Multistep Growth Mechanism of Calcium Phosphate in the Earliest

ARTICLE pubs.acs.org/Langmuir

Multistep Growth Mechanism of Calcium Phosphate in the Earliest Stage of Morphology-Controlled Biomineralization Takayuki Nonoyama,† Takatoshi Kinoshita,*,† Masahiro Higuchi,‡ Kenji Nagata,‡ Masayoshi Tanaka,† Kimiyasu Sato,§ and Katsuya Kato*,§ †

Department of Frontier Materials and ‡Department of Materials Science and Engineering, Graduate School of Engineering, Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya, Aichi 466-8555, Japan § National Institute of Advanced Industrial Science and Technology, 2266-98 Anagahora, Shimo-Shidami Moriyama-ku, Nagoya, Aichi 463-8560, Japan

bS Supporting Information ABSTRACT: We studied the effect of surface-functional-group position on precipitate morphology in the earliest stage of calcium phosphate biomineralization and determined the detailed mechanism of precipitation starting from nucleation to precipitate growth. The biomineralization template was a β-sheet peptide scaffold prepared by adsorption with carboxyl groups arranged at strict 7 Å intervals. Phosphate was then introduced. Within 10 s, highly ordered embryos of calcium phosphate were formed and confined by a peptide nanofiber pattern. They repeatedly nucleated and dissolved, with the larger embryos absorbing the smaller ones in a clear demonstration of an Ostwald-ripening-like phenomenon, then aggregated in a line pattern, and finally formed highly ordered nanofibers of amorphous calcium phosphate. This multistep growth process constitutes the earliest stage of biomineralization.

’ INTRODUCTION Biomineralization is a biological synthesis process for growing hard tissues such as bones, teeth, and shells at ordinary temperatures and pressures. The study of biomineralization not only facilitates the clean and efficient industrial synthesis of specific functional materials1 5 but also helps to clarify fundamental theories of biology. To investigate the essence of biomineralization during nucleation and early crystal growth, we require a mineralization template that enables us to manipulate or fine tune the arrangement of functional groups on a membrane substrate on an angstrom scale. Previous in vitro studies used templates of a general polymer film or a lipid membrane, and it is difficult to fabricate a template with a functional-groups-arranged surface. Therefore, we chose a simple sequence of peptides for our template. Peptides have two main advantages over polymers and lipids. First, the functional-group position on a peptide can be completely determined by self-assembly6 8 because peptides are known to form stable secondary structures based on their amino acid sequences (their so-called primary structures) and external conditions (such as pH, temperature, and concentration).9 14 For example, β-sheet conformational peptides are secondary stretch structures of peptide main chains with functional groups derived from amino acid side chains positioned at regular 7 Å intervals.15 In addition, peptides are known to self-assemble and form a nanofiber by intermolecular hydrogen bonding.16 18 r 2011 American Chemical Society

Second, because a peptide consists of a chain of amino acids, various functional groups derived from its side chains can be readily introduced into the peptide sequence. We previously fabricated a self-assembled monolayer of highly ordered peptide nanofibers using a simple β-sheet peptide poly(ethylene glycol) (peptide PEG) diblock copolymer on a mica substrate.19 Functional groups of amino acid side chains are arranged at strict 7 Å intervals on the surface of the peptide membrane, and the nucleation site of the peptide part is separated by PEG on the angstrom scale. We also investigated the effect of the functional-group position on the crystal phase and morphology of a CaP mineralization system grown on the peptide template.20 We found that highly ordered amorphous CaP (ACP) nanofibers formed. Normally, ACP precipitates do not grow anisotropically and their morphology is difficult to control.21,22 However, precipitates grown on a peptide template clearly differ in morphology from those grown on a bare mica substrate. On a peptide template, embryos of CaP precipitate, which are less stable because of their small size, are confined in a linear alignment. Although this result is interesting in the confinement effect induced by the organic substance, the embryos do not have a uniform growth rate. Rather, their growth rate differs Received: February 23, 2011 Revised: April 25, 2011 Published: May 13, 2011 7077

dx.doi.org/10.1021/la2006953 | Langmuir 2011, 27, 7077–7083

Langmuir

ARTICLE

Chart 1. (LE)8-PEG70 Peptidea

(a) Molecular structure. (b) β-Sheet structure. (c) Hierarchical self-assembly of the β-sheet peptide on the monolayer.

