Development of a novel DNA chip based on a bipolar semiconductor microchip system

Biosensors and Bioelectronics 22 (2007) 1447–1453

Development of a novel DNA chip based on a bipolar semiconductor microchip system Joon Myong Song a,∗ , Min-Sung Yang b , Ho Taik Kwan b a

Research Institute of Pharmaceutical Sciences and College of Pharmacy, Seoul National University, Seoul 151-742, South Korea b 8F-8, Hightech Center, Gyeonggi Technopark, Celltek Co., Ltd., Ansan-si 426-170, South Korea Received 11 February 2006; received in revised form 19 June 2006; accepted 23 June 2006 Available online 4 August 2006

Abstract We have applied an integrated circuit photodiode array (PDA) chip system to a DNA chip. The PDA chip system, constructed using conventional bipolar semiconductor technology, acts as a solid transducer surface as well as a two-dimensional photodetector. DNA hybridization was performed directly on the PDA chip. The target DNA, the Bacillus subtilis sspE gene, was amplified by polymerase chain reaction (PCR). The 340-bp PCR product was labeled using digoxigenin (DIG). A silicon nitride layer on the photodiode was treated with poly-l-lysine to immobilize the DNA on the surface of the photodiode detection elements. Consequently, the surface of the photodiode detector became positively charged. An anti-DIGalkaline phosphatase conjugate was reacted with the hybridized DIG-labeled DNA. A color reaction was performed based on the enzymatic reaction between nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl-phosphate (NBT/BCIP) staining solution and a DNA complex containing antibodies. A blue precipitate was formed on the surfaces of the photodiode detection elements. Successful quantitative analysis of the hybridized PCR products was achieved from the light absorption properties of the blue enzymatic reaction product that was produced after a series of reaction processes. Our DNA chip system avoids the complicated optical alignments and light-collecting optical components that are usually required for an optical DNA chip device. As a result, a simple, compact, portable and low-cost DNA chip is achieved. This system has great potential as an alternative system to the conventional DNA reader. © 2006 Elsevier B.V. All rights reserved. Keywords: DNA chip; Bipolar semiconductor technology; On-chip bioassay; Enzymatic reaction

1. Introduction A miniaturized biochip system requires a bioreceptor with a dense patterned organization, on a solid transducer surface (DeLouise et al., 2005; Fouque et al., 2005; Kohn and Plaxco, 2005; Pazos et al., 2005; Storhoff et al., 2004). Two-dimensional imaging detectors, such as charge-coupled-device (CCD) or complementary metal oxide semiconductor (CMOS) detectors, are used to observe the optical transduction that occurs on the transducer surface. Although organized microarray systems allow a high-throughput parallel biosensing capability, operational complexity is unavoidable in the biochip. Therefore, the development of methods to address and trigger specific biomolecules in predesigned arrays is essential.

∗

Corresponding author. Tel.: +82 2 880 7841; fax: +82 2 871 2238. E-mail address: [email protected] (J.M. Song).

0956-5663/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bios.2006.06.026

Two cases to achieve optical correspondence between the microarray and the photodetector can be considered. In some systems, a beam of light is focused onto a microarray, and the microarray is scanned from spot to spot (Nakauchi et al., 2002). In the other case, each spot of the microarray is optically transduced and detected simultaneously, using a two-dimensional detector. In this case, the optical alignment must be such that a 1:1 optical correspondence occurs between each spot of the microarray and a pixel of the photodetector (Allain et al., 2001). Furthermore, the light source must be irradiated onto each spot of the microarray in a uniform manner. In both cases, the microarray is spatially separated from the photodetector, and the biochip must contain delicate optical components to achieve optical correspondence between the microarray and the photodetector (Peter et al., 2001; Reichert et al., 2000). To construct a miniaturized high-throughput high-speed biochip system, the photodetector must meet

