder including arrhythmia and autism.
      ,
      • Splawski I.
      • Timothy K.W.
      • Decher N.
      • Kumar P.
      • Sachse F.B.
      • Beggs A.H.
      • et al.
      Severe arrhythmia disorder caused by cardiac L-type calcium channel mutations.
      ,
      • Barrett C.F.
      • Tsien R.W.
      The Timothy syndrome mutation differentially affects voltage- and calcium-dependent inactivation of CaV1.2 L-type calcium channels.
      ,
      • Dick I.E.
      • Joshi-Mukherjee R.
      • Yang W.
      • Yue D.T.
      Arrhythmogenesis in Timothy Syndrome is associated with defects in Ca(2+)-dependent inactivation.
      ,
      • Raybaud A.
      • Dodier Y.
      • Bissonnette P.
      • Simoes M.
      • Bichet D.G.
      • Sauvé R.
      • et al.
      The role of the GX9GX3G motif in the gating of high voltage-activated Ca2+ channels.
      ). In contrast, the inactivation kinetics of the gain-of-function TS CaV1.2 G419R variant classified as a pathogenic TS variant (
      • Tarnovskaya S.I.
      • Kostareva A.A.
      • Zhorov B.S.
      L-type calcium channel: predicting pathogenic/likely pathogenic status for variants of uncertain clinical significance.
      ) were slightly faster than CaV1.2 WT (
      • Kelu Bisabu K.
      • Zhao J.
      • Mokrane A.E.
      • Segura É.
      • Marsolais M.
      • Grondin S.
      • et al.
      Novel gain-of-function variant in CACNA1C associated with Timothy syndrome, multiple accessory pathways, and noncompaction cardiomyopathy.
      ). The faster inactivation kinetics were associated with increased peak current density and a leftward shift in the voltage of activation, leading to an increased probability of channel being open at physiological voltages without any significant change in the voltage dependence of inactivation (Table 1). Glycine residues close to the pore (e.g., Gly402 and Gly406) appear to be essential to convey the movement of the inactivation gate, whereas inserting glycine residues further away and closer to the high-affinity binding domain for CaVβ (Fig. 1) yielded opposite results (
      • Findeisen F.
      • Minor Jr., D.L.
      Disruption of the IS6-AID linker affects voltage-gated calcium channel inactivation and facilitation.
      ,
      • Papa A.
      • Kushner J.
      • Hennessey J.A.
      • Katchman A.N.
      • Zakharov S.I.
      • Chen B.X.
      • et al.
      Adrenergic Ca(V)1.2 activation via Rad phosphorylation converges at α(1C) I-II loop.
      ). Increased flexibility within this stretch has been argued to loosen up the interaction between CaVβ and CaV1.2 (
      • Papa A.
      • Kushner J.
      • Hennessey J.A.
      • Katchman A.N.
      • Zakharov S.I.
      • Chen B.X.
      • et al.
      Adrenergic Ca(V)1.2 activation via Rad phosphorylation converges at α(1C) I-II loop.
      ). We validated that the substitution of the glycine residue at position 449 (rabbit numbering) does not impair the interaction with the canonical CaVβ and CaVα2δ1 subunits (Fig. 2). The latter observation is in line with the recent demonstration that interaction with CaVα2δ1 involves extracellular loops of CaV1.2 (
      • Segura E.
      • Bourdin B.
      • Tétreault M.P.
      • Briot J.
      • Allen B.G.
      • Mayer G.
      • et al.
      Proteolytic cleavage of the hydrophobic domain in the Ca(V)α2δ1 subunit improves assembly and activity of cardiac Ca(V)1.2 channels.
      ,
      • Bourdin B.
      • Briot J.
      • Tétreault M.P.
      • Sauvé R.
      • Parent L.
      Negatively charged residues in the first extracellular loop of the L-type Ca(V)1.2 channel anchor the interaction with the Ca(V)α2δ1 auxiliary subunit.
      ,
      • Briot J.
      • Mailhot O.
      • Bourdin B.
      • Tétreault M.P.
      • Najmanovich R.
      • Parent L.
      A three-way inter-molecular network accounts for the Ca(V)α2δ1-induced functional modulation of the pore-forming Ca(V)1.2 subunit.
      ). We thus turned to investigate functional regulation by the ubiquitous CaM (
      • Saimi Y.
      • Kung C.
      Calmodulin as an ion channel subunit.
      ). Disease-causing mutations at CaM proteins lead to major cardiac dysfunction, and in turn, mutations at the CaM-binding site of ion channels have been associated with a host of diseases (
      • Urrutia J.
      • Aguado A.
      • Muguruza-Montero A.
      • Núñez E.
      • Malo C.
      • Casis O.
      • et al.
      The crossroad of ion channels and calmodulin in disease.
      ).
      Table 1Electrophysiological properties of CaV1.2 WT and G449R with W-13
      Cav1.2CaMn/NElectrophysiological properties
      Peak I (pA/pF)E0.5,act (mV)R100n/NE0.5,inact (mV)
      WTNative30/7−15 ± 4−10 ± 30.65 ± 0.0414/5−33 ± 3
      +W-1317/4−8 ± 2

      p = 0.002 versus control
      −12 ± 40.57 ± 0.02

      p < 0.001 versus control
      6/3−33 ± 3
      G449RNative31/4−33 ± 12

      p < 0.001 versus WT
      −17 ± 3

      p < 0.001 versus WT
      0.52 ± 0.03

      p < 0.001 versus WT
      13/4−35 ± 3
      +W-1312/1−38 ± 8−17 ± 30.51 ± 0.035/2−34 ± 2
      Effects of CaM inhibitor W-13 on the gating properties of CaV1.2 WT or G449R channels with native or endogenous CaM. CaV1.2 WT or G449R were coexpressed in HEKT cells with CaVβ2a and CaVα2δ1. Whole-cell currents were measured in the presence of 2 mM Ca2+ in the extracellular medium. E0.5,inact values were estimated after a 5 s long depolarizing pulse to 0 mV. Fractional currents were fitted to Boltzmann equations as described in the Experimental procedures section. The R100 values report the relative current decay observed 100 ms after the peak current. n/N refers to the number of cells/transfections measured in each condition of study. Mean ± SD are shown. Statistical analysis was carried out by one-way ANOVA and a Bonferroni post hoc test.
      Figure thumbnail gr2
      Figure 2G449R interacts with CaVβ proteins. HEKT cells were transiently transfected with CaV1.2 WT (WT) or CaV1.2 G449R (GR) with cMyc-tagged CaVβ3 or cMyc-tagged CaVβ2a. CaVα2δ1 was present throughout. Cell lysates were immunoprecipitated overnight with anti-cMyc magnetic beads (Pierce Anti-c-Myc Magnetic Beads; catalog no.: 88842, Thermo Fisher Scientific) to capture the CaVβ, eluted in a 2× Laemmli buffer and fractionated by 8% SDS-PAGE gels. A, immunoblotting was carried out on total proteins (20 μg) collected from the cell lysates before the immunoprecipitation assay (total proteins). The signal for the housekeeping protein GAPDH is shown below each blot. B, immunoblotting of “IP-proteins” was carried out after eluting the protein complexes from the beads. All immunoblots were carried out in parallel under the same transfection and extraction conditions. Western blotting was carried out with either anti-CaVβ3 (Alomone; catalog no.: ACC008, 1:10,000 dilution), anti-CaVβ2a (Alomone; catalog no.: ACC105, 1:1000 dilution), anti-CaV1.2 directed against CaVα1C (Alomone; catalog no.: ACC003, 1:250 dilution), anti-CaVα2δ1 (Alomone; catalog no.: ACC015, 1:1000 dilution), and GAPDH (Sigma; 1:10,000 dilution) with an anti-rabbit as secondary antibody (Jackson ImmunoResearch; 1:10,000 dilution). Signals were detected with the enhanced chemiluminescence substrate. Blots were visualized with the ChemiDoc Touch system (Bio-Rad). Molecular weights were estimated using Image Lab software, version 5.2 (Bio-Rad) by linear regression of standard molecular weight markers. GAPDH, CaVβ3, CaVβ2a, CaVα2δ1, and CaV1.2 proteins migrated (in kilodalton) at 35, 60, 80, 175, and 250 kDa, respectively. From left to right in A and B: lane 1: CaV1.2 WT + CaVα2δ1 + CaVβ2a; lane 2: CaV1.2 G449R + CaVα2δ1 + CaVβ2a; lane 3: CaV1.2 WT + CaVα2δ1 + CaVβ3; and lane 4: CaV1.2 G449R + CaVa2δ1 + CaVβ3. HEKT, human embryonic kidney 293T cell line; IP, immunoprecipitation.

      CaM antagonist W-13 blocks CaV1.2 WT but not G449R whole-cell currents

      Functional regulation of CaV1.2 WT by endogenous CaM was examined with the membrane-permeable naphthalenesulfonamide derivative CaM antagonist W-13. Under our conditions, CaV1.2 WT currents activated at −35 mV and reached the peak inward current at +5 mV. As seen in Figure 3A, the peak current density of CaV1.2 WT was reduced by about 50% from −15 ± 4 pA/pF to −8 ± 2 pA/pF after adding 10 μM W-13 into the bath. Decay of the CaV1.2 WT current was accelerated in the presence of W-13, which reduced the noninactivating component of CaV1.2 at the end of 100 ms depolarization (R100) from 0.65 ± 0.04 to 0.57 ± 0.02 (p < 0.001) (Table 1). Of note, W-13 did not impair Ca2+-dependent facilitation in cardiac cells (
      • Tiaho F.
      • Piot C.
      • Nargeot J.
      • Richard S.
      Regulation of the frequency-dependent facilitation of L-type Ca2+ currents in rat ventricular myocytes.
      ). Under the same experimental conditions, the inhibitory effect of W-13 on the current amplitude and the acceleration of current decay were blunted in the G449R construct with −33 ± 12 versus −38 ± 8 pA/pF (Fig. 3B) suggesting that the glycine substitution prevents the channel modulation by endogenous CaM.
      Figure thumbnail gr3
      Figure 3CaV1.2 G449R is insensitive to W-13. A, representative CaV1.2 WT current traces recorded in the presence of native/endogenous CaM from HEKT cells before (left) and after (middle) the application of W-13. Peak current densities of CaV1.2 WT currents are plotted against the applied voltages and fitted by a Boltzmann equation (right). Incubation with 10 μM W-13 for 15 min reduced the CaV1.2 WT current density by approximately 50%, from −15 ± 4 pA/pF under control conditions versus −8 ± 2 pA/pF in the presence of W-13. B, representative CaV1.2 G449R current traces recorded in the presence of native/endogenous CaM from HEKT cells before (left) and after (middle) W-13 treatment. In contrast to CaV1.2 WT, CaV1.2 G449R currents were unaffected by the W-13 and did not display any inhibition in the peak current density (right). C, representative CaV1.2 WT current traces cotransfected with CaM WT recorded from HEKT cells before (left) and after (middle) W-13 treatment. Overexpression of CaM WT significantly enhanced the current density of CaV1.2 WT, whereas only approximately, 10% of peak CaV1.2 currents remained following W-13 treatment (right). Peak current densities of CaV1.2 WT coexpressed with CaM WT are plotted against the applied voltages and fitted by a Boltzmann-like equation. D, representative CaV1.2 G449R current traces cotransfected with CaM WT recorded from HEKT cells before (left) and after (middle) application of W-13. Unlike the CaV1.2 WT channels, overexpression of CaM WT did not alter the CaV1.2 G449R currents. Furthermore, inhibition of the CaV1.2 G449R peak currents by W-13 was undetectable (right). The vertical scale bars are 10 pA/pF, and the horizontal scale bars are 100 ms throughout. All biophysical values are reported in Table 1, Table 2, Table 3. CaM, calmodulin; HEKT, human embryonic kidney 293T cell line.

      CaM promotes the activity of CaV1.2 WT

      In a typical cellular environment, CaM targets could far exceed that of free endogenous CaM (
      • Villarroel A.
      • Taglialatela M.
      • Bernardo-Seisdedos G.
      • Alaimo A.
      • Agirre J.
      • Alberdi A.
      • et al.
      The ever changing moods of calmodulin: how structural plasticity entails transductional adaptability.
      ,
      • Persechini A.
      • Stemmer P.M.
      Calmodulin is a limiting factor in the cell.
      ). To further explore the regulation by CaM, CaM WT was overexpressed along with the complementary DNA (cDNA) coding for the channel subunits. Overexpressing CaM has been shown to compete with endogenous CaM WT and was successfully used to reveal the mechanistic actions of CaM on voltage-activated Ca2+ channels (
      • DeMaria C.D.
      • Soong T.W.
      • Alseikhan B.A.
      • Alvania R.S.
      • Yue D.T.
      Calmodulin bifurcates the local Ca2+ signal that modulates P/Q-type Ca2+ channels.
      ,
      • Simms B.A.
      • Souza I.A.
      • Zamponi G.W.
      Effect of the Brugada syndrome mutation A39V on calmodulin regulation of Cav1.2 channels.
      ,
      • Ravindran A.
      • Lao Q.Z.
      • Harry J.B.
      • Abrahimi P.
      • Kobrinsky E.
      • Soldatov N.M.
      Calmodulin-dependent gating of Ca(v)1.2 calcium channels in the absence of Ca(v)beta subunits.
      ,
      • Limpitikul W.B.
      • Dick I.E.
      • Joshi-Mukherjee R.
      • Overgaard M.T.
      • George Jr., A.L.
      • Yue D.T.
      Calmodulin mutations associated with long QT syndrome prevent inactivation of cardiac L-type Ca(2+) currents and promote proarrhythmic behavior in ventricular myocytes.
      ). Representative current traces from cells coexpressing CaV1.2 WT and CaM WT are shown in Figure 3C. As seen, under these conditions, the peak current density nearly doubled up from −15 ± 4 to −28 ± 8 pA/pF (p < 0.001 as compared with endogenous CaM) to reach values not significantly different than G449R under the same conditions (p > 0.05). CaM WT shifted the E0.5,act to hyperpolarized potentials (p < 0.05) and slightly accelerated the inactivation kinetics (Table 2). CaM enhanced the fraction of CaV1.2 WT currents that was inhibited by W-13, with about 90% reduction in peak current density, from 28 ± 8 pA/pF for control versus 3.3 ± 0.7 pA/pF for W-13 (p = 0.001). Overexpressing CaM WT caused undetectable changes in the peak current density, the voltage of activation, and the current decay of G449R that remained unaffected by the W-13 treatment (Fig. 3D) (Table 3). G449R functionally behaved like it intrinsically adopted a maximally active mode (
      • Nowycky M.C.
      • Fox A.P.
      • Tsien R.W.
      Long-opening mode of gating of neuronal calcium channels and its promotion by the dihydropyridine calcium agonist Bay K 8644.
      ).
      Table 2Electrophysiological properties of CaV1.2 WT with CaM WT and phosphorylation surrogates
      CaV1.2CaMn/NElectrophysiological properties
      Peak current density (pA/pF)E0.5,act (mV)R100
      CaV1.2 WTCaM WT28/6−28 ± 8