a

Figure 1. AFM image of an (LE)8-PEG70 peptide monolayer adsorbed onto a mica substrate. The peptides have self-assembled into nanofibers by intermolecular hydrogen bonding, and the nanofibers are patterned.

during nucleation and precipitate formation, suggesting that mineralization is a multistep process. In nature, it can be seen that biominerals such as calcium phosphate and calcium carbonate are formed and confined to their morphology and crystal phase by organic substances through a biomineralization process. Hydroxyapatite microcrystals are known to form and align themselves in a particular confinement on collagen fibrils.23 27 Similarly, calcium carbonate crystals such as calcite and aragonite are known to form and align themselves in a particular confinement on chitin frames.28,29 In vitro studies also show that biominerals can be controlled and confined via their character by organic substances. For example, Sommerdijk’s group synthesized calcite crystal on polymer or lipid templates.30 34 As described above in in vivo and in vitro studies, biominerals control their morphology and crystal phase by organic substances and interfacial geometry in biomineralization. In this article, to investigate the detailed mechanism of the multistep mineralization process, we investigated the timedependent weight changes during mineralization and the associated changes in precipitate morphology by the quartz

Figure 2. SEM images of the CaP precipitate on an (LE)8-PEG70 template, taken at various times after the phosphate source was dropped. (a) At 5 min, amorphous CaP nanofibers have grown. (b) At 10 min, the nanofibers are highly ordered. (c) At 15 min, precipitates have grown and bridged the spaces between the nanofibers.

crystal microbalance (QCM) and atomic force microscopy (AFM), respectively. From these data, we were able to clarify the detailed mechanism of multistep growth in the earliest stage of mineralization. 7078

dx.doi.org/10.1021/la2006953 |Langmuir 2011, 27, 7077–7083

Langmuir

’ EXPERIMENTAL SECTION Our goal in designing a peptide sequence for our template was that the peptides should contain well-regulated fundamental functional groups. β-Sheet structures can be reportedly constructed with alternately coupled hydrophilic and hydrophobic amino acids.35 Thus, for the peptide amino acids, we chose hydrophilic leucine (L) and hydrophobic glutamic acid (E), and for the standard peptide sequence, we chose the (LE)8 sequence. For the fundamental functional group, we chose the carboxyl group, which can reportedly bind calcium ions efficiently.36 39 In our previous study, we introduced PEG (degree of polymerization = 70) at the peptide’s C-terminal position, which enabled us to prepare β-sheet peptide monolayers easily. The role of PEG is 3-fold: (1) it maintains the interval between the peptide nanofibers; (2) it prevents the β-sheet peptide from forming an amyloidlike aggregate;40 and (3) it preserves calcium ions and supplies them during mineralization.41 On the basis of these concepts, we designed the (LE)8-PEG70 diblock copolymer molecule for use as our peptide monolayer template (Chart 1). We then synthesized (LE)8-PEG70 by combinatorial solidphase peptide synthesis42 and confirmed the synthesis by matrix-assisted

