Implicit membrane treatment of buried charged groups: Application to peptide translocation across lipid bilayers

Abstract

The energetic cost of burying charged groups in the hydrophobic core of lipid bilayers has been controversial, with simulations giving higher estimates than certain experiments. Implicit membrane approaches are usually deemed too simplistic for this problem. Here we challenge this view. The free energy of transfer of amino acid side chains from water to the membrane center predicted by IMM1 is reasonably close to all-atom free energy calculations. The shape of the free energy profile, however, for the charged side chains needs to be modified to reflect the all-atom simulation findings (IMM1-LF). Membrane thinning is treated by combining simulations at different membrane widths with an estimate of membrane deformation free energy from elasticity theory. This approach is first tested on the voltage sensor and the isolated S4 helix of potassium channels. The voltage sensor is stably inserted in a transmembrane orientation for both the original and the modified model. The transmembrane orientation of the isolated S4 helix is unstable in the original model, but a stable local minimum in IMM1-LF, slightly higher in energy than the interfacial orientation. Peptide translocation is addressed by mapping the effective energy of the peptide as a function of vertical position and tilt angle, which allows identification of minimum energy pathways and transition states. The barriers computed for the S4 helix and other experimentally studied peptides are low enough for an observable rate. Thus, computational results and experimental studies on the membrane burial of peptide charged groups appear to be consistent.

Comparative analysis reveals amino acids critical for anticancer activity of peptide CIGB-552

“…Membrane incorporation proceeds by first inserting the N-terminal, followed by partially solvated basic amino acids ( Figure 7A), while C-terminal carboxylates in metabolites 1a and 3 remain exposed to the solvent. This agrees with previously described mechanisms for basic peptide penetration [30], and is putatively caused by a higher energetic cost to dehydrate C-terminal moieties [31]. In contrast, CIGB-552 and metabolite 5 maintain the looped conformation in the membrane environment.…”

“…However, we observed that the presence of aromatic residues only present at the C‐termini of CIGB‐552 and metabolite 5 translates in the formation of a hydrophobic cluster that ultimately results in a looped structure stabilized by the formation of a stiff salt bridge between an arginine and the COO − moiety. Although the maximum affordable time scale of the simulations differs significantly from experimental times, this structural organization could help internalization by reducing the free energy needed to translocate the C‐terminal carboxylate . Moreover, the looped conformation conserved in the two longest peptides offers a putative explanation for the similar characteristics displayed in contrast with metabolites 1 and 3a in terms of enhanced membrane interaction, internalization, and cytotoxic activity.…”

Membrane Permeation of a Peptide: It Is Better to be Positive

“…Amino acid and peptide translocation are useful in modeling the mechanism of peptide and of protein insertion to membranes 5 . Arginine was studied most extensively due to its significant role in Cell Permeating Peptides (CPP).…”

The Treatment of Solvent in Multiscale Biophysical Modeling

“…The solvation parameters of all atoms now depend on the position along the membrane normal and are modeled as a linear combination of the values corresponding to water and cyclohexane. [190] IMM1 has been applied to the modeling of fusion peptide assemblies, [191] computation of potentials of mean force for peptides in membranes, [192] presenilin, [193] rhodopsin, [194] the transmembrane dimer of amyloid precursor protein, [195] transmembrane structures of amyloid oligomers, [196] dimerization of the lutropin receptor, [197] gating of MscL, [198] the translocon, [199] peptide translocation through bilayers, [200] and others. In addition, the distance-dependent dielectric is modified to account for strengthening of the electrostatic interactions in the membrane.…”

“…[188] The surface, transmembrane and dipole potentials are accounted for by a term of the form P i yðz i Þ q i , in which q i is the partial charge of atom i and y is the potential, obtained by GouyÀChapman theory, [183] explicit simulations, [187] analytical solutions to the PB equation, [189] or numerical solutions to the PB equation. [190] IMM1 has been applied to the modeling of fusion peptide assemblies, [191] computation of potentials of mean force for peptides in membranes, [192] presenilin, [193] rhodopsin, [194] the transmembrane dimer of amyloid precursor protein, [195] transmembrane structures of amyloid oligomers, [196] dimerization of the lutropin receptor, [197] gating of MscL, [198] the translocon, [199] peptide translocation through bilayers, [200] and others.…”