Significance
How membrane potential drives conformational change at the voltage-sensing domain (VSD) and regulates channel gating is a central theme for voltage-gated ion channels. To elucidate voltage-gating mechanisms, one needs to capture VSD structures in both the activated state and resting state, the latter of which is difficult to obtain because VSDs tend to adopt an activated state in the absence of membrane potential in most in vitro experiments. Here, we determined AtTPC1 structures in a closed conformation with a resting VSDII and an unbound EF-hand domain and in a partially open conformation with an activated VSDII and a Ca2+-activated EF-hand domain, elucidating the structural mechanisms of voltage gating, cytosolic Ca2+ activation, and their coupling in AtTPC1.
Keywords: TPC, voltage gating, Ca2+ activation, coupling
Abstract
Arabidopsis thaliana two-pore channel AtTPC1 is a voltage-gated, Ca2+-modulated, nonselective cation channel that is localized in the vacuolar membrane and responsible for generating slow vacuolar (SV) current. Under depolarizing membrane potential, cytosolic Ca2+ activates AtTPC1 by binding at the EF-hand domain, whereas luminal Ca2+ inhibits the channel by stabilizing the voltage-sensing domain II (VSDII) in the resting state. Here, we present 2.8 to 3.3 Å cryoelectron microscopy (cryo-EM) structures of AtTPC1 in two conformations, one in closed conformation with unbound EF-hand domain and resting VSDII and the other in a partially open conformation with Ca2+-bound EF-hand domain and activated VSDII. Structural comparison between the two different conformations allows us to elucidate the structural mechanisms of voltage gating, cytosolic Ca2+ activation, and their coupling in AtTPC1. This study also provides structural insight into the general voltage-gating mechanism among voltage-gated ion channels.
Voltage-gated ion channels (VGICs), such as voltage-gated potassium channel (Kv), sodium channel (Nav), and calcium channel (Cav), are activated by depolarization of membrane potential and play essential roles in electrical signal transduction (1–3). VGICs sense the membrane potential by voltage-sensing domains (VSDs), which consist of four transmembrane helices S1 to S4. In most VGICs, VSDs are stabilized in the resting state by hyperpolarizing (negative) membrane potential, and the channel gate stays closed. At depolarizing (relatively positive) membrane potential, VSDs are activated, and their depolarization-induced conformational changes are coupled to the S5–S6 pore domain, resulting in the opening of the channel gate.
While dozens of VGIC structures have been determined over the past two decades, only very few VSDs in these structures were captured in the resting state. That is because VGICs for structural studies in vitro are solubilized in detergent micelle or lipid nanodisc, making it difficult to recapitulate the resting state under hyperpolarizing (negative) membrane potential. The structure of plant two-pore channel (TPC) from Arabidopsis thaliana (AtTPC1) was among the first to capture a resting-state VSD (4, 5). In addition, mutagenesis combined with cross-linking, ion bridge, or toxin binding have been used to trap the structures of VGICs in resting state in several recent studies, including structures of the bacterial sodium channel NavAb (6), the eukaryotic sodium channel chimera Nav1.7-NavPaS (7), and the hyperpolarization-activated cyclic nucleotide-gated ion channel HCN (8). To fully understand the similarities and differences of the voltage-gating mechanism among different VGICs, it will be essential to visualize the structures of various VGICs in both activated and resting state. To this end, we are using AtTPC1 as a model to elucidate the structural mechanism of voltage gating.
TPCs belong to the VGIC superfamily and are ubiquitously expressed in animals and plants (9, 10). While animal TPCs (TPC1 and TPC2) are endolysosomal sodium channels, the plant TPC (TPC1) is a nonselective cation channel responsible for generating the slow vacuole (SV) current (11, 12). TPCs function as homodimer with each subunit comprising two homologous Shaker-like 6-transmembrane segment domains (6-TM I and 6-TM II) (9, 10, 13), thereby equivalent to a classical VGIC with four VSDs and one pore domain.
Plant TPC is involved in many important physiological processes, such as germination and stomatal opening (12), jasmonate biosynthesis (14, 15), modulation of Ca2+ waves induced by salinity stress (16), and plant–pathogen interaction (17). AtTPC1, the most well-studied plant TPC from A. thaliana, is activated by the membrane depolarization and cytosolic Ca2+ but inhibited by vacuolar Ca2+ (18, 19). Previously, we determined the crystal structure of AtTPC1 in closed state (Protein Data Bank [PDB]: 5E1J, AtTPC15E1J) (4). We demonstrated that between the two VSDs within each AtTPC1 subunit, only the second one (VSDII) senses the membrane potential and adopts a resting state in the structure whereas the first one (VSDI) lacks several key features essential for voltage sensing and therefore does not contribute to the voltage-dependent gating. Ca2+ activation occurs at the EF-hand domain containing two EF-hand motifs. However, Ca2+ binding at EF hand 1 appears to play a structural role and does not contribute to Ca2+ activation; Ca2+ binding at EF hand 2 is central for Ca2+ activation and it adopts an unbound state in the structure (4, 19). We also identified the luminal divalent inhibition site in AtTPC1 where Ca2+ or Ba2+ binding can stabilize the voltage-sensing VSDII in a resting state. Based on our structural and electrophysiological analysis, we proposed that the conformational changes triggered by the binding of Ca2+ to cytosolic EF-hand domain are coupled with the pair pore-lining inner helices from the 6-TM I (IS6), whereas the conformational changes of VSDII activated by membrane potential are coupled with the pair of inner helices from the 6-TM II (IIS6) (4). In order to understand the structural basis underlying multistimuli gating of AtTPC1, here we determined AtTPC1 structures in both closed and partially open conformation under different Ca2+ conditions, revealing the structural mechanism of voltage gating and Ca2+ modulation of AtTPC1.
Results
Structure Determinations of AtTPC1 in Different States.
