Alanine-rich amphiphilic peptide containing the RGD cell adhesion motif: a coating material for human fibroblast attachment and culture

Abstract

We studied the self-assembly of peptide A₆RGD (A: alanine, R: arginine, G: glycine, D: aspartic acid) in water, and the use of A₆RGD substrates as coatings to promote the attachment of human cornea stromal fibroblasts (hCSFs). The self-assembled motif of A₆RGD was shown to depend on the peptide concentration in water, where both vesicle and fibril formation were observed. Oligomers were detected for 0.7 wt% A₆RGD, which evolved into short peptide fibres at 1.0 wt% A₆RGD, while a co-existence of vesicles and long peptide fibres was revealed for 2–15 wt% A₆RGD. A₆RGD vesicle walls were shown to have a multilayer structure built out of highly interdigitated A₆ units, while A₆RGD fibres were based on β-sheet assemblies. Changes in the self-assembly motif with concentration were reflected in the cell culture assay results. Films dried from 0.1–1.0 wt% A₆RGD solutions allowed hCSFs to attach and significantly enhanced cell proliferation relative to the control. In contrast, films dried from 2.5 wt% A₆RGD solutions were toxic to hCSFs.

---

Introduction

Surfactant-like peptides (SLPs)^1–3 comprise a headgroup, consisting of a short sequence of charged peptide residues, attached to a tailgroup of neutral amino acids. For bio-functional purposes, the headgroup is usually designed from hydrophilic biologically-active epitopes, while a hydrophobic tail group drives peptide self-assembly in aqueous medium. The bio-functionality and bioactivity afforded by sequences of amino acids, together with their amphiphilicity, lead to the remarkable potential of SLPs in the development of novel biomaterials.⁴

Zhang and coworkers pioneered the study of amphiphilic peptides including those with poly-alanine sequences such as A₆ ^5–8 (A: alanine). The self-assembly of peptides such as A₆K (K: lysine) with a cationic lysine headgroup has been studied by several groups.^9–16 In a previous work we described the self-assembly of the amphiphilic peptide A₆R (R: arginine) into ultrathin free-floating nanosheets in dilute aqueous solution.¹⁷ The sheets comprise opposed dimers of A₆R molecules that are closely-packed but not hydrogen-bonded into β-sheets. We showed that A₆R self-assembly is driven by the amphiphilic sequence design of the peptide with the unique conformational and electrostatic properties of the cationic R headgroup. At high concentration, A₆R forms helically wrapped ribbons coexisting with nanotubes, and β-sheet formation occurs concomitantly.¹⁷

Here, we functionalise the A₆ sequence by attaching the RGD (G: glycine, D: aspartic acid) tripeptide. More than three decades ago the peptide epitope RGD was first shown to promote cell adhesion¹⁸ and it is now widely used as a model cell adhesion domain. Direct interactions occur between the RGD motif and integrins, namely those composed of the α5, α8, or αV subunits. Integrins are a family of α,β-heterodimeric transmembrane receptors that interlace transmembrane ties between the extracellular matrix or cell surface molecules and the cytoskeleton,¹⁹ thus triggering intracellular signalling pathways involved in the regulation of mitosis and cell proliferation.^20,21 Several SLPs have been functionalized with a RGD sequence over the past three decades.^22–30 The great potential in the development of novel biomaterials, presented by a broad range of RGD-SLPs, has been widely explored in academic research.^{3,19–21,31–35}

In this work we studied the self-assembly of A₆RGD in water and its use as films to create coated surfaces for cell attachment and growth. We first used a combination of spectroscopic, microscopic, and X-ray scattering techniques to investigate the self-assembly of A₆RGD in water. We then evaluated the biocompatibility of A₆RGD, as well as its ability to promote adhesion of human cornea stromal fibroblasts (hCSFs) when used as a dry film coating.

Experimental

Materials

Peptides A₆RGD, C₁₆-ETTES (E: glutamic acid, T: threonine, S: serine) and cRGD (cyclo [RGD(f)-NMe-V], V: valine, f: D-phenylalanine, NMe: methylated amide group) were custom-synthesized by CS Bio (Menlo Park, USA) as TFA salts. The purity of the peptides was determined by analytical HPLC in a TFA-water–acetonitrile gradient, while the molecular weight (M_W) was obtained by electrospray-mass spectrometry. A₆RGD purity was 99.15%, with analyzed M_W = 771.9 Da (expected M_W = 772.83 Da). In this work, weighed amounts of A₆RGD were dissolved in ultrapure water from a Barnstead Nanopure system to the desired concentration. C₁₆-ETTES and cRGD were used for cell culture assays. Purity was 96.51% and 97.58% for C₁₆-ETTES and cRGD, respectively, with corresponding M_W = 803.4 and 588.15 Da (expected M_W = 803.94 and 588.66 Da, respectively).

Thioflavin (ThT) and pyrene (Pyr) fluorescence spectroscopy. Fluorescence spectra were recorded with a Varian Cary Eclipse Fluorescence Spectrometer with samples in 4 mm inner width quartz cuvettes. For the ThT assay, the spectra were recorded from 460 to 600 nm using an excitation wavelength λ_ex = 440 nm, and the peptide was dissolved in a 5.0 × 10⁻³ wt% ThT solution. For the Pyr assay, the spectra were recorded from 360 to 550 nm using an excitation wavelength λ_ex = 339 nm. Pyr assays were performed using a 1.2 × 10⁻⁵ wt% Pyr solution as a diluent.

