An amperometric enzyme-channeling immunosensor

0 downloads 0 Views 675KB Size Report
Keywords: separation-free, rapid immunoassay, electrochemical, enzyme immu- noassay, antibody ... example of a 'pseudo-homogeneous' electro- chemical ...
Biosensors & Bioelectronics Vol. 12, No. 3, pp. 195-204, 1997 © 1997 Elsevier Science Limited Printed in Great Britain. All rights reserved 0956-5663/97/$17.00

ELSEVIER ADVANCED TECHNOLO~i'~

An amperometric enzyme-channeling immunosensor J. Rishpon* & D. Ivnitski Department of Molecular Microbiology & Biotechnology, Tel-Aviv University, Ramat-Aviv 69978, Israel Fax: [972]-3-640-9407 Email: [email protected] (Received 30 April 1996; accepted 30 September 1996)

Abstract: This paper presents a new disposable amperometric, enzyme-channeling immunosensor for a quantitative, rapid, separation-free enzyme immunoassay (EIA) that can be used in clinical diagnostics, as well as in biomedical, biochemical, and environmental research. The sensor consists of a disposable, polymer-modified, carbon electrode on which enzyme 1 is coimmobilized with a specific antibody that binds the corresponding antigen in a test solution. The solution also contains a conjugate of enzyme 2. An immunological reaction brings the two enzymes into close proximity at the electrode surface, and the signal is amplified through enzyme channeling. The localization of both enzymes on the electrode surface limits the enzymatic reactions to the polymer/membrane/electrode interface. The sensor overcomes the problem of discriminating between the signal that is produced by the immuno-bound enzyme label on the electrode surface and the background level of signal that emerges from the bulk solution. Combining enzyme-channeling reactions, optimizing hydrodynamic conditions, and electrochemically regenerating mediators within the membrane layer of the antibody electrode significantly increased the signalto-noise ratio of the sensor. The amperometric enzyme-channeling immunosensor enabled the performance of separation-free EIAs without washing steps, resulting in a relatively short assay time of 5-30 rain for the complete immunoassay, compared with at least 1-3 h for ELISA methods. Model systems using peroxidase-antibody, biotin-avidin, viral antigens (CD4-gpl20), and bacteria (Staphylococcus aureus) were investigated. S. aureus cells were detected in pure culture at concentrations as low as 1000 cells/ml. © 1997 Elsevier Science Limited. All rights reserved. Keywords: separation-free, rapid immunoassay, electrochemical, enzyme immunoassay, antibody electrode, enzyme channeling.

INTRODUCTION Modifying quantitative immunoassays for rapid, 'on-the-spot' performance in the field is important for clinical diagnostics, biochemical and biotechn-

*To whom correspondence should be addressed.

ological research, as well as for environmental monitoring. At present, quantitative immunoassays are usually performed only in major laboratories, which are equipped with complex instrumentation and highly qualified technical staff. A device operating in such remote sites as a physician's office, an emergency ward, or in field testing of environmental pollutants should not 195

