Article pubs.acs.org/Langmuir
Drop-Casting Hydrogels at a Liquid Interface: The Case of Hydrophobic Dipeptides Tao Li,† Michail Kalloudis,‡ Andre Zamith Cardoso,§ Dave J. Adams,§ and Paul S. Clegg*,† †
School of Physics and Astronomy, University of Edinburgh, Mayfield Road, Edinburgh EH9 3JZ, U.K. Institute for Materials and Processes, School of Engineering, University of Edinburgh, Mayfield Road, Edinburgh EH9 3JL, U.K. § Department of Chemistry, University of Liverpool, Crown Street, Liverpool L69 7ZD, U.K. ‡
S Supporting Information *
ABSTRACT: Hydrophobic dipeptide molecules have been induced to self-assemble into thin interfacial films at the air− water interface via drop-casting. The mechanism involves fiberlike strands, which exist in the high-pH spreading solvent, becoming intertwined at the surface of a low-pH subphase. Atomic force microscopy (AFM) reveals that the strands are ∼40 nm wide and ∼20 nm high and are woven together to form layers that can be up to ∼800 nm thick. The use of Thioflavin T (ThT) fluorescence suggests that the dipeptides are ordered in a β-sheet configuration irrespective of whether they form an interfacial film, while Fourier transform infrared spectroscopy (FTIR) shows the protonation effect for those which do form an interfacial film. The entanglement between protonated strands results in the formation of an elastic sheet. The interfacial films buckled under compression in a Langmuir trough and have the ability to convey long-term stability to large air bubbles.
■
probed only down to 12 amino acid sequences.15 The chosen sequence is known to form amyloid fibrils in the bulk. Interfacial films were prepared at an air−water interface by first dissolving the peptide in trifluoroacetic acid (TFA)/CHCl3 or hexafluoroisopropanol (HFIP)/CHCl3. Pressure−area isotherms of the resulting films exhibited hysteresis unless the surface pressure was kept below 10 mN/m. This indicates that an irreversible phenomenon (e.g., formation of 3D structures) occurs at around this surface pressure. Atomic force microscopy (AFM) studies demonstrated that the interface was (heterogeneously) coated in rod-shaped assemblies with a thickness of 0.5−1 nm, a width of 50−100 nm, and a length of 1 μm. The thickness makes these structures inconsistent with fibrils. Fourier transform infrared spectroscopy (FTIR) revealed the almost exclusive presence of antiparallel β-sheets. The authors called the rods self-assembled bricks, suggesting that the elongated nanodomains are generic building blocks.15 A subsequent study showed that these nanodomains are built up from smaller nanostrips. X-ray diffraction was used to demonstrate both transverse and longitudinal crystallinity in the nanodomains. An HFIP/CHCl3 spreading solvent with 10 mM CaCl2 in the subphase gave long-range crystallinity in the interfacial packing.16 We are interested in much shorter peptide sequences tethered to an organic group. In a previous study, FmocFF has been deposited at an air−water interface as one component
INTRODUCTION Many dipeptides are known to form hydrogels in solution as the solvent conditions are changed using pH, salt, etc.1−4 Of particular interest, both as model systems and for applications as biomimetic materials, are N-protected dipeptides (Fmoc-, Boc-, Cbz-, or naphthalene-protected).4−10 One means of inducing gelation is in response to a pH change from high to low due to the carboxylic acid becoming protonated and hence the molecule becoming increasingly hydrophobic. The behavior is fascinating because the organized supramolecular strands are woven together to create the hydrogel. In the light of this bulk behavior, it is important to determine what would happen if a dipeptide in a high-pH solvent was drop-cast on a liquid subphase with low pH. Previously, studies of peptides at interfaces have mainly focused on amphiphilic short sequences created by tethering an alkyl tail to a hydrophilic sequence of peptides via a several glycine linker section. These are found to self-assemble into stable rods in solution with β-sheet formation observed closest to the alkyl chain.11−13 Sodium salts of Fmoc-amino acids have also been investigated as potential surfactants.14 They have been shown to reduce the interfacial tension (albeit to a relatively modest extent) and then to form assemblies above a critical micelle concentration (CMC). Nuclear magnetic resonance spectroscopy (NMR) was used to show that the Fmoc group was buried in the center of the assembly. A controlling factor is the length of the alkyl chain associated with the chosen peptide. Salts based on peptides Val, Leu, and Ile all had readily determined CMCs.14 An alternative phenomena is large-scale peptide self-assembly at interfaces which has been © 2014 American Chemical Society
Received: April 2, 2014 Revised: May 19, 2014 Published: May 20, 2014 13854
dx.doi.org/10.1021/la501182t | Langmuir 2014, 30, 13854−13860
Langmuir
Article
Figure 1. (a) Naphthalene dipeptide molecules used in this study. (b, c) Atomic force microscopy images of strands of NapFF dipeptide drop-cast on a subphase of pH 3.0. (b) Submonolayer region, prepared from a concentration 0.5 wt % dispersion, shows that the strand width is ∼40 nm and the height is ∼20 nm. (c) Concentration 0.5 wt % giving a film thickness of ∼90 nm. (d) Fluorescence confocal microscopy of a NapFF dipeptide with ThT showing the film on the scale of several tens of micrometers. Inset: a micrograph of a film prepared from a different dipeptide (BrNapFF). Here the subphase used is at pH 2.0.
