Specificity and Biomineralization Activities of Ti-Binding Peptide-1

Ken-Ichi Sano, Hiroyuki Sasaki, and Kiyotaka Shiba*. Department of ...... Akio Kuroda , Maxym Alexandrov , Tomoki Nishimura , Takenori Ishida. Biotech...
0 downloads 0 Views 433KB Size
3090

Langmuir 2005, 21, 3090-3095

Specificity and Biomineralization Activities of Ti-Binding Peptide-1 (TBP-1) Ken-Ichi Sano,† Hiroyuki Sasaki,‡ and Kiyotaka Shiba*,† Department of Protein Engineering, Cancer Institute, Japanese Foundation for Cancer Research, 1-37-1 Kami-Ikebukuro, Toshima, Tokyo 170-8455, Japan, CREST, JST, Toshima, Tokyo 170-8455, Japan, and Department of Molecular Cell Biology, Institute of DNA Medicine, The Jikei University School of Medicine, Minato, Tokyo, Japan Received October 18, 2004. In Final Form: January 6, 2005 Numerous peptide aptamers that recognize inorganic materials have been isolated using in vitro peptide evolution systems. However, it remains unknown how peptides interact with inorganic materials or how specific those interactions are. We, therefore, assessed the target specificities of the peptide aptamer TBP-1 (RKLPDAPGMHTW) by monitoring its ability to bind 10 different metals. We found that phages displaying TBP-1 bound to Ti, Si, and Ag surfaces but not to Au, Cr, Pt, Sn, Zn, Cu, or Fe. As previously seen with Ti, binding to Si and Ag was diminished by R1A, P4A, or D5A mutation, suggesting that the same molecular mechanism underlies TBP-1 binding to all three materials. We also observed that a synthetic TBP-1 peptide mediated mineralization of both silica and Ag. It, thus, appears that although the overall chemical characteristics of Ti, Si, and Ag surfaces are dissimilar, they share a common subnanometric structure that is recognized by TBP-1.

Introduction During the course of evolution, combinatorics of 20 amino acid species gave rise to a bewildering array of proteins (polypeptides) displaying unparalleled functionality, among which are a wide variety of catalysts (enzymes), specific binding and structural molecules, and molecular motors, just to name a few. An understanding of the rules governing construction of natural proteins that would enable us to create novels proteins with desired functions is the ultimate goal in the field of protein science. With the exception of several admirable achievements, however, “the rational design” of novel artificial polypeptides having specific functions has met with little success.1 Instead, it has demonstrated our need for a deeper understanding of the relationship between the amino acid sequences of proteins and their functions. By contrast, the irrational approach, that is, “selection from random sequences”,2 has been used successfully to create a variety of artificial polypeptides,3 although their sizes are much smaller and their functions much simpler (they were mostly created as binders) than those of natural polypeptides. With the irrational approach, a molecularly diverse pool of peptides is first prepared by combinatorial polymerization of amino acids, after which clones possessing desired functions are selected from the pool. One of the most commonly employed formulas within this methodology is the “peptide-phage display system,” in which phages * To whom correspondence should be addressed at Japanese Foundation for Cancer Research. Phone/fax: +81-3-5394-3903. E-mail: [email protected]. † Japanese Foundation for Cancer Research and CREST, JST. ‡ The Jikei University School of Medicine. (1) (a) Dahiyat, B. I.; Mayo, S. L. Science 1997, 278, 82-87. (b) Harbury, P. B.; Plecs, J. J.; Tidor, B.; Alber, T.; Kim, P. S. Science 1998, 282, 1462-1467. (c) Kuhlman, B.; Dantas, G.; Ireton, G. C.; Varani, G.; Stoddard, B. L.; Baker, D. Science 2003, 302, 1364-1368. (d) Kaplan, J.; DeGrado, W. F. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 1156611570. (2) (a) Szostak, J. W. Trends Biochem. Sci. 1992, 17, 89-93. (b) Kauffman, S. A. J. Theor. Biol. 1992, 157, 1-7. (3) Keefe, A. D.; Szostak, J. W. Nature 2001, 410, 715-718.

