Colorimetric Detection of Lysozyme Based on Electrostatic Interaction

Their effective surface charges were further evaluated by capillary electrophoresis (CE). A linear calibration curve with the range of 0.1−1 μM was...
0 downloads 0 Views 281KB Size
Download PDF
3654

Langmuir 2008, 24, 3654-3660

Colorimetric Detection of Lysozyme Based on Electrostatic Interaction with Human Serum Albumin-Modified Gold Nanoparticles Yi-Ming Chen,† Cheng-Ju Yu,† Tian-Lu Cheng,‡,§ and Wei-Lung Tseng*,†,§ Department of Chemistry, National Sun Yat-sen UniVersity, Taiwan, Faculty of Biomedical Science and EnVironmental Biology, Kaohsiung Medical UniVersity, Kaohsiung, Taiwan, and National Sun Yat-sen UniVersity-Kaohsiung Medical UniVersity Joint Research Center, Kaohsiung, Taiwan ReceiVed NoVember 6, 2007. In Final Form: December 18, 2007 In this study, an aqueous solution of 13-nm gold nanoparticles (AuNPs) covalently bonded with human serum albumin (HSA) was used for sensing lysozyme (Lys). HSA molecules were good stabilizing agents for AuNPs in high-salt solution and exhibited the ability to bond with Lys electrostatically. The aggregation of HSA-AuNPs was achieved upon the addition of high-pI proteins, such as Lys, R-chymotrypsinogen A, and conalbumin. Not the same was achieved, however, when low-pI proteins such as ovalbumin, bovine serum albumin, and R-lactalbumin were added. Matrix-assisted desorption/ionization mass spectrometry was used to demonstrate the interaction between HSA-AuNPs and Lys. It was found that the sensitivity of HSA-AuNPs for Lys was highly dependent on the HSA concentration. The Lys-induced aggregation of HSA-AuNPs was suggested based on the London-van der Waals attractive force. We further improved the selectivity of the probe by adjusting the pH solution to 8.0. Under the optimum conditions, the selectivity of this system for Lys over other proteins in high-salt solutions was remarkably high, even when their pI was very close to the Lys. The lowest detectable concentration of Lys in this approach was 50 nM. The applicability of the method was validated through the analyses of Lys in chicken egg white.

Introduction Gold nanoparticles (AuNPs) are important colorimetric indicators because they provide large surface areas and distancedependent optical properties. More importantly, they possess high extinction coefficients, which are over 1000 times larger than those of organic dyes.1 Moreover, AuNPs exhibit selective binding toward thiol-containing biomolecules, such as glutathione and cytochrome C.2 When compared with dispersed AuNPs having red color, the aggregation of AuNPs having interparticle distances substantially less than their average particle diameter results in a rapid color change from red to purple. Therefore, the target analyte should be designed as a linker between the dispersed AuNPs, to which the receptor molecules have been covalently bonded. This principle has been applied for detecting duplex DNA formation,3 antibody-antigen recognitions,4 metal ionchelant complexes,5 and protein-carbohydrate interactions.6 On * Corresponding author. Address: Department of Chemistry, National Sun Yat-sen University, 70, Lien-hai Road, Kaohsiung, Taiwan 804. E-mail: [email protected]. Fax: 011-886-7-3684046. † National Sun Yat-sen University. ‡ Kaohsiung Medical University. § National Sun Yat-sen University-Kaohsiung Medical University Joint Research Center. (1) (a) Mulvaney, P. Langmuir 1996, 12, 788-800. (b) Link, S.; Mohamed, M. B.; El-Sayed, M. A. J. Phys. Chem. B 1999, 103, 3073-3077. (c) Kelly, K. L.; Coronado, E.; Zhao, L. L.; Schatz, G. C. J. Phys. Chem. B 2003, 107, 668677. (2) (a) He, X.; Zhong, Z.; Guo, Y.; Lv, J.; Xu, J.; Zhu, M.; Li, Y.; Liu, H.; Wang, S.; Zhu, Y.; Zhu, D. Langmuir 2007, 23, 8815-8819. (b) Chah, S.; Hammond, M. R.; Zare, R. N. Chem. Biol. 2005, 12, 323-328. (3) (a) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J. Nature 1996, 382, 607-609. (b) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078-1081. (c) Reynolds, R. A., III; Mirkin, C. A.; Letsinger, R. L. J. Am. Chem. Soc. 2000, 122, 3795-3796. (4) (a) Jiang, Z.; Sun, S.; Liang, A.; Huang, W.; Qin, A. Clin. Chem. 2006, 52, 1389-1394. (b) Thanh, N. T. K.; Rosenzweig, Z. Anal. Chem. 2002, 74, 1624-1628. (c) Hirsch, L. R.; Jackson, J. B.; Lee, A.; Halas, N. J.; West, J. L. Anal. Chem. 2003, 75, 2377-2381. (5) (a) Huang, C. C.; Yang, Z.; Lee, K. H.; Chang, H. T. Angew. Chem., Int. Ed. 2007, 46, 6824-6828. (b) Kim, Y.; Johnson, R. C.; Hupp, J. T. Nano Lett. 2001, 1, 165-167. (c) Lin, S.-Y.; Liu, S.-W.; Lin, C.-M.; Chen, C.-h. Anal. Chem. 2002, 74, 330-335.

