Article pubs.acs.org/Langmuir
Self-Assembling Peptide/Thermoresponsive Polymer Composite Hydrogels: Effect of Peptide−Polymer Interactions on Hydrogel Properties A. Maslovskis,† J.-B. Guilbaud,† I. Grillo,‡ N. Hodson,§ A. F. Miller,*,† and A. Saiani*,∥ †
Manchester Institute of Biotechnology and School of Chemical Engineering & Analytical Science, The University of Manchester, Oxford Road, Manchester, M13 9PL, U.K. ‡ Institut Laue-Langevin, 6 rue Jules Horowitz, 38042 Grenoble, France § BioAFM Facility, Stopford Building, The University of Manchester, Oxford Road, Manchester, M13 9PL, U.K. ∥ School of Materials and Manchester Institute of Biotechnology, The University of Manchester, Oxford Road, Manchester, M13 9PL, U.K. ABSTRACT: We have investigated the effect of doping the self-assembling octapeptide FEFEFKFK (F, phenylalanine; E, glutamic acid; K, lysine) hydrogels with various amounts of thermoresponsive conjugate of FEFEFKFK and poly(Nisopropylacrylamide) (PNIPAAm) in order to create novel hydrogels. The samples were characterized using a range of techniques including microdifferential scanning calorimetry (μDSC), oscillatory rheology, transmission electron microscopy (TEM), atomic force microscopy (AFM), and small angle neutron scattering (SANS). The peptide from the conjugate was shown to be incorporated into the peptide fiber, resulting in the polymer being anchored to the peptide fiber. The conjugation of the polymer to the peptide and its anchoring to the peptide fibers did not affect its lower critical solution temperature (LCST). On the other hand, it did result in a decrease in the LCST enthalpy and a significant increase in the G′ of the hydrogels, suggesting the presence of hydrogen bond interactions between the peptide and the polymer. As a result, the polymer was found to adopt a fibrillar arrangement tightly covering the peptide fiber. The polymer was still found to go through a conformational change at the LCST, suggesting that it collapses onto the peptide fiber. On the other hand, the fibrillar network was found to be mainly unaffected by the polymer LCST. These changes at the LCST were also found to be fully reversible. The nature of the interaction between the polymer and the peptide was shown to have a significant effect on the conformation adopted by the polymer around the fibers and the mechanical properties of the hydrogels.
■
INTRODUCTION
amino acids, the properties of the hydrogels can be tailored to the intended application.3,9−17 Self-assembling peptide systems can also be considered as stimuli-responsive materials, as gelation (sol-to-gel transition) and conformational changes can be triggered by various environmental stimuli, such as pH, ionic strength, and the presence of enzymes. In some cases, temperature has been used as a stimulus to trigger gelation, but for the majority of systems, only gel melting is observed upon heating.18,19 Peptides exhibiting a lower critical solution temperature (LCST) have been reported. For example, short elastin-like peptides based on GVGVP (G, glycine; V, valine; P, proline) sequence exhibit LCST behavior in water; however, the LCST values are very high, i.e., ∼100−200 °C.20 For biomedical applications, much lower LCST temperatures are required.
Self-assembling peptide-based materials have attracted significant attention over the past decade due to their inherent biocompatibility, biodegradability, and ability to easily manipulate their chemical functionality and properties simply by changing their primary structure. Such materials are typically formed from short oligopeptides which self-assemble into various secondary structures, including β-sheets, β-hairpins, and α-helices. These tend to further assemble under certain conditions into a variety of higher ordered structures such as micellar aggregates, nanotubes, membranes, or fibrils.1−8 Design rules to control such self-assembling pathways are emerging, and fibril formation, in particular, has been the focus of many studies. Under appropriate conditions, above a critical concentration, these fibrils can self-associate and form selfsupporting hydrogels which have been shown to have potential in a number of applications including controlled drug delivery and tissue regeneration. It has been shown that, by controlling peptide concentration and the distribution and sequence of © 2014 American Chemical Society
Received: June 17, 2014 Revised: August 4, 2014 Published: August 6, 2014 10471
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Langmuir
Article
Figure 1. Schematic representation of the self-assembly mechanism of the β-sheet forming peptide/peptide−polymer conjugate mixtures.
conjugate would incorporate into the fibrillar structure of the matrix, thus decorating the peptide fibers with PNIPAAm polymer, as shown schematically in Figure 1. In this paper, we decided to investigate the effect that doping the peptide hydrogels with peptide−PNIPPAm conjugate has on the hydrogel properties, mechanical and morphological. In particular, we investigated the role played by peptide−polymer interactions on the peptide fibrillar network morphology and on the LCST behavior of the polymer. For this purpose, the composite hydrogels were investigated using a combination of techniques including microdifferential scanning calorimetry (μDSC) oscillatory rheology, transmission electron microscopy (TEM), atomic force microscopy (AFM), and small angle neutron scattering (SANS).
One strategy to introduce thermoresponsiveness in hydrogels consists of introducing thermoresponsive micelles or nanoparticles into a nonthermoresponsive gel matrix, or into an already thermoresponsive gel to accentuate its thermoresponsiveness and improve gel network swelling/shrinking and mechanical properties. Nguyen et al.21 developed a thermoresponsive composite gel by incorporating poly(N-isopropylacrylamide) (PNIPAAm) nanoparticles into poly(ethylene oxide) (PEG) matrix, and the results showed that these gels were able to release bovine serum albumin (BSA) above the LCST. Later, the same group prepared thermoresponsive hydrogels, where poly(N-isopropylacrylamide-co-acrylamide) (PNIPAAm-AAm) nanoparticles, preloaded with drug, were incorporated into a photo-cross-linkable poly(ethylene glycol) diacrylate (PEGDA) matrix.22 Above the LCST, the drug release occurred due to the collapse of PNIPAAm. In one other study, poly(N-isopropylacrylamide)/poly(ethylene glycol) diacrylate (PNIPAAm/PEG-DA) microgels were incorporated into PNIPAAm hydrogels. This additive improved gel swelling/ shrinking ratios and thermal behavior.23 Zhang et al.24 reported composite gels, containing poly(N-isopropylacrylamide)-bpoly(methyl methacrylate) (PNIPAAm-b-PMMA) micelles embedded in a PNIPAAm hydrogel. Gels showed faster shrinking and slower swelling rates with the increase of micelle content. When drug loaded micelles were used, drug release occurred in a more controlled fashion in comparison to the pure PNIPAAm gel. Hudson et al.25 prepared interpenetrating networks of PNIPAAm and silk fibroin. These networks were reported to possess improved mechanical stability and faster deswelling kinetics than ordinary PNIPAAm gels. One emerging strategy to introduce thermoresponsiveness into peptide hydrogels is to conjugate peptides to responsive polymers such as PNIPAAm. Such peptide hybrids represent a class of materials with interesting properties and are attractive for biomaterials applications, as they combine the controlled chemical, mechanical, and thermal properties of a polymer with the structural properties and functionality of a peptide.26−30 In this study, we have combined the non-LCST-possessing FEFEFKFK peptide matrix with the thermoresponsive conjugate of FEFEFKFK and PNIPAAm to develop a composite gel. The prototypical thermoresponsive polymer PNIPAAm has been chosen, as it possesses a sharp coil-toglobule conformational change (i.e., LCST) at temperatures close to body temperature.31,32 The octapeptide FEFEFKFK was selected, as it is well-known to exhibit self-assembly in aqueous solution and form β-sheet-rich fibrillar networks and hydrogels at low concentrations (∼10−20 mg mL−1).33−39 It was postulated that the peptide part of the polymer−peptide
■
MATERIALS AND METHODS
Peptide Synthesis. The peptide FEFEFKFK (F, phenylalanine; E, glutamic acid; K, lysine) was synthesized on a ChemTech ACT 90 peptide synthesizer (Advance ChemTech Ltd, Cambridgeshire, U.K.) using N-methyl-2-pyrrolidone (NMP) as solvent and standard solid phase peptide synthesis protocols.40 The side chain protected amino acids, amino acid activator (HCTU), and preloaded Fmoc-Lys(Boc)Wang resin (bead size, 100−200 mesh; substitution, 0.70 mmol g−1) were purchased from Novabiochem (Merck), 3-mercaptopropionic acid was purchased from Acros Organics, and all other solvents were purchased from Aldrich. All chemicals were used as received. The functionalization of the peptide with a thiol end-group (SHFEFEFKFK) was performed directly on the resin using a procedure described in detail elsewhere.29 Once synthesized, both peptides were cleaved from the resin and deprotected using a mixture of trifluoroacetic acid (TFA) and anisole (95:5). The resin was removed by filtration, and the peptides were precipitated in cold diethyl ether. The peptides were redissolved in water and washed a further four times in diethyl ether before they were isolated via freeze-drying. Reverse-phase high performance liquid chromatography (RP-HPLC), mass spectrometry (MS), and proton nuclear magnetic resonance spectroscopy (1H NMR) were used to confirm the peptide structures and determine their purities which were typically >90%. Conjugate Synthesis. The PNIPAAm−FEFEFKFK conjugate was synthesized via free radical polymerization (FRP) using the thiolfunctionalized peptide SH-FEFEFKFK as a chain transfer agent.28 NIsopropylacrylamide (NIPAAm) (Aldrich, 97%) and azo-iso-butyronitrile (AIBN) (Fisher Scientific, laboratory grade) were recrystallized prior to use from hexane and methanol, respectively. 4.42 mmol of NIPAAm, 0.0614 mmol of AIBN, and 0.0935 mmol of SH-FEFEFKFK were dissolved in 30 mL of deoxygenated dimethyl sulfoxide (DMSO, Aldrich, used as received) solution. The reaction was performed at 65 °C for 24 h under inert atmosphere (N2) and constant stirring. The conjugate/polymer mixture was purified by diluting the reaction mixture to a total volume of 0.5 L using deionized water and dialyzing for 4 days using a membrane with a molecular weight cutoff of 3500 g 10472
