Multiplex Immunoassay Using Fluorescent-Surface Enhanced Raman

Dec 31, 2008 - Corresponding authors. Dae Hong Jeong, Ph.D., Department of Chemistry Education, Seoul National University, Seoul 151-742, Korea...
0 downloads 0 Views 2MB Size
Download PDF
Anal. Chem. 2009, 81, 1008–1015

Multiplex Immunoassay Using Fluorescent-Surface Enhanced Raman Spectroscopic Dots for the Detection of Bronchioalveolar Stem Cells in Murine Lung Min-Ah Woo,† Sang-Myung Lee,‡ Gunsung Kim,§ JongHo Baek,§ Mi Suk Noh,† Ji Eun Kim,† Sung Jin Park,† Arash Minai-Tehrani,† Se-Chang Park,† Yeong Tai Seo,| Yong-Kwon Kim,| Yoon-Sik Lee,*,‡ Dae Hong Jeong,*,§ and Myung-Haing Cho*,† College of Veterinary Medicine and Interdisciplinary Program in Nano-Science and Technology, School of Chemical and Biological Engineering, Department of Chemistry Education, and School of Electrical Engineering and Computer Science, Seoul National University, Seoul 151-742, Korea Immunoassays using nanomaterials have been rapidly developed for the analysis of multiple biomolecules. Highly sensitive and biocompatible surface enhanced Raman spectroscopy-active nanomaterials have been used for biomolecule analysis by many research groups in order to overcome intrinsic problems of conventional immunoassays. We used fluorescent surface-enhanced Raman spectroscopic dots (F-SERS dots) to detect biomolecules in this study. The F-SERS dots are composed of silver nanoparticle-embedded silica nanospheres, organic Raman tagging materials, and fluorescent dyes. The F-SERS dots demonstrated highly sensitive, selective, and multifunctional characteristics for multiplex targeting, tracking, and imaging of cellular and molecular events in the living organism. We successfully applied F-SERS dots for the detection of three cellular proteins, including CD34, Sca1, and SP-C. These proteins are simultaneously expressed in bronchioalveolar stem cells (BASCs) in the murine lung. We analyzed the relative expression ratios of each protein in BASCs since external standards were used to evaluate SERS intensity in tissue. Quantitative comparisons of multiple protein expression in tissue were first attempted using SERS-encoded nanoprobes. Our results suggested that immunoassays using F-SERS dots offered significant increases in sensitivity and selectivity. Such immunoassays may serve as the primary next-generation labeling technologies for the simultaneous analysis of multiple biomolecules. * Corresponding authors. Dae Hong Jeong, Ph.D., Department of Chemistry Education, Seoul National University, Seoul 151-742, Korea. Fax: (+82) 2-8890749. E-mail: [email protected]. Yoon-Sik Lee, Ph.D., School of Chemical and Biological Engineering, Seoul National University, Seoul 151-742, Korea. Fax (+82) 2-888-1604. E-mail: [email protected]. Myung-Haing Cho, D.V.M., Ph.D., Laboratory of Toxicology, College of Veterinary Medicine, Seoul National University, Seoul 151-742, Korea. Fax: (+82) 2-873-1268. E-mail: mchotox@ snu.ac.kr. † College of Veterinary Medicine and Interdisciplinary Program in NanoScience and Technology. ‡ School of Chemical and Biological Engineering. § Department of Chemistry Education. | School of Electrical Engineering and Computer Science.

