VOLUME 5, NUMBER 1, JANUARY 2005 © Copyright 2005 by the American Chemical Society
Self-Assembled Peptide Amphiphile Nanofibers Conjugated to MRI Contrast Agents Steve R. Bull,† Mustafa O. Guler,† Rafael E. Bras,‡ Thomas J. Meade,†,§,|,⊥ and Samuel I. Stupp*,†,‡,§ Department of Chemistry, Department of Materials Science and Engineering, Department of Biochemistry and Molecular and Cell Biology, Neurobiology and Physiology, Feinberg School of Medicine, Northwestern UniVersity, 2145 Sheridan Road, EVanston, Illinois Received September 15, 2004; Revised Manuscript Received November 17, 2004
ABSTRACT Self-assembled peptide amphiphile nanofibers have been investigated for their potential use as in vivo scaffolds for tissue engineering and drug delivery applications. We report here the synthesis of magnetic resonance (MR) active peptide amphiphile molecules that self-assemble into spherical and fiber-like nanostructures, enhancing T1 relaxation time. This new class of MR contrast agents can potentially be used to combine high-resolution three-dimensional MR fate mapping of tissue-engineered scaffolds with targeting of specific cellular receptors.
Magnetic resonance imaging (MRI) has evolved as one of themostimportantdiagnostictechniquesinclinicalradiology.1-3 The advent of high magnetic fields, improved gradient coils and pulse sequences has provided the means to obtain threedimensional images of whole animals at near cellular resolution.4 Typically, intrinsic contrast is augmented by the use of paramagnetic agents that increase the spin lattice relaxation rate (T1) of water protons.5 The majority of MR agents use the paramagnetic metal ion Gd(III) because it has seven unpaired electrons and a long electronic relaxation * Corresponding author. E-mail:
[email protected] † Department of Chemistry. ‡ Department of Materials Science and Engineering. § Feinberg School of Medicine. | Department of Biochemistry and Molecular and Cell Biology. ⊥ Neurobiology and Physiology. 10.1021/nl0484898 CCC: $30.25 Published on Web 12/09/2004
© 2005 American Chemical Society
time. When chelated to a suitable ligand the Gd(III) ion is detoxified and provides a powerful tool for a number of clinical and experimental applications that include tumor identification,6 perfusion analysis,7 cell tracking,8 and gene expression.9,10 A principle barrier to the development of new generations of MR agents is inherent lack of in vivo sensitivity. To obtain significant contrast over long periods of time, the observed relaxivity of MR agents must be increased. A common method of relaxivity enhancement is to increase the rotational correlation time (τr).1 This is accomplished by increasing the molecular weight of the agent by conjugation to proteins, polymers, or the preparation of micellar structures.1,4,11 Our approach to increasing the relaxivity of new MR agents utilizes self-assembling peptide amphiphiles (PAs),
Figure 1. Monomer structures of a chelate conjugated to 1, an RGD bioactive epitope and 2, a cross-linkable PA scaffold. Both structures self-assemble into nanostructures upon raising the pH above 7.0; molecule 1 assembles into fibers and molecule 2 assembles into spherical micelles.
