Direct and Selective Immobilization of Proteins by Means of an

Dec 10, 2009 - Binding Analysis of ZnOBP-Fused GFP onto Inorganic Material ..... (33) In the present study, N1−7 ZnOBP, which had the highest affini...
0 downloads 0 Views 297KB Size
Download PDF
480

J. Phys. Chem. B 2010, 114, 480–486

Direct and Selective Immobilization of Proteins by Means of an Inorganic Material-Binding Peptide: Discussion on Functionalization in the Elongation to Material-Binding Peptide Nozomi Yokoo,† Takanari Togashi,† Mitsuo Umetsu,*,‡,§,| Kouhei Tsumoto,⊥ Takamitsu Hattori,‡ Takeshi Nakanishi,‡ Satoshi Ohara,[ Seiichi Takami,† Takashi Naka,† Hiroya Abe,[ Izumi Kumagai,‡ and Tadafumi Adschiri† Institute of Multidisciplinary Research for AdVanced Materials, Tohoku UniVersity, Katahira 2-1-1, Aoba-ku, Sendai 980-8577, Japan, Department of Biomolecular Engineering, Graduate School of Engineering, Tohoku UniVersity, Aoba 6-6-11, Aramaki, Aoba-ku, Sendai 980-8579, Japan, Center for Interdisciplinary Research, Tohoku UniVersity, Aoba 6-3, Aramaki, Aoba-ku, Sendai 980-8578, Japan, Bio- and Electromechanical Autonomous Nano Systems (BEANS) Laboratories, New Energy and Industrial Technology DeVelopment Organization (NEDO), Japan, Department of Medical Genome Sciences, Graduate School of Frontier Sciences, The UniVersity of Tokyo, Kashiwanoha 5-1-5, Kashiwa 277-8562, Japan, Joining and Welding Research Institute, Osaka UniVersity, Mihogaoka 11-1, Ibaraki 567-0047, Japan ReceiVed: August 11, 2009; ReVised Manuscript ReceiVed: NoVember 15, 2009

Using an artificial peptide library, we have identified a peptide with affinity for ZnO materials that could be used to selectively accumulate ZnO particles on polypropylene-gold plates. In this study, we fused recombinant green fluorescent protein (GFP) with this ZnO-binding peptide (ZnOBP) and then selectively immobilized the fused protein on ZnO particles. We determined an appropriate condition for selective immobilization of recombinant GFP, and the ZnO-binding function of ZnOBP-fused GFP was examined by elongating the ZnOBP tag from a single amino acid to the intact sequence. The fusion of ZnOBP with GFP enabled specific adsorption of GFP on ZnO substrates in an appropriate solution, and thermodynamic studies showed a predominantly enthalpy-dependent electrostatic interaction between ZnOBP and the ZnO surface. The ZnOBP’s binding affinity for the ZnO surface increased first in terms of material selectivity and then in terms of high affinity as the GFP-fused peptide was elongated from a single amino acid to intact ZnOBP. We concluded that the enthalpy-dependent interaction between ZnOBP and ZnO was influenced by the presence of not only charged amino acids but also their surrounding residues in the ZnOBP sequence. Introduction The immobilization of biomolecules on solid substrates is an essential technique used for biosensors, lab-on-a-chip systems, and nanoimaging. DNA molecules have been immobilized on nanoparticles1 and plates2 for use in gene diagnosis and single nucleotide polymorphism analysis, and materials bearing immobilized proteins have been utilized for antibody arrays,3 proteomics,4 and imaging.5 Many methods of protein immobilization have been proposed, and each method has both advantages and drawbacks. In general, traditional physical adsorption is the easiest method of immobilization, but the interaction force in such cases is too weak to adequately retain proteins on solid substrates. Chemical coupling between functional groups on proteins and solid substrates can irreversibly immobilize proteins;6 but the coupling reactions can be complicated, and such methods require the fabrication of solid surfaces with specific functional groups. Furthermore, both physical adsorption and chemical coupling methods often lead to a decrease in the activity of the immobilized proteins because neither of these methods offers control of protein orientation at * To whom correspondence should be addressed. Phone: +81-22-7957276. Fax: +81-22-795-7276. E-mail: [email protected]. † Institute of Multidisciplinary Research for Advanced Materials, Tohoku University. ‡ Graduate School of Engineering, Tohoku University. § Center for Interdisciplinary Research, Tohoku University. | NEDO. ⊥ The University of Tokyo. [ Osaka University.

