Coupling Reaction-Based Ultrasensitive Detection of Phenolic ...

ARTICLE pubs.acs.org/ac

Coupling Reaction-Based Ultrasensitive Detection of Phenolic Estrogens Using Surface-Enhanced Resonance Raman Scattering Xiao Xia Han,*,† Prompong Pienpinijtham,‡ Bing Zhao,§ and Yukihiro Ozaki*,† †

Department of Chemistry and Research Center for Single Molecule Vibrational Spectroscopy, School of Science and Technology, Kwansei Gakuin University, Sanda, Hyogo 669-1337, Japan ‡ Sensor Research Unit, Department of Chemistry, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand § State Key Laboratory of Supramolecular Structure and Materials, Jilin University, Changchun 130012, People’s Republic of China

bS Supporting Information ABSTRACT: Studies have shown that many adverse health effects are associated with human exposure to dietary or environmental estrogens. Therefore, the development of rapid and highly sensitive detection methods for estrogens is very important and necessary to maintain hormonal concentration below the safety limit. Herein, we demonstrate a simple and rapid approach to detect trace amounts of phenolic estrogen based on surface-enhanced resonance Raman scattering (SERRS). Because of a coupling reaction between diazonium ions and the phenolic estrogens, azo compounds are formed with strong SERRS activity, which allows phenolic estrogen recognition at subnanomolar levels in solution. The proposed protocol has multiplexing capability, because each SERRS fingerprint of the azo dyes specifically corresponds to the related estrogen. Moreover, it is universal and highly selective, not only for phenolic estrogens but also for other phenolic molecules, even in complex systems.

E

strogens are a broad class of compounds functioning as primary female sex hormones that exert many physiological effects. The action of estrogens is mediated by estrogen receptors, nuclear proteins that bind to DNA, and control gene expression.1 Natural estrogens are steroid hormones, while some synthetic or estrogen-like hormones are nonsteroid.2 Increasing evidence has indicated that many adverse health consequences are associated with relatively high exposures to environmental estrogens and estrogen-like compounds,3 and the link between some diseases (e.g., breast cancer) and hormones, estrogens in particular, has been a source of great concern among scientists.4,5 In 2002, a clinical trial of estrogen plus progestin treatment therapy was terminated, because of an increased risk of coronary heart disease, breast cancer, and stroke.6 Thus, the development of rapid and highly sensitive detection methods for estrogens is very important and necessary to control hormonal concentration below safety standards. Current direct-detection approaches for estrogens, such as high-performance liquid chromatography (HPLC) with tandem quadruple mass spectrometry7 and liquid chromatography mass spectrometry (LC-MS),8 are rather complicated and timeconsuming for sample pretreatment. Some indirect methods with high sensitivity, such as enzyme-linked immunosorbent assay,9 immunochip,10 high-performance liquid chromatography-radioimmunoassay (HPLC-RIA),11 and plasmon resonance biosensors,12 have also been developed to detect estrogens. However, these immunochemical methods require complicated antibody preparation and cannot always discriminate between specific and nonspecific binding,13 especially when the estrogen r 2011 American Chemical Society

concentration is very low, which may lead to false results and thus restrain their practical applications. Therefore, developing a detection technique that allows simple, rapid, selective, and sensitive estrogen detection is very important and challenging. Surface-enhanced Raman scattering (SERS) is ultrasensitive down to the single-molecule level, under favorable circumstances, and is capable of identifying multiple components in a mixture without separation.14 18 Ultrasensitive detection methods for some small molecules19 22 and biological samples23 27 have been successfully developed using SERS. For SERS-based estrogen detection, a major problem is that the affinity of these phenolic molecules to silver or gold surfaces is rather weak,28 30 which results in poor SERS signals when mixing the analytes with silver nanoparticles (AgNPs) directly. There is a coupling reaction between phenol and diazonium ions, producing products known as azo dyes. Thus, Pauly’s reagent is often used for tyrosine detection, based on the optical absorption of the products, azo compounds, which have been proven to have strong Raman activity and a propensity toward binding to AgNPs.31 Coupled with surface-enhanced resonance Raman scattering (SERRS), these azo dyes derived from phenolic estrogens would make it possible for ultrasensitive phenolic estrogen detection with multiplexing capability, because different SERRS “fingerprint” spectra of azo dyes correspond to different phenolic estrogens. Received: July 29, 2011 Accepted: October 12, 2011 Published: October 12, 2011 8582