the other hand, the aggregation of AuNPs can be induced by the London-van der Waals attractive force between the nanoparticles, while the surface charges of receptor-modified AuNPs become neutral upon the addition of target analyte.7 Under an identical mechanism, the stabilized molecules desorbed from the surface of AuNPs also lead to an unstable colloidal solution.8 Recently, without the presence of a linker DNA, the aggregation of probe DNA-modified AuNPs occurs in high-salt solution when the target DNA is fully complementary to the probe.7a In addition, an aqueous solution of 13-nm AuNPs, whose nonionic surfactant has been adsorbed noncovalently, is used for the sensing of aminothiols under high-salt conditions.8a The concentration of salt plays an important role in the determination of the rate of the aggregation process because the electrostatic interaction could be effectively screened. Moreover, the analyte-induced aggregation of AuNPs is performed based on the electrostatic and hydrogen-bonding interaction without molecular recognition.9 The amino acids with two amino groups, such as lysine and histidine, show strong electrostatic interaction with citrate-capped AuNPs. The AuNPs attached by aminothiols aggregate as a result of the interparticle interaction between their amino acid groups.10 The network of citrate-capped AuNPs and bacteriophages assembled through opposite-charge interaction are employed as probes to investigate peptide ligand binding to receptors on the cell surface.11 The difference in electrostatic properties for single-strand and double(6) de la Fuente, J. M.; Penade´s, S. Biochim. Biophys. Acta 2006, 1760, 636651. (7) (a) Sato, K.; Hosokawa, K.; Maeda, M. J. Am. Chem. Soc. 2003, 125, 8102-8103. (b) Zhao, W.; Chiuman, W.; Lam, J. C. F.; Brook, M. A.; Li, Y. Chem. Commun. 2007, 3729-3731. (8) (a) Lu, C.; Zu, Y.; Yam,V. W.-W. Anal. Chem. 2007, 79, 666-672. (b) Lu, C.; Zu, Y.; Yam, V. W.-W. J. Chromatogr. A 2007, 1163, 328-332. (9) Selvakannan, P.; Mandal, S.; Phadtare, S.; Pasricha, R.; Sastry, M. Langmuir 2003, 19, 3545-3549. (10) (a) Huang, Y.-F.; Chang, H.-T. Anal. Chem. 2006, 78, 1485-1493. (b) Zhang, F. X.; Han, L.; Israel, L. B.; Daras, J. G.; Maye, M. M.; Ly, N. K.; Zhong, C-J. Analyst 2002, 127, 462-465. (c) Lim, I. S.; Ip, W.; Crew, E.; Njoki, P. N.; Mott, D.; Zhong, C.-J.; Pan, Y.; Zhou, S. Langmuir 2007, 23, 826-833.

10.1021/la7034642 CCC: $40.75 © 2008 American Chemical Society Published on Web 02/16/2008

