An integrated microfluidic biochemical detection system for protein analysis with magnetic bead-based sampling capabilities

Abstract

This paper presents the development and characterization of an integrated microfluidic biochemical detection system for fast and low-volume immunoassays using magnetic beads, which are used as both immobilization surfaces and bio-molecule carriers. Microfluidic components have been developed and integrated to construct a microfluidic biochemical detection system. Magnetic bead-based immunoassay, as a typical example of biochemical detection and analysis, has been successfully performed on the integrated microfluidic biochemical analysis system that includes a surface-mounted biofilter and electrochemical sensor on a glass microfluidic motherboard. Total time required for an immunoassay was less than 20 min including sample incubation time, and sample volume wasted was less than 50 μl during five repeated assays. Fast and low-volume biochemical analysis has been successfully achieved with the developed biofilter and immunosensor, which is integrated to the microfluidic system. Such a magnetic bead-based biochemical detection system, described in this paper, can be applied to protein analysis systems.

---

Introduction

In the past few years, a large number of microfluidic prototype devices and systems have been developed, specifically for biochemical warfare detection systems and portable diagnostic applications.^1–7 The BioMEMS team at the University of Cincinnati has been developing a remotely accessible generic microfluidic system for biochemical detection and biomedical analysis, based on the concepts of both surface-mountable microfluidic motherboards and electrochemical detection techniques.¹ The basic concepts and development of several types of generic microfluidic subsystems were demonstrated, focusing on microfluidic performance. In this paper, however, in addition to developing a new integrated microfluidic biochemical detection and analysis system, a full magnetic bead-based immunoassay is reported as a typical example of biochemical analysis.

The limited goal of this work is to develop a generic MEMS-based microfluidic system and to apply the fluidic system to detect bio-molecules such as specific proteins and/or antigens in liquid samples. Fig. 1 illustrates the schematic diagram of a generic microfluidic system for biochemical detection using a magnetic bead approach for both sampling and manipulating the target bio-molecules.⁸ The analytical concept is based on sandwich immunoassay and electrochemical detection as illustrated in Fig. 2.

Fig. 1 Schematic diagram of a generic microfluidic system for biochemical detection.

Fig. 2 Analytical concept based on sandwich immunoassay and electrochemical detection.

Magnetic beads are used as the solid phase for the capture of antibodies, and as carriers of captured target antigens.^8-11 A simple concept of magnetic bead-based bio-sampling with an electromagnet for the case of sandwich immunoassay is shown in Fig. 3. Antibody coated beads are introduced to the electromagnet and separated by applying a magnetic field. While holding the antibody-coated beads, antigens are injected into the channel. Only target antigens are immobilized, and thus separated onto the magnetic bead surface due to antibody/antigen reaction. Other antigens get washed out with the flow. Next, enzyme-labeled secondary antibodies are introduced and incubated with the immobilized antigens. The chamber is then rinsed to remove all unbound secondary antibodies. Substrate solution, which will react with enzyme, is injected into the channel and the electrochemical detection is performed. Finally the magnetic beads are released to the waste chamber and the bio-separator is ready for another immunoassay.

Fig. 3 Conceptual illustration of bio-sampling and immunoassay procedure using magnetic bead approach: (a) injection of magnetic beads; (b) separation and holding of beads; (c) flowing sample; (d) immobilization of target antigen; (e) flowing labeled antibody; (f) electrochemical detection after adding enzyme substrate; and (g) washing out magnetic beads and ready for another immunoassay.

This paper describes the magnetic bead-based sampling and immunoassays obtained from the realized integrated microfluidic biochemical detection system, which includes a biofilter,⁸ an electrochemical sensor,¹² and microfluidic channels.

