1848
Bioconjugate Chem. 2009, 20, 1848–1852
Gold Nanoparticle Arrangement on Viral Particles through Carbohydrate Recognition: A Non-Cross-Linking Approach to Optical Virus Detection Kenichi Niikura,*,‡ Keita Nagakawa,† Noriko Ohtake,† Tadaki Suzuki,§ Yasutaka Matsuo,‡ Hirofumi Sawa,§,| and Kuniharu Ijiro‡ Graduate School of Science and Research Institute for Electronic Science, Hokkaido University, N21W10, Sapporo 001-0021, Japan, Research Center for Zoonosis Control and Global COE program, Hokkaido University, N20W10, Sapporo 001-0020, Japan. Received June 9, 2009; Revised Manuscript Received August 8, 2009
We propose a new approach to optical virus detection based on the spatial assembly of gold nanoparticles on the surface of viruses. Since JC virus-like particles (VLPs) comprise a repeating viral capsid protein that binds to sialic acid, the conjugation of sialic acid-linked Au particles with VLPs enables the spatial arrangement of Au particles on the VLP surface. This structure produced a red shift in the absorption spectrum due to plasmon coupling between adjacent Au particles, leading to the construction of an optical virus detection system. Our system depends not on the simple cross-linking of VLPs and Au particles, but on an ordered Au structure covering the entire surface of the VLPs and can be applied to various virus detection systems using the inherent ligand recognition of animal viruses.
Plasmonic nanostructures have received increasing attention due to their potential application to the optical sensing of biological and chemical analytes. A metal nanoparticle sensor is based on the analyte-induced assembly of metal nanoparticles (1-4). The assembly of nanoparticles leads to a red shift in the peak wavelength of the surface plasmon resonance frequency. This principle has been applied to various sensing systems, including DNA (5), lectin (6), protein A (7), potassium ion (8), heavy metal ion (9), and nitrate detection (10). These methods require the cross-linking of nanoparticles by an analyte to form aggregations, in which the gap between nanoparticles is small enough to allow interparticle plasmon coupling. However, the analytes are much larger in size than the metallic nanoparticle, such as with viruses and bacteria; the nanoparticles cannot easily aggregate within the plasmon coupling distance. In this paper, we propose a novel approach to the optical detection of viruses using gold nanoparticles. Recently, inorganic nanostructures prepared with viruses as templates have been often reported (11-16). However, there have been no reports on the use of their inherent molecular recognitions for the spatially ordered arrangement of metal nanoparticles, which can be applied to virus detection. Since the virus has multiple recognition sites for ligands on its surface, the ligand-displaying nanoparticles could be specifically arranged on the viral surface (Figure 1A). In this case, viruses can be detected from the plasmon shift derived from the assembled gold nanoparticles on the virus surface rather than the analyte-induced aggregation of metallic nanoparticles. The surface plasmon resonance shift strongly depends on both the interparticle distance and the number of coupled particles (17). The plasmon shift for multiple (more than two) nanoparticles in linear arrangement is larger than that for two nanoparticles; thus, the multiple coupling of nanoparticles on the virus is expected to cause a large shift of the plasmon band. In addition, the assembly of metallic * Corresponding author. Tel/Fax: +81-11-706-9370. E-mail:
[email protected]. ‡ Research Institute for Electronic Science. † Graduate School of Science. § Research Center for Zoonosis Control. | Global COE program.