      p < 0.001 versus native CaM
      −14 ± 3

      p = 0.002 versus native CaM
      0.60 ± 0.03

      p < 0.001 versus native CaM
      +W-139/2−3.3 ± 0.7

      p = 0.001 versus control
      −14 ± 30.65 ± 0.03

      p = 0.002 versus control
      CaM T79A18/2−13 ± 5

      p < 0.001 versus CaM WT, T79D, S81D
      −13 ± 20.69 ± 0.03

      p < 0.01 versus native CaM

      p < 0.01 versus CaM T79D, S81D
      +W-1316/2−14 ± 3−14 ± 30.56 ± 0.03

      p < 0.001 versus control
      CaM T79D15/4−29 ± 7

      p < 0.001 versus native CaM

      p < 0.001 versus CaM T79A, S81A
      −15 ± 3

      p < 0.001 versus native CaM
      0.55 ± 0.03

      p < 0.001 versus native CaM

      p < 0.001 versus CaM WT, T79A, S81A, S81D
      +W-1310/2−4 ± 1

      p < 0.001 versus control
      −12 ± 20.71 ± 0.02

      p < 0.001 versus control
      CaM S81A17/2−13 ± 4

      p < 0.001 versus CaM WT, T79D, S81D
      −13 ± 20.66 ± 0.03

      p < 0.01 versus CaM WT, T79D, S81D
      +W-138/1−12 ± 2−9 ± 30.61 ± 0.02

      p = 0.02 versus control
      CaM S81D14/1−32 ± 8

      p = 0.001 versus native CaM

      p = 0.001 versus CaM T79A, S81A
      −13 ± 30.62 ± 0.02

      p < 0.01 versus CaM T79A, T79D, S81A
      +W-139/1−5 ± 1

      p < 0.001 versus control
      −12 ± 20.70 ± 0.02

      p < 0.001 versus control
      Effects of CaM inhibitor W-13 on the biophysical properties of CaV1.2 WT channels. CaV1.2 WT was coexpressed with CaVβ2a, CaVα2δ1, and CaM WT or phosphoresistant and phosphomimetic variants (T79A, T79D, S81A, or S81D). Activation properties (E0.5,act) were estimated from the I–V relationships and fitted as described in the Experimental procedures section. The R100 values report the relative current decay observed 100 ms after the peak current. n/N refers to the number of cells/transfections measured in each condition of study. Mean ± SD are shown. Statistical significance of observed differences was evaluated using one-way ANOVA and Bonferroni test (p < 0.05). As seen, coexpression with CaM WT, CaM T79D, or T81D potentiated Ca2+ currents that were sensitive to CaM antagonists and sped up current decay. In contrast, coexpression with CaM T79A or CaM S81A (phosphoresistant analogs) produced Ca2+ currents similar to the ones measured in the presence of endogenous/native CaM in terms of peak current density and current decay. Nonetheless, CaM T79A and CaM S81A did not prevent the robust activation and rendered Ca2+ currents insensitive to inhibition by W-13.
      Table 3Electrophysiological properties of CaV1.2 G449R with CaM WT and phosphorylation surrogates
      Cav1.2CaMn/NElectrophysiological properties
      Peak current density (pA/pF)E0.5,act (mV)R100
      CaV1.2 G449RCaM WT20/3−35 ± 10−16 ± 20.50 ± 0.03
      +W-139/2−35 ± 9−17 ± 30.54 ± 0.01

      p = 0.04 versus control
      CaM T79A15/2−32 ± 10−16 ± 20.50 ± 0.02
      +W-1314/2−33 ± 10−17 ± 30.45 ± 0.03

      p < 0.001 versus control
      CaM T79D24/3−35 ± 9−18 ± 40.50 ± 0.02
      +W-138/1−32 ± 9−16 ± 20.50 ± 0.02
      CaM S81A13/1−34 ± 9−18 ± 20.48 ± 0.02

      p = 0.002 versus native CaM

      p = 0.02 versus CaM S81D
      +W-139/1−33 ± 7−19 ± 20.51 ± 0.02
      CaM S81D9/1−35 ± 8−17 ± 30.52 ± 0.02

      p = 0.02 versus CaM S81A
      +W-1310/1−33 ± 10−15 ± 30.50 ± 0.02
      Effects of CaM inhibitor W-13 on the biophysical properties of CaV1.2 G449R channels. CaV1.2 G449R was coexpressed with CaVβ2a, CaVα2δ1, and CaM WT or phosphoresistant and phosphomimetic variants (T79A, T79D, S81A, or S81D). Activation properties (E0.5,act) were estimated from the I–V relationships and fitted as described in the Experimental procedures section. The R100 values report the relative current decay observed 100 ms after the peak current. n/N refers to the number of cells/transfections measured in each condition of study. Mean ± SD are shown. Statistical analysis was evaluated using one-way ANOVA and Bonferroni post hoc test. As seen, all experimental conditions yielded whole-cell Ca2+ currents that were not significantly different from one another (p > 0.05).
      CaM was previously shown to bind to the I–II linker in addition to other intracellular sites within CaV1.2 (
      • Asmara H.
      • Minobe E.
      • Saud Z.A.
      • Kameyama M.
      Interactions of calmodulin with the multiple binding sites of Cav1.2 Ca2+ channels.
      ). Pull-down assays demonstrated that CaM is tethered to the WT and the G449R channel complex (Fig. 4). In fact, the protein signal for G449R appeared to be more intense, hinting that it could maintain a stronger interaction with CaM. Enhanced channel activity could arise because of improved activation gating and/or increase in the relative cell surface protein expression/stability. Previous studies have reported that CaM enhances trafficking of CaV1.2 in human embryonic kidney (HEK) cells (
      • Wang H.G.
      • George M.S.
      • Kim J.
      • Wang C.
      • Pitt G.S.
      Ca/Calmodulin regulates trafficking of Cav1.2 Ca channels in cultured hippocampal neurons.
      ). CaM-induced increases in peak current density may reflect an improved surface expression of channel complexes. To sort this issue, we performed a series of cell fractionation assays. As seen in Figure 5 in the presence of endogenous CaM, the signal for CaV1.2 WT was stronger in the total membrane protein fraction (Fig. 5A, lane 3) than in the cell surface protein fraction (Fig. 5A, lane 4). Under the same conditions, the signal for CaV1.2 G449R was stronger in the cell surface protein fraction suggesting that G449R is better trafficked or more stable than channel complexes including the WT protein and endogenous CaM. Differences in the relative channel expression were obliterated when the channel complexes were overexpressed with CaM WT (Fig. 5B). Under these conditions, the WT and G449R channel complexes are similarly found in the cell surface fraction. Overexpression of CaM enhanced the cell surface trafficking of CaV1.2 WT, which can account in part for the increased peak current density and possibly the increase in the activation gating.
      Figure thumbnail gr4
      Figure 4Calmodulin (CaM) pulls down the L-type calcium channel. HEKT cells were transiently transfected with CaVβ2a, CaM WT, and either CaV1.2 WT (WT) or G449R (GR). CaM was captured by the anti-His–coated beads. A, proteins were homogenized, and a fraction of this solution (referred to as total) was set aside to validate protein expression. B, coimmunoprecipitation was carried out with anti-His magnetic beads. The bound proteins were eluted (referred to as pull-down) and electrophoresed on a 6% SDS-polyacrylamide gel or a 10% SDS-polyacrylamide gel for CaM and GAPDH before being transferred onto a nitrocellulose membrane. Western blotting was carried out with anti-CaVβ2 (Alomone; catalog no.: ACC105, 1:1000 dilution), anti-CaV1.2 (Alomone; catalog no.: ACC003, 1:250 dilution) with an anti-rabbit as secondary antibody (Jackson ImmunoResearch, 1:10,000 dilution), and anti-CaM (Millipore; catalog no.: 05-193, 1:1000 dilution) with an anti-mouse as secondary antibody (Jackson ImmunoResearch, 1:10,000 dilution). Molecular weights were estimated using Image Lab software, version 5.2 (Bio-Rad) by linear regression of standard molecular weight markers. CaV1.2 WT and G449R, CaVβ2a, and CaM proteins were translated at the expected molecular masses of 250, 70, and 18 to 24 kDa, respectively. CaV1.2 WT and G449R were successfully pulled indicating that CaV1.2 G449R interacts with CaVβ2a and CaM. This result was successfully obtained from four independent transfections carried out over the course of 3 months. HEKT, human embryonic kidney 293T cell line.
      Figure thumbnail gr5
      Figure 5CaM promotes the cell surface localization of CaV1.2 WT but not CaV1.2 G449R. A, HEKT cells were transiently transfected with CaV1.2 WT + CaVα2δ1 + CaVβ2a (left) and CaV1.2 G449R + CaVα2δ1 + CaVβ2a (right) in the presence of native/endogenous CaM. B, HEKT cells were transiently transfected with CaV1.2 WT + CaVα2δ1 + CaVβ2a + CaM WT (left) and CaV1.2 G449R + CaVα2δ1 + CaVβ2a + CaM WT (right). Two days after transfection, the cells were lysed, and cell fractions were obtained through preparative ultracentrifugation as described in the section. Western blotting was carried out for the four protein fractions found in lanes 1 to 4; lane 1: total proteins; lane 2: cytoplasmic proteins; lane 3: total membrane proteins; and lane 4: plasma membrane proteins. The proteins were probed with the following antibodies: CaV1.2 (Alomone; catalog no.: ACC003, 1:250 dilution) with anti-rabbit (1:10,000 dilution); CaVα2δ1 (Alomone; catalog no.: ACC015, 1:1000 dilution) with anti-rabbit (1:10,000 dilution); CaVβ2a (Alomone; catalog no.: ACC105, 1:1000 dilution) with anti-rabbit (1:10,000 dilution); CaM (Millipore; catalog no.: 05-193, 1:1000 dilution) with anti-mouse (1:10,000 dilution); His (Invitrogen; catalog no.: 71700, 1:1000 dilution) with anti-mouse (1:10,000 dilution); and cadherin (Pan-cadherin; Thermo Fisher; catalog no.: 71-7100, 1:1000 dilution) with anti-rabbit (1:10,000). Cadherin was used as a marker for the plasma membrane. The membrane was cut at 115 and 28 kDa to probe first CaV1.2, CaVβ2a, and CaM. Membranes were then stripped and reprobed with antibodies against the proteins: CaVα2δ1, cadherin, and housekeeping GAPDH (Sigma; catalog no.: G9545, 1:10,000 dilution with anti-rabbit [1:10,000 dilution]). Each lane was loaded with 20 μg proteins. The lines to the left of the blots indicate the position of the molecular markers, and the value is provided in kilodalton. The molecular masses were estimated by linear regression and interpolation from the molecular mass markers using the Image Lab software, version 5.2 (Bio-Rad). As seen in A, in the presence of endogenous CaM, the signal for CaV1.2 WT was stronger in the total membrane protein fraction (lane 3) than in the cell surface protein fraction (lane 4). Under the same conditions, the signal for CaV1.2 G449R was stronger in the cell surface protein fraction. B, demonstrates that under conditions where CaM was overexpressed, the signal for CaV1.2 WT and CaV1.2 G449R is stronger in the cell surface protein fraction. Along with CaV1.2, CaVα2δ1 (as previously reported (
      • Segura E.
      • Bourdin B.
      • Tétreault M.P.
      • Briot J.
      • Allen B.G.
      • Mayer G.
      • et al.
      Proteolytic cleavage of the hydrophobic domain in the Ca(V)α2δ1 subunit improves assembly and activity of cardiac Ca(V)1.2 channels.
      )) and CaVβ2a are found in the cell surface fraction but not CaM and GAPDH. This observation suggests that the interaction of CaM with the pore-forming subunit is very robust during the cell surface export and is compatible with the binding and unbinding kinetics of CaM in other cell types (
      • Persechini A.
      • Stemmer P.M.
      Calmodulin is a limiting factor in the cell.
      ). No significant signal was found in the cytoplasmic fraction for CaVα2δ1 and the membrane-anchored CaVβ2a. This result was successfully obtained from two independent transfections carried out over the course of 2 months. CaM, calmodulin; HEKT, human embryonic kidney 293T cell line.

