Stimulation of the metabotropic GABAB receptor by γ-aminobutyric acid (GABA) results in prolonged inhibition of neurotransmission, which is central to brain physiology1. GABAB belongs to family C of the G-protein-coupled receptors, which operate as dimers to transform synaptic neurotransmitter signals into a cellular response through the binding and activation of heterotrimeric G proteins2,3. However, GABAB is unique in its function as an obligate heterodimer in which agonist binding and G-protein activation take place on distinct subunits4,5. Here we present cryo-electron microscopy structures of heterodimeric and homodimeric full-length GABAB receptors. Complemented by cellular signalling assays and atomistic simulations, these structures reveal that extracellular loop 2 (ECL2) of GABAB has an essential role in relaying structural transitions by ordering the linker that connects the extracellular ligand-binding domain to the transmembrane region. Furthermore, the ECL2 of each of the subunits of GABAB caps and interacts with the hydrophilic head of a phospholipid that occupies the extracellular half of the transmembrane domain, thereby providing a potentially crucial link between ligand binding and the receptor core that engages G proteins. These results provide a starting framework through which to decipher the mechanistic modes of signal transduction mediated by GABAB dimers, and have important implications for rational drug design that targets these receptors.
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
All data generated or analysed during this study are included in the Article and its Supplementary Information. Cryo-EM maps of GABAB heterodimer and GABAB1b homodimer have been deposited in the Electron Microscopy Data Bank under accession codes EMD-21533 and EMD-21534, respectively. The atomic coordinates of GABAB heterodimer and GABAB1b homodimer have been deposited in the Protein Data Bank under the accession codes 6W2X and 6W2Y, respectively.
Bettler, B., Kaupmann, K., Mosbacher, J. & Gassmann, M. Molecular structure and physiological functions of GABAB receptors. Physiol. Rev. 84, 835–867 (2004).
Mannoury la Cour, C., Herbelles, C., Pasteau, V., de Nanteuil, G. & Millan, M. J. Influence of positive allosteric modulators on GABAB receptor coupling in rat brain: a scintillation proximity assay characterisation of G protein subtypes. J. Neurochem. 105, 308–323 (2008).
Franek, M. et al. The heteromeric GABA-B receptor recognizes G-protein α subunit C-termini. Neuropharmacology 38, 1657–1666 (1999).
Robbins, M. J. et al. GABAB2 is essential for G-protein coupling of the GABAB receptor heterodimer. J. Neurosci. 21, 8043–8052 (2001).
Galvez, T. et al. Mutagenesis and modeling of the GABAB receptor extracellular domain support a Venus flytrap mechanism for ligand binding. J. Biol. Chem. 274, 13362–13369 (1999).
Hepler, J. R. & Gilman, A. G. G proteins. Trends Biochem. Sci. 17, 383–387 (1992).
Enna, S. J. GABAB receptor agonists and antagonists: pharmacological properties and therapeutic possibilities. Expert Opin. Investig. Drugs 6, 1319–1325 (1997).
Malcangio, M. GABAB receptors and pain. Neuropharmacology 136 (Pt A), 102–105 (2018).
Pin, J. P., Galvez, T. & Prézeau, L. Evolution, structure, and activation mechanism of family 3/C G-protein-coupled receptors. Pharmacol. Ther. 98, 325–354 (2003).
Bräuner-Osborne, H., Wellendorph, P. & Jensen, A. A. Structure, pharmacology and therapeutic prospects of family C G-protein coupled receptors. Curr. Drug Targets 8, 169–184 (2007).
Chun, L., Zhang, W. H. & Liu, J. F. Structure and ligand recognition of class C GPCRs. Acta Pharmacol. Sin. 33, 312–323 (2012).
Geng, Y., Bush, M., Mosyak, L., Wang, F. & Fan, Q. R. Structural mechanism of ligand activation in human GABAB receptor. Nature 504, 254–259 (2013).
Geng, Y. et al. Structural mechanism of ligand activation in human calcium-sensing receptor. eLife 5, e13662 (2016).