Colorimetric Detection of Lysozyme Using HSA-AuNPs

strand DNA interacted with citrate-capped AuNPs can be conducted in the analysis of single-base-pair mismatches.12 However, there is still a problem in the selectivity of this type of nanosensor. Lysozyme (Lys) is a high-pI enzyme that hydrolyzes the polysaccharide wall of bacteria so that it is widely distributed in body tissues and secretions. Nevertheless, low concentrations of Lys are found in urine and serum. It has been discovered that increased Lys concentrations in urine and serum is associated with leukemia,13 renal diseases,14 and meningitis.15 Additionally, Lys obtained from chicken egg white is commonly used as an antimicrobial agent in the food industry.16 Many foods, such as fresh vegetables, cheese, and wine have been preserved by coating the surface of the food with Lys or adding Lys to the food.17 Thus, Lys must be declared on the label accordingly. The majority of the reported methods for the analysis of Lys are turbidimetric kinetic18 and lysoplate methods.19 The quantification of Lys is often accomplished by separation techniques such as polyacrylamide gel electrophoresis20 and high-performance liquid chromatography,21 as well as immunoassay, such as immunoelectrophoresis22 and enzyme-linked immunosorbent assay.23 Recently, anti-Lys apatmer has been developed and has been used as a biosensor for the detection of Lys based on the different detection techniques.24 In contrast to these studies, we report a simple approach for the selective detection of Lys by human serum albumin (HSA)capped AuNPs (HSA-AuNPs) based on electrostatic interaction. The Lys is first bonded to neutralize the HSA-AuNPs, resulting in non-cross-linking aggregation driven by the London-van der Waals attractive force. The selectivity of HSA-AuNPs toward Lys can be effectively improved by increasing the pH solution and salt concentration. The interaction between HSA-AuNPs and Lys in high-salt solution was demonstrated by matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). Their effective surface charges were further evaluated by capillary electrophoresis (CE). A linear calibration curve with the range of 0.1-1 µM was constructed by monitoring the increases in the ratio of absorption at 600-530 nm. To demonstrate its practicality, the present approach was applied to the determination of Lys in chicken eggs. (11) Souza, G. R.; Christianson, D. R.; Staquicini, F. I.; Ozawa, M. G.; Snyder, E. Y.; Sidman, R. L.; Miller, J. H.; Arap, W.; Pasqualini, R. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 1215-1220. (12) Li, H.; Rothberg, L. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 1403614039. (13) Levinson, S. S.; Elin, R. J.; Yam, L. Clin. Chem. 2002, 48, 11311132. (14) Harrison, J. F.; Lunt, G. S.; Scott, P.; Blainey, J. D. Lancet 1968, 1, 371-375. (15) Klockars, M.; Reitamo, S.; Weber, T.; Kerttula, Y. Acta Med. Scand. 1978, 203, 71-74. (16) Chang, H.-M.; Yang, C.-C.; Chang, Y.-C. J. Agric. Food Chem. 2000, 48, 161-164. (17) Mine, Y.; Ma, F.; Lauriau, S. J. Agric. Food Chem. 2004, 52, 10881094. (18) (a) Houser, M. T. Clin. Chem. 1983, 29, 1488-1493. (b) Liao, Y.-H.; Brown, M. B.; Martin, G. P. J. Pharm. Pharmacol. 2001, 53, 549-554. (19) (a) Jenzano, J. W.; Hogan, S. L.; Lundblad, R. L. J. Clin. Microbiol. 1986, 24, 963-967. (b) Gupta, D. K.; von Figura, K.; Hasilik, A. Clin. Chim. Acta 1987, 165, 73-82. (20) Weth, F.; Schroeder, T.; Buxtorf, U. P. Z. Lebensm.-Unters.-Forsch. 1988, 187, 541-545. (21) Liao, Y. H.; Brown, M. B.; Martin, G. P. J. Pharm. Pharmacol. 2001, 53, 549-554. (22) Johansson, B. G.; Malmquist, J. Scand. J. Clin. Lab. InVest. 1971, 27, 255-261. (23) Taylor, D. C.; Cripps, A. W.; Clancy, R. L. J. Immunol. Methods 1992, 146, 55-61. (24) (a) Cheng, A. K. H.; Ge, B.; Yu, H.-Z. Anal. Chem. 2007, 79, 51585164. (b) Kawde, A. N.; Rodriguez, M. C.; Lee, T. M. H.; Wang, J. Electrochem. Commun. 2005, 7, 537-540. (c) Rodriguez, M. C.; Kawde, A. N.; Wang, J. Chem. Commun. 2005, 4267-4269.

Langmuir, Vol. 24, No. 7, 2008 3655 Table 1. Physical Properties of Proteins protein

source

abbreviation

human serum albumin lysozyme ribonuclease A cytochrome C trypsinogen R-chymotrypsinogen A myoglobin hemoglobin conalbumin R-lactalbumin ovalbumin bovine serum albumin β-casein glucose oxidase insulin

human plasma chicken egg white bovine pancreas bovine heart bovine pancreas bovine pancreas horse heart bovine blood chicken egg white bovine milk chicken egg white bovine plasma bovine milk Aspergillus niger bovine insulin

HSA Lys Rib Cyt Try R-Chy Myo Hem Con R-Lac Ova BSA β-cas GO Ins

a

MWa

pIa

36 700 4.7 14 300 11.1 13 700 9.3 12 400 10.2 24 000 9.3 25 000 9.2 17 500 6.8 64 500 6.8 78 000 6.0 14 200 4.8 45 000 4.7 69 000 4.7 23 500 5.1 160 000 4.2 5 700 5.5

Reference 35.

