Author's Accepted Manuscript

A real-time microfluidic multiplex electrochemical loop-mediated isothermal amplification chip for differentiating bacteria Juan Luo, Xueen Fang, Daixing Ye, Huixiang Li, Hui Chen, Song Zhang, Jilie Kong

www.elsevier.com/locate/bios

PII: DOI: Reference:

S0956-5663(14)00254-1 http://dx.doi.org/10.1016/j.bios.2014.03.073 BIOS6701

To appear in:

Biosensors and Bioelectronics

Received date: 13 February 2014 Revised date: 27 March 2014 Accepted date: 31 March 2014 Cite this article as: Juan Luo, Xueen Fang, Daixing Ye, Huixiang Li, Hui Chen, Song Zhang, Jilie Kong, A real-time microfluidic multiplex electrochemical loop-mediated isothermal amplification chip for differentiating bacteria, Biosensors and Bioelectronics, http://dx.doi.org/10.1016/j.bios.2014.03.073 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A real-time microfluidic multiplex electrochemical loop-mediated isothermal amplification chip for differentiating bacteria

Juan Luo, Xueen Fang, Daixing Ye, Huixiang Li, Hui Chen, Song Zhang, Jilie Kong∗ Department of Chemistry and Institutes of Biomedical Sciences, Fudan University, Shanghai 200433, P.R. China

                                                              ∗ Corresponding author. Address: Department of Chemistry and Institutes of Biomedical Sciences, Fudan University, Shanghai 200433, PR China. Tel.: +86 21 65642138 ; fax: +86 21 65641740 E-mail address: [email protected] 1   

ABSTRACT

This report shows that loop-mediated isothermal amplification (LAMP) of nucleic acid can be integrated on a laser etched indium tin oxide (ITO) electrode-based multiplex microfluidic chip for real-time quantitative differentiation of bacteria; we call this technique microfluidic multiplex electrochemical LAMP (μME-LAMP) system. Three important acute upper respiratory tract infection (URTI) related bacteria, namely Mycobacterium tuberculosis (MTB), Haemophilus influenza (HIN), and Klebsiella pneumonia (KPN) were chosen for this study. We monitored the amplification process by measuring and analyzing the electrochemical signal of methylene blue (MB) through eight etched ITO electrochemical reactors. The results indicated that this assay with the ability of analyzing multiple genes qualitatively and quantitatively is highly specific, operationally simple, and cost/time effective. It exhibits high sensitivity with detection limits of 28, 17, and 16 copies μL-1 for MTB, HIN, and KPN, respectively. The whole differentiation can be finished in a short time of 45 minutes, which has the potential to apply in clinical diagnosis.

Keywords: Microfluidic chip; Loop mediated isothermal amplification (LAMP); Electrochemistry; Multiple; ITO electrode

2   

1. Introduction

About 78% of patients of acute upper respiratory tract infection (URTI) receive incorrect or inappropriate antibiotic therapy for the belief that antibiotics may prevent secondary bacterial infections (Kenealy and Arroll 2013). However, 80% of URTI is caused by virus, not the bacteria (Heikkinen and Jarvinen 2003). And it is known to us that antibiotics have no role in the treatment of virus induced URTI (Shefrin and Goldman 2009). More importantly, the abuse of antibiotics overwhelmingly enhances the resistance of bacteria, and more and more drug resistance bacteria appear (Furuya and Lowy 2006; Levy and Marshall 2004). Point-of-care diagnostics for the discrimination of the type of URTI would be very important for doctors to direct the appropriate treatment. Clinically diagnosis of URTI is mainly based on serological methods (Principi et al., 2010), immunoassays (Olenec et al., 2010), and various polymerase chain reactions (PCRs), which either consume unacceptable long time or demand sophisticated instruments for routine and point-of-care detection (Eccles 2005; Heikkinen and Jarvinen 2003). Therefore, effective approaches to detect the bacteria that cause URTI qualitatively and quantitatively are critical for prescribing the adequate remedy and on-demand administration. Loop mediated isothermal amplification (LAMP) is an outstanding gene amplification procedure (Tomita et al., 2008), which amplifies DNA/RNA under isothermal conditions (60-65 oC) for 1 h or less with high specificity and sensitivity using a set of six specially designed primers and Bst DNA polymerase (Mori and Notomi 2009). Without the need to accurately toggle the reaction mixture between different temperatures normally required for PCR (Gao et al., 2013), LAMP is a 3   

