Electrochemical Microbead-Based Immunoassay Using an (η5

Electrochemical Microbead-Based Immunoassay Using an (η5-Cyclopentadienyl)tricarbonylmanganese Redox Marker Bound to Bovine Serum Albumin...
0 downloads 0 Views 200KB Size
506

Langmuir 2006, 22, 506-511

Electrochemical Microbead-Based Immunoassay Using an (η -Cyclopentadienyl)tricarbonylmanganese Redox Marker Bound to Bovine Serum Albumin 5

Magdale´na Hromadova´,*,† Miche`le Salmain,‡ Nathalie Fischer-Durand,‡ Lubomı´r Pospı´sˇil,† and Ge´rard Jaouen‡ J. HeyroVsky´ Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, DolejsˇkoVa 3, 182 23 Prague, Czech Republic, and Laboratoire de Chimie Organome´ tallique, UMR CNRS 7576, Ecole Nationale Supe´ rieure de Chimie de Paris, 11 rue Pierre et Marie Curie, 75231 Paris Cedex 05, France ReceiVed August 11, 2005. In Final Form: October 17, 2005 A first example of the solid-phase immunoassay of a high-weight antigen bovine serum albumin (BSA) using an (η5-cyclopentadienyl)tricarbonylmanganese (cymantrene) redox probe is presented. The electrochemical detection is based on the impedance measurements of a one-electron reversible reduction of the organometallic probe. The microbeadbased immunoassay is discussed for two types of microbeads with different diameters (2.5 and 90 µm) and capabilities to bind the immunoglobulins (2.4 and 10 µg/mg of beads). The use of larger agarose microbeads allows the formation of an antigen-antibody complex at the surface of microbeads directly dispersed in the analyzed solution. No additional separation step is necessary for the electrochemical competitive immunoassay analysis of BSA. The presence of agarose beads in the analyzed solution has no effect on the electrochemical signal from labeled BSA released from the antigen-antibody complex.

1. Introduction Immunoassays represent a group of very useful analytical methods for quantification of the extremely low concentrations of analytes contained in body fluids (serum and urine). Most of the immunological methods utilize biomolecules labeled by special markers detectable by suitable physical methods. Current trends in the immunoassay analysis strive for the replacement of the radioisotopic labels. The electroactive species represent an attractive group of such alternative markers.1 The development of electrochemical immunoassay methods is currently based on three main approaches. The first one may be called collectively the electroenzymatic approach, in which the antigen or the antibody conjugated to a suitable enzyme generates the redox species detected by a variety of the electrochemical methods.2-9 The second approach makes use of the electroactive species directly attached to the antigen or the antibody. Examples of such an approach include the use of colloidal metal particles,10 metal ions attached to the protein through a chelating agent,11-14 or the covalently attached organometallic compounds. Ferrocene15 and the organic mol† ‡

Academy of Sciences of the Czech Republic. Ecole Nationale Supe´rieure de Chimie de Paris.

(1) Warsinke, A.; Benkert, A.; Scheller, F. W. Fresenius’ J. Anal. Chem. 2000, 366, 622. (2) Bourdillon, Ch.; Demaille, Ch.; Moiroux, J.; Save´ant, J. M. J. Am. Chem. Soc. 1999, 121, 2401. (3) Ferna´ndez-Sa´nchez, C.; Costa-Garcı´a, A. Anal. Chim. Acta 1999, 402, 119. (4) Medyantseva, E. P.; Khaldeeva, E. V.; Glushko, N. I.; Budnikov, H. C. Anal. Chim. Acta 2000, 411, 13. (5) Abad-Villar, E. M.; Ferna´ndez-Abedul, M. T.; Costa-Garcı´a, A. Anal. Chim. Acta 2000, 409, 149. (6) Wijayawardhana, C. A.; Wittstock, G.; Halsall, H. B.; Heineman, W. R. Electroanalysis 2000, 12, 640. (7) Toda, K.; Tsuboi, M.; Sekiya, N.; Ikeda, M.; Yoshioka, K. I. Anal. Chim. Acta 2002, 463, 219. (8) Dong, Y.; Shannon, C. Anal. Chem. 2000, 72, 2371. (9) Aguilar, Z. P.; Vandaveer, W. R.; Fritsch, I. Anal. Chem. 2002, 74, 3321. (10) Dequaire, M.; Degrand, C.; Limoges, B. Anal. Chem. 2000, 72, 5521. (11) Hayes, F. J.; Halsall, H. B.; Heineman, W. R. Anal. Chem. 1994, 66, 1860. (12) Wang, J.; Tian, B.; Rogers, K. R. Anal. Chem. 1998, 70, 1682.