the surface and gets around denaturation of proteins.7 To attain high-performance protein-immobilized plates, both the solid substrate surface and the conjugation system should be optimized to immobilize target proteins in active orientation and minimize nonspecific interactions between immobilized proteins and the surface. To control the orientation of proteins on solid substrates, proteins have been modified with various affinity tags. For example, proteins with poly histidine tags fused at the C- or N-terminus can be orientationally immobilized on solid substrates with nickel-resin surfaces,8 and the antigen peptides, such as c-myc,9 FLAG peptides,10 and biotinylated peptides,11 can also be used to control protein orientation at surfaces. The use of tag-fused proteins in conjunction with a surface containing complementary antibodies results in highly active immobilized protein systems. Recently, several peptides with an affinity for nonbiological materials have been identified by means of a combinatorial library approach,12,13 and these peptides are promising for use in bottom-up fabrication procedures in the field of bionanotechnology. Through the use of these peptides, proteins can be directly immobilized with desired orientations and without the need for surface fabrication of substrate or complicated conjugation processes.14 In addition, direct electron transfer can occur between immobilized redox proteins and unmodified inorganic surfaces because no interface membranes exist between proteins and substrate.15 In biological assays, metals such as gold and platinum are preferred owing to their conductivity and the

10.1021/jp907731b  2010 American Chemical Society Published on Web 12/10/2009

Immobilization of Proteins utilization of metal-thiol linkages in immobilizing proteins.16 Furthermore, the nonconductive or semiconductive properties of metal oxide materials make them attractive for use in fieldeffect transistors17 and as electrodes for protein redox reactions.18-20 Combinatorial identification of material-binding peptides shows the potential of them in biological assay systems using various material substrates; however, the binding mechanisms of material-binding peptides have not been elucidated. By describing the physicochemical mechanism, these peptides could be further optimized for use in a variety of biological applications. In this study, we used a ZnO-binding peptide (ZnOBP)21 for selective and orderly immobilization of recombinant proteins on a ZnO substrate, and we physicochemically analyzed the mechanism of the immobilization of proteins via ZnOBP. Green fluorescent protein (GFP) with ZnOBP fused at the N-terminus showed specificity for adsorption on ZnO substrates without any nonspecific protein adsorption on the substrate when an appropriate solution was used. Consequently, we could quantitatively analyze the binding of ZnOBP-fused proteins on solid materials. Using the ZnOBP-ZnO system, we performed thermodynamic analysis on the binding properties of intact and shortened ZnOBP tags to describe how the ZnO-binding function was expressed as the tag was elongated from a single amino acid to the intact ZnOBP sequence. Materials and Methods Expression and Preparation of ZnOBP-Fused GFP. The DNA sequences coding ZnOBP (EAHVMHKVAPRP), a linker sequence of GGGSAGSAAGSGEF,22 and GFP in that order from the N-terminus, were amplified from the pGFP vector (Clontech Laboratories, Inc., Mountain View, CA) by means of an overlap extension polymerase chain reaction with KODplus DNA polymerase, and the sequences were inserted into thepET20bvectorsbyNdeI-EcoRIdigestion(pET20b-ZnOBP-GFP). For GFP with shortened ZnOBP, the DNA sequences coding shortened ZnOBP were inserted into the NdeI-BamHI fragment of the pET20b-ZnOBP-GFP vector without ZnOBP. Escherichia (E) coli strain BL21 (DE3) cells were transformed with the expression plasmid encoding GFP fused with ZnOBP or shortened ZnOBP. E. coli cells harboring the plasmid were incubated in lysogeny broth medium at 28 °C, and expression of recombinant GFP was induced by adding 1 mM isopropyl β-D-thiogalactopyranoside. The harvested cells were centrifuged at 5800g for 30 min and suspended in 50 mM Tris-HCl (pH 8.0) buffer containing 200 mM NaCl. The cells were lysed by sonication in a Kubota 201 M sonicator operated at ∼180 W for 15 min at 4 °C, and then the suspension was centrifuged at 5800g for 30 min at 4 °C. After NaCl was removed from the supernatant by dialysis, the solution was loaded on a 5-mL HiTrap Q XL column (GE Healthcare, Tokyo, Japan) with a 0.1-1 M gradient of NaCl in 50 mM Tris-HCl buffer (pH 9.0). Finally, the fractionated recombinant GFP was refined by gel filtration chromatography (HiLoad 26/60 Superdex 75 prep grade, Sephacryl S-200; GE Healthcare, Tokyo, Japan). The molecular weight of refined GFP was confirmed from mass spectra obtained with a REFLEX III matrix-assisted laser desorption/ionization-time-of-flight mass spectrometer (Bruker Analytische, GmbH, Germany) equipped with a nitrogen laser (337 nm). Binding Analysis of ZnOBP-Fused GFP onto Inorganic Material Surfaces. Bovine serum albumin (BSA) and lysozyme (Sigma-Aldrich Japan Co., Tokyo, Japan) were dissolved in a Tris-HCl solution (50 mM Tris, 30 mM NaCl, 0.05% Tween20, pH 7.5) or a phosphate solution (10 mM or 30 mM NaH2PO4,