specifically binding to defined target molecules via peptides displayed on their virus particles are selected from a phage library displaying random peptide sequences.4 This type of peptide evolution system was first established to create aptamers that target biological macromolecules, such as receptors, antibodies, and enzymes.5 Contributing to the specific, tight binding of these peptide aptamers to proteinous macromolecules are the constellations of electron clouds attributed to both the side chains and main chains of their amino acids. In comparison to biomacromolecules, the structures of the electron clouds of inorganic materials are simple and rather monotonic, making it questionable whether a specific peptide aptamer could be created to bind to inorganic material. This question was elegantly answered in 1992 by S. Brown who constructed a bacterial (Escherichia coli) library displaying random peptides by inserting concatamers of random oligonucleotides into a gene encoding an outer membrane protein and then selected the clones that bound to an Fe2O3 surface.6 He then expanded his exploration to isolate a separate Au binding peptide.7 Later, Belcher’s group applied a peptide-phage display system to isolate aptamers that bound to semiconductive materials, including GaAs (100).8 At present, numerous peptide aptamers targeting a variety of inorganic materials have been created using peptide-phage display systems or other related artificial evolution systems. Among these are aptamers that bind to Ag,9 Pt,10 Ti,11 Pd,10 Cr2O3,12 PbO2,12 CoO,12 MnO2,12 (4) Barbas, C. F., III; Burton, D. R.; Scott, J. K.; Silverman, G. J. Phage display : a laboratory manual; Cold Spring Harbor Laboratory Press: New York, 2001. (5) (a) Cwirla, S. E.; Peters, E. A.; Barrett, R. W.; Dower, W. J. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 6378-6382. (b) Scott, J. K.; Smith, G. P. Science 1990, 249, 386-390. (c) Devlin, J. J.; Panganiban, L. C.; Devlin, P. E. Science 1990, 249, 404-406. (6) Brown, S. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 8651-8655. (7) Brown, S. Nat. Biotechnol. 1997, 15, 269-272. (8) Whaley, S. R.; English, D. S.; Hu, E. L.; Barbara, P. F.; Belcher, A. M. Nature 2000, 405, 665-668. (9) Naik, R. R.; Stringer, S. J.; Agarwal, G.; Jones, S. E.; Stone, M. O. Nat. Mater. 2002, 1, 169-172. (10) Sarikaya, M.; Tamerler, C.; Jen, A. K.; Schulten, K.; Baneyx, F. Nat. Mater. 2003, 2, 577-585. (11) Sano, K.; Shiba, K. J. Am. Chem. Soc. 2003, 125, 14234-14235.

10.1021/la047428m CCC: $30.25 © 2005 American Chemical Society Published on Web 02/23/2005