On the one hand, AtTPC1 activation requires both depolarization of membrane potential and cytosolic Ca2+. On the other hand, Ca2+ inhibits the channel from the luminal side by binding and stabilizing VSDII in the resting state (equivalent to imposing a hyperpolarization membrane potential). This converse effect of Ca2+ makes it difficult to stabilize the biochemically purified AtTPC1 channel in the activated state (Fig. 1A). Our previous study has structurally and functionally defined the luminal Ca2+ inhibition site formed by residues Asp240, Asp454, and Glu528 (4). The mutant construct AtTPC1ΔCai with triple mutations of the three Ca2+-coordinating residues (D240A/D454A/E528A) can mitigate the luminal Ca2+ inhibition (20), and therefore is being used to obtain the open-state AtTPC1 structure with the presence of Ca2+ (Fig. 1B). Based on electrophysiology, 0.3 mM cytosolic Ca2+ is sufficient to fully activate AtTPC1 at 0 mV membrane potential (4). To ensure a full Ca2+ occupancy at the EF-hand domain, we determined the structures of AtTPC1ΔCai in higher Ca2+ concentrations, 1 mM (AtTPC1ΔCai/1Ca) and 50 mM Ca2+ (AtTPC1ΔCai/50Ca), respectively (SI Appendix, Fig. S2). For structural comparison, we also determined the structures of wild-type (WT) AtTPC1 in 1 mM (AtTPC1WT/1Ca) and 50 mM Ca2+ (AtTPC1WT/50Ca), which represent the Ca2+ inhibited closed state (SI Appendix, Fig. S1).
Fig. 1.
Schematic illustration of the proposed voltage gating and Ca2+ modulation of AtTPC1. (A) For the purified WT AtTPC1 sample in the presence of Ca2+, the luminal Ca2+ binding inhibits the channel by stabilizing its VSDII in a resting state and thereby prevents the channel from opening despite the presence of cytosolic Ca2+ that can activate the channel by binding to EF-hand domain. (B) For the purified AtTPC1ΔCai mutant in the presence of Ca2+, the removal of the luminal Ca2+ inhibitory site allows the channel to adopt an open state with activated VSDII and Ca2+-bound EF-hand domain. Triple mutations of the three Ca2+-coordinating residues (D240A/D454A/E528A) at the luminal Ca2+ inhibitory site are highlighted in red.
The AtTPC1WT/1Ca Structure.
The AtTPC1WT/1Ca structure was determined at 3.3-Å resolution (Fig. 2 A–C and SI Appendix, Fig. S1). The cryoelectron microscopy (cryo-EM) density map is of high quality, allowing us to build the model of major parts of the protein (SI Appendix, Fig. S1). The AtTPC1WT/1Ca structure is essentially the same as the previously determined crystal structure AtTPC15E1J—both adopt a closed conformation with a resting-state VSDII and an unbound EF hand 2 (SI Appendix, Fig. S3A) (4). Nevertheless, with improved map quality of AtTPC1WT/1Ca, we were able to correct several errors in both the crystal structure and the recent EM study due to their resolution limit (4, 21) and also reveal additional structural features at the Ca2+-activating EF-hand domain and the voltage-sensing VSDII that were unresolved in previous studies. Thus, AtTPC1WT/1Ca provides a much better structural model representing a closed AtTPC1 with an unbound EF-hand domain and a resting-state VSDII.
Fig. 2.
Overall structure of AtTPC1WT/1Ca. (A) 3D reconstruction of AtTPC1WT/1Ca with each subunit individually colored. (B) Cartoon representation of AtTPC1WT/1Ca in the same orientation as the map in A. Ca2+ ions are shown as blue spheres. SN, the first helix at the N terminus; S0, the second helix at the N terminus; 6-TMI, transmembrane segments IS1 to S6; P1-P2, pore helices; E1-F1, EF hand 1; E2-F2, EF hand 2; L1-2, the loop linking EF-hand domain and VSDII; 6-TMII, transmembrane segments IIS1 to S6; SU, unidentified segment. (C) Topology diagram of AtTPC1WT/1Ca. (D) Structure of one subunit of AtTPC1WT/1Ca, with individual elements colored as in C. (Inset) Zoomed-in view of EF-hand domain. (E) Structure of AtTPC1WT/1Ca VSDII with IIS1 omitted for clarity. 310 helix and side chains of gating charges are shown in cyan and orange, respectively. Side chains of key residues for voltage gating and vacuolar Ca2+ inhibition are shown as sticks.
In the EF-hand domain, the N-terminal S0 helix covering residue Thr29-Asp43 is tightly packed with the E1, F1, and F2 helices. In addition, we were also able to visualize an additional short helix at the N terminus (SN helix, Arg20-His27) sitting on top of S0 helix in AtTPC1WT/1Ca (Fig. 2D). Between E2 and F2, density of an extra helix was observed (Fig. 2 A and D). The map quality will not allow us to unambiguously assign the side chains of the helix, and therefore we only modeled it as a poly-Ala helix named SU (unidentified segment) (Fig. 2D). We suspect that this density comes from a predicted C-terminal helix after IIS6, the only missing structured region in the model. In fact, the predicted C-terminal helix appears to be essential for AtTPC1 activity as its deletion mutation resulted in the loss of channel function (22). Following F2 helix, the loop (L1-2, Gln400-Pro405) links EF-hand domain and VSDII by forming interactions with F1–E2 linker, S0, and IIS2–IIS3 linker (Fig. 2D).
The voltage-sensing VSDII can be completely modeled in the AtTPC1WT/1Ca structure. VSDII contains three gating charges in IIS4, namely Arg537 (R3), Arg540 (R4), and Arg543 (R5) (4), and adopts a resting state in AtTPC1WT/1Ca, with the first gating charge Arg537 (R3) positioned in the gating charge transfer center formed by residues Tyr475, Glu478, and Asp500 (23) whereas the other two exposed to the cytosolic solvent (Fig. 2E). Part of IIS4 adopts a 310 helix covering residues Arg537 to Met546 (Fig. 2E). The IIS3–IIS4 linker (Pro518–Asn526) that undergoes conformational changes with IIS4 during voltage gating is also clearly modeled. Glu528 at the N terminus of IIS4 participates in the luminal Ca2+ coordination, as was demonstrated in the crystal structure AtTPC15E1J (Fig. 2E) (4).
The AtTPC1ΔCai/50Ca Structure.