Microfluidic components and system

For the successful immunoassay, the biofilter and the electrochemical sensor were fabricated separately and integrated together as shown in Fig. 4. The reaction and sensing chamber volume is 750 nl. The integrated biofilter and electrochemical sensor were surface-mounted using a fluoropolymer bonding technique¹³ on a microfluidic motherboard, which contains microchannels fabricated by a glass etching and glass to glass direct bonding technique. All microchannels are 400 μm wide and 100 μm deep. The basic structural concept of the fluidic interconnection is illustrated in Fig. 5. This design enables the five different solutions that contain the reagents needed for an immunoassay to be sequentially injected into the chamber with the biofilter and the electrochemical sensor.

Fig. 4 Microphotograph of the integrated biofilter and biosensor. The volume of the fluidic chamber for biofiltration, reaction, and detection was calculated to be 750 nl.

Fig. 5 Illustration of microfluidic interconnection and surface-mounting technique.

Microchannel structures were designed to minimize interactions between biochemical reagents to increase detection and analysis sensitivity. Each inlet and outlet was connected to sample reservoirs through custom-designed microvalves. Fig. 6 shows the integrated microfluidic biochemical detection system for magnetic bead-based immunoassay.

Fig. 6 Photograph of the fabricated microfluidic biochemical detection system for magnetic bead-based immunoassay.

Electrochemical detection schemes

Electrochemical detection techniques are well suited for biosensor systems that depend on chemical reactions for biological recognition. Since the electrochemical techniques measure the results of the biological recognition in terms of electrical parameters, they can be directly interfaced to an electronic computational system for further signal processing as well as to an integrated microfluidic biochemical detection and analysis system. Electrochemical detection is coming into widespread use for the trace determination of oxidizable and reducible organic and inorganic compounds because electrochemical cells are easily miniaturized for incorporation in a microfluidic system.^14–16

Interdigitated array (IDA) microelectrodes are particularly interesting because they are geometrically different from conventional electrodes and this results in particular electrochemical behavior.^17–19 IDAs have various advantages. These include their ability to enhance the current response while retaining the properties of a single microelectrode, which gives them high sensitivity as detectors for flow injection analysis and liquid chromatography. Additional sensitivity can be gained by maintaining the two interdigitated electrodes at different potentials to cause redox cycling. For this reason, an IDA was fabricated for use as the electrochemical detector for this device. In this work the IDA with both electrodes held at the same potential was used for detection of the magnetic bead-based immunoassay.

Alkaline phosphatase (AP) and p-aminophenyl phosphate (PAPP) were chosen as enzyme and electrochemical substrate, respectively. Alkaline phosphatase converts PAPP to the electrochemical product, p-aminophenol (PAP). By applying an oxidizing potential to the IDA electrode, PAP is converted into 4-quinoneimine (4-QI) by a 2-electron oxidation. Fig. 7 illustrates the electrochemical detection principle. Alternatively, one set of fingers can be used for this oxidation process and the other set held at a more negative potential to reduce 4-QI to PAP (i.e. redox cycling). This mode was not used in this work.

Fig. 7 Enzymatic kinetics for electrochemical detection of the immunosensor. Oxidation of PAP at the IDA.

Magnetic bead-based immunoassay

After fluidic sequencing tests, immunoassays were performed in the integrated microfluidic system to demonstrate the magnetic bead-based biochemical detection and sampling function.

Magnetic beads (Dynabeads^® M-280, Dynal Biotech Inc.) coated with biotinylated sheep anti-mouse immunoglobulin G (IgG) were injected into the reaction chamber and captured on the surface of the biofilter by applying a magnetic field. While holding the magnetic beads, antigen (mouse IgG) was injected into the chamber and incubated. Then secondary antibody with label (rat anti-mouse IgG conjugated alkaline phosphatase) and enzyme substrate (PAPP) were injected sequentially and incubated to produce PAP. For electrochemical detection of PAP, both sets of interdigitated electrodes were maintained at the same potential. At a fixed time after adding PAPP, a potential of 290 mV was applied to the IDA and the current was monitored with time as the output signal. After detection, magnetic beads with all reagents were washed away and the system was ready for another immunoassay. The sequence used for the immunoassay is summarized in Table 1. This sequence was repeated for every new immunoassay. The flow rate was set to 20 μl min⁻¹ in every step.