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Langmuir
Article
Table 1. Molar Volume, Molar Mass, Scattering Length, and Contrast Factor of Compound Used
a b
compound
scattering length (10−12 × cm)
molar mass (g mol−1)
molar volume (mL mol−1)
contrast factor (10−3 × cm2 g−1)
H2O D2 O FEFEFKFK in H2O FEFEFKFK in D2O NIPAAm in H2O NIPAAm in D2O NIPAAm-d7 in H2O NIPAAm-d7 in D2O
−0.17 1.91 27.1 43.7b 1.39 2.44b 9.73 8.68b
18 20b 1120 1136b 113 114b 120 121b
18.0 18.1 871a
N/A N/A 0.60 1.09 0.22 1.86 3.75 0.03
83 83.8
The peptide molar volume was estimated by adding the molar volume values reported by Zaccai et al. for each amino acid in the peptide sequence. The scattering lengths and molar masses were calculated assuming that in D2O all labile hydrogens are exchanged with deuterium.44
mol−1 (Medicell Ltd., U.K.). The dialysis allowed the removal of short conjugate/polymer chains and any unreacted reagents and peptides. The nonconjugated polymer was removed by centrifugation using a procedure described in detail elsewhere.28 The conjugate fraction was collected and freeze-dried to give a white powder. The conjugate molecular weights were measured by gel permeation chromatography (GPC) using PEG/PEO standards. Mass-average (Mw) and numberaverage (Mn) molecular weights of 8800 ± 500 and 2500 ± 500 g mol−1 were obtained, respectively. The peptide weight fraction in the conjugates was estimated by 1H NMR using physical mixtures of peptide and NIPAAm as the reference following a procedure described in detail in elsewhere.28 The weight fraction of peptide in the conjugate was estimated to be 0.12. The deuterated conjugate was prepared using the exact same methodology as described above by replacing NIPAAM by its deuterated counterpart, NIPAAm-d7 (Polymer Source Inc., 98%). Macroscopic gelation behavior and μDSC were used to ensure that the deuterated conjugate behaved in the same fashion as its protonated partner. Sample Preparation and Phase Behavior. Samples were prepared by dissolving the desired quantities of material in 1 mL of deionized water at room temperature to avoid any interference by the polymer LCST on the gelation process. The samples’ pH was then adjusted to 4 by adding a small amount of a 1 M NaOH solution. The samples were subsequently left to equilibrate at room temperature for 12 h prior to any measurement. The macroscopic phase behavior of the samples was studied as a function of temperature by placing the samples in a temperature-controlled water bath and allowing 15 min of equilibration at each temperature before visual inspection. Microdifferential Scanning Calorimetry (μDSC). μDSC measurements were performed on a SETARAM μDSC III instrument. The sample cell was filled with 0.6 mL of solution using a micropipette. The reference cell was filled with water and its weight adjusted using a microbalance to ensure an identical mass of solution was present in both the sample and reference cells. μDSC thermographs were recorded using a scanning rate of 1.0 °C min−1 in the temperature range 20−80 °C. Four heating/cooling cycles were performed for each sample. Onset temperature and transition enthalpies were determined using Setsoft 2000 software supplied with the instrument. Each experiment was repeated at least three times to ensure reproducibility. Oscillatory Rheology. Rheological studies were performed on a stress-controlled Bohlin C-CVO rheometer, equipped with a Peltier temperature controller. A parallel plate geometry (⌀ = 20 mm; 0.25 mm gap) was used with a solvent trap to minimize water evaporation. Initially, strain amplitude sweeps (γ = 0.001−10) were performed at constant angular frequency (ω = 1 rad s−1) to identify the linear viscoelastic region (LVR). Subsequently, frequency sweeps (ω = 0.1− 10 rad s−1) at constant strain amplitude (γ = 0.01) were performed to determine the elastic (G′) and viscous (G″) moduli within the LVR. All measurements were performed at 20 °C, and all measurements were repeated at least three times to ensure reproducibility. Small Angle Neutron Scattering (SANS). SANS experiments were performed on the D22 beamline at the Institut Laue-Langevin (ILL) in Grenoble, France. A wavelength of λ = 6 Å was used, and the data were collected on a (3He) multidetector with an active area of 100
× 100 cm2 and a pixel size of 0.8 × 0.8 cm2. By varying the sample− detector distance, the available q was in the range 0.1 < q (nm−1) < 5, with q = (4π/λ) sin(θ/2), where θ is the scattering angle. For a binary system (scattering objects/solvent), the coherent intensity scattered by the scattering objects, in our case the peptide or monomer depending on the sample used (see below for more details), in absolute units is
IA(q) =
1 (I N(q) − (1 − C p)ID(q) − Ib) K
(1)
where IN(q) and ID(q) are the normalized intensity scattered by the sample and the solvent, respectively, Cp is the scattering object concentration in g cm−3, Ib is the background scattering due to incoherent scattering of the hydrogenated peptide, and K is the contrast factor:
K=
(a p − Ypsas)2 NA mp2
(2)
where ap and as are the scattering lengths of the scattering object and the solvent, respectively, Yps is the molar volume ratio between the scattering object and the solvent (υp/υs), NA is the Avogadro number, and mp is the scattering object molar mass.41−43 The peptides’ molar volumes were estimated by adding the molar volume values reported by Jacrot and Zaccai for each amino acid in the sequence, and the peptides’ scattering lengths were calculated using the scattering lengths listed by the same authors for each amino acid.44 For the sample prepared in D2O, all labile hydrogens on the peptides and polymer were assumed to be exchanged with deuterium. For the sample prepared in H2O/D2O mixtures, the scattering length was assumed to be the weighted average between the scattering length density calculated in pure H2O (no exchange) and in D2O (full exchange). The background scattering, Ib, was estimated using the Porod law which gives the scattered intensity of a two-phase system at high q values:41−43
I(q) =
Kp q4
+ Ib
(3)
where Kp is the Porod constant. Ib was estimated by fitting the last 10 data points of the scattering curves using a Porod representation: q4I(q) vs q4 (data not shown). For clarity purposes, the SANS data presented in this article are limited to the q region where the coherent scattering is dominant, while Ib was evaluated at high q values where the background scattering is dominant. To investigate the peptide and polymer structural properties independently, two sets of samples were prepared. First, the peptide fibers were studied by preparing samples using the deuterated conjugate in pure D2O. In this case, the contrast factor between the peptide and the solvent is 2 orders of magnitude higher than the contrast between the deuterated polymer and D2O (Table 1). Therefore, it can be assumed that under these conditions the scattering measured is dominated by the peptide and that the scattering originating from the polymer can be neglected (i.e., the scattering object is the peptide). Second, the behavior of the polymer was studied by preparing samples using the deuterated conjugate in a 10473
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Langmuir
Article
42/58 (vol %) H2O/D2O mixture. In this case, the contrast between the peptide and the solvent is 0 (i.e., the peptide is contrast matched and the scattering object is the monomer) and the scattering observed originates from the polymer component only. Samples were prepared in 2 or 5 mm path quartz cells. The scattering patterns were measured below and above the LCST of PNIPAAm: at 25, 35, and 45 °C upon heating. The samples were then cooled down to 30 °C and their scattering pattern measured once more. The temperature was controlled by an external thermostat with ±3 °C accuracy. Data were corrected for background scattering (intensity scattered by the empty cell and the solvent) and normalized by the incoherent water scattering according to ILL standard procedures (additional information can be obtained from the ILL upon request). Transmission Electron Microscopy (TEM). TEM micrographs were taken using a JEM-1220 electron microscope (JEOL) equipped with a Gatan digital camera. Samples were prepared by diluting the samples to a final concentration of 0.5 mg mL−1 of peptide. Carboncoated copper grids (400 mesh size, Agar scientific) were placed with the carbon-coated face down on top of a 10 μL drop of solution for 10 s and then blotted using Whatman 50 filter paper. The grid was then washed in deionized water by placing it on top of a 10 μL drop of deionized water for 10 s and blotted again. The sample was negatively stained by placing the grids on top of a 10 μL drop of 4% (w/v) solution of uranyl acetate for 60 s and then blotted. After air drying, the sample was examined by TEM at 100 kV accelerating voltage. Atomic Force Microscopy (AFM). The morphology of the composite gel fibers was studied using a Veeco Multimode atomic force microscope with a Nanoscope IIIa controller. Muscovite mica and metal specimen discs (Agar Scientific, U.K.) were used for sample preparation. Samples were prepared by diluting the composite gels to a final concentration of 0.5 mg mL−1 of peptide. A 80 μL aliquot of the sample was deposited onto a freshly cleaved mica surface for 60 s. Excess liquid was removed by capillary action and the surface washed twice with 1 mL of distilled water before being left to air-dry prior to imaging. Olympus high aspect ratio etched silicon probes with a nominal spring constant of 42 N m−1 (Bruker AXS S.A.S., France) were used for imaging. Samples were imaged by tapping mode at a relative humidity 0.5 μm) that scatter visible light. The significantly lower absorbance of the 20−10 sample above the LCST indicates that