developed over the past few years as scaffolds for regenerative medicine,12-14 coupled to a modified MR agent. We chose these PAs due their biocompatibility and bioactivity, and because of their well characterized self-assembly into cylindrical nanofibers and gels.12-14 Here, we describe the synthesis of two modified self-assembling PAs covalently linked to a derivative of 1,4,7,10-tetraazacyclododecane1,4,7,10 tetraacetic acid (DOTA) and their supramolecular structure. This class of peptide-amphiphile contrast agent (PACA) conjugates can form either self-assembled nanofibers or spherical micelles and can be cross-linked through disulfide bonds.12 The PACA systems described here are examples of self-assembling PAs in which the peptide sequence is modified to enable tracking of the molecule by MRI (Figure 1). Ideally one would like to extend the benefits of MRI to the PA gel scaffolds for three-dimensional noninvasive visualization. These PA monomers self-assemble into nanofibers with diameters on the order of six to eight nanometers and form self-supporting bioactive gels.12,14 The in vivo use of these gels would benefit greatly from the ability to detect and track their fate, migration and degradation by MRI. The branched PAs were prepared by Fmoc solid-phase peptide synthesis (SPPS) on a 0.1 mmol scale. The amino acids were purchased from NovaBiochem and the reagents from Sigma-Aldrich and used without further purification. The branching of the peptide headgroup was achieved using orthogonal protection group chemistry for the amines at the R and positions of the lysine residue.15 First, Fmoc-Lys(Mtt)-OH, (Mtt: 4-methyl trityl) was coupled to MBHA rink amide resin, followed by cleavage of the Mtt protecting group on the amine for palmitic acid coupling without affecting the Fmoc protection.16 This was followed by Fmoc removal on the R amine to grow the peptide segment of the PA. The branching point in PAs 1 and 2 was introduced at a lysine dendron using Fmoc-Lys(Mtt)-OH. To grow the first arm 2
of PAs, Fmoc on the R amine was removed before Mtt. For PACA 1 and 2, Boc-Lys(Boc)-OH was coupled at the end of the first branch. Both R and amine positions of the lysine were blocked with Boc protection as it is more stable under the cleavage conditions employed for the removal of Fmoc and Mtt. Later, Mtt was removed and the other branch was grown using Boc-Lys(Fmoc)-OH. The peptide sequence RGDS,17 important in cell adhesion, was built on the branch (using the amine) in order to combine bioactivity and MR functionality. The DOTA moiety was also coupled on SPPS to the N terminus of PAs using a tert-butyl ester protected DOTA molecule. The peptide portion of PACA 2 was designed with four consecutive cysteine residues for potential cross-linking of the self-assembled system. PACA 1 was cleaved from the resin in 95:2.5:2.5 TFA/TIS/H2O and PACA 2 was cleaved by 95:2:2:1 TFA/EDT/H2O/TIS solution (1,2-ethanedithiol (EDT) was used to prevent sulfur cross-linking). Excess TFA was evaporated under reduced pressure, and crude PA solutions were triturated using cold ether. 1 and 2 were dried under vacuum and characterized by MS-MALDI with a single peak found at 2308.66 (calcd. 2308.87) and 1920.25 (calcd. 1920.12), respectively. The final product was obtained by the addition of GdCl3 stirred at pH 6.5 for 2 days and purified by dialysis in deionized water for 3 days (1000MWC Spectrum Laboratories Inc.). Samples were lyophilized and reconstituted in deionized water, and relaxivity experiments performed on a Bruker mq60 NMR analyzer (Bruker Canada, Milton, Ont., Canada) in pH 7.41 buffer at 37° C.18 Structure 2 was designed to have reversibly cross-linkable thiol bonds that can contribute to the stability of the supramolecular complex (Figure 1). Of the two derivatives, only 1 forms nanofibers in solution at a pH greater than 7.0. This is the result of β-sheet forming amino acids present in the peptide sequence of this PA.19 To elucidate the size deviations from the typical PA nanofibers, more structural investigations are needed, which Nano Lett., Vol. 5, No. 1, 2005
Figure 2. AFM and TEM images of molecule 1 with Gd(III) (a) and without Gd(III) (b). Panel b is negatively stained with PTA to provide contrast, without staining no distinguishable structures are visible due to lack of electron density contrast. The unstained TEM of molecule 1 with Gd(III) can be found in the Supporting Information.
