Colloidal Au-Enhanced Surface Plasmon Resonance Immunosensing

and an antigen-colloidal Au conjugate comprise the binding pair. A highly specific ... plasmon angle shift is presented, thereby providing for a direc...
7 downloads 0 Views 154KB Size
Download PDF
Anal. Chem. 1998, 70, 5177-5183

Colloidal Au-Enhanced Surface Plasmon Resonance Immunosensing L. Andrew Lyon, Michael D. Musick, and Michael J. Natan*

Department of Chemistry, The Pennsylvania State University, University Park, Pennsylvania 16802

Surface plasmon resonance (SPR) biosensing using colloidal Au enhancement is reported. Immobilization of ∼11-nm-diameter colloidal Au to an evaporated Au film results in a large shift in plasmon angle, a broadened plasmon resonance, and an increase in minimum reflectance. The incorporation of colloidal Au into SPR biosensing results in increased SPR sensitivity to proteinprotein interactions when a Au film-immobilized antibody and an antigen-colloidal Au conjugate comprise the binding pair. A highly specific particle-enhanced analogue of a sandwich immunoassay is also demonstrated by complexing the Au particle to a secondary antibody. A tremendous signal amplification is observed, as addition of the antibody-Au colloid conjugate results in a 25-fold larger signal than that due to addition of a free antibody solution that is 6 orders of magnitude more concentrated. Picomolar detection of human immunoglobulin G has been realized using particle enhancement, with the theoretical limits for the technique being much lower. Finally, a quasi-linear relationship between particle coverage and plasmon angle shift is presented, thereby providing for a direct correlation between plasmon shift and solution antigen concentration. Together, these results represent significant advances in the generality and sensitivity of SPR as it is applied to biosensing. Surface plasmon resonance (SPR) is a general method for the detection of chemical changes occurring at the surface of a thin noble metal film. Immobilization of monolayer or multilayer quantities of material onto a SPR-active film, as well as subsequent chemical modification of that layer, can be detected as a small change in refractive index at the metal surface. As a result of the generality of the technique, SPR has proven to be useful for the characterization of a wide variety of chemistries1 and as a * To whom correspondence should be addressed: [email protected]. (1) (a) Hayashida, O.; Shimizu, C.; Fujimoto, T.; Aoyama, Y. Chem. Lett. 1998, 13-14. (b) Sigal, G. B.; Mrksich, M.; Whitesides, G. M. Langmuir 1997, 13, 2749-2755. (c) Caruso, F.; Rodda, E.; Furlong, D. N.; Haring, V. Sens. Actuators B 1997, 41, 189-197. (d) Hanken, D. G.; Naujok, R. R.; Gray, J. M.; Corn, R. M. Anal. Chem. 1997, 69, 240-248. (e) Peterlinz, K. A.; Georgiadis, R. M.; Herne, T. M.; Tarlov, M. J. J. Am. Chem. Soc. 1997, 119, 3401-3402. (f) Bier, F. F.; Kleinjung, F.; Scheller, F. W. Sens. Actuators B 1997, 38-39, 78-82. (g) Hanken, D. G.; Corn, R. M. Anal. Chem. 1995, 67, 3767-3774. (h) Frey, B. L.; Jordan, C. E.; Kornguth, S.; Corn, R. M. Anal. Chem. 1995, 67, 4452-4457. (i) Mrksich, M.; Sigal, G. B.; Whitesides, G. M. Langmuir 1995, 11, 4383-4385. 10.1021/ac9809940 CCC: $15.00 Published on Web 11/06/1998

