471
Electrophoresis 2015, 36, 471–474
Shanshan Li1,2,3 ∗ Yukun Ren1 ∗ Haochen Cui2 Quan Yuan2 Jie Wu2 ∗∗ Shigetoshi Eda4 Hongyuan Jiang1 1 School
of Mechatronics Engineering, Harbin Institute of Technology, Harbin, China 2 Department of Electrical Engineering and Computer Science, The University of Tennessee, Knoxville, TN, USA 3 School of Mechanical Engineering, Hebei University of Technology, Tianjin, China 4 Department of Forestry, Wildlife and Fisheries, The University of Tennessee, Knoxville, TN, USA
Received June 11, 2014 Revised September 6, 2014 Accepted September 8, 2014
Short Communication
Alternating current electrokinetics enhanced in situ capacitive immunoassay A rapid in situ capacitive immunoassay is presented herein. Conventional immunoassay typically relies on diffusion for transport of analytes in many cases causing long detection time and lack of sensitivity. By integrating alternating current electrokinetics (ACEK) and impedance sensing, this work provides a rapid in situ capacitive affinity biosensing. ACEK induces both fluid flow and particle motion, conveying target molecules toward electrodes immobilized with probes, resulting in rapid enrichment of target molecules and a capacitance change at the ‘‘electrode-fluid’’ interface. The benefit of ACEK enhanced immunoassay was demonstrated using the antigen and antibody from Johne’s disease (JD) as an example. To clarify the importance of DEP and ACET effects for binding reaction, two different electrode pattern designs for capacitive immunoassay are studied. The asymmetric array and symmetric electrodes exhibit very similar response at lower electric field due to DEP effects, while asymmetric array has remarkable higher response at high-electric field because the convection becomes more important at high field. The disease positive and negative serum samples are distinguished in few minutes. Keywords: AC electrothermal effect / Capacitive affinity sensing dielectrophoresis / Immunosensor DOI 10.1002/elps.201400284
Additional supporting information may be found in the online version of this article at the publisher’s web-site
Immunoassay is a macromolecule detection technique widely used in immunological study, infectious disease diagnostics, detection of biological warfare agents, and environmental diagnostics. Conventional immunoassays, such as microarrays or ELISA, rely on molecular diffusion of analytes to bind with immobilized receptor to achieve detection. This could lead to long incubation time, ranging from minutes to several hours, and could pose limitations to sensitivity and throughput of immunoassay [1]. Moreover, florescent label or specific enzyme is necessary for signal reporting. For diffusion-limited heterogeneous reaction, external stirring is highly desired for enhancing the transport of macromolecules, so as to improve the detector sensitivity and assay time [2]. ACEK effects, inducing particle or fluid
Correspondence: Professor Hongyuan Jiang, School of Mechatronics Engineering, Harbin Institute of Technology, Harbin, 150001 P. R. China E-mail: [email protected] Fax: +86-451-86402658
Abbreviations: ACEK, alternating current electrokinetics; ACEO, alternating current electroosmosis; ACET, alternating current electrothermal; JD, Johne’s disease; SAW, surface acoustic wave C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
motion by externally applied AC electric fields, can be readily employed to accelerate the movement of macromolecules toward sensing areas, which has been reported by several groups [3–5]. In the aforementioned reports, the ACEK concentration step is separate from the detection step, making the sensor more complicated to operate than the capacitive immunoassay reported here. In this work, a rapid and straightforward immunosensing technique is presented, which merges ACEK effects with impedance measurement to achieve a single-step operation without any wash steps. The capacitive immunosensing mechanism is illustrated in Fig. 1A. Prior to tests, the microelectrode surface will be functionalized with receptor. Protein-protein interactions could lead to macromolecule deposition at the electrode surface, which may cause a change in the thickness or dielectric properties at the electrolyte–electrode interface. Figure 1B shows a generic equivalent circuit network to represent the impedance of an electrode cell, where Rct , Cint , Rs , Cs stand for charge transfer resistance at the fluid/electrode interface, interfacial capacitance, fluid
∗
These authors have contributed equally to this work. Additional corresponding author: Dr. Jie Wu, E-mail: [email protected]
∗∗
Colour Online: See the article online to view Figs. 1–3 in colour.