ecules possessing one or more electroactive groups, for example, 2,4-dinitrophenol,16-19 are frequently used. The third approach provides detection of binding between the antigen and antibody by the electrochemical methods using biomolecules without any marker.20-27 This approach explores changes in the interfacial capacitance upon adsorption of large biomolecules at the electrode surface. We describe a new electrochemical immunoassay method based on the use of a novel organometallic (η5-cyclopentadienyl)tricarbonylmanganese redox label (cymantrene) bound to bovine serum albumin (BSA). Recently, we introduced transition metal carbonyl complexes as organometallic labels. We have demonstrated their suitability for the immunoanalysis of drugs28-30 and pesticides.31 The method was introduced as the carbonyl (13) Guo, W.; Song, J. F.; Zhao, M. R.; Wang, J. X. Anal. Biochem. 1998, 259, 74. (14) Doyle, M. J.; Halsall, H. B.; Heineman, W. R. Anal. Chem. 1982, 54, 2318. (15) Suzawa, T.; Ikariyama, Y.; Aizawa, M. Anal. Chem. 1994, 66, 3889. (16) Sa´nchez-Sua´rez, M. D.; Costa-Garcı´a, A. Talanta 1997, 44, 909. (17) Ferna´ndez-Bobes, C.; Ferna´ndez-Abedul, M. T.; Costa-Garcı´a, A. Instrum. Sci. Technol. 2001, 29, 65. (18) Costa-Garcı´a, A.; Ferna´ndez-Abedul, M. T.; Tun˜o´n-Blanco, P. Talanta 1994, 41, 1191. (19) Zayats, M.; Raitman, O. A.; Chegel, V. I.; Kharitonov, A. B.; Willner, I. Anal. Chem. 2002, 74, 4763. (20) Flores, J. R.; Smyth, M. R. J. Electroanal. Chem. 1987, 235, 317. (21) Smyth, M. R.; Buckley, E. Analyst 1988, 113, 31. (22) Kuramitz, H.; Matsuda, M.; Thomas, J. H.; Sugawara, K.; Tanaka, S. Analyst 2003, 128, 182. (23) Maupas, H.; Soldatkin, A. P.; Martelet, C.; Jaffrezic-Renault, N.; Mandrand, B. J. Electroanal. Chem. 1997, 421, 165. (24) Wang, M.; Wang, L.; Yuan, H.; Li, X.; Sun, Ch.; Ma, L.; Bai, Y.; Li, T.; Li, J. Electroanalysis 2004, 16, 757. (25) Berggren, C.; Johansson, G. Anal. Chem. 1997, 69, 3651. (26) Sˇ nejda´rkova´, M.; Csaderova´, L.; Reha´k, M.; Hianik, T. Electroanalysis 2000, 12, 940. (27) Song, J. F.; Liu, Y. Q.; Guo, W. Anal. Biochem. 2003, 314, 212. (28) Salmain, M.; Vessie`res, A.; Brossier, P.; Butler, I. S.; Jaouen, G. J. Immunol. Methods 1992, 148, 65. (29) Varenne, A.; Vessie`res, A.; Salmain, M.; Brossier, P.; Jaouen, G. J. Immunol. Methods 1995, 186, 195. (30) Salmain, M.; Vessie`res, A.; Varenne, A.; Brossier, P.; Jaouen, G. J. Organomet. Chem. 1999, 589, 92.