© 1998 American Chemical Society

complementary technique for other analytical methods.2 The technique has also been expanded to imaging, microscopy, and holography formats,3 while fiber-optic and miniaturized SPR sensors have been introduced for remote monitoring applications.4 Perhaps the most widely studied subset of chemistries studied by SPR is protein-protein interactions,5,6 where binding event signal transduction is difficult or impossible to accomplish by traditional optical spectroscopies. The vast majority of papers on biological SPR describe experiments conducted on commercial instrumentation that utilize an extended coupling matrix to decrease nonspecific binding and to increase surface loading of biomolecules.7 Such measurements typically begin with one protein immobilized on proprietary substrates comprising a modified dextran (extended coupling) matrix atop a thin evaporated Au film.7 Protein binding events leading to small changes in the refractive index of the dextran layer are detected via correspondingly small changes in the angle-dependent attenuated total reflectance.8 Despite the signal amplification afforded by the (2) (a) Nelson, R. W.; Krone, J. R.; Jansson, O. Anal. Chem. 1997, 69, 43634368. (b) Krone, J. R.; Nelson, R. W.; Dogruel, D.; Williams, P.; Granzow, R. Anal. Biochem. 1997, 244, 124-132. (c) Wu, X.-Z.; Nakagawa, M.; Nagamori, C.; Uchiyama, K.; Hobo, T. Bull. Chem. Soc. Jpn. 1996, 69, 19691974. (d) Jory, M. J.; Bradberry, G. W.; Cann, P. S.; Sambles, J. R. Sens. Actuators B 1996, 35-36, 197-201. (e) Chinowsky, T. M.; Saban, S. B.; Yee, S. S. Sens. Actuators B 1996, 35-36, 37-43. (f) Jung, C. C.; Saban, S. B.; Yee, S. S.; Darling, R. B. Sens. Actuators B 1996, 32, 143-147. (3) (a) Jordan, C. E.; Corn, R. M. Anal. Chem. 1997, 69, 1449-1456. (b) Jordan, C. E.; Frutos, A. G.; Thiel, A. J.; Corn, R. M. Anal. Chem. 1997, 69, 49394947. (c) Thiel, A. J.; Frutos, A. G.; Jordan, C. E.; Corn, R. M.; Smith, L. M. Anal. Chem. 1997, 69, 4948-4956. (d) Hickel, W.; Kamp, D.; Knoll, W. Nature 1989, 339, 186. (e) Rothenhausler, B.; Knoll, W. Nature 1988, 332, 615-617. (f) Hickel, W.; Knoll, W. J. Appl. Phys. 1990, 67, 35723575. (g) Knobloch, H.; Knoll, W. Makromol. Chem., Macromol. Symp. 1991, 46, 389-393. (h) Flatgen, G.; Krischer, K.; Pettinger, B.; Doblhofer, K.; Junkes, H.; Ertl, G. Science 1995, 269, 668-671. (i) Kim, Y.-K.; Ketterson, J. B.; Morgan, D. J. Opt. Lett. 1996, 21, 165-167. (j) Maruo, S.; Nakamura, O.; Kawata, S. Appl. Opt. 1997, 36, 2343-2346. (4) (a) Melendez, J.; Carr, R.; Bartholomew, D.; Taneja, H.; Yee, S.; Jung, C.; Furlong, C. Sens. Actuators B 1997, 38-39, 375-379. (b) Melendez, J.; Carr, R.; Bartholomew, D. U.; Kukanskis, K.; Elkind, J.; Yee, S.; Furlong, C.; Woodbury, R. Sens. Actuators B 1996, 35, 1-5. (c) Abdelghani, A.; Chovelon, J. M.; Jaffrezic-Renault, N.; Ronot-Trioli, C.; Veillas, C.; Gagnaire, H. Sens. Actuators B 1997, 38-39, 407-410. (d) Abdelghani, A.; Chovelon, J. M.; Jaffrezic-Renault, N.; Veillas, C.; Gagnaire, H. Anal. Chim. Acta 1997, 337, 225-232. (e) Tubb, A. J. C.; Payne, F. P.; Millington, R. B.; Lowe, C. R. Sens. Actuators B 1997, 41, 71-79. (f) Jorgenson, R. C.; Yee, S. S. Sens. Actuators B 1993, 12, 213-220. (g) Niggemann, M.; Katerkamp, A.; Pellmann, M.; Bolsmann, P.; Reinbold, J.; Cammann, K. Sens. Actuators B 1996, 34, 328-333. (h) Matsubara, K.; Kawata, S.; Minami, S. Appl. Spectrosc. 1988, 42, 1375-1379. (5) For recent reviews see, for example: (a) Schuck, P. Annu. Rev. Biophys. Biomol. Struct. 1997, 26, 541-566. (b) Garland, P. B. Q. Rev. Biophys. 1996, 29 (1), 91-117. (c) Lo¨fås, S. Pure Appl. Chem. 1995, 67, 829-834.

