Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Jul;571(7764):279-283.
doi: 10.1038/s41586-019-1286-0. Epub 2019 Jun 5.

Cryo-EM structure of oxysterol-bound human Smoothened coupled to a heterotrimeric Gi

Affiliations

Cryo-EM structure of oxysterol-bound human Smoothened coupled to a heterotrimeric Gi

Xiaofeng Qi et al. Nature. 2019 Jul.

Abstract

The oncoprotein Smoothened (SMO), a G-protein-coupled receptor (GPCR) of the Frizzled-class (class-F), transduces the Hedgehog signal from the tumour suppressor Patched-1 (PTCH1) to the glioma-associated-oncogene (GLI) transcription factors, which activates the Hedgehog signalling pathway1,2. It has remained unknown how PTCH1 modulates SMO, how SMO is stimulated to form a complex with heterotrimeric G proteins and whether G-protein coupling contributes to the activation of GLI proteins3. Here we show that 24,25-epoxycholesterol, which we identify as an endogenous ligand of PTCH1, can stimulate Hedgehog signalling in cells and can trigger G-protein signalling via human SMO in vitro. We present a cryo-electron microscopy structure of human SMO bound to 24(S),25-epoxycholesterol and coupled to a heterotrimeric Gi protein. The structure reveals a ligand-binding site for 24(S),25-epoxycholesterol in the 7-transmembrane region, as well as a Gi-coupled activation mechanism of human SMO. Notably, the Gi protein presents a different arrangement from that of class-A GPCR-Gi complexes. Our work provides molecular insights into Hedgehog signal transduction and the activation of a class-F GPCR.

PubMed Disclaimer

Conflict of interest statement

Competing interests The authors declare no competing financial interests.

