Free-energy cost for translocon-assisted insertion of membrane proteins

Abstract

Nascent membrane proteins typically insert in a sequential fashion into the membrane via a protein-conducting channel, the Sec translocon. How this process occurs is still unclear, although a thermodynamic partitioning between the channel and the membrane environment has been proposed. Experiment- and simulation-based scales for the insertion free energy of various amino acids are, however, at variance, the former appearing to lie in a narrower range than the latter. Membrane insertion of arginine, for instance, requires 14-17 kcal/mol according to molecular dynamics simulations, but only 2-3 kcal/mol according to experiment. We suggest that this disagreement is resolved by assuming a two-stage insertion process wherein the first step, the insertion into the translocon, is energized by protein synthesis and, therefore, has an effectively zero free-energy cost; the second step, the insertion into the membrane, invokes the translocon as an intermediary between the fully hydrated and the fully inserted locations. Using free-energy perturbation calculations, the effective transfer free energies from the translocon to the membrane have been determined for both arginine and leucine amino acids carried by a background polyleucine helix. Indeed, the insertion penalty for arginine as well as the insertion gain for leucine from the translocon to the membrane is found to be significantly reduced compared to direct insertion from water, resulting in the same compression as observed in the experiment-based scale.

Fig. 1.

Side view of the simulated systems. In all panels, the polyleucine helix is shown in cyan; red spheres on the helix indicate the position of hybrid leucine/arginine residues used for FEP. The extent of water in the periodic system is illustrated as a transparent gray surface. Lipids are shown as teal lines with phosphorus and nitrogen atoms of their headgroups indicated as orange and blue spheres, respectively. (A) Pure-membrane/polyleucine system. (B) SecYEβ/polyleucine system. SecYEβ is shown in gray (SecY), orange (E), and yellow (β). The lateral gate helices 2b and 7 are shown in green. (C) Close-up of the pore region of SecY from B. SecYEβ is displayed as a molecular surface, colored as in B, indicating the exposure of the polyleucine helix to the membrane.

Fig. 2.

Thermodynamic cycle for TM insertion. The upper cycle represents the transfer of a polyleucine helix with a solvent-exposed arginine residue from SecY to membrane. The lower cycle illustrates the same transfer but with the arginine residue located at the center of SecY and the membrane. The red box encapsulates the second stage of the proposed insertion process. Free-energy differences ΔG(poly-L_aq.→mem.), ΔG(poly-L_aq.→SecY), ΔG(Arg_aq.→mem.), and ΔG(Arg_aq.→SecY) were determined from FEP simulations, with the transfers from the aqueous state derived based on transfers from a vacuum state (see SI Appendix, Table S1 and Fig. S7). ΔG(poly-L_SecY→mem.) and ΔG(poly-L∶Arg_SecY→mem.) were calculated by completing the cycles.

Fig. 3.

Deformation of the membrane by snorkeling of the arginine residue toward the membrane-water interface. The lower leaflet of the membrane, water, and polyleucine helix are shown as in Fig. 1A. The arginine residue is shown in licorice representation with the coordinating lipid phosphorus atom shown as a yellow sphere. Water that has penetrated the bilayer to solvate arginine is shown in red and white. (A) Membrane deformation for a pure membrane system. (B) Deformation of the membrane by arginine in the SecY-membrane system. The view is perpendicular to SecY’s lateral gate.