Lipopeptides for the Fragment-Based Pharmaceutics Design

This paper describes the synthesis of peptide fragments for use in a new type of combinatorial discovery technology, in which the building blocks are brought together by non-covalent interactions, rather than direct chemical bonding. The building blocks of interest—in this case different amino acids—are converted to amphiphiles by attachment to lipid tails. The amphiphiles, when mixed together in aqueous phase, are designed so that they aggregate spontaneously to form micelles. The building blocks form the headgroups of each of the amphiphiles, and these headgroups cover the surface of the micelle in a dynamic close-packed fluid mosaic array. These building blocks come together so closely that twoor three-dimensional structures are created on the surface of the micelles, and these can be screened in biological assays to find out which combination of building blocks is able to elicit a biological response. Lipopeptides consisting of two residues of lipoamino acid and other amino acids moieties have been designed, synthesized, characterized and the ability of these constructs to form supra-molecular assemblies is demonstrated.


Introduction
Conventional methods of drug discovery often employ combinatorial chemistry as part of the methodology to create new structures as potential ligands for binding to biological targets [1][2][3][4].When using amino acids as building blocks, the individual subunits are linked together via peptide bonds.Unfortunately, these techniques suffer from a number of disadvantages [5].The created peptides are time-consuming and expensive to make and purify, even using parallel synthesis, and the quantities available for testing are severely limited.Furthermore, the need for chemical linkages introduces physical constraints on the molecules under study.They are usually linear, and do not have free rotation about the bonds which joining them together so that even if the right combination of building blocks is selected, it is possible that they will not have the right orientation with respect to each other to interact appropriately with the target.
For these reasons, we have devised a combinatorial screening system in which amino-acid building blocks are brought together without the use of covalent linkages.The amino acids are conjugated to lipid tails so that amphiphiles are formed, in which the amino-acid building blocks made up the headgroups of the amphiphiles.A series of different amphiphiles has been constructed, where each member of the series has a different headgroup, but the lipid tails are identical throughout.The amphiphiles form micelles when mixed in aqueous phase, thus bringing the headgroups together sufficiently closely to form structures capable of binding to biological targets (Figure 1).In order to achieve the right degree of spacing between headgroups on the surface of the micelles, design of the lipid tail is important [6,7].Here, the entire amphiphile is synthesized in the form of a linear peptide, where the lipid tail is provided by two lipoamino acids in tandem in which the side-chain residues are extended straight-chain hydrocarbons.In addition, amino acids, usually glycine and serine, are used a spacers between the lipid tail and headgroup.

Results and Discussion
We designed and synthesized amphiphilic peptides 1 and 2, containing lipophilic amino acidic residues (Figure 2) combined with representatives of coded amino acids.The lipoamino acids are alpha amino acids containing long hydrocarbon side chains [8,9].
Lipopetides 1a-d possessed only natural peptide bonds, whereas the lipopeptide derivatives (2a-c) were synthesized with N-methylated dilipidic residue to increase the biological stability.In both cases, to prevent the degradation of the building blocks by exopeptidases, glycolic acid was attached to the N-terminus and C-terminus was amidated.While solid phase synthesis was used to produce compounds 1a-d, solution-phase synthesis was favored for compounds 2a-c.The synthesized amphiphilic peptides were designed to aggregate spontaneously to form micelles when codispersed in aqueous solution.The ability of the amphiphiles to form supra-molecular assemblies was assessed by isothermal titration calorimetry (ITC), dynamic light scattering (DLS) and then was confirmed by transmission electron microscopy (TEM).
The particles formed during the ITC experiments were used for the DLS analysis.We detected two sets of size distributions, the first was in the range of 30 -110 nm for the mixture 1a-d and of 25 -100 nm for 2a-c, and the second was in the range from 200 to over 1 μm for both mixtures.
TEM experiments confirmed the results obtained from the DLS measures and confirmed the formation of supramolecular structures (Figure 3).TEM images of the mixture 1a-d showed nanoparticles (20 -60 nm) with tendency to aggregate to larger particles.Images of the mixture 2a-c showed nanoparticle formation in the size ranges from 20 -40 nm to particles of hundreds nm.
The aim of the approach described here was to facilitate the studies of ligand-receptor interactions and to provide a new tool for drug discovery.We developed the described library of lipopeptides to address the structural diversity and variability of short peptide sequences.The lipopeptide building blocks were designed to form micelles on which surface the headgroups/amino acids are displayed in such manner that they are mimicking a random peptide sequence.Thus, a variety of amino acids were incorporated into the building blocks.It was desirable to keep the length of the peptide to a minimum to avoid eliciting unwanted immune response.We focused on the development of chemical bonds linking the head groups to the lipidic tails.Lipopetides 1a-d were synthesized with natural peptide bonds exhibiting a considerable degree of flexibility, while lipopeptide 2a-c were synthesized with N-methylated dilipidic residue to study its influence on peptide conformation and biological activity.The mixtures of synthesized compounds were able to form nanoparticles.However, an extensive aggregation was also observed.The tendency to form large aggregates can be explained by the relatively high hydrophobicity of synthesized amphiphiles [13,14].To obtain more uniform particles, the nanoparticle might be selfassembled by dialysis of a solution of the lipopeptides in organic solvent against water [15,16].

