Fluorescent Nanometer Microspheres as a Reporter for Sensitive

We succeeded in using 40 nm FRET (fluorescence resonance energy transfer) microspheres conjugated to antibodies as the fluorescent reporters to perfor...
0 downloads 0 Views 196KB Size
Download PDF
Bioconjugate Chem. 2005, 16, 194−199

194

Fluorescent Nanometer Microspheres as a Reporter for Sensitive Detection of Simulants of Biological Threats Using Multiplexed Suspension Arrays L. Wang,*,† K. D. Cole,† A. K. Gaigalas,† and Yu-Zhong Zhang‡ Biotechnology Division, National Institute of Standards and Technology, Gaithersburg, Maryland 20899-8312, and Molecular Probes, Inc., 4849 Pitchford Avenue, Eugene, Oregon 97402-9165. Received August 18, 2004; Revised Manuscript Received November 18, 2004

We succeeded in using 40 nm FRET (fluorescence resonance energy transfer) microspheres conjugated to antibodies as the fluorescent reporters to perform the multiplexing suspension array measurements on two simulants of biological threats, ricin (A chain) and a crude spore preparation of Bacillus globigii (Bg). The microspheres were impregnated with two types of fluorophores in equal number (∼140 fluorophores in total per microsphere) and displayed bright PE-like fluorescence via a fluorescence resonance energy transfer mechanism. Activated microspheres (aldehyde groups) were directly coupled to antibodies and used to form sandwich-type immunoassays in a suspension array. For the crude preparations of Bg, the assay sensitivity using antibody-conjugated microspheres is an order of magnitude higher than that using the conventional fluorescent reporter, R-phycoerythrin (PE). Using the microspheres, Bg at the concentration of 5 ng/mL can be easily detected. For ricin, the assay sensitivity was similar to that obtained using PE as the reporter, but washing the reaction mixtures resulted in the fluorescence signals that were 2-3 times higher compared to those using PE. Ricin at a concentration of 1 ng/mL can be readily identified. Importantly, the two simulants do not interfere with each other in the multiplexing experiments. The 40 nm FRET microspheres are a new sensitive alternative as fluorescent reporters for detection in suspension arrays.

INTRODUCTION

Over the past 20 years, flow cytometry has become an indispensable measurement tool for clinical diagnostics and basic research. It is exquisitely suited to multiplexing analysis because the technology is capable of analyzing multiple wavelengths of emitted fluorescence from thousands of particles per second with excellent sensitivity (1, 2). When compared to conventional flow cytometers with multiple lasers for the respective fluorophores, the latest design of the suspension array flow cytometer (Luminex 100 from Luminex Corp.) provides high sensitivity, high throughput, and multiplex detection at a lower cost. The instrument is designed to take advantage R-phycoerythrin (PE) as the reporter. The suspension arrays utilize polystyrene microbeads (5.5 µm in diameter) that are impregnated with precise ratios of red and infrared fluorescent dyes yielding an array of 100 bead sets, each with a unique intensity address in the instrument. A 635 nm, 10 mW red diode laser excites the two classification fluorophores embedded in the microbeads, and a 532 nm, 13 mW yttrium aluminum garnet (YAG) laser provides the excitation of the fluorescent reporter PE. The immunoassays adopt a typical sandwich immunoassay format where antigen-specific capture antibodies are immobilized on the microbeads, antigen is then added and allowed to bind to capture beads, and the bound antigen is subsequently recognized by the biotinylated secondary antibodies that are identified by streptavidin* Corresponding author: Lili Wang, National Institute of Standards and Technology, 100 Bureau Dr., Stop 8312, Gaithersburg, MD 20899-8312, (301) 975-2447, [email protected]. † National Institute of Standards and Technology. ‡ Molecular Probes, Inc.

