Self-bonding and the electrochemical properties of silica-coated nanowires composed of cobalt-coordinated peptide bundles

Kazuki Murai *a, Yusuke Yamamoto b, Takatoshi Kinoshita c, Kenji Nagata b and Masahiro Higuchi *b
aDepartment of Materials Science and Technology, Faculty of Industrial Science and Technology, Tokyo University of Science, 6-3-1 Niijuku, Katsushika-ku, Tokyo 125-8585, Japan. E-mail: murai.kazuki@rs.tus.ac.jp
bDepartment of Life Science and Applied Chemistry, Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya 466-8555, Japan. E-mail: higuchi.masahiro@nitech.ac.jp
cNagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya 466-855, Japan

Received 24th April 2017 , Accepted 20th June 2017

First published on 21st June 2017


We propose a method for self-bonding between electrodes using the self-organizational processes of a silica-coated peptide hybrid nanowire. We designed and synthesized a 23-mer α-helical peptide having functional sites that serve as catalytic sites for silica mineralization (glutamic acid [Glu] and lysine [Lys]) and binding (serine [Ser]), as well as for the formation of a Co(II) ion complex. The peptides formed nanowires composed of α-helical bundles that showed axial connection because of complexation between Co(II) and the histidine (His) residues, and macro-dipole interactions of the α-helical peptides. The nanowires self-bonded between the electrodes, and the surface of the nanowires was coated with silica using mineralization. The silica coating of the surface of the Co(II)-coordinated peptide nanowires induced two kinds of phenomena: (1) structural stabilization of the peptide nanowire component in the composite and (2) an increase in conductivity compared with that of the non-coated peptide nanowires.


Introduction

Fabrication of functional nanomaterials using the “bottom-up” approach has attracted interest as an environmentally-friendly process. Self-organization via the “bottom-up” approach is a very important technique for nanostructure formation. Organic matter-based self-organization affords two- and three-dimensional nanomaterials with precise shapes and sizes by non-covalent interactions such as hydrogen bonding, electrostatic interactions, coordinate bonding, and dipole–dipole interactions. Self-organized nanomaterials play an important role in engineering, medicine, and nanoelectronics.1–5 In particular, there has been increased interest in nanowires that connect arbitrary devices in the field of nanoelectronics. Aono et al. have reported the fabrication of nanowires at arbitrary positions by partial polymerization of a self-assembled diacetylene monolayer on the substrate, which can be observed using a scanning tunneling microscope.6 However, the structural regularity of self-assembled organic matter is limited because the assemblies are formed by simple interactions. In recent years, biomolecules such as peptides and DNA have been investigated for use as building blocks for self-organized matter.7–10 Peptides and DNA form an ordered hierarchical assembly owing to specific interactions such as hydrogen bonding and complementary base-pairing that regulate the distance and direction. Erbe and Liedl et al. reported DNA origami-based Au aggregates composed of one-dimensionally arranged Au nanoparticles on the DNA origami as a template.11,12 We have focused on peptides as the building blocks for self-organized nanomaterials. Note that peptides are designable molecules. The primary and second-order structures of the peptides can be controlled, and these structures and specific interactions determine the high-order structure of the peptide assemblies. Particularly, introduction of various functional groups into specific parts of the peptide assembly is possible. We have previously reported the electron transfer capability of a self-organized metal-coordinated α-helical peptide assembly.13–15 The Co(II)–His complexes in the peptide assembly accelerated electron transfer through the assembly. However, organic nanomaterials fabricated by self-organization have several drawbacks. As opposed to inorganic materials, self-organized organic materials do not have sufficient durability and stability to withstand external stimuli such as heat and light. To solve this problem, we examined biominerals found in living matter (e.g., bones, seashells, and teeth). Biominerals have numerous functions owing to their hierarchically organized, organic–inorganic nanohybrid structures, formed through a unique self-organization process. In previous studies, we have reported that coating with inorganic materials stabilized the self-organized nanostructure of peptides because of mineralization.16,17 Therefore, organic–inorganic nanocomposites, fabricated by mineralization, would have a novel designable nanostructure and functionality provided by the self-assembled organic matter, while the inorganic coating would stabilize the functional structure of the organic matter assembly. In this study, we report the fabrication of stable nanowires having self-bonding ability. We propose a method for self-bonding between electrodes using peptide nanowires composed of Co(II)-coordinated α-helical bundles that connect axially. We used silica mineralization on the surface of the peptide nanowires to stabilize the structure and improve the conductivity of the composite nanowires.

Experimental

Preparation of peptide assembly by self-organization

For complexation between Co(II) and His residues of NH2-LESEHEKLKSKHKSKLKEHESEL-COOH (Pep23), 2 μL of the cobalt acetate ethanol solution (600 mM) was serially added to 8 mL of the Pep23 ethanol solution (1.0 mM). Next, before adding the cobalt acetate ethanol solution, changes in the UV-vis spectra at 530 nm were measured based on the complex formation between Co(II) and the His residues. When UV-vis spectral changes in the mixture could not be observed, we added 2 μL of the cobalt acetate ethanol solution (600 mM). An ethanolic solution of cobalt acetate was added until the molar ratio of Co(II) to His residues, [Co(II)]/[His], reached 3.0. We designated the Co(II)–peptide complex obtained at [Co(II)]/[His] = 0.25 as Co(II)-coordinated Pep23. This value is the stoichiometric ratio for complex formation, which involves the octahedral coordination between the four imidazole groups of the His residues in square-planar geometry and the two apical CH3COO groups capping the square-planar surface.