Table 1 Sequence of the magnetic bead-based immunoassay

(1)	Injection of primary antibody coated magnetic beads (biotinylated sheep anti-mouse IgG on magnetic beads) for 2 min
(2)	Flowing buffer for 30 s
(3)	Injection of antigen (mouse IgG) for 30 s
(4)	Incubation for 5 min
(5)	Flowing buffer for 30 s
(6)	Injection of labeled antibody (alkaline phosphatase labeled rat anti-mouse IgG) for 30 s
(7)	Incubation for 5 min
(8)	Flowing buffer for 30 s
(9)	Injection of substrate (PAPP) for 30 s
(10)	Incubation for 5 min
(11)	Detection for 1 min
(12)	Flushing everything out
(13)	The system is ready for another assay
(14)	Assay time: Less than 20 min at 20 µl min^−1 flow rate

---

Experimental results and discussion

Before the actual magnetic bead-based immunoassays, the electrochemical sensor was tested and calibrated by flowing standard PAP solution into the sensing chamber. The calibration curve of the electrochemical immunosensor for different PAP concentrations is shown in Fig. 8. After calibrating the electrochemical sensor, full immunoassays were performed following the sequence in Table 1 for different antigen concentrations. The concentrations of primary antibody-coated magnetic beads and conjugated secondary antibody were 1.02 × 10⁷ beads ml⁻¹ and 0.7 μg ml⁻¹, respectively. Immunoassay results for different antigen concentrations are shown by the current–time signals in Fig. 9.

Fig. 8 Calibration curve of the electrochemical immunosensor for PAP, the enzyme product.

After injection of PAPP and incubation, PAP was detected by chronoamperometry. In this technique, the potential of the IDA was stepped to 290 mV (time zero in Fig. 9) and a large anodic (negative) current occurred due to rapid oxidation of accumulated PAP. This current decayed toward zero current as PAP was consumed by oxidation to 4-QI. During this time the solution in the detection chamber was quiescent and mass transport of PAP to the IDA was by diffusion. Then injection of washing reagent was initiated, which caused an abrupt increase in anodic current due to the enhanced mass transport of PAP to the IDA by convection. The current peaked and then decayed rapidly toward zero as the remaining PAP was flushed from the chamber. Finally the current stabilized at a residual current for that potential in the absence of PAP. Two methods for measuring the current signal were evaluated. Both used the residual current after washout as a baseline for measurement. In one method the decayed anodic current was measured just before injection of washing reagent; in the other method the anodic peak current that accompanies injection of washing reagent was measured. In both cases, a substantial increase in current (signal) was observed in going from 50 ng ml⁻¹ to 100 ng ml⁻¹ of the analyte mouse IgG, as expected for a sandwich immunoassay. Further increases in mouse IgG concentration resulted in a plateau in the signal, which is consistent with saturation of the capture antibody with captured IgG. Thus, it would appear that the working range for this assay is below 100 ng ml⁻¹.

Fig. 9 Immunoassay results measured by chronoamperometric detection of PAP. Sample consumed during one immunoassay was 10 μl (20 μl min⁻¹ × 30 s) and total assay time was less than 20 min including all incubation and detection steps.

Sample consumed during one immunoassay was 10 μl (20 μl min⁻¹ × 30 s) and the total assay time was less than 20 min, including all incubation and detection steps. The assay time and volume depend on the flow rate and incubation time, so the optimization of the immunoassay condition with microprocessor control is currently being investigated.

These results demonstrate that all of the steps required for a complete immunoassay can be achieved with this microfluidic system acting in concert with magnetic beads as a mobile antibody substrate. Several key factors for immunoassay were demonstrated. The ability to magnetically trap the beads in the reaction chamber and then subject them to the several transfers of solution during an immunoassay was demonstrated. PAP was detected electrochemically by measuring current in the nanoampere range. Beads and reagents could be flushed from the reaction chamber sufficiently well for repetitive assays.