The AtTPC1ΔCai mutant structures obtained in 1 mM (AtTPC1ΔCai/1Ca) and 50 mM Ca2+ (AtTPC1ΔCai/50Ca) are virtually the same (SI Appendix, Fig. S3B). The latter one was refined to 2.8 Å with higher map quality and therefore will be used in the structural analysis (Fig. 3 A and B and SI Appendix, Figs. S2 and S3B). In comparison with closed AtTPC1WT/1Ca, AtTPC1ΔCai/50Ca exhibits significant structural differences at the EF-hand domain, VSDII, and the pore domain, exemplifying major conformational changes at these key gating modules. In the EF-hand domain of AtTPC1ΔCai/50Ca, both the EF hand 1 and EF hand 2 are in Ca2+-bound state (Fig. 3C). The N-terminal SN and the unidentified SU helices likely dissociate from the EF-hand domain and become mobile as they are no longer visible in the AtTPC1ΔCai/50Ca structure (Fig. 3C). In addition, a third Ca2+ binding site (S0 site) is identified at the C-terminal end of S0 helix (Fig. 3C). This Ca2+ is mainly coordinated by side chains of Asp39 and Asp43 and main chain carbonyls of Asp39 and Leu45. Mutations on residues Asp39 and Asp43 markedly reduce the cytosolic Ca2+ activation of AtTPC1 (Fig. 3D), suggesting that this Ca2+ site is an integral part of part of the EF-hand domain and plays a central role in cytosolic Ca2+ activation together with EF hand 2 (4, 19).
Fig. 3.
Overall structure of AtTPC1ΔCai/50Ca. (A) 3D reconstruction of AtTPC1ΔCai/50Ca with each subunit individually colored. (B) Cartoon representation of AtTPC1ΔCai/50Ca in the same orientation as the map in A. Ca2+ ions are shown as blue spheres. (C) Structure of one subunit of AtTPC1ΔCai/50Ca in the same orientation as Fig. 2D, with S0 and EF-hand domains colored as in Fig. 2C. (Inset) Zoomed-in view of EF-hand domain. (D) The whole-cell currents of AtTPC1 with the presence of 0.1 or 0.3 mM cytosolic Ca2+. Mutations at S0 Ca2+ site largely reduce the cytosolic Ca2+-activated currents. (E) Structure of AtTPC1ΔCai/50Ca VSD2 with IIS1 omitted for clarity. 310 helix and side chains of gating charges are shown in cyan and orange, respectively. Side chains of key residues for voltage gating are shown as sticks.
With the elimination of the luminal Ca2+ inhibition site in AtTPC1ΔCai mutant, VSDII adopts a fully activated conformation in AtTPC1ΔCai/50Ca structure (Fig. 3E), in which the last gating charge Arg543 (R5) is positioned at the gating charge transfer center whereas the other two (Arg540 and Arg537) are exposed to the luminal solution. The side chains of Arg540 and Arg537 are stabilized by several negatively charged residues at the luminal-facing cavity of VSDII, including Glu450 in IIS1, Glu468 in IIS2, and Glu511 in IIS3 (Fig. 3E). In AtTPC1ΔCai/50Ca, the 310 helix in IIS4 is one turn shorter than that in the resting VSDII of AtTPC1WT/1Ca and contains residues Arg540 to Met546 (Fig. 3E).
The Voltage-Gating Mechanism.
The voltage activation of AtTPC1 involves two kinds of conformational changes at VSDII: one local movement mainly involving the translation of IIS4 helix within VSDII and the other global movement involving the rotation of VSDII in the context of whole channel. To reveal the local conformational changes within VSDII in response to membrane depolarization, we first align the VSDIIs of AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca using IIS1 to IIS3 (Fig. 4A). While little conformational change occurs at IIS1 and IIS2 during voltage gating (Fig. 4B), IIS3 tilts by ∼15° around its middle part, with its N-terminal end moving away from IIS1/IIS2 and its C-terminal end close to IIS1/IIS2 (Fig. 4C), likely caused by the concerted movement with IIS4 N-terminal end and IIS3–S4 linker. The membrane depolarization-induced conformational changes of IIS4 display several features. First, in the middle of IIS4, the Cα atoms of gating charge residues move up by 6 to 7 Å, about 1.5 helical turns (Fig. 4C). Although in the resting state R3 points to the gating charge transfer center and in the activated state R5 points to the gating charge transfer center, the actual vertical shift is less than two helical turns because the side chain of R3 in the resting state is positioned down to the gating charge transfer center. Second, the 310 helix becomes one turn shorter in the activated state (Arg540 to Arg546) than that in the resting state (Arg537 to Met546), and residues Arg537 to Leu539 adopt a regular α helix after they pass through the gating charge transfer center (Fig. 4C). Third, the IIS4 bends by ∼20° at the Arg537 (R3) in the resting state whereas it is almost a straight helix in the activated state (Fig. 4C). Fourth, the N-terminal end of IIS4 moves closer to the IS5 of the other subunit by 2 to 3 Å (Fig. 4C). With these local conformational changes within IIS4, the vertical shift in the middle of IIS4 is partially converted to a lateral movement at its N-terminal end. The structural rearrangements of IIS4 N-terminal end and IIS3 C-terminal end ensure that during voltage gating the hydrophobic region of IIS3/IIS4 will not be exposed to the luminal solution and the hydrophilic region of IIS3–S4 linker (Pro518 to Asn526) will not be embedded into the membrane.
Fig. 4.
Conformational changes of VSDII between AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca. (A) Top view of superposition of VSDII in AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca when they are aligned at IIS1 to IIS3. (B) Side view of the superposition in A with IIS3 and IIS4 omitted for clarity. (C) Side view of the superposition in A with IIS1 and IIS2 omitted for clarity. 310 helices are shown in magenta (AtTPC1WT/1Ca) and deep green (AtTPC1ΔCai/50Ca). The dash line shows distance between Cα atoms of R3 (R537) in AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca. (D) Top (luminal) view of superposition of AtTPC1WT/1Ca (pink) and AtTPC1ΔCai/50Ca (green) when they are aligned at the entire channel. (Inset) Zoomed-in view of VSDII. Arrows indicate the rotation of VSDII and IIS4–S5 linker. (E) Side view of the superimposition of IIS4 and IIS4–S5 linker in AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca when they are aligned at the entire channel. Pink-green arrows show rotations of IIS4 and IIS4–S5 linker. Pink angle shows the sharp turn formed by IIS4 and IIS4–S5 linker in resting state. The red arrow shows the repulsion imposed on IIS4 by IIS4–S5 linker during its outward rotation.
