| STUDY |
Nano Letters, ASAP Article 10.1021/nl0347536 S1530-6984(03)04753-2 Web Release Date: December 16, 2003 Copyright © 2003 American Chemical Society Genetically Driven Assembly of Nanorings Based on the M13 Virus Ki Tae Nam, Beau R. Peelle,# Seung-Wuk Lee, and Angela M. Belcher*# Department of Materials Science and Engineering, Department of Biology, Division of Biological Engineering, Institute for Soldier Nanotechnologies, Massachusetts Institute of Technology, 16-244, Cambridge, Massachusetts 02139-4307, and Department of Chemistry, University of Texas at Austin, Austin, Texas 78712 Received September 8, 2003 Revised September 30, 2003 Abstract: One-dimensional ring structures from M13 viruses were constructed by two genetic modifications encoding binding peptides and synthesis of a heterobifunctional linker molecule. The bifunctional viruses displayed an anti-streptavidin peptide and hexahistidine peptide at opposite ends of the virus as pIII and pIX fusions. Stoichiometric addition of the streptavidin-NiNTA linker molecule led to the reversible formation of virus-based nanorings with circumferences corresponding to lengths of the packagable DNAs. These virus-based ring structures will be further engineered to nucleate inorganic materials and form metallic, magnetic, or semiconductor nanorings using trifunctionalized viruses. -------------------------------------------------------------------------------- Biological self-assembly and biomolecular interactions continue to inspire novel approaches for the development of nanostructured materials.1-5 Furthermore, the remarkable ability of biomolecules to recognize and nucleate inorganic materials such as semiconductors, magnetic materials, and metals has broadened the possible applications in nanoelectronics and nanobiotechnology.6-9 Our previous work, along with that of others,10-14 has shown that biomolecules, in our case genetically engineered M13 bacteriophage (virus), can be used as a molecular building block to nucleate and arrange quantum dots,15 template semiconductor nanowires,16,17 and build multidimensional liquid crystals and films.15,18-20 Other self-assembling peptide and protein systems have been used to make wires,21 fibers,22 and other structures incorporating inorganic materials.23 However, the potential of these systems for assembling devices is limited in part by difficulties in programming distinct structural size and geometric control into the self-assembling components. Thus we set out to explore the possibility of genetically encoding size and shape information into self-assembling multifunctional viruses. Here we demonstrate the one-dimensional (1D) formation of a ring structure from a M13 virus genetically engineered to display fused functional binding peptides at each end. The wild-type filamentous M13 virus is approximately 6.5 nm in diameter and 880 nm in length.24 The length of the cylinder reflects the length of the packaged single stranded DNA genome size. At one end of the M13 virus, there are approximately five molecules each of protein VII (pVII) and protein IX (pIX). The other end has about five molecules each of protein III (pIII) and protein VI (pVI), totaling 10-16 nm in length. The wild-type M13 virus coat is composed of roughly 2800 copies of the major coat protein VIII (pVIII) stacked in units of 5 in a helical array. Ring structures of M13 viruses were constructed by two genetic modifications of the M13 virus and synthesis of a heterobifunctional linker molecule, which was designed to specifically bind each modified virus end (Figure 1). At one end of the virus, the anti-streptavidin peptide (SWDPYSHLLQHPQ-), identified through a phage library screen for streptavidin binding,19 was displayed at the N-terminus of pIII. The anti-streptavidin sequence was encoded in the M13 virus genome, thus, all copies of the pIII minor coat protein on the engineered virus display the anti-streptavadin peptide. At the other end of the virus, a hexahistidine peptide (AHHHHHH-), which binds strongly to Ni(II)-nitrilotriacetic acid complex (Ni-NTA), was fused to the N-terminus of pIX.25 These bifunctional viruses are easily amplified to large quantities using standard bacterial amplification procedures. -------------------------------------------------------------------------------- Figure 1 (a) Schematic representation of engineered M13 virus. His6 peptide displayed as pIX fusion shown in red, anti-streptavidin peptide displayed as pIII