Non-covalent stabilization of a -hairpin peptide into liposomes Dennis W. P. M. Löwik,* Jeffrey G. Linhardt, P. J. Hans M. Adams and Jan C. M. van Hest Department of Organic Chemistry, University of Nijmegen, Toernooiveld 1 – U177, 6525 ED Nijmegen, The Netherlands. E-mail: lowik@sci.kun.nl; Fax: +31(0)243653393; Tel: +31(0)243652325 Received 4th April 2003, Accepted 7th May 2003 First published as an Advance Article on the web 12th May 2003 An oligopeptide modified on both the N- and C-termini with hydrophobic moieties was prepared on a solid phase and anchored into a liposome, stabilizing the fold of the peptide into a -hairpin, which would otherwise be a random coil. Much effort has been put in the design of peptide-based mole- cules that are able to interact selectively with proteins by mimicking a natural binding partner. 1 This design has proven to be a non-trivial task and the enormous diversity of potential targets requires a flexible approach. Moreover, it has been clearly demonstrated that such interactions do not only depend on the primary amino acid sequence, but also on the three- dimensional structure that sequence adopts. Additionally, often multiple epitopes are involved in binding, in which they are usually displayed at the turn of β-hairpin structures. 2 Therefore, many researchers have recognized the importance not only to introduce the specific recognition site for a template, but also to mimic the structural environment in which this oligopeptide is incorporated. Since β-hairpins are ubiquitous in molecular recognition events and biological activity of proteins, much research has focused on the stabilization of these structures in oligopeptides. Various approaches have been described in the literature to achieve this goal: e.g. variations in primary amino acid sequence, 3 introduction of stabilizing linkages such as cysteine bridges, 4 amide bonds, 5 metal complexation, 6 double bonds 7 using non-natural amino acids, 8 or attachment to a scaf- fold. 9,10 These methods have been successful in introducing the desired fold, however, their limitation is that they require the positioning of certain amino acids at specific sites in the sequence. Additionally, the covalent approaches might restrict the peptide’s flexibility to such an extent that it could impair its biological activity! The goal of our current research is to stabilize short β-hairpin peptides onto a dynamic liposome scaffold by means of a non-covalent approach. Oligopeptides capable of forming β-hairpins can be modified on both the N- and C-termini with hydrophobic moieties, 11,12 allowing the peptides to be anchored at both ends in a dynamic bilayer, as shown schematically in Fig. 1. This in our view simple but novel approach, results in stabilization of the folding pattern without much interference with the dynamic character of the peptide, contrary to covalent methods used to impose secondary structure. 13 Furthermore, the method presented here does not require the incorporation of specific amino acid residues nor any specialized chemistry but conventional solid-phase peptide synthesis. Finally, modifi- cation of the liposome periphery by non-covalent interactions and exploiting the fluidic nature of the bilayer allows for the preparation of combinatorial libraries of epitopes. In our initial experiments, the β-turn sequence from the CS protein of the malaria parasite, plasmodium falciparum, 14 was chosen as an ideal candidate for stabilization by anchoring into a lipid bilayer. This fold has already been carefully studied by Robinson and coworkers using covalent attachment to a syn- thetic template. They demonstrated that a short peptide based on the NPNA sequence of the CS protein could be stabilized in the β-turn conformation on an -Pro–-Pro scaffold creating a cyclic peptide. 10,12,15 In order to study the effect of N- and C-terminal alkyl chains on peptide conformation, peptides 1–3 (Fig. 1) were syn- thesized. Peptides 1 and 2 were prepared on a Wang resin using standard Fmoc chemistry 16 and, in the case of peptide 2, a final coupling with stearic acid was performed before cleavage. The synthesis of peptide 3, which required the incorporation of lipophilic tails at both termini of a peptide, was accom- plished as shown in Scheme 1. The preparation commenced with a reductive amination reaction on a commercially available aldehyde modified resin, 17 followed by standard Fmoc synthesis of the peptide and subsequent capping with stearic acid, to afford the peptide modified at both termini. This generic solid phase strategy allows one to prepare the amphiphilic peptides completely on a solid phase, enabling quick production of a variety of lipidated peptides. The folding characteristics of the peptides were studied with several techniques. First, the behaviour of peptide amphiphiles 2 and 3 was investigated at the air–water interface using a Langmuir trough. 18 The isotherms of peptides 2 and 3 and those of stearic acid and distearoyl phosphatidyl choline (DSPC) are depicted in Fig. 2. Comparing the behaviour of peptide 2 with that of stearic acid, we conclude the behaviour of 2 not to be determined by its alkyl chain but by the peptide head group. However, no further conclusion could be drawn since a stable monolayer was not formed, probably due to the partial solubility of the peptide in water. In contradistinction, peptide 3, containing two alkyl chains, formed a stable mono- layer. From the observed plot a molecular area was extra- polated of just over double the size of a stearic acid molecule, as can be seen in Fig. 2. The first rise of the curve compared favorably with DSPC, a neutral amphiphile with two stearoyl tails. It is surprising to see that the molecular area occupied by 3 is solely determined by the lipophilic alkyl chains. Therefore, we conclude the peptide must be able to adopt such a conform- ation that it is possible for the alkyl chains to closely pack in a similar fashion as for the model phospholipid. A molecular model of compound 3, as shown in Fig. 3a, indicates that such a Fig. 1 Stabilization of a β-hairpin by attachment of terminal alkyl tails. DOI: 10.1039/ b303749e This journal is © The Royal Society of Chemistry 2003 Org. Biomol. Chem. , 2003, 1, 1827–1829 1827