Fully Automated Sample Preparation Microsystem for Genetic

All the solutions are aspirated out of the chip from the waste outlet to a syringe pump (PSD/4, Hamilton), on which a 500-μL Hamilton glass syringe i...
1 downloads 15 Views 6MB Size
Download PDF
Article pubs.acs.org/ac

Fully Automated Sample Preparation Microsystem for Genetic Testing of Hereditary Hearing Loss Using Two-Color Multiplex AlleleSpecific PCR Bin Zhuang,Δ,†,‡ Wupeng Gan,Δ,†,‡,∥,⊥ Shuaiqin Wang,†,‡ Junping Han,§ Guangxin Xiang,∥,⊥ Cai-Xia Li,# Jing Sun,# and Peng Liu*,†,‡,¶ †

Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, 100084, China Medical Systems Biology Research Center, School of Medicine, Tsinghua University, Beijing, 100084, China § Chinese Peoples Public Security University, Beijing, 100038, China ∥ CapitalBio Corporation, Beijing, 102206, China ⊥ National Engineering Research Center for Beijing Biochip Technology, Beijing, 102206, China # Institute of Forensic Science, Key Laboratory of Forensic Genetics, Ministry of Public Security, Beijing, 100038, China ¶ Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Hangzhou, Zhejiang 310003, China ‡

S Supporting Information *

ABSTRACT: A fully automated microsystem consisting of a disposable DNA extraction and PCR microchip, as well as a compact control instrument, has been successfully developed for genetic testing of hereditary hearing loss from human whole blood. DNA extraction and PCR were integrated into a single 15-μL reaction chamber, where a piece of filter paper was embedded for capturing genomic DNA, followed by in-situ PCR amplification without elution. Diaphragm microvalves actuated by external solenoids together with a “one-way” fluidic control strategy operated by a modular valve positioner and a syringe pump were employed to control the fluids and to seal the chamber during thermal cycling. Fully automated DNA extractions from as low as 0.3-μL human whole blood followed by amplifications of 59-bp β-actin fragments can be completed on the microsystem in about 100 min. Negative control tests that were performed between blood sample analyses proved the successful elimination of any contamination or carryover in the system. To more critically test the microsystem, a two-color multiplex allele-specific PCR (ASPCR) assay for detecting c.176_191del16, c.235delC, and c.299_300delAT mutations in GJB2 gene that accounts for hereditary hearing loss was constructed. Two allele-specific primers, one labeled with TAMRA for wild type and the other with FAM for mutation, were designed for each locus. DNA extraction from blood and ASPCR were performed on the microsystem, followed by an electrophoretic analysis on a portable microchip capillary electrophoresis system. Blood samples from a healthy donor and five persons with genetic mutations were all accurately analyzed with only two steps in less than 2 h.

H

mutation detection,3−5 allele-specific PCR (ASPCR) coupled with a detection technique, such as microarray2,6 or capillary electrophoresis,7 is one of the most widely used methods. However, the current ASPCR-based genetic testing is still labor-intensive, time-consuming, and is one that requires welltrained personnel and dedicated instruments.8 There is no doubt that this testing could be significantly improved if the entire operation was fully automated and integrated on a single instrument.8,9 Micro-Total Analysis System (μTAS) could provide an excellent platform for reaching this goal because of

earing loss is one of the most common congenital anomalies, affecting one in 1000 live births, and at least 60% of these cases are hereditary.1,2 Connexin 26 gene (also known as GJB2 gene) which encodes the gap junction protein connexin 26 accounts for up to 50% of recessive nonsyndromic deafness.2 According to the previous study on Chinese nonsyndromic hearing impairment (NSHI) patients, c.235delC mutation was found in 88.8% of the patients who carry the GJB2 mutation. c.299_300delAT and c.176_191del16 mutation account for 24.8% and 7.6% of these patients, respectively.2 The rapid detection of these mutations is highly desired in clinical diagnosis and neonatal screening, so that early clinical intervention or consultation can be put in place. Among all the techniques that have been developed for © 2014 American Chemical Society

Received: October 12, 2014 Accepted: December 11, 2014 Published: December 11, 2014 1202

DOI: 10.1021/ac5039303 Anal. Chem. 2015, 87, 1202−1209

Analytical Chemistry

■

Article

MATERIALS AND METHODS DNA Extraction and Amplification Chip. As illustrated in Figure 1A, the disposable microfluidic device for DNA

