Molecular dynamics simulations of the Cx26 hemichannel: evaluation of structural models with Brownian dynamics

Abstract

The recently published crystal structure of the Cx26 gap junction channel provides a unique opportunity for elucidation of the structure of the conductive connexin pore and the molecular determinants of its ion permeation properties (conductance, current-voltage [I-V] relations, and charge selectivity). However, the crystal structure was incomplete, most notably lacking the coordinates of the N-terminal methionine residue, which resides within the pore, and also lacking two cytosolic domains. To allow computational studies for comparison with the known channel properties, we completed the structure. Grand canonical Monte Carlo Brownian dynamics (GCMC/BD) simulations of the completed and the published Cx26 hemichannel crystal structure indicate that the pore is too narrow to permit significant ion flux. The GCMC/BD simulations predict marked inward current rectification and almost perfect anion selectivity, both inconsistent with known channel properties. The completed structure was refined by all-atom molecular dynamics (MD) simulations (220 ns total) in an explicit solvent and POPC membrane system. These MD simulations produced an equilibrated structure with a larger minimal pore diameter, which decreased the height of the permeation barrier formed by the N terminus. GCMC/BD simulations of the MD-equilibrated structure yielded more appropriate single-channel conductance and less anion/cation selectivity. However, the simulations much more closely matched experimentally determined I-V relations when the charge effects of specific co- and posttranslational modifications of Cx26 previously identified by mass spectrometry were incorporated. We conclude that the average equilibrated structure obtained after MD simulations more closely represents the open Cx26 hemichannel structure than does the crystal structure, and that co- and posttranslational modifications of Cx26 hemichannels are likely to play an important physiological role by defining the conductance and ion selectivity of Cx26 channels. Furthermore, the simulations and data suggest that experimentally observed heterogeneity in Cx26 I-V relations can be accounted for by variation in co- and posttranslational modifications.

Figure 1.

Structural models and pore dimensions of Cx26 hemichannels. (A) Top view of the Cx26 hemichannel corresponding to the crystal structure (Protein Data Bank accession no. 2ZW3; Maeda et al., 2009). The six protein subunits, depicted in different colors, surround a central aqueous pore. (B) Side view of the structure shown in A, illustrating the TM architecture and channel pore relative to two opposite subunits. The extracellular entrance of the channel (z = 45 Å) is at the top of the structure. (C) Pore radius of the Cx26 “crystal” hemichannel determined by HOLE plotted as a function of the z coordinate. (D) Top view of the “completed” crystal structure including the atomic coordinates of the N-terminal methionine residue (Met1), the side chains of K15, S17, and S19, as well as residues comprising the CL and the C terminus. (E) Side view of the completed crystal structure, showing the same subunits as in B. (F) Pore radius of the completed crystal structure. (G) Top view of the channel conformation that has the smallest RMSD from the average equilibrated structure, average pore dimension, and pore-lining probability. We define this structure as the “average equilibrated structure.” (H) Side view of the average equilibrated structure. (I) Pore dimensions of the average equilibrated structure. The mean pore radius is depicted by the black line. The red lines represent the dynamic variation by plotting the mean ± the standard deviation determined from the four 20-ns production-phase simulations.

Figure 2.

Skeleton drawings of selected portions of the crystal and average of all equilibrated structures to show gross changes resulting from structural equilibration by MD simulations. Superimposition of the skeleton of the crystal structure (blue) and the average of all equilibrated structures (red).

Figure 3.

Heterogeneity of I-V relations observed for human Cx26 undocked hemichannels in Xenopus oocytes obtained with outside-out patch-clamp recordings in 100 mM of symmetric KCl. Five I-V relations ranging from slight inward to slight outward are illustrated. (A) Inward/sigmoidal I-V relation obtained with a ±150-mV voltage ramp. 2 of 10 sequential current traces from one record are superimposed. Current traces from −110 to 100 mV show the open-channel I-V relation. (B) Slightly inward I-V relation obtained with a ±70-mV voltage ramp. 3 of 10 sequential traces are superimposed. (C) Linear I-V relation obtained with a ±150-mV voltage ramp. 2 of 10 sequential current traces are superimposed. Current traces are shown from −125 to 100 mV. (D) Slightly outward I-V relation obtained with a ±70-mV voltage ramp. 2 of 10 sequential current traces are superimposed. (E) Outward I-V relation obtained with a ±70-mV voltage ramp. 2 of 10 sequential current traces are superimposed. In all cases, the single-channel records display both forms of voltage-dependent gating reported for Cx26 undocked hemichannels (González et al., 2006; Sánchez et al., 2010). Loop- or slow-gating transitions that can result in full channel closure were observed at large inside negative potentials, and V_j- or fast-gating transitions that result in closure to subconductance states were observed at inside positive potentials. These events correspond to the upward current deflections from the fully open state at negative potentials and downward deflections from the fully open state at positive potentials, respectively, in the records shown. Full channel closures at negative potentials were used to subtract leak conductance in single-channel records.