www.electrophoresis-journal.com
472
S. Li et al.
Electrophoresis 2015, 36, 471–474
frequency; subscript p and m denote particle and medium, respectively. a and are particle diameter and fluid viscosity, respectively. For nanoscale macromolecules, DEP may be effective only when molecules are located within a very short distance to the electrodes. So DEP enrichment of macromolecules would benefit from fluid movements that carry molecules to the electrodes for binding. AC electric field can generate microflows through one or both of two ACEK flow mechanisms known as ACEO [9] and ACET [10] effect. ACEO typically dominates at low ionic strength. ACEO flow velocity has been observed to decrease significantly with increasing fluid conductivity and eventually drop to zero above 0.085 S/m, unsuitable for most bioassays [5]. This work uses 15 mM Na+ PBS with 0.18 S/m, so ACEO flow will be negligible. In comparison, ACET effect could generate microflows as [10, 11]: Figure 1. (A) Conceptual diagram of ACEK enhanced capacitive biosensing and interfacial capacitance change by the binding process. Dimensions are not drawn to scale. (B) Equivalent circuit of the ‘‘electrode-fluid’’ system. Cint : interfacial capacitance, Rct : charge transfer resistance, Rs : electrolyte resistance, Cs : electrolyte capacitance. The dashed line Cs is regarded as open circuit under our frequency.
resistance, and fluid capacitance, respectively [6]. When macromolecules are adsorped onto the electrode surface and add to the thickness of the surface layer, the interfacial capacitance change is expressed as ⌬ C/Cint,0 = −dana /[(drec + dana ) + edl ε p /ε s ], where Cint, 0 is the initial interfacial capacitance, dana , drec , and edl are the thickness of analyte layer, receptor layer, and electric double layer, which naturally forms at the interface between solid and electrolyte. ε p , εs are the dielectric constants of macromolecules and sample fluids. Based on impedimetric characterization of the sensing electrodes, the electrode cell is adequately represented by a serial connection of Rs and Cint from 10 kHz and 1 MHz [6], shown as the solid lines in Fig. 1B. Coincidentally, enhancement of binding reaction is found to be the most obvious around 100 kHz [7]. Therefore, our ACEK capacitive sensing is conducted at 100 kHz, and the changes in Cint is conveniently monitored by direct measurement of the sensor capacitance. Here besides capacitive sensing, AC signal has another function for concurrent analyte enrichment. When an inhomogeneous AC electric field is applied through microelectrodes to an aqueous solution, particle and fluid movement could be induced around the microelectrodes to achieve in situ concentration of macromolecules. Direct particle movement can be caused by DEP, and particle can also be carried by microflows such as alternating current electroosmosis (ACEO) or AC electrothermal (ACET) flows to reach the microelectrodes. DEP velocity of a spherical particle is expressed as [8]: ∗ ε ∗p − εm a 2 εm uDEP = (1) ∇|E |2 , Re ∗ ∗ 6 ε p + 2εm where ε ∗ = ε − j / is the complex permittivity with ε, , and being permittivity, conductivity, and angular C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
uACET = f et l 2 / ∝ |E |2 ∇T ∝ |E |4 ,
(2)
in which l is the characteristic length of the device, typically on the order of electrode spacing, fet is the body force. Combining DEP and ACET effect, nanoscale particles is concentrated over a large distance to realize accelerated detection. The magnitudes and ranges of DEP and ACET velocities depend on the electrode geometries. Previously, we reported a capacitive immunosensor using microelectrodes modified from off-the-shelf surface acoustic wave (SAW) resonator chips [12]. This work studies a different electrode design for capacitive immunoassay to better exploit ACEK effects for sensitivity and acceleration in immunoassay. Microelectrode array used here is planar thin film electrodes on silicon wafer with asymmetric widths of 5 and 25 m separated by 5 and 25 m gaps, hence denoted as 5/5/25/25 electrodes. This design is able to induce ACET microflows for biofluid convection effectively [10,11,13–15], and yet has adequate electrode surface for interfacial capacitance sensing. The electrodes were functionalized with antigen extracted from Mycobacterium avium subsp. paratuberculosis, a causative agent of Johne’s disease (JD). The quality of prepared electrodes was also checked by measuring the interfacial capacitance, as described in [16]. A microchamber for loading the serum sample (5 L) was formed by sealing a PDMS microchannel (L × W × H = 1 mm × 300 m × 40 m) over the electrodes, as shown in Fig. 2B. The sample fluid quickly reaches static equilibrium in the microchannel. The microchannel is sufficiently high since electric fields and ACEK effects are mostly limited within 25 m from the electrodes, shown by numerical simulation [5]. Then its capacitance was measured at 100 kHz by an impedance analyzer (Agilent 4294A). The whole detection process finished at the end of capacitance measurement, usually within 2 min, much simpler and faster than multiplewash/incubation, hour-long operation of ELISA. ACEK capacitive sensing method uses dimensionless change in the measured capacitance (capacitance change divided by the initial capacitance at time zero) as the metric to indicate antibody deposition on the electrode, which is the slope of www.electrophoresis-journal.com
Electrophoresis 2015, 36, 471–474
Microfluidics and Miniaturization
473
Figure 2. Experimental setup and proof of concept results. (A) Schematic of experimental set up for ACEK capacitive immunoassay. Interfacial capacitance of microelectrode sensor is read by impedance analyzer connected to a laptop via GPIB interface. (B) Schematic side-view (top) and 3D view (bottom) of the microelectrode sensing chamber. (C) Change of dimensionless interfacial capacitance with time for disease positive/negative serum, and control buffer.