PE (3). Because the reporter fluorescence does not overlap with the classification signal, no fluorescent compensation is required. High-throughput suspension arrays have been used for disease diagnostics (4) and for detections of biological threats because of the sensitivity, specificity, and multiplex capabilities (5, 6). McBride and co-workers (6) have developed and tested a fully autonomous pathogen detection system for monitoring airborne biological threat agents. This system is designed to automate the sample collection, preparation, and multiplex detection of biological threat agents (bacteria, viruses, and toxins) using suspension arrays. The technological characteristics exhibit advantages over other immunoassay formats, such as enzyme-linked immunosorbent assay (ELISA) and time-resolved fluorescence (TRF) assays (7). To fulfill the need of the public health against the biological threats, it’s important to develop simulants of biothreat agents for the purposes of rapid and accurate detection, remediation, and personnel training. In the present study, two simulants of biological threats, ricin (A chain) and a spore preparation of Bacillus globigii (Bg), have been chosen as the model systems for detection using suspension arrays. Ricin, a glycoprotein toxin derived from castor plant beans, is one of the most poisonous naturally occurring substances known and has great potential for harm due to its wide availability and superior stability. The A chain portion of the ricin heterodimer has the enzymatic activity that binds and depurinates a specific adenine in the 28S subunit of ribosomes. The B chain portion of ricin is required for it to efficiently enter cells (8, 9). Bg is a nonpathogenic relative of Bacillus anthracis and the spore size and properties are similar to Bacillus anthracis

10.1021/bc0498020 CCC: $30.25 © 2005 American Chemical Society Published on Web 12/23/2004

Fluorescent Nanometer Microspheres

spores. The dormant spores can endure a wide range of extreme environmental stresses while retaining the capacity to rapidly return to vegetative growth (germination process) when environmental conditions are favorable (10). These two simulants (Bg and ricin A-chain) can safely be used as internal controls to increase the confidence of the results obtained from suspension arrays. PE has an extremely high extinction coefficient and high fluorescence quantum yield and is widely used in conventional sandwich immunoassays. The Luminex 100 instrument was specifically designed to use PE as the fluorescent reporter. Since new types of brightly fluorescent reporter molecules are being developed continuously, the choice of the fluorescent reporter is likely to play an important role in increasing the sensitivity and reliability of immunoassays using liquid suspension arrays (manuscript in preparation). Although fluorescent microspheres (µm size) are widely used as standards for flow cytometry and fluorescence microscopy (11), nanometer-sized microspheres imbedded with fluorophores have been developed and conjugated to antibodies or streptavidin for immunology studies using flow cytometry, fluorescence microscopy, and ELISAs (12-15). These brightly fluorescent microspheres possess many advantages, such as high photostability and signal-to-noise ratio, environmental insensitivity, and incorporation of multiple dyes to make a set of probes that fluoresce at different wavelengths with a single excitation wavelength. In the present study, 40 nm FRET microspheres impregnated with two types of fluorophores were used as the fluorescent reporter for the sandwich assays detecting both ricin A chain and Bg simultaneously. We have focused on the application of these new FRET microspheres for multiplex assays and compared the results to PE, that is the reporter with the best sensitivity and greatest number of previous applications. MATERIALS AND METHODS

An aqueous suspension of 40 nm FRET microspheres (∼3.0 × 1014/mL) with aldehyde functional groups on the surface was from Molecular Probes (Eugene, OR).1 These microspheres are impregnated with two types of orange BODIPY dyes in equal number (∼140 fluorophores in total per microsphere) and display bright PE-like fluorescence via a fluorescence resonance energy transfer mechanism. The dried preparation of Bacillus globigii (Bg, Bacillus atropheus) spores (lot Sept. 17, 2002) was obtained from U. S. Army Edgewood Chemical Biological Center (Edgewood, MD). The spore preparation gave an average of 1 × 105 colonies per µg of powder when suspended in water or buffer. Stock spore suspensions in PBS were stored at 4 °C. Ricin A chain (1 mg/mL, #L1190, Vector Laboratories, Burlingame, CA) was diluted in PBS containing 5 mM dithiothreitol and 1% (w/w) BSA. Diluted solutions of the Bg spore and ricin A chain were prepared in PBS and used as the multiplex antigen samples. The polyclonal antibody against Ricinus communis agglutinin I & II was also purchased from Vector Laboratories. The antibody stock solution was prepared by adding 0.5 mL of water to obtain a final concentration of 2 mg/mL in phosphate buffer, pH 7.8 (10 mM phos1 Certain commercial equipment, instruments, and materials are identified in this paper to specify adequately the experimental procedure. In no case does such identification imply recommendation or endorsement by the National Institute of Standards and Technology, nor does it imply that the materials or equipment are necessarily the best available for the purpose.)