dx.doi.org/10.1021/ac2019766 | Anal. Chem. 2011, 83, 8582–8588

Analytical Chemistry

Figure 1. Structures of three natural estrogens (estrone (E1), estradiol (E2), and estriol (E3)) and three synthetic estrogens (bisphenol A (BPA), hexestrol (HES), and diethylstilbestrol (DES)).

Herein, we describe the development of a SERRS-based approach for recognition of phenolic estrogens. A key feature of this study is the coupling reaction between phenolic estrogens and Pauly’s reagent, which we used not only to enhance the affinity of the analytes to metal nanoparticle surfaces but also to amplify SERS signals by the SERRS effect of the reaction products. The proposed method is simple (mixing the estrogens, Pauly’s reagent, and silver colloid without separation or purification), rapid (completion of the coupling reaction and SERS measurement within 2 min), ultrasensitive (down to the subnanomolar level in solution), highly selective due to phenol-based coupling and the SERRS effect, and applicable to most molecules with phenol groups. Moreover, unlike immunochemistry-based methods, nonspecific binding and false positive results can be eliminated, because each SERRS spectrum of the azo dyes is specific for its corresponding estrogen, which enables the proposed method to detect multiple phenolic estrogens simultaneously in a mixed solution.

’ EXPERIMENTAL METHODS Materials. Estrone (E9750, E1), β-estradiol (E8875, E2), estriol (E1253, E3), bisphenol A (239658, BPA), hexestrol (H7753, HES), diethylstilbestrol (D4628, DES), human serum (H4522), and silver nitrate were purchased from Sigma Aldrich Co., Ltd. The estrogens were dissolved in ethanol or dioxane and used without further purification. Infant formula was from Meiji Co., Ltd. 4-Aminophenol, 4-chlorophenol, 4-ethylphenol, and all other chemicals were from Wako Co., Ltd., and ultrapure water (18 MΩ cm) was used throughout the study. Silver Colloid. Colloidal silver (Ag) was prepared by aqueous reduction of Ag nitrate (10 3 M, 200 mL) with trisodium citrate (1%, 4 mL).32 The plasmon absorption maximum of this Ag colloid was ∼430 nm, and the average Ag nanoparticle (AgNP) size was ∼50 nm. Pauly’s Reagent and Coupling Reaction Reagent A: sulfanilic acid (4.5 g) + 12 M HCl (5 mL) for 500 mL of solution (stored at 4 °C) Reagent B: 5% sodium nitrite (stored at 4 °C) Reagent C: 10% sodium carbonate Coupling reaction: Reagents A + B + C + estrogen (1:1:1:2, v/v/v/v) SERS Measurement. After the coupling reaction, 25 μL of each sample mixed with 25 μL of Ag colloid was deposited

ARTICLE

Figure 2. Raman spectra of pure E2 and BPA and their SERS spectra in silver colloid; E2 was dissolved in dioxane (100 ppm), and BPA was dissolved in ethanol (100 ppm).

dropwise into an aluminum pan (No. 0219-0041, Perkin Elmer), and the mixture was exposed to a laser beam for 30 s before each SERS measurement. A Raman spectrophotometer (Model RS-2100, Photon Design, Inc.) equipped with a charge-coupled device (CCD) (Princeton Instruments) was used. Radiation with wavelengths of 514.5, 488, and 458 nm from an Ar+-ion laser (Spectra Physics) and a HoloSpec f/1.8i spectrograph (Kaiser Optical Systems, Inc.) with a 785-nm nearinfrared (NIR) diode laser (Invictus) were used for SERS excitation with a power of 15 mW at the sample. The typical exposure time for each Raman/SERS measurement in this study was 15 s with two accumulations.

