Sensitive and Simultaneous Detection of Cardiac Markers in Human

Oct 12, 2011 - Bio Lab, Samsung Advanced Institute of Technology, Samsung Electronics Co., Ltd., San #14-1, Nongseo-dong, Giheung-gu, Yongin-si, Gyeon...
3 downloads 8 Views 2MB Size
Download PDF
ARTICLE pubs.acs.org/ac

Sensitive and Simultaneous Detection of Cardiac Markers in Human Serum Using Surface Acoustic Wave Immunosensor Joonhyung Lee,† Youn-Suk Choi,† Yeolho Lee,† Hun Joo Lee,†,‡ Jung Nam Lee,† Sang Kyu Kim,† Kyung Yeon Han,† Eun Chol Cho,† Jae Chan Park,† and Soo Suk Lee*,† †

Bio Lab, Samsung Advanced Institute of Technology, Samsung Electronics Co., Ltd., San #14-1, Nongseo-dong, Giheung-gu, Yongin-si, Gyeonggi-do, Republic of Korea ‡ Interdisciplinary Program of Integrated Biotechnology, Sogang University, Shinsu-dong, Mapo-gu, Seoul, Republic of Korea ABSTRACT: We present a rapid and sensitive surface acoustic wave (SAW) immunosensor that utilizes gold staining as a signal enhancement method. A sandwich immunoassay was performed on sensing area of the SAW sensor, which could specifically capture and detect cardiac markers (cardiac troponin I (cTnI), creatine kinase (CK)-MB, and myoglobin). The analytes in human serum were captured on gold nanoparticles (AuNPs) that were conjugated in advance with detection antibodies. Introduction of these complexes to the capture antibody-immobilized sensor surface resulted in a classic AuNP-based sandwich immunoassay format that has been used for signal amplification. In order to achieve further signal enhancement, a gold staining method was performed, which demonstrated that it is possible to obtain gold staining-mediated signal augmentation on a masssensitive device. The sensor response due to gold staining varied as a function of cardiac marker concentration. We also investigated effects of increasing operating frequency on sensor responses. Results showed that detection limit of the SAW sensor could be further improved by increasing the operating frequency.

D

etection of proteins in a sensitive and rapid manner is essential in clinical applications.1 3 Today, most of the common biosensing platforms require the use of labels, such as fluorophores and radiolabels. However, these techniques require long sample preparation time and additional cost due to the labeling process.4 6 Label-free techniques, such as surface plasmon resonance (SPR),7,8 quartz crystal microbalance (QCM),9,10 and surface acoustic wave (SAW),11,12 have been developed to alleviate this concern. SAW sensor has been widely investigated and used to detect a variety of target analytes due to its high sensitivity, low cost, and reliability.11,12 Among surface acoustic wave biochemical platforms, shear-horizontal (SH) SAW sensor has been demonstrated to be effective in recognizing binding events of various biomolecules in liquid phase, which operates on the basis of changes in mass, temperature, viscosity, and pH.13,14 Unlike Rayleigh surface acoustic wave sensors that possess a displacement component perpendicular to the substrate, shear-horizontal waves can propagate in liquid without coupling excess acoustic energy, avoiding severe attenuation and high insertion loss in aqueous environment.15,16 In particular, guided SH-SAW sensor (also known as Love wave sensor) that consists of an SH-SAW substrate with an overlayer having a lower shear wave velocity has been shown to be one of the most promising platforms for biosensor applications due to its high sensitivity.17 19 The role of the overlayer is to trap the acoustic energy near the sensing surface, thus yielding high sensitivity to any physical perturbation on the surface, such as changes in mass density, mechanical stiffness, viscosity, pressure, or temperature. Furthermore, the overlayer can also protect interdigital transducer (IDT) electrodes from the liquid environment. r 2011 American Chemical Society