To reveal the global conformational change of VSDII, we next align the structures of AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca using the entire channel (Fig. 4D). While the pore domains and VSDIs align well between the two structures, VSDII of AtTPC1ΔCai/50Ca (activated state) undergoes an ∼25° clockwise rotation around the central pore in comparison with that of AtTPC1WT/1Ca (resting state) when viewed from the luminal side (Fig. 4D). The global movement of VSDII is initiated by the vertical translation of IIS4, which induces an ∼20° outward rotation of IIS4–IIS5 linker around the joint of IIS4–S5 linker and IIS5 (Fig. 4D). Because in resting state (AtTPC1WT/1Ca) IIS4 and IIS4–IIS5 linker are tightly packed by forming an ∼40° sharp turn, the outward rotation of IIS4–IIS5 linker will inevitably push IIS4 away from its original position, which otherwise will spatially constrain the rotation of IIS4–IIS5 linker (Fig. 4E). The concerted movement of IIS4 and IIS4–IIS5 linker results in the rotation of IIS4 and the entire VSDII (Fig. 4D).
The voltage-gating mechanism revealed by structural comparison of AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca supports the conventional helix translation model and provides more details of conformational change of IIS4, including vertical translation in the middle, 310–α helix transition, straightening or bending of the helix, and lateral movement at the N-terminal end of IIS4. In AtTPC1, membrane depolarization drives two gating charges to pass through the gating charge transfer center and an ∼20° outward rotation of IIS4–S5 linker (Fig. 5 and Movie S1). However, this rotation induces subtle movement of IIS6 at the activation gate (Fig. 5C). It is possible that under physiological conditions, the activation of VSDII will cause the IIS6 helix pair to dilate further away from the central axis to fully open the channel.
Fig. 5.
Voltage-gating mechanism in AtTPC1. (A) 310-to-α helix transition and gating charges movement in IIS4 helix from resting (AtTPC1WT/1Ca) to activated state (AtTPC1ΔCai/50Ca). 310 helix and α helix in IIS4 are shown in cyan and magenta, respectively. Side chains of gating charges (orange) and residues forming the gating charge transfer center (light blue) are shown as sticks. IIS1 is omitted for clarity. (B) Side view of structural transition between resting (AtTPC1WT/1Ca) and activated (AtTPC1ΔCai/50Ca) states. IIS4 and IIS4–IIS5 linker are highlighted in magenta and blue, respectively. For clarity, only one VSDII is shown in each state. (C) Bottom (cytosolic) view of structural transition between resting (AtTPC1WT/1Ca) and activated (AtTPC1ΔCai/50Ca) states. IS1 to IS3 and IIS1 to IIS3 are omitted for clarity.
The Cytosolic Ca2+ Activation Mechanism.
The cytosolic Ca2+ activation of AtTPC1 also involves two kinds of conformational changes at EF-hand domain: one local movement mainly involving the translation of E2 helix within EF-hand domain, and the other global rotation and translation of EF-hand domain in the context of whole channel. To reveal the Ca2+-induced local conformational changes within the EF-hand domain, we first align structures of the EF-hand domains from AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca using S0 helix (Fig. 6 A and B). While S0, E1, and F1 remain almost unchanged, E2 moves close to F2 and pushes its N-terminal end by ∼3 Å toward E1, along with the dissociation of the unidentified SU helix from EF-hand domain in the Ca2+-bound EF-hand domain of AtTPC1ΔCai/50Ca (Fig. 6 A and B). The dissociation of SU helix appears to be prerequisite for Ca2+ binding at EF hand 2 and the ensued tight packing between E2 and F2 helices. This local conformational change within EF hand 2 is likely triggered by the rotation of VSDII in activated state, which pulls F2 to kick off SU (Fig. 6B). Indeed, Ca2+ binding and activation of the channel at EF-hand domain require the VSDII of AtTPC1 to be activated as discussed in The Coupling of VSDII Depolarization and Cytosolic Ca2+ Activation.
Fig. 6.
Cytosolic Ca2+ activation mechanism in AtTPC1. (A) Superposition of EF-hand domain in AtTPC1WT/1Ca (pink) and AtTPC1ΔCai/50Ca (green) when they are aligned at the S0 helix. The pink-green arrow indicates the translation of E2 helix. For clarity, SN in AtTPC1WT/1Ca is omitted. Ca2+ ions binding at S0 and EF hand 2 in AtTPC1ΔCai/50Ca are shown as blue spheres, and Ca2+ ions binding at EF hand 1 in AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca are shown as orange and green spheres, respectively. (B) The channels in A are rotated by 150° around the x-axis. Pink-green arrows indicate movements of E2 and F2 N terminus. The blue arrow indicates the dragging force imposed on the loop L1-2 by VSDII. The red arrow indicates force imposed on SU by F2. (C) Side view of superposition of AtTPC1WT/1Ca (pink) and AtTPC1ΔCai/50Ca (green) when they are aligned at the entire channel. (D) Zoomed-in view of EF-hand domain of AtTPC1WT/1Ca in C. For clarity, the channel in C is rotated by 10° around the y-axis. SN and SU are shown in purple and yellow, respectively. The blue arrow indicates the dragging force imposed on the loop L1-2 by VSDII. Red arrows indicate force imposed on SN by IS1 and on SU by F2 during the rotation of EF-hand domain. The pink rotational arrow indicates the rotation of the entire EF-hand domain around E1. (E) Zoomed-in view of EF-hand domain of AtTPC1ΔCai/50Ca in C. For clarity, the channel in C is rotated by 10° around the y-axis. Ca2+ are shown as blue spheres. (F) Zoomed-in view of EF-hand domain in C. For clarity, the channels in C are rotated by 10° around the y-axis and 40° around the x-axis. Ca2+ are shown as blue spheres. Arrows indicate the rotation of EF-hand domains and translation of S0, E1, and F1. For clarity, SN, EF hand 2, and SU are not shown.
To reveal the cytosolic Ca2+-induced global conformational change of EF-hand domain, we then align structures of AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca using the entire channel (Fig. 6C). Almost all helices in EF-hand domain of AtTPC1ΔCai/50Ca, including E1, F1, E2, F2, and S0 undergo major structural rearrangement compared with those in AtTPC1WT/1Ca (Fig. 6 C–E). We propose that the global conformational change of EF-hand domain is also initiated by the rotation of VSDII upon voltage activation, which drags EF-hand domain to undergo a global rotation around E1 via the loop L1-2 (Fig. 6 C and D). During the rotation, SN helix is pushed away from EF-hand domain by IS1 N-terminal end, making the Ca2+ site at S0 helix accessible (Fig. 6 D and E). Ca2+ binding at S0 site strengthens the packing between EF-hand domain and IS1, allowing EF-hand domain to undergoes global conformational change along with IS1 (Fig. 6E). The binding of Ca2+ at S0 and EF hand 2, together with dragging by VSDII, causes an ∼40° rotation of EF-hand domain and ∼4-Å shift of E1 away from its original position (Fig. 6F and Movie S2). Since E1 and IS6 form a single helix, the motion of E1 directly causes an outward movement and a rotational motion of IS6, resulting in an enlarged pore along the pair of IS6 direction.
