Chemoresponsive surface-tethered polypeptide brushes based on switchable secondary conformations

Abstract

Surface-tethered chemoresponsive polypeptides prepared by surface-initiated vapor deposition polymerization (SI-VDP) were used to investigate the conversion efficiency between α-helical and β-sheet conformations of poly(γ-methyl-L-glutamate) (PMLG) homopolypeptide and PMLG-b-poly(β-benzyl-L-glutamate) (PBLG) copolypeptide brushes. The end-grafted PMLG brush predominantly adopted a β-sheet conformation after exposure to formic acid vapor, and it was converted back to a predominantly α-helical conformation after treatment with a mixture of dichloroacetic acid (DCA) and chloroform (CHCl₃). This conformational inter-conversion process was reversible and repeatable for a number of cycles. Factors that affected the conversion efficiency between α-helix and β-sheet were studied, including PMLG chain mobility, concentration of DCA, and temperature of formic acid vapor. For end-grafted PMLG brushes with one end chain restriction, 60% conversion efficiency from α to β was achieved with 40 °C formic acid vapor treatment for 2 days, while for the PMLG in end-grafted PMLG-b-PBLG brushes, 40% conversion was achieved after 9 days, which was slower than end-grafted PMLG brushes due to the additional chain restriction from the PBLG block. In comparison, in spin-coated PMLG films in which the PMLG chains had no end chain restrictions, 80% conversion was achieved within 2 days in the same condition. The effects of solvent on the morphologies of surface-grafted PMLG brushes were also observed. The changes of the properties of these films, including surface wettability, thickness, and surface morphologies, were characterized by contact angle measurement, ellipsometry, and atomic force microscopy, respectively.

---

Introduction

Surface-initiated ring opening polymerization (SI-ROP) has been developed as a powerful synthetic strategy to prepare polypeptides brushes on solid substrates in the last two decades.^1,2 The study of stimuli-responsive polypeptides at interfaces, perhaps inspired by their behavior observed in different biological systems, has been of considerable interest.^3,4 For example, because proteins in cell membranes often help to regulate the rate of transport of materials into and out of the cell, the permeability of synthetic peptides has been widely investigated for potential applications in artificial membranes and biosensors.^5–8 In addition, the aggregation or mal-folding of peptides, involving a conformational transition, is associated with particular genetic diseases, such as the β-amyloid with Alzheimer's disease,^9–11 or the prion with Creutzfeldt–Jakob disease in humans, as well as bovine spongiform encephalopathy (known as “mad cow disease”) in cattle.^12,13 The observation of changes in biological activity due to conformational changes of peptides implies that differences in the fundamental physical properties exist between normal and aberrant forms. Inspired by natural phenomena, complex architectures based on β-sheet peptides are employed to mimic and investigate these biological systems.^14,15 Our focus in this study is to explore the transformation behavior of biological and bio-related materials on surfaces, because we anticipate that the surface properties could be greatly influenced by simply altering their conformation.^16–19

To simplify the system, we started with poly(γ-methyl-L-glutamate) (PMLG) and PMLG-b-poly(β-benzyl-L-glutamate) (PBLG) polypeptides as model compounds and studied their secondary structure transformation on solid substrates. In the polyglutamates (PG) family, the side chain structures of PBLG and PMLG differ by a benzyl group; PBLG has the greatest rigidity and α-helical stability, while PMLG has the least.^20–22 Many attempts have been made to alter the secondary conformation of PBLG and PMLG in films through exposure to various solvents, with the objective of modifying the surface properties such as the packing density, viscosity, mechanical strength, or surface energy.^23–26 The conformation of PBLG film is not affected by the chosen solvents. In contrast, the predominant conformation of PMLG at the air–water interface or in solid film was determined by the particular solvent used to form the sample.²⁷ For example, the α-helical conformation predominates in PMLG when helicogenic solvents such as CHCl₃, dimethyl formamide (DMF), or DCA/CHCl₃ (v/v = 1 : 9) is applied. However, the β-sheet conformation predominates when a pyridine-rich solvent, such as trifluoroacetic acid (TFA) or formic acid, is used.^24,28–32 Besides, the non-peptide oligomer blending was employed to investigate the secondary conformations of PBLG and PMLG. It was found that α-helical conformation is greatly influenced by the side chain rigidity of polypeptides as well as the intermolecular hydrogen bonding induced by the blending oligomer.³³