      Phosphomimetic surrogates of CaM are associated with increased activity of CaV1.2 WT

      The precise mechanism though which W-13 inhibits CaM is not currently known, but it has been shown that W-13 bends the flexible linker of CaM between Met78 and Glu82 (
      • Osawa M.
      • Swindells M.B.
      • Tanikawa J.
      • Tanaka T.
      • Mase T.
      • Furuya T.
      • et al.
      Solution structure of calmodulin-W-7 complex: the basis of diversity in molecular recognition.
      ,
      • Osawa M.
      • Kuwamoto S.
      • Izumi Y.
      • Yap K.L.
      • Ikura M.
      • Shibanuma T.
      • et al.
      Evidence for calmodulin inter-domain compaction in solution induced by W-7 binding.
      ), a region that harbors two important phosphorylation sites Thr79 and Ser81 (
      • Quadroni M.
      • James P.
      • Carafoli E.
      Isolation of phosphorylated calmodulin from rat liver and identification of the in vivo phosphorylation sites.
      ). Phosphorylation at these two sites causes structural changes in the relative orientation of the C- and N-lobes, which in turn modulate the interaction of CaM with its protein targets (
      • Villalobo A.
      The multifunctional role of phospho-calmodulin in pathophysiological processes.
      ,
      • Tabernero L.
      • Taylor D.A.
      • Chandross R.J.
      • VanBerkum M.F.
      • Means A.R.
      • Quiocho F.A.
      • et al.
      The structure of a calmodulin mutant with a deletion in the central helix: implications for molecular recognition and protein binding.
      ). To evaluate the structural properties of the flexible linker, we introduced phosphomimetic and phosphoresistant mutations on CaM by individually changing phosphorylation sites Thr79 and Ser81 to alanine (A) or aspartate (D), respectively, the latter mimicking the negative charge change induced by post-translational modification.
      Overexpression of phosphoresistant CaM T79A (Fig. 6A, left) or CaM S81A (Fig. 6C, left) abrogated the upregulation of Cav1.2 WT currents by CaM WT. The peak current densities of CaV1.2 WT were −13 ± 5 pA/pF for CaM T79A (p < 0.001) and −13 ± 4 pA/pF for CaM S81A (p < 0.001) as compared with −28 ± 8 pA/pF when coexpressed with CaM WT (Table 2). The two phosphoresistant mutations failed to increase the peak current density and activation gating. Overexpression with phosphomimetic CaM variants T79D (Fig. 6B, left) or S81D (Fig. 6D, left) produced peak current densities and activation potentials comparable to those obtained with CaV1.2 WT + CaM WT (Fig. 6, E and F). The structural properties of the flexible linker were also shown to regulate the activity of Ca2+-activated SK2 channels, although in this latter case, CaM T79D reduced channel activity (
      • Bildl W.
      • Strassmaier T.
      • Thurm H.
      • Andersen J.
      • Eble S.
      • Oliver D.
      • et al.
      Protein kinase CK2 is coassembled with small conductance Ca(2+)-activated K+ channels and regulates channel gating.
      ).
      Figure thumbnail gr6
      Figure 6Phosphomimetic CaM T79D and S81D upregulate CaV1.2 WT channels. A and C, middle, CaV1.2 WT was coexpressed with the “phosphoresistant” CaM mutations T79A or S81A. Overexpression of CaM T79A or S81A failed to enhance the currents and was insensitive to W-13. A and C, right, average I–V curves of CaV1.2 WT coexpressed with CaM T79A or S81A. The peak current densities were not different between control and W-13 treatment. B and D, left, middle, CaV1.2 WT current traces recorded from HEKT cells after coexpression with phosphomimetic CaM T79D or S81D. Overexpression of CaM T79D or CaM S81D boosted CaV1.2 peak currents that were sharply abolished by the extracellular application of W-13. The vertical scale bars are 10 pA/pF, and the horizontal scale bars are 100 ms throughout. B and D, right, average I–V curves of CaV1.2 WT coexpressed with CaM T79D or S81D for control and W-13 treatment. E and F, the distribution of the peak current densities and E0.5,act for control and W-13 are summarized as filled circles for CaV1.2 WT coexpressed with either CaM WT (black), T79A (red), T79D (blue), S81A (green), or S81D (light purple). The mean data ± SD are shown as gray hyphens. The values of the average peak current densities and E0.5,act are listed in . CaM, calmodulin; HEKT, human embryonic kidney 293T cell line.
      CaM inhibitor W-13 substantially diminished CaV1.2 + CaM T79D or CaV1.2 + CaM S81D currents by around 85% (Fig. 6, B and D, middle) with values of −4 ± 1 pA/pF (Fig. 6B, right) and −5 ± 1 pA/pF, respectively (Fig. 6D, right). Treatment with W-13 did not however further inhibit CaV1.2 channels coexpressed with phosphoresistant CaM T79A and S81A (Fig. 6A and C, middle and right; Fig. 6E) and had little impact on the activation gating under any of these conditions (Fig. 6F and Table 2).
      The modulation of CaM WT on the function of CaV1.2 WT was equivalent to the action of phosphomimetic surrogates CaM T79D and S81D, suggesting that the phosphorylated form of CaM is responsible for the functional upregulation of CaV1.2. Roughly 10 to 45% of endogenous CaM is constitutively phosphorylated in vivo by casein kinase II (CK2) (
      • Quadroni M.
      • James P.
      • Carafoli E.
      Isolation of phosphorylated calmodulin from rat liver and identification of the in vivo phosphorylation sites.
      ,
      • Plancke Y.D.
      • Lazarides E.
      Evidence for a phosphorylated form of calmodulin in chicken brain and muscle.
      ,
      • Quadroni M.
      • L'Hostis E.L.
      • Corti C.
      • Myagkikh I.
      • Durussel I.
      • Cox J.
      • et al.
      Phosphorylation of calmodulin alters its potency as an activator of target enzymes.
      ), and in vitro studies confirmed that CaM Thr79 and Ser81 are the most likely targets (
      • Sacks D.B.
      • Davis H.W.
      • Crimmins D.L.
      • McDonald J.M.
      Insulin-stimulated phosphorylation of calmodulin.
      ,
      • Sacks D.B.
      • Mazus B.
      • Joyal J.L.
      The activity of calmodulin is altered by phosphorylation: modulation of calmodulin function by the site of phosphate incorporation.
      ). Experiments were thus performed in the presence of 4,5,6,7-tetrabromobenzotriazole (TBB; Tocris, Bio-Techne), a specific inhibitor of CK2. As shown in Table 4, TBB significantly decreased the peak current density by ≈70% and right shifted the activation gating of whole-cell currents recorded in the presence of CaV1.2 WT with native CaM. Furthermore, TBB annihilated the impact of overexpressing CaM WT on the peak current density of CaV1.2 WT. The impact of TBB was comparable to the disrupting effect of W-13 and much greater than the coexpression with either CaM T79A or S81A.
      Table 4Electrophysiological properties of CaV1.2 WT and G449R with TBB
      Cav1.2CaMn/NElectrophysiological properties
      Peak current density (pA/pF)E0.5,act (mV)R100
      CaV1.2 WTNative CaM30/7−15 ± 4−10 ± 30.65 ± 0.04
      +TBB6/2−4 ± 1

      p = 0.001 versus control
      −5 ± 1

      p = 0.001 versus control
      0.80 ± 0.01

      p < 0.001 versus control
      CaM WT28/6−28 ± 8−14 ± 30.60 ± 0.03
      +TBB5/1−4.9 ± 0.7

      p < 0.001 versus control
      −8 ± 2

      p = 0.001 versus control
      0.73 ± 0.01

      p < 0.001 versus control
      CaV1.2 G449RNative CaM31/4−33 ± 12−17 ± 30.52 ± 0.03
      +TBB7/2−15 ± 6

      p = 0.001 versus control
      −15 ± 30.55 ± 0.03
      CaM WT20/3−35 ± 10−16 ± 20.50 ± 0.03
      +TBB7/2−13 ± 5

      p < 0.001 versus control
      −13 ± 4

      p = 0.06 versus control
      0.57 ± 0.03

      p < 0.01 versus control
      Effects of TBB, the cell-permeable inhibitor of CK2 on the biophysical properties of CaV1.2 WT and CaV1.2 G449R channels. CaV1.2 (WT or G449R) was coexpressed with CaVβ2a, CaVα2δ1, and CaM WT as indicated. Two days after transfection, experiments were performed in the presence of 2.5 μM TBB, usually regarded as a membrane-permeable specific inhibitor of CK2. Activation properties (E0.5,act) were estimated from the I–V relationships and fitted as described in the Experimental procedures section. The R100 values report the relative current decay observed 100 ms after the peak current. n/N refers to the number of cells/transfections measured in each condition of the study. Mean ± SD are shown. Statistical analysis was evaluated using one-way ANOVA and Bonferroni post hoc test. As seen, TBB significantly decreased the channel peak current density under all conditions. It also significantly right shifted the activation gating of CaV1.2 WT but not of CaV1.2 G449R.
      Ca2+ binding to CaM remains a prerequisite step for driving the channel complex into its higher functioning mode. Overexpression of the Ca2+-free form of CaM (CaM1234 or CaM D20A/D56A/D93A/D129A) decelerated, as expected, the CDI kinetics (Table 5). It also abrogated the increased peak current density and restored its activation gating to the level observed in the presence of endogenous CaM.
      Table 5Effect of CaM1234 on electrophysiological properties of CaV1.2 WT and CaV1.2 G449R
      CaV1.2CaMn/NElectrophysiological properties
      Peak current density (pA/pF)E0.5,act (mV)R100
      CaV1.2 WTCaM123410/2−9 ± 2

      p < 0.001 versus CaM WT
      −6 ± 2

      p < 0.001 versus CaM WT
      0.76 ± 0.02

      p < 0.001 versus CaM WT
      +W-137/1−4 ± 2

      p =0.04 versus control
      −5 ± 20.71 ± 0.02

      p < 0.001 versus control
      CaV1.2 G449RCaM12344/2−17 ± 4

      p < 0.001 versus CaM WT
      −10 ± 2

      p < 0.001 versus CaM WT
      0.73 ± 0.01

      p < 0.001 versus CaM WT
      +W-136/1−19 ± 2−10 ± 20.63 ± 0.01

      p < 0.001 versus control
      Whole-cell currents were recorded from HEKT cells transiently transfected with CaV1.2 WT or variants coexpressed with CaVβ2a, CaVα2δ1, and CaM1234. Activation properties (E0.5,act) were estimated from the I–V relationships and fitted to a BoltzIV equation as described in the Experimental procedures section. The R100 values report the relative current decay observed 100 ms after the peak current. n/N refers to the number of cells/transfections measured in each condition of study. Mean ± SD are shown. Statistical analysis was carried out against the values measured for CaM WT. Herein “control” refers to the data collected in the presence of CaM1234 in the absence of W-13.

      The gain of function in CaV1.2 G449R requires the Ca2+-bound CaM form

      Unlike CaV1.2 WT, coexpressing either phosphoresistant CaM T79A (Fig. 7A, left) and S81A (Fig. 7C, left) or phosphomimetic CaM T79D (Fig. 7B, left) and S81D (Fig. 7D, left) with Cav1.2 G449R did not appreciably affect the peak current density, activation gating kinetics (E0.5,act), and current decay (R100) of CaV1.2 G449R (Fig. 7, E and F and Table 3). As observed in the presence of CaM WT, the peak current densities (Fig. 7, AD, middle, right; Fig. 7E), the E0.5,act (Fig. 7F) were not altered by the application of W-13. This sharply contrasts with the results obtained with the CaV1.2 WT channel complex. Nonetheless, preventing the phosphorylation of all CaM molecules with TBB reduced by 50% the peak current density measured under all other conditions (Table 4) save for CaM1234 (Table 5). Indeed, limiting Ca2+ binding to CaM with the CaM1234 variant not only impaired the CDI of CaV1.2 G449R but also prevented the leftward shift in activation gating and the increase in peak current density (Table 5).
      Figure thumbnail gr7
      Figure 7CaV1.2 G449R is not modulated by CaM or CaM inhibitor W-13. Representative CaV1.2 G449R current traces were recorded from HEKT cells in the presence of 2 mM Ca2+. AD, left, CaV1.2 G449R was coexpressed with CaM WT, with the phosphoresistant CaM (T79A or S81A) or with phosphomimetic CaM (T79D or S81D) as shown. AD, middle, CaV1.2 G449R channels coexpressed with either CaM WT, T79A, T79D, S81A, or S81D are resistant to block by W-13. The vertical scale bars are 10 pA/pF, and the horizontal scale bars are 100 ms throughout. AD, right, average I–V curves of CaV1.2 G449R coexpressed with CaM T79A, T79D, S81A, or S81D. The peak current densities were not different between control and W-13 treatment. E and F, the distribution of the peak current densities and E0.5,act for control conditions and after W-13 treatment are summarized individually as filled circles for Cav1.2 G449R coexpressed with either CaM WT (black), T79A (red), T79D (blue), S81A (green), or S81D (light purple). The mean data ± SD are shown as gray hyphens. The complete set of values is found in . CaM, calmodulin; HEKT, human embryonic kidney 293T cell line.

      Alanine substitutions in the hinge region of CaM are not disrupting interaction with CaV1.2

      We next evaluated whether CaM variants T79A and S81A alter the interaction of CaM with the pore-forming CaVα1C subunit (Fig. 8). Whether for CaV1.2 WT or CaV1.2 G449R, the pull-down assays failed to reveal a correlation between the signal intensity and any of the tested CaM-substituted proteins indicating that phosphomimetic analogs of Ca2+-bound CaM impact channel function rather than protein interaction. Nonetheless, coimmunoprecipitation assays performed over a 1-year period consistently revealed a stronger signal for G449R proteins than for CaV1.2 WT proteins suggesting that the glycine to arginine substitution at position 449 could increase the affinity of CaM for CaV1.2.
      Figure thumbnail gr8
      Figure 8CaM T79A and T79D coimmunoprecipitate CaV1.2 WT and G449R. HEKT cells were transiently transfected with CaVβ2a in the presence of CaV1.2 WT or CaV1.2 G449R and either CaM WT, CaM T79A, or CaM T79D. A, total proteins are shown. B, coimmunoprecipitation was carried out with anti-His magnetic beads. Immunoblotting was carried out after elution of the bound proteins using the antibodies described in the legend of . As seen, CaV1.2 WT and G449R, CaVβ2a, and CaM proteins were translated at the expected molecular masses of 250, 70, and 18 to 24 kDa, respectively. There was no significant difference between the signals measured in the presence of either CaM WT, CaM T79A, or CaM T79D. The signals were nonetheless systematically stronger for CaV1.2 G449R than for CaV1.2 WT despite equivalent loading and similar signals for the total proteins. Similar data were obtained from three independent transfections carried out over the course of 2 months with protein extraction carried out with digitonin or CHAPS. CaM, calmodulin; HEKT, human embryonic kidney 293T cell line.

