Abstract
Recently there has been a flurry of interest in the regulation of the homo-dimeric calcium-activated chloride channel ANO1 (also known as TMEM16A) by phosphatidylinositol (4,5)-bisphosphate (PI(4,5)P2). These recent studies show that upon Ca2+ binding, PI(4,5)P2 cooperates to maintain the conductive state of ANO1. PI(4,5)P2 does so by binding to sites or modules on the protein’s cytosolic side. These findings add a new function to the PI(4,5)P2 repertoire and a new dimension to ANO1 gating.
Keywords: phospholipids, anion channels, ion channel gating, calcium, ion channel regulation
Graphical Abstract

1. Introduction
The ANO1 anion channel (also known as TMEM16A) controls a mesmerising number of important physiological functions [1–7]. Elevation of cytosolic Ca2+ concentration opens ANO1 to conduct Cl− ions, but relatively little is known about other pathways that regulate the channel. Recently, considerable attention has become focussed on ANO1 regulation by phosphatidylinositol (4,5)-bisphosphate (PI(4,5)P2) [8–14]. PI(4,5)P2 is a low-abundance phospholipid in the inner leaflet of the plasma membrane that has a remarkable number of cellular functions including regulation of cytoskeletal organization, membrane trafficking, and ion channel gating [15–18]. While the role of PI(4,5)P2 in cation channel regulation is well-known, there is relatively little information about its role in anion channel gating.
Studies aimed at understanding gating and regulation of ANO1 were given a tremendous boost in 2008 by its cloning [19–21] followed in 2017 by cryo-EM structure determinations [22–24]. The structures reveal dimers with each subunit consisting of 10 transmembrane segments with the dimer interface formed in large part by the C-termini of the subunits. Each monomer has a hydrophilic pore surrounded by transmembrane helices TM4–7 (Fig 1). The large cytoplasmic and extracellular vestibules narrow to form a constricted pore with a radius of ~ 1 Å near the cytoplasmic side of the membrane. Structural analysis, which confirmed earlier mutagenesis and patch clamp studies, revealed the presence of two Ca2+ ions coordinated by sidechain oxygens of glutamic and aspartic acid residues located in a hydrophilic pocket situated close to the cytosolic end of the pore. A comparison of apo and Ca2+-bound structures suggested that Ca2+ binding induced a major conformational change of the cytoplasmic half of TM6 [22–25]. At the same time, binding of Ca2+ reduces the electronegative field within the narrow neck of the pore because the binding pocket is situated in the pore wall [26]. The helix rearrangement sandwiches Ca2+ ions between TM6, TM7, and TM8 and at the same time opens the permeation pathway. From this finding, one might fantasize that Ca2+ binding stabilizes an open conformation of ANO1, but the mechanisms are certainly more complicated because high Ca2+concentrations also cause open channels rapidly to inactivate. Indeed, the pores of Ca2+-bound cryo-EM structures are too small to conduct Cl− ions and probably represent inactivated conformations. Surprisingly, application of a water-soluble analog of PI(4,5)P2 (diC8-PI(4,5)P2) to inside-out excised patches nearly abolishes Ca2+-induced inactivation, whereas depletion of PI(4,5)P2 from the plasma membrane using Danio rerio voltage-sensitive phosphatase decreases ANO1 activity [11, 12]. PI(4,5)P2 could likely be the cofactor necessary to maintain the conductive state of ANO1.
Figure 1.
Homo-dimeric structure of Ca2+-bound ANO1. The structure with two independent subunits (maroon and dark cyan) is shown in the plane of the membrane. The dimer interface formed by the apposition of the two C-terminus is perpendicular to the membrane. Ten transmembrane domains per monomer are depicted. TM6 indicated by light blue arrow rotates from TM4 towards TM8 upon Ca2+ binding to reach a straight conformation. The anion permeation pathway with large vestibules at both ends within each monomer is revealed by water molecules (grey). The Ca2+ pocket, located towards the intracellular side almost in contact with the permeation pathway, houses two Ca2+ ions (red spheres).