Koehl, A. et al. Structural insights into the activation of metabotropic glutamate receptors. Nature 566, 79–84 (2019).
Margeta-Mitrovic, M., Jan, Y. N. & Jan, L. Y. A trafficking checkpoint controls GABAB receptor heterodimerization. Neuron 27, 97–106 (2000).
Mukherjee, R. S., McBride, E. W., Beinborn, M., Dunlap, K. & Kopin, A. S. Point mutations in either subunit of the GABAB receptor confer constitutive activity to the heterodimer. Mol. Pharmacol. 70, 1406–1413 (2006).
Robertson, M. J., van Zundert, G. C. P., Borrelli, K. & Skiniotis, G. GemSpot: a pipeline for robust modeling of ligands into cryo-EM maps. Structure 28, 707–716 (2020).
Conklin, B. R., Farfel, Z., Lustig, K. D., Julius, D. & Bourne, H. R. Substitution of three amino acids switches receptor specificity of Gqα to that of Giα. Nature 363, 274–276 (1993).
Trinquet, E. et al. d-myo-Inositol 1-phosphate as a surrogate of d-myo-inositol 1,4,5-tris phosphate to monitor G protein-coupled receptor activation. Anal. Biochem. 358, 126–135 (2006).
Liu, J. et al. Molecular determinants involved in the allosteric control of agonist affinity in the GABAB receptor by the GABAB2 subunit. J. Biol. Chem. 279, 15824–15830 (2004).
Audet, M. & Stevens, R. C. Emerging structural biology of lipid G protein-coupled receptors. Prot. Sci. 28, 292–304 (2019).
Doré, A. S. et al. Structure of class C GPCR metabotropic glutamate receptor 5 transmembrane domain. Nature 511, 557–562 (2014).
Frangaj, A. & Fan, Q. R. Structural biology of GABAB receptor. Neuropharmacology 136 (Pt A), 68–79 (2018).
Gasparini, F. & Spooren, W. Allosteric modulators for mGlu receptors. Curr. Neuropharmacol. 5, 187–194 (2007).
Pagano, A. et al. C-terminal interaction is essential for surface trafficking but not for heteromeric assembly of GABAB receptors. J. Neurosci. 21, 1189–1202 (2001).
Balasubramanian, S., Teissére, J. A., Raju, D. V. & Hall, R. A. Hetero-oligomerization between GABAA and GABAB receptors regulates GABAB receptor trafficking. J. Biol. Chem. 279, 18840–18850 (2004).
Hyland, N. P. & Cryan, J. F. A gut feeling about GABA: focus on GABAB receptors. Front. Pharmacol. 1, 124 (2010).
Ng, T. K. & Yung, K. K. Differential expression of GABABR1 and GABABR2 receptor immunoreactivity in neurochemically identified neurons of the rat neostriatum. J. Comp. Neurol. 433, 458–470 (2001).
Burman, K. J. et al. GABAB receptor subunits, R1 and R2, in brainstem catecholamine and serotonin neurons. Brain Res. 970, 35–46 (2003).
Calver, A. R. et al. The expression of GABAB1 and GABAB2 receptor subunits in the cNS differs from that in peripheral tissues. Neuroscience 100, 155–170 (2000).
Gassmann, M. et al. Redistribution of GABAB(1) protein and atypical GABAB responses in GABAB(2)-deficient mice. J. Neurosci. 24, 6086–6097 (2004).
Margeta-Mitrovic, M., Jan, Y. N. & Jan, L. Y. Ligand-induced signal transduction within heterodimeric GABAB receptor. Proc. Natl Acad. Sci. USA 98, 14643–14648 (2001).
Xue, L. et al. Rearrangement of the transmembrane domain interfaces associated with the activation of a GPCR hetero-oligomer. Nat. Commun. 10, 2765 (2019).