dimensionless capacitance versus time and found by least square linear fitting method. Each test was repeated for five times and the CV was calculated. More details are described in [12, 17]. Figure 2C demonstrates how the dimensionless interfacial capacitance typically behaves with time, for a JD positive/negative serum sample and a control sample (15 mM PBS). Measurements were performed using a 100 kHz AC signal of 500 mV. Here, the positive serum exhibited a capacitance change rate of −81‰/min, while the change rates from the negative serum and control sample were less than −5‰/min. To demonstrate the effect of AC electric field on immunosensing, JD positive samples were tested with AC potentials from 50 mV to 1 V for 2 min. The corresponding sensor responses were plotted in Fig. 3A, with electric field strength estimated as AC potential/electrode gap. As can be seen, the capacitance change rate increases rapidly with increasing electric field strength, while the control sample exhibited consistently low change rate. The JD positive sample showed significantly higher change rates than the control samples except for the results at 10 mV/m (i.e. 50 mV), which is too low to induce any ACEK effects to enhance specific binding. Therefore, if a low voltage (i.e. 50 mV or lower) is used as in conventional impedance sensing, no detection result can be reached within 2 min due to too small a difference between disease-positive and control sample. For comparison, also plotted in Fig. 3A are the capacitance change rates from applying AC voltages of 5 to 200 mV on symmetric SAW electrodes (gap/width = 1.1 m/1.4 m). The capacitance change rates are very similar between the two electrodes when electric field strength is below 60 mV/m. At higher electric field strength, the response from 5/5/25/25
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
electrodes continues to rise, while there is no increase from the SAW electrodes after 100 mV/m. At 100 mV/m, the capacitance change rates from the 5/5/25/25 electrodes and SAW electrodes are (−77.75 ± 8.30)‰/min with a CV of 10.675%, and (−48.32 ± 7.81)‰/min with a CV of 16.16%, respectively. Figure 3B shows a capacitance change rate of 5.09‰ ± 0.38‰ for 1 ng/mL sample, which is more than 3 standard deviations from that of control sample, −0.95‰ ± 0.90‰. Therefore, the LOD was determined to be 1 ng/mL, a noticeable improvement from the SAW electrodes’ LOD of 10 ng/mL [12, 17], also at 500 mV. To clarify the importance of DEP and ACET effects for binding reaction, numerical simulation using COMSOL Multiphysics was performed for these two different electrode designs (see Supporting Information), with procedures similar to the description in [18, 19]. The simulation also reveals that the difference in sensor response is attributed to the electric field distribution induced by different electrode design. For SAW electrodes, the electric fields are concentrated around the electrode corners and between the electrodes, which is caused by an electrode thickness of 150 nm and a separation of 1.1 m. In comparison, 5/5/25/25 electrodes have a thickness of 100 nm and a separation of 5 m. Therefore, with 5/5/25/25 electrodes electric field reaches further into the fluids and produces much larger ACET vortices than SAW electrodes. Based on Eqs. (1) and (2), DEP velocity uDEP ∝ |E |2 and ACET velocity uACET ∝ |E |4 , so ACET convection becomes more important at higher electric field. Thus 5/5/25/25 and SAW electrodes exhibit very similar responses at lower field (⬍60 mV/m), and noticeably higher response for 5/5/25/25 electrode at higher field. To further our understanding of ACET capacitive sensing, numerical simulation was also performed using commercial finite element software Comsol Multiphysics 3.5a,