Figures

Extended Data Fig. 1
Extended Data Fig. 1. Assembly of hSMO–Gi–Fab complex
a, GTPγS causes the dissociation of 24(S),25-EC mediated hSMO–Gi complex. b, Size-exclusion chromatogram and SDS–PAGE gel of the purified hSMO–Gi–Fab complex. Molecular standards are indicated on left side of the gel.
Extended Data Fig. 2
Extended Data Fig. 2. Data processing.
a, A representative electron micrograph at −2.0 μm defocus. b, The data processing work-flow for the complex with the full map. The cryo-EM 2D classification from RELION is shown. The subtracted parts were indicated by dash circles. c, The data processing work-flow for the complex with the subtracted map. Class 3 of the full map and Class 4 of the subtracted map were used for the final refinement; Class 4 of the full map and Class 1 of the subtracted map failed to have sufficient structural features in the final refinement. Masks used for the refinement are shown. The cryo-EM map after Frealign refinement sharpened using BFACTOR.EXE (author: Nikolaus Grigorieff) with a resolution limit of 4 Å or 3.9 Å and a B-factor value of −100 Å2. Each subunit is colored.
Extended Data Fig. 3
Extended Data Fig. 3. The model quality assessment.
a, Fourier shell correlation (FSC) curve of the structure without the CRD and half Fab with FSC as a function of resolution using Frealign output. b, The FSC curves calculated between the refined structure and the half map used for refinement (blue), the other half map (red) and the full map (black). c, Density maps of structure colored by local resolution estimation using Blocres.
Extended Data Fig. 4
Extended Data Fig. 4. cryo-EM map of structural elements in the complex.
a, The major helices of hSMO. b, The major structural elements of Gi protein. EM density map and model of the complex are shown in mesh and cartoon. c, the putative ligand.
Extended Data Fig. 5
Extended Data Fig. 5. Comparison of the maps in the ligand-binding pocket of hSMO.
a, The extra density within the TMD ligand-binding pocket in the hSMO crystal structure (PDB: 5L7D). The density is shown in green at 3σ level and indicated by arrow. b, The density of the ligand in the Gi–hSMO complex. The density is shown in purple mesh at 5σ level at 3.9Å and indicated by arrow.
Extended Data Fig. 6
Extended Data Fig. 6. Comparison of the binding sites of different SMO ligands.
a, SAG1.5 bound hSMO (PDB:4QIN). b, Cyclopamine bound hSMO (PDB: 4O9R). c, Vismodegib bound hSMO (PDB: 5L7I). d, SANT1 bound hSMO (PDB:4N4W). e, LY2940680 bound hSMO (PDB:4JKV). Structures of hSMO with different ligands viewed from the side of the membrane. f, Superimposition of the ligands that bind the pocket in the transmembrane domain of hSMO.
Extended Data Fig. 7
Extended Data Fig. 7. Comparisons of the TM6s and cytosolic sites of hSMO and μOR.
a, Structural comparison of TM6s of hSMO, μOR and GLP-1R in the inactive and G protein-bound states. Left: hSMO, inactive SMO in pink (PDB: 5L7D); Middle: μOR, inactive μOR in light orange (PDB: 4DKL), Gi-μOR in light cyan (PDB: 6DDE); Right: GLP-1R, inactive GLP-1R in red (PDB: 5VEW), Gs-GLP-1R (PDB: 6B3J) in dark blue. b, Electrostatic surface representations of the cytosolic side of SMO and μOR complex with Gαi-α5.
Extended Data Fig. 8
Extended Data Fig. 8. The structures of Gi-bound Class-A GPCRs.
a, Rhodopsin-Gi complex (PDB: 6CMO). b, A1R-Gi complex (PDB: 6D9H). c, μOR-Gi complex (PDB: 6DDE). d, CB1-Gi complex (PDB: 6N4B).
Extended Data Fig. 9
Extended Data Fig. 9. Comparison of the Gi coupled hSMO, inactive hSMO and apo hFZD4.
The Gi coupled hSMO is in blue, the inactive hSMO is in pink (PDB: 5L7D) and apo hFZD4 is in yellow (PDB: 6BD4). Structures are viewed from the side of the membrane.
Fig. 1
Fig. 1. Functional characterization of PTCH1-associated oxysterols in HH signaling.
a, The sterol-like densities in the SHH-N mediated PTCH1 dimer. Sterol-like densities at 5s level at 3.5 Å resolution in the domains of ECD-I, SSDs and near TM-12 are colored in green, red and purple, respectively. b, HPLC-MS quantitation of oxysterols extracted from purified PTCH1 protein. Data are mean ± s.d. (n = 3 biologically independent experiments). Oxysterol structures are shown. c, Oxysterol-mediated HH signaling. The SHH-light II cells were treated with vehicle (0.3 mM MCD), MCD complexed with 30μM sterol or SHH-N conditioned media. d, GTPγS binding assay. Basal represents the hSMO basal activity without ligand. All ligands were used at a saturating concentration of 50 μM. e, PTX decreases the 24(S),25-EC mediated HH signaling. HH activity was measured by dual-luciferase assay. Each assay in c-e was repeated at least three times with similar results and data are mean ± s.d. (n = 3 biologically independent experiments). *P ≤ 0.05, **P ≤ 0.01, two-sided t-test using GraphPad Prism 7.