      Substitution with an alpha-helix breaker in CaV1.2 antagonizes channel function

      The structural properties of the I–II linker near the high-affinity binding site for CaVβ have been consistently shown to modulate the gating properties of CaV1 and CaV2 channels (
      • Almagor L.
      • Chomsky-Hecht O.
      • Ben-Mocha A.
      • Hendin-Barak D.
      • Dascal N.
      • Hirsch J.A.
      The role of a voltage-dependent Ca2+ channel intracellular linker: A structure-function analysis.
      ). In CaV1.2, most, if not all substitutions, tested at position 449 altered the channel properties. Stronger activation gating and faster inactivation kinetics characterized CaV1.2 G449A, G449D, and G449K in the presence of endogenous CaM (Fig. S1 and Table 6). All these substituted channels activated at more hyperpolarized voltages than CaV1.2 WT with a threshold at −40 mV and currents peaking between 0 and −5 mV. These data could suggest that α-helix-enhancing residues and/or positively charged residues increase the channel affinity for CaM. Hence, the CaM–channel complex would be very stable in the presence of endogenous CaM as to avert the impact of CaM mutants. To test the role of the secondary structure, position 449 in CaV1.2 was substituted with proline, recognized as α-helix breaker (
      • Li S.C.
      • Goto N.K.
      • Williams K.A.
      • Deber C.M.
      Alpha-helical, but not beta-sheet, propensity of proline is determined by peptide environment.
      ). G449P produced whole-cell peak currents (−2.5 ± 0.8 pA/pF, n = 10, N = 2, p < 0.001 versus CaV1.2 WT) that were five times smaller than CaV1.2 WT but significantly different than voltage-activated inward Ca2+ currents measured in nontransfected cells. The activation gating of G449P was right shifted when compared with CaV1.2 WT. In contrast to G449R and G449K, CaV1.2 G449P was modulated by CaM phosphomimetic variants (Fig. 9 and Table 6). Peak currents of G449P nearly tripled in the presence of CaM WT, CaM T79D, or CaM S81D and were not significantly altered by coexpressing CaM T79A or CaM S81A (Fig. 9, A and B). Remarkably, the activation of the G449P channel was left shifted in the presence of the phosphor-silenced CaM variants (Fig. 9D), the only occurrence where the larger peak currents were not associated with stronger activation gating. Altogether, these observations support a strong mechanistic link between the structural properties of the I–II linker near the binding site for CaVβ and the modulation of the channel activation gating by CaM. In particular, the channel propensity to adopt a longer α-helix in this region appears to improve the activation gating of the channel and to supersede the modulation by the phosphorylated forms of CaM.
      Table 6Electrophysiological properties of CaV1.2 Gly449 variants with CaM phosphorylation surrogates
      CaV1.2CaMn/NElectrophysiological properties
      Peak current density (pA/pF)E0.5,act (mV)R100
      CaV1.2 G449ANative12/2−35 ± 8−18 ± 30.55 ± 0.03
      CaM WT22/4−31 ± 8−15 ± 20.57 ± 0.03
      CaM T79A7/1−11 ± 2

      p < 0.001 versus native CaM

      p < 0.001 versus CaM WT
      −13 ± 2

      p = 0.002 versus native CaM
      0.70 ± 0.01

      p < 0.001 versus native CaM

      p < 0.001 versus CaM WT
      CaM T79D6/1−28 ± 8−18 ± 20.55 ± 0.02
      CaM S81A13/2−14 ± 3

      p < 0.001 versus native CaM

      p < 0.001 versus CaM WT
      −13 ± 2

      p < 0.001 versus native CaM
      0.70 ± 0.03

      p < 0.001 versus native CaM

      p < 0.001 versus CaM WT
      CaM S81D19/2−30 ± 9−14 ± 3

      p = 0.007 versus native CaM
      0.58 ± 0.03
      CaV1.2 G449DNative17/2−14 ± 4−15 ± 20.62 ± 0.03
      CaM WT17/2−27 ± 8

      p < 0.001 versus native CaM
      −17 ± 20.56 ± 0.02

      p < 0.001 versus native CaM
      CaM T79A16/1−15 ± 4

      p < 0.001 versus CaM WT
      −16 ± 30.59 ± 0.02

      p = 0.04 versus native CaM

      p = 0.002 versus CaM WT
      CaM T79D14/1−29 ± 6

      p < 0.001 versus native CaM
      −18 ± 2

      p = 0.007 versus native CaM
      0.55 ± 0.02

      p < 0.001 versus native CaM
      CaM S81A21/2−16 ± 4

      p < 0.001 versus CaM WT
      −15 ± 20.60 ± 0.02

      p < 0.001 versus CaM WT
      CaM S81D10/1−33 ± 7

      p < 0.001 versus native CaM
      −18 ± 2

      p = 0.03 versus native CaM
      0.52 ± 0.02

      p < 0.001 versus native CaM

      p = 0.01 versus CaM WT
      CaV1.2 G449PNative10/2−2.5 ± 0.8−4 ± 10.72 ± 0.02
      CaM WT8/2−6 ± 1

      p < 0.001 versus native CaM
      1.0 ± 2

      p < 0.001 versus native CaM
      0.65 ± 0.02

      p < 0.001 versus native CaM
      CaM T79A3/1−1.6 ± 0.3

      p = 0.001 versus CaM WT
      −9 ±1

      p = 0.04 versus native CaM

      p = 0.002 versus CaM WT
      0.76 ± 0.01

      p < 0.001 versus CaM WT
      CaM T79D4/1−11 ± 3

      p < 0.001 versus native CaM, CaM WT
      −0.5 ± 1.5

      p =0.01 versus native CaM
      0.57 ± 0.01

      p < 0.001 versus native CaM

      p < 0.001 versus CaM WT
      CaM S81A4/1−1.8 ± 0.5

      p < 0.001 versus CaM WT
      −10 ± 2

      p < 0.001 versus native CaM

      p < 0.001 versus CaM WT
      0.73 ± 0.02

      p < 0.001 versus CaM WT
      CaM S81D3/1−9.7 ± 0.7

      p < 0.001 versus native CaM

      p = 0.01 versus CaM WT
      3.5 ± 0.7

      p < 0.001 versus native CaM

      p = 0.01 versus CaM WT
      0.69 ± 0.03
      CaV1.2 G449KNative18/2−23 ± 5−15 ± 30.60 ± 0.02
      CaM WT26/6−26 ± 6−15 ± 30.61 ± 0.03
      CaM T79A8/1−28 ± 7−17 ± 20.57 ± 0.02

      p = 0.006 versus CaM WT
      CaM T79D12/1−30 ± 6

      p = 0.02 versus native CaM

      p = 0.002 versus CaM WT
      −17 ± 20.54 ± 0.02

      p < 0.001 versus native CaM

      p < 0.001 versus CaM WT
      CaM S81A15/2−27 ± 7−16 ± 30.56 ± 0.03

      p < 0.001 versus native CaM

      p < 0.001 versus CaM WT
      CaM S81D15/2−29 ± 5

      p = 0.01 versus CaM WT
      −17 ± 30.56 ± 0.03

      p < 0.001 versus native CaM

      p < 0.001 versus CaM WT
      CaV1.2 Gly449 variants were coexpressed with CaVβ2a, CaVα2δ1, and CaM WT or CaM T79A, CaM T79D, CaM S81A, or CaM S81D. Activation properties (E0.5,act) were estimated from the I–V relationships as described in the Experimental procedures section. n/N refers to the number of cells/transfections measured in each condition of study. Mean ± SD are shown. Statistical analysis was carried out against CaM WT or against endogenous/native CaM. As seen, CaV1.2 WT, G449D, and G449P were modulated by CaM phosphovariants, whereas the properties of G449K remained unaffected. G449A was not upregulated by CaM WT but was downregulated by phosphoresistant CaM variants.
      Figure thumbnail gr9
      Figure 9CaV1.2 G449P is modulated by CaM phosphorylation surrogates. A, whole-cell currents were recorded from HEKT cells transiently transfected with CaV1.2 G449P coexpressed with Cavβ2a and Cavα2δ1 and either CaM WT, with the phosphoresistant CaM (T79A or S81A), or with phosphomimetic CaM (T79D or S81D) as indicated. Exemplar traces are shown (from left to right) for CaV1.2 G449P + CaM WT, G449P + CaM T79A, G449P + CaM T79D, G449P + CaM S81A, and G449P + CaM S81D. The vertical scale bars are 10 pA/pF, and the horizontal scale bars are 100 ms throughout. B, the corresponding peak current densities are plotted as a function of applied voltage. C and D, the summarized distribution of the peak current densities and the midpotential of activation E0.5,act. Peak whole-cell currents and E0.5,act are reported individually as black circles. The mean data ± SD are shown as red hyphens. Values of peak current densities and E0.5,act are reported in . CaM, calmodulin; HEKT, human embryonic kidney 293T cell line.

      Discussion

      Ca2+–CaM modulates the activity of L-type CaV1.2 through multifaceted mechanisms

      The ubiquitous multifunctional Ca2+-binding protein CaM is a two-lobe protein with each of two hydrophilic pockets for Ca2+ sensing separated by a flexible central linker. It is regulating the function of many voltage-gated ion channels, such as Kv7.2 (
      • Ambrosino P.
      • Alaimo A.
      • Bartollino S.
      • Manocchio L.
      • De Maria M.
      • Mosca I.
      • et al.
      Epilepsy-causing mutations in Kv7.2 C-terminus affect binding and functional modulation by calmodulin.
      ), NaV1.4 (
      • Adams P.J.
      • Ben-Johny M.
      • Dick I.E.
      • Inoue T.
      • Yue D.T.
      Apocalmodulin itself promotes ion channel opening and Ca(2+) regulation.
      ), and in particular, voltage-gated CaV channels (
      • Adams P.J.
      • Ben-Johny M.
      • Dick I.E.
      • Inoue T.
      • Yue D.T.
      Apocalmodulin itself promotes ion channel opening and Ca(2+) regulation.
      ,
      • Ben-Johny M.
      • Yue D.T.
      Calmodulin regulation (calmodulation) of voltage-gated calcium channels.
      ) (for review, see Ref. (
      • Ben-Johny M.
      • Yue D.T.
      Calmodulin regulation (calmodulation) of voltage-gated calcium channels.
      )). At least two CaM molecules can simultaneously bind to the C-terminal region of CaV1.2 (
      • Kim E.Y.
      • Rumpf C.H.
      • Fujiwara Y.
      • Cooley E.S.
      • Van Petegem F.
      • Minor J.
      Structures of CaV2 Ca2+/CaM-IQ domain complexes reveal binding modes that underlie calcium-dependent inactivation and facilitation.
      ,
      • Van Petegem F.
      • Chatelain F.C.
      • Minor Jr., D.L.
      Insights into voltage-gated calcium channel regulation from the structure of the CaV1.2 IQ domain-Ca2+/calmodulin complex.
      ), but additional binding sites in the N-terminal region and the first intracellular linker of CaVα1C have been identified (
      • Asmara H.
      • Minobe E.
      • Saud Z.A.
      • Kameyama M.
      Interactions of calmodulin with the multiple binding sites of Cav1.2 Ca2+ channels.
      ,
      • Kim E.Y.
      • Rumpf C.H.
      • Fujiwara Y.
      • Cooley E.S.
      • Van Petegem F.
      • Minor J.
      Structures of CaV2 Ca2+/CaM-IQ domain complexes reveal binding modes that underlie calcium-dependent inactivation and facilitation.
      ,
      • Dick I.E.
      • Tadross M.R.
      • Liang H.
      • Tay L.H.
      • Yang W.
      • Yue D.T.
      A modular switch for spatial Ca2+ selectivity in the calmodulin regulation of CaV channels.
      ,
      • Simms B.A.
      • Souza I.A.
      • Zamponi G.W.
      A novel calmodulin site in the Cav1.2 N-terminus regulates calcium-dependent inactivation.
      ,
      • Ivanina T.
      • Blumenstein Y.
      • Shistik E.
      • Barzilai R.
      • Dascal N.
      Modulation of L-type Ca2+ channels by gbeta gamma and calmodulin via interactions with N and C termini of alpha 1C.
      ). The overall structural organization of CaM within the CaV1.2 channel complex remains to be established. CaM-binding sites were not resolved in the cryo-electron microscopy structure of the homologous CaV1.1 channel (
      • Wu J.
      • Yan Z.
      • Li Z.
      • Qian X.
      • Lu S.
      • Dong M.
      • et al.
      Structure of the voltage-gated calcium channel Cav1.1 at 3.6 Å resolution.
      ).
      In CaV1.2 channels, Ca2+ binding to CaM contributes to CDI and Ca2+-dependent facilitation (
      • Adams P.J.
      • Ben-Johny M.
      • Dick I.E.
      • Inoue T.
      • Yue D.T.
      Apocalmodulin itself promotes ion channel opening and Ca(2+) regulation.
      ,
      • Zühlke R.D.
      • Pitt G.S.
      • Deisseroth K.
      • Tsien R.W.
      • Reuter H.
      Calmodulin supports both inactivation and facilitation of L-type calcium channels.
      ,
      • Zühlke R.D.
      • Pitt G.S.
      • Tsien R.W.
      • Reuter H.
      Ca2+-sensitive inactivation and facilitation of L-type Ca2+ channels both depend on specific amino acid residues in a consensus calmodulin-binding motif in the(alpha)1C subunit.
      ). Either process requires the binding of incoming Ca2+ ions to CaM preassociated to the isoleucine–glutamine motif in the C-terminal region of the pore-forming CaVα1C subunit (
      • Pate P.
      • Mochca-Morales J.
      • Wu Y.
      • Zhang J.Z.
      • Rodney G.G.
      • Serysheva I.I.
      • et al.
      Determinants for calmodulin binding on voltage-dependent Ca2+ channels.
      ,
      • Romanin C.
      • Gamsjaeger R.
      • Kahr H.
      • Schaufler D.
      • Carlson O.
      • Abernethy D.R.
      • et al.
      Ca(2+) sensors of L-type Ca(2+) channel.
      ,
      • Pitt G.S.
      • Zühlke R.D.
      • Hudmon A.
      • Schulman H.
      • Reuter H.
      • Tsien R.W.
      Molecular basis of calmodulin tethering and Ca2+-dependent inactivation of L-type Ca2+ channels.
      ,
      • Mouton J.
      • Feltz A.
      • Maulet Y.
      Interactions of calmodulin with two peptides derived from the c-terminal cytoplasmic domain of the Ca(v)1.2 Ca2+ channel provide evidence for a molecular switch involved in Ca2+-induced inactivation.
      ). The potentiating form of CaM-dependent facilitation or upregulation is observed in native cardiac L-type channels during trains of depolarization (
      • Noble S.
      • Shimoni Y.
      The calcium and frequency dependence of the slow inward current ‛staircase’ in frog atrium.
      ,
      • Marban E.
      • Tsien R.W.
      Enhancement of calcium current during digitalis inotrophy in mammalian heart: positive feed-back regulation by intracellular calcium?.
      ) but usually not reported in recombinant systems with the intact CaV1.2 WT channel (
      • Zühlke R.D.
      • Pitt G.S.
      • Deisseroth K.
      • Tsien R.W.
      • Reuter H.
      Calmodulin supports both inactivation and facilitation of L-type calcium channels.
      ,
      • Zühlke R.D.
      • Pitt G.S.
      • Tsien R.W.
      • Reuter H.
      Ca2+-sensitive inactivation and facilitation of L-type Ca2+ channels both depend on specific amino acid residues in a consensus calmodulin-binding motif in the(alpha)1C subunit.
      ,
      • Findeisen F.
      • Minor Jr., D.L.
      Disruption of the IS6-AID linker affects voltage-gated calcium channel inactivation and facilitation.
      ). We herein report that phosphomimetic analogs of CaM stimulate Ca2+ influx and promotes the activation gating of CaV1.2. CaM promotes the cell surface trafficking of CaV1.2 and stimulates function through an increase in peak current density and a leftward shift in the activation gating. In our hands, the latter actions of CaM require Ca2+ as it was impaired in the presence of the constitutively Ca2+-free form CaM1234 where the four Ca2+-binding sites are invalidated. This observation is compatible with data from Kim et al. (
      • Kim J.
      • Ghosh S.
      • Nunziato D.A.
      • Pitt G.S.
      Identification of the components controlling inactivation of voltage-gated Ca(2+) channels.
      ), who reported that the interaction between the CaM-bound C-terminal peptide and the I–II linker is disrupted in the complete absence of Ca2+. CaM1234 prevented the increase in peak current density, failed to promote channel activation gating, and as expected, slowed down the CDI kinetics by 30%. Nonetheless, Ca2+ binding is not sufficient to account for the wide-ranging impact of CaM on channel function. The structural properties of the flexible linker region of CaM contribute to the channel response to CaM. Coexpression with CaM T79A or CaM S81A averted the boost in peak current density (although it did not alter the activation gating). In contrast, coexpression with either CaM WT or phosphomimetic CaM T79D or CaM S81A yielded similar results suggesting that phosphorylation of either site participates to the modulation of CaV1.2 by CaM. Indeed, preventing the phosphorylation of native and overexpressed CaM by incubating the cells with TBB, a membrane-permeable inhibitor of CK2, nearly abrogated channel function. Hence, Ca2+-bound CaM modulates the function of the CaV1.2 channel complex in a fashion reminiscent of the ancillary subunits CaVβ and CaVα2δ, which like CaM may also modulate other ion channels (
      • Campiglio M.
      • Flucher B.E.
      The role of auxiliary subunits for the functional diversity of voltage-gated calcium channels.
      ).

