Article

Structure of a volume-regulated anion channel of the LRRC8 family

Received:
Accepted:
Published:

Abstract

Volume-regulated anion channels are activated in response to hypotonic stress. These channels are composed of closely related paralogues of the leucine-rich repeat-containing protein 8 (LRRC8) family that co-assemble to form hexameric complexes. Here, using cryo-electron microscopy and X-ray crystallography, we determine the structure of a homomeric channel of the obligatory subunit LRRC8A. This protein conducts ions and has properties in common with endogenous heteromeric channels. Its modular structure consists of a transmembrane pore domain followed by a cytoplasmic leucine-rich repeat domain. The transmembrane domain, which is structurally related to connexin proteins, is wide towards the cytoplasm but constricted on the outside by a structural unit that acts as a selectivity filter. An excess of basic residues in the filter and throughout the pore attracts anions by electrostatic interaction. Our work reveals the previously unknown architecture of volume-regulated anion channels and their mechanism of selective anion conduction.

  • Subscribe to Nature for full access:

    $199

    Subscribe

Additional access options:

Already a subscriber?  Log in  now or  Register  for online access.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. 1.

    Jentsch, T. J. VRACs and other ion channels and transporters in the regulation of cell volume and beyond. Nat. Rev. Mol. Cell Biol. 17, 293–307 (2016).

  2. 2.

    Hoffmann, E. K., Lambert, I. H. & Pedersen, S. F. Physiology of cell volume regulation in vertebrates. Physiol. Rev. 89, 193–277 (2009).

  3. 3.

    Okada, Y., Sato, K. & Numata, T. Pathophysiology and puzzles of the volume-sensitive outwardly rectifying anion channel. J. Physiol. (Lond.) 587, 2141–2149 (2009).

  4. 4.

    Pasantes-Morales, H., Murray, R. A., Sánchez-Olea, R. & Morán, J. Regulatory volume decrease in cultured astrocytes. II. Permeability pathway to amino acids and polyols. Am. J. Physiol. 266, C172–C178 (1994).

  5. 5.

    Kimelberg, H. K., Goderie, S. K., Higman, S., Pang, S. & Waniewski, R. A. Swelling-induced release of glutamate, aspartate, and taurine from astrocyte cultures. J. Neurosci. 10, 1583–1591 (1990).

  6. 6.

    Hyzinski-García, M. C., Rudkouskaya, A. & Mongin, A. A. LRRC8A protein is indispensable for swelling-activated and ATP-induced release of excitatory amino acids in rat astrocytes. J. Physiol. (Lond.) 592, 4855–4862 (2014).

  7. 7.

    Nilius, B. et al. Properties of volume-regulated anion channels in mammalian cells. Prog. Biophys. Mol. Biol. 68, 69–119 (1997).

  8. 8.

    Qiu, Z. et al. SWELL1, a plasma membrane protein, is an essential component of volume-regulated anion channel. Cell 157, 447–458 (2014).

  9. 9.

    Voss, F. K. et al. Identification of LRRC8 heteromers as an essential component of the volume-regulated anion channel VRAC. Science 344, 634–638 (2014).

  10. 10.

    Sawada, A. et al. A congenital mutation of the novel gene LRRC8 causes agammaglobulinemia in humans. J. Clin. Invest. 112, 1707–1713 (2003).

  11. 11.

    Kubota, K. et al. LRRC8 involved in B cell development belongs to a novel family of leucine-rich repeat proteins. FEBS Lett. 564, 147–152 (2004).

  12. 12.

    Abascal, F. & Zardoya, R. LRRC8 proteins share a common ancestor with pannexins, and may form hexameric channels involved in cell-cell communication. BioEssays 34, 551–560 (2012).

  13. 13.

    Lee, C. C., Freinkman, E., Sabatini, D. M. & Ploegh, H. L. The protein synthesis inhibitor blasticidin S enters mammalian cells via leucine-rich repeat-containing protein 8D. J. Biol. Chem. 289, 17124–17131 (2014).

  14. 14.

    Planells-Cases, R. et al. Subunit composition of VRAC channels determines substrate specificity and cellular resistance to Pt-based anti-cancer drugs. EMBO J. 34, 2993–3008 (2015).

  15. 15.

    Lutter, D., Ullrich, F., Lueck, J. C., Kempa, S. & Jentsch, T. J. Selective transport of neurotransmitters and modulators by distinct volume-regulated LRRC8 anion channels. J. Cell Sci. 130, 1122–1133 (2017).

