What makes a good graphene-binding peptide? Adsorption of amino acids and peptides at aqueous graphene interfaces

Abstract

Investigation of the non-covalent interaction of biomolecules with aqueous graphene interfaces is a rapidly expanding area. However, reliable exploitation of these interfaces in many applications requires that the links between the sequence and binding of the adsorbed peptide structures be clearly established. Molecular dynamics (MD) simulations can play a key role in elucidating the conformational ensemble of peptides adsorbed at graphene interfaces, helping to elucidate these rules in partnership with experimental characterisation. We apply our recently-developed polarisable force-field for biomolecule-graphene interfaces, GRAPPA, in partnership with advanced simulation approaches, to probe the adsorption behaviour of peptides at aqueous graphene. First we determine the free energy of adsorption of all twenty naturally occurring amino acids (AAs) via metadynamics simulations, providing a benchmark for interpreting peptide-graphene adsorption studies. From these free energies, we find that strong-binding amino acids have flat and/or compact side chain groups, and we relate this behaviour to the interfacial solvent structuring. Second, we apply replica exchange with solute tempering simulations to efficiently and widely sample the conformational ensemble of two experimentally-characterised peptide sequences, P1 and its alanine mutant P1A3, in solution and adsorbed on graphene. For P1 we find a significant minority of the conformational ensemble possesses a helical structure, both in solution and when adsorbed, while P1A3 features mostly extended, random-coil conformations. In solution this helical P1 configuration is stabilised through favourable intra-peptide interactions, while the adsorbed structure is stabilised via interaction of four strongly-binding residues, identified from our metadynamics simulations, with the aqueous graphene interface. Our findings rationalise the performance of the P1 sequence as a known graphene binder.

Predictions of Pattern Formation in Amino Acid Adlayers at the In Vacuo Graphene Interface: Influence of Termination State

“…The water molecules were described with the CHARMM‐modified version of the TIP3P water model . This combination of force‐fields have been previously implemented by Hughes et al to study aqueous graphene/amino acid interfaces . In brief, after consideration and testing of a range of simulation approaches, SAMD was used to model these adlayers.…”

“…Alternatively, molecular dynamics (MD) simulations have the potential to address some of these questions, and have demonstrated a complementary route to elucidating the atomic‐scale details required to enable clear interpretation of experimental data relating to surface organization of biomolecules on graphene/graphite . To this end, numerous theoretical studies have reported the adsorption modes and configurations of amino acids on graphene/graphite surfaces using MD simulation …”

“…[9,[41][42][43] To this end, numerous theoretical studies have reported the adsorption modes and configurations of amino acids on graphene/graphite surfaces using MD simulation. [8,[43][44][45][46][47][48][49][50] For example, Hughes et al reported that tryptophan adsorbed to the aqueous graphene surface via the aromatic ring of the side-chain and also aligned flat on graphene. [8,45] However, most of the previous simulation studies focused on the adsorption of an isolated amino acid, and many considered adsorption in aqueous solution, not UHV conditions.…”

Understanding Noncovalent Interactions of Small Molecules with Carbon Nanotubes

“…[27][28][29] Amongt hem, Grimme's dispersion-correction (so called D3) is the most populara bi nitio approacht od ealing with the weak noncovalenti nteractions for whicht he "gold-standard" coupled-cluster CCSD(T) becomes prohibitive. Free-energy techniques have been used to calculate the adsorption affinities of amino acids on model gold, [32][33][34][35] silver, [36] zinc oxide surfaces, [37] and graphenic materials, [38,39] as well as for contaminants on amorphous silica. Alternatively,c lassical molecular dynamics simulations with atomistic models and explicit solvent have arisen as ap romising approach to elucidating interactionsb etween commonn anomaterials and organic molecules.…”

Adsorption and Conformational Evolution of Alpha-Helical BSA Segments on Graphene: A Molecular Dynamics Study

“…In this case, there was an abrupt increase in the interaction energy of the arginine residue at about 10 ns, almost doubling its value and exceeding the energy of −80 kJ/mol exhibited by the tryptophan residue. This event was likely to be caused by the rapid structural change in an arginine sidechain as it adsorbed onto graphene within a short period of time, as arginine residues were previously shown to interact strongly with graphene [Hughes et al, 2015]. The conformational difference is highlighted in the insets of Figs.…”