Bioconjugate Chem., Vol. 16, No. 1, 2005 195

phate, 0.15 M NaCl, 0.08% sodium azide, and sucrose). The polyclonal antibody against Bg was provided by Tetracore Inc. (Gaithersburg, MD) and was dissolved in phosphate-buffered saline, pH 7.4, (0.01 M phosphate, 0.138 M NaCl, 0.0027 M KCl, 0.05% (w/w) NaN3) at a concentration of 3.6 mg/mL. The immobilization of the polyclonal antibodies to Luminex carboxylated microspheres was accomplished through formation of a carbodiimide bond and carried out according to the manufacturer-recommended procedures. A FluoReporter Mini-Biotin-XX-Protein Labeling Kit from Molecular Probes was used for biotinylation of the antibodies, and the coupling reaction of biotins to the antibody was carried out according to the manufacturerrecommended procedures. The conjugation of the antibodies to 40 nm FRET microspheres was through the formation of an amide bond between amine groups of the antibodies and aldehyde groups on the surface of microspheres (16) and was carried out as follows. A 20 µL of antibody stock solution was diluted into 400 µL water, and then a proper volume of 40 nm FRET microsphere suspension was slowly added to the antibody solution. The labeling reaction carried out overnight with gentle vortexing. To prepare antibody-conjugated microspheres with different molar ratios between antibody and microsphere, the volume of microsphere suspension added was adjusted while the amount of antibody was kept constant. The agarose gel electrophoresis was used to detect the fluorescent products of the labeling reactions and help out finding of the optimal molar ratio of antibody and microsphere. A commercial Luminex 100 instrument (Luminex Corp., Austin, TX) was used to perform the multiplexing sandwich assays detecting surrogates of biological threat agents, ricin A chain and Bg. Polyclonal antibodies coupled to specific Luminex beads (1500-3000 beads, 14 µL) were used for the primary analyte-capture step. After incubation with analytes (20 µL) for 20 min with gentle vortexing, the biotin-labeled antibodies (30 µL, 0.8 µg) were added and incubated for 20 min to form the sandwich-type complexes. PE-labeled streptavidin (10 µL, 1.5 µg) was then added as the conventional fluorescent reporter for the detection of ricin toxin and Bg. The final volume of the reaction mixture was 74 µL. On the other hand, microsphere-conjugated antibodies were utilized as an alternative, bright fluorescent reporter to detect the two analytes. The addition of the prior labeled antibodies combines the last two steps in conventional Luminex sandwich assays (Figure 1). When washing steps were employed, the assays were performed in 1.5 µL centrifuge tubes. The sandwich reaction mixtures were centrifuged at a g-force of 9000 for 2 min. The supernatant was discarded and 500 µL of blocking buffer, pH 7.4, (0.01 M phosphate, 0.138 M NaCl, 0.0027 M KCl, 1% (w/w) bovine serum albumin, 0.05% (w/w) NaN3) was added. The centrifugation step described above was repeated, and the supernatant was again discarded. After the addition of 74 µL of blocking buffer, the reaction mixtures were vortexed and transferred to a 96-well plate (Model P, Corning Incorporated, Corning, NY) and prepared for detection. For the multiplexing sandwich assays, the blank signals were obtained in the absence of the antigens and were subtracted from the fluorescence signals of the sandwich complexes in the presence of the two analytes. RESULTS AND DISCUSSION

The various steps of sandwich assays are illustrated in Figure 1. The analyte is recognized by the polyclonal

196 Bioconjugate Chem., Vol. 16, No. 1, 2005

Wang et al.