its facile miniaturizing and scaling capability, as well as its extraordinary potential of integrating multiple analytical steps on a single microdevice.10,11 In 2006, Easley et al. integrated a solid-phase extraction column for DNA purification with PCR and CE on a single device for pathogen detection for the first time.12 Later, Liu et al. reported an integrated system for forensic short tandem repeat (STR) analysis.13 And Kim et al. described a lab-on-a-disc microdevice that integrated DNA extraction, isothermal recombinase polymerase amplification, and lateral flow strip together for Salmonella detection from spiked milk samples.14 While this progress validates the concept of the “sample-in-answer-out”, the practical application of such microsystems is still not common. The major challenge in the development of a more feasible and affordable microsystem for genetic analysis comes from the sample preparation process, which usually includes sample preprocess, cell lysis, nucleic acid extraction, impurity wash, DNA/RNA elution, PCR reagent loading, and PCR amplification.15−17 This complicated procedure involves a variety of chemicals, requires precise fluidic and temperature controls, and demands efficient integration of multiple microstructures. The diverse sample types that the system need to process make the integration work even more difficult. Nevertheless, as the first step, various microfabricated devices have been developed to translate the DNA extraction to the microchip platform. For examples, Landers’ group has developed several different solidphase DNA extraction microchips using either silica-based or pH-induced DNA adsorption methods.18,19 Novel DNA capture materials, such as nanoporous aluminum oxide membrane (AOM) and filter paper, have also been utilized for DNA purification.20,21 To integrate all the sample preparation steps together, Shaw et al. developed an integrated DNA extraction and PCR amplification device which can directly accept buccal swab samples.22 Landers’ group integrated solid phase extraction with PCR without the aid of valves for forensic DNA typing.23 Later, they employed an enzyme-based DNA preparation method for on-chip analysis and successfully shortened the time of DNA extraction and PCR to less than 45 min.24 More recently, our group developed a filter-paper based microdevice which can extract and amplify DNA from a variety of raw samples, such as blood stains, buccal swabs, and cigarette butts.25 Unfortunately, the sample preparation systems demonstrated by far were either not fully automated or too expensive to be widely used in a daily base. Here, we present a fully automated sample preparation microsystem consisting of a low-cost, disposable plastic chip for DNA extraction and PCR amplification with a compact control instrument for chip operations. Using a filter paper-based DNA extraction structure developed previously,25 DNA isolated from human whole blood can be captured in a filter disc located in the extraction chamber, where an in-situ PCR amplification was carried out directly without elution. Because of the use of externally actuated microvalves and a “one-way” fluidic operation strategy, the microsystem achieved full automation and integration at a low cost. To critically test the performance of the system, DNA extractions and two-color multiplex allelespecific PCR amplifications of three GJB2 gene mutations (c.176_191del16, c.235delC, and c.299_300delAT) that lead to inherited hearing loss were conducted from human whole blood. It is a significant step toward the development of a “sample-in-answer-out” microsystem allowing an individual with minimal training to perform rapid nucleic acid analysis.

Figure 1. Disposable plastic microchip for DNA sample preparation. (A) Exploded view of the microdevice with integrated DNA extraction and PCR. (B) Schematic view of the microchip structure. (C) Expanded view of the filter disc, the top and the bottom compartment forming the DNA extraction and PCR chamber. (D) Photograph of the microchip with dimensions 88 × 20 mm.

extraction and PCR amplification is comprised of two PMMA layers with three PDMS membrane discs (BISCO HT-6240, Rogers, Woodstock, CT) for valve actuation and a filter paper disc (Fusion 5, GE Healthcare, Pittsburgh, PA) for DNA extraction. Figure 1B shows that each microchip contains a reference temperature unit and a DNA extraction and amplification unit. The reference temperature unit is designed for fixing a thermocouple (TT-K-36-SLE, OMEGA, Stamford, CT) in the chip. Once the temperature calibration is accomplished, this thermocouple is no longer needed in the following experiments. The DNA extraction and amplification unit includes a DNA extraction inlet, a PCR inlet, a waste outlet, a loading chamber, a DNA extraction and PCR chamber, a sampling chamber, and three mechanically actuated diaphragm microvalves. The inlets and the outlet are formed by drilling via holes on the ends of channels. The loading chamber for pipetting samples into the chip consists of a through hole with a 2 mm diameter on the upper layer and a 300-μm-deep round compartment with the same diameter on the lower PMMA layer. A Microseal B adhesive tape (Bio-Rad, Hercules, CA) is employed to seal the chamber once the samples are loaded. The sampling chamber has the same structure as that of the loading chamber, except the diameter is increased to 3 mm and the compartment is 1 mm deep. This chamber is sealed by a piece of BarSeal tape (Thermo Fisher, Waltham, MA), which can be peeled off easily when taking PCR products out. The DNA extraction and PCR chamber with a volume of 15 μL is constructed by aligning two compartments together, located on the upper and the lower layers, respectively. The top compartment with a round end on one side and a tapered end on the other is 4 mm in length, 2.6 mm in width, and 0.5 mm in height. The bottom compartment 1203