Subnanometric Structure Biological discrimination

ZnO,13,14 CaCO3,15 SiO2,16 ZnS,17 Cu2O,13 carbon nanotubes,18 carbon nanohorns,19 C60,20 and zeolites.21 Peptide aptamers that bind to the surface of defined inorganic materials might have their greatest impact in the field of bionanotechnology. This is because with sufficient understanding of the mechanism underlying the molecular interaction between peptide aptamers and their target materials it should be possible to develop sophisticated composite nanomaterials using peptide aptamers. Thus, key questions are how do peptide aptamers interact with inorganic materials and how specific are those interactions. Much of this information could be provided by high-resolution X-ray diffractions of cocrystals containing a peptide and its target molecule.22 However, it is generally very difficult to obtain cocrystals containing a peptide and material substrate that give diffractions suitable for building molecular models. Mutational analyses can also provide useful insights into the binding mechanism. For instance, we substituted with alanine each position of a 12-mer peptide, TBP-1 (RKLPDAPGMHTW), which was isolated as an aptamer targeting a Ti surface.11 Because the side chain of alanine is an uncharged methyl group, we reasoned that if a particular position was critical for binding, the alaninesubstituted mutant would show reduced binding ability. From these mutational analyses, we learned that the arginine in the first position (R1), the proline in the fourth position (P4), and the aspartic acid in the fifth position (D5) are all critical for TBP-1 binding to Ti. Moreover, because proline is known to introduce a kink in the main chain of peptides, R1 and D5 would be expected to be directed toward the same surface, resulting in a Lewis base and a Lewis acid with a proximity to one another measurable on a nanometric scale. Bearing that in mind, it is noteworthy that the surface of Ti is oxidized in a biological environment and displays hydroxyl groups that can act as either an acid (-O-) or a base (-OH2+),23 making interionic attraction the most likely mode of interaction between TBP-1 and a Ti surface. Although such mutational studies are commonly used in fields of biology, its use to date with aptamers targeting inorganic materials has been very limited.18 It is well-known that under normal ambient conditions the surfaces of metals are often covered by an oxide film, that these material-oxide surfaces are generally hydroxylated, and that the hydroxyl groups can function as acids or bases.24 This raises a question: Does TBP-1 bind (12) Schembri, M. A.; Kjærgaard, K.; Klemm, P. FEMS Microbiol. Lett. 1999, 170, 363-371. (13) Kjærgaard, K.; Sørensen, J. K.; Schembri, M. A.; Klemm, P. Appl. Environ. Microbiol. 2000, 66, 10-14. (14) Thai, C. K.; Dai, H.; Sastry, M. S.; Sarikaya, M.; Schwartz, D. T.; Baneyx, F. Biotechnol. Bioeng. 2004, 87, 129-137. (15) Gaskin, D. J. H.; Starck, K.; Vulfson, E. N. Biotechnol. Lett. 2000, 22, 1211-1216. (16) Naik, R. R.; Brott, L. L.; Clarson, S. J.; Stone, M. O. J. Nanosci. Nanotechnol. 2002, 2, 95-100. (17) Lee, S. W.; Mao, C.; Flynn, C. E.; Belcher, A. M. Science 2002, 296, 892-895. (18) Wang, S.; Humphreys, E. S.; Chung, S. Y.; Delduco, D. F.; Lustig, S. R.; Wang, H.; Parker, K. N.; Rizzo, N. W.; Subramoney, S.; Chiang, Y. M.; Jagota, A. Nat. Mater. 2003, 2, 196-200. (19) Kase, D.; Kulp, J. L. I.; Yudasaka, M.; Evans, J. S.; Iijima, S.; Shiba, K. Langmuir 2004, 20, 8939-8941. (20) Morita, Y.; Ohsugi, T.; Iwasa, Y.; Tamiya, E. J. Mol. Catal. B: Enzym. 2004, 28, 185-190. (21) Nygaard, S.; Wendelbo, R.; Brown, S. Adv. Mater. 2002, 14, 18531856. (22) Livnah, O.; Stura, E. A.; Johnson, D. L.; Middleton, S. A.; Mulcahy, L. S.; Wrighton, N. C.; Dower, W. J.; Jolliffe, L. K.; Wilson, I. A. Science 1996, 273, 464-471. (23) Jones, F. H. Surf. Sci. Rep. 2001, 42, 75-205. (24) Bolger, J. C. Annu. Tech. Conf.sSoc. Plast. Eng. 1972, 18, 402407.