ARTICLE

laser desorption/ionization time-of-flight mass spectrometry (MALDITOF-MS) (Figure S1). Preparation of a Peptide Monolayer Template. We prepared a peptide monolayer template by adsorption at the solid water interface.20 The peptide was dissolved in distilled water to a concentration of 1.0  10 5 M, and the pH was adjusted to 11.0. A fresh mica plate was soaked in the peptide aqueous solution at 25 °C for 18 h. Surplus adsorbed molecules were removed by rinsing with distilled water, and the substrate was dried at room temperature. Mineralization of Calcium Phosphate on the Template. We carried out the mineralization of CaP on the peptide monolayer template by what we call the alternate dropping method.20 For the calcium and phosphate sources, we used aqueous solutions of calcium acetate (Ca(CH3COO)2) and diammonium phosphate ((NH4)2HPO4). First, 20 μL of a 50 mM Ca(CH3COO)2 aqueous solution was dropped onto the template over the course of 1 min. Then, the template was rinsed and dried. The washing process removes the surplus calcium ions on the peptide surface and prevents homogeneous nucleation when the phosphate source is dropped. The drying process removes fluid on the template to prevent a concentration change in the phosphate source. An (NH4)2HPO4 aqueous solution (30 mM) was dropped onto the template in the same way. After 5, 10, and 15 min, the substrate was washed and dried to produce the final samples. Observation of Surface Morphology. We observed the surface morphologies of the peptide monolayer and the CaP embryo by AFM (Nano Scope IV, Veeco Instruments) with a silicon cantilever (length = 125 μm, tip radius = 12 nm) at a scanning line frequency of 1 Hz in an ordinary atmosphere. We observed the surface morphology of the CaP precipitate by field emission scanning electron microscopy (FE-SEM, S4300, Hitachi) at an accelerating voltage of 10 kV and a power current of 3 μA. All samples were coated with platinum nanoparticles by ion sputtering.

Measurement of the Time-Dependent Weight Change. Figure 3. AFM image of CaP embryos on an (LE)8-PEG70 template taken 10 s after the phosphate source was dropped. The embryos are formed and confined in a line pattern; the minimum embryo diameter is about the width of a peptide nanofiber.

We measured the weight change of the precipitate by QCM analysis (HP/Agilent 53131A universal counter; Seiko EG&G QA-CL4 electrode holder) with the following parameters: fundamental frequency of the quartz crystal = 9 MHz; electrode area = 0.152 cm2; shear modulus

Figure 4. QCM isotherm of CaP mineralization on an (LE)8-PEG70 template; at t = 0, the phosphate source was dropped. The isotherm is divided into four regions: (a) nucleation, (b) induction (nucleation and dissolution), (c) precipitate growth, and (d) the end of mineralization. 7079

dx.doi.org/10.1021/la2006953 |Langmuir 2011, 27, 7077–7083

Langmuir

ARTICLE

Figure 5. AFM images of CaP embryos on the (LE)8-PEG70 template, whose QCM isotherm is shown in Figure 4, taken 10, 20, 50, 100, and 150 s after the phosphate source was dropped. (See the time axis in Figure 4.) All embryos are confined in a line pattern; the larger embryos are increasing and the smaller embryos are decreasing gradually in size.

Figure 6. CaP embryo measurements taken at 10, 20, 50, 100, and 150 s; at t = 0, the phosphate source was dropped. (a) Size distribution. (b) Modulus of total volume change. The total volumes were calculated using values from Table 1; the total volume at 10 s was considered to be the base volume and was set to 1. of quartz = 2.95  106 dyn/cm2; and density of quartz = 2.65 g/cm3. We prepared the peptide monolayer by the same adsorption method on a silicon oxide-coated quartz device. The peptide monolayer-coated QCM chip was mounted onto the QCM apparatus, and 500 μL of distilled water was placed on the surface of the peptide-coated chip. There was an initial 30 s warm-up period for frequency stabilization. Subsequently, 500 μL of a 100 mM Ca(CH3COO)2 aqueous solution was added

gently, and the QCM isotherm was recorded. After calcium adsorption, the QCM chip was rinsed with water, dried, and subjected to the QCM measurement again. In the successive measurement, a 60 mM (NH4)2HPO4 aqueous solution was added instead of the calcium solution. The temperature of the cell was kept at 24 °C throughout the measurements. We then applied the Sauerbrey equation to estimate the precipitate weight. 43 7080

