How do T-type calcium channels control low-threshold exocytosis?

Abstract

Low-voltage-activated T-type calcium channels act as a major pathway for calcium entry near the resting membrane potential in a wide range of neuronal cell types. Several reports have uncovered an unrecognized feature of T-type channels in the control of vesicular neurotransmitter and hormone release, a process so far thought to be mediated exclusively by high-voltage-activated calcium channels. However, the underlying molecular mechanisms linking T-type calcium channels to vesicular exocytosis have remained enigmatic. In a recent study, we have reported that Ca_v3.2 T-type channel forms a signaling complex with the neuronal Q-SNARE syntaxin-1A and SNAP-25. This interaction that relies on specific Ca_v3.2 molecular determinants, not only modulates T-type channel activity, but was also found essential to support low-threshold exocytosis upon Ca_v3.2 channel expression in MPC 9/3L-AH chromaffin cells. Overall, we have indentified an unrecognized regulation pathway of T-type calcium channels by SNARE proteins, and proposed the first molecular mechanism by which T-type channels could mediate low-threshold exocytosis.

Keywords: :

• Ca_v3.2 channel
• calcium channel
• chromaffin cell
• exocytosis
• MPC cell
• neuron
• SNAP-25
• SNARE
• syntaxin-1A
• T-type

This article refers to:

Depolarization-evoked synaptic transmission relies on the calcium (Ca²⁺)-regulated release of quantal packets of neurotransmitters following fusion of synaptic vesicles with the presynaptic plasma membrane.¹ It is well established that neuronal voltage-gated Ca²⁺ channels, by converting electrical signals into intracellular Ca²⁺ concentration elevations, play a key role in triggering evoked neurotransmitter release.^2-4 Hence, Ca²⁺ entry through high-voltage-activated (HVA) channels (N-, P/Q-, and in some extent and particular cell populations, L- and R-type) into presynaptic nerve terminals in response to action potentials supports a transient Ca²⁺ microdomain⁵ essential for synaptic exocytosis. However, the observation that some neurons can release functionally significant amounts of neurotransmitter below the threshold of action potentials⁶ questioned the possible involvement of another source of Ca²⁺ ions, independent of HVA channels activation.

In contrast to HVA channels, low-voltage-activated (LVA) T-type Ca²⁺ channels activate in response to subthreshold membrane depolarizations between -65 mV and -50 mV and thus represent an important source of Ca²⁺ entry near the resting membrane potential. Hence, besides controlling important physiological processes by regulating neuronal excitability, pacemaker activity and post-inhibitory rebound burst firing, mounting evidences from various neuronal cell types suggest an efficient role of T-type channels in fast and low-threshold exocytosis.^7-10 Until now, however, the mechanism whereby these channels support exocytosis events at the molecular level remained a mystery. In a recent study, we provided compelling evidence of the existence of a Ca_v3.2/syntaxin-1A molecular complex essential for T-type-dependent exocytosis.¹¹

In mammalian synapses, interaction of several members of the vesicle-docking / release machinery (including syntaxin-1A/1B and SNAP-25) onto a synprint (synaptic protein interaction site) domain located within the intracellular loop between domains II and III of the Ca_v2.1¹² and Ca_v2.2¹³ channels (Fig. 1A, top panel) ensures a close localization of the secretory vesicles near the Ca²⁺ source. In turn, both synatxin-1A/1B and SNAP-25 modulate calcium channel activity^14,15 to fine tune Ca²⁺ entry and synaptic strength. However, in contrast to HVA channels, all of the three T-type channel members (i.e Ca_v3.1, Ca_v3.2 and Ca_v3.3) lack the consensus synprint site, making the molecular understanding of the involvement of these channels in the exocytosis process quite difficult. Our observation that Ca_v3.2 channels associate with syntaxin-1A in central neurons prompted us to investigate the possible existence of specific Ca_v3.2 channel molecular determinants other than the consensus HVA synprint domain. Using biochemical and cellular trafficking approaches, we demonstrated that syntaxin-1A, as well as SNAP-25, interact with the C-terminal domain of Ca_v3.2 channel (Fig. 1A, lower panel). Moreover, using patch-clamp recordings performed on tsA-201 cells expressing Ca_v3.2 channels, we demonstrated that co-expression of a syntaxin-1A in its “closed” conformational state (i.e the conformation adopted by the syntaxin-1A in isolation or in interaction with Munc18)^16,17 potently decreases Ca_v3.2 channel availability by shifting the voltage-dependence of inactivation toward more hyperpolarized membrane potentials, similarly to what was previously reported for N- and P/Q-type channels.^14,15,18-20 Interestingly, this regulation was abolished upon co-expression of SNAP-25, and not observed with a constitutively “open” syntaxin-1A (i.e the conformation adopted upon its association with SNAP-25)¹⁸ (Fig. 1B). Given that syntaxin-1A undergoes a conformational switch from a “closed” to an “open” conformation during the vesicle release cycle,^16,21,22 this suggests that syntaxin-1A may be able to dynamically regulate T-type channel availability during various stages of exocytosis. Interestingly, although T-type channels utilize distinct molecular determinants to interact with SNARE proteins (the C-terminal domain vs. the classical synprint of the II-III linker), they are subjected to a similar SNARE regulation. Does this observation question the molecular mechanism by which binding of syntaxin-1A produces changes in channel gating? Earlier reports have shown that reorganization of intramolecular interactions among the main intracellular loops of Ca_v2 channels critically influence channel inactivation.^23-30 Mapping the intramolecular interactions of T-type channels along with the characterization of the minimal sequence engaged in the interaction with SNARE proteins will provide important structural information on how syntaxin-1A modulates channel gating.