10.1021/la052188b CCC: $33.50 © 2006 American Chemical Society Published on Web 11/17/2005

Electrochemical Microbead-Based Immunoassay Scheme 1. Reaction of BSA with Organometallic Label

metalloimmunoassay (CMIA), and FTIR spectroscopy was used as the detection method. Immunoanalytical methods using electrochemical detection become increasingly popular due to their flexibility in terms of sensitivity and miniaturization of the fabricated sensors. We focused our attention on the electrochemical detection of a cymantrene redox marker in the form of cymantrenyl methyl imidate conjugated to the protein bovine serum albumin (BSA) via the -amino group of the protein lysine residues (see Scheme 1). The applicability of such an organometallic label for the electrochemical detection of BSA has been demonstrated in our previous work.32 The aim of this paper is to demonstrate the suitability of cymantrene label for the electrochemical immunoanalysis of BSA. To achieve the analysis without any additional manipulation or separation steps, microbeads coated with anti-BSA antibody were used in conjunction with BSA labeled with several cymantrene markers. Such a method suggested in this paper does not require any complicated procedure to separate the antibody-antigen complex from the electrochemical cell, which should allow further miniaturization of the sensor design or use in the flow-through systems. Beads with immobilized enzymes or proteins have already been used in the flow-through arrangement33-35 and as renewable electrode materials.36,37 Magnetic beads proved to be especially useful for preconcentration of analytes.38-41 2. Experimental Section 2.1. Reagents and Materials. Rabbit IgG (MW ) 150 000 g/mol), rabbit anti-bovine serum albumin (IgG fraction), BSA (grade VI, immunoglobulin-free, MW ) 67 000 g/mol), goat anti-rabbit IgG conjugated to horseradish peroxidase (HRP), and CNBr-activated Sepharose 4B were purchased from Sigma. The immunoglobulins were reconstituted by addition of Nanopure water (Millipore Q system) or 0.02% NaN3 in water. The concentrations of stock BSA and IgG solutions were determined spectrophotometrically42 at 280 nm using (BSA) ) 44 000 M-1‚cm-1 and (IgG) ) 210 000 M-1‚cm-1. N,N,N′N′-Tetramethyl-O-(N-succinimidyl)uronium tetrafluoroborate (TSTU) was purchased from Fluka. Estapor carboxylated latex microbeads (2.5 µm diameter, functionalization degree 4 (31) Vessie`res, A.; Fischer-Durand, N.; Le Bideau, F.; Janvier, P.; Heldt, J. M.; Ben Rejeb, S.; Jaouen, G. Appl. Organomet. Chem. 2002, 16, 669. (32) Hromadova´, M.; Salmain, M.; Sokolova´, R.; Pospı´sˇil, L.; Jaouen, G. J. Organomet. Chem. 2003, 668, 17. (33) Farrell, S.; Ronkainen-Matsuno, N. J.; Halsall, H. B.; Heineman, W. R. Anal. Bioanal. Chem. 2004, 379, 358. (34) Hayes, M. A.; Polson, N. A.; Phayre, A. N.; Garcia, A. A. Anal. Chem. 2001, 73, 5896. (35) Mayer, M.; Ruzicka, J. Anal. Chem. 1996, 68, 3808. (36) Sole´, S.; Merkoc¸ i, A.; Alegret, S. Trends Anal. Chem. 2001, 20, 102. (37) Besselink, G. A. J.; Schasfoort, R. B. M.; Bergveld, P. Biosens. Bioelectron. 2003, 18, 1109. (38) Purushothama, S.; Kradtap, S.; Wijayawardhana, C. A.; Halsall, H. B.; Heineman, W. R. Analyst 2001, 126, 337. (39) Dequaire, M.; Degrand, Ch.; Limoges, B. Anal. Chem. 1999, 71, 2571. (40) Thomas, J. H.; Kim, S. K.; Hesketh, P. J.; Halsall, H. B.; Heineman, W. R. Anal. Biochem. 2004, 328, 113. (41) Thomas, J. H.; Ronkainen-Matsuno, N. J.; Farrell, S.; Halsall, H. B.; Heineman, W. R. Microchem. J. 2003, 74, 267. (42) Stoschek, C. M. Methods Enzymol. 1990, 182, 50.