J. Rishpon & D. Ivnitski

only provide fast, sensitive measurements but also be simple to operate and inexpensive. In this respect, an electrochemical device presents an attractive option. Consequently, devices that combine the high sensitivity and the relative simplicity of electroanalytical techniques with the high specificity of immunoenzymatic assays have shown a powerful analytical capability. Various types of electrochemical immunosensors have been developed (Jenkins et al., 1991; Foulds et al., 1990; Mirhabibollahi et al., 1990; Ngo, 1987; Hadas et al., 1992; Rishpon et al., 1992). The limiting factor in the development of rapid and separation-free electrochemical immunosensors, however, is the background noise induced by non-specific reactions. Such noise is most often due to the excess of enzyme-antigen conjugate in solution, which makes it difficult to discriminate between a signal that is obtained from small amounts of immunobound enzyme label and high background levels of signal that emanate from the conjugate in the bulk solution. The most general approaches to overcome the non-specific signal of the conjugate in solution are to link the enzyme label catalytically to an additional system, such as a substrate cycle, or to other enzymes to form 'cascades' (DiGleria et al., 1989; Harris, 1984; Litman et al., 1980). This viewpoint has led to the development of an enzyme-channeling immunoassay, where an immunological reaction brings two enzymes into immediate proximity at a surface. A novel, separation-free, sandwich-type enzyme immunoassay (EIA) using microporous gold electrodes and selfassembled monolayer/immobilized antibodies was described by Duan & Meyerhoff (1994). An example of a 'pseudo-homogeneous' electrochemical immunoassay without a washing step was described by McNeil et al. (1995). This assay was based on enzyme-channeling with direct electron transfer between an enzyme-conjugated antigen and the electrode. In previous research, we developed a rapid, one-step, separation-free, amperometric enzyme immunosensor for electrochemically monitoring in situ reactions with rabbit IgG and with human luteinizing hormone in human serum (Ivnitski & Rishpon, 1996). The sensor allowed preferential measurement of surface-bound conjugate relative to the excess enzyme-labeled regent in the bulk sample solution. The amplification of the analytical signal was achieved by employing an anionexchange PEI-GOD-antibody modified electrode 196

Biosensors & Bioelectronics

with iodide-ion mediators in solution and a peroxidase labeled conjugate. When performing mulitple in situ immunoassays, however, one problem that may be encountered is reusing the electrochemical cell after exposure to contaminated samples. A disposable electrochemical cell for use in such measurements would be invaluable. In this report we describe the construction of a disposable, electrochemical cell and polymer-modified, carbon electrode and its application in competition, displacement, and sandwich assays using diverse compounds and microorganisms.

EXPERIMENTAL

Materials Glucose oxidase (GOD) (EC 1.1.3.4, from Aspergillus niger, 150 IU/mg), horseradish peroxidase (HRP) (EC 1.11.1.7, RZ 3.1), goat anti-horseradish peroxidase, Staphylococcus aureus (formalin treated), bovine serum albumin (BSA), biotin, glutaraldehyde (50%) and polyethylenimine (50%) (PEI) were obtained from the Sigma Chemical Company (St Louis, MO). Recombinant, soluble CD4 and gpl20 proteins were obtained from American Bio-Technologies, Inc. (USA). GOD-conjugated egg-white avidin, CromPure rabbit IgG, whole-molecule (RblgG), AffiniPure goat anti-rabbit IgG (c~RblgG) and horseradish peroxidase-conjugated Affini-Pure goat antirabbit IgG (c~-RblgG-HRP) were obtained from the Jackson Immuno Research Laboratories, Inc. (West Grove, PA). Other chemicals of analytical grade were obtained from standard sources.

Preparation of the PEI membrane antibody electrode Disposable graphite electrodes in cylindrical form (made from pencil lead, HB 0.9 mm) were used as the working electrodes. The electrodes were cleaned in methanol, rinsed with double-distilled water, dried, and then dipped for 15 min in a 0.2% methanol solution of PEI and air-dried for 8 h. The electrodes were washed with methanol to remove excess unbound polymer and then immersed for 4 h in a 2.5% glutaraldehyde solution (pH 7.0). GOD and antibody were immobilized covalently to PEI film on the graphite electrode surface. The electrodes were then immersed

Biosensors & Bioelectronics

in 0.1 M phosphate buffer (pH 7) containing GOD (0.5 mg/ml) and antibody (0.5 mg/ml) and incubated for 8 h at 4°C. Unreacted aldehyde groups were blocked by soaking the electrodes for 1 h in a solution of 0.1 M glycine in 0.1 M phosphate buffer (pH 7). The electrodes were washed and then stored in 0.1 M phosphate buffer (pH 7) at 4°C. The activity of HRP was monitored spectrophotometrically at 403 nm, using an extinction coefficient of 1 × 105 1 mol -t cm -~ (Chen et al., 1984). Protein concentration was determined by UV absorbency at 280 nm using an extinction coefficient of 2.16 × 1051 mol -~ c m -1.

