Anal. Chem. 1999, 71, 2571-2577
An Immunomagnetic Electrochemical Sensor Based on a Perfluorosulfonate-Coated Screen-Printed Electrode for the Determination of 2,4-Dichlorophenoxyacetic Acid Murielle Dequaire, Chantal Degrand, and Benoıˆt Limoges*
Electrosynthe` se et Electroanalyse Bioorganique, UMR CNRS 6504, Universite´ Blaise Pascal de Clermont-Ferrand, 24 Avenue des Landais, 63177 Aubie` re, France
A disposable immunomagnetic electrochemical sensor involving a magnetic particle-based solid phase and a Nafion film-coated screen-printed electrode (Nafion-SPE) stuck at the bottom of a polystyrene cylinder (microwell of 300 µL) was developed and evaluated in a competitive immunoassay of the widely used herbicide 2,4-dichlorophenoxyacetic acid (2,4-D). The competitive binding of 2,4-D and 2,4-D labeled with alkaline phosphatase (AP) for a limited amount of polyclonal anti-2,4-D antibodycoated magnetic beads was monitored electrochemically by measuring the AP label activity bound to the beads. The phosphoric acid ester of [[(4-hydroxyphenyl)amino]carbonyl]cobaltocenium hexafluorophosphate was used as the AP substrate. This anionic substrate (S-) is enzymatically transformed at pH 9.0 into a cationic phenol derivative (P+) which can be easily accumulated in the polyanionic Nafion coating and determined by cyclic voltammetry. During the enzyme reaction, the AP-associated beads were localized on the surface of the NafionSPE with the aid of a magnet, thus effectively increasing the concentration of P+ in the Nafion-modified electrode vicinity. The enzyme generation of P+ close to the electrode surface, and thereby to the Nafion film, resulted in a high amplification of the response. A detection limit of 0.01 µg L-1 2,4-D was thus achieved. The performance of the sensor was successfully evaluated on river water samples spiked with 2,4-D, indicating that this convenient and sensitive technique offers great promise for decentralized environmental applications. The wide use of pesticides in agriculture has led to a growing concern about the potential contamination of groundwater. The European Community regulations state that the maximum admissible concentration of a single pesticide in drinking water is 0.1 µg L-1 (0.1 ppb) and the total combined concentration of all of the pesticides is 0.5 µg L-1 (0.5 ppb).1 Conventional methods for the analysis of pesticides in water are generally based on high-performance liquid chromatography or gas chromatography. This requires the use of relatively expensive and specialized instrumentation and generally does not produce rapid results. Alternative methods based on immunoas(1) Wittmam C.; Hock B. J. Agric. Food. Chem. 1991, 39, 1194-200. 10.1021/ac990101j CCC: $18.00 Published on Web 05/29/1999
© 1999 American Chemical Society
says provide inexpensive and rapid screening techniques for water quality monitoring and support to conventional techniques for both laboratory and field analysis.2 Moreover, immunoassay methods are simple, sensitive, reliable, and relatively selective for pesticide testing. Among the high number of immunoassay techniques, the enzyme-linked immunosorbent assays (ELISA) combined with a colorimetric end-point measurement are the most widely used,3-12 and at least two pesticide ELISA kits are available on the market, i.e., Immunosystems (Millipore) and Ohmicron (Prolabo). The antibodies are adsorbed on the walls of tubes or microtiter plates in the Millipore kit, whereas they are covalently linked to magnetic particles in the Ohmicron kit.5-9 Immunosensors with either the antibody or antigen directly immobilized at the sensor surface (transducer) offer great promise,13 especially for on-site environmental analysis, but to our knowledge they have not yet been commercialized in this field. Many types of transducers have been used in immunosensor technology, which exploit changes in mass, heat, electrochemical, or optical properties.13 Among them, electrochemical immunosensors with an amperometric transducer have gained considerable attention in recent years.13,14 They usually rely on the use of an enzyme label that generates an electroactive product close to the electrode surface.14-19 The formation of a relatively high local concentration of the enzyme product15,16 and its eventual involvement in a bioelectrocatalytic reaction17-19 lead to a significant (2) Van Emon, J. M.; Lopez-Avila, V. Anal. Chem. 1992, 64, 79A-99A. (3) Hall, C J.; Deschamps, J. A. R.; Krieg, K. K. J. Agric. Food. Chem. 1989, 37, 981-4. (4) Franek, M.; Kolar, V.; Granatova, M.; Nevorankova, Z. J. Agric. Food. Chem. 1994, 42, 1369-74. (5) Hottenstein, C. S.; Rubio, F. M.; Herzog, D. P.; Fleeker, J. R.; Lawruk, T. S. J. Agric. Food. Chem. 1996, 44, 3576-81. (6) Lawruk, T. S.; Hottenstein, C. S.; Fleeker, J. R.; Hall, J. C.; Herzog, D. P.; Rubio, F. M. Bull. Environ. Contam. Toxicol. 