J. Phys. Chem. B, Vol. 114, No. 1, 2010 481 30 mM NaCl, 0.05% Tween20, pH 7.5), respectively, and each solution was mixed with 25 mg of gold particles (Sigma-Aldrich Japan Co., Tokyo, Japan) or 25 mg of ZnO particles (Hosokawa Powder Technology Research, Osaka, Japan) for 10 min. After centrifugation at 10000g for 15 min, the protein concentration in the supernatants was estimated by absorbance at 280 nm. One milliliter of the solution of recombinant GFPs with a series of ZnOBP fragments in phosphate solutions (10-50 mM NaH2PO4, 30 mM NaCl, 0.05% Tween20, pH 7.5) was mixed with 25 mg of ZnO particles, which had a Brunauer-EmmettTeller (BET) surface area of 9.7 m2/g (average diameter: 109 nm, Hosokawa Powder Technology Research, Osaka, Japan), at various concentrations of GFP for 10 min. After centrifugation at 10000g for 15 min, the fluorescence from GFP in the supernatants was measured on the FP-6500 fluorescence spectrometer with an excitation wavelength of 490 nm (Jasco Inc., Tokyo, Japan). We also performed binding analysis on recombinant GFP against 7 mg of TiO2 particles (BET surface area, 39 m2/g; average diameter, 42 nm; C.I. Kasei. Co., Ltd., Tokyo, Japan), 11 mg of SnO2 particles (BET surface area, 22 m2/g; average diameter, 40 nm; Hosokawa Powder Technology Research, Osaka, Japan), 8 mg of Fe2O3 particles (BET surface area, 34 m2/g; average diameter, 34 nm; C.I. Kasei. Co., Ltd., Tokyo, Japan), and 5 mg of Al2O3 particles (BET surface area, 51 m2/g; diameter, 33 nm; C.I. Kasei. Co., Ltd., Tokyo, Japan). By adding the mentioned weight of each particle, about 0.25 m2 surface areas were supplied in all the binding experiments. Reflectometric Interference Spectroscopy. A biosensor array system with reflectometric interference spectroscopy (Fluidware Technology Inc., Kawaguchi, Japan) was used to measure the binding of recombinant GFP on a ZnO film. Recombinant GFP (1 µM) in phosphate buffer was flowed at a rate of 10 µL/min for 220 s onto a ZnO film deposited on a silicon plate, and then phosphate buffer was flowed onto the plate to rinse. The change in wavelength with minimum reflection intensity was measured to observe the adsorption of proteins on the ZnO film. Fluorescence Microscopy. Fluorescence microscopy images were obtained with an inverted microscope (IX 81, Olympus Corp., Japan) equipped with a luminous xenon source (ULH75XEAPO, Olympus Corp., Japan) at room temperature. After recombinant GFP proteins in 30 mM phosphate solution were mixed with 25 mg of ZnO particles or Fe2O3 particles, the mixture was centrifuged at 10000g for 15 min, and the precipitates were washed several times with the 30 mM phosphate solution. The particles were suspended in 30 mM phosphate solution, and then the suspension was deposited on a glass plate to measure the fluorescence from GFP on the particles by means of fluorescence microscopy. The fluorescence from the excited suspension was observed through a 530-nm filter (XF22, Omega Optical Inc., USA), using a UPLSAPO 100XO lens and a charge-coupled device camera (CoolSNAP ER, Olympus Corp., Japan). Results Nonspecific Adsorption of Proteins on Gold and ZnO Particles. Figure 1 shows the adsorption of BSA and hen egg white lysozyme proteins on gold particles and ZnO particles, as estimated from the protein concentration of the supernatant after mixing the proteins and particles. In a 50 mM Tris-HCl solution at pH 7.5, BSA (pI ∼5) was adsorbed on the gold particles at surface coverages as high as 130 nmol/m2 (open circles in Figure 1a), whereas the surface coverage on the ZnO particles was one-third that on the gold particles (open squares

482

J. Phys. Chem. B, Vol. 114, No. 1, 2010

Yokoo et al.

Figure 1. Adsorption of BSA (a) and lysozyme (b) on gold and ZnO particles. The amounts of BSA adsorbed from a 50 mM Tris-HCl solution (pH 7.5) on gold and ZnO particles are plotted as open circles and squares, respectively, with solid lines. The amounts of BSA adsorbed on ZnO from phosphate solutions (pH 7.5) are plotted as open squares with a dashed line (10 mM phosphate) and open squares with a dotted-dashed line (50 mM phosphate). The amounts of lysozyme adsorbed on gold and ZnO from the Tris-HCl solution are plotted as closed circles (gold) and squares (ZnO) with solid lines. The amounts of lysozyme adsorbed on ZnO from phosphate solutions are plotted as closed squares with a dashed line (10 mM phosphate).