Conclusion
New fragment-based lipopeptide building blocks were designed, synthesized and characterized.The lipid-modified peptidic amphiphiles were capable of forming a supramolecular assemblies bound by weak non-covalent associations.These compounds will be applied for the study of the receptor-ligand interactions.Biological evaluation of these lipopeptidic constructs is underway, and results will be reported elsewhere.Applying our dynamic concept might provide the new generation of therapeutics.and reagents, trifluoroacetic acid (TFA) and diisopropylethylamine (DIPEA), were purchased from Auspep (Melbourne, VIC, Australia).O-Benzotriazole-N,N,N', N'-tetramethyl-uronium-hexafluoro-phosphate (HBTU) and di-tert-butyl dicarbonate were obtained from GL Biochem Ltd (Shanghai, China).N-Boc-protected amino acids and pMBHA resin were supplied by Novabiochem (Läufelfingen, Switzerland).Palladium (10% wt on carbon) was purchased from Lancaster Synthesis (Lancashire, England).Ultra pure gases (N 2 , H 2 , Ar) were supplied by BOC Gases (Brisbane, QLD, Australia).Silica for flash chromatography (Silica gel 60, 230 -400 mesh) was obtained from Lomb Scientific (Taren Point, NSW, Australia).Deuterated solvents (d 1 -CDCl 3 and DMSO-d 6 ) were manufactured by Cambridge Isotope Laboratories Inc. (Andover, MA, USA).All other reagents were purchased in analytical grade or higher purity from Sigma-Aldrich (Castle Hill, NSW, Australia) or Merck Pty Ltd (Kilsyth, VIC, Australia).Solvents were freshly distilled and dried prior to use and all moisture-sensitive reactions were carried out under inert atmosphere (N 2 /Ar) using oven-dried glassware.
1 H NMR spectra were recorded at 297 K using a Bruker spectrometer at 400 MHz and 500 MHz.The instrument operating at 400 MHz and 500 MHz for 1 H used CDCl 3 , CD 3 OD or Acetone-d 6 as solvent and tetramethylsilane as an internal standard, unless stated otherwise.Coupling constants were provided in Hz. 13 C spectra were measured at 100.62 MHz and 125.77MHz and referenced to CDCl 3 (77.0Hz), CD 3 OD (49.0 Hz) or Acetone-d 6 (30.5 Hz).Homo-and heteronuclear 2D NMR spectroscopy was performed using Bruker standard software.Thin layer chromatography (TLC) was performed on Merck pre-coated aluminum sheets (Silica Gel 60F254); spots were detected by spraying with H 2 SO 4 .Amine derivatives were detected by ninhydrin spray, followed by heating the sheets.Column chromatography was performed on silica gel columns (size A, 28 × 2; B, 30 × 2.5; and C, 43 × 4 cm; silica gel 0.040 -0.063 mm).Analytical RP-HPLC was performed on a Shimadzu instrument (LC-10AT liquid chromatograph, SC L -10A system controller, SPD-6A UV detector, a SIL-6B auto injector with a SCL-6B system controller, and a C18-HPLC-column using an acetonitrile/water/0.1% TFA gradient as well as an isopropanol/water/0.1% TFA gradient.HPLC purification was done on a Waters HPLC system (Model 600 controller, 490E UV detector, F pump, and 0.46 × 15 cm Vydac RP-C18 column with 0.005 mm particle size) using a acetonitrile/water/0.1% TFA gradient.He-gas was applied for degassing of HPLCsolvents.Mass spectra were obtained on a quatropolelectrospray-MS (Perkin Elmer API 3000 instrument) in the positive ion mode.Concentration of solutions was performed at reduced pressure and at temperatures < 40˚C.
High-resolution mass spectrometry (HRMS) data were obtained on a QStar Pulsar instrument (Applied Biosystems) operating in positive-ion electrospray mode.The thermodynamic interactions were calculated using a Mi-croCal VP-ITC Microcalorimeter (Northampton, MA, USA) with Origin 5.0 software and VPViewer 2000.The particle size measurements were done by using a Zetasizer Nano ZP instrument (Malvern Instruments, UK) with DTS software.
Transmission electron microscopy photographs were done on a JEOL-1010 microscope operating at an acelerating voltage of 100 kV.The sample images were taken and analyzed using the AnalySIS ® software (Soft Imaging Systems, Megaview III, Munster, Germany).

