NANO LETTERS
Presentation and Recognition of Biotin on Nanofibers Formed by Branched Peptide Amphiphiles
2005 Vol. 5, No. 2 249-252
Mustafa O. Guler,† Stephen Soukasene,‡ James F. Hulvat,‡ and Samuel I. Stupp*,†,‡,§ Department of Chemistry, Department of Materials Science and Engineering, Feinberg School of Medicine, Northwestern UniVersity, 2220 Campus DriVe, EVanston, Illinois 60208 Received October 25, 2004; Revised Manuscript Received December 10, 2004
ABSTRACT A branched peptide amphiphile system was designed for enhanced recognition of biotin on nanofibers formed by self-assembly of these molecules. Branching at a lysine residue was used to design peptide amphiphiles that are capable of presenting more than one epitope per molecule. We found that biotinylated branched structures form nanofibers that enhance recognition by the avidin protein receptor relative to similar nanostructures formed by linear peptide analogues. Biotin−avidin binding to the supramolecular nanofibers was characterized by measurement of fluorescence from nanofibers incubated with chropmophore-conjugated avidin.
Molecular recognition among ligands and receptors in biology requires appropriate presentation of epitopes.1 Cellular adhesion ligands in extracellular matrix play a critical role in cell adhesion and attachment, which affect cell proliferation, differentiation, and maintaining regular metabolic activities. Recently, there has been great interest in designing scaffolds that mimic cellular structures with artificial epitopes in order to trigger biological events important in regenerative medicine or targeted chemotherapy.2 Differences in cellular response have been reported with changes in distribution and structural presentation of the signals on these artificial cell scaffolds.3 For, example, varying the nanoscale separation between cell adhesion ligands has been found to improve the recognition of signals and subsequent proliferation of the cells.3 Among the various methodologies used to synthesize biomaterials, self-assembly is a particularly attractive tool to create scaffolds from solutions of molecules that can encapsulate cells and assemble in situ.4,5 In this work, we have synthesized a novel series of molecules to study presentation of biotin in selfassembling, fiber-like nanostructures consisting of branched and linear architectures of peptide amphiphiles (PAs). These PA molecules consist of a hydrophilic peptide sequences containing the epitope of interest as well as other amino acids that promote secondary structure formation through hydrogen bonding. The molecule is terminated at one end by an alkyl * Corresponding author. E-mail:
[email protected]. † Department of Chemistry. ‡ Department of Materials Science and Engineering. § Feinberg School of Medicine. 10.1021/nl048238z CCC: $30.25 Published on Web 12/29/2004
© 2005 American Chemical Society
segment to drive self-assembly in aqueous media through hydrophobic collapse.5 The author’s laboratory has previously reported on linear PAs that self-assemble into cylindrical nanofibers, triggered by changes in pH,5b electrostatic interactions,5c and addition of electrolytes.5e The series of molecules described here demonstrates a novel branched PA architecture, designed to increase the accessibility of epitopes to receptors on nanofiber surfaces by using a bulky, sterically hindered peptide structure. The molecules contain lysine dendron moieties and, similar to previously reported linear PAs, self-assemble to form aqueous gels formed by a network of nanofibers. As shown in Figure 1, the molecules contain the peptide sequence KXXXAAAK (X)G or L) followed by a saturated sixteencarbon alkyl segment, with the dendron branch introduced at a lysine residue. Alanine, glycine, and leucine residues were used to promote hydrogen-bonded β-sheet formation, which should favor aggregation into extended structures such as cylindrical nanofibers, rather than into spherical micelles as is more typically observed in amphiphilic self-assembly.6 The well-known biological epitope RGDS is present in cell binding domains of extracellular proteins such as fibronectin and vitronectin7 and was used as a proof-of-concept for the incorporation of bioactive peptides into these PA systems. The epitope is known to bind to integrin receptors, and this molecular recognition event plays a critical role in adhesion of cells to the extracellular matrix and in the complex cascade of signaling that follows.8 Molecule 1 was synthesized with a linear peptide structure to compare epitope availability with that in branched PAs. A lysine residue was introduced to
Chart 1. Structures of Linear and Branched Peptide Amphiphiles 1-3a
a
Biotin is shown in green, RGDS is shown in red.
