Regulation of Ca_V2 calcium channels by G protein coupled receptors

Abstract

Voltage gated calcium channels (Ca²⁺ channels) are key mediators of depolarization induced calcium influx into excitable cells, and thereby play pivotal roles in a wide array of physiological responses. This review focuses on the inhibition of Ca_V2 (N- and P/Q-type) Ca²⁺-channels by G protein coupled receptors (GPCRs), which exerts important autocrine/paracrine control over synaptic transmission and neuroendocrine secretion. Voltage-dependent inhibition is the most widespread mechanism, and involves direct binding of the G protein βγ dimer (Gβγ) to the α1 subunit of Ca_V2 channels. GPCRs can also recruit several other distinct mechanisms including phosphorylation, lipid signaling pathways, and channel trafficking that result in voltage-independent inhibition. Current knowledge of Gβγ-mediated inhibition is reviewed, including the molecular interactions involved, determinants of voltage-dependence, and crosstalk with other cell signaling pathways. A summary of recent developments in understanding the voltage-independent mechanisms prominent in sympathetic and sensory neurons is also included.

1. Voltage gated calcium channels

Voltage gated calcium channels (Ca²⁺ channels) are key mediators of depolarization induced calcium influx into excitable cells, which in turn mediates a wide array of physiological responses including the activation of calcium dependent enzymes, smooth muscle contraction, pacemaker activity and neurotransmitter release [1-8]. Ca²⁺ channels are also associated with a wide range of pathologies, including pain, epilepsy, migraine, cardiac arrhythmias and autism [9-14]. It is widely known that there are subtypes of Ca²⁺ channels with different pharmacological and biophysical properties, and distinct cellular and physiological functions [15-17]. In neurons, certain L-type Ca²⁺ channel isoforms are expressed at cell bodies and dendrites, and one of their key functions is the initiation of calcium dependent gene transcription events [18-22]. Other L-type channel subtypes are expressed in cochlear hair cells and photoreceptor nerve terminals where they regulate neurotransmitter release at ribbon synapses [23, 24]. T-type calcium channels are expressed in cell bodies as well as dendrites and one of their key functions is to regulate cellular excitability and neuronal firing properties [25-27], in addition to participating in secretion [28-30]. N-type and P/Q-type calcium channels are expressed at synaptic nerve terminals where their opening results in the release of neurotransmitters [1, 19, 31-34].

All Ca²⁺ channels are comprised of a pore forming Cavα1 subunit that contains the major structural features required for permeation, activation, and inactivation. The mammalian genome encodes ten different Cavα1 subunits that fall into three major families - Cav1 (L-type channels), Cav2 (N, P/Q- and R-types), and Cav3 (T-types) [17, 35]. The Ca_V1 and Ca_V2 families are high voltage activated (HVA) channels, and are heteromers comprised of a pore forming Cavα1 subunit as well as Cavα2-δ and Cavβ subunits [36-38] (Fig 1). In addition, these channels associate with calmodulin which is now considered part of the HVA channel macromolecular complex [39-44]. The Cavα1 subunit determines the Ca²⁺ channel subtype and is a large (~175-225 kDa) protein with four homologous transmembrane domains that are connected by cytoplasmic loops and bracketed by cytoplasmic N- and C-termini [37] (Fig 1). These cytoplasmic regions are key targets for second messenger regulation including protein kinases and G proteins, as we discuss here in detail. The Cavβ subunits are cytoplasmic proteins that associate with HVA α1 subunits at a highly conserved region within the domain I-II linker (termed the Alpha Interaction Domain – AID) [45-47]. These subunits are encoded by four different genes (for review see [48, 49]). The Cavα2-δ subunits are transcribed from one of four different Cavα2-δ genes, proteolytically cleaved and then reconnected via a disulfide bond (for review, see [50]). The α2 portion is located at the extracellular side of the channel, whereas the δ portion either spans the membrane or may be linked to the extracellular leaflet of the plasma membrane through a glycosylphosphatidylinositol (GPI) anchor [51]. The function of these ancillary subunits is to regulate channel properties and promote Cavα1 subunit trafficking to and stabilization at the plasma membrane [52-54] (for reviews see [48, 49, 55-57]. As we will outline below, Cavβ subunits also alter second messenger regulation the channel complex [58-61]. Finally, it should be noted that most Ca²⁺ channel subunits are subject to alternate mRNA splicing, thus greatly increasing the functional diversity of calcium channels [62-65]. Recently described RNA editing events that alter channel function add further complexity [66]. This then makes it challenging to precisely reconstitute all specific features of native calcium currents in transient expression systems. In this review, we focus on the Ca_V2 family of channels, and in particular their regulation by G protein coupled receptors.

Figure 1.

Schematic depiction of the topology and subunit composition of Ca_V2 voltage-gated Ca²⁺ channels. (A) Cartoon showing the 3D topology along with channel auxiliary subunits. The intracellular β subunit interacts through its guanylate kinase-like domain (GK) with the I-II linker of the α1 subunit (at the α-interaction domain or AID). The α2δ subunit is largely extracellular and likely GPI-anchored to the plasma membrane. (B) Topology of the pore forming α₁ subunit. Four homologous repeats (domain I through domain IV) each consist of six transmembrane spanning α-helices (S1-S6) (blue or green cylinders) and a ‘P-loop’ between S5 and S6. The S5-S6 helices and P-loop comprise the pore domain of the channel (colored green), while S1-S4 (in particular S4 that has multiple charged residues) comprises the voltage sensor (colored blue). The intracellular N- and C-termini and the cytoplasmic loops connecting domains I-IV are important for interaction with other proteins including the auxiliary β subunit, synaptic proteins, Gβγ, GPCRs, calmodulin and other Ca²⁺ binding proteins (CaBP1, VILIP). These cytoplasmic domains are also targeted by second messenger pathways including phosphorylation by PKC, CaMKII, and tyrosine kinases. Alternative splicing greatly increases the functional diversity of the channels. For example, alternative splicing of exon37 on the proximal C-terminus controls inhibition of Ca_V2.2 channels by GPCRs in sensory neurons (see section 12 for more details).