Figure 1. The schematic of immunological assays implemented in the study. As a first step, an analyte such as ricin A chain, for example, is bound to beads coated with a layer of polyclonal antibodies raised against ricin. Subsequently, biotinylated antibodies are added to the suspension in conventional immunoassays (2). Finally R-phycoerythrin (PE)-labeled streptavidin as the fluorescent reporter is added to the suspension to conclude the conventional sandwich assay (3). Alternatively, we have employed antibodyconjugated 40 nm FRET microspheres (MS) as the fluorescent reporter for the sandwich assay (2′). The 40 nm FRET microsphere is impregnated with as many as 140 fluorophores and is highly fluorescent compared to PE. The sizes of beads (5.5 µm) and 40 nm microspheres are not proportional in the illustration.

antibody immobilized on the surface of the beads (Step 1). Used in conventional immunoassays, biotin-labeled antibodies are subsequently bound to the analytes to form sandwich-type complexes (Step 2). Last, R-phycoerythrin (PE)-labeled streptavidin is added as the fluorescent reporter for intensity measurements by the instrument (Step 3). Alternatively, the antibody conjugated microspheres (40 nm in diameter) replaces the biotin-labeled antibodies and the strepavidin-PE steps in a single step (Step 2′). Since each 40 nm FRET microsphere, on the average, contains about 140 fluorophores, and a resonance energy transfer method (equal numbers of donor and acceptor) is used, about 70 fluorophores are sufficient for the final emission. It is known that a single PE is equivalent to ∼20 fluorophores. Therefore, when the microspheres are used as the fluorescent label, in theory, the fluorescence signal should be about 3.5 times higher than that using R-PE as the fluorescent reporter. Since the 40 nm FRET microspheres are largely made of hydrophobic polystyrene, however, effort may be needed for preventing nonspecific adsorption. Because of light scattering by the 40 nm polystyrene spheres, the absorbance of the dye molecules cannot be accurately measured through simple subtraction of the absorbance from the blank microspheres. The absorbance below 500 nm is affected by the light scattering. Wittmershaus and co-workers have measured the absorbance of different BODIPY dyes embedded in microspheres via the method of index matching (17). Given that the absorption of dye molecules impregnated in microspheres should be proportional to the fluorescence signal obtained at the same excitation wavelength when the emission

wavelength is held constant (18), we measured the excitation spectrum of the dye-embedded microspheres. The excitation and emission spectra of the microspheres are shown in Figure 2. The maximal absorbance and fluorescence are at 538 nm and at 570 nm, respectively. These properties are ideal for applications using Luminex 100 instruments that are equipped with a 532 nm laser beam and a 575 ( 12 nm band-pass filter for the excitation and detection of fluorescent reporters, respectively. As shown in the inset of Figure 2, two types of BODIPY dyes are needed to maximize fluorescence signal for detections using Luminex 100 instrument. Upon excitation, the first BODIPY fluorophores whose emission spectrum shown as dash dot line in the inset will transfer its energy to the second BODIPY dye molecules (shown as dash line) via dipole-dipole interactions with ∼95% transfer efficiency (19). Because these fluorophores are imbedded in the microspheres, their photostabilities are superior to the commonly used R-PE (17, 20). To minimize nonspecific adsorption of the microspheres to various hydrophobic assay components, the reactive aldehyde groups on the outer surfaces of the microspheres were saturated with antibodies. We have used agarose gel electrophoresis to detect the fluorescent products of the labeling reactions between antibodies and aldehyde microspheres. Figure 3 shows the gel electrophoresis of the conjugation reactions with various molar ratios of antibody and microsphere as well as pure reactive microspheres as a control (lane 1 in Figure 3A). Since the polystyrene microspheres are neutral on the outer surface, they hardly move into the gel. At relative low molar ratios of ricin antibody and microsphere from

Fluorescent Nanometer Microspheres

Figure 2. Normalized excitation (short dash line) and emission (solid line) spectra of the 40 nm FRET microspheres suspended in water. The fluorescence was monitored at 570 nm for the excitation spectrum and the excitation wavelength was set at 532 nm for the emission spectrum, respectively. The inset displays emission spectra (arbitrary units) of 40 nm microspheres with one type of BODIPY dye (dash and dash dot lines) and FRET microspheres imbedded with two types of BODIPY dye (solid line). The excitation wavelength was set at 490 nm for the spectral comparison. The Stokes shifts for both types of BODIPY dye are about 20 nm. The slits for excitation and emission monochromators were set at 4 nm for these measurements.