DOI: 10.1021/ac5039303 Anal. Chem. 2015, 87, 1202−1209

Article

Analytical Chemistry is a 2.6 mm-diameter blind hole with a depth of 300 μm. To accommodate a 200-μm-thick Fusion 5 filter disc in the chamber, the bottom compartment has a 200-μm-deep circular step, as shown in Figure 1C. A 0.5 mm-diameter pillar protruding from the ceiling of the top compartment is designed to press the filter disc in place. Microfluidic channels with a cross section of 300 × 300 μm link all the microstructures. The mechanically actuated diaphragm microvalve, as illustrated in Figure 2, consists of three components: a housing

stack. The pressure is maintained during the temperature ramping process. Once the temperature reached the setting, a prebonding step was performed with a pressure of 2 bar for 5 min. After that, the temperature was increased to 249 °F under the pressure of 1.5 bar, followed by a final bonding step with a 2-bar pressure for 5 min. Then, the chip was cooled down to room temperature while maintaining the pressure at 0.5 bar. Prior to use, 160 mg/mL polyethylene glycol (PEG, MW 10 000, Sigma-Aldrich, St. Louis, MO) was loaded into the chip and incubated for 5 min. After dried completely by aspiration, the chip was stored at room temperature. Compact Microchip Control Instrument. The selfcontained control instrument used to operate the microchip is shown in Figure 3A. The instrument contains a microchip

Figure 2. Schematics of the on-chip microvalve. (A) Cross-section view of the microvalve structure in the open status. (B) Cross-section view in the close status. (C) Exploded view of the microvalve components. (D) Expanded view of the valve base, the chamfer, and the microchannels.

compartment, an elastic PDMS membrane disc, and a valve base. The housing compartment on the upper PMMA layer is a 2.5 mm-diameter through hole with a 4.5 mm-diameter circular step on the bottom side for accommodating the PDMS membrane disc. The 0.5 mm-thick PDMS membrane disc is glued to the step and embedded between two PMMA layers during the thermal bonding process. The valve base together with flow channels are designed on the lower PMMA layer. The valve base is a 2 mm-diameter blind hole with a 45° chamfer on the edge. The diaphragm valve is actuated by a plunger, which is controlled by solenoids. The end of the plunger also has a 45° chamfer, providing a flush contact between the PDMS and the valve base when closing. Microdevice Fabrication. The device fabrication was performed as follows: first, the structures on the PMMA layers were milled by a milling machine (MODEL 5410, Sherline, Vista, CA). To assemble the microvalve structure, a silicone sealant (HT902, Huitian, Xiangyang, China) was first spread evenly on the circular step of the housing compartment, and the PDMS disc was then placed immediately. Once the sealant was fully solidified, all the layers were sequentially cleaned with ethanol, detergent, and deionized water (DI water). After completely dried by N2, they were carefully assembled in a clean hood. The thermal bonding of two PMMA layers was performed on a manual hydraulic press (15−1-HT, GRIMCO, Paterson, NJ). First, the assembled PMMA stack was sandwiched between two polished steel plates and then put into the hydraulic press. The temperature of the press machine was set to 205 °F and a pressure of 1.5 bar was applied to the chip

Figure 3. Compact microchip control system for DNA extraction and PCR amplification. (A) Photograph of the control instrument. The analysis system box has dimensions 30 × 30 × 16 cm. (B) Top view of the microchip fixture located on the top of the instrument. Red circle indicates the place where the PCR Master Mix tube is on the fixture. (C) Photograph of the microchip fixture when it is open. (D) Schematic diaphragm of the fluidic control system.

fixture, a fluidic system, and electronics for temperature and fluidic control. A LabVIEW program (NI, Austin, TX) is used to control the instrument through a NI USB 6259 OEM board. The microchip fixture consists of a chip platform and a manifold. As shown in Figure 3B and C, the microdevice is placed into a recessed area on the top of the chip platform and held in place with the manifold using screws. The frames of the fixture are made of Teflon, providing excellent chemical 1204