Circular dichroism (CD). Spectra were recorded using a Chirascan spectropolarimeter (Applied Photophysics, UK). The sample was placed in a cover slip cuvette (0.01 mm thick). Spectra are presented with absorbance A < 2 at any measured point with a 0.5 nm step, 1 nm bandwidth, and 1 second collection time per step at 20 °C.

Fourier transform infra-red (FTIR) spectroscopy. Spectra were measured on a Nicolet Nexus spectrometer with DTGS detector. FTIR data was measured for SLP solutions in D₂O. Samples were sandwiched between two CaF₂ plate windows (spacer 0.0125 mm). Spectra were scanned 128 times over the range of 900–4000 cm⁻¹.

UV-vis absorption. Spectra were recorded using a Varian Cary 300 Bio UV/Vis spectrometer. Samples were analyzed in quartz cuvettes with a 5.0 mm path length and were baseline corrected with respect to a blank cell with the appropriate solvent. For the Congo red assay, the spectra were recorded from 300 to 800 nm using a 2 × 10⁻⁴ wt% Congo red solution as a diluent.

Small-angle X-ray scattering (SAXS). Experiments were performed on beamline ID02 at the ESRF (Grenoble, France). Samples were placed in a glass capillary mounted in a brass block for temperature control. Micropumping was used to minimise beam damage by displacing a drop of the sample by 0.01–0.1 mm for each exposure. The sample-to-detector distance was 1 m, and the X-ray energy was 12.46 keV. The q = 4π sin θ/λ range was calibrated using silver behenate. Data processing (background subtraction, radial averaging) was performed using the software SAXSUtilities.

SAXS theory. The SAXS intensity from a dilute system of disordered particles is dominated by the particle form factor. In our model, the form factor was fitted to a model of Gaussian bilayers co-existing with long cylinders. The details of the bilayer model are given elsewhere.³⁶ The model assumes an electron density profile (Fig. 5a) comprising one Gaussian function for each headgroup on either side of the bilayer electron density (ρ_H), and one Gaussian function for the chains in the core of the bilayer electron density (ρ_C). The density ρ_H has peaks centred at z_H and −z_H with width σ_H. The density ρ_C has a peak centred at z_C with width σ_C. Our model assumes that the bilayer is centred at z = z_C = 0. The model for a long cylinder assumed an electron density ρ_cyl, a cylinder radius R and a cylinder length L (Fig. 5b). We used a Gaussian distribution of z_H and R, with associated degrees of polydispersity Δ_2z^H and Δ_R, respectively. The background was fitted according to the Porod law³⁷C₁ + (C₂/q^C3). The fitting parameters of the model are the total bilayer thickness l_T = 2(σ_H + z_H) ± Δ_2z^H, ρ_H, ρ_C, σ_C, ρ_cyl, R, Δ_R, C₁, C₂ and C₃.

Cryo-transmission electron microscopy (cryo-TEM). Experiments were carried out using a field emission cryo-electron microscope (JEOL JEM-3200FSC). Images were taken using bright-field mode and zero-loss energy filtering (omega type) with a slit width 20 eV. Micrographs were recorded using a Gatan Ultrascan 4000 CCD camera. Vitrified specimens were prepared using an automated FEI Vitrobot device using Quantifoil 3.5/1 holey carbon copper grids with 3.5 μm hole sizes. Grids were cleaned using a Gatan Solarus 9500 plasma cleaner just prior to use and then transferred into an environmental chamber of FEI Vitrobot at room temperature and 100% humidity. Thereafter, 3 μl of solution was applied on the grid, blotted once for 1 s and then vitrified in a 1/1 mixture of liquid ethane and propane at −180 °C. Grids with vitrified sample solutions were maintained in a liquid nitrogen atmosphere and then cryo-transferred into the microscope. The cryo-electron microscope was operated at 300 kV for 0.1, 1 and 2.5 wt% A₆RGD and at 200 kV for 15 wt% A₆RGD. A₆RGD solutions were maintained at −187 °C during the imaging. 0.1, 1.0 and 2.5 wt% A₆RGD solutions were heated from −187 °C to −60 °C at ∼10⁻⁵ Pa, before being imaged at −187 °C. The heating process from −187 to −60 °C, equivalent to a freeze drying process in the microscope, allowed for the sublimation of the ice from the sample and removed the vitrified water. Therefore, 0.1, 1 and 2.5 wt% A₆RGD are not classical cryo-TEM samples with a vitrified ice layer, but water is removed and only the self assembled A₆RGD is present.

X-ray diffraction (XRD). X-ray diffraction was performed on stalks prepared by suspending a drop of the peptide solutions between the ends of wax-coated capillaries and allowing them to dry. The stalk was mounted vertically onto the four axis goniometer of a RAXIS IV++ X-ray diffractometer (Rigaku) equipped with a rotating anode generator. The XRD data was collected using a Saturn 992 CCD camera. The same setup was used to measure XRD from a solution containing 12 wt% A₆RGD which was prodded into a borosilicate capillary with D = 1 mm internal diameter and 0.01 mm wall thickness. The sample-to-detector distance was 50 mm.