the formation of large aggregates is significantly reduced. This result again suggests that the peptide from the conjugate participates in the formation of the β-sheet-rich fibers, resulting in the polymer chains being anchored to the peptide fibers, as schematically shown in Figure 1. As a result, the formation of large polymer aggregates above the LCST is hindered and no macroscopic phase separation is observed. It should be noted that although the 10−3 and 10−5 samples are not gels they are highly viscous solutions. Indeed, the formation of elongated βsheet-rich fibers by these peptides was shown to occur at very low concentration, typically 0.5 um) formed.33,46 Therefore, although this process can contribute to the increase in the mechanical properties of the hydrogel discussed above, it is not thought to be dominant. At low q values, an underlying q−1 scattering behavior is observed, which is in agreement with the formation of fibrillar structures by the peptide. Due to the presence of the structural peak (i.e., intermolecular scattering effects), fiber dimensions cannot be extracted from this data. In our previous work, we performed SANS on more diluted samples (i.e., in the diluted regime where intermolecular scattering effects become negligible) and fiber sizes of 3−4 nm were obtained in agreement with the TEM micrographs (Figure 6) and the adoption of a β-sheet conformation by this peptide.34 To investigate the conformation adopted by the polymer, a second set of samples was prepared using a 42/58 (vol %) H2O/D2O mixture as solvent. In this case, the peptide is contrast matched and the scattering observed originates from the polymer only (see the Materials and Methods section for more information). As can be seen from Figure 7B, a q−1 behavior is observed for all samples in the series, suggesting the adoption by the polymer of a fibrillar conformation/arrangement.41−43,47 This result clearly suggests that the polymer does not simply “dangle” on the side of the peptide fiber in a Gaussian conformation but rather surrounds and covers the peptide fiber. This arrangement of the polymer chains agrees with the presence of hydrogen bonding interactions between the peptide fiber and the polymer. In this case, the scattering observed was found to be proportional to the polymer
⎛ q 2R 2 ⎞ σ ⎟ qI(q) = πCpμL exp⎜ − 2 ⎠ ⎝
(4)
where Cp and μL are the polymer concentration and the rod linear mass, respectively. Rσ can be estimated from the slope of the linear section of the curve obtained at low q in a ln[qIA(q)] versus q2 representation (data not shown). For a plain cylinder model, Rσ is related to the diameter, d, of the cylinder through Rσ = (d2/8)1/2.43 In our case, a diameter of d = 5.3 ± 0.5 nm was obtained. This is in good agreement with the fiber diameter estimated for the peptide fiber of 3−4 nm as the polymer covers the peptide fibers. These results clearly point toward the existence of strong interactions between the polymer and the peptide fiber and suggest a relatively “tight” arrangement of the polymer chains around the peptide fibers. These are thought to have a significant effect on the intrinsic properties of the fibers and result in their stiffening. This is thought to be the dominant mechanism explaining the increase in the mechanical properties of the hydrogels with the addition of conjugate. To confirm the polymer still goes through a conformational change at the LCST, SANS experiments were performed on the 20−10 sample as a function of temperature. Samples were heated up to 45 °C and then cooled back down to 30 °C to confirm that the transition is reversible. The scattering patterns obtained for both series of samples are given in Figure 7C and D. The scattering profiles in Figure 7C are a result of scattering from the peptide only and reveal that there is minimal change before and after heating. As the sample is heated to 45 °C, the structural peak intensity decreases, revealing the q−1 behavior typical of the scattering of rod-like structures in agreement with our previous work.34 The decrease of the structural scattering peak intensity was also observed in the pure peptide sample (data not shown). The detailed investigation of the behavior of these peptide hydrogels as a function of temperature is beyond the scope of this article. These results clearly show that the polymer LSCT does not significantly affect the topology of the peptide network. On the other hand, Figure 7D shows the SANS curves where scattering originates from the polymer component only and a significant change in the scattering profile is observed in this case when heating the sample above the LCST. At 45 °C, a 10478
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Langmuir
Article
the LCST of the polymer which is still observed could potentially be used as a trigger for the delivery of drugs or signals.
strong scattering feature is observed which is thought to correspond to a broad structural peak. The structural origin of this strong structural feature has not been fully elucidated, and further work is being undertaken to do so. What these results show, however, is that at the LCST the polymer still goes through a conformational change. The assumption is, as mentioned above, that below the LCST the polymer chains cover the peptide fibers but are swollen by water molecules. At the LCST, the polymer becomes hydrophobic and hence water is expelled and the polymer collapses onto the peptide fibers. Whether the polymer keeps a rod-like arrangement along the peptide fiber or collapses into a tight coil on its side remains an open question, though it is reasonable to think that, if the polymer is already relatively tightly wrapped around the peptide fiber, a simple collapse along the fiber is more favorable. The transition was found to be fully reversible in agreement with our μDSC results. As can be seen from Figure 7C and D, upon cooling, the original scattering patterns are recovered. The LCST transition was found to occur at a slightly higher temperature in the SANS experiments (>35 °C). This is due to the use of deuterated PNIPAAn and D2O. It is well-known that deuterium atoms form stronger hydrogen bonds than hydrogen atoms, leading to the case of thermoresponsive polymers to higher LCSTs.48,49 Finally, the mechanical properties of the composite hydrogel were investigated as a function of temperature to see whether the polymer LCST affected the G′ of the samples. As can be seen from Figure 8, the shear moduli of the hydrogels were found to remain constant when heating the samples from 25 to 45 °C. These results support the interpretations made above. Indeed, as strong interactions already exist between the polymer and the peptide, leading to the polymer chains being already tightly wrapped around the peptide fibers, the further collapse of the polymer chain onto the peptide fibers at the LCST is not significantly further affecting the mechanical properties of the fibers and therefore of the hydrogels. These results also confirm that the stiffening of the peptide fibers by the polymer chains through strong hydrogen bond interactions at room temperature is the main mechanism behind the increase in mechanical properties observed when doping the hydrogels with conjugates.
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail: [email protected]. Phone: +44 (0) 161 306 5781. *E-mail: [email protected]. Phone: +44 (0) 161 306 5981. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors acknowledge the European Commission for funding this work through projects FP7-EST Polyelectrolyte and FP7-NMP BIOCENT. The authors are grateful to the ILL for the beam time allocated and all the staff on D22 for the support with the SANS experiments.
■
REFERENCES
(1) Bowerman, C. J.; Nilsson, B. L. Review self-assembly of amphipathic beta-sheet peptides: Insights and applications. Biopolymers 2012, 98, 169−184. (2) Ulijn, R. V.; Smith, A. M. Designing peptide based nanomaterials. Chem. Soc. Rev. 2008, 37, 664−675. (3) Maude, S.; Ingham, E.; Aggeli, A. Biomimetic self-assembling peptides as scaffolds for soft tissue engineering. Nanomedicine 2013, 8, 823−847. (4) Gelain, F.; Horii, A.; Zhang, S. G. Designer self-assembling peptide scaffolds for 3-D tissue cell cultures and regenerative medicine. Macromol. Biosci. 2007, 7, 544−551. (5) Saracino, G. A. A.; Cigognini, D.; Silva, D.; Caprini, A.; Gelain, F. Nanomaterials design and tests for neural tissue engineering. Chem. Soc. Rev. 2013, 42, 225−262. (6) Woolfson, D. N.; Ryadnov, M. G. Peptide-based fibrous biomaterials: some things old, new and borrowed. Curr. Opin. Chem. Biol. 2006, 10, 559−567. (7) Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Peptide-amphiphile nanofibers: A versatile scaffold for the preparation of self-assembling materials. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 5133−5138. (8) Gazit, E. Peptides as nanomaterials: Self-assembiv and technological applications of peptide nanotubes, nanospheres, hydrogels and other nanostructures. J. Pept. Sci. 2008, 14, 35−35. (9) Tang, C.; Miller, A. F.; Saiani, A. Peptide hydrogels as mucoadhesives for local drug delivery. Int. J. Pharm. 2014, 465, 427−435. (10) Mujeeb, A.; Miller, A. F.; Saiani, A.; Gough, J. E. Self-assembled octapeptide scaffolds for in vitro chondrocyte culture. Acta Biomater. 2013, 9, 4609−4617. (11) Iwasaki, M.; Wilcox, J. T.; Nishimura, Y.; Zweckberger, K.; Suzuki, H.; Wang, J.; Liu, Y.; Karadimas, S. K.; Fehlings, M. G. Synergistic effects of self-assembling peptide and neural stem/ progenitor cells to promote tissue repair and forelimb functional recovery in cervical spinal cord injury. Biomaterials 2014, 35, 2617− 2629. (12) Liu, Y.; Ye, H.; Satkunendrarajah, K.; Yao, G. S.; Bayon, Y.; Fehlings, M. G. A self-assembling peptide reduces glial scarring, attenuates post-traumatic inflammation and promotes neurological recovery following spinal cord injury. Acta Biomater. 2013, 9, 8075− 8088. (13) Brunton, P. A.; Davies, R. P. W.; Burke, J. L.; Smith, A.; Aggeli, A.; Brookes, S. J.; Kirkham, J. Treatment of early caries lesions using biomimetic self-assembling peptides - a clinical safety trial. Br. Dent. J. 2013, 215, E6.