DOI: 10.1021/ac5039303 Anal. Chem. 2015, 87, 1202−1209

Article

Analytical Chemistry

using the syringe pump in exact 6 s at a speed of 250 μL/min. The microchip is then taken out from the instrument and the amplicons are pipetted out from the sampling chamber. Singleplex PCR for System Characterization. K562 standard DNA (Promega, Madison, WI) as well as a human whole blood sample obtained from a volunteer with informed consent was employed to test the performance of the system. A set of primers (forward, 5′-TGAGCGCGGCTA CAGCTT-3′, Reverse: 5′-TCCTTAATGTCACGCACGATT T-3′) (Sangon Biotech, Shanghai, China) amplifying a 59-bp fragment of βactin gene from human genomic DNA was used to verify onchip PCR amplifications. The 100-μL PCR mixture was comprised of 2 μL each primer (10 μM), 20 μL PCR buffer (10× , Roche, Indianapolis, IN), 2 μL dNTP (10 mM each), 2.5 μL Roche FastTaq polymerase, 2 μL bovine serum albumin (BSA, 50 μg/μL, Sigma-Aldrich), 3 μL PEG (160 μg/μL), and 66.5 μL DI water. The PCR protocol includes an initial activation of polymerases at 95 °C for 5 min, followed by 35 cycles of 95 °C for 30 s, 60 °C for 30 s and 72 °C for 30 s, and a final extension step for 10 min at 72 °C. K562 DNA was employed as a positive control for PCR operations. All the experiments were performed on brand new microchips and the positive control was run on a conventional thermal cycler (Nexus, Eppendorf, Hamburg, Germany). PCR products together with a DNA Marker DL2000 (Takara Bio, Shiga, Japan) were electrophoretically analyzed on a 2% agarose gel at 150 V. In the sensitivity test, amplicons obtained using TAMRA-labeled primers were analyzed on a portable microchip capillary electrophoresis instrument developed by our group (details listed in the Supporting Information). Two-Color Multiplex Allele-Specific PCR for Detecting Hereditary Hearing Loss. Human whole blood samples from a healthy donor and five persons with genetic mutations were obtained from CapitalBio Corporation (Beijing, China) with informed consent. Two-color multiplex allele-specific PCR for detecting three mutations in GJB2 gene (c.176_191del16, c.235delC, and c.299_300delAT) was constructed using three sets of primers kindly provided by CapitalBio. These three loci share the same reverse primer, while each locus has its own two allele-specific (AS) primers labeled with two different fluorescence dyes, TAMRA for wild type and FAM for mutation, respectively. The sizes of the amplicons are listed in Table 1. Each 100-μL PCR Master Mix is composed of 21

resistance and thermal insulation. The manifold contains three fluidic microport connectors (PK1/16-LSM01, Wenhao, Suzhou, China) to provide leakage-proof connections to the chip. The manifold also contains a heating block for PCR thermal cycling, which is constructed by sandwiching two 3 × 3 cm Peltiers between a polished oxygen-free copper (OFC) plate and a copper heat sink. A thermocouple is inserted into the center of the OFC plate to measure the block temperature. Temperature control is accomplished through a PID module within the LabVIEW program. To actuate the on-chip microvalves, self-locking solenoids (ZHK-0521, Zonhen Electric Appliances, Shenzhen, China) installed on the manifold are employed to drive plungers, which can protrude into the housing compartments of the microvalves to press the PDMS membrane discs. The schematic of the fluidic system in the instrument is shown in Figure 3D. There are three reagent tubes in the tube rack, including a 20-mM sodium hydroxide (NaOH) tube for cell lysis, a 1-mM hydrochloric acid (HCl) tube for neutralization, and a DI water tube for rinsing. These tubes are connected to three input ports of a modular valve positioner (MVP) (Hamilton, Bonaduz, Switzerland) through PTFE tubing. The fourth input port of the MVP is left open to air (blank). The MVP sequentially switches each reagent tube to the DNA extraction inlet of the microchip. As indicated in Figure 3B by a red circle, a 200-μL Eppendorf tube containing PCR reagents is placed on the manifold, from which PCR reagent is drawn into the chip through a short piece of tubing in order to save the reagents. All the solutions are aspirated out of the chip from the waste outlet to a syringe pump (PSD/4, Hamilton), on which a 500-μL Hamilton glass syringe is installed. The waste port of the pump is connected to a waste tube on the tube rack. Procedure of DNA Extraction and PCR. The detailed operation procedure for DNA extraction and amplification can be found in Supporting Information Figure S-4. Briefly, blood sample is pipetted into the loading chamber, which is then sealed by a piece of adhesive tape. After that, the chip is loaded into the microchip fixture on the instrument and a preprogrammed protocol is carried out automatically. First, the DNA extraction process, which is exactly the same as that in our previous work,25 begins with a washing step. 200 μL of DI water is aspirated through the filter disc by the syringe pump while closing valve 1 and opening valve 2 and 3. This washing step brings the blood sample in the loading chamber into the DNA extraction and PCR chamber, where cells in blood can be blocked by the filter disc. After that, 100 μL of NaOH is drawn into the chamber and incubated for 200 s to lyse cells completely, followed by a neutralization step using 60-μL HCl and a final washing step using 60-μL DI water. All the reagents are automatically switched and aspirated through the filter disc using the MVP and the syringe pump at a speed of 800 μL/ min. Between each reagent loading step, an air plug is introduced into the tubing by switching the MVP to the blank port, followed by the reset of the syringe pump to its fully infused position to expel the solution in the syringe to the waste tube. After the DNA extraction, the PCR Master Mix is aspirated into the chip via the PCR inlet at a rate of 250 μL/ min by opening valve 1 and closing valve 2. All the valves are then tightly closed to seal the PCR chamber completely and the thermal cycling is conducted. Once the PCR stops, valve 1 and 3 are reopened, and the PCR products are transported from the DNA extraction and PCR chamber to the sampling chamber

Table 1. Sizes of the Amplicons in the Multiplex AlleleSpecific PCR wild type mutation

c.299_300delAT

c.235delC

c.176_191del16

109 109

173 172

221 214

μL of primer set, 10 μL of PCR buffer (10 × , Roche), 3 μL of dNTP (10 mM each), 8 μL of MgCl2 (25 mM), 5 μL of Roche FastTaq polymerase, 10 μL of BSA (50 μg/μL), 3 μL of PEG (160 μg/μL), and 40 μL of DI water. The thermal cycling protocol is the same as that described above. PCR products were analyzed on the microchip capillary electrophoresis instrument.