  16. 16.

    Syeda, R. et al. LRRC8 proteins form volume-regulated anion channels that sense ionic strength. Cell 164, 499–511 (2016).

  17. 17.

    Nilius, B., Prenen, J., Voets, T., Eggermont, J. & Droogmans, G. Activation of volume-regulated chloride currents by reduction of intracellular ionic strength in bovine endothelial cells. J. Physiol. (Lond.) 506, 353–361 (1998).

  18. 18.

    Maeda, S. et al. Structure of the connexin 26 gap junction channel at 3.5 A resolution. Nature 458, 597–602 (2009).

  19. 19.

    Bennett, B. C. et al. An electrostatic mechanism for Ca2+-mediated regulation of gap junction channels. Nat. Commun. 7, 8770 (2016).

  20. 20.

    Oshima, A., Tani, K. & Fujiyoshi, Y. Atomic structure of the innexin-6 gap junction channel determined by cryo-EM. Nat. Commun. 7, 13681 (2016).

  21. 21.

    Bella, J., Hindle, K. L., McEwan, P. A. & Lovell, S. C. The leucine-rich repeat structure. Cell. Mol. Life Sci. 65, 2307–2333 (2008).

  22. 22.

    Huttlin, E. L. et al. A tissue-specific atlas of mouse protein phosphorylation and expression. Cell 143, 1174–1189 (2010).

  23. 23.

    Zhou, H. et al. Toward a comprehensive characterization of a human cancer cell phosphoproteome. J. Proteome Res. 12, 260–271 (2013).

  24. 24.

    Ullrich, F., Reincke, S. M., Voss, F. K., Stauber, T. & Jentsch, T. J. Inactivation and anion selectivity of volume-regulated anion channels (VRACs) depend on C-terminal residues of the first extracellular loop. J. Biol. Chem. 291, 17040–17048 (2016).

  25. 25.

    Sobolevsky, A. I., Rosconi, M. P. & Gouaux, E. X-ray structure, symmetry and mechanism of an AMPA-subtype glutamate receptor. Nature 462, 745–756 (2009).

  26. 26.

    Matthies, D. et al. Cryo-EM structures of the magnesium channel CorA reveal symmetry break upon gating. Cell 164, 747–756 (2016).

  27. 27.

    Valiunas, V., Cohen, I. S. & Brink, P. R. Defining the factors that affect solute permeation of gap junction channels. Biochim. Biophys. Acta 1860, 96–101 (2018).

  28. 28.

    Goldberg, G. S., Valiunas, V. & Brink, P. R. Selective permeability of gap junction channels. Biochim. Biophys. Acta 1662, 96–101 (2004).

  29. 29.

    Nilius, B., Oike, M., Zahradnik, I. & Droogmans, G. Activation of a Cl current by hypotonic volume increase in human endothelial cells. J. Gen. Physiol. 103, 787–805 (1994).

  30. 30.

    Gaitán-Peñas, H. et al. Investigation of LRRC8-mediated volume-regulated anion currents in Xenopus oocytes. Biophys. J. 111, 1429–1443 (2016).

  31. 31.

    Pedersen, S. F., Okada, Y. & Nilius, B. Biophysics and physiology of the volume-regulated anion channel (VRAC)/volume-sensitive outwardly rectifying anion channel (VSOR). Pflugers Arch. 468, 371–383 (2016).

  32. 32.

    Bryan-Sisneros, A., Sabanov, V., Thoroed, S. M. & Doroshenko, P. Dual role of ATP in supporting volume-regulated chloride channels in mouse fibroblasts. Biochim. Biophys. Acta 1468, 63–72 (2000).

  33. 33.

    Gradogna, A., Gaitán-Peñas, H., Boccaccio, A., Estévez, R. & Pusch, M. Cisplatin activates volume sensitive LRRC8 channel mediated currents in Xenopus oocytes. Channels (Austin) 11, 254–260 (2017).

  34. 34.

    Voets, T. et al. Regulation of a swelling-activated chloride current in bovine endothelium by protein tyrosine phosphorylation and G proteins. J. Physiol. (Lond.) 506, 341–352 (1998).

  35. 35.

    Geertsma, E. R. & Dutzler, R. A versatile and efficient high-throughput cloning tool for structural biology. Biochemistry 50, 3272–3278 (2011).

  36. 36.