■
CONCLUSION We have investigated the effect of doping FEFEFKFK hydrogels with PNIPAAm−FEFEFKFK conjugate. The peptide from the conjugate was shown to be incorporated into the peptide fiber, resulting in the polymer being anchored to the peptide fiber. The conjugation of the polymer to the peptide and its anchoring does not affect the LCST temperature of the polymer but does result in a decrease in the enthalpy of the transition. In addition, the introduction of the conjugate also results in a significant increase in the G′ of the hydrogels. These results suggest the presence of hydrogen bonding interactions between the peptide and the polymer which consequently lead to the polymer adopting a fibrillar arrangement around the surface of the peptide fiber. The polymer was still found to go through a conformational change at the LCST, suggesting that it collapses onto the peptide fiber, while the fibrillar network was found to be mainly unaffected by the polymer LCST. These changes at the LCST were also found to be fully reversible. Through this work, we have shown that peptideresponsive polymer conjugate can be used to modify the mechanical properties of peptide-based hydrogels. In addition, 10479
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Langmuir
Article
(14) Tian, Y. F.; Devgun, J. M.; Collier, J. H. Fibrillized peptide microgels for cell encapsulation and 3D cell culture. Soft Matter 2011, 7, 6005−6011. (15) Roberts, D.; Rochas, C.; Saiani, A.; Miller, A. F. Effect of Peptide and Guest Charge on the Structural, Mechanical and Release Properties of beta-Sheet Forming Peptides. Langmuir 2012, 28, 16196−16206. (16) Szkolar, L.; Guilbaud, J. B.; Miller, A. F.; Gough, J. E.; Saiani, A. Enzymatically triggered peptide hydrogels for 3D cell encapsulation and culture. J. Pept. Sci. 2014, 20, 578−584. (17) Castillo Diaz, L. A.; Saiani, A.; Gough, J. E.; Miller, A. F. Human osteoblasts within soft peptide hydrogels promote mineralisation in vitro. J. Tissue Eng. 2014, 5, 2041731414539344. (18) Choi, Y. Y.; Joo, M. K.; Sohn, Y. S.; Jeong, B. Soft Matter 2008, 4, 2383−2387. (19) Larsen, T. H.; Branco, M. C.; Rajagopal, K.; Schneider, J. P.; Furst, E. M. Macromolecules 2009, 42, 8443−8450. (20) Nuhn, H.; Klok, H.-A. Secondary Structure Formation and LCST Behavior of Short Elastin-Like Peptides. Biomacromolecules 2008, 9, 2755−2763. (21) Ramanan, R. M. K.; Chellamuthu, P.; Tang, L.; Nguyen, K. T. Biotechnol. Prog. 2006, 22, 118−125. (22) Sabnis, A.; Wadajkar, A. S.; Aswath, P.; Nguyen, K. T. Nanomed.: Nanotechnol., Biol. Med. 2009, 5, 305−315. (23) Zhang, X.-Z.; Chu, C.-C. Polymer 2005, 46, 9664−9673. (24) Xu, X.-D.; Wei, H.; Zhang, X.-Z.; Cheng, S.-X.; Zhuo, R.-X. J. Biomed. Mater. Res., Part A 2006, 81A, 418−426. (25) Gil, E. S.; Hudson, S. M. Biomacromolecules 2007, 8, 258−264. (26) Mano, J. F. Stimuli-Responsive Polymeric Systems for Biomedical Applications. Adv. Eng. Mater. 2008, 10, 515−527. (27) Trzebicka, B.; Robak, B.; Trzcinska, R.; Szweda, D.; Suder, P.; Silberring, J.; Dworak, A. Thermosensitive PNIPAM-peptide conjugate - Synthesis and aggregation. Eur. Polym. J. 2013, 49, 499−509. (28) Maslovskis, A.; Tirelli, N.; Saiani, A.; Miller, A. F. PeptidePNIPAAm conjugate based hydrogels: synthesis and characterisation. Soft Matter 2011, 7, 6025−6033. (29) Stoica, F.; Alexander, C.; Tirelli, N.; Miller, A. F.; Saiani, A. Selective synthesis of double temperature-sensitive polymer-peptide conjugates. Chem. Commun. 2008, 37, 4433−4435. (30) Molawi, K.; Studer, A. Reversible switching of substrate activity of poly-N-isopropylacrylamide peptide conjugates. Chem. Commun. 2007, 48, 5173−5175. (31) Heskins, M.; Guillet, J. E. Solution Properties of Poly(Nisopropylacrylamide). J. Macromol. Sci., Part A: Pure Appl. Chem. 1968, 2, 1441−1455. (32) Schild, H. G. Poly(N-isopropylacrylamide): experiment, theory and application. Prog. Polym. Sci. 1992, 17, 163−249. (33) Boothroyd, S.; Miller, A. F.; Saiani, A. From fibres to networks using self-assembling peptides. Faraday Discuss. 2013, 166, 195−207. (34) Saiani, A.; Mohammed, A.; Frielinghaus, H.; Collins, R.; Hodson, N.; Kielty, C. M.; Sherratt, M. J.; Miller, A. F. Self-assembly and gelation properties of alpha-helix versus beta-sheet forming peptides. Soft Matter 2009, 5, 193−202. (35) Lee, N. R.; Bowerman, C. J.; Nilsson, B. L. Effects of Varied Sequence Pattern on the Self-Assembly of Amphipathic Peptides. Biomacromolecules 2013, 14, 3267−3277. (36) Lee, N. R.; Bowerman, C. J.; Nilsson, B. L. Sequence length determinants for self-assembly of amphipathic beta-sheet peptides. Biopolymers 2013, 100, 738−50. (37) Caplan, M. R.; Schwartzfarb, E. M.; Zhang, S. G.; Kamm, R. D.; Lauffenburger, D. A. Control of self-assembling oligopeptide matrix formation through systematic variation of amino acid sequence. Biomaterials 2002, 23, 219−227. (38) Caplan, M. R.; Schwartzfarb, E. M.; Zhang, S. G.; Kamm, R. D.; Lauffenburger, D. A. Effects of systematic variation of amino acid sequence on the mechanical properties of a self-assembling, oligopeptide biomaterial. J. Biomater. Sci., Polym. Ed. 2002, 13, 225− 236.
(39) Caplan, M. R.; Moore, P. N.; Zhang, S. G.; Kamm, R. D.; Lauffenburger, D. A. Self-assembly of a beta-sheet protein governed by relief of electrostatic repulsion relative to van der Waals attraction. Biomacromolecules 2000, 1, 627−631. (40) Chan, W. C.; White, P. D. Fmoc Solid Phase Peptide Synthesis. A Practical Approach; Oxford University Press: Oxford, U.K., 2000. (41) Guenet, J.-M. Thermoreversible Gelation of Polymers and Biopolymers; Academic Press: London, 1992. (42) Higgins, J. S.; Benoit, H. C. Polymer and Neutron Scattering; Clarendon Press: Oxford, U.K., 1994. (43) Guilbaud, J.-B.; Saiani, A. Using small angle scattering (SAS) to structurally characterise peptide and protein self-assembled materials. Chem. Soc. Rev. 2011, 40, 1200−1210. (44) Jacrot, B.; Zaccai, G. Determination of Molecular-Weight by Neutron-Scattering. Biopolymers 1981, 20, 2413−2426. (45) Schild, H. G. Poly (N-Isopropylacrylamide) - Experiment, Theory and Application. Prog. Polym. Sci. 1992, 17, 163−249. (46) Guilbaud, J.-B.; Rochas, C.; Miller, A. F.; Saiani, A. Effect of Enzyme Concentration of the Morphology and Properties of Enzymatically Triggered Peptide Hydrogels. Biomacromolecules 2013, 14, 1403−1411. (47) Guinier, A.; Fournet, G. Small-Angle Scattering of X-rays; John Wiley & Sons, Inc.: New York, 1955. (48) Sun, B.; Lin, Y.; Wu, P.; Siesler, H. W. A FTIR and 2D-IR Spectroscopic Study on the Microdynamics Phase Separation Mechanism of the Poly(N-isopropylacrylamide) Aqueous Solution. Macromolecules 2008, 41, 1512−1520. (49) Zhang, Y.; Furyk, S.; Bergbreiter, D. E.; Cremer, P. S. Specific Ion Effects on the Water Solubility of Macromolecules: PNIPAM and the Hofmeister Series. J. Am. Chem. Soc. 2005, 127, 14505−14510.