TABLE 1: Estimated Association Equilibrium Constants (Ka) and Saturation Amounts (Ws) for Immobilized ZnOBP-Fused GFP on ZnO in Phosphate Solutions at Concentrations of 10 mM, 30 mM, and 50 mMa phosphate concentration Ka/(105 M-1) Ws/(nmol/m2)

10 mΜ

20 mM

30 mΜ

50 mΜ

59 87

30 78

6.5 77

1.9 41

a Values were estimated from the Langmuir adsorption isotherm equation.

Figure 2. Adsorption of native GFP (circles) and ZnOBP-fused GFP (squares) on ZnO particles in 10 mM (open circles and squares) and 30 mM (closed circles and squares) phosphate solutions.

with solid line in Figure 1a). The surface coverage of BSA on the ZnO particles decreased further in a 10 mM phosphate solution (open squares with dashed line in Figure 1a), and an increase in phosphate concentration to 50 mM completely suppressed the nonspecific adsorption of the BSA (open squares with dotted-dashed line in Figure 1a). This observed suppression of adsorption might have been caused by interactions between phosphate groups and the ZnO surface; such interactions would compete with BSA adsorption. When the adsorption of lysozyme, which has a high pI of ∼11 and contains eight cysteines forming four disulfide bonds, was examined for gold and ZnO, behavior similar to that observed for BSA was seen (Figure 1b). These results indicate that the use of phosphate solution can completely suppress nonspecific adsorption of proteins on ZnO particles, regardless of the proteins’ charge. Protein Immobilization on ZnO with ZnOBP. To selectively immobilize GFP at ZnO via the ZnO-binding peptide (ZnOBP), GFP with ZnOBP fused at the GFP N-terminus (ZnOBP-fused GFP) was mixed with ZnO particles in phosphate solution at pH 7.5 (Figure 2). In 10 mM phosphate solution, only a small amount of native GFP was adsorbed on the ZnO particles (open circles), whereas ZnOBP-fused GFP was immobilized to a substantially greater extent than was native GFP

on ZnO (open squares). The selective immobilization of ZnOBPfused GFP was sufficient even in 30 mM phosphate solution (closed squares), in which the nonspecific adsorption of native GFP was completely suppressed (closed circles). However, this increase in the phosphate concentration caused partial dissociation of ZnOBP-fused GFP from the ZnO surface. These results imply that the utilization of ZnO as a substrate and ZnOBP as a tag in phosphate solution is effective for selective immobilization of proteins on material surfaces. Using the Langmuir adsorption isotherm eq (1/W ) (1/Ws) + (1/KaWs)(1/Ceq)), we estimated the association equilibrium constants (Ka) and saturation amounts (Ws) for immobilized ZnOBP-fused GFP on ZnO at several phosphate concentrations (Table 1). All the adsorption isotherms were sufficiently fitted by the equation, and the fitting results showed that an increase in phosphate concentration decreased the Ka and Ws values. To examine the immobilization of GFP on ZnO in flow conditions, native and ZnOBP-fused GFP solutions were flowed onto a ZnO film deposited on a Si plate, and the immobilization was detected by reflectometric interference spectroscopy (RIfS). Figure 3 shows the time-dependent change of the wavelength with minimum intensity in the reflected spectrum for various sample solutions. For a flowing 10 mM phosphate solution, little native GFP was adsorbed on the ZnO film (dotted line), and the use of a 30 mM phosphate solution completely suppressed the adsorption of native GFP (dashed line). In contrast, we observed spontaneous immobilization of ZnOBP-fused GFP even in flowing 30 mM phosphate solution. The surface coverage of immobilized protein in this case was 89 nmol/m2,

Immobilization of Proteins

Figure 3. RIfS sensorgrams for the interaction of 1 µM solutions of native and ZnOBP-fused GFP with a ZnO film. Experimental conditions: native GFP in 10 mM phosphate solution (dotted line), native GFP in 30 mM phosphate solution (dashed line), ZnOBP-fused GFP in 30 mM phosphate solution (solid line). Flow rate for all samples was 10 µL/min.

Figure 4. Adsorption of recombinant GFP with N1-3 ZnOBP (open circles), N1-4 ZnOBP (closed circles), N1-6 ZnOBP (open squares), N1-7 ZnOBP (closed squares), N1-10 ZnOBP (open triangles), N1-11 ZnOBP (closed triangles), and intact ZnOBP (crosses) on ZnO particles in a 30 mM phosphate solution.