    Keefe, A. D., Wilson, D. S., Seelig, B. & Szostak, J. W. One-step purification of recombinant proteins using a nanomolar-affinity streptavidin-binding peptide, the SBP-Tag. Protein Expr. Purif. 23, 440–446 (2001).

  37. 37.

    Bindels, D. S. et al. mScarlet: a bright monomeric red fluorescent protein for cellular imaging. Nat. Methods 14, 53–56 (2017).

  38. 38.

    Reeves, P. J., Callewaert, N., Contreras, R. & Khorana, H. G. Structure and function in rhodopsin: high-level expression of rhodopsin with restricted and homogeneous N-glycosylation by a tetracycline-inducible N-acetylglucosaminyltransferase I-negative HEK293S stable mammalian cell line. Proc. Natl Acad. Sci. USA 99, 13419–13424 (2002).

  39. 39.

    Hacker, D. L. et al. Polyethyleneimine-based transient gene expression processes for suspension-adapted HEK-293E and CHO-DG44 cells. Protein Expr. Purif. 92, 67–76 (2013).

  40. 40.

    Schuck, P. Size-distribution analysis of macromolecules by sedimentation velocity ultracentrifugation and Lamm equation modeling. Biophys. J. 78, 1606–1619 (2000).

  41. 41.

    Strop, P. & Brunger, A. T. Refractive index-based determination of detergent concentration and its application to the study of membrane proteins. Protein Sci. 14, 2207–2211 (2005).

  42. 42.

    Kabsch, W. Automatic processing of rotation diffraction data from crystals of initially unknown symmetry and cell constants. J. Appl. Crystallogr. 26, 795–800 (1993).

  43. 43.

    McCoy, A. J. et al. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674 (2007).

  44. 44.

    Adams, P. D. et al. PHENIX: building new software for automated crystallographic structure determination. Acta Crystallogr. D 58, 1948–1954 (2002).

  45. 45.

    Yang, J. et al. The I-TASSER Suite: protein structure and function prediction. Nat. Methods 12, 7–8 (2015).

  46. 46.

    Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D 60, 2126–2132 (2004).

  47. 47.

    Mastronarde, D. N. Automated electron microscope tomography using robust prediction of specimen movements. J. Struct. Biol. 152, 36–51 (2005).

  48. 48.

    Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy. Nat. Methods 14, 331–332 (2017).

  49. 49.

    Rohou, A. & Grigorieff, N. CTFFIND4: Fast and accurate defocus estimation from electron micrographs. J. Struct. Biol. 192, 216–221 (2015).

  50. 50.

    Scheres, S. H. RELION: implementation of a Bayesian approach to cryo-EM structure determination. J. Struct. Biol. 180, 519–530 (2012).

  51. 51.

    Kimanius, D., Forsberg, B. O., Scheres, S. H. & Lindahl, E. Accelerated cryo-EM structure determination with parallelisation using GPUs in RELION-2. eLife 5, e18722 (2016).

  52. 52.

    Bai, X. C., Rajendra, E., Yang, G., Shi, Y. & Scheres, S. H. Sampling the conformational space of the catalytic subunit of human γ-secretase. eLife 4, e11182 (2015).

  53. 53.

    Chen, S. et al. High-resolution noise substitution to measure overfitting and validate resolution in 3D structure determination by single particle electron cryomicroscopy. Ultramicroscopy 135, 24–35 (2013).

  54. 54.

    Rosenthal, P. B. & Henderson, R. Optimal determination of particle orientation, absolute hand, and contrast loss in single-particle electron cryomicroscopy. J. Mol. Biol. 333, 721–745 (2003).

  55. 55.

    Scheres, S. H. & Chen, S. Prevention of overfitting in cryo-EM structure determination. Nat. Methods 9, 853–854 (2012).

  56. 56.

    Heymann, J. B. & Belnap, D. M. Bsoft: image processing and molecular modeling for electron microscopy. J. Struct. Biol. 157, 3–18 (2007).

  57. 57.

    Cardone, G., Heymann, J. B. & Steven, A. C. One number does not fit all: mapping local variations in resolution in cryo-EM reconstructions. J. Struct. Biol. 184, 226–236 (2013).

  58. 58.

    Murshudov, G. N. et al. REFMAC5 for the refinement of macromolecular crystal structures. Acta Crystallogr. D 67, 355–367 (2011).

  59. 59.

    Brown, A. et al. Tools for macromolecular model building and refinement into electron cryo-microscopy reconstructions. Acta Crystallogr. D 71, 136–153 (2015).

