Article pubs.acs.org/Biomac

A Self-Assembling Peptide Scaffold for the Multivalent Presentation of Antigens Elsa Zacco,† Chakkumkal Anish,†,‡,⊥ Christopher E. Martin,†,‡ Hans v. Berlepsch,†,§ Enrico Brandenburg,† Peter H. Seeberger,†,‡ and Beate Koksch*,† †

Institute of Chemistry and Biochemistry, Freie Universität Berlin, Takustrasse 3, 14195 Berlin, Germany Department of Biomolecular Systems, Max Planck Institute of Colloids and Interfaces, Am Mühlenberg 1, 14424 Potsdam, Germany § Research Center of Electron Microscopy and Core Facility BioSupraMol, Institute of Chemistry and Biochemistry, Freie Universität Berlin, Fabeckstraße 36a, 14195 Berlin, Germany ⊥ Bacterial Vaccines Discovery and Early Development, Janssen Pharmaceuticals (Johnson & Johnson), Leiden, The Netherlands ‡

S Supporting Information *

ABSTRACT: Self-assembling peptides can be used to create tunable higher-order structures for the multivalent presentation of a variety of ligands. We describe a novel, fiber-forming coiled-coil-based peptide that assembles to display, simultaneously, carbohydrate and peptide ligands recognized by biomacromolecules. Preassembly decoration of the scaffold with a diphtheria toxin peptide epitope or a mannose motif did not interfere with self-assembly of the nanostructure. The resulting multivalent display led to tighter binding by antidiphtheria toxin antibodies and mannose-specific carbohydrate binding proteins, respectively. The potential of this self-assembling peptide to display ligands in bioanalytical assays is illustrated by its decoration with a disaccharide glycotope from the Leishmania parasite. Carbohydrate-specific antibodies produced in response to a Leishmania infection are detected more sensitively in human and canine sera due to the multivalent presentation on the self-assembled scaffold. Thus, nanofibers based on coiled-coil peptides are a powerful tool for the development of bioassays and diagnostics.

■

INTRODUCTION The high surface-area-to-volume ratio and chemical diversity of nanoscale structures render them of great interest to the biomedical and pharmaceutical fields. Inspired by molecular recognition processes in nature, a host of synthetic nanoscaffolds have been developed including branched polymers and dendrimers, PNA, LNA, and self-assembling peptides.1−4 Compared to other nanoscale bioassemblies, peptides offer the advantage of a facile de novo design by fine-tuning of the primary sequence to obtain morphologies such as peptide nanotubes, amyloid fibrils, peptide nanoparticles, and α-helical nanofibers.5−9 For instance, a starting point for the synthesis of newly designed amyloid fibrils is the Aβ peptide from the APP protein,6 while collagen-mimetic peptides imitate an α-helical fiber supercoil occurring in nature.10 The predictable structural behavior of self-assembling peptides, dictated by electrostatic and hydrophobic interactions, can be modulated by changes in the environment (solvent, pH, metal ions, and temperature).11−14 Furthermore, peptides are easier to synthesize and characterize than other polymeric structures, are biocompatible, and offer distinct functional groups to which a molecule of interest can be conjugated with arbitrary stoichiometry.15,16 The architectures adopted by peptide nanoassemblies are often retained in the presence of guest molecules, and their potential as biomolecular carriers has © XXXX American Chemical Society

found applications in regenerative medicine and in the delivery of peptide epitopes and growth factors.17−19 Self-assembling peptide nanostructures based on the coiledcoil motif are more water-soluble and less cytotoxic than amyloidogenic peptides for biomedical applications.20 Moreover, coiled-coil peptides are highly thermally stable and can be covalently modified, concomitant with their synthesis, without perturbing their secondary structure.21 The motif consists of a core of seven amino acids, called the heptad repeat, denoted by the letters a, b, c, d, e, f, and g; positions a and d are generally occupied by hydrophobic residues, while e and g often present complementary charges to stabilize the helices. Two or more helices can arrange themselves around each other into a supercoil configuration, and their degree of oligomerization can be synthetically modulated to obtain highly aggregated systems,22 which makes this type of self-assembling peptide an attractive nanoscaffold for the presentation of ligands in a multivalent fashion. One requirement of such a system is that it be biomimetic: the ligand must be presented on the nanoscaffold in a manner that conforms with the specificity and reversibility of Received: April 29, 2015 Revised: May 26, 2015