Although these early studies successfully demonstrated the influence of the solvent on the resulting film conformation, it is difficult to determine whether such treatments are reversible when films of polypeptides are physically adsorbed on the substrate, because vigorous chemical post-treatment on films can easily cause films detachment from the substrate. To counteract the issue of detachment, Wieringa et al.^34–36 utilized a surface-grafting technique in DMF to fabricate PMLG and PBLG-b-PMLG block copolypeptides, and studied the helical orientation of these polypeptides with respect to the substrate. However, a prolonged reaction time was required to achieve an equivalent film thickness (24 h, 40 nm) of PMLG as compared with our case (1 h, 40 nm). In addition, due to the β-sheet aggregates are formed during the ROP in the solution phase a confined space for the control of β-sheet secondary conformation is in demand.³⁷ Recently, PMLG thin film was fabricated as a field effect transistor gate dielectric because of its highly ordered structure and ferroelectric characteristics.³⁸ Additionally, the piezoelectricity can be significantly enhanced by α-helical conformation in PMLG/poly(vinylidene fluoride) composite fibers, suggesting that the materials properties can also be greatly influenced by the secondary structures of polypeptides.³⁹

In this work, our strategy for studying the secondary structure transformation behavior of the polypeptide film was to synthesize polypeptides that were chemically grafted onto a selected substrate. Because of the covalent bond between the polypeptide and the substrate, we could then treat the polypeptide film with a greater variety of chemistries than we could with unbounded films, without running the risk of dissolving the polypeptides from the surface. PMLG was used as a model synthetic polypeptide in this study, and was synthesized with surface-initiated vapor deposition polymerization (SI-VDP).⁴⁰ We analyzed the ability of the surface grafted PMLG brushes to undergo chain conformations based on exposure to helicogenic and sheet-supporting solvents, as has been shown with spin-coated PMLG films.³² In particular, we demonstrate that end-grafted PMLG brushes can undergo reversible transformation between α-helical conformation and β-sheet without severe film depletion, and we analyze the effects of temperature and solvent concentration on these conversions. This work could eventually lead to the development of an in vitro assay to help solve existing biological questions as well as establishing a new class of “smart”, stimuli-responsive thin film materials.

Experimental section

Materials

All the solvents and reagents were purchased from Sigma-Aldrich and Acros and used without further purification. Four kinds of substrates were used in this study, depending on the intended characterization technique. Undoped, double-polished, wedged silicon (100) wafers (Harrick Scientific Corp., with a wedge angle of 0.25°, and a dimension of 10/16′′ × 7/16′′ × 500 microns) were used for transmission Fourier transform infrared spectroscopy (FTIR) and ellipsometric measurements. A 100 nm-thick layer of aluminum, deposited on the silicon wafers by physical vapor deposition, was used in external reflection FTIR (ER-FTIR). Undoped, single-polished silicon (100) wafers (Siltec Silicon, 500 μm) were used for water contact angle measurements. Quartz plates (Starna Cells Inc., with a dimension of 45 × 12.5 × 1 mm) were used for CD measurements.

Sample preparation

To generate the monomers for brush formation from the SI-VDP grafting process, the N-carboxyanhydrides (NCAs) of γ-methyl L-glutamate and γ-benzyl L-glutamate were synthesized following the method of Daly and Poché⁴¹ with a slight modification: in our case, the crude product was purified by recrystallization 3 times in a glove box with tetrahydrofuran (THF) and hexane. All the solid substrates were cleaned with piranha solution (H₂SO₄/H₂O₂ = 7/3 (v/v)) at 120 °C for 30 min (Caution: piranha solution is highly corrosive. Extreme care should be taken when handling it.), followed by rinsing 3 times with deionized water and acetone. The substrates were then dried under a stream of nitrogen and directly used. The NCA amount used in the SI-VDP setup was 8 mg, and the vacuum was 10⁻⁵ to 10⁻⁴ Torr. The SI-VDP reaction conditions for synthesizing PMLG brushes consisted of a substrate temperature setting of 98 °C, an NCA evaporation temperature of 98 °C and a reaction time of 1 h. The reaction conditions for the synthesis of the block copolypeptide PMLG-b-PBLG brushes were similar to previous work.⁴⁰ After the completion of the reaction, the substrates were ultrasonicated in CHCl₃/DCA (v/v = 8 : 2) solution for 5 min, followed by rinsing with fresh CHCl₃, and drying under a stream of nitrogen.

