Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

BAK core dimers bind lipids and can be bridged by them

Nature Structural & Molecular Biology volume 27, pages 1024–1031 (2020)Cite this article

Subjects

Abstract

BAK and BAX are essential mediators of apoptosis that oligomerize in response to death cues, thereby causing permeabilization of the mitochondrial outer membrane. Their transition from quiescent monomers to pore-forming oligomers involves a well-characterized symmetric dimer intermediate. However, no essential secondary interface that can be disrupted by mutagenesis has been identified. Here we describe crystal structures of human BAK core domain (α2–α5) dimers that reveal preferred binding sites for membrane lipids and detergents. The phospholipid headgroup and one acyl chain (sn2) associate with one core dimer while the other acyl chain (sn1) associates with a neighboring core dimer, suggesting a mechanism by which lipids contribute to the oligomerization of BAK. Our data support a model in which, unlike for other pore-forming proteins whose monomers assemble into oligomers primarily through protein–protein interfaces, the membrane itself plays a role in BAK and BAX oligomerization.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Structure of the BAK core domain hexamer in complex with bacterial lipids.
Fig. 2: Phospholipid binding modes.
Fig. 3: Phospholipid binding and lipid-mediated oligomerization of BAK dimers.
Fig. 4: Molecular modeling of the BAK core domain dimer in a lipid bilayer.
Fig. 5: Mutation of lipid-binding residues impairs BAK oligomerization and cell death.

Similar content being viewed by others

Data availability

Atomic structures have been deposited in the PDB with accession codes 6UXM (BAK core dimer in complex with E. coli lipids), 6UXN (BAK core dimer in complex with 8:0–8:0 PS), 6UXO (BAK core dimer in complex with DDM), 6UXP (BAK core dimer in complex with 8:0–8:0 PG), 6UXQ (BAK core dimer in complex with POPC and C8E4) and 6UXR (BAK core dimer in complex with LysoPC). Mass spectrometry data have been deposited with the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD020582. Source data are provided with this paper.

References

  1. Czabotar, P. E. et al. Bax crystal structures reveal how BH3 domains activate Bax and nucleate its oligomerization to induce apoptosis. Cell 152, 519–531 (2013).

    CAS  PubMed  Google Scholar 

  2. Brouwer, J. M. et al. Bak core and latch domains separate during activation, and freed core domains form symmetric homodimers. Mol. Cell 55, 938–946 (2014).

    CAS  PubMed  Google Scholar 

  3. Dewson, G. et al. Bak activation for apoptosis involves oligomerization of dimers via their ɑ6 helices. Mol. Cell 36, 696–703 (2009).

    CAS  PubMed  Google Scholar 

  4. Ma, S. et al. Assembly of the Bak apoptotic pore: a critical role for the Bak protein ɑ6 helix in the multimerization of homodimers during apoptosis. J. Biol. Chem. 288, 26027–26038 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  5. Bleicken, S. et al. Structural model of active Bax at the membrane. Mol. Cell 56, 496–505 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  6. Aluvila, S. et al. Organization of the mitochondrial apoptotic BAK pore: oligomerization of the BAK homodimers. J. Biol. Chem. 289, 2537–2551 (2014).

    CAS  PubMed  Google Scholar 

  7. Uren, R. T. et al. Disordered clusters of Bak dimers rupture mitochondria during apoptosis. Elife 6, e19944 (2017).

    PubMed  PubMed Central  Google Scholar 

  8. Grosse, L. et al. Bax assembles into large ring-like structures remodeling the mitochondrial outer membrane in apoptosis. EMBO J. 35, 402–413 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  9. Salvador-Gallego, R. et al. Bax assembly into rings and arcs in apoptotic mitochondria is linked to membrane pores. EMBO J. 35, 389–401 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  10. Basañez, G. et al. Bax-type apoptotic proteins porate pure lipid bilayers through a mechanism sensitive to intrinsic monolayer curvature. J. Biol. Chem. 277, 49360–49365 (2002).

    PubMed  Google Scholar 

  11. van der Wel, C. et al. Lipid membrane-mediated attraction between curvature inducing objects. Sci. Rep. 6, 32825 (2016).