Immobilization of connecting peptides onto electrodes and self-bonding between connecting peptides by peptide nanowires

To form connecting points on the electrodes for the peptide nanowires, we prepared a mixed self-assembled monolayer (SAM) composed of lipoic acid-LESEHEKLKSKHKSKLKEHESEL-COOH (Lipo-Pep23) and the n-butyl disulfide (C4 spacer) on an interdigitated array (IDA) Au electrode without a passivation membrane (Code 012259, BAS Ltd). The IDA electrode has 65 pairs of gold electrodes, whose interval, width, and length are 2 μm, 2 μm, and 2.5 mm, respectively. The IDA electrodes were immersed in ethanol containing Lipo-Pep23 (4.0 × 10−6 mM) and the C4 spacer (0.1 mM) and incubated for 24 h. The molar ratio of C4 spacer to Lipo-Pep23 was 25[thin space (1/6-em)]000[thin space (1/6-em)]:[thin space (1/6-em)]1. To regulate the lateral density of the connecting points for the self-bonding of the Co(II)-coordinated Pep23 nanowire between the electrodes, we prepared mixed SAMs composed of the C4 spacer and Lipo-Pep23; the molar ratio of the C4 spacer to Lipo-Pep23 was 2500[thin space (1/6-em)]:[thin space (1/6-em)]1 and 250[thin space (1/6-em)]000[thin space (1/6-em)]:[thin space (1/6-em)]1, respectively. The concentration of the C4 spacer in ethanol solution for the preparation of the mixed SAMs was fixed at 0.1 mM. The IDA electrodes, equipped with the Lipo-Pep23-modified IDA electrode connecting points, were washed with ethanol several times and used immediately for the next step. For comparison, we prepared a C4 spacer-modified IDA electrode, which was a SAM composed only of the C4 spacer on the IDA electrode surfaces; this C4 spacer-modified IDA electrode served as a connecting point according to the above procedure.

Self-bonding of the Co(II)-coordinated Pep23 nanowires between the electrodes was carried out as follows. The SAM-immobilized IDA electrodes, the Lipo-Pep23-modified IDA electrode, and the C4 spacer-modified IDA electrode, were immersed in 2 mL of 1.0 mM Pep23 ethanol solution and incubated for eight days. We added 3 μL of 100 mM Co(II)–ethanol solution to the electrodes immersed in Pep 23 ethanol solutions at intervals of eight days; the total amount of Co(II)–ethanol solution was 15 μL. After the Co(II)-coordinated Pep23 nanowires self-bonded between the electrodes, the incubated IDA electrodes were washed with ethanol several times and stored in a desiccator until use for the electrochemical measurements.

Silica mineralization on the surface of peptide nanowires

To fabricate silica-coated Pep23 nanowires, we carried out silica mineralization using the Co(II)-coordinated Pep23 nanowires as the organic template. Silica mineralization was performed as follows. 1.5 mL of tetraethoxysilane (TEOS) was added to 1.5 mL of ethanol solution containing Co(II)-coordinated Pep23 nanowires. The concentration of Co(II)-coordinated Pep23 was 1.0 mM. The mixture was stirred for 15 days at 25 °C. After mineralization, the obtained precipitate was collected by centrifugation, washed with ethanol several times, and then dried in vacuo. For comparison, we also performed silica mineralization in a Co(II)-free Pep23 ethanol system, as well as in peptide-free systems with and without Co(II) ions. The protocol for mineralization was the same as above. The concentration of Co(II)-free Pep23 in ethanol was 1.0 mM. The concentration of Co(II) ions in the peptide-free ethanol solution was 1.3 mM, which was identical to that in the above-mentioned ethanol solution containing Co(II)-coordinated Pep23 nanowires.

To mineralize silica on the surface of Co(II)-coordinated Pep23 nanowires between the connecting points of IDA electrodes, the electrodes were incubated in 50 vol% TEOS ethanol solution (3.0 mL) for 7 days at 25 °C, and then washed with ethanol several times and dried under ambient conditions.

TEM observations

We observed the morphology of peptide assemblies before and after silica mineralization using a transmission electron microscope (TEM; JEM-z2500, JEOL. Ltd) equipped with an Ultra Scan CCD camera (Orius Camera, Gatan Inc.).

The peptides (Pep23 aggregates and Co(II)-coordinated Pep23 nanowires) maintained in an ethanol solution before mineralization (1.0 mM, 2 μL) were adsorbed onto an elastic carbon-coated scanning TEM (STEM) grid, and excess solution was removed by adsorption onto a filter paper. TEM observations were performed under defocused conditions using unstained samples, at an accelerating voltage of 100 kV. Additionally, TEM observations of samples negatively stained with ammonium molybdate were performed. The negatively stained samples were prepared as follows. An aliquot of aqueous solution containing ammonium molybdate (2 wt%) was placed on the STEM grid on which the Co(II)-coordinated Pep23 nanowires were adsorbed. After negative staining for 5 min, the excess solution was removed by adsorption onto a filter paper. The accelerating voltage for the negative staining TEM observations was 200 kV.

The silica-coated Co(II)-coordinated Pep23 nanowires and silica-coated Pep23 aggregates, obtained by mineralization, were re-dispersed in ethanol. The silica-coated Co(II)-coordinated Pep23 nanowires and silica-coated Pep23 aggregates were adsorbed onto STEM grids according to the above procedure. For these observations, the samples were unstained, and TEM observations were conducted at 200 kV.

Electrochemical measurements

Electrochemical measurements of the Co(II)-coordinated Pep23 nanowires, self-bonded between the electrodes of an IDA, were performed at room temperature using a Pach/Whole Cell Clamp Amplifier (CEZ-2400, Nihon Kohden Co., Ltd) connected with a Triangular Wave Generator (SET-2100, Nihon Kohden Co., Ltd) and a digital data recorder (Power Lab 2/25, AD Instruments Co., Ltd). The response currents to the triangular voltage signal (applied voltage; +500 to −500 mV, sweep rate; 1 mV s−1) were recorded on a digital data recorder. The conductivity of the silica-coated Co(II)-coordinated Pep23 nanowires was measured as follows. First, the conductivity of the Co(II)-coordinated Pep23 nanowires connected between the electrodes of the IDA was measured, as described earlier. Then, the surfaces of the Co(II)-coordinated Pep23 nanowires, connected between the electrodes, were coated with silica by immersing the IDA electrode into the TEOS ethanol solution. We measured the conductivity of the silica-coated Co(II) coordinated Pep23 nanowires to assess the effect of inorganic coatings by comparing the conductivities before and after mineralization.