Langmuir, Vol. 21, No. 7, 2005 3091

to any metal with an oxidized surface, or does it specifically recognize the geometric constellation of the charged hydroxyl groups on Ti? Also unknown is the extent to which aptamers that bind Ti will cross-react with other inorganic materials. For instance, Schembri et al. mentioned that aptamers targeting either PbO2 or CoO also cross-reacted with the other, though details were not provided.12 In this article we first describe cross-reactivity of TBP-1 and mutational analysis was provided, and then we explore biomineralization of TBP-1. Finding that crossreactivity of TBP-1 is suggested, peptide aptamers targeting inorganic materials could serve as novel molecular probes with which to explore the subnanometric structure of material surfaces. Materials and Methods Phage Binding Assay. Atomized Ti (purity, 99.78%; particle size, 99.9%; 99.9%; 98%; 10 µm), Cu (>99%; 75-150 µm), Fe (99.9%; 150 µm), Pt (>99.9%; 75 µm), Si (99%; 99.9%; 99.9%; 150 µm) were purchased from Kojundo Chemical Lab Co., Ltd. (Saitama). To evaluate the capacity of TBP-1 to bind to various metals, 1.0 × 1010 pfu of phage were incubated with 10 mg of metal particles that had been pre-equilibrated in 1 mL of Tris-buffer saline (TBS; pH 7.5) containing 1% bovine serum albumin and 0.5% polyoxyethylene sorbitan monolaurate (Tween 20, Sigma) for 1 h at room temperature with slow rotation. The particles were then washed 10 times with TBS containing 0.5% Tween 20. Bound phages were eluted from the particles by incubation in 1 mL of 0.2 M glycine-HCl (pH 2.2) for 10 min at room temperature. After neutralizing the supernatant (containing the eluted phage) using 150 µL of 1 M Tris-HCl (pH 9.1), the number of phages eluted was estimated by infecting the E. coli ER2738 strain. We used a phage that expressed a wild-type pIII protein (no display of foreign peptide) as a control, and material selectivity was expressed as the ratio of the numbers of peptide phages bound to the numbers of control phages. Determination of Langmuir Adsorption Parameters. TBP-1 was synthesized using the Fmoc method (Wako Chemicals, Tokyo), after which a 3 mM stock solution was made by dissolving TBP-1 in buffer containing 50 mM N-2-hydroxyethylpiperazineN′-2-ethanesulfonic acid (HEPES)-NaOH (pH 7.0) and 150 mM NaCl. The concentration was determined by UV absorption at 280 nm by calculating the extinction coefficient of the peptide using the ProtPram program available on the ExPAsy web site (http://tw.expasy.org/tools/protparam.html). The binding of the indicated concentrations of peptide to Ti or Si particles (10 mg) was carried out for 2 h at room temperature in buffer containing 50 mM HEPES-NaOH (pH 7.0) and 150 mM NaCl. After removing the particles by centrifugation, the concentration of TBP-1 in the supernatant (unbound peptide) was determined using fluoraldehyde (Pierce). The relative surface areas of the Ti and Si particles were then calculated to be 0.0197 and 0.3727 m2/g, respectively, by measuring their densities using an Accupyc 1330 (Micromeritics, Norcross) and the particle size distribution in 50% glycerol using a Mastersizer S (Malvern, Worcestershire). Biomineralization. Ag mineralization was assayed as described by Naik et al.9 Silification was assayed as described previously by us with slight modification. Briefly, the indicated amounts of TBP-1 were incubated for 5 min with 2 µL of 0.1 M prehydrolyzed tetramethoxysilane (TMOS) in TBS or phosphate buffered saline (PBS).25 For transmission and scanning electron microscopy, one drop of sample suspension was placed on a copper grid with a supporting film of collodion and air-dried. Transmission electron images and electron diffractions were obtained with a Hitachi H-7500 transmission electron microscope at 100 KV. For secondary electron images, samples on the grids were sputter coated with gold-palladium and observed using a Hitachi S-4500 (25) Sambrook, J.; Russell, D. W. Molecular Cloning A Laboratory Manual, 3rd ed.; Cold Spring Harbor Laboratory Press: New York, 2001.

3092

Langmuir, Vol. 21, No. 7, 2005

Sano et al.

Figure 1. Binding specificity of TBP-1. (a) Material selectivity of φTi-12-3-1, a phage displaying TBP-1. The ratios of the numbers of φTi-12-3-1 bound to the numbers of control phage (no displayed peptide) are indicated on the ordinate. Note that in experiments with Fe and Cu, the number of phages bound were less than 10 per 10 mg of material. (b) Effect of alanine substitution on material binding. scanning electron microscope at 10 kV. The spectrum by energy dispersive spectroscopy was obtained using a GENESIS2000 attached to the Hitachi S-4500.