Various dielectric materials, such as silicon dioxide (SiO2),20,21 zinc oxide (ZnO),22 and polymers,23 can be used as waveguide material. In many studies, silicon dioxide (SiO2) has been widely used as the guiding layer due to its low acoustic loss, high mechanical and chemical resistance, and ease of functionalization with biomolecules.11 However, despite its high sensitivity, analytes in small quantities are still difficult to detect with label-free technique alone, pointing to a need for robust signal enhancement schemes. The metal staining method has been proven to enhance signals for detecting various analytes due to its superb sensitivity.24,25 Mass loading caused by gold nanoparticles (AuNPs) and subsequent catalyzed deposition of gold or silver leads to signal enhancement. Although other signal amplification methods including immuno-PCR and rolling circle amplification also enable highly sensitive detection of various target molecules, metal staining has several advantages over these methods. Metal staining does not require expensive, complicated, and multistep protocols, which makes it adaptable to low-cost and robust biosensing. In addition, it has been shown that gold staining provides higher sensitivity than silver staining.24 In this study, gold staining was used as a robust signal amplification technique to detect small amounts of target analytes. Cardiac markers play an essential role in the diagnosis, prognosis, monitoring, and risk stratification of suspected heart attack patients.26 30 For example, cardiac markers have been used as an Received: August 9, 2011 Accepted: October 12, 2011 Published: October 12, 2011 8629

dx.doi.org/10.1021/ac2020849 | Anal. Chem. 2011, 83, 8629–8635

Analytical Chemistry

ARTICLE

Figure 1. (a) Top and cross-sectional views of the single-packaged Love wave SAW sensor. (b) Top view of the Love wave SAW sensor array.

important factor in ruling in or ruling out acute myocardial infarction (AMI) that is the world’s leading cause of morbidity and mortality. Among currently used cardiac markers, myoglobin, which appears 1 to 2 h after symptom onset, has been proven to be valuable with high sensitivity within the first few hours after presentation. However, myoglobin has low specificity for cardiac necrosis in patients due to its abundant presence in myocardial as well as skeletal muscle cells. Creatine kinase (CK)-MB appears 5 6 h after the onset of chest pain and remains abnormally elevated for 12 24 h after the onset of AMI. It has low specificity due to its presence in skeletal muscle. However, its measurement is still effective in evaluation of AMI in multimarker assays and useful in monitoring reinfarction. Detection of cardiac troponins (cardiac troponin I (cTnI), cardiac troponin T (cTnT)) has been regarded as current biochemical “gold standard” for the diagnosis of AMI due to their excellent specificity and sensitivity. Since it takes approximately 4 h for cardiac troponins to reach detectable concentrations in blood, they cannot be considered early markers. However, they remain elevated for 4 10 days after the onset of AMI, indicating their capability to remain elevated for a reasonable length of time to allow a suitable diagnostic window. Given that these cardiac markers have different characteristics, including clinical sensitivity and specificity, release time after symptom onset, clinical cutoff level (myoglobin 70 200 ng/mL; CK-MB 3.5 10 ng/mL; cTnI 0.06 1.5 ng/mL),31,32 and capability to remain elevated for a reasonable length of time, a rapid, accurate, and simultaneous measurement of these cardiac markers is important in reducing detection time, decreasing cost of patient treatment, and saving patient lives. This study describes a highly sensitive 200 MHz Love wave SAW biosensor that can achieve individual and simultaneous detection of multiple cardiac markers (myoglobin, CK-MB, and cTnI) in human serum below clinical cutoff levels. An immunosandwich assay in combination with gold staining was used as a basic detection methodology. Capture antibodies were adsorbed nonspecifically on the sensing area. Cardiac markers in human serum were captured on gold nanoparticles (AuNPs) conjugated with detection antibodies. Introduction of these cardiac markerAuNP complexes to the capture antibody-immobilized surface resulted in an immuno-sandwich assay, in which subsequent catalyzed deposition of gold onto AuNP led to signal enhancement. In this fashion, we significantly increased sensitivity of the assay by reducing minimum detectable concentrations of the cardiac markers below their clinical cutoff levels. Next, in order to investigate the possibility to lower the limit of detection, this

detection strategy was applied to a SAW sensor with a higher operating frequency (400 MHz). Slope of frequency response vs cTnI concentration increased by a factor of 2 compared to the 200 MHz sensor, indicating that increasing the operating frequency can lead to more sensitive detection of cardiac markers.