Native polyacrylamide gel electrophoresis. The products of A₆RGD aggregation were analyzed using non-reducing, non-denaturing (SDS-free) polyacrylamide gel electrophoresis (PAGE). This method is often used to separate complex-forming molecules, or when the native conformation needs to be maintained. Due to the density of the polyacrylamide gel and the size of A₆RGD (<1 kDa), this technique was only suitable to resolve large SLP aggregates. A₆RGD solutions at 0.1, 1.0, and 2.5 wt% were sampled for a total amount of 50 μg of peptide (50, 5, and 2 μl aliquots, respectively), and injected directly into an 8% native Tris-glycine gel. The ColorBurst molecular ladder (Sigma) was used as molecular weight. Gels were stained with Coomassie G250, and then washed in ddH₂O–acetic acid–methanol solution (7 : 1 : 2) for a day before imaging.

Biocompatibility and cell adhesion assays. A₆RGD solutions at 0.01–2.5 wt% were applied as 50 μl drops onto tissue culture plates and allowed to evaporate overnight at room temperature to obtain a dry film. Films were then washed three times with sterile phosphate buffer saline (PBS). To evaluate the bioactivity and compatibility of A₆RGD, human cornea stromal fibroblasts (hCSFs) isolated from corneal rings were seeded at 7.5 × 10³ cells per cm² onto the various peptide films coating polystyrene tissue culture plates (Nunc, Thermo Scientific, USA). To evaluate the specificity of cell adhesion to the RGD motif of the A₆RGD peptide, 90%-confluent hCSFs were passaged by trituration using PBS to avoid cleavage of membrane proteins, incubated for 15 min at 37 °C in 0.1 ml culture medium containing 5 nM of integrin blockers, cRGD or anti-integrin αV antibody (ab76609, Abcam) or corresponding mock treatments, C₁₆-ETTES or goat anti-mouse IgG-HRP antibody (R&D Systems). Treated cells were seeded (1.5 × 10⁴ cells per cm²) onto ultra-low attachment plates (Costar, Corning, USA) previously coated with 1 wt% A₆RGD. Cell quantification was performed each day for 5 days (proliferation) or 24 h post-seeding (specificity of adhesion) using the Alamar Blue assay. Briefly, cells were incubated with 1 : 10 resazurin reagent in culture media for 4 h, after which samples were taken and culture media replaced. Samples were analysed by spectrofluorometry (λ_ex = 540 nm, λ_em = 590 nm), and fluorescence signals plotted to give a fluorescence vs. cell number calibration curve. For each assay, data statistics from three independent replicates was performed using two-way analysis of variance (ANOVA) with Bonferroni's post-hoc test.

Results and discussion

A₆RGD consists of a hydrophobic A₆ block and the cell adhesion peptide motif RGD, and was designed to simultaneously ensure solubility in water and specific binding to cells. The good solubility of A₆RGD in water was quantified by an octanol–water partition coefficient log P = −5.757 calculated using the software Molinspiration.³⁸

The isoelectric point of A₆RGD in water was calculated as pH 6.78 using Innovagen software.³⁹ ESI Fig. 1† shows the measured pH dependence with concentration for the samples studied in this work, together with the chemical structure of A₆RGD. Our study focused on solutions with pH below the isoelectric point, and therefore the SLP was expected to be positively charged in solution due to the arginine residue.

Fig. 1 shows ThT and Pyr fluorescence emission assays performed to determine the critical concentration for the formation of A₆RGD aggregates. The emission fluorescence spectra (λ_ex = 440 nm) for A₆RGD solutions diluted in ThT at 5.0 × 10⁻³ wt% showed a broad peak centred at ∼486 nm (Fig. 1a, inset). The dependence of I₄₈₆ on A₆RGD concentration is shown in Fig. 1a. ThT fluorescence showed a dramatic increase of I₄₈₆ emission band intensity upon amyloid formation.^40,41 In agreement with that information, the critical aggregation concentration (c.a.c.) was measured as 2.1 wt% A₆RGD from the inflection point of the data in Fig. 1a.

Fig. 1 Determination of c.a.c., from (a) ThT and (b) Pyr fluorescence assays. The insets show representative fluorescence emission spectrum for (a) ThT and (b) Pyr.

A Pyr fluorescence assay was performed to provide additional information about the c.a.c. The emission fluorescence spectra (λ_ex = 339 nm) for samples containing A₆RGD diluted in Pyr at 1.3 × 10⁻⁵ wt% displayed a band centred at ∼374 nm (Fig. 1b, inset). The dependence of the intensity I₃₇₄ with the peptide concentration is shown in Fig. 1b. The inflection point for the data in Fig. 1b denoted a change of environment for the Pyr molecule, and was identified as the c.a.c. Indeed, Fig. 1b shows that, for concentrations higher than 1.4 wt% A₆RGD, the Pyr molecule was surrounded by a hydrophobic environment corresponding to the core of the peptide aggregate. The averaged value for A₆RGD c.a.c. = 1.8 ± 0.5 wt% was calculated from the c.a.c. values reported in Fig. 1 using ThT and Pyr assays.