nanoparticles into a shell-shaped structure is expected to lead to a large plasmon shift. Metallic clusters formed on a spherical body, such as silica particles (diameter∼100 nm) (18) or liposomes (diameter∼50 nm) (19), show a large red shift (∼200 nm) of the plasmon band. The virus-templated assembly of metallic nanoparticles in a shell-shaped structure is thought to contribute to this large plasmon shift. To test our approach to the detection of viruses, we used JC virus, which belongs to the Polyomavirus family. This animal virus is a non-enveloped virus that infects humans through an N-glycoprotein with R2,6-linked sialic acid on the cell surface (20). The capsids contain 360 copies of viral protein 1 (VP1), arranged in 72 VP1 pentamers in an icosahedral lattice (21), and each VP1 monomer can bind to the R2,6-linked sialic acid. Since, without the inner DNA genome, the VP1s self-assemble to form virus-like particles (VLPs) in Vitro, we used the VLPs as a model of native viruses. Using the binding affinity of VP1 to sialic acid, it is expected that the nanoparticles can be arranged on the virus surface corresponding to the spatial pattern of the VP1s (Figure 1A). Herein, using scanning transmission electron microscopy (STEM), we demonstrate that sialic acid-linked gold nanoparticles (sialyl-Au particles) can be assembled on the surface of VLPs. With the aid of a macromolecular crowding reagent, which promotes the association between VLPs and sialyl-Au particles, the interparticle plasmon coupling of the sialyl-Au particles on the viral surface could be detected in a solution as a red shift of the surface plasmon resonance band. On the basis of previous literature (22, 23), we synthesized a sialic acid-linked lipid (Figure 1B). Synthetic details are shown in the Supporting Information. On the surface of commercially available gold nanoparticles, a ligand-exchange reaction was carried out in the presence of the sialic acid-linked lipid (10 µM) in water. We prepared three sizes of sialyl-Au particles with diameters of 5, 10, and 15 nm, respectively. The binding of the sialic acid-linked lipid to the Au nanoparticles was qualitatively confirmed by FT-IR spectrometry (see Supporting Information Figure S1). JC capsid protein VP1s recombinantly expressed in Escherichia coli spontaneously self-assemble into virus-like particles (VLPs) as described in our previous report (24). The conjugation
10.1021/bc900255x CCC: $40.75 2009 American Chemical Society Published on Web 09/11/2009
Communications
Bioconjugate Chem., Vol. 20, No. 10, 2009 1849
Figure 1. Schematic representation of the detection of virus particles. (A) The formation of the sialyl-Au particle array based on virus-like particles (VLPs) through viral ligand recognition. (B) Structures of the sialic acid-linked and carboxyl lipids used in the preparation of sialyl-Au and carboxylAu particles.
Figure 2. STEM images of (A) VLPs alone, (B) VLPs incubated with 10-nm sialyl-Au particles, and (C) VLPs with 10 nm carboxyl-Au particles. The scale bars represent 100 nm. Sialyl-Au particles were arranged on VLPs, and the diameters of the sialyl-Au particles were (D) 5 nm, (E) 10 nm, and (F) 15 nm, respectively. The scale bar represents 20 nm. The samples were stained with 2% phosphotungstic acid.
of sialyl-Au particles with VLPs was confirmed by STEM. VLP solutions were mounted and immobilized on a carbon-coated TEM grid. After bovine serum albumin (BSA) blocking, a solution of sialyl-Au particles was placed on the VLP-immobilized grid. It is important to note that, for effective binding, this solution was gradually concentrated by evaporation of the solvent on the TEM grid. After washing the grid surface with pure water to remove unbound nanoparticles, the samples were stained with 2% phosphotungstic acid. We observed a homogeneous population of VLPs of around 40 nm in diameter by STEM (Figure 2A). After incubation of the sialyl-Au particles (10 nm) with VLPs for 90 min, binding was confirmed by visualization (Figure 2B). In a control experiment, Au nanoparticles covered with carboxyl lipids were found not to bind onto the VLPs (Figure 2C), indicating that the binding is based on specific recognition between the sialyl-Au particles and the VLPs. The dark background observed in Figure 2B,C is due to BSA blocking. Figure 2D,E,F shows the typical binding behavior between VLPs and sialyl-Au particles of 5 nm, 10 nm, and 15 nm, respectively. Using 5 and 10 nm sialyl-Au particles, the only observable nanoparticle assembly is in the form of a “ring” structure surrounding the VLPs. These “ring” TEM images have