10480
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Self-Assembling Peptide/Thermoresponsive Polymer Composite Hydrogels: Effect of Peptide−Polymer Interactions on Hydrogel Properties A. Maslovskis,† J.-B. Guilbaud,† I. Grillo,‡ N. Hodson,§ A. F. Miller,*,† and A. Saiani*,∥ †
Manchester Institute of Biotechnology and School of Chemical Engineering & Analytical Science, The University of Manchester, Oxford Road, Manchester, M13 9PL, U.K. ‡ Institut Laue-Langevin, 6 rue Jules Horowitz, 38042 Grenoble, France § BioAFM Facility, Stopford Building, The University of Manchester, Oxford Road, Manchester, M13 9PL, U.K. ∥ School of Materials and Manchester Institute of Biotechnology, The University of Manchester, Oxford Road, Manchester, M13 9PL, U.K. ABSTRACT: We have investigated the effect of doping the self-assembling octapeptide FEFEFKFK (F, phenylalanine; E, glutamic acid; K, lysine) hydrogels with various amounts of thermoresponsive conjugate of FEFEFKFK and poly(Nisopropylacrylamide) (PNIPAAm) in order to create novel hydrogels. The samples were characterized using a range of techniques including microdifferential scanning calorimetry (μDSC), oscillatory rheology, transmission electron microscopy (TEM), atomic force microscopy (AFM), and small angle neutron scattering (SANS). The peptide from the conjugate was shown to be incorporated into the peptide fiber, resulting in the polymer being anchored to the peptide fiber. The conjugation of the polymer to the peptide and its anchoring to the peptide fibers did not affect its lower critical solution temperature (LCST). On the other hand, it did result in a decrease in the LCST enthalpy and a significant increase in the G′ of the hydrogels, suggesting the presence of hydrogen bond interactions between the peptide and the polymer. As a result, the polymer was found to adopt a fibrillar arrangement tightly covering the peptide fiber. The polymer was still found to go through a conformational change at the LCST, suggesting that it collapses onto the peptide fiber. On the other hand, the fibrillar network was found to be mainly unaffected by the polymer LCST. These changes at the LCST were also found to be fully reversible. The nature of the interaction between the polymer and the peptide was shown to have a significant effect on the conformation adopted by the polymer around the fibers and the mechanical properties of the hydrogels.
■
INTRODUCTION
amino acids, the properties of the hydrogels can be tailored to the intended application.3,9−17 Self-assembling peptide systems can also be considered as stimuli-responsive materials, as gelation (sol-to-gel transition) and conformational changes can be triggered by various environmental stimuli, such as pH, ionic strength, and the presence of enzymes. In some cases, temperature has been used as a stimulus to trigger gelation, but for the majority of systems, only gel melting is observed upon heating.18,19 Peptides exhibiting a lower critical solution temperature (LCST) have been reported. For example, short elastin-like peptides based on GVGVP (G, glycine; V, valine; P, proline) sequence exhibit LCST behavior in water; however, the LCST values are very high, i.e., ∼100−200 °C.20 For biomedical applications, much lower LCST temperatures are required.
Self-assembling peptide-based materials have attracted significant attention over the past decade due to their inherent biocompatibility, biodegradability, and ability to easily manipulate their chemical functionality and properties simply by changing their primary structure. Such materials are typically formed from short oligopeptides which self-assemble into various secondary structures, including β-sheets, β-hairpins, and α-helices. These tend to further assemble under certain conditions into a variety of higher ordered structures such as micellar aggregates, nanotubes, membranes, or fibrils.1−8 Design rules to control such self-assembling pathways are emerging, and fibril formation, in particular, has been the focus of many studies. Under appropriate conditions, above a critical concentration, these fibrils can self-associate and form selfsupporting hydrogels which have been shown to have potential in a number of applications including controlled drug delivery and tissue regeneration. It has been shown that, by controlling peptide concentration and the distribution and sequence of © 2014 American Chemical Society
Received: June 17, 2014 Revised: August 4, 2014 Published: August 6, 2014 10471
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Langmuir
Article
Figure 1. Schematic representation of the self-assembly mechanism of the β-sheet forming peptide/peptide−polymer conjugate mixtures.
conjugate would incorporate into the fibrillar structure of the matrix, thus decorating the peptide fibers with PNIPAAm polymer, as shown schematically in Figure 1. In this paper, we decided to investigate the effect that doping the peptide hydrogels with peptide−PNIPPAm conjugate has on the hydrogel properties, mechanical and morphological. In particular, we investigated the role played by peptide−polymer interactions on the peptide fibrillar network morphology and on the LCST behavior of the polymer. For this purpose, the composite hydrogels were investigated using a combination of techniques including microdifferential scanning calorimetry (μDSC) oscillatory rheology, transmission electron microscopy (TEM), atomic force microscopy (AFM), and small angle neutron scattering (SANS).
One strategy to introduce thermoresponsiveness in hydrogels consists of introducing thermoresponsive micelles or nanoparticles into a nonthermoresponsive gel matrix, or into an already thermoresponsive gel to accentuate its thermoresponsiveness and improve gel network swelling/shrinking and mechanical properties. Nguyen et al.21 developed a thermoresponsive composite gel by incorporating poly(N-isopropylacrylamide) (PNIPAAm) nanoparticles into poly(ethylene oxide) (PEG) matrix, and the results showed that these gels were able to release bovine serum albumin (BSA) above the LCST. Later, the same group prepared thermoresponsive hydrogels, where poly(N-isopropylacrylamide-co-acrylamide) (PNIPAAm-AAm) nanoparticles, preloaded with drug, were incorporated into a photo-cross-linkable poly(ethylene glycol) diacrylate (PEGDA) matrix.22 Above the LCST, the drug release occurred due to the collapse of PNIPAAm. In one other study, poly(N-isopropylacrylamide)/poly(ethylene glycol) diacrylate (PNIPAAm/PEG-DA) microgels were incorporated into PNIPAAm hydrogels. This additive improved gel swelling/ shrinking ratios and thermal behavior.23 Zhang et al.24 reported composite gels, containing poly(N-isopropylacrylamide)-bpoly(methyl methacrylate) (PNIPAAm-b-PMMA) micelles embedded in a PNIPAAm hydrogel. Gels showed faster shrinking and slower swelling rates with the increase of micelle content. When drug loaded micelles were used, drug release occurred in a more controlled fashion in comparison to the pure PNIPAAm gel. Hudson et al.25 prepared interpenetrating networks of PNIPAAm and silk fibroin. These networks were reported to possess improved mechanical stability and faster deswelling kinetics than ordinary PNIPAAm gels. One emerging strategy to introduce thermoresponsiveness into peptide hydrogels is to conjugate peptides to responsive polymers such as PNIPAAm. Such peptide hybrids represent a class of materials with interesting properties and are attractive for biomaterials applications, as they combine the controlled chemical, mechanical, and thermal properties of a polymer with the structural properties and functionality of a peptide.26−30 In this study, we have combined the non-LCST-possessing FEFEFKFK peptide matrix with the thermoresponsive conjugate of FEFEFKFK and PNIPAAm to develop a composite gel. The prototypical thermoresponsive polymer PNIPAAm has been chosen, as it possesses a sharp coil-toglobule conformational change (i.e., LCST) at temperatures close to body temperature.31,32 The octapeptide FEFEFKFK was selected, as it is well-known to exhibit self-assembly in aqueous solution and form β-sheet-rich fibrillar networks and hydrogels at low concentrations (∼10−20 mg mL−1).33−39 It was postulated that the peptide part of the polymer−peptide
■
MATERIALS AND METHODS
Peptide Synthesis. The peptide FEFEFKFK (F, phenylalanine; E, glutamic acid; K, lysine) was synthesized on a ChemTech ACT 90 peptide synthesizer (Advance ChemTech Ltd, Cambridgeshire, U.K.) using N-methyl-2-pyrrolidone (NMP) as solvent and standard solid phase peptide synthesis protocols.40 The side chain protected amino acids, amino acid activator (HCTU), and preloaded Fmoc-Lys(Boc)Wang resin (bead size, 100−200 mesh; substitution, 0.70 mmol g−1) were purchased from Novabiochem (Merck), 3-mercaptopropionic acid was purchased from Acros Organics, and all other solvents were purchased from Aldrich. All chemicals were used as received. The functionalization of the peptide with a thiol end-group (SHFEFEFKFK) was performed directly on the resin using a procedure described in detail elsewhere.29 Once synthesized, both peptides were cleaved from the resin and deprotected using a mixture of trifluoroacetic acid (TFA) and anisole (95:5). The resin was removed by filtration, and the peptides were precipitated in cold diethyl ether. The peptides were redissolved in water and washed a further four times in diethyl ether before they were isolated via freeze-drying. Reverse-phase high performance liquid chromatography (RP-HPLC), mass spectrometry (MS), and proton nuclear magnetic resonance spectroscopy (1H NMR) were used to confirm the peptide structures and determine their purities which were typically >90%. Conjugate Synthesis. The PNIPAAm−FEFEFKFK conjugate was synthesized via free radical polymerization (FRP) using the thiolfunctionalized peptide SH-FEFEFKFK as a chain transfer agent.28 NIsopropylacrylamide (NIPAAm) (Aldrich, 97%) and azo-iso-butyronitrile (AIBN) (Fisher Scientific, laboratory grade) were recrystallized prior to use from hexane and methanol, respectively. 4.42 mmol of NIPAAm, 0.0614 mmol of AIBN, and 0.0935 mmol of SH-FEFEFKFK were dissolved in 30 mL of deoxygenated dimethyl sulfoxide (DMSO, Aldrich, used as received) solution. The reaction was performed at 65 °C for 24 h under inert atmosphere (N2) and constant stirring. The conjugate/polymer mixture was purified by diluting the reaction mixture to a total volume of 0.5 L using deionized water and dialyzing for 4 days using a membrane with a molecular weight cutoff of 3500 g 10472