A spin-coated PMLG film was also prepared for comparison. A dilute solution of PMLG (Sigma, molecular weight 10 000–50 000) in CHCl₃ was spin-coated onto the silicon substrate at 3000 rpm for 30 s. The thickness of the film depended on the concentration of the solution. In this study, 48 nm PMLG-coated films were prepared.

To study the effects of formic acid in the vapor phase (denoted by “_υ”) on both spin-coated films and grafted brushes, we placed 48 nm-thick PMLG spin-coated films and 40 nm-thick PMLG grafted brushes in a sealed Petri dish, as shown in previous work.³² The films were taken out of the Petri dish after each reaction condition and then quickly dried overnight in vacuo at room temperature before we examined their conformational changes by ex situ atomic force microscopy (AFM) as well as FTIR and CD spectroscopy.

Characterization methods

The secondary structure of PMLG was studied by FTIR (Nicolet Magna-IR 860 spectrometer in transmission mode with a pure silicon wafer as a reference, 32 scans; resolution, 4 cm⁻¹). The α-helical conformation has an absorption band for amide I at 1653 cm⁻¹ and for amide II at 1550 cm⁻¹. For the β-sheet conformation, the absorption bands for amide I and amide II are at 1632 cm⁻¹ and 1529 cm⁻¹, respectively.^32,42 The helix orientation of end-grafted PMLG brushes on the aluminum substrate was recorded by external reflection FTIR (ER-FTIR, Nicolet Magna-IR 860 spectrometer with a clean sample as the reference, 32 scans; resolution, 4 cm⁻¹). The incident angle was 10°.

CD spectroscopy (JASCO J-60 spectropolarimeter) was used to examine the conformations of surface-grafted polypeptides on quartz substrates. The spectra were recorded at a bandwidth of 1 nm and a scanning rate of 20 nm min⁻¹. A typical right-handed α-helix conformation is indicated by two negative bands at 208 nm and 222 nm. A positive absorption at 198 nm and a negative band at 217 nm represent the β-sheet structure.

The ellipsometric measurements were performed with a Rudolph Research-IV Auto EL ellipsometer at λ = 632.8 Å and an angle of incidence of 70°. Five different spots were measured and averaged for each test. Water contact angles were determined with a OCA-15 PLUS goniometer (Future Digital Scientific Corp.) with a water droplet (3.0 μL). All the measurements were performed five times, and the average values with standard deviations were obtained.

Surface morphologies of PMLG brushes were observed by AFM (Asylum Research, MFP-3D) performed in tapping mode. Cantilevers with a spring constant of 10–130 N m⁻¹ (NanosensorsTM SSS-NCHR-20) were used for measurement at 25 °C in air. Topography, amplitude and phase images were simultaneously obtained.

Results and discussion

The PMLG brush films were synthesized and treated with formic acid _(υ) (95% reagent grade) to convert the α-helical conformation to the β-sheet conformation. Conversely, DCA/CHCl₃ (v/v = 1 : 9), or DCA (99%+) followed by a CHCl₃ rinse were applied to revert the β-sheet structure of PMLG back to the α-helix form.^24,28–32 As shown in Fig. 1, the conversion of PMLG in spin-coated films and as end-tethered brushes shows some of the similarities of PMLG brushes to un-tethered, spin-coated PMLG chains. The 48 nm-thick, spin-coated PMLG films, the as-cast PMLG was in the α-helix conformation, as evidenced by the amide I and II peaks at 1652 cm⁻¹ and 1547 cm⁻¹, respectively, as shown in the FTIR spectra in Fig. 1A. The FTIR spectra of a 40 nm-thick film of PMLG brushes (Fig. 1B) had the same amide I and II peaks. Treating both spin-coated PMLG and the PMLG brushes with formic acid _(υ) at 60 °C for 16 h resulted in the amide I and amide II peaks shifting to 1624 cm⁻¹ and 1520 cm⁻¹, respectively, indicative of the conversion of the PMLG chains to the β-sheet conformation. Along with the similarities in the FTIR spectra of the PMLG brushes and the PMLG film, we also measured water contact angles of the PMLG brushes, which are interchangeable between 60 ± 2° (α-helix) and 72 ± 3° (β-sheet). These contact angle measurements were in good agreement with the spin-coated PMLG films,²⁹ which showed that the spin-coated PMLG and the PMLG brushes have similar surface energies.

