Evaluating Force Fields for the Computational Prediction of Ionized Arginine and Lysine Side-Chains Partitioning into Lipid Bilayers and Octanol

Abstract

Abundant peptides and proteins containing arginine (Arg) and lysine (Lys) amino acids can apparently permeate cell membranes with ease. However, the mechanisms by which these peptides and proteins succeed in traversing the free energy barrier imposed by cell membranes remain largely unestablished. Precise thermodynamic studies (both theoretical and experimental) on the interactions of Arg and Lys residues with model lipid bilayers can provide valuable clues to the efficacy of these cationic peptides and proteins. We have carried out molecular dynamics simulations to calculate the interactions of ionized Arg and Lys side-chains with the zwitterionic 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) lipid bilayer for 10 widely used lipid/protein force fields: CHARMM36/CHARMM36, SLIPID/AMBER99SB-ILDN, OPLS-AA/OPLS-AA, Berger/OPLS-AA, Berger/GROMOS87, Berger/GROMOS53A6, GROMOS53A6/GROMOS53A6, nonpolarizable MARTINI, polarizable MARTINI, and BMW MARTINI. We performed umbrella sampling simulations to obtain the potential of mean force for Arg and Lys side-chains partitioning from water to the bilayer interior. We found significant differences between the force fields, both for the interactions between side-chains and bilayer surface, as well as the free energy cost for placing the side-chain at the center of the bilayer. These simulation results were compared with the Wimley-White interfacial scale. We also calculated the free energy cost for transferring ionized Arg and Lys side-chains from water to both dry and wet octanol. Our simulations reveal rapid diffusion of water molecules into octanol whereby the equilibrium mole fraction of water in the wet octanol phase was ∼25%. Surprisingly, our free energy calculations found that the high water content in wet octanol lowered the water-to-octanol partitioning free energies for cationic residues by only 0.6 to 0.7 kcal/mol.

Atomistic Molecular Simulations Suggest a Kinetic Model for Membrane Translocation by Arginine-Rich Peptides

“…16 We did find that generally consistent results were obtained with the above force fields with respect to the interaction free energies of the side-chains with a zwitterionic lipid bilayer. Furthermore, these calculated interactions were consistent with the Wimley-White interfacial scales.…”

Maximally asymmetric transbilayer distribution of anionic lipids alters the structure and interaction with lipids of an amyloidogenic protein dimer bound to the membrane surface

“…Finally, the results of any MD simulations depend on the force fields (Poger and Mark, 2012; Kukol, 2015; Sun et al, 2015). Our primary incentive of choosing GROMOS87/Berger lipid force field was to compare our results with the published studies using the identical forces fields.…”

Multistep Molecular Dynamics Simulations Identify the Highly Cooperative Activity of Melittin in Recognizing and Stabilizing Membrane Pores

“…Our simulated pore structure is very different to the "disordered toroidal pore" reported by Sengupta et al, 15 wherein the melittin peptide adopts randomly tilted orientations. We recall that Sengupta et al employed the Berger lipid and GROMOS protein force fields, which predicts quite strong interactions between the lysine amino acid and zwitterionic lipid heads 38 . This is somewhat inconsistent with experimental interfacial hydrophobicity scales.…”

“…By contrast, the all-atom CHARMM force field predicts a weak interaction between lysine and the DPPC lipid 20 heads. 38 Hence, in these simulations, melittin will likely adopt a perpendicular orientation in the pore to maximise hydrophobic contacts within the membrane of the pore surface (as described above). …”