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Langmuir
Article
Table 1. Molar Volume, Molar Mass, Scattering Length, and Contrast Factor of Compound Used
a b
compound
scattering length (10−12 × cm)
molar mass (g mol−1)
molar volume (mL mol−1)
contrast factor (10−3 × cm2 g−1)
H2O D2 O FEFEFKFK in H2O FEFEFKFK in D2O NIPAAm in H2O NIPAAm in D2O NIPAAm-d7 in H2O NIPAAm-d7 in D2O
−0.17 1.91 27.1 43.7b 1.39 2.44b 9.73 8.68b
18 20b 1120 1136b 113 114b 120 121b
18.0 18.1 871a
N/A N/A 0.60 1.09 0.22 1.86 3.75 0.03
83 83.8
The peptide molar volume was estimated by adding the molar volume values reported by Zaccai et al. for each amino acid in the peptide sequence. The scattering lengths and molar masses were calculated assuming that in D2O all labile hydrogens are exchanged with deuterium.44
mol−1 (Medicell Ltd., U.K.). The dialysis allowed the removal of short conjugate/polymer chains and any unreacted reagents and peptides. The nonconjugated polymer was removed by centrifugation using a procedure described in detail elsewhere.28 The conjugate fraction was collected and freeze-dried to give a white powder. The conjugate molecular weights were measured by gel permeation chromatography (GPC) using PEG/PEO standards. Mass-average (Mw) and numberaverage (Mn) molecular weights of 8800 ± 500 and 2500 ± 500 g mol−1 were obtained, respectively. The peptide weight fraction in the conjugates was estimated by 1H NMR using physical mixtures of peptide and NIPAAm as the reference following a procedure described in detail in elsewhere.28 The weight fraction of peptide in the conjugate was estimated to be 0.12. The deuterated conjugate was prepared using the exact same methodology as described above by replacing NIPAAM by its deuterated counterpart, NIPAAm-d7 (Polymer Source Inc., 98%). Macroscopic gelation behavior and μDSC were used to ensure that the deuterated conjugate behaved in the same fashion as its protonated partner. Sample Preparation and Phase Behavior. Samples were prepared by dissolving the desired quantities of material in 1 mL of deionized water at room temperature to avoid any interference by the polymer LCST on the gelation process. The samples’ pH was then adjusted to 4 by adding a small amount of a 1 M NaOH solution. The samples were subsequently left to equilibrate at room temperature for 12 h prior to any measurement. The macroscopic phase behavior of the samples was studied as a function of temperature by placing the samples in a temperature-controlled water bath and allowing 15 min of equilibration at each temperature before visual inspection. Microdifferential Scanning Calorimetry (μDSC). μDSC measurements were performed on a SETARAM μDSC III instrument. The sample cell was filled with 0.6 mL of solution using a micropipette. The reference cell was filled with water and its weight adjusted using a microbalance to ensure an identical mass of solution was present in both the sample and reference cells. μDSC thermographs were recorded using a scanning rate of 1.0 °C min−1 in the temperature range 20−80 °C. Four heating/cooling cycles were performed for each sample. Onset temperature and transition enthalpies were determined using Setsoft 2000 software supplied with the instrument. Each experiment was repeated at least three times to ensure reproducibility. Oscillatory Rheology. Rheological studies were performed on a stress-controlled Bohlin C-CVO rheometer, equipped with a Peltier temperature controller. A parallel plate geometry (⌀ = 20 mm; 0.25 mm gap) was used with a solvent trap to minimize water evaporation. Initially, strain amplitude sweeps (γ = 0.001−10) were performed at constant angular frequency (ω = 1 rad s−1) to identify the linear viscoelastic region (LVR). Subsequently, frequency sweeps (ω = 0.1− 10 rad s−1) at constant strain amplitude (γ = 0.01) were performed to determine the elastic (G′) and viscous (G″) moduli within the LVR. All measurements were performed at 20 °C, and all measurements were repeated at least three times to ensure reproducibility. Small Angle Neutron Scattering (SANS). SANS experiments were performed on the D22 beamline at the Institut Laue-Langevin (ILL) in Grenoble, France. A wavelength of λ = 6 Å was used, and the data were collected on a (3He) multidetector with an active area of 100
× 100 cm2 and a pixel size of 0.8 × 0.8 cm2. By varying the sample− detector distance, the available q was in the range 0.1 < q (nm−1) < 5, with q = (4π/λ) sin(θ/2), where θ is the scattering angle. For a binary system (scattering objects/solvent), the coherent intensity scattered by the scattering objects, in our case the peptide or monomer depending on the sample used (see below for more details), in absolute units is
IA(q) =
1 (I N(q) − (1 − C p)ID(q) − Ib) K
(1)
where IN(q) and ID(q) are the normalized intensity scattered by the sample and the solvent, respectively, Cp is the scattering object concentration in g cm−3, Ib is the background scattering due to incoherent scattering of the hydrogenated peptide, and K is the contrast factor:
K=
(a p − Ypsas)2 NA mp2
(2)
where ap and as are the scattering lengths of the scattering object and the solvent, respectively, Yps is the molar volume ratio between the scattering object and the solvent (υp/υs), NA is the Avogadro number, and mp is the scattering object molar mass.41−43 The peptides’ molar volumes were estimated by adding the molar volume values reported by Jacrot and Zaccai for each amino acid in the sequence, and the peptides’ scattering lengths were calculated using the scattering lengths listed by the same authors for each amino acid.44 For the sample prepared in D2O, all labile hydrogens on the peptides and polymer were assumed to be exchanged with deuterium. For the sample prepared in H2O/D2O mixtures, the scattering length was assumed to be the weighted average between the scattering length density calculated in pure H2O (no exchange) and in D2O (full exchange). The background scattering, Ib, was estimated using the Porod law which gives the scattered intensity of a two-phase system at high q values:41−43
I(q) =
Kp q4
+ Ib
(3)
where Kp is the Porod constant. Ib was estimated by fitting the last 10 data points of the scattering curves using a Porod representation: q4I(q) vs q4 (data not shown). For clarity purposes, the SANS data presented in this article are limited to the q region where the coherent scattering is dominant, while Ib was evaluated at high q values where the background scattering is dominant. To investigate the peptide and polymer structural properties independently, two sets of samples were prepared. First, the peptide fibers were studied by preparing samples using the deuterated conjugate in pure D2O. In this case, the contrast factor between the peptide and the solvent is 2 orders of magnitude higher than the contrast between the deuterated polymer and D2O (Table 1). Therefore, it can be assumed that under these conditions the scattering measured is dominated by the peptide and that the scattering originating from the polymer can be neglected (i.e., the scattering object is the peptide). Second, the behavior of the polymer was studied by preparing samples using the deuterated conjugate in a 10473
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Langmuir
Article
42/58 (vol %) H2O/D2O mixture. In this case, the contrast between the peptide and the solvent is 0 (i.e., the peptide is contrast matched and the scattering object is the monomer) and the scattering observed originates from the polymer component only. Samples were prepared in 2 or 5 mm path quartz cells. The scattering patterns were measured below and above the LCST of PNIPAAm: at 25, 35, and 45 °C upon heating. The samples were then cooled down to 30 °C and their scattering pattern measured once more. The temperature was controlled by an external thermostat with ±3 °C accuracy. Data were corrected for background scattering (intensity scattered by the empty cell and the solvent) and normalized by the incoherent water scattering according to ILL standard procedures (additional information can be obtained from the ILL upon request). Transmission Electron Microscopy (TEM). TEM micrographs were taken using a JEM-1220 electron microscope (JEOL) equipped with a Gatan digital camera. Samples were prepared by diluting the samples to a final concentration of 0.5 mg mL−1 of peptide. Carboncoated copper grids (400 mesh size, Agar scientific) were placed with the carbon-coated face down on top of a 10 μL drop of solution for 10 s and then blotted using Whatman 50 filter paper. The grid was then washed in deionized water by placing it on top of a 10 μL drop of deionized water for 10 s and blotted again. The sample was negatively stained by placing the grids on top of a 10 μL drop of 4% (w/v) solution of uranyl acetate for 60 s and then blotted. After air drying, the sample was examined by TEM at 100 kV accelerating voltage. Atomic Force Microscopy (AFM). The morphology of the composite gel fibers was studied using a Veeco Multimode atomic force microscope with a Nanoscope IIIa controller. Muscovite mica and metal specimen discs (Agar Scientific, U.K.) were used for sample preparation. Samples were prepared by diluting the composite gels to a final concentration of 0.5 mg mL−1 of peptide. A 80 μL aliquot of the sample was deposited onto a freshly cleaved mica surface for 60 s. Excess liquid was removed by capillary action and the surface washed twice with 1 mL of distilled water before being left to air-dry prior to imaging. Olympus high aspect ratio etched silicon probes with a nominal spring constant of 42 N m−1 (Bruker AXS S.A.S., France) were used for imaging. Samples were imaged by tapping mode at a relative humidity 0.5 μm) that scatter visible light. The significantly lower absorbance of the 20−10 sample above the LCST indicates