is beyond the scope of this manuscript. The relaxivity of 1 in its self-assembled state is 14.7 mM-1 s-1.20 The relaxivity of the uncrosslinked 2 was 22.8 mM-1 s-1 and became 20.8 mM-1 s-1 upon cross linking (normalized to ICP measurements) in a buffer solution at pH 7.41.20 It is possible the cross linking of 2 does not appreciably change the relaxivity due to the spherical nature of the structure. We do not know at this time why cross linking does not change the relaxivity appreciably. One possibility is the spherical nature of the aggregates. The observed relaxivities of these derivatives are significantly higher than those of typical monomer contrast agents.17 We believe this increased relaxivity is due to the self-assembly of the monomers into nanofibers in basic conditions, thereby decreasing τr. We believe 2 has a higher relaxivity than 1 due to its lower degree of molecular flexibility, since the DOTA moiety is closer to the core of the nanofiber. The oxidation modulation of cysteines in 2 was through the addition of excess aqueous iodine to covalently cross-link the monomers after the self-assembling pH change and the addition of excess dithiothreitol to dissociate the cross linking.13 Further work on systems with different epitopes is currently underway as well as varying the position of the DOTA derivative. Transmission electron microscopy (TEM) images21 and atomic force microscopy (AFM) reveal the supramolecular structure of aggregates from 1 and 2. The contrast observed is provided by the Gd(III) on the surface of the PACAs (unstained) and is further proof that Gd(III) is indeed chelated (Figure 3). Structure 1 forms uniform fibers with lengths beyond 100 nm and widths corresponding to 22 ( 2 nm, which can be explained by the tendency of nanofibers to bundle (Figure 2). In contrast, uncrosslinked 2 does not selfassemble into fibers but rather into spherical micelles with diameters corresponding to 20 ( 2 nm (Figure 3).20 Circular dichroism spectroscopy (CD)22 confirmed the difference in supramolecular structure seen in 1 and 2 (Figure 4). 1 shows an intense Cotton effect corresponding to a β-sheet structural motif, and 2, in contrast, reveals a less intense structural signal that is a combination of random coil, R-helix, and β-sheet. The less defined molecular conformation in the selfassembled state is consistent with the formation of spherical Nano Lett., Vol. 5, No. 1, 2005
Figure 3. AFM (a) drop cast on mica and TEM (b) unstained of molecule 2. These images depict a spherical nanostructure, and the contrast in panel b is due to the Gd(III) chelation.
Figure 4. CD spectra of molecules 1 and 2 with and without Gd(III) chelation taken at 0.025 mM. PACA 2 without Gd(III) (blue curve), PACA 2 with Gd(III) (pink curve), PACA 1 without Gd(III) (green curve), and PACA 1 with Gd(III) (orange curve). The spectra show an atypical β-sheet motif in molecule 1 and a weaker and structurally undefined signal for molecule 2.
micelles.19 On the other hand, the nanofiber formation observed in 1 by AFM and TEM supports the CD signature for an extended β-sheet structure.23 We have described the design and synthesis of two MRI active PA monomers that self-assemble into different supramolecular structures. The PA structure and architecture allow for an increase in τr as a result of self-assembly in water and will allow the integration of additional bioactive functions in these systems for variability in bioactivity of the amino acid sequence. We envision using these molecules to noninvasively track PA gel scaffolds in vivo, allowing for a complete time-series of