Spectroscopic measurements

The second-order structure of the peptides was determined by circular dichroism (CD) measurements. CD spectra of the peptides in ethanol were recorded using a spectropolarimeter (J-820, JASCO Co., Ltd) at room temperature under a nitrogen atmosphere. The CD experiments were performed in a quartz cell with a 1 mm path length over the range of 190–260 nm. The spectra were averaged from eight consecutive scans and subtracted from the background. The fraction of the second-order structure was calculated using a curve-fitting method.18

To investigate the effect of silica formation on the peptide assembly and second-order structure of the peptide in a hybrid material, we performed transmission Fourier transform infrared (TM-FTIR) spectroscopy (Spectrum 2000, Perkin-Elmer Co., Ltd) using a Hg/Cd/Te detector (wavenumber: 1800–900 cm−1, resolution: 4 cm−1, number of scans: 32). The sample solutions were evaporated and dried in vacuo to obtain samples for measurements. TM-FTIR sample pellets were prepared by mixing the samples with KBr. The fraction of the second-order structure of the peptide was calculated by peak deconvolution of the amide I and amide II bands.19 The 1800–1400 cm−1 regions of the spectra were analyzed as the sum of the individual bands. When the Gaussian/Lorentzian ratio was 9[thin space (1/6-em)]:[thin space (1/6-em)]1, the sum of the calculated individual bands was best fit to the experimental spectra. Moreover, silica formation was estimated from the dominant band, attributable to Si–O–Si bonding at around 1070 cm−1.

To investigate the formation of a Co(II)-coordinated Pep23 peptide, we performed UV-vis measurements on a peptide–ethanol solution containing Co(II) ions. UV-vis spectra were recorded using a UV-3600 (Shimadzu Co., Ltd) system, in a quartz cell with a 1 cm path length over the range 400–700 nm. The obtained UV-vis spectra were analyzed as the sum of the spectra of the absorption and scattering components. The sum of the calculated components was best fit to the experimental spectra.

Results and discussion

Design of the peptide as a building block of the nanowire

We designed the amino acid sequence -LESEHEKLKSKHKSKLKEHESEL- of the peptide as a building block of the nanowire (Scheme 1). When a peptide assumes an α-helical conformation, it forms an amphiphilic structure. The imidazole group of His and the isobutyl group of Leu face the hydrophobic surface of the α-helix. The three imidazole groups of 5His, 12His, and 19His, acting as metal ligands oriented in the direction of the hydrophobic surface of the α-helix (Scheme 1b), can form a metal complex between the Co(II) ion and peptides, affording an α-helical bundle (Scheme 1c).20 Conversely, the hydrophilic side chains of Glu, Lys, and Ser residues are located on the opposite hydrophobic surface of the α-helical peptide. Wallace et al. reported that silica mineralization is promoted at the interface between the carboxyl and amino groups.21 Furthermore, we have reported that silica mineralization is promoted by the charge relay effect between the functional groups of the acidic and basic amino acid side chains on the peptide surface.22,23 Therefore, we chose Glu, having a carboxyl group, and Lys, having an amino group, as the hydrophilic amino acids. We chose Ser since it has a hydroxyl group that can act as the binding site for silica on the hydrophilic surface, which is on the opposite side of the His residue (Scheme 1a). Additionally, lipoic acid was introduced into the N-terminus of the peptide to serve as a binding site for the Au electrode, and the peptide forms a connecting point between the nanowire and the electrode.
image file: c7tb01118k-s1.tif
Scheme 1 Schematic images of amino acid arrangement in the α-helical Pep23 (a) top view and (b) side view. (c) Schematic image of the nanowire formation consisting of Pep23s by Co(II) coordination.

The 23-mer peptides, NH2-LESEHEKLKSKHKSKLKEHESEL-COOH (Pep23) and lipoic acid-LESEHEKLKSKHKSKLKEHESEL-COOH (Lipo-Pep23), were synthesized using solid-phase peptide synthesis on cross-linked ethoxylate acrylate (CLEAR)-acid resin using 9-fluorenylmethoxycarbonyl chemistry.24,25 The peptides were characterized using matrix laser desorption/ionization time-of-flight mass spectroscopy (MALDI-TOF-MS) on JMS-S3000 (JEOL). The Pep23 and Lipo-Pep23 peptides had molecular weights of 2775.1 and 2962.8, respectively, which were in agreement with the calculated values of 2775.1 and 2962.4, respectively (Fig. S1, ESI). Thus, using MALDI-TOF-MS, we confirmed the successful solid-phase peptide synthesis of the designed peptides.

Self-organization of peptide nanowires composed of Co(II)-coordinated α-helical bundles

The second-order structure of the peptides in ethanol was determined using CD spectroscopy. Fig. S2 (ESI) shows the CD spectra of Pep23 and Lipo-Pep23 in ethanol. The concentration of each peptide was fixed at 1.0 mM. The CD spectrum exhibited two negative bands at 208 and 222 nm, which showed the typical CD patterns for the α-helical conformation.18 A fraction of each peptide’s second-order structure was calculated by the curve-fitting method using a linear combination of the typical CD spectra of the dispersed α-helix, β-sheet, and random coil conformations. Pep23 and Lipo-Pep23 assumed an α-helical conformation with a considerable degree of random coil and β-sheet structures, in the ethanol solution, respectively. The helicity of Lipo-Pep23 at 47% was approximately the same as that of the Pep23 peptide at 50%. This result indicates that introducing lipoic acid to the N-terminus of the peptide main-chain did not influence the second-order structure in ethanol solution.