2. G protein coupled receptors and heterotrimeric G proteins

G protein coupled receptors (GPCRs) are a large family of membrane proteins encoded by almost 800 human genes, and represent an important class of therapeutic targets [67, 68]. GPCRs are characterized by an extracellular N-terminus, seven transmembrane spanning alpha helices, and an intracellular C-terminus which couples to heterotrimeric G proteins. Extracellular ligand binding to the receptor leads to activation of the G proteins and a myriad of downstream intracellular signaling cascades. In human, sixteen genes encode G protein α subunits (Gα), and these are classified into four major families: G_s, G_i, G_q, and G₁₂, in addition to transducin (Gαt) which is found in the retina. Five genes encode Gβ subunits, and twelve genes encode Gγ subunits (for reviews see[69-72]. Binding of agonist to the GPCR catalyzes the exchange of GDP to GTP on Gα causing conformational changes/dissociation of the Gα and Gβγ heterodimer [71, 73]. The liberated Gα and Gβγ are both capable of signaling to multiple downstream effectors, including voltage-gated Ca²⁺-channels as discussed in this article. Signaling is terminated by intrinsic GTPase activity of Gα and subsequent reassociation of the Gα-GDP subunit with the Gβγ heterodimer. This GTPase activity can be accelerated by a family of RGS proteins (regulator of G protein signaling) which thus influence the extent and duration of downstream events [74]. Receptor desensitization in the continued presence of agonist can also terminate signaling. Desensitization is complex, involving phosphorylation by PKA, PKC, or G protein coupled receptor kinases (GRKs) and uncoupling of the receptor from the downstream G proteins. Endocytic removal of the GPCR from the plasma membrane can also occur. GRKs recruited by Gβγ phosphorylate the C-terminus of the GPCR leading to recruitment of arrestins and the endocytic machinery [75, 76]. As discussed below (section 10), direct interaction of GPCRs and Ca²⁺ channels might result in co-internalization adding another dimension to channel modulation.

4. Inhibition of Ca_V2 channels by G protein coupled receptors

Neurotransmitter mediated inhibition of Ca²⁺ channels was first demonstrated ~30 years ago by Dunlap and Fischbach who reported that norepinephrine reduced the duration of action potentials [77] and the amplitude of I_Ca [78] in chick sensory neurons. It is now apparent that a variety of different neurotransmitters/neuromodulators acting on their cognate GPCRs inhibit I_Ca, and that this is important for controlling neurosecretion (for reviews see [79-85]). It is also known that GPCRs can recruit several distinct signaling pathways that converge on Ca²⁺ channels. The most widespread and intensively studied of these involves direct binding of Gβγ to the α1 subunit of Ca_V2 channels. As detailed below (section 5), Gβγ-mediated inhibition shifts the voltage-dependence of channel activation, is less prominent at depolarized membrane potentials, and is transiently relieved by large depolarizing voltage steps. Consequently, this mechanism is often termed voltage-dependent inhibition. GPCRs can also elicit voltage-independent inhibition of I_Ca which is mediated by several other distinct and generally less well characterized pathways including phosphorylation, lipid signaling pathways, and channel trafficking (see sections 10-12). While voltage-dependent inhibition is widespread throughout the nervous system, voltage-independent inhibition is more variable in extent and mechanism but seems particularly prominent in sensory and sympathetic neurons. In this review we first consider Gβγ-mediated inhibition, including recent developments and crosstalk with other cell signaling pathways. Then we outline some of the voltage-independent mechanisms prominent in sympathetic and sensory neurons.

5. Voltage-dependent inhibition mediated by Gβγ

Voltage-dependent inhibition primarily targets Ca_V2.1 (P/Q-type) and Ca_V2.2 (N-type) channels, although Ca_V2.3 channels are also inhibited by similar mechanisms (see section 5.4 below). The voltage-dependent nature of the inhibition was first demonstrated by Bean [86], who showed that the decrease in current amplitude was not due to a loss of channels per se, but rather a shift in the gating properties that could be overcome by strong depolarization. Several hallmarks are characteristic of this voltage-dependent mechanism: In whole cell recordings, the inhibition of peak I_Ca amplitude is diminished at depolarized membrane potentials; activation kinetics are slowed; the voltage-dependence of activation is shifted to more depolarized potentials; a conditioning prepulse to depolarized potentials relieves most of the inhibition and normalizes channel kinetics (termed prepulse relief or prepulse facilitation). Figure 2 shows an example of voltage-dependent inhibition of I_Ca. Prominent slowing of activation kinetics and prepulse relief of the inhibition is clearly seen. Voltage-dependent relief of the inhibition can also occur at least to some extent during more physiologically relevant stimuli such as high frequency trains of action potential-like waveforms [87-92]. In turn, this might contribute to short term synaptic plasticity at some synapses [93].

Figure 2.

Functional effects of voltage-dependent inhibition on Ca_V2 channels. (A) “Whole cell” patch clamp recording of I_Ca from an adrenal chromaffin cell which express purinergic P2Y autoreceptors. Application of a P2Y receptor agonist (red trace) inhibited I_Ca compared to control conditions (black trace) with the hallmark features of voltage-dependent inhibition. Peak amplitude was reduced with prominent slowing of the activation kinetics, and both of these effects were reversed by a conditioning prepulse to +100mV (green trace). (B) Voltage-dependent relief of inhibition reflects transient dissociation of Gβγ from the channel. Shown is an example of “whole cell” I_Ca recorded from recombinant Ca_V2.2 channels expressed with β1b, α2δ and Gβγ in HEK293 cells. Gβγ produced tonic inhibition of I_Ca that was reversed by a conditioning prepulse to +100 mV. The magnitude of this reversal (prepulse facilitation) diminished as the interval between prepulse and test pulse (Δ) was increased (examples shown are with Δ = 10 ms, 25 ms, and 200 ms). As illustrated by the inset cartoon, prepulse facilitation is thought to reflect dissociation of Gβγ from an inhibitory binding site on the channel at the depolarized membrane potential. Upon return to the hyperpolarized membrane potential, Gβγ rebinds to (and re-inhibits) the channel. This re-inhibition of I_Ca is monoexponential, and the rate depends on the local concentration of Gβγ.

Bean also introduced the “willing and reluctant” model to explain these functional effects [86], a framework that persists to this day [94-97]. The channels exhibit two functional gating states, “willing” and “reluctant”. In the absence of Gβγ, the “willing” state predominates, whilst binding of Gβγ favors the “reluctant” state which displays the shifts in channel gating noted above. Voltage-dependent relief of the inhibition is thought to reflect a shift of the channels from “reluctant” to “willing” due to transient dissociation of Gβγ (Fig 2B). This was supported by kinetic analyses of prepulse relief as a function of agonist or Gβγ concentration. Increasing the concentration of Gβγ did not alter the rate of relief during the prepulse, but did accelerate the rate of reinhibition following the prepulse [98-101], as expected for voltage-dependent dissociation and rebinding of Gβγ. Further investigations revealed that the kinetics of reinhibition were consistent with binding and unbinding of a single Gβγ dimer with the channel [101].