      Multiple mechanisms converge toward CaV1.2 G449R

      The missense variant, glycine to arginine, was identified from a patient with prolonged QT interval (≈500 ms) and features associated with the TS, but its heterologous expression revealed a novel phenotype where the gain of function resulted from increased peak current density, a negative shift in the activation potential, and no decrease in the channel current decay (
      • Kelu Bisabu K.
      • Zhao J.
      • Mokrane A.E.
      • Segura É.
      • Marsolais M.
      • Grondin S.
      • et al.
      Novel gain-of-function variant in CACNA1C associated with Timothy syndrome, multiple accessory pathways, and noncompaction cardiomyopathy.
      ). The hyperactive mode of the variant expressed in HEK293T (thereafter referred to as HEKT) cells was mimicked by the coexpression of CaV1.2 WT with CaM WT or phosphorylated surrogates CaM T79D or CaM S81D. The functional properties of the clinical glycine to arginine variant remained remarkably insensitive to pharmacological inhibition by W-13 and by overexpression with phosphoresistant CaM analogs (T79A and S81A). The impact of the phosphorylation of CaM appears to be limited to function. CaV1.2 G449R was pulled down equally by CaM WT, T79A, and T79D. Preventing the phosphorylation of CaM with TBB, an inhibitor of CK2, significantly reduced the peak current density of CaV1.2 G449R by ≈50% without a significant alteration in the channel activation voltage as compared with the control conditions. The rate-limiting factor appears to be Ca2+ binding to CaM. Coexpression of G449R with the CaM1234 variant not only impaired the CDI and the increased peak current density but also prevented the leftward shift in activation gating. Overexpression of the CaM1234 variant obliterated the gain in the function of CaV1.2 G449R yielding an activity profile akin to CaV1.2 WT in the presence of endogenous/native CaM. The stronger activity of CaV1.2 G449 thus minimally requires the direct or indirect action of the Ca2+-bound CaM form.
      These observations suggest that the higher channel activity of G449R could result from a stronger affinity for native CaM. Though not measured in this article, the affinity between the two full-length proteins can be roughly approximated by the relative intensity of the signal measured in coimmunoprecipitation assays. Within all the limitations of this exercise, the protein signal obtained for G449R in coimmunoprecipitation assays was indeed systematically stronger than the signal measured for the WT channel complex when measured under the same experimental conditions and this over the course of 12 months. This interpretation is compatible with the cell surface fractionation assays showing that G449R was more likely to be found in the cell surface fraction than the WT channel complex in the presence of endogenous CaM, whereas this differential localization was not discernable when the cells were saturated with overexpressed CaM. CaM bound to the C-terminal region of CaV1.2 has been previously reported to interact in a Ca2+-dependent manner with the cytosolic I–II loop, where is located the glycine to arginine variant (
      • Kim J.
      • Ghosh S.
      • Nunziato D.A.
      • Pitt G.S.
      Identification of the components controlling inactivation of voltage-gated Ca(2+) channels.
      ). It is thus conceivable that the higher “intrinsic” activity of G449R results from a stronger interaction with endogenous CaM. In this model, the cellular availability of CaM could modulate the operating window of CaV1.2.
      CaV1.2 G449R is located in a structural region involved in activation gating (
      • Beyl S.
      • Depil K.
      • Hohaus A.
      • Stary-Weinzinger A.
      • Timin E.
      • Shabbir W.
      • et al.
      Physicochemical properties of pore residues predict activation gating of CaV1.2: a correlation mutation analysis.
      ), inactivation kinetics (
      • Dafi O.
      • Berrou L.
      • Dodier Y.
      • Raybaud A.
      • Sauvé R.
      • Parent L.
      Negatively charged residues in the N-terminal of the AID helix confer slow voltage dependent inactivation gating to CaV1.2.
      ), protein stability, ubiquitination (
      • Hu Z.
      • Li G.
      • Wang J.-W.
      • Chong S.Y.
      • Yu D.
      • Wang X.
      • et al.
      Regulation of blood pressure by targeting CaV1.2-galectin-1 protein interaction.
      ), and cell surface trafficking (
      • Bourdin B.
      • Marger F.
      • Wall-Lacelle S.
      • Schneider T.
      • Klein H.
      • Sauvé R.
      • et al.
      Molecular determinants of the CaVbeta-induced plasma membrane targeting of the CaV1.2 channel.
      ). The proximal segment of the first intracellular linker hosts the high-affinity binding site for CaVβ (
      • Buraei Z.
      • Yang J.
      The β subunit of voltage-gated Ca2+ channels.
      ) and plays a role in networking with direct partners such as galectin (
      • Hu Z.
      • Li G.
      • Wang J.-W.
      • Chong S.Y.
      • Yu D.
      • Wang X.
      • et al.
      Regulation of blood pressure by targeting CaV1.2-galectin-1 protein interaction.
      ) or Ras/Rad proteins through CaVβ (
      • Papa A.
      • Kushner J.
      • Hennessey J.A.
      • Katchman A.N.
      • Zakharov S.I.
      • Chen B.X.
      • et al.
      Adrenergic Ca(V)1.2 activation via Rad phosphorylation converges at α(1C) I-II loop.
      ,
      • Finlin B.S.
      • Crump S.M.
      • Satin J.
      • Andres D.A.
      Regulation of voltage-gated calcium channel activity by the Rem and Rad GTPases.
      ,
      • Liu G.
      • Papa A.
      • Katchman A.N.
      • Zakharov S.I.
      • Roybal D.
      • Hennessey J.A.
      • et al.
      Mechanism of adrenergic CaV1.2 stimulation revealed by proximity proteomics.
      ). Glycine residues are unique in their lack of side-chain steric interference, permitting a higher flexibility to protein structures. Increasing flexibility by inserting glycine residues (
      • Findeisen F.
      • Minor Jr., D.L.
      Disruption of the IS6-AID linker affects voltage-gated calcium channel inactivation and facilitation.
      ,
      • Papa A.
      • Kushner J.
      • Hennessey J.A.
      • Katchman A.N.
      • Zakharov S.I.
      • Chen B.X.
      • et al.
      Adrenergic Ca(V)1.2 activation via Rad phosphorylation converges at α(1C) I-II loop.
      ) decreases channel function. In contrast, decreasing flexibility of this region by removing glycine residues promoted channel function (
      • Kelu Bisabu K.
      • Zhao J.
      • Mokrane A.E.
      • Segura É.
      • Marsolais M.
      • Grondin S.
      • et al.
      Novel gain-of-function variant in CACNA1C associated with Timothy syndrome, multiple accessory pathways, and noncompaction cardiomyopathy.
      ,
      • Almagor L.
      • Chomsky-Hecht O.
      • Ben-Mocha A.
      • Hendin-Barak D.
      • Dascal N.
      • Hirsch J.A.
      The role of a voltage-dependent Ca2+ channel intracellular linker: A structure-function analysis.
      ,
      • Almagor L.
      • Avinery R.
      • Hirsch J.A.
      • Beck R.
      Structural flexibility of CaV1.2 and CaV2.2 I-II proximal linker fragments in solution.
      ,
      • Almagor L.
      • Chomsky-Hecht O.
      • Ben-Mocha A.
      • Hendin-Barak D.
      • Dascal N.
      • Hirsch J.A.
      CaV1.2 I-II linker structure and Timothy syndrome.
      ). The presence of a glycine residue proximal to the α-interacting domain in CaV1.2 WT could thus explain the requirement of a stronger depolarization in CaV1.2 WT versus G449R channels. The same position is already occupied by an arginine residue in CaV2.2 (
      • Almagor L.
      • Chomsky-Hecht O.
      • Ben-Mocha A.
      • Hendin-Barak D.
      • Dascal N.
      • Hirsch J.A.
      The role of a voltage-dependent Ca2+ channel intracellular linker: A structure-function analysis.
      ) and CaV2.3 channels whose activation is left shifted when compared with CaV1.2 under the same expression conditions (
      • Shakeri B.
      • Bourdin B.
      • Demers-Giroux P.O.
      • Sauvé R.
      • Parent L.
      A quartet of leucine residues in the guanylate kinase domain of CaVβ determines the plasma membrane density of the CaV2.3 channel.
      ).
      The high-affinity binding site of CaVβ adopts an α-helical structure in vitro (
      • Opatowsky Y.
      • Chen C.C.
      • Campbell K.P.
      • Hirsch J.A.
      Structural analysis of the voltage-dependent calcium channel beta subunit functional core and its complex with the alpha 1 interaction domain.
      ). The relative rigidity α-helix could promote a strong van der Waals interaction between the guanylate domain of CaVβ and hydrophobic residues of CaV1.2 (
      • Findeisen F.
      • Minor Jr., D.L.
      Disruption of the IS6-AID linker affects voltage-gated calcium channel inactivation and facilitation.
      ,
      • Van Petegem F.
      • Clark K.A.
      • Chatelain F.C.
      • Minor Jr., D.L.
      Structure of a complex between a voltage-gated calcium channel beta-subunit and an alpha-subunit domain.
      ,
      • Chen Y.H.
      • Li M.H.
      • Zhang Y.
      • He L.L.
      • Yamada Y.
      • Fitzmaurice A.
      • et al.
      Structural basis of the alpha1-beta subunit interaction of voltage-gated Ca2+ channels.
      ). In the native protein, this α-helix breaks at the glycine located at position 449 (
      • Almagor L.
      • Avinery R.
      • Hirsch J.A.
      • Beck R.
      Structural flexibility of CaV1.2 and CaV2.2 I-II proximal linker fragments in solution.
      ). Crystallographic and circular dichroism spectroscopic studies demonstrated that the arginine substitution prolongs the α-helix (
      • Almagor L.
      • Chomsky-Hecht O.
      • Ben-Mocha A.
      • Hendin-Barak D.
      • Dascal N.
      • Hirsch J.A.
      The role of a voltage-dependent Ca2+ channel intracellular linker: A structure-function analysis.
      ). We also report that substitution with other α-helix-promoting residues, such as alanine (
      • Pace N.C.
      • Scholtz M.J.
      A helix propensity scale based on experimental studies of peptides and proteins.
      ), produced channels with strong activation properties, and from the contrary, substitution with proline, regarded as a α-helix breaker, was found to curb channel activation. The substituted channels however manifested distinct electrophysiological signatures in the presence of the phosphomimetic and phosphoresistant CaM proteins, from a complete indifference (G449K) to impaired peak current density in the presence of phosphoresistant CaM variants (G449A, G449D, G449P, and G449Q). Our data are compatible with the proposition that the longer α-helix enhances the coupling of the I–II linker with the inner pore responsible for channel activation. The intracellular linker would contribute to electromechanical coupling in CaV1.2 either through its intrinsic structural properties or following interaction with CaM.
      The structural properties of the clinical variant could be envisioned to facilitate the interplay between accessory CaM proteins bound onto the C terminus of CaV1.2 and channel function as it was postulated for AKAP150 (
      • Dixon R.E.
      • Cheng E.P.
      • Mercado J.L.
      • Santana L.F.
      L-type Ca2+ channel function during Timothy syndrome.
      ). In this context, the LQTS phenotype associated with the glycine to arginine substitution in the I–II linker could result from either process: an intrinsically stronger activation of CaV1.2 that renders the channel insensitive to cellular variations in phosphorylated CaM or else a higher affinity to CaM that causes the channel to be maximally activated at near endogenous concentration of CaM.

