Cryo-EM analysis of PIP2 regulation in mammalian GIRK channels

  1. Yiming Niu
  2. Xiao Tao
  3. Kouki K Touhara
  4. Roderick MacKinnon  Is a corresponding author
  1. Laboratory of Molecular Neurobiology and Biophysics, The Rockefeller University, Howard Hughes Medical Institute, United States
5 figures, 2 tables and 1 additional file

Figures

Figure 1 with 3 supplements
GIRK2 adopts an extended conformation in the absence of PIP2.

(A) Conformational changes upon PIP2 binding in the Kir2.2 channel viewed from side with the extracellular side above (Left: the extended conformation without PIP2, PDB: 3JYC. Right: the docked conformation upon PIP2 binding, PDB: 3SPI). Four PIP2 molecules are shown as sticks and colored according to atom type: carbon, yellow; phosphorous, orange; and oxygen, red. The CTD translates towards the TMD by 6 Å upon PIP2 binding. A set of reference atoms (Asp72 and Lys220 α-carbons) are highlighted as blue spheres in each structure. (B) Side and top views of the cryo-EM structure of the GIRK2 channel in an extended conformation.

Figure 1—figure supplement 1
Cryo-EM analysis of the GIRK2 channel in the extended conformation, related to Figure 1.

(A) Representative cryo-EM image of the GIRK2 channel in the absence of PIP2. (B) Selected 2D-class averages of the GIRK2 channel. The scale bar is 17.5 nm. (C) Cryo-EM data processing workflow. (D) Gold-standard FSC curve after correction for masking effects. The resolution was estimated based on the FSC = 0.143 criterion. (E) Cross-validation FSC curves for the GIRK2 channel: blue curve, refined model versus half map one used for refinement (Work); red curve, refined model versus half map two not used for refinement (Free); black curve, refined model versus the combined map (Full). The similarity of the ‘work’ and ‘free’ curves suggests no substantial over-fitting. The correlation is above 0.5 up to a resolution of 4.1 Å. (F) Local resolution map calculated using Relion3.

Figure 1—figure supplement 2
Cryo-EM densities for selected regions of the GIRK2 extended conformation (contour level 6.5 in Coot), related to Figure 1.
Figure 1—figure supplement 3
Structural comparison of the apo GIRK2 determined by cryo-EM (gray) and X-ray crystallography (salmon, PDB: 3SYO), related to Figure 1.

(A) Overall conformational changes viewed from side with the extracellular solution above. The lipid bilayer boundaries are shown as grey bars. A set of reference atoms (Thr80 and Leu229 α-carbons) are highlighted as blue spheres. The CTD in the crystal structure translates towards the TMD by 6 Å. (B and C) Conformational changes between cryo-EM and X-ray structures of apo GIRK2 in TMD (B) and CTD (C). (D) Crystal packing in the CrystalApo structure. The channel tetramer is colored salmon and symmetry-related tetramers are colored orange. The crystal contact interface is boxed and a zoomed-in view is shown.

Figure 2 with 1 supplement
GIRK2 conformation as a function of PIP2 concentration.

(A) Structural titration image analysis workflow. Representative images of GIRK2 channels recorded in the presence of 0, 0.25, 0.5, 0.75, 0.875 or 1 mM C8-PIP2. GIRK2 particles were automatically selected from the images (green circles). The extracted particles from the respective images were combined into a single titration dataset for 3D refinement with C4 symmetry in RELION. Using the angles and translations obtained from the 3D refinement, the particles from the titration dataset were classified into five classes. The extended classes are colored red and the docked class blue. (B) Representative cryo-EM images of GIRK2 in the presence of 0, 0.25, 0.5, 0.75, 0.875 or 1 mM C8-PIP2. Particles marked with a red circle were classified as extended, and those with a blue circle were classified as docked. (C) The fraction of particles classified as the docked conformation of GIRK2 is plotted against the concentration of C8-PIP2.

Figure 2—figure supplement 1
Reproducibility of the 3D classification of extended and docked classes of the GIRK2 channel, related to Figure 2.