We examined the formation of an α-helical bundle based on the complexation between the imidazole groups of the His residues and Co(II). Formation of the α-helical bundle, induced by the addition of Co(II) ions, was investigated based on the UV-vis absorption spectral changes attributable to complex formation, and the CD and TM-FTIR spectral changes based on second-order structural changes. In the UV-vis spectrum of a mixture containing Pep23 and Co(II) ions, an absorption band was observed at 530 nm, which corresponded to the octahedral coordination geometry around the Co(II) atom (Fig. S3, ESI). These results suggest that Co(II) formed a complex with four imidazole groups of the His residues in square-planar geometry, and that two apical acetate groups capped the square-planar face in the vertically oriented Pep23 assembly (Fig. S3 (ESI) inset). In the above measurements, we added the Co(II) ions all at once; the molar ratio of the Co(II) ion to the His residue of Pep23 was 0.25. We then investigated whether the complex formation was dependent on the concentration of Co(II). Fig. S4 (ESI) shows the UV-vis spectral changes in the mixture at 530 nm, when Co(II) ions were serially added to Pep23 ethanol solution. The changes in the UV-vis spectra indicated that the incubation time required to reach equilibrium was approximately 8 days at a low molar ratio of [Co(II)]/[His] = 0.05. The incubation time was decreased with an increase in the molar ratio of [Co(II)]/[His]. These results indicate that complex formation was dependent on the concentration of His residues and Co(II). This phenomenon can be explained as follows. In the initial state, which involves Co(II)-free Pep23 in solution, individual Pep23 units were dispersed in ethanol. Therefore, complexation between Co(II) and the four imidazole groups of the four peptide molecules takes a long time to reach equilibrium. The incompletely Co(II)-coordinated Pep23 assembly, formed at a lower [Co(II)]/[His] molar ratio, had an empty binding site for Co(II) coordination in the assembly. The imidazole groups at the empty binding site quickly formed a complex with the newly added Co(II). Therefore, complex formation was promoted by increasing the [Co(II)]/[His] molar ratio.

Fig. 1a shows the changes in the UV-vis spectra of Pep23 after equilibrium was reached by the stepwise addition of Co(II) ions. The spectra revealed absorption (Fig. 1b) and scattering components (Fig. 1c) based on the formation of the Co(II)-coordinated assembly of Pep23. The spectra were separated into absorption and scattering components by peak deconvolution. The UV-vis spectral measurements indicated an increased absorbance with the addition of Co(II) and a redshift of the absorption band due to complexation between the imidazole groups of the His residues and Co(II). Additionally, the scattering intensity increased with the addition of Co(II). The increased absorbance (Fig. 2a) and shift of the absorption band (Fig. 2b) indicated that equilibrium was achieved when [Co(II)]/[His] was approximately 0.25. The [Co(II)]/[His] molar ratio of 0.25 agreed with the stoichiometric ratio for the formation of the square-planar complex between one Co(II) and four His; the value indicated the formation of a four α-helix bundle composed of Co(II)-coordinated Pep23s. The redshift of the absorption band, caused by the formation of the Co(II) complex, implied that the complexes conjugate in a bundle, and that the conjugate length is increased by the addition of Co(II). The increase in the conjugate length suggested that the α-helical bundles were connected axially because of the macro-dipole interactions between the α-helical peptides. This result was supported by the increase in the scattering intensity observed with the addition of Co(II). Fig. 2c shows the changes in the scattering intensity at 400 nm, which implied that the Co(II)-coordinated Pep23 assembly grew larger with complexation. The sudden increase in the scattering intensity above [Co]/[His] = 0.25 indicated that the formation of large aggregates was induced by the interaction between excess Co(II) ions and the carboxyl groups of Glu on the peptide assemblies.


image file: c7tb01118k-f1.tif
Fig. 1 (a) UV-vis spectra of the Co(II)-coordinated Pep23 bundle in ethanol by sequential addition of Co(II), after reaching equilibrium. The (b) absorption and (c) scattering components based on the Co(II) complex obtained by peak deconvolution of (a).

image file: c7tb01118k-f2.tif
Fig. 2 [Co(II)]/[His] dependence of (a) absorbance, (b) absorption band based on the Co(II) complex, and (c) scattering intensity at 400 nm.

Next, we investigated the changes in the CD and TM-FTIR spectra based on structural changes in the peptide caused by the addition of Co(II) ions. The CD spectra of Pep23 showed that continuous addition of Co(II) ions caused a decrease in the molar ellipticity (Fig. 3a). The CD spectra of Pep23 after the addition of Co(II) did not fit the linear combination of typical CD spectra of the dispersed α-helical, β-sheet, and random coil conformations. The CD spectra of the aggregated α-helical peptides were flattened and decreased at a band of 208 nm, and redshifted at a band of 222 nm band,26–28 compared with those of the typically dispersed α-helical peptides. We used the molar ellipticity ratio of 222 nm, [θ]222, to that at 208 nm, [θ]208, ([θ]222/[θ]208) as an indicator of peptide bundle formation by complexation between Co(II) and the imidazole groups of the His residues. Fig. 3b shows the [Co(II)]/[His] dependence of [θ]222/[θ]208. The [θ]222/[θ]208 value increased by the addition of Co(II) ions and reached equilibrium at a stoichiometric ratio of [Co(II)]/[His] = 0.25, where the four-α-helix bundle was formed by complexation with Co(II). We could not obtain an accurate ratio for the second-order structure of Co(II)-coordinated Pep23 from the CD measurements; therefore, we performed TM-FTIR measurements to obtain the helicity of the four α-helical bundle induced by Co(II) complexation. Fig. 4 shows the TM-FTIR spectra of Co(II)-free Pep23 and Co(II)-coordinated Pep23, respectively. The molar ratio of Co(II) to His, [Co(II)]/[His], was fixed at 0.25. The helicity of Pep23 in the dry state was 53%, which was in agreement with that of Pep23 at 50% (Fig. S2, ESI), in ethanol solution (Fig. 4a). The helicity of Co(II)-coordinated Pep23 increased to 60% (Fig. 4b). Further, the absorbance at 1590 cm−1, assigned to the carbonyl stretching band of the CH3COO group, which was the apical capping group of the Co(II) complex, was observed in the spectrum of Co(II)-coordinated Pep23 (Fig. 4b). Hence, the formation of the Pep23 bundle by Co(II) complexation stabilized the α-helical conformation.