Scanning electron microscopy Electron microscopy of S. a u r e u s cells on the electrode surface was performed with a JEOL840a scanning electron microscope with an acceleration voltage of 25 kV, a magnification of 10,000, and an objective aperture of 1/xm.

Electrochemical measurements Figure 1 shows the schematic layout of the electrochemical cell. All electrochemical measurements were performed in the three-electrode cell (Vsolution=0.3ml) with a rotating cylindrical graphite antibody working electrode (0.9mm diameter), a graphite ink counter, and Ag/AgC1 ink reference electrodes, using 0.1 M acetate buffer (pH 5.6) containing 0.1 M NaCl and 3 mM KI. Standard electrochemical equipment for amperometric measurements can be applied in the monitoring of the sensor response. We used an EG & G Par 273 potentiostat interfaced to a PC 486 computer system with PAR M270 software. The rotation of the graphite electrode was performed with a Pine Instruments rotator and with an MSRS speed controller. The kinetic measurements were performed at a fixed electrode potential (0.0V versus Ag/AgC1) without washing steps. Cyclic voltammetry was conducted at a 20 mV/s scan rate with the RbIgG electrode.

Separation-free, amperometric immunoassay protocol A disposable antibody graphite electrode, with both immobilized GOD and antibody (or avidin) on the PEI film, was placed in a cell (V = 0.3 ml) containing 0.1 M acetate buffer (pH 5.6) with 0.1 M NaC1, 3 mM KI, and the sample of analyte

Amperometric enzyme-channeling immunosensor

and conjugate. The electrode was then rotated at a controlled speed (100-500 r.p.m.) and preincubated for 5-10 min. After a fixed period of incubation, the background current was stabilized for 2 min at 0-0 mV versus Ag/AgCI at 20°C. Glucose (0.01 ml, 0.01 M) was added to the cell, and changes in the current were monitored continuously and recorded for 1 min. The slope of the binding curve (A//At), which was directly visualized on a computer screen, was proportional to the amount of analyte present in the sample. A blank experiment was performed with the working electrode that was modified only by GOD and BSA. The concentrations of glucose and mediator in the buffer were chosen so that the current change was close to zero in the absence of a bound conjugate on the electrode surface.

GP120 assay procedure The graphite electrode, with both immobilized CD4 and GOD on the PEI film, was placed in a cell ( V = 0 . 3 ml) containing 0.1 M acetate buffer (pH 5-6), 0.1 M NaC1, and 3 mM KI. Aliquots (10/~1) of purified gpl20 at various concentrations in the interval 100-500 ng/ml or buffer alone (control) were injected into the cell and allowed to react with immobilized CD4 on the electrode surface at a controlled speed (500 r.p.m.) for 4 min at 0.0V versus Ag/AgC1. Conjugate (10/~1) (gpl20-HRP 10/xg protein/ml) was then added to the same cell, and after the background current reached stabilization (2 min), 10/xl of 100 mM glucose was added to the solution, and the catalytic current was monitored for 1 min.

Determination of The

S. a u r e u s

enzyme-channeling

immunoassay

of

S.

a u r e u s is based on the principle of a sandwich

assay. The working graphite electrode was coated by a PEI film with immobilized GOD and RblgG via a glutaraldehyde linker, and remaining sites were blocked with 0.1 M glycine as described above. Standard solutions of S. a u r e u s (102-106 cells/ml) were prepared in PBS. Two procedures were tested in the S. a u r e u s assay; four replicates of each procedure were performed. (1) The RblgG electrode was placed in the 197

