XB-ART-56031
J Gen Physiol
2019 Aug 05;1518:986-1006. doi: 10.1085/jgp.201812285.
Show Gene links
Show Anatomy links
Phosphoinositides modulate the voltage dependence of two-pore channel 3.
Shimomura T
,
Kubo Y
.
???displayArticle.abstract???
Two-pore channels, or two-pore Na+ channels (TPCs), contain two homologous domains, each containing a functional unit typical of voltage-dependent cation channels. Each domain is considered to be responsible for either phosphoinositide (PI) binding or voltage sensing. Among the three members of the TPC family, TPC1 and TPC2 are activated by PI(3,5)P2, while TPC3 has been thought not to be affected by any PIs. Here, we report that TPC3 is sensitive to PI(3,4)P2 and PI(3,5)P2, but not to PI(4,5)P2, and that the extremely slow increase in TPC3 currents induced by depolarization in Xenopus oocytes is due to the production of PI(3,4)P2 Similarly to TPC1, the cluster of basic amino acid residues in domain I is critical for PI sensitivity, but with a slight variation that may allow TPC3 to be sensitive to both PI(3,4)P2 and PI(3,5)P2 We also found that TPC3 has a unique PI-dependent modulation mechanism of voltage dependence, which is achieved by a specific bridging interaction between domain I and domain II. Taken together, these findings show that TPC3 is a unique member of the TPC family that senses PIs and displays a strong coupling between PI binding and voltage-dependent gating.
???displayArticle.pubmedLink??? 31182502
???displayArticle.pmcLink??? PMC6683669
???displayArticle.link??? J Gen Physiol
Species referenced: Xenopus
Genes referenced: aopep ins
???attribute.lit??? ???displayArticles.show???
Figure 1. Depolarization-induced modulation of the voltage dependence of XtTPC3 expressed in Xenopus oocytes. (A) A representative current trace of induction of XtTPC3 in Xenopus oocytes. Currents were elicited by changing the membrane voltage from −60 mV holding potential to 50 mV for 20 s. Two sequential stimuli were applied, and then membrane voltage was maintained at the holding potential for 10 min until the third stimulus. (B) A brief protocol to check the extent of induction (left) and the resultant current traces of XtTPC3 before (black) and after (red) the induction stimulus of 80 mV for 20 s. (C) The protocol used to record the G–V relationships (top) and the resultant current traces (middle). The region indicated by a black bar is shown in an expanded view (bottom). Before induction, each step pulse was preceded by an interval of −60 mV holding potential, generally for 1 min, to maintain the channels in a pre-induction state. The interval time was varied depending on the mutant under study. To confirm that the channels were in a pre-induction state, the current amplitude elicited by a 100 mV pulse for 50 ms was monitored, to confirm stability of the amplitude within a series of sweeps. After a set of traces was obtained, an 80-mV depolarizing pulse was applied for 20 s, and then the current traces after induction were obtained. In the case after induction, to maintain the induced condition, 100-mV prepulses were applied for 500 ms before each step pulse. (D) G–V relationships of XtTPC3 before (filled) and after (open) induction. G/Gmax is the normalized tail current amplitude. Error bars represent the SD of six independent experiments. | |
Figure 2. Comparison of endogenous “inducible” Na+ current and exogenously expressed XtTPC3 current recordings. (A) Representative current traces obtained with a +50 mV long depolarizing pulse from oocytes with or without XtTPC3 cRNA injection. (B) Representative current traces obtained with a +100 mV depolarization pulse from the −60-mV holding potential before and after induction. Two traces from oocytes in which XtTPC3 cRNA was injected (left) or not injected (right) are shown. The induction stimulus was +80 mV for 20 s. (C) Representative current traces for obtaining G–V relationships. Currents were elicited by the same protocols as shown in Fig. 1 C. Traces from the same oocytes before (left) and after (right) induction are shown. Black arrows indicate the time points at which tail current measurements were made. (D) Plots of tail current amplitude versus prepulse voltage before (closed) and after (open) induction. Error bars represent the SD of five independent experiments. | |
Figure 3. The effects of wortmannin pretreatment on the G–V relationships before and after induction. (A) Representative traces evoked by a 100-mV step pulse before (black) and after (red) induction. (B) Representative current traces from the 30 µM wortmannin-treated or nontreated oocytes for obtaining the G–V relationships. The currents before and after induction were elicited by the same protocols as Fig. 1 C, although the tail currents were obtained at 50 mV because of the small amplitude. The 100-mV prepulse in each condition was omitted for clarity. (C) G–V relationships of XtTPC3 with (square) or without (circle) wortmannin pretreatment. Data before (filled) and after induction (open) are shown. (D) Plots of V1/2 before (black) and after (red) induction with or without wortmannin pretreatment. Error bars represent the SD of five to seven independent experiments. n.s., not significant at P > 0.05; **, P < 0.01. wort., wortmannin. | |
Figure 4. Modulation of induction and activation kinetics of XtTPC3 by Ci-VSP. (A) A scheme showing the voltage-dependent switch of the phosphatase activity of Ci-VSP. Ci-VSP dephosphorylates PIP3 to PI(3,4)P2 and PI(4,5)P2 at lower membrane voltage, and even to PI(4)P at higher membrane voltage. (B) Representative current traces of XtTPC3 only (black), with Ci-VSP WT (red), or with Ci-VSP F161W/R232K (blue). (C) The protocol to obtain the voltage-dependent effects of Ci-VSP on time constants of activation in XtTPC3. The effect of a prepulse to various membrane voltages for 2 s, which is short enough not to cause the induction, was monitored during the subsequent 100-mV test pulse for 500 ms. To verify recovery to the basal condition after 5 min at holding potential, the current amplitude generated by a short 30-mV pulse was checked before the prepulses. (D–F) Representative current traces of XtTPC3 only (D), with Ci-VSP WT (E) and with Ci-VSP F161W/R232K (F). The whole traces elicited by the protocol in (C; left) and the partial traces elicited by 100 mV test pulse (right) are shown. (G) The relationships between time constants of activation and prepulse voltages. Time constants were obtained by fitting each trace with a single exponential function. The time constant labeled “induced” (the separate plot on the right) was obtained from the current traces evoked after a 20-s, 80-mV induction stimulus in WT. Error bars represent the SD of six independent experiments. | |
Figure 5. Voltage dependence of induction. (A) The protocol (top) and resultant representative current traces (middle) to obtain the voltage dependence of induction. The prepulses with different membrane voltage were given for 20 s, and then the membrane was kept at the holding potential for 20 s. The effect of the prepulses was monitored during the subsequent 100-mV test pulse. The currents evoked by 100-mV test pulses, in the red dotted box, are shown in an expanded view (bottom). (B) The relationship between time constant of activation and prepulse voltage, obtained from A. The current traces evoked by 100-mV test pulses were fitted by a single exponential function. V1/2 (mV) is 32.9 ± 2.7 mV. Error bars represent the SD of six independent experiments. | |
Figure 6. The effects of insulin pretreatment on XtTPC3. (A) The protocol for examining the potentiation of XtTPC3. (B) Representative current traces evoked by the protocol (A), without (top) or with (bottom) 10 µM insulin pretreatment. The pulse protocol was given before (black) and after (red) an induction stimulus (80 mV for 20 s). (C) The plot of normalized amplitude evoked by a 50-mV stimulus with (blue) or without (black) insulin pretreatment. The current amplitude before induction was normalized by that after induction as depicted in B. (D) The plot of time constant of activation at 100 mV. Error bars represent the SD of five independent experiments. n.s., not significant at P > 0.05; **, P < 0.01. | |
Figure 7. Simultaneous recording of XtTPC3 currents and fluorescence signals of FRET-based PI sensors. (A and C) Representative recordings of currents (middle) and fluorescence signals (bottom) elicited by step pulses (top), in XtTPC3 coexpressed with Ci-VSP and F-TAPP (A) or F-PLC (C). The current traces elicited by a 70-mV step pulse often appeared as inward currents, possibly because of a positive shift of the reversal potential caused by incubation in the low Na+ solution. (B) Comparative plots of the time constants at 70 mV between current and fluorescence. The recordings of currents and fluorescence were fitted to a double exponential function, and the kinetics of the increasing (top) components and decreasing (bottom) components were compared between current and fluorescence. "I" and "F" in each axis indicate the current and fluorescence, respectively. (D and E) Representative recordings of currents (middle) and fluorescence (bottom) elicited by +50-mV step pulses (top), in XtTPC3 with F-TAPP (D) or F-PLC (E), without Ci-VSP. (F) Plots of the normalized peak value of ΔFRET as shown in D and E. Error bars represent the SD of four or five independent experiments. | |
Figure 8. Effects of PI injection in oocytes expressing XtTPC3. (A) Representative current traces in the PI injection experiments. The protocol to obtain the two parameters at 50 mV and 100 mV (top), and representative current traces injected with water (middle) or 0.2 mM PI(3,4)P2 (bottom) are shown. For each injection, the traces before injection (black; 0 s), after injection (red; 700 s) and after an induction stimulus (blue; 800 s) are shown. (B) Time courses of the effects of 0.2-mM PI injection on time constants of activation at 100 mV (top) and the amplitude of inward currents (bottom). PIs or water were injected after the second pulse. After the eighth pulse, the 80-mV depolarization stimulus was given twice for 10 s to fully induce XtTPC3. The current amplitude at each time was normalized by that after induction. The points of injection and induction are indicated by black arrows. The colors of the circles above the plots correspond to those in A. (C) Time constants of activation at 100 mV before injection (black; 0 s) and after injection (red; 700 s) for each type of injected PI. (D) The normalized amplitudes elicited by 50 mV at 700 s for each type of injected PI. Error bars represent the SD of five independent experiments. *, P < 0.05; **, P < 0.01. | |
Figure 9. PI perfusion of inside-out patches expressing XtTPC3. (A) Representative current recordings before (black) and after (red) PI perfusion in inside-out patches from Xenopus oocytes expressing XtTPC3. (B) Sequential perfusion of PI(3,4)P2 and PI(3,5)P2 to XtTPC3. Step pulses were given every 10 s. The dotted inset is an expanded view of the tail currents. (C) A plot of the fold increase of tail current amplitude by each PI perfusion (0.5 µM) as shown in A. *, P < 0.05. | |
Figure 10. Identification of the residues critical for both induction and PI(3,4)P2 recognition. (A) Overall structure of domain I of the XtTPC3 model. The regions shown in green correspond to those shown in C. The two dotted lines indicate the supposed lipid bilayers. The view from the eye symbol is expanded in B. (B) The putative PI binding site in the XtTPC3 model. The MmTPC1 structure (PDB accession no. 6C9A) was superimposed on the XtTPC3 model, and only some side chains and the PI molecule are depicted. (C) Sequence alignment of XtTPC3 and MmTPC1. The numbers of the sequences are shown at the top (XtTPC3) and at the bottom (MmTPC1). Basic residues are shown in blue. (D and E) Representative traces evoked by a 100-mV test pulse (D) and G–V relationships before (closed) and after (open) induction (E) in WT, R187Q, and R297Q. (F) Plots of V1/2 in a series of mutants before (black) and after (red) induction. **, P < 0.05; n.s., not significant at P > 0.05. (G) The relationships between time constants of activation and prepulse voltages in R187Q only (black) and coexpressed with Ci-VSP (red). The dotted line indicates the data of WT with Ci-VSP shown in Fig. 4 G. (H) Representative current traces from the R187Q mutant in an inside-out patch. Currents were elicited by a +120-mV step pulse from the −30-mV holding potential. (I) Fold increase in tail current amplitude by 0.5 µM PI(3,4)P2 and PI(3,5)P2 perfusion. Error bars represent the SD of five independent experiments. | |
Figure 11. Alignment of amino acid sequences of TPC1-3 orthologues. Secondary structures were assigned based on the cryo-EM structure of MmTPC1. The filled circles above the XtTPC sequence indicate the residues that are less important (black) or critical (red) for induction, respectively. The aqua circles above the MmTPC1 sequence indicate the residues that are important for PI(3,5)P2 recognition. The species of the orthologues are as follows: zebrafish (D. rerio, Dr), dog (Canis lupus, Cl), cattle (Bos taurus, Bt), chicken (Gallus gallus, Gg), horse (Equus caballus, Ec), rabbit (Orytolagus cuniculus, Oc). and Arabidopsis thaliana (At). | |
Figure 12. Effects on induction of mutation of residues in the H1-IS0 loop. (A) Structural view of the H1-IS0 loop and surrounding region in the XtTPC3 model. (B) Sequence alignment of the H1-IS0 loop region in the N-terminal cytoplasmic region of XtTPC3 and MmTPC1. The numbering of the sequences is shown at the top (XtTPC3) and at the bottom (MmTPC1). (C) G–V relationships of XtTPC3 with mutations at Asn52, Arg53, and Asn54. (D) Plots of V1/2 in a series of mutants before (black) and after (red) induction. (E) Representative current traces of R53D/N54K in oocytes injected with vehicle and PI(3,4)P2. In each case, the traces before injection (black), after injection (red), and after induction (blue) were obtained by the same protocol as shown in Fig. 8 A. (F) Time constants of activation at 100 mV before (black) and after (red) injection are shown for each type of injected PI. Error bars represent the SD of five to seven independent experiments. n.s., not significant at P > 0.05; *, P < 0.05; **, < 0.01. | |
Figure 13. Critical importance of the interaction between IS6 and IIS6 to induction. (A) A view of the XtTPC3 structural model in the PI(3,4)P2-bound open state, depicting the regions related to PI(3,4)P2 binding and coupling with voltage dependence. Domain I and domain II in one monomer are depicted in green and aqua, respectively. Those in the other monomer are shown in white and gray, and the VSDs are omitted for clarity. VSDs and pore domains are shown as ribbon and cylinder models, respectively. The region indicated by the dotted circle shows the interaction that bridges the DI-S6 and DII-S6 (Tyr293, Arg297, and Glu665). (B) Structural view of the proximal region between the cytosolic parts of IS6 and IIS6 in the superimposed structures of XtTPC3 and MmTPC1. The carbon atoms of side chains are shown in green for XtTPC3 and in white for MmTPC1. Dashed lines indicate putative hydrogen bonds with distances of 2.6–3.3 Å in the model structures. (C) Representative current traces of a series of channels with mutations at Tyr293, Arg297, and Glu665 before (black) and after (red) induction. Currents were evoked by a 100-mV pulse from −60 mV. (D) G–V relationships of channels with mutations at Tyr293, Arg297, or Glu665. (E) Plots of V1/2 in a series of mutants of Tyr293, Arg297, and Glu665 before (black) and after (red) induction. #, the V1/2 values before induction in E665Q and Y293H could not be obtained because the currents were too small. Error bars represent the SD of five to seven independent experiments. **, P < 0.01. | |
Figure 14. Extracellular pH dependent change of XtTPC3 current in the whole cell configuration. (A) Representative current recordings from Xenopus oocytes expressing XtTPC3 by two-electrode voltage clamp in extracellular pH (pHo) 7.4 (left, top), in pHo 4.6 (right, top) and upon return to pHo 7.4 (left, bottom). Currents were elicited from −60-mV holding potential to various step pulses in the same protocol to obtain the G–V relationships after induction as shown in Fig. 1 C, right. (B) Plots of the tail current amplitude normalized to that before pH change. Error bars represent the SD of five independent experiments. | |
Figure 15. A structural model for induction, PI(3,4)P2-induced modulation of voltage dependence. The induction model. The model depiction is based on the structure shown in Fig. 13 A. In the apo-state, or before induction (left), a strong voltage stimulus (red arrow) is required to direct both DI-S6 and DII-S6, which are thought to be weakly linked, to the open conformation. In the PI(3,4)P2-bound state, or after induction (right), PI(3,4)P2 binding facilitates the opening of the activation gate through the tight interaction between DI-S6 and DII-S6, which is formed in this state. As some of the required energy to move both S6s is provided by PI(3,4)P2 binding (blue arrow), less voltage stimulus is required to open the activation gate than in the case of the apo state. Vm indicates membrane voltage. |
References [+] :
Aggarwal,
Contribution of the S4 segment to gating charge in the Shaker K+ channel.
1996, Pubmed,
Xenbase
Aggarwal, Contribution of the S4 segment to gating charge in the Shaker K+ channel. 1996, Pubmed , Xenbase
Armstrong, Charge movement associated with the opening and closing of the activation gates of the Na channels. 1974, Pubmed
Bae, Profilin1 regulates PI(3,4)P2 and lamellipodin accumulation at the leading edge thus influencing motility of MDA-MB-231 cells. 2010, Pubmed
Balla, Phosphoinositides: tiny lipids with giant impact on cell regulation. 2013, Pubmed
Barish, A transient calcium-dependent chloride current in the immature Xenopus oocyte. 1983, Pubmed , Xenbase
Baud, Sodium channels induced by depolarization of the Xenopus laevis oocyte. 1982, Pubmed , Xenbase
Boucrot, Endophilin marks and controls a clathrin-independent endocytic pathway. 2015, Pubmed
Braccini, PI3K-C2γ is a Rab5 effector selectively controlling endosomal Akt2 activation downstream of insulin signalling. 2015, Pubmed
Brailoiu, Essential requirement for two-pore channel 1 in NAADP-mediated calcium signaling. 2009, Pubmed
Brailoiu, An ancestral deuterostome family of two-pore channels mediates nicotinic acid adenine dinucleotide phosphate-dependent calcium release from acidic organelles. 2010, Pubmed
Cai, Degeneration of an intracellular ion channel in the primate lineage by relaxation of selective constraints. 2010, Pubmed
Calcraft, NAADP mobilizes calcium from acidic organelles through two-pore channels. 2009, Pubmed
Cang, mTOR regulates lysosomal ATP-sensitive two-pore Na(+) channels to adapt to metabolic state. 2013, Pubmed
Cang, The voltage-gated sodium channel TPC1 confers endolysosomal excitability. 2014, Pubmed
Cang, A non-inactivating high-voltage-activated two-pore Na⁺ channel that supports ultra-long action potentials and membrane bistability. 2014, Pubmed , Xenbase
Cross, A fast block to polyspermy in frogs mediated by changes in the membrane potential. 1980, Pubmed
Dong, PI(3,5)P(2) controls membrane trafficking by direct activation of mucolipin Ca(2+) release channels in the endolysosome. 2010, Pubmed
García-Rúa, Endolysosomal two-pore channels regulate autophagy in cardiomyocytes. 2016, Pubmed
Grey, An electrical block is required to prevent polyspermy in eggs fertilized by natural mating of Xenopus laevis. 1982, Pubmed , Xenbase
Grimm, Allosteric substrate switching in a voltage-sensing lipid phosphatase. 2016, Pubmed , Xenbase
Guo, Structure of the voltage-gated two-pore channel TPC1 from Arabidopsis thaliana. 2016, Pubmed
Hehl, Phosphoinositide 3-kinase-gamma induces Xenopus oocyte maturation via lipid kinase activity. 2001, Pubmed , Xenbase
Ho, Molecular mechanism for sodium-dependent activation of G protein-gated K+ channels. 1999, Pubmed , Xenbase
Hogan, The phosphoinositol 3,4-bisphosphate-binding protein TAPP1 interacts with syntrophins and regulates actin cytoskeletal organization. 2004, Pubmed
Kintzer, Structure, inhibition and regulation of two-pore channel TPC1 from Arabidopsis thaliana. 2016, Pubmed
Kume, Two mutations at different positions in the CNBH domain of the hERG channel accelerate deactivation and impair the interaction with the EAG domain. 2018, Pubmed , Xenbase
Kurokawa, 3' Phosphatase activity toward phosphatidylinositol 3,4-bisphosphate [PI(3,4)P2] by voltage-sensing phosphatase (VSP). 2012, Pubmed
Li, Phosphatidylinositol (3,4) bisphosphate-specific phosphatases and effector proteins: A distinct branch of PI3K signaling. 2015, Pubmed
Li, Phosphatidylinositol-3,4-Bisphosphate and Its Binding Protein Lamellipodin Regulate Chemotaxis of Malignant B Lymphocytes. 2016, Pubmed
Lin, Lysosomal two-pore channel subtype 2 (TPC2) regulates skeletal muscle autophagic signaling. 2015, Pubmed
Liu, Molecular cloning of an amphibian insulin receptor substrate 1-like cDNA and involvement of phosphatidylinositol 3-kinase in insulin-induced Xenopus oocyte maturation. 1995, Pubmed , Xenbase
Manning, AKT/PKB Signaling: Navigating the Network. 2017, Pubmed
Marat, Phosphatidylinositol 3-phosphates-at the interface between cell signalling and membrane traffic. 2016, Pubmed
Murata, Phosphoinositide phosphatase activity coupled to an intrinsic voltage sensor. 2005, Pubmed , Xenbase
Patel, The endo-lysosomal system as an NAADP-sensitive acidic Ca(2+) store: role for the two-pore channels. 2011, Pubmed
Patel, NAADP-evoked Ca2+ signals through two-pore channel-1 require arginine residues in the first S4-S5 linker. 2017, Pubmed
Peiter, The vacuolar Ca2+-activated channel TPC1 regulates germination and stomatal movement. 2005, Pubmed
Peres, Sodium conductance and the activation potential in Xenopus laevis eggs. 1985, Pubmed , Xenbase
Pochynyuk, Molecular determinants of PI(4,5)P2 and PI(3,4,5)P3 regulation of the epithelial Na+ channel. 2007, Pubmed
Posor, Spatiotemporal control of endocytosis by phosphatidylinositol-3,4-bisphosphate. 2013, Pubmed
Ramos, Two-pore channels function in calcium regulation in sea star oocytes and embryos. 2014, Pubmed
Ratzan, Voltage sensitive phosphoinositide phosphatases of Xenopus: their tissue distribution and voltage dependence. 2011, Pubmed , Xenbase
Rohács, Distinct specificities of inwardly rectifying K(+) channels for phosphoinositides. 1999, Pubmed , Xenbase
Rohács, Specificity of activation by phosphoinositides determines lipid regulation of Kir channels. 2003, Pubmed , Xenbase
Ruas, Purified TPC isoforms form NAADP receptors with distinct roles for Ca(2+) signaling and endolysosomal trafficking. 2010, Pubmed
Sakurai, Ebola virus. Two-pore channels control Ebola virus host cell entry and are drug targets for disease treatment. 2015, Pubmed
Sasaki, The PtdIns(3,4)P(2) phosphatase INPP4A is a suppressor of excitotoxic neuronal death. 2010, Pubmed
Schwede, SWISS-MODEL: An automated protein homology-modeling server. 2003, Pubmed
Seoh, Voltage-sensing residues in the S2 and S4 segments of the Shaker K+ channel. 1996, Pubmed , Xenbase
She, Structural insights into the voltage and phospholipid activation of the mammalian TPC1 channel. 2018, Pubmed
She, Structural mechanisms of phospholipid activation of the human TPC2 channel. 2019, Pubmed
Tong, Direct activation of the epithelial Na(+) channel by phosphatidylinositol 3,4,5-trisphosphate and phosphatidylinositol 3,4-bisphosphate produced by phosphoinositide 3-OH kinase. 2004, Pubmed
Wang, TPC proteins are phosphoinositide- activated sodium-selective ion channels in endosomes and lysosomes. 2012, Pubmed
Xu, Lysosomal physiology. 2015, Pubmed
Yu, The VGL-chanome: a protein superfamily specialized for electrical signaling and ionic homeostasis. 2004, Pubmed
Zeng, Structural determinants and specificities for ROMK1-phosphoinositide interaction. 2002, Pubmed , Xenbase
Zhang, Depolarization increases phosphatidylinositol (PI) 4,5-bisphosphate level and KCNQ currents through PI 4-kinase mechanisms. 2010, Pubmed , Xenbase
Zhang, Phosphoinositide isoforms determine compartment-specific ion channel activity. 2012, Pubmed
Zong, The two-pore channel TPCN2 mediates NAADP-dependent Ca(2+)-release from lysosomal stores. 2009, Pubmed