’ RESULTS AND DISCUSSION Three natural steroid estrogens (estrone (E1), estradiol (E2), and estriol (E3)) and three nonsteroid estrogens (bisphenol A (BPA), hexestrol (HES), and diethylstilbestrol (DES)) were used as model phenolic estrogens in the study. As shown in Figure 1, there were slight differences in the substituents and the number of hydroxyl groups among the three natural estrogens; for the nonsteroid group, the structures varied in terms of carbon bridges of two phenol groups. As mentioned previously, the affinity of these phenolic estrogens to AgNPs was rather weak, resulting in low SERS sensitivity. As shown in Figure 2, no SERS signals of the estrogens were detectable, even at high estrogen concentration, in silver colloid, except for Raman peaks from the solvents. Coupling Reaction. To achieve highly sensitive detection of phenolic estrogens, before SERS measurements, we changed the phenolic estrogens to azo dyes by a coupling reaction between the estrogens and diazonium ions (see Figure 3). According to the mechanism of the coupling reaction,33 the phenol groups of target estrogens may form azo dyes preferentially at the 4-position with diazobenzenesulfonic acid via electrophilic aromatic substitution. As a result, we observed significant color changes immediately after adding the estrogens to the Pauly’s reagent (the inset of Figure 4). Most of these azo dyes are red or dark red when viewed by the naked eye, while the product of BPA is somewhat yellow. The color changes were more clearly distinguished by UV vis spectroscopy (Figure 4). It is noted that five of the azo dyes had a 8583

dx.doi.org/10.1021/ac2019766 |Anal. Chem. 2011, 83, 8582–8588


ARTICLE

Figure 3. Coupling reaction between estradiol and diazonium ions.

Figure 5. Resonance Raman spectra of the azo dyes derived from the six estrogens at a concentration of 100 ppm.

Figure 4. Visual color change after coupling reactions between the estrogens E1 (a), E2 (b), E3 (c), BPA (d), HES (e), DES and diazonium ions (f), and UV vis spectra of the corresponding azo dyes; the estrogens at a concentration of 1 mg/mL were dissolved in dioxane (spectra a, b, and c), with OA as a control, and in ethanol (spectra d, e, and f), with OB as a control.

similar maximum absorption at ∼500 nm, which differed from that of the azo product from BPA (∼450 nm). Laser excitation at 514.5 nm was used for all estrogens when considering the resonance effect of the samples and the plasmon resonance of silver nanoparticles with the laser. Furthermore, we found that, under our experimental conditions, the coupling reaction was completed very fast (within 1 min) and its maximum absorption was almost stable within 1 h (see SI-Figure 1 in the Supporting Information). Resonance Raman (RR) Spectra. As shown in Figure 5, RR spectra of the azo dyes derived from the six estrogens were observed with a high fluorescence background. From the fluorescence background, obvious spectral overlap among five azo dyes derived from E1, E2, E3, HES, and DES can be observed, similar to their UV vis spectra, which indicates the limitation of