J. Rishpon & D. lvnitski

Biosensors & Bioelectronics

rotor

W / " k

polystyrene tube test solution

r--y

---C

J

5 mm Fig. 1. Schematic layout of the electrochemical cell. W, working pencil lead electrode; C, counter electrode; R, reference electrode.

sample solution containing 0.1 M acetate buffer, 0.1 M NaCi, 3 mM KI, and 0.01% BSA. An aliquot (10/xl) of a S. aureus standard solution was added, and the RbIgG-GOD electrode was rotated at 1000r.p.m. for 10 rain at room temperature. Then the rotation speed was changed to 100r.p.m., and an aliquot (10/xl) of RbIgG-HRP (4/xg/ml) was added to the cell. After 10 min of incubation, an aliquot (10/xl) of glucose (0.1 M) was added to the cell, and the response to glucose was recorded. A control experiment was performed using a BSA- and GOD-modified working electrode. (2) The second procedure included separation and washing steps. The RbIgG-GOD electrode was incubated directly in the standard solutions. Immediately after the incubation with the bacteria, the electrode surface was thrice rinsed for 2 min per rinse. The next step was incubating the electrode with the 198

RblgG-peroxidase conjugate. The electrode was placed in the electrochemical cell containing 0-1 M acetate buffer, 0.1 M NaC1, 3 mM KI, and 130 ng/ml RblgG-HRP conjugate and incubated for another 10 min at 1000r.p.m. Five replicates of this procedure were performed. The background signal was measured and stabilized at 0.0 V versus Ag/AgC1 for 2 min at a controlled speed, typically 100r.p.m. At this point 10/xl of glucose (0.1 M) was added to the cell, and the catalytic current was monitored continuously for 1 min. The bacterial concentration in the stock suspension was counted with a hemocytometer (BS748, UK).

RESULTS AND DISCUSSION Figure 2 shows the principle of the pseudo-homogeneous enzyme-channeling immunoassay with

Biosensors & Bioelectronics

Amperometric enzyme-channeling immunosensor

ELECTRODE

(pencil lead)

GOD--

\ \

I

/ ~ I V ~ e _

Glucose +02 ) - ~ H202 + gluconic acid AB ~ A g - H R P

PEI FILM

\

'~ ,

I2+ 2 H 2 0 " ~ b "-xa--

H202 + 2I" +2H +

AB PEI FILM

Fig. 2. Schematic illustration of the principle of the separation-free, amperometric enzyme-channeling immunoassay with immobilized GOD and antibody on the PEl-modified electrode surface.

amperometrical detection. The active surface of the immunoelectrode consists of an antibody and glucose oxidase, which are coimmobilized on a PEI membrane. The test solution contains the antigen (the analyte), the HRP conjugate, and iodide. An immunological reaction brings the HRP conjugate into close proximity with the GOD on the electrode. The glucose solution reacts with GOD on the electrode surface and the product of this enzymatic reaction, H202, reacts with iodide ions in the presence of a surface-bound, peroxidase label. The iodine thus formed is monitored at the electrode. The concentration of H202, which is high at the electrode surface, is diluted when leaving the electrode surface to the solution. The PEI membrane plays a crucial role in

discriminating between the analytical signal (resulting from the immunocomplex at the electrode) and the background signal (resulting from the reaction of the HRP conjugate with H202 that leaches out from the electrode surface to the solution). Figure 3 shows the cyclic voltammograms of a bare electrode and a PEI-modified electrode. Clearly, the iodine/iodide electrode reaction is significantly enhanced in the PEImodified electrode because the iodide accumulates in the cationic polymer. Consequently, accumulation in the PEI membrane further increases the discrimination between the reaction of iodide ions with the HRP in the solution and at the electrode. The electrochemical response also depends on the concentration of the PEI in the organic solution that is used for the membrane preparation, show199

J. Rishpon & D. Ivnitski 70

l

I

Biosensors & Bioelectronics

I

I

I

I

I

I

0

i

I

I

-40

50

gl