dx.doi.org/10.1021/la2006953 |Langmuir 2011, 27, 7077–7083

Langmuir

ARTICLE

’ RESULTS AND DISCUSSION Morphology of the Peptide Monolayer Template. Figure 1 shows an AFM image of the fabricated (LE)8-PEG70 monolayer template. Highly ordered peptide nanofibers are evident over an area of 1  1 μm2. The fibers are 6 nm wide, as estimated by crosssectional analysis, which is consistent with the calculated β-sheet peptide width (Chart 1). Figure S2 shows the surface-enhanced Raman scattering (SERS) spectrum of the (LE)8-PEG70 monolayer template. The strong peaks at 1628 and 1685 cm 1 are assigned to amide I of the antiparallel β-sheet and the β-plated sheet, respectively.44,45 Thus, the peptide nanofibers mainly form a β-sheet structure. By water-contact-angle measurement based on our previous work,20 we confirmed that the carboxyl groups of the glutamic acid side chains are positioned on the outermost layer of the template. Thus, the peptide monolayer has well-regulated functional groups on its surface and is an ideal template for our purpose, which is to investigate the relationship between the functional-group position on the template and the crystal phase of the precipitate. Calcium Phosphate Mineralization on the Template. We performed CaP mineralization on the (LE)8-PEG70 monolayer template by the alternate dropping method. Figure 2 shows the sequence of events after the phosphate source is dropped. At 5 min, patterned short nanofibers or dots have formed. At 10 min, patterned nanofibers have formed. At 15 min, precipitates have grown and bridged the spaces between the nanofibers.

Table 1. Embryo Sum Numbers and Average Sizes 10, 20, 50, 100, and 150 s after the Phosphate Source Was Droppeda time/s

a

10

20

50

100

150

sum number/count

1611

1003

1114

958

760

average of embryo diameter/nm

5.3

6.9

6.7

8.0

8.0

Profiles were estimated from the AFM images in Figure 5.

The crystal phase of the nanofibers is ACP as measured by electron diffraction based on our previous work.20 Figure 3 shows an AFM image of the template surface 10 s after the phosphate source was dropped. In this earliest stage of mineralization, patterned embryos have formed with a minimum diameter of 5 nm, which is almost equal to the width of a peptide nanofiber. This behavior differs from that observed on a bare mica substrate on which random precipitates form.20 Thus, the morphology of ACP grown on a peptide monolayer template clearly reflects the template’s design because the calcium ions adsorb onto the surface functional groups. Of course, on a bare mica substrate, calcium ions similarly adsorb onto surface hydroxyl groups. However, the peptide monolayer template differs from the mica substrate in the position of the bound calcium. On the peptide monolayer, calcium ions bind to patterned peptide nanofibers; therefore, the embryos formed are confined in the line pattern. Multistep Growth. CaP mineralization is affected by the position of the functional groups on the template surface as early as nucleation. However, the embryo significantly changes in size during the period from nucleation to precipitate formation. The embryo diameter is about the width of a peptide nanofiber but much less than the width of an ACP nanofiber. Thus, we speculate that the mode of precipitation continuously changes during this period. To clarify our speculation, we measured the time-dependent change in the precipitate weight by QCM and simultaneously observed the surface morphology by AFM. Figure 4 shows a multistep QCM isotherm. At t = 0, the phosphate source is dropped. At t = 900 s, the final surface concentrations are as follows: precipitate, 6300 ng/cm2; adsorbed calcium ions, 2600 ng/cm2 (Figure S3). The number of calcium ions is large relative to the number of carboxyl groups at the surface, suggesting that calcium ions also adsorb onto PEG, which then serves as a source of calcium ions during mineralization. If the weight change from dropping the phosphate source is caused only by adsorbed PO4, then the Ca/P ratio calculated

