The ß subunit of voltage-gated Ca2+ channels

Abstract

Calcium regulates a wide spectrum of physiological processes such as heartbeat, muscle contraction, neuronal communication, hormone release, cell division, and gene transcription. Major entryways for Ca(2+) in excitable cells are high-voltage activated (HVA) Ca(2+) channels. These are plasma membrane proteins composed of several subunits, including α(1), α(2)δ, β, and γ. Although the principal α(1) subunit (Ca(v)α(1)) contains the channel pore, gating machinery and most drug binding sites, the cytosolic auxiliary β subunit (Ca(v)β) plays an essential role in regulating the surface expression and gating properties of HVA Ca(2+) channels. Ca(v)β is also crucial for the modulation of HVA Ca(2+) channels by G proteins, kinases, and the Ras-related RGK GTPases. New proteins have emerged in recent years that modulate HVA Ca(2+) channels by binding to Ca(v)β. There are also indications that Ca(v)β may carry out Ca(2+) channel-independent functions, including directly regulating gene transcription. All four subtypes of Ca(v)β, encoded by different genes, have a modular organization, consisting of three variable regions, a conserved guanylate kinase (GK) domain, and a conserved Src-homology 3 (SH3) domain, placing them into the membrane-associated guanylate kinase (MAGUK) protein family. Crystal structures of Ca(v)βs reveal how they interact with Ca(v)α(1), open new research avenues, and prompt new inquiries. In this article, we review the structure and various biological functions of Ca(v)β, with both a historical perspective as well as an emphasis on recent advances.

Fig. 1

Molecular organization of voltage-gated Ca²⁺ channels. A: subunit composition of high-voltage activated (HVA) Ca²⁺ channels. B: schematic representation of the predicted transmembrane topology of Ca_vα₁, with the location of the α-interaction domain (AID) marked. C: Ca²⁺ channel current types and the corresponding α₁ subunits of the channels that produce them. D: list of all cloned auxiliary HVA Ca²⁺ channel subunits. E: amino acid sequence alignment of the AID from the indicated Ca_vα₁. Residues involved in interactions with Ca_vβ are marked in red, with the most critical residues underlined. Residue numbers are indicated on both sides of the sequence.

Fig. 2

Ca_vβ crystal structure. A: crystal structure of the β₃ core in complex with the AID (PDB accession code 1VYT). This structure reveals the following regions: the NH₂ terminus (light blue, residues –59), an SH3 domain (gold, residues 60–120 and 170–175), a HOOK region (purple, residues –169), and a GK domain (green, residues 176–360). Residues 137–166 were disordered and are not included. Residues 226–244 (forming the α₄ helix of the GK domain) were disordered in this molecule but were well-resolved in another one in the same asymmetric unit. Residues 422–446 of Ca_v1.2 containing the entire AID are colored in orange. B: same structure as in A but with the BID (β₃ residues K163-T193) highlighted in dark blue. The BID spans parts of the SH3-HOOK-GK motif but is not directly involved in binding the AID. C: close-up of the interface between β₃ and AID. Some residues involved in the interactions are shown. [Adapted from Chen et al. (84)].

Fig. 3

Amino acid sequence alignment of Ca_vβ subtypes. The four included subtypes are β_1b (GenBank accession number, NP-000714), β_2a (M80545), β₃ (M88751), and β_4a (L02315). Light blue indicates the NH₂ terminus, gold the SH3 domain, purple the HOOK region, green the GK domain, and gray the COOH terminus. Secondary structure elements are indicated in the top line as arrows for β sheets and solid lines for α helices (based on the crystal structure of β₃). Residues involved in interactions with the AID are marked in red.

Fig. 4

Structural model of a partial Ca_vα₁/Ca_vβ complex on the plasma membrane. A side view and an inside-to-outside view are presented. The partial structure of Ca_vα₁ includes only the S5, P-loop, and S6 segments and is based on a Ca_vα₁ homology model developed in Stary et al. (411). IS5 is colored orange, and IS6 is red. The IS6-AID linker from Ca_v1.2 is modeled as an α-helix and is joined with IS6 at its NH₂ terminus and the AID at its COOH terminus. The structure of Ca_vβ is based on the crystal structure of the β₄ core region (84) and the NMR structure of the β₄ NH₂ terminus (451); there is no Ca_vβ COOH terminus. Since the structure of the β₄-AID complex is not available, we docked the AID to β₄ based on the crystal structure of the β₃ core-AID complex (84). The regions of Ca_vβ are color coded as in Figures 2 and 3 (NH₂ terminus in light blue, SH3 in gold, HOOK in purple, and GK in green).

Fig. 5

Human Ca_vβ splice variants. Fourteen Ca_vβ exons (13 for β₃) are color-coded based on the regions they give rise to: the NH₂ terminus (light blue), the SH3 domain (gold), the HOOK (purple), the GK domain (green), and the COOH terminus (gray). Exons are numbered, and some exons have additional letters to indicate alternatively spliced variants. The thick full and dashed lines at the very top indicate highly or somewhat conserved exons, respectively. Of the weakly conserved regions, similar exons are placed in the same column (e.g., β₁ exon 2 is homologous to β₂ exon 2A). Exons 13 and 14 of β₁ were originally designated as 13a and 13b, respectively (222). The names of splice variants are, from left to right columns, those used in this article, those proposed by Foell et al. (166), and those proposed by Yang and Berggren (486). β_2a is the only splice variant that can be palmitoylated (wave). The jagged edge (e.g., exon 6 of β_1d) indicates missing amino acids resulting from exon skipping and/or frame-shifts. Striped exons (e.g., exon 8 of β_1d) are translated with a frame shift; hence, their amino acid sequence is unrelated to the “conventional” sequence produced by that exon. **Direct submission by M. E. Williams, 1997. ***AK316045; direct submission by T. Isogai and J. Yamamoto, 2008.