(A) Fraction of particles in the docked class from the titration dataset is plotted as a function of PIP2 concentration for five independent 3D refinement and classification runs. (B) Plot of the fraction of particles classified n times into the docked class during five independent 3D refinement and classification runs.

Figure 3 with 1 supplement
GIRK1/4 channels also form extended and docked conformations.

(A and B) Side views of the cryo-EM density map of the extended (A) and docked (B) conformations of the GIRK1/4 channel. (C) Top view of the CTD regions aligned with respect to the TMD reveals a 35° rotation from extended to docked conformations.

Figure 3—figure supplement 1
Cryo-EM analysis of the GIRK1/4 channel in the presence of 0.5 mM C8-PIP2, related to Figure 3.

(A) Representative cryo-EM image of the GIRK1/4 channel. (B) Selected 2D-class averages of the GIRK1/4 channel. The scale-bar is 17.5 nm. (C) Cryo-EM data processing workflow for the GIRK1/4 channel. (D) Gold-standard FSC curve after correction for masking effects. The resolution was estimated based on the FSC = 0.143 criterion.

Figure 4 with 3 supplements
Conformational changes between the docked and extended GIRK2 channel upon PIP2 binding.

(A) Side and top views of the cryo-EM density map of the docked conformation of GIRK2 channel (cyan) with four bound PIP2 molecules (orange). The PIP2 acyl chains were only partially resolved. (B) Close-up view of the PIP2 binding pocket. PIP2 is shown as sticks and colored according to atom type (carbon, yellow; phosphorous, orange; and oxygen, red). The PIP2 interacting residues are also shown as sticks (carbon, cyan and nitrogen, blue). (C) Comparison of the extended (gray, no PIP2) and the docked (cyan, with 4 PIP2 bound) structures. The channel is viewed from the side with the extracellular side above. The lipid bilayer boundaries are shown as grey bars. Four PIP2 molecules are shown as sticks and colored as in panel (B). The PIP2 molecule in a similar orientation as in (B) is outlined by a black box. Upon PIP2 binding, the CTD in the extended structure translates towards the TMD by 6 Å accompanied by a 35° rotation viewed from the extracellular side. (D and E) Local conformational changes at the binding site for Gβγ in the βL-βM loop of GIRK2 CTD are shown by structural superposition. The CTD region of GIRK2 CryoEMApo, CryoEMPIP2 and CrystalGIRK-Gβγ structures are colored gray, cyan, and blue, respectively. Rearrangement of the βL-βM loop is indicated by an arrow.

Figure 4—figure supplement 1
Cryo-EM analysis of the GIRK2 channel in the docked conformation, related to Figures 2 and 4.

(A) Cryo-EM data processing workflow for the GIRK2 channel. The 155 K particles correspond to all the particles classified as the docked class (class 5) in Figure 2A. (B) Gold-standard FSC curve after correction for masking effects. The resolution was estimated based on the FSC = 0.143 criterion. (C) Cross-validation FSC curves: blue curve, refined model versus half map one used for refinement (Work); red curve, refined model versus half map two not used for refinement (Free); black curve, refined model versus the combined final map (Full). The similarity of the ‘work’ and ‘free’ curves suggests no substantial over-fitting. The correlation is above 0.5 up to a resolution of 3.5 Å. (D) Local resolution map calculated using Relion3.

Figure 4—figure supplement 2
Cryo-EM densities for selected regions of the GIRK2 docked conformation (contour level 8.0 in Coot), related to Figures 2 and 4.
Figure 4—figure supplement 3
PIP2 binding pocket, related to Figure 4.

(A and B) Comparison of the PIP2 binding site in the CryoEMApo (white, panel (A)) and CryoEMPIP2 (cyan, panel (B)) structures. Cα atoms of residues involved in PIP2 binding are shown as spheres. Distances between K194 and K64 (Cα to Cα) are labeled.

PIP2 serves as an allosteric regulator to permit Gβγ binding.

