Molecular Dynamics Simulation of a Spin-Labeled Membrane Protein

Abstract

We have developed a computational molecular dynamics technique to simulate the motion of the TOAC spin label bound to the cytosolic and transmembrane domains of monomeric phospholamban (PLB) in a lipid bilayer. In cardiac muscle, the integral membrane protein PLB binds to and inhibits the Sarco/Endoplasmic Reticulum Ca2+ ATPase (SERCA), decreasing calcium sequestration into the SR lumen. Previous electron paramagnetic resonance (EPR) experiments showed strong immobilization of the spin probe when placed near the C-terminus of PLB, indicating a stable and highly ordered transmembrane helix. It was shown that the cytosolic domain samples two distinct conformations, an ordered T state with moderately restricted motion and a dynamically disordered R state with nearly unrestricted isotropic motion. Phospholamban regulatory function requires an order-to-disorder transition in the cytosolic domain, with phosphorylation at Ser16 shifting the equilibrium toward the mobile R state and relieving SERCA inhibition. In the present study, monomeric PLB was simulated in a POPC lipid bilayer with the newly parameterized TOAC spin label replacing amino acids at positions 11 and 36. Unlike cysteine-reactive spin labels, TOAC directly reports the motion of the peptide backbone, making it possible to computationally analyze the coupled motion of the transmembrane helix. The simulation for each mutant is performed with and without phosphorylation of Ser16. We optimized molecular dynamics simulation conditions and the resulting trajectories are used to calculate order parameters, orientational distributions, rotational correlation times, and helical tilt angles for comparison with previous experimental results. The dynamics of the spin-labeled protein are compared with the wild type monomer to investigate the effect of the spin probe on the peptide backbone. This computational method informs our current experimental model of TOAC-protein dynamics and strengthens our molecular dynamics modeling technique for performing future spin probe experiments in silico.