Experimental Section Chemicals. All proteins were obtained from Sigma (Louis, MO), and their physical properties are listed in Table 1. Sinapic acid was also obtained from Sigma and prepared in analytical-grade acetonitrile, which was obtained from J.T. Baker (Phillipsburg, NJ). Hydrogen tetrachloroaurate (III) dehydrate, trisodium citrate, N,Ndimethylformamide, sodium chloride, sodium chloride, glycine, H3PO4, NaH2PO4, Na2HPO4, and Na3PO4 were purchased from Aldrich (Milwaukee, WI). Dextran sulfate (MW ) 500 000) was purchased from Fluka (Buchs, Switzerland). Buffer solutions are 20 mM Na2HPO4 and 10 mM glycine. The pH of the solution was adjusted by adding Na3PO4. Milli-Q ultrapure water was used in all of the experiments. Characterization of AuNPs. A double-beam UV-visible spectrophotometer (Cintra 10e, GBC Scientific Equipment Pty Ltd., Dandenong, Victoria, Australia) was used for the measurement of absorption spectra of the AuNPs. A H7100 transmission electron microscope (TEM) (Hitachi High-Technologies Corp., Tokyo, Japan) operating at 75 keV was used to collect TEM images of as-prepared AuNPs. Capillary Electrophoresis. A commercial UV-vis absorbance detector (ECOM, Germany) was used to detect HSA and HSAAuNPs while the detection wavelength was set at 220 nm. A 0.02% N,N-dimethylformamide was used as the electroosmotic flow marker. Electrophoresis was driven by a high-voltage power supply (Bertan, Hicksville, NY). The high-voltage end of the separation system was put in a laboratory-made plexiglass box for safety. Data acquisition (10 Hz) and control were conducted by the use of DataApex Software (DataApex, Prague, Czech Republic). The fused-silica capillaries (Polymicro Technologies, Phoenix, AZ) were 60-cm long and had an i.d. of 75-µm (effective length: 40 cm). Prior to analysis, capillaries were filled with a solution of 1% dextran sulfate, which was prepared in 10 mM phosphate buffer with a pH range of 8.0-12.0. Note that the adsorption of protein on the capillary wall could be suppressed upon the addition of dextran sulfate to the background electrolyte.25 The capillary was equilibrated with separation buffer for 20 min at 250 V/cm before the first injection. Subsequently, samples were injected by hydrodynamic injection at 20-cm height for 20 s. All separations were performed at 250 V/cm. Lysozyme Capture. The protein (500 µL) was added to different concentrations (37.4-1122.0 nM; 500 µL each) of HSA-AuNPs solutions and equilibrated for 1 h. The resulting mixture was then subjected to centrifugation at 14000 rpm for 20 min. The obtained supernatants and precipitate were mixed with the 2% sinapic acid (prepared in 70% acetonitrile), respectively, and then pipetted into the wells of the steel plate and dried in air at room-temperature prior to MALDI-MS. Note that the sinapic acid was used as a comatrix for MALDI-MS analysis. MS experiments were performed in the positive-ion mode on a reflectron-type time-of-flight (TOF) mass (25) Stathakis, C.; Arriaga, E. A.; Lewis, D. F.; Dovichi, N. J. J. Chromatogr. A 1998, 817, 227-232.

3656 Langmuir, Vol. 24, No. 7, 2008

Chen et al.