Fig. 1 FTIR spectra of (A) a spin-coated PMLG film (B) and an end-grafted PMLG brushes (a) and (b) are the spectra before and after the formic acid _(υ) treatment at 60 °C for 16 h (c) denotes the spectrum of the end-grafted PMLG brushes taken after subsequent DCA/CHCl₃ (v/v = 1 : 9) treatment at room temperature for 16 h.

While there were similarities between the spin-coated PMLG films and the PMLG brushes, there were some important differences. For example, while most of the spin-coated PMLG was transformed from the α-helix conformation to the β-sheet conformations by the formic acid _(υ) treatment, a relatively larger fraction of the PMLG brushes remained in the α-helical conformation, as evidenced by the smaller peaks at 1652 cm⁻¹ and 1547 cm⁻¹ in the FTIR spectrum of the PMLG brushes after treating with formic acid _(υ). A computer-generated peak deconvolution of the FTIR spectra of PMLG brushes, using a mixture of Gaussian and Lorentz fits based on the shifts of amide I and the definition of α-helix content, was obtained by taking the ratio of the peak areas at 1652 cm⁻¹.⁴³ We determined the change of α-helix content of films with the time of formic acid _(υ) treatment as shown in Fig. 2. The α-helix content of the spin-coated PMLG decreased from 90% to less than 20% within 5 h of formic acid _(υ) treatment at 60 °C (Fig. 2A), and the α-helix content of the PMLG brushes slowly decreased over a period of 25 h under formic acid _(υ) treatment and stabilized at ∼40% helical content. The reversible conversion of the β-sheet PMLG brushes back to the α-helix by treatment with DCA/CHCl₃ (v/v = 1 : 9) was also incomplete. As seen in Fig. 2B, treatment of the PMLG brushes with formic acid _(υ) at elevated temp at 70 °C accelerated the conversion of the original α-helical structure from ∼85% initially to 40% in 10 h; immersing the PMLG brush in DCA/CHCl₃ (v/v = 1 : 9) at 20 °C quickly converted the PMLG to an α-helical rich conformation within 2 h, but the α-helical content reached a plateau of ∼70%. The lower conversion efficiency of the PMLG brushes, compared to the spin-coated PMLG, was probably due to the restriction of chain mobility caused by the tethering of one end of the PMLG molecular chain on the solid substrate.

Fig. 2 Incomplete conformation conversion of PMLG brushes. (A) Comparison of the chain conformation conversion of spin-coated PMLG and PMLG brushes caused by the treatment with formic acid _(υ) at 60 °C. (B) α-helix fraction versus time of 40 nm-thick PMLG brushes treated first with formic acid _(υ) at 70 °C, and then with DCA/CHCl₃ (v/v = 1 : 9).