5.1. Single channel investigations

Single channel studies provided early evidence that the inhibition did not involve a diffusible second messenger. In the “cell-attached” (“on-cell”) recording configuration, bath application of agonist did not inhibit the channels whereas agonist in the patch pipette did [99, 102, 103]. This led to the conclusion that the inhibition was “direct” or “membrane delimited”. Single channel recording also directly revealed “reluctant” gating of inhibited channels. Upon membrane depolarization, the latency (delay) to first channel opening was increased during inhibition whereas there was little impact on other single channel parameters [95, 104]. As a result, the inhibited (“reluctant”) channels appeared essentially silenced, unable to open until Gβγ dissociated and the channels shifted to the “willing” state. Subsequently it has been reported that Ca_V2.2 (N-type) but not Ca_V2.1 (P/Q-type) channels can display very brief channel openings from the “reluctant” state (i.e. without Gβγ unbinding), although the probability of such events was low [96, 97].

Overall, the dominant effects of inhibition observed in all studies are the shift in activation and prolonged latency to first channel opening. The slow activation kinetics seen in whole cell recording (Fig 2) and longer latency in single channel recordings reflect the conformational changes and subsequent dissociation of Gβγ from the channel upon membrane depolarization. This diminished binding of Gβγ at depolarized potentials also results in little inhibition of whole cell I_Ca when neurotransmitter agonists are rapidly applied during a depolarizing voltage-step [105].

5.2. Alteration of gating currents by Gβγ

Further evidence for altered activation comes from recording of channel “gating currents”. Gating currents are not due to ionic flux through the channel pore, but rather reflect movement of the charged voltage-sensor domain of the channels in response to membrane potential changes. Expression of recombinant Ca_V2.2 in HEK293 cells enables recording of these gating currents in isolation as the cells lack other endogenous voltage-gated channels. G proteins were found to reduce the amplitude, and shift the voltage-dependence of gating currents to more depolarized potentials [106]. G proteins also produced a significant separation in the voltage-dependent activation of gating current and ionic current [106]. These data suggest that Gβγ binding slows movement of the voltage-sensor and uncouples this movement from opening of the activation gate. Modulation of gating currents by G proteins has also been reported in rat sympathetic neurons [107, 108].

5.3. Gβγ and channel inactivation

In addition to these dominant effects on channel activation, evidence supports the idea that Gβγ can also modulate inactivation of Ca_V2.2 channels [109, 110]. Inactivation of Ca²⁺ channels is complex and mediated by several voltage-dependent and Ca²⁺-dependent mechanisms [111-113]. The precise molecular correlates remain somewhat unclear, but fast voltage-dependent inactivation might involve a “hinged lid” type mechanism in which the intracellular loop connecting domains I and II of the α₁-subunit serves as the “inactivation gate” [112, 114] (but see [115]). The I-II loop is also an important binding site for Gβγ on the channel [116-119] (Fig 1) (see section 7 for more discussion). Therefore, it is possible that binding of Gβγ disrupts movement of this putative inactivation gate, or its interaction with other channel domains. Inactivation of Ca_V2.2 can also occur from intermediate closed state(s) of the channel favored during trains of brief repetitive stimuli [104]. If Gβγ were to reduce the probability that the channels populate this state (from which inactivation is preferred) it might reduce the cumulative inactivation throughout a stimulus train. Further investigations are needed to determine quite how G protein modulation and channel inactivation interact.

Ca²⁺-dependent inactivation is mediated by calmodulin bound to the C-terminus of the channel α1 subunit [42, 120-123]. Strong intracellular Ca²⁺ buffering (EGTA or BAPTA in the patch pipette solution) blocks Ca²⁺-dependent inactivation of Ca_V2 channels indicating that it is mediated by a “global” elevation rather than a “local” microdomain of Ca²⁺. The reduction of Ca²⁺-dependent inactivation by Gβγ [109] might therefore result from fewer channels opening and a diminished “global” Ca²⁺ signal, although more complex interactions are also possible, and direct in vitro binding of Ca²⁺-calmodulin to Gβγ has been reported [124].

5.4. Differential inhibition of Ca_V2 channels by Gβγ

Originally demonstrated for N-type channels (Ca_V2.2) in sensory and sympathetic neurons (for example, [78, 125-127]), it subsequently became clear that Ca_V2.1 (P/Q-type) channels are also modulated by Gβγ in a similar manner [128]. Initially it was thought that Ca_V2.3 channels were insensitive to G proteins [129-131], although other studies did find some degree of inhibition [132-135]. Chimeric approaches suggested that the lack of (or poor) responsiveness of Ca_V2.3 resided in several regions within the N-terminus, domain I, and I-II linker of the channels [136-138]. Subsequently it was discovered that alternative splicing of the N-terminus conferred G protein sensitivity to the channels [139]. Truncation of 50 amino acids abolished inhibition, whereas a splice variant with full length N-terminus did display inhibition, albeit to a lesser extent than Ca_V2.2 channels [139]. While inhibition of Ca_V2.3 channels can occur (depending on splice variation), it is generally to a lesser extent and remains less well understood than for Ca_V2.1 and Ca_V2.2 channels. In part this might be due to difficulty in isolating these channels in neuronal cell types. The focus on Ca_V2.1 and Ca_V2.2 channels is also driven by their prominence in triggering neurotransmitter release, and most of the following discussion revolves around those two channels.

Although the basic mechanism of inhibition is similar for Ca_V2.1 and Ca_V2.2 (direct binding of Gβγ to the channel), subtle differences have emerged. As noted above, single channel recording showed that Ca_V2.2 but not Ca_V2.1 channels display very brief duration, low probability “reluctant openings” [96, 97]. Differences are also apparent with macroscopic (whole-cell) recordings: activation of GPCRs or expression of Gβγ reduces the peak amplitude of I_Ca to a significantly greater extent for Ca_V2.2 than Ca_V2.1 [129, 140, 141]. Moreover, trains of action potential-like stimuli reverse a greater proportion of Ca_V2.1 inhibition than Ca_V2.2 inhibition [92]. These effects can be explained by differences in the affinity of Gβγ binding to the channels. The apparent affinity of Gβγ for the channel can be inferred from prepulse relief and re-inhibition experiments, and is quite similar for the two channels at hyperpolarized or very depolarized potentials. However, at moderately depolarized potentials (< +30mV), within the physiologically relevant range of action potentials, there is a significant divergence in the affinity of Gβγ binding to the two channel types [96]. Subtle differences in binding of Gβγ to the channels is also suggested when comparing the inhibition produced by different Gβ subunits (Gβ_1-5), all paired with the same Gγ₂ subunit. Such experiments revealed a different rank order of inhibition for Ca_V2.1 and Ca_V2.2 channels [142]. Point mutations on the Gβ₁ subunit also have distinct effects on the inhibition of Ca_V2.1 and Ca_V2.2 channels [143].