of a liquid marble.17 Additionally, the formation of films has been demonstrated via electrochemistry using the molecule FmocLG, which is known to form a hydrogel in the bulk when the solvent pH is lowered to below pH 4. Film growth was successfully induced by creating a localized pH drop at an electrode surface by the oxidation of hydroquinone.8 In this article, we report the self-assembly of hydrophobic dipeptides at the air−water interface via drop-casting. Our experiments focus mainly on the NapFF dipeptide (Figure 1a); we demonstrate that it becomes a thin interfacial film when brought into contact with a low-pH subphase. AFM and FTIR spectra are used to study the structure of these films. The mechanism for interfacial trapping is also discussed. Furthermore, we demonstrate that the interfacial film has the character of an elastic sheet which can be compressed in a Langmuir trough. More interestingly, these interfacial films are able to stabilize large air bubbles for several days.
■
solution was formed. The dipeptide/NaOH mol ratio is 1:1.5. To prepare the interfacial film, 2 μL of a 5 mg/mL NapFF solution was deposited dropwise at the free surface (∼12 cm2) of an aqueous HCl subphase at room temperature. In most cases, the pH of the subphase was 3.0, as measured with an electronic pH meter (Mettler-Toledo AG, 8603 Schwerzen-bach, Switzerland) and calibrated using standard buffer solutions at pH 4.0 and 7.0. Atomic Force Microscopy. A Bruker AFM Multimode/Nanoscope IIIa (Bruker, Santa Barbara, CA, USA) was used to study the morphology of the dipeptide thin films in tapping mode. RTESP Bruker cantilevers with a nominal spring constant and resonance frequency of 40 N/m and 300 kHz, respectively, were used to image the samples. In order to minimize the interaction force between the tip and the substrate (but without losing contact), light tapping was used by keeping the set-point amplitude ratio rsp = Asp/A0 close to 1 (where Asp and A0 are the free oscillation amplitude and the reduced scanning set-point amplitude of the cantilever, respectively). Images were processed using WSxM software (Nanotec Electronica S.L., Madrid, Spain). Confocal Microscopy and Thioflavin T Dye (ThT) Fluorescence. AFM and confocal microscopy studies were often carried out on different regions of the same film. The ThT concentration used was at or below 2 μM to avoid dye self-assembly.18 Confocal microscopy was performed using a Zeiss Observer.Z1 inverted microscope in conjunction with a Zeiss LSM 700 scanning system 9 and a 63× 1.40
EXPERIMENTAL SECTION
Napthalene Dipeptide Synthesis. The NapFF, BrNapFF, and BrNapAV molecules (Figure 1a) used in this study were prepared using the procedures described in ref 2. The dipeptide was suspended in deionized water at different concentrations of 0.05−8 mg/mL. NaOH (0.1 M, aq) was added, and the liquid was stirred until a clear 13855
dx.doi.org/10.1021/la501182t | Langmuir 2014, 30, 13854−13860
Langmuir
Article
Figure 2. (a) Dispersion of NapFF dipeptide deposited on the interface of pH-adjusted water. The dipeptide is dyed with sodium fluorescein (1.0 wt %), making it yellow. At low subphase pH, the dipeptide and dye are trapped at the interface; as the subphase pH is raised, the yellow deposit can be seen to sink through the interface. (b) Demonstration of the colocalization of sodium fluorescein and NapFF. The subphase has a pH of 2.0; a 1.0 wt % solution of sodium fluorescein was drop-cast into the left-hand dish, and a combined dispersion of NapFF and sodium fluorescein (again at 1.0 wt %) were drop-cast onto the interface of the right-hand dish. NA oil-immersion objective. The ThT fluorescence was excited using the 488 nm line from an Ar ion laser. Naphthalene Fluorescence. Aqueous HCl (36 mL) with different pH values (2.0−7.0) was added to a separation funnel as subphases. NapFF solutions (2 μL, 5 mg/mL, without fluorescent dye) were deposited dropwise on the surface of these subphases. After a relaxation time of 5 min, 10 mL of the lower subphase was isolated. Before the fluorescence measurement, the UV−visible absorption spectra of NapFF solution were recorded with a Cary 1E spectrophotometer. Afterward, the fluorescence spectra from the subphase were recorded on a Varian Cary Eclipse spectrophotometer. In all cases, excitation was carried out at 220 nm, and a 3 mL quartz cell with an optical path length of 10 mm was utilized. FTIR Measurements. The dipeptide film formed at the interface was transferred to a low-emissivity (low-e) glass slide and then freezedried. Infrared spectra were measured using a Smiths Illuminat IR module coupled to a Renishaw inVia Raman microscope. Each spectrum was scanned between 4000 and 650 cm−1 at a resolution of 2 cm−1. The spectrum of the blank low-e slide was used as the