Figure 3. Agarose gel electrophoresis of microsphere preparations. The samples were electrophoresed for 3 h at 6 V/cm in 0.75% agarose using 0.5 × TBE buffer (45 mM TRIS, 45 mM boric acid, 1 mM EDTA, pH 8.3). Samples with diluted with 4% Ficoll 400 and a trace of bromphenol blue before loading. The arrow indicts the sample loading position. (A) The samples are microspheres (lane 1) and antibody-conjugated microspheres with various molar ratios of antibody against ricin and microsphere: 4:1 (lane 2), 8:1 (lane 3), 16:1 (lane 4), 24:1 (lane 5), 32:1 (lane 6), 48:1 (lane 7), 64:1 (lane 8), 48:1 from earlier preparation (lane 9), and 64:1 from earlier preparation (lane 10). (B) The conjugation reactions with four different molar ratios of antibody against Bg and microsphere: 32:1 (lane 1), 48:1 (lane 2), 64:1 (lane 3), 96:1 (lane 4). The gel was placed on a Dark Reader illuminator (output 400 nm to 500 nm, Clare Chemical Research, Denver, CO) and photographed using the supplied filter.

4:1 to 32:1 (Figure 3A: lane 2 to lane 6), not all microspheres are labeled with antibodies given that bare microspheres are seen near the loading well. At the ratios of 48:1 (Figure 3A: lane 7 and 9) and 64:1 (Figure 3A: lane 8 and 10), almost all the microspheres are conjugated with antibodies that are shown by the distinct

Bioconjugate Chem., Vol. 16, No. 1, 2005 197

Figure 4. Standard curves of the immunoassays for ricin A chain generated by using two different fluorescent reporters, PE (circle and solid line) and antibody-conjugated microspheres with the molar ratio of antibody and microsphere of 48:1 (triangle and solid line), and implementing the two-step washing after completion of the sandwich complexes with the use of antibody-conjugated microspheres (open triangle and dot line) or PE (open circle and dot line). The error bars in the figure show the standard deviations obtained from multiple experimental repeats (N ) 4). The inset shows the standard curves in the low concentration region of ricin A chain where the fluorescence signal corresponds linearly to ricin concentration. Various lines in the inset are the linear fitting curves.

fluorescent band that moved into the gel. The conjugation reactions (Figure 3A: lane 9 and 10) were made two weeks earlier than those (Figure 3A: lane 7 and 8), and the antibody-conjugated microspheres were stable as shown in the figure. Both ratios (48:1 and 64:1) were used to perform the sandwich assays and gave similar results. Microspheres conjugated with the antibodies against Bg run more slowly in agarose gel than those labeled with anti-ricin antibodies (Figure 3B). At two ratios of Bg antibody and microspheres, 32:1 (Figure 3B, lane 1) and 48:1 (Figure 3B, lane 2), there were excess microspheres present near the loading wells and the coupling reactions were inefficient. With the increase of this ratio to 64:1 (Figure 3B, lane 3) and 96:1 (Figure 3B, lane 4), however, most microspheres were conjugated with antibodies, which could be visualized as a fluorescent band in the gel. The conjugated microspheres with the ratio of 64:1 were used for Bg to carry out the multiplexing experiments. The excitation and emission spectral shapes of these conjugated microspheres are the same as those shown in Figure 2. Figure 4 displays the standard curves of the multiplexing immunoassays for ricin A chain in the presence of Bg using two different fluorescent reporters, PE (circle and solid line) and antibody-conjugated microspheres at the molar ratio of ricin antibody and microsphere of 48:1 (triangle and solid line). The two standard curves agree with each other. At low ricin concentrations (e20 ng/mL) the fluorescence signal reacts with ricin in a linear fashion, and at high concentrations of ricin (g20 ng/mL) the signal slowly approaches a plateau. Because of the possibility of the nonspecific adsorption by the microspheres, we further explored the effect of implementing a two-step washing step prior to the fluorescence detection. With the addition of the washing step, the standard curve with the use of antibody-conjugated microspheres (open triangle and dot line) was obtained with a highly improved sensitivity. In contrast, with the use of PE, the