image file: c7tb01118k-f3.tif
Fig. 3 (a) CD spectral changes of the Lipo-Pep23 in ethanol by sequential addition of Co(II), after reaching equilibrium. (b) [Co(II)]/[His] dependence of the ratio of molar ellipticity at 222 nm to that at 208 nm, [θ]222/[θ]208.

image file: c7tb01118k-f4.tif
Fig. 4 TM-FTIR spectra of the (a) Pep23 and (b) Co(II)-coordinated Pep23 nanowires. Dotted lines show the peak deconvolution of the amide I and amide II bands from random coil, α-helical, and β-sheet conformations.

Based on the results of UV-vis, CD, and TM-FTIR spectral analysis, we proposed a mechanism for the self-organization of Pep23 induced by Co(II) complexation. Pep23s form a four α-helical bundle via complexation between the Co(II) and four imidazole groups of the four Pep23 molecules. The bundle formation stabilizes the α-helical structure of Pep23. Co(II) complexation and macro-dipole interactions between the α-helical Pep23s induce axial connection of the peptide bundles; as a result, peptide nanowires, having conjugated Co(II) complexes in the interior, are formed (Scheme 1c).

We directly observed the morphology of the Co(II)-coordinated Pep23 assembly using TEM (accelerating voltage: 100 kV). The molar ratio of Co(II) to His was fixed at 0.25. The observations were performed under unstained and defocused conditions. For comparison, we used the morphology of Co(II)-free Pep23. The TEM observations indicated that Co(II)-free Pep23 peptides formed globular aggregates that were ca. 30 nm in diameter (Fig. 5a). Conversely, the morphology of Co(II)-coordinated Pep23 bundles was clearly observed in the nanowires (Fig. 5b). However, TEM observations of the unstained organic matter under defocused conditions did not clarify characteristics such as fiber width. Using TEM (accelerating voltage: 200 kV) and negative staining with ammonium molybdate, we found that the fiber width of the Co(II)-coordinated Pep23 nanowire was 3.7 nm (Fig. S5a, ESI). Fig. S5b (ESI) shows a schematic image (side view) of the α-helical Pep23 rod. Fig. S5c and d (ESI) show schematic images of the Co(II)-coordinated bundle composed of four Pep23 peptides, as well as the anti-parallel and twisted arrangement of the α-helical rods, respectively. The diameter of the α-helical peptide and the four α-helical Pep23 bundle was 1.38 nm29 and 3.3 nm, respectively, when the four helical Pep23 rods in anti-parallel arrangement formed a tetragonal pack. We used the diameter of α-helical poly(leucine [Leu]) for that of the α-helical Pep23, because the Leu residues were oriented toward the hydrophobic surface between the Pep23 rods in the bundle. The distance between two α-helical rods facing each other was 0.57 nm. This is slightly greater than the length of the coordination bond of Co(II) between two imidazole groups (0.4 nm); the typical length of a Co(II)–N coordination bond in the equatorial plane of the octahedral coordination geometry of a Co(II) complex is 0.2 nm. Importantly, the three imidazole groups of Pep23 did not undergo complexation in the anti-parallel arrangement of the Pep23 bundle (Fig. S5c, ESI). However, the imidazole groups of the His side chains in the α-helical Pep23 were slanted by 20° with respect to the helical axis of the peptide (Scheme 1b). For the oblique arrangement of the imidazole groups, the four twisted helical rods enabled the stoichiometric formation of the complex between Co(II) and the four imidazole groups (Scheme S5b, ESI). The Co(II)-coordinated four α-helical bundle had a diameter of 3.2 nm. The diameter of the Pep23 bundle was larger than the calculated value because Pep23 has Lys residues, the side chains of which are longer than those of Leu, on the opposing hydrophilic surface. The nanowire width, 3.7 nm, obtained from negative staining TEM observations agreed with the calculated value.


image file: c7tb01118k-f5.tif
Fig. 5 TEM images of the (a) Pep23 aggregates and (b) Co(II)-coordinated Pep23 nanowires. These samples were prepared using the adsorption method. TEM observation was carried out at 100 kV under the unstained condition.

Silica coating of Pep23 nanowires by mineralization

Co(II)-Coordinated Pep23 nanowires may have potential problems in structural stability with respect to the conformational transition of the peptide component and complex dissociation induced by external stimuli. To address this issue, we used silica mineralization to apply an inorganic coating to the surface of the Pep23 nanowire. The designed Pep23 has many catalytic sites composed of functional group pairs (Lys as a basic amino acid and Glu as an acidic amino acid) for silica mineralization. We performed silica mineralization under two conditions: (1) in ethanol containing Co(II)-free Pep23 and Co(II)-coordinated Pep23 nanowires; (2) in peptide-free ethanol solution with and without Co(II) ions. After silica mineralization for 15 days at 25 °C, precipitates were observed only in systems containing Co(II)-free Pep23 and Co(II)-coordinated Pep23 nanowires. The amounts of mineralized silica on the Co(II)-free Pep23 and Co(II)-coordinated Pep23 nanowire systems were 1.7 and 2.1 mg silica per mg peptide, respectively. These results imply that silica mineralization is catalyzed by the peptide. Silica mineralization is induced by the peptide in a two-step reaction mechanism. The first step is hydrolysis of the ethoxy groups of TEOS used as the silica precursor. TEOS is hydrolyzed to orthosilicic acid, having four silanol groups in a solvent containing H2O. However, we used ethanol as the solvent. We believe that hydrolysis of the ethoxy groups of TEOS in ethanol is triggered by the traces of water present in ethanol. The second step is a dehydration condensation reaction between the silanol groups catalyzed by the basic and acidic side chains pairs of the peptide.