1994, 52, 538-45. (7) Lawruk, T. S.; Lachman, C. E.; Jourdan, S. W.; Fleeker, J. R.; Herzog, D. P.; Rubio, F. M. J. Agric. Food. Chem. 1993, 41, 747-52. (8) Lawruk, T. S.; Lachman, C. E.; Jourdan, S. W.; Fleeker, J. R.; Herzog, D. P.; Rubio, F. M. J. Agric. Food. Chem. 1993, 41, 1426-31. (9) Lawruk, T. S.; Hottenstein, C. S.; Herzog, D. P.; Rubio, F. M. Bull. Environ. Contam. Toxicol. 1992, 48, 643-50. (10) Gascon, J.; Martinez, E.; Barcelo, D. Anal. Chim. Acta 1995, 311, 357-64. (11) Aga, D. S.; Thurman, E. M. Anal. Chem. 1993, 65, 2894-8. (12) Gruessner, B.; Shambaugh, N. C.; Watzin, C. M. Environ. Sci. Technol. 1995, 29, 251-4. (13) Morgan, C. L.; Newman, D. J.; Price, C. P. Clin. Chem. 1996, 42, 193209. (14) Skadal, P. Electroanalysis 1997, 9, 737-45.
Analytical Chemistry, Vol. 71, No. 13, July 1, 1999 2571
signal amplification. The main problem associated with immunosensors is their cost, and consequently, regeneration of the sensing surface by exploiting the reversibility of the antibody/ antigen reaction is desirable.20 However, this regeneration step is time-consuming and gives irreproducible results, particularly when the immunocomplex possess a high affinity constant. Moreover, the drastic regeneration conditions can damage and release the immunoreagent bound to the surface of the transducer.20,21 The regeneration step can be avoided with single-use screen-printed electrochemical immunosensors.22-26 This approach, which is promising for decentralized assays, was very recently applied to the assay of the 2,4-dichlorophenoxyacetic acid herbicide (2,4-D) by Kalab and Skadal.24 In their work, the authors investigated a competitive ELISA format with 2,4-D labeled by acetylcholinesterase (AChE) and with monoclonal antibodies entrapped within a thin perfluorosulfonated film deposited on the surface of a screen-printed electrode (SPE). The amount of the labeled 2,4-D bound to the antibody at the electrode surface was determined amperometrically by oxidation of thiocholine enzymatically generated from the acetylthiocholine substrate present in solution. The high turnover number of AChE allows less than 0.01 µg L-1 2,4-D to be detected, but an inhibition of AChE by other pesticides potentially present in the sample (e.g., organophosphate or carbamate) provides a false presence of 2,4-D.24 Another approach avoiding regeneration consists of using disposable antibody- or antigen-coated magnetic beads and building up in situ the immunosensing surface by localizing the immunomagnetic beads (IMBs) on the electrode area with the aid of a magnet.27-29 Robinson and co-workers27 were the first ones to explore this strategy in a sandwich electrochemical immunoassay of the human choriogonadotropin hormone. The authors used a reusable pyrolitic graphite electrode connected to a bar magnet, and they determined the amount of the bound enzyme localized at the electrode surface by bioelectrocatalysis. A similar system was described by Weetall and Hotaling,28 who rather than devise a reusable electrode chose to exploit a disposable SPE, and it was applied for the determination of immunoglobulin G. More recently, Gehring et al.29 developed an enzyme-linked immunomagnetic electrochemical detection of Salmonella typhinurium. In this (15) Rishpon, J.; Gezundhajt, Y.; Soussan, L.; Rosen-Margalit, I.; Hadas, E. In Biosensor Design and Application; Mathewson, P. R., Finley, J. W., Eds.; ACS Symposium Series 511; American Chemical Society: Washington, DC, 1992; pp 59-70. (16) Duan, C. M.; Meyerhoff, M. E. Anal. Chem. 1994, 66, 1369-77. (17) Ho, W. O.; Athey, D.; McNeil, C. J. Biosens. Bioelectron. 