Charge interactions along the ER/K α-helix backbone. (A) Fraction helix as a function of temperature from experiment (CD melt) and computation (REMD). Blue, REMD data starting with α-helical conformation; red, REMD data starting with random initial conformation; black, CD-melt data from Lyu et al. (16); green, CD melt data for the MT ER/K α-helix in myosin VI. CD data were converted to fraction helix as per Rohl and Baldwin (28). (B) With reference to an (E₄K₄)₂ helix (E1 to K16), sample variation of distance (Å) between N_ζ atom of K7 (i) and center of mass of O_ε atoms of E4 (i − 3) (Right) and E11 (i + 4) (Left) with time (ns) during MD simulations (Left). (C) (Left) Snapshot of two orthogonal views of a section of the ER/K α-helix (residues 24–55) in mannosyltransferase from the MD simulation, showing side-chain interactions and residue numbers. For reference, the backbone carbonyl oxygens corresponding to the illustrated side chains are also shown. (Right) The charge interaction map (see SI Methods) shown to the right illustrates the pattern of E↔R/K interactions along the vertical axis (see yellow boxed groups of residues in columns). Arrows show boundaries of the segment of the α-helix displayed in Left. (D) Snapshot of two orthogonal views of a section of the ER/K helix (residues 12–51) in upf2 regulator of nonsense transcripts homolog (Right), with its charge interaction map (Left).

Side-chain interaction distances in the (E₄K₄)₂ α-helix. (A) Histogram of minimum distance, at every time step of the MD simulation, between K7-N_ζ and the centers of mass of O_ε in neighboring E residues. For the entire simulation time, K7-N_ζ is closest to one of E3, E4, E10, and E11. (B) Histogram of minimum distances, at every time step of the MD simulation, between N_ζ of each K residue and centers of mass of O_ε in all neighboring E residues and vice versa. The red line shows a maximum likelihood fit to the sum of three Gaussians, which yields the relative time spent under the 3, 5, and 8-Å peaks along with their standard deviations. (C) Snapshot of the (E₄K₄)₂ α-helix showing sample side-chain conformations during 3-Å (Left) and 5-Å (Right) separations between N_ζand O_ε.

Lateral bending and stability of ER/K α-helix. (A) Water density as a function of distance from the α-helix axis for the (E₄K₄)₂ α-helix. The vertical dotted line at 2.2 Å shows the average location of C_α atoms, whereas the vertical dotted line at 6 Å shows the average location of N_ζ/O_ε atoms of the K/E side chains. Note the exclusion of water from the α-helical backbone resulting from the presence of side chains. (B) Exaggerated schematic representation showing bending of the ER/K α-helix in mannosyltransferase. The α-helix backbone is shown in ribbon representation (yellow), along with selected E and K side chains. To illustrate the effect of bending on side-chain interaction distances, the bent α-helix is shown as a transparent structure. The α-helix bending approximately the neutral axis results in an increase in E↔K distances on the convex side (top face), with a corresponding decrease in E↔K distances on the concave side (bottom face).

Experimental verification of stability and bending rigidity of the MT ER/K α-helix in myosin VI. (A) Schematic of the lateral bending of the MT ER/K α-helix during the 180° swing of the myosin VI lever arm. (Upper) The poststroke state is indicated with the prestroke state shown as a transparent structure (side view along the actin filament). (Lower) The poststroke state is indicated with the prestroke state and lever arm in transition shown as transparent structures (top view along the actin filament). Distances shown in the schematic representation are derived from the known myosin VI crystal structure with modeled IQ-CAM (22) and predicted structures of the PT and MT. Although the angle between the PT and MT is unknown, the orientation shown in the schematic representation is consistent with the SAXS structures of PT-MT and PT-MT-DT published previously (8), and aligns the MT parallel to the F-actin filament to allow dimerization of the cargo binding domains, resulting in ≈36-nm processive stepping of myosin VI. (B) Two orthogonal views of the predicted structure of a fragment of myosin VI including the IQ motif and PT with bound calmodulin docked into its SAXS envelope. The PT is predicted to contribute only ≈1.5 nm to the lever arm of myosin VI (Upper). Orthogonal views of the predicted structure of the MT ER/K α-helix in myosin VI docked into its SAXS envelope. The end–end distance of the SAXS envelope (10 nm) is consistent with the ER/K motif forming an isolated, stable α-helix in solution. N and C termini are marked (Lower). (C) Histogram of bead displacement in response to binding of myosin VI to an actin filament, followed by the myosin VI power stroke. Mean displacement is a measure of the stroke size of the myosin VI lever arm without (Upper) and with (Lower) the MT ER/K α-helix.