Conformational Changes at the Activation Gate.
The activation gate of AtTPC1WT/1Ca is closed by constrictions formed by hydrophobic residues from both IS6 (Leu301 and Tyr305) and IIS6 (Val668, Leu672, and Phe676). The global conformational changes upon voltage and Ca2+ activation observed in AtTPC1ΔCai/50Ca converge to the movement of pore-lining IS6 and IIS6 helices to open the gate. The conformational change triggered by Ca2+ binding at EF hand is directly coupled to the pair of IS6 helices, driving constriction-forming Leu301 and Tyr305 to dilate and rotate away from the central axis and resulting in the widening of the constrictions from 5.1 to 7.3 Å and from 6.7 to 11.7 Å, respectively, when measuring the diagonal atom-to-atom distances (Fig. 7 A and B and Movie S3). The conformational change at VSDII upon voltage activation is coupled to the IIS6 helices. However, the IIS6 movement appears to be quite subtle between closed AtTPC1WT/1Ca and activated AtTPC1ΔCai/50Ca with the increase of the diagonal atom-to-atom distances at constriction-forming residues Val668, Leu672, and Phe676 from 6.9 to 8.6 Å, from 3.8 to 4.7 Å, and from 7.3 to 7.4 Å, respectively (Fig. 7 B and C). While the cross-sections at the activation gate between IS6 pair and IIS6 pair are both enlarged in AtTPC1ΔCai/50Ca, the constriction at Leu672 is still quite narrow (Fig. 7D). The van der Waals radius of AtTPC1ΔCai/50Ca pore calculated with HOLE program remains at about 1 Å at the narrowest part of the activation gate and expands to over 2 Å at the lower part (Fig. 7 E–G). Therefore, we propose that AtTPC1ΔCai/50Ca is in a partially open, ion-impermeable conformation. The IIS6 helix pair needs to dilate further away from the central axis to have a fully opened pore. Interestingly, tubule-shaped bulky densities, likely arising from lipid or detergent molecules, fill up the activation gate from the intracellular side, similar to those observed in the open-state structure of human TPC2 (Fig. 7H) (24).
Fig. 7.
Ion conduction pores in AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca. (A) Side view of the bundle crossing formed by IS6 pair in AtTPC1WT/1Ca (Left) and AtTPC1ΔCai/50Ca (Right). Numbers are diagonal atom-to-atom distances (in Å) of the constriction points. (B) Bottom (cytosolic) view of superposition of activation gates in AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca in two sections: L301/V668/L672 (top) and Y305/F676 (below). (C) Side view of the bundle crossing formed by IIS6 pair in AtTPC1WT/1Ca (Left) and AtTPC1ΔCai/50Ca (Right). Numbers are diagonal atom-to-atom distances (in Å) of the constriction points. (D) Bottom (cytosolic) view of superposition of pore domains in AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca. (E) Solvent-accessible pathway along the pore mapped using the HOLE program for AtTPC1WT/1Ca. The pore domain of AtTPC1WT/1Ca is shown as a pink ribbon. (F) Solvent-accessible pathway along the pore mapped using the HOLE program for AtTPC1ΔCai/50Ca. The pore domain of AtTPC1ΔCai/50Ca is shown as a green ribbon. (G) Comparison of pore radii (calculated with the program HOLE) for AtTPC1WT/1Ca and AtTPC1ΔCai/50Ca. (H) Side view of the EM density (blue mesh) in the cytosolic gate of AtTPC1ΔCai/50Ca.
The Coupling of VSDII Depolarization and Cytosolic Ca2+ Activation.
We initially anticipated that with the presence of high Ca2+ concentration, we would capture the structure of WT AtTPC1 in an intermediate state with activated EF-hand domain but inhibited voltage sensor (resting-state VSDII). To our surprise, the structure of WT AtTPC1 even in 50 mM Ca2+ (AtTPC1WT/50Ca) is virtually the same as that in 1 mM Ca2+ (AtTPC1WT/1Ca)—it has no Ca2+ binding at both EF hand 2 and S0 sites and a resting VSDII due to luminal Ca2+ inhibition, implying that a resting VSDII prohibits Ca2+ binding and activation of EF-hand domain (SI Appendix, Figs. S1 and S3C). By contrast, in the structure of AtTPC1ΔCai mutant whose luminal Ca2+ inhibition site is mutated and the VSDII is activated, 1 mM Ca2+ is sufficient to fully activate EF-hand domain (in AtTPC1ΔCai/1Ca) (SI Appendix, Fig. S3B). Therefore, the activation of VSDII and EF-hand domain is cooperatively regulated. We suspect that the coupling between VSDII and EF-hand domain is mediated by the loop L1-2, SU, and SN helices. When the VSDII is in resting state, Ca2+ binding sites at S0 and EF hand 2 are blocked by the SN and SU, respectively (Figs. 6D and 8). The activation and lateral rotation of VSDII induces a global rotation of EF-hand domain via the loop L1-2, when SU and SN are kicked off by F2 and IS1 N-terminal end, respectively, allowing the two Ca2+ binding sites to be accessible (Figs. 6 D and E and 8 and Movie S2). Without the activation of VSDII, both Ca2+ binding sites are likely blocked, preventing Ca2+ activation of EF-hand domain even in the presence of Ca2+ at high concentration.
Discussions
Voltage-Gating Mechanisms in Different VGICs.