  60. 60.

    Pettersen, E. F. et al. UCSF Chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612 (2004).

  61. 61.

    Burnley, T., Palmer, C. M. & Winn, M. Recent developments in the CCP-EM software suite. Acta Crystallogr. D 73, 469–477 (2017).

  62. 62.

    Smart, O. S., Neduvelil, J. G., Wang, X., Wallace, B. A. & Sansom, M. S. HOLE: a program for the analysis of the pore dimensions of ion channel structural models. J. Mol. Graph. 14, 354–360, 376 (1996).

  63. 63.

    Berka, K. et al. MOLEonline 2.0: interactive web-based analysis of biomacromolecular channels. Nucleic Acids Res. 40, W222–W227 (2012).

  64. 64.

    Sanner, M. F., Olson, A. J. & Spehner, J. C. Reduced surface: an efficient way to compute molecular surfaces. Biopolymers 38, 305–320 (1996).

  65. 65.

    Biasini, M. et al. SWISS-MODEL: modelling protein tertiary and quaternary structure using evolutionary information. Nucleic Acids Res. 42, W252–W258 (2014).

  66. 66.

    Brooks, B. R. et al. CHARMM: a program for macromolecular energy, minimization, and dynamics calculations. J. Comput. Chem. 4, 187–217 (1983).

  67. 67.

    Im, W., Beglov, D. & Roux, B. Continuum solvation model: Electrostatic forces from numerical solutions to the Poisson-Bolztmann equation. Comput. Phys. Commun. 111, 59–75 (1998).

Download references

Acknowledgements

This research was supported by a grant from the Swiss National Science Foundation (No. 31003A_163421). We thank O. Medalia and M. Eibauer, the Center for Microscopy and Image Analysis (ZMB) of the University of Zurich, and the Mäxi foundation for support and access to the electron microscopes, S. Klauser and S. Rast for their help in establishing the computer infrastructure, T. J. Jentsch for providing the LRRC8−/− HEK cell line, and B. Blattmann and C. Stutz-Ducommun for help with crystallization screening of the LRRC8ALRRD construct. X-ray data were collected at the X10SA Beamline at the Swiss Light Source of the Paul Scherrer Institute. We thank J. D. Walter for comments on the manuscript and all members of the Dutzler laboratory for their help at various stages of the project.

Reviewer information

Nature thanks S.-L. Shyng, T. Stauber and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Author information

Author notes

  1. These authors contributed equally: Dawid Deneka, Marta Sawicka.

Affiliations

  1. Department of Biochemistry, University of Zurich, Zurich, Switzerland

    • Dawid Deneka
    • , Marta Sawicka
    • , Andy K. M. Lam
    • , Cristina Paulino
    •  & Raimund Dutzler
  2. Department of Structural Biology, Groningen Biomolecular Sciences and Biotechnology Institute, University of Groningen, Groningen, The Netherlands

    • Cristina Paulino

Authors

  1. Search for Dawid Deneka in:

  2. Search for Marta Sawicka in:

  3. Search for Andy K. M. Lam in:

  4. Search for Cristina Paulino in:

  5. Search for Raimund Dutzler in:

Contributions

D.D. generated expression constructs, purified proteins and determined the X-ray structure of the cytoplasmic domain. M.S. prepared the samples for cryo-EM, collected cryo-EM data and proceeded with structure determination. A.K.M.L. recorded and analysed electrophysiology data. C. P. established data processing infrastructure and performed initial screening of conditions for cryo-EM studies. D.D., M.S., A.K.M. L. and R.D. jointly planned experiments, analysed the data and wrote the manuscript.

Competing interests

The authors declare no competing interests.

Corresponding author

Correspondence to Raimund Dutzler.

Extended data figures and tables

  1. Extended Data Fig. 1 Sequence alignment and topology.

    a, Multiple sequence alignment of mouse paralogues of LRRC8. Shown are the sequences of LRRC8A (NCBI: NP_808393), LRRC8B (NCBI: NP_001028722), LRRC8C (NCBI: NP_598658), LRRC8D (NCBI: NP_001116240) and LRRC8E (NCBI: NP_082451). Residues strictly conserved in all subunits are coloured in blue. Segments of LRRC8A that are not defined in the cryo-EM density are highlighted in grey. Secondary structure elements of LRRC8A are shown below (pore domain, green; LRRD, blue; LRβ16 and LRα16 correspond to a less conserved non-canonical repeat). Numbering corresponds to LRRC8A. The coloured symbols represent the following: Red circles, two threonine residues facing the pore at the extracellular membrane boundary; blue triangles, basic residues of the ESD of LRRC8A facing the pore; yellow inverted triangles, cysteine residues involved in disulfide bridges; green inverted triangles, residues at the extracellular side that have a role in desensitization; black squares, annotated phosphorylation sites (Uniprot: Q80WG5). The two paralogues LRRC8A and LRRC8C investigated in this study are of similar length and share 58% of identical and 72% of homologous residues. b, Topology of the pore domain with secondary structure elements indicated.