Scheme 1. Schematic Representation of the Aggregation of HSA-AuNPs in the Presence of Lys

spectrometer (Autoflex, Bruker) equipped with a 3-m flight tube. Desorption/ionization was obtained by using a 337-nm nitrogen laser with a 3 ns pulse width. The available accelerating voltages existed in the range from +20 to -20 kV. Each mass spectrum was generated by averaging 300 laser pulses. Nanoparticle Synthesis. The citrate-capped AuNPs were prepared by the chemical reduction of metal salt precursor (hydrogen tetrachloroaurate, HAuCl4) in a liquid phase.26 Briefly, 200 mM HAuCl4 (250 µL) was added rapidly to a solution of 4 mM sodium citrate (50 mL) that was heated under reflux. Heating under reflux was continued for an additional 15 min, during which time the color changed to deep red. The TEM images (not shown) confirmed that the diameter of the AuNPs was 12.1 ( 0.6 nm. To estimate the concentration of the nanoparticles, we assumed that the reduction from gold (III) to gold atoms was 100% complete. The concentration of the original citrate-capped AuNPs was ∼18.7 nM (1.13×1013 particles/mL). A solution of HSA-AuNPs (50 mL) was obtained when 500 µL of 100 µM HSA was added into 18.7 nM citratecapped AuNPs. The resulting mixture was equilibrated at 4 °C overnight and then subjected to two repeated cycles of centrifugation/ washing cycles to remove any excess. It should be noted that the adsorbed HSA molecules were present at 30 molecules per 12-nm diameter AuNP. Centrifugation was conducted at 14 000 rpm for 20 min, and phosphate buffer (45 mL) was used to redisperse the sediment. Note that the phosphate buffer contained 100 mM NaCl. Sample Preparation. A stock solution of protein (1 mM) was prepared in deionized water and diluted with phosphate buffer as necessary. Different concentrations of proteins (500 µL) were added to a solution (500 µL) of (a) 37.4 nM HSA-AuNPs, 100 mM NaCl, and 20 mM phosphate or (b) 37.4 nM citrate-capped AuNPs and 10 mM glycine. The resulting mixture was equilibrated for the optimal incubation time. Subsequently, UV-vis absorption spectra of the solution were recorded. Chicken eggs were obtained from the local supermarket. The egg white and egg yolk were separated. The egg white was diluted with 20 mM phosphate buffer (pH 8.0) in a 1:20 ratio before use. The as-prepared samples (500 µL) were added to a solution (500 µL) of 37.4 nM HSA-AuNPs, 100 mM NaCl, and 20 mM phosphate (pH 8.0) and were then equilibrated for 40 min. Subsequently, the concentration of Lys in egg white was determined by both UV-vis absorption spectra and capillary electrophoresis with UV absorbance.

) 10.3).27 The reconstituted myoglobin was designed to bind cationic Cyt by means of electrostatic interaction. To begin with, it is well-known that Lys exhibits the ability to electrostatically bind HSA and other negatively charged proteins.28 To design a colorimetric detection of Lys (pI ) 11.0) based on electrostatic

Results and Discussion

Figure 1. (A) UV-vis absorption spectra of HSA-AuNPs in the (a) absence and (b) presence of 10 µM Lys. Buffer: 20 mM phosphate solution containing 100 mM NaCl, pH 7.0. (B) Time-course measurement of change in the SPR peaks of HSA-AuNPs upon the addition of 10 µM Lys. The concentration of AuNPs is 18.7 nM. (C) TEM images of solution containing HSA-AuNPs in the (a) absence and (b) presence of 10 µM Lys.

The Mechanism of Lys-Induced Aggregation of HSAAuNPs. The protein-protein electrostatic interaction was shown in the case of cytochrome C peroxidase (pI ) 5.3) and Cyt (pI (26) Lee, P. C.; Meisel, D. J. Phys. Chem. 1982, 86, 3391-3395.

Colorimetric Detection of Lysozyme Using HSA-AuNPs

Langmuir, Vol. 24, No. 7, 2008 3657

Figure 2. Mass spectra of the supernatant obtained from a mixture of (A) 0, (B) 18.7, (C) 187.0, and (d) 561.0 nM HSA-capped AuNPs and 10 µM Lys. Supernatant obtained after centrifuging the mixture at 14 000 rpm for 20 min. A total of 200 pulsed laser shots were applied under a laser fluence set at 46.9 µJ.