often been observed in noncovalent DNA interaction-driven nanoparticle assemblies (25, 26). In the case of the 15 nm sialylAu particles, however, the particles were observed over the entire VLP surface (Figure 2F). We believe that the larger contact area of the 15 nm particles produces an increased number of sialic acid-VP1 interactions affording a stronger affinity between the VLPs and Au particles than that of 5 nm and 10 nm particles. When Au nanoparticles are assembled, their surface plasmons combine, and the aggregate can be considered a single large particle. This causes a red shift and a broadening of the surface plasmon band. However, the dissociation constant of the native sialic acid-containing glycoprotein and VLPs is on the order of 10-7 M (24); thus, the concentration used in this study (nM order) was too low for the conjugation of Au nanoparticles with VLPs in solution. Thus, we added dextran as a macromolecular crowding reagent. Molecular crowding enhances the equilibrium association of dilute macromolecules (27-29). We expected the addition of dextran to induce complex formation, leading to a clear red shift in the plasmon spectra. Figure 3A shows the effect of the molecular crowding reagent on the red shift of the surface plasmon band. At above 30 wt %, dextran induced effective association between the 15 nm sialyl-Au particles and VLPs. As a control, without VLPs, no such shifts were observed, supporting the notion that the red shift can be attributed to the specific interaction between sialyl-Au particles and VLPs. The color of the sialyl-Au particle solution was changed from red to purple after the addition of VLPs (Supporting Information Figure S2). The surface plasmon spectrum of the 15 nm sialylAu particles in the presence of VLPs and 40 wt % dextran is shown together with the control spectra in Figure 3B. A 22 nm red shift was observed when 15 nm sialyl-Au particles were incubated with VLPs. To exclude the possibility of nonspecific binding driven by electrostatic interaction, we prepared other negatively charged gold nanoparticles coated with a carboxyl lipid (carboxyl-Au particles). The zeta potential of the sialylAu and carboxyl-Au particles were almost the same (-31 ( 1.9 mV and -29 ( 3.8 mV, respectively), indicating that the carboxyl-Au particles afford a suitable control for the negatively charged sialyl-Au particles. The addition of carboxyl-Au particles to a solution of VLPs did not cause a plasmon shift.
1850 Bioconjugate Chem., Vol. 20, No. 10, 2009
Communications
Figure 3. (A) The effect of dextran on the red shift of the surface plasmon band of sialyl-Au particles with or without VLPs ([sialyl-Au particles] ) 4 nM, [VLPs] ) 0.8 nM). (B) UV-vis spectra for sialyl-Au and carboxyl-Au particles with VLPs in the presence of 40 wt % dextran. As negative controls, 15 nm sialyl-Au particle solutions without VLPs in the presence of 40 wt % dextran and 15 nm sialyl-Au particle solutions containing 5.5 mM sialyllactose as an inhibitor were measured. (C) The red shift of the surface plasmon band of 15 nm sialyl-Au and carboxyl-Au particles in 40 wt % dextran-containing buffered solution as a function of the ratio of Au particles to VLPs. VLP concentration remained at 0.8 nM. (D) UV-vis spectra of solutions of 5 nm and 10 nm sialyl-Au particles containing 40 wt % dextran, with or without VLPs. (E) Schematic representations of the binding of each size of Au nanoparticle to the VLP structure (35). Red dots on VLP represent sialic acid binding sites, with each single VP1 pentamer having five binding sites in total. The yellow areas represent the area occupied by each size of Au nanoparticles. The 15 nm Au nanoparticle can be seen to overlap with the adjacent VP1 pentamers. The scale bar represents 10 nm.