  2. Extended Data Fig. 2 Electrophysiology and biochemical characterization.

    a, Time-dependent whole-cell currents of non-transfected HEK293T cells recorded at 100 mV (dark blue) and −100 mV (light blue) evoked in response to low intracellular ion concentration (100 mM NMDG-Cl). b, Representative current response of activated channels at different voltages at a symmetric Cl concentration of 100 mM. c, Current–voltage relationships in symmetric (left, n = 6) and asymmetric (right, n = 4) ionic conditions. The negative reversal potential corresponding to the Nernst potential of Cl in the asymmetric condition is indicative of an anion-selective current. a and b show a representative recording of data shown in c, left. d, e, Time-dependent response (d) and representative traces and current–voltage relationships (e) of the LRRC8 knockout cell line LRRC8−/− at the indicated ion concentrations (100o/100i, n = 5; 100o/50i, n = 5). d and e, left show representative recordings of data shown in e, right. f, g, Time-dependent response (f) and current response (g) of LRRC8−/− cells expressing LRRC8A/C at the indicated ion concentrations. f and g show representative recordings of data shown in Fig. 1b. h, i, Time-dependent response (h) and current response (i) of LRRC8−/− cells expressing LRRC8A at the indicated ion concentrations. h and i show representative recordings of data shown in Fig. 1c. Dashed lines show the corresponding responses of LRRC8−/− cells at the indicated ion concentrations. Data in ai were recorded under voltage-clamp conditions in the whole-cell configuration. Current–voltage relationships show averages of the indicated number of independent experiments, errors are s.e.m. In a, d, f, h, the black triangle indicates the point of break-in. In b, e, g, i, traces show representative data recorded at the plateau at the indicated ion concentrations. j, Plasma membrane expression of LRRC8 constructs probed by surface biotinylation. Representative western blot (n = 3) shows avidin-purified chemically biotinylated proteins at the cell surface, which are detected by an anti-Myc antibody recognizing a fusion tag on the overexpressed LRRC8 constructs. Lanes show M, marker with molecular weights indicated; NT, non-transfected biotinylated LRRC8−/− cells; NB, LRRC8A/C transfected, non-biotinylated LRRC8−/− cells; A, LRRC8A transfected, biotinylated LRRC8−/− cells; AC, LRRC8A/C transfected, biotinylated LRRC8−/− cells. The blot shows comparable surface expression of LRRC8A and LRRC8A/C constructs, which is in agreement with previous reports9. The experiment was repeated three times as independent biological replicates with similar results. k, Elution profile of LRRC8A (red) and LRRC8APD (blue) from a Superose 6 10/300 column. The first peak between 13 and 14 ml corresponds to a hexamer of LRRC8A, the smaller peak at around 17 ml corresponds to the cleaved fluorescent fusion protein. l, Elution profile and derived molecular weight for LRRC8APD purified in DDM. The continuous black line shows normalized absorbance at 280nm. The calculated molecular weight of the entire complex and its protein and detergent components based on MALS experiments are shown in purple, blue and red respectively. The calculated values are consistent with a hexameric assembly of LRRC8APD. m, Elution profile of LRRC8ALRRD from a Superdex 75 10/300 column. The main peak around 12 ml corresponds to a monomer of the cytoplasmic domain. In k, m, arrows indicate the void and bed volume. Source data