A

DOI: 10.1021/acs.biomac.5b00572 Biomacromolecules XXXX, XXX, XXX−XXX

Article

Biomacromolecules

Abz−OH, selective deprotection of K17 was performed. This free amine was used as a starting point for the orthogonal synthesis of Ep01. The mannose derivative selected to be coupled to Ep01 and FF03-Ep01 was the 5-amino-pentanyl-α-D-mannopyranoside, and it was synthesized as previously reported.29 The synthesis of 5-aminopentanyl β-D-galactopyranosyl-(1 → 2)-α-D-mannopyranoside derived from the leishmanial lipophosphoglycan was performed according to the protocol previously established.36 The addition of the carbohydrate moieties to Ep01 and FF03-Ep01 was carried out by first switching the functionality on resin from −NH2 to −COOH with five-fold excess of glutaric anhydride and a catalytic amount of DIPEA for 3 h. The further activation of the free acid with 3 equiv HATU and 6 equiv DIPEA for 5 min allowed the coupling of the sugars on resin using sixfold excess. Final cleavage from the resin was performed using a solution of 95% TFA, 3% water, and 2% TIS (5 mL) with shaking for 3 h. The resin was washed with 1 mL of TFA and 1 mL of DCM, and excess solvent was removed by evaporation. The peptides were precipitated in cold diethyl ether and collected by centrifugation. Peptide Purification. The peptides were purified via analytical high-performance liquid chromatography (HPLC). This system was preferred over the preparative HPLC for the difficulty in the separation of the main products from the side products (monoglycosylation did not lead to a relevant difference in the elution time). The purification was performed on a LaChrom-ELITE-HPLC-System (VWR) consisting of two HPLC pumps (L-2130) with a solvent degasser, an auto sampler (L-2200), a diode flow detector (L-2455), and a high pressure gradient mixer. Data analysis was performed with EZ Chrom ELITE software. All runs were performed with a flow rate of 1.0 mL/ min using acetonitrile (ACN, 0.1% TFA) and Millipore H2O (0.1% TFA). Yield and purity of the final products were calculated according to the areas of all peaks during the purification procedure. Peptide Concentration Determination. Peptide concentration was estimated by UV spectroscopy using the absorption maximum at 280 nm in 6 M GndHCl, as previously reported.38 For the peptide FF03 alone, the absorption of Abz at 320 nm was used. Peptide Characterization. Pure peptides were characterized via mass spectroscopy (MS), circular dichroism (CD), transmission electron microscopy (TEM), and size-exclusion chromatography coupled to static light scattering (SEC−SLS). The quality of the synthesized glycopeptides was determined using ESI-TOF 6210 Agilent (USA, CA-95051−7201, Santa Clara). All samples were dissolved in a mixture of water and ACN before injection, and the data were collected on positive ion mode. For the CD experiments, peptides were dissolved in a 100 mM phosphate buffer, and the pH was adjusted to 7.4. Spectra were recorded on a JASCO-8−10 spectropolarimeter with a temperature controller set at 20 °C. The CD data were registered using 0.1 cm path length quartz cuvettes and normalized according to extinction coefficient, optical path length, peptide concentration, and number of residues. A constant N2 flush of 3.0 L/min was provided. Samples for TEM were prepared by absorbing 5 μL aliquots of peptide solution onto glow-discharged carbon-coated collodium films on 400-mesh copper grids. The grids were blotted, stained with 1% phosphotungstic acid (PTA), and air-dried. TEM micrographs were taken at a primary magnification of 58 300 using a defocus of 0.8 μm. SEC was performed using a WTC-015S5 column (5 μM, 150 Å, 7.8 × 300 mm2, Wyatt Technology) connected to a HPLC workstation (La Chrom, VWR, Hitachi, L-2130). The separation was carried out for 50 min with a 0.3 mL/min flow rate at room temperature in PBS buffer consisting of 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, and 1.8 mM KH2PO4 at pH 7.4. Elution of peptide was monitored by UV detection (VWR, Hitachi, L-2400) at 230 nm. Analysis was performed with 100 μM and 50 μM peptide concentrations and with an injection volume of 100 μL. SLS experiments were performed at 25 °C by using a Dawn Heleos 8 light scattering photometer and an Optilab rEX refractive index detector (Wyatt Technology). Molecular weight values were calculated by using a dn/dc value of 0.185 mL/g. All data were analyzed with ASTRA software version 5.3.4.20 (Wyatt Technology).

biomolecular recognition events that are key to many biological processes.23,24 In nature, proteins often form multimeric architectures that contain several recognition sites to compensate for weak individual noncovalent interactions.25 Similarly, oligosaccharide and peptide ligands are often presented multivalently.26 Thus, molecular recognition is enabled by a synergistic combination of several cooperative binding events. Two aspects of the immune response may serve as good examples of this principle. On one hand, immune system receptors, such as pattern recognition receptors, have evolved to distinguish among binding partners based on antigen density.27 On the other hand, invading pathogens can be detected according to distinct ligand arrays, comprising, for example, repetitive glycoprotein determinants on virions.28 Glycans and peptide epitopes are useful diagnostic probes in bioassays and immunogens in vaccines29,30 and, because of their central role in the immune response, represent interesting classes of ligands to be explored in the context of self-assembled scaffolds.31 Recent advances in this area have demonstrated that simulating native antigen organization and recognition is an important design principle for nanomolecular structures for biomedical applications. With these criteria in mind, we designed a peptide nanoscaffold based on the coiled-coil folding motif that selfassembles into α-helical fibers under physiological conditions. Prior to assembly, the peptide scaffold was conjugated with a glycopeptide ligand using a novel all-on-solid-phase (AOSP) approach for the monomeric building block. The highly immunogenic epitope from Diphtheria toxin32,33 (DT) was selected as the peptide unit of the ligand and the mannosederivative 5-amino-pentanyl-α-D-mannopyranoside34 as the carbohydrate moiety. The peptide epitope is recognized by anti-DT antibodies when linked to the self-assembling scaffold. The carbohydrate is recognized in vitro by the mannose binding protein (lectin) concanavalin A (Con-A), and the availability of the glycan was confirmed in a whole-cell assay using a mannose-binding bacterial strain (E. coli ORN178).35 To evaluate the ability of the new coiled coil-based nanoscaffold to efficiently present the ligands on its surface for use in in vitro diagnostic assays, a conserved disaccharide glycotope from the surface of Leishmania parasites36 was appended to detect anticarbohydrate antibodies in clinical samples.

■

MATERIALS AND METHODS

Peptide Synthesis. Resins for solid-phase peptide synthesis were purchased from Novabiochem. Preloaded Fmoc-Leu-NovaSyn TGA resin and preloaded Fmoc-Gly-NovaSyn TGT resin, both with 0.2 mmol/g amino acid loading, were used for the synthesis of FF03 and Ep01, respectively. The synthesis was performed with standard Fmoc/ tBu chemistry, on a 0.05 mmol scale, using fluorenylmethyloxycarbonyl (Fmoc)-protected amino acids purchased from Orpegen and removing the Fmoc at each step with 20% piperidine in DMF (2 × 10 min and 2 × 5 min with 5 mL of deprotecting solution). The synthesis of Ep01 was performed manually with single 1-h couplings using eight-fold excess of amino acid, HOAt and DIC, relative to resin loading. The choice of the manual synthesis was dictated by the need to avoid histidine racemization. The synthesis of FF03 and FF03-Ep01 was performed in two main steps: FF03 was synthesized with 2-h double couplings using a SyroXP-I peptide synthesizer (Multi-SynTech GmbH) with eight-fold excess of amino acid, TBTU/HOBt as coupling agents, and eight-fold excess of DIPEA relative to resin loading. The lysine on position 17 was protected with N-methyltrityl (Mtt), an amine protective group removable under mild acidic conditions.37 After the full-length FF03 sequence was synthesized and the N-terminus was blocked with BocB