We further investigated the amyloid-like properties of A₆RGD aggregates by measuring the binding to another amyloid dye, Congo red. Congo red alone had an absorbance maximum at 498 nm (Fig. 2). Upon increasing the A₆RGD concentration in the Congo red solution, the absorbance maximum progressively red-shifted, reaching a stable value of ∼549 nm for 0.07–2.0 wt% A₆RGD (Fig. 2). This result indicated that, for concentrations below the c.a.c. and as low as 0.07 wt%, A₆RGD formed peptide oligomers.

Fig. 2 Representative results for the absorption spectrum of Congo red in the presence of 0, 0.01 and 2 wt% A₆RGD. Inset: dependence of the Congo red absorption peak position as a function of the A₆RGD concentration.

The products of A₆RGD aggregation were analyzed using native polyacrylamide gel electrophoresis (PAGE). Fig. 3 shows the results obtained for A₆RGD solutions at 0.1, 1.0 and 2.5 wt% run by PAGE in non-reducing, non-denaturing conditions. During the run, a large smear was observed in the front of migration in lanes corresponding to 1.0 and 2.5 wt%, but not for 0.1 wt% A₆RGD. After de-staining, a diffuse group of bands corresponding to high molecular weight (>80 kDa) species were resolved for A₆RGD at 1.0 and 2.5 wt% (Fig. 3). In particular, 2.5 wt% A₆RGD migrated as ∼250 kDa species, whereas bands from 1.0 wt% were resolved within the 80–250 kDa regions. In contrast, no bands were observed for 0.1 wt% A₆RGD. Although A₆RGD aggregates of smaller sizes (20–60 kDa) were probably responsible for the smears occurring during electrophoresis, no bands were observed at this molecular weight range after de-staining. A₆RGD at 0.1 wt% did not form stained aggregates in the resolution range of the polyacrylamide gel (>20 kDa) (Fig. 3).

Fig. 3 Non-reducing native PAGE of A₆RGD.

Taken together, the results shown in Fig. 2 and 3 were consistent, although obtained using very different analytical methods. The self-assembly of A₆RGD into large fibrillar aggregates takes place at ∼1 wt% of A₆RGD, while lower concentrations in the 0.07–1.0 wt% range are dominated by oligomeric species.

CD was used to access the secondary structure of A₆RGD in solution below and above the c.a.c. CD spectra measured for 0.5 and 5.0 wt% A₆RGD, as displayed in ESI Fig. 2.† The CD spectra suggest contributions from several secondary structures. The spectrum for 0.5 wt% A₆RGD was characterized by a negative band at ∼194 nm and a weak positive band at ∼217 nm (ESI Fig. 2†), characteristic of a predominant polyproline II conformation.⁴² However, the spectrum for 5 wt% A₆RGD showed a much shallower minimum and maximum, which indicates the presence of mainly disordered conformation.⁴³

FTIR experiments were also performed to probe the secondary structure of A₆RGD in solution. The amide I region of the FTIR spectra measured for A₆RGD at 0.5–15.5 wt% is displayed in ESI Fig. 3.† The spectra contain two main bands at 1672 and 1649 cm⁻¹. The band at 1672 cm⁻¹ is due to the TFA dissolved in the solution,^44,45 while the contribution of the band centred at 1649 cm⁻¹ indicate a disordered component⁴⁶ in the A₆RGD structure, in good agreement with the CD data in ESI Fig. 1.†

Fibre X-ray diffraction was performed to investigate the self-assembled morphology at the level of the β-sheet structure. The one-dimensional integration of the 2D XRD spectra was performed in order to quantify peak positions. ESI Fig. 4† shows the profile obtained for a stalk dried from a 10 wt% A₆RGD sample, and for a 12 wt% A₆RGD solution. The wide peak with spacing d = 4.7 Å, measured for the stalk indicated a distribution of β-sheet spacing.^47,48 The XRD pattern for 12 wt% A₆RGD (ESI Fig. 4†) presented eight spacings d = 27.7, 13.2, 11.1, 9.3, 7.2, 5.5, 4.8, and 4.5 Å. Possibly, d = 27.7, 13.2, and 9.3 Å corresponded to the first, second and third order of a lamellar structure with cell parameter d_o = 27.7 Å. The spacings d = 4.8 and 4.5 Å from A₆RGD at 12 wt% were assigned to a β-sheet spacing and the spacing of alanine residues in non β-sheet structures, respectively.^17,49 The latter is consistent with the substantial disordered component observed by FTIR (ESI Fig. 3†). In addition, and based on the indexation of similar XRD patterns previously reported by us for A₆K, A₆R, and A₁₂R₂ in water,^13,14,17,49 the d = 5.5 Å spacing found for 10 wt% A₆RGD was assigned to the stacking distance between β-sheets. The relatively small β-sheet stacking distance is due to the efficient packing of alanine residues⁵⁰ and is consistent with a compact structure as observed for other alanine-rich peptides.⁵⁰ The XRD from the stalk presented a lower number of reflections than that from A₆RGD in solution (ESI Fig. 4†), probably due to a loss of order caused by the drying process.