    PubMed  PubMed Central  Google Scholar 

  12. Last, N. B., Schlamadinger, D. E. & Miranker, A. D. A common landscape for membrane-active peptides. Protein Sci. 22, 870–882 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  13. Dewson, G. et al. To trigger apoptosis, Bak exposes its BH3 domain and homodimerizes via BH3:groove interactions. Mol. Cell 30, 369–380 (2008).

    CAS  PubMed  Google Scholar 

  14. Trott, O. & Olson, A. J. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem. 31, 455–461 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  15. Mandal, T. et al. Assembly of Bak homodimers into higher order homooligomers in the mitochondrial apoptotic pore. Sci. Rep. 6, 30763 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  16. Tanaka, K., Caaveiro, J. M., Morante, K., González-Mañas, J. M. & Tsumoto, K. Structural basis for self-assembly of a cytolytic pore lined by protein and lipid. Nat. Commun. 6, 6337 (2015).

    PubMed  Google Scholar 

  17. Gupta, K. et al. The role of interfacial lipids in stabilizing membrane protein oligomers. Nature 541, 421–424 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

  18. Iyer, S. et al. Bak apoptotic pores involve a flexible C-terminal region and juxtaposition of the C-terminal transmembrane domains. Cell Death Differ. 22, 1665–1675 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  19. Bleicken, S., Landeta, O., Landajuela, A., Basañez, G. & García-Sáez, A. J. Proapoptotic Bax and Bak proteins form stable protein-permeable pores of tunable size. J. Biol. Chem. 288, 33241–33252 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  20. Westphal, D. et al. Apoptotic pore formation is associated with in-plane insertion of Bak or Bax central helices into the mitochondrial outer membrane. Proc. Natl Acad. Sci. USA 111, E4076–E4085 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  21. Ader, N. R. et al. Molecular and topological reorganizations in mitochondrial architecture interplay during Bax-mediated steps of apoptosis. Elife 8, e40712 (2019).

    PubMed  PubMed Central  Google Scholar 

  22. Hauseman, Z. J. et al. Homogeneous oligomers of pro-apoptotic BAX reveal structural determinants of mitochondrial membrane permeabilization. Mol. Cell 79, 68–83 (2020).

    CAS  PubMed  PubMed Central  Google Scholar 

  23. Bleicken, S. et al. Topology of active, membrane-embedded Bax in the context of a toroidal pore. Cell Death Differ. 25, 1717–1731 (2018).

    CAS  PubMed  PubMed Central  Google Scholar 

  24. Lawrence, M. C. & Colman, P. M. Shape complementarity at protein/protein interfaces. J. Mol. Biol. 234, 946–950 (1993).

    CAS  PubMed  Google Scholar 

  25. Moldoveanu, T. et al. The X-ray structure of a BAK homodimer reveals an inhibitory zinc binding site. Mol. Cell 24, 677–688 (2006).

    CAS  PubMed  Google Scholar 

  26. Kabsch, W. XDS. Acta Crystallogr. D Biol. Crystallogr. 66, 125–132 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  27. Evans, P. R. & Murshudov, G. N. How good are my data and what is the resolution? Acta Crystallogr. D Biol. Crystallogr. 69, 1204–1214 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

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

    CAS  PubMed  PubMed Central  Google Scholar 

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

    PubMed  Google Scholar 

  30. Adams, P. D. et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D Biol. Crystallogr. 66, 213–221 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  31. Moriarty, N. W., Grosse-Kunstleve, R. W. & Adams, P. D. electronic Ligand Builder and Optimization Workbench (eLBOW): a tool for ligand coordinate and restraint generation. Acta Crystallogr. D Biol. Crystallogr. 65, 1074–1080 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  32. Liebschner, D. et al. Polder maps: improving OMIT maps by excluding bulk solvent. Acta Crystallogr. D Struct. Biol. 73, 148–157 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

  33. Gelin, M. et al. Combining ‘dry’ co-crystallization and in situ diffraction to facilitate ligand screening by X-ray crystallography. Acta Crystallogr. D Biol. Crystallogr. 71, 1777–1787 (2015).