One significant difference between the spin-coated PMLG and the PMLG brushes was that in the case of the PMLG brushes the conversion from α-helix to β-sheet was reversible, whereas for spin-coated PMLG, it was not. The conformational transition of the spin-coated PMLG from β-sheet to α-helix was unsuccessful because the unbound β-sheet PMLG film quickly dissolved in the DCA/CHCl₃ (v/v = 1 : 9) mixture. In contrast, the PMLG brushes could be transformed back to the α-helix conformation by the DCA/CHCl₃ (v/v = 1 : 9) treatment because they were surface grafted. As seen in the FTIR spectra in Fig. 1B(c), after treatment with DCA/CHCl₃ (v/v = 1 : 9), most of the α-helix structure was restored, as evidenced by the amide peaks at 1652 cm⁻¹ and 1547 cm⁻¹. The time dependence of helical restoration was shown in Fig. 2B. Additional CD spectra taken after solvent treatments were also shown in Fig. 3, where the PMLG brushes after formic acid _(υ) treatment had a minimum at around 217 nm, clearly indicative of a β-sheet conformation, and after treatment with DCA/CHCl₃ (v/v = 1 : 9), the spectra had 2 minima, at 208 and 224 nm, indicative of an α-helix conformation. The conformations were cycled from α-helix to β-sheet and back to α-helix a total of 3 times as shown in Fig. 4. Treatment 1 represents the data collected after formic acid _(υ) treatment at 60 °C, where a large portion of α-helix was transformed to β-sheet structure, and treatment 2 was the DCA/CHCl₃ (v/v = 1 : 9) treatment, which converted the β-sheet back to α-helix. The graph in Fig. 4 shows that the cycle can be repeated for at least 3 times, with very little film depletion during this process. It is interesting to point out that the transition was slowest during the first cycle. For the second and third cycles, the conversion was completed within the initial 2 hours of treatment. The faster transition time after the first cycle might be due to the re-organization of entangled PMLG macromolecules in the films during the first cycle. Furthermore, we observed that the α-helical content gradually reduced over the repeating cycles. Though we cannot completely rule out the degradation or depletion of PMLG chains during the process, the interconvertibility of two conformations were maintained in the remaining films. This observation is similar to a previous study on the sequential treatment of poly (β-benzyl-L-aspartate) between L and R conformations where the R conformation is preferred after several cycles.⁴⁴

Fig. 3 CD spectra of conformational changes of surface-tethered PMLG brushes induced from (a) α-helix with DCA/CHCl₃ (v/v = 1 : 9), to (b) β-sheet conformations with formic acid _(υ).

Fig. 4 Kinetic study of conformational transition of air-dried 40 nm-thick, end-grafted PMLG brushes on a silicon wafer. 1 denotes treatment with 60 °C formic acid _(υ) and 2 denotes treatment with 20 °C DCA/CHCl₃ (v/v = 1 : 9). Thickness was measured by ellipsometry.

The detailed mechanism of the helix-to-sheet conversion is still not fully understood. The conversion from α-helix to β-sheet is a process involving the dissociation and then the reformation of hydrogen bonds. A β-sheet-induction reagent should have the ability to dissociate the intramolecular hydrogen bonds, and support the formation of intermolecular hydrogen bonds. It was found that simple organic acids, especially formic acid, have this kind of ability. Nakajima et al.⁴⁵ proposed a model to explain the mechanism of helix-to-sheet transformation induced by formic acid _(υ): the dimeric formic acid forms hydrogen bonds with PMLG chains and, after the evaporation of formic acid molecules, the intermolecular chain hydrogen bonds should be formed, yielding an anti-parallel β-sheet structure. The hydrogen bond reformation required a high degree of intra- and inter-molecular mobility in the molecular chains. The conformation conversion results in Fig. 2A point to a decrease of chain mobility of the end-grafted PMLG brushes, compared to the free chains in a spin-coated film. We hypothesized that if we were to further decrease the mobility of the PMLG brushes, the conversion efficiency would decrease compared to the standard PMLG brushes. We decreased the mobility of the PMLG brushes by capping the end-tethered PMLG chain brushes with another bulky synthetic polypeptide, PBLG. As with the synthesis of the PMLG brushes, SI-VDP was used to synthesize the PMLG-b-PBLG, creating block copolypeptide brushes with the PMLG and PBLG blocks having lengths of 50 and 11 nm, respectively.⁴⁰