’ EXPERIMENTAL SECTION Materials. Cardiac troponin I (cTnI) was purchased from Fitzgerald (Acton, MA). Two cTnI monoclonal antibodies were obtained from Hytest (Turku, Finland). CK-MB, CK-MB monoclonal antibody, and CK-MB polyclonal antibody were purchased from Fitzgerald (Acton, MA). Myoglobin and two different types of myoglobin monoclonal antibodies were also acquired from Fitzgerald (Acton, MA). Thirty nm diameter AuNPs were obtained from BBI (Cardiff, UK). Bovine serum albumin (BSA) and trimethoxy(octadecyl) silane were purchased from Sigma-Aldrich (St. Louis, MO). Gold(III) chloride trihydrate and hydroxylamine hydrochloride were purchased from Sigma-Aldrich. PBS buffer was obtained from Invitrogen (Carlsbad, CA). Human multisera normal (Human serum) was acquired from Bio Clinical System (Anyang, Republic of Korea). Phosphate buffer (PB) consisted of sodium phosphate monobasic and sodium phosphate dibasic that were purchased from Sigma-Aldrich. Design and Fabrication of the SAW Sensor. Aluminum interdigital transducer (IDT; 3000 Å thick) electrodes deposited by sputtering were patterned on a 36°YX-LiTaO3 substrate (Yamaju Ceramics Co. Ltd., Anada-Cho Seto City, Japan) using a conventional photolithography as shown in Figure 1. The input and output IDT electrodes consisted of 72 finger pairs with an electrode width of 5.0 μm to obtain 200 MHz center frequency or 36 finger pairs with an electrode width of 2.5 μm to achieve 400 MHz center frequency. A SiO2 guiding layer (5.2 and 2.6 μm thick for 200 and 400 MHz sensor, respectively) was deposited on the IDT patterned substrate by plasma-enhanced chemical vapor deposition (P-500; Applied Material, Inc., Santa Clara, CA), followed by wet etching with buffered oxide etchant to open the contact pads for electric connection. After dicing the fabricated wafer, the SAW sensors were mounted on ceramic surface mount device (SMD) packages (Figure 1a) having one sensing spot or on printed circuit board (PCB) having four sensing spots (Figure 1b) to form a single SAW sensor or a foursensor array, respectively, followed by aluminum wire bonding for electric connection. 8630

dx.doi.org/10.1021/ac2020849 |Anal. Chem. 2011, 83, 8629–8635

Analytical Chemistry

Figure 2. Block diagram of the measuring system for the SAW sensor.

Figure 3. Diagram of the fluidic cell.

Measurement System. The four-SAW sensor array system consisted of a laboratory-built oscillator, a frequency counter, and two multiplexers with channel controller as shown in Figure 2. The SAW sensors are positioned as the frequency determining element of the oscillator, and the additional use of the multiplexers allows a sequential switching of the four SAW sensors by operating with the channel controller. The surface acoustic waves, which were passed through the delay lines, were transmitted to the oscillator and fed back to the sensors again. The frequency counter measured the frequency of SAW signals using field-programmable gate array (FPGA). The flow cell was constructed as shown in Figure 3 with a peristaltic pump (ISM597; ISMATEC, Glattbrugg, Switzerland) and a custom-made fluidic block and silicon gasket. In the array sensor, each spot is separated from the other spots with a silicon gasket, which allows for separate reaction chambers. Also, each spot is connected to the peristaltic pump via a tube. To exclude effects of temperature change on SAW sensor signal, temperature controller was installed below the chamber to keep the temperature at 25 °C. Antibody Conjugation to the Gold Nanoparticles. Each antibody (10 μL of 1 mg/mL) was added to 1 mL of 30 nm colloidal AuNPs, followed by incubation at room temperature for 30 min. To block unreacted sites on the AuNP surface, 0.1 mL of 1% BSA in deionized water was introduced to the AuNPantibody mixture. After additional incubation at room temperature for 30 min, the mixture was centrifuged at 13 00 rpm for 10 min at 4 °C. Supernatant was removed, and then, the antibodyconjugated AuNPs were dispersed in 0.5 mL of PBS solution