It would appear that subtle differences in the binding affinity of Gβγ to the Ca_V2.1 and Ca_V2.2 channels results in differential inhibition: Ca_V2.2 I_Ca is inhibited to greater extent and this inhibition is more resistant to reversal by high frequency bursts of action potentials. The relative expression level of the two channel types varies between neurons, and even between neighboring synapses arising from the same neuron. Therefore, differential inhibition of Ca_V2.1 and Ca_V2.2 could lead to cell and/or synapse specific neuromodulation by GPCRs. Functional differences might also arise from variable interactions or crosstalk with other signaling pathways such as PKC (see section 9).

6. Structural determinants on Gβγ that govern modulation of Ca_V2 channels

Gβγ is thought to be an obligate heterodimer and there are several high resolution crystal structures in isolation or bound to interacting proteins including Gα, GRK2, and phosducin [144-148]. Figure 3 shows a rendering of the heterotrimer (Gα_iβ₁γ₂) (Fig 3A), and heterodimer (the Gβ₁γ_{2 )} (Fig 3B) based on the structure reported by Wall et al [144] (PDB ID: 1GP2). Gβ adopts a seven blade β-propeller structure with an α-helical N-terminal domain that binds to the α-helical N-terminus of the Gγ subunit (Fig 3B). In the heterotrimeric complex, Gα interacts with multiple residues on the top face of Gβ and the side aspect of propeller blade 1 (Fig 3A). Gβγ interacts with multiple downstream effectors and mutagenesis approaches have been used to map the interaction sites important for binding to these targets. Many effectors bind to a protein interaction “hot spot” on the surface of Gβ that interacts with Gα, with overlapping subsets of residues involved in binding to different effectors [149]. A number of residues identified in mutagenesis studies to contribute to inhibition of Ca²⁺ channels are highlighted in figure 4. Most of these are on the Gα interacting surface (Fig 3C) and are masked when Gα is present [85, 109, 150-152], although residues on the reverse face of Gβ₁ have also been implicated [152-154] (Fig 3D). Also of note, Asn³⁵ and Asn³⁶ on Gβ₁ mediate the ability of PKC to antagonize inhibition of Ca_V2.2 [153]. Thr⁴²² on the rat Ca_V2.2 I-II linker has been identified as the phosphorylation site for PKC that mediates this effect [155], so it is tempting to speculate that this region of the channel and Gβγ come into close proximity with one another (see section 7).

Figure 3.

Structural determinants on Gβγ that govern modulation of Ca_V2 channels. (A, B) Ribbon diagram renderings of the heterotrimeric G protein structure in panel A, and the Gβγ dimer in panel B (Gα_i - green; Gβ₁ - red; and Gγ₂ blue). Gβ adopts a seven blade β-propeller structure with an α-helical N-terminal domain that binds to the α-helical N-terminus of Gγ. Gα interacts with multiple residues on the top face of Gβ and the side aspect of propeller blade 1. Many effectors bind to a protein interaction “hot spot” on the surface of Gβ that is masked by Gα in the heterotrimer. (C, D) Molecular surface rendering of the Gβγ dimer (Gβ - red; Gγ - blue). Panel C shows the Gα interacting face of Gβγ, and panel D is rotated ~180° to show opposite face of Gβγ. Residues marked in yellow have been reported to disrupt inhibition of Ca_V2 channels. Residues marked in green are involved in crosstalk between Gβ1 and PKC phosphorylation of Ca_V2.2. Molecular graphics images based on data reported by Wall et al [144] (PDB ID: 1GP2) were produced using the UCSF Chimera package [256, 257] from the Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco.

Figure 4.

Model depicting the molecular interactions that underlie Gβγ-mediated inhibition of Ca_V2 channels. Panels A and B (upper three images) depict a channel with a Ca_Vβ subunit, while panel C (lower images) depicts the situation in which the Ca_Vβ subunit is absent. Currently, data suggest the binding site for Gβγ is comprised from multiple sites on the N-terminus, I-II linker, and probably C-terminus of the channel. Binding of Gβγ causes a conformational shift that promotes interaction of the N-terminus “inhibitory module” with the initial one-third of the I-II-linker (panel Aii). This (and perhaps other interactions) shifts gating charge movement to more depolarized potentials and uncouples voltage-sensor movement from channel activation. Strong membrane depolarization (panel B) leads to conformational changes that result in unbinding of Gβγ and loss of interaction between the N-terminus and I-II linker. This depends upon binding of a Ca_Vβ subunit to the AID on the I-II linker that induces a rigid α-helical connection to the upstream IS6 region of the pore and voltage-sensor. In the absence of Ca_Vβ subunit binding, inhibition still occurs (panel Ci) but cannot be reversed by strong depolarization (panel Cii).

Another study reported that a peptide mimicking the N-terminal 25 amino acids of Gβ₂ reduced inhibition of Ca_V2.1 [156]. The Gβ N-terminal peptide disrupted FRET interaction between the Gβ₂ and Gγ₃ subunits suggesting a conformational shift or reorientation of the heterodimer that could disrupt interaction with the channels. A few studies have also shown that the subtype of Gγ within the Gβγ heterodimer can influence the extent of inhibition, with Gγ₂ generally eliciting greater inhibition than Gγ₁, Gγ₃ or Gγ₁₃ [157, 158]. The molecular basis for why the Gγ subtype influences inhibition of I_Ca is not clear, but it is interesting to note that the II-III linker (of the channel α1 subunit) contains a G-gamma-like (GGL) domain [159].

7. Structural determinants on the channel α1 subunit that govern modulation by Gβγ

Although there is currently no crystal structure for voltage-gated calcium channels that could be used to visualize their interactions with G proteins, site directed mutagenesis, chimeric, and biochemical approaches have been used to elucidate channel structural determinants involved in modulation. The first investigations involved chimeras between Cav2.1 and Cav2.2 channels [140]. These chimeras were expressed in Xenopus oocytes and their sensitivities to G proteins assessed via two electrode voltage clamp. These experiments identified domain I as a key determinant of G protein inhibition, along with the C-terminus of the channel. Subsequent biochemical studies using in vitro translated Gβγ subunits revealed two spatially distinct regions on the I-II linker of CaV2.1 as possible Gβγ targets [117]. The existence of two separate Gβγ binding domains in the domain I-II linker was also observed in functional assays. Zamponi et al. [160] showed that intracellular dialysis of tsA-201 cells with ~20 amino acid peptides directed against different regions of the I-II linker of both Cav2.1 and Cav2.2 channels prevented the ability of exogenously delivered Gβγ subunits to mediate voltage dependent inhibition of the channels. The first site contains a QXXER consensus sequence (QQIER in all three Ca_V2 family members) found in other Gβγ binding partners. This site also overlapped partially with the putative Cavβ subunit binding domain on the channel (the AID). Subsequent co-crystal structures of the Cavβ subunit bound to its interaction site on the isolated domain I-II linker revealed that only part of the 20 amino acid stretch forming the putative Gβγ interaction site is likely to be accessible in the presence of a bound Cavβ subunit [46]. This may suggest two possibilities: Either the Cavβ subunit partially dissociates from regions involved in Gβγ binding, or alternatively Gβγ interacts with those residues that remain exposed after Cavβ docking.