1008

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

Immunofluorescence assays have been widely used to elucidate the distribution and localization of expressed proteins in biological tissues. Fluorescence-based materials have been widely used in biological applications.1-4 Variable fluorescent dyes can be used for multiplex tissue analysis. However, limited numbers of available fluorescent agents and broad emission profiles leading to peak overlapping restricts multiplex detection in multiple labeling with fluorescent molecules. Also, fluorescence is insufficiently sensitive due to background autofluorescence emanating from superficial tissue layers as well as a rapid photobleaching problem.5 Surface-enhanced Raman scattering (SERS) has been used for the analysis of molecular information through sharp and easily distinguishable vibrational bands and also for bioanalyte detection.6-12 The use of normal Raman scattering as a readout method for immunoassays is limited by low intensity and lack of sensitivity. This may be circumvented by the use of SERS and metal nanoparticles which may overcome these limitations. The SERS technique provides 106-1014-fold enhancement in Raman signal intensity, and this is sufficient to detect pico- to femtomolar amounts of biomolecules.7,10,12 Furthermore, SERS is currently utilized as (1) Pirrung, M. C.; Connors, R. V.; Odenbaugh, A. L.; Montague-Smith, M. P.; Walcott, N. G.; Tollett, J. J. J. Am. Chem. Soc. 2000, 122, 1873–1882. (2) Duhachek, S. D.; Kenseth, J. R.; Casale, G. P.; Small, G. J.; Porter, M. D.; Jankowiak, R. Anal. Chem. 2000, 72, 3709–3716. (3) He, B.; Burke, B. J.; Zhang, X.; Zhang, R.; Regnier, F. E. Anal. Chem. 2000, 73, 1942–1947. (4) Peruski, A. H.; Johnson, L. H.; Peruski, L. F. J. Immunol. Methods 2002, 263, 35–41. (5) Keren, S.; Zavaleta, C.; Cheng, Z.; de la Zerda, A.; Gheysens, O.; Gambhir, S. S. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 5844–5849. (6) Kneipp, J.; Kneipp, H.; Kneipp, K. Chem. Soc. Rev. 2008, 37, 1052–1060. (7) Lin, C.-C.; Yang, Y.-M.; Chen, Y.-F.; Yang, T.-S.; Chang, H.-C. Biosens. Bioelectron. 2008, 24, 178–183. (8) Nithipatikom, K.; McCoy, M. J.; Hawi, S. R.; Nakamoto, K.; Adar, F.; Campbell, Y. B. Anal. Biochem. 2003, 322, 198–207. (9) Swain, R. J.; Stevens, M. M. Biochem. Soc. Trans. 2007, 35, 544–549. (10) Lee, S. Y.; Kim, S. Y.; Choo, J. B.; Shin, S. Y.; Lee, Y. H.; Choi, H. Y.; Ha, S. H.; Kang, K. H.; Oh, C. H. Anal. Chem. 2007, 79, 916–922. (11) Lutz, B.; Dentinger, C.; Sun, L.; Nguyen, L. C.; Zhang, J. W.; Chmura, A. J.; Allen, A.; Chan, S.; Knudsen, B. J. Histochem. Cytochem. 2008, 56, 371– 379. (12) Sun, L.; Sung, K. B.; Dentinger, C.; Lutz, B.; Nguyen, L.; Zhang, J. G.; Qin, H.; Yamakawa, M.; Cao, M.; Lu, Y.; Chmura, A. J.; Zhu, J.; Su, X.; Berlin, A. A.; Chan, S.; Knudsen, B. Nano Lett. 2007, 7, 351–356. 10.1021/ac802037x CCC: $40.75  2009 American Chemical Society Published on Web 12/31/2008

an effective tool in quantitative analysis.13-17 Functionalized SERS nanoparticles in the presence of Raman-active molecules are typically used in detection, sensing, or imaging of biological samples such as DNA, proteins, cells, and tissues. These particles have great potential as rapid and sensitive diagnostic tools.8,10,18 Previously, our group introduced new surface-enhanced Raman spectroscopic dots (SERS dots).19 These SERS dots are silver nanoparticle-embedded silica spheres with incorporated organic Raman label compounds. Therefore, CD10 and HER-2 antibodyconjugated SERS dots have been successfully applied toward specific targeting of cellular labeling in cancer. Many research groups have focused on the development of multiplexed immunoassays using various nanoprobes as labels.20-22 The SERS technique has strong points in its capability for multiplex targeting using individual spectral signatures or its combinations.23 Technologies of multiplex tissue analysis using SERS-active nanoparticles are currently under further development.24 Simultaneous detection of two or more analytes (multiplex tissue analysis) is a powerful tool to study protein coexpression and spatial distribution with less consumption of tissue samples.7,12,19 We previously introduced a new variety of SERS dots known as the F-SERS dots (fluorescent-surface-enhanced Raman spectroscopic dots).25 Biocompatible and multifunctional F-SERS dots are composed of silver nanoparticle-embedded silica spheres with fluorescent organic dyes and specific SERS active chemicals (encoding chemicals). Immunoassays were performed with FSERS dots, which were fabricated with two different encoding chemicals, for multiplex targeting, tracking, and imaging of two apoptosis protein expression in tissue. Our data suggested that F-SERS dots have the capability for effective detection of multiple targets but is still not optimized for the quantification of multiple biomarkers. We therefore examined the potential application of F-SERS dots in quantitative analysis using SERS intensity. Bronchioalveolar stem cells (BASCs) play a critical role in the maintenance of bronchiolar Clara cell and alveolar cell homeo(13) Smith, W. E.; Faulds, K.; Graham, D. Top. Appl. Phys. 2006, 103, 381– 396. (14) Haynes, C. L.; Yonzon, C. R.; Zhang, X.; Van Duyne, E. R. J. Raman Spectrosc. 2005, 36, 471–484. (15) Pinzaru, S. C.; Pavel, I.; Leopold, N.; Kiefer, W. J. Raman Spectrosc. 2004, 35, 338–346. (16) Narayanan, R.; Lipert, R. J.; Porter, M. D. Anal. Chem. 2008, 80, 2265– 2271. (17) Ackermann, K. R.; Henkel, T.; Popp, J. ChemPhysChem 2007, 8, 2665– 2670. (18) Sha, M. Y.; Xu, H.; Penn, S. G.; Cromer, R. Nanomedicine 2007, 2, 725– 734. (19) Kim, J.-H.; Kim, J.-S.; Choi, H. J.; Lee, S.-M.; Jun, B.-H.; Yu, K.-N.; Kuk, E. Y.; Kim, Y.-K.; Jeong, D. H.; Cho, H.-M.; Lee, Y.-S. Anal. Chem. 2006, 78, 6967–6973. (20) Jun, B.-H.; Kim, J.-H.; Park, H. M.; Kim, J.-S.; Yu, K.-N.; Lee, S.-M.; Choi, H. J.; Kwak, S.-Y.; Kim, Y.-K.; Jeong, D. H.; Cho, M.-H.; Lee, Y.-S. J. Comb. Chem. 2007, 9, 237–244. (21) Wu, X.; Liu, H.; Liu, J.; Haley, K. N.; Treadway, J. A.; Larson, J. P.; Ge, N.; Peale, F.; Bruchez, M. P. Nat. Biotechnol. 2003, 21, 41–46. (22) Xing, Y.; Chaudry, Q.; Shen, C.; Kong, K. Y.; Zhau, H. E.; Chung, L. W.; Petros, J. A.; O’Regan, R. M.; Yezhelyev, M. V.; Simons, J. W.; Wang, M. D.; Nie, S. Nat. Protoc. 2007, 2, 1152–1165. (23) Brown, L. O.; Doorn, S. K. Langmuir 2008, 24, 2277–2280. (24) Su, X.; Zhang, J. W.; Sun, L.; Koo, T. W.; Chan, S.; Sundararajan, N.; Yamakawa, M.; Berlin, A. A. Nano Lett. 2005, 5, 49–54. (25) Yu, K. N.; Lee, S.-M.; Han, J. Y.; Park, H. M.; Woo, M.-A.; Noh, M. S.; Hwang, S.-K.; Kwon, J.-T.; Jin, H.; Kim, Y.-K.; Hergenrother, P. J.; Jeong, D. H.; Lee, Y.-S.; Cho, M.-H. Bioconjugate Chem. 2007, 18, 1155–1162.