ARTICLE

(0.1 mM Na2HPO4, 1.8 mM KH2PO4, 137 mM NaCl, pH 7.4) containing 0.1% BSA. Sandwich Immunoassay with Gold Staining. Silicon dioxide (SiO2) sensing area was sequentially rinsed with deionized water and absolute ethanol and then dried under a nitrogen gas. The chip was next placed in a UV/ozone chamber (144AX-220; Jelight Company Inc., Irvine, CA, USA) for 10 min, rinsed with deionized water, and again dried under a nitrogen gas. The chip was incubated in a solution of 5% (v/v) fresh trimethoxy(octadecyl) silane in toluene with 0.5% n-butylamine as a catalyst for 1 h, followed by rinsing with toluene for 2 min and drying under a nitrogen gas. The silanized sensor was then baked in an oven at 110 °C for 1 h, followed by rinsing with toluene for 2 min and drying under a nitrogen gas. For the SAW single sensor assay, antibodies (100 μg/mL in phosphate buffer (PB), 7.4 pH) to each cardiac marker (cTnI, CK-MB, or myoglobin) were immobilized nonspecifically on the sensing area of the guiding layer for 1 h. For the SAW sensor array, antibodies to cTnI, CK-MB, and myoglobin were immobilized in separate spots. In order to acquire a larger amount of data for the detection of cTnI, which requires higher sensitivity for clinical applications, 2 spots were functionalized with antibodies to cTnI. Afterward, unreacted sites on the sensing area were blocked with 5% BSA in PB solution. Human serum (50 μL) containing a specific type of cardiac marker was mixed with 50 μL of its specific antibody-AuNP conjugate for 3 min. Considering binding affinities and required cutoff values of these cardiac markers with their specific detection antibodies, for the SAW sensor array assay, cTnI antibody-AuNP, CK-MB antibody-AuNP, and myoglobin antibody-AuNP conjugates were mixed in a 6:3:1 ratio, respectively, prior to use. Then, 80 μL of the conjugate mixtures was added to 80 μL of human serum containing cTnI, CK-MB, and myoglobin, which were allowed to bind to their specific antibody-AuNP conjugates for 3 min. Then, these complexes were introduced to the sensor surface and allowed to bind to the immobilized antibodies for 3 min. After washing with PBS solution for 1 min, gold staining solution that consisted of 50 μL of gold(III) chloride trihydrate (10 mM) and 50 μL of hydroxylamine hydrochloride (20 mM) was incubated with AuNPs on the sensor surface for 2 min, which resulted in catalyzed deposition of gold onto the AuNPs captured on the sensing surface. Then, the sensing surface was rinsed again with PBS solution for 1 min. All experiments were performed in human serum that was spiked with the cardiac markers. In order to figure out original concentrations of the analytes, human serum was analyzed five times with ADVIA Centaur (SIEMENS, Erlangen, Germany), IMMULITE 2000 Immunoassay System (SIEMENS), and Modular Analytics E 170 Module (Roche, Basel, Switzerland). Concentrations of cTnI, CK-MB, and myoblogin in human serum were 14 pg/mL, 1.1 ng/mL, and 16.0 ng/mL, respectively.

