Dynamic role of the tether helix in PIP2-dependent gating of a G protein–gated potassium channel

G protein–gated inwardly rectifying potassium (GIRK) channels are activated by the phospholipid phosphatidylinositol 4,5 bisphosphate (PIP2). Using functional and computational experiments, Lacin et al. reveal that PIP2 interacts with the tether helix of the neuronal GIRK channel in a dynamic way.


Electrophysiology
Whole-cell patch-clamp recordings were made as described previously (Bodhinathan and Slesinger, 2013). Currents were recorded with an Axopatch 200B (Molecular Devices, Axon Instruments) amplifier, compensated electronically for cell capacitance and series resistance (75-100%), filtered at 1 kHz with an 8-pole Bessel filter, and digitized at 5 kHz with a Digidata 1320 interface (Molecular Devices, Axon Instruments). Currents were elicited with voltage ramp protocol, from −120 mV to +50 mV, delivered at 0.5 Hz. Basal K + currents (Ba 2+ sensitive) were quantified at −120 mV by applying 1 mM BaCl 2 in the extracellular 20K solution and measuring the amplitude of the current inhibited by Ba 2+ . The effect of various modulators were measured at −100 mV or −120 mV by averaging current amplitudes from at least three consecutive sweeps at steady state and subtracting the mean basal current before and after application of the modulator. For studying the effect of depleting membrane PIP 2 , Dr-Vsp was activated by a voltage step from a holding potential of −40 mV to +100 mV for varying lengths of time, and the amplitude of the GIRK currents were then measured at −120 mV. This three-step protocol was repeated every 2 seconds (i.e., Dr-Vsp activation protocol; Bodhinathan and Slesinger, 2013). The extent of Dr-Vsp inhibition was calculated as a fraction of the Ba-sensitive basal current. The decrease in current (i.e., depletion) and recovery of current following Dr-Vsp activation were fit with a single exponential to determine the tau.