    CAS  PubMed  Google Scholar 

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

    CAS  PubMed  Google Scholar 

  35. Lydic, T. A., Busik, J. V. & Reid, G. E. A monophasic extraction strategy for the simultaneous lipidome analysis of polar and nonpolar retina lipids. J. Lipid Res. 55, 1797–1809 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  36. Ryan, E. & Reid, G. E. Chemical derivatization and ultrahigh resolution and accurate mass spectrometry strategies for “shotgun” lipidome analysis. Acc. Chem. Res. 49, 1596–1604 (2016).

    CAS  PubMed  Google Scholar 

  37. Fahy, E., Cotter, D., Sud, M. & Subramaniam, S. Lipid classification, structures and tools. Biochim. Biophys. Acta 1811, 637–647 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  38. R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2016).

  39. Kirby, N. M. et al. A low-background-intensity focusing small-angle X-ray scattering undulator beamline. J. Appl. Crystallogr. 46, 1670–1680 (2013).

    CAS  Google Scholar 

  40. Weijman, J. F. et al. Structural basis of autoregulatory scaffolding by apoptosis signal-regulating kinase 1. Proc. Natl Acad. Sci. USA 114, E2096–E2105 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

  41. Konarev, P. V., Volkov, V. V., Sokolova, A. V., Koch, M. H. J. & Svergun, D. I. PRIMUS: a Windows PC-based system for small-angle scattering data analysis. J. Appl. Crystallogr. 36, 1277–1282 (2003).

    CAS  Google Scholar 

  42. Svergun, D. I. Determination of the regularization parameter in indirect-transform methods using perceptual criteria. J. Appl. Crystallogr. 25, 495–503 (1992).

    CAS  Google Scholar 

  43. Svergun, D., Barberato, C. & Koch, M. H. J. CRYSOL—a program to evaluate X-ray solution scattering of biological macromolecules from atomic coordinates. J. Appl. Crystallogr. 28, 768–773 (1995).

    CAS  Google Scholar 

  44. Oh, K. J. et al. Conformational changes in BAK, a pore-forming proapoptotic Bcl-2 family member, upon membrane insertion and direct evidence for the existence of BH3–BH3 contact interface in BAK homo-oligomers. J. Biol. Chem. 285, 28924–28937 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  45. Alsop, A. E. et al. Dissociation of Bak ɑ1 helix from the core and latch domains is required for apoptosis. Nat. Commun. 6, 6841 (2015).

    CAS  PubMed  Google Scholar 

  46. Wessel, D. & Flugge, U. I. A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Anal. Biochem. 138, 141–143 (1984).

    CAS  PubMed  Google Scholar 

  47. Kapp, E. A. et al. MSCypher: an integrated database searching and machine learning workflow for multiplexed proteomics. Preprint at bioRxiv https://doi.org/10.1101/397257 (2018).

  48. Webb, B. & Sali, A. Comparative protein structure modeling using MODELLER. Curr. Protoc. Bioinformatics 54, 5.6.1–5.6.37 (2016).

    Google Scholar 

  49. Monticelli, L. et al. The MARTINI coarse-grained force field: extension to proteins. J. Chem. Theory Comput. 4, 819–834 (2008).

    CAS  PubMed  Google Scholar 

  50. Abraham, M. J. et al. GROMACS: high performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1–2, 19–25 (2015).

    Google Scholar 

  51. Marrink, S. J., Risselada, H. J., Yefimov, S., Tieleman, D. P. & de Vries, A. H. The MARTINI force field: coarse grained model for biomolecular simulations. J. Phys. Chem. B 111, 7812–7824 (2007).

    CAS  PubMed  Google Scholar 

  52. Bussi, G., Donadio, D. & Parrinello, M. Canonical sampling through velocity rescaling. J. Chem. Phys. 126, 014101 (2007).

    PubMed  Google Scholar 

  53. Berendsen, H. J. C., Postma, J. P. M., Vangunsteren, W. F., DiNola, A. & Haak, J. R. Molecular dynamics with coupling to an external heat bath. J. Chem. Phys. 81, 3684–3690 (1984).

    CAS  Google Scholar 

  54. Gapsys, V., de Groot, B. L. & Briones, R. Computational analysis of local membrane properties. J. Comput. Aided Mol. Des. 27, 845–858 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  55. Essmann, U. et al. A smooth particle mesh Ewald method. J. Chem. Phys. 103, 8577–8593 (1995).