The self-assembled structures of A₆RGD at 0.1, 1.0, 2.5, and 15 wt% were further investigated by cryo-TEM (Fig. 4). Cryo-TEM images for 0.1 wt% A₆RGD did not show the formation of aggregates (results not shown). Indeed, 0.1 wt% A₆RGD presented oligomeric species in solution, which might be extremely difficult to visualize using cryo-TEM. The images in Fig. 4 suggest that the self-assembled structure was dependent on concentration. Images obtained for 1.0 wt% A₆RGD (Fig. 4a) showed the formation of short fibrillar structures. At higher concentrations, a network of fibres ∼11 ± 3 nm thick was observed for 2.5 wt% A₆RGD (Fig. 4b). Cryo-TEM images for 15 wt% A₆RGD (Fig. 4c,d) indicated the co-existence of structures consisting of fibrils and vesicles with variable sizes. The peptide fibrils were 14 ± 3 nm thick (Fig. 4c). On the other hand, A₆RGD vesicles had very thin walls, and were highly polydisperse in size, with an average diameter of 135 ± 62 nm (Fig. 4d). According to these results, A₆RGD formed fibres at lower concentrations, while vesicles co-exist with fibrils at higher concentrations.

Fig. 4 Cryo-TEM images measured for (a) 1, (b) 2.5 and (c–d) 15 wt% A₆RGD. (c) shows the co-existence of fibrils with vesicles for 15 wt% peptide, while (d) corresponds to a region in the 15 wt% A₆RGD grid, where only A₆RGD vesicles were present.

The concentration-dependent morphology of the peptide aggregates is probably due to different mechanisms of aggregation. Changes in the concentration, correlated to the number of peptide clusters seeding the aggregation process, determine the mechanism governing fibrillization and the final morphology of the peptide aggregates. SAXS curves measured for 2, 5, and 10 wt% A₆RGD (Fig. 5) were used to identify the self-assembly motifs of the A₆RGD in water.

Fig. 5 Models for the (a) Gaussian bilayer and (b) long cylinder fitted to the SAXS curves. (c) SAXS data for 2, 5 and 10 wt% A₆RGD, fitted according to a co-existence of Gaussian bilayers and long cylinders. (d) Representative bimodal distribution for 10 wt% A₆RGD obtained from the fits shown in (c), corresponding to the radius of the cylinder in (b) and the thickness of the bilayers in (a). I₁ and I₂ in (d) correspond to the bilayer and cylinder contributions to the form factor. The inset in (d) displays the dependence of I₂/I₁ with concentration.

The SAXS data were fitted to a co-existence of long cylinders and a bilayer structure, using the software SASfit.⁵¹ The fitted SAXS curves are displayed in Fig. 5c. The parameters obtained from the SAXS model, already described in the Experimental section, are listed in ESI Table 1.† The length of the cylinder was fixed to 10⁶ Å for all the fits and it behaved as a scale factor because R ≫ L in the form factor.

Fig. 5d displays the bimodal Gaussian distribution function weighted by R³ (R = intra-particular distance) used to model the SAXS data for 10 wt% A₆RGD in Fig. 5c. I₁ and I₂ in Fig. 5d corresponds to the intensities for the distributions in bilayer thickness and cylinder radius, respectively. The inset in Fig. 5d shows the values obtained for I₂/I₁ as a function of A₆RGD concentration. According to the inset in Fig. 5d, the fraction of fibres relative to bilayers decreases with increasing A₆RGD concentration.

On the basis of the poor water solubility of the A₆ block, we expect that the Gaussian bilayers and long cylinders will be built from A₆RGD bilayers, with highly interdigitated hydrophobic A₆ ‘tails’ and positively charged RGD groups exposed to water. Similarly, the cylinders will comprise an A₆ core and RGD corona. A similar bilayer structure has been proposed by us for free-floating sheets of A₆R¹⁷ and A₁₂R₂ nanotapes⁴⁹ in aqueous solution. This idea was confirmed by the calculated layer spacing listen in ESI Table 1,†l_T = 37 Å ≪ 2 × l_E (l_E: length of the A₆RGD molecule in a β-sheet conformation = 9 × 3.4 Å = 30.6 Å; 3.4 Å = distance between residues comprising the β-sheet⁵²), which corresponds to bilayers of interdigitated A₆ blocks.

The multilayer cell parameter d_o = 27.7 Å measured from the XRD data (ESI Fig. 4†) is within the experimental error of the bilayer spacing l_T (ESI Table 1†). We propose that the multilayer order deduced from the XRD data for the A₆RGD solution (ESI Fig. 4†) is associated to the ordering at the wall of the SLP vesicles. According to these results, the lamellar order was lost after the drying process and only the β-sheet order prevailed for the 10 wt% A₆RGD stalk.

Cryo-TEM images for a 2.5 wt% A₆RGD solution did not show the formation of vesicles, probably because the population of bilayers was ∼88 times lower than the population of fibres (Fig. 5d, inset).

To test the use of A₆RGD in cell culture, solutions of peptide at 0.01–2.5 wt% were drop-spotted on polystyrene cell culture plates and allowed to dry, forming thin films (Fig. 6a) that were then tested for their properties as biocompatible promoters of cell adhesion and proliferation. Human cornea stromal fibroblasts (hCSFs) seeded onto surfaces produced by 0.01–1.0 wt% A₆RGD adhered strongly to the surface and maintained high viability of cultured cells. Cells growing on 1 wt% A₆RGD films acquired a fusiform morphology with a substantial number of process extensions (Fig. 6b), which constituted an indication of viable cells with appropriate phenotype.