Analytical Chemistry, Vol. 70, No. 24, December 15, 1998 5177

dextran matrix, the immobilization of small (18 MΩ with a Barnstead Nanopure system and filtered through a 200-µm particulate filter. The following biomolecules were obtained from Sigma Immunochemicals: human IgG purified from serum, anti-human IgG (6) (a) Heyse, S.; Ernst, O. P.; Dienes, Z.; Hofmann, K. P.; Vogel, H. Biochemistry 1998, 37, 507-522. (b) Tung, J.-S.; Giminez, J.; Przysiecki, C. T.; Mark, G. J. Pharm. Sci. 1998, 87, 76-80. (c) Berger, C. E. H.; Beumer, T. A. M.; Kooyman, R. P. H.; Greve, J. Anal. Chem. 1998, 70, 703-706. (d) Li, J.; Cook, R.; Doyle, M. L.; Hensley, P.; McNulty, D. E.; Chaiken, I. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 6694-6699. (e) Autiero, M.; Gaubin, M.; Mani, J.-C.; Castejon, C.; Martin, M.; El Marhomy, S.; Guardiola, J.; Piatier-Tonneau, D. Eur. J. Biochem. 1997, 245, 208-213. (f) Myszka, D. G.; Morton, T. A.; Doyle, M. L.; Chaiken, I. M. Biophys. Chem. 1997, 64, 127-137. (g) Lange, C.; Koch, K.-W. Biochemistry 1997, 36, 12019-12026. (h) Ladbury, J. E.; Lemmon, M. A.; Zhou, M.; Green, J.; Botfield, M. C.; Schlessinger, J. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 3199-3203. (i) Ward, L. D.; Howlett, G. J.; Hammacher, A.; Weinstock, J.; Yasukawa, K.; Simpson, R. J.; Winzor, D. J. Biochemistry, 1995, 34, 2901-2907. (j) Minunni, M. Anal. Lett. 1995, 28, 933-944. (k) Bernard, A.; Bosshard, H. R. Eur. J. Biochem. 1995, 230, 416-423. (l) O’Brien, D. P.; Kemball-Cook, G.; Hutchinson, A. M.; Martin, D. M. A.; Johnson, D. J. D.; Byfield, P. G. H.; Takamiya, O.; Tuddenham, E. G. D.; McVey, J. H. Biochemistry 1994, 33, 14162-14169. (7) (a) Karlsson, R.; Fa¨lt, A. J. Immunol. Methods 1997, 200, 121-133. (b) Adamczyck, M.; Gebler, J. C.; Gunasekera, A. H.; Mattingly, P. G.; Pan, Y. Bioconjugate Chem. 1997, 8, 133-145. (c) Liedberg, B.; Lundstro ¨m, I.; Stenberg, E. Sens. Actuators B 1993, 11, 63-72. (d) Johne, B.; Gadnell, M.; Hansen, K. J. Immunol. Methods 1993, 160, 191-198. (e) O’Shannessy, D. J.; Brigham-Burke, M.; Soneson, K. K.; Hensley, P.; Brooks, I. Anal. Biochem. 1993, 212, 457-468. (f) Toyama, S.; Ikariyama, Y. Chem. Lett. 1997, 1083-1084. (8) (a) Pockrand, I. Surf. Sci. 1978, 72, 577-588. (b) Economou, E. N. Phys. Rev. 1969, 182, 539-554. (c) Surface Plasmons on Smooth and Rough Surfaces and on Gratings; Raether, H., Ed.; Springer Tracts in Modern Physics 111; Springer-Verlag: Berlin, 1988. (9) (a) Okun, I.; Veerapandian, P. Nature Biotechnol. 1997, 15, 287-288. (b) Major, J. S. J. Recept., Signal Transduction Res. 1995, 15 (1-4), 595-607. (10) (a) Colloidal Gold: Principles, Methods and Applications; Hyatt, M. A., Ed.; Academic Press: New York, 1989; Vols. 1-3. (b) DeRoe, C.; Courtoy, P. J.; Baudhuin, P. J. Histochem. Cytochem. 1987, 35, 1191-1198.