1448

J.M. Song et al. / Biosensors and Bioelectronics 22 (2007) 1447–1453

certain requirements, such as miniaturization and multiple photosensors that correspond to each spot of the microarray (Culha et al., 2004; Stratis-Cullum et al., 2003). An integrated circuit (IC) photodetector microchip is an effective solution that provides the required miniaturization and multiple photosensors (Hwang et al., 2004; Gosch et al., 2004; Song et al., 2005). Photodetection involves photosensing, in which a photocurrent is produced in response to light. This photocurrent is amplified and then converted to a digitized voltage. Current semiconductor technology is well developed and permits the integration of all these components on a small IC microchip. In this work, we report a bipolar IC photodiode array (PDA) microchip detection system that can be used as a miniaturized DNA chip based on an on-chip bioassay. An on-chip bioassay usually involves many reactions and washing steps. To perform a DNA assay on the surface of a photodiode, the photodiode must be exposed directly to bulk solutions and bioreagents. Reaction conditions should be optimized to minimize any damage to the photodiode surface arising from its direct exposure to the bulk medium. Simultaneously, the efficiency of DNA detection should also be considered. Thus, even under optimized conditions, an on-chip bioassay may still cause a certain level of damage to the photodiode resulting from unavoidable interference from the bulk medium. This becomes a particularly important issue for a PDA chip that is used repeatedly. A reusable PDA chip is exposed to many types of biomolecules under various experimental conditions, e.g., at different pH. Accordingly, each photodiode in the PDA chip must have an operational capacity to maintain its activity during the on-chip bioassay. The bipolar microchip system that we have constructed can be used under different conditions. Compared with CMOS technology, bipolar semiconductor technology has the advantage of allowing high-performance analog signal processing of low noise levels. Many biochemical reagents in reaction solutions may damage the chip surface during the onchip bioassay. Under such conditions, the bipolar PDA chip is expected to be more resistant to the interference that may occur during the on-chip bioassays and allow high-performance analog signal processing of low noise levels. The use of bipolar transistors to amplify the photodiode current allows the device to be operated at high frequencies, with low power consumption. With these advantages, our bipolar microchip system is thought to be more appropriate than the CMOS chip for repeated onchip bioassays, and our circuit was constructed using bipolar semiconductor technology instead of CMOS technology. To further protect the photodiode from any damage during the on-chip bioassay, the PDA chip was fabricated so that each photodiode is covered with a layer of silicon nitride (Si3 N4 ). Compared with silicon oxide, silicon nitride has a less porous structure, which confers on the PDA chip greater resistance to damage from direct exposure to the bulk medium. Based on an enzymatic color reaction (Wang et al., 2002) and the light absorption properties of the enzymatic reaction product, our bipolar microchip system has great potential as a miniaturized portable DNA chip. The bipolar microchip system is also suitable for on-chip bioassays.