Fig. 6 Adhesion and morphology of human cornea stromal fibroblasts (hCSFs) cultured on A₆RGD coatings. A₆RGD in solution at 0.01, 0.1, 1 and (a) 2.5 wt% were drop-spotted and dried overnight to create dense films (f; inset) that stably-coated the underlying cell culture polystyrene surface (u). Cells seeded and grown for three days on 1.0 wt% A₆RGD films (b) adhered to the coating, whereas those seeded onto 2.5 wt% A₆RGD films (c) failed to attach, and assumed a round morphology. Scale bars = 100 μm (inset = 50 μm).

In addition, cells growing on 1.0 wt% A₆RGD films achieved a significantly (p < 0.001) higher cell density when compared to uncoated controls, corresponding to a 1.2-fold increase in cell number at day 5 (Fig. 7). This is also equivalent to values recently obtained with the same cells on fibronectin-coated surfaces routinely used in tissue culture.⁵³

Fig. 7 Bioactivity of A₆RGD. Human cornea stromal fibroblasts (hCSFs) were grown on films produced from 0.01–2.5 wt% A₆RGD solutions or uncoated polystyrene culture plate controls (0 wt% A₆RGD) for five days. Cells were quantified (mean ± S.D.) using the Alamar Blue assay (n = 3; *** corresponded to p < 0.001; n.d., not determined due to absence of cells).

However, cells grown on 2.5 wt% A₆RGD films assumed a round morphology (Fig. 6c) and adhered poorly compared to the uncoated control (Fig. 7), both constituting evident signs of cytotoxicity. Furthermore, after five days in culture, no viable cells were found on films produced from A₆RGD at 2.5 wt%, in contrast to the still-proliferating hCSFs on control uncoated surfaces (Fig. 6 and 7).

To verify the specificity of cell adhesion to the RGD motif, inhibition assays were performed using cRGD or anti-integrin αV antibody as soluble factors to block the predominant integrins expressed at the surface of hCSFs (Fig. 8). Cells were incubated with soluble anti-integrin αV antibody or cRGD prior to seeding onto 1.0 wt% A₆RGD films coating ultra-low attachment plates, while soluble anti-mouse IgG-HRP antibody or the peptide amphiphile C₁₆-ETTES were used as corresponding mock treatments. Results showed that, 24 h post-seeding, the number of integrin-blocked hCSFs attached to A₆RGD coatings was significantly (p < 0.001) reduced to 27 ± 4% (+Antibody) and 27 ± 11% (+cRGD) of that from corresponding mock treatments (Fig. 8). In addition, no cells were shown to attach to uncoated low-attachment surfaces. These results demonstrated that the enhanced cell adhesion observed for the A₆RGD coating was a specific effect, and involved the direct interaction between the RGD motif and integrins.

Fig. 8 A₆RGD inhibition assays. Human cornea stromal fibroblasts (hCSFs) treated with soluble anti-integrin αV antibody or cRGD (blockers) and corresponding mock substitutes were seeded on 1 wt% A₆RGD films coating ultra-low attachment surfaces. Cells were quantified (mean ± S.D.) 24 h post-seeding using the Alamar Blue assay (n = 3; *** corresponded to p < 0.001).

These results indicate that films produced from A₆RGD solutions containing up to 1.0 wt% SLP promote adhesion and enhanced proliferation of hCSFs, probably through mechanisms mediated by integrin clustering and downstream signaling.⁵⁴ However, 2.5 wt% A₆RGD induces cytotoxicity that limited cell adhesion and prevents subsequent proliferation. This effect can be explained by either a direct or indirect action of the RGD motif on mechanisms of cell apoptosis. Previous studies have shown that incubation with 1 × 10⁻³ M of soluble RGD-containing linear peptides directly activates caspase signaling, inducing apoptosis and reducing proliferation of human lymphocytes³⁵ and umbilical vein endothelial cells.³⁴ Alternatively, cell death could result from the uptake of SLP vesicles able to integrate cell membranes and causing formation of pores, thus increasing lipid bilayer conductance and causing membrane disruption and apoptosis.^55,56 Long fibers formed from 2.5 wt% A₆RGD provide a compact coating of the cell culture plates (Fig. 6a). Excess A₆RGD molecules might then be released from the SLP coating to the media, thus blocking integrin-mediated cell attachment and inducing caspase activation and apoptosis. However, it is unlikely that apoptosis is caused by the presence of SLP vesicles, due to the small population of vesicles present in a ∼2.5 wt% A₆RGD solution (inset Fig. 5d).

Our results showed that the self-assembly motif of A₆RGD depended on its concentration. In particular, we propose that 0.1 wt% A₆RGD self-assembles into oligomers, while at 1.0 and 2.5 wt% A₆RGD forms short and long peptide fibrils respectively. Both 0.1 and 1.0 wt% SLP promoted adhesion and enhanced proliferation of hCSFs, while higher A₆RGD concentrations were toxic to the cells.