’ RESULTS AND DISCUSSION Detection of Cardiac Markers via Gold Staining. Figure 4 highlights the sandwich immunoassay procedure used in this study, including immobilization of capture antibodies, complex formation between antibody-AuNP conjugates and cardiac markers in human serum, capture of these complexes to the immobilized antibodies, and subsequent gold staining-based signal enhancement strategy. The capture antibodies were immobilized nonspecifically on the SiO2 guiding layer. A major concern inherent 8631

dx.doi.org/10.1021/ac2020849 |Anal. Chem. 2011, 83, 8629–8635

Analytical Chemistry

ARTICLE

Figure 4. Schematic of the sandwich immunoassay format utilized in this study in combination with gold staining.

in protein detection assays is potential background interference from other biomolecules and chemical species. Provided that target cardiac markers were present in human serum, it was important to assess impact other human serum proteins would have in our detection strategy. Accordingly, all experiments were performed with human serum that was spiked with cardiac markers (TnI, CK-MB, and myoglobin). The cardiac markers present in human serum were captured on antibody-AuNP conjugates, which bound to the immobilized capture antibodies in a classic sandwich immunoassay format. Then, introduction of gold staining solution that consisted of gold(III) chloride trihydrate and hydroxylamine hydrochloride resulted in catalyzed deposition of gold onto the captured AuNPs on the sensor surface. Figure 5 shows sensor response to introduction of the cardiac markers (10 ng/mL of cTnI, 5.5 ng/mL of CK-MB, and 13 ng/ mL of myoglobin) in human serum captured with antibodyAuNP conjugates and subsequent addition of gold staining mixture. As expected, addition of the cardiac marker-AuNP complexes in human serum to the antibody-immobilized surface resulted in decrease in frequency due to specific adsorption of the complexes as well as nonspecific adsorption of human serum proteins to the functionalized surface. Addition of PBS buffer caused a minor response. This frequency change was indicative of loosely adsorbed complexes and human serum proteins on the functionalized surface, which were readily removed by washing with PBS solution. Introduction of low pH gold staining solution led to frequency increase due to a sudden decrease in pH, as reported in previous studies.33 However, the frequency decreased over time as catalyzed deposition of gold on the captured AuNPs resulted in significant mass increase over the sensor surface. Hence, we demonstrate potential of the SAW sensor to perform rapid detection (∼12 min) of the cardiac markers. However, it is necessary to investigate the sensor response to variation in cardiac marker concentrations. Dependence of Sensor Response on Cardiac Markers Concentration in the Single Sensor SAW System. We investigated dependence of sensor response due to gold staining on applied cardiac marker concentration in a single sensor chip system. Measurements were performed five times for each concentration of cardiac markers, and the results are displayed in Figure 6. As concentration of cTnI, CK-MB, and myoglobin increased from 20 pg/mL to 100 ng/mL, from 1.1 ng/mL to 100.5 ng/mL, and from 16.0 ng/mL to 1016 ng/mL, respectively, the sensor response due to gold staining also increased, indicating that amount of the adsorbed cardiac marker-AuNP complexes was proportional to the applied cardiac marker concentration. On the other hand, the sensor response due to the addition of the cardiac marker-AuNP complexes to the antibody-immobilized surface was not dependent on the applied cardiac marker

Figure 5. Sensor response due to introduction of the cardiac markers in human serum captured with antibody-AuNP conjugates and subsequent gold staining reagents. Frequency decrease represents an increase in effective mass of the single sensor chip. (a) cTnI: 10 ng/mL. (b) CKMB: 5.5 ng/mL. (c) myoglobin: 26 ng/mL.

concentration (data not shown). Surface acoustic wave (SAW) sensor is a mass-sensitive sensor that measures the property that scales proportionally to mass associated with or bound to its sensitivity. When the cardiac markers in human serum, which were captured on antibody-AuNP conjugates, were introduced on the capture antibody-immobilized surface, the amount of the adsorbed complexes (cardiac markers-AuNP conjugates) onto 8632

dx.doi.org/10.1021/ac2020849 |Anal. Chem. 2011, 83, 8629–8635

Analytical Chemistry

ARTICLE

Figure 6. Variation of the frequency response due to gold staining with the cardiac marker concentration in the single sensor chip system. Measurements were performed five times for each concentration of cardiac markers. (a) cTnI: 20 pg/mL 100 ng/mL. (b) CK-MB: 1.1 100.5 ng/mL. (c) myoglobin: 16 1016 ng/mL.