fusion shown in blue. (b) Tetrameric streptavidin shown in blue conjugated with four nickel-nitrilotriacetic acid (Ni-NTA) groups. -------------------------------------------------------------------------------- The engineered viruses were produced using a phagemid system,24 which used a plasmid separate from the M13 genome to express the engineered His6-pIX fusion. Single-stranded DNA produced from this plasmid inside the Escherichia coli host packages into viruses, called phagemids, as does the M13 genome. Since the length of the virus is proportional to the size of its packaged DNA, the phagemid-based viruses are typically observed to be 300-600 nm in length, shorter than wild-type M13 viruses. In this phagemid expression system, both wild-type and His6-pIX were produced, and the resulting virus may display between zero and five copies24 of the His-tagged pIX per virus. The dual-end modified viruses were amplified by infection of ER2738 E. coli (New England Biolabs) harboring the His6-pIX phagemid with the anti-streptavidin-pIII containing M13 virus, purified from centrifuge clarified supernatant by PEG fractionation and resuspended in Tris buffered saline, pH 7.5. A linker molecule consisting of streptavidin conjugated with Ni-NTA was also synthesized (Figure 1b). A NTA ligand (Qiagen) bearing a primary amine was reacted with carboxylate groups on streptavidin (New England Biolabs) in the presence of EDC catalyst, then charged with Ni, and purified.26 Addition of a linker molecule triggered the ring formation reaction, which was therefore programmed and inducible. Since the linker must bind both ends of the virus, heterobifunctionality was important in order to prevent saturation of binding sites on the linker by binding peptides displayed multivalently at a single end of the virus. Before observing the nanostructures, synthesis of the linker was verified using matrix assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry on a Voyager DE-Star instrument in linear mode. Streptavidin is a tetrameric protein made up of four identical subunits each with a single biotin binding site.27 The approximate total molecular weight of the streptavidin used was 52.8 kDa, corresponding to a mature truncated form of streptavidin. The intensity distribution indicated that the streptavidin was fully dissociated into monomer subunits, and momomer, dimer, trimer, and tetramer were present, as seen by others.28 In the case of the NiNTA-streptavidin linker molecule, the peak maximum corresponding to tetramer was shifted by approximately 1.2 kDa, and monomer by about 0.3 kDa (data not shown). Considering the additional 301.7 Da mass of a single conjugated Ni-NTA, each monomer of streptavidin appeared to have on average one Ni-NTA molecule attached. However, if a stochastic distribution of conjugated products was present, they were not resolved in the mass spectra. The viruses were observed to form a ring structure when the interdistance of each M13 virus in solution was greater than a few times the virus length and the relative concentration of the M13 virus to NiNTA-streptavidin was 1:1 (1011 phage/mL: 1011 molecules/mL). The bifunctional virus (in 1 mM Tris HCl, 1.5 mM NaCl, pH 7.5) and linker molecule (in H2O) were stoichiometrically combined, mixed by vortexing, and incubated at 23 C for 1 day. Figure 2 shows atomic force microscopy (AFM) images of virus ring structures on mica substrate observed with a NanoscopeIV (Digital Instruments) in tapping mode. The virus-based rings were found scattered over a large area of the sample. The radii of rings predominantly ranged between 60 and 90 nm. The range of circumferences observed corresponded to the size of packagable DNA from the phagemid plasmid and virus genome. Additional engineering of the phagemid system to package a single DNA or of the viral genome to incorporate the modified His6-pIX gene should lead to a greater monodispersity of ring structures. -------------------------------------------------------------------------------- Figure 2 M13 virus-based ring structures observed by AFM on mica surface. -------------------------------------------------------------------------------- The ring formation verifies the successful engineering of the pIII and pIX to generate a bifunctional M13 virus. This to our knowledge is the first engineered bifunctional phage. Other experiments were performed to verify the requirement of the engineered components to form the rings. A monofunctionalized