DOI: 10.1021/acs.biomac.5b00572 Biomacromolecules XXXX, XXX, XXX−XXX

Article

Biomacromolecules

Figure 1. FF03 fiber formation at neutral pH. Cylinders represent single helices; circles within each cylinder represent the 26 residues of the peptide sequence. (A) Helical wheel representation of FF03; (B) sticky end effect; (C) electrostatic interaction contribution to fiber formation. Yellow, hydrophobic residues; red, positively charged residues; blue, negatively charged residues; gray, most solvent-exposed residues. Letters within circles give the single-letter code for amino acids (A) or indicate heptad repeat positions (B and C). Enzyme-Linked Immunosorbent Assay (ELISA). One micromolar stock solutions Ep01 and FF03-Ep01 were prepared in PBS and incubated overnight to allow self-assembling process to complete. Samples (50 μL) were immobilized on 96-well Nunclon flat bottom transparent polystyrene plates at three different concentrations (1 μM, 100 nM, 10 nM) in triplicate. As a positive control, also the Diphtheria protein CRM19736 was immobilized in triplicate at the same concentrations. After incubation, the supernatant was aspirated, and 50 μL of blocking solution (1% BSA in PBS) was added to the wells and incubated for 1 h at 37 °C. The wells were then washed three times with PBST (PBS with 0.05% tween). A mix of mouse polyclonal sera against CRM197 was added to the wells with five different dilutions, from 1:1000 to 1:10 000 000, for 1 h and 30 min at 37 °C. Details about immunization can be found in our previous publication.36 After aspiration of supernatants, the wells were washed for 15 min with 150 mM NH4SCN, then PBST (two times). Goat anti-mouse IgG Fc antibody, horseradish peroxidase (HRP)-conjugate (Dianova), diluted 1:20 000 in PBS (50 μL), was incubated for 1 h at 37 °C and then removed. After washing again with PBST (three times), 50 μL of 3,3′,5,5′ tetramethylbenzidine substrate solution (1Step Ultra TMB-ELISA) was added as substrate for the peroxidase and left to react for 3 min. The reaction was stopped by addition of 2% sulfuric acid (50 μL). The plates were then read three times at 450 nm with the multimode microplate reader Infinite 200 PRO by Tecan. Analysis was performed using Microsoft Excel software for data plotting and p-value calculation (Student t-test type, and the number of tails: 1). Confocal Laser Scanning Microscopy (CLSM). Fluorescent images were acquired with the instrument Zeiss LSM 700 using the laser at a wavelength of 488 nm (2.0%) for FITC excitation. The selected objective was Plan Apochromat 63X/1.4 oil DIC, and the images were processed with the software ZEN 2009 by Zeiss. ConA-FITC Binding Assay. The two mannose-containing peptides and their controls without sugar (1.5 mg/mL) were dissolved in lectin-binding buffer (10 mM HEPES, 1 mM MgCl2, 1 mM CaCl2, pH 7.4) and spotted twice on poly lysine-coated microscope slides. The samples were left overnight to self-assemble, and 15 μL of blocking solution (10 mg/mL of BSA in PBS) was added to each spot and incubated for 1 h. A drop of ConA from Canavalia ensiformis, FITC-conjugated (5 μL, 5 mg/mL, purchased from Sigma-Aldrich) in lectin-binding buffer, was added and left for 1 h in the dark to interact with the peptides. The spots were washed twice with 15 μL of PBS,

and another drop of buffer was added before the results were observed with the CLSM. Mannose-Binding E. coli Assay. E. coli strain ORN178 was grown in Luria−Bertani broth in the presence of 12.4 μg/mL tetracycline at 37 °C with shaking to an OD595 nm of 0.8. A 50 μL culture was added to a 96-well plate (in triplicate), and the cells were collected by centrifugation (4 °C, 500 g, 20 min). The bacteria were washed three times with ice-cold PBS and incubated for 1 h at room temperature with 50 μL of PBS containing FITC (500 nmol final concentration). The cells were again washed three times, and 20 μL of the suspension was spotted onto poly lysine microscope plates, upon which 20 μL of a 1.5 mg/mL solution of FF03-Ep01 or FF03-Ep01Man in PBS had been left overnight to self-assemble. After incubation at room temperature, the cells were observed with CLSM. Microdot Assay. The peptide library, CRM197, and the carbohydrates were immobilized on CodeLink N-hydroxyl succinimide (NHS) ester activated glass slides (SurModics Inc., Eden Prairie) with a piezoelectric spotting device (S3; Scienion). The subsaturation concentration of 10 μM and the oversaturation concentration of 100 μM of compounds to react with NHS groups on the slide were chosen to minimize the effect of difference in antibody reactivity due to possible disparity in immobilization levels. On each slide, 64 replicate array grids were printed. The slides were incubated in a humid chamber for 24 h until reaction completion and quenched with 100 mM ethanolamine in 50 mM sodium phosphate buffer at pH 9 for 1 h at 50 °C. A washing procedure (three times) with deionized water followed, and the slides were dried via centrifugation for 5 min at 1200 g. Before use, the slides were blocked with 1% BSA in PBS overnight at 4 °C, washed three times with PBS, and dried again. With the help of the 64-well grid (ProPlate multiarray system), two different slides were treated to test the reactivity of the primary antibodies from sera obtained from five human patients and five dogs infected with the Leishmania parasite as previously reported.36 The slides were incubated with 40 μL of four dilutions (from 1:100 to 1:800) of the appropriate serum in blocking buffer (20 μL) for 1 h at 37 °C. The slides were washed three times with 40 μL of PBST and incubated for 1 h at 37 °C with 40 μL of secondary antibodies solution in 1% BSA in PBS. The secondary antibodies used are as follows: AlexaFluor 647 goat α-human IgG (Life Technologies) for the human sera, AlexaFluor 635 goat α-mouse IgG (Invitrogen) for the mouse sera (in the case of the CRM197 control), and DyLight 488 rabbit α-dog IgG (Fuller Laboratories) for the canine sera. The slides were washed four times C

DOI: 10.1021/acs.biomac.5b00572 Biomacromolecules XXXX, XXX, XXX−XXX

Article

Biomacromolecules

Figure 2. Peptide library. (A) Ep01; (B) Ep01-Man; (C) Ep01-Man-Gal; (D) FF03-Ep01; (E) FF03-Ep01-Man; (F) FF03-Ep01-Man-Gal.