Fig. 2
Fig. 2. Structure of hSMO–Gi–Fab complex.
a, Ribbon representation of the complex structure. Primary structure of hSMO is on the top. Residues 556–787 of hSMO were removed for protein expression and the CRD domain (gray) was not determined in the cryo-EM map. hSMO, Gα, Gβ, Gγ and Fab-G50 are colored in blue, green, magenta, dark teal and orange; the putative 24(S),25-EC is shown as yellow sticks. b, The ligand-binding pocket. The putative ligand and its bound residue are shown as sticks. c, HH signaling in Smo−/− MEFs transfected with pcDNA3.1, full-length hSMO-wild type (WT) or full-length hSMO-N521A mutant and response to SAG or 24(S),25-EC via luciferase activity. d, The GTPγS binding competition assay using cells overexpressing hSMO-ΔCRD. The assays were set up as Fig. 1d with various ligand concentrations. Each assay in c-d was repeated at least three times with similar results and data are mean ± s.d. (n = 3 biologically independent experiments). *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, two-sided t-test using GraphPad Prism 7.
Fig. 3
Fig. 3. Structural comparison of the Gi-bound hSMO with inactive hSMO and cyclopamine bound xSMO.
a, Superimposition of TMs of hSMO molecules. Gi-coupled hSMO is colored in blue, inactive hSMO (PDB: 5L7D) is colored in pink. The movements of structural elements are indicated. b, Movement of TM6 in the Gi-coupled hSMO due to ligand binding. c, Movements of TMs 5–7 in the Gi-coupled hSMO compared with the inactive hSMO. The related residues are shown as sticks. d, Superimposition of 7-TMs of hSMO (blue) and xSMO (PDB: 6D32, gray). e, Comparison of R451 and W535 in Gi-coupled hSMO with the corresponding residues in xSMO. f, Comparison of the ICL1 of hSMO and xSMO. R261 and its corresponding residue in xSMO are shown. g, The putative tunnel in hSMO is shown as red mesh.
Fig. 4
Fig. 4. Conformational changes in Gi upon coupling to hSMO.
a, The interaction details between hSMO (blue) and Gαi (green). The structural elements involved in the interaction are indicated. b, R257 in hSMO-ICL1 binds D312 of Gβ. Both residues are labeled and shown. c and d, Comparison of GDP-bound Gαi (PDB: 1GP2, gray) and nucleotide-free Gαi (green) from hSMO–Gi complex. GDP is shown as yellow sticks. e, Structural rearrangement of Gαi-AHD domain after coupling to hSMO.
Fig. 5
Fig. 5. Distinct orientations of heterotrimeric Gi proteins after coupling to hSMO and Class-A GPCRs.
a, Structural comparison of hSMO-Gi complex with μOR-Gi complex (PDB: 6DDE, light cyan). The Gαi-α5 and Gαi-αN are indicated. hSMO, Gα, Gβ and Gγ in hSMO-Gi complex are colored as Fig. 2a. b, The comparison of the ICL2 and ICL3 among the five GPCR-Gi complexes. The structural elements from Rhodopsin-Gi (PDB: 6CMO) are in orange; from A1R-Gi (PDB: 6D9H) are in pink, from μOR-Gi are in light cyan and from CB1-Gi (PDB: 6N4B) are in cyan. c, The comparison of the Gαi-α5 among the five GPCR-Gi complexes. d, The structural comparison of Gi proteins after coupling μOR and hSMO. e, The structural comparison of Gi proteins after coupling rhodopsin and hSMO. The major differences of Gi protein orientations are indicated.

References

    1. Jiang J & Hui CC Hedgehog signaling in development and cancer. Developmental cell 15, 801–812, doi: 10.1016/j.devcel.2008.11.010 (2008). - DOI - PMC - PubMed
    1. Petrov K, Wierbowski BM & Salic A Sending and Receiving Hedgehog Signals. Annual review of cell and developmental biology 33, 145–168, doi: 10.1146/annurev-cellbio-100616-060847 (2017). - DOI - PubMed
    1. Arensdorf AM, Marada S & Ogden SK Smoothened Regulation: A Tale of Two Signals. Trends in pharmacological sciences 37, 62–72, doi: 10.1016/j.tips.2015.09.001 (2016). - DOI - PMC - PubMed
    1. Taipale J, Cooper MK, Maiti T & Beachy PA Patched acts catalytically to suppress the activity of Smoothened. Nature 418, 892–897, doi: 10.1038/nature00989 (2002). - DOI - PubMed
    1. Qi X, Schmiege P, Coutavas E, Wang J & Li X Structures of human Patched and its complex with native palmitoylated sonic hedgehog. Nature 560, 128–132, doi: 10.1038/s41586-018-0308-7 (2018). - DOI - PMC - PubMed

MeSH terms