interactions with protein-capped AuNPs, a low pI (4.9) of HSA was chosen for this study for the following reasons: (a) HSA with one free cysteine residue could be spontaneously attached onto Au surfaces by means of the formation of Au-S bonds29 and (b) HSA-AuNPs are very stable, even in the presence of high salt concentration. Although it is expected to provide poor selectivity toward Lys with HSA-AuNPs, this problem could be addressed by controlling the solution pH and the salt concentration. For example, 0.65% dextran sulfate was routinely used to selectively precipitate high-density lipoproteins in the presence of 0.2 M MnCl2.30 The HSA-AuNPs changed color in the presence of Lys, as shown in Scheme 1. Prior to the binding of Lys, the (27) Bayraktar, H.; Ghosh, P. S.; Rotello, V. M.; Knapp, M. J. Chem. Commun. 2006, 1390-1392. (28) (a) Moon, Y. U.; Curtis, R. A.; Anderson, C. O.; Blanch, H. W.; Prausnitz, J. M. J. Solution Chem. 2000, 29, 699-718. (b) Miller, S. M.; Kato, A.; Nakai, S. J. Agric. Food Chem. 1982, 30, 1127-1132. (29) Wang, A.; Wu, C.-J.; Chen, S.-H. J. Proteome Res. 2006, 5, 1488-1492.

HSA-AuNPs tended to be negatively charged at neutral pH. A small difference in electrophoretic mobility between HSA and HSA-AuNPs was observed under neutral pH, as determined by capillary electrophoresis (Supporting Information, Figure S1). This result provided clear evidence that HSA-AuNPs still had negatively charged residues when the HSA molecules were absorbed on the surface of nanoparticles. On the contrary, upon the binding of Lys, the surface charges of HSA-AuNPs became neutral and resulted in the aggregation of AuNPs via the Londonvan der Waals attractive force. Another reason governing the aggregation of AuNPs might be that Lys molecules acted as bridges that linked HSA-AuNPs together. The Interaction of Lys with HSA-AuNPs. To test our concept, a series of AuNP solutions were prepared in the presence and absence of Lys. Compared with HSA-AuNPs prepared in (30) Burstein, M.; Scholnick, H. R.; Morfin, R. J. Lipid Res. 1970, 11, 583595.

3658 Langmuir, Vol. 24, No. 7, 2008

Figure 3. (A) UV-vis absorption spectra of AuNPs with (a) 1, (b) 3, and (c) 10 µM HSA as capping agents upon the addition of 10 µM Lys. (B) Effect of the HSA concentration on the values of A600nm/525nm ratios for AuNPs.

Chen et al.

deionized water, spectrum a in Figure 1A displays that the preparation of HSA-AuNPs in 100 mM NaCl that resulted in an ∼4 nm red-shift of the surface plasmon resonance (SPR) band. The main reason for this is a change in the refractive index of the surrounding medium.31 Upon the addition of Lys (10 µM), the SPR band of HSA-AuNPs shifted dramatically to the red in the presence of 100 mM NaCl, as depicted in spectrum b (Figure 1A). This result suggests that the presence of a high salt concentration does not affect the interaction between HSA-AuNPs and Lys. Note that the presence of 300 mM NaCl hinders the interaction of HSA-AuNPs with Lys. Inset A in Figure 1A indicates a visible color change from red to deep purple after the addition of 10 µM Lys to the solution of HSA-AuNPs and 100 mM NaCl. Figure 1B suggests that this reaction reached completion within 40 min when monitoring the absorption band at 548 nm. Evidence of Lys-induced aggregation can be seen in the TEM image (Figure 1C). The interaction between HSAAuNPs and Lys was also examined using the MALDI-MS analysis. Under neutral conditions (pH 7.0), different concentrations of HSA-AuNPs were employed to trap opposite charged Lys from 100 mM NaCl solution. The resulting mixtures were then subjected to centrifugation, and the obtained supernatants were detected by MALDI-MS using an organic matrix. A decrease in signal intensities for [Lys + H]+ ions was found upon the addition of an increasing concentration of HSA-AuNPs from 18.7 to 561.0 nM (Figure 2A-D). This result demonstrates that the binding occurred between HSA-AuNPs and Lys under neutral conditions. To further understand the aggregation mechanism, the effect of the concentration of the capping agent (HSA) was tested in this study. At an HSA concentration of 1 µM, fast aggregation

Figure 4. Values of (A) A587nm/520nm ratios for citrate-capped AuNPs at pH 7.0, (B) A570nm/525nm ratios for HSA-AuNPs at pH 7.0, and (C) A600nm/525nm ratios for HSA-AuNPs at pH 8.0 upon the addition of 10 µM proteins. Buffer: (A) 10 mM glycine, pH 7.0, and (B, C) 20 mM phosphate solution containing 100 mM NaCl. (D) Photographs of HSA-AuNPs with added 10 µM proteins. From left to right: no analyte, Lys, Rib, Cyt, Try, R-Chy, Myo, Hem, and Con. The concentration of AuNPs is 18.7 nM. The proteins were incubated with the AuNPs for 40 min.