    CAS  Google Scholar 

  56. Lee, J. et al. CHARMM-GUI membrane builder for complex biological membrane simulations with glycolipids and lipoglycans. J. Chem. Theory Comput. 15, 775–786 (2019).

    CAS  PubMed  Google Scholar 

  57. Huang, J. et al. CHARMM36m: an improved force field for folded and intrinsically disordered proteins. Nat. Methods 14, 71–73 (2017).

    CAS  PubMed  Google Scholar 

  58. Evans, D. J. & Holian, B. L. The Nose–Hoover thermostat. J. Chem. Phys. 83, 4069–4074 (1985).

    CAS  Google Scholar 

  59. Parrinello, M. & Rahman, A. Polymorphic transitions in single crystals: a new molecular dynamics method. J. Appl. Phys. 52, 7182–7190 (1981).

    CAS  Google Scholar 

  60. Hess, B. P-LINCS: a parallel linear constraint solver for molecular simulation. J. Chem. Theory Comput. 4, 116–122 (2008).

    CAS  PubMed  Google Scholar 

  61. Humphrey, W., Dalke, A. & Schulten, K. VMD: Visual Molecular Dynamics. J. Mol. Graph. 14, 33–38 (1996).

    CAS  PubMed  Google Scholar 

  62. Haberkant, P. et al. In vivo profiling and visualization of cellular protein–lipid interactions using bifunctional fatty acids. Angew. Chem. Int. Ed. 52, 4033–4038 (2013).

    CAS  Google Scholar 

Download references

Acknowledgements

We thank R. Kluck, G. Lessene, A. Robin, R. Birkinshaw, C. Robinson and K. Gupta for discussions and comments on the manuscript and Z. Liu for assistance with mitochondrial experiments. This research was undertaken in part using the MX1 and MX2 crystallography and SAXS-WAXS beamlines at the Australian Synchrotron and made use of the ACRF Detector. We thank the staff at this facility for assistance with data collection and the staff at the Collaborative Crystallisation Centre for assistance with crystallization experiments. We acknowledge an Australian Postgraduate Award for A.D.C. and NHMRC fellowships to P.E.C., P.M.C. and J.M.M. (1079700, 1116934 and 1172929). Our work is supported by the NHMRC (project grants 1079706 and 1059290 and program grant 1113133), the ARC (Linkage Infrastructure, Equipment and Facilities Grant LE160100015), the Australian Cancer Research Foundation, the Leukemia and Lymphoma Society (US) (SCOR grant 7001-03), the Victorian State Government Operational Infrastructure Support and the Australian Government NHMRC IRIISS (9000587). Part of this work was undertaken using resources from the National Computational Infrastructure, which is supported by the Australian government and provided through Intersect Australia under LIEF grant LE170100032 and through the HPC-GPGPU Facility, which was established with the assistance of LIEF Grant LE170100200.

Author information

Authors and Affiliations

  1. Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia

    Angus D. Cowan, Jarrod J. Sandow, James M. Murphy, Jason M. Brouwer, Jonathan P. Bernardini, Michael J. Roy, Ahmad Z. Wardak, Iris K. Tan, Andrew I. Webb, Jacqueline M. Gulbis, Grant Dewson, Peter M. Colman & Peter E. Czabotar

  2. Department of Medical Biology, University of Melbourne, Parkville, Victoria, Australia

    Angus D. Cowan, Jarrod J. Sandow, James M. Murphy, Jason M. Brouwer, Jonathan P. Bernardini, Michael J. Roy, Andrew I. Webb, Jacqueline M. Gulbis, Grant Dewson, Peter M. Colman & Peter E. Czabotar