■

RESULTS AND DISCUSSION Microdevice Design Strategies. To achieve fully automated sample preparation in the microsystem at a low cost, several design strategies were employed. First, a loading

1205

DOI: 10.1021/ac5039303 Anal. Chem. 2015, 87, 1202−1209

Article

Analytical Chemistry

timing issue, resulting in more reliable and reproducible controls. Microvalve Characterization. The microvalves integrated in our microchip have two functions: switching the fluids during the fluidic loading process and sealing the chamber during PCR. Both functions require the valve be able to withstand high pressures without any leakage. To characterize the hydrodynamic properties of the on-chip microvalve, we first measured the flow rates of water passing through the valve at different driving pressures when the valve was at its fully open status. As shown in Figure 4, the water flow rate in the channel

chamber was designed upstream the DNA extraction and PCR chamber in our system as a macro-to-micro interface for accepting human whole blood samples directly. Meanwhile, a sampling chamber was located downstream the DNA extraction and PCR chamber as a micro-to-macro interface, where PCR products were pipetted out after each run. The macro-to-micro interface is important because it determines what kinds of preprocess operations are needed prior to loading samples into the chip. With this loading chamber, the manual operation in our system only includes loading a sample into the chip, sealing the chip, and putting the chip into the instrument, and the rest of the process is completely conducted by the instrument under the control of a LabVIEW program. Likewise, the sampling chamber was also designed for ease of use. In addition, since the heating block in our system can only cover a limited chip surface around the DNA extraction and PCR chamber, only ∼15 μL of the PCR solution in the chamber undergo the thermal cycling while the rest of ∼25 μL solution in the flow channels, valve bases, reservoirs, the loading and the sampling chambers remains unreacted. As a result, if taken out directly from the waste outlet, PCR products will be diluted at least 2fold. In our system, a pumping step which moves PCR products from the PCR chamber to the sampling chamber is performed after thermal cycling. From the sampling chamber, 15 μL of the solution, roughly equal to the volume of the PCR chamber, is taken out for the downstream analysis, partially alleviating the dilution problem. Second, on-chip diaphragm microvalves actuated by external solenoids and a “one-way” fluidic control operated by a modular valve positioner and a syringe pump constitute the valving and pumping system of the instrument. Although various microvalves and micropumps have been developed, many of these on-chip components require complicated microfabrication, making the microchip expensive and unlikely to be disposable.26,27 In our system, the expensive valve control parts and the pump are separated from the chip and integrated into the instrument for repeated use in order to keep the microchip structure as simple as possible. Thus, the single time use of the microchip was achieved at a low cost. In addition, to eliminate contamination and carryover, we designed a “oneway” fluid control strategy, that is, all the fluids are kept to flow from the inlets to the outlet in the chip to prevent any reflux of contaminants (verification results shown below). By using this method coupled with the simple design of the chip, the reliability of the fluidic control is significantly improved while the cost per run stays low. Third, an in-situ amplification concept was employed for the integration of DNA extraction and PCR. Currently, the majority of on-chip extractions employs a bind-wash-elute protocol, which requires the transportation of eluted DNA from the extraction structure to the PCR chamber in a timing way followed by complete mixing of DNA template with PCR reagents.12,23 Such operations often cause the dilution of extracted DNA, and demand complicated microstructures for sample mixing, as well as a delicate control that is difficult to repeat. In our system, the DNA extraction and PCR were combined together in a single DNA extraction and PCR chamber, where PCR amplification was performed directly without elution. This in-situ amplification method can dramatically simplify the structures of the integrated system. In addition, PCR reagents can be easily loaded into the chamber and thoroughly mixed with the entire DNA extract without the worry of sample loss or dilution, as well as the

Figure 4. Characterization of the on-chip microvalve. (A) The plot of the flow rate vs the driving pressure when the valve is open. (B) The plot of the flow rate vs the driving pressure when the valve is close. The valve burst pressure was measured to be 44 kPa.

was linearly increased with the increase of the pressure. In contrast, when the valve was closed by the solenoids, the flow rate stayed zero until the pressure reached 45.73 kPa, at which the flow rate was 0.1 mL/min with an error bar of ±0.496 kPa. After that, the flow rate was increased to 0.4 mL/min under a pressure of 50.03 kPa with an error bar of ±1.817 kPa. Apparently, when the pressure exceeded 45 kPa, the flow rate increased along with the pressure, indicating that the valve leakage happened under such pressure. Therefore, we concluded that the closed valve is able to sustain a pressure as high as 44 kPa. During PCR thermal cycling, when the temperature increases from the room temperature to 95 °C, the increased pressure is usually less than 33.4 kPa according to the ideal gas equation, which is far below the burst pressure of the microvlave. On-Chip PCR Amplifications. In our system, the PCR temperature control was achieved by sensing the temperature on the surface of the PCR heating block. Therefore, the temperature disparities between the PCR chamber (TC) and the heating block (TB) must be determined and remedied. Two thermocouples were employed in the calibration process: one was fixed inside of the microchip to measure the chamber 1206