background reference. Surface Pressure Measurements. These studies were carried out in a Langmuir trough (KSV) with symmetric barriers and a usable area of 313 × 54 mm2. The surface pressure was measured using a Wilhelmy plate (using filter paper to give a wetting angle of 0°) which was aligned perpendicular to the barriers. The air−water interface was considered clean if the surface pressure of the interface changes less than 0.1 mN/m when the barriers are moved in. A dipeptide solution (6 μL) of the desired composition was then spread dropwise at the free surface of an aqueous HCl subphase at room temperature and was left for 20 min prior to the start of measurements. The barriers were typically moved together at a rate of 2 mm/min. 1 H NMR Experiment. A stock solution of NapFF was prepared by diluting 100 mg in 9.78 mL of doubly distilled H2O and 0.22 mL of 1 M NaOH. The 1 M NaOH was freshly prepared and filtered through a 0.2 μm syringe (Minisart RL 15, Sartorius Stedim) before use. The pH was measured using a FC200 pH probe (HANNA Instruments) with a (6 mm × 6 mm) conical tip. The stated accuracy of the pH measurements is ±0.1. Solutions at final concentrations of 1.0, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.09, 0.08, 0.07, 0.06, 0.05, and 0.04 wt % were prepared from this stock solution (all adjusted to pH 10.5 using a 1 M NaOH solution). Aliquots of 0.5 mL of each were transferred to an NMR tube (NE-ML5-8, NEW ERA Enterprises). A reference capillary with ethanol in D2O was prepared to allow a precise quantification and shimming of the NMR spectrometer. The capillary was filled with an aliquot of 6 μL of ethanol in 1 mL of D2O solution and sealed with PTFE tape. The capillary was calibrated with an 8 mg/ mL L-alanine solution in H2O. 1H NMR spectra of NapFF solutions in H2O were obtained at 25 °C using a 500 MHz Bruker Avance-III.
Before each measurement, the reference capillary was inserted with a PTFE holder into the NMR tube. A standard Bruker proton experiment (zg30) was performed with 16 scans, 64K data points, and an acquisition time for each scan of 3.27 s for the chemical shift analysis. Peak positions were measured by analyzing the spectra using Topspin 3.
■
RESULTS AND DISCUSSION
Figure 1 shows films prepared by drop-casting a dispersion of NapFF dipeptide molecules (typically ≥0.5 wt %) in a high-pH (∼10) spreading solvent (Experimental section) onto a subphase at pH 3.0. The as-cast films are then transferred onto a mica substrate by pulling them through the coated interface. Imaging is subsequently carried out using AFM. The films consist of nanoscale-width strands of material which can be many micrometers in length. For higher concentrations of dipeptide, the films also become thick. Partial monolayer films (such as the one shown in Figure 1b) were used to measure the dimensions of the strands. Averaged over more than 60 strands, we find that the in-plane width of a single fiber is 35−40 nm. TEM studies of bulk hydrogels show somewhat thinner strands.2 Some dipeptide dispersions were labeled with ThT, and the resulting films were studied using both AFM and confocal microscopy. The observed ThT fluorescence enhancement demonstrates that the ThT molecule has been immobilized.19 This is commonly associated with the dyed species forming β-sheets. A similar enhancement of the ThT fluorescence has previously been observed for naphthalene dipeptides in bulk hydrogels.2 Further investigations with BrNapFF and BrNapAV (Figure 1a) show that film formation is not unique to NapFF nor is it a universal feature of naphthalene-based dipeptides. BrNapFF was found to form interfacial films following a similar protocol to that employed with NapFF. The only significant change was that the subphase pH needed to be below 2, i.e., somewhat lower than for NapFF. Again, the use of ThT dye supports the idea that the interfacial strands of BrNapFF have a β-sheet structure (inset to Figure 1d). By contrast, BrNapAV does not form interfacial films for the range of concentrations and subphase pH values that we explored (Supporting Information). This difference in behavior can be linked to the micellar structures formed at high pH. We have previously shown by TEM, SAXS, and viscosity measurements that NapFF and 13856
dx.doi.org/10.1021/la501182t | Langmuir 2014, 30, 13854−13860
Langmuir
Article