Voltage-gating mechanisms of AtTPC1 share some common features with those of other reported VGICs such as NavAb, Nav1.7-NavPaS, Nav1.7-NavAb chimera, Nav1.5, and HCN (6–8, 25–27). In all cases, upon membrane depolarization, S4 undergoes vertical translation with gating charge residues passing through gating charge transfer center while S1 and S2 remain almost unchanged (SI Appendix, Fig. S4). In AtTPC1, NavAb, and Nav1.7-NavPaS with domain-swapped conformation, vertical movement of S4 brings an outward rotation of S4–S5 linker around the joint of S4–S5 linker and S5 and will further induce outward movement and (or) rotation of S6 at the activation gate (SI Appendix, Fig. S4). In different VGICs, the numbers of gating charge residue and the lengths of 310 helices may be different. Consequently, the magnitude of S4 movement and the total gating charges across gating charge transfer center, as well as the degree of S4–S5 linker rotation, vary among VGICs. Different from other VGICs, in AtTPC1 we observed a profound rotation of the entire VSDII during voltage gating. In resting state, VSDII forms a sharp turn with the pore domain at the joint of IIS4 and IIS4–S5 linker, resulting in a rectangular shape of the channel viewed from the luminal side (Fig. 4D). Along with VSDII activation and IIS4 translation, the entire VSDII rotates by ∼25° around the pore domain. This rotation is essential for the activation of the EF-hand domain by cytosolic Ca2+. Rotation of the VSD around the pore domain during voltage activation has also been reported in NavAb structures but at a much smaller extend than that observed in AtTPC1 (6, 28, 29). It is worth noting that the structure of the resting VSDII in AtTPC1 is stabilized by the luminal Ca2+, yielding a highly rectangular-shaped channel dimer. It is possible that the observed VSDII rotation between the resting and activated states is partly contributed by the constrain of the resting VSDII when chelated with the luminal Ca2+.
Structural Conservation of Activated IIS4 within TPC Family.
Previously, we determined mouse TPC1 (MmTPC1), which is a voltage-gated, PI(3,5)P2-activated channel, in apo closed (MmTPC1apo) and PI(3,5)P2-bound (MmTPC1PI(3,5)P2) states (30). In both states, the VSDII of MmTPC1 is in activated state. A major difference between MmTPC1 and AtTPC1 is that the distance between the two VSDs within the same subunit is larger in MmTPC1 than that in AtTPC1WT/1Ca or AtTPC1ΔCai/50Ca, (SI Appendix, Fig. S5 A and B). In comparison with the activated IIS4 in AtTPC1ΔCai/50Ca, IIS4 in MmTPC1 adopts different conformation—it bends by ∼30° at Arg540 in the middle. However, when AtTPC1ΔCai/50Ca and MmTPC1PI(3,5)P2 are aligned at IIS1 to S3, the entire VSDIIs are superimposed well, and the distance between Cα atoms of the last gating charge residues that point to the gating charge transfer center is within 2 Å, suggesting high structural conservation within TPC family (SI Appendix, Fig. S5C).
Comparison with Related Work.
Recently, Kintzer et al. reported cryo-EM structures of AtTPC1ΔCai in saposin A in the presence of 1 mM Ca2+, with Fab antibody (21). The VSDII in their 3.3-Å resolution structure (PDB: 6E1K, AtTPC16E1K) is not resolved, and two low-resolution structures of VSDII with different conformations (PDB: 6E1N, AtTPC16E1N; PDB: 6E1P, AtTPC16E1P) are reconstructed after focused three-dimensional (3D) classification. In light of low local resolution (less than 5 Å) of VSDII in both structures, the secondary structures assignment and amino acid registry in VSDII are not accurate. Contrary to our AtTPC1ΔCai/50Ca or AtTPC1ΔCai/1Ca structures, no Ca2+-induced conformational changes at the EF-hand domain were observed in the AtTPC16E1K structure. We suspect that these differences may be caused by different sample preparations. AtTPC1ΔCai/50Ca structure was determined using protein sample stabilized in the LMNG detergent, whereas AtTPC16E1K was determined using protein sample reconstituted in saposin A with Fab bound at the EF-hand domain.
Insight into the Physiological Activity of AtTPC1.
Our AtTPC1 structures were determined with Ca2+ at 1 to 50 mM, which is much higher than the physiological Ca2+ concentration. In vivo, the cytosolic Ca2+ concentration is about 100 nM, whereas the vacuolar Ca2+ concentration is between 0.2 and 2 mM (31). Some biotic and abiotic stimuli can significantly elevate local cytosolic Ca2+ concentration (32, 33). Moreover, cytosolic Mg2+ has been previously shown to potentiate the Ca2+ sensitivity of TPC1 to the low nanomolar range and shift the voltage dependence of the channel to more physiological potentials (34, 35). In our current recording system, AtTPC1 can still be activated in the presence of 0.1 to 1.0 mM extracellular (vacuolar) Ca2+ (4). In vivo with the increase of local cytosolic Ca2+ concentration and the presence of cytosolic Mg2+, the channel may still conduct Ca2+ even with the inhibition of vacuolar Ca2+. We therefore believe that AtTPC1 can mediate Ca2+ release from vacuole or Ca2+ uptake into vacuole, depending on the local electrical and Ca2+ gradient across the vacuolar membrane (36). Physiologically, AtTPC1 has to be strictly regulated because of its high abundance in vacuolar membrane and Ca2+ permeability.
Conclusion
In summary, by combining mutagenesis at the vacuolar luminal Ca2+ inhibitory site and Ca2+ concentrations in protein samples, we are able to capture AtTPC1 structures in a closed conformation with a resting VSDII and an unbound EF-hand domain and in a partially open conformation with an activated VSDII and a Ca2+-activated EF-hand domain. By comparing the high-quality structures in two different conformations, we reveal some features of voltage-gating mechanism in AtTPC1, including the rotation of VSDII, 310–α helix transition within part of IIS4, and straightening or bending of IIS4 helix that associates with its up–down translation. We also elucidate the cytosolic Ca2+ activation mechanism and its coupling with voltage gating, providing mechanistic insights into how AtTPC1 integrates multiple cellular stimuli for its gating regulation (Fig. 8 and Movies S1 and S2).
Fig. 8.
The voltage gating, cytosolic Ca2+ activation, and their coupling mechanism in AtTPC1. (A) Cartoon representation of working model for voltage gating and Ca2+ modulation of AtTPC1. Black arrows indicate upward movement of IIS4 and rotation of IIS4–S5 linker. (B) The channels in A are rotated by 150° around the y-axis. Black arrows indicate the rotation of VSDII and movements of E2, IS6, and IIS6. Also reference Movies S1–S3.
Materials and Methods
Protein Purification and Structure Determination.