Results and Discussions Target Specificity of TBP-1. We addressed the question of TBP-1 cross-reactivity by assessing the ability of a phage displaying TBP-1 (φTi-12-3-1) to bind 10 metals: Ti, Si, Ag, Au, Cr, Pt, Sn, Zn, Cu, and Fe (Figure 1a). The target specificity was evaluated by scoring the ratios of the numbers of φTi-12-3-1 adsorbed onto the metal surface to the numbers of adsorbed control phages that displayed no peptide. We found that φTi-12-3-1 bound not only to Ti but also to Ag and Si surfaces. No interaction was observed between φTi-12-3-1 and any of the other metals tested. We then tested the effect of substituting an alanine for R1, P4, or D5 of TBP-1 on the binding of φTi-12-3-1 to Si and Ag surfaces (Figure 1b). As previously observed with Ti, all of three alanine substitutions reduced binding to Ag and Si in comparison with wild-type peptide, although the effects on Si binding were rather mild. By contrast, alanine substitution of the lysine in the second position (K2) had no effect on the binding of TBP-1 to Ag or Si. Taken together, these findings suggest that similar electrostatic attraction governs the interaction between TBP-1 and Ti, Ag, and Si surfaces. Because free peptide often behaves differently than peptide attached to a phage body,26 we confirmed the binding of TBP-1 to Ti and Si using a synthetic 12-mer TBP-1 peptide. We followed the binding of the synthetic peptide to atomized Ti and Si using the Langmuir absorption isotherm (Figure 2). Through curve fitting we determined that the maximal amounts of TBP-1 adsorbed onto Ti and Si were 2.5 ( 0.3 × 10-6 and 2.1 ( 0.3 × 10-7 mol/m2, respectively, and that the dissociation constants for Ti and Si were 13.2 ( 4.0 and 11.1 ( 2.8 µM, respectively. The selective binding of TBP-1 to Ti, Ag, and Si suggests that the peptide recognizes a common structure shared by all three. In that regard, although Ti, Ag, and Si surfaces all are oxidized and display hydroxyl groups under the ambient conditions of our experiments, this is not sufficient to explain the selective TBP-1 binding, as Cr, Sn, Zn, Cu, (26) Yayon, A.; Aviezer, D.; Safran, M.; Gross, J. L.; Heldman, Y.; Cabilly, S.; Givol, D.; Katchalski-Katzir, E. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 10643-10647.

Figure 2. Binding of synthetic TBP-1 peptide to Ti and Si particles. The indicated concentrations of peptide were incubated with Ti or Si particles for 2 h at room temperature in buffer containing 50 mM HEPES-NaOH (pH 7.0) and 150 mM NaCl. Bound peptides were estimated by measuring the peptide in supernatants. Experiments were repeated 4-5 times; bars represent standard deviations.

and Fe surfaces are also well oxidized24 but do not bind TBP-1. Moreover, oxidized Fe surfaces contain both acid (-O-) and basic (-OH2+) hydroxyl groups, as does Ti.27 Notably, when taken as a mass Ti, Ag, and Si surfaces differ markedly in terms of their isoelectric points (6.7, 10.4 and 2.0, respectively)28 and crystal structures (a hexagon, a cube with a centered face, and a diamond, respectively).29 This suggests that the epitope, or set of molecules recognized by TBP-1, reflects a local electrochemical geometry on the oxide surface that is shared among Ti, Ag, and Si. TBP-1 Assisted Biomineralization of Silver. It was recently shown that peptide aptamers targeting inorganic materials often possess the ability to accelerate the mineralization of their target elements,9,16,30 a characteristic that may enable bioprocessing of nanocrystals.31 (27) Fowkes, F. M.; Dwight, D. W.; Cole, D. A. J. Non-Cryst. Solids 1990, 120, 47-60. (28) (a) Kosmulski, M. Adv. Colloid Interface Sci. 2002, 99, 255264. (b) Chau, L.-K.; Porter, M. D. J. Colloid Interface Sci. 1991, 145, 283-286. (c) Parks, G. A. Chem. Rev. 1965, 65, 177-198. (29) (a) Pawar, R. R.; Deshpande, V. T. Acta Crystallogr., A 1968, 24, 316-317. (b) Liu, L.; Bassett, W. A. J. Appl. Phys. 1973, 44, 14761479. (c) Hubbard, C. R.; Swanson, H. E.; Mauer, F. A. J. Appl. Crystallogr. 1975, 8, 45-48.

Subnanometric Structure Biological discrimination

Langmuir, Vol. 21, No. 7, 2005 3093

Figure 3. Mineralization of Ag by TBP-1. (a, b) Nanocrystals obtained by mineralization of Ag with 0.28 mg/mL (a) or 0.56 mg/mL (b) TBP-1 for 48 h at room temperature. Scale bar represents 500 nm in part a and 200 nm in part b. (c) Electron diffraction pattern of a biomineralized Ag particle.