Table 1. Peptide Sequences and Description of the Peptide Library. The Particular Lysine to Which the Different Ligands Are Attached Is Shown in Bold

Figure 3. Peptide CD spectra collected 1 h after sample preparation at concentrations between 100 and 300 μM. Values were normalized on the basis of the extinction coefficient, the path length, the peptide concentration, and the number of residues. (A) Solid line, Ep01; short dashes, Ep01Man; long dashes, Ep01-Man-Gal. (B) Solid line, FF03; short−long dashes, FF03-Ep01; short dashes, FF03-Ep01-Man; long dashes, FF03-Ep01Man-Gal. with 40 μL of PBST and dried as previously before scanning with a GenePix 4300A scanner (Bucher Biotec) with excitation wavelengths of either 488 nm, for FITC or DyLight 488, or 633 nm for AlexaFluor 647 and AlexaFluor 635; emission was detected by means of filters set to either 530 or 660 nm. The data were acquired with GenePix Pro 7 software (Bucher Biotec). Data and p-value calculation were performed using Microsoft Excel software. Student t-test type and number of tails selected were “1” and “2”, respectively.

axis. Furthermore, fiber elongation is promoted by “head-totail” stacking of the helices. In this conformation, complementary charges easily interact with each other (Figure 1B).40 Within each heptad, the a and d positions are occupied by leucine residues, providing the hydrophobic core of the helices.41 To guarantee the formation of self-assembling fibers, lysine is present at all e positions and glutamate at all g positions, while the b and c positions carry these two charged amino acids alternatively and complementarily. Thus, when a given heptad contains lysine at the b position and glutamate at the c position, the two heptads adjacent to it will carry glutamate at the b position and lysine at the c position. This design aims to induce the helices to interact with each other as well as to stabilize the coiled-coil motif. Figure 1 depicts a model for how fibers of FF03 could form. Electrostatic interaction between different dimers is likely the main factor supporting growth of the fibers perpendicular to the helix axis (Figure 1C). In fact, the e and g positions stabilize the core

■

RESULTS AND DISCUSSION Peptide Design and Synthetic Strategy. The peptide scaffold, here referred to as FF03, was designed following the rules of the α-helical coiled-coil peptide architecture.39 Three sets of seven amino acids, defined as heptad repeats and conventionally labeled with the letters a, b, c, d, e, f, and g, constitute the core of the peptide sequence. An extra hydrophobic residue is placed at the C-terminus and half a heptad (d, e, f, and g) at the N-terminus to create “sticky ends”8 that facilitate longitudinal assembly of the fibers along the helix D

DOI: 10.1021/acs.biomac.5b00572 Biomacromolecules XXXX, XXX, XXX−XXX

Article

Biomacromolecules

TEM. TEM analysis subsequent to self-assembly was carried out to determine whether FF03 and its derivatives are in fact able to form α-helical fibers (Figure 4). As a control, Ep01 and

coiled-coil motif and higher oligomeric states, while the b and c positions play a role in bundle formation by enabling association between fibers. In all heptads, the f position, the most solvent-exposed position, is occupied by lysine residues. This amino acid was chosen for its side-chain functionality. The side chain of the lysine, selected to bear the glycopeptide ligand, was protected with a group that can be removed under mild acidic conditions. After the selective deprotection of this particular lysine at position 17, the free amino group is used to build up the peptide ligand, Ep01. After functionality of the Ep01 Nterminus is switched by means of glutaric anhydride, the resulting free carboxylic acid was activated when still on resin, and the sugar moiety coupled via a short amino-linker. This strategy allows for the synthesis of glycopeptide conjugates (shown in Figure 2 and Table 1) entirely on solid phase. Peptide purity was verified with HPLC and mass spectrometry (MS) (Supporting Information, Figure S2 and Table S2). Evaluation of Peptide Structural Features. The peptide structure plays a fundamental role for the realization of the desired multivalent effect. FF03 was evaluated for its ability to retain its structure when preassembly loaded with the different ligands and for its suitability to serve as a stable scaffold for their multivalent presentation. The Ep01structure was also explored because the epitope had not yet been isolated and studied outside the DT context. The secondary structure of all library members was investigated by CD; the presence of aggregates was detected by TEM, and SEC−SLS was used to determine the oligomerization state. CD. Ep01 appears to adopt a combination of secondary structures (Figure 3A). The global minimum between 200 and 205 nm suggests either a combination of random-coil and βsheet conformations, or the presence of β-turns,42,43 while the maximum at 195 nm could indicate the coexistence of α-helical species. The spectrum of Ep01-Man differs from that of the parental peptide only in a shift of the minimum at 205 nm toward shorter wavelengths, whereas Ep01-Man-Gal displays no maximum at 195 nm but an otherwise similar tendency to the other two traces. In all cases, after more than 3 days at concentrations higher than 300 μM, a shift in the minimum from 200 to 215 nm is observed (data not shown) indicating a possible underlying amyloid aggregation process.44 Ep01 was used exclusively at low concentrations within 24 h after sample preparation to avoid aggregation into amyloids. FF03 and its Ep01 or glycosylated Ep01 conjugates show a CD profile typical of an α-helical secondary structure, with minima at 208 and 222 nm and a maximum at 195 nm (Figure 3B). There are, however, subtle differences between unmodified FF03 and the three modified species. The parent peptide FF03 displays a less intense shorter-wavelength minimum, which is likely attributable to helical assemblies formation.8 FF03-Ep01, FF03-Ep01-Man, and FF03-Ep01-Man-Gal display a shorter-wavelength minimum with increased intensity at 208 nm, which results in roughly equal intensities for the two minima at 208 and 222 nm. This difference in CD behavior could be due to either reduced fiber formation in the modified FF03 species or to the presence of Ep01, which itself has a global minimum between 200 and 205 nm. The latter hypothesis is highly likely since it is in accordance with the simulated CD spectrum obtained by combining the separate FF03 and Ep01 curves (Figure S4).