      Experimental procedures

      Recombinant DNA techniques

      The CaVα1C subunit of CaV1.2 (GenBank accession number: X15539), CaVβ2a (GenBank accession number: NM_001398773), and CaVα2δ1 (GenBank accession number: NM_000722) was subcloned in commercial vectors under the control of the cytomegalovirus (CMV) promoter as described elsewhere (
      • Bourdin B.
      • Briot J.
      • Tétreault M.P.
      • Sauvé R.
      • Parent L.
      Negatively charged residues in the first extracellular loop of the L-type Ca(V)1.2 channel anchor the interaction with the Ca(V)α2δ1 auxiliary subunit.
      ,
      • Briot J.
      • Mailhot O.
      • Bourdin B.
      • Tétreault M.P.
      • Najmanovich R.
      • Parent L.
      A three-way inter-molecular network accounts for the Ca(V)α2δ1-induced functional modulation of the pore-forming Ca(V)1.2 subunit.
      ,
      • Bourdin B.
      • Marger F.
      • Wall-Lacelle S.
      • Schneider T.
      • Klein H.
      • Sauvé R.
      • et al.
      Molecular determinants of the CaVbeta-induced plasma membrane targeting of the CaV1.2 channel.
      ,
      • Bourdin B.
      • Shakeri B.
      • Tétreault M.P.
      • Sauvé R.
      • Lesage S.
      • Parent L.
      Functional characterization of CaVα2β mutations associated with sudden cardiac death.
      ). The cDNA sequence of the rabbit clone is near identical to the human clone save for an additional 30 amino acids in its N terminus, accounting for the +30 residue shift in residue numbering. The human CaM (GenBank accession number: M27319), subcloned in pcDNA3.1 (Thermo Fisher Scientific) vector with consecutive histidine (His-His-His-His-His-His) and cMyc (Glu-Gln-Lys-Leu-Iso-Ser-Glu-Glu-Asp-Leu) tags in C-terminal region, was a gift from Dr Rémy Sauvé, Université de Montréal. The cDNA mutations of CaM were introduced in this vector. CaM is numbered as reported (
      • Lee C.H.
      • MacKinnon R.
      Activation mechanism of a human SK-calmodulin channel complex elucidated by cryo-EM structures.
      ) to take into account that the mature protein lacks N-terminal Met residue. All cDNA mutations in CaVα1C of CaV1.2 and CaM were produced with the Q5 Site-Directed Mutagenesis Kit (New England Biolabs, Inc) according to the manufacturer’s instructions. Briefly, substitutions of nucleotides were created by incorporating the desired mutation in the center of the forward primer, and the reverse primer is designed so that the 5′ ends of the two primers anneal back to back. Following the PCR, the amplified DNA is circularized, and the template is removed with a kinase–ligase–DpnI enzyme mixture, before transformation into high-efficiency NEB DH5-α competent Escherichia coli. All constructs were verified by automated double-stranded sequence analysis (“Centre d’expertise et de services Génome Québec”). The protein expression at the expected molecular weight was confirmed by standard Western blot analysis for each construct.

      Gene transfection and cell culture

      HEKT cells were grown using standard tissue culture conditions (5% CO2, 37 °C) in high-glucose Dulbecco's modified Eagle's medium supplemented with fetal bovine serum (10%), l-glutamine (2 mM), penicillin (100 U/ml), and streptomycin (10 mg/ml) as described before (
      • Bourdin B.
      • Briot J.
      • Tétreault M.P.
      • Sauvé R.
      • Parent L.
      Negatively charged residues in the first extracellular loop of the L-type Ca(V)1.2 channel anchor the interaction with the Ca(V)α2δ1 auxiliary subunit.
      ,
      • Briot J.
      • Mailhot O.
      • Bourdin B.
      • Tétreault M.P.
      • Najmanovich R.
      • Parent L.
      A three-way inter-molecular network accounts for the Ca(V)α2δ1-induced functional modulation of the pore-forming Ca(V)1.2 subunit.
      ,
      • Bourdin B.
      • Marger F.
      • Wall-Lacelle S.
      • Schneider T.
      • Klein H.
      • Sauvé R.
      • et al.
      Molecular determinants of the CaVbeta-induced plasma membrane targeting of the CaV1.2 channel.
      ). Using Lipofectamine 2000 (Invitrogen), as per the manufacturer's instructions, HEKT cells (80% confluence, 35 mm petri dish) were transiently transfected with cDNA plasmids, namely pCMV-CaV1.2 WT or variants (4 μg), pCMV-Cavβ2a (4 μg), pCMV-CaVα2δ1 (4 μg), and in some experiments, pcDNA3-HisB-cMyc-CaM WT or variants (2 μg), with a weight ratio of 1:1:1:0.5 for a total of 12 to 14 μg cDNAs. The molar ratio was 7:1 for CaM and CaV1.2. Unless otherwise noted, the plasmids pCMV-CaVβ2a, pCMV-CaVα2δ1, and pcDNA3-HisB-cMyc-CaM WT are simply referred to as CaVβ2a, CaVα2δ1, and CaM WT in the text and figures. cDNA coding for peGFP (0.2 μg) was included in the cDNA mixture as a marker of successful transfection for patch-clamp experiments (
      • Segura E.
      • Bourdin B.
      • Tétreault M.P.
      • Briot J.
      • Allen B.G.
      • Mayer G.
      • et al.
      Proteolytic cleavage of the hydrophobic domain in the Ca(V)α2δ1 subunit improves assembly and activity of cardiac Ca(V)1.2 channels.
      ,
      • Shakeri B.
      • Bourdin B.
      • Demers-Giroux P.O.
      • Sauvé R.
      • Parent L.
      A quartet of leucine residues in the guanylate kinase domain of CaVβ determines the plasma membrane density of the CaV2.3 channel.
      ). The culture medium was changed, and cells were detached with 0.05% trypsin before being replated on 35 mm petri dishes 6 h post-transfection. Whole-cell patch clamp experiments were performed 24 to 32 h after transfection.

      Coimmunoprecipitation

      HEKT cells were transiently transfected with the appropriate constructs (as indicated later), and protein extraction proceeded 2 days after transfection. Experiments described in Figure 2 were carried out as follows. HEKT cells were transiently transfected with CaV1.2 WT or CaV1.2 G449R with pCMV-CaVα2δ1 and cMyc-tagged versions of CaVβ3 or CaVβ2a using, respectively, the pCMV-Tag5-CaVβ3 or the pCMV-Tag5-CaVβ2a plasmids. CaVβ acted as the bait. Cell lysates were immunoprecipitated overnight with anti-cMyc magnetic beads (Pierce Anti-c-Myc Magnetic Beads; catalog no.: 88842, Thermo Fisher Scientific) to capture the given CaVβ. In the experiments shown in Figures 4 and 5, the constructs were pCMV-CaVβ2a with pCMV-CaV1.2 WT or G449R and pcDNA3-HisB-cMyc-CaM WT and used CaM as the bait. Cell lysates were immunoprecipitated overnight with anti-His magnetic beads (code no.: MBL-D29111). The procedure was otherwise similar for the three experimental groups. Two different detergents have been used to compare extraction efficiency between digitonin (a nonionic saponin detergent) and CHAPS–Na (zwitterionic detergent). Both extraction conditions have produced the same results and were thus combined, for three independent experiments over the course of 2 months. Two days after transfection, cells were homogenized in 20 mM Na–Mops (pH 7.4), 300 mM NaCl, and 1% digitonin or 0.5% CHAPS–Na, supplemented with protease inhibitors without EDTA (Thermo Fisher Scientific). Homogenates were sonicated, incubated for 1 h at 4 °C, and centrifuged at 13,000 rpm for 30 min. A fraction (20 μg) of the homogenates or starting material was set aside as representative of total proteins and was immunoblotted to confirm normal protein expression. Coimmunoprecipitation was carried out using 200 μg homogenates diluted in 150 μl of 20 mM Na–Mops (pH 7.4) and 300 mM NaCl. The 200 ± 20 μl protein solution was incubated overnight with the appropriate antibody-coated magnetic beads that were collected using a PureProteome magnetic rack (Millipore). The magnetic beads were washed three times with a buffer containing 20 mM Na–Mops (pH 7.4), 300 mM NaCl, and 0.2% digitonin or alternatively 20 mM Na–Mops (pH 7.4), 300 mM NaCl, without additional detergent for the extraction under the “CHAPS conditions.” The bound proteins were eluted with Laemmli buffer (20 μl) at 95 °C for 5 min, electrophoresed on a 6% or 10% SDS-polyacrylamide gel, and transferred onto a nitrocellulose membrane for Western blotting. Antibodies are described in the figure legends. Signals were detected with the enhanced chemiluminescence substrate. Blots were visualized with the ChemiDoc Touch system (Bio-Rad). Molecular weights were estimated using Image Lab software, version 5.2 (Bio-Rad) by linear regression of standard molecular weight markers.

      Cell surface fractionation assay

      Four different protein fractions (total cell lysates, cytosolic, total membrane, and plasma membrane fractions) were prepared as explained before (
      • Segura E.
      • Bourdin B.
      • Tétreault M.P.
      • Briot J.
      • Allen B.G.
      • Mayer G.
      • et al.
      Proteolytic cleavage of the hydrophobic domain in the Ca(V)α2δ1 subunit improves assembly and activity of cardiac Ca(V)1.2 channels.
      ). Briefly, transfected HEKT cells cultured in 100 mm dishes were homogenized at 4 °C in a Tris-based solution containing a mixture of protease inhibitors (Sigma) at pH 7.4. The cell homogenate was aliquoted into three tubes. After a 2 h incubation period at 4 °C with 1% (v/v) Triton X-100, the first tube was centrifuged at 10,000g for 10 min to remove cell debris, nuclei, and mitochondria. The supernatant was kept as the total protein fraction (whole-cell lysates). The second tube was centrifuged at 200,000g and 4 °C for 20 min. The supernatant is referred to as the cytosolic fraction. The pellet was resuspended in homogenizing buffer containing 1% (v/v) Triton X-100. After 30 min of incubation on ice, a second centrifugation was performed at 200,000g. The resulting supernatant is referred to as the total membrane protein fraction. The third tube was centrifuged at 10,000g for 10 min. The supernatant obtained was centrifuged at 200,000g and 4 °C for 20 min. The pellet was resuspended in the homogenizing buffer containing 0.6 M KCl. Subsequent centrifugations were performed at 200,000g and 4 °C for 20 min to wash out the KCl. The final pellet was resuspended in the homogenizing buffer and is considered to be enriched in plasma membrane proteins. Proteins (20 μg) were electrophoresed on a 10% SDS-polyacrylamide gel.

      Whole-cell patch-clamp recordings and data analysis

      Whole-cell Ca2+ currents from transfected HEKT cells were recorded using pCLAMP software 11.2 (Molecular Devices) and an Axopatch 200B amplifier (Molecular Devices). Patch electrodes were pulled from borosilicate glass (Corning; code: 8161) and heat-polished to a final resistance about 3.0 to 3.5 MΩ when filled with the intracellular solution. Whole-cell currents were low-pass filtered at 2 kHz, digitized at a sampling rate of 100 μs during acquisition, and stored on a microcomputer equipped with an AD converter (Axon Digidata 1440A; Molecular Devices). Electrodes were filled with a solution containing (in millimolar) 140 CsCl, 0.6 NaGTP, 3 MgATP, 10 EGTA, 10 Hepes, titrated to pH 7.4 with NaOH. HEKT cells were bathed in a modified Earle’s saline solution (in millimolar) as follows: 135 NaCl, 20 tetraethylammonium chloride, 2 CaCl2, 1 MgCl2, 10 Hepes, titrated to pH 7.4 with potassium hydroxide. Stock solution of the cell-permeable CaM antagonists W-13 N-(4-aminobutyl)-5-chloro-2-naphthalenesulfonamide and monohydrochloride (Tocris, Bio-Techne) was prepared in distilled water, diluted to its final concentration just before use, and added directly in the bath solution. Cells were incubated for 15 min prior to whole-cell recordings. A few experiments were performed in the presence of 2.5 μM TBB, a cell-permeable inhibitor of CK2. Stock solution of TBB (5 mM) was prepared in dimethylsulfoxide, diluted to its final concentration just before use, and added directly in the bath solution. Whole-cell currents were recorded 15 min after drug equilibration. All experiments were carried out at room temperature (23–25 °C). Cellular capacitance was estimated by measuring the time constant of current decay evoked by a depolarizing pulse pf 10 mV applied to the cell from a holding potential of −100 mV.
      Whole-cell Ca2+ currents were elicited from a holding potential of −100 mV and depolarized to potentials ranging from −80 to 65 mV in 5 mV increments lasting 450 ms for each step. Ca2+ current densities (pA/pF) were obtained by dividing the peak current by the cell capacitance. Average I–V curves were obtained by plotting the peak current densities versus the voltage and fitted to a BoltzIV equation, which is a transformed Boltzmann function for I–V data of the following form:
      I=(VmVrev).Gmax1+e(VE0.5.act)/dx


      where I is the current, Vm is the applied voltage, E0.5,act is the voltage at which channels are half-maximally activated, dx is the steepness of the slope, Gmax is the maximal conductance, and Vrev is the reversal potential. Steady-state activation curves were constructed by dividing the peak I–V data by the driving force. The R100 ratio of CaV1.2 current was defined as the peak current remaining after a 100 ms depolarizing pulse (I100ms/Ipeak) and was used as an indicator of the inactivation kinetics. n/N refers to the number of cells/transfections measured in each condition of study.
      The steady-state inactivation was determined using a two-step protocol in which conditioning prepulses were applied from a holding potential of −100 mV to a range of potentials from −100 to 40 mV in 10 mV increments for 5 s, immediately followed by a test pulse to 5 mV for 100 ms. For the construction of inactivation curves, the peak current amplitudes during the test pulses were normalized to the maximum peak current amplitude measured at −100 mV and plotted against the conditioning pulse. Steady-state inactivation curves were fitted to a modified Boltzmann equation:
      I/Imax=A1A21+e(VE0.5,inact)/dx+A2


      where I/Imax is the relative current measured at the test pulse, A1 and A2 represent, respectively, the maximum relative current value and the fraction of the noninactivated current, Vm is the voltage applied during the conditioning pulse, E0.5,inact is the voltage at which channels are half-maximally inactivated, and dx is the steepness of the slope.