  3. La Trobe Institute for Molecular Science, La Trobe University, Melbourne, Victoria, Australia

    Nicholas A. Smith & Brian J. Smith

  4. Department of Biochemistry and Molecular Biology, University of Melbourne, Parkville, Victoria, Australia

    Eugene A. Kapp, Yepy H. Rustam & Gavin E. Reid

  5. School of Chemistry, University of Melbourne, Parkville, Victoria, Australia

    Eugene A. Kapp, Yepy H. Rustam & Gavin E. Reid

  6. Bio21 Molecular Science and Biotechnology Institute, University of Melbourne, Parkville, Victoria, Australia

    Eugene A. Kapp, Yepy H. Rustam & Gavin E. Reid

Authors
  1. Angus D. Cowan
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Nicholas A. Smith
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jarrod J. Sandow
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Eugene A. Kapp
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Yepy H. Rustam
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. James M. Murphy
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Jason M. Brouwer
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Jonathan P. Bernardini
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Michael J. Roy
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Ahmad Z. Wardak
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Iris K. Tan
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Andrew I. Webb
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Jacqueline M. Gulbis
    View author publications

    You can also search for this author in PubMed Google Scholar

  14. Brian J. Smith
    View author publications

    You can also search for this author in PubMed Google Scholar

  15. Gavin E. Reid
    View author publications

    You can also search for this author in PubMed Google Scholar

  16. Grant Dewson
    View author publications

    You can also search for this author in PubMed Google Scholar

  17. Peter M. Colman
    View author publications

    You can also search for this author in PubMed Google Scholar

  18. Peter E. Czabotar
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

A.D.C., J.J.S., E.A.K., Y.H.R., J.M.M., J.M.B., J.P.B., M.J.R., A.Z.W., I.K.T. and G.D. performed experiments under supervision of A.I.W., G.D., G.E.R., P.M.C. and P.E.C. N.A.S. and B.J.S. performed molecular modeling experiments. A.D.C., N.A.S., J.J.S., E.A.K., Y.H.R., J.M.M., J.M.B., J.P.B., M.J.R., A.Z.W., I.K.T., A.I.W., J.M.G., B.J.S., G.E.R., G.D., P.M.C. and P.E.C. contributed to data interpretation. A.D.C., P.M.C. and P.E.C. wrote the paper.

Corresponding authors

Correspondence to Peter M. Colman or Peter E. Czabotar.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Anke Sparmann was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team.

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

Extended data

Extended Data Fig. 1 Unlatching and homodimerisation of BAK.

a, The structure of monomeric BAK (PDB code: 2IMT25) shown as pipes and ribbon representation in upper and lower panels, respectively. The α1 helix is colored light blue, the BH3 domain (α2) is colored yellow, the rest of the core domain (α3-α5) is colored purple and the latch domain (α6-α8) is colored salmon. b, Activation by transient interaction with BH3-only proteins causes separation of the core and latch domains (PDB code: 4U2U2, α1 not displayed). c, Freed core domains, colored as in a, b for one protomer and light gray for the other, homodimerise by reciprocal exchange of their BH3 domains (α2) into the canonical hydrophobic grooves (α3’-α5’) of their partner molecule forming a symmetric homodimer (PDB code: 4U2V2).

Extended Data Fig. 2 Evidence for lipid in BAK core dimers.

a, Comparison of BAK core domain homodimers from Fig.1b (colored shades of blue) and from Brouwer et al. 2014 (colored shades of gray, PDB code: 4U2V). Helices α2–5 are labeled on the light blue and light gray protomers of the dimer as in Extended Data Fig. 1c. Though the overall fold is maintained, the new structures are somewhat ‘flatter’. b, a view of the lipophilic α4α5α5’α4’ surfaces of the BAK core dimers from a. Small differences in the detailed atomic contacts between α5 and α5’ across the dyad axis between the structures reported here (upper panel) and previously (lower panel) suggest softness in this hydrophobic interface and lead to different surface topography in the site that accommodates the sn2 acyl chain of the lipid in the new structure, indicated by arrows. c, Density (2Fo-Fc in gray and Fo-Fc in green contoured at 1 and 3 σ, respectively) resembling a diacyl phospholipid (modeled at the time as a sulphate ion, yellow) was observed during the early stages of refinement at 6 identical positions in the α5α5’ grooves of the 3 homodimers present in the BAK core domain hexamer structure. d, Common chemical structure of phospholipids using stereospecific numbering (sn). Attached to the glycerol core are two acyl chains of variable lengths and saturation levels at the sn1 and sn2 positions, and the sn3 phosphate connected to variable group R (for example choline, ethanolamine etc.). e, Pairwise interatomic distance distribution (P(r)) determined with GNOM from SAXS data42. f, Guinier plot of the SAXS data for q.Rg ≤ 1.3 performed with PRIMUS41. The linearity of a Guinier plot of BAK α2–5 scattering data shows that higher molecular weight aggregates do not measurably contribute to scattering. g, h, Experimental SAXS curve measured for BAK α2–5 (black) overlayed with the theoretical scattering curve (red) generated with CRYSOL43 using the crystal structure of the BAK(α2–5) hexamer including PE lipids with acyl chains of lengths based on mass spec measurements (rather than length resolved in electron density) (g) and with lipids deleted from the model (h). The χ values of 0.939 and 1.592 indicate that the inclusion of lipids better represents the structure of the complex in solution.