5178 Analytical Chemistry, Vol. 70, No. 24, December 15, 1998

(a-h-IgG, γ-chain and Fc specific, isolated from goat), bovine serum albumin, human IgA (colostrum), and human serum (IgG free). Also purchased from Sigma were disodium phosphate heptahydrate, monosodium phosphate monohydrate, 3-mercaptopropionic acid (MPA), NaCl, 1-ethyl-3-(3-(dimethylamino)propyl)carbodiimide hydrochloride (EDC), and 2-mercaptoethylamine (MEA). Sulfo-N, hydroxysuccinimide (S-NHS) and monoclonal anti-human IgG1 (Fc specific, isolated from mouse) were purchased from Pierce. (3-Mercaptopropyl)trimethoxysilane was purchased from United Chemical and stored in a vacuum desiccator until just prior to use. CH3CH2OH and poly(ethylene glycol) (MW ∼20 000) were purchased from Pharmco and Fluka, respectively. Acids and bases used for pH adjustment (NaOH, HCl, H2SO4, H3PO4) were obtained from J. T. Baker. Two separate batches of colloidal Au were synthesized as previously described.11 Average particle diameter and standard deviations were determined by particle size analysis of TEM images using NIH Image version 1.62 software.12 The two colloidal sols were calculated to consist of 11.5 ( 1.5 and 10 ( 1.1 nm sized particles, respectively. Optical spectra with λmax ) 516 nm and a full peak width at half-maximum of ∼84 nm were recorded using a Hewlett-Packard 8452 spectrophotometer. Protein Conjugates. A flocculation curve was constructed for both h-IgG and a-h-IgG (Fc) Au colloid conjugates to determine the amount of protein that was necessary to coat the exterior of the Au particle.10 For h-IgG, solutions were prepared from 0.75 mg/mL stock solution aliquots (0-72 µg) of h-IgG and were added in 6-µg increments to cuvettes containing 1.0 mL of 10-nmdiameter colloid solution adjusted to pH 8.0 (EM Science colorpHast indicator strips, pH 2-9) using 0.1 M NaOH. The samples were volume-corrected to 1.150 mL with deionized H2O and 150 µL of 1.0 M NaCl was added to each. The solutions were agitated and optical spectra recorded after 10 min. For a 10-nm colloid, 24 µg of h-IgG/1 mL of colloid was required to prevent aggregation (as determined by an absence of broadening in the visibleregion absorbance band). The procedure was repeated for a-hIgG (Fc), and 31.5 µg of a-h-IgG/mL of colloid was determined to inhibit aggregation. Protein-Au conjugates were synthesized by the addition of either h-IgG or a-h-IgG (Fc) to 5 mL of pH-adjusted colloid followed by incubation on ice with periodic gentle mixing for 1 h. The conjugate was then divided into 1-mL fractions in 1.5-mL microcentrifuge tubes and centrifuged at 12500g for 45 min (Heraeus PicoBiofuge). The clear to pink supernatant was removed and the soft pellet resuspended to an optical density of ∼3 into H2O or 40 mM phosphate buffer (pH 7.0). Conjugates can be stored between 2 and 8 °C for several days without loss of activity. Au Film Preparation. Thin (47-50 nm) Au films were prepared by thermal evaporation of Au shot (99.99%, Johnson Mathey) from a resistively heated molybdenum boat (Kurt J. Lesker) in a diffusion-pumped Edwards Auto 306 thin-film fabrication system. Evaporation substrates were 1 in. × 1 in. × 0.02 in. pieces of polished SF11 glass (n ) 1.78, Schott Glass Technologies) that had been exposed to a 10% (v/v) (3-mercaptopropyl)(11) (a) Turkevich, J.; Stevenson, P. C.; Hillier, J. Discuss. Faraday Soc. 1951, 11, 55-75. (b) Frens, G. Nature Phys. Sci. 1973, 241, 20-22. (c) Grabar, K. C.; Freeman, R. G.; Hommer, M. B.; Natan, M. J. Anal. Chem. 1995, 67, 735-743. (12) Available on the Internet at zippy.nimh.nih.gov by anonymous FTP.