198 Bioconjugate Chem., Vol. 16, No. 1, 2005

Figure 5. Standard curves of the immunological assays for Bacillus globigii (Bg) generated by using two fluorescent reporters, PE (circle and solid line) and antibody-conjugated microspheres with the molar ratio of antibody and microsphere of 64:1 (triangle and solid line), and employing the washing step after formation of the sandwich complexes in the presence of the antibody-conjugated microspheres (open triangle and dot line) or PE (open circle and dot line). The error bars show the standard deviations obtained from multiple experimental repeats (N ) 4). The inset shows the standard curves in the low concentration region of Bg where the fluorescence signal responds to antigen concentration in a linear fashion. The two lines in the inset are the linear fitting curves.

assay sensitivity has dropped (open circle and dot line). The inset of the figure shows the standard curves in the low concentration region of the analyte where the fluorescence signal from the reporter responds linearly to analyte concentration. The larger slope of the linear fitting curve demonstrates the high sensitivity of ricin A chain detection by using the antibody-conjugated microspheres. The fluorescence enhancement achieved by the washing step is likely due to the minimization of the selfquenching between 40 nm dye-imbedded microspheres. As we anticipated, the fluorescence signals from the sandwich complexes with conjugated microspheres are 2 to 3 times larger than those using PE as the fluorescent reporter. Additionally, the presence of Bg did not interfere with the detection of ricin A chain since the fluorescence signals of the ricin sandwich complexes are about the same in the presence and absence of Bg (data not shown). The error bars in the figure show the standard deviations obtained from multiple experimental repeats. The samples of Bg spores used in this study are crude preparations that would be representative of an environmental sample. The Bg samples must be constantly mixed to prevent settling during sampling and during the antigen capture step. The biological activity of these samples can be expressed in colony forming units (∼1 × 105 colonies per µg), but comparison to other results done with different preparations of spores is not very meaningful because the total antigen content is not well defined in these samples. There is a real need for developing well-defined spore samples in both highly pure and crude forms. In the absence of standardized spore preparations, it makes the most sense to look at the relative differences in the detection limits when comparing the different reporters using the same samples. The standard curves of the sandwich assays for Bg in the presence of ricin A chain are shown in Figure 5 with the use of two fluorescent reporters, PE (circle and solid line) and antibody-conjugated microspheres at the molar ratio of Bg antibody and microsphere of 64:1 (triangle

Wang et al.

and solid line). The assay sensitivity by using antibodyconjugated microspheres is much higher than that using the conventional PE. For instance, the fluorescence signals of the sandwich complexes at the Bg concentration of 5 ng/mL are distinctively above the signal of the negative control. However, with the use of PE as the reporter, the fluorescence signals of the formed complexes and of the negative control are about the same at the Bg concentration of 50 ng/mL. McBride and co-workers (5) showed that the sensitivity of Bg detection (using PE as the reporter) in suspension arrays was more than an order of magnitude higher compared to that using ELISAs. We further explored the detection of Bg spores using the microspheres by adding a washing step. The fluorescence signals were only slightly improved by the washing step (open triangle and dot line) compared to the signals without the washing step (triangle and solid line). Similar to ricin assays shown in Figure 4, applying the washing steps with the use of PE as the reporter results in decrease of signals (open circle and dot line) when compared to the signals without the washing step (circle and solid line). The high fluorescence signals obtained with the microspheres are likely attributed to several factors. As described earlier, each microsphere could emit, in principle, a ∼3.5 times higher fluorescence signal than that of a single PE molecule. Additionally, because of the large size of Bg spore (∼ 1 µm long and 0.7 µm wide) (21), it could be recognized by multiple polyclonal antibodies raised against Bg. With the use of microsphere-labeled antibodies, the signals of formed Bg complexes at high Bg concentrations are about 6 times higher than that using PE-conjugated antibodies (Figure 5). This implies that self-quenching between PE molecules is likely at high analyte concentration. The FRET microspheres coated with a layer of Bg antibodies may be less susceptible to self-quenching and therefore more versatile in recognizing the analyte (a crude spore preparation of Bg). The inset of the figure shows the standard curves in the low concentration region of the analyte where the fluorescence signal from antibody-conjugated microspheres responds to Bg concentration in a linear fashion. Using PE as the fluorescent reporter, the signals were hardly distinguishable from the background at the Bg concentration of 50 ng/mL. CONCLUSIONS