Extended Data Fig. 3 Lipidomic analysis of BAK core domain protein, crystals and the cells in which it was expressed.

Mole percent composition of lipid classes (pie chart) and the 40 most abundant individual lipid species (bar chart) identified in a; purified BAK(α25) and b; BAK(α25) crystals. c, Mole percent composition of lipid classes (left) and the 40 most abundant individual lipid species (right) identified in E. coli expressing BAK(α25) 3 h after IPTG addition. Data are represented and mean ± SD for 3 technical replicates. Though PE is the lipid represented in our BAK core domain E. coli lipid structure, a lack of headgroup specificity suggests that many different lipid classes could occupy the α5α5’ grooves.

Extended Data Fig. 4 Lipid and detergent binding to BAK core domain dimers.

a, Three BAK core homodimers (surface representation, colored shades of blue, purple, and orange) formed a hexameric funnel in complex with 8:0–8:0 PS (yellow spheres). The exterior surface of the funnel is hydrophobic, made up of the lipophilic α4α5α5’α4’ surfaces of the homodimers with bound 8:0–8:0 PS. b, A near-isomorphic crystal structure in which DDM (gray spheres) replaces 8:0–8:0 PS. In both the 8:0–8:0 PS- and DDM-mediated crystals, the core dimer interface Sc is ~0.77 over ~1250 Å2 whilst the inter-dimer protein-protein contacts around the local trimer axis have Sc of ~0.67 over ~190 Å2. Despite the geometrical appeal of a funnel shaped oligomer with a hydrophilic pore as a potential intermediate in a larger BAK pore-forming oligomer, the small surface area involved in this crystal contact questions any biological relevance. c, The lipophilic α4α5α5’α4’ surface of each dimer binds 2 or more DDM molecules. The primary α5α5’ binding grooves are always occupied by the alkyl chains of DDM, while the secondary α4α5 crevices are sporadically occupied. d, Two crystallographically distinct D2 sandwich-like particles (ribbon representation, particle 1 colored shades of blue and purple and particle 2 colored shades of orange and pink) in the asymmetric unit of the crystal structure of BAK core domains in complex with 8:0–8:0 PG. Two 8:0–8:0 PG are well resolved in electron density in complex with particle 1, bridging the two dimers via the phosphate and sn2 acyl chain binding to the α5α5’ groove one dimer and the sn1 chain binding to the α4α5 crevice of the second dimer. Density for acyl chains is observed in the α5α5’ grooves of each dimer where a full 8:0–8:0 PG is not resolved. e, Side-by-side dimers in ribbon (upper panel) and surface (lower panel) representation colored shades of blue and purple from the crystal structure of BAK core domains in complex with POPC (orange spheres) and C8E4 (gray spheres). Two C8E4 molecules sit at the junction of the dimers, each with its alkyl chain bound to an α5α5’ groove and the ethylene glycol moiety bridging the homodimers. At the α5α5’ groove site distal to the dimer junction, POPC and C8E4 are present on the left- and right-hand dimers, respectively. Additional C8E4 molecules surround the POPC and bind to shallow clefts on the right-hand dimer. f, BAK core homodimer in the asymmetric unit (ribbon representation, shades of blue) forms a D2 sandwich with a symmetry mate (ribbon representation, gray) in complex with 16:0–0:0 LysoPC (magenta spheres) and what could be either PEG molecules or LysoPC sn1 acyl chains. The headgroups of the lipid were not resolved in the electron density, suggesting an intact sn2 chain (not present in 16:0–0:0 LysoPC) is required for sn3 phosphate binding.