trimethoxysilane/CH3OH solution for 30 min in order to increase the adhesion of Au to the glass.13 Au was deposited at a pressure of 1 × 10-6 mbar at 0.5 nm/s with constant sample rotation to ensure uniform deposition. Following evaporation, the films were annealed in a home-built oven at 300 °C for 5-10 min under a constant flow of argon to decrease the surface roughness of the evaporated layer. Atomic Force Microscopy. AFM images acquired were obtain using a Digital Instruments Nanoscope IIIA operated in tapping mode with an acquisition frequency of 1.5 Hz and line density of 512. Standard 200-µm etched silicon probes were used. SPR Instrumentation. Schematic diagrams of the highresolution scanning SPR instrument and the flow cell are available in the Supporting Information. Excitation of the surface plasmon is accomplished using the Kretschmann geometry where a 1-in.diameter hemispherical prism (SF11 glass, Esco Products) is index-matched (Cargille immersion oil, n ) 1.78) to a SF11 substrate onto which Au had been previously evaporated. This assembly is then affixed to a home-built flow cell (volume ∼100 µL) with the Au film exposed to solution. The SPR excitation source is a cylindrical 5 mW, 500:1 polarization extinction HeNe laser (632.8 nm, Melles Griot) which is further polarized by a 500:1 visible-optimized linear polarizer (Newport, 10-LP VIS). An optical chopper (Stanford Research Systems) is used to modulate the optical signal at a frequency of 2 kHz, which was then correlated with detection via a lock-in amplifier (Stanford Research Systems, SR530). The beam is focused by a 100-mm focal length (fl) planoconvex (PCX) lens and recollimated by a 25-mm fl PCX lens, thereby reducing the beam size to 0.4 ( 0.1 mm in diameter. A hemispherical lens is then used to focus the beam such that it is recollimated by the hemispherical prism-sample assembly. The reflected beam is then passed through an iris and focused onto a silicon photodiode detector (Thor Labs), the signal from which is then measured with the lock-in amplifier that is in-phase with the excitation source. Angular positioning of the sample is accomplished with a home-built θ-2θ stage consisting of two highresolution (0.001°) servo-drive rotation stages (Newport, RTM80CC and RTM160CC annular rotary stages) that are mounted together such that their axes of rotation are collinear. The prism/sample/ flow cell assembly is then mounted on the θ-2θ stage such that the center of the Au/glass sample is at the axis of rotation. Stage rotation and data collection are controlled through a computer interface that was developed in-house with the LabVIEW programming language (version 4.01, National Instruments). A typical SPR scan was run at either 0.1 or 0.01° resolution, a stage rotation rate of 0.5° s-1, and a lock-in time constant of 0.3 s. Au Film Derivitization. Annealed Au films were modified with alkanethiols (MPA and MEA) from 10 mM ethanolic solutions for a period of 30 min. Longer immersion times did not noticeably improve the efficiency or density of subsequent derivitization steps. Immobilization of colloidal Au onto aminemodified (MEA-coated) surfaces was performed from undiluted aqueous sols at room temperature. Biosensing experiments utilized carboxylate-modified (MPA-coated) surfaces to covalently attach proteins to the Au film via traditional carbodiimide coupling to protein free amine moieties.14 A typical protein immobilization (13) Goss, C. A.; Charych, D. A.; Majda, M. Anal. Chem. 1991, 63, 85-88. (14) Grabarek, Z.; Gergely, J. Anal. Biochem. 1990, 185, 131-135.

Figure 1. SPR curves (reflectance vs angle) for a 2-mercaptoethylamine-coated 50-nm evaporated gold film modified with 11-nmdiameter colloidal Au for varying exposure times: 0 s (s), 60 s (‚‚‚), and 60 min (- -). Intermediate exposure times are omitted for clarity. Inset: change in reflectance at the plasmon angle of a 65-nm evaporated gold film measured during a 10-s exposure to 11-nm Au colloid.