Because TBP-1 binds to both Ag and Si, we anticipated that it would mineralize both Ag and silica under our ambient conditions. To test this possibility, we first incubated 0.14-0.56 mg/mL TBP-1 with 0.1 mM AgNO3 for 48 h at room temperature. As we expected, increasing amounts of orange-tan deposits were formed as we added increasing amounts of TBP-1 to the solution. Transmission electron microscopy revealed that these deposits were made up of nanoparticles ranging from 300 to 500 nm in size (Figure 3a,b); the electron diffraction patterns showed the particles to be crystalline. The above finding is consistent with that of Naik et al. who isolated peptide aptamers targeting Ag particles and showed that the peptides mineralized Ag nanocrystals from AgNO3.9 Interestingly, the C-terminal half of one of their Ag aptamers, AG4 (RYLPSD), is similar to the N-terminal half of TBP-1 (RKLPDA). In particular, R1, P4, and D5 of TBP, which are important for binding Ti, Ag, and Si (Figure 1), are well conserved in AG4, although the position of the aspartic acid (D) is shifted by one residue in AG4. The sequence similarity between the Ti aptamer TBP-1 and the Ag aptamer AG4 and the fact that TBP-1 also binds to and mineralizes Ag in an R1, P4, and D5dependent manner implies that TBP-1 and AG4 recognize a common subnanometric chemical structure shared by Ti and Ag surfaces and that their recognition of the surface is somehow related to their ability to mediate biomineralization. In that context, the earlier report that arginine interacts with the Ag ion is interesting, as the aspartic acid might have mediated the interaction with the positively charged Ag surface.32 As compared to the Ag nanocrystals obtained with AG4, the sizes of crystals formed by TBP-1 were rather larger (300-500 nm vs 60-150 nm).9 We do not yet know what factors caused the difference in crystal size, but appendix sequences and/or the relative position of the aspartic acid would seem to be likely contributors. TBP-1 Assisted Silification. Silification from TMOS is reportedly catalyzed under ambient conditions by the synthetic R5 peptide (SSKKSGSYSGSKGSKRRIL) designed by mimicking silaffin-1A, the natural biomineralization peptide of the diatom Cylindrotheca fusiformis.33 The peptide is enriched in positively charged arginine (30) Brown, S.; Sarikaya, M.; Johnson, E. J. Mol. Biol. 2000, 299, 725-735. (31) (a) Yu, L.; Banerjee, I. A.; Matsui, H. J. Am. Chem. Soc. 2003, 125, 14837-14840. (b) Mao, C.; Flynn, C. E.; Hayhurst, A.; Sweeney, R.; Qi, J.; Georgiou, G.; Iverson, B.; Belcher, A. M. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 6946-6951. (c) Mao, C.; Solis, D. J.; Reiss, B. D.; Kottmann, S. T.; Sweeney, R. Y.; Hayhurst, A.; Georgiou, G.; Iverson, B.; Belcher, A. M. Science 2004, 303, 213-217. (32) Gruen, L. C. Biochim. Biophys. Acta 1975, 386, 270-274. (33) Kro¨ger, N.; Deutzmann, R.; Sumper, M. Science 1999, 286, 11291132.

and lysine residues, and the C-terminal RRIL motif is known to be important for mineralization.34 The fact that polylysine or polyarginine also has been demonstrated to mediate silification35 provides further evidence that positively charged amino acids play a key role in mineralization. Because TBP-1 binds to oxidized Si surfaces and because such binding was associated with TBP-1-mediated mineralization of Ag, we anticipated that the peptide would also mediate silification at ambient temperature. And, if so, we wondered whether mutations that impaired binding would also affect silification. To address these questions, we first incubated various concentrations (which were used in the standard silification experiments)33,34 of TBP-1 with TMOS. As expected, in the presence of TBP-1, the TMOS solution rapidly formed white precipitates. Transmission and scanning electron microscopy revealed these precipitates to be made up of spherical particles (Figure 4a-h), which electron diffraction analysis showed to be silica (Figure 4i). We used both TBS and PBS for these experiments because the buffer used reportedly can affect the silification in vitro.36 We found that both buffers yielded similar 500-nm silica spheres when the peptide concentration was 6 mg/mL (Figure 4a-c,g,h). On the other hand, smaller spheres with rougher surfaces were obtained at a higher TBP-1 concentration (10 mg/mL) in TBS (Figure 4d-f), whereas the shape and texture of the bio-silica formed with 10-12 mg/mL TBP-1 in PBS were the same as at lower concentrations (data not shown). We do not yet know the mechanism responsible for the changes in the silica spheres, but these experiments confirmed that the buffer used can significantly affect biomineralization. We next investigated the effect of substituting alanine for R1, P4, or D5 of TBP-1 on silification. If the presence of positively charged residue(s) is necessary and sufficient for silification, alanine substitution of P4 and D5 would be expected to have little or no effect on silification. The results, however, showed that the P4 substitution dramatically impaired silification from TMOS under ambient conditions (Figure 5), indicating P4 to be somehow involved in the mineralization activity of the TBP-1 peptide. On the other hand, the D5 substitution, which abolished the binding of TBP-1 to Ti and Ag surfaces and had a milder but significant effect on the binding to Si, had no effect on silification. (34) Knecht, M. R.; Wright, D. W. Chem. Commun. 2003, 30383039. (35) Patwardhan, S. V.; Clarson, S. J. Silicon Chem. 2002, 1, 207214. (36) (a) Kro¨ger, N.; Lorenz, S.; Brunner, E.; Sumper, M. Science 2002, 298, 584-586. (b) Rodriguez, F.; Glawe, D. D.; Naik, R. R.; Hallinan, K. P.; Stone, M. O. Biomacromolecules 2004, 5, 261-265.