2. Experimental 2.1. Reagents Tris(hydroxymethyl)aminomethane (Tris) was purchased from J.T. Baker Inc. (Phillipsburg, NJ). Ethidium bromide and Alexa Fluor 647 carboxylic acid, succinimidyl ester (Alexa Fluor 647) were purchased from Molecular Probes Inc. (Eugene, OR). The polymerase chain reaction (PCR) kit was obtained from Applied Biosystems (Foster City, CA). The digoxigenin (DIG) DNA labeling and detection kit was purchased from Boehringer Mannheim Biochemica (Mannheim, Germany). 2.2. PCR protocol and dioxigenin DNA labelling The target DNA, the Bacillus subtilis sspE gene, was amplified by PCR. Two primers, sspE-1, 3 -AGGAATAGCTATACGATCAC-5 , and sspE-2, 3 -AGTGCTTTTTTCTGTTTCAG5 were used to obtain a 340-bp PCR product. The PCR reagent mixture consisted of: (a) 0.5 ␮L of 100 ng/␮L genomic DNA, (b) 0.5 ␮L of 5 units/␮L Taq polymerase, (c) 10 ␮L of 10 × PCR buffer, (d) 8 ␮L of dNTP mixture (2.5 mM of each dNTP), (e) 1 ␮L of 10 pmol/␮L each primer and (f) 79 ␮L of sterile water. The above components were stored on ice during the preparation of the PCR mixture. The total volume of the PCR mixture was 100 ␮L. The conditions used for the PCR were: preheating at 95 ◦ C for 5 min, 35 cycles of denaturation at 95 ◦ C for 30 s, annealing at 52 ◦ C for 30 s and extension at 72 ◦ C for 30 s. The efficacy of the PCR reaction was determined using 2% agarose gel electrophoresis. Ethidium bromide was used to stain the PCR product during gel electrophoresis. The PCR product was labeled with DIG using a DIG DNA labeling and detection kit (Hsu et al., 2005). The PCR product was denatured by heating for 10 min at 95 ◦ C, and was then rapidly cooled on ice. An aliquot of 100 ng/␮L fresh denatured DNA, dNTP labeling solution and hexanucleotide solution were mixed in sterile redistilled water and 2 units/␮L of Klenow enzyme was added. The solution was centrifuged for a short period, and incubated for 2 h at 37 ◦ C. The reaction was stopped by the addition of EDTA solution (2 mM, pH 8.0). The labeled DNA was precipitated with 2.5 ␮L of LiCl (4 M) and 75 ␮L of cold ethanol at −20 ◦ C, and was then vortexed. The solution was incubated for 2 h at −20 ◦ C, and then centrifuged at 12,000 × g. The sediment was washed with 70% (v/v) cold ethanol, then dried under vacuum and finally dissolved in TE buffer. 2.3. PDA chip device The PDA chip used in this work was fabricated using conventional bipolar semiconductor technology. Fig. 1 shows a schematic diagram of the photodiode element of the PDA chip. In addition to the p–n junction, another n-type impurity (n+) was doped into the n-type silicon to induce a higher leakage current. The photocurrent from the photodiode flows into the current amplifier through a piece of aluminum (the first metal). Another piece of aluminum (the second metal) was used to protect the current amplifier from exposure to ambient light. The

J.M. Song et al. / Biosensors and Bioelectronics 22 (2007) 1447–1453

1449

Fig. 1. On-chip bioassay on a DNA chip fabricated using bipolar semiconductor technology. The photodiode was coated with a layer of silicon nitride. The polyl-lysine layer on the silicon nitride induced a positive charge. The DIG-labeled target DNA attached to the probe DNA immobilized on the silicon nitride layer by hybridization. An enzymatic color reaction between alkaline phosphatase and NBT/BCIP was performed, and a blue precipitate was produced on the surface of the photodiode. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

probe DNA is directly immobilized on the silicon nitride layer. The PDA chip consists of a 12 × 12 array of photodiodes, an array of current amplifiers and a photodiode element addressing the circuit, which was integrated into a single IC chip, as shown in Fig. 2. Each photodiode had dimensions of 300 ␮m × 300 ␮m and the photodiode-to-photodiode distance was 100 ␮m. The signal from the PDA chip was transferred to a test board via the 4 × 12 pin connectors located on the back edge of the PDA chip. The PDA chip was placed at the center of the test board,

and consisted of a transistor-transistor logic (TTL) circuit, an analog-to-digital converter, and a power supply circuit. The PDA detection elements were individually addressed and were read using digital I/O lines and an analog-to-digital conversion channel provided by an RS232 interface to a laptop computer. The data acquisition process was controlled using custom software written in the C language. Fig. 3 shows the dependence of the photodiode current on the light intensity, which was measured by the electronic circuit

Fig. 2. (a) A photo image of the constructed bipolar microchip system. (b) A schematic representation of the PDA chip. The photodiode array was composed of a 12 × 12 photodiode array. The photocurrent from each photodiode was amplified and converted into an output voltage by an amplifier and current-to-voltage converter.