virus displaying only the anti-streptavidin peptide on pIII was stoichiometrically mixed 1:1 with linker (1011 phage/mL: 1011 molecules/mL). Also, the bifunctional virus was mixed 1:1 with streptavidin lacking the NiNTA modification. Neither of these control samples formed rings, and only linear viruses were observed by AFM (data not shown). Thus, modified pIX on the virus and the NiNTA functionalization of the linker were essential to ring formation. Earlier experiments identifying anti-streptavidin peptides displayed on pIII viruses and attaching streptavidin conjugates to such viruses have demonstrated the importance of a displayed peptide for binding streptavidin.19,29 Ring formation was also observed by transmission electron microscopy (TEM) on a JEL 200CX instrument operating at 200 kV, as shown Figure 3a. The darker region of about 5 nm (indicated with the arrow in upper part of ring) was believed to be the linker. The size of the observed ring by TEM was similar to that of the ring structure in AFM image. Antibody labeling was used to enhance the contrast of the virus rings for TEM imaging (Figure 3b). A polyclonal pVIII primary antibody was used to bind the major coat of the virus and anti-rabbit IgG secondary antibody conjugated to 10 nm gold nanoparticle was used as the label.30 The large effective diameter of the antibody conjugated gold particles potentially distorts the size of the observed virus rings. However, ring-shaped assembly of gold nanoparticles were observed in the sample. -------------------------------------------------------------------------------- Figure 3 (a) TEM image of an individual virus-based ring structure stained with 2% uranyl acetate. The arrow indicates the darker region believed to be the linker. (b) TEM image of a virus-based ring structure where virus is labeled with antibody conjugated gold nanoparticles. -------------------------------------------------------------------------------- To form a ring, the two ends of the virus must meet and be held together via the linker molecule. An increase of strain energy and some decrease in entropy would accompany this process. Thermal fluctuation, mechanical forces induced by vortexing or hydrodynamic fluid motion including Brownian motion, may contribute to overcoming the required activation energy to bend the virus into a ring. Once formed, the ring structure is believed to be stable due to exceptionally strong binding between the displayed peptides and the linker. In fact, the dissociation constant (Kd) of His6 to Ni-NTA has been measured at pH 8 to be 10-13 M.34 This is stronger than most antibody bindings, of which the Kd values range from 10-7 M to 10-10 M. The binding of streptavidin and anti-streptavidin peptide, with its HPQ biotin-like motif, is expected to be slightly weaker than that of streptavidin and biotin, which has a Kd of 10-15 M.35 Furthermore, avidity effects should strengthen interaction since multiple copies of the binding peptides were displayed at each end of the virus and the linker molecule possesses on average four binding sites for each peptide. -------------------------------------------------------------------------------- Figure 4 AFM image of engineered phage mixed with linker molecule at different stoichiometric ratios. At 10:1 virus/linker, (a) radially aggregated viruses and (b) linearly linked viruses were observed. -------------------------------------------------------------------------------- This virus-based ring structure is expected to function as a biological template for the nucleation and formation of metalized, magnetic, or semiconductor nanorings using a trifunctionalized virus. Magnetic ring structures have unique magnetic properties32 and are of interest for use as magnetic data storage devices.33 Modification of the pVIII with peptides that nucleate ZnS, CdS, or FePt nanowires has already been demonstrated.16,17,31 Incorporating pVIII display into the bifunctional viruses by changing conditions of virus amplification will enable nucleation and ring formation by the viruses. When the concentration of the virus was high enough for multiple viruses to collide with a linker, linear or radial linkages of viruses were observed. Virus and linker were mixed at a ratio of 10:1 (1012 units/mL: 1011 units/mL) and imaged by AFM, as shown in Figure 4. The resultant structures (Figure 4a) can be explained through the presence of multiple binding sites on the linker. NiNTA-streptavidin