1995, 10, 68391. (18) Rishpon, J.; Ivnitski, D. Biosens. Bioelectron. 1997, 12, 195-204. (19) Gyss, C.; Bourdillon, C. Anal. Chem. 1987, 59, 2350-5. (20) Bright, F. V.; Litwiler, K. S.; Vargo, T. G.; Gardella, J. A. Anal. Chim. Acta 1992, 262, 323-30. (21) Blanchard, G. C.; Taylor, C. G.; Busey, B. R.; Williamson, M. L. J. Immunol. Methods 1990, 130, 263. (22) Wang, J.; Pamidi, P. V. A. Anal. Chem. 1998, 70, 1171-5. (23) Wang, J.; Tian, B.; Rogers, K. R. Anal. Chem. 1998, 70, 1682-5. (24) Kalab, T.; Skladal, P. Electroanalysis 1997, 9, 293-7. (25) Hart, J. P.; Pemberton, R. M.; Luxton, R.; Wedge, R. Biosens. Bioelectron. 1997, 12, 1113-21. (26) Schreiber, A.; Feldbru ¨ gge, R.; Key, G.; Glatz, J. F. C.; Spener, F. Biosens. Bioelectron. 1997, 12, 1131-7. (27) Robinson, G. A.; Hill, H. A. O.; Gear, J. M.; Rattle, S. J.; Forrest, G. C. Clin. Chem. 1985, 31, 1449-52. (28) Weetall, H. H.; Hotaling, T. Biosensors 1987, 3, 57-63. (29) Gehring, A. G.; Crawford, C. G.; Mazenko, R. S.; Van Houten, L. J.; Brewster J. D. J. Immunol. Methods 1996, 195, 15-25.
2572 Analytical Chemistry, Vol. 71, No. 13, July 1, 1999
system, the bacteria were sandwiched between antibody-coated magnetic beads and alkaline phosphatase (AP)-conjugated antibody, and the beads were magnetically localized onto the surface of a bare graphite SPE. The AP label localized at the bare electrode surface was able to hydrolyze p-aminophenyl phosphate to p-aminophenol, which was monitored by electrochemical oxidation. However, the immunoreaction was carried out in a separate tube in order to avoid the contamination of the bare SPE by the sample, the antiserum constituents, and/or the enzyme conjugate, and so it is only after magnetic separation, washing, and resuspension in buffer that the IMBs were transferred into a microwellshaped electrochemical cell for enzyme reaction and detection.29 Interferences at the naked electrode due to the adsorption of hydrophobic species are not new and cannot be ignored when biological or environmental samples are involved. It was established that some polyanionic films such as Nafion coatings are efficient antifouling membranes in biological fluids.30-33 Moreover, it was shown in our group that the cation-exchange properties of Nafion can be beneficially exploited to preconcentrate an enzyme product at the surface of the modified electrode, which thus offers an amplified voltammetric current response.34-38 More precisely, the widely used enzyme label AP was indirectly determined at femtomolar levels using a single-use Nafion-film coated SPE (Nafion-SPE) and the substrate S-, which is hydrolyzed to P+ by AP.38
The goal of the present work is to address various challenges associated with the adaptation of the electrochemical detection method previously developed for AP38 to a competitive immunomagnetic assay of 2,4-D in a multielectrochemical microwell plate format. This strategy combines the convenient use of immunomagnetic beads with the sensitive AP determination at a NafionSPE. The principle, whose key step is depicted in Figure 1, involves, successively, the competitive immunoreaction between the analyte and the AP-analyte conjugate for a limited amount of antibodies-coated magnetic beads, a rinsing step, and the addition of the anionic substrate S-, the AP-associated beads being magnetically localized on the Nafion-SPE (Figure 1). The anionic substrate S- is converted into the cationic electroactive product P+ close to the electrode surface, and so it is immediately (30) Hoyer, B.; Florence, T. M. Anal. Chem. 1987, 59, 2839-42. (31) Kristensen, E. W.; Kuhr, W. G.; Wightman, R. M. Anal. Chem. 1987, 59, 1752-7. (32) Le Gal La Salle, A.; Limoges, B.; Anizon, J. Y.; Degrand, C. J. Electroanal. Chem. 1993, 350, 329-35. (33) Le Gal La Salle, A.; Limoges, B.; Rapicault, S.; Degrand, C.; Brossier, P. Anal. Chim. Acta 1995, 311, 301-8. (34) Le Gal La Salle, A.; Limoges, B.; Degrand, C. J. Electroanal. Chem. 1994, 379, 281-91. (35) Le Gal La Salle, A.; Limoges, B.; Degrand, C. Anal. Chem. 1995, 67, 124553. (36) Rapicault, S.; Limoges, B.; Degrand, C. Electroanalysis 1996, 8, 880-4. (37) Limoges, B.; Degrand, C. Anal. Chem. 1996, 68, 4141-8. (38) Bagel, O.; Limoges, B.; Scho¨llhorn, B.; Degrand, C. Anal. Chem. 1997, 69, 4688-94.