2. PI(4,5)P2 helps ANO1 gating to evade Ca2+-induced inactivation
ANO1 is gated by physiological intracellular Ca2+ concentrations (0.1–5 μM) in a voltage-dependent manner. Under this condition, repetitive depolarisations activate a current that reaches a steady plateau without decaying. However, simultaneous activation of the depolarization-activated phosphatase to dephosphorylate PI(4,5)P2 rapidly induces current decay (Fig 2A). This phenomenon is reversed by shutting off the phosphatase by holding cells at negative voltages, providing that ATP is present in the intracellular solution [11, 12]. Presumably, under these conditions PI(4,5)P2 is re-synthetized to restore its membrane content. Similarly, when ANO1 is activated in excised inside-out patches exposed to low Ca2+, PI(4,5)P2 fully rescues channels that have previously been stripped of PI(4,5)P2 using poly-lysine [11, 12]. The effect of PI(4,5)P2 is greatest at lower Ca2+ concentrations and decreases with depolarization. However, the ability of PI(4,5)P2 to regulate ANO1 at high intracellular Ca2+ (i.e. 100 μM) is different. Under this condition, fully active ANO1 channels rapidly inactivate, but inactivation is prevented when PI(4,5)P2 is provided along with high intracellular Ca2+ (Fig 2B). Surprisingly, PI(4,5)P2 is unable to rescue channels previously inactivated by high Ca2+ [11, 12]. Somehow high Ca2+ induces a profound inactivated state, presumably in a process that involves removing PI(4,5)P2.
Figure 2.
ANO1 opening requires PI(4,5)P2. A: Depletion of PI(4,5)P2 from the plasma membrane induces a progressive decay (from black to red) of the ANO1 Ca2+-activated Cl− current in a HEK cell also expressing the Danio rerio voltage-sensitive phosphatase (Dr-VSP). The cell was dialyzed with 0.2 μM Ca2+ and currents were induced by applying a +80 mV step from a holding potential of −30 mV every 2s during 250 s. B: Ca2+-induced inactivation of ANO1 is dampened by diC8-PIP2, a PI(4,5)P2 analog. Currents were induced by applying 100 μM Ca2+ (top bar) to inside-out patches held at +100 mV in the absence (black) and in the presence of 25 μM diC8-PIP2 (blue).
3. Several PI(4,5)P2 binding sites exist on ANO1
PI(4,5)P2 could facilitate ANO1 gating by electrostatically attracting Ca2+ to the channel or by directly altering gating, as has been shown previously for certain potassium channels [15]. Guided by the Ca2+-bound cryo-EM structure of mouse TMEM16A, Huanghe Yang’s lab [12] performed an alanine scanning analysis of basic lysine and arginine residues located in the vicinity of the plasma membrane inner leaflet. To gauge PI(4,5)P2 sensitivity, they used the rate of channel inactivation induced by 100 μM Ca2+. They found that residues R451, K461, R482, K567, R575, and K579 (numbering based on mouse ANO1 variant a, lacking the EAVK segment) form a putative PI(4,5)P2 binding site near the cytosolic interface of TMs 3–5. Accordingly, mutations in the PI(4,5)P2 binding site were nearly insensitive to PI(4,5)P2 application and the mutant channels inactivated faster than wild type channels [12]. Ko and co-workers have confirmed that R486 (equivalent to R482 in [12]) in the first intracellular loop is part of a putative PI(4,5)P2 binding site [27].
In addition to this site identified by Le et al, other PI(4,5)P2 binding sites have been found on ANO1. The Hartzell lab performed a glutamine scanning analysis combined with patch clamp and molecular dynamics simulation of lysine and arginine residues located within 10 Å of the cytoplasmic membrane interface, including those residues located in the cytoplasmic N-terminus predicted to interact with PI(4,5)P2 [14]. In contrast to Le et al., they measured the stimulatory effect of 10 μM diC8-PI(4,5)P2 on wild type and mutant ANO1 currents activated by low (270 nM) Ca2+ in inside-out patches. This approach yielded three putative binding sites: site A near the dimer interface, site B at the cytoplasmic end of TM6, and site C in the short intracellular loop between TM2 and TM3. Of these sites, site A is close to the site located by Le et al., and near a lipid-filled cavity at the dimer interface according to the cryo-EM structure of TMEM16A. Interestingly, some of the residues in sites B and C identified as being important in regulation channel amplitude in low Ca2+ by the Hartzell lab were not found by the Yang lab to be important in regulating inactivation. This difference may be explained if PI(4,5)P2 has different effects on the fully open and partially open conformations of the channel as proposed by Le et al. [12].