which was comparable to the amount that was estimated from the Langmuir isotherm (Table 1). These results indicate that the implementation of the ZnOBP tag allowed GFP to be directly immobilized on a ZnO film under flow conditions. Expression of ZnO-Binding Function in the Elongation of the ZnOBP Sequence. To analyze the ZnO-binding function of ZnOBP, we prepared a series of GFPs in which the peptide sequence at the GFP N-terminus was elongated from a single glutamine residue to the intact ZnOBP sequence, and the interactions between the peptide tag and ZnO particles were observed in a 30 mM phosphate solution (Figure 4). The amount of immobilized GFP increased as the peptide was elongated, but the trend was not linear. Peptides with fewer than three amino acid residues (N1 ZnOBP, E; N1-2 ZnOBP, EA; N1-3 ZnOBP, EAH) showed no binding of GFP to ZnO particles (open circles), but N1-4 ZnOBP (EAHV) allowed for adsorption onto the ZnO particles (closed circles), and a similar amount of GFP adsorption was observed when the sequence was elongated to N1-6 ZnOBP (EAHVMH) (open squares). For the elongation from N1-6 ZnOBP, the amount of adsorbed GFP remarkably increased at two specific points: the seventh lysine residue (closed squares) and the 11th arginine residue (closed triangles). Figure 5 shows the Ka and Ws values for adsorption of GFP with shortened ZnOBP on ZnO particles. For the elongation from N1-4 ZnOBP to N1-7 ZnOBP, the Ka values of fused peptide tags increased as the number of charged amino acid residues increased; most notably, the addition of the seventh

J. Phys. Chem. B, Vol. 114, No. 1, 2010 483 lysine residue resulted in N1-7 ZnOBP having the highest affinity for ZnO of all the fragmented ZnOBP tags, suggesting that a “hot spot” for ZnO affinity existed around the seventh lysine residue of ZnOBP. The addition of the seventh lysine residue also increased the Ws value, which corresponds to an increase in the number of binding sites on ZnO, further demonstrating the importance of the sequence around the seventh lysine residue for adsorption on ZnO. When N1-7 ZnOBP was elongated to N1-8 ZnOBP, the Ka value decreased and subsequently remained nearly constant as the tag was elongated to intact ZnOBP. In terms of Ws values, little changes were observed in the elongation from N1-7 ZnOBP to N1-10 ZnOBP; however, the addition of the 11th arginine residue increased the Ws value with little change in Ka. These results suggest that the N8-12 sequence (VAPRP) structurally interfered with the binding of the N1-7 sequence but that the segment around the 11th arginine residue had another low-affinity binding ZnO site different from that of N1-7 ZnOBP. Selectivity of ZnOBP Tag for Immobilization of Proteins on Other Materials. To analyze the affinity of the ZnOBP tag for materials other than ZnO, we measured the optical and fluorescent microscope images of ZnO and Fe2O3 particles that had been mixed with ZnOBP-fused GFP and then washed as well as images of a mixture of both types of particles (Figure 6). When Fe2O3 particles were absent from this mixture, fluorescence derived from GFP was observed on aggregated ZnO particles, but the addition of Fe2O3 particles resulted in a decrease in observed fluorescence intensity because GFP did not adsorb on the Fe2O3 particles. These results indicate that the ZnOBP tag discriminated between ZnO and nontarget substrates in a mixture of inorganic materials. Additionally, we measured the adsorption isotherms on several materials for GFP with intact ZnOBP, N1-7 ZnOBP, and N1-4 ZnOBP (Figure 7). Although a few GFP proteins with intact ZnOBP were immobilized on NiO particles, no GFP immobilization was observed on TiO2, SnO2, Fe2O3, or Al2O3 (Figure 7a). These results indicate that intact ZnOBP was clearly selective toward ZnO inorganic materials. Notably, this selectivity was observed for N1-7 ZnOBP and N1-4 ZnOBP, as well (Figure 7b and c). Therefore, even sequences as short as N1-4 had a selective affinity for ZnO. To elucidate the thermodynamic mechanism for protein immobilization via ZnOBP, we measured the adsorption isotherms at several temperatures for ZnOBP-fused GFP on ZnO particles and estimated the thermodynamic properties of ZnOBPfused GFP from van’t Hoff plots (Table 2). The thermodynamic properties of N1-4 ZnOBP, N1-7 ZnOBP, and intact ZnOBP tags showed that all the immobilizations of GFP via ZnOBP tags were predominantly driven by an enthalpy change (∆H) and that the variation in Ka values that resulted from shortening ZnOBP correlated with the change in ∆H (∆∆H) rather than with any changes in entropy (∆∆S). These results demonstrate that the binding of ZnOBP tags on the ZnO surface occurred via electrostatic interactions. Notably, the ∆∆H value for N1-7 ZnOBP increased despite the observed decrease in the N8-12 amino acid sequence (VAPRP) on the C-terminus side. This defect in the C-terminus sequence might have resulted in an increase in hydrogen or coordination bonding between the N1-7 sequence and the ZnO surface; consequently, the increase in the electrostatic interaction may have restricted the peptide’s mobility, thus decreasing the ∆S value.

484

J. Phys. Chem. B, Vol. 114, No. 1, 2010

Yokoo et al.

Figure 5. Estimated association equilibrium constants Ka (a) and saturation amounts Ws (b) of recombinant GFPs with various shortened ZnOBPtags for adsorption on ZnO particles.