1450

J.M. Song et al. / Biosensors and Bioelectronics 22 (2007) 1447–1453

Fig. 3. Photoresponse of the photodiode. (a) The circuit diagram used to measure the photocurrent from a photodiode exposed to a light source. A digital multimeter (DMM) was used to measure the photocurrent. (b) A plot of the photocurrent vs. light irradiance obtained using the circuit shown in (a). The chip-to-chip variances in photocurrent were negligible at identical intensities of irradiated light. This is mainly because the concentration of silicon was uniform on the fabricated chips.

using a constructed photodiode with a diameter of 1000 ␮m. A near fourth-order linear response could be obtained with respect to light intensity. For use as a photodetector, the fabricated PDA chip must have linear response that is proportional to the light intensity, and this is important for quantitative analysis on the PDA chip.

on the PDA chip coated with poly-l-lysine was performed using the PCR product and the DIG-labeled target DNA (Hsu et al., 2005). The DNA hybridization protocol was as follows. After denaturing at 95 ◦ C for 10 min and rapid cooling on ice, 1 ␮L of the PCR product solution was immediately dotted onto the PDA chip. The PDA chip was then baked at 80 ◦ C in a drying oven for 2 h under vacuum. The PDA chip was blocked using a prehybridization solution (5× salinesodium citrate buffer (SSC), 0.02% sodium dodecyl sulfate (SDS), 0.1% N-lauroylsarcosine and 0.5% blocking reagent) for 30 min at 45 ◦ C. The freshly denatured DIG-labeled target DNA was added to the prehybridization solution, and this was incubated at 52 ◦ C for 16–20 h for hybridization. Before the DNA was detected, the PDA chip was washed in cleaning solution (2× SSC, 0.1% SDS) for 5 min, and then washed twice in Tris buffer (50 mM Tris–HCl, 150 mM NaCl, pH 7.6) for 5 min each. The staining reaction was performed in the dark without shaking or stirring. The PDA chip was washed in maleic-acid-buffered saline (MBS) (0.1 M maleic acid, 0.15 M NaCl, pH 7.5) and incubated for 30 min with blocking solution (1:10 stock blocking solution in MBS; final concentration, 1% blocking reagent). Anti-DIG-alkaline phosphatase conjugate was diluted to 150 mU/mL (1:5000 dilution) in blocking solution, and the PDA chip was incubated for 30 min in the diluted antibody–conjugate solution. The antibody–conjugate solution was then washed away by rinsing the PDA chip three times in MBS solution for 5 min each. The PDA chip was then allowed to equilibrate in Tris buffer (50 mM Tris–HCl, 100 mM NaCl, 50 mM MgCl2 , pH 9.5) for 2 min at room temperature, and was subsequently incubated in nitroblue tetrazolium/5-bromo4-chloro-3-indolyl-phosphate (NBT/BCIP) staining solution to visualize the antibody binding. The reaction was stopped by the addition of Tris–EDTA buffer solution (10 mM Tris–HCl, 1 mM EDTA, pH 8.0) to the PDA chip. The reaction was also confirmed by the reduction in the PDA signal that occurred as a result of light absorption by the color reaction product when the light source was irradiated onto the PDA chip that had already been covered with the color reaction product.

2.4. DNA hybridization and detection on a PDA chip 2.5. Fluorescence detection on the PDA chip DNA hybridization was performed directly on the surface of the silicon nitride (Si3 N4 ) layer, as shown in Fig. 1. The surface of the PDA chip was coated with poly-l-lysine and thus became positively charged (Eisen and Brown, 1999; Hessner et al., 2004). The poly-l-lysine was applied to the PDA chip at room temperature. To remove any dust on the PDA chip, the surface of the PDA chip was cleaned with 60% absolute ethanol for a period of 5 min, and then rinsed with sterile redistilled water for another 5 min. The PDA chip was then immersed in poly-llysine coating solution (10%, v/v solution in 1× PBS; pH 7.4) and incubated with gentle shaking for 1 h. After the PDA chip was coated with poly-l-lysine, it was rinsed once again with sterile redistilled water for 5 min, centrifuged at 1300 rpm for a period of 3 min to remove any water on the PDA chip, and then dried at 45 ◦ C in a drying oven for 30 min. The PDA chip was finally stored in a box to protect it from dust. DNA hybridization

