Do Valine Side Chains Have an Influence on the Folding Behavior ofβ-Substitutedβ-Peptides?

Synthesis and characterization of β-peptide helices as transmembrane domains in lipid model membranes

Aggregation, orientation and dynamics of transmembrane helices are of relevance for protein function and transmembrane signaling. To explore the interactions of transmembrane helices and the interdependence of peptide structure and lipid composition of the membranes, β-peptides were explored as model transmembrane domains. Various hydrophobic β-peptide sequences were synthesized by solid phase peptide synthesis. Conformational analyses of β-peptide helices were performed in organic solvents (methanol and 2,2,2-trifluoroethanol) and in large unilamellar liposomes (dimyristoylphosphatidylcholine, dipalmitoylphosphatidylcholine and dioleoylphosphatidylcholine) indicating 12- and 14-helix conformations, depending on β³ -amino acid sequences. The intrinsic tryptophan fluorescence of β³ -homotryptophan units inserted in the center or near the end of the sequence was used to verify the membrane insertion of the β-peptides. A characteristic blue shift with peripheral β³ -homotryptophan compared with β-peptides with central tryptophan served as indication for a transmembrane orientation of the β-peptides within the lipid bilayer. Copyright © 2016 European Peptide Society and John Wiley & Sons, Ltd.

Synthesis of a new β-amino acid with a 3-deoxy-L-ara furnaoside side chain: the influence of the side chain on the conformation of α/β-peptides

The important role of side chains in the stabilization of helical folds in peptidic foldamers containing C-linked carbo-β-amino acids (β-Caa), an interesting class of β-amino acids, with carbohydrate side chains has been extensively elaborated. As a pragmatic approach to alleviate the interference of substituents in the side chains on the folding propensities of the peptides, they are often modified or removed. The present study reports the synthesis of a new β-Caa with a 3-deoxy-L-ara furanoside side chain, [(R)-β-Caa(da)], from D-glucose, and its use in the synthesis of α/β-peptides in 1 : 1 alternation with D-Ala. The synthesis of peptides using (R)-β-Caa(da), was facile unlike those from (R)-β-Caa(a) having the L-ara furanoside side chain. The detailed NMR, molecular dynamics (MD) and CD studies on the new α/β-peptides showed the presence of robust left-handed 11/9-mixed helices. The study demonstrates that the new (R)-β-Caa(da), behaves differently compared to the other two related monomers, (R)-β-Caa(x) with the D-xylo furanoside side chain and (R)-β-Caa(a).

Establishing Effective Simulation Protocols for β- and α/β-Mixed Peptides. I. QM and QM/MM Models

A quantum mechanical (QM) model for non-natural β- and α/β-mixed peptides is investigated using an approximate density functional method (called SCC-DFTB). In the gas phase the predictions of the model for cyclic and acyclic dipeptides and several acyclic heptapeptides are compared to ab initio B3LYP and LMP2 calculations. The SCC-DFTB reproduces the global minimum of the configurations with the root-mean-square (rms) error in the key dihedral angles of less than 14 degrees. The relative energies of different conformers are also well described in general, with the typical rms error of 2-3 kcal/mol relative to LMP2 energies at either B3LYP or LMP2 optimized structures. The dipole moments are reproduced with a systematic underestimate of less than 15%. The QM model is also used with a molecular mechanical (MM) model of the solvent. For a tetrameric α/β-peptide in water, the SCC-DFTB/MM energies are well correlated with B3LYP/6-31+G**/MM single point energies for a wide range of structures sampled in 2 ns of SCC-DFTB/MM molecular dynamics. For an octameric α/β-peptide in methanol the predicted structures are in qualitative agreement with experimental NOE data. These results suggest that the SCC-DFTB model provides a fairly accurate representation of the structure and thermodynamics of these peptides.