Colorimetric Detection of Lysozyme Using HSA-AuNPs

of HSA-AuNPs was observed in the presence of 10 µM Lys and reached the completion within 40 min (spectrum a in Figure 3A). It should be noted that the addition of 0.5 µM to the solution of AuNPs made the nanoparticles very unstable in high-salt solution. When the HSA concentration was increased to 5 µM, less aggregation of HSA-AuNPs was found upon the addition of the same concentration of Lys (Figure 3B). Furthermore, no obvious aggregation was observed at an HSA concentration of above 10 µM. It is suggested that the surface charges of the AuNPs using low-concentration HSA as a capping agent could be effectively screened in the presence of Lys, resulting in fast aggregation kinetics. However, once the HSA concentration went above 10 µM, the electrostatic repulsion force between HSA-AuNPs still existed, even in the presence of Lys. Therefore, upon the binding of Lys, the HSA-AuNPs became neutral and then aggregated through the London-van der Waals attraction force. The Selectivity of HSA-AuNPs. To assess the selectivity of the citrate-capped AuNPs (citrate-AuNPs) and HSA-AuNPs toward Lys, we studied the changes in their absorption spectra that occurred within 40 min of adding some possible interference proteins, including Rib, Cyt, Try, R-Chy, Myo, Hem, Con, R-Lac, TI, Ova, BSA, β-Cas, GO, and Ins. In the case of citrate-AuNPs, we found that the presence of Rib, Cyt, Try, R-Chy, Myo, and Hem resulted in significant increases in the ratio of absorption at 587 to 520 nm while the low-pI proteins exhibited no effect at pH 7.0 (Figure 4A). It was not surprising that low-pI proteins did not bind to HSA-AuNPs because of the electrostatic repulsion force between them. In comparison, the use of HSA as capping agents provided a significant improvement of the selectivity of the AuNP probe toward Lys when monitoring the ratio of adsorption at the plasmon peak (525 nm) and at 570 nm (Figure 4B). Except for R-Chy, Hem, and Con, the high-pI proteins such as Cyt and Try that might have electrostatic interactions with the negatively charged HSA-AuNPs did not cause interference, further indicating the specificity of the present probe. Note that Cyt and Lys possess almost the same molecular weight and a similar ratio of basic and acidic residues on their surfaces.32 The observation shows that the pI of protein is not the only parameter affecting the ability of protein solution to initiate the aggregation. Additionally, it is found that the electrophoretic mobility of Lys (1.1 × 10-4 cm2 V-1 s-1) is larger than that of Cyt (0.8 × 10-4 cm2 V-1 s-1) at pH 7.0. This result also supported the idea that the pI is not the only parameter for evaluating the protein’s charge. To further overcome the interference of R-Chy, Hem, and Con, we performed a series of binding experiments in 20 mM phosphate buffer containing 100 mM NaCl, with the pH ranging from 8.0 to 12.0. Figure 4C demonstrates that HSAAuNPs provide high specificity toward Lys by means of adjusting the pH solution to 8.0. The relative responses to the absorption ratio (A600nm/525nm) of the interfering proteins are negligible when compared to those from Lys. However, no aggregation of HSAAuNPs induced by Lys was observed above a pH of 10.0 since the solution pH is very close to the pI of Lys. On the basis of these results, we suggest that the use of HSA as a capping agent is necessary to improve the selectivity of AuNPs toward Lys and the optimal pH condition at 8.0. Sensitivity and Application. As indicated in Figure 5A, the intensity of the absorbance signal was sensitive to Lys and increased at 600 nm as the concentration of Lys increased. When plotting absorption ratios (A600nm/525nm) against the Lys concentration, it is found that the correlation coefficients (R2) were (31) McFarland, A. D.; Van Duyne, R. P. Nano Lett. 2003, 3, 1057-1062. (32) Renner, C.; Piehler, J.; Schrader, T. J. Am. Chem. Soc. 2006, 128, 620628.

Langmuir, Vol. 24, No. 7, 2008 3659

Figure 5. (A) Absorption spectral changes of the HSA-AuNPs (18.7 nM) upon the addition of Lys concentration within the range of 0.1-1 µM. (B) Plot of A600nm/525nm ratios of the HSA-AuNPs as a function of the Lys concentration. Buffer: 20 mM phosphate solution containing 100 mM NaCl, pH 8.0. The sample was incubated with the AuNPs for 40 min.