Conclusions

In this work we studied the nanostructure of A₆RGD in water. The self-assembly motif was shown to depend on the concentration of SLP. Initial fluorescence assays, together with Congo red absorption experiments and native PAGE assays confirmed the presence of A₆RGD oligomers for 0.1 wt% SLP, short peptide fibrils for 1.0 wt% SLP and a network of long peptide fibres for 2.5 wt% A₆RGD. Cryo-TEM and SAXS confirmed that a small population of vesicles co-exists with fibres from 2 to 15 wt% A₆RGD. The population of vesicles relative to fibres increases with concentration. This is, to our knowledge, the first study to report the formation of vesicles by an alanine-rich SLP in solution.

A₆RGD vesicle walls are built out of highly interdigitated A₆ blocks, presumably with positively charged RGD epitope groups exposed to water. An efficient packing of alanine residues into β-sheets was also revealed by XRD. A₆RGD fibres are also likely to present an A₆ core surrounded by RGD groups exposed to water.

Changes in the A₆RGD self-assembly motif with concentration have implications in the use of this SLP to fabricate substrates for cell culture assays. At concentrations below the c.a.c., films of A₆RGD are compatible with cell viability and proliferation. Films produced from A₆RGD solutions at 0.1–1.0 wt%, cell adhesion proliferation was enhanced relative to uncoated controls.

As such, A₆RGD was shown to constitute a promising SLP for the manufacture of biocompatible dry film substrates for tissue engineering and cell biology applications.

Acknowledgements

This work was supported by EPSRC grants EP/F048114/1 and EP/G026203/1 and BBSRC BB/I008187/1. We would like to acknowledge T. Narayan for support during the beamtime at ID02 (Project Number SC-3235). We would like to acknowledge the University of Reading (UK) for access to the Chemical Analysis Facility.