We compared the α-helix to β-sheet conversion efficiency of the PMLG-b-PBLG brushes to the PMLG brushes and to two spin-coated films of PMLG, as shown in Fig. 5. For the spin-coated PMLG films, the α-helix fraction dropped to 20% within 2 days with the treatment of 40 °C formic acid _(υ). In contrast, the α-helix fraction of the grafted PMLG and PMLG-b-PBLG decreased only to 50% and 60%, respectively, even after 9 days. The results suggest that while the grafted PMLG brushes could not fully convert to the β-sheet structure, the molecular structure of the PMLG-b-PBLG brushes was even further limited. These results were analogous to the right-to-left conversion of screw sense observed for PBLA-b-PBLG copolypeptide brushes in our previous study.⁴⁴

Fig. 5 Influence of chain mobility on the transition from α-helix to β-sheet with 40 °C formic acid _(υ) treatment. The fitted, solid lines are included solely to guide the eye. Thickness was measured by ellipsometry.

The conversion of the PMLG brushes from α-helix to β-sheet was also strongly dependent upon temperature of formic acid _(υ), both thermodynamically and kinetically, as illustrated in Fig. 6. PMLG brushes were exposed to formic acid _(υ) at temperatures of 20, 40, 60, and 75 °C. At regular intervals the α-helix fraction of each sample was determined by FTIR. At 75 °C, the helical content quickly converted to β-sheet, while at the lower temperatures (20 °C and 40 °C), a comparable content of β-sheet structure could not be achieved.

Fig. 6 Temperature effect of formic acid _(υ) treatment on the conformational transition of 40 nm-thick, end-grafted PMLG brushes. The fitted, solid lines are included solely to guide the eye. Thickness was measured by ellipsometry.

We also examined the effect of DCA concentration on the conversion of the PMLG brushes from having a preponderance of the β-sheet conformation, after the formic acid _(υ) treatment, back to having a majority of the α-helix conformation. The conversion from β-sheet conformation to α-helix was very sensitive to the DCA concentration, as shown in Fig. 7. While 10% of DCA converted the β-sheet to α-helices in 2 h at 20 °C, with a maximum helical content of 70%, the 7.5% DCA can never achieve the same level even after 40 h of treatment.

Fig. 7 DCA concentration effect on the conversion of 40 nm-thick PMLG brushes to an α-helix conformation. The fitted, solid lines are included solely to guide the eye. Thickness was measured by ellipsometry.

We used AFM to study whether the surface morphology could be cycled from rough to smooth by solvent treatments. We chose CHCl₃ followed by ethanol (CHCl₃–ethanol), and ethanol followed by acetone (ethanol–acetone) as the good–bad, and bad–bad solvent pairs, respectively, to treat the PMLG brush films. As observed in AFM images and the corresponding sectional analysis (red lines drawn on the images), the surface morphologies of PMLG brushes were clearly influenced by the solvent treatments. The surface treated with CHCl₃–ethanol pair showed a particulate surface with the peak-to-valley difference of ∼30 nm (the cross-sectional profile along the red line), and the root-mean-square (rms) roughness of 8 nm (the 5 × 5 μm image size, Fig. 8A). Interestingly, after treating with ethanol–acetone solvent pair, the surface became amorphous and featureless with the reduction of peak-to-valley difference to ∼8 nm, and rms to 3 nm (Fig. 8B). The ER-FTIR spectra of the PMLG sample after the treatments of solvents pairs show no clear distinction with respect to the absorbance peak locations and intensity, indicating that these treatments did not alter the α-helical conformation and the average molecular orientation in the millimeter scale, the spot size of the IR beam employed (Fig. 8C). Instead, the solvent pair simply alters the aggregated state of PMLG brushes in the submicron level. In the case of CHCl₃–ethanol treatment, the use of CHCl₃ (a good solvent) first solvates the PMLG brushes, and then the quick rinse by ethanol (a bad solvent) consequently “quenches” the elongated state of PMLG brushes, forming a particulate surface. On the contrary, bad solvent pairs such as the ethanol–acetone pair has very minimal interaction with the PMLG brushes, so the PMLG brushes collapse, forming a flat and smooth surface. The results of the “quench” experiment are similar to what we have studied earlier in the PBLG brush systems.⁴⁶ Due to the nature of covalently-bound of each polypeptide chain, we were able to demonstrate not only the reversible conformational transitions but also the reversible surface morphologies by simply controlling the molecular interaction with the external stimuli.