www.electrophoresis-journal.com
474
S. Li et al.
Electrophoresis 2015, 36, 471–474
Figure 3. (A) Dimensionless capacitance change rate as a function of electric field strength for JD-positive serum and control samples. Data from two types of electrodes, 5/5/25/25 and SAW electrodes, were compared; (B) Dimensionless capacitance change rate for concentration from 0.2 to 10 ng/mL. IgG antibody binding to immobilized IgG whole molecules were measured with 500 mV, 100 kHz AC signal for 2 min.
the model settings and simulation results are provided in the Supporting Information. To sum up, ACEK capacitive sensing method can provide rapid, sensitive, and easy-to-operate detection of specific proteins in biological relevant samples. The electrode design presented here demonstrated an improvement in detection sensitivity over our prior report [12]. We acknowledge the financial support of the U.S. National Science Foundation (ECS-0448896), China Natural Science Foundation (No. 11372093, 51305106), the University of Tennessee Research Foundation Maturation Fund, M-CERV seed grant and AgResearch Innovation Grant. Microfabrication was conducted at the Center for Nanophase Materials Sciences, which is sponsored at Oak Ridge National Laboratory by the Scientific User Facilities Division, Office of Basic Energy Sciences, U. S. Department of Energy. The authors have declared no conflict of interest.
References [1] Zhao, W., Zhang, L., Jing, W., Liu, S., Tachibana, H., Cheng, X., Sui, G., Biomicrofluidics 2013, 7, 011101. [2] Hamada, R., Takayama, H., Shonishi, Y., Mao, L., Nakano, M., Suehiro, J., Sens. Actuators B: Chem. 2013, 181, 439–445. [3] Cheng, I.-F., Yang, H.-L., Chung, C.-C., Chang, H.-C., Analyst 2013, 138, 4656–4662. [4] Islam, N., Lian, M., Wu, J., Microfluid Nanofluidics 2007, 3, 369–375.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[5] Wu, J., Ben, Y., Battigelli, D., Chang, H.-C., Ind. Eng. Chem. Res. 2005, 44, 2815–2822. [6] Li, S., Yuan, Q., Morshed, B. I., Ke, C., Wu, J., Jiang, H., Biosens. Bioelectron. 2013, 41, 649–655. [7] Liu, X., Yang, K., Wadhwa, A., Eda, S., Li, S., Wu, J., Sens. Actuators A: Phys. 2011, 171, 406–413. [8] Castellanos, A., Ramos, A., Gonzalez, A., Green, N. G., Morgan, H., J. Phys. D Appl. Phys. 2003, 36, 2584. [9] Ramos, A., Morgan, H., Green, N., Castellanos, A., J. Phys. D Appl. Phys. 1998, 31, 2338. ´ [10] Green, N. G., Ramos, A., Gonzalez, A., Castellanos, A., Morgan, H., J. Electrostat. 2001, 53, 71–87. [11] Wu, J., Lian, M., Yang, K., Appl. Phys. Lett. 2007, 90, 234103. [12] Li, S., Cui, H., Yuan, Q., Wu, J., Wadhwa, A., Eda, S., Jiang, H., Biosens. Bioelectron 2014, 51, 437–443. [13] Yuan, Q., Wu, J., Biomed. Microdevices 2013, 15, 125–133. [14] Yuan, Q., Yang, K., Wu, J., Microfluidics Nanofluidics 2014, 16, 167–178. [15] Yang, K., Wu, J., Biomicrofluidics 2010, 4, 034106. [16] Cui, H., Cheng, C., Wu, J., Eda, S., Sensors, 2013 IEEE, IEEE 2013, pp. 1–4. [17] Cui, H., Li, S., Yuan, Q., Wadhwa, A., Eda, S., Chambers, M., Ashford, R., Jiang, H., Wu, J., Analyst 2013, 138, 7188–7196. [18] Li, S., Ren, Y., Jiang, H., RSC Adv. 2014, 4, 9064–9071. [19] Hu, G., Gao, Y., Li, D., Biosens. Bioelectron. 2007, 22, 1403–1409.