Moreover, no red shift of the sialyl-Au particles was observed in the presence of VLPs and an excess amount of sialyllactose (Figure 3B). This indicates that the red shift in Figure 3B is due to specific binding between the sialyl-Au particles and VLPs. Keating et al. have demonstrated the effect of macromolecular crowding reagents, such as dextran and poly(ethylene glycol) (PEG) solutions, on the thermodynamics of DNA-linked nanoparticle assembly (30). They reported that the macromolecular reagents induced not only DNA hybridizationdriven nanoparticle assembly, but also undesired nonspecific nanoparticle aggregation. In our experiment on carbohydratevirus interaction, we clearly demonstrate that the macromolecular crowding reagent is effective in enhancing the specific association of the nanoparticle-biomolecule conjugation. The
degree of the red shift was increased in a dose-dependent manner as a function of sialyl-Au particle concentration (Figure 3C), indicating that the interparticle distances were gradually shortened and collective surface plasmon resonance (SPR) changed through modulation of dipole coupling. As an important control, the addition of carboxyl-Au particles to VLPs did not affect the wavelength of the plasmon peak. On the other hand, the addition of dextran to solutions of 5 nm and 10 nm sialyl-Au particles in the presence of VLPs had no observable effect on the red shift (Figure 3D). Figure 3E shows the difference in size between each Au-particle and a VLP. Each VLP is composed of 72 VP1 pentamers, the diameter of which is approximately 8 nm. A single 5 nm Au particle is able to cover five binding sites of a single VLP pentamer. A 10 nm Au particle is larger than a VP1
Communications
pentamer, but it is not large enough to reach the binding sites located on adjacent VP1 pentamers. A 15 nm Au particle disposed on a particular VP1 pentamer, however, can reach six adjacent VP1 pentamers. The number of linkages between the single sialyl-Au particle and the recognition sites of the VLP proportionally influences the binding enthalpy due to cooperative interaction. The larger Au nanoparticles, which have a larger contact area, form a larger number of sialic acid-VLP linkages. Therefore, we can attribute the large plasmon shift of the 15 nm sialyl-Au particles in solution to the increased affinity. The main factors controlling plasmon coupling are interparticle gap and particle size, and the relationship between these which can be quantitatively expressed as D/(2r), where D is the distance between the centers of adjacent particles, and 2r is the diameter of the metal nanoparticle (31). It is known that strong interparticle coupling between a pair of adjacent nanoparticles occurs when D/(2r) is smaller than 1.2. In the case of sialyl-Au particles, since the molecular length of the sialic acid lipid is approximately 4.5 nm, the minimum gap when two particles are attached to each other between the metal cores of two sialyl-Au particles is calculated to be 9 nm (D/(2r) ) 1.6 for the 15 nm Au particles) (Supporting Information Figure S3). According to previous theoretical and experimental data (17, 32), a plasmonic-coupled pair of Au particles having a D/(2r) value of 1.6 cannot explain the pronounced shift (22 nm) observed in this study. Arrayed colloidal nanoparticles, however, have been reported to demonstrate near-field coupling, although the D/(2r) value was larger than 1.2, leading to coherent interactions between multiple adjacent colloidal nanoparticles (33, 34). Therefore, the shift observed here might be caused not by the interparticle coupling from a single pair of metal particles, but by multiple particle arrays. In summary, we found that the 15 nm sialyl-Au particles are suitable to the optical detection of JC virus VLPs through sialic acid recognition with the aid of a crowding reagent. STEM images supported the notion of multiple Au particle interactions on the surface of a single VLP, leading to a red plasmon shift. Since animal viruses often show inherent molecular recognition, for example, recognizing carbohydrate molecules, the use of ligand-specific interactions could provide a versatile approach to virus detection. Furthermore, we propose that this method is not only applicable to virus detection but also as a simple approach to the investigation of the optical properties of a nanospace surrounded by multiple metal nanoparticles.
ACKNOWLEDGMENT This study was supported in part by grants from the Ministry of Education, Culture, Sports, Science and Technology; the Ministry of Health, Labor and Welfare, Japan; and the Program of Founding Research Centers for Emerging and Reemerging Infectious Diseases, MEXT, Japan. Nagakawa and Ohtake appreciate the financial support provided by JSPS. The analysis of STEM was carried out at the OPEN FACILITY, Hokkaido University Sousei Hall and HINTS. Supporting Information Available: Experimental details for the synthesis of sialic acid-linked lipid. Sample preparations for Au-nanoparticles-VLP conjugation. This material is available free of charge via the Internet at http:// pubs.acs.org.
LITERATURE CITED (1) Rosi, N. L., and Mirkin, C. A. (2005) Nanostructures in biodiagnostics. Chem. ReV. 105, 1547–1562.