DOI: 10.1021/ac5039303 Anal. Chem. 2015, 87, 1202−1209

Article

Analytical Chemistry temperature; the other was inserted into the center of the heating block. To find out the relationship between TC and TB, we recorded both temperatures by raising the block temperature TB from 50 °C up to 105 °C at a 5 °C interval. This calibration process was repeated four times on each microchip, and total four chips were tested. The relationship between these two temperatures was found completely linear (shown in Supporting Information Figure S-5), validating the feasibility of achieving an accurate temperature control through an off-chip thermocouple. During thermal cycling, the temperature ramp rates were measured as 2.53 °C/s for heating and 2.37 °C/s for cooling. Once the temperature was calibrated, we tested the performance of on-chip PCR amplifications using K562 standard genomic DNA. As shown in Figure 5A, 59-bp short fragments of β-actin gene were successfully amplified from 15ng K562 DNA in three independent experiments on the chips along with a positive and a negative control performed off-chip. Minor nonspecific amplifications or primer dimers were observed in the low molecular weight region of the gel image. The thermal cycling was finished in about 86 min. On-Chip DNA Extraction and Amplification from Blood. Human whole blood samples were employed to test the automated DNA extraction and PCR on the chip. In this experiment, 2 μL of blood was first pipetted into the loading chamber. Once the chamber was sealed with a piece of adhesive tape, the microchip was manually loaded into the instrument and the DNA extraction and PCR were started by simply clicking a button in the LabVIEW program. The whole extraction and amplification process, including ∼10 min for DNA extraction and ∼86 min for PCR, was completed automatically without any human intervention. As demonstrated in Figure 5B (lane B1, B3, B5), the 59-bp amplicons can be repeatedly obtained using a brand new microchip each time on the instrument. In addition, between the tests using blood samples, negative controls using DI water as template were conducted on new microchips. The blank results shown in Figure 5B (lane W2, W4, W6) proved the success elimination of any contamination and carryover by employing the “oneway” fluidic control method without changing the pump and tubing after each run. For the sake of consistence, the tubing and the microports were changed and the syringe was cleaned using detergent every ten runs in the study. To evaluate the sensitivity of the DNA extraction and PCR amplification, human whole blood samples with different input amounts ranging from 0.3 to 1 μL were performed on the microsystem. In parallel, 5 ng K562 DNA was amplified off-chip as a positive control. As shown in Figure 5C, even with only 0.3 μL of input blood, the 59-bp β-actin amplicons still can be successfully obtained using the microchip capillary electrophoresis, demonstrating the excellent efficiency of the on-chip DNA extraction and PCR amplification. Genetic Testing for Inherited Hearing Loss. To critically test our microsystem and to demonstrate one of its potential applications, a two-color multiplex allele-specific PCR assay for detecting c.176_191del16, c.235delC, and c.299_300delAT mutations in GJB2 gene was constructed in collaboration with CapitalBio Corporation. Human whole blood obtained from a healthy donor as a control was first processed on the microsystem. DNA extraction from 2-μL blood and multiplex allele-specific PCR were performed on the microsystem, followed by an electrophoretic analysis on a portable microchip capillary electrophoresis system. As shown in panel A of Figure 6, three TAMRA-labeled peaks, including

Figure 5. DNA extraction and PCR amplification performed on the microsystem. (A) Electropherograms of 59-bp β-actin amplicons from 15-ng standard K562 genomic DNA. (Lane M: DL 2000 DNA marker. NC: Negative control. PC: Positive control. OS1−3: PCR products obtained on the instrument from standard K562 DNA in three independent runs.) (B) On-chip analysis of 2-μL blood samples and negative control from DI water conducted alternatively on the microsystem. These results demonstrated that blood can be reproducibly analyzed and the contaminations or carryovers were eliminated completely. (Lane M: DL 2000 DNA marker. B1, B3, B5: on-chip PCR products obtained from 2-μL blood samples. W2, W4, W6: negative control performed on the system from DI water. NC: Negative control performed off-chip. P1, P2: Positive control with 10 and 20 ng K562 DNA.) (C) Sensitivity test of DNA extraction and singleplex amplification of the 59-bp DNA fragments from blood using the microsystem. Even with only 0.3 μL of input blood, the amplicons still can be obtained using the microchip capillary electrophoresis system.