Figure 2. Schematic design of the multielectrochemical microwells aligned over the magnet holding block. Figure 1. Schematic representation of the key step of the magnetic electrochemical immunoassay.
entrapped within the Nafion film, owing to the high affinity of P+ for the anionic polymer. To illustrate the principle, we have employed the anti-2,4-D antibody-coated magnetic beads available from the commercial 2,4-D kit provided by Ohmicron (colorimetric assay), and we have performed the entire assay protocol in the multielectrochemical microwells. The 2,4-D compound is the best representative of the chlorophenoxy herbicide class and it is widely used in agriculture for the control of broad-leaved weeds. This hydrosoluble salt is readily adsorbed in soil and can contaminate groundwater and drinking water. Since it is suspected as a cancer-causing agent, the monitoring of environmental samples for the presence of 2,4-D is desirable. EXPERIMENTAL SECTION Material and Reagents. A 5 wt % Nafion solution (EW 1100) was purchased from Aldrich. The carbon ink was obtained from Emerson & Cuming (Minico M3021-1RS). Flat-bottomed polystyrene microwells plates were provided from Polylabo (Nunc, ref 198-9902R876). The synthesis of [[(4-hydroxyphenyl)amino]carbonyl]cobaltocenium hexafluorophosphate (P+) and of the corresponding phosphoric acid ester (S-) was previously published.38 Lyophilized AP from bovine intestinal mucosa (1800 units mg-1, VII-NL, P8647), lyophilized phophatase alkaline-biotinamidocaproyl (B-AP, 3000 units mg-1, biotin content 3-6 mol/mol of protein, P1318), 2,4-dichlorophenoxyacetic acid, 2-(5-norbornene-2,3-dicarboxiimide)-1,1,3,3-tetramethyluronium (TNTU), N-methylmorpholine (NMM), bovine serum albumin (BSA, fraction V, A-3059), sodium azide, and tris(hydroxymethyl)aminomethane (TRIS) were obtained from Sigma. Streptavidin immobilized on magnetic beads (S-MBs) with a binding capacity of 4.4 nmol mL-1 of free biotin, was purchased from Dynal (M-280 Dynabeads, 6.7 × 108 units mL-1 in phosphate-buffered saline containing 0.1% BSA and 0.02% NaN3). Magnetic beads coated by polyclonal anti-2,4-D antibodies (Ab-MBs) suspended in buffered saline containing preservatives and stabilizers were available from a Ohmicron kit purchased from Prolabo.37 TRIS buffer (TB: 50 mM TRIS, 1 mM MgCl2‚6H2O, and 50 mM NaCl, pH 9.0) and phosphate-buffered saline (PBS: 4.3 mM NaH2PO4, 15.1 mM Na2HPO4, and 50 mM NaCl, pH 7.4) were
prepared using deionized and doubly distilled water. Stock solutions of B-AP (750 units mL-1) in PBS containing 1% BSA and 0.02% NaN3, and of 2,4-D (100 mg mL-1 in absolute ethanol), were stored at 4 °C. A fresh S- stock solution (10-2 M in TB) was daily prepared as previously described.38 Other chemicals used were of reagent grade. Apparatus, Electrodes, and Multielectrochemical Microwells. An Autolab potentiostat (EcoChemie) interfaced to a PC system with a GPSE version 4.4 software was used for cyclic voltammetry (CV). An array of four SPEs (1.5-mm diameters on 9-mm centers) was printed on a flexible polyester film (Staedler film for overhead projection) by forcing the commercial ink to penetrate through the mesh of a screen stencil (120 threads cm-1). Each SPE consisted of a disk (1.5-mm diameter), a conductive track (8 × 0.7 mm2), and a square extremity (12 mm2) for the electrical contact.39 After drying for 1 h at 100 °C, an insulator layer was spread manually over the conductive track, leaving the working disk area ready for Nafion modification. The Nafion coating was made by syringing 0.2 µL of a Nafion solution on the working disk area, and then the solvent was allowed to evaporate for 1 h at 60 °C under vacuum. The flexible electrode array was stuck on a rigid PVC substrate in order to be more conveniently handled. Polystyrene cylinders (300 µL) were precut in microtiter plates, and their bottoms were immersed in toluene for