the fluorescence as well as UV vis spectroscopy in identifying analytes with similar structures. In contrast to broad fluorescence bands, sharp Raman peaks of these azo dyes allowed spectral comparison in detail and clear identification of target analytes with slight structural differences. SERRS Spectra. Our experiments showed that RR spectra were almost undetectable at lower estrogen concentrations (e10 ppm). To improve the sensitivity of this azo-dye based protocol for the target estrogen detection, silver nanoparticles were included in the azo dye solution for SERS measurements. As shown in Figure 6, strong SERRS signals were observed due to adsorption of the azo dyes to the Ag surface and silver aggregation. The six estrogens can be easily identified by their corresponding azo dyes, based on their structural differences. The strongest band between 1500 and 1300 cm 1 was observed at ∼1435 cm 1 and was assigned to the NdN stretching vibration34 of the trans isomer in all samples except the BPA derivative, indicating the existence of a trans isomer of NdN in these azo dyes. Compared with other SERRS spectra of the estrogen-derived azo dyes investigated in this study, SERRS spectra of the BPA-derived azo dye is unique, in terms of a weaker band at 1441 cm 1 and a new band at 1414 cm 1, which probably originated from the stereo structure of BPA and its steric effects on NdN stretching. In addition, the SERRS spectra of the azo dyes were similar to the corresponding RR spectra, indicating physical adsorption of the azo 8584



ARTICLE

dyes on the silver surface. Detailed band assignments are listed in Table 1.34 38 Moreover, normal Raman and SERS spectra of diazonium ions before reacting with phenolic estrogens are very weak, which almost have no interference with SERRS spectra of the estrogen-derived azo dyes (see SI-Figure 2 in the Supporting Information). Laser-Dependent SERS Spectra. As shown in Figure 4, the absorption maximum of the BPA-derived azo dye was ∼450 nm. To examine the effect of laser wavelength on SERRS intensity, laser-dependent SERRS spectra of the BPA-derived azo dye were investigated (Figure 7). As expected, no SERRS spectra of the azo dyes were observed with the 785-nm laser wavelength

because the laser line was far from their absorption maxima. From the SERS spectrum, we observed new Raman peaks, in addition to the vibrational information of the BPA-derived azo dye. Therefore, with the red 785-nm laser, the azo dyes could not be selectively detected. In contrast, except for the slight difference in relative band intensity, similar SERRS spectra were observed using laser wavelengths of 458, 488, and 514.5 nm, respectively. As we know, the SERS enhancement depends not only on the resonance effect of the azo-dye but also on plasmon coupling in Ag aggregates,23,39 and therefore, the clearer SERRS spectrum with the 514.5-nm laser was the result of a combination of the two effects. The laser-dependent SERS spectra of the BPA-derived azo dye indicate that the 514.5-nm laser wavelength was as (or slightly more) sensitive than the other laser wavelengths (458 and 488 nm) for the detection of BPA in the present study. Moreover, HCl used in the Pauly’s reagent did not shift the surface plasmon resonance of Ag nanoparticles to longer wavelength (see SI-Figure 3 in the Supporting Information). Therefore, we used the 514.5-nm laser wavelength as an excitation source for target estrogen detection. Sensitivity. In the mixture of AgNPs and azo dyes, small amounts of silver aggregates were observed by the confocal microscopy of the Raman spectrometer, because of the metal salts used for the coupling reaction in solution. Further silver aggregation occurred when the mixture was exposed to a laser beam, which induced stronger SERS enhancement due to stronger surface plasmon resonance between the AgNP aggregates

Figure 6. SERRS spectra of the azo dyes derived from the estrogens at a concentration of 50 ppm.

Figure 7. Laser-dependent averaged SERRS spectra of the BPA-derived azo dye (40 ppm BPA).

Table 1. Assignments of SERRS Bands of the Six Estrogen-Derived Azo Dyesa E1

E2

E3

BPA

HES

DES

1593

1594

1587

1593

1593

1591

ν(CC) within phenol and/or phenyl rings

1438

1437

1434

1441

1435

1432

ν(NN) of trans isomers ν(NN)

1414

a

assignments

1391

1390

1385

1383

1386

1383

ν(CC) within phenyl rings; δ(CH) + δ(OH) from phenol groups

1316

1320

1326

1305

1310

1329

δ(CCH) + δ(NCC) within phenyl rings

1218

1213

1206

1193

1204

1202

ν(CN) + δ(CNN) (phenyl N); δ(CH) + δ(OH) + ν(CC) from phenol groups

1150 1115

1161 1117

1158 1112

1163 1117

1150 1110

1147 1112

ν(CN) + δ(CNN) (phenyl N); δ(CH) + ν(CC) from phenol groups ν(CC) within phenol and/or phenyl rings

δ denotes in-plane bending; ν, stretching. 8585



Figure 8. E1 and BPA concentration-dependent SERRS spectra of the corresponding azo dyes in dioxane and the intensity ratio of two SERRS bands versus the estrogen concentration.