A diode laser beam (λ = 635 nm) was focused onto a capillary fixed to a capillary holder, using a planoconvex lens with a 40 mm focal length (Song et al., 2005). The polyimide coating of the capillary was removed at the laser irradiation position. Aqueous Alexa Fluor 647 solution was injected into the capillary using a syringe. The up and down movements of the capillary were controlled using an x–y translational stage, so that the laser beam passed through the capillary. A 5× microscope objective (Nikon, 0.1 NA), placed perpendicular to the capillary array, was used to collect the fluorescence from the capillary and focus it onto a photodiode element in the bipolar microchip. A narrow band-pass filter (Edmund Industrial Optics; central wavelength = 656 nm, FWHM = 10 nm) was placed in front of the bipolar microchip to remove any laser scattering.

J.M. Song et al. / Biosensors and Bioelectronics 22 (2007) 1447–1453

3. Results and discussion Fig. 4 shows the electric circuit designed for current amplification in the fabricated PDA chip. The labels Q1A, Q2A, Q3A and Q4A denote the current mirrors, where equivalent currents flow. The reference circuit, consisting of Q4A, Q6A, Q27A, Q28A and DP4A, was used to obtain the background signal induced by the light source irradiating the bare photodiode upon which no DNA hybridization had been performed. On the other hand, other detection photodiodes, such as DP1A, DP2A and DP3A, were tested for DNA hybridization. The color reaction product displayed a certain degree of absorption of the light irradiating the detection photodiodes. Therefore, the amount of light that reached the detection photodiodes covered with this product was less than that reaching the bare photodiodes. Consequently, the photocurrent from the detection photodiode was reduced. On the photodiode, the target DNA with incorporated DIG-dUTP was detected by anti-DIG antibody conjugated with alkaline phosphatase. The enzymatic reaction of BCIP and NBT salts with alkaline phosphatase produces a blue precipitate. The source light (white light) was absorbed by the blue precipitate. To measure the current gain in the constructed PDA chip, we used an electric circuit consisting of a photodiode. The resulting current gains of the transistor were 100:1. Assuming that the current gains of the transistors were equal, and that the photocurrent produced at photodiode DP4A exposed to the light source was 30 nA, then the current flowing at Q27A was 3 ␮A because of the 100× amplification. Because the current was also amplified 100 times at Q28A, the current flowing at Q4 was 300 ␮A.

1451

When no DNA hybridization is performed, and there is no blue precipitate on the detection photodiode DP1A, the 300 ␮A current sinks to Q21A and Q22A. As a result, all the current provided by Q1A must flow into Q21A and Q22A. Accordingly, the base voltage at QO1 is Vcc /2 (Vcc = 4.9 V), and the final output voltage at column 1 is (Vcc /2)–0.7 = 1.75 V. However, when a blue precipitate forms on the detection photodiode DP1A after the DNA hybridization reaction and the enzymatic color reaction, the output voltage at column 1 increases because the amount of light reaching the photodiode is reduced, and a lower photocurrent is produced. If the photocurrent at DP1A is 25 nA, then a 250 ␮A current sinks to Q21A and Q22A. In this situation, of the 300 ␮A current supplied by Q1A, 250 ␮A flows to Q21A and Q22A and 50 ␮A flows across resistor R17. The base voltage at QO1 can then be calculated as follows: VB(QO1) = (50 kΩ × I1) + (50 kΩ × 50 ␮A)

(1)

(50 kΩ × I1) = 4.9 V − VB(QO1)

(2)

where I1 is the current flowing at R1 and VB(QO1) is the base voltage at QO1. From the above equations, VB(QO1) was calculated to be 3.7 V. This calculation shows that the output voltage of the detection photodiode increases when the amount of blue precipitate on the detection photodiode increases. As shown in Fig. 4, the amplification circuit is simple when it is composed of bipolar transistors, and this allows the amplification circuit area of each photodiode to be small. DNA detection based on the light absorption by the enzymatic products at the surface of the constructed PDA chip has several advantages. A conventional DNA chip allows the simultaneous

Fig. 4. Circuit diagram of the amplification of the photocurrent produced by the photodiode array.