a very short period (< 1s) and then pressed onto the electrode array (Figure 2). Prior to the streptavidin/biotin affinity assay and the 2,4-D immunoassay, the polystyrene microwells equipped with a Nafion-SPE were precoated with BSA by adding 350 µL of PBS containing 0.1% (w/v) BSA (PBS/BSA) and incubating either for 2-3 h at room temperature or overnight at 4 °C. A platinum wire counter electrode and a Ag/AgCl wire reference electrode were inserted in the microwell-shaped electrochemical cell for electrochemical measurements, which were carried out at room temperature. After an accumulation period, the P+ salt was determined by CV (scan rate of 50 mV s-1) and the resulting anodic peak current (ip) corresponding to the oxidation of the phenolic function of P+ at ∼0.5 V vs Ag/AgCl was taken as the analytical response after blank subtraction.38 In a few experiments, a Vortex mixer (Heidolph Vibramax 110) from Bioblock Scientific was used to shake the multielectrochemical microwells. (39) Bagel, O. Thesis, University of Clermont-Ferrand, D.U. 1039, 1998.
Analytical Chemistry, Vol. 71, No. 13, July 1, 1999
2573
Construction of the Magnet Holding Block. A magnet holding block was constructed by piercing a PVC block (1.5 × 12 × 12 cm3) with four series of four cylindrical holes (1.8-mm diameters on 9-mm centers) and then inserting a cylindrical magnet (Francosid, 1.8-mm diameter) into each hole, making possible to position simultaneously four arrays of four electrodes, as shown in Figure 2 for one array. The precise alignment of the permanent magnet with the electrode disk area allowed the beads to be precisely reassembled onto the working electrode surface. Synthesis of AP-Labeled 2,4-D. The synthesis of 2,4-D labeled with AP (2,4-D-AP) was performed following the procedure described by Bauer et al.40 Briefly, 2,4-D (71.5 mM), TNTU (1 equiv), and NMM (1 equiv) were incubated for 15 min in 3 mL of dry DMF at room temperature. The activated 2,4-D was diluted 1:100 in DMF, and 22 µL of this solution was added to 200 µL of a buffer solution (pH 8.2; 0.1 M NaHCO3) containing 1800 units mL-1 AP, i.e., a 12-fold molar excess of activated ester over enzyme. After a 2-h reaction at room temperature, the reaction mixture was dialyzed overnight against 0.1 M TRIS, pH 7.6, containing 1 mM MgCl2, and then 2,4-D-AP was desalted with the same buffer on a Sephadex column. The purification was followed by colorimetry, i.e., by cutting off 1 µL of eluent every 0.5 mL and diluting each sample in 200 µL of TB containing the disodium salt of 4-nitrophenyl phosphate (10-3 M). The main fraction (∼1.5 mL) was stored at 4 °C (stock solution). The enzyme activities of 2,4-D-AP and AP were monitored electrochemically in TB (200 µL) containing S- (10-4 M) in a microwell equipped with a Nafion-SPE. After 30 min of enzyme incubation, the determination of P+ was carried out by CV and the activity of 2,4-D-AP in the main fraction was estimated to be equivalent to 36 units mL-1 free AP. Assuming that the main fraction contained ∼75% of the conjugate, it could be roughly estimated that the AP activity in 2,4-D-AP was 5-fold weaker than in the case of free AP. Streptavidin/Biotin Electrochemical Magnetic ParticleBased Assay. All of the reactions were performed at room temperature. The streptavidin-coated beads were washed with PBS/BSA before use to remove the NaN3 preservative, as recommended by the manufacturer. A suspension of 5 µL of the cleaned magnetic particles was introduced in each BSA-precoated electrochemical microwell and then allowed to bind with 10 µL of the diluted B-AP stock solution (7.5 × 10-3 unit mL-1) in 185 µL of PBS/BSA for 30 min. After incubation, the microwells were positioned on the magnet holding block, the supernatant was removed, and the particles were washed twice with 200 µL of PBS/ BSA, and twice with 280 µL of TB (pH 9.0). Each washing consisted