Binet, V. et al. The heptahelical domain of GABAB2 is activated directly by CGP7930, a positive allosteric modulator of the GABAB receptor. J. Biol. Chem. 279, 29085–29091 (2004).
Galvez, T. et al. Allosteric interactions between GB1 and GB2 subunits are required for optimal GABAB receptor function. EMBO J. 20, 2152–2159 (2001).
Guan, X. M., Kobilka, T. S. & Kobilka, B. K. Enhancement of membrane insertion and function in a type IIIb membrane protein following introduction of a cleavable signal peptide. J. Biol. Chem. 267, 21995–21998 (1992).
Mastronarde, D. N. Automated electron microscope tomography using robust prediction of specimen movements. J. Struct. Biol. 152, 36–51 (2005).
Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy. Nat. Methods 14, 331–332 (2017).
Zhang, K. Gctf: real-time CTF determination and correction. J. Struct. Biol. 193, 1–12 (2016).
Rohou, A. & Grigorieff, N. CTFFIND4: fast and accurate defocus estimation from electron micrographs. J. Struct. Biol. 192, 216–221 (2015).
Zivanov, J. et al. New tools for automated high-resolution cryo-EM structure determination in RELION-3. eLife 7, e42166 (2018).
Pettersen, E. F. et al. UCSF Chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612 (2004).
Jacobson, M. P. et al. On the role of the crystal environment in determining protein side-chain conformations. J. Mol. Biol. 320 (3), 597–608 (2002).
Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. D 66, 486–501 (2010).
Liebschner, D. et al. Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix. Acta Crystallogr. D 75, 861–877 (2019).
Lomize, M. A., Pogozheva, I. D., Joo, H., Mosberg, H. I. & Lomize, A. L. OPM database and PPM web server: resources for positioning of proteins in membranes. Nucleic Acids Res. 40, D370–D376 (2012).
Jo, S., Kim, T., Iyer, V. G. & Im, W. CHARMM-GUI: a web-based graphical user interface for CHARMM. J. Comput. Chem. 29, 1859–1865 (2008).
Humphrey, W., Dalke, A. & Schulten, K. VMD: visual molecular dynamics. J. Mol. Graph. 14, 33–38 (1996).
Phillips, J. C. et al. Scalable molecular dynamics with NAMD. J. Comput. Chem. 26, 1781–1802 (2005).
Robertson, M. J., Tirado-Rives, J. & Jorgensen, W. L. Improved peptide and protein torsional energetics with the OPLSAA force field. J. Chem. Theory Comput. 11, 3499–3509 (2015).
Ryckaert, J.-P., Ciccotti, G. & Berendsen, H. J. C. Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. J. Comput. Phys. 23, 327–341 (1977).
Miyamoto, S. & Kollman, P. A. Settle – an analytical version of the shake and rattle algorithm for rigid water models. J. Comput. Chem. 13, 952–962 (1992).
Laurent, B. et al. Epock: rapid analysis of protein pocket dynamics. Bioinformatics 31, 1478–1480 (2015).
Ballesteros, J. A. & Weinstein, H. in Methods in Neurosciences vol. 25 (ed. Sealfon, S. C.) 366–428 (Academic, 1995).
Hilger, D. et al. Structural insights into ligand efficacy and activation of the glucagon receptor. Preprint at https://www.biorxiv.org/content/10.1101/660837v1 (2019).
Towns, J. et al. XSEDE: accelerating scientific discovery. Comput. Sci. Eng. 16, 62–74 (2014).
Scheres, S. H. Processing of structurally heterogeneous cryo-EM data in RELION. Methods Enzymol. 579, 125–157 (2016).
Afonine, P. V. et al. New tools for the analysis and validation of cryo-EM maps and atomic models. Acta Crystallogr. D 74, 814–840 (2018).
Kato, K., Goto, M. & Fukuda, H. Regulation by divalent cations of 3H-baclofen binding to GABAB sites in rat cerebellar membranes. Life Sci. 32, 879–887 (1983).