Helix stability of oligoglycine, oligoalanine, and oligo-β-alanine dodecamers reflected by hydrogen-bond persistence

Helices are important structural/recognition elements in proteins and peptides. Stability and conformational differences between helices composed of α- and β-amino acids as scaffolds for mimicry of helix recognition has become a theme in medicinal chemistry. Furthermore, helices formed by β-amino acids are experimentally more stable than those formed by α-amino acids. This is paradoxical because the larger sizes of the hydrogen-bonding rings required by the extra methylene groups should lead to entropic destabilization. In this study, molecular dynamics simulations using the second-generation force field, AMOEBA (Ponder, J.W., et al., Current status of the AMOEBA polarizable force field. J Phys Chem B, 2010. 114(8): p. 2549-64.) explored the stability and hydrogen-bonding patterns of capped oligo-β-alanine, oligoalanine, and oligoglycine dodecamers in water. The MD simulations showed that oligo-β-alanine has strong acceptor+2 hydrogen bonds, but surprisingly did not contain a large content of 3(12) -helical structures, possibly due to the sparse distribution of the 3(12) -helical structure and other structures with acceptor+2 hydrogen bonds. On the other hand, despite its backbone flexibility, the β-alanine dodecamer had more stable and persistent <3.0 Å hydrogen bonds. Its structure was dominated more by multicentered hydrogen bonds than either oligoglycine or oligoalanine helices. The 3(1) (PII) helical structure, prevalent in oligoglycine and oligoalanine, does not appear to be stable in oligo-β-alanine indicating its competition with other structures (stacking structure as indicated by MD analyses). These differences are among the factors that shape helical structural preferences and the relative stabilities of these three oligopeptides.

Establishing effective simulation protocols for beta- and alpha/beta-peptides. III. Molecular mechanical model for acyclic beta-amino acids

All-atom molecular mechanics (MM) force field parameters are developed for the backbone of acyclic beta-amino acid using an improved version of the multiobjective evolutionary algorithm (MOEA). The MM model is benchmarked using beta(3)-homo-Alanine (beta(3)-hAla) diamide in water with SCC-DFTB/MM simulations as the reference. Satisfactory agreements are found between the MM and SCC-DFTB/MM results regarding the distribution of key dihedral angles for the beta(3)-hAla diamide in water. The MM model is further applied to a beta-hepta-peptide in methanol solution. The calculated NOE values and (3)J coupling constants averaged over different trajectories are consistent with experimental data. By contrast, simulations using parameters directly transferred from the CHARMM22 force field for proteins lead to much worse agreement, which highlights the importance of careful parameterization for non-natural peptides, for which the improved MOEA is particularly useful. Finally, as an initial application of the new force field parameters, the behaviors of a short random copolymer consisting of beta amino acids in bulk solution and membrane/water interface are studied using a generalized Born implicit solvent model (GBSW). Results for four selected sequences show that segregation of hydrophobic and cationic groups occur easily at the membrane/solution interface for all sequences. The sequence that features alternating short blocks exhibits signs of lower stability at the interface compared to other sequences. These results confirm the hypothesis in recent experimental studies that beta-amino-acid based random copolymers can develop a high degree of amphiphilicity without regular three-dimensional structure.

Structure determination of a flexible cyclic peptide based on NMR and MD simulation 3J-coupling

Molecular dynamics (MD) simulations, in which experimental values such as nuclear Overhauser effects (NOEs), dipolar couplings, (3)J-coupling constants or crystallographic structure factors are used to bias the values of specific molecular properties towards experimental ones, are often carried out to study the structure refinement of peptides and proteins. However, (3)J-coupling constants are usually not employed because of the multiplicity of torsional angle values (phi) corresponding to each (3)J-coupling constant value. Here, we apply the method of adaptively enforced restraining using a local-elevation (LE) biasing potential energy function in which a memory function penalizes conformations in case both the average <(3)J> and the current (3)J-values deviate from the experimental target value. Then, the molecule is forced to sample other parts of the conformational space, thereby being able to cross high energy barriers and to bring the simulated average <(3)J> close to the measured <(3)J> value. Herein, we show the applicability of this method in structure refinement of a cyclo-beta-tetrapeptide by enforcing the (3)J-value restraining with LE on twelve backbone torsional angles. The resulting structural ensemble satisfies the experimental (3)J-coupling data better than the NMR model structure derived using conventional single-structure refinement based on these data. Thus, application of local-elevation search MD simulation in combination with biasing towards (3)J-coupling makes it possible to use (3)J-couplings quantitatively in structure determination of peptides.

Electrochemistry of ferrocenoyl beta-peptide monolayers on gold

The electrochemistry of self-assembled monolayers (SAMs) on gold containing a lipoic acid linker, the beta-peptide sequence (beta(3)Val-beta(3)Ala-beta(3)Leu)(n) for n = 1, 2, and a terminal ferrocenyl group has been described for the first time. Circular dichroism (CD), NMR, and molecular modeling were used to evaluate the beta-peptide structure in solution, while the monolayer film organization and electron-transfer kinetics were evaluated by cyclic voltammetry, chronoamperometry (CA), and ellipsometry. The peptides were assembled from trifluoroethanol solutions, where they are linear (n = 1) or helical (n = 2) based on CD, NMR, ellipsometry, and modeling evidence. The structure of the SAMs is less well understood. There is evidence for noncompact layers that allow electrolyte ions to approach the interface. Electron-transfer rates for n = 1, 2 were found to be 2500 and 1200 s(-1), respectively, and CA evidence indicated that the transfer is based on the hopping mechanism.