Bioconjugate Chem., Vol. 20, No. 10, 2009 1851 (2) Hong, S., Choi, I., Lee, I., Yang, Y. I., Kang, T., and Yi, J. (2009) Sensitive and colorimetric detection of the structural evolution of superoxide dismutase with gold nanoparticles. Anal. Chem. 81, 1378–1382. (3) Storhoff, J. J., Elaghantan, R., Mucic, R. C., Mirkin, C. A., and Letsinger, R. L. (1998) One-pot colorimetric differentiation of polynucleotides with single base imperfections using gold nanoparticle probes. J. Am. Chem. Soc. 120, 1959–1964. (4) Aili, D., Enander, K., Baltzer, L., and Liedberg, B. (2008) Assembly of polypeptide-functionalized gold nanoparticles through a heteroassociation- and folding-dependent bridging. Nano Lett. 8, 2473–2478. (5) Mirkin, C. A., Lestinger, R. L., and Mucic, R. C. (1996) A DNA-based method for rationally assembling nanoparticles into macroscopic materials. Nature 382, 607–609. (6) Otsuka, H., Akiyama, Y., Nagasaki, Y., and Kataoka, K. (2001) Quantitative and reversible lectin-induced association of gold nanoparticles modified with R-lactosyl-ω-mercapto-poly(ethyleneglycol). J. Am. Chem. Soc. 123, 8226–8230. (7) Thanh, N. T. K., and Rosenzweig, Z. (2002) Development of an aggregation-based immunoassay for anti-protein a using gold nanoparticles. Anal. Chem. 74, 1624–1628. (8) Lin, S., Liu, S., Lin, C., and Chen, C. (2002) Recognition of potassium ion in water by 15-crown-5 functionalized gold nanoparticles. Anal. Chem. 74, 330–335. (9) Kim, Y., Johnson, R. C., and Hupp, J. T. (2001) Gold nanoparticle-based sensing of “spectroscopically silent” heavy metal ions. Nano Lett. 1, 165–167. (10) Daniel, W. L., Han, M. S., Lee, J., and Mirkin, C. A. (2009) Colorimetric nitrite and nitrate detection with gold nanoparticle probes and kinetic end points. J. Am. Chem. Soc. 131, 6362– 6363. (11) Blum, A. S., Soto, C. M., Wilson, C. D., Cole, J. D., Kim, M., Gnade, B., Chatterji, A., Ochoa, W. F., Lin, T., Johnson, J. E., and Ratna, B. R. (2004) Cowpea mosaic virus as a scaffold for 3-D patterning of gold nanoparticles. Nano Lett. 4, 867–870. (12) Tseng, R. J., Tsai, C., Ma, L., Ouyang, J., Ozkan, C. S., and Yang, Y. (2006) Digital memory device based on tobacco mosaic virus conjugated with nanoparticles. Nat. Nanotechnol. 1, 72– 77. (13) Radloff, C., Vaia, R. A., Brunton, J., Bouwer, G. T., and Ward, V. K. (2005) Metal nanoshell assembly on a virus bioscaffold. Nano Lett. 5, 1187–1191. (14) Slocik, J. M., Naik, R. R., Stone, M. O., and Wright, D. W. (2005) Viral templates for gold nanoparticle synthesis. J. Mater. Chem. 15, 749–753. (15) Dujardin, E., Peet, C., Stubbs, G., Culver, J. N., and Mann, S. (2003) Organization of metallic nanoparticles using tobacco mosaic virus templates. Nano Lett. 3, 413–417. (16) Liu, A., Abbineni, G., and Mao, C. (2009) Nanocomposite films assembled from genetically engineered filamentous viruses and gold nanoparticles: nanoarchitecture- and humidity-tunable surface plasmon resonance spectra. AdV. Mater. 21, 1001–1005. (17) Zhong, Z., Patskovskyy, S., Bouvrette, P., Luong, J. H. T., and Gedanken, A. (2004) The surface chemistry of Au colloids and their interactions with functional amino acids. J. Phys. Chem. B 108, 4046–4052. (18) Oldenburg, S. J., Averitt, R. D., Westcott, S. L., and Halas, N. J. (1998) Nanoengineering of optical resonances. Chem. Phys. Lett. 288, 243–247. (19) Troutman, T. S., Barton, J. K., and Romanowski, M. (2009) Biodegradable plasmon resonant nanoshells. AdV. Mater. 20, 2604–2608. (20) Gee, G. V., Tsomaia, N., Mierke, D. F., and Atwood, W. J. (2004) Modeling a sialic acid binding pocket in the external loops of JC virus VP1. J. Biol. Chem. 279, 49172–49176. (21) Griffith, J. P., Griffith, D. L., Rayment, I., Murakami, W. T., and Casper, D. L. D. (1992) Inside polyomavirus at 25-Å resolution. Nature 355, 652–654.