Further support for the involvement of the I-II linker came from scanning mutagenesis of the amino acids in each of the two binding regions in rat Ca_V2.2 channels [161]. Mutation of two residues (Arg³⁷⁶ and Val⁴¹⁶ to alanine) out of thirty tested significantly reduced the magnitude of voltage-dependent inhibition while mutation of Arg³⁷⁶ to phenylalanine increased inhibition. Irrespective of the precise nature of the Gβγ interaction on the domain I-II linker, this general region has been implicated as being important for functional channel inhibition by a number of other groups [116, 131, 136]. These studies contrast with work from Qin and colleagues [134] whose data implicated the C-terminus rather than the domain I-II linker as the critical element for G protein modulation. While likely playing an auxiliary role, the C-terminus region does not appear to be essential for N-type channel inhibition as large parts can be deleted with only small consequences on the extent of receptor mediated voltage-dependent modulation [155, 162, 163]

Several other groups attributed an important role to the N-terminus of the channel based on site directed mutagenesis work [136, 138, 151, 164]. The Dolphin lab identified the N-terminal 55 amino acids of Ca_V2.2, and in particular an eleven amino acid stretch (45-55) that is predicted to form an α-helix [165], to be critical for Gβγ-mediated inhibition of the channels. The Yue group demonstrated direct interaction of the N-terminus with Gβγ [151] and that the N-terminus (residues 56-95) also binds directly to the I-II linker. Thus, the N-terminus contributes both to binding of Gβγ, and as an “inhibitory module” which binds the I-II linker to perhaps mediate the shift from willing to reluctant gating states. Finally, a recent study revealed that a point mutation (S218L) in the domain I S4-S5 linker of Cav2.1 that is found in patients with familial hemiplegic migraine (FHM) facilitates recovery of the channels from Gβγ inhibition, perhaps by facilitating the dissociation of the G protein dimer [166]. Two other FHM mutations (R192Q, Y1245C) have also been reported to diminish Gβγ-mediated inhibition [167, 168].

Taken together, several sites on both the Ca_V2 α₁ subunit and the Gβγ heterodimer have been implicated in voltage-dependent inhibition. On the Cav2 α1 subunit, the domain I-II linker and N-terminus are essential structural elements (Fig 4). Ultimately crystal structure data will be needed to precisely determine how G protein subunits interact with these channel loci in the presence and the absence of the Cavβ subunit.

8. Contribution of the Ca_Vβ subunit to voltage-dependent inhibition

The subtype of Ca_Vβ can influence the extent and kinetics of Gβγ-mediated inhibition and this depends on the subtype of Gβ involved [169, 170]. However, the precise role of Ca_Vβ subunits in voltage-dependent inhibition of I_Ca has been unclear (for reviews see [49, 79]). Overlapping binding sites for the two proteins have been identified on the I-II linker, and one fundamental question that arose was whether Ca_Vβ and Gβγ can bind to the channel at the same time, or whether they compete in a mutually exclusive manner. Seemingly contradictory data including FRET analyses suggested either competition [171] or synergistic binding [172]. Some of this confusion might stem from endogenous Ca_Vβ subunits found in some heterologous expression systems (including Xenopus oocytes), or confounding shifts in the voltage-dependence of activation by some Ca_Vβ subunits (see [79]). Evidence from the Dolphin and Yang labs outlined below now suggest that both proteins can interact with the channel simultaneously, and that binding of the Ca_Vβ subunit is required to confer voltage-dependent reversal to Gβγ-mediated inhibition (Fig 4) [173-176].

The Dolphin lab introduced a mutation (W391A) into the AID on the I-II linker of Ca_V2.2 channels which reduces CaVβ subunit binding affinity by ~1000 fold. [174]. While the extent of Gβγ-mediated inhibition was similar for mutant (W391A) and wild-type channels, prepulse reversal of the inhibition was almost abolished in the mutant. Expression of wild type Ca_V2.2 along with α₂δ but without Ca_Vβ resulted in similar findings, and the voltage-independent inhibition in the absence of the Ca_Vβ was blocked by overexpression of transducin which acts to scavenge free Gβγ subunits [175]. Thus, in the absence of Ca_Vβ binding to the I-II linker, Gβγ-mediated inhibition of the channels was still present but could no longer be reversed in a voltage-dependent manner. The experiments outlined above used the β_1b subunit, but when β_2a was expressed with the W391A channels voltage-dependent relief of the Gβγ mediated inhibition was restored. Unlike β_1b, the β_2a subunit is palmitoylated at two N-terminal cysteine residues, and mutation of these residues led to loss of voltage-dependent relief (i.e. the data resembled β_1b). The authors proposed that palmitoylation increased the local plasma membrane concentration of β2a such that low affinity interaction with α₁ could still take place and permit voltage-dependent relief of the inhibition.

The Yang lab came to similar conclusions for Ca_V2.1 channels [176]. In this case the authors mutated Ca_Vβ to reduce the affinity for the AID. The channels were expressed in Xenopus oocytes and macroscopic currents were recorded from giant inside-out patches that contained many channels. Washing the cytoplasmic face of the patches resulted in dissociation of Ca_Vβ (due to the reduced binding affinity of the mutant), and this was confirmed by the expected shifts in channel kinetics compared to wild type. In these channels lacking Ca_Vβ, purified Gβγ still inhibited the currents but prepulse reversal was abolished.