Figure 6. Optical and fluorescent microscope images of 25 mg ZnO particles (a), an equal mixture of ZnO (12 mg) and Fe2O3 (12 mg) particles (b), and 25 mg Fe2O3 particles (c). Each set of particles was mixed with 1 mL of 10 µM ZnOBP-fused GFP (30 mM phosphate) solution and then washed with phosphate solution.

Discussion Advances in molecular evolution have enabled the identification of various peptides with affinity for inorganic material surfaces, and therefore, relatively large numbers of studies have been carried out on the utilization of such peptides for the immobilization and patterning of proteins and nanomaterials at inorganic surfaces.23-27 The immobilization of specific proteins via peptides with an affinity for inorganic materials is attractive because such methods eliminate the need for special substrate treatments and complicated immobilization processes; however, nonspecific adsorption of proteins on the exposed surface of inorganic materials must also be inhibited. In this study, we found that the combination of solution and material substrate used in the immobilization process was important for inhibiting nonspecific adsorption of proteins on used materials. At pH 7.5, the surface of ZnO, which has a ζ potential of 8-9, is positively charged, whereas BSA (pI ) ∼5) and lysozyme (pI ) ∼11) are negatively and positively charged, respectively. Therefore, lysozyme is expected to be adsorbed on the surface of ZnO at pH 7.5, but BSA should not be; however, in this study, both proteins were adsorbed on ZnO in Tris-HCl buffer. In contrast, in the phosphate solutions used here, the nonspecific adsorption of BSA and lysozyme was completely inhibited, and the extent

of inhibition ability increased with increasing phosphate concentration. This inhibition effect was also observed for GFP at ZnO. Notably, the conjugate acids in Tris-HCl and in phosphate buffer are cationic and anionic at pH 7.5, respectively. Therefore, in phosphate solution, the anion pair (H2PO4- and HPO42-) may have interacted with the positively charged ZnO surface so that nonspecific adsorption of proteins on the ZnO surface competed with electrostatic interaction between HxPO4y- with ZnO. To avoid nonspecific adsorption, neutral surfactants, such as Tween and Triton, have been often used; however, the increase in Tween concentration to 0.5% in the Tris-HCl solution used here did not substantially decrease the nonspecific adsorption of proteins on ZnO (data not shown). Therefore, we concluded that electrostatic interaction was one of the major factors controlling nonspecific adsorption in this system. Using the appropriate combination of ZnO surface and phosphate solution, we physicochemically analyzed the binding properties of ZnOBP on ZnO particles and calculated the Gibbs free energy change (∆G) associated with the binding of ZnOBP on the ZnO surface. Recently, kinetic studies using a quartz crystal microbalance and surface plasmon resonance techniques have been undertaken to examine the binding mechanisms of material-binding peptides. The reported ∆G values for gold-,

Immobilization of Proteins

J. Phys. Chem. B, Vol. 114, No. 1, 2010 485

Figure 7. Adsorption of recombinant GFP with intact ZnOBP (a), N1-7 (b), and N1-4 (c) for ZnO (open circles), NiO (closed circles), TiO2 (open squares), SnO2 (closed squares), Fe2O3 (open triangles), and Al2O3 (closed squares) particles in a 30 mM phosphate solution. The concentrations of surface area supplied from added particles were adjusted to 0.25 m2.

TABLE 2: Thermodynamic Parameters Calculated from van’t Hoff Plots for GFP Adsorption onto ZnO Particles via Intact ZnOBP, N1-4 ZnOBP, and N1-7 ZnOBP

N1-4 N1-7 ZnOBP

∆G, (∆∆G)/ (kJ/mol)

∆H, (∆∆H)/ (kJ/mol)

T∆S, (T∆∆S)/ (kJ/mol)

-30.1 (3.0) -34.4 (-1.3) -33.1 (-)

-27.3 (4.5) -38.8 (-7.0) -31.8 (-)

2.9 (1.5) -4.4 (5.7) 1.3 (-)

platinum-, and quartz-binding linear peptides28-31 are comparable to the values we obtained for ZnOBP, suggesting that linear inorganic-material-binding short peptides generally have ∆G values of 25-35 kJ/mol. In addition, we used van’t Hoff equations to estimate the ∆H and ∆S values associated with adsorption of ZnOBP-fused GFP. The results described that ∆H was the major contributor to ∆G, which implies that the electrostatic interaction between ZnOBP and the ZnO surface was enthalpy-driven, involving hydrogen bonding. This implication is also supported by the competitive assay results, which showed that an increase in phosphate concentration inhibited the binding of ZnOBP on ZnO (Table 1). Several other reports have suggested that electrostatic interaction by means of hydrogen bonding is critical for the binding of peptides to inorganic materials. Patwardhan et al. used molecular modeling to demonstrate the importance of basic