Bowery, N. G., Hill, D. R. & Hudson, A. L. Characteristics of GABAB receptor binding sites on rat whole brain synaptic membranes. 1983. Br. J. Pharmacol. 120 (Suppl), 452–467, discussion 450–451 (1997).
Haga, K. et al. Structure of the human M2 muscarinic acetylcholine receptor bound to an antagonist. Nature. 482, 547–551 (2012).
Marheineke, K., Grünewald, S., Christie, W. & Reiländer, H. Lipid composition of Spodoptera frugiperda (Sf9) and Trichoplusia ni (Tn) insect cells used for baculovirus infection. FEBS Lett. 441, 49–52 (1998).
Ma, J. C. & Dougherty, D. A. The cation–π interaction. Chem. Rev. 97, 1303–1324 (1997).
This work is supported by National Institutes of Health (NIH) grant R01 NS092695 (G.S. and J.M.M.) and used the Extreme Science and Engineering Discovery Environment (XSEDE)56 resource comet-gpu through sdsc-comet allocation TG-MCB190153, which is supported by National Science Foundation grant number ACI-1548562. We thank Q. Qu for advice and assistance with cryo-EM.
The authors declare no competing interests.
Peer review information Nature thanks Ryan Hibbs and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 Sample preparation, cryo-EM processing and reconstruction of GABAB heterodimer.
a, Purification scheme for GABAB. b, c, Representative cryo-EM micrograph of 10,315 collected (b) and 2D class averages (c) of GABAB dimers. d, Flow chart outlining the cryo-EM processing workflow using RELION57, the global resolutions of the full-length structure and VFT focused structures were 3.6 Å and 3.5 Å, respectively, at 0.143 Fourier shell correlation (FSC) as calculated by RELION. e, f, Gold-standard FSC curve of half-maps calculated using RELION and Phenix Mtriage58 (e), and map-to-model validation curves generated through Phenix Mtriage (f). g, Local resolution of cryo-EM maps. h, Angular distribution of projections used in final cryo-EM reconstruction. i, Ordered cryo-EM densities (light yellow), probably corresponding to GDN and/or cholesteryl hemisuccinate, are found at the TM5 interface of GABAB1 and GABAB2.
a, Electron microscopy density and model for GABAB heterodimer complex; transmembrane helices of GABAB1, transmembrane helices of GABAB2, linker region, bound phosphatidylethanolamine (PE) and ligand CGP55845. Densities visualized within UCSF Chimera42 and zoned at 2.2 with threshold set to 0.0142, with the exception of the following: GABAB1-bound lipid,GABAB1-bound CGP55845 and GABAB2 linker, in which thresholds of 0.01, 0.0189 and 0.0127 were used, respectively. b, Electron microscopy density and model for GABAB1 homodimer; transmembrane helices and linker region of both protomers, 7TM-bound PE and ligand CGP55845. Densities were zoned at 2.2 and threshold set to 0.016, apart from CGP55845 and the ECL2 in both protomers, in which a threshold of 0.03 or 0.02 was used, respectively.
a, Model of CGP55845 within the VFT of GABAB1b. The entire ligand is confined by W65 and W278 of GABAB1, which form hydrophobic interactions with the chlorinated ring of CGP55845. b, Schematic of interacting residues on GABAB1b with the inhibitor, CGP55845. GABAB1b residues S153 and S130 form hydrogen bonds with oxygen atoms of the phosphate group, and H170 and E349 form a hydrogen bond and a salt bridge with the amine group of the ligand, respectively. π–π stacking occurs between the chlorinated ring structure of CGP55845 and W278, and W65 provides hydrophobic packing on the opposing side of the ring. S130, H170, E349 and W65 are all substantially different residues in GABAB2, precluding ligand binding. Residues are colour-coded corresponding to their properties: light blue, hydrophilic; orange, anionic; green, hydrophobic; yellow, glycine; and grey, cysteine. Interaction lines are also colour-coded according to their type: light blue, side-chain hydrogen bonding; grey, backbone hydrogen bonding; blue–red gradient, salt bridge; and green, π–π stacking. c, Overlay of CGP55845-bound GABAB cryo-EM structure (tan and teal) with CGP54626-bound GABAB crystal structure (pink) (PDB 4MR7) or apo-state GABAB crystal structure (grey) (PDB 4MQE), resulting in r.m.s.d. values of 1.30 Å and 1.28 Å, respectively12. d, Comparison of ligand pose between CGP55845 (yellow) and CGP54626 (blue), which differs from CGP55845 only in a substitution of an aromatic ring in place of cyclohexane. e, Spherical density surrounded by anionic residues within the VFT supports a cation (magenta) at that site. The presence of a metal ion would be consistent with the observation that calcium and other divalent ions affect ligand affinity59,60, and examination of the deposited scattering factors for the high-resolution VFT crystal structure (PDB 4MR712) reveals positive difference density also consistent with a cation at this site.