Langmuir, Vol. 22, No. 1, 2006 507 µequiv/g of beads) were purchased as a 10% suspension (w/v) from Merck Eurolab. Phosphate-buffered saline (PBS) was prepared by dissolving 0.34 g of KH2PO4, 1.42 g of NaH2PO4‚2H2O, 0.2 g of KCl, and 8 g of NaCl in 1 L of water. MES buffer was prepared from 2.13 g of 2-(N-morpholino)ethanesulfonic acid in 100 mL of water adjusted to pH 7.5 with 1 M NaOH. HEPES buffer was prepared from 1.19 g of N-(2-hydroxyethyl)piperazine-N′-(2-ethyl)sulfonic acid in 100 mL of water and was adjusted to pH 7.5 with 1 M NaOH. 2.2. Preparation of Labeled BSA. BSA was labeled by cymantrene entities (redox marker) according to a previously reported procedure.32 The concentration of cymantrenyl amidinium groups bound to BSA was assayed by UV-visible spectroscopy at 337 nm ( ) 1500 M-1‚cm-1). UV-visible spectra were recorded on a UV/ mc2 spectrometer (Safas). The coupling ratio CR of the conjugates was calculated as a concentration ratio of the cymantrenyl amidinium groups to the protein concentration. Labeled BSA samples with redox marker coupling ratios CR ) 7, 11, and 15 were prepared in this manner. They will be further called BSA7, BSA11, and BSA15, respectively. The protein solutions were freeze-dried on a Speedvac concentrator (Savant) and kept at -20 °C until use. On the basis of our previous findings,32 the electrochemical detection of labeled BSA is not sensitive to the coupling ratio of the redox marker; however, we find it appropriate to report the coupling ratio as part of the experimental conditions. 2.3. Coupling of Anti-BSA to Latex Microbeads. Microbeads (30 mg) were washed twice with 100 mM MES buffer at pH 5.5 and suspended in 1.4 mL of a 31 mM solution of TSTU in MES buffer. The suspension was shaken for 10 min at room temperature; then a 1.8 mg/mL solution of anti-BSA (0.1 mL; 180 µg) was added and the mixture was shaken for a further 2 h. The suspension was centrifuged, the supernatant was taken, the beads were suspended in 1 M ethanolamine at pH 8.5, and the suspension was shaken for 1 h. After centrifugation, the beads were washed with PBS at pH 7.5 containing 10% ethylene glycol (v/v) and then three times with PBS. The amount of immunoglobulin in the supernatants was assayed spectrophotometrically at 280 nm taking A280nm ) 1.4 for a 1 mg/mL solution of IgG. The beads were kept as a 2% suspension in PBS + 0.02% NaN3 at 4 °C until use.43 2.4. Coupling of Anti-BSA to Agarose Beads. Reconstituted anti-BSA was diluted in water to a final concentration of 5 mg/mL and dialyzed in 1 L of 0.2 M carbonate buffer at pH 8.3 + 0.5 M NaCl overnight at 4 °C. CNBr-activated agarose beads (300 mg) were suspended in cold 1 mM HCl and washed 3 times with this solution. The supernatants were discarded, anti-BSA (5 mg/mL, 10 mg) was added to the beads, and the suspension was shaken for 2 h at room temperature. The beads were filtered on a fritted glass (Schott no. 4) and further treated with 1 M ethanolamine at pH 8.5 for 2 h. The suspension was filtered and washed first with 100 mM acetate buffer at pH 6.0 + 0.5 M NaCl and then with 50 mM HEPES buffer at pH 7.5. The amount of immunoglobulin in the supernatants was assayed spectrophotometrically at 280 nm taking A280nm ) 1.4 for a 1 mg/mL solution of IgG. The beads were kept as a 10% suspension (w/v) in HEPES buffer pH 7.5 at 4 °C until use.44 2.5. Competitive Enzyme-Linked Immunosorbent Assay (ELISA) Experiment. Microtiter plates (Dynatech Immulon IV) were coated with BSA by treatment with 1 µg/mL solution in carbonate buffer pH 9.5 overnight at 4 °C. Nonspecific binding was further blocked by PBS + 4% skimmed milk (w/v). A variable amount of BSA or BSA labeled with 11 cymantrene units (BSA11) in the range 0-500 pmol and a fixed amount of anti-BSA (0.1 µg) in solution in PBS + 2% skimmed milk (w/v) was applied to the wells in duplicate and incubated in the dark for 3 h at room temperature. After the wells were washed 4 times with PBS + 0.05% Tween 20 (PBS-T), goat anti-rabbit IgG-HRP conjugate in PBS + (43) Danielson, S. J.; Specht, D. P. Method and kit for attaching compounds to carboxylated substrates using uronium salt. European patent application No. 932012488, 1993. (44) Hill, C. R.; Thompson, L. G.; Keeney, A. C. Immunopurification. In Protein purification methods; Harris, E. L. V., Angel, S., Eds.; IRL Press: Oxford, U.K., 1989; p 282.