To detect small amount of biological threats, it is critical to search for better fluorescent reporters that give brightest possible fluorescence for detection using highthroughput suspension arrays. In the present study, we have used 40 nm FRET microsphere-conjugated antibodies as the fluorescent reporters that emit about 3 times higher fluorescence than that of the conventional PE reporter used in suspension arrays to perform the multiplexing array experiments. In the case of Bg, the assay sensitivity using antibody-conjugated microspheres is an order of magnitude higher than that using the conventional PE without the implementation of the washing step. With the use of antibody-coupled microspheres, Bg at the concentration of 5 ng/mL can be easily detected. In the case of ricin A chain, the assay sensitivity is as good as that using PE as the fluorescent reporter without the washing step. While employing the washing step, however, the fluorescence signals of the sandwich complexes are 2-3 times higher than those using PE. Ricin A chain at concentration of 1 ng/mL can be readily

Bioconjugate Chem., Vol. 16, No. 1, 2005 199

Fluorescent Nanometer Microspheres

identified. Importantly, the two simulants do not interfere with each other in the multiplexing experiments. Thus, the present 40 nm FRET microspheres can be visualized as a unique class of fluorescence reporters for sensitive detections using suspension arrays. ACKNOWLEDGMENT

The authors are indebted to D. Hancock for making antibody-coated Luminex beads and biotinylated antibodies. LITERATURE CITED (1) Martinez, A., Pittaluga, S., Villamor, N., Colomer, D., Rozman, M., Raffeld, M., Montserrat, E., Campo, E., and Jaffe, E. S. (2004) Clonal T-cell Populations and Increased Risk for Cytotoxic T-cell Lymphomas in B-CLL Patients: Clinicopathologic Observations and Molecular Analysis. Am. J. Surg. Pathol. 28, 849-858. (2) Xu, H., Sha, M. Y., Wong, E. Y., Uphoff, J., Xu, Y., Treadway, J. A., Truong, A., O′Brien, E., Asquith, S., Stubbins, M., Spurr, N. K., Lai, E. H., and Mahoney, W. (2003) Multiplexed SNP Genotyping Using the Qbeads System: a Quantum Dotencoded Microsphere-based Assay. Nucleic Acids Res. 31, e43. (3) Kellar, K. L.; Iannone M. A. (2002) Multiplexed Microspherebased Flow Cytometric Assays. Exp. Hematol. 30, 1227-1237. (4) Pickering, J. W., Martins, T. B., Carl Schroder, M., and Hill, H. R. (2002) Comparison of a Multiplex Flow Cytometric Assay with Enzyme-Linked Immunosorbent Assay for Quantitation of Antibodies to Tetanus, Diphtheria, and Haemophilus influenzae Type b. Clin. Diagn. Lab. Immunol. 9, 872876. (5) McBride, M. T., Gammon, S., Pitesky, M., O′Brien, T. W., Smith, T., Aldrich, J., Langlois, R. G., Colston, B., and Venkateswaran, K. S. (2003) Multiplexed Liquid Arrays for Simultaneous Detection of Simulants of Biological Warfare Agents. Anal. Chem. 75, 1924-1930. (6) McBride, M. T., Masquelier, D., Hindson, B. J., Makarewicz, A. J., Brown, S., Burris, K., Metz, T., Langlois, R. G., Tsang, K. W., Bryan, R., Anderson, D. A., Venkateswaran, K. S., Milanovich, F. P., and Colston, B. W., Jr. (2003) Autonomous Detection of Aerosolized Bacillus anthracis and Yersinia pestis. Anal. Chem. 75, 5293-5299. (7) Peruski, A. H., Johnson, L. H., III., and Peruski, L. F., Jr. (2002) Rapid and Sensitive detection of Biological Warfare Agents Using Time-Resolved Fluorescence Assays. J. Immunol. Methods 263, 35-41. (8) Robertus, J. D. (1991) The Structure and Action of Ricin, a Cytotoxic N-glycosidase. Semin. Cell Biol. 2, 23-30.