Figure 9. SERRS spectra of E1-derived azo compounds with different E1 concentrations under different experimental conditions.

and the laser.40,41 The inset of Figure 8 shows E1- and BPAdependent SERRS spectra of the corresponding azo dyes, which indicate a tendency of increasing bands at 1391 and 1383 cm 1 with an increase in estrogen concentration. Using a dioxane

ARTICLE

Figure 10. SERRS spectra of estrogen-derived azo dyes from estrogen mixtures (E1 and E3 in dioxane; BPA, HES, DES, and E2 in ethanol) and their individual spectra (each estrogen was at a concentration of 50 ppm).

vibration band at 1310 cm 1 as an internal standard, the intensity ratios of the two bands (I1391/I1310 and I1383/I1310) versus the two estrogen concentration were plotted (Figure 8). The limit of detection (LOD) of the proposed method for target estrogen detection in solution was ∼0.1 ppb (0.1 μg/kg), 500-fold lower than the BPA safety limit (the current U.S. Environmental Protection Agency (USEPA) safety limit is 50 μg/kg/day) and comparable to the lowest value reported using mass spectrometry and surface plasmon resonance (SPR)-based methods.7,12 As shown in the inset of Figure 8, at very low estrogen concentration, SERS bands are dominated by the vibrational information of the solvent (dioxane), which hides the relatively weak SERRS bands of the azo compounds in solution. Phenolic estrogens can be further identified, even at very low estrogen concentration, since clearer SERRS fingerprints of the azo dyes can be observed after solvent evaporation (see Figure 9). For quantitative detection of estrogens in solid substrates, the inhomogeneity of the silver aggregates requires selection of an appropriate internal standard to eliminate instrumental and experimental variables.40,42 Alternatively, the azo compounds can be assembled on highly regular and reproducible SERS-active substrates.43 Moreover, soaking solid SERS-active substrates in azo dye solutions may increase analyte adsorption and lower LOD.44 Routine and practical use of this method for quantitative 8586



ARTICLE

Figure 12. SERRS spectra of three phenol derivative-derived azo dyes (4-aminophenol, 4-chlorophenol, and 4-ethylphenol), each at a concentration of 50 ppm.

Figure 11. SERRS spectra of E2-derived azo dyes from the extracts of human serum and infant formula.

estrogen detection below nanomolar levels on solid substrates is now in progress in our group. Multiplexing Capability. To investigate the multiplexing capability of the proposed method, we investigated SERRS spectra of the azo dyes in an estrogen mixture. Our experiments showed that E2 was soluble in ethanol at room temperature but E1 and E3 were not; therefore, we divided the target estrogens into two groups, ethanol soluble and dioxane soluble, after considering the practical applications of the study. Because of their azo structure, the comparable affinity of the azo dyes to the Ag nanoparticles allowed for parallel adsorption followed by overlap of the SERRS bands. As shown in Figure 10, composite SERRS bands were clearly observed in the azo dyes of the estrogen mixture, confirming the application potential of the present method for the detection of an estrogen mixture. Practical Feasibility. To further confirm the feasibility of the proposed method for the detection of estrogens in real-life applications, we examined the SERRS response of human serum and infant formula mixed with E2 after the coupling reaction. Human serum and infant formula (50 mg/mL in aqueous solution) were diluted with dioxane (5 dilution). Most proteins in the human serum and formula were aggregated by dioxane and were separated by centrifugation, and E2 was subsequently added to the supernatants. As shown in Figure 11, the human serum extract had a low background, which allowed for clear azo SERRS spectra. For the infant formula extract, the background was relatively high, because of the possible existence of phenolic molecules in the infant formula, but the spectral changes were