Mechanical stability of helical beta-peptides and a comparison of explicit and implicit solvent models

Synthetic beta-peptide oligomers have been shown to form stable folded structures analogous to those encountered in naturally occurring proteins. Literature studies have speculated that the conformational stability of beta-peptides is greater than that of alpha-peptides. Direct measurements of that stability, however, are not available. Molecular simulations are used in this work to quantify the mechanical stability of four helical beta-peptides. This is achieved by subjecting the molecules to tension. The potential of mean force associated with the resulting unfolding process is determined using both an implicit and an explicit solvent model. It is found that all four molecules exhibit a highly stable helical structure. It is also found that the energetic contributions to the potential of mean force do not change appreciably when the molecules are stretched in explicit water. In contrast, the entropic contributions decrease significantly. As the peptides unfold, a loss of intramolecular energy is compensated by the formation of additional water-peptide hydrogen bonds. These entropic effects lead in some cases to a loss of stability upon cooling the peptides, a phenomenon akin to the cold denaturing of some proteins. While the location of the free energy minimum and the structural helicity of the peptides are comparable in the implicit-solvent and explicit-water cases, it is found that, in general, the helical structure of the molecules is more stable in the implicit solvent model than in explicit water.

Theoretical analysis of secondary structures of beta-peptides

Unlike alpha-amino acids, peptides formed from beta-amino acids (beta-peptides) display stability toward enzymatic degradation and may form turns and helices with as few as four residues. Because both the C alpha and C beta of the beta-amino acid may bear substituents, a large number of beta-amino acids can be synthesized. Beta-peptides form various well-defined secondary structures, including 14-helix, 12-helix, 10/12-helix, 10-helix, 8-helix, turn structures, sheets, and hairpins. For all of these reasons, beta-amino acids have been increasingly used as building blocks for molecular design and pharmaceutical applications. To explain the conformational features of beta-peptides, several quantum mechanics and molecular dynamics studies that rationalize the observed conformational features have been reported. However, a systematic account that unifies various factors critical to the conformational features is still lacking. In this Account, we present a detailed analysis of the conformational features of various beta-peptides. We start by studying the basic local conformational features of beta-peptides using di- and tripeptide models. Then, various secondary structures of unsubstituted beta-peptides with differing numbers of residues are investigated using a repeating unit approach to derive the intrinsic backbone conformational features. We find that the 10/12-helix is intrinsically most stable for the beta-peptide backbone. The 14-helix, 12-helix, and 10-helix structures have similar stabilities for beta-peptide backbones of four to six residues. The substituent effects on the stabilities of beta-peptide secondary structures are then analyzed. Combined with the substituent effect and the intrinsic backbone preferences, all experimental observations of secondary structure formation can be understood. For example, the 10/12-helix is favored for like-beta(2)/beta(3)-peptides, unlike-beta(3)/beta(3)-peptides, and beta(3)/beta-hGly-peptides because these substitution patterns do not cause steric problems for the 10/12-helix. Beta(3)-peptides, beta(2)-peptides, and beta (2,3)-peptides favor the 14-helix because the substituents in these peptides benefit the 14-helix the most but significantly destabilize the 10/12-helix. Because the 10/12-helix is intrinsically favored and has two favorable positions in each residue for substituents, many more hybrid beta-peptides are predicted to exist in this secondary structure, which suggests the need for further experiments. These results are valuable for determining the best use of these building blocks in the design of well-structured molecules with desirable chemical functions.