Figure 4.

(A) Schematic representation of the GCMC/BD system for Cx26 hemichannels. A Cx26 hemichannel (yellow) inserted into the explicit POPC membrane system was integrated into the implicit membrane system of the GCMC/BD program to ensure a tight seal between the irregular conical surface of the Cx26 channel and the implicit membrane. The boundaries of the implicit and explicit membranes are indicated by the dotted green lines. The simulation box (66 × 66 × 120 Å) was separated by the channel/membrane into two compartments, each containing 100 mM KCl. The buffer region maintains the chemical potential in the two compartments by creating and destroying ions with a GCMC algorithm. The upper compartment (extracellular part of the channel) was defined as the ground in voltage applications. 20 replicate 450-ns simulations were performed at each of seven voltages, ±150, ±100, ±50, and 0 mV, to plot the I-V relations. Blue circles, K⁺; red circles, Cl⁻. (B) Positions of modified residues identified by Locke et al. (2009) that would alter the distribution of charge in the Cx26 channel pore, shown in a side view of two opposite subunits of the completed crystal structure. The positions of acetylated residues are colored as follows: blue, Met1; red, K15; green, K102, K103, K105, K108, K112, and K116 in CL/TM2; orange, γ-carboxyglutamated residues E42, E47, and E114.

Figure 5.

GCMC/BD simulations of the average equilibrated structure. (A) The I-V relation obtained computationally for the multi-ion case (100 mM of symmetric KCl). The black line is the total current, the red line is the Cl⁻ current, the blue line is the K⁺ current, and the green line is the total current of the completed crystal structure corresponding to Fig. S5 A. (B) The pore radius of the average equilibrated structure is shown in red, and the completed crystal structure is in blue. The positions of Cα of Met1, the amide group of K41, the carbonyl of E42, and the Cα of A49 are indicated. (C) The PMF of K⁺ at three voltages, 150 mV (blue), 0 mV (green), and −150 mV (red), are plotted against the z coordinate. The blue, green, and red lines are the PMF determined at 150, 0, and −150 mV, respectively. (D) The PMF of Cl⁻ determined from simulations at the same three voltages.

Figure 6.

Superimposition of experimental I-V relations (gray traces) and those computed by GCMC/BD simulations with the charge effects of the indicated protein modifications. (A) The correspondence between slight outward rectification (Fig. 3 E) and simulated channel with neutralization of Met1 and six internal lysine residues in TM2/CL. The charge neutralizations mimic acetylation. The black line is the total current, the blue line is the K⁺ current, the red line is the Cl⁻ current, and the gray line is the single-channel current trace. (B) The pore radius of the simulated channel structure plotted against the z coordinate. The positions of the modified atoms are shown. The six acetylated lysine residues at the tip of TM2/CL are grouped into one position and labeled “CL-Acet 6x.” (C) The PMF of K⁺ at three voltages, 150 mV (blue), 0 mV (green), and −150 mV (red), plotted against the z coordinate. (D) The PMF of Cl⁻ determined at the same three voltages. (E–H) Outwardly rectifying experimental I-V relation corresponds to simulated I-V relations in which Met1, K15, and six lysine residues in TM2/CL are neutral-mimicking acetylation. (I–L) Inward sigmoidal experimental I-V relation corresponds to simulated I-V relations in which the six lysine residues in TM2/CL are neutralized (acetylated) and the three glutamate residues are modified by γ-carboxyglutamation (Gla). (M–P) Slightly inward rectifying experimental I-V relations correspond to simulated I-V relations in which Met1, K15, and K116 are neutralized (acetylated) and the three glutamate residues are modified by γ-carboxyglutamation (Gla).