(A and B) A cartoon depiction of PIP2 regulation of GIRK channels. The blue shape depicts the GIRK channel. a, b, c, d, and e indicate the selectivity filter, inner helix gate, TMD-CTD linker, the G-loop gate, and the βL-βM loop, respectively. Circular arrow indicates the rotation about the pore axis with respect to the TMD and perpendicular arrow indicates CTD translation upon PIP2 binding. The ‘gg’ label represents the geranylgeranyl lipid modification at the C terminus of Gγ. In the absence of PIP2 (A), GIRK2 channel adopts the extended conformation and is not positioned for Gβγ binding. Upon PIP2 binding (B), the GIRK2 channel transits to the docked conformation, allowing Gβγ binding to occur. (C) Plot of pore diameter (between van der Waals surfaces, calculated with Hole) for the extended (gray) and docked (cyan) structures. (D and E) Superposition of the channel TMD from the extended (gray) and docked (cyan) structures. Conformational changes at the inner helix gate (D) and G-loop gate (E) are boxed and zoomed-in details are shown. Sidechains of the inner helix gate-forming residue Phe192 are shown as sticks. Cα atoms of the G-loop gate constriction residue Met313 are shown as spheres.

Tables

Key resources table
Reagent type
(species) or
resource
DesignationSource or
reference
IdentifiersAdditional
information
Gene (Mus musculus GIRK2)GIRK2syntheticSynthesized at GeneWiz.
Gene (Homo sapiens GIRK1)GIRK1syntheticSynthesized at GeneWiz.
Gene (Homo sapiens GIRK4)GIRK4syntheticSynthesized at GeneWiz.
Strain, strain background (Escherichia coli)DH10BacThermoFisher10361012
Recombinant DNA reagentpPICZ-GIRK2https://doi.org/10.1016/j.cell.2011.07.046Maintained at the Mackinnon lab
Recombinant DNA reagentGIRK1-His10-pEG BacMamhttps://doi.org/10.7554/eLife.15750.001Maintained at the Mackinnon lab
Recombinant DNA reagentGIRK4-1D4-pEG BacMamhttps://doi.org/10.7554/eLife.15750.001Maintained at the Mackinnon lab
Cell line (Pichia pastoris)SMD1163InvitrogenC17500
Cell line
(Spodoptera frugiperda)
Sf9ATCCCat# CRL-1711
Cell line (Homo sapiens)HEK293S GnTI-ATCCCat# CRL-3022
Chemical compound, drugSF-900 II SFM mediumGIBCOCat# 10902–088
Chemical compound, drugL-Glutamine (100x)GIBCOCat# 25030–081
Chemical compound, drugPen StrepGIBCOCat# 15140–122
Chemical compound, drugGrace’s insect mediumGIBCOCat# 11605–094
Chemical compound, drugFreestyle 293 mediumGIBCOCat# 12338–018
Chemical compound, drugFetal bovine serumGIBCOCat# 16000–044
Chemical compound, drugCellfectin II reagentInvitrogenCat# 10362100
Chemical compound, drugCholesteryl hemisuccinate (CHS)AnatraceCH210
Chemical compound, drugn-Dodecyl-β-D-Maltopyranoside (DDM)AnatraceD310S
Chemical compound, drugn-Decyl-β-D-Maltopyranoside (DM)AnatraceD322S
Chemical compound, drug1,2-dioctanoyl-sn-glycero-3-phospho-(1'-myo-inositol-4',5'-bisphosphate) (ammonium salt) (C8-PIP2)Avanti Polar Lipids850185P
Chemical compound, drug(1H, 1H, 2H, 2H-Perfluorooctyl)phosphocholine (FFC8)AnatraceF300F
Commercial assay or kitCNBr-activated Sepharose beadsGE HealthcareCat# 17-0430-01