The full-length WT or mutant AtTPC1 were expressed in Pichia pastoris GS115 strain in MMH medium (1.34% YNB, 0.5% methanol, 4 × 10−5% biotin, 0.004% Histidine) at 30 °C for 2 d. The cells were lysed by Continuous High Pressure Cell Disrupter (GNBIO) at 1,500 bar for four times. The AtTPC1 protein was extracted with 1.5% (weight/volume) n-Dodecyl-β-D-Maltopyranoside (DDM, ExceedBio) by gentle agitation for 3 h at 4 °C. After purified with Talon cobalt (Clontech) affinity chromatography and size-exclusion chromatography on a Superose 6 10/300 GL column (GE Healthcare), the final sample was concentrated to 5 to 8 mg/mL for single-particle analysis. The cryo-EM grids were prepared using a Mark IV Vitrobot (FEI). Data were acquired at a Titan Krios microscope (FEI) operated at 300 kV with a K2 direct electron detector (Gatan). Data were dose fractionated to 40 frames with a dose rate of 8 e−/pixel/s and a total exposure time of 8 s, corresponding to a total dose of about 64 e−/Å2. Images processing and 3D reconstruction were performed following standard procedures, and the final resolution was estimated by gold-standard Fourier shell correlation = 0.143 criterion (SI Appendix, Figs. S1 and S2 and Table S1). Atomic model of AtTPC1 (PDB: 5E1J) was used as a reference for model building of WT and mutant AtTPC1 cryo-EM structures.
Electrophysiology.
The WT or mutant AtTPC1 open reading frame was cloned into pEGFP-C1 vector at SalI/SmaI sites (Clontech), and plasmid was transfected into HEK293 cells using Lipofectamine 2000 (Life Technology). To measure AtTPC1 current in HEK293 cells expressing GFP-AtTPC1, patch clamp in the whole-cell configuration was employed. The standard bath solution contained the following (in mM): 145 sodium methanesulfonate (Na-MS), 5 NaCl, 10 Hepes buffered with Tris, pH 7.4. The pipette solution contained the following (in mM): 150 Na-MS, 2.5 MgCl2, 10 Hepes buffered with Tris, pH 7.4. Data were acquired using an AxoPatch 200B amplifier (Molecular Devices) and a low-pass analog filter set to 1 kHz. The holding potential was set to −70 mV. The membrane was stepped from the holding potential (−70 mV) to various testing potentials (−100 to +100 mV with a step of 10 mV) for 0.8 s and then returned to the holding potential.
Detailed methods of protein purification, structure determination, and electrophysiology are provided in SI Appendix, Supplementary Materials and Methods.
Supplementary Material
Acknowledgments
Single-particle cryo-EM data were collected at the Center of Cryo-Electron Microscopy at Zhejiang University. We thank Dr. Xing Zhang and Dr. Shenghai Chang for support in facility access and data acquisition. This work was supported in part by the Ministry of Science and Technology (2020YFA0908501 and 2018YFA0508100 to J.G.), the National Natural Science Foundation of China (31870724 to J.G.; 82030108, 31872796, and 81571127 to W.Y.), National Major Special Project on New Drug Innovation of China (2018ZX09711001-004-005 to W.Y.), Zhejiang Provincial Natural Science Foundation (LR19C050002 to J.G.; LR16H090001 to W.Y.), and the Fundamental Research Funds for the Central Universities (2021FZZX001-28 to J.G.). J.G. is supported by Ministry of Education Frontier Science Center for Brain Science and Brain-Machine Integration, Zhejiang University. Y.J. is supported by the HHMI (Y.J.) and by grants from the NIH (R35GM140892 to Y.J.) and the Welch Foundation (Grant I-1578 to Y.J.).
Footnotes
The authors declare no competing interest.
This article is a PNAS Direct Submission.
This article contains supporting information online at https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2113946118/-/DCSupplemental.
Data Availability
The cryo-EM density maps and coordinates of AtTPC1WT/1Ca, AtTPC1WT/50Ca, AtTPC1ΔCai/1Ca, and AtTPC1ΔCai/50Ca have been deposited in the Electron Microscopy Data Bank under accession nos. EMD-31585–EMD-31588, respectively, and in the Research Collaboratory for Structural Bioinformatics (RCSB) PDB under accession nos. 7FHK (37), 7FHL (38), 7FHN (39), and 7FHO (40), respectively.
References
- 1.Hodgkin A. L., Huxley A. F., A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol. 117, 500–544 (1952). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Yu F. H., Catterall W. A., The VGL-chanome: A protein superfamily specialized for electrical signaling and ionic homeostasis. Sci. STKE 2004, re15 (2004). [DOI] [PubMed] [Google Scholar]
- 3.Catterall W. A., Wisedchaisri G., Zheng N., The chemical basis for electrical signaling. Nat. Chem. Biol. 13, 455–463 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Guo J., et al. , Structure of the voltage-gated two-pore channel TPC1 from Arabidopsis thaliana. Nature 531, 196–201 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kintzer A. F., Stroud R. M., Structure, inhibition and regulation of two-pore channel TPC1 from Arabidopsis thaliana. Nature 531, 258–262 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wisedchaisri G., et al. , Resting-state structure and gating mechanism of a voltage-gated sodium channel. Cell 178, 993–1003.e12 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Clairfeuille T., et al. , Structural basis of α-scorpion toxin action on Nav channels. Science 363, eaav8573 (2019). [DOI] [PubMed] [Google Scholar]
- 8.Lee C. H., MacKinnon R., Voltage sensor movements during hyperpolarization in the HCN channel. Cell 179, 1582–1589.e7 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ishibashi K., Suzuki M., Imai M., Molecular cloning of a novel form (two-repeat) protein related to voltage-gated sodium and calcium channels. Biochem. Biophys. Res. Commun. 270, 370–376 (2000). [DOI] [PubMed] [Google Scholar]
- 10.Furuichi T., Cunningham K. W., Muto S., A putative two pore channel AtTPC1 mediates Ca(2+) flux in Arabidopsis leaf cells. Plant Cell Physiol. 42, 900–905 (2001). [DOI] [PubMed] [Google Scholar]
- 11.Hedrich R., Neher E. J. N., Cytoplasmic calcium regulates voltage-dependent ion channels in plant vacuoles. Nature 329, 833–836 (1987). [Google Scholar]
- 12.Peiter E., et al. , The vacuolar Ca2+-activated channel TPC1 regulates germination and stomatal movement. Nature 434, 404–408 (2005). [DOI] [PubMed] [Google Scholar]