Figure 1. SNARE proteins modulate high- and low-voltage-gated calcium channels via distinct molecular determinants. (A) Membrane topology of voltage-gated calcium channels highlighting the localization of the synprint site located within the intracellular linker between domains II and III of Ca_v2.1/Ca_v2.2 channels (top panel), and the “synprint like” domain of Ca_v3.x channels (bottom panel) located within the C-terminal domain of the channel. (B) Voltage-dependence of Ca_v3.x channel availability during the conformational switch of syntaxin-1A.

It is well known that direct interaction of SNARE proteins with Ca_v2.1 and Ca_v2.2 channels is critical for depolarization-evoked neurotransmitter release. Hence, disruption of the Ca²⁺ channel-SNARE proteins coupling by deletion of the synprint domain or by peptides derived from the synprint sequence, alters synaptic transmission.^31-34 We revealed that similarly to HVA channels, T-type channels-mediated exocytosis relies on a channel-SNARE protein interaction. Indeed, membrane capacitance recordings performed on MPC 9/3L-HA chromaffin cells expressing Ca_v3.2 channels revealed robust voltage-dependent exocytosis which was totally prevented by co-expression of the Ca_v3.2 C-terminal domain (i.e the synaptic protein interaction site of Ca_v3.2). Ablation of Ca_v3.2-dependent exocytosis most likely results from the specific uncoupling of the channel with SNARE proteins and not from a side alteration of the exocytosis machinery by itself because no alteration was observed when exocytosis was induced by direct intracellular Ca²⁺ elevation. Hence, we showed that similarly to HVA channels, a physical coupling between SNARE proteins and T-type channels is critical for T-type-dependent exocytosis. Considering the relative small conductance of T-type channels³⁵ and the restricted diffusion of Ca²⁺ due to the high Ca²⁺ buffering capacity of neuronal cells,³⁶ it is conceivable that this interaction allows the close localization of the vesicle-docking / release machinery in close proximity to the Ca²⁺ source in order to efficiently sense Ca²⁺ elevation. However, we cannot exclude the possibility that interaction of T-type channels with SNARE proteins could form a macromolecular complex through which channel conformational changes following membrane depolarization would work as an on/off molecular switch of secretion by controlling the ultimate conformational change of the releasing complex as previously proposed for HVA channels.^37,38 Although this concept still requires further investigation, the use on a non-conducting channel to investigate T-type-dependent secretion would definitively provide interesting information about the functional importance of T-type channel interaction with SNARE proteins.

Overall, we revealed an unrecognized regulation of low-voltage-activated T-type Ca²⁺ channels by SNARE proteins, and provide the first evidence for a molecular mechanism by which these channels could mediate low-threshold exocytosis. We revealed that, although T-type Ca²⁺ channels differ from HVA channels by their molecular constituents, they possess the same ability to functionally interact with SNARE proteins, highlighting a key evolutionary mechanism for specialized fast and spatially delimited exocytosis.

Acknowledgments

N.W is supported by a postdoctoral fellowship from Alberta Innovates Health Solutions (AIHS) and Hotchkiss Brain Institute. G.W.Z is funded by the Canadian Institutes of Health Research, is a Canada Research Chair and AIHS Scientist.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.