Commercial assay or kitSuperdex 200 Increase 10/300 GLGE Healthcare Life Sciences28990944
Commercial assay or kitR1.2/1.3 400 mesh Au holey carbon gridsQuantifoil1210627
Commercial assay or kitSuperose 6 Increase 10/300 GLGE Healthcare Life Sciences29091596
Software, algorithmRELION 3.0https://doi.org/10.7554/eLife.42166.001http://www2.mrc-lmb.cam.ac.uk/relion
Software, algorithmRELION 3.1https://doi.org/10.1101/798066http://www2.mrc-lmb.cam.ac.uk/relion
Software, algorithmMotionCor2https://doi.org/10.1038/nmeth.4193http://msg.ucsf.edu/em/software/motioncor2.html
Software, algorithmGctf 1.0.6https://doi.org/10.1016/j.jsb.2015.11.003https://www.mrc-lmb.cam.ac.uk/kzhang/Gctf/
Software, algorithmCtfFind4.1.8https://doi.org/10.1016/j.jsb.2015.08.008http://grigoriefflab.janelia.org/ctffind4
Software, algorithmGautomatchhttps://www.mrc-lmb.cam.ac.uk/kzhang/Gautomatch/
Software, algorithmCryoSPARC 2.4.0https://doi.org/10.7554/eLife.46057.001https://cryosparc.com/
Software, algorithmPyemhttps://github.com/asarnow/pyem
Software, algorithmCOOThttps://doi.org/10.1107/S0907444910007493http://www2.mrc-lmb.cam.ac.uk/personal/ pemsley/coot
Software, algorithmPHENIXhttps://doi.org/10.1107/S0907444909052925https://www.phenix-online.org
Software, algorithmAdobe Photoshop version 16.0.0 (for figure preparation)Adobe Systems, Inc
Software, algorithmGraphPad Prism version 8.0GraphPad Software
Software, algorithmMacPyMOL: PyMOL v2.0 Enhanced for Mac OS XSchrodinger LLChttps://pymol.org/edu/?q=educational/
Software, algorithmChimerahttps://doi.org/10.1002/jcc.20084https://www.cgl.ucsf.edu/chimera/download.html
Software, algorithmSerial EMhttps://doi.org/10.1016/j.jsb.2005.07.007http://bio3d.colorado.edu/SerialEM
Software, algorithmHOLEhttps://doi.org/10.1016/S0263-7855(97)00009-Xhttp://www.holeprogram.org
Table 1
Cryo-EM data collection and refinement statistics, related to Figures 1, 2 and 4.
GIRK2ExtendedGIRK2DockedGIRK1/4ExtendedGIRK1/4Docked
EMDB IDEMD-22199EMD-22200EMD-22201EMD-22202
PDB ID6XIS6XIT
Data collection
MicroscopeTitan Krios
DetectorK3 summitK2 summit
Voltage (kV)300
Pixel size (Å)0.430.50.515
Total electron exposure
(e-2)
103.380.075.4
Defocus range (μm)1.0 to
3.0
1.5 to
2.5
1.5 to
3.5
Micrographs collected210311,3493415
Reconstruction
Final particle images112,517155,12857,64448,757
Pixel size (Å)1.2911.031.03
Box size (pixels)256400256256
Resolution (Å)
(FSC = 0.143)
3.93.37.94.6
Map Sharpening B-factor (Å2)−26−12-−192
Model composition
Non-hydrogen atoms946010,252
Protein residues124013,08
Ligands04
Metals03
Refinement
Model-to-map CC (mask)0.620.72
Model-to-map CC (volume)0.640.75
R.m.s deviations
Bond length (Å)0.0060.009
Bond angles (°)1.31.3
Validation
MolProbity score2.001.80
Clash score8.559.91
Ramachandran plot
Outliers (%)00
Allowed (%)1.74.0
Favored (%)98.396.0
Rotamer outliers (%)1.400.74
C-beta deviations (%)00

Additional files

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Yiming Niu
  2. Xiao Tao
  3. Kouki K Touhara
  4. Roderick MacKinnon
(2020)
Cryo-EM analysis of PIP2 regulation in mammalian GIRK channels
eLife 9:e60552.
https://doi.org/10.7554/eLife.60552