Fig. 8 The AFM images (5 μm × 5 μm) and the corresponding cross-sectional height profiles (the red line in each image) of a PMLG brush film on a silicon wafer. The image was taken after the sample was treated (A) CHCl₃–ethanol pair, then air-dried, followed by (B) ethanol–acetone pair, then air-dried. (C) ER-FTIR spectra of the air-dried PMLG brushes on the aluminum substrate treated sequentially with (a) CHCl₃–ethanol pair, (b) ethanol–acetone pair. The average film thickness of the sample was 40 nm measured by ellipsometry.

Conclusions

In this study, we have demonstrated the reversible conformation transformation of polypeptide in grafted ultrathin film by solvent treatments. Because traditional thin film fabrication processes are unable to prepare polypeptide film with sufficiently high stability to sustain a wide range of chemical treatments, studies on the conformation transformation behavior are lacking. By chemically grafting one end of the polypeptide molecules onto surfaces, the film stability can be greatly enhanced by the covalent bond. At the same time, the molecular chains still retain a high degree of mobility, compared to multi-grafting point approaches. The results presented here should encourage further exploration on other peptide systems, because the immobilized peptide surface can be a powerful tool to understand the factors influencing peptide conformation. Moreover, such inter-convertible behavior may be utilized as a switch in the applications of membrane or biosensors. For this purpose, more in situ examinations on the instantaneous conformational responses to factors such as temperature, pH, and ionic strength should be emphasized in the future.

Acknowledgements

We thank National Science Council, 98-3114-M-003-001 and Genomics Research Center, Academia Sinica, Taiwan for the financial support.