obvious when E2 was added. These results indicate high selectivity of SERRS for these azo dyes, even in complicated mixtures, and the feasibility of the approach to practical applications. Universality. The coupling reaction between phenolic molecules and diazonium ions is universal. To confirm the universality of the approach for versatile phenolic molecule detection, three phenol derivatives were investigated. After the coupling reaction, the three phenolic molecules (4-aminophenol, 4-chlorophenol, and 4-ethylphenol) changed in color to dark brown, dark red, and bright red, respectively. As shown in Figure 12, SERRS spectra of the products were significantly different and easily discerned. The color and SERRS spectra of the 4-aminophenol-derived azo compound differed significantly from other phenolic molecules investigated in our study, probably because of the formation of a polyazo compound between 4-aminophenol and diazonium ions. Moreover, besides phenolic derivatives, diazonium ions can also react with other benzene derivatives and this proposed protocol is also useful for their identification (see SI-Figure 4 in the Supporting Information).

’ CONCLUSIONS In summary, a SERRS-based indirect approach for phenolic estrogen detection was developed in the present study. Before SERS measurement, target estrogens were changed to azo dyes by a coupling reaction with diazonium ions, which significantly enhanced the SERRS activity of the analytes. Highly specific SERRS fingerprints of the estrogen-derived azo dyes enabled multiplexing estrogen detection due to structural differences of the analytes. The promising technique with its advantages of simplicity, rapidness, universality, high sensitivity and selectivity, has great potential in applications of phenolic molecule detection in food and environmental safety control. ’ ASSOCIATED CONTENT

bS

Supporting Information. UV-vis spectra of E2-derived azo dye during the coupling reaction process and the control spectrum of diazonium ions without E2. (SI-Figure 1), Raman and SERS spectra of diazonium ions before reacting with phenolic derivatives (SI-Figure 2), UV-vis spectra of the silver colloid, BPA-derived azo dye before and after the addition of Ag

8587


Analytical Chemistry (SI-Figure 3), and SERRS spectra of para-aminothiophenol and estradiol-derived azo compounds (SI-Figure 4). (PDF) This material is available free of charge via the Internet at http:// pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mails: [email protected] (X.X.H.), ozaki@kwansei. ac.jp (Y.O.).