It appears that PI(4,5)P2 regulation may be a general feature of the ANO family. There are 10 ANO genes in mammals and while two of them (ANO1 and ANO2) are clearly Ca2+-activated Cl− channels, others, notably ANO6, are phospholipid scramblases [28]. Previous studies from the Jan lab have shown that ANO6 is regulated by PI(4,5)P2 and they have identified a PI(4,5)P2 binding site in the N-terminus of the protein [29]. The Hartzell lab found amino acids in the N-terminus of ANO1 that alter PI(4,5)P2 effects when they are mutated [14], but the molecular mechanisms by which these residues regulate the channel remain uncertain because the cryo-EM structures do not fully model this region of the protein. However, the multiplicity of PI(4,5)P2 “binding sites” raises interesting questions. Both structural and electrophysiological data suggest that ANO1 apparently can adopt a number of different conformations. Different conformational states likely have different PI(4,5)P2 binding properties. Mutations that stabilize different conformational states (such as K579 [12] or R579 [14]), but do not directly coordinate PI(4,5)P2, could have profound effects on the interpretation of assays that rely on channel function to assess PI(4,5)P2 binding. This emphasizes the importance of melding electrophysiological studies with structural, computational, and biochemical approaches.
4. How can PI(4,5)P2 prevent ANO1 inactivation and exactly how does it participate in ANO1 gating?
While ANO1 gating requires Ca2+ binding to the Ca2+ pocket, high Ca2+ concentrations could trigger depletion of PI(4,5)P2 to induce ANO1 inactivation as has been suggested for ANO6 [29]. Inactivation could be caused by pore collapse as suggested by the Ca2+-bound cryo-EM structures showing a pore with a radius smaller than that of Cl− ions [22–24]. Indirect evidence from the Yang lab gathered using highly permeant anions, pore blockers, and pore mutants suggest that an occupied pore inactivates slower than an empty pore probably by precluding pore collapse [12]. A direct interaction of PI(4,5)P2 with the gating machinery of ANO1 was suggested by molecular dynamics simulations. The simulation suggests that binding of PI(4,5)P2 to sites A and B/C induces a rotation of the cytoplasmic half of TM6 that dilates the conduction pathway ~30%, but still smaller than a Cl− ion [14]. Co-incidentally, the Ca2+-bound structure of ANO1 shows that the cytoplasmic end of TM6 adopts a straight conformation and forms one side of the channel pore. The fact that PI(4,5)P2 alters the conformational state of the Ca2+ bound structure suggests that PI(4,5)P2 acts by inhibiting Ca2+-induced pore collapse and that the cooperative action of PI(4,5)P2 and Ca2+ stabilize the conducting conformation. Further analysis is necessary to determine if pore collapse is prevented by rotation of TM6.
5. Paradigm shift of ANO1 structure and gating: modular architecture and cooperative gating.
From these studies a new, clearer picture of ANO is starting to emerge: a modular structure of the pore and a gating mechanism that depends on the cooperative action of PI(4,5)P2 and Ca2+. Le et al found that channels with mutations in the PI(4,5)P2 binding site displayed rapid inactivation but unchanged apparent Ca2+ sensitivity [12]. Surprisingly, mutating residues near the Ca2+ binding pocket decreased the apparent Ca2+ affinity but did not change inactivation. This result led to the idea that two modules control the anion permeation pathway: a PI(4,5)P2 binding regulatory module (TM3-TM5) responsible for inactivation and a Ca2+ binding gating module (TM6-TM8) responsible for channel activation (Fig. 3A). These modules are arranged around the pore but they obviously do not operate in isolation. Le at al. [12] show that with wild type ANO1, PI(4,5)P2 alters the Ca2+ sensitivity of the channel and that inactivation may be induced by a Ca2+-dependent dissociation of PI(4,5)P2 from the regulatory module. In this structural scenario, gating would depend on direct binding of PI(4,5)P2 and Ca2+ to their respective modules (Fig. 3B). Under resting Ca2+ conditions ANO1 directly interacts with PI(4,5)P2 via a complex set of binding sites. A sudden Ca2+ increase will rapidly activate ANO1 by occupying the Ca2+ pocket. Sustained activation will depend on both binding of PI(4,5)P2 and Ca2+ sitting in the hydrophilic pocket. The combined actions of PI(4,5)P2 and Ca2+ will ensure an open state by holding the cytoplasmic end of TM6 from collapsing the pore and by partially neutralizing the electronegative field in the narrowest section of the pore. Structural information about different conformational states of ANO1 together with molecular dynamic simulations will help resolve these remaining questions about ANO1 gating.
Figure 3.
Modular architecture and cooperative gating of ANO1. A: Modular structure of the ANO1 pore. The scheme shows one pore filled with water molecules (grey) surrounded by the PI(4,5)P2 regulatory module (TM3-TM5 in green) and by the Ca2+ binding module (TM6-TM8 in yellow). B: Hypothetical co-operative ANO1 gating mechanism. (1) When intracellular Ca2+ is low, ANO1 is in a closed state with PI(4,5)P2 bound. In this state, ANO1 can be opened by strong depolarization [30] and removal of PI(4,5)P2 decreases current amplitude. The current strongly rectifies outwardly because of the strong electronegative field at the intracellular end of the channel due to the acidic residues in the Ca2+ binding pocket and bound PI(4,5)P2 molecules [26]. (2) When Ca2+ increases and binds to the Ca2+ pocket, the cytoplasmic half of TM6 straightens with the help of PI(4,5)P2 [14]. This relaxation may be partly a result of neutralization of the negative electric field by Ca2+ binding. The open state will be determined by the balance between the electric field and the position of the cytoplasmic end of TM6. The current will be less rectifying as intracellular Ca2+ reaches near 5 μM. Dephosphorylation of membrane PI(4,5)P2 with the voltage-sensitive phosphatase or removal of PI(4,5)P2 by poly-L-lysine will cause the pore to partially close because PI(4,5)P2 is not available to adjust the conformation of TM6. But, the pore would not be inactivated (collapsed) because the electric field has not been fully neutralized. Also, in the absence of PI(4,5)P2, the voltage dependence of open probability has shifted such that a strong depolarization or an increase in Ca2+ is necessary to open the pore. (3) Binding of additional Ca2+ will fully neutralize the electric field and presumably remove PI(4,5)P2. This condition will inactivate the channel by collapsing the pore.
Highlights.
The Ca2+-activated anion channel, ANO1 or TMEM16A, is regulated by phosphatidylinositol 4,5-bisphosphate.
Acknowledgements
Thanks to Huanghe Yang, Son Le, and Steven Foltz for helpful comments. Research on ANO 1 channels is supported by grants FC 2016-01-1955 and 219949 from CONACYT, Mexico (JA) and by grants EY0114852, AR067786, and GM132598 from the National Institutes of Health (HCH).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Pedemonte N and Galietta LJ, Structure and Function of TMEM16 Proteins (Anoctamins). Physiol Rev, 2014. 94(2): p. 419–59. [DOI] [PubMed] [Google Scholar]
- 2.Ji Q, et al. , Recent advances in TMEM16A: Structure, function, and disease. J Cell Physiol, 2019. 234(6): p. 7856–7873. [DOI] [PubMed] [Google Scholar]
- 3.Crottes D and Jan LY, The multifaceted role of TMEM16A in cancer. Cell Calcium, 2019. 82: p. 102050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Whitlock JM and Hartzell HC, Anoctamins/TMEM16 Proteins: Chloride Channels Flirting with Lipids and Extracellular Vesicles. Annu Rev Physiol, 2017. 79: p. 119–143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kunzelmann K, et al. , Modulating Ca(2+) signals: a common theme for TMEM16, Ist2, and TMC. Pflugers Arch, 2016. 468(3): p. 475–90. [DOI] [PubMed] [Google Scholar]
- 6.Oh U and Jung J, Cellular functions of TMEM16/anoctamin. Pflugers Arch, 2016. 468(3): p. 443–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Picollo A, Malvezzi M, and Accardi A, TMEM16 Proteins: Unknown Structure and Confusing Functions. J Mol Biol, 2015. 427(1): p. 94–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Tian Y, et al. , Control of TMEM16A by INO-4995 and other inositolphosphates. Br J Pharmacol, 2013. 168(1): p. 253–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Pritchard HA, et al. , Inhibitory role of phosphatidylinositol 4,5-bisphosphate on TMEM16A-encoded calcium-activated chloride channels in rat pulmonary artery. Br J Pharmacol, 2014. 171(18): p. 4311–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ta CM, et al. , Contrasting effects of phosphatidylinositol 4,5-bisphosphate on cloned TMEM16A and TMEM16B channels. Br J Pharmacol, 2017. 