Silica is formed in the second step of the reaction between the silanol groups of the Si(OH)4 molecules. We focused on the Ser residue having a hydroxyl group in the side-chain as a binding site for silica. We investigated the possibility of a dehydration condensation reaction between the hydroxyl group of the Ser side-chain and the silanol group. We selected trimethylethoxysilane (TMEOS), which has only one ethoxy group that is available for the dehydration condensation reaction. The Pep23 ethanol solution (1.0 mM, 1.5 mL) was added to 1.5 mL of TMEOS, and the mixture was stirred for 15 days at 25 °C. After the reaction, the molecular weights of the products in the reaction solution were analyzed using MALDI-TOF-MS. The MALDI-TOF-MS spectrum showed a peak at 2774.8 from [Pep23 + H]+ and a new peak at 2847.5 (Fig. S6, ESI). The new peak was in fair agreement with the molecular weight (2846.2) of Pep23, with one TMEOS resulting from a dehydration condensation reaction between the hydroxyl groups of the Ser side-chain and the silanol group of the hydrolyzed TMEOS. We revealed that the hydroxyl group of the Ser side-chain acted as a binding site for the silica formed by mineralization.

We investigated the formation of composites consisting of the peptide and silica using TM-FTIR measurements. Fig. 6 shows the TM-FTIR spectra of silica-coated (a) Pep23 and (b) Co(II)-coordinated Pep23 nanowires obtained using silica mineralization, respectively. For comparison, the spectra of (c) pristine Pep23 and (d) Co(II)-coordinated Pep23 nanowires before silica mineralization are shown. The precipitates obtained by mineralization gave a new absorption peak at 1070 cm−1 as compared with the TM-FTIR spectra of the pristine peptides. This peak was assigned to the Si–O–Si bond generated by silica mineralization.30–32 We used TM-FTIR analysis to examine the second-order structure of the peptides (Pep23 and Co(II)-coordinated Pep23) in the precipitates obtained after silica mineralization (Fig. 7). After silica mineralization, the helicities of Pep23 and Co(II)-coordinated Pep23 nanowires increased, respectively. In particular, the helicity of Pep23 in the Co(II)-coordinated Pep23 nanowires became 79%, implying stabilization of the helical structure by the silica coating. We have reported that mineralization of the peptide template induced the stabilization of the second-order structure because the peptide template surface was coated by the inorganic material.16,17 The inorganic shell, formed by silica mineralization, suppressed the molecular motion of Pep23. This effect led to stabilization of the α-helical structure of Pep23 in the silica-coated Co(II)-coordinated Pep23 nanowires.


image file: c7tb01118k-f6.tif
Fig. 6 TM-FTIR spectra of the peptide–silica nanocomposites obtained by mineralization using (a) Pep23 and (b) Co(II)-coordinated Pep23 nanowires and spectra of non-silica-coated (c) Pep23 aggregates and (d) Co(II)-coordinated Pep23 nanowires, respectively.

image file: c7tb01118k-f7.tif
Fig. 7 TM-FTIR spectra of the (a) silica-coated Pep23 and (b) silica-coated Co(II)-coordinated Pep23 nanowires. Dotted lines show the peak deconvolution of the amide I and amide II bands from random coil, α-helical, and β-sheet conformations.

We observed the morphology of the silica-coated Pep23 and silica-coated Co(II)-coordinated Pep23 nanowires using TEM (accelerating voltage was 200 kV in the unstained samples). The silica-coated Pep23 (Fig. 8a) had the morphology of a globular particle, while the silica-coated Co(II)-coordinated Pep23 (Fig. 8b) showed a nanowire structure. Although the sizes of the assemblies had increased, the morphology of each peptide assembly was the same as that before silica mineralization (Fig. 5). This result indicates that mineralized silica was fixed at the Ser residues, on the surface of Pep23 and Co(II)-coordinated Pep23 nanowires, by the template effect. The silica-coated Pep23 nanowires are expected to be promising candidates for self-bondable conductive nanowires because of the conjugated cobalt complexes aligned in the direction of the wire axis in the interior of the complex.


image file: c7tb01118k-f8.tif
Fig. 8 TEM images of the (a) silica-coated Pep23 aggregates and (b) silica-coated Co(II)-coordinated Pep23 nanowires after mineralization. The composites were re-dispersed in ethanol and prepared using the adsorption method. These observations were carried out without staining at 200 kV.

Self-bonding and the electrochemical properties of silica-coated peptide nanowires

We investigated the self-bonding ability between electrodes using the self-organization of Pep23 by Co(II) complexation. Pep23 does not have any binding site for the gold electrodes of the IDA. To introduce binding sites on the gold electrodes, we designed a new Lipo-Pep23 having lipoic acid at the N-terminus of the Pep23 main-chain. We constructed a mixed SAM containing Lipo-Pep23 and the C4 spacer, n-butyl disulfide, on the gold electrodes to separate the connecting points for the nanowires. We prepared the mixed SAM-modified IDA electrode, the Lipo-Pep23 modified IDA, by a 24 hour incubation in a solution of ethanol containing Lipo-Pep23 and the C4 spacer. The molar ratio of the C4 spacer to Lipo-Pep23 was fixed at 25[thin space (1/6-em)]000[thin space (1/6-em)]:[thin space (1/6-em)]1. For comparison, we prepared an IDA electrode modified only with the C4 spacer (C4 spacer-modified IDA). Fig. 9a[i] shows the current–voltage (IV) curve of the Co(II)-coordinated Pep23 nanowires self-connected between Lipo-Pep23-modified IDA electrodes.
image file: c7tb01118k-f9.tif
Fig. 9 IV curves of the (a) Co(II)-coordinated Pep23 nanowires bonded between the SAM-modified IDA electrodes and (b) silica-coated Co(II)-coordinated Pep23 nanowires bonded between Lipo-Pep23-modified IDA electrodes. Curves (i) and (ii) in (a) denote the response current of the Lipo-Pep23 nanowires bonded between the Lipo-Pep23-modified and C4 spacer-modified IDA electrodes.

