Negatively charged residues in the first extracellular loop of the L-type CaV1.2 channel anchor the interaction with the CaVα2δ1 auxiliary subunit

Abstract

Voltage-gated L-type Ca_V1.2 channels in cardiomyocytes exist as heteromeric complexes. Co-expression of Ca_Vα2δ1 with Ca_Vβ/Ca_Vα1 proteins reconstitutes the functional properties of native L-type currents, but the interacting domains at the Ca_V1.2/Ca_Vα2δ1 interface are unknown. Here, a homology-based model of Ca_V1.2 identified protein interfaces between the extracellular domain of Ca_Vα2δ1 and the extracellular loops of the Ca_Vα1 protein in repeats I (IS1S2 and IS5S6), II (IIS5S6), and III (IIIS5S6). Insertion of a 9-residue hemagglutinin epitope in IS1S2, but not in IS5S6 or in IIS5S6, prevented the co-immunoprecipitation of Ca_V1.2 with Ca_Vα2δ1. IS1S2 contains a cluster of three conserved negatively charged residues Glu-179, Asp-180, and Asp-181 that could contribute to non-bonded interactions with Ca_Vα2δ1. Substitutions of Ca_V1.2 Asp-181 impaired the co-immunoprecipitation of Ca_Vβ/Ca_V1.2 with Ca_Vα2δ1 and the Ca_Vα2δ1-dependent shift in voltage-dependent activation gating. In contrast, single substitutions in Ca_V1.2 in neighboring positions in the same loop (179, 180, and 182-184) did not significantly alter the functional up-regulation of Ca_V1.2 whole-cell currents. However, a negatively charged residue at position 180 was necessary to convey the Ca_Vα2δ1-mediated shift in the activation gating. We also found a more modest contribution from the positively charged Arg-1119 in the extracellular pore region in repeat III of Ca_V1.2. We conclude that Ca_V1.2 Asp-181 anchors the physical interaction that facilitates the Ca_Vα2δ1-mediated functional modulation of Ca_V1.2 currents. By stabilizing the first extracellular loop of Ca_V1.2, Ca_Vα2δ1 may up-regulate currents by promoting conformations of the voltage sensor that are associated with the channel's open state.

Keywords: calcium channel; electrophysiology; gating; molecular modeling; protein-protein interaction.

Figure 1.

Three-dimensional cryo-electron microscopy structure of the Ca_V1.1 channel. A, surface representation of the rabbit Ca_V1.1 channel (cyan) in complex with Ca_Vα2δ1 (deep blue) and Ca_Vβ2 (violet) (PDB: 5GJV). The structural domains of Ca_Vα2δ1 and the relative orientation of repeats I–IV in the pore-forming Ca_Vα1 subunit of Ca_V1.1 are identified. For the correlation between the primary sequence and the structural domains of Ca_Vα2δ1, see Fig. 2 in Ref. . The extensive interface between the two proteins is surrounded by a dashed black square, and this region is shown enlarged in B and C. B, structural details of the VWA domain of Ca_Vα2δ1 are emphasized, and the residues forming the MIDAS are identified by stars. C, extracellular loops of Ca_V1.1 forming the interface with Ca_Vα2δ1 are shown. Only the main chains are shown. The S1S2 loop in repeat I (IS1S2, residues 71–82) is in red; the turret and external pore region S5S6 in repeat I (IS5S6, residues 219–278) is in black; the turret and external pore region S5S6 in repeat II (IIS5S6, residues 581–600) is in yellow, and the turret and external pore region S5S6 in repeat III (IIIS5S6, residues 950–998) is in orange. Images were produced with PyMOL (Molecular Graphics System, Version 1.8 Schrödinger, LLC).

Figure 2.