Establishing effective simulation protocols for beta- and alpha/beta-peptides. II. Molecular mechanical (MM) model for a cyclic beta-residue

All-atom molecular mechanical (MM) force field parameters are developed for a cyclic beta-amino acid, amino-cyclo-pentane-carboxylic acid (ACPC), using a multi-objective evolutionary algorithm. The MM model is benchmarked using several short, ACPC-containing alpha/beta-peptides in water and methanol with SCC-DFTB (self consistent charge-density functional tight binding)/MM simulations as the reference. Satisfactory agreements are found between the MM and SCC-DFTB/MM results regarding the distribution of key dihedral angles for the tetra-alpha/beta-peptide in water. For the octa-alpha/beta-peptide in methanol, the MM and SCC-DFTB/MM simulations predict the 11- and 14/15-helical form as the more stable conformation, respectively; however, the two helical forms are very close in energy (2-4 kcal/mol) at both theoretical levels, which is also the conclusion from recent NMR experiments. As the first application, the MM model is applied to an alpha/beta-pentadeca-peptide in water with both explicit and implicit solvent models. The stability of the peptide is sensitive to the starting configuration in the explicit solvent simulations due to their limited length ( approximately 10-40 ns). Multiple ( approximately 20 x 20 ns) implicit solvent simulations consistently show that the 14/15-helix is the predominant conformation of this peptide, although substantially different conformations are also accessible. The calculated nuclear Overhauser effect (NOE) values averaged over different trajectories are consistent with experimental data, which emphasizes the importance of considering conformational heterogeneity in such comparisons for highly dynamical peptides.

Simulation ofβ-depsipeptides: The effect of missing hydrogen-bond donors on their folding equilibria

beta-Depsipeptides are beta-peptides in which one or more peptide linkages are replaced by ester linkages, resulting in a loss of a hydrogen-bond donor (N--H) and weakening of the corresponding carbonyl hydrogen-bond acceptor moiety. The effects of three of such peptide by ester substitutions in a hepta-beta-peptide upon its (un)folding equilibrium in methanol solution are investigated using molecular dynamics simulations and compared to experimental data from NMR spectroscopy. The simulated conformational ensembles largely reproduce the experimentally measured NOE and 3J-coupling constant data for the three different hepta-beta-peptides, and confirm the relative stabilities of the 3(14)-helical conformation, which is most weakened by substitution of the 4th peptide linkage and least by substitution of the 6th peptide linkage. The simulations are complementary to the experimental data by providing detailed insight into the conformational distributions that are compatible with the experimentally measured average values of observables.

Thermodynamic stability of beta-peptide helices and the role of cyclic residues

Beta-peptides are emerging as an attractive class of peptidomimetic molecules. In contrast to naturally occurring alpha-peptides, short oligomers of beta-amino acids (comprising just 4-6 monomers) exhibit stable secondary structures that make them amenable for quantitative, concerted experimental and theoretical studies of the effects of particular chemical interactions on structure. In this work, molecular simulations are used to study the thermodynamic stability of helical conformations formed by beta-peptides containing varying proportions of acyclic (beta(3)) and cyclic (ACH) residues. More specifically, several beta-peptides differing only in their content of cyclic residues are considered in this work. Previous computational studies of beta-peptides have relied mostly on energy minimization of molecular dynamics simulations. In contrast, our study relies on density-of-states based Monte Carlo simulations to calculate the free energy and examine the stability of various folded structures of these molecules along a well-defined order parameter. By resorting to an expanded-ensemble formalism, we are able to determine the free energy required to unfold specific molecules, a quantity that could be measured directly through single-molecule force spectroscopy. Simulations in both implicit and explicit solvents have permitted a systematic study of the role of cyclic residues and electrostatics on the stability of secondary structures. The molecules considered in this work are shown to exhibit stable H-14 helical conformations and, in some cases, relatively stable H-12 conformations, thereby suggesting that solvent quality may be used to manipulate the hydrogen-bonding patterns and structure of these peptides.

Helices and other secondary structures of β- and γ-peptides

The principal secondary structural motifs adopted by peptides assembled from beta-amino acid units are discussed: the 14-, 12-, 10-, 12/10-, and 8-helices, as well as the hairpin turn, extended structures, stacks, and sheets. Features that promote a particular folding propensity are outlined and illustrated by structures determined in solution (NMR) and in the solid-state (x-ray). The N-C(beta)-C(alpha)-CO dihedral angles from molecular dynamics simulations, which are indicative of a particular secondary structure, are presented. A brief description of a helix and a turn of gamma-peptides is also given.