Figure 1. (a) Negatively stained TEM micrograph and (b) AFM image of nanofibers formed by peptide amphiphile 2B (AFM image is 1 µm wide).
create asymmetrically branched molecules 2A and 2B, thereby altering structural presentation of the bioactive peptide sequence after self-assembly. Molecules 3A and 3B were synthesized to introduce symmetrical branches in a fashion similar to 2A and 2B and to investigate the presentation of multiple epitopes by a single PA molecule. Furthermore, in molecules 2 and 3, the effect of hydrophobic side chains on structural accessibility of the epitope was studied by exchanging glycine residues with leucine residues. To examine recognition and availability of epitopes, the RGDS sequence on each PA was terminated with a biotin group, and biotin accessibility was then probed using the binding of either a BODIPY-NeutrAvidin or an FITC250
avidin. It is well-known that avidin has a very high affinity for biotin, with four biotin binding sites per protein.10 This binding affinity has been previously used to study surface availability of monolayers by varying the number of biotin moieties presented.9 Interactions between the fluorophore and amino acid residues in the biotin binding site of the avidin cause quenching of fluorescence.10 Therefore, binding of the biotinylated PA with fluorescently labeled avidin should lead to a significant fluorescence recovery by weakening the quenching interactions.10a The branched PAs were prepared using solid-phase peptide synthesis (SPPS). Branching of the peptide segment was achieved by using orthogonal protecting group chemistry.11 Fmoc, Boc, and 4-methyl trityl (Mtt) protecting groups on the amines of the lysine residues were used to control the design of peptides, as each of these protecting groups can be manipulated independently. Fmoc-protected amines were used to couple amino acids onto the peptide, Boc protecting groups were used to block lysine branches, and Mtt was used for selective deprotection and growth of asymmetrical branches. The RGDS epitope was coupled to the amine of the lysine residue to enhance the epitope’s conformational freedom, due to the flexible four-carbon linker. Biotinylation of the PAs was achieved via SPPS by coupling a biotin to the end of the peptide sequence (see Supporting Information). Nano Lett., Vol. 5, No. 2, 2005
Figure 2. Increase in fluorescence emission at 514 nm upon binding of BODIPY-NeutrAvidin to the biotinylated branched and linear peptide amphiphiles, relative to unbound BODIPY-NeutrAvidin at the same concentration (5 × 10-8 M). Error bars are a 99% confidence interval on the mean for each set of samples (*** ) p < 0.001).
All PAs were soluble in water at pH 4 and formed selfsupporting gels at concentrations greater than 0.5 wt % when pH was increased above 6.5. Gel formation was found to be fully reversible with pH change. Transmission electron microscopy (TEM), atomic force microscopy (AFM), FTIR, and circular dichroism (CD) spectroscopy were used to characterize the self-assembly of branched PA molecules. TEM micrographs of self-assembled PAs 1-4 at pH 7.4 revealed the formation of uniform, high aspect ratio nanostructures with diameters of 7 ( 1 nm and ranging from hundreds of nanometers to several micrometers in length (see Figure 1 and Supporting Information). The FT-IR of lyophilized (freeze-dried) gels of all PAs indicates hydrogen bonding between the peptides, based on N-H stretching peaks at 3280-3285 cm-1. Amide I peaks at 1628-1632 cm-1 are consistent with a predominantly β-sheet-like character for the peptide secondary structure, with some R-helix and random coil conformations, indicated by peaks in the range of 1650-1675 cm-1.12 Additionally, a shift of νa (CH2) from ca. 2932 to ca. 2921 cm-1 indicates a high degree of ordering in the palmitoyl hydrophobic segment.13 