(9) Vitetta, E. S. and Thorpe, P. E. (1991) Immunotoxins Containing Ricin or Its A Chain. Semin. Cell Biol. 2, 47-58. (10) Moir, A., Corfe, B. M., and Behravan, J. (2002) Spore Germination. Cell. Mol. Life Sci. 59, 403-409. (11) Vogt, R. F., Cross, G. D., Phillips, D. L., Henderson, L. O., and Hannon, W. H. (1991) Interlaboratory Study of Cellular fluorescence Intensity Measurements with FluoresceinLabeled Microbead Standards. Cytometry 12, 525-536. (12) Bhalgat, M. K., Haugland, R. P., Pollack, J. S., Swan, S., and Haugland, R. P. (1998) Green- and Red-fluorescent Nanospheres for the Detection of Cell Surface Receptors by Flow Cytometry. J. Immunol. Methods 219, 57-68. (13) Zhao, X., Hilliard, L. R., Mechery, S. J., Wang, Y., Bagwe, R. P., Jin, S., and Tan, W. (2004) A Rapid Bioassay for Single Bacterial Cell Quantitation Using Bioconjugated Nanoparticles. Proc. Natl. Acad. Sci. U.S.A. 101, 15027-15032. (14) Va¨isa¨nen, V., Ha¨rma¨, H., Lilja, H., and Bjartell, A. (2000) Time-resolved Fluorescence Imaging for Quantitative Histochemistry Using Lanthanide Chelates in Nanoparticles and Conjugated to Monoclonal Antibodies. Luminescence 15, 389397. (15) Schaertl, S., Meyer-Almes, F., Lopez-Calle, E., Siemers, A., and Kra¨mer, J. (2000) A Novel and Robust Homogeneous Fluorescence-based Assay Using Nanoparticles for Pharmaceutical Screening and Diagnostics. J. Biomol. Screen. 5, 227237. (16) Hermanson, G. T. (1996) Bioconjugate Techniques, Chapter 7, Academic Press, San Diego, CA. (17) Wittmershaus, B. P., Skibicki, J. J., McLafferty, J. B., Zhang, Yu-Zhong, and Swan, S. (2001) Spectral Properties of Single BODIPY Dyes in Polystyrene Microspheres and in Solutions. J. Fluoresc. 11, 119-127. (18) Schwartz, A., Wang, L., Early, E., Gaigalas, A., Zhang, Yuzhong, Marti, G. E., and Vogt, R. F. (2002) Quantitating Fluorescence Intensity from Fluorophore: The Definition of MESF Assignment. J. Res. Natl. Inst. Stand. Technol. 107, 83-91. (19) Roberts, D. V., Wittmershaus, B. P., Zhang, Yu-Zhong, Swan, S., and Klinosky, M. P. (1998) Efficient Excitation Energy Transfer among Multiple Dyes in Polystyrene Microspheres. J. Luminesc. 79, 225-231. (20) White, J. C. and Stryer, L. (1987) Photostability Studies of Phycobiliprotein Fluorescent Labels. Anal. Biochem. 161, 442-452. (21) Chada, V. G. R., Sanstad, E. A., Wang, R., and Driks, A. (2003) Morphogenesis of Bacillus Spore Surfaces. J. Bacteriol. 185, 6255-6261.

BC0498020