  3. Extended Data Fig. 3 Cryo-EM reconstruction of LRRC8A at 5.01 Å.

    a, Representative cryo-EM micrograph of dataset 1 acquired with an FEI Tecnai G2 Polara microscope. The image shows well-dispersed single particles with occasional micro-aggregates. b, 2D class averages of LRRC8A. c, Angular distribution plot of all particles included in the final C3-symmetrized reconstruction showing uniform orientations. The length and the colour of cylinders correspond to the number of particles with respective Euler angles. d, The data-processing workflow. Seven out of nine classes generated during 3D classification reveal a well-defined, C6-symmetric pore domain and structural heterogeneity in the LRRDs, which reflects their intrinsic mobility. The second class from the left shows additional density representing a part of the cytoplasmic subunits of a neighbouring particle. Particles assigned to the boxed class were used in further refinement. The distribution of all particles (%) and the resolution of each class is indicated. e, FSC plot of the final refined unmasked (orange), masked (pink), phase-randomized (green) and corrected for mask convolution effects (blue) cryo-EM density map of LRRC8A. The resolution at which the FSC curve drops below the 0.143 threshold is indicated. The inset shows the atomic model within the mask that was applied for calculations of the resolution estimates. f, Final 3D reconstruction coloured according to local resolution.

  4. Extended Data Fig. 4 Cryo-EM reconstruction of LRRC8A at 4.25 Å and its pore domain at 3.66 Å.

    a, Representative cryo-EM micrograph of dataset 2 obtained at an FEI Titan Krios. b, 2D class averages of LRRC8A. c, Angular distribution plots of all particles included in the final C3-symmetrized full-length reconstruction (upper panel) and the final C6-symmetrized pore domain reconstruction from focused refinement (lower panel) show uniform orientations. The length and colour of cylinders correspond to the number of particles with respective Euler angles. d, The data-processing workflow. 3D classification was performed directly on the initial set of particles to include less-populated orientations, which could be otherwise excluded if subjected to prior 2D classification. One of two predominant classes (boxed) revealed a subset of homogeneous particles, which were used in further processing. The distribution of all particles (percentage) and the resolution of each class is indicated. The focused refinement with C6 symmetry improved the resolution of the pore domain. The inset shows the mask applied to the map during refinement, in which the particle signal outside the mask, corresponding to the LRRDs, was subtracted e, FSC plot of the final refined unmasked (orange), masked (pink), phase-randomized (green) and corrected for mask convolution effects (blue) cryo-EM density map of C3-symmetric LRRC8A. The resolution at which the FSC curve drops below the 0.143 threshold is indicated. The inset shows the atomic model within the mask that was applied for calculations of the resolution estimates. f, Final 3D reconstruction of full-length LRRC8A coloured according to its local resolution. g, as in e, but for the 3D reconstruction of the C6-symmetric pore domain. h, as in f, but for the 3D reconstruction of the pore domain.

  5. Extended Data Fig. 5 Comparison of cryo-EM 3D reconstructions of LRRC8A and LRRC8A/C.

    a, Cα-representation of the refined model of LRRC8A with cryo-EM density at 4.25 Å at low (4.0 σ, left) and high (6.0 σ, right) contour threshold superimposed. b, Cα-representation of the refined model of LRRC8A with cryo-EM density at 5.01 Å at low (7.0 σ, left) and high (9.7 σ, right) contour threshold superimposed. c, Representative cryo-EM micrograph of dataset 3 acquired with an FEI Tecnai G2 Polara microscope. d, 2D class averages of the LRRC8A/C assembly. LRRC8A/C heteromers were obtained from the co-expression of the two constructs in HEK293 cells followed by a tandem affinity purification to select channels containing both subunits. e, The angular distribution plot of all particles included in the final C3-symmetrized reconstruction shows uniform orientations. The length and the colour of cylinders correspond to the number of particles with respective Euler angles. f, The data-processing workflow for LRRC8A/C. After 3D classification with either C1 or C3 symmetry imposed, the three predominant classes show a symmetric pore domain and LRRDs with higher flexibility. Cryo-EM density represents an average of LRRC8A and C subunits which, owing to their strong sequence relationship, are essentially identical in structure at the current resolution of the data. The final refined C3-symmetrized reconstructions reveal the overall architecture of the hexameric protein, which exhibits very similar features to the LRRC8A homo-hexamer. Although the final C3-symmetric model generated from non-symmetrized 3D classes contains twice the number of particles, its resolution is lower compared to the model calculated from the predominant symmetrized 3D class. We therefore used the latter for structural descriptions. The distribution of all particles (percentage) and the resolution of each class is indicated. g, FSC plot of the final refined unmasked (orange), masked (pink), phase-randomized (green) and corrected for mask convolution effects (blue) cryo-EM density map of LRRC8A/C. The resolution at which the FSC curve drops below the 0.143 threshold is indicated. The inset shows the cryo-EM density within the mask that was applied for calculations of the resolution estimates. h, Final 3D reconstruction at high (left) and low (right) thresholds, coloured according to local resolution.