’ ACKNOWLEDGMENT This work was supported by Support Project to Assist Private Universities in Developing Bases for Research (Research Center for Single Molecule Vibrational Spectroscopy) by Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT), and KAKENHI (Grant-in-Aid for Scientific Research) on Priority Area “Strong Photon-Molecule Coupling Fields (No. 470, 20043032)” from MEXT. This work was also supported by the 111 project (B06009), China. X.X.H. acknowledges the support from the Japanese Society for Promotion of Science (JSPS) (No. P09248). ’ REFERENCES (1) Lin, C. Y.; Strom, A.; Vega, V. B.; Kong, S. L.; Yeo, A. L.; Thomsen, J. S.; Chan, W. C.; Doray, B.; Bangarusamy, D. K.; Ramasamy, A.; Vergara, L. A.; Tang, S. S.; Chong, A.; Bajic, V. B.; Miller, L. D.; Gustafsson, J. A.; Liu, E. T. Genome Biol. 2004, 5, R66. (2) Fang, H.; Tong, W. D.; Shi, L. M.; Blair, R.; Perkins, R.; Branham, W.; Hass, B. S.; Xie, Q.; Dial, S. L.; Moland, C. L.; Sheehan, D. M. Chem. Res. Toxicol. 2001, 14, 280–294. (3) Kavlock, R. J.; Daston, G. P.; DeRosa, C.; FennerCrisp, P.; Gray, L. E.; Kaattari, S.; Lucier, G.; Luster, M.; Mac, M. J.; Maczka, C.; Miller, R.; Moore, J.; Rolland, R.; Scott, G.; Sheehan, D. M.; Sinks, T.; Tilson, H. A. Environ. Health Perspect. 1996, 104, 715–740. (4) Chlebowski, R. T.; Hendrix, S. L.; Langer, R. D.; Stefanick, M. L.; Gass, M.; Lane, D.; Rodabough, R. J.; Gilligan, M. A.; Cyr, M. G.; Thomson, C. A.; Khandekar, J.; Petrovitch, H.; McTiernan, A. JAMA, J. Am. Med. Assoc. 2003, 289, 3243–3253. (5) Kuller, L. H.; Cauley, J. A.; Lucas, L.; Cummings, S.; Browner, W. S. Environ. Health Perspect. 1997, 105, 593–599. (6) Rossouw, J. E.; Anderson, G. L.; Prentice, R. L.; LaCroix, A. Z.; Kooperberg, C.; Stefanick, M. L.; Jackson, R. D.; Beresford, S. A. A.; Howard, B. V.; Johnson, K. C.; Kotchen, M.; Ockene, J. JAMA, J. Am. Med. Assoc. 2002, 288, 321–333. (7) Mahmoud, I. Y.; Alkindi, A. Y.; Khan, T.; Al-Bahry, S. N. J. Exp. Zool. Part A 2011, 315A, 170–174. (8) Jimenez-Diaz, I.; Zafra-Gomez, A.; Ballesteros, O.; Navea, N.; Navalon, A.; Fernandez, M. F.; Olea, N.; Vilchez, J. L. J. Chromatogr. B 2010, 878, 3363–3369. (9) De Meulenaer, B.; Baert, K.; Lanckriet, H.; Van Hoed, V.; Huyghebaert, A. J. Agric. Food. Chem. 2002, 50, 5273–5282. (10) Gao, Z.; Liu, N.; Cao, Q.; Zhang, L.; Wang, S.; Yao, W.; Chao, F. Biosens. Bioelectron. 2009, 24, 1445–1450. (11) Geisler, J.; Berntsen, H.; Lonning, P. E. J. Steroid Biochem. Mol. Biol. 2000, 72, 259–264. (12) Hegnerova, K.; Piliarik, M.; Steinbachova, M.; Flegelova, Z.; Cernohorska, H.; Homola, J. Anal. Bioanal. Chem. 2010, 398, 1963–1966. (13) Dai, J. H.; Baker, G. L.; Bruening, M. L. Anal. Chem. 2006, 78, 135–140. (14) Aroca, R. Surface-Enhanced Vibrational Spectroscopy; John Wiley & Sons, Ltd.: Chichester, U.K., 2006. (15) Kneipp, K.; Moskovits, M.; Kneipp, H. Surface-Enhanced Raman Scattering: Physics and Applications; Springer: Berlin, 2006.