In order to probe the organization of the molecules further, we have used FTIR. Figure 4a shows the signal from NapFF before (black) and after (red) it has formed an interfacial film. The samples were freeze-dried prior to characterization. The original (high-pH) solution has a peak at 1630 cm−1, which is often assigned to β-sheet formation,2,22 although recent data for related materials has suggested that this assignment might not be so straightforward for such short peptides.23,24 No peak was observed at 1720 cm−1, indicating that the carboxylic acid group is deprotonated. Peaks appear at both 1720 cm−1 and at 1650 cm−1 once a film has formed. The former relates to the carboxylic acid group becoming protonated (and hence to the dipeptide becoming more hydrophobic) while the latter may well be related to the formation of a random coil structure.2 Both the original solution and the interfacial film exhibit an absorption peak at around 1540 cm−1 which is indicative of βsheet formation. This spectroscopic investigation suggests that β-sheet formation has already occurred in the high-pH spreading solvent. Organized strands stick together on impact with the interface, rather than isolated molecules. This result is supported by bright-field microscopy images from dispersions of NapFF as the concentration is increased (Figure 4b). For concentrations >0.35 wt %, we see large, fiberlike assemblies of dipeptide molecules at the pH of the spreading solvent, which have been shear-aligned during loading into the imaging cell. By contrast, at lower NapFF concentration there is some indication of the existence of small spherical assemblies. These two regimes meet at the concentration threshold at which we begin to observe interfacial film formation from fiberlike strands. The microscopy observations are corroborated by NMR measurements probing the proton environment of the naphthalene group (Figure 4c). Here a transition is observed at around the same concentration, suggesting a change from small micelles at low concentrations to large fiberlike structures at higher concentrations. The inset to Figure 4a shows a comparison between the FTIR spectra for the solution of NapFF (which forms an interfacial film on impact with a low-pH subphase) and that for a solution of BrNapAV which does not form an interfacial film under any of the conditions probed here. Again a β-sheet organization is observed in the dispersion, hence this molecular order alone is not a predictor of interfacial assembly (there is also a small chance that this organization might be a side effect of drying25). This observed organization is perhaps unsurprising given that spherical structures are known to be formed at high pH by BrNapAV.20 We have shown that appropriately chosen dipeptide molecules form thin films at acid−air interfaces. To expand the realm of applicability of this observation, we have developed a method for stabilizing large air bubbles with these molecular assemblies. Air was bubbled through an aqueous solvent at the appropriate pH. As bubbles formed, NapFF in a high-pH spreading solvent was dripped onto the interface as the bubble reached it. Bubbles were stabilized for several days; they did not coalesce with the top surface of the sample or with other bubbles on impact (Figure 5) and they did not exhibit significant Ostwald ripening. Our AFM and confocal microscopy results (Figure 1) strongly suggest that we do not have complete coverage of the air interface; we infer that the long-term stability of these bubbles is a consequence of the elastic properties of the interfacial film. To confirm that the interfacial film has the character of an elastic sheet, we have
BrNapFF form wormlike micelles at high pH. On the other hand, BrNapAV forms spherical structures.20 However, we note that it is currently unclear whether the structures formed at high pH are simply “trapped” when the pH is lowered or whether a structural reorganization occurs. That BrNapAV does not form films may be a function of the hydrophobicity as well as the nature of the micellar aggregates at high pH. However, this is difficult to isolate as we have previously shown that the presence of wormlike micelles correlates with the hydrophobicity of the molecules.21 Due to the similarity between the hydrogels formed previously in the bulk and the interfacial layer observed here at the interface, it is essential to demonstrate that the dipeptides are indeed interfacially trapped. To this end, we use the fluorescence signal, first from added sodium fluorescein and second from the naphthalene end group, to establish the location of the dipeptide molecules as the subphase pH is modified. Figure 2a shows a series of Petri dishes at different pH values; into each has been drop-cast NapFF dispersions containing sodium fluorescein (1 wt %). Clearly for pH 2 and 3.3, the dyed dispersion remains at the interface; by contrast, at pH 3.8 and 5.2 substantial quantities of the yellow dispersion are seen to penetrate the subphase (Figure 2b). Figure 3 shows
Figure 3. Napthalene fluorescence signal from dipeptide in the subphase. Following excitation at 220 nm, the intensity of the observed emission signal was a strong function of the subphase pH. At pH