Figure 4. TEM micrographs of FF03 series acquired at concentrations of 120−250 μM 24 h after sample preparation. Samples were dissolved in 100 mM phosphate buffer at pH 7.4 and stained with a 1% PTA solution. (A) FF03; (B) FF03 single fiber; (C) FF03-Ep01; (D) FF03Ep01-Man; (E) FF03-Ep01-Man-Gal.

its glycosylated versions were also analyzed; however, no aggregates were found. The micrographs clearly demonstrate that all peptides from the FF03 series are able to form α-helical fibers; in all cases, individual 2.5 nm thick and >100 nm long fibers were observed alongside large bundles of fibers. An example of an isolated α-helical fiber formed by peptide FF03, the simplest unit of the bundles, is shown in Figure 4, panel B. The fibers formed by FF03 are exclusively linear and nonbranching, and the fiber formation process probably starts immediately after the peptide is dissolved in neutral buffer and is under completion already after 24 h. TEM micrographs acquired 7 days from sample preparation show fibers of comparable type, length, and thickness (see Figure S7). The formation of bundles appears to be a signature of the FF03 series and indicates strong interfiber interactions.45 The isolation of a single fiber is therefore often not feasible. Neither the preassembly coupling of Ep01 nor its glycosylation influences the structure of FF03 enough to abrogate self-assembly or lead to the formation of other types of supramolecular structures. This indicates that the fibers formed by the coiled-coil-based scaffold FF03 are stable enough to ensure the multivalent presentation of both peptide Ep01 and carbohydrate ligands. SEC−SLS. By employing SEC and detection based on light scattering, it was possible to determine the oligomerization state of Ep01at neutral pH in PBS. FF03 was also analyzed with SEC−SLS, but the rapid formation of the fibers does not allow for the isolation and separation of the building blocks (data not shown). E

DOI: 10.1021/acs.biomac.5b00572 Biomacromolecules XXXX, XXX, XXX−XXX

Article

Biomacromolecules

Figure 5. SEC−SLS data showing UV absorbance, RI, and molecular mass. (A) Calculated and experimental molecular weight for each peak detected. (B) Comparison between RI, UV, and MW within the peaks retention time range: black solid line, Abs230 nm; short dashes, RI values; red solid line, MW for peak 1; green line, MW for peak 2; purple line, MW for peak 3. (C) Absorbance recorded at 230 nm over the course experiment. Samples concentration: 100 μm.

with the cross-reacting material 197 (CRM197),32 a nontoxic mutant of DT. Subsequent to primary antibody incubation, the plates were repeatedly washed with NH4SCN as a dissociating agent to reduce nonspecific binding.46 Epitope recognition was determined by means of an antimouse IgG horseradish peroxidase-conjugate. The results of the ELISA for serial dilutions of the peptide are shown in Figure 6. We observed significantly higher absorbance values for FF03Ep01 compared to Ep01 alone, indicating more efficient

Figure 5 shows average curves comparing UV detection (Abs230 nm) with refractive index (RI); the experimentally obtained molecular masses are also shown. The absorbance and refractive index traces reveal the existence of a predominant species of a small oligomerization state (Peak 1, Figure 5B). Another peak is visible shortly after the first (Peak 2, Figure 5B) with a shoulder toward higher molecular weight species (Peak 3, Figure 5B). The retention time for each peak is shown in Figure 5, panel A. According to the RI, a molecular mass value is given for each point within the selected area, and the values in the table (Figure 5A) reflect the average of all experimentally obtained molecular weights. The main peak (Peak 1) is due to a species with an average weight of about 1961 Da, close to the calculated mass for monomeric Ep01. The molecular weight obtained for Peak 2, around 3510 Da, suggests that a small percentage of the peptide dimerizes within the first 24 h after sample preparation. The shoulder of Peak 2, here referred to as Peak 3, indicates the existence of a tetrameric species, although the error in this calculation is high due to the similar elution characteristics of the dimeric and tetrameric complexes. This study is in agreement with the results of the CD experiments described earlier: although Ep01 is mainly present as a monomer in solution, this peptide has a tendency to form higher-order oligomers in a time- and concentration-dependent manner; therefore, in the following sections, we will refer to it as the “monomeric system” and compare it to the “multivalent system”, where Ep01 is covalently linked to the fiber-forming FF03 peptide. Interaction of the Peptide Ligand with Anti-DT Antibodies. An ELISA was used to determine how efficiently the monomeric Ep01 system and the multivalent FF03-Ep01 are bound by polyclonal antibodies against DT. The peptide Ep01 and the conjugate FF03-Ep01were immobilized on microtiter plates and incubated with sera from mice immunized

Figure 6. Antidiphtheria toxin antibody binding to the epitope presented in a monomeric (Ep01, striped columns) or multivalent (FF03-Ep01, solid columns) regime. The figure refers to the assay performed with 1 μM, 0.1 μM, and 0.01 μM Ep01 and FF03-Ep01 and a 1:10 000 serum dilution. The mean AU at 450 nm is plotted against peptide concentration, and these values are given at the top of each column. Asterisks indicate P values (∗∗∗, P < 0.005; ∗, P < 0.05). F

DOI: 10.1021/acs.biomac.5b00572 Biomacromolecules XXXX, XXX, XXX−XXX

Article

Biomacromolecules

Figure 7. CLSM images of the self-assembled scaffold after incubation with FITC-labeled ConA. Green fluorescence associated with FITC indicates efficient binding of ConA to the fiber-forming peptide. (A) FF03-Ep01; (B) FF03-Ep01-Man.