109 bp at c.299_300delAT, 173 bp at c.235delC, and 221 bp at c.176_191del16, were all successfully obtained along with the sizing standards in the ROX detection channel. Figure 6B and C show that two blood samples from hearing impaired patients with homozygous mutations in c.299_300delAT and c.235delC, respectively, were accurately analyzed, showing FAM-labeled 109-bp and 172-bp peaks accordingly. After that, another three blood samples with heterozygous mutations in 1207

DOI: 10.1021/ac5039303 Anal. Chem. 2015, 87, 1202−1209

Article

Analytical Chemistry

Compared to other conventional ASPCR-based methods, such as ASPCR with microarray,2 our fully automated sample preparation microsystem coupled with the microchip capillary electrophoresis instrument has successfully simplified the entire procedure of a genetic testing to just two steps: sample preparation and CE detection. And the analytical time was reduced from 5 h to less than 2 h with minimum manual operations, including 10 min for DNA extraction, 86 min for multiplex PCR amplification, and 15 min for CE separation. Although the microsystem is currently low-throughput, it can be readily expanded by arraying the DNA extraction and PCR microstructures up to 16, which will match up with the throughput of the most widely used capillary electrophoresis system. Our study clearly proved that this sample preparation system has a great potential to be integrated into a genetic analysis for rapid and automated mutation detection.

■

CONCLUSIONS

■

ASSOCIATED CONTENT

A fully automated microsystem consisting of a low-cost, disposable plastic chip for DNA extraction and PCR amplification operated on a compact control instrument was successfully developed. DNA extraction was performed on the chip by adopting a filter paper-based method, followed by an in-situ PCR that was carried out directly in the same extraction chamber without elution. Fully automated sample preparation from human whole blood was successfully achieved due to the overcome of three critical issues: first, a loading and a sampling chamber were designed into the chip to function as a macro-tomicro and a micro-to-macro interface, respectively. Second, a low-cost fluidic control system consisting of diaphragm microvalves, a modular valve positioner, and a syringe pump for precise “one-way” fluidic manipulation was developed. Third, an in-situ PCR concept was adopted to dramatically simplify the microchip structure, thus to improve the reliability of the overall system. By coupling this sample preparation microsystem with a portable chip-based capillary electrophoresis system, DNA extractions from blood samples, twocolor multiplex allele-specific PCR amplifications, and on-chip electrophoretic analyses for detecting three GJB2 gene mutations were successfully performed in less than 2 h, demonstrating the great potential of our system for rapid, automated genetic analysis. Currently, the sample preparation and the CE detection were performed on two separated instruments, and the samples were manually transferred between these two steps. Since both systems are based on microchip formats, it is possible to simply combine them together with an additional microvalve to form a truly “sample-in−answer-out” system for clinical diagnosis, forensic human identification, food safety, etc.

Figure 6. Genetic testing of three GJB2 gene mutations for hereditary hearing loss. (A) Analysis of 2-μL whole blood sample from a healthy donor. Wild type allele peaks labeled with TAMRA were all obtained. (B and C) Analyses of two blood samples with homozygous mutations in c.299_300delAT and c.235delC, respectively. FAM-labeled 109-bp and 172-bp peaks were obtained in these two loci accordingly. (D, E and F) Analyses of three blood samples with heterozygous mutations in c.299_300delAT, c.235delC, and c.176_191del16, showing two peaks in each heterozygous locus.

S Supporting Information *

Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

c.299_300delAT, c.235delC, and c.176_191del16 were tested on the system. As shown in Figure 6D, E, and F, two peaks, one labeled with TAMRA from wild type allele and the other labeled with FAM from mutation, in each heterozygous locus were successfully obtained.

*E-mail: [email protected]. Phone: +86-10-62798732. Fax: +86-10-62798732. Author Contributions Δ

1208

B. Z. and W. G. contributed equally to this work. DOI: 10.1021/ac5039303 Anal. Chem. 2015, 87, 1202−1209

Article

Analytical Chemistry Notes

(24) Lounsbury, J. A.; Karlsson, A.; Miranian, D. C.; Cronk, S. M.; Nelson, D. A.; Li, J.; Haverstick, D. M.; Kinnon, P.; Saul, D. J.; Landers, J. P. Lab Chip 2013, 13, 1384−1393. (25) Gan, W.; Zhuang, B.; Zhang, P.; Han, J.; Li, C. X.; Liu, P. Lab Chip 2014, 14, 3719−3728. (26) Grover, W. H.; Skelley, A. M.; Liu, C. N.; Lagally, E. T.; Mathies, R. A. Sens. Actuators, B 2003, 89, 315−323. (27) Unger, M. A.; Chou, H. P.; Thorsen, T.; Scherer, A.; Quake, S. R. Science 2000, 288, 113−116.

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Prof. Jing Cheng, Prof. Wanli Xing, and Prof. Guoliang Huang at the Department of Biomedical Engineering, Tsinghua University for their valuable discussions. This research was supported by National Instrumentation Program (No. 2013YQ190467) from the Ministry of Science and Technology of China, Basic Research Project (No. 2012JB012) from the Institute of Forensic Science of China, and the research project (No. Z111100067311003) from the Beijing Municipal Science and Technology Commission.