Circular dichroism spectra from the self-assembled PAs reveals a broad peak (nπ* transition) between 200 and 230 nm,14 which can be interpreted as a signature for the predominant presence of β-sheets, as well as minor contributions from R-helical and random coil conformations (see Supporting Information). IR and CD results are consistent with a highly ordered assembly of hydrogen-bonded PAs with β-sheet character, resulting in densely packed molecules within the nanofibers. Dilute samples of biotinylated PAs were prepared at pH 7.4 to investigate the influence of binding with BODIPYNeutrAvidin and FITC-avidin. Interestingly, Figure 2 shows a significant increase in fluorescence emission upon binding of BODIPY-NeutrAvidin to biotinylated branched PAs, relative to linear PA 1. Similar results were observed with FITC-avidin (see Supporting Information). These results suggest that, despite the structural similarity observed by TEM (see Supporting Information), avidin has greater Nano Lett., Vol. 5, No. 2, 2005
accessibility to the biotin on the surface of nanofibers made up of branched molecules compared with those made up of linear molecules. In linear PA systems, we propose that dense hydrogen bonding may result in more compact packing of the epitopes on the surface of nanofibers, thus hindering binding of avidin to biotin, resulting in less recovery of fluorescence emission. However, in the sterically hindered branched systems, enhanced availability of biotin to the avidin receptor may indicate less effective packing of molecules on the fiber surface. In addition, incorporation of hydrophobic side chains on the PA structure altered the availability of the epitopes as well. Biotin availability on 2B and 3B was significantly higher than on 2A and 3A, respectively. Therefore, hydrophobic side chains in these molecules may also be affecting the nature of packing in the assembly and consequently epitope availability. As a control, a biotinylated RGDS peptide (see Supporting Information) was added to the BODIPY-NeutrAvidin to distinguish any effect of the peptide segment itself on biotinavidin binding. Biotin and biotin-RGDS titrations into solutions of BODIPY-NeutrAvidin resulted in similar fluorescence dependence with concentration, revealing saturation with 4 equivalents of biotin (see Supporting Information). In addition, nonbiotinylated versions of PA 1 and 2B were prepared and tested with FITC-avidin under the same conditions. No significant change in the fluorescence of FITC-avidin was observed, indicating that the increased fluorescence is not due to nonspecific binding of avidin to the PA nanofibers. The number of biotins bound to each avidin could not be calculated from these experiments. However, our results confirm the proposed effect of branching and hydrophobic side chains on epitope availability at the periphery of the nanofibers. Biological experiments are underway to establish if structural differences in RGDS epitope presentation on the nanofibers influences in similar fashion the more complex recognition process of this peptide sequence by cells cultured with the peptide amphiphile nanofibers. We have reported here on cylindrical nanostructures formed by branched peptide amphiphile molecules that presents high densities of binding sites on their surfaces. The branched covalent architecture of these molecules leads to greater accessibility of binding sites to a probing protein receptor. This observation will be useful in supramolecular design of bioactivity in synthetic nanoscale materials for biology and medicine. Acknowledgment. This work was funded by the U.S. Department of Energy under award no. DE-FG02-00ER54810. J.F.H. is supported by the Institute for BioNanotechnology in Medicine (IBNAM). We thank the Electron Probe Instrumentation Center for use of its Hitachi H-8100 transmission electron microscope, Nanoscale Integrated Fabrication, Testing and Instrumentation Center (NIFTI) for use of AFM, and the Keck Biophysics Facility for use of its Jasco J-715 CD spectrometer and PC1 spectrofluorometer at Northwestern University. We also thank Bryan Rabatic, Randal Claussen, Dan Harrington, and Ben Messmore of the author’s laboratory for their helpful discussions. 251