Inserting epitopes within the first extracellular loop of Ca_V1.2 prevents functional modulation of whole-cell current by Ca_Vα2δ1. Stable recombinant HEKT cells expressing Ca_Vβ3 were transiently transfected with pmCherry–Ca_Vα2δ1–HA (or pmCherry–Ca_Vα2δ1 WT) and pCMV–Ca_V1.2 WT or mutants as indicated over each current trace. In all cases, Ca_Vα2δ1 was tagged in the C terminus by an mCherry epitope. The epitopes HA or BTX were inserted after the identified residues without altering the primary sequence on each side of the insertion site. A, representative whole-cell Ca²⁺ current traces obtained after recombinant expression of Ca_V1.2 WT + Ca_Vα2δ1 WT, Ca_V1.2 WT + Ca_Vα2δ1–HA, Ca_V1.2-HA (Ser-182) + Ca_Vα2δ1–HA, Ca_V1.2–HA (Ser-182) + Ca_Vα2δ1 WT, Ca_V1.2–BTX (Ser-182) + Ca_Vα2δ1–HA, and Ca_V1.2–HA (Glu-331) + Ca_Vα2δ1–HA (from left to right). Ca_V1.2 currents were modulated to the same extent in the presence of mCherry–Ca_Vα2δ1–HA or mCherry–Ca_Vα2δ1. Furthermore, similar results were obtained with Ca_V1.2–HA (Glu-331). Currents were recorded in the presence of 2 mm Ca²⁺ from a holding potential of −100 mV. Time scale is 100 ms throughout. The current density scale ranged from 2 to 10 pA/pF, as indicated. B, averaged current-voltage relationships. The absolute peak current densities measured with mCherry–Ca_Vα2δ1–HA varied from −4 to −46 pA/pF over the 12-month recording period with a mean of −18 ± 1 pA/pF (n = 243). Averaged peak current densities obtained with the mock mCherry vector are shown. Co-expression with Ca_Vα2δ1 left-shifted the voltage dependence of activation of Ca_V1.2 WT/Ca_Vβ3 from E_{0.5, act} = +8 ± 2 mV (n = 25) to E_{0.5, act} = −8.3 ± 0.2 mV (n = 243) for Ca_V1.2 WT/Ca_Vβ3 with mCherry–Ca_Vα2δ1–HA. Statistical analyses were performed with a one-way ANOVA test: *, p < 0.01, and **, p < 0.001, against the mock vector. See Table 1 for details. C, distribution of the free energies of activation. The free energies of activation (ΔG_act) measured in the presence of the mock vector and in the presence of mCherry–Ca_Vα2δ1–HA are centered at 0.5 ± 0.1 and −0.78 ± 0.03 kcal mol⁻¹, respectively. The distribution of the ΔG_act values for the following combinations Ca_V1.2–HA (Ser-182) with Ca_Vα2δ1–HA, Ca_V1.2-HA (Ser-182) with Ca_Vα2δ1 WT, and Ca_V1.2–BTX (Ser-182) with Ca_Vα2δ1–HA overlapped with the ΔG_act values measured in the presence of the mock vector (no Ca_Vα2δ1).

Figure 3.

Epitope insertion in the first extracellular loop of Ca_V1.2 impairs the co-immunoprecipitation of Ca_Vα2δ1 with Ca_V1.2/Ca_Vβ3 proteins. HEKT cells were transiently transfected with pmCherry–Ca_Vα2δ1–HA and pCMV–Ca_Vβ3–c-Myc and either pCMV–Ca_V1.2 WT, pCMV–Ca_V1.2–HA (Ser-182), pCMV–Ca_V1.2–HA (Glu-331), or pCMV–Ca_V1.2–HA (Asp-710). Cell lysates were immunoprecipitated (IP) overnight with anti-c-Myc magnetic beads to capture Ca_Vβ3, eluted in a Laemmli buffer, and fractionated by SDS-PAGE using 8% gels. A, immunoblotting was carried out on total proteins (20 μg) collected from the cell lysates for each of the three proteins (Ca_V1.2, Ca_Vα2δ1, and Ca_Vβ3) before the immunoprecipitation assay (Input) to confirm that each protein was translated at the expected molecular mass. Each experimental condition is identified by the specific Ca_V1.2 construct. The signal for the housekeeping protein GAPDH is shown below each blot. B, immunoblotting (IB) was carried out after eluting the protein complexes from the anti-c-Myc beads with anti-Ca_V1.2, anti-Ca_Vα2δ1, and anti-Ca_Vβ3 antibodies (from top to bottom, as indicated). Images for Ca_Vα2δ1 were captured after short (1 s) or long exposure times (20 and 200 s). Ca_Vβ3 and Ca_V1.2 proteins migrated at 60 and 250 kDa, respectively. All Ca_Vα2δ1 proteins migrated at ≈175 kDa, which is consistent with the molecular mass of the mCherry–Ca_Vα2δ1–HA in previous studies (30). All immunoblots were carried out in parallel under the same transfection and extraction conditions. C, proteins that did not bind to the antibody-bead complex (referred to as the flow-through fraction) were collected, diluted in a Laemmli buffer, and fractionated by SDS-PAGE using an 8% gel and revealed with the anti-Ca_Vα2δ1. As seen, mCherry–Ca_Vα2δ1–HA is present in the flow-through fraction at the expected molecular mass (175 kDa) confirming that the proteins were appropriately translated and were present in the preparation in detectable quantities throughout. These experiments were carried out three times with the mutants and 10 times for the WT construct over a period of 5 months and yielded qualitatively similar results.