www.electrophoresis-journal.com
Electrophoresis 2015, 36, 471–474
Shanshan Li1,2,3 ∗ Yukun Ren1 ∗ Haochen Cui2 Quan Yuan2 Jie Wu2 ∗∗ Shigetoshi Eda4 Hongyuan Jiang1 1 School
of Mechatronics Engineering, Harbin Institute of Technology, Harbin, China 2 Department of Electrical Engineering and Computer Science, The University of Tennessee, Knoxville, TN, USA 3 School of Mechanical Engineering, Hebei University of Technology, Tianjin, China 4 Department of Forestry, Wildlife and Fisheries, The University of Tennessee, Knoxville, TN, USA
Received June 11, 2014 Revised September 6, 2014 Accepted September 8, 2014
Short Communication
Alternating current electrokinetics enhanced in situ capacitive immunoassay A rapid in situ capacitive immunoassay is presented herein. Conventional immunoassay typically relies on diffusion for transport of analytes in many cases causing long detection time and lack of sensitivity. By integrating alternating current electrokinetics (ACEK) and impedance sensing, this work provides a rapid in situ capacitive affinity biosensing. ACEK induces both fluid flow and particle motion, conveying target molecules toward electrodes immobilized with probes, resulting in rapid enrichment of target molecules and a capacitance change at the ‘‘electrode-fluid’’ interface. The benefit of ACEK enhanced immunoassay was demonstrated using the antigen and antibody from Johne’s disease (JD) as an example. To clarify the importance of DEP and ACET effects for binding reaction, two different electrode pattern designs for capacitive immunoassay are studied. The asymmetric array and symmetric electrodes exhibit very similar response at lower electric field due to DEP effects, while asymmetric array has remarkable higher response at high-electric field because the convection becomes more important at high field. The disease positive and negative serum samples are distinguished in few minutes. Keywords: AC electrothermal effect / Capacitive affinity sensing dielectrophoresis / Immunosensor DOI 10.1002/elps.201400284
Additional supporting information may be found in the online version of this article at the publisher’s web-site
Immunoassay is a macromolecule detection technique widely used in immunological study, infectious disease diagnostics, detection of biological warfare agents, and environmental diagnostics. Conventional immunoassays, such as microarrays or ELISA, rely on molecular diffusion of analytes to bind with immobilized receptor to achieve detection. This could lead to long incubation time, ranging from minutes to several hours, and could pose limitations to sensitivity and throughput of immunoassay [1]. Moreover, florescent label or specific enzyme is necessary for signal reporting. For diffusion-limited heterogeneous reaction, external stirring is highly desired for enhancing the transport of macromolecules, so as to improve the detector sensitivity and assay time [2]. ACEK effects, inducing particle or fluid
Correspondence: Professor Hongyuan Jiang, School of Mechatronics Engineering, Harbin Institute of Technology, Harbin, 150001 P. R. China E-mail: [email protected] Fax: +86-451-86402658
Abbreviations: ACEK, alternating current electrokinetics; ACEO, alternating current electroosmosis; ACET, alternating current electrothermal; JD, Johne’s disease; SAW, surface acoustic wave C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
motion by externally applied AC electric fields, can be readily employed to accelerate the movement of macromolecules toward sensing areas, which has been reported by several groups [3–5]. In the aforementioned reports, the ACEK concentration step is separate from the detection step, making the sensor more complicated to operate than the capacitive immunoassay reported here. In this work, a rapid and straightforward immunosensing technique is presented, which merges ACEK effects with impedance measurement to achieve a single-step operation without any wash steps. The capacitive immunosensing mechanism is illustrated in Fig. 1A. Prior to tests, the microelectrode surface will be functionalized with receptor. Protein-protein interactions could lead to macromolecule deposition at the electrode surface, which may cause a change in the thickness or dielectric properties at the electrolyte–electrode interface. Figure 1B shows a generic equivalent circuit network to represent the impedance of an electrode cell, where Rct , Cint , Rs , Cs stand for charge transfer resistance at the fluid/electrode interface, interfacial capacitance, fluid
∗
These authors have contributed equally to this work. Additional corresponding author: Dr. Jie Wu, E-mail: [email protected]
∗∗
Colour Online: See the article online to view Figs. 1–3 in colour.