a, GABAB protomers share similar secondary and overall structure. b, Comparison of mGlu5 and GABAB 7TM and ECL2–linker shown from side view. c, Top-down view of GABAB1, GABAB2, mGlu5 (PDB 6N5214) and family A M2 acetylcholine receptor (PDB 3UON61) with sidechains corresponding to the toggle-switch motif shown. Phospholipid space-filling model is included in grey within GABAB1 and GABAB2. d, Sequence alignment of human GABAB receptors with mGlu5 and M2 receptors, comparing canonical GPCR activation motifs: TM3–TM6 ionic lock (blue), toggle-switch motif (green) and FXPKXY motif (yellow). Sequences are aligned to motifs within each transmembrane helix and transmembrane helical secondary structure is underlined. Residues in GABAB sequences differing from canonical motifs are outlined in pink. The cysteine residue that replaces the toggle-switch tryptophan is highlighted in pink.
a, The results of three out of seven total simulations of GABAB 7TM and linker in the absence of core-bound lipid after 200 ns. The results show the extent of lipid (green and purple) entry in GABAB2 (grey) in the simulations versus the experimental structure (tan) with PE (orange). One of the trajectories was extended an additional 100 ns (rightmost panel) and lipid entry was observed to progress towards the core. b, Representative side view of the GABAB ribbon and stick model from simulations at 200 ns, showing persistence of the ECL2 and β-sheet structure even in absence of the VFTs. c, d, Violin plots of ensemble distances between GABAB1 (d) and GABAB2 (e) TM helices in simulations with and without core-bound lipid. Distances were measured from the Cα atoms of the following residues in GABAB1: L550 (TM3), W611 (TM4), L653 (TM5), M707 (TM6) and A716 (TM7); and the following residues in GABAB2: T557 (TM3), W615 (TM4), L657 (TM5), F711 (TM6) and Q720 (TM7). Simulations were run over 200 ns and violin plots represent n = 37,500 data points and 5 simulations per condition. e, 7TM cavity volume measured over time for individual simulations. Average cavity volume of phospholipid-bound receptor is shown as dashed line, thick lines indicate rolling averages of 5.33 ns and thin lines represent raw data.