Figure 4.

Three-dimensional model of the extracellular loops of the rabbit Ca_V1.2 in complex with the VWA domain of the rat Ca_Vα2δ1 protein. The 3D model of the region spanning the first transmembrane helix S1 to the fourth transmembrane helix S4 in the first repeat in Ca_V1.2 (IS1S4) is shown in cyan, and the VWA domain of Ca_Vα2δ1 is shown in deep blue. Residues Pro-178, Glu-179, Asp-180, Asp-181, and Ala-184 of Ca_V1.2 and residues Ser-261, Gly-262, Ser-263, and Glu-366 in Ca_Vα2δ1 (three of the five residues in the MIDAS) are shown in stick representation with oxygen and nitrogen atoms colored in red and in blue, respectively. The model does not predict strong electrostatic interactions between Ca_V1.2 Asp-180 and residues in Ca_Vα2δ1. Intramolecular interactions with residues in the extracellular IS3S4 loop are not ruled out. Ca_V1.2 Asp-181 appears to be appropriately oriented to form electrostatic interactions with Gly-262 and Ser-263 in Ca_Vα2δ1. Modeling was achieved with Modeler 9.17. The figure was produced using PyMOL.

Figure 5.

Mutations of the aspartate residue at position 181 impair the co-immunoprecipitation of Ca_Vα2δ1 with Ca_V1.2/Ca_Vβ3 proteins. A–C, HEKT cells were transiently transfected with pmCherry–Ca_Vα2δ1–HA and pCMV–Ca_Vβ3–c-Myc and either pCMV–Ca_V1.2 WT, pCMV–Ca_V1.2 E179A, pCMV–Ca_V1.2 D180A, or pCMV–Ca_V1.2 D181A. Cell lysates were immunoprecipitated (IP) overnight with anti-c-Myc magnetic beads to capture Ca_Vβ3, eluted in a Laemmli buffer, and fractionated by SDS-PAGE using 8% gels. A, immunoblotting was carried out on total proteins (20 μg) collected from the cell lysates for each of the three proteins (Ca_V1.2, Ca_Vα2δ1, and Ca_Vβ3) before the immunoprecipitation assay (Input) to confirm that each protein was translated at the expected molecular mass. Each experimental condition is identified by the specific Ca_V1.2 construct. The signal for the housekeeping protein GAPDH is shown below each blot. B, immunoblotting was carried out after eluting the protein complexes from the beads with anti-Ca_V1.2, anti-Ca_Vα2δ1, and anti-Ca_Vβ3 antibodies (from top to bottom, as indicated). Images for Ca_Vα2δ1 were captured after short (1 s) or longer exposure times (20 and 200 s). All immunoblots were carried out in parallel under the same transfection and extraction conditions. C, proteins that did not bind to the antibody-bead complex (flow-through fraction) were collected, diluted in a Laemmli buffer, and fractionated by SDS-PAGE using an 8% gel and revealed with the anti-Ca_Vα2δ1. D–F, HEKT cells were transiently transfected with pmCherry–Ca_Vα2δ1–HA and pCMV–Ca_Vβ3–c-Myc and either pCMV–Ca_V1.2 WT, pCMV–Ca_V1.2 D180A, pCMV–Ca_V1.2 D180E, pCMV–Ca_V1.2 D181A, or pCMV–Ca_V1.2 D181E. Cell lysates were immunoprecipitated overnight with anti-c-Myc magnetic beads to capture Ca_Vβ3, eluted in a Laemmli buffer, and fractionated by SDS-PAGE using 8% gels. D, immunoblotting was carried out on total proteins (20 μg) collected from the cell lysates for each of the three proteins (Ca_V1.2, Ca_Vα2δ1, and Ca_Vβ3) before the immunoprecipitation assay (Input) to confirm that each protein was translated at the expected molecular mass. Each experimental condition is identified by the specific Ca_V1.2 construct. The signal for the housekeeping protein GAPDH is shown below each blot. E, immunoblotting (IB) was carried out with anti-Ca_V1.2, anti-Ca_Vα2δ1, and anti-Ca_Vβ3 antibodies (from top to bottom, as indicated) after eluting the protein complexes from the beads. Images for Ca_Vα2δ1 were captured after short (1 s) or longer exposure times (20 and 200 s). All immunoblots were carried out in parallel under the same transfection and extraction conditions. F, proteins that did not bind to the antibody-bead complex (flow-through fraction) were collected, diluted in a Laemmli buffer, and fractionated by SDS-PAGE using an 8% gel and revealed with the anti-Ca_Vα2δ1. As seen, mCherry–Ca_Vα2δ1–HA is present in the flow-through fraction at the expected molecular mass (175 kDa) confirming that the proteins were appropriately translated and were present in the preparation in detectable quantities throughout. These experiments were carried out four times with the mutants and 10 times for the WT construct over a period of 5 months and yielded reproducible results.