1452

J.M. Song et al. / Biosensors and Bioelectronics 22 (2007) 1447–1453

detection of genes immobilized on a solid support and essentially requires a complicated optical alignment. Compared with a conventional system, our system can perform high-throughput parallel DNA immobilization and sensing on the PDA chip without the need for an additional solid transducer surface. As a consequence, the constructed system eliminates the need for a complicated optical alignment, such as a 1:1 correspondence between the microarray, a two-dimensional detector, and the optical components for emission collection. This is very advantageous for the production of a portable and miniaturized DNA chip. The PDA chip can identify each pixel using the addressing circuit. Accordingly, different types of genes immobilized on each photodiode can be accessed and analyzed simultaneously. Furthermore, semiconductor technology can provide a variety of coating materials on silicon wafers so that the PDA chip can have specific properties for specific biological samples. Currently, a conventional biochip reader is very costly. Another advantage of semiconductor technology is its suitability for mass production, which makes possible low-cost PDA chips. Our PDA chip is cheap, and reusable as well as replaceable. The hybridized target DNA is detected by the enzymatic reaction between NBT/BCIP and the anti-DIG-alkaline phosphatase conjugate, and poly-l-lysine is used to immobilize the DNA on the detection photodiodes. The surfaces of the detection photodiodes are positively charged after the poly-l-lysine treatment. The DNA is immobilized on the detection photodiode as the result of an electrostatic interaction with the poly-l-lysine. Although a quantitative analysis of the hybridized DNA can be made based on the light absorption properties of the enzymatic product, simple visualization shows the efficacy of the entire DNA bioassay. An increase in blue precipitate increases the light absorption, and produces a higher output voltage from the PDA chip. This visualization is a useful tool to rapidly confirm whether DNA hybridization has occurred (Lizard et al., 1997; Soumet et al., 1995), even before the output voltage is read. The enzymatic color reaction also provides an output optical density that is proportional to the amount of blue precipitate, which allows an approximate quantitative analysis. An accurate quantitative analysis can be made from the digitized output voltage. The silicon nitride layer on the PDA chip surface is covered with blue precipitate after the enzymatic reaction. Silicon dioxide is commonly used as an insulating layer, as a mask, and as a sacrificial material. Silicon dioxide is often used because it is convenient. However, amorphous oxides have an open structure. A variety of impurities, such as sodium and potassium ions, can readily diffuse through a porous silicon dioxide structure, and this happens particularly rapidly when the oxide is hydrated (Bhushan, 2004). On the other hand, silicon nitride provides an efficient passivation barrier to the diffusion of water and mobile ions, particularly Na+ (Madou, 2002). Moreover, it oxidizes slowly compared with silicon. In our work, the PDA chip was exposed to reaction solutions during the DNA hybridization and enzymatic reactions. Therefore, it was important that the silicon nitride layer was not damaged during the reactions. Compared with a silicon oxide layer which has a more open structure, the silicon nitride layer allows the on-chip bioassay to be performed on the

Fig. 5. The diode laser-induced fluorescence of Alexa Fluor 647 with 10−7 M. The fluorescence measurements were made before and after the full on-chip bioassay. After the 10th bioassay, the fluorescence was measured to determine any variance in the fluorescent signal. The signal on the Y-axis corresponds to the relative fluorescence intensity of Alexa Fluor 647. The relative standard deviations before and after a full on-chip bioassay, and after 10 full on-chip bioassays were 4.0, 15.1 and 18.0 mV, respectively, based on four measurements.