508 Langmuir, Vol. 22, No. 1, 2006 2% skimmed milk was applied to the wells (1/8000) and was incubated 2 h at room temperature. Wells were washed 4 times with PBS-T, and a 0.7 mg/mL solution of o-phenylenediamine in citratephosphate buffer at pH 5.0 + 0.012% H2O2 (v/v) was applied to each well. The color was developed for 10 min, and the enzymatic reaction was stopped by addition of 2.5 M H2SO4. The absorbance was read at 490 nm with a microtiter plate reader (Biorad). 2.6. Electrochemical Detection Method. Electrochemical measurements were done using a system for cyclic voltammetry, phasesensitive ac voltammetry, and ac and dc polarography. It consisted of a fast rise-time potentiostat45 and a lock-in amplifier (Stanford Research, model SR830). The instruments were interfaced to a personal computer via an IEEE-interface card (PC-Lab, AdvanTech model PCL-848) and a data acquisition card (PCL-818) using 12-bit precision. The AC signal for the cell was derived from the internal oscillator of the lock-in amplifier. The amplitude was 10 mV (p-p). A three-electrode electrochemical cell was used. The working electrode was a valve-operated static mercury electrode (SMDE2, Laboratornı´ Pøı´stroje, Prague) with an area 1.13 × 10-2 cm2. The reference electrode was an Ag wire covered with electrochemically deposited AgCl salt. The reference electrode was immersed directly into the aqueous KCl solution. The auxiliary electrode was platinum wire. Oxygen was removed from the solution with a stream of argon. Potassium chloride for the supporting electrolyte preparation was of the best quality available from Aldrich Chemicals. All aqueous solutions were prepared using Nanopure water (Millipore Q system). The electrochemical cell and solutions were kept at constant temperature of 20 °C during all measurements. We have shown recently that the redox cymantrene marker is reduced reversibly at the potential of -1.84 V against the Ag/AgCl reference electrode on the mercury electrode. The redox marker signal was taken as the faradaic maximum of the real component (baseline-subtracted) of the admittance of the electrochemical system. The same procedure for determination of the analytically useful signal is employed in this work.32

HromadoVa´ et al.

Figure 1. Competitive ELISA experiment. Semilogarithmic plot of the relative absorbance at λ ) 490 nm as a function of the different concentration of the nonlabeled BSA (b) and labeled BSA11 (O) competitor. For experimental details, see section 2.5.

3. Results and Discussion The choice of cymantrene as a redox label was done with respect to the requirement to couple its methyl imidate derivative to the protein and use such a conjugate in the aqueous medium. Only few organometallic compounds can fulfill this requirement. BSA was labeled with cymantrene units according to a previously published procedure.32 Cymantrene bound to BSA yields the reversible redox exchange at -1.84 V against the Ag/AgCl reference electrode, which overlaps with the hydrogen evolution catalyzed by the presence of a protein. The discrimination of both electron transfers is most conveniently achieved on the basis of very different electron transfer rates for cymantrene and the proton reduction. Ac voltammetry at a suitably chosen frequency of 16 Hz was used for this purpose. The choice of frequency was made on the basis of the impedance spectra measured in the frequency range 1 Hz to 100 kHz. The electroanalytical signal used for the construction of the calibration curves was obtained from the in-phase admittance after subtraction of the baseline corresponding to the protein-catalyzed evolution of hydrogen. 3.1. Competitive ELISA Experiment. To ascertain that the labeling of BSA molecule with cymantrene marker had no effect on its recognition by the antibody, a competitive ELISA experiment was performed involving either nonlabeled BSA or BSA11 as competitors. In the experiment, the antibody binds either to BSA (or BSA11) in solution or to BSA adsorbed on the microtiter plates. The amount of bound antibody was further quantified spectrophotometrically by the addition of anti-rabbit IgG-HRP conjugate. The more BSA or BSA11 was in the solution, the smaller amount of anti-rabbit IgG-HRP conjugate was detected (45) Pospı´sˇil, L.; Fiedler, J.; Fanelli, N. ReV. Sci. Instruments 2000, 71, 1805.