■

REFERENCES

(1) Ballana, E.; Ventayol, M.; Rabionet, R.; Gasparini, P.; Estivill, X. Connexins and deafness homepage. http://davinci.crg.es/deafness. (2) Li, C. X.; Pan, Q.; Guo, Y. G.; Li, Y.; Gao, H. F.; Zhang, D.; Hu, H.; Xing, W. L.; Mitchelson, K.; Xia, K.; Dai, P.; Cheng, J. Human Mutat. 2008, 29, 306−314. (3) Hurd, C. M.; Cavanagh, G.; Schuh, A.; Ouwehand, W. H.; Metcalfe, P. Vox Sang. 2002, 83, 1−12. (4) Newton, C. R.; Graham, A.; Heptinstall, L. E.; Powell, S. J.; Summers, C.; Kalsheker, N.; Smith, J. C.; Markham, A. F. Nucleic Acids Res. 1989, 17, 2503−2516. (5) Lin, D.; Goldstein, J. A.; Mhatre, A. N.; Lustig, L. R.; Pfister, M.; Lalwani, A. K. Hum. Mutat. 2001, 18, 42−51. (6) Zhang, G. B.; Li, C. X.; Lu, Y.; Hu, H.; Xiang, G. X.; Liang, Z. Q.; Liao, P.; Dai, P.; Xing, W. L.; Cheng, J. Biosens. Bioelectron. 2011, 26, 4708−4714. (7) Gomez-Llorente, C.; Antunez, A.; Blanco, S.; Suarez, A.; GomezCapilla, J. A.; Farez-Vidal, M. E. Eur. J. Haematol. 2004, 72, 121−129. (8) Liu, P.; Mathies, R. A. Trends Biotechnol. 2009, 27, 572−581. (9) Kovarik, M. L.; Ornoff, D. M.; Melvin, A. T.; Dobes, N. C.; Wang, Y. L.; Dickinson, A. J.; Gach, P. C.; Shah, P. K.; Allbritton, N. L. Anal. Chem. 2013, 85, 451−472. (10) Arora, A.; Simone, G.; Salieb-Beugelaar, G. B.; Kim, J. T.; Manz, A. Anal. Chem. 2010, 82, 4830−4847. (11) Culbertson, C. T.; Mickleburgh, T. G.; Stewart-James, S. A.; Sellens, K. A.; Pressnall, M. Anal. Chem. 2014, 86, 95−118. (12) Easley, C. J.; Karlinsey, J. M.; Bienvenue, J. M.; Legendre, L. A.; Roper, M. G.; Feldman, S. H.; Hughes, M. A.; Hewlett, E. L.; Merkel, T. J.; Ferrance, J. P. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 19272− 19277. (13) Liu, P.; Li, X.; Greenspoon, S. A.; Scherer, J. R.; Mathies, R. A. Lab Chip 2011, 11, 1041−1048. (14) Kim, T. H.; Park, J.; Kim, C. J.; Cho, Y. K. Anal. Chem. 2014, 86, 3841−3848. (15) Chen, L.; Manz, A.; Day, P. J. R. Lab Chip 2007, 7, 1413−1423. (16) Chang, C.; Chang, W.; Wang, C.; Wang, J.; Mai, J. D.; Lee, G. Lab Chip 2013, 13, 1225−1242. (17) Sin, M. L. Y.; Mach, K. E.; Wong, P. K.; Liao, J. C. Expert Rev. Mol. Diagn. 2014, 14, 225−244. (18) Breadmore, M. C.; Wolfe, K. A.; Arcibal, I. G.; Leung, W. K.; Dickson, D.; Giordano, B. C.; Mary, E.; Ferrance, J. P.; Feldman, S. H.; Norris, P. M. Anal. Chem. 2003, 75, 1880−1886. (19) Cao, W.; Easley, C. J.; Ferrance, J. P.; Landers, J. P. Anal. Chem. 2006, 78, 7222−7228. (20) Kim, J.; Mauk, M.; Chen, D. F.; Qiu, X. B.; Kim, J.; Gale, B.; Bau, H. H. Analyst 2010, 135, 2408−2414. (21) Jangam, S. R.; Yamada, D. H.; McFall, S. M.; Kelso, D. M. J. Clin. Microbiol. 2009, 47, 2363−2368. (22) Shaw, K. J.; Joyce, D. A.; SDocker, P. T.; Dyer, C. E.; Greenway, G. M.; Greenman, J.; Haswell, S. J. Lab Chip 2011, 11, 443−448. (23) Hagan, K. A.; Reedy, C. R.; Bienvenue, J. M.; Dewald, A. H.; Landers, J. P. Analyst 2011, 136, 1928−1937. 1209

DOI: 10.1021/ac5039303 Anal. Chem. 2015, 87, 1202−1209