stasis and are linked to lung adenocarcinoma initiation.26 Studies implicating BASCs in tumor initiation may provide more effective stem cell-specific markers and therapeutic strategies for lung cancer through the targeting of BASCs.27 Multifunctional F-SERS dots were synthesized by two fluorescent dyes (FITC and AF647) to prescan BASCs at bronchioalveolar duct junctions (BADJ) for immunoassays, and three Raman chemicals were used to differentiate three proteins (CD34, Sca-1, and SP-C). We used this strategy to identify BASCs through both fluorescence-based imaging and SERS intensity mapping for signal detection at different targets. The relative expression ratios of the CD34, Sca1, and SP-C proteins in BASCs were determined through quantitative analysis with SERS intensity and by multiplex targeting. Here, we report that F-SERS dots are able to detect BASCs coexpressed CD34, Sca-1, and SP-C proteins successfully. Our results suggest that F-SERS dots are useful in the sensitive detection of multiple tissue proteins and also provide cellular characterization and information effectively compared to the conventional fluorescent probe system. Furthermore, F-SERS dots could be applied to the multiplex immunoassays for smart biological analysis. EXPERIMENTAL SECTION Preparation of Thiolated Silica Nanoparticle Templates. Silica nanoparticles of ∼120 nm size were prepared by the Sto¨ber method.30 A 1 mL portion of ammonium hydroxide (27%) was added to 78 mL of 95% ethanol and 10 mL of H2O, and 5 mL of (22.4 mmol) of tetraethylorthosilicate (TEOS) was added to this solution. The solution was stirred with a magnetic bar vigorously for 12 h at 25 °C. The resulting silica colloids were centrifuged and washed with ethanol several times. To obtain thiolated silica nanoparticles, silica nanoparticles (300 mg) were dispersed in ethanol and 150 µL of (3-mercaptopropyl) trimethoxysilane (MPTS) and 300 µL of ammonium hydroxide were added to the colloidal solution. The mixture was stirred with a magnetic bar vigorously for 12 h at 25 °C. The resulting thiolated silica nanoparticles were centrifuged and washed with ethanol several times to remove excess reagents. Preparation of Silver-Doped Silica Nanoparticles. Silverdoped silica nanoparticles were prepared as previously described.25,31 Briefly, 160 mg of thiolated silica nanoparticles were dispersed in a water and ethanol solution (15 to 15 mL), and 100 µL of trifluoroacetic acid (TFA) and 100 mg of SnCl2 were added to the suspension. The mixture was stirred vigorously for 1 h at 25 °C, and the resulting silica nanoparticles were centrifuged and washed with ethanol (20 mL, 3 times). The Sn2+-coated silica nanoparticles were redispersed in 250 mL of water. Silver nitrate (50 mL of 6 mM AgNO3) was added dropwise to the solution and was vigorously stirred with a magnet at 25 °C. The dispersion was centrifuged and washed with H2O (30 mL, (26) Kim, C. F.; Jackson, E. L.; Woolfenden, A. E.; Lawrence, S.; Babar, I.; Vogel, S.; Crowley, D.; Bronson, R. T.; Jacks, T. Cell 2005, 121, 823–835. (27) Kim, C. F. Am. J. Physiol.: Lung Cell. Mol. Physiol. 2007, 293, 1092–1098. (28) Giangreco, A.; Reynolds, S. D.; Stripp, B. R. Am. J. Pathol. 2002, 161, 173– 182. (29) Nolen-Walston, R. D.; Kim, C. F.; Mazan, M. R.; Ingenito, E. P.; Gruntman, A. M.; Tsai, L.; Boston, R.; Woolfenden, A. E.; Jacks, T.; Hoffman, A. M. Am. J. Physiol.: Lung Cell. Mol. Physiol. 2008, 294, 1158–1165. (30) Stober, W.; Fink, A.; Bohn, E. J. Colloid Interface Sci. 1968, 26, 62–69. (31) Kobayashi, Y.; Tadaki, Y.; Nagao, D.; Konno, M. J. Colloid Interface Sci. 2005, 283, 601–604.

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

1009

2 times) and ethanol (30 mL, 2 times) to remove excess reagents. Incorporation of Fluorescence and Raman Tagging Materials into Silver-Doped Silica Nanoparticles. The molecules of mercaptotoluene (MT), benzenethiol (BT), and naphthalenethiol (NT) were used as Raman-encoding chemicals, and dye molecules such as fluorescein isothiocyanate (FITC) and Alexafluoro 647 (AF647) were used for fluorescence signals in order to introduce dual signals of SERS and fluorescence to the silica nanoparticles. The Raman-active chemicals were combined with (3-mercaptopropyl)trimethoxysilane (MPTS) and allowed to form a self-assembled monolayer on the silver surface of silica nanoparticles. Raman-active chemicals (10 mL of 5 mM in ethanol) and MPTS (10 mL of 50 mM in ethanol) were mixed and added to a 50 mg portion of silver embedded silica nanoparticles. The dispersion was stirred vigorously for 1 h at 25 °C. The colloids were centrifuged and washed with ethanol (20 mL, 3 times) and water (20 mL). Silica nanoparticles were dispersed in 20 mL of water, and 30 µL of aqueous sodium silicate (27%) was added into the dispersion and stirred vigorously for 12 h at 25 °C. The resulting colloids were centrifuged and washed with water (20 mL, 3 times) and ethanol (20 mL) to remove excess reagents. A fluorescent layer was introduced in the next step onto silica nanoparticles labeled with Raman signals. Conjugates of 3-aminopropyltriethoxysilane (APS) and fluorescence dyes were synthesized. FITC (10 µL of 8 mM in DMSO) was added to APS (100 µL of 19.2 mM in ethanol), and the resulting solution was stirred for 6 h at 25 °C. The resulting FITC-APS conjugate was mixed with 20 µL of tetraethylorthosilicate (TEOS) and 40 µL of aqueous ammonium hydroxide (25%), and this solution was added to the 20 mg portion of silica nanoparticles with Raman labels. The mixture was stirred vigorously for 12 h at 25 °C. The resulting nanoparticles (F-SERS dots) were centrifuged and thoroughly washed with ethanol. Surface Functionalization of F-SERS Dots. The purified nanoparticles were coated with a polyethylene glycol (PEG) spacer to reduce nonspecific protein binding and to introduce amino groups to the nanoparticles. The methoxy-PEG spacer, 2-[methoxy(polyoxy-ethylene)propyl] trimethoxysilane (mPEG600-Si; MW, ∼600) (11 mg, 0.5 mM) and APTS (0.22 mg, 0.1 mM) were dissolved together in 10 mL of ethanol with 100 µL of ammonium hydroxide solution and added to the dispersion of 10 mg of nonfunctionalized F-SERS dots. The mixture was stirred for 12 h at room temperature, and the final mPEG600/-NH2 functionalized F-SERS dots were washed with ethanol. Conversion of the amine terminal group to a carboxylic terminal group was facilitated by the treatment of 1 mg of F-SERS dots with 50 mM succinic anhydride and 50 mM N,N-diisopropyethylamine (DIEA) in N,N-dimethylformamide (DMF) for 2 h at room temperature. The resulting nanoparticles were sequentially washed with DMF, ethanol, and deionized water. Preparation of Antibody-Conjugated F-SERS Dots. F-SERS dots were conjugated with CD34 (Abcam, Cambridge, U.K.), Sca-1 (R&D Systems, MN) and SP-C (Santa Cruz Biotechnology, CA) monoclonal antibodies. A general bioconjugation method between the carboxylic acid and amine groups was followed. The carboxylated F-SERS dots (1 mg) were dispersed in 1 mL of 2-(N1010

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

morpholino)ethanesulfonic acid (MES) buffer solution (pH 6.0). The solution was mixed with 50 µL of the 50 mg/mL 1-ethyl-3[3-dimethylaminopropyl] carbodiimide hydrochloride (EDC) in MES buffer and 50 µL of the 50 mg/mL N-hydroxysuccinimide (NHS) in MES buffer. The mixture was stirred for 1 h at room temperature. The activated F-SERS dots were rinsed with MES solution once and redispersed in MES buffer solution. Monoclonal antibody (40 µg) was added into the dispersion and stirred for 1 h at room temperature. The antibody-conjugated F-SERS dots were sequentially rinsed with PBST (phosphate buffer saline including 1% Tween20) and PBS and stored at 4 °C in the dark. The size and morphology of F-SERS dots were observed by a highresolution transmission electron microscope (HR-TEM, JEM-3010, Jeol Inc., NY). Measurement of SERS Intensity of F-SERS Dots. F-SERS dots (20 µg in 100% ethanol) were spun by microcentrifugation. The F-SERS dot pellet was photoexcited after drying the ethanol in order to measure SERS intensity. The Raman scattering signal was collected in 180° scattering geometry and detected by a spectrometer equipped with a thermoelectrically cooled CCD detector in this Raman system (LabRam 300, JY-Horiba). A 514.5 nm laser line from a continuous wave Ar ion laser (Melles Griot, 35-MAP-321) was used as a photoexcitation source with a laser power of ∼2 mW at the sample. Raman scattered light was collected with a ×10 microscope objective (Olympus, 0.25 NA) that was also used to focus the excitation laser light. The strong Rayleigh scattered light was rejected using a holographic notch filter. Acquisition times for all spectra of F-SERS dots were 20 s, and measured sites were randomly selected. Cell Culture and F-SERS Dots Treatment. The Abelson murine leukemia virus-induced tumor cell line RAW 264.7 was obtained from ATCC (The American Type Culture Collection, Manassas, VA). The cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM, PAA Laboratories, Pasching, Austria) with 10% fetal bovine serum (FBS, Hyclone, Logan, UT) and 100 units/mL penicillin-streptomycin (Invitrogen, CA). The cells were incubated in a humidified environment with 5% CO2 and 95% room air at 37 °C. Cells were detached from the culture flask after growth in media with a trypsin/EDTA solution (Invitrogen) and harvested by centrifugation. Cells were washed with 1× PBS solution (8.1 mM Na2HPO4, 1.2 mM KH2PO4, 138 mM NaCl, 2.7 mM KCL, pH 7.4) and fixed with 4% paraformaldehyde for 2 h at 4 °C. The fixative agent was removed by centrifugation, and the cell suspension was incubated with 3% bovine serum albumin (BSA) in 1× PBS for 1 h at room temperature in order to block nonspecific binding sites. Nuclear staining was performed after blocking with 1 µg/mL 4′,6diamidino-2-phenylindole (DAPI, Sigma-Aldrich, MO), and cells were washed with 1× PBS solution. Sca-1 antibody-conjugated F-SERS dots and CD34 antibody-conjugated F-SERS dots were both added to different cell suspensions (in 1% BSA solution), and the cells were incubated for overnight at 4 °C. The cells were incubated, extensively washed, and observed with a confocal laser scanning microscope (CLSM, Nikon, Tokyo, Japan). Tissue Preparation and F-SERS Dots Treatment. Lung tissue samples were obtained from 15-week K-ras null mice, which are used as a laboratory animal model of nonsmall lung cancer

Figure 1. Peak height analysis of F-SERS dots: (a) 1593 cm-1 band of MT, (b) 997 cm-1 band of BT, and (c) 1378 cm-1 band of NT. The dashed line in the upper image indicates a specific Raman band of each Raman label. The lower image is a magnified image of the SERS band in the dashed line in the upper image. The red lines indicate regions as the left and right base at 8 cm-1 band intervals. The blue line connects between the averaged intensity value of the left and right base. The green line indicates the intensity difference between the peak top and the middle point of the blue line. Each spectrum in Figure 1 shows the actual SERS signal of one point extracted from Raman mapping of the tissue sample.

(NSLC). Formalin-fixed and paraffin-embedded tissue sections were cut at 4 µm and transferred to microscope slides (Fisher Scientific, Pittsburgh, PA). The tissue sections were deparaffinized in xylene and rehydrated with an alcohol gradient. Sections were incubated in 150 mL of proteinase K for antigen retrieval, washed with 1× TBS (100 mM Tris-Cl, 150 mM NaCl, pH 7.5), and fixed with an ice-cold methanol-acetone mixture (v:v 1:1) for 10 min. Tissue sections were washed with 1× TBS, rinsed for 10 min in 0.025% Triton X-100 in 1× TBS to reduce surface tension, and incubated with 3% BSA in TBST (Tris buffer saline including 1% Tween20) for 1 h at room temperature to block nonspecific binding sites. Nuclear staining was performed with 1 µg/mL DAPI. Tissue sections were washed with TBST for 10 min. An F-SERS dot mixture was diluted in 1× PBS with 10% glycerol and 0.003% Triton X-100. The mixture was then added to tissue slides. The slides were incubated overnight under dark conditions at 4 °C. Coverslips were mounted with mounting medium (DAKO, Carpinteria, CA) after incubation and extensive washing. Tissues were observed with a CLSM. Raman Intensity Analysis of Cell and Tissue Sample. Multiplex targets were characterized by Raman intensity analysis using a confocal Raman system after CLSM site identification. Laser power at the sample was ∼500 µW with a ×100 microscope objective. We used a motorized X-Y stage to detect a Raman signal in the selected target cell or tissue region. The scanning step size was 1 µm in the x and y directions, and the collection time was 10 s for each step. RESULTS AND DISCUSSION Synthesis of F-SERS Dots. Immunoassays for multiplex biomarker detections strongly require improvement in multifunc-

tional probes for biocompatibility, sensitivity, selectivity, and spatial localization. We showed that F-SERS dots were nontoxic, sensitive, and selective in a previous study.25 To perform multiple targeting of three proteins in BASCs, F-SERS dots were successfully synthesized, following our methods (Scheme S-1 in the Supporting Information). We confirmed Raman spectra of F-SERS dots labeled with MT, BT, and NT before the fluorescent silica layers were coated (Figure S-1a in the Supporting Information). Also after treating fluorescent dyes, we obtained TEM images of F-SERS dots (Figure S-1b in the Supporting Information). Peak Height Analysis of F-SERS Dots. We used an analytic method for the calculation of the peak height of the specific Raman band to generalize SERS intensity measurement (Figure 1). We estimated the peak height from the peak value subtracted with baseline value in order to systematically extract the SERS intensity map from the SERS spectra of the tissue sample. Specific Raman bands from respective F-SERS dots were selected at a 1593 cm-1 MT band, a 997 cm-1 BT band, and a 1378 cm-1 NT band without overlapping of spectra. Irregular peaks occur at both sides of specific Raman bands, and we appointed each range of the left and right base at a 8 cm-1 band interval in order to equalize irregular peaks. Average values of band intensities were included in that range. Each point at both ends of the blue line included the average intensity values within each range of the left and right base in parts a-c of Figure 1. The green line denoted peak height and connected the peak top and the middle point of the blue line. The middle point of the blue line is equal to the mean intensities of the left and right bases. (Table S-1 in the Supporting Information provides the summary of calculations for peak height analysis). Quantification analysis Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

1011

using SERS intensities could be used for effective analytical methods in this study since we chose specific bands of Raman labels under the nonoverlapping conditions. Specific Cellular Targeting by F-SERS Dots. We evaluated the capability of F-SERS dots for specific targeting. RAW 264.7 cells were chosen as a model system. These cells express the Sca-1 antigen on their cell surfaces but have no CD34 antigen.32 Sca-1 antibody-conjugated F-SERS dots and CD34 antibodyconjugated F-SERS dots were treated with two different cell suspensions including bovine serum albumin for the prevention of nonspecific adsorption and blocking of free carboxyl groups. F-SERS dots(BT-AF647-Sca1), which were tagged with BT as the Raman active chemical and AF647 (red fluorescent dye) and conjugated with Sca-1 antibody, were bound to the cell surface (Figure 2a). The red fluorescence signal from F-SERS dots was detected at most cell surfaces. The fluorescence signal was not observed in cells treated with F-SERS dots(MT-FITC-CD34) which were tagged with MT as the Raman probe, anchored with FITC (green color), and conjugated with CD34 antibody (Figure 2d). Bright field optical images denote the scanning area in the upper images of parts b and e of Figure 2. The lower images in parts b and e of Figure 2 are SERS intensity maps for the 997 cm-1 BT band (Figure 2b) and the 1593 cm-1 MT band (Figure 2e) corresponding to the region shown in the upper images. The red blocks in Figure 2b represent the SERS intensity of the 997 cm-1 BT band (signal of Sca-1), and the green blocks in Figure 2e represents the 1593 cm-1 MT band (signal of CD34). Parts c and f of Figure 2 depict the Raman spectra for cells treated with F-SERS dots(BT-AF647-Sca1) (Figure 2b) and F-SERS dots(MT-FITC-CD34) (Figure 2e), represented by yellow dots. Cells were treated with F-SERS dots(BT-AF647-Sca1), and Sca-1 proteins on the cell surfaces were detected by fluorescence. SERS signals exhibited strongly intense and highly corrected cell morphologic distributions. However, CD34 proteins were not detected in any cells that were treated with F-SERS dots(MT-FITC-CD34). These results demonstrated that F-SERS dots conjugated with antibodies had the potential of specific and selective protein targeting. CLSM Analysis and SERS Intensity Mapping of BADJ Treated with Three F-SERS Dots(MT-FITC-CD34/BT-AF647-Sca1/NT-AF647-SPC). BASCs are a regional pulmonary stem cell population and exhibit self-renewal, multipotency, and cancer development in the lung. Such cells have been identified at bronchioalveolar duct junctions (BADJ).26-29 The detection and identification of BASCs in tissue under current experimental conditions require fluorescent signal detection of markers, and this may limit its use in some cases such as multiplex targeting or quantification due to its broad emission profiles and it can be photobleached rapidly.26,28,29 Kim et al.26 identified BASCs in murine lung tissue by FACS-based immunofluorescence assay with two surface markers. We identified BASCs coexpressing three markers in single tissue samples with the aid of F-SERS dots. A multiplex immunoassay was performed to detect BASCs in murine lung tissue using different F-SERS dots(MT-FITCCD34/BT-AF647-Sca1/NT-AF647-SPC). Figure 3a showed that a terminal bronchiole (TB) and BADJ in lung tissue were localized by CLSM. The BADJ region was magnified 800 times for more definite recognition of BASCs (Figure 3b). The FITC green color indicated CD34 protein (32) Brachvogel, B.; Pausch, F.; Farlie, P.; Gaipl, U.; Etich, J.; Zhou, Z.; Camerone, T.; Mark, K. V.; Bateman, J. F.; Po ¨schl, E. Exp. Cell Res. 2007, 313, 2730– 2743.

1012

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

expression (Figure 3c), and the AF647 red color indicated Sca-1 or SP-C protein expression (Figure 3d). BASCs coexpress CD34, Sca1, and SP-C, but cells coexpressing only CD34 and Sca-1 are also regarded as BASCs.26,27 We regarded all two-color-merged cells as potential BASC candidates. Several cells contained both green and red colors together in the BADJ region, and the cells were likely classified as BASCs (white arrows, Figure 3f). We could not distinguish between Sca-1 and SP-C proteins through simple CLSM analysis because they revealed AF647 red color fluorescence signals. A portion of BADJ was selected as an area for Raman analysis as white dashed lines in parts c-f of Figure 3 to obtain identical information about each protein. SERS intensity mapping of the areas for Raman analysis selected in CLSM images was performed to identify CD34, Sca-1, and SP-C proteins in BASCs. Maps were differentiated by MT, BT, and NT signals. Maps were provided by the baseline-corrected intensities of the 1593 cm-1 MT band corresponding to the CD34 protein (Figure 4a), the 997 cm-1 BT band corresponding to the Sca-1 protein (Figure 4b), and the 1378 cm-1 NT band corresponding to the SP-C protein (Figure 4c). The maps were very similar to their fluorescent images as indicated by CLSM images. The same brightness scales for the three SERS intensity maps were used to visualize their different intensities. Merged image (Figure 4d) of parts a, b, and c of Figure 4 were taken together with the DAPI-staining image of Figure 3e to illustrate the CD34, Sca-1, and SP-C protein distributions. We tracked and designated the identical scanning area for SERS intensity mapping through a bright field image by CLSM and optical image by microRaman. Therefore, the area for Raman analysis illustrated in Figure 3 and the area in the white dashed line in Figure 4 are all identical. Figure 4d was slightly magnified for the clearness of each mapping result. However, the positions of these three different molecular markers are not identical. This is quite natural since the targeted position of each marker at one cell can be different from one another. The yellow circles in Figure 4 are hypothetical cell regions based on DAPI-staining for the nucleus. BASCs are found in the distal portion of the BADJ.26,28 We selected four cells located in the distal portion of the BADJ with all the MT, BT, and NT signals represented by yellow circles (numbers 1, 2, 3, and 4) and seen in Figure 4d. SERS intensity maps provided tracking information for three BASC proteins through specific spectra differentiated by each SERS signal. Most of SERS signals appeared on the cell surface or cytoplasm region except the nuclei region because of the original expression position of CD34, Sca-1, and SP-C proteins in the cell. In the case of CD34 and Sca-1, they are expressed on the cell surface and SP-C is expressed on the cytoplasm. There is no protein expressed in the nuclei among the three proteins we targeted. In accordance with the positional character of CD34, Sca-1, and SP-C in the cell, they were mostly detected on the outside of the nuclei as shown in Figure 4d. Figure 4e revealed several kinds of Raman spectra. Specific bands from each Raman label was detected due to distinguishable bands without overlap (Figure 4e). In the event, F-SERS dots detected multiple BASC lung tissue targets with SERS intensity mapping. Immunoassays with F-SERS dots required a single tissue slide in rapid, simple, and efficient methodologies. Fluorescence signals of three F-SERS dots were effective for the initial detection and localization of BASCs, and the Raman spectral

Figure 3. CLSM images of murine lung treated with three F-SERS dots(MT-FITC-CD34/BT-AF647-Sca1/NT-AF647-SPC): (a) Bright field image (×400). TB, Terminal bronchiole; BADJ, bronchioalveolar duct junction. Black square indicates the region for BADJ magnification. (b) Highmagnification image of BADJ (×800). Black line indicates an area for Raman analysis. (c) Fluorescent image of FITC. (d) Fluorescent image of AF647. (e) Fluorescent image of DAPI. (f) The merged image of images c, d, and e. White dashed lines indicate areas for Raman analysis. White arrows indicate cell positions with both green and red merged colors for BADJ.

Figure 2. Specific cellular targeting by F-SERS dots. (a) Fluorescent image of RAW 264.7 cells incubated with F-SERS dots(BT-AF647-Sca1). (b) Bright field optical image denoting the scanning area (upper image) and the SERS intensity map for the 997 cm-1 band (lower image). The scale bar dimension is 5 µm. The line at the rim of the cells indicates an area for SERS intensity mapping. (c) Representative Raman spectra at several points indicated by yellow dots in Figure 2b. (d) The fluorescent image of RAW 264.7 cells incubated with F-SERS dots(MT-FITC-CD34). (e) The bright field optical image denoting the scanning area (upper image) and the SERS intensity map for 1593 cm-1 band (lower image). The scale bar dimension is 5 µm. (f) Representative Raman spectra at several points indicated by yellow dots in Figure 2e. Each spectrum was separated by a different offset value to get rid of spectrum overlap.

fingerprints of MT, BT, and NT corresponding to CD34, Sca-1, and SP-C proteins by SERS intensity mapping were very useful for stem cell marker characterization. Raman Analysis for Relative Quantification of Three Proteins Coexpressed in BASCs. Internal and external standards are necessary for objective analysis in the quantitative analysis of Raman measurements.33,34However, in a tissue sample, it is very difficult to introduce internal standard due to several problems associated with strongly absorbing impurity,33 homogeneity, stability, spectral overlapping, and so on. Therefore, we took an average of SERS intensity values of specific Raman MT, BT, and NT bands from all the synthesized F-SERS dots(MT-FITC-CD34/BT-AF647-Sca1/NT-AF647-SPC)to obtain external standards as the denominator. For the quality of the external standard, we determined the optimum experimental condition (33) Bell, S. E.; Sirimuthu, N. M. Chem. Soc. Rev. 2008, 37, 1012–1024. (34) Liu, G. L.; Chen, F. F.; Ellman, J. A.; Lee, L. P. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2006, 1, 795–798.