The Ca_Vβ subunit consists of SH3 and GK domains separated by a variable HOOK region [48, 49]. Binding to the AID on the I-II linker of α₁ is mediated by the GK domain, although interaction between the SH3 and HOOK domains elsewhere on the α₁ subunit might also modulate functional properties. In terms of Gβγ effects, voltage-dependent reversal was restored even by binding of the isolated GK domain of Ca_Vβ to the AID [175, 176]. In the absence of such binding the AID adopts a random coil, but the presence of Ca_Vβ induces an α-helical conformation that extends back to the interface with IS6 [46, 47, 177, 178]. The Yang lab introduced seven glycines between the AID and IS6 to disrupt this α-helical structure and found that this prevented the ability of Ca_Vβ to confer voltage-dependence to the inhibition [176]. Conversely, introducing seven alanines (not expected to disrupt the α-helix) maintained the ability of Ca_Vβ to confer voltage-dependence to Gβγ-mediated inhibition. It is possible that binding of Ca_Vβ to the AID induces a rigid α-helical link with domain IS6, and this transmits movement of the voltage-sensor and activation gate (including IS6) to the I-II linker to alter the Gβγ binding pocket at depolarized potentials. It is also worth noting that Gβγ-mediated inhibition was still present in both channel types lacking Ca_Vβ, and in the Ca_V2.1 channels containing the seven glycine insert [175, 176]. Apparently the rigid α-helical link to the upstream activation gate and voltage-sensor is not required per se to transduce binding of Gβγ into functional inhibition. Ca_Vβ might also influence Gβγ-mediated inhibition in other ways. For example, deletion of the HOOK domain promoted tonic inhibition of Ca_V2.2 channels, perhaps due to increased affinity for the basal level of free Gβγ in the cells [175].

9. Crosstalk between N-type channels, Gβγ, kinases and synaptic proteins

9.1. Protein kinase C

Most cell signaling events do not occur in isolation, but instead in an integrated fashion. G protein regulation of voltage-gated calcium channels is no exception. This is exemplified by the modulation of voltage dependent Gβγ inhibition of N-type channels by protein kinase C ( PKC). In peripheral neurons, activation of PKC was shown to reduce the extent of subsequent G protein modulation by a number of different receptor pathways, including GABA-B, adenosine and muscarinic receptors [179-181]. Such an interference with G protein inhibition could be due to PKC dependent phosphorylation of the G protein interaction site on the channel, the G protein coupled receptor, or the G protein itself. The first hint supporting the first mechanism came from experiments showing that in vitro phosphorylated domain I-II linker peptides could no longer effectively interact with Gβγ peptides [160]. Subsequent work showed that a threonine residue within the putative I-II linker Gβγ interaction site was responsible for this effect. When phosphorylated, or substituted for glutamic acid, this residue destabilizes the interaction of the channel with Gβγ, and its substitution for alanine precludes the antagonistic effects of PKC [155]. Interestingly, only Gβ1 mediated signaling (not other Gβ isoforms) was subject to this type of PKC crosstalk [182], and this was attributed to a single locus unique to Gβ1 [153] (Fig 3C). This observation suggests that activation of Gq-coupled receptors can modulate signaling of certain types of Gβ1 linked receptors to N-type calcium channels. It should also be noted that PKC activation not only results in antagonistic effects on Gβγ-mediated inhibition but, depending on N-type channel splice variant, can also promote direct enhancement of current activity [155]. This is mediated by phosphorylation of both the above noted threonine residue and an adjacent serine and adds further complexity to the PKC-G protein signaling crosstalk.

9.2. Synaptic proteins

The two types of calcium channels that are most susceptible to the effects of Gβγ also control neurotransmitter release at CNS synapses [1]. Both Cav2.1 and Cav2.2 channels physically associate with proteins that are involved in synaptic vesicle release, such as syntaxin 1A and SNAP25. These SNARE proteins bind directly to a synaptic protein interaction (synprint) site on the II-III linker (Fig 1) which serves to bring the channels into close proximity of the synaptic vesicle release sites [183-189]. RIMs (rab3 interacting molecules) have also emerged as important organizers of the presynaptic active zone [190], and bind Ca_V2 channels both directly and through RIM binding proteins to control their density and localization at release sites [191-195]. Binding of syntaxin 1A to both Cav2.1 and Cav2.2 also results in a hyperpolarizing shift in the voltage dependence of channel inactivation [185, 196-198] (for review see [199]). In addition to this effect on channel gating, syntaxin 1A modulates G protein regulation of the channels. Coexpression of syntaxin 1A with N-type channels in tsA-201 cells induces tonic inhibition mediated by Gβγ [200]. Syntaxin 1A physically associates with Gβγ at a site distinct from that involved in binding to the N-type channel [198, 201], suggesting the possibility that syntaxin 1A serves to colocalize the channels and Gβγ to ultimately promote a form of tonic inhibition. In contrast, syntaxin 1B does not mediate such an effect, even though it is capable of binding to both the channel and Gβγ [202]. This may suggest that the spatial orientation of the syntaxin/G protein complex relative to the N-type channel complex is critical for functional modulation. Gβγ interaction with SNARE proteins might also serve to directly regulate neurotransmitter release in both lamprey and mammalian neurons and neuroendocrine cells [203-206].

Several other types of synaptic proteins have been shown to alter G protein regulation of N-and P/Q-type channels. Cysteine string protein (CSP) interacts with G proteins and the synprint site and mediates an effect similar to that seen with syntaxin 1A [207]. In addition, CSP appears to stimulate Gα subunit activity by promoting the exchange of GTP for GDP in a receptor independent manner [208]. In contrast to these enhancing effects of syntaxin 1A and CSP, coexpression of Rim1 with Cav2.2 in HEK293 cells promotes “deinhibition” (recovery from Gβγ-mediated inhibition during depolarization) in addition to substantially slowing channel inactivation [209].

9.3. Calcium channel γ subunits

Another protein of note is stargazin, a member of the calcium channel γ subunit family. Skeletal muscle Ca_V1.1 channels have been shown to associate with a γ1 subunit in addition to β and α2δ subunits. Several neuronal γ subunits have been identified although it remains uncertain that these constitute bona fide channel subunits. Indeed the γ2 isoform (also called stargazin) and related proteins (γ3-7) associate with and modulate glutamatergic AMPA receptors [210]. However, it has been shown that stargazin can bind Gβγ in vitro, and acts to scavenge Gβγ and reduce inhibition of Ca_V2.2 channels in Xenopus oocytes [211]. Altogether, these findings highlight the notion that Gβγ modulation of calcium channels does not occur in isolation, but is tightly controlled by a wide range of cellular processes and signalling pathways.

10. Direct GPCR/N-type calcium channel interactions

Efficient signaling necessitates close proximity between GPCRs and effectors such as ion channels. This can be accomplished through the formation of large macromolecular signaling complexes between receptors, channels, G proteins, and kinase anchoring proteins [212-215]. In addition, a physical association between receptors and channels provides for a possible mechanism by which receptors can control channel function in an agonist independent manner. This was first shown by Kitano and colleagues [216] who identified a physical association of Cav2.1 channels with metabotropic glutamate receptors that results in altered P/Q-type channel function. For N-type channels, the formation of a signaling complex between Cav2.2 and the NOP (a.k.a. nociceptin) receptor was demonstrated in dorsal root ganglion neurons, and shown to promote tonic voltage-dependent modulation in the absence of receptor ligand, presumably reflecting constitutive receptor activity [217]. Similar observations have been reported for δ- and μ-opioid receptors coexpressed with N-type channels in tsA-201 cells [218, 219].