Extended Data Fig. 5 Mass spectrometric analysis UV-induced BAK-lipid crosslinking.

a, UV-activatable pacFA-18:1 phosphatidylcholine (PC) probe62. The probe contains a photoactivatable and crosslinkable fatty acyl (pacFA) chain within the sn1 chain. b, BAK ΔN22ΔC25ΔCys 6His was added to mitochondrial mix liposomes doped with 5% nickel-chelating lipid. For mitochondrial mix liposomes, the PC component was either 16:0–18:1 PC (POPC) or the pacFA-18:1 PC probe. BAK was activated on liposomes with BID for 2 h then UV irradiated for 1 min to activate the diazirine group for crosslinking. Following UV irradiation, the samples were subjected to a click reaction to attach a TAMRA fluorophore. The UV fluorescence gel on the left demonstrates robust BAK labeling in the BID-activated and UV-irradiated samples containing the lipid probe. The western blot on the right confirms the band as BAK protein. Gaps in the western blot figure separate blots that were run simultaneously with samples prepared in the same experiment. c, Liposome dye release experiments confirm mitochondrial liposomes are still permeabilised by BAK when POPC is substituted with pacFA-18:1 PC. Normal mitochondrial liposomes and pacFA-18:1 PC-substituted mitochondrial liposomes containing the self-quenching dye 5(6)-carboxyfluorescein were incubated with BAK ΔN22ΔC25ΔCys 6His ± BID BH3 peptide. Data are normalised to BAK + BID BH3 peptide in normal mitochondrial liposomes and represented as mean ± SEM, n = 3. Means of untreated and BID BH3-treated samples were compared for POPC- vs. pacFA-18:1 PC-containing liposomes by unpaired two-tailed t-tests in Prism (GraphPad Software), ns = not significant. d, Representative mass spectrum and annotated peptide sequence for the highest scoring BAK peptide labeled by the lipid probe demonstrating high-confidence localization of the pacFA-18:1 PC probe sn1 linkage site. Lowercase residues indicate site of modification; n - cross-linked pacFA-18:1 PC, m – oxidation. Uncropped gel images and data for graphs are shown in the Source Data online.

Source data

Extended Data Fig. 6 Representative mass spectra of different pacFA-18:1 PC linkage sites related to Extended Data Fig. 6.

Representative annotated MS/MS spectra of 2 regions where cross-linked pacFA-18:1 PC was observed. Lowercase residues indicate site of modification; f, t, n - cross-linked pacFA-18:1 PC, m – oxidation.

Extended Data Fig. 7 All-atom molecular modelling of BAK WT and mutant core domain dimers in lipid membranes.

a, Density calculation for position of lipid acyl chains bound to the BAK α5α5’ binding site throughout simulation. The α5α5’ groove shows a marked preference for binding the sn2 (pink; upper) acyl chain over sn1 (purple; lower). Lipids within the groove are calculated with the VMD volmap plugin from MD simulation. The protein is shown for reference in blue and orientated with the lipophilic α4α5α5’α4’ surface towards the bottom. b, Position of sn3 phosphate of lipid bound to the α5α5’ binding site across simulation of WT BAK (40 snapshots throughout simulation from red-gray-blue). The RMSF (root mean square fluctuation) of the phosphate throughout the trajectory was 2.4 Å. c, Position of sn3 phosphate of lipid bound to the α5α5’ binding site throughout simulation of BAK mutant (W125P/R137A) (40 snapshots throughout simulation from red-gray-blue). The RMSF of the phosphate throughout the trajectory was 3.8 Å. The varied interactions between the sn3 phosphate and the α5α5’ binding site, along with the disparity in the molecular topology between native and mutant BAK dimers at the binding site (that is the absence of the R137’ side chain and W125 backbone amide), preclude a comparative calculation of distances between the sn3 phosphate and the relative binding sites. d, Top-down view of the time averaged local membrane thickness of the BAK mutant (W125P/R137A) dimer. Membrane directly under the dimer thins to approximately 30 Å, similar to that seen for wild type (Fig. 4b). This isn’t unexpected as while there is no preference for sn1 vs sn2 at the α5α5’ groove, and the sn3 headgroup is no longer retained at the headgroup binding site, lipid tails still engage the α4α5α5’α4’ surface thus drawing them away from the center of the bilayer leading to thinning. Thus, the modeling suggests that the primary defect of the mutant is not due to an impact on thinning but instead on altered engagement with the lipid headgroup leading to decreased bridging between dimers. e, The BAK W125P/R137A double mutant, equivalent to that used for in silico modelling, has reduced activity in cell killing assays as seen for single mutants (presented in Fig. 5c). Inset; western blot of Bak protein levels for cells used in Extended Data Fig. 7e, VDAC blot included as loading control. Data presented in Fig. 5c and Extended Data Fig. 7e are from experiments performed at the same time under the same conditions and the same data are presented for WT and MKR in both figures. Data represent mean ± SEM of 4 independent experiments. f, The BAK W125P/R137A double mutant forms dimers (2x) but not higher order oligomers (4x, 6x and above). Thus, like the single mutants (presented in Fig. 5d), the double mutant can form the core dimer but these have a reduced capacity to assemble into larger complexes. Performed under the same conditions as that described in the legend for Fig. 5d. Representative of 3 independent experiments. Uncropped gel images and data for graphs are shown in the Source Data online.