174(18): p. 2984–2999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.De Jesus-Perez JJ, et al. , Phosphatidylinositol 4,5-bisphosphate, cholesterol, and fatty acids modulate the calcium-activated chloride channel TMEM16A (ANO1). Biochim Biophys Acta, 2017. 1863(3): p. 299–312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Le SC, et al. , Molecular basis of PIP2-dependent regulation of the Ca(2+)-activated chloride channel TMEM16A. Nat Commun, 2019. 10(1): p. 3769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Tembo M, et al. , Phosphatidylinositol 4,5-bisphosphate (PIP2) and Ca(2+) are both required to open the Cl(−) channel TMEM16A. J Biol Chem, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yu K, et al. , A network of phosphatidylinositol 4,5-bisphosphate binding sites regulates gating of the Ca(2+)-activated Cl(−) channel ANO1 (TMEM16A). Proc Natl Acad Sci U S A, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hille B, et al. , Phosphoinositides regulate ion channels. Biochim Biophys Acta, 2015. 1851(6): p. 844–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hansen SB, Lipid agonism: The PIP2 paradigm of ligand-gated ion channels. Biochim Biophys Acta, 2015. 1851(5): p. 620–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Logothetis DE, et al. , Phosphoinositide control of membrane protein function: a frontier led by studies on ion channels. Annu Rev Physiol, 2015. 77: p. 81–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Suh BC and Hille B, PIP2 is a necessary cofactor for ion channel function: how and why? Annu Rev Biophys, 2008. 37: p. 175–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Caputo A, et al. , TMEM16A, A Membrane Protein Associated With Calcium-Dependent Chloride Channel Activity. Science, 2008. 322: p. 590–594. [DOI] [PubMed] [Google Scholar]
- 20.Yang YD, et al. , TMEM16A confers receptor-activated calcium-dependent chloride conductance. Nature, 2008. 455: p. 1210–1215. [DOI] [PubMed] [Google Scholar]
- 21.Schroeder BC, et al. , Expression Cloning of TMEM16A as a Calcium-Activated Chloride Channel Subunit. Cell, 2008. 134: p. 1019–1029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Paulino C, et al. , Activation mechanism of the calcium-activated chloride channel TMEM16A revealed by cryo-EM. Nature, 2017. 552(7685): p. 421–425. [DOI] [PubMed] [Google Scholar]
- 23.Paulino C, et al. , Structural basis for anion conduction in the calcium-activated chloride channel TMEM16A. Elife, 2017. 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Dang S, et al. , Cryo-EM structures of the TMEM16A calcium-activated chloride channel. Nature, 2017. 552(7685): p. 426–429. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Peters CJ, et al. , The Sixth Transmembrane Segment Is a Major Gating Component of the TMEM16A Calcium-Activated Chloride Channel. Neuron, 2018. 97(5): p. 1063–1077 e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Lam AK and Dutzler R, Calcium-dependent electrostatic control of anion access to the pore of the calcium-activated chloride channel TMEM16A. Elife, 2018. 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Ko WJ, S.-R.;Park C-G; Nam JH;Hille B;Suh BC, Allosteric Modulation of Ca2+-Activated Cl- Channels TMEM16A by PIP2 and CAMKII Biophysical Journal, 2019. 114: p. 41a. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Suzuki J, et al. , Calcium-dependent phospholipid scrambling by TMEM16F. Nature, 2010. 468(7325): p. 834–8. [DOI] [PubMed] [Google Scholar]
- 29.Ye W, et al. , Phosphatidylinositol-(4, 5)-bisphosphate regulates calcium gating of small-conductance cation channel TMEM16F. Proc Natl Acad Sci U S A, 2018. 115(7): p. E1667–E1674. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Xiao Q, et al. , Voltage- and calcium-dependent gating of TMEM16A/Ano1 chloride channels are physically coupled by the first intracellular loop. Proc. Natl. Acad. Sci, 2011. 108(21): p. 8891–8896. [DOI] [PMC free article] [PubMed] [Google Scholar]



