Solvation thermodynamics of amino acid side chains on a short peptide backbone

Abstract

The hydration process of side chain analogue molecules differs from that of the actual amino acid side chains in peptides and proteins owing to the effects of the peptide backbone on the aqueous solvent environment. A recent molecular simulation study has provided evidence that all nonpolar side chains, attached to a short peptide backbone, are considerably less hydrophobic than the free side chain analogue molecules. In contrast to this, the hydrophilicity of the polar side chains is hardly affected by the backbone. To analyze the origin of these observations, we here present a molecular simulation study on temperature dependent solvation free energies of nonpolar and polar side chains attached to a short peptide backbone. The estimated solvation entropies and enthalpies of the various amino acid side chains are compared with existing side chain analogue data. The solvation entropies and enthalpies of the polar side chains are negative, but in absolute magnitude smaller compared with the corresponding analogue data. The observed differences are large; however, owing to a nearly perfect enthalpy-entropy compensation, the solvation free energies of polar side chains remain largely unaffected by the peptide backbone. We find that a similar compensation does not apply to the nonpolar side chains; while the backbone greatly reduces the unfavorable solvation entropies, the solvation enthalpies are either more favorable or only marginally affected. This results in a very small unfavorable free energy cost, or even free energy gain, of solvating the nonpolar side chains in strong contrast to solvation of small hydrophobic or nonpolar molecules in bulk water. The solvation free energies of nonpolar side chains have been furthermore decomposed into a repulsive cavity formation contribution and an attractive dispersion free energy contribution. We find that cavity formation next to the peptide backbone is entropically favored over formation of similar sized nonpolar side chain cavities in bulk water, in agreement with earlier work in the literature on analysis of cavity fluctuations at nonpolar molecular surfaces. The cavity and dispersion interaction contributions correlate quite well with the solvent accessible surface area of the nonpolar side chains attached to the backbone. This correlation however is weak for the overall solvation free energies owing to the fact that the cavity and dispersion free energy contributions are almost exactly cancelling each other.

The Multiple Origins of the Hydrophobicity of Fluorinated Apolar Amino Acids

“…A weak correlation between SASA and hydration free energy of amino acids has also been observed by others, 50 and does not reflect poor model quality. Evidence that there is no fundamental flaw in the models is presented in Figure 4B, which demonstrates the excellent correlation between the experimental retention times of Fmoc-protected amino acids on a functionalized silica HPLC column 23,43,48 (and present work) and the SASA calculated from our models.…”

Enthalpic and Entropic Contributions to Hydrophobicity

“… Free energy, enthalpy and entropy values for all uncharged amino acids obtained with the GIST approach (y-values) with respect to the values obtained by Hajari 15 (x-values) with thermodynamic integration. The values of the linear regression for every water model are given in the top left corner.…”

“…Therefore, only one simulation is necessary to get the solvation entropy (for a rigid solute). For TI on the other hand, entropic changes are typically derived indirectly from the temperature dependence of the free energy (using multiple simulations at different temperatures) as done by Hajari et al 15 or via the difference of the free energy and the enthalpy. While Hajari et al used 26 or more λ-values, 6 different temperatures, simulating every window for 5 ns, resulting in a total simulation time of ≥600 ns, we only needed 200 ns of simulation time (potentially even less without significant loss of accuracy).…”

“…Therefore, hydrophobic properties of side chain analogues may differ from those of amino acids, including the backbone atoms. 15 , 42 , 43 …”

Hydrophobicity Revisited: a Molecular Story

“…The free energy of hydrating of polar secondary group ASA remains largely unchanged from free amino-acids to amino-acids in short peptides stemming from peptide bond water interactions leading to a entropy-enthalpy compensating balance. ( [47]) The molecular effects of the peptide bond and secondary group ASA interactions with adjacent aqueous clathrate shells The peptide bond is capable of strong dipole interactions, hydrogen bonding and also strong Van Der Waals dispersion force attractions (from a 40% PI bond QM superposition of states) depending upon the context of the molecular ensemble environment the peptide bond is found within. Polar ASA of amino-acid secondary groups strongly attracts (dipole, hydrogen bonding) water which significantly mitigates the strength of aqueous clathrate shell surface tension at these interfaces.…”