Source data

Extended Data Fig. 8 Phospholipase A2 digestion of BAK oligomers and liposome dye release by BAK lipid binding mutants.

a, S200 gel filtration chromatogram of oligomeric BAK ΔN22ΔC25 C166S 6His generated at a 1:20 protein:lipid ratio on Ni-POPC liposomes either untreated (blue trace), treated with phospholipase A2 (PLA2) inactivated by EDTA (pink trace) or treated with active PLA2 (orange trace). The peak shift to a smaller size indicates that the oligomer decreased in molecular weight following treatment with active PLA2. b, Digestion of BAK oligomers generated at a 1:20 protein:lipid ratio on Ni-Mito and Ni-POPC liposomes with PLA2 visualised by BN-PAGE. Purified BAK oligomers were treated with increasing concentrations of PLA2 for up to 1 h. As a negative control, PLA2 was preincubated with 20 mM EDTA prior to addition of BAK oligomer (“Inactivated PLA2”). As in Fig. 5a, BAK oligomers were reduced to smaller species over time by the activity of the phospholipase. The oligomers generated on Ni-POPC liposomes broke down further than those generated on Ni-Mito liposomes, to roughly the size of a dimer (though a tetramer sized band dominates). c, Schematic explaining banding patterns for Fig. 5b (based on ref. 3). Top panel here is equivalent to the experimental set up of the top panel for Fig. 5b and for Fig. 5d. In this scenario crosslinking can occur within (M71C-K113C) and between (H164C) dimers, high order species represent larger oligomers composed of dimers. Lower panel here is equivalent to set up of Fig. 5b lower panel. In this scenario crosslinking only occurs between (H164C) dimers within the higher order oligomer. In Fig. 5b treatment with lipase results in a reduction in the presence of higher order species in the top panel and of dimer in the lower panel, consistent with a reduction in the presence of higher order oligomers but not BH3:groove homodimers.

Supplementary information

Supplementary Information

Supplementary Fig. 1 and Supplementary Table 1

Reporting Summary

Supplementary Video 1

Coarse-grained molecular simulations of membrane self-assembly in combination with the BAK core domain dimer.

Supplementary Video 2

All-atom modeling of BAK core domain dimer embedded in membrane.

Supplementary Video 3

All-atom modeling of BAK W125P/R137A mutant core domain dimer embedded in membrane.

Source data

Source Data Fig. 5

Unprocessed western blots.

Source Data Fig. 5

Cell death data.

Source Data Extended Data Fig. 5

Unprocessed western blots.

Source Data Extended Data Fig. 5

Liposome data.

Source Data Extended Data Fig. 7

Unprocessed western blots.

Source Data Extended Data Fig. 7

Cell death data.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Cowan, A.D., Smith, N.A., Sandow, J.J. et al. BAK core dimers bind lipids and can be bridged by them. Nat Struct Mol Biol 27, 1024–1031 (2020). https://doi.org/10.1038/s41594-020-0494-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41594-020-0494-5

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing