Synthesis, Characterization, and Application of Eu ... - ACS Publications

Feb 22, 2005 - ACS Applied Materials & Interfaces 2014 6 (15), 12012-12021 ... Harri Härmä. Analytical Chemistry 2012 84 (11), 4950-4956 ... Bioconj...
0 downloads 0 Views 189KB Size
Anal. Chem. 2005, 77, 2643-2648

Technical Notes

Synthesis, Characterization, and Application of Eu(III), Tb(III), Sm(III), and Dy(III) Lanthanide Chelate Nanoparticle Labels Petri Huhtinen, Mirja Kivela 1 ,† Outi Kuronen, Virve Hagren, Harri Takalo, Heikki Tenhu,† Timo Lo 1 vgren, and Harri Ha 1 rma 1*

Department of Biotechnology, University of Turku, Tykisto¨katu 6, FIN-20520 Turku, Finland

Preparation and characterization of europium(III), terbium(III), samarium(III), and dysprosium(III) polystyrene nanoparticle labels with lanthanide-specific fluorescence properties has been presented. Emulsion copolymerization of styrene and acrylic acid was used to synthesize uniform-sized nanoparticles ∼45 nm in diameter. Europium(III) and samarium(III) lanthanides were chelated with 2-naphthoyltrifluoroacetone and trioctylphosphine oxide to dye the spherical particles, whereas terbium(III) and dysprosium(III) chelate complexes contained a newly synthesized ligand, 4-(2,4,6-tridecyloxyphenyl)pyridine-2,6-dicarboxylic acid. The fluorescence properties of the four lanthanidessincluding a wide Stokes shift, a narrow emission peak, and long fluorescence lifetimeswere retained despite the incorporation into the nanoparticles. Furthermore, the nanoparticles, containing more than 1000 lanthanide chelates, were detectable at label concentrations 3 orders of magnitude lower than the corresponding soluble lanthanide chelate labels. The applicability of the labels prepared was demonstrated by a heterogeneous sandwich-type immunoassay for human prostate-specific antigen, where the lowest limits of detection of 1.6, 2.4, 10.1, and 114.2 ng/L were achieved using europium(III), terbium(III), samarium(III), and dysprosium(III) nanoparticles, respectively. The spectral and functional properties of the lanthanideembedded polystyrene nanoparticles developed here suggest that the technology is applicable for high-sensitivity multicolor assays. Time-resolved fluorometry (TRF) employing long-lifetime fluorescent lanthanide chelate labels has been routinely applied in in vitro diagnostics for two decades. The benefits of TRF technology are based on the spectral and temporal resolution of label detection allowing highly sensitive assays of biological samples. The fluorescence of lanthanide chelates features a wide Stokes shift and a narrow emission band. In addition, the * To whom correspondence should be addressed. Telephone: +358-2-3338063. Fax: +358-2-333-8050. E-mail: [email protected]. † Current address: Laboratory of Polymer Chemistry, University of Helsinki, PB 55, FIN-00014, Helsinki, Finland. 10.1021/ac048360i CCC: $30.25 Published on Web 02/22/2005

© 2005 American Chemical Society

fluorescence has a long lifetime enabling a delayed fluorescence measurement on a scale of microseconds. The delayed measurement increases the sensitivity of the assay since most of the background fluorescence occurring due to biological materials has a short lifetime that decays prior to triggering the detection.1-3 An optimal chelate label would be capable of strong complex formation with a high thermodynamic stability and kinetic inertness. It has a high photochemical stability and quantum yield, it is soluble in aqueous solution, and it possesses a reactive group allowing covalent attachment to biomolecules. Moreover, the functional properties of chelate label-conjugated biomolecules should not be affected by the label.4 Despite thorough research, these properties are rarely all present in a single chelate. Thus compromises are required. β-Diketones, a group of ligands forming two-dentate fluorescent chelates with lanthanide ions,5 have been successfully utilized in combination with Eu(III), Tb(III), and Sm(III) ions in fluoroimmunoassays.6-8 An optimized Eu(III) lanthanide-based chelate complex possessed the highest fluorescence intensity when an aromatic 2-naphthoyltrifluoroacetone (2-NTA) β-diketone was employed as a fluorescenceenhancing ligand generating a luminescence yield of 26 320.9-11 However, aromatic β-diketones are not suitable for Tb(III) measurements. Instead, a corresponding Tb(III) ion enhancement can be achieved using a trimethoxyphenyl derivative of dipicolinic acid as a chelating ligand12,13 producing chelate complex with a luminescence yield of 9048.14 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

Ekins, R. P.; Dakubu, S. Pure Appl. Chem. 1985, 57, 473-482. Jackson, T. M.; Ekins, R. P. J. Immunol. Methods 1986, 87, 13-20. Soini, E.; Lo ¨vgren, T. CRC Crit. Rev. Anal. Chem. 1987, 18, 105-154. Hemmila¨, I. Scand. J. Clin. Lab. Invest. 1988, 48, 389-400. Filipescu, N.; Sager, W. F.; Serafin, F. A. J. Phys. Chem. 1964, 68, 33243346. Siitari, H.; Hemmila¨, I.; Soini, E.; Lo¨vgren, T.; Koistinen, V. Nature 1983, 301, 258-260. Hemmila¨, I. Anal. Chem. 1985, 57, 1676-1681. Bador, R.; Dechaud, H.; Claustrat, F.; Desuzinges, C. Clin. Chem. 1987, 33, 48-51. Dakubu, S.; Ekins, R. P. Anal. Biochem. 1985, 144, 20-26. Hemmila¨, I.; Dakubu, S.; Mukkala, V.-M.; Siitari, H.; Lo ¨vgren, T. Anal. Biochem. 1984, 137, 335-343. Mukkala, V.-M.; Helenius, M.; Hemmila¨, I.; Kankare, J.; Takalo, H. Helv. Chim. Acta 1993, 76, 1361-1378. Dakubu, S.; Hale, R.; Lu, A.; Quick, J.; Solas, D.; Weinberg, J. Clin. Chem. 1988, 34, 2337-2340.

Analytical Chemistry, Vol. 77, No. 8, April 15, 2005 2643

Various luminescent micro- and nanosized particles have been applied as labels in diverse detection technologies. The employment of luminescent particle labels in highly sensitive assays is based on their specific optical properties. For example, semiconductor nanoparticles have a wide excitation range and size- and material-dependent emission spectra,15 up-converting materials are excited at wavelengths above their respective emission wavelengths,16,17 and lanthanide-dyed nanospheres possess the longlifetime fluorescence properties of the respective lanthanide ions.18 The characteristics of commercial 107-nm Eu(III)-dyed polystyrene particles have been thoroughly researched,18,19 and they have been employed as labels in several assay applications to improve the sensitivity and kinetic properties of the assays.20-23 Moreover, the syntheses of nanosized particulate lanthanide labels have been presented recently.24-27 Functionalized europium oxide nanoparticles were prepared by employing microwave-assisted surface chemistry and applied as labels in a competitive high-throughput immunoassay for atrazine.25 Correspondingly, uniform-sized 50nm silica-based Eu(III) and Tb(III) nanoparticle labels were synthesized using a covalent binding-copolymerization method. Furthermore, the Tb(III)- and Eu(III)-dyed particles were utilized in immunometric assays for prostate-specific antigen and hepatitis B surface antigen, respectively.26,27 In the present work, the preparation and characterization of four differently dyed long-lifetime fluorescence nanoparticle labels ∼45 nm in diameter is described for the first time. The nanoparticles were synthesized using styrene and acrylic acid monomers to produce polystyrene nanospheres with a negatively charged surface. The aromatic β-diketone 2-NTA, and a dipicolinic acid derivative 4-(2,4,6-tridecyloxyphenyl)pyridine-2,6-dicarboxylic acid were applied as ligands in the chelate complex formation with europium(III) or samarium(III) and terbium(III) or dysprosium(III), respectively. In combination with 2-NTA, a trioctylphosphine oxide (TOPO) was utilized as a synergistic fluorescence-enhancing agent. The chelate complexes were incorporated into the polystyrene nanoparticles, and the physical and fluorescent properties of the particulate labels were characterized. The results obtained were analyzed with respect to the properties of corresponding (13) Hemmila¨, I.; Mukkala, V.-M.; Latva, M.; Kiilholma, P. J. Biochem. Biophys. Methods 1993, 26, 283-290. (14) Latva, M.; Takalo, H.; Mukkala, V.-M.; Matachescu, C.; Rodriguez-Ubis, J. C.; Kankare, J. J. Luminesc. 1997, 75, 149-169. (15) Chan, W. C.; Nie, S. Science 1998, 281, 2016-2018. (16) Ullman, E. F.; Kirakossian, H.; Singh, S.; Wu, Z. P.; Irvin, B. R.; Pease, J. S.; Switchenko, A. C.; Irvine, J. D.; Dafforn, A. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 5426-5430. (17) Zijlmans, H. J.; Bonnet, J.; Burton, J.; Kardos, K.; Vail, T.; Niedbala, R. S.; Tanke, H. J. Anal. Biochem. 1999, 267, 30-36. (18) Ha¨rma¨, H.; Soukka, T.; Lo ¨vgren, T. Clin. Chem. 2001, 47, 561-568. (19) Soukka, T.; Paukkunen, J.; Ha¨rma¨, H.; Lo ¨nnberg, S.; Lindroos, H.; Lo ¨vgren, T. Clin. Chem. 2001, 47, 1269-1278. (20) Ha¨rma¨, H.; Soukka, T.; Lo ¨nnberg, S.; Paukkunen, J.; Tarkkinen, P.; Lo ¨vgren, T. Luminescence 2000, 15, 351-355. (21) Soukka, T.; Ha¨rma¨, H.; Paukkunen, J.; Lo ¨vgren, T. Anal. Chem. 2001, 73, 2254-2260. (22) Kokko, L.; Sandberg, K.; Lo ¨vgren, T.; Soukka, T. Anal. Chim. Acta, 2004, 503, 155-162. (23) Soukka, T.; Antonen, K.; Ha¨rma¨, H.; Pelkkikangas, A. M.; Huhtinen, P.; Lo ¨vgren, T. Clin. Chim. Acta 2003, 328, 45-58. (24) Tamaki, K.; Shimomura, M. Int. J. Nanosci. 2002, 1, 533-537. (25) Feng, J.; Shan, G.; Maquieira, A.; Koivunen, M. E.; Guo, B.; Hammock, B. D.; Kennedy, I. A. Anal. Chem. 2003, 75, 5282-5286. (26) Ye, Z.; Tan, M.; Wang, G.; Yuan, J. Anal. Chem. 2004, 76, 513-518. (27) Hai, X.; Tan, M.; Wang, G.; Ye, Z.; Yuan, J.; Matsumoto, K. Anal. Sci. 2004, 20, 245-246.

2644

Analytical Chemistry, Vol. 77, No. 8, April 15, 2005

soluble lanthanide labels. The applicability of the labels prepared was demonstrated in a heterogeneous sandwich-type immunoassay for human prostate-specific antigen (PSA), where nanoparticles were coated with monoclonal detection antibodies and used for signal generation. The assay for the model analyte chosen showed a high sensitivity and good accuracy. EXPERIMENTAL SECTION Synthesis of Dipicolinic Acid Derivative Ligands. Three structurally analogous dipicolinic acid derivative ligands, 4-(2,4,6trimethoxyphenyl)pyridine-2,6-dicarboxylic acid (1a), 4-(2,4,6tributoxyphenyl)pyridine-2,6-dicarboxylic acid (1b), and 4-(2,4,6tridecyloxyphenyl)pyridine-2,6-dicarboxylic acid (1c), were synthesized as described.28 The ligands were used for chelation with Tb(III) and Dy(III) lanthanides. 1H NMR spectra of the ligands were recorded at 400 MHz on a JNM-GX-400 spectrometer (JEOL, Peabody, MA) or at 200 MHz on an AM200 spectrometer (Bruker, Ta¨by, Sweden). A Mariner System 5272 mass spectrometer (Applied Biosystems, Foster City, CA) was used for highresolution mass spectrometry. Synthesis and Characterization of Nanoparticles. Emulsion copolymerization of styrene (7 mmol) and acrylic acid (0.58 mmol) monomers was carried out in a sealed round-bottom flask equipped with a magnetic stirrer and an oil bath to control the reaction temperature. Sodium dodecyl sulfate (0.44 mmol) was dissolved in 9 mL of water in the reaction flask after which the monomers were added and the flask was sealed with a septum. The mixture was bubbled with nitrogen to remove radical polymerization inhibiting oxygen and stirred at room temperature for 20 min. The nitrogen in- and outlets were removed, and the flask was placed into the oil bath at 65 °C. Polymerization was initiated after 20 min by injecting potassium peroxodisulfate (0.074 mmol) dissolved in 1 mL of water into the reaction mixture. After 5 h of reaction, the flask was unsealed and the product was cooled to room temperature. The product was filtered through a filter paper (Whatman 2V, pore size 8 µm, Whatman Plc., Brentford, U.K.) and purified by dialyzing using a regenerated cellulose tubular membrane, CelluSep 4 (MWCO 12 000-14 000 g/mol, Membrane Filtration Products Inc., Seguin, TX) for 7 days against distilled water. The dialyzed aqueous particle dispersion was extracted several times with n-hexane and stored as such at 4 °C. The polymer concentration of aqueous particle dispersion was obtained by drying a weighed sample of aqueous dispersion to equilibrium weight in a vacuum. A corresponding reaction was also performed without acrylic acids to synthesize the polystyrene nanoparticles employed as a reference. Fourier transform infrared (FT-IR) spectra of the nanoparticles were measured from freeze-dried particle samples using a Spectrum One FT-IR spectrometer (PerkinElmer Life and Analytical Sciences, Boston, MA). The mean hydrodynamic diameter and the surface charge of particles in 0.1 g/L aqueous solution, pH 9.0, were obtained using a Malvern Zetasizer 3000HS (Malvern Instruments Ltd., Worcestershire, U.K.). Dyeing of the Nanoparticles. Eu(III), Tb(III), Sm(III), and Dy(III) chelates were embedded into nanosized polymer particles using two different chelating structures. Eu(III) and Sm(III) polystyrene nanoparticles were prepared by incubating 300 µL, (28) Hale, R. L.; Solas, D. W. U.S. Patent 4,761,481, 1988.

2.8 mass %, of nanoparticles in 1650 µL of 10 mM carbonate buffer, pH 9.5, containing EuCl3 (3.0 µmol; Alfa Aesar, Ward Hill, MA) or SmCl3 (3.0 µmol, Alfa Aesar), 2-NTA (9.0 µmol), and TOPO (9.0 µmol). Preparation of Tb(III) and Dy(III) nanoparticles was carried out using TbCl3 (3.4 µmol, Alfa Aesar) or DyCl3 (3.4 µmol, Alfa Aesar) and dipicolinic acid derivative (12.6 µmol) in 1350 µL of the same carbonate buffer. The embedding was performed at room temperature overnight. The lanthanide chelate nanoparticles were dialyzed against 8 L of water using a Spectra/Por poly(vinylidene difluoride) membrane (MWCO 300 000 kDa, Spectrum Laboratories, Rancho Dominiquez, CA) overnight. The number of particles was calculated by comparing the scattering signals of the prepared particles to a known Eu(III) particle solution using a Coulter N4 Plus Submicron Particle Size Analyzer (Beckman Coulter, Fullerton, CA). The same sample dilutions and instrument were used for the measurement of the mean particle diameters after the dyeing. The number of lanthanide chelates per particle was calculated by measuring long-lifetime fluorescence signal against a lanthanide calibrator in a corresponding chelating solution using a Victor2 fluorometer (PerkinElmer Life and Analytical Sciences) in time-resolved mode. The measurement protocols set by the manufacturer were applied having the emission wavelengths of 615, 642, 545, and 572 nm; delay times of 400, 500, 50, and 30 µs; and window times of 400, 1400, 100, and 30 µs for Eu(III), Tb(III), Sm(III), and Dy(III), respectively. All four labels were excited at a wavelength of 340 nm. Characterization of Label Fluorescence Properties. A Cary Eclipse fluorescence spectrophotometer (Varian, Inc., Palo Alto, CA) was employed for the fluorescent spectra and lifetime determinations of the nanoparticle labels. Eu(III), Tb(III), Sm(III), and Dy(III) spectra scans were performed using 400, 400, 250, and 100 µs delay times; 400, 1400, 750, and 800 µs window times; 340, 320, 340, and 320 nm excitation wavelengths; and 614, 545, 654, and 485 nm emission wavelengths, respectively. The spectral data obtained were normalized, plotted, and compared to the respective results of the soluble chelate labels. The lifetimes of Eu(III), Tb(III), Sm(III), and Dy(III) particulate labels were measured with excitation wavelengths of 340, 320, 340, and 320 nm and emission wavelengths of 614, 544, 645, and 485 nm, respectively. The lifetimes of three different Tb(III) chelate complexes were determined using an excitation wavelength of 320 nm and an emission wavelength of 545 nm. Cary Eclipse Scan Application, Version 1.1 and Lifetimes Application, Version 1.1 software was employed for the data processing. The lowest detectable label concentrations of nanoparticles and soluble chelate labels were determined by measuring the longlifetime fluorescence of the dilution series using the Victor2 fluorometer in time-resolved mode by applying the manufacturer’s measurement protocols. The particulate labels were diluted in aqueous 0.1% Triton X-100 solution, whereas the soluble Eu(III) and Sm(III) chelate labels were diluted in 200 µL of Delfia Enhancement Solution (PerkinElmer Life and Analytical Sciences) and soluble Tb(III) and Dy(III) chelate labels in a mixture solution of Delfia Enhancement Solution (200 µL) and Delfia Enhancer (50 µL). Differently dyed nanoparticles were visualized with a Nikon Eclipse E600 light microscope (Nikon Corp., Tokyo, Japan) equipped with a Signifier time-resolved imaging apparatus29

(PerkinElmer Life and Analytical Sciences). The nanoparticle labels were diluted with water to obtain the same fluorescence intensity levels for all particle suspensions. Furthermore, a mixture including Eu(III), Tb(III), and Sm(III) nanoparticles was prepared using the same dilutions. One-microliter droplets of the nanoparticle label dilutions were dispensed on a polycarbonate slide. The transmission and time-resolved fluorescence imaging was performed using an ×10 objective and 5-s and 200-ms exposure times, respectively. Moreover, long-lifetime fluorescence imaging was executed by closing the light path with a rotating chopper for a 10-µs excitation at 340 nm and a delay. Thereafter, the rotating chopper opened the light path at lanthanide-specific emission wavelengths, which were 615 nm for Eu(III), 545 nm for Tb(III), 642 nm for Sm(III), and 572 nm for Dy(III), for 600 µs. The images were analyzed with Image-Pro Plus, Ver. 4.0 (Media Cybernetics, Inc., Silver Spring, MD) and Corel Photo-Paint 11 (Corel Corp., Ottawa, Canada) image analysis software. PSA Immunoassay. The nanoparticles prepared were employed as labels in a heterogeneous, sandwich-type fluoroimmunoassay for PSA. PSA-specific monoclonal antibodies H117 and 5A10, which were produced and characterized in our laboratory,30,31 were used for capture and detection, respectively. The nanoparticle labels were coated with 5A10 as described previously.21 The assay was performed by incubating 5 µL of PSA calibrators (PerkinElmer Life and Analytical Sciences) and 25 µL of assay buffer (InnoTrac Diagnostics, Turku, Finland) in H117coated microtitration wells (900 rpm, +23 °C, 30 min). The wells were washed twice, and 1.0 × 109 nanoparticles were added into wells in a 40-µL volume of the same buffer (900 rpm, +23 °C, 2 h). After washing the wells four times, the long-lifetime fluorescence of the nanoparticle labels was measured directly from the surface of wells using the Victor2 fluorometer in time-resolved mode. The manufacturer’s protocols for Eu(III), Tb(III), Sm(III), and Dy(III) measurements were employed. The sensitivities of the nanoparticle label assays were determined with the lowest limits of detection (LLDs), calculated as the analyte concentrations corresponding to the mean of the zero calibrator + 3 SD. RESULTS AND DISCUSSION Synthesis of the Chelates. Three structurally analogous dipicolinic acid derivative ligands were synthesized for the Tb(III) and Dy(III) chelate complex formation (Figure 1). The synthesis of 1a was quantitative, and the yields of syntheses of 1b and 1c were 47% and 40%, respectively. The high-resolution mass spectrometry characterization results of the ligands were analogous with the corresponding computational mass values (1a: calculated 332.08, found 332.09. 1b: calculated 458.22, found 458.23. 1c: calculated 710.50, found 710.53). The lifetimes of the Tb(III)-1a, -1b, and -1c chelate complexes were 667, 734, and 759 µs, respectively, whereas the lifetimes of Dy(III) chelates were not detectable due to short lifetimes and instrumental limitations. (29) Lo ¨vgren T.; Heinonen, P.; Lehtinen, P.; Hakala, H.; Heinola, J.; Harju, R.; Takalo, H.; Mukkala, V.-M.; Schmid, R.; Lo ¨nnberg, H.; Pettersson, K.; Iitia¨, A. Clin. Chem. 1997, 43, 1937-1943. (30) Pettersson, K.; Piironen, T.; Seppa¨la¨, M.; Liukkonen, L.; Christensson, A.; Matikainen, M.-T.; Suonpa¨a¨, M.; Lo¨vgren, T.; Lilja, H. Clin. Chem. 1995, 41, 1480-1488. (31) Piironen, T.; Villoutreix, B. O.; Becker, C.; Hollingsworth, K.; Vihinen, M.; Bridon, D.; Qiu, X.; Rapp, J.; Dowell, B.; Lo ¨vgren, T.; Pettersson, K.; Lilja, H. Protein Sci. 1998, 7, 259-269.

Analytical Chemistry, Vol. 77, No. 8, April 15, 2005

2645

Table 1. Properties of Long-Lifetime Fluorescence Lanthanide Chelate Labels lanthanide

Figure 1. Structures of the dipicolinic acid derivative ligands: (1a) 4-(2,4,6-trimethoxyphenyl)pyridine-2,6-dicarboxylic acid, (1b) 4-(2,4,6tributoxyphenyl)pyridine-2,6-dicarboxylic acid, and (1c) 4-(2,4,6tridecyloxyphenyl)pyridine-2,6-dicarboxylic acid.

label type

Eu(III) particle soluble Tb(III) particle soluble Sm(III) particle soluble Dy(III) particle soluble

lanthanide chelate/particle (chelates) 1400 1000 250 2000

detectable main peaks label lifeconcn excitation emission time a (pM) (nm) (nm) (µs) 0.00048 0.2 0.00082 1.4 0.022 3.3 0.4 110.0

340

614

618

320

544

695

341

645

89

318

485

a The molarities of the nanoparticle labels are calculated by multiplying the number of particles by the Avogadro constant.

Figure 2. FT-IR spectra of the synthesized PSAAc (s) and PS (‚‚‚) nanoparticles. The characteristic peak for the carbonyl group of the acrylic acid is found at wavenumber 1704 cm-1 (arrow).

Preparation and Properties of Lanthanide Chelate-Dyed Nanoparticles. Monomodal negatively charged polystyreneacrylic acid nanoparticles (PSAAc) and polystyrene nanoparticles (PS), used as a reference, were synthesized. The carboxylic groups of acrylic acids gave a greater negative charge allowing the high colloidal stability and functionality necessary for the capture of antibodies. The FT-IR spectra of the PSAAc confirmed the presence of acrylic acid at wavenumber 1704 cm-1, whereas the reference did not show a corresponding peak (Figure 2). Further, PSAAc resulted in a lower ζ-potential than the reference particles, -59.1 ( 1.0 and -54.6 ( 1.1 mV, respectively. Before dyeing, the mean hydrodynamic diameters of the PS and PSAAc nanoparticles were 52 ( 2 and 45 ( 6 nm, respectively. Eu(III) and Sm(III) dyeing was performed together with 2-NTA and TOPO, whereas Tb(III) and Dy(III) were chelated with ligand 1c possessing the highest hydrophobicity and, therefore, favoring the embedding. After dyeing, the mean particle diameters, measured using photon correlation spectroscopy, were 54 ( 11, 41 ( 11, 51 ( 9, 46 ( 6, and 46 ( 17 nm for the PS-Eu(III), PSAAc-Eu(III), -Tb(III), -Sm(III), and -Dy(III), respectively. Hence, the dyeing did not affect the average particle diameter. Moreover, the nanoparticle labels prepared were monomodal unlike commercial nanoparticles of analogous size. Previously we have primarily studied commercial monomodal polystyrene nanoparticles of 107 nm in diameter.18,20,21 The nanoparticles prepared in this study offer an ∼15-fold reduction in particle volume and a 2.5-fold reduction in radius. The radius is directly proportional to the diffusivity according to the Einstein-Stokes equation D ) kT/ 6(π)rµ, where D is Brownian diffusivity, k is Boltzmann’s constant, T is absolute temperature, r is radius, and µ is viscosity of the 2646 Analytical Chemistry, Vol. 77, No. 8, April 15, 2005

medium. Thus, the size reduction achieved should improve the kinetic properties of nanoparticles ∼2-fold. This theoretical assumption is supported by our ongoing studies. Furthermore, in homogeneous nanoparticle label-based energy-transfer assays, diminished particle volume favors the specific signal of the label by lowering the fractional proportion of the nanoparticle core volume that is generating only background signal.22 The lanthanide chelate content of an individual nanoparticle label was 1000-2000 chelates. For an unknown reason, Sm(III) nanoparticles contained not more than 250 chelates/particle (Table 1). The lanthanide chelate quantity determined was clearly lower than the previously reported 31 000 chelates per commercial 107nm Eu(III) nanoparticle.18 Nevertheless, when the volume ratio of 107-nm commercial and ∼50-nm PSAAc nanoparticles is taken into consideration, the commercial particles have less than 2-fold higher label chelate density. A chelate density analogous to our results has been reported with silica-based 40-nm Tb(III) nanoparticles.26 Characterization of the Fluorescence Properties of the Nanoparticles. The fluorescence excitation and emission spectra of the dyed nanoparticles (Figure 3, Table 1) were identical to the respective soluble lanthanide chelate label spectra. Thus, the embedding of lanthanide chelates into nanoparticles had no effect on the optical properties of the lanthanides. We compared the dilution series of nanoparticle and soluble lanthanide chelate labels. All four nanoparticle labels were detectable at concentrations 2-3 orders of magnitude lower than the corresponding soluble lanthanide chelate labels (Table 1). Eu(III) and Tb(III) nanoparticles possessed the highest ratio of specific signal to concentration. The ability to detect soluble Dy(III) and Sm(III) labels was limited by low fluorescence signals caused by their short lifetimes and low quantum yields. However, the high lanthanide chelate density of nanoparticles increased their specific activity, and low concentrations were also detectable with the two lanthanides. Sm(III) labels possessed very low background fluorescence signals based on the instrumental settings, producing high signal-to-background ratios. Accordingly, Sm(III) labels were detectable at lower concentrations than expected in proportion to the other three lanthanides.4 Time-resolved fluorescence images were taken from the particulate label suspensions using a Nikon Eclipse E600 light microscope equipped with a Signifier time-resolved imaging apparatus. In Figure 4A, a transmission photograph shows the positioning of the four droplets on the polycarbonate slide. Three

Figure 5. Calibration curves of the four different nanoparticle labelbased immunoassays for PSA. The dashed lines indicate the detection limits calculated as the mean of the zero calibrator + 3 SD.

Figure 3. Normalized excitation (A) and emission (B) spectra of four lanthanide chelate nanoparticle labels measured in aqueous 0.1% Triton X-100 solution.

Figure 4. Visualization of Eu(III), Tb(III), and Sm(III) chelate dyed nanoparticles. A transmission picture of the droplets of Eu(III), Tb(III), and Sm(III) nanoparticles in water and a droplet of a nanoparticle label mixture solution containing the same three nanoparticles (A). False-color photographs of time-resolved images taken with 615-, 545-, and 642-nm filters to visualize Eu(III), Tb(III), and Sm(III) nanoparticle labels, respectively (B)-(D).

of the spots represent individual Eu(III), Tb(III), and Sm(III) particulate label solutions, and the fourth spot is a mixture of the three particulate label suspensions. The false-color photographs (Figures 4B-D) taken with 615-, 545-, and 642-nm filters visualize Eu(III), Tb(III), and Sm(III) nanoparticles, respectively. Interfer-

ence was not detected from the adjacent label droplets, nor were the distinct label signals of the mixture spot affected by other labels presented in the mixture droplet. The signal intensities of the individual visualized label droplets and the mixture droplet were analogous. A visualization of the Dy(III) nanoparticle labels was not performed. Low particle stock concentration, short fluorescence lifetime, and instrumental limitations were the main reasons for low signal yield. PSA Immunoassay. The potential of the long-lifetime fluorescence particulate labels prepared was demonstrated in a heterogeneous sandwich-type immunoassay for PSA. Typical calibration curves, based on three replicates of calibrators, are presented in Figure 5. The equations of curves were y ) 23076x0.934 (R2 ) 0.998), y ) 50307x0.981 (R2 ) 0.996), y ) 879x1.003 (R2 ) 0.998), and y ) 616x0.991 (R2 ) 0.994) employing Eu(III), Tb(III), Sm(III), and Dy(III) nanoparticles, respectively; thus, good linearity was achieved. The concentrations CVs were from 1.1% to 18.7% except for the lowest calibrator of each curve, whose concentration CVs were ∼30%. The LLDs, calculated as the mean of zero calibrator + 3 SD, were 1.6, 2.4, 10.1, and 114.2 ng/L using Eu(III), Tb(III), Sm(III), and Dy(III) nanoparticles, respectively. The values correspond to 50 fM-3.6 pM concentrations or 250 zmol-18 amol chemical amounts. The sensitivity is based on the multivalent character of the nanoparticle bioconjugates. The affinity is increased and the specific activity is also high enough even with the smaller nanoparticles.21 The LLD values achieved with Eu(III)- and Tb(III)-dyed nanoparticles are analogous with the highsensitivity PSA assays introduced before.23,26,29,32,33 Sm(III)- and Dy(III)-dyed particles showed high sensitivity, too. Hence, the nanoparticles could be employed as labels in a quadruple-label fluorometric immunoassay achieving improved sensitivity compared to an earlier soluble lanthanide label-based assay application.34 Previously, multicolor assays with time-resolved fluorometry have been employed, e.g., in hybridization analysis,35,36 immu(32) Ferguson, R. A.; Yu, H.; Kalyvas, M.; Zammit, S.; Diamandis, E. P. Clin. Chem. 1996, 42, 675-684. (33) Black, M. H.; Grass, L.; Leinonen, J.; Stenman, U.-H.; Diamandis, E. P. Clin. Chem. 1999, 45, 347-354.

Analytical Chemistry, Vol. 77, No. 8, April 15, 2005

2647

noassays,37,38 and immunocytochemistry.39,40 The nanoparticle labels presented are applicable in the corresponding multicolor applications with lowered detection limits. Moreover, the multiformity could be further increased by employing, for example, a prompt fluorophore and phosphorescent metalloporphyrins palladium(II) and platinum(II), and it would still be possible to record spectrally well-resolved emission peaks of all the labels used. CONCLUSIONS The preparation and characterization of monomodal, negatively charged, nanosized polystyrene particles and their dyeing with four long-lifetime fluorescent lanthanide chelate labels, Eu(III), Tb(III), Sm(III), and Dy(III), is presented. The nanoparticles are employed as labels in a heterogeneous sandwich-type fluoroimmunoassay for PSA. To our knowledge, this is the first publication describing the preparation and application of four spectrally distinct polystyrene nanoparticle labels. (34) Xu, Y. Y.; Pettersson, K.; Blomberg, K.; Hemmila¨, I.; Mikola, H.; Lo ¨vgren, T. Clin. Chem. 1992, 38, 2038-2043. (35) Iitia¨, A.; Liukkonen, L.; Siitari, H. Mol. Cell. Probes 1992, 6, 505-512. (36) Samiotaki, M.; Kwiatkowski, M.; Ylitalo, N.; Landegren, U. Anal. Biochem. 1997, 253, 156-161. (37) Mitrunen, K.; Pettersson, K.; Piironen, T.; Bjo¨rk, T.; Lilja, H.; Lo¨vgren, T. Clin. Chem. 1995, 41, 1115-1120. (38) Heinonen, P.; Iitia¨, A.; Torresani, T.; Lo¨vgren, T. Clin. Chem. 1997, 43, 1142-1150. (39) Bjartell, A.; Siivola, P.; Hulkko, S.; Pettersson, K.; Rundt, K.; Lilja, H.; Lo ¨vgren, T. Prostate Cancer Prostatic Dis. 1999, 2, 140-147. (40) Soini, A. E.; Kuusisto, A.; Meltola, N. J.; Soini, E.; Seveus, L. Microsc. Res. Technol. 2003, 62, 396-407.

2648

Analytical Chemistry, Vol. 77, No. 8, April 15, 2005

The characterization of the dyed nanoparticles confirmed that the properties of the lanthanides were not affected by the dyeing procedure. The 100-1000-fold lower concentrations of nanoparticle labels than the corresponding soluble labels were detectable. Highly dyed detection antibody-coated nanoparticles were applicable as labels in high-sensitivity heterogeneous fluoroimmunoassays. The assay sensitivities achieved were comparable with the highest sensitivities reported before. The nanoparticle labels presented enable interesting possibilities for multiplex protein as well as nucleic acid assay applications. Enhanced assay kinetics is achievable due to the reduced diameter of the nanoparticle labels, and direct measurement of the particle fluorescence leads to a simplification in measurement procedures. Furthermore, the number of labels in an assay could be further increased with suitable long lifetime and prompt dyes possessing spectrally individual emission peaks. ACKNOWLEDGMENT We thank Mr. Mark Smith for revising the language of the manuscript. This work was supported by the National Technology Agency of Finland (TEKES).

Received for review November 5, 2004. Accepted January 20, 2005. AC048360I