1852 Bioconjugate Chem., Vol. 20, No. 10, 2009 (22) Barrientos, A. G., de la Fuente, J. M., Rojas, T. C., Ferna´ndez, A., and Penade´s, S. (2003) Gold glyconanoparticles: Synthetic polyvalent ligands mimicking glycocalyx-like surfaces as tools for glycobiological studies. Chem.sEur. J. 9, 1909–1921. (23) Niikura, K., Nishio, T., Akita, H., Matsuo, Y., Kamitani, R., Kogure, K., Harashima, H., and Ijiro, K. (2007) Accumulation of O-GlcNAc-displaying CdTe quantum dots in cells in the presence of ATP. ChemBioChem 8, 379–384. (24) Ohtake, N., Niikura, K., Suzuki, T., Nagakawa, K., Sawa, H., and Ijiro, K. (2008) Enhanced cellular uptake of virus-like particles through immobilization on a sialic acid-displaying solid surface. Bioconjugate Chem. 19, 507–515. (25) Mucic, R. C., Storhoff, J. J., Mirkin, C. A., and Lestinger, R. L. (1998) DNA-directed synthesis of binary nanoparticle network materials. J. Am. Chem. Soc. 120, 12674–12675. (26) Sebba, D. S., Mock, J. J., Smith, D. R., LaBean, T. H., and Lazarides, A. A. (2008) Reconfigurable core-satellite nanoassemblies as molecularly-driven plasmonic switches. Nano Lett. 8, 1803–1808. (27) Minton, A. P. (2006) How can biochemical reactions within cells differ from those in test tubes? J. Cell Sci. 119, 2863– 2869. (28) Minton, A. P. (2001) The influence of macromolecular crowding and macromolecular confinement on biochemical reactions in physiological media. J. Biol. Chem. 276, 10577– 10580.
Communications (29) Ellis, R. J. (2001) Macromolecular crowding: obvious but underappreciated. Trends Biochem. Sci. 26, 597–604. (30) Goodrich, G. P., Helfrich, M. R., Overberg, J. J., and Keating, C. D. (2004) Effect of macromolecular crowding on DNA: Au nanoparticle bioconjugate assembly. Langmuir 20, 10246–10251. (31) Zhao, L., Kelly, K. L., and Schatz, G. C. (2003) The extinction spectra of silver nanoparticle arrays: influence of array structure on plasmon resonance wavelength and width. J. Phys. Chem. B 107, 7343–7350. (32) Jain, P. K., Huang, W., and El-sayed, M. A. (2007) On the universal scaling behavior of the distance decay of plasmon coupling in metal nanoparticle pairs: a plasmon ruler equation. Nano Lett. 7, 2080–2088. (33) Kang, Y., Erickson, K. J., and Taton, T. A. (2005) Plasmonic nanoparticle chains via a morphological, sphere-to-string transition. J. Am. Chem. Soc. 127, 13800–13801. (34) Yockell-Lelie`vre, H., Desbiens, J., and Ritcey, A. M. (2007) Two-dimensional self-organization of polystyrene-capped gold nanoparticles. Langmuir 23, 2843–2850. (35) Carrillo-Tripp, M., Shepherd, C. M., Borelli, I. A., Venkataraman, S., Lander, G., Natarajan, P., Johnson, J. E., Brooks III, C. L., and Reddy, V. S. (2009) VIPERdb2: an enhanced and web API enabled relational database for structural virology. Nucleic Acids Res. 37, D436–D442. BC900255X