      Data analysis and statistics

      Data were analyzed using a combination of pCLAMP software 11.2, Microsoft Excel, and OriginPro 2020 (OriginLab Corporation). Data in the tables are expressed as mean ± SD. Statistical significance was determined by one-way ANOVA and Bonferroni post hoc test in OriginPro 2020. The level of statistical significance was set at p < 0.05.

      Data availability

      All data are contained within the article.

      Supporting information

      This article contains supporting information.

      Conflict of interest

      The authors declare that they have no conflicts of interest with the contents of this article.

      Acknowledgments

      We are grateful for the ongoing collaboration with Dr Rafik Tadros from the Cardiovascular Genetics Center at the Montreal Heart Institute, and we thank Dr Rémy Sauvé for stimulating discussions and critical reading of the article.

      Author contributions

      L. P. conceptualization; J. Z., E. S., M. M., and L. P. methodology; J. Z., and L. P. validation; J. Z., and E. S. formal analysis; J. Z., E. S., and M. M. investigation; J. Z, E. S., and L. P. writing–original draft; L. P. supervision; L. P. project administration; L. P. funding acquisition; J. Z., E. S., M. M., and L. P. writing–review & editing.

      Funding and additional information

      This work was completed with the operating grant 159556 from the Canadian Institutes of Health Research to L.P. E.S. is the recipient of a PhD award from “Fonds de la recherche du Québec en nature et technologies.”

      Supporting information

      • Supplemental Figure S1

        CaV1.2 channel activity is modulated by substitutions at position Gly449. Panel A. Whole-cell currents were recorded from HEKT cells transiently transfected with CaV1.2 WT or variants co-expressed with Cavβ2a and Cavα2δ1. Exemplar traces are shown (from left to right) for CaV1.2 WT, G449A, G449D, G449P, and G449R. CaV1.2 currents were elicited from a holding potential of −100 mV and were depolarized to potentials ranging from −80 to 65 mV in 5 mV increments lasting 450 ms for each step (depicted above each series of recordings). The vertical scale bars are 10 pA/pF and the horizontal scale bars are 100 ms throughout. Panel B. Peak current densities of Cav1.2 WT, G449A, G449D, G449P, and G449R currents are plotted as a function of applied voltage and fitted by a Boltzman equation described in Experimental procedures. The activation curves were generated using the same protocol as in Panel A. Panels C and D show the summarized distribution of the peak current densities and the mid-potential of activation E0.5, act. Peak whole-cell currents and E0.5, act are reported individually as black circles. The mean data ± S.D. are shown as red hyphens. Values of peak current densities and E0.5, act are reported in Table 6.