of a resuspension of the IMBs in the buffer solution followed by separation with the magnet holding block for 3 min and removal of the liquid. Thereafter, 4 µL of S- stock solution and 200 µL of TB were added and the enzyme generation and accumulation of P+ were allowed to proceed simultaneously for 30 min. The accumulated P+ salt was finally determined by CV. In a few experiments, the incubation and washing steps were performed in a separate microtiter plate, and after resuspension in 200 µL of TB, the AP-associated beads were transferred into the microwell-shaped electrochemical cell containing S-, followed by enzyme reaction and CV measurements. (40) Bauer, C. G.; Eremenko, A. V.; Ehrentreich-Fo ¨rster, E.; Bier, F. F.; Makover, A.; Halsall, H. B.; Heineman, W. R.; Scheller, F. W. Anal. Chem. 1996, 68, 2453-8.
2574 Analytical Chemistry, Vol. 71, No. 13, July 1, 1999
Anti-2,4-D Antibody Titration Curve Procedure. To (250 - x) µL of PBS were added successively 50 µL of 2,4-D-AP (1:100 diluted 2,4-D-AP stock solution in PBS) and x µL of AbMBs (ranging from 5 to 200 µL). After thorough mixing, the solution was incubated for 40 min, and then the beads were magnetically separated for 3 min before removing the supernatant. The magnetic particles were then washed/resuspended once with 250 µL of PBS and twice with 250 µL of TB. The enzyme generation and accumulation of P+ (30 min), as well as its electrochemical detection, were carried out as described for the streptavidin/biotin procedure. 2,4-D Enzyme-Linked Immunomagnetic Electrochemical Procedure. A 210-µL aliquot of sample or standard solution of 2,4-D was introduced in the BSA-precoated electrochemical microwell and mixed thoroughly with 25 µL of 2,4-D-AP (1:100 diluted 2,4-D-AP stock solution in PBS) and 15 µL of Ab-MBs. After a 40-min incubation period, the beads were magnetically separated for 3 min and the liquid was removed. The magnetic particles were then washed/resuspended twice with 300 µL of TB followed by magnetic separation and removal of the supernatant after each washing. The enzyme generation and accumulation of P+ (30 min), as well as its electrochemical detection, were carried out as described above. RESULTS AND DISCUSSION The experiments were carried out at room temperature in the microwell-shaped 300-µL electrochemical cells shown in Figure 2. They are equipped with a small-size Nafion-SPE (1.5-mm diameter) occupying only 6% of the microwell bottom. A preliminary study concerning the reproducibility of the Nafion-SPEs was conducted by determining the CV response of the phenolic cobaltocenium salt (200 µL of 10-6 M P+ in TB) after a 30-min accumulation period under vortex agitation (∼600 rpm) and depletion of ∼12% P+.39 An average anodic peak current ip of 390 ( 35 nA corresponding to the oxidation of the phenolic function of P+ was obtained with a relative standard deviation (RSD) of 9% from two random batches of seven Nafion-SPEs. The main source of error was probably related to the manual fabrication procedure of the Nafion-SPEs. Electrochemical Characterization of the Synergetic Effect between IMBs and Nafion. To show the beneficial effect of the enzyme-linked immunomagnetic assay with electrochemical detection at a Nafion-SPE, the well-established streptavidin/biotin interaction (dissociation constant of 10-15) was first investigated as a model system to monitor AP-labeled biotin (B-AP) bound to streptavidin magnetic beads (S-MBs). Four main steps were involved in the affinity assay, i.e., (1) affinity reaction between S-MBs suspended in solution and B-AP, (2) magnetic separation and washing step, (3) addition of S- followed by the concomitant enzyme reaction and accumulation of P+ at a Nafion-modified SPE, and (4) CV detection of accumulated P+. The entire assay protocol was performed in microwell-shaped electrochemical cells precoated with BSA and using the magnet holding block (Figure 2, magnet of 1.8-mm diameter), unless otherwise stated. The selected experimental conditions took into account a series of preliminary experiments carried out in order to evaluate the influence of several parameters such as the working area of the SPE, the diameter of the magnet, the amounts of reagents, the reaction and accumulation times, and other previously optimized param-