Scheme 1. Multistep Growth Mechanism of Cap Mineralization from Nucleation to Precipitationa

a

(a) Embryos are instantaneously forming everywhere. (b) Embryos are undergoing repeated multistep nucleation and dissolution, and larger embryos are absorbing smaller ones. (c) Precipitates are growing in the direction of the surface arrangement. (d) Mineralization has ended because the calcium source is depleted. 7081

dx.doi.org/10.1021/la2006953 |Langmuir 2011, 27, 7077–7083

Langmuir from the QCM profile is 0.97, which agrees well with the ratio measured by energy-dispersive X-ray (EDX) spectroscopy based on our previous work.20 The isotherm contains four distinct regions: (1) At 0 < t < 10 s, nucleation occurs, during which embryos form instantaneously on the surface. (2) At 10 < t < 230 s, equilibrium or induction occurs, during which embryos nucleate and dissolve repeatedly. (3) At 230 < t < 800 s, precipitation of the embryos into the ACP nanofibers occurs. (4) At t > 800 s, mineralization ceases because of the depletion of the calcium source. Figure 5 shows AFM images of the template surface 10, 20, 50, 100, and 150 s after the phosphate source was dropped. Figure 6a shows a plot of the embryo size distribution, and Table 1 lists the associated numbers and average sizes based on estimates from the AFM images. During this portion of induction (10 < t < 150 s), the number of embryos decreases but their average size increases; the larger embryos increase gradually in size, and the smaller embryos decrease gradually in size. Intriguingly, by 150 s, a large precipitate has grown across the peptide nanofibers. Thus, during induction, the larger embryos absorb the smaller ones. In our previous report,20 we speculated that an Ostwald ripening-like phenomenon might occur in the earliest stage of mineralization. Here, we see conclusively that the earliest stage of mineralization demonstrates an Ostwald ripening-like phenomenon at the surface.46 49 To confirm this initial clarification of the earliest precipitation mechanism of mineralization, we investigated the relationship between the QCM and AFM data. The QCM profile shows that the precipitate weight does not change during induction (10 < t < 230 s). Assuming that a phase transition does not occur during this period, the total volume of the embryos also does not change. We calculated the modulus of volume change for the embryos from their total number and average size, assuming that they are hemispherical in shape and by estimating their radius from the AFM data. Figure 6b shows a plot of the modulus of volume change as a function of time. The volume change remains almost constant in the range of 10 < t < 150 s, supporting our speculation that an Ostwald ripening-like phenomenon occurs. Scheme 1 shows a proposed multistep growth mechanism. (1) Calcium ions are introduced and bind to carboxyl groups on the template surface, phosphate ions are introduced and are attracted to the bound calcium ions, and calcium phosphate nucleation occurs on the peptide nanofibers. (2) Calcium phosphate embryos form and then nucleate and dissolve repeatedly. (3) Over time, large embryos (so-called critical nuclei) form stochastically, absorb the smaller embryos, and grow along the arrangement direction of the template. (4) The embryo microaggregates transform into stable patterned ACP nanofibers.

’ CONCLUSIONS Using simple β-sheet peptide monolayer templates having surface carboxyl groups, we clarified the effect of the surfacefunctional-group position on precipitate morphology and the detailed mechanism of precipitation in the earliest stage of biomineralization. The precipitate is affected by the position of the functional groups on the peptide template surface as early as nucleation. Embryos grow into microaggregates on the template surface by a mechanism of an Ostwald ripening-like effect. The microaggregate growth direction is confined along the arrangement direction of the template and transformed into patterned ACP nanofibers. Knowledge of this biomimetic growth mechanism

ARTICLE

should enhance our understanding of biomineralization and enable us to create more structure-controllable materials.

’ ASSOCIATED CONTENT

bS

Supporting Information. MALDI-TOF-MS spectrum of (LE)8-PEG70. SERS spectrum of an (LE)8-PEG70 monolayer. QCM isotherm of calcium ion adsorption on an (LE)8-PEG70 monolayer. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected] and katsuya-kato@ aist.go.jp.