Association of channels with receptors also provides for an additional level of control through regulation of channel density in the plasma membrane. NOP receptors coexpressed with N-type channels not only promote the cell surface expression of the channels (Fig 5A), but also trigger an agonist mediated co-internalization of the channel/receptor complex into lysosomes [219, 220], thus giving rise to a new form of voltage-independent inhibition (Fig 5B). The extent to which this occurs in neurons is up to some debate. While imaging studies show a clear NOP receptor mediated internalization of channels in cultured DRG neurons, and reduced calcium entry in response to prolonged activation of receptors [220], there do not appear to be clear effects on whole cell current densities in nociceptin treated neurons [221]. It is possible that receptor mediated internalization is offset by kinase pathways that augment the activities of channels remaining in the plasma membrane. D1 and D2 dopamine receptors also associate with N-type calcium channels [222, 223], but while NOP receptors and Cav2.2 channels interact via their C-termini, D1 and D2 receptors also interact with other regions of the Cav2.2 channel α1 subunit such as the domain II-III linker. As with the NOP receptor, D1 or D2 receptor coxpression facilitates trafficking of the channels to the plasma membrane, and allows for receptor-channel co-internalization. In prefrontal cortex neurons, the D1 receptor appears to target N-type channels to dendritric sites [222]. It is likely that other types of GPCR may form complexes with N-type and perhaps P/Q-type calcium channels, however this will need to be confirmed experimentally. Altogether, the formation of macromolecular signaling complexes between receptors and channels provides for previously unrecognized means for controlling channel activity/density.

Figure 5.

Trafficking-mediated modulation of Ca_V2.2 channels due to direct interaction with GPCRs. (A) Nociceptin receptors (NOP) interact directly with the Cav2.2 α1 subunit via their C-termini. D1 and D2 receptors also interact with additional regions of the α1 subunit such as the domain II-III linker. Coexpression of these GPCRs with Ca_V2.2 facilitates trafficking of the channels to the plasma membrane, and the D1 receptor appears to target N-type channels to dendritric sites in prefrontal cortex [222]. (B) Prolonged NOP agonist application has been reported to promote co-internalization of the receptor/channel complex into lysosomes in cultured sensory neurons, thus giving rise to a new form of voltage-independent inhibition [220] (but see [221]).

11. Voltage-independent inhibition of Ca_V2 channels by Gq-coupled GPCRs

Sympathetic neurons have been used extensively to investigate modulation of Ca_V2.2 (N-type) channels, and at least two distinct pathways have been identified: the “fast” pathway, mediated by pertussis toxin sensitive Gi/o-coupled GPCRs, is due to direct, voltage-dependent inhibition by Gβγ; a “slow”, voltage-independent pathway takes tens of seconds to develop, is mediated by Gq-coupled GPCRs, involves a diffusible second messenger, and is sensitive to intracellular [BAPTA] (for reviews see [224, 225]). Typical Gq-coupled signaling pathways downstream of phospholipase Cβ including IP3/Ca²⁺, diacylglycerol, and PKC were shown not to mediate the inhibition, and the pathway remained elusive for quite some time. In the past decade evidence has mounted implicating depletion of plasma membrane PIP₂ and/or generation of arachidonic acid as possible mediators of this inhibition [226-229]. Here we outline the proposed mechanisms along with some recent developments.

PIP₂ is required for a variety of ion channels to function (for reviews see [229-232]). The first evidence that this included Ca²⁺ channels was the demonstration that time-dependent “rundown” (loss) of Ca_V2.1 channel activity in excised membrane patches was slowed by application of PIP₂ and accelerated by depleting or sequestering PIP₂[233]. Similar effects were subsequently reported for N-type (Ca_V2.2) channels, along with evidence that the “slow” inhibition by Gq-coupled GPCRs in sympathetic neurons was due to phospholipase C mediated PIP₂ hydrolysis [234]. For example, inhibition by muscarinic receptors was blunted by including PIP₂ in the patch pipette, whereas recovery from inhibition was slowed by blocking PI-4 kinases which replenish the depleted PIP₂. The overall picture that has emerged is that PIP₂ is required for channels to open in response to membrane potential changes. This may involve dynamic low affinity interaction of PIP₂ with the channels and perhaps additional higher affinity binding to a distinct channel domain. It has been postulated that such interactions might “crosslink” hydrophobic and hydrophilic domains and favor protein conformations conducive to active channel states. Similar to Gq-mediated modulation of M-type potassium channels [226, 230], this model proposes that depletion of local PIP₂ by phospholipase C mediated hydrolysis removes this permissive interaction and is both necessary and sufficient to inhibit channel activity.

An alternative, although related, lipid signaling pathway has been proposed by the Rittenhouse lab [227] who reported that arachidonic acid elicits bidirectional modulation of N-type channels; I_Ca was enhanced at relatively hyperpolarized test potentials and inhibited at more depolarized potentials [235, 236]. The enhancement seems to involve extracellular actions of arachidonic acid [237], whereas the inhibition is mediated at the cytoplasmic face of the membrane. Arachidonic acid can be produced either by the action of phospholipase A2 on PIP₂ and other membrane phospholipids, or by the action of diacylglycerol-lipase on diacylglycerol [227]. It is postulated that activation of these lipases by muscarinic receptors cleaves PIP₂ and generates arachidonic acid which binds to the channel. This binding has the opposite effect to PIP₂ such that it stabilizes closed/inactivated states of the channel and thus leads to inhibition. The involvement of arachidonic acid is a matter of some debate, in part due to conflicting reports on the ability of DAG-lipase inhibitors to block channel inhibition [226, 227, 234, 236].

A recent study from the Hille lab has provided evidence in support of the PIP₂ depletion model [238]. To avoid downstream and parallel signaling pathways subsequent to Gq-coupled receptor activation, the authors used controlled activation of exogenous polyphosphoinositide 5-phosphatases which convert PIP₂ into PI(4)P [239, 240]. In one approach, cells were transfected with a voltage-sensitive phosphatase (VSP) that enabled rapid (~1s) and reversible depletion of PIP₂. Another approach involved chemical dimerization to translocate transfected yeast INP54p 5-phosphatase to the membrane and irreversibly deplete PIP₂. In both cases I_Ca was inhibited by PIP₂ depletion, and the rate/extent of recovery from this inhibition tracked the rate/extent of PIP₂ resynthesis [238]. These data support the idea that depletion of PIP₂ in of itself is sufficient to inhibit I_Ca. However, it was noted that the magnitude of I_Ca inhibition was less than that produced by muscarinic receptors, even though the predicted depletion of PIP₂ is comparable. Therefore, it is possible that another signal, perhaps arachidonic acid, also contributes to the Gq-mediated inhibition, perhaps through synergistic actions with PIP₂ depletion [226, 227].