the capture antibody led to the frequency shift. However, the amount of adsorbed macromolecules such as albumin, antibodies, and other globulins in human serum onto the surface also caused the frequency shift as a background signal, which was not negligible. Hence, although the amount of the adsorbed complexes onto the capture antibody depended on the cardiac marker concentration, the sensor response was not dependent on the cardiac marker concentration. Next, minimum detectable concentrations of cTnI, CK-MB, and myoglobin were 20 pg/mL, 1.1 ng/mL, and 16.0 ng/mL, respectively, which were below the clinical cutoff levels (myoglobin 70 200 ng/mL; CK-MB 3.5 10 ng/mL; cTnI 0.06 1.5 ng/mL). The minimum detectable concentrations were determined that can clearly be distinguished from the samples with the lowest concentrations (myoglobin 16.0 ng/mL; CK-MB 1.1 ng/mL ng/mL; cTnI 14 pg/mL) present in human serum in terms of frequency shifts. A recent study that investigated human cTnI detection using a label free immunoassay with a fiber optic-based surface plasmon resonance (SPR) sensor demonstrated detection sensitivity of 1.4 nM.34

Figure 7. Variation of the frequency response due to gold staining with the cardiac marker concentration ((a) cTnI: 50 pg 100 ng. (b) CKMB: 1.1 100.6 ng. (c) myoglobin: 16 1016 ng) in the array sensor chip system. Measurements were performed five times for each concentration of cardiac markers.

Another study that used an SPR sensor in combination with a sandwich immunoassay reported detection of human cTnI with a concentration range of 0.5 to 20 nM.35 Due to the gold stainingmediated signal enhancement, detection limit of our SAW sensor system was better by 1 to 2 orders of magnitude. Dependence of Sensor Response on Cardiac Markers Concentration in the Array SAW Sensor System. We also analyzed effects of applied cardiac marker concentration on the sensor response due to gold staining in an array system. Human serum was spiked with the three cardiac markers, enabling simultaneous detection of cTnI, CK-MB, and myoglobin. Measurements 8633

dx.doi.org/10.1021/ac2020849 |Anal. Chem. 2011, 83, 8629–8635

Analytical Chemistry

ARTICLE

sensor chip. As shown in Figure 8, the 400 MHz sensor chip, compared with the 200 MHz sensor chip, produced approximately 2 times higher slope of frequency response versus cTnI concentration in the range of 0.02 50 ng/mL, which was consistent with the calculated results. These results highlight the potential that the detection limit of the SAW sensor can be further improved by increasing the operating frequency with accompanying noise level and stability taken into account.

Figure 8. Effects of variation of operating frequency (∼400 MHz) on the sensor response (higher operating frequency results in higher mass sensitivity).

were carried out five times for each concentration of the three cardiac markers. Standard deviation values were similar to those of the single sensor chip system. As shown in Figure 7, however, average frequency shifts in the array sensor chip system were lower than those in the single sensor chip system. In addition, the minimum detectable concentration of cTnI was 50 pg/mL, which was higher than that from the single sensor chip system. The three cardiac markers in human serum sample competed for specific binding to their corresponding detection antibodies immobilized on the AuNPs as well as the capture antibodies immobilized on the sensor surface, leading to lower amount of the AuNPs on the surface and, therefore, lower frequency shifts from gold staining. Also, as expected, increasing the concentration of cTnI, CK-MB, and myoglobin resulted in higher sensing response due to gold staining, indicating that the amount of cardiac marker-AuNP varied as a function of the applied cardiac marker concentration. The minimum detectable concentrations of cTnI, CK-MB, and myoglobin were 50 pg/mL, 1.1 ng/mL, and 16.0 ng/mL, respectively, which were below clinical cutoff levels. Hence, we demonstrate potential of the SAW biosensor to perform sensitive and simultaneous detection of multiple cardiac markers. Detection of cTnI Using 400 MHz Single SAW Sensor System. In order to improve the detection limit, we investigated effects of increasing operating frequency on the sensor response. It has been demonstrated that higher operating frequency in acoustic wave sensor depends on mass sensitivity.36 Also, in a Love wave type SAW sensor, the thickness of the guiding layer has a strong influence on the mass sensitivity and an optimal guiding layer thickness exists, at which maximum mass sensitivity is obtained for a given wavelength.11 In our previous study, the mass sensitivity of the SAW sensor at 200 MHz was calculated theoretically as a function of the SiO2 guiding layer thickness.11 In the same way, the maximum mass sensitivity of the SAW sensor at 400 MHz was calculated theoretically at the optimal guiding layer (∼2.6 μm). The mass sensitivity of 400 MHz sensor at the optimal guiding layer is about two times higher than was seen in 200 MHz sensor at the optimal guiding layer (∼5.2 μm). Furthermore, a SAW sensor with higher operating frequency offers significant advantages in terms of cost as well as ease of construction due to lower thickness of the guiding layer. A single sensor chip was designed to operate at a center frequency of 400 MHz, and dependence of the sensor response due to gold staining on the applied cTnI concentration was investigated. Afterward, results were compared to those of the 200 MHz single