was expected to have four binding sites for the anti-streptavidin peptide, inferred from the known tetrametric structure of the biotin-streptavidin complex.36 Additionally, as determined by mass spectrometry (data not shown) and shown schematically in Figure 1b, each native linker had on average a total of four Ni-NTA ligands and thus multiple binding sites for the His6 peptides. Binding of anti-streptavidin or His6 peptide of one virus and anti-streptavidin or His6 of another virus to the same linker molecule caused M13 virus to arrange linearly as shown in Figure 4b. When the virus concentration (1012 phage/mL) is greatly exceeded by linker (1013 molecules/mL), all phage-displayed binding peptides were seemingly passivated with linker such that viruses remained independent (data not shown). As a result, by simply controlling the concentration of the virus and the linker in this system, various self-assembling structures were achieved. -------------------------------------------------------------------------------- Figure 5 Engineered phage mixed 1:1 with linker molecule imaged by AFM. (a) Virus based ring structures were observed. (b) After addition of 50 mM imidazole to the same suspension only linear viruses were observed. (c) Linear viruses were also observed with addition of 5 mM biotin to the suspension. -------------------------------------------------------------------------------- In addition, ring formation was designed to be reversible through regulation of His6-NiNTA binding via control over free imidazole concentration.37 The sample was prepared 1:1 virus to linker as in Figure 2, then imidazole was added to a final concentration of 50 mM (5 uL virus suspension + 5 uL 100 mM imidazole in H2O) and samples imaged by AFM (Figure 5). Indeed, when imidazole was added to the ring-forming sample (Figure 5a), no rings were observed (Figure 5b). Furthermore, addition of biotin to a final concentration of 5 mM (5 uL virus suspension + 5 uL 10 mM biotin in H2O) to the sample in Figure 5a also produced only linear viruses (Figure 5c). Combined, these results demonstrate that ring formation is dependent upon highly specific binding of the virus-displayed peptides to polyhistidine and biotin binding sites on the linker. This and other reversible biomolecular interactions may enable engineered virus to function as a biomolecule based switch, which is similar in concept to a supramolecular switch.38 Further, similar systems may be constructed which genetically incorporate both binding domain and binding site functions directly into the virus. Heterofunctional viruses may allow us to do several things, such as bind or grow different materials in spatially distinct areas, assemble multiple different viruses in specific orders, and regulate specific individual interconnects. We expect that more complex assemblies and arrangements of biomolecules may be achieved by developing additional "directionally interconnecting" components that capitalize upon the diversity and specificity of biomolecular recognition. Acknowledgment The authors thank Chung-Yi Chiang for his helpful discussion about conjugating the linker molecule, Dr. Christine Flynn for editing this paper for clarity, and Manlin Luo in the MIT Biopolymers Laboratory for mass spectrometry support. This research was supported in part by the U.S. Army through the Institute for Soldier Nanotechnologies, under Contract DAAD-19-02-D-0002 with the U.S. Army Research Office, the National Science Foundation Nanotechnologies Interdisciplinary Research Team, and the Air Force Office of Scientific Research. The content does not necessarily reflect the position of the government, and no official endorsement should be inferred. * Corresponding author. E-mail: belcher@mit.edu These authors contributed equally to this work. Department of Materials Science and Engineering, MIT. Institute for Soldier Nanotechnologies, MIT. Department of Biology, MIT. # Division of Biological Engineering, MIT. Department of Chemistry, University of Texas at Austin, Austin, TX 78712 1. Mann, S. Biomimetic Materials Chemistry; VCH: New York, 1996. 2. Ball, P. Nature 2001, 413, 667-668. 3. Seeman, N. C. Nature 2003, 421, 427-431.[Medline] 4. Flynn, C. E.; Lee, S.-W.; Peelle, B. R.; Belcher, A. M. Acta Mater. 2003, in press. 5. Niemeyer, C. M.; Adler, M.; Gao, S.; Chi, L. Angew. Chem., Int. Ed. Engl. 2000, 39, 3055-3059. 6. Brown, S. Nature Biotechnol. 1997, 15, 269-272. 