Figure 2. Real component of the cell admittance for 7.3 × 10-6 M IgG in 0.1 M KCl aqueous solution (b), 7.3 × 10-6 M IgG in 0.1 M aqueous KCl solution containing 0.018% of NaN3 (O), 0.7 × 10-6 M BSA11 in 0.1 M KCl solution (4), and 0.3 × 10-6 M BSA11 and 6.0 × 10-6 M IgG in 0.1 M KCl solution containing 0.016% NaN3 (2). Data acquisition of ac voltammetric measurements started at -1.0 V, and potential was scanned negatively at the scan rate 2 mV/s using a peak-to-peak amplitude of 10 mV and frequency 16 Hz. The cell temperature was kept at 20 °C.

on the microtiter plate. The competition curves for both nonlabeled and labeled BSA are shown in Figure 1. Linearization of the curves on the basis of the logit B/B0 versus ln [BSA] plot46 allows the evaluation of IC50 of both competitors. Values of (3.5 ( 1.1) × 10-8 M for BSA and (2.3 ( 1.6) × 10-8 M for BSA11 clearly indicate that the covalent binding of a large number of organometallic probes to the BSA antigen does not affect its recognition by the specific antibody. 3.2. Detection of the Antigen-Antibody Interaction by Phase-Sensitive Ac Voltammetry. Figure 2 shows the real component of the admittance Y′ in the potential range of the cymantrene label reduction. Each measurement was obtained on a fresh mercury drop electrode created at the potential of -1.0 V. The potential was scanned immediately toward the more negative values at the scan rate 2 mV/s. The height of the peak (46) Chard, T. An introduction to radioimmunoassay and related techniques, 2nd ed.; Elsevier Biomedical Press: Amsterdam, 1982.

Electrochemical Microbead-Based Immunoassay

Figure 3. Calibration curve for labeled BSA15 in 0.1 M KCl aqueous solution. The faradaic part of the real component of the cell admittance was obtained at the ac frequency 16 Hz and at the scan rate 2 mV/s. The cell temperature was kept at 20 °C.

after baseline subtraction was used as an analytical signal. There are four real admittance curves in Figure 2 showing response of BSA11 in 0.1 M KCl aqueous solution (4), anti-BSA IgG in 0.1 M KCl solution (b), anti-BSA IgG in 0.1 M KCl solution containing 0.018% of sodium azide (O), and the mixture of BSA11 and IgG in 0.1 M KCl containing 0.016% of sodium azide (2). As expected, the Y′ admittance component has a peak around -1.84 V corresponding to the reversible redox marker reduction, when labeled BSA is used. The curve does not show any peak in the case of the IgG solution. However, when the IgG solution contained sodium azide as a preservative, a practice very common in biochemical research, a new peak develops corresponding to the reduction of azide. Even though it may be far enough from the peak of interest, we opted for the use of IgG solutions free of sodium azide. The calibration curves for determination of labeled BSA concentration were obtained and the corresponding data are shown in Figure 3 for BSA15. The relationship between the electrochemical signal and the concentration is linear in a reasonably wide concentration range (0.1-1.0 µM), and this range was used in the immunoassay experiment. As was shown in our previous work, the actual signal is not linear within the entire range measurable by the electrochemical method,32 but it can be linearized through the double-logarithmic plot. Linearization follows the Freundlich isotherm expression and is in line with our previous finding that the signal is observed only when the protein adsorbs at the electrode surface bringing the electroactive label to a distance appropriate for a fast electron transfer. We did not find it necessary to use such a double-logarithmic plot here, since the measured electrochemical signal was always within the range shown in Figure 3. 3.3. Homogeneous Immunoassay Experiment. On the basis of previously reported works47-49 we thought it possible that the formation of the immuno complexes between labeled BSAx and anti-BSA immunoglobulin would decrease the admittance signal because of the hindered accessibility of the redox probe to the electrode. Hence a homogeneous-type of immunoassay analysis was tried. (47) Weber, S.; Purdy, W. Anal. Lett. 1979, 12, 1. (48) Di Gleria, K.; Hill, H. O. A.; McNeil, C. J. Anal. Chem. 1986, 58, 1203. (49) Forrow, N. J.; Foulds, N. C.; Frew, J. E.; Law, J. T. Bioconjugate Chem. 2004, 15, 137.