We observed relatively high conductivity in the Co(II)-coordinated Pep23 nanowires connected between the Lipo-Pep23-modified IDA electrodes. Conversely, we could not observe the responsive current when the C4 spacer-modified IDA electrode was used as the self-bonding electrode for Co(II)-coordinated Pep23 nanowires (Fig. 9a[ii]). This result indicated that Pep23s formed nanowires composed of Co(II)-coordinated Pep23 bundles by Co(II) complexation between the connecting points; furthermore, it was demonstrated that the nanowires physically and electrochemically bonded between Lipo-Pep23s on the electrodes. The molar ratio of the C4 spacer to Lipo-Pep23 was 25[thin space (1/6-em)]000[thin space (1/6-em)]:[thin space (1/6-em)]1. The lateral density of the perpendicularly oriented Lipo-Pep23 in the mixed SAM was ca. 1.0 × 104 nm2 per molecule, assuming that Lipo-Pep23 and the C4 spacer form an ideal mixture, and that the occupied area of the C4 spacer and perpendicularly oriented Lipo-Pep23 is 0.4 nm2[thin space (1/6-em)]33 and 1.5 nm2,29 respectively. This value of 1.0 × 104 nm2 per molecule, which indicated the density of Lipo-Pep23 on the electrode, implies that the averaged distance between the connecting points is 0.1 μm, which is shorter than the interval (2 μm) between the IDA electrodes. We prepared mixed SAMs in which the molar ratio of the C4 spacer and Lipo-Pep23 was 250[thin space (1/6-em)]000[thin space (1/6-em)]:[thin space (1/6-em)]1 and 2500[thin space (1/6-em)]:[thin space (1/6-em)]1, on the IDA electrodes, respectively. However, on the mixed SAM-modified IDA electrodes, self-bonding of the Co(II)-coordinated Pep23 nanowires did not occur. This result suggested that on the electrode with a lower density of Lipo-Pep23 (1.0 × 105 nm2 per molecule), the probability of bonding by the nanowires between the connecting points decreases. On the electrode with a higher density of Lipo-Pep23 (1.0 × 103 nm2 per molecule), nanowires bonding on one electrode are predominant over those bonding between the electrodes.

Next, we investigated the conductivity of self-bonded silica-coated Co(II)-coordinated Pep23 nanowires. The number of peptide nanowires that self-connected between the gold electrodes differed in each experiment, because self-bonding was stochastically formed. The conductivities of the nanowires fabricated under the same conditions were different. Fig. S7 (ESI) shows the IV curves of the peptide nanowires that were prepared under the same conditions. We investigated the IV properties of the same peptide nanowires before and after mineralization to evaluate the effects of mineralization. We used the same Lipo-Pep23-modified IDA electrode, connected by the Co(II)-coordinated Pep23 nanowires, to investigate the effect of silica coatings on the nanowires. After IV measurements, the Lipo-Pep23-modified IDA electrode, connected by the nanowires, was immersed in TEOS ethanol solution to coat the nanowire surfaces with silica via mineralization. The conductivity of the silica-coated Co(II)-coordinated Pep23 nanowires was 10 times higher than that before silica coating (Fig. 9b). Furthermore, relatively high hysteresis was observed in the IV curve of the Co(II)-coordinated Pep23 nanowires (Fig. 9a[i]); conversely, the hysteresis was considerably reduced after silica coating (Fig. 9b). The existence of hysteresis in the IV curve implies that before silica coating, the nanowires had dielectric-like properties. After silica coating, the conductor-like properties of the nanowires changed based on Ohm's law. An increase in the persistence length of the conjugated Co(II) complexes in the nanowires might be induced by the stabilization of the α-helical bundle structure induced by the silica coating. Such an increase in the persistence length is suggested to improve the nanowire conductivity. We conclude that a significant improvement in the conductivity of the silica-coated peptide nanowires was induced by structural stabilization of the inner Co(II)-coordinated α-helical bundles owing to composite formation by silica mineralization.

Conclusion

Here, we propose a novel method for self-bonding between electrodes by silica-coated Co(II)-coordinated peptide nanowires using self-organizational processes (Co(II) complexation, macro-dipole interactions, and silica mineralization) of an α-helical Pep23 peptide. The four Pep23s formed a four α-helical bundle having Co(II) complexes in the interior; the bundles grew axially to form nanowires by macro-dipole interactions of the α-helical peptides. This self-organized nanowire formation occurred between the connecting points that consisted of the sequence peptide, Lipo-Pep23, fixed on the electrodes. Silica coating on the self-bonded Co(II)-coordinated peptide nanowires induced the conductor-like properties, leading to high conductivity. The formation of highly conductive nanowires is explained as follows. Co(II) complexes, arranged along the wire axis within the nanowires, are conjugated, and the persistence length of the conjugated Co(II) complexes is increased by silica coatings. Based on the results of this study, we propose a novel approach for the fabrication of functional assemblies using peptide sequence design. This knowledge is expected to trigger innovations in the field of nanoelectronics, including the development of functional nanodevices by environmentally-friendly processes.