Figure 8. CLSM images of self-assembled scaffold after incubation with FITC-labeled E. coli ORN178. Green fluorescence associated with FITC indicates efficient binding of E. coli to the fiber-forming peptide. (A) FF03; (B) FF03-Ep01-Man.

weak individual intermolecular interactions that are formed between lectins and carbohydrates, these proteins have evolved to detect glycans that are organized in a multivalent fashion and to discriminate among ligands based on density.49 To determine whether a fluorescently labeled lectin (ConAFITC) is able to recognize the mannose moieties presented on the self-assembled FF03 multivalent scaffold, we used CLSM. Green fluorescence of ConA-FITC is clearly observed in the presence of mannosylated fibers (Figure 7B), while the images of the control peptide FF03-Ep01 show negligible fluorescence (Figure 7A). The fluorescence signal clearly defines the shape and organization of the macro-structures, which suggest that the carbohydrate moiety is highly available for lectin binding and that the coiled-coil-based scaffold is an appropriate system for facilitating protein−glycan interactions. Multivalent Presentation of the Carbohydrate Ligand to Native Receptors. Ligand density and orientation are key factors for the success of in vitro diagnostic assays. To test whether our fiber-forming scaffold can compete with the standard carrier proteins currently used in diagnostic assays, we employed CLSM and the FITC-labeled E. coli ORN178, a

antibody binding. Since the same amount of epitope was incubated with the 96-well plate in both cases, the observed difference is likely due to improved antibody avidity and more suitable organization of the epitope when presented on the fiber-forming scaffold. This result is particularly intriguing in the context of conventional assay platforms, which tend to have limitations at low concentrations of the immobilized species. However, it is possible that immobilization efficiency and differential resistance to washing conditions could play a role in the observed differences in AU. Assuming that such effect would be small, antigen platforms of the kind offered by peptide FF03 could be used to enhance the sensitivity of such assays, regardless of whether high local concentration or the regularity of the spatial distribution of the ligand is the key factor. Interaction of the Carbohydrate Ligand with Lectins. ConA-Binding Assay. Aberrations in glycosylation and unique glycan motifs expressed on transformed cells and pathogens are biomarkers for disease states.47 Lectins are naturally occurring carbohydrate-binding proteins that are often employed to probe cell surface changes in vitro.48 Because of the relatively G

DOI: 10.1021/acs.biomac.5b00572 Biomacromolecules XXXX, XXX, XXX−XXX

Article

Biomacromolecules

Figure 9. Anti-β-Gal-(1−4)-α-Man antibody binding to unconjugated (Man-Gal), monomeric (Ep01-Man-Gal), and multivalent antigens (FF03Ep01-Man-Gal) at the oversaturating concentration of 100 μm. (A) Mean fluorescence intensity of DyLight 488 indicating binding of canine anti-βGal-(1 → 4)-α-Man IgG. (B) Mean fluorescence intensity of AlexaFluor 647 indicating binding of human anti-β-Gal-(1 → 4)-α-Man IgG. (C) Mean fluorescence intensity of FITC indicating binding of ConA to 5-amino-pentanyl-α-D-mannopyranoside. Asterisks indicate P values (∗∗∗, P < 0.005; ∗, P < 0.05). Under each column, a representative example of wells corresponding to the specific array is given (duplicates). 0, buffer only; 1, ManGal; 2, Ep01-Man-Gal; 3, FF03-Ep01-Man-Gal; 4, Man; 5, Ep01-Man; 6, FF03-Ep01-Man.

Binding of antileishmanial antibodies, from canine or human sources, to the antigen was detected using fluorescently labeled secondary antibodies, and native CRM197 protein was used as a positive control (example given in Figure S13). In Figure 9, panels A and B, binding of the human and canine polyclonal antibodies to the leishmanial antigen is depicted. When this Leishmania-derived disaccharide is coupled to the Ep01 monomer, antibody recognition is comparable to that of the unconjugated antigen. Binding of the multivalently presented antigen was two- or three-times higher for the canine and human sera, respectively, than in the monomeric system. These results demonstrate the potential of this coiled-coilbased self-assembling peptide as a scaffold for the presentation of peptide and carbohydrate antigens. High density antigen presentation platforms likely improve the sensitivity of numerous bioassays.

strain expressing a certain pattern of lectins that enables specific binding to mannose (Figure 8). Analogous to the results obtained from the ConA-binding assay, the green-fluorescent signal is much more intense and localized when mannose is present on the FF03-Ep01 scaffold, while faint, unspecific binding of the FITC-labeled E. coli can be observed on the mannose-free FF03-Ep01. This finding confirms that the coiledcoil-based self-assembling system efficiently presents a monosaccharide ligand to a cell surface receptor and prompted further investigations where a disaccharide glycotope was tested instead of mannose. Presentation of the Antigenic Disaccharide to Specific Antibodies. The ELISA, ConA, and bacterial-binding experiments described earlier demonstrate that the coiled-coil-based self-assembling peptide is a suitable scaffold for the multivalent presentation of a peptide epitope and a monosaccharide. We evaluated the accessibility of a naturally occurring antigen by means of a microdot assay to further investigate the utility of this system. Compared to the more conventional ELISA, the microdot assay allows for a high-number repetitions of the same conditions from the exact same sample. This reduces the error due to the use of low volumes and gives a broader range of samples for a higher statistical significance. β-Gal-(1 → 4)-αMan disaccharide, the minimal epitope of a lipophosphoglycan found on pathogenic Leishmania parasites, was selected as ligand. This antigen was immobilized on a glass slide in unconjugated (Man-Gal), monomeric (Ep01-Man-Gal), or multivalent (FF03-Ep01-Man-Gal) forms. The concentrations were carefully selected to have subsaturation and oversaturation levels of peptides to react with the NHS ester-activated glass slides. In this manner, the error that could derive from differences in the immobilization degree of the peptides and carbohydrates was minimized. Moreover, we verified the validity of the microdot assay and the success of the immobilization by including in our investigation the binding of ConA-FITC to the mannose presented by Ep01 and FF03Ep01 (Figure 9C).