amino acid residues and hydrophilic residues,32 and Tamerler et al. speculated that hydroxyl and amine groups in methionine, lysine, threonine, glutamine, and serine residues may play an important role for enhancing the adsorptive properties of goldbinding peptides.28 Willett et al. examined the adhesion of homopeptides composed of a single amino acids to various inorganic materials, and they showed that charged amino acids interacted strongly with material surfaces.33 In the present study, N1-7 ZnOBP, which had the highest affinity for ZnO, contained four of the five charged amino acid residues found in intact ZnOBP, and the seventh Lys residue was observed to critically influence the Ka value. Therefore, in terms of thermodynamics, charged amino acids also played an important role in the interaction of ZnOBP with ZnO. The examination of ZnO-binding assays using a series of shortened ZnOBP tags demonstrated how the selective binding function of ZnOBP for ZnO is expressed. By elongating the peptide tag of GFP from a single glutamic acid residue to the intact ZnOBP, we found that the tripeptide EAH hardly bound to ZnO, but the tetrapeptide EAHV (N1-4 ZnOBP) bound with selectivity toward ZnO. The binding of the EAHV tetrapeptide was thermodynamically dependent on enthalpy changes, whereas the side chain in the valine residue can be considered to have no contribution to these changes. We concluded that the addition of valine at the C-terminus of EAH induced a conformational change of the EAH sequence to promote its binding onto the ZnO surface. As charged residues were added in the elongation from EAHV to N1-7 ZnOBP (EAHVMHKV), the peptide’s affinity for ZnO increased, but the addition of valine residue decreased the affinity. Therefore, the addition of amino acids next to charged residues critically influences the binding function of the N1-8 ZnOBP. The elongation from N1-8 ZnOBP to intact ZnOBP resulted in little change in the peptide’s affinity for ZnO, but the addition of the charged amino acid arginine increased the amount of peptides adsorbed on ZnO. This increase in peptide adsorption despite a relatively constant affinity suggests that the segment surrounding the 11th arginine residue weakly binds on another ZnO site different from that of N1-7 ZnOBP segment. It should be noted that the bindings of all the measured ZnOBP peptides are specific for ZnO surface. Our thermodynamic analysis implies electrostatic interactions between all the applied peptides and ZnO surface; however, these results cannot directly elucidate the specificity of peptide recognition because other materials also have surfaces that can form charge-charge interactions and hydrogen bonds. The ZnO-binding assays using a series of shortened ZnOBP tags suggested that the conformation of peptide is critical to the binding to ZnO, and Seker et al. also showed the importance of peptide conformation for the binding to material surface.29 Peptides might be able to recognize the configuration of hydroxyl group and exposed metal atom on material surface using their conformation. In conclusion, we selectively immobilized recombinant GFP via ZnOBP on ZnO substrates, avoiding nonspecific protein adsorption by selecting a buffer solution with an appropriate phosphate concentration. Thermodynamically quantitative analyses showed a predominantly enthalpy-dependent electrostatic interaction between the ZnOBP-fused GFP and ZnO, and the results obtained from the elongation of the fused peptide from a single glutamine acid to the intact ZnOBP demonstrated the enhancement of the peptide’s affinity for ZnO after the selective recognition of the surface was achieved. We observed that the presence of charged amino acids and their next residues induced

486

J. Phys. Chem. B, Vol. 114, No. 1, 2010

critical points for selective affinity, indicating the importance of steric conformation for material-binding function of the peptides. Acknowledgment. This work was supported by a Scientific Research Grant from the Ministry of Education, Science, Sports, and Culture of Japan (M.U.), by the New Energy and Industrial Technology Development Organization (NEDO) of Japan (M.U.), and by Precursory Research for Embryonic Science and Technology from the Japan Science and Technology Agency (JST). This research was also supported in part by the Association for the Progress of New Chemistry foundation (M.U.). References and Notes (1) Storhoff, J. J.; Elghanian, R.; Mucic, R. C.; Mirkin, C. A.; Letsinger, R. L. J. Am. Chem. Soc. 1998, 120, 1959–1964. (2) Taton, T. A.; Lu, G.; Mirkin, C. A. J. Am. Chem. Soc. 2001, 123, 5164–5165. (3) Berger, R.; Delamarche, E.; Lang, H. P.; Gerber, Ch.; Gimzewski, J. K.; Meyer, E.; Gu¨ntherodt, H.-J. Science 1997, 276, 2021–2024. (4) Lomas, L. O. In Protein Microarray Technology; Kambhampati, D., Ed.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, 2004; pp 153-163. (5) Weissleder, R.; Moore, A.; Mahmood, U.; Bhorade, R.; Benveniste, H.; Chiocca, E. A.; Basilion, J. P. Nat. Med. 2000, 6, 351–355. (6) Scha¨ferling, M.; Schiller, S.; Paul, H.; Kruschina, M.; Pavlickova, P.; Meerkamp, M.; Giammasi, C.; Kambhampati, D. Electrophoresis 2002, 23, 3097–3105. (7) Norde, W. AdV. Colloid Interface Sci. 1986, 25, 267–340. (8) Dietrich, C.; Boscheinen, O.; Scharf, K.-D.; Schmitt, L.; Robert Tampe´, R. Biochemistry 1996, 35, 1100–1105. (9) Evan, G. I.; Lewis, G. K.; Ramsay, G.; Bishop, J. M. Mol. Cell. Biol. 1985, 5, 3610–3616. (10) Knappik, A; Plu¨ckthun, A. Biotechniques 1994, 17, 754–761. (11) Gaja, T.; Meyera, S. C.; Ghosh, I. Protein Exp. Purif. 2007, 56, 54–61. (12) Whaley, S. R.; English, D. S.; Hu, E. L.; Barbara, P. E.; Belcher, A. M. Nature 2000, 405, 665–668. (13) Sarikaya, M.; Tamerler, C.; Jen, A. K.-Y.; Schulten, K.; Baneyx, F. Nat. Mater. 2003, 2, 577–585.