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

1013

Figure 5. Raman analysis for relative quantification of three proteins coexpressed in BASCs. (a) Average SERS intensities of F-SERS dots labeled with MT, BT, or NT. Average SERS intensities of MT, BT, and NT are, respectively, 1832, 2109, and 4871 au. Data represent mean standard deviation (n ) 5). (b) Relative expression ratios of CD34, Sca-1, and SP-C proteins in each BASC. The graph represents data in Table S-4 in the Supporting Information. Data represent mean standard deviation.

Figure 4. SERS intensity mapping of BADJ: (a) Maps produced through the use of baseline-corrected intensities of the 1593 cm-1 MT band corresponded to the CD34 protein. (b) The 997 cm-1 BT band corresponded to the Sca-1 protein. (c) The 1378 cm-1 NT band corresponded to SP-C protein. (d) The overlaid images of all three maps with nuclear staining images for the illustration of protein cellular distribution. White dashed lines indicate areas for Raman analysis. The yellow circles with Arabic numbers indicate boundaries of BASCs coexpressing CD34, Sca-1, and SP-C. (e) Representative Raman spectra at several points indicated by white dots with Roman alphabet letters in Figure 4d. (i) NT signal, (ii) MT signal, (iii) NT and MT signals, (iv) BT and NT signals, (v) BT and MT signals, and (vi) BT, NT, and MT signals. 1014

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

and applied it to all experiments. The stability of the SERS intensity was confirmed by the specific intensity of the silicon wafer. Each bar in Figure 5a is an average of the SERS intensity values of each of the F-SERS dots labeled with MT, BT, or NT. The pellet consisting of spun-down F-SERS dots was laser-focused, and SERS signals were randomly measured at five pellet sites. We took an average of five SERS intensity values calculated at the specific Raman bands (1593 cm-1 for MT, 997 cm-1 for BT, and 1378 cm-1 for NT). Experiments were conducted for each F-SERS dot at similar concentration and conditions since spontaneous Raman measurements required a number of optimized and controlled factors for reliable SERS measurements.34 To evaluate the amount of CD34, Sca-1, and SP-C proteins in BASCs, we separately added up the SERS intensities of the MT, BT, and NT signal in each BASC (cells 1, 2, 3, and 4) (Table S-2 in the Supporting Information). The peak specific Raman band heights measured in tissues were obtained from the analytic method in Figure 1. Then, the total SERS intensity values

in Table S-2 in the Supporting Information were normalized by dividing with the external standards of 1832 (au) for MT, 2109 (au) for BT, and 4871 (au) for NT, respectively indicated in Figure 5a. By using the normalized results (Table S-3 in the Supporting Information), we calculated the occupying portion of one protein in four cells for the total quantity of each protein in four cells (Table S-4 in the Supporting Information). As shown in Table S-4 in the Supporting Information, the means are distribution ratios of each protein in four cells and the standard deviations are calculated by applying the same ratio of standard deviations in Figure 5a. Figure 5b represents Table S-4 in the Supporting Information results with standard deviations. As a result, we could compare and analyze the quantitative difference of three proteins among four different BASCs. These results showed that the number 2 cell contained the most CD34 proteins compared to other cells, and that the number 4 cell had the highest levels of SP-C and Sca-1 proteins compared with other cells. In addition, the number 3 cell had low expression levels over all proteins. Quantitative Raman analysis with SERS intensity allowed the analysis of the cellular characterization of BASCs coexpressing multiple proteins (CD34, Sca-1, and SP-C). In view of BASCs that may serve as a critical target in lung cancer because their expansion parties are involved in tumor initiation,26 the studies describing BASC-specific markers and identification of BASCs using F-SERS dots with multiple markers have great biological potential. CONCLUSION This study demonstrated the potential of F-SERS dots as an effective tool for targeting and quantitative analysis of multiple proteins in the characterization and identification of BASCs. The

F-SERS dots conjugated with BASC-specific markers were successfully applied to multiplex immunoassays for the detection of BASCs and provided a potential application for relative quantification of target proteins by measurement of SERS intensity. In addition, multiplex immunoassays using F-SERS dots could provide informative characterization of BASCs. Our results strongly suggest that multiplex immunoassays by F-SERS dots are objective and reproducible because we showed analytic standards for SERS peak analysis. Therefore, methods described in current studies can be applicable to multiple target analysis. The targeting of stem cells clarifies more effective stem cellspecific markers and therapy strategies for disease, and this study in detecting BASCs by multifunctional F-SERS dots may contribute to the development of targeting, tracking, and imaging systems for labeling stem cells. Further, immunoassays with F-SERS dots will be applicable to high-throughput screening studies for a variety of biomolecules in the future. ACKNOWLEDGMENT M.-A.W., S.-M.L., and G.K. contributed equally to this work. This work is partly supported by the Nano Systems InstituteNational Core Research Center (NCI-NCRC) program of KOSEF and BK21 Program for Veterinary Science. SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review September 25, 2008. Accepted December 10, 2008. AC802037X

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

1015