References

1. S. H. Wibowo, A. Sulistio, E. H. Wong, A. Blencowe and G. G. Qiao, Chem. Commun., 2014, 50, 4971–4988.
2. Y. Shen and H.-A. Klok, Chin. J. Polym. Sci., 2015, 33, 931–946.
3. J. Huang and A. Heise, Chem. Soc. Rev., 2013, 42, 7373–7390.
4. Y. Shen, X. Fu, W. Fu and Z. Li, Chem. Soc. Rev., 2015, 44, 612–622.
5. T. Rezai, B. Yu, G. L. Millhauser, M. P. Jacobson and R. S. Lokey, J. Am. Chem. Soc., 2006, 128, 2510–2511.
6. J. E. Bock, J. Gavenonis and J. A. Kritzer, ACS Chem. Biol., 2012, 8, 488–499.
7. Y.-R. Kim, S. Jung, H. Ryu, Y.-E. Yoo, S. M. Kim and T.-J. Jeon, Sensors, 2012, 12, 9530–9550.
8. A. M. Oelker, S. M. Morey, L. G. Griffith and P. T. Hammond, Soft Matter, 2012, 8, 10887–10895.
9. C. J. Barrow and M. G. Zagorski, Science, 1991, 253, 179–182.
10. T. Dyrks, E. Dyrks, T. Hartmann, C. Masters and K. Beyreuther, J. Biol. Chem., 1992, 267, 18210–18217.
11. I. Hamley, Chem. Rev., 2012, 112, 5147–5192.
12. S. B. Prusiner, Science, 1991, 252, 1515–1522.
13. S. B. Prusiner, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 13363–13383.
14. S. Zhang, M. A. Greenfield, A. Mata, L. C. Palmer, R. Bitton, J. R. Mantei, C. Aparicio, M. O. de La Cruz and S. I. Stupp, Nat. Mater., 2010, 9, 594–601.
15. J. H. Ortony, C. J. Newcomb, J. B. Matson, L. C. Palmer, P. E. Doan, B. M. Hoffman and S. I. Stupp, Nat. Mater., 2014, 13, 812–816.
16. J. C. Wu, Y. L. Wang, C. C. Chen and Y. C. Chang, Chem. Mater., 2008, 20, 6148–6156.
17. D. Y. Wang, T. S. Teng, Y. C. Wu, Y. C. Lee, K. H. Chen, C. H. Chen, Y. C. Chang and C. C. Chen, J. Phys. Chem. C, 2009, 113, 13498–13504.
18. C. T. Yang, Y. L. Wang, S. Yu and Y. C. I. Chang, Biomacromolecules, 2009, 10, 58–65.
19. T. Borase, M. Iacono, S. I. Ali, P. D. Thornton and A. Heise, Polym. Chem., 2012, 3, 1267–1275.
20. J. H. Bradbury and M. D. Fenn, J. Mol. Biol., 1968, 36, 231–246.
21. E. T. Samulski, Liquid Crystalline Order in Polypeptides, Academic, New York, 1978, p. 167.
22. H. Block, Poly (g-benzyll-L-glutamate) and Other Glutamic Acid Containing Polymers, Gordon and Breach Science Publishers, New York, 1st edn, 1983.
23. T. Fukuzawa and I. Uematsu, Polym. J., 1974, 6, 431–437.
24. D. W. Goupil and F. C. Goodrich, J. Colloid Interface Sci., 1977, 62, 142–148.
25. R. Gehani, J. Watanabe, S. Kasuya and I. Uematsu, Polym. J., 1980, 12, 871–882.
26. K. Ito, M. Ogami, T. Kajiyama and M. Takayanagi, J. Macromol. Sci., Part B: Phys., 1982, 21, 1–18.
27. H. Gillgren, A. Stenstam, M. Ardhammar, B. Nordén, E. Sparr and S. Ulvenlund, Langmuir, 2002, 18, 462–469.
28. G. I. Loeb and R. E. Baier, J. Colloid Interface Sci., 1968, 27, 38–45.
29. R. E. Baier and W. A. Zisman, Macromolecules, 1970, 3, 70–79.
30. K. Ito, T. Kajiyama and M. Takayanagi, Polym. J., 1977, 9, 355–369.
31. D. G. Cornell, J. Colloid Interface Sci., 1979, 70, 167–180.
32. K. Kugo, M. Okuno, K. Kitayama, T. Kitaura, J. Nishino, N. Ikuta, E. Nishio and M. Iwatsuki, Biopolymers, 1992, 32, 197–207.
33. S.-W. Kuo and C.-J. Chen, Macromolecules, 2011, 44, 7315–7326.
34. R. H. Wieringa, E. A. Siesling, P. F. M. Geurts, P. J. Werkman, E. J. Vorenkamp, V. Erb, M. Stamm and A. J. Schouten, Langmuir, 2001, 17, 6477–6484.
35. R. H. Wieringa, E. A. Siesling, P. J. Werkman, H. J. Angerman, E. J. Vorenkamp and A. J. Schouten, Langmuir, 2001, 17, 6485–6490.
36. R. H. Wieringa, E. A. Siesling, P. J. Werkman, E. J. Vorenkamp and A. J. Schouten, Langmuir, 2001, 17, 6491–6495.
37. S. H. Wibowo, A. Sulistio, E. H. Wong, A. Blencowe and G. G. Qiao, Adv. Funct. Mater., 2015, 25, 3147–3156.
38. L. Liang, T. Fukushima, K. Nakamura, S. Uemura, T. Kamata and N. Kobayashi, J. Mater. Chem. C, 2014, 2, 879–883.
39. C.-T. Pan, C.-K. Yen, H.-C. Wu, L. Lin, Y.-S. Lu, J. C.-C. Huang and S.-W. Kuo, J. Mater. Chem. A, 2015, 3, 6835–6843.
40. Y. L. Wang and Y. C. Chang, Langmuir, 2002, 18, 9859–9866.
41. W. H. Daly and D. Poché, Tetrahedron Lett., 1988, 29, 5859–5862.
42. T. Miyazawa and E. R. Blout, J. Am. Chem. Soc., 1961, 83, 712–719.
43. N. H. Lee, L. M. Christensen and C. W. Frank, Langmuir, 2003, 19, 3525–3530.
44. C. T. Yang, Y. L. Wang and Y. C. Chang, Biomacromolecules, 2010, 11, 1308–1313.
45. A. Nakajima, T. Fujiwara, T. Hayashi and K. Kaji, Biopolymers, 1973, 12, 2681–2690.
46. Y. L. Wang and Y. C. Chang, J. Am. Chem. Soc., 2003, 125, 6376–6377.

---