www.electrophoresis-journal.com
472
S. Li et al.
Electrophoresis 2015, 36, 471–474
frequency; subscript p and m denote particle and medium, respectively. a and are particle diameter and fluid viscosity, respectively. For nanoscale macromolecules, DEP may be effective only when molecules are located within a very short distance to the electrodes. So DEP enrichment of macromolecules would benefit from fluid movements that carry molecules to the electrodes for binding. AC electric field can generate microflows through one or both of two ACEK flow mechanisms known as ACEO [9] and ACET [10] effect. ACEO typically dominates at low ionic strength. ACEO flow velocity has been observed to decrease significantly with increasing fluid conductivity and eventually drop to zero above 0.085 S/m, unsuitable for most bioassays [5]. This work uses 15 mM Na+ PBS with 0.18 S/m, so ACEO flow will be negligible. In comparison, ACET effect could generate microflows as [10, 11]: Figure 1. (A) Conceptual diagram of ACEK enhanced capacitive biosensing and interfacial capacitance change by the binding process. Dimensions are not drawn to scale. (B) Equivalent circuit of the ‘‘electrode-fluid’’ system. Cint : interfacial capacitance, Rct : charge transfer resistance, Rs : electrolyte resistance, Cs : electrolyte capacitance. The dashed line Cs is regarded as open circuit under our frequency.
resistance, and fluid capacitance, respectively [6]. When macromolecules are adsorped onto the electrode surface and add to the thickness of the surface layer, the interfacial capacitance change is expressed as ⌬ C/Cint,0 = −dana /[(drec + dana ) + edl ε p /ε s ], where Cint, 0 is the initial interfacial capacitance, dana , drec , and edl are the thickness of analyte layer, receptor layer, and electric double layer, which naturally forms at the interface between solid and electrolyte. ε p , εs are the dielectric constants of macromolecules and sample fluids. Based on impedimetric characterization of the sensing electrodes, the electrode cell is adequately represented by a serial connection of Rs and Cint from 10 kHz and 1 MHz [6], shown as the solid lines in Fig. 1B. Coincidentally, enhancement of binding reaction is found to be the most obvious around 100 kHz [7]. Therefore, our ACEK capacitive sensing is conducted at 100 kHz, and the changes in Cint is conveniently monitored by direct measurement of the sensor capacitance. Here besides capacitive sensing, AC signal has another function for concurrent analyte enrichment. When an inhomogeneous AC electric field is applied through microelectrodes to an aqueous solution, particle and fluid movement could be induced around the microelectrodes to achieve in situ concentration of macromolecules. Direct particle movement can be caused by DEP, and particle can also be carried by microflows such as alternating current electroosmosis (ACEO) or AC electrothermal (ACET) flows to reach the microelectrodes. DEP velocity of a spherical particle is expressed as [8]: ∗ ε ∗p − εm a 2 εm uDEP = (1) ∇|E |2 , Re ∗ ∗ 6 ε p + 2εm where ε ∗ = ε − j / is the complex permittivity with ε, , and being permittivity, conductivity, and angular C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
uACET = f et l 2 / ∝ |E |2 ∇T ∝ |E |4 ,
(2)
in which l is the characteristic length of the device, typically on the order of electrode spacing, fet is the body force. Combining DEP and ACET effect, nanoscale particles is concentrated over a large distance to realize accelerated detection. The magnitudes and ranges of DEP and ACET velocities depend on the electrode geometries. Previously, we reported a capacitive immunosensor using microelectrodes modified from off-the-shelf surface acoustic wave (SAW) resonator chips [12]. This work studies a different electrode design for capacitive immunoassay to better exploit ACEK effects for sensitivity and acceleration in immunoassay. Microelectrode array used here is planar thin film electrodes on silicon wafer with asymmetric widths of 5 and 25 m separated by 5 and 25 m gaps, hence denoted as 5/5/25/25 electrodes. This design is able to induce ACET microflows for biofluid convection effectively [10,11,13–15], and yet has adequate electrode surface for interfacial capacitance sensing. The electrodes were functionalized with antigen extracted from Mycobacterium avium subsp. paratuberculosis, a causative agent of Johne’s disease (JD). The quality of prepared electrodes was also checked by measuring the interfacial capacitance, as described in [16]. A microchamber for loading the serum sample (5 L) was formed by sealing a PDMS microchannel (L × W × H = 1 mm × 300 m × 40 m) over the electrodes, as shown in Fig. 2B. The sample fluid quickly reaches static equilibrium in the microchannel. The microchannel is sufficiently high since electric fields and ACEK effects are mostly limited within 25 m from the electrodes, shown by numerical simulation [5]. Then its capacitance was measured at 100 kHz by an impedance analyzer (Agilent 4294A). The whole detection process finished at the end of capacitance measurement, usually within 2 min, much simpler and faster than multiplewash/incubation, hour-long operation of ELISA. ACEK capacitive sensing method uses dimensionless change in the measured capacitance (capacitance change divided by the initial capacitance at time zero) as the metric to indicate antibody deposition on the electrode, which is the slope of www.electrophoresis-journal.com
Electrophoresis 2015, 36, 471–474
Microfluidics and Miniaturization
473
Figure 2. Experimental setup and proof of concept results. (A) Schematic of experimental set up for ACEK capacitive immunoassay. Interfacial capacitance of microelectrode sensor is read by impedance analyzer connected to a laptop via GPIB interface. (B) Schematic side-view (top) and 3D view (bottom) of the microelectrode sensing chamber. (C) Change of dimensionless interfacial capacitance with time for disease positive/negative serum, and control buffer.