On the Influence of Charged Side Chains on the Folding–Unfolding Equilibrium of β-Peptides: A Molecular Dynamics Simulation Study

The influence of charged side chains on the folding-unfolding equilibrium of beta-peptides was investigated by means of molecular dynamics simulations. Four different peptides containing only negatively charged side chains, positively charged side chains, both types of charged side chains (with the ability to form stabilizing salt bridges) or no charged side chains were studied under various conditions (different simulation temperatures, starting structures and solvent environment). The NMR solution structure in methanol of one of the peptides (A) has already been published; the synthesis and NMR analysis of another peptide (B) is described here. The other peptides (C and D) studied herein have hitherto not been synthesized. All four peptides A-D are expected to adopt a left-handed 3(14)-helix in solution as well as in the simulations. The resulting ensembles of structures were analyzed in terms of conformational space sampled by the peptides, folding behavior, structural properties such as hydrogen bonding, side chain-side chain and side chain-backbone interactions and in terms of the level of agreement with the NMR data available for two of the peptides. It was found that the presence of charged side chains significantly slows down the folding process in methanol solution due to the stabilization of intermediate conformers with side chain-backbone interactions. In water, where the solvent competes with the solute-solute polar interactions, the folding process to the 3(14)-helix is faster in the simulations.

The GROMOS software for biomolecular simulation: GROMOS05

We present the latest version of the Groningen Molecular Simulation program package, GROMOS05. It has been developed for the dynamical modelling of (bio)molecules using the methods of molecular dynamics, stochastic dynamics, and energy minimization. An overview of GROMOS05 is given, highlighting features not present in the last major release, GROMOS96. The organization of the program package is outlined and the included analysis package GROMOS++ is described. Finally, some applications illustrating the various available functionalities are presented.

Solution structure of a beta-peptide ligand for hDM2

We recently reported a beta-peptide foldamer, beta53-1, that folds into a 14-helix in aqueous solution, binds the oncoprotein hDM2 with submicromolar affinity, and potently inhibits the interaction of hDM2 with a peptide derived from the activation domain of p53 (p53AD). Here, we present the solution structure of beta53-1 in methanol. Details of the structure illustrate fundamental and novel elements of beta-peptide folding and recognition. These elements include the detailed arrangement of a complex, 14-helix-stabilizing salt bridge on one helical face, and a unique "wedge into cleft" packing interaction along a second. The structure also reveals how a subtle distortion in the beta53-1 14-helix geometry alters the presentation of its recognition epitope, rendering it particularly well suited for alpha-helix mimicry. The solution structure of beta53-1 demonstrates that well folded beta-peptide oligomers can effectively present an extended, highly variable surface that could be used as a general platform for targeting critical protein-protein interfaces.

Peptide Motifs for Cell-Surface Intervention

The discovery of new antimicrobial and anticancer drugs, and overcoming the problem of resistance to current anti-infective and anticancer drug therapies require innovation in the pharmaceutical and scientific research community. A further challenge of drug design is to make the therapeutic agent specific, long lasting, of minimal toxicity, and affordable. Microbial and cancer cell surfaces present molecular features that can differentially prefocus drugs within the human host. This property can localize drugs near cell-surface targets, thereby reducing opportunities for adverse effects, or the emergence of drug resistance caused by intracellular drug and target modification and by the induction of drug efflux pumps. The solubility demands on cell-surface targeting drugs should also be less stringent than for those drugs requiring transmembrane transport or internalization in order to reach intracellular targets. Cationic peptides have provided an increasingly important research focus in this regard. Although the cationic antimicrobial peptides are distributed widely in nature and provide localized primary defenses against microbial attack, the susceptibility of L-peptides to proteolysis and the known properties of successful antimicrobials have led to a focus on circularized peptides, D,L-peptides, and peptides containing unusual amino acids. New on the scene as lead antifungal agents are D-octapeptides and their derivatives that were developed from a combinatorial library produced through solid-phase peptide synthesis protocols. These peptides contain an amidated C-terminal tri-arginine motif, which confers membrane impermeability and focuses the peptides near the fungal cell surface. To date, the octapeptides and their derivatives also require some aromaticity, preferably the indole ring of tryptophan. In some cases, a single 4-methoxy-2,3,6-trimethylbenzenesulfonyl moiety remaining on the peptide after incomplete cleavage of the peptide from the solid phase produces a peptide with activity, whereas the parent shows little or no activity in the screen. Recent research advances that support the polycationic cell surface approach include the RGD (Arg-Gly-Asp) tripeptide and its mimetics, as well as aminoglycoside arginine drugs (e.g. neomycin coupled to small arginine polymers) and prodrugs. In the case of polycationic peptides, D-peptides could be used for intravenous injection and direct-surface drug applications, but mimetics will probably be needed for oral delivery.