3094

Langmuir, Vol. 21, No. 7, 2005

Sano et al.

Figure 4. Mineralization of silica by TBP-1. (a, d, g) Transmission electron micrographs of silica precipitants obtained by incubating TMOS and TBP-1. (b, c, e, f, h) Scanning electron micrographs of silica precipitants. Mineralization was carried out in TBS (a-f) or PBS (g, h). Concentrations of TBP-1 used were 6 mg/mL (a-c, g, h) and 10 mg/mL (d-f). Scale bars represent 1 µm in parts a, b, and d- g and 500 nm in parts c and f. (i) Energy dispersive spectrum induced by electron beam irradiation for a sample obtained with 6 mg/mL TBP-1 in TBS. The Cu and C signals were derived from the grid, and the Au and Pd signals were from gold-palladium coating.

Figure 5. Effects of TBP-1 mutation on silica mineralization. TBP-1 (10 mg/mL) R1A, P3A, and D4A mutants or R5 peptide were incubated with 0.1 M TMOS in TBS for 5 min at ambient temperature. Reactions were carried out in a 96-well flat-bottom microplate. After the reaction, the plate was scanned by an image scanner equipped with transparency unit. When precipitates formed, the well became gray in the image.

The isolation of peptide aptamers using a peptide-phage display system must be carried out under aqueous conditions, where the surface of the metals is oxidized to varying degrees.24 It, therefore, seems plausible that the epitopes for the peptide aptamers would contain oxygen atoms. Furthermore, involvement of peptide aptamers in mineralization is likely related to their interaction with a precursor for biomineralization. In the case of TBPmediated Ag mineralization, the arginine of TBP-1 should interact with the hydroxylated Ag ion serving as the precursor for mineralization. That TBP-1 may recruit the hydroxylated Ag ion to the crystal growth core means that crystal growth may be controlled by TBP-1 binding at the surface of the Ag crystal. We are currently investigating this possibility. Conclusion Peptide Aptamers in Bionanotechnology. Peptide aptamers that recognize inorganic materials, including

Subnanometric Structure Biological discrimination

metals and semiconductors, are attracting increasing attention in the field of nanobiotechnology.10 This is because these peptides can be used not only to allocate various molecules on patterned substrates 8 but also to form nanocrystals of inorganic material under ambient conditions.31 As a prerequisite for establishing sophisticated novel systems that take advantage of such aptamers, the molecular mechanisms of their recognition needs to be understood. The data presented in this paper clearly show that a peptide aptamer can distinguish between groups of metals. TBP-1 recognizes Ti, Si, and Ag surfaces but not Au, Cr, Pt, Sn, Zn, Cu, or Fe. Because the overall net charges and crystal structures of Ti, Si, and Ag differ

Langmuir, Vol. 21, No. 7, 2005 3095

markedly, TBP-1 must recognize an as-yet-unknown nanometric chemical structure, suggesting that, in addition to their other uses, peptide aptamers targeting inorganic materials could serve as novel molecular probes with which to explore the surface of materials. Such examples would include the peptide probe that is conjugated on the cantilever of atomic force microscope. Acknowledgment. We thank Dr. Y. Takaoka for coordinating the surface area analyses of Ti and Si particles at the Iwate Industrial Research Institute. LA047428M