ARTICLE

(16) Stiles, P. L.; Dieringer, J. A.; Shah, N. C.; Van Duyne, R. R. Annu. Rev. Anal. Chem. 2008, 1, 601–626. (17) Nie, S.; Emory, S. R. Science 1997, 275, 1102–1106. (18) Blackie, E. J.; Le Ru, E. C.; Etchegoin, P. G. J. Am. Chem. Soc. 2009, 131, 14466–14472. (19) Tsoutsi, D.; Montenegro, J. M.; Dommershausen, F.; Koert, Ul.; Liz-Marzan, L. M.; Parak, W. J.; Alvarez-Puebla, R. A. ACS Nano 2011, DOI: 10.1021/nn2025176. (20) Dasary, S. S. R.; Singh, A. K.; Senapati, D.; Yu, H. T.; Ray, P. C. J. Am. Chem. Soc. 2009, 131, 13806–13812. (21) Zhao, Y.; Newton, J. N.; Liu, J.; Wei, A. Langmuir 2009, 25, 13833–13839. (22) Sanles-Sobrido, M.; Rodriguez-Lorenzo, L.; Lorenzo-Abalde, S.; Gonzalez-Fernandez, A.; Correa-Duarte, M. A.; Alvarez-Puebla, R. A.; Liz-Marzan, L. M. Nanoscale 2009, 1, 153–158. (23) Bell, S. E. J.; Sirimuthu, N. M. S. J. Am. Chem. Soc. 2006, 128, 15580–15581. (24) Han, X. X.; Jia, H. Y.; Wang, Y. F.; Lu, Z. C.; Wang, C. X.; Xu, W. Q.; Zhao, B.; Ozaki, Y. Anal. Chem. 2008, 80, 2799–2804. (25) Han, X. X.; Huang, G. G.; Zhao, B.; Ozaki, Y. Anal. Chem. 2009, 81, 3329–3333. (26) He, L. L.; Rodda, T.; Haynes, C. L.; Deschaines, T.; Strother, T.; Diez-Gonzalez, F.; Labuza, T. P. Anal. Chem. 2011, 83, 1510–1513. (27) Huang, G. G.; Han, X. X.; Hossain, M. K.; Ozaki, Y. Anal. Chem. 2009, 81, 5881–5888. (28) Lezna, R. O.; Detacconi, N. R.; Centeno, S. A.; Arvia, A. J. Langmuir 1991, 7, 1241–1246. (29) Rodriguez-Lorenzo, L.; Alvarez-Puebla, R. A.; de Abajo, F. J. G.; Liz-Marzan, L. M. J. Phys. Chem. C 2010, 114, 7336–7340. (30) Barnett, S. M.; Vlckova, B.; Butler, I. S.; Kanigan, T. S. Anal. Chem. 1994, 66, 1762–1765. (31) Moore, B. D.; Stevenson, L.; Watt, A.; Flitsch, S.; Turner, N. J.; Cassidy, C.; Graham, D. Nat. Biotechnol. 2004, 22, 1133–1138. (32) Lee, P. C.; Meisel, D. J. Phys. Chem. 1982, 86, 3391–3395. (33) Szele, I.; Zollinger, H. Top. Curr. Chem. 1983, 112, 1–66. (34) Wu, Y. Q.; Zhao, B.; Xu, W. Q.; Li, B. F.; Jung, Y. M.; Ozaki, Y. Langmuir 1999, 15, 4625–4629. (35) Biswas, N.; Umapathy, S. J. Phys. Chem. A 1997, 101, 5555– 5566. (36) Stuart, C. M.; Frontiera, R. R.; Mathies, R. A. J. Phys. Chem. A 2007, 111, 12072–12080. (37) Wilson, H. W.; Macnamee, R. W.; Durig, J. R. J. Raman Spectrosc. 1981, 11, 252–255. (38) Lampert, H.; Mikenda, W.; Karpfen, A. J. Phys. Chem. A 1997, 101, 2254–2263. (39) Macaskill, A.; Chernonosov, A. A.; Koval, V. V.; Lukyanets, E. A.; Fedorova, O. S.; Smith, W. E.; Faulds, K.; Graham, D. Nucleic Acids Res. 2007, 35, e42. (40) Bell, S. E. J.; Sirimuthu, N. M. S. Chem. Soc. Rev. 2008, 37, 1012–1024. (41) Han, X. X.; Chen, L.; Ji, W.; Xie, Y. F.; Zhao, B.; Ozaki, Y. Small 2011, 7, 316–320. (42) Han, X. X.; Xie, Y.; Zhao, B.; Ozaki, Y. Anal. Chem. 2010, 82, 4325–4328. (43) Wang, H.; Levin, C. S.; Halas, N. J. J. Am. Chem. Soc. 2005, 127, 14992–14993. (44) Pienpinijtham, P.; Han, X. X.; Ekgasit, S.; Ozaki, Y. Anal. Chem. 2011, 83, 3655–3662.

8588