that the formation of large aggregates is significantly reduced. This result again suggests that the peptide from the conjugate participates in the formation of the β-sheet-rich fibers, resulting in the polymer chains being anchored to the peptide fibers, as schematically shown in Figure 1. As a result, the formation of large polymer aggregates above the LCST is hindered and no macroscopic phase separation is observed. It should be noted that although the 10−3 and 10−5 samples are not gels they are highly viscous solutions. Indeed, the formation of elongated βsheet-rich fibers by these peptides was shown to occur at very low concentration, typically 0.5 um) formed.33,46 Therefore, although this process can contribute to the increase in the mechanical properties of the hydrogel discussed above, it is not thought to be dominant. At low q values, an underlying q−1 scattering behavior is observed, which is in agreement with the formation of fibrillar structures by the peptide. Due to the presence of the structural peak (i.e., intermolecular scattering effects), fiber dimensions cannot be extracted from this data. In our previous work, we performed SANS on more diluted samples (i.e., in the diluted regime where intermolecular scattering effects become negligible) and fiber sizes of 3−4 nm were obtained in agreement with the TEM micrographs (Figure 6) and the adoption of a β-sheet conformation by this peptide.34 To investigate the conformation adopted by the polymer, a second set of samples was prepared using a 42/58 (vol %) H2O/D2O mixture as solvent. In this case, the peptide is contrast matched and the scattering observed originates from the polymer only (see the Materials and Methods section for more information). As can be seen from Figure 7B, a q−1 behavior is observed for all samples in the series, suggesting the adoption by the polymer of a fibrillar conformation/arrangement.41−43,47 This result clearly suggests that the polymer does not simply “dangle” on the side of the peptide fiber in a Gaussian conformation but rather surrounds and covers the peptide fiber. This arrangement of the polymer chains agrees with the presence of hydrogen bonding interactions between the peptide fiber and the polymer. In this case, the scattering observed was found to be proportional to the polymer
⎛ q 2R 2 ⎞ σ ⎟ qI(q) = πCpμL exp⎜ − 2 ⎠ ⎝
(4)
where Cp and μL are the polymer concentration and the rod linear mass, respectively. Rσ can be estimated from the slope of the linear section of the curve obtained at low q in a ln[qIA(q)] versus q2 representation (data not shown). For a plain cylinder model, Rσ is related to the diameter, d, of the cylinder through Rσ = (d2/8)1/2.43 In our case, a diameter of d = 5.3 ± 0.5 nm was obtained. This is in good agreement with the fiber diameter estimated for the peptide fiber of 3−4 nm as the polymer covers the peptide fibers. These results clearly point toward the existence of strong interactions between the polymer and the peptide fiber and suggest a relatively “tight” arrangement of the polymer chains around the peptide fibers. These are thought to have a significant effect on the intrinsic properties of the fibers and result in their stiffening. This is thought to be the dominant mechanism explaining the increase in the mechanical properties of the hydrogels with the addition of conjugate. To confirm the polymer still goes through a conformational change at the LCST, SANS experiments were performed on the 20−10 sample as a function of temperature. Samples were heated up to 45 °C and then cooled back down to 30 °C to confirm that the transition is reversible. The scattering patterns obtained for both series of samples are given in Figure 7C and D. The scattering profiles in Figure 7C are a result of scattering from the peptide only and reveal that there is minimal change before and after heating. As the sample is heated to 45 °C, the structural peak intensity decreases, revealing the q−1 behavior typical of the scattering of rod-like structures in agreement with our previous work.34 The decrease of the structural scattering peak intensity was also observed in the pure peptide sample (data not shown). The detailed investigation of the behavior of these peptide hydrogels as a function of temperature is beyond the scope of this article. These results clearly show that the polymer LSCT does not significantly affect the topology of the peptide network. On the other hand, Figure 7D shows the SANS curves where scattering originates from the polymer component only and a significant change in the scattering profile is observed in this case when heating the sample above the LCST. At 45 °C, a 10478
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Langmuir
Article
the LCST of the polymer which is still observed could potentially be used as a trigger for the delivery of drugs or signals.
strong scattering feature is observed which is thought to correspond to a broad structural peak. The structural origin of this strong structural feature has not been fully elucidated, and further work is being undertaken to do so. What these results show, however, is that at the LCST the polymer still goes through a conformational change. The assumption is, as mentioned above, that below the LCST the polymer chains cover the peptide fibers but are swollen by water molecules. At the LCST, the polymer becomes hydrophobic and hence water is expelled and the polymer collapses onto the peptide fibers. Whether the polymer keeps a rod-like arrangement along the peptide fiber or collapses into a tight coil on its side remains an open question, though it is reasonable to think that, if the polymer is already relatively tightly wrapped around the peptide fiber, a simple collapse along the fiber is more favorable. The transition was found to be fully reversible in agreement with our μDSC results. As can be seen from Figure 7C and D, upon cooling, the original scattering patterns are recovered. The LCST transition was found to occur at a slightly higher temperature in the SANS experiments (>35 °C). This is due to the use of deuterated PNIPAAn and D2O. It is well-known that deuterium atoms form stronger hydrogen bonds than hydrogen atoms, leading to the case of thermoresponsive polymers to higher LCSTs.48,49 Finally, the mechanical properties of the composite hydrogel were investigated as a function of temperature to see whether the polymer LCST affected the G′ of the samples. As can be seen from Figure 8, the shear moduli of the hydrogels were found to remain constant when heating the samples from 25 to 45 °C. These results support the interpretations made above. Indeed, as strong interactions already exist between the polymer and the peptide, leading to the polymer chains being already tightly wrapped around the peptide fibers, the further collapse of the polymer chain onto the peptide fibers at the LCST is not significantly further affecting the mechanical properties of the fibers and therefore of the hydrogels. These results also confirm that the stiffening of the peptide fibers by the polymer chains through strong hydrogen bond interactions at room temperature is the main mechanism behind the increase in mechanical properties observed when doping the hydrogels with conjugates.
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail: [email protected]. Phone: +44 (0) 161 306 5781. *E-mail: [email protected]. Phone: +44 (0) 161 306 5981. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors acknowledge the European Commission for funding this work through projects FP7-EST Polyelectrolyte and FP7-NMP BIOCENT. The authors are grateful to the ILL for the beam time allocated and all the staff on D22 for the support with the SANS experiments.
■
REFERENCES
(1) Bowerman, C. J.; Nilsson, B. L. Review self-assembly of amphipathic beta-sheet peptides: Insights and applications. Biopolymers 2012, 98, 169−184. (2) Ulijn, R. V.; Smith, A. M. Designing peptide based nanomaterials. Chem. Soc. Rev. 2008, 37, 664−675. (3) Maude, S.; Ingham, E.; Aggeli, A. Biomimetic self-assembling peptides as scaffolds for soft tissue engineering. Nanomedicine 2013, 8, 823−847. (4) Gelain, F.; Horii, A.; Zhang, S. G. Designer self-assembling peptide scaffolds for 3-D tissue cell cultures and regenerative medicine. Macromol. Biosci. 2007, 7, 544−551. (5) Saracino, G. A. A.; Cigognini, D.; Silva, D.; Caprini, A.; Gelain, F. Nanomaterials design and tests for neural tissue engineering. Chem. Soc. Rev. 2013, 42, 225−262. (6) Woolfson, D. N.; Ryadnov, M. G. Peptide-based fibrous biomaterials: some things old, new and borrowed. Curr. Opin. Chem. Biol. 2006, 10, 559−567. (7) Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Peptide-amphiphile nanofibers: A versatile scaffold for the preparation of self-assembling materials. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 5133−5138. (8) Gazit, E. Peptides as nanomaterials: Self-assembiv and technological applications of peptide nanotubes, nanospheres, hydrogels and other nanostructures. J. Pept. Sci. 2008, 14, 35−35. (9) Tang, C.; Miller, A. F.; Saiani, A. Peptide hydrogels as mucoadhesives for local drug delivery. Int. J. Pharm. 2014, 465, 427−435. (10) Mujeeb, A.; Miller, A. F.; Saiani, A.; Gough, J. E. Self-assembled octapeptide scaffolds for in vitro chondrocyte culture. Acta Biomater. 2013, 9, 4609−4617. (11) Iwasaki, M.; Wilcox, J. T.; Nishimura, Y.; Zweckberger, K.; Suzuki, H.; Wang, J.; Liu, Y.; Karadimas, S. K.; Fehlings, M. G. Synergistic effects of self-assembling peptide and neural stem/ progenitor cells to promote tissue repair and forelimb functional recovery in cervical spinal cord injury. Biomaterials 2014, 35, 2617− 2629. (12) Liu, Y.; Ye, H.; Satkunendrarajah, K.; Yao, G. S.; Bayon, Y.; Fehlings, M. G. A self-assembling peptide reduces glial scarring, attenuates post-traumatic inflammation and promotes neurological recovery following spinal cord injury. Acta Biomater. 2013, 9, 8075− 8088. (13) Brunton, P. A.; Davies, R. P. W.; Burke, J. L.; Smith, A.; Aggeli, A.; Brookes, S. J.; Kirkham, J. Treatment of early caries lesions using biomimetic self-assembling peptides - a clinical safety trial. Br. Dent. J. 2013, 215, E6.