PDA chip. Bipolar transistor amplification also provides a lower noise level with respect to the photocurrent from the photodiode exposed to reaction solutions. The stability of the PDA chip over the entire on-chip bioassay was investigated by observing the variance in output voltage of a photodiode exposed to reaction solution at each reaction step. The background signal of the PDA chip remained constant over the entire reaction period and was obtained from 12 photodiodes at each reaction step. To confirm that the photoresponse of the photodiode acting as a photodetector was maintained after exposure to a large volume of reaction solution, the fluorescence intensity of Alexa Fluor 647 was measured at a photodiode exposed to reaction solutions, using the diode laser-induced fluorescence technique. The results are shown in Fig. 5. The concentration of Alexa Fluor 647 was 10−7 M. Measurements were made before and after the full on-chip bioassay and the results of each were compared. The photoresponse of the exposed photodiode was also measured after repeated use (i.e., 10 full on-chip bioassays). The observed fluorescence intensity of Alexa Fluor 647 was the same, irrespective of the number of exposures of the chip. This means that the PDA chip fabricated using bipolar semiconductor technology is adequately protected from any damage that may occur during an on-chip bioassay, and is therefore an appropriate photodetector for on-chip bioassays. Fig. 6 shows the photosignals obtained as a function of DNA concentration. As a result of the color reaction, blue precipitate spread over the detection photodiodes where the DIG-labeled DNAs were spotted. The DNA concentrations were prepared by two-fold serial dilutions of the initial DNA stock solution, and spotted in five different areas. The light source irradiated the detection photodiodes covered with blue precipitate. An optical filter was placed in front of the bipolar microchip to narrow the wavelength range of the light source, where there was strong absorption by the blue product. To obtain an output voltage from the chip for four different concentrations, the average output voltages from five different detection photodiodes were collected at each concentration. This was repeated

J.M. Song et al. / Biosensors and Bioelectronics 22 (2007) 1447–1453

1453

microchip was carried out, and successfully complemented the results of the color reaction. This detection system eliminates the need for any complicated optical alignment or optical components, which are usually required for high-speed and highthroughput optical transduction based DNA chips. The detection limit of the target gene was 9.4 × 10−7 M. This system has great potential in applications to diverse fields, including protein chips based on on-chip bioassays. Acknowledgement This research is sponsored by Korean Ministry of Commerce, Industry and Energy under contract 10023590-0000-00 program with Seoul National University. Fig. 6. Photodiode signal vs. the concentration of target DNA. As the concentration of the target DNA solution decreased, the photodiode signal also decreased due to the decrease in the number of blue precipitates. The signal on the Y-axis corresponds to the intensity of the light reaching the photodiode. The linear regression coefficient (R2 ) was 0.99 from 1.6 × 10−6 to 1.3 × 10−5 M. The relative standard deviations from the lowest to the highest concentrations were 6.7, 10.4, 19.8, 24.2 and 29.5 mV, based on four measurements.

with four different chips. Then the output voltage values at the same concentration were averaged for the four different chips, and output voltage versus concentration was plotted. The data are shown in Fig. 6. The error bars in Fig. 6 correspond to the chip-to-chip signal variance at each concentration. The output voltage showed a linear dependence on the DNA concentration from 1.6 × 10−6 to 1.3 × 10−5 M (linear regression coefficient, R2 = 0.99). This result is in good agreement with the linear response of the detection photodiode with respect to the light intensity, as shown in Fig. 3. A detection limit of 9.4 × 10−7 M (at S/N = 3) was obtained with our DNA chip. A background noise level of 9 mV was obtained from the control enzymatic color reaction performed without target DNA. The standard deviation of the output voltage obtained from four different chips at the same DNA concentration was within 10%, which reflects the consistent properties of the fabricated PDA chips. 4. Conclusions The sspE gene was detected with a PDA chip fabricated using bipolar semiconductor technology. DNA hybridization was performed on the surface of the PDA chip. The bipolar microchip successfully demonstrated the direct detection of a target sspE gene on the surface of the PDA. The bipolar microchip allowed a high-performance analog process with a low noise level, which is suitable for on-chip bioassays, particularly with a microchip intended for repeated use. The principle underlying our target DNA detection was the light absorption by the enzymatic product produced by DNA hybridization followed by an enzymatic reaction. Quantitative analysis of the hybridized target DNA using the bipolar