a, Comparative normalized surface-expression levels of constructs. Surface-expression levels of wild-type and mutant GABAB receptors were determined using a direct ELISA against the N-terminal GABAB1 Flag tag and the N-terminal GABAB2 HA tag. Surface-expression levels were normalized to the surface expression of a wild-type GABAB receptor when 6.3 ng Flag-tagged GABAB1 and 6.3 ng HA-tagged GABAB2 DNA was used for transfection. The line x = y indicates similar surface expression of GABAB subunits as evident from transfection with lower amounts of Flag-tagged GABAB1 and HA-tagged GABAB2. Values above the line have greater GABAB2 surface expression relative to GABAB1, and values below the line have greater GABAB1 surface expression relative to GABAB2. GABAB1 did not reach the cell surface (data point hidden behind triangle at the coordinate (0,0) in the left subpanel). b, c, Mutations of ionic residues forming the interface of GABAB1 (b) and GABAB2 (c) result in increased constitutive activity of the receptor. To achieve a range of expression levels of the mutants in b and c, the DNA amounts transfected were serially diluted twofold from 3.3 ng (mutants in b) or 6.3 ng (mutants in c) of each subunit, respectively. d, e, GABAB2(H579A/E677A) expressed without GABAB1 shows a moderate increase in basal activity over wild-type GABAB2 expressed alone (d), although surface expression (e) of the wild-type GABAB2 subunit was higher than of GABAB2(H579A/E677A). f, g, Mutation of lipid coordinating residues (blue triangle) increase the constitutive activity of GABAB1 (f) and GABAB2 (g), whereas mutations displacing the lipid tails from the 7TM core (purple triangle) result in decreased basal activity of the receptor when compared to wild-type receptor of similar receptor surface expression. h, Key to colours and symbols used in a, f, and g with DNA transfection amounts indicated in parentheses. Data are representative of one experiment performed in triplicate and repeated independently at least three times with similar results (n = 3 independent experiments), error bars represent mean ± s.e.m.
a, Schematic of GABAB1 residues interacting with the polar headgroup of PE. The terminal amine (-NH3+) forms a salt bridge with residue D714 in ECL3, and the phosphate group is coordinated by S710 of ECL3, R549 of TM3 and the backbone nitrogen of the C644 of ECL2. b–d, As our receptor purification did not contribute additional lipid, we considered the known lipid composition of Sf9 insect cells. Four primary phospholipids are present in Sf9 insect cells: phosphatidylcholine (43%), phosphatidylethanolamine (32%), phosphatidylinositol (23%) and cardiolipin (4%)62. It was apparent from the map that the lipid had only two carbon chains, immediately excluding cardiolipin as it has four hydrocarbon tails. A comparison of the map and the binding site residues led to the decision to model PE (b) into the pocket using the GemSpot pipeline17, which produced good cross-correlation with favourable interactions. To further confirm our selection, analysis of overlays of phosphatidylcholine (c) and phosphatidylinositol (d) over the docked model revealed phosphatidylcholine is unlikely given that the interactions with the cation appear to be primarily salt bridges, rather than the cation–π interactions that more commonly coordinate choline in proteins63. Although phosphatidylinositol may make favourable interactions, our map does not appear to support such a large moiety in the headgroup position. Thus, PE is the most likely lipid to reside in the structure and was therefore used in the models.
a, Flow chart outlining the cryo-EM processing of the GABAB homodimer. b, c, Representative micrograph of 5,602 collected (b) and 2D class averages (c). d, Local resolution of cryo-EM maps. e, Angular distribution of projections used in the final cryo-EM reconstruction. f, g, Gold-standard FSC curve of half-maps calculated using RELION and Phenix Mtriage58 (f), and map-to-model validation curves generated through Phenix Mtriage (g). Global indicated resolutions of the full-length structure and VFT structure were 3.2 Å and 3.1 Å, respectively, at 0.143 FSC as calculated by RELION. h, Electron microscopy maps of active mGlu5 (purple) (EMD-034514), and homodimeric GABAB1 (green). The GABAB1 homodimer adopts an overall architecture similar to that of active mGlu5.
Details of statistical analysis and reproducibility for cellular signalling assays.
Ensemble data of molecular dynamic simulation of GABAB without 7TM core lipid. Representative ensemble of ten structures from time points taken starting at 40 ns, and then every 20 ns thereafter. Provided as a composite PDB file.
Ensemble data of molecular dynamic simulation of GABAB with PE bound in 7TM core. Representative ensemble of ten structures from time points taken starting at 40 ns, and then every 20 ns thereafter. Provided as a composite PDB file.
About this article
Cite this article
Papasergi-Scott, M.M., Robertson, M.J., Seven, A.B. et al. Structures of metabotropic GABAB receptor. Nature 584, 310–314 (2020). https://doi.org/10.1038/s41586-020-2469-4