  6. Extended Data Fig. 6 Cryo-EM and X-ray electron density.

    a, Sections of the cryo-EM density (grey, 6.5 σ) superimposed on the refined structure of the pore domain. Secondary structure elements are labelled. b, Cryo-EM density (grey, 6.0 σ) around disulfide bonds in the ESD. c, Stereo view of cryo-EM density (6.5 σ) of the pore region in the ESD with front subunits removed for clarity. ac show cryo-EM density of the C6-symmetrized pore domain of LRRC8A at 3.66 Å. d, e, Section (d) and stereo (e) views of the isolated LRRD of LRRC8A. Electron density (1.0 σ) of the LRRD determined by X-ray crystallography at 1.8 Å is shown superimposed on the model. f, g, Section (f) and stereo (g) views of the LRRD in the full-length LRRC8A channel. C3-symmetrized cryo-EM density (4.0 σ) is shown superimposed on the model. Although the resolution of the entire reconstruction is 4.25 Å, the local resolution of the LRRDs is lower. In ag, structures are shown as sticks.

  7. Extended Data Fig. 7 X-ray crystallography and cryo-EM statistics and model validation.

    a, Statistics of X-ray data collection, phasing and refinement. b, Statistics of cryo-EM data collection, processing and model refinement. c, d, FSC plots of refined atomic models against the cryo-EM density maps of the full-length protein at 4.25 Å (c, dataset 2) and the pore domain of LRRC8A at 3.66 Å (d, focus-refined dataset 2a). FSCsum (orange) is calculated for the full masked map and the model refined against the complete dataset, FSCwork (blue) is calculated for the masked half-map 1 and the shaken model refined against the dataset comprising the half map 1, FSCfree (pink) is calculated for the masked half-map 2 and the shaken model refined against the half map 1. FSC thresholds at 0.5 and 0.143 were used for FSCsum and FSCwork/FSCfree, respectively.

  8. Extended Data Fig. 8 Homology model and electrostatics.

    a, Model of LRRC8A used for electrostatic calculations. The high sequence conservation between paralogues and the structural similarity between homo- and heteromeric channels has enabled the construction of a homology model of LRRC8A/C. Whereas in this model the general features of LRRC8C are reliably represented, the alternate arrangement of A and C subunits in the hexameric channel, although plausible, is at this stage speculative. In both structures, the missing residues at the respective N termini were modelled in such a way that they do not obstruct the pore. Membrane boundaries (C, hydrophobic core and HG, head-group regions) are indicated. b, Electrostatic potentials along the pores of LRRC8A (red) and LRRC8A/C (blue), as determined by a numerical solution of the Poisson–Boltzmann equation at 150 mM of monovalent ions in the solvent and the pore region. c, d, Part of the ESD of the LRRC8A/C model that lines the pore, viewed from within the membrane with front subunits removed for clarity (c) and from the extracellular side (d). The protein is displayed as a ribbon representation, with selected side chains as shown as sticks. Subunits A and C are shown in green and orange, respectively. Although most of the interaction interface is conserved in the pore domain of this model, differences in pore-lining residues alter the electrostatic properties of the ion conduction path. e, Residual density (5.0 σ) in the pore region of the C3-symmetrized cryo-EM map of LRRC8A. The region around T44 and T48 is highlighted, with front subunits removed for clarity. The view is as in c and the protein is displayed as sticks. f, The electrostatic potential along the pore of the LRRC8A(R103A)/C mutant (green), as determined by a numerical solution of the Poisson–Boltzmann equation in the absence (left) and at 150 mM (right) of monovalent ions in the solvent and the pore region. In b, f, the membrane boundaries and the border between the pore domain and the LRRD are indicated.