Conclusion

The integrated microfluidic biochemical detection system has been successfully developed and fully tested for fast and low volume immunoassays using magnetic beads, which are used as both immobilization surfaces and bio-molecule carriers. Magnetic bead-based immunoassay, as a typical example of biochemical detection and analysis, has been performed on the integrated microfluidic biochemical analysis system that includes a surface-mounted biofilter and immunosensor on a glass microfluidic motherboard. Protein sampling capability has been demonstrated by capturing target antigens.

The microfluidic system and magnetic bead capture methodology, which has been developed in this work, can also be applied to generic bio-molecule detection and analysis systems by replacing antibody/antigen with appropriate bioreceptors/reagents, such as DNA fragments or oligonucleotides, for detection and analysis of a wide variety of biological materials as well as to DNA analysis and high throughput protein analysis.

Acknowledgement

This research was fully supported by a DARPA grant under contract AF F30602-97-2-0202 from MicroFlumes Program, MTO/DoD, USA.

References

1. C. H. Ahn, H. T. Henderson, W. R. Heineman and H. B. Halsall, Proceedings of the Micro Total Analysis Systems ’98 (μ-TAS ’98) Workshop, pp. 225–230, 1998.
2. C. Bisson, J. Campbell, R. Cheadle, M. Chomiak, J. Lee, C. Miller, C. Milley, P. Pialis, S. Shaw, W. Weiss and C. Widrig, Technical Digest of Solid-State Sensor and Actuator Workshop 1998, 1998, pp. 1–6.
3. S. C. Jakeway, A. J. de Mello and E. L. Russell, Fresenius’ J. Anal. Chem., 2000, 366, 525.
4. G. H. W. Sanders and A. Manz, Trends Anal. Chem., 2000, 19, 364.
5. R. D. Oleschuk and D. J. Harrison, Trends Anal. Chem., 2000, 19, 379.
6. P. Bergveld, Biomed. Microdevices, 2000, 2, 185.
7. O. J. A. Schueller, D. C. Duffy, J. A. Rogers, S. T. Brittain and G. M. Whitesides, Sensors Actuators A, 1999, 78, 149.
8. J.-W. Choi, C. H. Ahn, S. Bhansali and H. T. Henderson, Sensors Actuators B, 2000, 64, 34.
9. G. M. Whitesides, R. J. and L. Josephson, Trends Biotechnol., 1983, 1, 144.
10. M. Zborowski, L. Sun, L. R. Moore, P. S. Williams and J. J. Chalmers, J. Magn. Magn. Mater., 1999, 194, 224.
11. H. Andersson, W. van der Wijngaart, P. Enoksson and G. Stemme, Sensors and Actuators B, 2000, 67, 203.
12. W. R. Heineman, H. B. Halsall, M. Cousino, C. A. Wijayawardhana, S. Purushothama, S. Kradtap, K. Schlueter, J.-W. Choi, C. H. Ahn and H. T. Henderson, Paper presented at Pittcon ’99, Orlando, FL, USA, March 7–12, 1999.
13. A. Han, K. W. Oh, S. Bhansali, H. T. Henderson and C. H. Ahn, Proceedings of IEEE MEMS ’00, IEEE, New York, 2000, pp. 414–418.
14. A. T. Woolley, K. Lao, A. N. Glazer and R. A. Mathies, Anal. Chem., 1998, 70, 684.
15. G. T. A. Kovacs, R. J. Reay, A. F. Flannery, S. P. Kounaves and C. W. Storment, Sensors Actuators B, 1996, 34, 450.
16. T. Neufeld, I. Eshkenazi, E. Cohen and J. Rishpon, J. Biosensors Bioelectron., 2000, 15, 323.
17. O. Niwa, Y. Xu, H. B. Halsall and W. R. Heineman, Anal. Chem., 1993, 65, 1559.
18. M. Paeschke, U. Wollenberger, T. Lisec, U. Schnakenberg and R. Hintsche, Sensors Actuators B, 1995, 26–27, 394.
19. A. E. Cohen and R. R. Kunz, Sensors Actuators B, 2000, 62, 23.

---