Synthesis
2-(R/S)-Aminododecanoic acid and its Boc-protected intermediate 3 were synthesized according to the published method [17,18].Tritylation of glycolic acid was performed by reaction of glycolic acid with tritylchloride and 4-dimethylaminopyridine (DMAP) in pyridine according to the literature [11].

Synthesis of Peptides 1a-d General Protocol for the synthesis of peptides 1a-d (SPPS method):
MBHA resin (4-Methyl benzhydrylamine, substitution ratio: 1.03 mmol/g, 3.0 mmol scale) was swelled in dimethylformamide in a sintered glass peptide synthesis vessel for 90 min.Each amino acid coupling cycle consisted of Boc-deprotection with neat TFA (2 × 1 min), a 1 min DMF flow wash, followed by 20 min coupling with the pre-activated amino acid.An activation mixture consisting of Boc-amino acid (3 eq. per mol amino-group), HBTU (2-(1H benzotriazole-1-yl)-1,3,3-tetramethyl-uronium hexafluoro phosphate, 0.5 M in DMF, 3 eq.)and DIPEA (0.442 mL, 4 eq.) was shaken for 12 min.Coupling efficiency was monitored by a quantitative ninhydrin test (≥99.8%).Upon completion of the synthesis and removal of the terminal Boc groups, the resin was washed with DMF, methanol and DCM.The resin was dried to constant weight over KOH in vacuo.The peptides were cleaved from the resin using HF, and p-cresol as a scavenger.The cleaved peptides were precipitated, filtered and washed thoroughly with ice-cold diethyl ether.The crude peptide was obtained as an amorphous powder after redissolving the cleaved peptides in acetonitrile-water (1:1) with 0.1% TFA and lyophilisation.200 mg of each crude peptide was separated using Sephadex LH-20 (70 × 3 cm) and acetonitrile-water (1:1) as a solvent.Peptide-positive fractions were determined on silica gel by spraying with ninhydrin-reagent.Analytical RP-HPLC (C4, 25 cm Vydac C4, C18 column with 5 nm pore size and 4.6 mm internal diameter) was performed using two different solvent systems in order to check the peptides' purity.Solvent system 1 comprised; Solvent A (H 2 O, 0.1% TFA), and Solvent B (90% CH 3 CN, 10% H 2 O, 0.1% TFA); and Solvent system 2 comprised; Solvent A (H 2 O, 0.1% TFA), Solvent B (90% Methanol, 10% H 2 O, 0.1% TFA).The gradient was 0% -100% B within 20 min; the flowrate was 1 mL/min; and the wave length was 214 nm.The purity of all peptides analyzed by both systems on HPLC was over 95%.Peptides were also characterized using ESI-MS.The resulting peptides were used as diastereomeric mixtures.

TEM Experiments
The samples were used after self-assembly process from microcalorimeter sample cell.A drop of mixture 1a-d or 2a-c solution was allowed to air-dry onto a 200 mesh copper grid, and the sample was stained with 10% of ammonium molybdate for 5 minutes.

Figure 1 .
Figure 1.New concept of combinatorial discovery technology, in which the building blocks are brought together by non-covalent interactions.