procedure begins with formation of an active ester at the surface by reacting 100 µL of a 100 mM, pH 5.5, EDC solution with the carboxylated surface for 15 min. A 50-µL aliquot of S-NHS (40 mM, pH 7) is then injected into the flow cell to stabilize the reactive surface through displacement of the imide moiety. After 15 min of incubation with S-NHS, the cell is rinsed with 10 mL of buffer (85 mM phosphate, pH 7) and a 1.0 mg/mL solution of the antibody to be immobilized is injected. The surface is incubated with the antibody solution for 30 min and is then washed with another 10 mL of phosphate buffer. Unreacted sites on the surface are then blocked by reaction with a 10 mg/mL solution of bovine serum albumin. After another buffer wash, immunochemical reactions may be performed; typical incubation times for antigen and conjugate solutions range from 5 to 30 min. All SPR curves were acquired in buffer to eliminate slight index of refraction differences that may exist between protein solutions. RESULTS AND DISCUSSION We have previously shown that two-dimensional arrays of colloidal Au nanoparticles11c,15 offer significant advantages relative to bulk Au films with respect to substrate reproducibility and biocompatibility.16 Accordingly, Au colloid-modified surfaces were evaluated for use in SPR biosensors. Figure 1 illustrates that immobilization of colloidal Au particles onto a ∼50-nm-thick evaporated Au film results in a significant perturbation of the SPR response. The SPR curve of the bare Au film is characterized by a sharp plasmon minimum at an angle of 43.7°. Exposure of a 2-mercaptoethylamine-derivatized Au film to a 17 nM sol of 11nm-diameter colloidal Au for 1 min results in a significant increase (15) Freeman, R. G.; Grabar, K. C.; Allison, K. J.; Bright, R. M.; Davis, J. A.; Guthrie, A. P.; Hommer, M. B.; Jackson, M. A.; Smith, P. C.; Walter, D. G.; Natan, M. J. Science 1995, 267, 1629-1632. (16) (a) Grabar, K. C.; Smith, P. C.; Musick, M. D.; Davis, J. A.; Walter, D. G.; Jackson, M. A.; Guthrie, A. P.; Natan, M. J. J. Am. Chem. Soc. 1996, 118, 1148-1153. (b) Brown, K. R.; Fox, A. P.; Natan, M. J. J. Am. Chem. Soc. 1996, 118, 1154-1157. (c) Brown, K. R.; Keating, C. D.; Grabar, K. C.; Smith, P. C.; Botha, G. H.; Natan, M. J. In Surface Modification of Polymeric Biomaterials; Ratner, B. D., Castner, D. G., Eds.; Plenum Press: New York, 1996; pp 193-201.

Analytical Chemistry, Vol. 70, No. 24, December 15, 1998

5179

in the minimum reflectance, a broadening of the curve, and a 0.2° shift in the plasmon angle. Longer exposure times served to increase the magnitude of these changes; a 60-min exposure resulted in a plasmon angle that had shifted to 46.3° while the minimum reflectance had increased by nearly 40% over that of the bare Au film. These results resemble the dramatic shifts induced by thin carbon coatings evaporated onto Ag surfaces, where the strong absorptivity of the film induced a significant degree of plasmon damping.8a It seems likely that a similar absorptive damping process is responsible for the SPR changes due to colloid adsorption, as these particles possess a nonzero imaginary dielectric component. A second possible mechanism for damping involves the coupling of the localized surface plasmon of the colloid with the propagating plasmon in the Au film, a situation that is similar to effects observed at roughened Au films8c and other particle-modified films.17 While the results presented here do not allow us to distinguish between these two mechanisms, it is reasonable to assume that both occur to some degree. Current efforts are focused on the elucidation of these complex particlesurface interactions. The magnitude of this effect is further illustrated in the inset to Figure 1. When the time course of the reflectance changes is monitored at the reflectance minimum, a >1% reflectance change is easily observed in the first 100 ms of colloid adsorption. Using the previously described kinetics of immobilization of 12-nmdiameter colloidal Au, the coverage at this time is roughly 0.1% of a close-packed monolayer.16a Further inspection of the data suggests that the limits of detection could potentially be much lower. As stated above, at 0.1% of a monolayer a 0.2° shift is observed. If we assume that the plasmon angle changes linearly with colloid coverage (a reasonable assumption for low coverages, vida infra), and we consider an instrumental angular resolution limit of 0.005°, a 40-fold (0.2°/0.005°) decrease in colloid coverage should be detectable. Accordingly, 0.0025% of a monolayer of 12nm-diameter colloidal Au (2.0 × 107 particles/cm2) should be observable. An immunochemical molecular recognition event was chosen to illustrate the utility of Au colloid-enhanced biosensing.17,18 As previously demonstrated, surface confinement of one of the two binding partners allows for monitoring of antibody-antigen association by SPR.5-7 Formation of an electrostatic conjugate between the remaining binding partner and 10-nm-diameter colloidal Au results in the introduction of particle enhancement to a traditional immunosensing scheme.10 Two architectures for (17) (a) Takemori, T.; Inoue, M.; Ohtaka, K. J. Phys. Soc. Jpn. 1987, 56, 15871602. (b) Hayashi, S.; Kume, T.; Amano, T.; Yamamoto, K. Jpn. J. Appl. Phys. 1996, 35, L331-L334. (c) Kume, T.; Nakagawa, N.; Hayashi, S.; Yamamoto, K. Solid State Commun. 1995, 93 (2), 171-175. (d) Shchegrov, A. V.; Novikov, I. V.; Maradudin, A. A. Phys. Rev. Lett. 1997, 78, 4269-4272. (e) Holland, W. R.; Hall, D. G. Phys. Rev. B 1983, 27, 7765-7767. (18) A few theoretical17 and experimental studies of colloid-surface interactions as measured by evanescent wave methods have been reported previously. There have been no reports, however, of Au colloid-amplified SPR at welldefined Au surfaces. (a) Leung, P.-T.; Pollard-Knight, D.; Malan, G. P.; Finlan, M. F. Sens. Actuators B 1994, 22, 175-180. (b) Johne, B.; Hansen, K.; Mørk, E.; Holtlund, J. J. Immunol. Methods 1995, 183, 167-174. (c) Buckle, P. E.; Davies, R. J.; Kinning, T.; Yeung, D.; Edwards, P. R.; Pollard-Knight, D. Biosens. Bioelectron. 1993, 8, 355-363. (d) Kubischko, S.; Spinke, J.; Bruckner, T.; Pohl, S.; Oranth, N. Anal. Biochem. 1997, 253, 112-122. (e) Wink, T.; van Zuilen, S. J.; Bult, A.; van Bennekom, W. P. Anal. Chem. 1998, 70, 827-832.