Yokoo et al. (14) Woodbury, R. G.; Wendin, C.; Clendenning, J.; Melendez, J.; Elkind, J.; Bartholomew, D.; Brown, S.; Furlong, C. E. Biosens. Bioelectron. 1998, 13, 1117–1126. (15) Li, Y.-F.; Liu, Z.-M.; Liu, Y.-L.; Yang, Y.-H.; Shen, G.-L.; Yu, R.-Q. Anal. Biochem. 2006, 349, 33–40. (16) Xiao, Y.; Patolsky, F.; Katz, E.; Hainfeld, J. F.; Willner, I. Science 2003, 299, 1877–1881. (17) Luo, X.-L.; Xu, J.-J.; Zhao, W.; Chen, H.-Y. Sens. Actuators, B 2004, 97, 249–255. (18) Topoglidis, E.; Cass, A. E. G.; Gilardi, G.; Sadeghi, S.; Beaumont, N.; Durrant, J. R. Anal. Chem. 1998, 70, 5111–5113. (19) Topoglidis, E.; Cass, A. E. G.; O’Regan, B.; Durrant, J. R. J. Electronanal. Chem. 2001, 517, 20–27. (20) Lin, K.-C.; Chen, S.-M. Biosens. Bioelectron. 2006, 21, 1737–1745. (21) Umetsu, M.; Mizuta, M.; Tsumoto, K.; Ohara, S.; Takami, S.; Watanabe, H.; Kumagai, I.; Adschiri, T. AdV. Mater. 2005, 17, 2571–2575. (22) Waldo, G. S.; Standish, B. M.; Berendzen, J.; Terwilliger, T. C. Nat. Biotechnol. 1999, 17, 691–695. (23) Sano, K.; Yoshii, S.; Yamashita, I.; Shiba, K. Nano Lett. 2007, 7, 3200–3202. (24) Park, T. J.; Lee, S. Y.; Lee, S. J.; Park, J. P.; Yang, K. S.; Lee, K.-B.; Ko, S.; Park, J. B.; Kim, T.; Kim, S. K.; Shin, Y. B.; Chung, B. H.; Ku, S.-J.; Kim, D. H.; Choi, I. S. Anal. Chem. 2006, 78, 7197–7205. (25) Krauland, E. M.; Peelle, B. R.; Wittrup, K. D.; Belcher, A. M. Biotechnol. Bioeng. 2007, 97, 1009–1020. (26) Kacar, T.; Ray, J.; Gungormus, M.; Oren, E. E.; Tamerler, C.; Sarikaya, M. AdV. Mater. 2009, 21, 295–299. (27) Wei, J. H.; Kacar, T.; Tamerler, C.; Sarikaya, M.; Ginger, D. S. Small 2009, 5, 689–693. (28) Termerler, C.; Oren, E. E.; Duman, M.; Venkatasubramanian, E.; Sarikaya, M. Langmuir 2006, 22, 7712–7718. (29) Seker, U. O. S.; Wilson, B.; Dincer, S.; Kim, I. W.; Oren, E. E.; Evans, J. S.; Tamerler, C.; Sarikaya, M. Langmuir 2007, 23, 7895–7900. (30) Hnilova, M.; Oren, E. E.; Seker, U. O. S.; Wilson, B. R.; Coliino, S.; Evans, J. S.; Tamerler, C.; Sarikaya, M. Langmuir 2008, 24, 12440– 12445. (31) Seker, U. O. S.; Wilson, B.; Sahin, D.; Tamerler, C.; Sarikaya, M. Macromolecules 2009, 10, 250–257. (32) Patwardhan, S. V.; Patwardhan, G.; Perry, C. C. J. Mater. Chem. 2007, 17, 2875–2884. (33) Willett, R. L.; Baldwin, K. W.; West, K. W.; Pfeiffer, L. N. Proc. Natl. Acad. Sci., U.S.A. 2005, 102, 7817–7822.

JP907731B