Notes and references

  1. C. Farcau, H. Moreira, B. Viallet, J. Grisolia and L. Ressier, ACS Nano, 2010, 4, 7275–7282 CrossRef CAS PubMed.
  2. M. Shimomura, A. Tsuruma, M. Tanaka, H. Sunami and S. Yamamoto, J. Surf. Sci. Soc. Jpn., 2006, 27, 170–175 CrossRef CAS.
  3. T. Nonoyama, M. Tanaka, Y. Inai, M. Higuchi and T. Kinoshita, ACS Nano, 2011, 5, 6174–6183 CrossRef CAS PubMed.
  4. O. Carny, D. E. Shalev and E. Gazit, Nano Lett., 2006, 6, 1594–1597 CrossRef CAS PubMed.
  5. R. J. Mart, R. D. Osborne, M. M. Stevens and R. V. Ulijin, Soft Matter, 2006, 2, 822–835 RSC.
  6. Y. Okawa and M. Aono, Nature, 2001, 409, 683–684 CrossRef CAS PubMed.
  7. C. Zhan, X. Xue, Q. Luo, Y. Li, K. Yang, X. Zhuang, Y. Jiang, J. Zhang, J. Liu, G. Zou and X.-J. Liang, ACS Nano, 2014, 8, 11715–11723 CrossRef PubMed.
  8. S. Sathaye, H. Zhang, C. Sonmez, J. P. Schneider, C. M. MacDermaid, C. D. V. Bargen, J. G. Saven and D. J. Pochan, Biomacromolecules, 2014, 15, 3891–3900 CrossRef CAS PubMed.
  9. A. Gopinath, E. Miyazono, A. Faraon and P. W. K. Rothemund, Nature, 2016, 535, 401–405 CrossRef CAS PubMed.
  10. A. A. Rafat, T. Pirzer, M. B. Scheible, A. Kostina and F. C. Simmel, Angew. Chem., Int. Ed., 2014, 53, 7665–7668 CrossRef PubMed.
  11. B. Teschome, S. Facsko, T. Schönherr, J. Kerbusch, A. Keller and A. Erbe, Langmuir, 2016, 32, 10159–10165 CrossRef CAS PubMed.
  12. R. Schreiber, J. Do, E.-M. Roller, T. Zhang, V. J. Schüller, P. C. Nickels, J. Feldmann and T. Liedl, Nat. Nanotechnol., 2014, 9, 74–78 CrossRef CAS PubMed.
  13. X. Wang, S. Fukuoka, R. Tsukigawara, K. Nagata and M. Higuchi, J. Colloid Interface Sci., 2013, 390, 54–61 CrossRef CAS PubMed.
  14. X. Wang, K. Nagata and M. Higuchi, Thin Solid Films, 2012, 520, 2884–2891 CrossRef CAS.
  15. X. Wang, K. Nagata and M. Higuchi, Langmuir, 2011, 27, 12569–12574 CrossRef CAS PubMed.
  16. K. Murai, M. Higuchi, T. Kinoshita, K. Nagata and K. Kato, Chem. Commun., 2013, 49, 9947–9949 RSC.
  17. K. Murai, T. Kinoshita, K. Nagata and M. Higuchi, Langmuir, 2016, 32, 9351–9359 CrossRef CAS PubMed.
  18. N. Greenfield and G. D. Fasman, Biochemistry, 1969, 8, 4108–4116 CrossRef CAS PubMed.
  19. T. Miyazawa and E. R. Blout, J. Am. Chem. Soc., 1961, 83, 712–719 CrossRef CAS.
  20. M. Higuchi, Thin Solid Films, 2008, 516, 4312–4318 CrossRef CAS.
  21. A. F. Wallace, J. J. DeYoreo and P. M. Dove, J. Am. Chem. Soc., 2009, 131, 5244–5250 CrossRef CAS PubMed.
  22. T. Kuno, T. Nonoyama, K. Hirao and K. Kato, Langmuir, 2011, 27, 13154–13158 CrossRef CAS PubMed.
  23. K. Murai, M. Higuchi, T. Kuno and K. Kato, ChemPlusChem, 2014, 79, 531–535 CrossRef CAS.
  24. C. D. Chang and J. Meienhofer, Int. J. Pept. Protein Res., 1978, 11, 246–249 CrossRef CAS PubMed.
  25. R. B. Merrifeld, J. Am. Chem. Soc., 1963, 85, 2149–2154 CrossRef.
  26. K. Rosenheck and A. S. Schneider, Proc. Natl. Acad. Sci. U. S. A., 1973, 70, 3458–3462 CrossRef CAS.
  27. O. Pieroni, A. Fissi, J. L. Houben and F. Ciardelli, J. Am. Chem. Soc., 1985, 107, 2990–2991 CrossRef CAS.
  28. H. Maeda, H. Kato and S. Ikeda, Biopolymers, 1984, 23, 1333–1346 CrossRef CAS.
  29. T. Doi, T. Kinoshita, Y. Tsujita and H. Yoshimizu, Bull. Chem. Soc. Jpn., 2001, 74, 421–425 CrossRef CAS.
  30. Y. F. Lu, H. Y. Fan, S. Stump, T. L. Ward, T. Rieker and C. J. Brinker, Nature, 1999, 398, 223–226 CrossRef CAS.
  31. V. Vamvakaki, M. Hatzimarinaki and N. Chaniotakis, Anal. Chem., 2008, 80, 5970–5975 CrossRef CAS PubMed.
  32. A. Altunbas, N. Sharma, M. S. Lamm, C. Yan, R. P. Nagarkar, J. P. Schneider and D. J. Pochan, ACS Nano, 2011, 4, 181–188 CrossRef PubMed.
  33. H. E. Reis, Sci. Am., 1961, 204, 152–165 CrossRef.

Footnote

Electronic supplementary information (ESI) available: MALDI-TOF-MS spectra, CD and UV-vis spectra, TEM images, and other supplementary data. See DOI: 10.1039/c7tb01118k

This journal is © The Royal Society of Chemistry 2017