MD simulations
The crystal structure of GIRK2 channel (PDB: 4KFM) bound to a PIP 2 head group was used. We used the PDB: 4KFM structure for simulations, because this channel is in a preopen conformation (Whorton and MacKinnon, 2013). We compared the 3SYA and 4KFM structures, and we determined there is a small RMSD MacKinnon, 2011, 2013), with the largest movements in the LM loop and a 0.41 Å RMSD for the c-alpha in the PIP 2 pocket, formed by residues 88 to 92 (from the slide helix), 192 to 203 (HBC gate and tether helix), and 62 to 64 (N-terminal b-loop). Four 1-stearoyl-2-arachidonoyl-sn-glycero-3-phosphoinositol 4,5 bisphosphate (18:0, 22:4 PIP 2 ) molecules were aligned to the crystal structure of PIP 2 , one at each binding site. Gβγ subunits were not included in the calculations. The GIRK2-PIP 2 system was embedded in a bilayer consisting of POPC (16:0, 18:1 PC) lipids using the CHA RMM-GUI membrane builder (Brooks et al., 2009;Wu et al., 2014). The protein-lipid system was then solvated with TIP3P water molecules and 150 mM KCl. The CHA RMM36 force field was employed (Best et al., 2012). Additionally, five potassium ions were placed in the ion conduction pathway and four sodium ions were placed in the sodium-binding site as seen in the crystal structures. A second system was set up with the same conditions but with a 6′Y mutation generated by in silico mutagenesis using CHA RMM-GUI. The two systems (WT and 6′Y) were subjected to stepwise decreased restraint equilibration used in CHA RMM-GUI membrane builder (Jo et al., 2007). The two systems (WT and 6′Y) were then equilibrated for 20 ns using positional restrain of 1,000 KJ/mol/nm 2 on all Cα atoms of the protein before the 400-ns unrestrained production simulations. We then repeated the 400-ns production simulations using different initial velocities for both systems.
Simulations were conducted using GRO MACS 4.6 (Hess et al., 2008) with a 2-fs integration time step. A periodic boundary condition was applied in all directions, and the temperature and semi-isotropic pressure with a compressibility of 4.5 × 10 −5 bar −1 were controlled by a velocity-rescale coupling algorithm and Parrinello-Rahman barostat (Bussi et al., 2007). The electrostatic interactions were calculated using the particle mesh Ewald with 0.12-nm Fourier spacing, and LIN CS constraints were applied to all bonds (Hess et al., 1997). All repeat simulations were performed with differences in initial velocity.
Simulations were visualized using VMD, Pymol (Humphrey et al., 1996), or Discovery Visualizer 4.0 (Accelrys). Hydrogen bond analysis was conducted using GRO MACS g-hbond with default cutoff of 3.5 Å and 30° between hydrogen bond donors (OH and NH) and acceptor oxygens. The probability of hydrogen bonds was determined by dividing the total number of hydrogen bonds by the simulation time. The relative PIP 2 association was calculated by taking the sum of hydrogen bonds for 0′K-5′-PO 4 , 5′K-5′-PO 4 , and 6′K-5′-PO 4 and subtracting hydrogen bonds for the non-PIP 2 association (e.g., K200-E203). Cross-distance diameter of F192 gate was calculated by measuring the distance of the center of mass of the F192 (−2′F) side chain between chain A-C and chain B-D and averaging the two. Water analysis was conducted by counting number of water in 15 × 15-Å box along the pore axis of the channel. Probability density of water was obtained by calculating water count at the pore throughout the 400 ns simulation and normalizing to the probability density of water at the bottom of the selectivity filter. Figure S1. Alignment of amino acids in the tether helix. (a) Sequence alignment for all members of the Kir channel family. Note the numbering scheme, where K 194 is the 0′ position. Arrow indicates gate in the HBC, F192 (−2′F). (b) Comparison of hydrogen bonds between basic residues in the tether helix and PIP 2 , based on the GIRK2/PIP 2 (PDB: 4KFM) and Kir2.2/PIP 2 structures (PDB: 3SPI; Hansen et al., 2011;MacKinnon, 2011, 2013). Figure S2. Mean percentage change in basal current with alcohol (EtOH, MPD) and GIRK2*, GIRK2*+MTS-HE, GIRK2*+MTS-F, or GIRK2*+MTS-Y. Note inhibition by MPD for all three modified channels. Bars represent mean ± SEM. Figure S3. Voltage protocol used to activate Dr-Vsp. Baseline current is measured at −120 mV with the test pulse protocol. Dr-Vsp is activated by voltage steps to +100 mV for varying length of time (Δt, milliseconds), and then GIRK current is measured at the end of the voltage step to −120 mV (Dr-Vsp activation protocol). Pulses are delivered at 0.5 Hz. Figure S4. Probability of hydrogen bond formation for 1'-PO 4 , 4′-PO 4 , and K64-9′E. (a) Probability of hydrogen bonds for R92 and 1′-PO 4 of PIP 2 (of 4 maximal bonds). (b) Probability of hydrogen bonds for 4′-PO 4 and R60, K64, 5′K and 6'K in the 6'K-WT and 6'Y simulations. (c) Probability of hydrogen bonds for 9′E and K64 for 6′K-WT and 6′Y simulations. Bars represent mean ± SEM. *, P < 0.05 using Student's t test. Figure S5. Time dependence of hydrogen bonds at all four PIP 2 -binding sites during two 400-ns simulations for GIRK2-6′K (blue) and GIRK2-6′Y (green). Note the scale bar on 1′-PO 4 interaction with R92 interaction, indicating the strong and relatively stable interaction of PIP 2 with the channel at the 1′-PO 4 position. Figure S6. Data for simulation 2 show a trend similar to simulation 1 in Fig. 6. (a) PIP 2 association number, (b) −2′F cross-distance, and (c) water profile of simulation 2. (d) Probability density of water for each simulation shows that WT has a decreased water density at the −2′F gate in both simulations compared with 6′Y mutant simulations. SF, selectivity filter.
Video 1. Video of the MD simulation of WT GIRK2 at the level of the HBC. Note the movement of the F192 in the HBC, as well as in PIP 2 .
Video 2. Video of the 400-ns MD simulation of the PIP 2 pocket in one subunit of the WT channel. All simulations are for 400 ns. The trajectories are 1-ns snapshots that were fit progressively on the c-α atoms using trjconv command in Gromacs. The trajectories were smoothened once using the "smooth" command in Pymol to suppress high-frequency vibrations to visualize the videos.
Video 3. Video of the MD simulation of GIRK2 6′Y at the level of the HBC. Note the movement of the −2′F (F192) in the HBC, as well as in PIP 2 .
Video 4. Video of the 400-ns MD simulation of the PIP 2 pocket in one subunit of the GIRK2 6′Y channel. All simulations are for 400 ns. The trajectories are 1-ns snapshots that were fit progressively on the c-α atoms using trjconv command in Gromacs. The trajectories were smoothened once using the "smooth" command in Pymol to suppress high-frequency vibrations to visualize the videos.