      References

        • Grant A.O.
        Cardiac ion channels.
        Circ. Arrhythm. Electrophysiol. 2009; 2: 185-194
        • Wang S.Q.
        • Song L.S.
        • Lakatta E.G.
        • Cheng H.
        Ca2+ signalling between single L-type Ca2+ channels and ryanodine receptors in heart cells.
        Nature. 2001; 410: 592-596
        • Wu J.
        • Yan Z.
        • Li Z.
        • Qian X.
        • Lu S.
        • Dong M.
        • et al.
        Structure of the voltage-gated calcium channel Cav1.1 at 3.6 Å resolution.
        Nature. 2016; 537: 191-196
        • Segura E.
        • Bourdin B.
        • Tétreault M.P.
        • Briot J.
        • Allen B.G.
        • Mayer G.
        • et al.
        Proteolytic cleavage of the hydrophobic domain in the Ca(V)α2δ1 subunit improves assembly and activity of cardiac Ca(V)1.2 channels.
        J. Biol. Chem. 2017; 292: 11109-11124
        • Colecraft H.M.
        • Alseikhan B.
        • Takahashi S.X.
        • Chaudhuri D.
        • Mittman S.
        • Yegnasubramanian V.
        • et al.
        Novel functional properties of Ca(2+) channel beta subunits revealed by their expression in adult rat heart cells.
        J. Physiol. 2002; 541: 435-452
        • Van Petegem F.
        • Duderstadt K.E.
        • Clark K.A.
        • Wang M.
        • Minor Jr., D.L.
        Alanine-scanning mutagenesis defines a conserved energetic hotspot in the CaValpha1 AID-CaVbeta interaction site that is critical for channel modulation.
        Structure. 2008; 16: 280-294
        • Adams P.J.
        • Ben-Johny M.
        • Dick I.E.
        • Inoue T.
        • Yue D.T.
        Apocalmodulin itself promotes ion channel opening and Ca(2+) regulation.
        Cell. 2014; 159: 608-622
        • Zühlke R.D.
        • Pitt G.S.
        • Deisseroth K.
        • Tsien R.W.
        • Reuter H.
        Calmodulin supports both inactivation and facilitation of L-type calcium channels.
        Nature. 1999; 399: 159-162
        • Zühlke R.D.
        • Pitt G.S.
        • Tsien R.W.
        • Reuter H.
        Ca2+-sensitive inactivation and facilitation of L-type Ca2+ channels both depend on specific amino acid residues in a consensus calmodulin-binding motif in the(alpha)1C subunit.
        J. Biol. Chem. 2000; 275: 21121-21129
        • Pate P.
        • Mochca-Morales J.
        • Wu Y.
        • Zhang J.Z.
        • Rodney G.G.
        • Serysheva I.I.
        • et al.
        Determinants for calmodulin binding on voltage-dependent Ca2+ channels.
        J. Biol. Chem. 2000; 275: 39786-39792
        • Romanin C.
        • Gamsjaeger R.
        • Kahr H.
        • Schaufler D.
        • Carlson O.
        • Abernethy D.R.
        • et al.
        Ca(2+) sensors of L-type Ca(2+) channel.
        FEBS Lett. 2000; 487: 301-306
        • Pitt G.S.
        • Zühlke R.D.
        • Hudmon A.
        • Schulman H.
        • Reuter H.
        • Tsien R.W.
        Molecular basis of calmodulin tethering and Ca2+-dependent inactivation of L-type Ca2+ channels.
        J. Biol. Chem. 2001; 276: 30794-30802
        • Mouton J.
        • Feltz A.
        • Maulet Y.
        Interactions of calmodulin with two peptides derived from the c-terminal cytoplasmic domain of the Ca(v)1.2 Ca2+ channel provide evidence for a molecular switch involved in Ca2+-induced inactivation.
        J. Biol. Chem. 2001; 276: 22359-22367
        • Jervell A.
        • Lange-Nielsen F.
        Congenital deaf-mutism, functional heart disease with prolongation of the Q-T interval and sudden death.
        Am. Heart J. 1957; 54: 59-68
        • Liberthson R.R.
        Sudden death from cardiac causes in children and young adults.
        N. Engl. J. Med. 1996; 334: 1039-1044
        • Chugh S.S.
        • Reinier K.
        • Teodorescu C.
        • Evanado A.
        • Kehr E.
        • Al Samara M.
        • et al.
        Epidemiology of sudden cardiac death: clinical and research implications.
        Prog. Cardiovasc. Dis. 2008; 51: 213-228
        • Schwartz P.J.
        • Stramba-Badiale M.
        • Crotti L.
        • Pedrazzini M.
        • Besana A.
        • Bosi G.
        • et al.
        Prevalence of the congenital long-QT syndrome.
        Circulation. 2009; 120: 1761-1767
        • Tester D.J.
        • Ackerman M.J.
        Postmortem long QT syndrome genetic testing for sudden unexplained death in the young.
        J. Am. Coll. Cardiol. 2007; 49: 240-246
        • Tester D.J.
        • Will M.L.
        • Haglund C.M.
        • Ackerman M.J.
        Compendium of cardiac channel mutations in 541 consecutive unrelated patients referred for long QT syndrome genetic testing.
        Heart Rhythm. 2005; 2: 507-517
        • Splawski I.
        • Timothy K.W.
        • Sharpe L.M.
        • Decher N.
        • Kumar P.
        • Bloise R.
        • et al.
        Ca(V)1.2 calcium channel dysfunction causes a multisystem disorder including arrhythmia and autism.
        Cell. 2004; 119: 19-31
        • Splawski I.
        • Timothy K.W.
        • Decher N.
        • Kumar P.
        • Sachse F.B.
        • Beggs A.H.
        • et al.
        Severe arrhythmia disorder caused by cardiac L-type calcium channel mutations.
        Proc. Natl. Acad. Sci. U. S. A. 2005; 102 (discussion 8086-8088): 8089-8096
        • Gillis J.
        • Burashnikov E.
        • Antzelevitch C.
        • Blaser S.
        • Gross G.
        • Turner L.
        • et al.
        Long QT, syndactyly, joint contractures, stroke and novel CACNA1C mutation: Expanding the spectrum of Timothy syndrome.
        Am. J. Med. Genet. A. 2012; 158a: 182-187
        • Po C.
        • Zordan R.
        • Vecchi M.
        • Cerutti A.
        • Sartori S.
        • Trevisson E.
        • et al.
        Photosensitive epilepsy and long QT: expanding Timothy syndrome phenotype.
        Clin. Neurophysiol. 2019; 130: 2134-2136
        • Colson C.
        • Mittre H.
        • Busson A.
        • Leenhardt A.
        • Denjoy I.
        • Fressard V.
        • et al.
        Unusual clinical description of adult with Timothy syndrome, carrier of a new heterozygote mutation of CACNA1C.
        Eur. J. Med. Genet. 2019; 62: 103648
        • Boczek N.J.
        • Ye D.
        • Jin F.
        • Tester D.J.
        • Huseby A.
        • Bos J.M.
        • et al.
        Identification and functional characterization of a novel CACNA1C-mediated cardiac disorder characterized by prolonged QT intervals with hypertrophic cardiomyopathy, congenital heart defects, and sudden cardiac death.
        Circ. Arrhythm. Electrophysiol. 2015; 8: 1122-1132
        • Han D.
        • Xue X.
        • Yan Y.
        • Li G.
        Highlight article: dysfunctional Cav1.2 channel in Timothy syndrome, from cell to bedside.
        Exp. Biol. Med. (Maywood). 2019; 244: 960-971
        • Hennessey J.A.
        • Boczek N.J.
        • Jiang Y.H.
        • Miller J.D.
        • Patrick W.
        • Pfeiffer R.
        • et al.
        A CACNA1C variant associated with reduced voltage-dependent inactivation, increased CaV1.2 channel window current, and arrhythmogenesis.
        PLoS One. 2014; 9e106982
        • Kelu Bisabu K.
        • Zhao J.
        • Mokrane A.E.
        • Segura É.
        • Marsolais M.
        • Grondin S.
        • et al.
        Novel gain-of-function variant in CACNA1C associated with Timothy syndrome, multiple accessory pathways, and noncompaction cardiomyopathy.
        Circ. Genom Precis Med. 2020; 13e003123
        • Findeisen F.
        • Minor Jr., D.L.
        Disruption of the IS6-AID linker affects voltage-gated calcium channel inactivation and facilitation.
        J. Gen. Physiol. 2009; 133: 327-343
        • Papa A.
        • Kushner J.
        • Hennessey J.A.
        • Katchman A.N.
        • Zakharov S.I.
        • Chen B.X.
        • et al.
        Adrenergic Ca(V)1.2 activation via Rad phosphorylation converges at α(1C) I-II loop.
        Circ. Res. 2021; 128: 76-88
        • Almagor L.
        • Chomsky-Hecht O.
        • Ben-Mocha A.
        • Hendin-Barak D.
        • Dascal N.
        • Hirsch J.A.
        The role of a voltage-dependent Ca2+ channel intracellular linker: A structure-function analysis.
        J. Neurosci. 2012; 32: 7602-7613
        • Barrett C.F.
        • Tsien R.W.
        The Timothy syndrome mutation differentially affects voltage- and calcium-dependent inactivation of CaV1.2 L-type calcium channels.
        Proc. Natl. Acad. Sci. U. S. A. 2008; 105: 2157-2162
        • Dick I.E.
        • Joshi-Mukherjee R.
        • Yang W.
        • Yue D.T.
        Arrhythmogenesis in Timothy Syndrome is associated with defects in Ca(2+)-dependent inactivation.
        Nat. Commun. 2016; 710370
        • Raybaud A.
        • Dodier Y.
        • Bissonnette P.
        • Simoes M.
        • Bichet D.G.
        • Sauvé R.
        • et al.
        The role of the GX9GX3G motif in the gating of high voltage-activated Ca2+ channels.
        J. Biol. Chem. 2006; 281: 39424-39436
        • Tarnovskaya S.I.
        • Kostareva A.A.
        • Zhorov B.S.
        L-type calcium channel: predicting pathogenic/likely pathogenic status for variants of uncertain clinical significance.
        Membranes. 2021; 11: 599
        • Bourdin B.
        • Briot J.
        • Tétreault M.P.
        • Sauvé R.
        • Parent L.
        Negatively charged residues in the first extracellular loop of the L-type Ca(V)1.2 channel anchor the interaction with the Ca(V)α2δ1 auxiliary subunit.
        J. Biol. Chem. 2017; 292: 17236-17249
        • Briot J.
        • Mailhot O.
        • Bourdin B.
        • Tétreault M.P.
        • Najmanovich R.
        • Parent L.
        A three-way inter-molecular network accounts for the Ca(V)α2δ1-induced functional modulation of the pore-forming Ca(V)1.2 subunit.
        J. Biol. Chem. 2018; 293: 7176-7188
        • Saimi Y.
        • Kung C.
        Calmodulin as an ion channel subunit.
        Annu. Rev. Physiol. 2002; 64: 289-311
        • Urrutia J.
        • Aguado A.
        • Muguruza-Montero A.
        • Núñez E.
        • Malo C.
        • Casis O.
        • et al.
        The crossroad of ion channels and calmodulin in disease.
        Int. J. Mol. Sci. 2019; 20: 400
        • Tiaho F.
        • Piot C.
        • Nargeot J.
        • Richard S.
        Regulation of the frequency-dependent facilitation of L-type Ca2+ currents in rat ventricular myocytes.
        J. Physiol. 1994; 477: 237-251
        • Villarroel A.
        • Taglialatela M.
        • Bernardo-Seisdedos G.
        • Alaimo A.
        • Agirre J.
        • Alberdi A.
        • et al.
        The ever changing moods of calmodulin: how structural plasticity entails transductional adaptability.
        J. Mol. Biol. 2014; 426: 2717-2735
        • Persechini A.
        • Stemmer P.M.
        Calmodulin is a limiting factor in the cell.
        Trends Cardiovasc. Med. 2002; 12: 32-37
        • DeMaria C.D.
        • Soong T.W.
        • Alseikhan B.A.
        • Alvania R.S.
        • Yue D.T.
        Calmodulin bifurcates the local Ca2+ signal that modulates P/Q-type Ca2+ channels.
        Nature. 2001; 411: 484-489
        • Simms B.A.
        • Souza I.A.
        • Zamponi G.W.
        Effect of the Brugada syndrome mutation A39V on calmodulin regulation of Cav1.2 channels.
        Mol. Brain. 2014; 7: 34
        • Ravindran A.
        • Lao Q.Z.
        • Harry J.B.
        • Abrahimi P.
        • Kobrinsky E.
        • Soldatov N.M.
        Calmodulin-dependent gating of Ca(v)1.2 calcium channels in the absence of Ca(v)beta subunits.
        Proc. Natl. Acad. Sci. U. S. A. 2008; 105: 8154-8159
        • Limpitikul W.B.
        • Dick I.E.
        • Joshi-Mukherjee R.
        • Overgaard M.T.
        • George Jr., A.L.
        • Yue D.T.
        Calmodulin mutations associated with long QT syndrome prevent inactivation of cardiac L-type Ca(2+) currents and promote proarrhythmic behavior in ventricular myocytes.
        J. Mol. Cell Cardiol. 2014; 74: 115-124
        • Nowycky M.C.
        • Fox A.P.
        • Tsien R.W.
        Long-opening mode of gating of neuronal calcium channels and its promotion by the dihydropyridine calcium agonist Bay K 8644.
        Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 2178-2182
        • Asmara H.
        • Minobe E.
        • Saud Z.A.
        • Kameyama M.
        Interactions of calmodulin with the multiple binding sites of Cav1.2 Ca2+ channels.
        J. Pharmacol. Sci. 2010; 112: 397-404
        • Wang H.G.
        • George M.S.
        • Kim J.
        • Wang C.
        • Pitt G.S.
        Ca/Calmodulin regulates trafficking of Cav1.2 Ca channels in cultured hippocampal neurons.
        J. Neurosci. 2007; 27: 9086-9093
        • Osawa M.
        • Swindells M.B.
        • Tanikawa J.
        • Tanaka T.
        • Mase T.
        • Furuya T.
        • et al.
        Solution structure of calmodulin-W-7 complex: the basis of diversity in molecular recognition.
        J. Mol. Biol. 1998; 276: 165-176
        • Osawa M.
        • Kuwamoto S.
        • Izumi Y.
        • Yap K.L.
        • Ikura M.
        • Shibanuma T.
        • et al.
        Evidence for calmodulin inter-domain compaction in solution induced by W-7 binding.
        FEBS Lett. 1999; 442: 173-177
        • Quadroni M.
        • James P.
        • Carafoli E.
        Isolation of phosphorylated calmodulin from rat liver and identification of the in vivo phosphorylation sites.
        J. Biol. Chem. 1994; 269: 16116-16122
        • Villalobo A.
        The multifunctional role of phospho-calmodulin in pathophysiological processes.
        Biochem. J. 2018; 475: 4011-4023
        • Tabernero L.
        • Taylor D.A.
        • Chandross R.J.
        • VanBerkum M.F.
        • Means A.R.
        • Quiocho F.A.
        • et al.
        The structure of a calmodulin mutant with a deletion in the central helix: implications for molecular recognition and protein binding.
        Structure. 1997; 5: 613-622
        • Bildl W.
        • Strassmaier T.
        • Thurm H.
        • Andersen J.
        • Eble S.
        • Oliver D.
        • et al.
        Protein kinase CK2 is coassembled with small conductance Ca(2+)-activated K+ channels and regulates channel gating.
        Neuron. 2004; 43: 847-858
        • Plancke Y.D.
        • Lazarides E.
        Evidence for a phosphorylated form of calmodulin in chicken brain and muscle.
        Mol. Cell Biol. 1983; 3: 1412-1420
        • Quadroni M.
        • L'Hostis E.L.
        • Corti C.
        • Myagkikh I.
        • Durussel I.
        • Cox J.
        • et al.
        Phosphorylation of calmodulin alters its potency as an activator of target enzymes.
        Biochemistry. 1998; 37: 6523-6532
        • Sacks D.B.
        • Davis H.W.
        • Crimmins D.L.
        • McDonald J.M.
        Insulin-stimulated phosphorylation of calmodulin.
        Biochem. J. 1992; 286: 211-216
        • Sacks D.B.
        • Mazus B.
        • Joyal J.L.
        The activity of calmodulin is altered by phosphorylation: modulation of calmodulin function by the site of phosphate incorporation.
        Biochem. J. 1995; 312: 197-204
        • Li S.C.
        • Goto N.K.
        • Williams K.A.
        • Deber C.M.
        Alpha-helical, but not beta-sheet, propensity of proline is determined by peptide environment.
        Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 6676-6681
        • Ambrosino P.
        • Alaimo A.
        • Bartollino S.
        • Manocchio L.
        • De Maria M.
        • Mosca I.
        • et al.
        Epilepsy-causing mutations in Kv7.2 C-terminus affect binding and functional modulation by calmodulin.
        Biochim. Biophys. Acta. 2015; 1852: 1856-1866
        • Ben-Johny M.
        • Yue D.T.
        Calmodulin regulation (calmodulation) of voltage-gated calcium channels.
        J. Gen. Physiol. 2014; 143: 679
        • Kim E.Y.
        • Rumpf C.H.
        • Fujiwara Y.
        • Cooley E.S.
        • Van Petegem F.
        • Minor J.
        Structures of CaV2 Ca2+/CaM-IQ domain complexes reveal binding modes that underlie calcium-dependent inactivation and facilitation.
        Structure. 2008; 16: 1455-1467
        • Van Petegem F.
        • Chatelain F.C.
        • Minor Jr., D.L.
        Insights into voltage-gated calcium channel regulation from the structure of the CaV1.2 IQ domain-Ca2+/calmodulin complex.
        Nat. Struct. Mol. Biol. 2005; 12: 1108-1115
        • Dick I.E.
        • Tadross M.R.
        • Liang H.
        • Tay L.H.
        • Yang W.
        • Yue D.T.
        A modular switch for spatial Ca2+ selectivity in the calmodulin regulation of CaV channels.
        Nature. 2008; 451: 830-834
        • Simms B.A.
        • Souza I.A.
        • Zamponi G.W.
        A novel calmodulin site in the Cav1.2 N-terminus regulates calcium-dependent inactivation.
        Pflugers Arch. 2014; 466: 1793-1803
        • Ivanina T.
        • Blumenstein Y.
        • Shistik E.
        • Barzilai R.
        • Dascal N.
        Modulation of L-type Ca2+ channels by gbeta gamma and calmodulin via interactions with N and C termini of alpha 1C.
        J. Biol. Chem. 2000; 275: 39846-39854
        • Noble S.
        • Shimoni Y.
        The calcium and frequency dependence of the slow inward current ‛staircase’ in frog atrium.
        J. Physiol. 1981; 310: 57-75
        • Marban E.
        • Tsien R.W.
        Enhancement of calcium current during digitalis inotrophy in mammalian heart: positive feed-back regulation by intracellular calcium?.
        J. Physiol. 1982; 329: 589-614
        • Kim J.
        • Ghosh S.
        • Nunziato D.A.
        • Pitt G.S.
        Identification of the components controlling inactivation of voltage-gated Ca(2+) channels.
        Neuron. 2004; 41: 745-754
        • Campiglio M.
        • Flucher B.E.
        The role of auxiliary subunits for the functional diversity of voltage-gated calcium channels.
        J. Cell Physiol. 2015; 230: 2019-2031
        • Beyl S.
        • Depil K.
        • Hohaus A.
        • Stary-Weinzinger A.
        • Timin E.
        • Shabbir W.
        • et al.
        Physicochemical properties of pore residues predict activation gating of CaV1.2: a correlation mutation analysis.
        Pflugers Arch. 2011; 461: 53-63
        • Dafi O.
        • Berrou L.
        • Dodier Y.
        • Raybaud A.
        • Sauvé R.
        • Parent L.
        Negatively charged residues in the N-terminal of the AID helix confer slow voltage dependent inactivation gating to CaV1.2.
        Biophys. J. 2004; 87: 3181-3192
        • Hu Z.
        • Li G.
        • Wang J.-W.
        • Chong S.Y.
        • Yu D.
        • Wang X.
        • et al.
        Regulation of blood pressure by targeting CaV1.2-galectin-1 protein interaction.
        Circulation. 2018; 138: 1431-1445
        • Bourdin B.
        • Marger F.
        • Wall-Lacelle S.
        • Schneider T.
        • Klein H.
        • Sauvé R.
        • et al.
        Molecular determinants of the CaVbeta-induced plasma membrane targeting of the CaV1.2 channel.
        J. Biol. Chem. 2010; 285: 22853-22863
        • Buraei Z.
        • Yang J.
        The β subunit of voltage-gated Ca2+ channels.
        Physiol. Rev. 2010; 90: 1461-1506
        • Finlin B.S.
        • Crump S.M.
        • Satin J.
        • Andres D.A.
        Regulation of voltage-gated calcium channel activity by the Rem and Rad GTPases.
        Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 14469-14474
        • Liu G.
        • Papa A.
        • Katchman A.N.
        • Zakharov S.I.
        • Roybal D.
        • Hennessey J.A.
        • et al.
        Mechanism of adrenergic CaV1.2 stimulation revealed by proximity proteomics.
        Nature. 2020; 577: 695-700
        • Almagor L.
        • Avinery R.
        • Hirsch J.A.
        • Beck R.
        Structural flexibility of CaV1.2 and CaV2.2 I-II proximal linker fragments in solution.
        Biophys. J. 2013; 104: 2392-2400
        • Almagor L.
        • Chomsky-Hecht O.
        • Ben-Mocha A.
        • Hendin-Barak D.
        • Dascal N.
        • Hirsch J.A.
        CaV1.2 I-II linker structure and Timothy syndrome.
        Channels (Austin). 2012; 6: 468-472
        • Shakeri B.
        • Bourdin B.
        • Demers-Giroux P.O.
        • Sauvé R.
        • Parent L.
        A quartet of leucine residues in the guanylate kinase domain of CaVβ determines the plasma membrane density of the CaV2.3 channel.
        J. Biol. Chem. 2012; 287: 32835-32847
        • Opatowsky Y.
        • Chen C.C.
        • Campbell K.P.
        • Hirsch J.A.
        Structural analysis of the voltage-dependent calcium channel beta subunit functional core and its complex with the alpha 1 interaction domain.
        Neuron. 2004; 42: 387-399
        • Van Petegem F.
        • Clark K.A.
        • Chatelain F.C.
        • Minor Jr., D.L.
        Structure of a complex between a voltage-gated calcium channel beta-subunit and an alpha-subunit domain.
        Nature. 2004; 429: 671-675
        • Chen Y.H.
        • Li M.H.
        • Zhang Y.
        • He L.L.
        • Yamada Y.
        • Fitzmaurice A.
        • et al.
        Structural basis of the alpha1-beta subunit interaction of voltage-gated Ca2+ channels.
        Nature. 2004; 429: 675-680
        • Pace N.C.
        • Scholtz M.J.
        A helix propensity scale based on experimental studies of peptides and proteins.
        Biophys. J. 1998; 75: 422-427
        • Dixon R.E.
        • Cheng E.P.
        • Mercado J.L.
        • Santana L.F.
        L-type Ca2+ channel function during Timothy syndrome.
        Trends Cardiovasc. Med. 2012; 22: 72-76
        • Bourdin B.
        • Shakeri B.
        • Tétreault M.P.
        • Sauvé R.
        • Lesage S.
        • Parent L.
        Functional characterization of CaVα2β mutations associated with sudden cardiac death.
        J. Biol. Chem. 2015; 290: 2854-2869
        • Lee C.H.
        • MacKinnon R.
        Activation mechanism of a human SK-calmodulin channel complex elucidated by cryo-EM structures.
        Science. 2018; 360: 508-513