dimensionless capacitance versus time and found by least square linear fitting method. Each test was repeated for five times and the CV was calculated. More details are described in [12, 17]. Figure 2C demonstrates how the dimensionless interfacial capacitance typically behaves with time, for a JD positive/negative serum sample and a control sample (15 mM PBS). Measurements were performed using a 100 kHz AC signal of 500 mV. Here, the positive serum exhibited a capacitance change rate of −81‰/min, while the change rates from the negative serum and control sample were less than −5‰/min. To demonstrate the effect of AC electric field on immunosensing, JD positive samples were tested with AC potentials from 50 mV to 1 V for 2 min. The corresponding sensor responses were plotted in Fig. 3A, with electric field strength estimated as AC potential/electrode gap. As can be seen, the capacitance change rate increases rapidly with increasing electric field strength, while the control sample exhibited consistently low change rate. The JD positive sample showed significantly higher change rates than the control samples except for the results at 10 mV/m (i.e. 50 mV), which is too low to induce any ACEK effects to enhance specific binding. Therefore, if a low voltage (i.e. 50 mV or lower) is used as in conventional impedance sensing, no detection result can be reached within 2 min due to too small a difference between disease-positive and control sample. For comparison, also plotted in Fig. 3A are the capacitance change rates from applying AC voltages of 5 to 200 mV on symmetric SAW electrodes (gap/width = 1.1 m/1.4 m). The capacitance change rates are very similar between the two electrodes when electric field strength is below 60 mV/m. At higher electric field strength, the response from 5/5/25/25
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
electrodes continues to rise, while there is no increase from the SAW electrodes after 100 mV/m. At 100 mV/m, the capacitance change rates from the 5/5/25/25 electrodes and SAW electrodes are (−77.75 ± 8.30)‰/min with a CV of 10.675%, and (−48.32 ± 7.81)‰/min with a CV of 16.16%, respectively. Figure 3B shows a capacitance change rate of 5.09‰ ± 0.38‰ for 1 ng/mL sample, which is more than 3 standard deviations from that of control sample, −0.95‰ ± 0.90‰. Therefore, the LOD was determined to be 1 ng/mL, a noticeable improvement from the SAW electrodes’ LOD of 10 ng/mL [12, 17], also at 500 mV. To clarify the importance of DEP and ACET effects for binding reaction, numerical simulation using COMSOL Multiphysics was performed for these two different electrode designs (see Supporting Information), with procedures similar to the description in [18, 19]. The simulation also reveals that the difference in sensor response is attributed to the electric field distribution induced by different electrode design. For SAW electrodes, the electric fields are concentrated around the electrode corners and between the electrodes, which is caused by an electrode thickness of 150 nm and a separation of 1.1 m. In comparison, 5/5/25/25 electrodes have a thickness of 100 nm and a separation of 5 m. Therefore, with 5/5/25/25 electrodes electric field reaches further into the fluids and produces much larger ACET vortices than SAW electrodes. Based on Eqs. (1) and (2), DEP velocity uDEP ∝ |E |2 and ACET velocity uACET ∝ |E |4 , so ACET convection becomes more important at higher electric field. Thus 5/5/25/25 and SAW electrodes exhibit very similar responses at lower field (⬍60 mV/m), and noticeably higher response for 5/5/25/25 electrode at higher field. To further our understanding of ACET capacitive sensing, numerical simulation was also performed using commercial finite element software Comsol Multiphysics 3.5a,