Langmuir, Vol. 22, No. 1, 2006 509

Figure 4. Change of the BSA11 concentration after addition of nonspecific rabbit IgG (O) and specific anti-BSA IgG (b) in 0.1 M KCl. The initial concentration of BSA11 was 0.7 × 10-6 M in the 0.1 M aqueous KCl solution. A redox marker signal was obtained as the faradaic part of the real component of the cell admittance at frequency 16 Hz, and the concentration of BSA15 was calculated from the calibration curve. Concentration is reported as the percentage of BSA11 in solution.

The initial BSAx concentration was chosen such as to provide a relatively high initial redox label signal. Afterward the solutions containing different concentrations of IgG were prepared and time for the completion of the immunochemical reaction was set at 15 min. Figure 4 shows the percentual decrease of the BSA11 concentration in solution as a function of the IgG concentration. In one case the rabbit IgG, i.e., nonspecific IgG, was used to see the response of the system in the case where no immunological reaction is expected (O). In the other case the specific IgG, rabbit anti-BSA was used (b). The electrochemical signal due to the labeled BSA decreases in the presence of specific IgG as would be expected. Unfortunately, it does decrease to a certain degree (see Figure 4) also in the presence of nonspecific IgG. Therefore, our initial assumption that the signal decrease would be due to the inaccessibility of the redox label toward the electron transfer in the BSA anti-BSA complex is faulty. A possible explanation of such a behavior is the competition of the large IgG molecules for the electrode surface that is occupied by the BSA molecules and subsequent decrease of the electrochemical signal. Supporting evidence for this hypothesis comes from the shape of the signal decrease upon addition of IgG and also from the necessity to use large amounts of IgG to achieve a significant decrease in the admittance signal. Therefore, this procedure was not further used and we worked up sample manipulation described below. 3.4. Heterogeneous Immunoassay Experiment. Previous experimental data pointed to the strong competitive adsorption of IgG and BSA for the electrode surface. To avoid the problems with adsorption of the IgG molecules at the electrode surface we have opted for the immobilization of IgG molecules onto microbeads. This way the interaction of the antigen and antibody is performed not on the electrode surface but on the bead interface. Resulting decrease of the concentration of labeled BSA in the bulk is monitored by the decrease in the electrochemical admittance signal at the mercury electrode. Immobilization of the anti-BSA and rabbit IgG to two kinds of chemically derivatized microbeads, i.e., carboxylate-modified latex microparticles (diameter 2.5 µm) and CNBr-activated agarose beads (diameter 90 µm), was performed using known procedures.43,44 The amount of protein immobilized onto the beads was calculated from the

510 Langmuir, Vol. 22, No. 1, 2006

HromadoVa´ et al.

Figure 5. Competitive immunoassay experiment employing antiBSA IgG attached to the latex microbeads (2.5 µm diameter). A redox marker signal was obtained as the faradaic part of the real component of the cell admittance at frequency 16 Hz normalized for the response of BSA11 before the anti-BSA addition. The concentration of the protein was 0.3 × 10-6 M BSA11 and 0.9 × 10-6 M of unlabeled BSA and 0.2 × 10-6 M anti-BSA IgG in 1 mL of 0.05 M KCl.

Figure 6. Change of the BSA15 concentration after addition of nonspecific IgG (O) and specific anti-BSA IgG (b) attached to the agarose beads (90 µm diameter). The initial concentration of BSA15 was 0.845 × 10-6 M in the 0.1 M aqueous KCl solution. A redox marker signal was obtained as the faradaic part of the real component of the cell admittance at frequency 16 Hz. BSA15 concentration was calculated from the calibration curve in Figure 3 and is reported as the percentage of BSA15 in solution.

Table 1. Covalent Binding of Anti-BSA IgG and Rabbit IgG to Microbeads (µg/mg of Beads) anti-BSA IgG rabbit IgG