■

CONCLUSIONS

We describe the design, synthesis, and characterization of a coiled coil-based, fiber-forming peptide. We have shown that it retains its structural properties when loaded prior to assembly with peptide and carbohydrate ligands and demonstrate its suitability as a scaffold for their multivalent presentation. Antibodies produced by mice immunized against Diphtheria toxin recognize the specific peptide epitope more efficiently when it is present as a ligand on the self-assembled scaffold than in its unconjugated form. In addition, the carbohydrate ligand mannose is optimally accessible to lectins that are either in solution or present on a bacterial cell surface. Polyclonal antibodies against the parasite Leishmania can be detected in canine and human sera when a leishmanial carbohydrate antigen is presented as ligand on the fiber-forming peptide scaffold. The novel self-assembling peptide described here offers an attractive platform for presenting antigens at high densities. H

DOI: 10.1021/acs.biomac.5b00572 Biomacromolecules XXXX, XXX, XXX−XXX

Article

Biomacromolecules

■

(10) Luo, J.; Tong, Y. W. ACS Nano 2011, 5, 7739−7747. (11) Baldauf, C.; Pagel, K.; Warnke, S.; von Helden, G.; Koksch, B.; Blum, V.; Scheffler, M. Chem.Eur. J. 2013, 19, 11224−11234. (12) Wagner, S. C.; Roskamp, M.; Pallerla, M.; Araghi, R. R.; Schlecht, S.; Koksch, B. Small 2010, 6, 1321−1328. (13) Pagel, K.; Wagner, S. C.; Samedov, K.; Berlepsch, H. V.; Böttcher, C.; Koksch, B. J. Am. Chem. Soc. 2006, 128, 2196−2197. (14) Pagel, K.; Vagt, T.; Kohajda, T.; Koksch, B. Org. Biomol. Chem. 2005, 3, 2500−2502. (15) Fasting, C.; Schalley, C. A.; Weber, M.; Seitz, O.; Hecht, S.; Koksch, B.; Dernedde, J.; Graf, C.; Knapp, E. W.; Haag, R. Angew. Chem., Int. Ed. 2012, 51, 10472−10498. (16) Artner, L. M.; Merkel, L.; Bohlke, N.; Beceren-Braun, F.; Weise, C.; Dernedde, J.; Budisa, N.; Hackenberger, C. P. R. Chem. Commun. 2012, 48, 522−524. (17) Lutolf, M. P.; Hubbell, J. A. Nat. Biotechnol. 2005, 1, 47−55. (18) Silva, G. A.; Czeisler, C.; Niece, K. L.; Beniash, E.; Harrington, D. A.; Kessler, J. A.; Stupp, S. I. Science 2004, 303, 1352−1355. (19) Schneider, A.; Garlick, J. A.; Egles, C. PLoS One 2008, 3, e1410. (20) Xu, J.; Chen, S.; Ahmed, S. H.; Chen, H.; Ku, G.; Goldberg, M. P.; Hsu, C. Y. J. Neurosci. 2001, 21, RC118. (21) Falenski, J. A.; Gerling, U. I. M.; Koksch, B. Bioorg. Med. Chem. 2010, 18, 3703−3706. (22) Pandya, M. J.; Spooner, G. M.; Sunde, M.; Thorpe, J. R.; Rodger, a.; Woolfson, D. N. Biochemistry 2000, 39, 8728−8734. (23) Zanetta, J. P.; Kuchler, S.; Lehmann, S.; Badache, A.; Maschke, S.; Didier, T.; Dufourcq, P.; Vincendon, G. Histochem. J. 1992, 24, 791−804. (24) Kieda, C. Adv. Exp. Med. Biol. 1998, 435, 75−82. (25) Cao, Z.; Partyka, K.; McDonald, M.; Brouhard, E.; Hincapie, M.; Brand, R. E.; Hancock, W. S.; Haab, B. B. Anal. Chem. 2013, 85, 1689−1698. (26) Gomez-Garcis, M.; Benito, J. M.; Rodríguez-Lucena, D.; Yu, J. X.; Chmurski, K.; Ortiz Mellet, C.; Gutiérrez Gallego, R.; Maestre, A.; Defaye, J.; García Fernández, J. M. J. Am. Chem. Soc. 2005, 127, 7970− 7971. (27) Mogensen, T. H. Clin. Microbiol. Rev. 2009, 22, 240−273. (28) Bachmann, M. F.; Zinkernagel, R. M. Annu. Rev. Immunol. 1997, 15, 235−270. (29) Xu, X. D.; Cheng, H.; Chen, W. H.; Cheng, S. X.; Zhuo, R. X.; Zhang, X. Z. Sci. Rep. 2013, 3, 2679. (30) Comber, J. D.; Karabudak, A.; Shetty, V.; Testa, J. S.; Huang, X.; Philip, R. Hepatitis Res. Treat. 2014, No. 860562. (31) Roytman, R.; Adler-Abramovich, L.; Kumar, K. S. A.; Kuan, T.C.; Lin, C.-C.; Gazit, E.; Brik, A. Org. Biomol. Chem. 2011, 9, 5755− 5761. (32) McCool, T. L.; Harding, C. V.; Greenspan, N. S.; Schreiber, J. R. Infect. Immun. 1999, 67, 4862−4869. (33) Adamo, R.; Nilo, A.; Castagner, B.; Boutureira, O.; Berti, F.; Bernardes, G. J. L. Chem. Sci. 2013, 4, 2995−3008. (34) Houseman, B. T.; Mrksich, M. Chem. Biol. 2002, 9, 443−454. (35) Harris, S. L.; Spears, P. A.; Havell, E. A.; Hamrick, T. S.; Horton, J. R.; Orndorff, P. E. J. Bacteriol. 2001, 183, 4099−4102. (36) Chakkumkal, A.; Martin, C. E.; Wahlbrink, A.; Bogdan, C.; Ntais, P.; Antoniou, M.; Seeberger, P. H. ACS Chem. Biol. 2013, 8, 2412−2422. (37) Li, D.; Elbert, D. L. J. Pept. Res. 2002, 60, 300−303. (38) Pace, C. N.; Vajdos, F.; Fee, L.; Grimsley, G.; Gray, T. Protein Sci. 1995, 4, 2411−2423. (39) Mason, J. M.; Arndt, K. M. ChemBioChem 2004, 5, 170−176. (40) Papapostolou, D.; Smith, A. M.; Atkins, E. D. T.; Oliver, S. J.; Ryadnov, M. G.; Serpell, L. C.; Woolfson, D. N. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 10853−10858. (41) Harbury, P. B.; Zhang, T.; Kim, P. S.; Alber, T. Science 1993, 262, 1401−1407. (42) Tatham, A. S.; Miflin, B. J.; Shewry, P. R. Cereal Chem. 1985, 62, 405−412. (43) Guijarro, J. I.; Sunde, M.; Jones, J. a.; Campbell, I. D.; Dobson, C. M. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 4224−4228.