www.electrophoresis-journal.com
474
S. Li et al.
Electrophoresis 2015, 36, 471–474
Figure 3. (A) Dimensionless capacitance change rate as a function of electric field strength for JD-positive serum and control samples. Data from two types of electrodes, 5/5/25/25 and SAW electrodes, were compared; (B) Dimensionless capacitance change rate for concentration from 0.2 to 10 ng/mL. IgG antibody binding to immobilized IgG whole molecules were measured with 500 mV, 100 kHz AC signal for 2 min.
the model settings and simulation results are provided in the Supporting Information. To sum up, ACEK capacitive sensing method can provide rapid, sensitive, and easy-to-operate detection of specific proteins in biological relevant samples. The electrode design presented here demonstrated an improvement in detection sensitivity over our prior report [12]. We acknowledge the financial support of the U.S. National Science Foundation (ECS-0448896), China Natural Science Foundation (No. 11372093, 51305106), the University of Tennessee Research Foundation Maturation Fund, M-CERV seed grant and AgResearch Innovation Grant. Microfabrication was conducted at the Center for Nanophase Materials Sciences, which is sponsored at Oak Ridge National Laboratory by the Scientific User Facilities Division, Office of Basic Energy Sciences, U. S. Department of Energy. The authors have declared no conflict of interest.
References [1] Zhao, W., Zhang, L., Jing, W., Liu, S., Tachibana, H., Cheng, X., Sui, G., Biomicrofluidics 2013, 7, 011101. [2] Hamada, R., Takayama, H., Shonishi, Y., Mao, L., Nakano, M., Suehiro, J., Sens. Actuators B: Chem. 2013, 181, 439–445. [3] Cheng, I.-F., Yang, H.-L., Chung, C.-C., Chang, H.-C., Analyst 2013, 138, 4656–4662. [4] Islam, N., Lian, M., Wu, J., Microfluid Nanofluidics 2007, 3, 369–375.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[5] Wu, J., Ben, Y., Battigelli, D., Chang, H.-C., Ind. Eng. Chem. Res. 2005, 44, 2815–2822. [6] Li, S., Yuan, Q., Morshed, B. I., Ke, C., Wu, J., Jiang, H., Biosens. Bioelectron. 2013, 41, 649–655. [7] Liu, X., Yang, K., Wadhwa, A., Eda, S., Li, S., Wu, J., Sens. Actuators A: Phys. 2011, 171, 406–413. [8] Castellanos, A., Ramos, A., Gonzalez, A., Green, N. G., Morgan, H., J. Phys. D Appl. Phys. 2003, 36, 2584. [9] Ramos, A., Morgan, H., Green, N., Castellanos, A., J. Phys. D Appl. Phys. 1998, 31, 2338. ´ [10] Green, N. G., Ramos, A., Gonzalez, A., Castellanos, A., Morgan, H., J. Electrostat. 2001, 53, 71–87. [11] Wu, J., Lian, M., Yang, K., Appl. Phys. Lett. 2007, 90, 234103. [12] Li, S., Cui, H., Yuan, Q., Wu, J., Wadhwa, A., Eda, S., Jiang, H., Biosens. Bioelectron 2014, 51, 437–443. [13] Yuan, Q., Wu, J., Biomed. Microdevices 2013, 15, 125–133. [14] Yuan, Q., Yang, K., Wu, J., Microfluidics Nanofluidics 2014, 16, 167–178. [15] Yang, K., Wu, J., Biomicrofluidics 2010, 4, 034106. [16] Cui, H., Cheng, C., Wu, J., Eda, S., Sensors, 2013 IEEE, IEEE 2013, pp. 1–4. [17] Cui, H., Li, S., Yuan, Q., Wadhwa, A., Eda, S., Chambers, M., Ashford, R., Jiang, H., Wu, J., Analyst 2013, 138, 7188–7196. [18] Li, S., Ren, Y., Jiang, H., RSC Adv. 2014, 4, 9064–9071. [19] Hu, G., Gao, Y., Li, D., Biosens. Bioelectron. 2007, 22, 1403–1409.
www.electrophoresis-journal.com