11.1. Ca_Vβ and intracellular Ca²⁺ modulate Gq-mediated inhibition

Notably, it has been reported that the Ca_V β2a subunit opposes inhibition of I_Ca by arachidonic acid or PIP₂ depletion. The β2a subunit is palmitoylated at two N-terminal residues and it is this lipidation that diminishes the inhibition. Indeed, N-type channels containing β2a are enhanced rather than inhibited by Gq-coupled GPCRs and arachidonic acid [60, 227, 241]. Rittenhouse and colleagues postulated that the palmitoyl groups interact directly with the α1 subunit of the channel and thereby mask an inhibitory binding site for arachidonic acid. Inhibition of I_Ca by PIP₂ depletion (using a voltage-sensitive phosphatase) is also diminished in channels containing a palmitoylated β2a subunit [61]. It is speculated that the palmitoyl groups of Ca_Vβ2a, the lipid tail of PIP₂, and arachidonic acid compete for binding to a site(s) on the α1 subunit of the channel. Binding of the palmitoyl groups or lipid tails of PIP₂ favors active channel conformations whereas loss of this interaction and/or binding of arachidonic acid favors closed/inactive channel conformations.

Highlighting the complexity of neuronal Ca²⁺ channel regulation, it is noteworthy that even in the same cell, not all Gq-coupled GPCRs elicit inhibition of I_Ca [234]. This correlates with the ability of the different receptors to elicit significant IP₃ mediated release of intracellular Ca²⁺ stores which likely relates to the proximity of the GPCR, IP3 receptors, and other components of a macromolecular signaling complex [226, 242]. In turn, such Ca²⁺ elevations are postulated to promote phosphatidylinositol 4-kinase activity thereby preventing local PIP₂ depletion and inhibition of I_Ca [226, 234, 243, 244]. Thus, the extent of this voltage-independent inhibition depends on the subunit composition of the channels (Ca_Vβ isoform), and colocalization of GPCRs and various phosphoinositide and Ca²⁺ signaling components in macromolecular signaling complexes. Adding further to this complexity, another type of “fast”, voltage-independent inhibition of N-type I_Ca mediated by a distinct pathway(s) perhaps involving both Gα and Gβγ signaling has also been described in these same sympathetic neurons [245, 246].

12. Kinase-mediated, voltage-independent inhibition of Ca_V2 channels in sensory neurons

Voltage-independent inhibition that appears to involve channel phosphorylation has also been described. For example, PKC has been implicated in the inhibition of N-type I_Ca in chick sensory neurons [247, 248] In frog and mammalian neurons PKC can also potentiate I_Ca, might target multiple phosphorylation sites with opposing actions, antagonize Gβγ mediated inhibition, or modulate channel trafficking (see sections 9, 10) [155, 180, 181, 249, 250]. It was also reported that rapid activation of a tyrosine kinase by GABA_B receptors resulted in voltage-independent inhibition of N-type I_Ca in chick sensory neurons [251]. More recently, the Lipscombe lab demonstrated that manifestation of this tyrosine kinase mediated inhibition in mammalian neurons is controlled by alternative splicing of the Ca_V2.2 C-terminus (Fig 6) [252, 253]. There are two mutually exclusive forms of exon 37 (e37a and e37b). Gβγ-mediated, voltage-dependent inhibition is identical in recombinant channels containing either e37a or e37b. However, channels containing exon 37a are also inhibited by another voltage-independent pathway. This second pathway involves rapid activation of pp60c-src tyrosine kinase and requires a tyrosine residue (Y1747) present in exon 37a that is replaced by phenylalanine in exon 37b. Of particular note, expression of exon 37a is restricted to dorsal root ganglia, and preferentially expressed in capsaicin sensitive, nociceptive neurons [254]. The gene encoding chicken Ca_V2.2 only has one exon 37 which is similar to e37a and includes a tyrosine residue [252]. This then explains the restriction of this pathway to nociceptive neurons in mammals and its prevalence in chick neurons. This expression pattern also suggests that the e37a splice variant might be tailored to play a role in pain transmission [255]. Recently, using an exon replacement strategy in mice, it was shown that basal nociceptive transmission was unaltered by loss of e37a, but the analgesic effects of intrathecal morphine were diminished [253].

Figure 6.

Alternative splicing of Ca_V2.2 controls voltage-independent inhibition of N-type I_Ca in sensory neurons. Two mutually exclusive forms of exon 37 encode the proximal C-terminus of Ca_V2.2. Expression of exon 37a is restricted to dorsal root ganglia, preferentially in nociceptive neurons, while exon 37b is widely expressed throughout the nervous system [254]. Gβγ-mediated, voltage-dependent inhibition of I_Ca is identical in channels containing either isoform of exon 37. An additional Gα-mediated, voltage-independent pathway involving pp60c-src tyrosine kinase inhibits channels containing exon 37a but not exon 37b. Thus alternative splicing of Ca_V2.2 results in cell-type specific alteration in the magnitude and mechanisms of GPCR-mediated inhibition.

13. Concluding remarks

In this review we have highlighted the complex inhibition of Ca_V2 channels by G protein coupled receptors. Voltage-dependent inhibition, mediated by direct binding of Gβγ to the Ca²⁺ channel α1 subunit, is the most common and best understood mechanism. Membrane potential, firing patterns, channel subunit composition/splice variants, and Gβγ heterodimer composition all modulate the extent and/or kinetics of voltage-dependent inhibition. Although less well understood and perhaps less widespread, there are also several mechanisms leading to voltage-independent inhibition of Ca_V2 channels. These include direct interaction with GPCRs, inhibition through lipid signaling pathways, and channel phosphorylation. Ca_V2 channels are also subject to a variety of other regulatory mechanisms, notably Ca²⁺-dependent feedback (both inactivation and facilitation). Thus, GPCRs in combination with Ca²⁺ channels sense and integrate a complex array of inputs in order to fine tune the spatiotemporal aspects of Ca²⁺ entry that play such pivotal roles in cellular physiology and synaptic transmission.

Highlights.

• Ca_V2 channels play pivotal roles in neurotransmitter and hormone release

• G protein coupled receptors orchestrate precise control of Ca_V2 channels

• Voltage-dependent inhibition is mediated by direct binding of Gβγ to the channels

• Voltage-independent inhibition is mediated by several other distinct pathways

• Current understanding of these important mechanisms is provided in this review