Figure 6.

Mutations at Asp-181 prevent up-regulation of Ca_V1.2 currents. HEKT cells were transiently transfected with pCMV–Ca_Vβ3 and pmCherry–Ca_Vα2δ1–HA with the Ca_V1.2 constructs (D181A, D181G, D181R, and D181E). A, whole-cell Ca²⁺ current traces were recorded in the presence of 2 mm Ca²⁺ from a holding potential of −100 mV for the constructs as identified. The current traces with the largest currents are shown for Ca_V1.2 constructs D181G and D181E. Time scale is 100 ms throughout. The current density scale is either 5 or 10 pA/pF as indicated. B, averaged current-voltage relationships. Peak current densities versus voltage relationships were measured for Ca_V1.2 WT and Ca_V1.2 mutants (as shown). Currents traces obtained with the empty mCherry (mock) vector are also shown. Ca_V1.2 constructs D181A, D181G, and D181R generated currents that were not significantly up-regulated by mCherry–Ca_Vα2δ1–HA WT. Statistical analyses were performed with a one-way ANOVA test: *, p < 0.01, and **, p < 0.001 against the mock vector. See Table 1 for details. C, distribution of the free energies of activation. The values for the free energy of activation (ΔG_act) measured for Ca_V1.2 constructs (D181A, D181G, D181E, and D181R) overlapped with the values measured for the mock vector. D, representative two-dimensional plots of mCherry versus FITC fluorescence. The cell-surface expression of the Ca_V1.2 mutants was evaluated by introducing the mutation in the mCherry–Ca_V1.2–HA construct. The surface fluorescence was estimated from the relative intensity of the fluorescence emitted by the fluorescein isothiocyanate (FITC)-conjugated anti-HA as measured using a flow cytometry assay (10,000 intact cells). The construct allows for detection of intracellular and extracellular fluorescence using FITC-conjugated anti-HA (“x axis”) and an anti-mCherry (“y axis”), respectively. The robust mCherry signal (y axis) confirms that the proteins were translated up to the end of the coding sequence. The cell-surface fluorescence for FITC, calculated as ΔMedFI as explained under “Experimental procedures,” was slightly lower for the Ca_V1.2 mutants (D181A, D181G, and D181R) than for the WT construct. Nonetheless, all constructs significantly fluoresced at the cell surface supporting the view that the absence of function did not result from a complete absence of trafficking to the cell membrane. Furthermore, the ΔMedFI signal for the total protein was similar for all WT and mutant constructs demonstrating that proteins were appropriately translated, an observation also obtained from carrying out routine Western blotting.

Figure 7.

Asp-180 in Ca_V1.2 controls the Ca_Vα2δ1-induced shift in voltage-dependent gating of Ca_V1.2 currents. A, representative whole-cell Ca²⁺ current traces obtained after the transient expression of pCMV–Ca_Vβ3 and pmCherry–Ca_Vα2δ1–HA with the Ca_V1.2 constructs (D180I, D180A, D180G, and D180E) in HEKT cells. Note that the leftmost current trace was obtained in the presence of Ca_V1.2 WT and the mock mCherry vector. Recordings were made in the presence of 2 mm Ca²⁺ from a holding potential of −100 mV. Time scale is 100 ms throughout. The current density scale ranged from 2 to 10 pA/pF as indicated. B, averaged current-voltage relationships. Peak current densities versus voltage relationships were measured for Ca_V1.2 WT and Ca_V1.2 mutants (as shown). Averaged peak current densities obtained with the mock mCherry vector are shown in light gray stars. All Ca_V1.2 constructs (D180I, D180A, D180G, and D180E) were up-regulated by mCherry–Ca_Vα2δ–HA albeit to variable extent. Statistical analyses were performed with a one-way ANOVA test: *, p < 0.01, and **, p < 0.001, against the mock mCherry vector. Nonetheless, the current-voltage relationships measured with the Ca_V1.2 mutants were clearly shifted to the right when compared with Ca_V1.2 WT. See Table 1 for details. C, distribution of the free energies of activation. The free energies of activation (ΔG_act) for Ca_V1.2 D180E, D180I, D180A, and D180G did not overlap with the values measured with Ca_V1.2 WT. ΔG_act values for Ca_V1.2 D180A and D180I were not significantly different from ΔG_act values measured for the mock vector, whereas ΔG_act values for Ca_V1.2 D180G and D180E were significantly different at p < 0.01.

Figure 8.

Extracellular loop S5S6 in repeat III has a modest impact on the interaction between Ca_Vα2δ1 and Ca_V1. 2/Ca_Vβ3 proteins. A, 3D model of IIIS5S6 in Ca_V1.2 (residues 1059–1204) is shown in cyan, and the cache1 domain (residues 104–223) in Ca_Vα2δ1 is shown in deep blue. The carboxyl group on the side chain of Arg-1119 in Ca_V1.2 is located close enough (< 3 Å) to the main-chain atoms of Glu-171 in Ca_Vα2δ1 to potentially contribute to the formation of hydrogen bonds. In contrast, the 3D model does not predict a favorable interaction between Lys-1100 in Ca_V1.2 and Glu-174 in Ca_Vα2δ1 with a minimum distance estimated to be at 4.6 Å. Modeling was achieved with Modeler 9.17. The figure was produced using PyMOL. B–D, HEKT cells were transiently transfected with pmCherry–Ca_Vα2δ1–HA WT and pCMV–Ca_Vβ3–c-Myc and either pCMV–Ca_V1.2 WT, pCMV–Ca_V1.2 R1119A, or pCMV–Ca_V1.2 K1100A. Cell lysates were immunoprecipitated overnight with anti-c-Myc magnetic beads to capture Ca_Vβ3, eluted in a Laemmli buffer, and fractionated by SDS-PAGE using 8% gels. B, immunoblotting (IB) was carried out on total proteins (20 μg) collected from the cell lysates for each of the three proteins (Ca_V1.2, Ca_Vα2δ1, and Ca_Vβ3) before the immunoprecipitation assay (Input) to confirm that each protein was translated at the expected molecular mass. Each experimental condition is identified by the specific Ca_V1.2 construct. C, immunoblotting was carried out as detailed earlier. Images for Ca_Vα2δ1 were captured after 1, 20, and 200 s of exposure. The signal for the anti-Ca_Vα2δ1 was detected only after a 200-s exposure when probed in the presence of Ca_V1.2 R1119A/Ca_Vβ3. D, protein lysates that ran through without binding to the antibody-bead complex (flow-through fraction) were collected, diluted in a Laemmli buffer, and fractionated by SDS-PAGE using an 8% gel and revealed with the anti-Ca_Vα2δ1. As seen, mCherry–Ca_Vα2δ1–HA is present in all flow-through fractions at the expected molecular mass (175 kDa) confirming that the proteins were appropriately translated and were present in the preparation in detectable quantities throughout.

Figure 9.

Double mutant analysis of charged residues in repeat III in Ca_V1.2. HEKT cells were transiently transfected with Ca_Vβ3, pCMV–Ca_V1.2 WT or mutant and pmCherry–Ca_Vα2δ1–HA WT or mutant. A, representative whole-cell Ca²⁺ current traces were recorded in the presence of 2 mm Ca²⁺ from a holding potential of −100 mV. From left to right: Ca_V1.2 WT with mCherry–Ca_Vα2δ1–HA WT; Ca_V1.2 R1119A with mCherry–Ca_Vα2δ1–HA WT; Ca_V1.2 WT with mCherry–Ca_Vα2δ1–HA D171A, and Ca_V1.2 R1119A with mCherry–Ca_Vα2δ1–HA D171A. Time scale is 100 ms throughout. The current density scale is either 5 or 10 pA/pF as indicated. B, averaged current-voltage relationships. Peak current densities versus voltage relationships were measured for the WT construct and the mutants (as shown). Current traces obtained with the empty mCherry vector (mock vector) are also shown. Statistical analyses were performed with a one-way ANOVA test: *, p < 0.01, and **, p < 0.001, against the mock vector. See Table 1 for details. C, distribution of the free energies of activation. The values of the free energy of activation (ΔG_act) for all conditions were significantly different from the values measured with the mock vector.