Supporting Information Available: Experimental protocols, circular dichroism data, fluorescence data and additional TEM images (PDF). This material is available free of charge via the Internet at http://pubs.acs.org. References (1) (a) Myszka, D. G.; Chaiken, I. M. Biochemistry 1994, 33, 23632372. (b) Omelyanenko, V.; Kopeckova, P.; Prakash, R. K.; Ebert, C. D.; Kopecek, J. Pharm. Res. 1999, 16, 1010-1019. (c) Wang, Y. C.; Kao, S. H.; Hsieh, H. J. Biomacromolecules 2003, 4, 224-231. (d) Lee, K. C.; Carlson, P. A.; Goldstein, A. S.; Yager, P.; Gelb, M. H. Langmuir 1999, 15, 5500-5508. (e) Gore, T.; Dori, Y.; Talmon, Y.; Tirrell, M.; Peled, H. B. Langmuir 2001, 17, 5352-5360. (f) Qi, K.; Ma, Q.; Remsen, E. E.; Clark, C. G., Jr.; Wooley, K. L. J. Am. Chem. Soc. 2004, 126, 6599-6607. (2) (a) Langer, R. Acc. Chem. Res. 2000, 33, 94-101. (b) Barbera, D. A.; Zylstra, E.; Landbury, P. T.; Langer, R. J. Am. Chem. Soc. 1993, 115, 11010-11011. (c) Nowak, A. P.; Breedveld, V.; Pakstis, L.; Ozbas, B.; Pine, D. J.; Pochan, D.; Deming, T. J. Nature 2002, 417, 424-428. (d) Welsh, E. R.; Tirrell, D. A. Biomacromolecules 2000, 1, 23-30. (3) (a) Houseman, B. T.; Mrksich, M. Biomaterials 2001, 22, 943-955. (b) Kato, M.; Mrksich, M. Biochemistry 2004, 43, 2699-2707. (c) Dillow, A. K.; Ochsenhirt, S. E.; McCarthy, J. B.; Fields, G. B.; Tirrell, M. Biomaterials 2001, 22, 1493-1505. (d) Lee. K. Y.; Alsberg, E.; Hsiong, S.; Comisar, W.; Linderman, J.; Ziff, R.; Mooney, D. Nano Lett. 2004, 4, 1501-1506. (4) (a) Klok, H. A.; Hwang, J. J.; Iyers, S.; Stupp, S. I. Macromolecules 2002, 35, 746-759.
252
(5) (a) Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Science 2001, 294, 1684-1688. (b) Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 5133-5138. (c) Niece, K. L.; Hartgerink, J. D.; Donners, J. J. J. M.; Stupp, S. I. J. Am. Chem. Soc. 2003, 125, 7146-7147. (d) Silva, G. A.; Czeisler, C.; Niece, K. L.; Beniash, E.; Harrrington, D. A.; Kessler, J. A.; Stupp, S. I. Science 2004, 303, 1352-1355. (e) Beniash, E.; Hartgerink, J. D.; Storrie, H.; Stupp, S. I., manuscript in preparation. (6) (a) Guler, M. O.; Claussen, R. C.; Stupp, S. I., manuscript in preparation. (b) Discher, D. E.; Eisenberg, A. Science 2002, 297, 967-973. (c) Zhang, L. F.; Yu, K.; Eisenberg, A. Science 1996, 272, 1777-1779. (d) Feiters, M. C.; Nolte, R. J. M. AdVances in Supramolecular Chemistry; Gokel, G. W., Ed.; Jai Press Inc.: Stamford CT, 2000; Vol. 6, pp 41-156. (7) Pierschbacher, M. D.; Ruoslahti, E. Nature 1984, 309, 30-33. (8) (a) Pytela, R.; Pierschbacher, M. D.; Ruoslahti, E. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 5766-5770. (b) Pytela, R.; Pierschbacher, M. D.; Ruoslahti, E. Cell 1985, 40, 191-198. (9) Zhao, S.; Reichert, W. M. Langmuir 1992, 8, 2785-2791. (10) (a)Emans, N.; Biwersi, J.; Verkman, A. S. Biophys J. 1995, 69, 716728. (b) Nicol, F.; Nir, S.; Szoka, F. C. Biophys J. 1999, 76, 21212141. (c) Green, N. M. AdV. Protein Chem. 1975, 29, 85-133. (11) (a) Bourel, L.; Carion, O.; Masse, H. G.; Melnyk, O. J. Peptide Sci. 2000, 45, 488-496. (b) Aletras, A.; Barlos, K.; Gatos, D.; Koutsogianni, S.; Mamos, P. J. Peptide Protein Res. 1995, 6, 264-270. (12) (a) Miyazawa, T.; Blout, E. R. J. Am. Chem. Soc. 1961, 83, 712718. (b) Haris, P.; Chapman, D. Bipolymers 1995, 37, 251-263. (13) Helmy, R.; Fadeev, A. Y. Langmuir 2002, 18, 8924-8928. (14) Greenfield, N.; Fasman, G. D. Biochemistry 1969, 8, 4108-4116.
NL048238Z
Nano Lett., Vol. 5, No. 2, 2005