  9. Extended Data Fig. 9 Electrophysiology of selectivity filter mutants.

    a, b, Time-dependent response (a) and representative traces and current–voltage relationship (b) of LRRC8−/− cells expressing LRRC8A(R103A) (n = 4). c, d, Time-dependent response (c) and representative traces and current–voltage relationships (d) of LRRC8−/− cells expressing LRRC8A/C(L105R) (n = 7). e, f, Time-dependent response (e) and representative traces and current–voltage relationships (f) of LRRC8−/− cells expressing LRRC8A(R103A)/C (n = 4). In a, c, e, the black triangle indicates the point of break-in. In b, d, f, traces show representative data recorded at the plateau at the indicated symmetric Cl concentrations. Current–voltage relationships show averages of the indicated number of independent experiments, errors are s.e.m. g, Current–voltage relationships of LRRC8A(R103A)/C (n = 5), LRRC8A/C (n = 4) and LRRC8A/C(L105R) (n = 5) at the indicated ionic gradient. h, Reversal potentials measured from the current–voltage relationship shown in g. In g, h, the Nernst potential of Cl is indicated. ik, Current–voltage relationships of the indicated constructs with different extracellular anions (LRRC8A(R103A)/C (i), n = 4; LRRC8A/C (j), n = 5; LRRC8A/C(L105R) (k), n = 6). l, Reversal potentials measured from the current–voltage relationship shown in i, j and k. m, Ratio of the outward (SO42−) and inward (Cl) current at ±100 mV measured from the current–voltage relationship shown in i, j and k. n, Relative inward current amplitude used to scale the normalized current–voltage relationship recorded in the presence of extracellular SO42− in i, j and k. The shift in the reversal potential under the bi-ionic SO42−o/Cli condition and the increase in relative SO42− current reflects the increase in the relative permeability of the divalent anion as the number of positive charges in the selectivity filter increases. In gn, the data show averages of the indicated number of independent experiments, errors are s.e.m. Data in an were recorded under voltage-clamp conditions in the whole-cell configuration. Source data

  10. Extended Data Fig. 10 Comparison with gap junction proteins.

    a, Superposition of subunits of LRRC8A (green) and connexin Cx26 (PDB: 2ZW3, red, left) and expanded view of the ESD (right). Selected secondary structure elements are indicated. b, Superposition of subunits of LRRC8A (green) and innexin-6 (PDB: 5H1Q, blue, left) and expanded view of the ESD (right). a, b, Proteins are displayed as ribbons; selected secondary structure elements are indicated. c, Pore radius along the symmetry axis. The dashed green line corresponds to the structure of LRRC8A containing the modelled N terminus used for electrostatic calculations. In the case of connexin Cx26, the difference in pore geometry arises from a shortening of helix E1H, and in C. elegans innexin-6 the difference arises from its larger, octameric assembly. df, Pore domains of LRRC8A with modelled N-terminal residues (d) connexin Cx26 (PDB: 2ZW3) (e) and innexin-6 (PDB: 5H1Q) (f), viewed from the extracellular side (top), and two opposing subunits of the respective channels shown from within the membrane with N termini indicated (bottom). In d–f, the proteins are shown as space-filling models.

Supplementary information

  1. Supplementary Figure 1

    Expression of LRRC8 constructs on the surface of LRRC8–/– cells probed by biotinylation. Results of three independent biological replicates are shown. a, Experiment 1; Raw image of the western blot displayed in Extended Data Fig. 2j. b, Experiment 2; For each condition, two aliquots were analyzed on the same western blot. c, Experiment 3. a, b, A Coomassie blue stained membrane of the western blot is shown at the top, a bright field image before staining at the center and a chemiluminescence image before staining at the bottom. c, A bright field image is shown at the top and a chemiluminescence image at the bottom. a–c, Lanes show 1, molecular weight marker; 2, non-transfected biotinylated HEK293 LRRC8–/– cells; 3, LRRC8A/C transfected, non-biotinylated LRRC8–/– cells; 4, LRRC8A transfected, biotinylated LRRC8–/– cells; 5, LRRC8A/C transfected, biotinylated LRRC8–/– cells; 6, LRRC8C transfected, biotinylated LRRC8–/– cells; 7, molecular weight marker.

  2. Reporting Summary

  3. Video 1

    Cryo-EM density map of a single subunit of the mLRRC8APD with the modelled structure superimposed

  4. Video 2

    X-ray electron density map of a single subunit of the isolated LRRD with the modelled structure superimposed

  5. Video 3

    Cryo-EM density map of a single subunit of the full-length mLRRC8A with the modelled structure superimposed

  6. Video 4: Ribbon representation of the homo-hexameric mLRRC8A structure

    The structure is first viewed from the extracellular side and then rotated to show the view from within the membrane. Two front subunits are removed for clarity. The zoom-in focuses on the residues T44, T48, K51 and R103, which line up the pore and are possibly involved in anion selectivity

  7. Source data Fig. 1

  8. Source data Fig. 6

  9. Source Data Extended Data Fig. 2

  10. Source Data Extended Data Fig. 9

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.