’ ACKNOWLEDGMENT We acknowledge the support of the SER program of the Institute of Ceramics Research and Education (ICRE) administered by Nagoya Institiute of Technology (Prof. M. Nogami and Prof. T. Kasuga). We are grateful for support from the Innovation School program by the National Institute of Advanced Industrial Science and Technology (AIST). ’ REFERENCES (1) Lowenstam, H. A. Science 1981, 211, 1126–1131. (2) Mann, S.; Archibald, D. D.; Didymus, J. M.; Douglas, T.; Heywood, B. R.; Meldrum, F. C.; Reeves, N. J. Science 1993, 261, 1286–1292. (3) Mann, S. Nature 1993, 365, 499–505. (4) Shen, X.; Belcher, A. M.; Hansma, P. K.; Stucky, G. D.; Morse, D. E. J. Biol. Chem. 1997, 272, 32472–32481. (5) Stupp, S. I.; Braun, P. V. Science 1997, 277, 1242–1148. (6) Pauling, L.; Corey, R. B.; Branson, H. R. Proc. Natl. Acad. Sci. U.S.A. 1951, 37, 205–211. (7) Pauling, L.; Corey, R. B. Proc. Natl. Acad. Sci. U.S.A. 1951, 37, 251–256. (8) Pauling, L.; Corey, R. B. Proc. Natl. Acad. Sci. U.S.A. 1951, 37, 729–740. (9) Higuchi, M.; Inoue, T.; Miyoshi, H.; Kawaguchi, M. Langmuir 2005, 21, 11462–11467. (10) Parete, R. A.; Nadasdi, L.; Subbarao, N. K.; Szoka, J. F. C. Biochemistry 1990, 29, 8713–8719. (11) Brown, C. R.; Hong-Brown, L. Q.; Welch, W. J. J. Clin Invest. 1997, 99, 1432–1444. (12) Chen, F. Y.; Lee, M. T.; Huang, H. W. Biophys. J. 2002, 82, 908–914. (13) Patchornik, A.; Amit, B.; Woodward, R. B. J. Am. Chem. Soc. 1970, 92, 6333–6335. (14) Bals, R.; Wang, X.; Wu, Z.; Freeman, T.; Bafna, V.; Zasloff, M.; Wilson, J. M. J. Clin. Invest. 1998, 102, 874–880. (15) Lietzow, M. A.; Hubbell, W. L. Biochemistry 2004, 43, 3137–3151. (16) Zhang, S. Nat. Biotechnol. 2004, 21, 1171–1178. (17) Koga, T.; Matsuoka, M.; Higashi, N. J. Am. Chem. Soc. 2005, 127, 17596–17597. (18) Zhao, Y.; Yokoi, H.; Tanaka, M.; Kinoshita, T.; Tan, T. Biomacromolecules 2008, 9, 1511–1518. (19) Hattori, M.; Hayashi, S.; Yokoi, H.; Zhang, S.; Takana, M.; Kinoshita, T. Trans. Mater. Res. Soc. Jpn. 2006, 31, 245–248. (20) Nonoyama, T.; Tanaka, M.; Kinoshita, T.; Nagata, F.; Sato, K.; Kato, K. Chem. Commun. 2010, 46, 6983–6985. (21) Eanes, E. D.; Termine, J. D.; Nylen, M. U. Calcif. Tissue Int. 1973, 12, 143–158. 7082