’ CONCLUSIONS In conclusion, we demonstrated that a Love wave type SAW biosensor in combination with gold staining signal enhancement strategy detected cardiac markers in a sensitive, rapid, and simultaneous manner. A sandwich immunoassay format was utilized; cardiac markers were captured with antibody-AuNP conjugates, which were bound to immobilized antibodies on the sensor surface, and gold staining resulted in signal enhancement. We showed that the sensor response depended on the cardiac marker concentration. In order to improve the detection limit, we investigated effects of increasing operating frequency on the sensor response, and the SAW sensor with higher operating frequency demonstrated higher mass sensitivity. Our future efforts will be focused on extending this platform to detection of other disease makers present in body fluids, such as blood and plasma. Due to its size and high sensitivity, we expect this platform to be useful in development of devices for point-of-care diagnostics. ’ AUTHOR INFORMATION Corresponding Author

*Phone: +82-31-280-6947. Fax: +82-31-280-6816. E-mail: [email protected].

’ ACKNOWLEDGMENT This work was supported by Bio Laboratory at Samsung Advanced Institute of Technology. ’ REFERENCES (1) Wu, G. H.; Datar, R. H.; Hansen, K. M.; Thundat, T.; Cote, R. J.; Majumdar, A. Nat. Biotechnol. 2001, 19, 856–860. (2) Savran, C. A.; Knudsen, S. M.; Ellington, A. D.; Manalis, S. R. Anal. Chem. 2004, 76, 3194–3198. (3) Lee, J.; Icoz, K.; Roberts, A.; Ellington, A. D.; Savran, C. A. Anal. Chem. 2010, 82, 197–202. (4) Yang, L. T.; Fung, C. W.; Cho, E. J.; Ellington, A. D. Anal. Chem. 2007, 79, 3320–3329. (5) Acharya, G.; Chang, C. L.; Doorneweerd, D. D.; Vlashi, E.; Henne, W. A.; Hartmann, L. C.; Low, P. S.; Savran, C. A. J. Am. Chem. Soc. 2007, 129, 15824–15829. (6) Savran, C. A.; Knudsen, S. M.; Ellington, A. D.; Manalis, S. R. Anal. Chem. 2004, 76, 3194–3198. (7) Lyon, L. A.; Musick, M. D.; Natan, M. J. Anal. Chem. 1998, 70, 5117–5183. (8) Soelberg, S. D.; Stevens, R. C.; Limaye, A. P.; Furlong, C. E. Anal. Chem. 2009, 81, 2357–2363. (9) Henne, W. A.; Doorneweerd, D. D.; Lee, J.; Low, P. S.; Savran, C. A. Anal. Chem. 2006, 78, 4880–4884. (10) Knudsen, S. M.; Lee, J.; Ellington, A. D.; Savran, C. A. J. Am. Chem. Soc. 2006, 128, 15936–15937. (11) Lee, H. J.; Namkoong, K.; Cho, E. C.; Ko, C.; Park, J. C.; Lee, S. S. Biosens. Bioelectron. 2009, 24, 3120–3125. 8634