Fig. 6

Amino acid sequence alignment of Ca_vβ splice variants. The 5 Ca_vβ regions, their corresponding exons, and the exon boundaries are marked. Color coding follows the same scheme as in previous figures, with the NH₂ terminus in light blue, the SH3 domain in gold, the HOOK in purple, the GK domain in green, and the COOH terminus in gray. Exon numbers are indicated in the color bar, and some exons have additional letters to indicate alternatively spliced variants. Arrows and bold amino acids mark exon boundaries. A single bold residue indicates that exon splicing occurs within its codon, whereas two bold residues indicate that splicing occurs between their codons. Shaded in black are missense sequences resulting from a frame-shift. # Indicates a premature stop codon. The GenBank accession number of each sequence is indicated at the end of the sequence, except for two sequences where the original reference is given. All sequences are from human except Cβ_4c, which is a chicken isoform. In regions where alternative splicing occurs (e.g., the NH₂ terminus of β₂), the amino acid sequence is aligned with its parent exon; thus the alignment in these regions does not necessarily indicate amino acid sequence similarity.

Fig. 6

Amino acid sequence alignment of Ca_vβ splice variants. The 5 Ca_vβ regions, their corresponding exons, and the exon boundaries are marked. Color coding follows the same scheme as in previous figures, with the NH₂ terminus in light blue, the SH3 domain in gold, the HOOK in purple, the GK domain in green, and the COOH terminus in gray. Exon numbers are indicated in the color bar, and some exons have additional letters to indicate alternatively spliced variants. Arrows and bold amino acids mark exon boundaries. A single bold residue indicates that exon splicing occurs within its codon, whereas two bold residues indicate that splicing occurs between their codons. Shaded in black are missense sequences resulting from a frame-shift. # Indicates a premature stop codon. The GenBank accession number of each sequence is indicated at the end of the sequence, except for two sequences where the original reference is given. All sequences are from human except Cβ_4c, which is a chicken isoform. In regions where alternative splicing occurs (e.g., the NH₂ terminus of β₂), the amino acid sequence is aligned with its parent exon; thus the alignment in these regions does not necessarily indicate amino acid sequence similarity.

Fig. 7

Modulation of Ca²⁺ channel gating by Ca_vβ. A: voltage dependence of activation of P/Q-type Ca²⁺ channels containing β_1b, β_2a, β₃, or β₄ or no β (β⁻). In this and all other panels, currents were recorded in cell-attached macropatches from oocytes expressing Ca_v2.1 and α₂δ, without or with the indicated β subunit. B: voltage dependence of inactivation. C: representative current traces evoked by a depolarization to ~30 mV, showing the kinetics of voltage-dependent inactivation. Currents are shown only from the first 2.5 s of a 25-s pulse. D and E: comparison of V_1/2 and t_1/2 of voltage-dependent inactivation of P/Q-type Ca²⁺ channels containing no β (β⁻) or the indicated β module: the GK domain, β core (SH3-HOOK-GK), or full-length (FL) β. V_1/2 is the membrane voltage at the midpoint of voltage-dependent inactivation, and t1/2 is the time for the current to inactivate to 50% of the peak value in C. Note the logarithmic scale of the y-axis in E. [All data from He et al. (206).]

Fig. 8

Model for the voltage dependence of G_βγ inhibition. The G_βγ-binding pocket in the holo-channel is postulated to be formed by a region of the I–II loop distal to the AID, the NH₂ terminus, and the COOH terminus of Ca_vα₁. A: WT channel: depolarization moves IS6; this movement is propagated through the rigid IS6-AID α-helix, consequently altering the conformation of the G_βγ-binding pocket and resulting in G_βγ dissociation. B: β-less channel: the AID relaxes into a random coil in the absence of Ca_vβ, uncoupling IS6 from the G_βγ-binding pocket. G_βγ can bind and inhibit the channel but does not dissociate in a voltage-dependent way. C: channel containing Ca_vβ but with a flexible IS6-AID linker: insertion of 3–7 glycine residues in the IS6-AID linker disrupts the α-helix, uncoupling IS6 from the G_βγ-binding pocket and abolishing voltage-dependent dissociation of G_βγ. [Adapted from Zhang et al. (502)].

Fig. 9

The “Ca_vβ-priming” model of Gem inhibition of surface HVA Ca²⁺ channels. Gem associates directly with Ca_vα₁ via an anchoring site in Ca_vα₁ (indicated by the purple patch). A: WT channel: binding of Ca_vβ to Ca_vα₁ induces an inhibitory site in Ca_vα₁ (indicated by the red patch), where Gem binds to induce inhibition. B: β-less channel: Gem can still associate with Ca_vα₁ via the anchoring site, but it does not inhibit the channel because Ca_vα₁ lacks the Ca_vβ-induced inhibitory site. C: WT channel with mutually noninteracting Ca_vβ and Gem: disrupting the interaction between Ca_vβ/Gem with mutations in the Ca_vβ/Gem interface does not affect Gem inhibition, since the interactions between Ca_vβ and Ca_vα₁ and between Gem and Ca_vα₁ remain intact. [Modified from Fan et al. (144).]