5180 Analytical Chemistry, Vol. 70, No. 24, December 15, 1998

Figure 2. In situ SPR curves of (A) an evaporated Au film modified with a-h-IgG(γ) (s) and then exposed to a 1.0 mg/mL solution of h-IgG (- -). (B) A film modified with a-h-IgG(γ) (s) and then exposed to h-IgG-10-nm Au colloid conjugate (- -).

Chart 1. Particle-Enhanced Biosensing Architectures

particle-enhanced SPR immunosensing are presented in Chart 1. The first involves direct binding of the antigen-Au conjugate to an antibody-derivatized surface, while the second comprises the same antibody-derivatized surface followed by binding of a free antigen and then a secondary antibody-Au conjugate. While the second geometry is more typical of traditional immunoassays, the efficacy of both cases will be discussed. SPR detection of antibody-antigen binding is demonstrated in Figure 2. Exposure of an evaporated Au film coated with γ-chain-specific monoclonal goat anti-human immunoglobulin G (a-h-IgG(γ)) to a 1.0 mg/mL solution of human immunoglobulin G results in a 0.1° shift in plasmon angle with no change in curve shape (Figure 2A). In contrast, a significantly larger shift (1.5°), a ∼2% increase in minimum reflectance, and noticeable broadening of the curve is observed upon exposure of an identically prepared surface to an electrostatically bound conjugate between h-IgG and

Figure 4. Differential SPR spectra obtained by subtraction of the SPR curve due to h-IgG from the curves due to a-h-IgG(Fc) (‚‚‚) and a-h-IgG(Fc)-10-nm Au colloid conjugate (s). The corresponding reflectance vs angle curves are shown in Figure 3.

Figure 3. In situ SPR curves of (A) an evaporated Au film modified with a-h-IgG(γ) (s) followed by sequential exposure to a 0.045 mg/ mL solution of h-IgG (‚‚‚) and a 8.5 mg/mL solution of a-h-IgG(Fc) (- -). (B) A film modified with a-h-IgG(γ) (s) followed by sequential exposure to a 0.045 mg/mL solution of h-IgG (‚‚‚) and a solution of a-h-IgG(Fc)-10-nm Au colloid conjugate (- -).