high-resolution three-dimensional MR images to reconstruct their fate. Acknowledgment. This work is supported by the U.S. Department of Energy (DOE) under award No. DE-FG0200ER45810 and the National Institute of Health (NIH) under award No. 7RO1 AI47003-4. We thank Kieth MacRenaris for his assistance with ICP analysis. We acknowledge the Electron Probe Instrumentation Center and the Keck Biophysics Facility at Northwestern University. Supporting Information Available: Additional data describing the relaxivity of molecules 1 and 2. This material 3
is available free of charge via the Internet at http:// pubs.acs.org. Note Added after ASAP Publication. Figure 1 was corrected. This paper was published ASAP on 12/9/04. The corrected version was posted on 12/14/04. References (1) Caravan, P.; Ellison, J. J.; McMurry, T. J.; Lauffer, R. B. Chem. ReV. 1999, 99, 2293-2352. (2) Victor R. Fuchs, H. C. S. Health Aff. 2001, 20, 30-43. (3) Meade, T. J.; Taylor, A. K.; Bull, S. R. Curr. Opin. Neurobiol. 2003, 13, 597-602. (4) Toth, E.; Helm, L.; Merbach, A. E. The Chemistry of Contrast Agents in Medical Magnetic Resonance Imaging, 1st ed.; John Wiley and Sons: New York, 2001; p 45-120. (5) Brucher, E. Top. Curr. Chem. 2002, 221, 103-122. (6) Hoegemann-Savellano, D.; Bos, E.; Blondet, C.; Sato, F.; Abe, T.; Josephson, L.; Weissleder, R.; Gaudet, J.; Sgroi, D.; Peters, P. J.; Basilion, J. P. Neoplasia 2003, 5, 495-506. (7) Ladd, D. L.; Hollister, R.; Peng, X.; Wei, D.; Wu, G.; Delecki, D.; Snow, R. A.; Toner, J. L.; Kellar, K.; Eck, J.; Desai, V. C.; Raymond, G.; Kinter, L. B.; Desser, T. S.; Rubin, D. L. Bioconjugate Chem. 1999, 10, 361-370. (8) Hueber, M. M.; Staubli, A. B.; Kustedjo, K.; Gray, M. H. B.; Shih, J.; Fraser, S. E.; Jacobs, R. E.; Meade, T. J. Bioconjugate Chem. 1998, 9, 242-249. (9) Li, W.-h.; Parigi, G.; Fragai, M.; Luchinat, C.; Meade, T. J. Inorg. Chem. 2002, 41, 4018-4024. (10) Louie, A. Y.; Huber, M. M.; Ahrens, E. T.; Rothbacher, U.; Moats, R.; Jacobs, R. E.; Fraser, S. E.; Meade, T. J. Nat. Biotechnol. 2000, 18, 321-325.
4
(11) Nicolle Gaelle, M.; Toth, E.; Eisenwiener, K.-P.; Macke Helmut, R.; Merbach Andre, E. J. Biol. Inorg. Chem. 2002, 7, 757-769. (12) Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Science 2001, 294, 16841688. (13) Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 5133-5138. (14) Silva, G. A.; Czeisler, C.; Niece, K. L.; Beniash, E.; Harrington, D. A.; Kessler, J. A.; Stupp, S. I. Science 2004, 303, 1352-1355. (15) Aletras, A.; Barlos, K.; Gatos, D.; Koutsogianni, S.; Mamos, P. Int. J. Pept. Protein Res. 1995, 45, 488-496. (16) De Leon-Rodriguez, L. M.; Kovacs, Z.; Dieckmann, G. R.; Sherry, A. D. Chem.-Eur. J. 2004, 10, 1149-1155. (17) Xiong, J. P.; Stehle, T.; Diefenbach, B.; Zhang, R. G.; Dunker, R.; Scott, D. L.; Joachimiak, A.; Goodman, S. L.; Arnaout, M. A. Science 2001, 294, 339-345. (18) 10 mM 3-(N-morpholino)propane sulfonic acid (MOPS), 100 mM sodium chloride, 20 mM sodium bicarbonate, and 4 mM sodium phosphate monobasic. (19) Claussen, R. C.; Rabatic, B. M.; Stupp, S. I. J. Am. Chem. Soc. 2003, 125, 12680-12681. (20) See Supporting Information. (21) Hitachi H-8100 TEM instrument. Samples were prepared on holey carbon coated TEM grids by dipping the grid in the PA solution, rinsing and wicking dry. The same solutions were used for relaxivity experiments. (22) Jasco J-715 CD spectrometer at concentrations of 2.50 × 10-5 M with a 1 mm cell path length. (23) This atypical beta sheet minimal is ∼212 nm and the shift is thought to be from the interference from the branching in the PA monomer.
NL0484898
Nano Lett., Vol. 5, No. 1, 2005