0.9890 for the determination of Lys in the concentration ranges of 100-1000 nM, as shown in Figure 5B. The calibration curve should be reconstructed when a new stock solution of AuNPs was prepared. Using this proposed method, the lowest detectable concentration of Lys is 50 nM, which is lower than the concentration (700 µg/mL; 49 µM) of Lys in chicken egg white.33 Having high sensitivity and selectivity, we evaluated the feasibility of an HSA-AuNP sensor for the determination of Lys in chicken egg white. It is noted that chicken egg white contains 54% Ova and 12% Con, which was shown to not interact with HSAAuNPs. The others are ovomucoid (11%) and ovomucin (3.5%), and ovoinhibitor (1.5%), which are low-pI proteins.34Although the total concentrations of Ova and Con are also high, it is expected that the determination of Lys (3.5%) in chicken egg white could be carried out in the presence of HSA-AuNPs, which are prepared in a solution of 20 mM phosphate and 100 mM NaCl (pH 8.0). We observed an apparent increase in the absorption ratios (A600nm/525nm) of HSA-AuNPs after the concentration of Lys spiked into the samples over the range of 0.1-1 µM (Figure 6A). By using a standard addition method, we estimated that the concentration of Lys in chicken egg white was 68.1 ( 2.2 µM (n ) 3) (Figure 6B). The result was further confirmed by capillary electrophoresis with UV absorbance. Consequently, we determined the concentration of Lys in chicken egg white to be 50.7 ( 3.0 µM (n ) 3) by using a standard addition method (Supporting (33) Kvasnicˇka, F. Electrophoresis 2003, 24, 860-864. (34) Hsieh, Y.-L.; Chen, T.-H.; Liu, C.-Y. Electrophoresis 2006, 27, 42884294. (35) (a) Gao, J.; Whitesides, G. M. Anal. Chem. 1997, 69, 575-580. (b) Krylov, S. N.; Dovichi, N. J. Anal. Chem. 2000, 72, 111R-1128R. (c) Huang, Y.-F.; Hsieh, M.-M.; Tseng, W.-L.; Chang, H.-T. J. Proteome Res. 2006, 5, 429-436.

3660 Langmuir, Vol. 24, No. 7, 2008

Chen et al.

good agreement with those from capillary electrophoresis. Additionally, the result provided clear evidence that the high concentrations of interfering proteins (such as Con and Ova) did not affect the selectivity of HSA-AuNPs for Lys.

Conclusions In this study, we describe simple assays based on the aggregation for the selective detection of Lys using HSA-AuNPs. We found that HSA is an effective capping agent to stabilize nanoparticles in high-salt solution and has electrostatic interaction with Lys, as demonstrated by the MALDI-MS analysis. When HSA molecules adsorbed on the AuNPs interacted with Lys, the surface charges of nanoparticles became neutral and resulted in the non-crosslinking aggregation. Our study also highlighted that the pH solution and the HSA concentration are both important parameters in determining the sensitivity and specificity of nanoparticle-based probes. In 20 mM phosphate containing 100 mM NaCl and 18.7 nM AuNPs (pH 8.0), the HSA-AuNP probe specifically and sensitively detected Lys with the lowest detectable concentration of 50 nM. In addition, the concentration of Lys (68.1 ( 2.2 µM) in the chicken egg white was estimated by our proposed method. Future efforts could be focused on developing a new sample preparation method before the MALDI-MS analysis. For example, the high-abundance Lys in tear samples could be removed after the pretreatment of HSA-AuNPs.

Figure 6. Colorimetric detection of Lys in chicken eggs using HSAAuNPs. The samples were spiked (a) without and (b) with 1 µM of Lys. Buffer: 20 mM phosphate solution containing 100 mM NaCl, pH 8.0. The sample is incubated with the AuNPs for 40 min. (B) Calibration curve for the detection of Lys in chicken eggs.

Information, Figure S2). On the basis of the F-test (95% confidence level), the results from our present approach are in

Acknowledgment. We would like to thank the National Science Council (NSC 96-2113-M-110-008-) and National Sun Yat-sen University-Kaohsiung Medical University Joint Research Center for the financial support of this study. Supporting Information Available: Graphs showing (1) the effect of pH on the electrophoretic mobility of HSA-AuNPs and HSA, and (2) the separation of egg-white proteins by capillary electrophoresis. This material is available free of charge via the Internet at http://pubs. acs.org. LA7034642