- 13.Papazian D., Schwarz T., Tempel B., Jan Y., Jan L. J. S., Cloning of genomic and complementary DNA from Shaker, a putative potassium channel gene from Drosophila. Science 237, 749–753 (1987). [DOI] [PubMed] [Google Scholar]
- 14.Bonaventure G., et al. , A gain-of-function allele of TPC1 activates oxylipin biogenesis after leaf wounding in Arabidopsis. Plant J. 49, 889–898 (2007). [DOI] [PubMed] [Google Scholar]
- 15.Bonaventure G., Gfeller A., Rodríguez V. M., Armand F., Farmer E. E., The fou2 gain-of-function allele and the wild-type allele of two pore channel 1 contribute to different extents or by different mechanisms to defense gene expression in Arabidopsis. Plant Cell Physiol. 48, 1775–1789 (2007). [DOI] [PubMed] [Google Scholar]
- 16.Choi W. G., Toyota M., Kim S. H., Hilleary R., Gilroy S., Salt stress-induced Ca2+ waves are associated with rapid, long-distance root-to-shoot signaling in plants. Proc. Natl. Acad. Sci. U.S.A. 111, 6497–6502 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Vincent T. R., et al. , Interplay of plasma membrane and vacuolar ion channels, together with BAK1, elicits rapid cytosolic calcium elevations in arabidopsis during aphid feeding. Plant Cell 29, 1460–1479 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Dadacz-Narloch B., et al. , A novel calcium binding site in the slow vacuolar cation channel TPC1 senses luminal calcium levels. Plant Cell 23, 2696–2707 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Schulze C., Sticht H., Meyerhoff P., Dietrich P., Differential contribution of EF-hands to the Ca2+-dependent activation in the plant two-pore channel TPC1. Plant J. 68, 424–432 (2011). [DOI] [PubMed] [Google Scholar]
- 20.Guo J., Zeng W., Jiang Y., Tuning the ion selectivity of two-pore channels. Proc. Natl. Acad. Sci. U.S.A. 114, 1009–1014 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kintzer A. F., et al. , Structural basis for activation of voltage sensor domains in an ion channel TPC1. Proc. Natl. Acad. Sci. U.S.A. 115, E9095–E9104 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Larisch N., et al. , The function of the two-pore channel TPC1 depends on dimerization of its carboxy-terminal helix. Cell. Mol. Life Sci. 73, 2565–2581 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Tao X., Lee A., Limapichat W., Dougherty D. A., MacKinnon R., A gating charge transfer center in voltage sensors. Science 328, 67–73 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.She J., et al. , Structural mechanisms of phospholipid activation of the human TPC2 channel. eLife 8, e45222 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Xu H., et al. , Structural basis of Nav1.7 inhibition by a gating-modifier spider toxin. Cell 176, 702–715.e14 (2019). [DOI] [PubMed] [Google Scholar]
- 26.Jiang D., et al. , Structural basis for voltage-sensor trapping of the cardiac sodium channel by a deathstalker scorpion toxin. Nat. Commun. 12, 128 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wisedchaisri G., et al. , Structural basis for high-affinity trapping of the NaV1.7 channel in its resting state by tarantula toxin. Mol. Cell 81, 38–48.e4 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Payandeh J., Scheuer T., Zheng N., Catterall W. A., The crystal structure of a voltage-gated sodium channel. Nature 475, 353–358 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Payandeh J., Gamal El-Din T. M., Scheuer T., Zheng N., Catterall W. A., Crystal structure of a voltage-gated sodium channel in two potentially inactivated states. Nature 486, 135–139 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.She J., et al. , Structural insights into the voltage and phospholipid activation of the mammalian TPC1 channel. Nature 556, 130–134 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Hedrich R., Marten I., TPC1-SV channels gain shape. Mol. Plant 4, 428–441 (2011). [DOI] [PubMed] [Google Scholar]
- 32.McAinsh M. R., Pittman J. K., Shaping the calcium signature. New Phytol. 181, 275–294 (2009). [DOI] [PubMed] [Google Scholar]
- 33.Roelfsema M. R., Hedrich R., Making sense out of Ca(2+) signals: Their role in regulating stomatal movements. Plant Cell Environ. 33, 305–321 (2010). [DOI] [PubMed] [Google Scholar]
- 34.Pei Z. M., Ward J. M., Schroeder J. I., Magnesium sensitizes slow vacuolar channels to physiological cytosolic calcium and inhibits fast vacuolar channels in fava bean guard cell vacuoles. Plant Physiol. 121, 977–986 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Carpaneto A., Cantù A. M., Gambale F., Effects of cytoplasmic Mg2+ on slowly activating channels in isolated vacuoles of Beta vulgaris. Planta 213, 457–468 (2001). [DOI] [PubMed] [Google Scholar]
- 36.Ward J. M., Schroeder J. I., Calcium-activated K+ channels and calcium-induced calcium release by slow vacuolar ion channels in guard cell vacuoles implicated in the control of stomatal closure. Plant Cell 6, 669–683 (1994). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.F. Ye, L. Xu, X. Li, Y. Jiang, J. Guo, Structure of AtTPC1 with 1 mM Ca2+. RCSB Protein Data Bank. https://www.rcsb.org/structure/7FHK. Deposited 29 July 2021. [Google Scholar]
- 38.F. Ye, L. Xu, X. Li, Y. Jiang, J. Guo, Structure of AtTPC1 with 50 mM Ca2+. RCSB Protein Data Bank. https://www.rcsb.org/structure/7FHL. Deposited 29 July 2021. [Google Scholar]
- 39.F. Ye, L. Xu, X. Li, Y. Jiang, J. Guo, Structure of AtTPC1 D240A/D454A/E528A mutant with 1 mM Ca2+. RCSB Protein Data Bank (PDB). https://www.rcsb.org/structure/7FHN. Deposited 29 July 2021. [Google Scholar]
- 40.F. Ye, L. Xu, X. Li, Y. Jiang, J. Guo, Structure of AtTPC1 D240A/D454A/E528A mutant with 50 mM Ca2+. RCSB Protein Data Bank. https://www.rcsb.org/structure/7FHO. Deposited 29 July 2021. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The cryo-EM density maps and coordinates of AtTPC1WT/1Ca, AtTPC1WT/50Ca, AtTPC1ΔCai/1Ca, and AtTPC1ΔCai/50Ca have been deposited in the Electron Microscopy Data Bank under accession nos. EMD-31585–EMD-31588, respectively, and in the Research Collaboratory for Structural Bioinformatics (RCSB) PDB under accession nos. 7FHK (37), 7FHL (38), 7FHN (39), and 7FHO (40), respectively.