Notes and references

1. S. Vauthey, S. Santoso, H. Gong, N. Watson and S. Zhang, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 5355.
2. S. Santoso, W. Hwang, H. Hartman and S. Zhang, Nano Lett., 2002, 2, 687.
3. I. W. Hamley, Soft Matter, 2011, 7, 4122.
4. S. G. Zhang, Nat. Biotechnol., 2003, 21, 1171.
5. S. Vauthey, S. Santoso, H. Gong, N. Watson and S. G. Zhang, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 5355.
6. G. von Maltzahn, S. Vauthey, S. Santoso and S. G. Zhang, Langmuir, 2003, 19, 4332.
7. R. G. Ellis-Behnke, Y. X. Liang, S. W. You, D. K. C. Tay, S. G. Zhang, K. F. So and G. E. Schneider, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 5054.
8. R. Das, P. J. Kiley, M. Segal, J. Norville, A. A. Yu, L. Y. Wang, S. A. Trammell, L. E. Reddick, R. Kumar, F. Stellacci, N. Lebedev, J. Schnur, B. D. Bruce, S. G. Zhang and M. Baldo, Nano Lett., 2004, 4, 1079.
9. C. X. Chen, F. Pan, S. Z. Zhang, J. Hu, M. W. Cao, J. Wang, H. Xu, X. B. Zhao and J. R. Lu, Biomacromolecules, 2010, 11, 402.
10. S. Bucak, C. Cenker, I. Nasir, U. Olsson and M. Zackrisson, Langmuir, 2009, 25, 4262.
11. A. Nagai, Y. Nagai, H. J. Qu and S. G. Zhang, J. Nanosci. Nanotechnol., 2007, 7, 2246.
12. U. Khoe, Y. L. Yang and S. G. Zhang, Macromol. Biosci., 2008, 8, 1060.
13. V. Castelletto, D. Nutt, I. W. Hamley, S. Bucak, C. Cenker and U. Olsson, Chem. Commun., 2010, 46, 6270.
14. D. A. Middleton, J. Madine, V. Castelletto and I. W. Hamley, Angew. Chem., Int. Ed., 2013, 52, 10537.
15. C. X. Chen, F. Pan, S. Z. Zhang, J. Hu, M. W. Cao, J. Wang, H. Xu, X. B. Zhao and J. R. Lu, Biomacromolecules, 2010, 11, 402.
16. H. Xu, J. Wang, S. Y. Han, J. Q. Wang, D. Y. Yu, H. Y. Zhang, D. H. Xia, X. B. Zhao, T. A. Waigh and J. R. Lu, Langmuir, 2009, 25, 4115.
17. I. W. Hamley, A. Dehsorkhi and V. Castelletto, Chem. Commun., 2013, 49, 1850.
18. M. D. Pierschbacher and E. Ruoslahti, Nature, 1984, 309, 30.
19. T. Pakalns, K. L. Haverstick, G. B. Fields, J. B. McCarthy, D. L. Mooradian and M. Tirrelli, Biomaterials, 1999, 20, 2265.
20. N. Oku, C. Koike, Y. Tokudome, S. Okada, N. Nishikawa, H. Tsukada, M. Kiso, A. Hasegawa, H. Fujii, J. Murata and I. Saiki, Adv. Drug Delivery Rev., 1997, 24, 215.
21. V. Marchi-Artzner, B. Lorz, C. Gosse, L. Jullien, R. Merkel, H. Kessler and E. Sackmann, Langmuir, 2003, 19, 835.
22. V. Castelletto, I. W. Hamley, C. Stain and C. J. Connon, Langmuir, 2012, 28, 12575.
23. G. Cheng, V. Castelletto, R. R. Jones, C. J. Connon and I. W. Hamley, Soft Matter, 2011, 7, 1326.
24. V. Castelletto, C. M. Moulton, G. Cheng, I. W. Hamley, M. R. Hicks, A. Rodger, D. E. López-Pérez, G. Revilla-López and C. Alemán, Soft Matter, 2011, 7, 11405.
25. R. Orbach, L. Adler-Abramovich, S. Zigerson, I. Mironi-Harpaz, D. Seliktar and E. Gazit, Biomacromolecules, 2009, 10, 2646.
26. M. Zhou, A. M. Smith, A. K. Das, N.W. Hodson, R. F. Collins, R. V. Ulijn and J. E. Gough, Biomaterials, 2009, 30, 2523.
27. M. O. Guler, L. Hsu, S. Soukasene, D. A. Harrington, J. F. Hulvat and S. I. Stupp, Biomacromolecules, 2006, 7, 1855.
28. T. Muraoka, C. Y. Koh, H. Cui and S. I. Stupp, Angew. Chem., Int. Ed., 2009, 48, 5946.
29. J. D. Hartgerink, E. Beniash and S. I. Stupp, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 5133.
30. J. D. Hartgerink, E. Beniash and S. I. Stupp, Science, 2001, 294, 1684.
31. H. W. Jun, S. E. Paramonov, H. Dong, N. Forraz, C. McGuckin and J. D. Hartgerink, J. Biomater. Sci., Polym. Ed., 2008, 19, 665.
32. M. J. Webber, J. Tongers, M.-A. Renault, J. G. Roncalli, D. W. Losordo and S. I. Stupp, Acta Biomater., 2010, 6, 3.
33. G. A. Silva, C. Czeisler, K. L. Niece, E. Beniash, D. A. Harrington, J. A. Kessler and S. I. Stupp, Science, 2004, 303, 1352.
34. M. S. Aguzzi, C. Giampietri, F. De Marchis, F. Padula, R. Gaeta, G. Ragone, M. C. Capogrossi and A. Facchiano, Blood, 2004, 103, 4180.
35. C. D. Buckley, D. Pilling, N. V. Henriquez, G. Parsonage, K. Threlfall, D. Scheel-Toellner, D. L. Simmons, A. N. Albar, J. M. Lord and M. Salmon, Nature, 1999, 397, 534.
36. G. Pabst, M. Rappolt, H. Amenitsch and P. Laggner, Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top., 2000, 62, 4000.
37. A. Guinier and G. Fournet, Small-Angle Scattering of X-rays, John Willey & Sons, Inc., New York, 1955.
38. Molinspiration Cheminformatics, Software package Molinspiration, 2010.
39. Innovagen, Software package Innovagen's Peptide Property Calculator, 2012.
40. H. LeVine, Protein Sci., 1993, 2, 404.
41. M. R. Nilsson, Methods, 2004, 34, 151.
42. A. F. Drake, G. Siligardi and W. A. Gibbons, Biophys. Chem., 1988, 31, 143.
43. S. E. Paramonov, H. W. Jun and J. D. Hartgerink, J. Am. Chem. Soc., 2006, 128, 7291.
44. H. Gaussier, H. Morency, M. C. Lavoie and M. Subirade, Appl. Environ. Microbiol., 2002, 68, 4803.
45. J. T. Pelton and L. R. McLean, Anal. Biochem., 2000, 277, 167.
46. P. Haris and D. Chapman, Biopolymers, 1995, 37, 251.
47. L. C. Serpell, Biochim. Biophys. Acta, Bioenerg., 2000, 1502, 16.
48. I. W. Hamley, Angew. Chem., Int. Ed., 2007, 46, 8128.
49. I. W. Hamley, A. Dehsorkhi, V. Castelletto, J. Seitsonen, J. Ruokolainen and H. Iatrou, Soft Matter, 2013, 9, 4794.
50. O. Rathore and D. Y. Sogah, J. Am. Chem. Soc., 2001, 123, 5231.
51. J. Kohlbrecher and I. Bressler, Software package SASfit for fitting small-angle scattering curves, 2011.
52. T. E. Creighton, Protein Folding, W.H. Freeman, New York, 1992.
53. R. M. Gouveia, V. Castelletto, S. G. Alcock, I. W. Hamley and C. J. Connon, J. Mater. Chem. B, 2013, 1, 6157.
54. F. G. Giancotti and E. Ruoslahti, Science, 1999, 285, 1028.
55. R. Kayed, Y. Sokolov, B. Edmonds, T. M. McIntire, S. C. Milton, J. E. Hall and C. G. Gable, J. Biol. Chem., 2004, 279, 46363.
56. R. Jelinek and T. Sheynis, in Lipids and Cellular Membranes in Amyloid Diseases, ed. R. Jelinek, Wiley-VCH Verlag & Co., Germany, 2011, ch. 12.

---

Footnote

† Electronic supplementary information (ESI) available: Displays chemical formula and contains pH, CD, FTIR and XRD data. Parameters obtained from the fittings to the SAXS data. See DOI: 10.1039/c3bm60232j

---