References Allain, L.R., Askari, M., Stokes, D.L., Vo-Dinh, T., 2001. Fresenius J. Anal. Chem. 371, 146–150. Bhushan, B. (Ed.), 2004. Springer Handbook of Nanotechnology. Springer, Berlin. Culha, M., Stokes, D.L., Griffin, G.D., Vo-Dinh, T., 2004. Biosens. Bioelectron. 15, 1007–1012. DeLouise, L.A., Kou, P.M., Miller, B.L., 2005. Anal. Chem. 77, 3222– 3230. Eisen, M., Brown, P., 1999. Methods Enzymol. 303, 179–205. Fouque, B., Schaack, B., Obeid, P., Combe, S., Getin, S., Barritault, P., Chaton, P., Chatelain, F., 2005. Biosens. Bioelectron. 20, 2335–2340. Gosch, M., Serov, A., Anhut, T., Lasser, T., Rocas, A., Basse, P.A., Popovic, R.S., Blom, H., Rigler, R., 2004. J. Biomed. Opt. 9, 913–921. Hessner, M.J., Meyer, L., Tackes, J., Muheisen, S., Wang, X., 2004. BMC Genom. 5, 53–60. Hsu, Y.C., Yeh, T.J., Chang, Y.C., 2005. J. Virol. Methods 128, 54–60. Hwang, S.H., Shih, Y.C., Wu, C.Y., Yuan, C.J., Yang, Y.S., Li, Y.K., Wu, T.K., 2004. Biosens. Bioelectron. 19, 1627–1633. Kohn, J.E., Plaxco, K.W., 2005. Proc. Natl. Acad. Sci. U.S.A. 102, 10841–10845. Lizard, G., Chignol, M.C., Roignot, P., Souchier, C., Chardonnet, Y., Schmitt, D., 1997. Histochem. J. 29, 545–554. Madou, M.J., 2002. Fundamentals of Microfabrication, second ed. CRC Press, New York. Nakauchi, G., Inaki, Y., Kitaoka, S., Yokoyama, C., Tanabe, T., 2002. Nucleic Acids Res. (Suppl. 2), 257–258. Pazos, M.J., Alfonso, A., Vieytes, M.R., Yasumoto, T., Botana, L.M., 2005. Chem. Res. Toxicol. 18, 1155–1160. Peter, C., Meusel, M., Grawe, F., Katerkamp, A., Cammann, K., Borchers, T., 2001. Fresenius J. Anal. Chem. 371, 120–127. Reichert, J., Csaki, A., Kohler, J.M., Fritzsche, W., 2000. Anal. Chem. 72, 6025–6029. Song, J.M., Culha, M., Kasili, P.M., Griffin, G.D., Vo-Dinh, T., 2005. Biosens. Bioelectron. 20, 2203–2209. Soumet, C., Ermel, G., Boutin, P., Boscher, E., Colin, P., 1995. Biotechniques 19, 792–796. Storhoff, J.J., Marla, S.S., Bao, P., Hagenow, S., Mehta, H., Lucas, A., Garimella, V., Patno, T., Buckingham, W., Cork, W., Muller, U.R., 2004. Biosens. Bioelectron. 19, 875–883. Stratis-Cullum, D.N., Griffin, G.D., Mobley, J., Vass, A.A., Vo-Dinh, T., 2003. Anal. Chem. 75, 275–280. Wang, R.F., Kim, S.J., Robertson, L.H., Cerniglia, C.E., 2002. Mol. Cell Probes 16, 341–350.