7. Whaley, S. R.; English, D. S.; Hu, E. L.; Barbara, P. F.; Belcher, A. M. Nature 2000, 405, 665-668.[Medline] 8. Mann, S.; Shenton, W.; Li, M.; Connolly, S.; Fitzmaurice, D. Adv. Mater. 2000, 12, 147-150. 9. Nygaard, S.; Wendelbo, R.; Brown, S. Adv. Mater. 2002, 14, 1853-1856. 10. Douglas, T.; Young, M. Nature 1998, 393, 152-155. 11. Shenton, W.; Douglas, T.; Young, M.; Stubbs, G.; Mann, S. Adv. Mater. 1999, 11, 253-256. 12. Knez, M.; Bittner, A. M.; Boes, F.; Wege, C.; Jeske, H.; Mai, E.; Kern, K. Nano Lett. 2003, 3, 1079-1082. 13. Li, Z.; Chung, S. W.; Nam, J. M.; Ginger, D. S.; Mirkin, C. A. Angew. Chem., Int. Ed. Engl. 2003, 42, 2306-2309. 14. Dujardin, E.; Peet, C.; Stubbs, G.; Culver, J. N.; Mann, S. Nano Lett. 2003, 3, 413-417. 15. Lee, S.-W.; Mao, C.; Flynn, C. E.; Belcher, A. M. Science 2002, 296, 892-895.[Medline] 16. Mao, C.; Flynn, C. E.; Hayhurst, A.; Sweeney, R.; Qi, J.; Williams, J.; Georgiou, G.; Iverson, B.; Belcher, A. M. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 6946-6941. 17. Flynn, C. E.; Mao, C.; Hayhurst, A.; Williams, J. L.; Georgiou, G.; Iverson, B.; Belcher, A. M. Synthesis and organization of nanoscale II-VI semiconductor materials using evolved peptide specificity and viral capsid assembly. J. Mater. Chem. 2003, 13, 2414-2421. 18. Lee, S.-W.; Wood, B. M.; Belcher, A. M. Langmuir 2003, 19, 1592-1598. 19. Lee, S.-W.; Lee, S.-K.; Belcher, A. M. Adv. Mater. 2003, 15, 689-692. 20. Fowler, C. E.; Shenton, W.; Stubbs, G.; Mann, S. Adv. Mater. 2001, 13, 1266. 21. Scheibel, T.; Parthasarathy, R.; Sawicki, G.; Lin, X. M.; Jaeger, H.; Lindquist, S. L. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 4527-4532. 22. Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Science 2001, 294, 1684-1688.[Medline] 23. Reches, M.; Gazit, E. Science 2003, 300, 625-627.[Medline] 24. Kay, B. K.; Winter, J.; McCafferty, J. Phage Display of Peptides and Proteins: A Laboratory Manual; Academic Press: San Diego, 1996. 25. A genetically encoded hexahistidine peptide and a flexible amino acid spacer sequence were fused upstream of M13 virus pIX gene in a pAK derived phagemid (ref 14, Mao 2003) by PCR cloning techniques. The mature protein sequence of expressed His6-pIX was AHHHHHHGQGGGVDMSVLVYSFASFVLGWCLRSGITYFTRLMETSS after cleavage of the pelB leader sequence in E. coli, as confirmed by DNA sequencing. 26. Streptavidin (1.5 mg/mL) dissolved in 1 mL 0.1 M MES [2-(N-morpholino) ethane sulfonic acid] and 0.5 M NaCl (pH 6.0), was activated by adding 0.01 M EDC [1-ethyl-3-(3-dimethylaminopropyl)carbodiimidie hydrochloride] and 0.01 M sulfo-NHS (N-hydroxysulfosuccinimide). After 15 min at room temperature, EDC was quenched with 2 uL of 2-mercaptoethanol. The buffer was exchanged to sodium phosphate buffer (0.3 mL, 0.1 M, pH 7.5) using a 10 kDa-cutoff spin column (Microcon). NTA ligand was added, and after 3 h at room temperature the buffer was exchanged to Tris buffered saline (pH 8.0). The conjugated NTA-streptavidin was incubated for 5 h in NiSO4 (5 mM in 0.5 mL TBS), spin purified again, and dissolved in distilled water. 27. Weber, P. C.; Ohlendorf, D. H.; Wendoloski, J. J.; Salemme, F. R. Science 1989, 243, 85-88.[Medline] 28. Cohen, L.; Strupat, K.; Hillenkamp, F. J. Am. Soc. Mass Spectrom. 1997, 8, 1046-1052. 29. Devlin, J. J.; Panganiban, L. C.; Devlin, P. E. Science 1990, 249, 404-406.[Medline] 30. Viruses were labeled by mixing 10 uL of 1: 100 diluted anti-fd bacteriophage antibody (Sigma-Aldrich, 7 mg/mL) with 10 uL of anti-rabbit 10 nm gold conjugate (Sigma-Aldrich, 1.4 × 1013 particles/mL) for 1 h, then adding 10 uL of 1: 1 virus-linker suspension. 31. Reiss, B. D.; Mao, C.; Solis, D. J.; Sweeney, R. Y.; Ryan, K. S.; Aggarwal, A.; Thomson, T.; Belcher, A. M., in preparation. 32. Castano, F. J.; Ross, C. A.; Frandsen, C.; Eilez, A.; Gil, D.; Smith, H. I.; Redjdal, M.; Humphrey, F. B. Phys. Rev. B 2003, 67, 184425. 33. Zhu, J.; Zheng, Y.; Prinz, G. J. Appl. Phys. 2000, 87, 6668-6673. 34. Schmitt, J.; Hess, H.; Stunnenberg, H. G. Mol. Biol. Rep. 1993, 18, 223-230. 35. Green, N. M. Adv. Protein Chem. 1975, 29, 85-133.[Medline] 36. Hendrickson, W. A.; Pahler, A.; Smith, J. L.; Satow, Y.; Merritt, E. A.; Phizackerley, R. P. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 2190-2194.[Medline] 37. Hochuli, E. Genet. Eng. 1990, 12, 87-98. 38. Lehn, J. M. Supramolecular chemistry: concepts and perspectives; VCH: Weinheim; New York, 1995 |
|
Want more information ? Interested in the hidden information ? Click here and do your request. |