dx.doi.org/10.1021/ac2020849 |Anal. Chem. 2011, 83, 8629–8635

Analytical Chemistry

ARTICLE

(12) L€ange, K.; Rapp, B. E.; Rapp, M. Anal. Bioanal. Chem. 2008, 391, 1509–1519. (13) Du, J.; Harding, G. L.; Ogilvy, J. A.; Dencher, P. R.; Lake, M. Sens. Actuators, A: Phys. 1996, 56, 211–219. (14) Jakoby, B.; Vellekoop, M. J. Sens. Actuators, A: Phys. 1998, 68, 275–281. (15) Lee, D. L. IEEE Trans. Sonics Ultrason. 1981, 28, 330–341. (16) Chang, R. C.; Chi, S. Y.; Hong, C. S.; Chuang, Y. T. Thin Solid Films 2006, 498, 146–151. (17) Gizeli, E.; Goddard, N. J.; Lowe, C. R.; Stevenson, A. C. Sens. Actuators, B 1992, 6, 131–137. (18) Gizeli, E.; Stevenson, A. C.; Goddard, N. J.; Lowe, C. R. IEEE Trans. Ultrason. 1992, 39, 657–659. (19) Stevenson, A. C.; Gizeli, E.; Goddard, N. J.; Lowe, C. R. Sens. Actuators, B 1993, 13 14, 635–637. (20) Kovacs, G.; Venema, A. Appl. Phys. Lett. 1992, 61, 639–641. (21) Kovacs, G.; Vellekoop, M. J.; Haueis, R.; Lubking, G. W.; Venema, A. Sens. Actuators, A: Phys. 1994, 43, 38–43. (22) Fu, Y. Q.; Luo, J. K.; Du, X. Y.; Flewitt, A. J.; Li, Y.; Markx, G. H.; Walton, A. J.; Milne, W. I. Sens. Actuators, B 2010, 143, 606–619. (23) Bender, F.; Cernosek, R.; Josse, F. Electron. Lett. 2000, 36, 1672–1673. (24) Kim, D.; Daniel, W. L.; Mirkin, C. A. Anal. Chem. 2009, 81, 9183–9187. (25) Kaur, J.; Singh, K. V.; Boro, R.; Thampi, K. R.; Raje, M.; Varshney, G. C.; Suri, C. R. Environ. Sci. Technol. 2007, 41, 5028–5036. (26) Kavsak, P. A.; MacRae, A. R.; Yema, M.; Jeffe, A. S. Clin. Chem. 2009, 55, 573–577. (27) Mohammed, M.; Desmulliez, M. Y. Lab Chip 2011, 11, 569–595. (28) Degefa, T. H.; Kwak, J. Anal. Chim. Acta 2008, 613, 163–168. (29) Yang, Z.; Zhou, D. M. Clin. Biochem. 2006, 39, 771–780. (30) Melanson, S. F.; Morrow, D. A.; Jarolim, P. Am. J. Clin. Pathol. 2007, 128, 282–286. (31) Friess., U.; Stark, M. Anal. Bioanal. Chem. 2009, 393, 1453–1462. (32) Christenson, R. H.; Azzazy, H. E. Clin. Biochem. 2009, 42, 150–157. (33) Kondoh, J.; Matsui, Y.; Shiokawa, S. Jpn. J. Appl. Phys. 1993, 32, 2376–2379. (34) Masson, J. F.; Obando, L.; Beaudoin, S.; Booksh, K. Talanta 2004, 62, 865–870. (35) Wei, J.; Mu, Y.; Song, D.; Fang, X.; Liu, X.; Bu, L.; Zhang, H.; Zhang, G.; Ding, J.; Wang, W.; Jin, Q.; Luo, G. Anal. Biochem. 2003, 321, 209–216. (36) Wu, C. NNIN REU 2006 Research Accomplishments 3 Biological Applications 2006, 28–29.

8635

dx.doi.org/10.1021/ac2020849 |Anal. Chem. 2011, 83, 8629–8635