10-nm-diameter colloidal Au [h-IgG-Au] (Figure 2B). On the basis of plasmon shift alone, this represents a 15-fold increase in SPR sensitivity to the biomolecular interaction event. A similar enhancement is obtained from the sandwich immunoassay format (Figure 3). Incubation of an a-h-IgG(γ)-modified film with a 0.045 mg/mL solution of h-IgG yields a 0.04° shift in plasmon angle (Figure 3A), while further exposure of this film to a 8.4 mg/mL solution of a secondary antibody, Fc-specific monoclonal goat anti-human immunoglobulin G [a-h-IgG(Fc)], results in a small (0.06°) plasmon angle shift. However, replacement of the free antibody with an electrostatic conjugate between 10-nm-diameter Au and a-h-IgG(Fc) [a-h-IgG(Fc)-Au] yields a 1.7° shift in plasmon angle, a 2% increase in minimum reflectance, and a noticeable broadening of the SPR curve (Figure 3B). This colloid-induced 1.7° shift is 28 times larger than that observed for the unamplified assay (0.06°). A significant (28-fold) increase in sensitivity is therefore afforded through the use of proteinAu colloid conjugates in this biosensing scheme. Biochemical specificity is also maintained under these conditions; surfaces modified with a-h-IgG(γ) show little or no reactivity with human serum (minus IgG), human immunoglobulin A, and [Au-a-h-IgG(Fc)] (Supporting Information). The a-h-IgG(γ)+h-IgG-modified surface is also highly specific, as exposure to a conjugate of 10nm-diameter Au with goat anti-human immunoglobulin G1 (ahIgG1-Au), leads to a small (0.08°) plasmon shift (Supporting Information). Differential SPR curves for the data shown in Figure 3 indicate that the sensitivity enhancement is actually larger when calculated in terms of reflectance changes (Figure 4). These curves are obtained by calculating the difference between the SPR curves obtained for the secondary antibody and antigen steps. The peak of the curve occurs at the largest difference between the two spectra and thus represents the optimal angle for monitoring reflectance changes during that reaction. For the case of the addition of free antibody (‚‚‚), a maximum differential signal of

Figure 5. Reflectance time courses corresponding to the secondary antibody reactions presented in Figure 3. Top curve: exposure to a 17 nM solution of a-h-IgG(Fc)-10-nm Au colloid conjugate. Bottom curve: exposure to a 8.5 mg/mL solution of a-h-IgG(Fc).

only ∼2% is obtained while a difference of ∼50% is observed for the addition of conjugate (s), a 25-fold increase in sensitivity. Reflectance time courses for the immobilization of both the free and Au-bound secondary antibodies illustrate this sensitivity enhancement for the case of real-time monitoring of biospecific interactions (Figure 5). When the reaction is monitored at an angle smaller than the plasmon angle (53.4°, 50% reflectance), a change in reflectance of > 40% is observed in the case of the Au conjugate, while immobilization of the uncomplexed antibody to an identically prepared surface results in less than a 100 nm) Analytical Chemistry, Vol. 70, No. 24, December 15, 1998

5181

Figure 6. In situ SPR curves of (A) an evaporated Au film modified with a-h-IgG(γ) (s) followed by sequential exposure to a 0.0045 mg/ mL solution of h-IgG (‚‚‚) and a solution of a-h-IgG(Fc)-10-nm Au colloid conjugate (- -). (B) Expansion of the area about the plasmon minimums for the a-h-IgG(γ)-modified Au surface before (O) and after ([) exposure to a 0.0045 mg/mL solution of h-IgG.

Figure 7. In situ SPR curves of evaporated Au films modified with a-h-IgG(γ) (s) followed by sequential exposure to solutions of h-IgG (- -) and a-h-IgG(Fc)-10-nm Au colloid conjugate (‚‚‚). The panels correspond to solution h-IgG concentrations of (a) 3.0 mM, (b) 0.3 mM, (c) 3.0 nM, and (d) 6.7 pM. In (c) and (d), the curves taken after h-IgG immobilization are indistinguishable from those taken prior to exposure to the antigen solution.

before complete decay, the functional form of this decay is exponential, suggesting that some distance dependence should be observable. The utility of particle enhancement becomes most evident at low concentrations of antigen. As shown in Figure 6, a h-IgG concentration of 4.5 µg/mL induces a plasmon angle shift that is too small to resolve (Figure 6A). This is evident upon further expansion of the region about the minimum (Figure 6B) where no significant difference in plasmon angle can be discerned at a stage resolution of 0.01°. However, the presence of bound antigen is confirmed by the colloid-linked immunoassay, as a significant shift is observed following exposure of the surface to [a-h-IgG5182 Analytical Chemistry, Vol. 70, No. 24, December 15, 1998

Figure 8. Tapping-mode AFM images of evaporated Au surfaces following colloidal Au-amplified sandwich immunoassays of solutions containing (A) 0.45 mg/mL h-IgG and (B) 4.5 ng/mL h-IgG. The scan area is 1 µm × 1 µm.

(Fc)-Au]. This result suggests that colloidal Au enhancement may be useful for the detection of low-molecular-weight (