Figure 3. CV curves (scan rate 50 mV s-1) recorded at (A) a bare SPE and (B) a Nafion-SPE immersed in 200 µL of TB (pH 9) containing 10-4 M S- and the complex B-AP/S-MBs. The enzyme generation and accumulation of P+ proceeded for 30 min, the beads being (a) magnetically localized at the electrode surface, (b) suspended in quiescent solution, and (c) suspended in an agitated solution (vortex, ∼600 rpm).
eters, i.e., the Nafion-film thickness.39 Consequently, a large excess of S-MBs (3.4 × 106 beads able to bind 2.2 × 10-11 mol of free biotin) was used in step 1 to capture a small amount of B-AP (1.6 × 10-16 mol) for 30 min in 200 µL of PBS/BSA (pH 7.4). Steps 3 and 4 proceeded in 200 µL of TB (pH 9.0) containing 10-4 M S(enzyme incubation and P+ accumulation for 30 min). Under these conditions, the blank signal obtained in the absence of B-AP was negligible. The beneficial effect of localizing the beads on the electrode surface is evidenced by the series of CV curves shown in Figure 3. These experiments were obtained by following the preceding protocol, except for steps 1 and 2, which were carried out in a separate microtiter plate and not directly in a microwell equipped with a SPE. Curves a and b in Figure 3A were recorded on a naked SPE, the beads being magnetically reassembled on the electrode surface during the 30-min incubation time for curve a. The anodic signal was ∼10 times higher than when the AP beads remained in suspension (curve b). The favorable effect of the magnetic accumulation was even more pronounced at a Nafion-SPE, as shown in Figure 3B, since the comparison of curves a and b indicates an 40-fold increase for the magnetic accumulation. Clearly, a dramatic enhancement in sensitivity (factor of 400) was reached in combining the use of magnet and Nafion (compare curve b in Figure 3A with curve a in Figure 3B), which allows
very sensitive enzyme immunoassays to be envisaged. The current response, thereby the sensitivity of the technique, was directly affected by the areas of both the Nafion-SPE and the magnet for a fixed amount of S-MBs. The peak current density, which reached 33 µA/cm2 under the conditions of curve a in Figure 3B, dropped to 2.6 µA/cm2 when a larger Nafion-SPE (working disk area of 9.6 mm2 occupying 34% area of the microwell bottom) was used in association with a plate-shaped magnet (5 × 1 cm2) positioned beneath the electrochemical microwells. This result can be explained by a lower coverage of the beads deposited on the sensing surface (only 34% beads was deposited on the large-size electrode). Taking into account the amount of beads (3.4 × 106 beads) and their diameter (2.8 ( 0.2 µm), it could be estimated that the beads would be packed in less than a monolayer (∼0.24 monolayer) on the large-size electrode, whereas in the case of the small Nafion-SPE (1.77 mm2) combined with a magnet of similar size, the beads would be roughly packed on ∼4 layers. The production of P+ close to the Nafion film surface did replace advantageously the bulk generation of P+ and its accumulation under vortexing conditions (compare curves a and c in Figure 3B). Indeed, the anodic peak current was ∼5 times lower in the latter case, which leads to the conclusion that the technique with magnetic accumulation is well-adapted to decentralized operations in a quiescent solution (no vortex agitation). When all of the steps proceeded in the same microwell-shaped electrochemical cell, an average electrode response of 690 ( 75 nA (eight assays, RSD