■
CONCLUSION We have investigated the effect of doping FEFEFKFK hydrogels with PNIPAAm−FEFEFKFK conjugate. The peptide from the conjugate was shown to be incorporated into the peptide fiber, resulting in the polymer being anchored to the peptide fiber. The conjugation of the polymer to the peptide and its anchoring does not affect the LCST temperature of the polymer but does result in a decrease in the enthalpy of the transition. In addition, the introduction of the conjugate also results in a significant increase in the G′ of the hydrogels. These results suggest the presence of hydrogen bonding interactions between the peptide and the polymer which consequently lead to the polymer adopting a fibrillar arrangement around the surface of the peptide fiber. The polymer was still found to go through a conformational change at the LCST, suggesting that it collapses onto the peptide fiber, while the fibrillar network was found to be mainly unaffected by the polymer LCST. These changes at the LCST were also found to be fully reversible. Through this work, we have shown that peptideresponsive polymer conjugate can be used to modify the mechanical properties of peptide-based hydrogels. In addition, 10479
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
Langmuir
Article
(14) Tian, Y. F.; Devgun, J. M.; Collier, J. H. Fibrillized peptide microgels for cell encapsulation and 3D cell culture. Soft Matter 2011, 7, 6005−6011. (15) Roberts, D.; Rochas, C.; Saiani, A.; Miller, A. F. Effect of Peptide and Guest Charge on the Structural, Mechanical and Release Properties of beta-Sheet Forming Peptides. Langmuir 2012, 28, 16196−16206. (16) Szkolar, L.; Guilbaud, J. B.; Miller, A. F.; Gough, J. E.; Saiani, A. Enzymatically triggered peptide hydrogels for 3D cell encapsulation and culture. J. Pept. Sci. 2014, 20, 578−584. (17) Castillo Diaz, L. A.; Saiani, A.; Gough, J. E.; Miller, A. F. Human osteoblasts within soft peptide hydrogels promote mineralisation in vitro. J. Tissue Eng. 2014, 5, 2041731414539344. (18) Choi, Y. Y.; Joo, M. K.; Sohn, Y. S.; Jeong, B. Soft Matter 2008, 4, 2383−2387. (19) Larsen, T. H.; Branco, M. C.; Rajagopal, K.; Schneider, J. P.; Furst, E. M. Macromolecules 2009, 42, 8443−8450. (20) Nuhn, H.; Klok, H.-A. Secondary Structure Formation and LCST Behavior of Short Elastin-Like Peptides. Biomacromolecules 2008, 9, 2755−2763. (21) Ramanan, R. M. K.; Chellamuthu, P.; Tang, L.; Nguyen, K. T. Biotechnol. Prog. 2006, 22, 118−125. (22) Sabnis, A.; Wadajkar, A. S.; Aswath, P.; Nguyen, K. T. Nanomed.: Nanotechnol., Biol. Med. 2009, 5, 305−315. (23) Zhang, X.-Z.; Chu, C.-C. Polymer 2005, 46, 9664−9673. (24) Xu, X.-D.; Wei, H.; Zhang, X.-Z.; Cheng, S.-X.; Zhuo, R.-X. J. Biomed. Mater. Res., Part A 2006, 81A, 418−426. (25) Gil, E. S.; Hudson, S. M. Biomacromolecules 2007, 8, 258−264. (26) Mano, J. F. Stimuli-Responsive Polymeric Systems for Biomedical Applications. Adv. Eng. Mater. 2008, 10, 515−527. (27) Trzebicka, B.; Robak, B.; Trzcinska, R.; Szweda, D.; Suder, P.; Silberring, J.; Dworak, A. Thermosensitive PNIPAM-peptide conjugate - Synthesis and aggregation. Eur. Polym. J. 2013, 49, 499−509. (28) Maslovskis, A.; Tirelli, N.; Saiani, A.; Miller, A. F. PeptidePNIPAAm conjugate based hydrogels: synthesis and characterisation. Soft Matter 2011, 7, 6025−6033. (29) Stoica, F.; Alexander, C.; Tirelli, N.; Miller, A. F.; Saiani, A. Selective synthesis of double temperature-sensitive polymer-peptide conjugates. Chem. Commun. 2008, 37, 4433−4435. (30) Molawi, K.; Studer, A. Reversible switching of substrate activity of poly-N-isopropylacrylamide peptide conjugates. Chem. Commun. 2007, 48, 5173−5175. (31) Heskins, M.; Guillet, J. E. Solution Properties of Poly(Nisopropylacrylamide). J. Macromol. Sci., Part A: Pure Appl. Chem. 1968, 2, 1441−1455. (32) Schild, H. G. Poly(N-isopropylacrylamide): experiment, theory and application. Prog. Polym. Sci. 1992, 17, 163−249. (33) Boothroyd, S.; Miller, A. F.; Saiani, A. From fibres to networks using self-assembling peptides. Faraday Discuss. 2013, 166, 195−207. (34) Saiani, A.; Mohammed, A.; Frielinghaus, H.; Collins, R.; Hodson, N.; Kielty, C. M.; Sherratt, M. J.; Miller, A. F. Self-assembly and gelation properties of alpha-helix versus beta-sheet forming peptides. Soft Matter 2009, 5, 193−202. (35) Lee, N. R.; Bowerman, C. J.; Nilsson, B. L. Effects of Varied Sequence Pattern on the Self-Assembly of Amphipathic Peptides. Biomacromolecules 2013, 14, 3267−3277. (36) Lee, N. R.; Bowerman, C. J.; Nilsson, B. L. Sequence length determinants for self-assembly of amphipathic beta-sheet peptides. Biopolymers 2013, 100, 738−50. (37) Caplan, M. R.; Schwartzfarb, E. M.; Zhang, S. G.; Kamm, R. D.; Lauffenburger, D. A. Control of self-assembling oligopeptide matrix formation through systematic variation of amino acid sequence. Biomaterials 2002, 23, 219−227. (38) Caplan, M. R.; Schwartzfarb, E. M.; Zhang, S. G.; Kamm, R. D.; Lauffenburger, D. A. Effects of systematic variation of amino acid sequence on the mechanical properties of a self-assembling, oligopeptide biomaterial. J. Biomater. Sci., Polym. Ed. 2002, 13, 225− 236.
(39) Caplan, M. R.; Moore, P. N.; Zhang, S. G.; Kamm, R. D.; Lauffenburger, D. A. Self-assembly of a beta-sheet protein governed by relief of electrostatic repulsion relative to van der Waals attraction. Biomacromolecules 2000, 1, 627−631. (40) Chan, W. C.; White, P. D. Fmoc Solid Phase Peptide Synthesis. A Practical Approach; Oxford University Press: Oxford, U.K., 2000. (41) Guenet, J.-M. Thermoreversible Gelation of Polymers and Biopolymers; Academic Press: London, 1992. (42) Higgins, J. S.; Benoit, H. C. Polymer and Neutron Scattering; Clarendon Press: Oxford, U.K., 1994. (43) Guilbaud, J.-B.; Saiani, A. Using small angle scattering (SAS) to structurally characterise peptide and protein self-assembled materials. Chem. Soc. Rev. 2011, 40, 1200−1210. (44) Jacrot, B.; Zaccai, G. Determination of Molecular-Weight by Neutron-Scattering. Biopolymers 1981, 20, 2413−2426. (45) Schild, H. G. Poly (N-Isopropylacrylamide) - Experiment, Theory and Application. Prog. Polym. Sci. 1992, 17, 163−249. (46) Guilbaud, J.-B.; Rochas, C.; Miller, A. F.; Saiani, A. Effect of Enzyme Concentration of the Morphology and Properties of Enzymatically Triggered Peptide Hydrogels. Biomacromolecules 2013, 14, 1403−1411. (47) Guinier, A.; Fournet, G. Small-Angle Scattering of X-rays; John Wiley & Sons, Inc.: New York, 1955. (48) Sun, B.; Lin, Y.; Wu, P.; Siesler, H. W. A FTIR and 2D-IR Spectroscopic Study on the Microdynamics Phase Separation Mechanism of the Poly(N-isopropylacrylamide) Aqueous Solution. Macromolecules 2008, 41, 1512−1520. (49) Zhang, Y.; Furyk, S.; Bergbreiter, D. E.; Cremer, P. S. Specific Ion Effects on the Water Solubility of Macromolecules: PNIPAM and the Hofmeister Series. J. Am. Chem. Soc. 2005, 127, 14505−14510.
10480
dx.doi.org/10.1021/la502358b | Langmuir 2014, 30, 10471−10480