latex microbeads

agarose beads

2.4

10 10

spectrophotometric assay of protein in supernatants and washing solutions. Characteristics of the modified beads are indicated in Table 1. The latex beads were added into the solution of labeled BSA and were allowed to interact with BSA for 15 min. Due to their small size the beads had to be eventually centrifuged off, and the supernatant was then transferred to the electrochemical cell for the measurements. Figure 5 shows the decrease in the BSA11 concentration in solution after its treatment with the anti-BSA attached to the latex beads. When the experiment was performed with the mixture of nonlabeled BSA and BSA11, the signal decreased much less as expected, due to the competition between the labeled and nonlabeled BSA for binding to anti-BSA IgG. Figure 5 clearly shows that the decrease of the electrochemical signal is due to the formation of the BSA anti-BSA complex on the latex beads and further supports the feasibility of the microbead approach to the electrochemical immunoanalysis using an (η5cyclopentadienyl)tricarbonylmanganese redox marker. Our further effort was aimed at the elimination of the tedious procedure of the bead separation from the BSA-containing analyte. Therefore, we used IgG molecules attached to agarose beads. This solid support has a diameter of 90 µm, and therefore its sedimentation is achieved within a few minutes in the aqueous solution. The IgG-coated agarose beads were put directly into the electrochemical cell containing labeled BSA molecules. No separation step was needed. An experiment showing the effect of the anti-BSA concentration on the electrochemical response of the BSA15 is shown in Figure 6. The initial concentration of the BSA15 was 0.845 µM in 0.1 M KCl solution. Each solution was deaerated for 2 min with a 13 min waiting time for the completion of the immunological reaction and bead settling. The response to the specific-IgG addition is linear within the whole concentration range needed for a signal decrease to a level below 10%. When similar additions are made using beads containing

Figure 7. Competitive immunoassay experiment employing antiBSA IgG attached to the agarose beads (90 µm diameter). A redox marker signal was obtained as the faradaic part of the real component of the cell admittance at frequency 16 Hz normalized for the response of BSA15 before the anti-BSA addition. The concentration of labeled albumin was 0.8 × 10-6 M BSA15 in all five samples presented. The concentration of the anti-BSA IgG was 1.5 × 10-6 M in sample A and 1.9 × 10-6 M in samples B-D. Samples C and D contained 0.6 × 10-6 and 1.2 × 10-6 M of nonlabeled BSA in 0.1 M KCl, respectively.

the nonspecific IgG, the signal remains unchanged (see the empty points in Figure 6). This is the key proof that the decrease of the electrochemical response is indeed due to the immunological reaction between the antigen and the antibody. The antibody titer is 7.2 × 10-6 M and is defined as the concentration of IgG that binds 50% of BSA15. The procedure was tested for the development of the microbeadbased competitive immunoanalysis. The experimental data are shown in Figure 7. The BSA15 signal in the absence of the beads was taken as 100%. The electrochemical response of the BSA15 in the solution containing two different anti-BSA IgG concentrations (different amounts of IgG-coated agarose beads) is shown as bars A and B in Figure 7. The decrease of BSA15 concentration in the bulk of the solution is due to the formation of a BSA15

Electrochemical Microbead-Based Immunoassay

anti-BSA complex. An addition of 0.6 × 10-6 M nonlabeled BSA increases the electrochemical signal to almost the original value. This happens due to the displacement of the labeled BSA15 in the immune complex by nonlabeled BSA on the microbeads. Figure 7 shows the feasibility of the use of the cymantrene organometallic label for the competitive electrochemical microbead-based immunoassay analysis of BSA.

Langmuir, Vol. 22, No. 1, 2006 511

onto the surface of the screen-printed sensor. The use of covalently attached antibodies to the beads has the advantage of not only better specificity but also the utilization of the wider concentration range of antibodies that can be used in the process of the effective sensor design. Another important advantage stems from the fact that it is not necessary to release the electroactive label from the tracer (BSA) as is many times required.

4. Conclusions The possibility to use (η5-cyclopentadienyl)tricarbonylmanganese (cymantrene) as a suitable redox label of proteins for their immunoassay detection has been demonstrated. The heterogeneous competitive immunoassay procedure utilized the formation of immune complex on the sepharose beads directly dispersed in the electrochemically analyzed solution. No additional separation step was necessary. The advantage of presented method stems from its greater flexibility compared to some previously suggested electrochemical detection procedures. For example, Wang et al.12 adsorbed the antibody noncovalently

Acknowledgment. The CNRS (France) and the Academy of Sciences of the Czech Republic are gratefully acknowledged for supporting the project with exchange grants (Project No. 12148 and Project Barrande 2005-06-005-1). This work was additionally supported by the Grant Agency of the Czech Republic (Grant 203/03/0821), the Grant Agency of the Academy of Sciences of the Czech Republic (Grant 400400505), and the Ministry of Education, Youth, and Sports of the Czech Republic (Grant LC510). LA052188B