ASSOCIATED CONTENT

S Supporting Information *

Additional experimental data are reported. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biomac.5b00572.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Author Contributions

B.K., P.H.S. and C.A. initiated this study. E.Z. performed chemical synthesis of peptides and glycoconjugates 1-7 (Ep01, Ep01-Man, Ep01-Man-Gal, FF03, FF03-Ep01, FF03-Ep01Man, FF03-Ep01-Man-Gal) and biochemical experiments with assistance from C.A. C.M. synthesized the carbohdrates 5amino-pentanyl-α-D-mannopyranoside and 5-aminopentanyl βD-galactopyranosyl-(1 → 2)-α-D-mannopyranoside. H.v.B. performed TEM experiments. E.B. contributed to the original design of the scaffold FF03. P.H.S., C.A., B.K. and E.Z. wrote the manuscript with assistance from C.M. and H.v.B. All authors read and approved the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the Collaborative Research Center 765 (SFB765/22014), the Deutsche Forschungsgemeinschaft (FG806HA2686/3-2), the Bundesministerium fü r Bildung und Forschung (Grant No. 0315447), and the German Federal Ministry of Education and Research as well as the Max-Planck Society for generous financial support. Furthermore, we would like to acknowledge the assistance of the Core Facility BioSupraMol supported by the DFG. Our thanks go also to P. Orndorff (North Carolina State University, USA) and M. Opitz (Ludwig Maximilians University, Munich, Germany) for providing the mannose-binding E. coli strain ORN178.

■

REFERENCES

(1) Bernardi, A.; Jiménez-Barbero, J.; Casnati, A.; De Castro, C.; Darbre, T.; Fieschi, F.; Finne, J.; Funken, H.; Jaeger, K. E.; Lahmann, M.; Lindhorst, T. K.; Marradi, M.; Messner, P.; Molinaro, A.; Murphy, P. V.; Nativi, C.; Oscarson, S.; Penadés, S.; Peri, F.; Pieters, R. J.; Renaudet, O.; Reymond, J. L.; Richichi, B.; Rojo, J.; Sansone, F.; Schäffer, C.; Turnbull, W. B.; Velasco-Torrijos, T.; Vidal, S.; Vincent, S.; Wennekes, T.; Zuilhof, H.; Imberty, A. Chem. Soc. Rev. 2013, 42, 4709−4727. (2) Nilsson, P.; O’Meara, D.; Edebratt, F.; Persson, B.; Uhlén, M.; Lundeberg, J.; Nygren, P. Anal. Biochem. 1999, 269, 155−161. (3) Astakhova, I. K.; Hansen, L. H.; Vester, B.; Wengel, J. Org. Biomol. Chem. 2013, 11, 4240−4249. (4) Stadler, L. K. J.; Tomlinson, D. C.; Lee, T.; Knowles, M. A.; Ko Ferrigno, P. Cell Death Dis. 2014, 5, e1037. (5) Hartgerink, J. D.; Granja, J. R.; Milligan, R. A.; Ghadiri, M. R. J. Am. Chem. Soc. 1996, 118, 43−45. (6) Madine, J.; Davies, H. A.; Shaw, C.; Hamley, I. W.; Middleton, D. A. Chem. Commun. 2012, 48, 2976−2978. (7) Boato, F.; Thomas, R. M.; Ghasparian, A.; Freund-Renard, A.; Moehle, K.; Robinson, J. A. Angew. Chem., Int. Ed. 2007, 46, 9015− 9018. (8) Bromley, E. H.; Channon, K. J.; King, P. J.; Mahmoud, Z. N.; Banwell, E. F.; Butler, M. F.; Crump, M. P.; Dafforn, T. R.; Hicks, M. R.; Hirst, J. D.; Rodger, A.; Woolfson, D. N. Biophys. J. 2010, 98, 1668−1676. (9) Dobson, C. M. Trends Biochem. Sci. 1999, 24, 329−332. I

DOI: 10.1021/acs.biomac.5b00572 Biomacromolecules XXXX, XXX, XXX−XXX

Article

Biomacromolecules (44) McPhie, P. Biopolymers 2004, 75, 140−147. (45) Hodges, R. S. Biochem. Cell Biol. 1996, 74, 133−154. (46) Almanzar, G.; Ottensmeier, B.; Liese, J.; Prelog, M. J. Immunol. Methods 2013, 387, 36−42. (47) Chen-Shiou, W.; Chia-Jui, Y.; Ruey-Hwang, C.; Shiou-Ting, L.; Wei-Chien, H.; Chien-Tai, R.; Chung-Yi, W.; Yung-Luen, Y. PLoS One 2012, 7, e39466. (48) Strauchen, J. A. Am. J. Hematol. 1982, 12, 227−232. (49) Dam, T. K.; Brewer, C. F. Glycobiology 2010, 3, 270−279.

J

DOI: 10.1021/acs.biomac.5b00572 Biomacromolecules XXXX, XXX, XXX−XXX