Ion-pulling simulations provide insights into the mechanisms of channel opening of the skeletal muscle ryanodine receptor

Abstract

The type 1 ryanodine receptor (RyR1) mediates Ca²⁺ release from the sarcoplasmic reticulum to initiate skeletal muscle contraction and is associated with muscle diseases, malignant hyperthermia, and central core disease. To better understand RyR1 channel function, we investigated the molecular mechanisms of channel gating and ion permeation. An adequate model of channel gating requires accurate, high-resolution models of both open and closed states of the channel. To this end, we generated an open-channel RyR1 model using molecular simulations to pull Ca²⁺ through the pore constriction site of a closed-channel RyR1 structure determined at 3.8-Å resolution. Importantly, we find that our open-channel model is consistent with the RyR1 and cardiac RyR (RyR2) open-channel structures reported while this paper was in preparation. Both our model and the published structures show similar rotation of the upper portion of the pore-lining S6 helix away from the 4-fold channel axis and twisting of Ile-4937 at the channel constriction site out of the channel pore. These motions result in a minimum open-channel pore radius of ∼3 Å formed by Gln-4933, rather than Ile-4937 in the closed-channel structure. We also present functional support for our model by mutations around the closed- and open-channel constriction sites (Gln-4933 and Ile-4937). Our results indicate that use of ion-pulling simulations produces a RyR1 open-channel model, which can provide insights into the mechanisms of channel opening complementing those from the structural data.

Keywords: calcium channel; calcium transport; conformational change; molecular dynamics; molecular modeling; ryanodine receptor.

Figure 1.

Pore opening progression from the initial closed structure (a, b) to the intermediate structure (c, d) to the final refined open structure (e, f). Top panels (a, c, e) show the side view of the pore, whereas lower panels (b, d, f) show the pore viewed from the cytoplasmic end of the channel. Side chains for amino acids bordering Ca²⁺ are shown as sticks and colored according to residue type, red, acidic; green, polar; and white, aliphatic. In the lower panels (b, d, f), Gln-4933 and Ile-4937 are shown in green and white, respectively. The position of Ca²⁺ is shown as a magenta sphere and waters within 4 Å of Ca²⁺ are shown in red spheres. The initial structure represents the time point in which Ca²⁺ was first moved to the position between Glu-4948 and Glu-4952, before water was allowed to equilibrate around the ion.

Figure 2.

Pore of the RyR1 open-channel model. a, pore profiles for the open-channel model, the open-channel RyR1, and the open-channel RyR2 are shown in blue, black, and gray, respectively. For the open-channel model the pore profile is the means ± S.E. calculated over 10 frames sampled evenly over the last 20 ns of molecular dynamics simulation. For the open-channel RyR1 the error bars are the mean ± S.E. calculated for the three open-channel RyR1 structures (PDB codes 5TA3, 5TAL, and 5T9V). The pore profile is shown aligned with structures for the (b) open-channel model, (c) open-channel RyR1 (PDB code 5TAL), and (d) open-channel RyR2 (PDB code 5GOA). Pore lining residues are shown as sticks and colored according to residue type: red, acidic; blue, basic; green, polar; and white, aliphatic. Residues Gly-4934 and Gly-4941 are colored black.

Figure 3.

The open-channel model of RyR1 agrees with the RyR1 open-channel cryo-EM density. Side (a) and cytosolic (b) views of the RyR1 open-channel model (cyan) aligned with the 3.8-Å resolution open-channel cryo-EM density (yellow surface) (EMD 8378). For comparison, side (c) and cytoplasmic (d) views of the closed structure (PDB code 3J8H; red) (21) are shown aligned with the same open-channel cryo-EM density (EMD 8378).

Figure 4.

Radial tilting and lateral twisting of the S6 helix contributes to channel opening. Cytosolic (a) and side (b) views of the of the closed-channel structure (PDB code 3J8H) (21). The red arrows show the direction of the first principal component between the closed-channel structure and 10 open-channel models sampled every 2 ns over the last 20 ns of simulation. The arrow lengths are scaled according to magnitude of the motion. Likewise, the structure is colored by the magnitude of the motion from smallest (blue) to largest (red). Black arrows in a denote the planes within which radial and lateral helical tilting angles were determined for the helix the arrows are centered on. The black arrows in b highlight the region over which the helical tilting angles were computed.

Figure 5.

Intersubunit contacts for the S6 helix. Contacts are shown for the RyR1 open-channel model (a), the RyR1 open-channel structure (PDB code 5TAL) (b), and the RyR1 closed-channel structure (PDB code 3J8H) (c). The contacting residues are shown in stick representation and colored by residue type: red, acidic; blue, basic; green, polar; and white, aliphatic. The two S6 glycines are colored black.

Figure 6.

Caffeine-induced Ca²⁺ release in HEK293 cells expressing wild-type and mutant RyR1s. Ca²⁺ transients were determined in Ca²⁺-free Krebs-Ringers Henseleit bath solution as changes of Fluo-4 fluorescence (F/F_o) before and following the addition of 8 mm caffeine to the cells (arrow).

Figure 7.

Single channel recordings of WT and mutant channels. Left panels, shown are representative single channel currents at −20 mV (upper and middle traces) or 0 mV (bottom traces) as downward deflections from the closed states (c–) in symmetrical 250 mm KCl with 2 μm Ca²⁺ in the cis chamber (upper traces) and following the subsequent addition of EGTA to yield free Ca²⁺ of 0.1 μm (middle traces) or the addition of 10 mm Ca²⁺ to the trans chamber (bottom traces). Right panels, representative current–voltage relationships in 250 mm symmetrical KCl (●) and after the addition of 10 mm trans Ca²⁺ (○). Averaged P_o values and ion permeation properties are summarized in Table 1.

Figure 8.

The Q4933A and Q4933V mutations increase both the minimum pore radius and Ca²⁺ solvation energy. a, structure of the open-channel pore surrounding position 4933. Pore lining residues are shown as sticks and colored according to residue type: red, acidic; green, polar; and white, aliphatic. Gln-4933, Q4933A, and Q4933V are colored green, black, and gray, respectively. b, RyR1 open-channel pore profiles for wild-type (green) and Q4933A (black), and Q4933V (gray) models. The pore profiles are the mean ± S.E. calculated over five structures sampled evenly over the last 20 ns of molecular dynamics simulation. c, mean change in Ca²⁺ solvation energies between mutant and wild-type RyR1 for Q4933A (black) and Q4933V (gray) mutations calculated over the same five structures from b. Error bars represent the 95% confidence interval for the difference between the means.