dx.doi.org/10.1021/la2006953 |Langmuir 2011, 27, 7077–7083

Langmuir

ARTICLE

(22) Layrolle, P.; Ito, A.; Tateishi, T. J. Am. Ceram. Soc. 1998, 81, 1421–1428. (23) Traub, W.; Arad, T.; Weiner, S. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 9822–9826. (24) White, S. W.; Hulmes, D. J. S.; Miller, A.; Timmins, P. A. Nature 1977, 266, 421–425. (25) Landis, W. J. J. Ultrastruct. Mol. Struct. Res. 1986, 94, 217–238. (26) Marino, A. A.; Becker, R. O. Nature 1969, 213, 697–698. (27) Nudelman, F.; Pieterse, K.; George, A.; Bomans, P. H. H.; Friedrich, H.; Brylka, L. J.; Hilbers, P. A. J.; With, G.; Sommerdijk, N. A. J. M. Nat. Mater. 2010, 9, 1004–1009. (28) Zaremba, C. M.; Belcher, A. M.; Fritz, M.; Li, Y.; Mann, S.; Hansma, P. K.; Morse, D. E.; Speck, J. S. Chem. Mater. 1996, 8, 679–690. (29) Fritz, M.; Belcher, A. M.; Radmacher, M.; Walters, D. A.; Hansma, P. K.; Stucky, F. D.; Morse, D. E.; Mann, S. Nature 1994, 371, 49–51. (30) Buijnsters, P. J. J.; Donners, J. J. J. M.; Hill, S. J.; Heywood, B. R.; Nolte, R. J. M.; Binne Zwanenburg, B.; Sommerdijk, N. A. J. M. Langmuir 2001, 17, 3623–3628. (31) Donners, J. J. J. M.; Nolte, R. J. M.; Sommerdijk, N. A. J. M. J. Am. Chem. Soc. 2002, 124, 9700–9701. (32) Popescu, D. C.; Smulders, M. M. J.; Pichon, B. P.; Chebotareva, N.; Kwak, S. Y.; Otto, L. J.; van Asselen, O. L. J.; Sijbesma, R. P.; DiMasi, E.; Sommerdijk, N. A. J. M. J. Am. Chem. Soc. 2007, 129, 14058–14067. (33) Pichon, B. P.; Paul, H. H.; Bomans, P. H. H.; Frederik, P. M.; Sommerdijk, N. A. J. M. J. Am. Chem. Soc. 2008, 130, 4034–4040. (34) Dey, A.; Bomans, P. H. H.; M€uller, F. A.; Will, J.; Frederik, P. M.; With, G.; Sommerdijk, N. A. J. M. Nat. Mater. 2010, 9, 1010–1014. (35) Holmes, T. C.; Lacalle, S. D.; Su, X.; Liu, G.; Rich, A. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 6728–6733. (36) Tanahasi, M.; Matsuda, T. J. Biomed. Mater. Res. 1997, 34, 305–315. (37) Sato, K.; Kogure, T.; Iwai, H.; Tanaka, J. J. Am. Ceram. Soc. 2002, 85, 3054–3058. (38) Sato, K.; Kumagai, Y.; Watari, K.; Tanaka, J. Langmuir 2004, 20, 2979–2981. (39) Sato, K. J. Ceram. Soc. Jpn. 2007, 115, 124–130. (40) Timothy, S. B.; Tammie, L. B.; David, N. M. J.; Klaas, H.; Stephen, C. M.; David, G. L. J. Am. Chem. Soc. 1998, 120, 7655–7656. (41) Horikoshi, K.; Hata, K.; Kawabata, N.; Ikawa, S.; Konaka, S. J. Mol. Struct. 1990, 239, 33–42. (42) Fields, G. B.; Noble, R. L. Int. J. Pept. Protein Res. 1990, 35, 161–214. (43) Sauerbrey, G. Z. Phys. 1955, 155, 206–222. (44) Podstawka, E. Biopolymers 2008, 89, 506–521. (45) Podstawka, E. Biopolymers 2008, 89, 980–992. (46) Yang, H. G.; Zeng, H. C. J. Phys. Chem. B 2004, 108, 3492–3495. (47) Eberl, D. D.; Srodon, J.; Kralik, M.; Taylor, B. E.; Peterman, Z. E. Science 1990, 248, 474–477. (48) Yao, J. H.; Elder, K. R.; Guo, H.; Grant, M. Phys. Rev. B 1993, 47, 14110–14125. (49) Wei, M.; Ruys, A. J.; Milthorpe, B. K.; Sorrell, C. C. J. Biomed. Mater. Res. 1999, 45, 11–19.

7083

dx.doi.org/10.1021/la2006953 |Langmuir 2011, 27, 7077–7083