novel and powerful tool for nucleic acid amplification and has been widely used for bacterial detection, such as Salmonella (Hara Kudo et al., 2005), Escherichia coli (Hill et al., 2008), Yersinia pseudotuberculosis (Horisaka et al., 2004) and so forth. Although LAMP is more convenient and effective than technologies based on serological methods and PCRs, most of the methods for monitoring the LAMP process are performed on expensive and sophisticated instruments, such as the Loopamp® turbidimeter system (Tani et al., 2007; Wang et al., 2013), which is not suitable for point-of-care applications, especially in resource-limited settings. And LAMP has the limitation of testing for a single target due to the incompatibility of different primer sets in the same reaction system (Fang et al., 2010). Therefore, developing a new detection method which maintains the merits of LAMP, and realizes the multiplex gene detection in a more convenient and economical way is required. In our previous work, we have developed an octopus-like microfluidic loop-mediated isothermal amplification chip (μ-LAMP) for the point-of-care analysis of multiple genes for three human influenza A viruses and eight important swine viruses (Fang et al., 2011; Fang et al. 2010). The quantitative measurement is performed by a digital fiber optical sensor system coupled with μ-LAMP. When compared with optical detectors, electrochemical detection method usually shows much more stable results in different conditions and higher accuracy than the former. Many publications also show excellent results with the real time nucleic acid amplification based sensors, such as the LAMP based pH sensor (Toumazou et al., 2013), the RT-LAMP (Veigas et al., 2014) and RT-PCR (Branquinho et al., 2011) based electrolyte-insulator-semiconductor (EIS) sensor, and real time helicase-dependent amplification (Kivlehan et al., 2011) based electrochemical 4   

sensor. However, they are limited either due to the low sensitivity or due to the harsh conditions. Electrochemical detector combined with LAMP offers compelling advantages of its easy manipulation, high sensitivity, and low cost. More importantly, batch-production of electrode array greatly increases the reproducibility of the measurement and ensures the accuracy of multiplex analysis (McNerney and Daley 2011). So, in this work, we integrated the merits of LAMP and electrochemistry methods to develop a microfluidic multiple electrochemical system, termed as μME-LAMP, and applied it to the real-time monitoring of Mycobacterium tuberculosis (MTB), Haemophilus influenza (HIN), and Klebsiella pneumonia (KPN) qualitatively and quantitatively with high sensitivity, specificity, and rapidity.

2. Materials and methods 2.1 . Materials The concentrations of standard DNA provided by Sangon Co., Ltd., Shanghai, China were 89 ng μL-1, 56 ng μL-1, and 55 ng μL-1 for Mycobacterium tuberculosis (MTB), Haemophilus influenza (HIN), and Klebsiella pneumonia (KPN), respectively. Conserved nucleic acid fragments were screened and cloned into the PUC57 plasmid by the company (the sequences of the DNA can be seen in Supplementary Information Table S1). The bacterial culture samples of MTB, HIN, and KPN from clinical sputum specimens were provided by the Shanghai Public Health Clinical Center. The three sets LAMP primers used in this work, as described by (Iwamoto et al., 2003), (Kim et al., 2011), and (Zhang et al., 2011), were synthesized by Invitrogen Co., Ltd., Shanghai, China (Supplementary Information Table S2). Bacterial lysis reagent was purchased from Huafeng Co., 5   

Ltd., Guangzhou, China. LAMP amplification reagent (Loopamp® DtNA Amplification kit) was purchased from Eiken Co., Ltd., Japan. Methylene blue (MB) was purchased from National Medicines Corporation Ltd. Polydimethylsiloxane (PDMS) used as the material for the microchips was purchased from Dow Corning Co., Ltd. Transparent hot melt adhesive used to seal the inlet/outlet holes was purchased from Jiexin Techonolgy Co., Ltd, Shenzhen, China. Operation of these active bacteria was performed in a P2/P3 lab because of the potential biohazard concern. The DNAs of the bacteria would not cause biohazard problems and could be used in common laboratories.

2.2. Design and fabrication of μME-LAMP chip The whole μME-LAMP chip (Fig. 1A) is assembled from two modules, the etched ITO glass and PDMS microchip. The etched ITO glass is designed like the Eight-Diagram tactics (Fig. 1B). And it was etched precisely by laser on a 4-cm-width, 1-mm-thick ITO glass (Weihua solar Co., Ltd. Xiamen, China) to form an Eight-Diagram tactics pattern. Only the etching parts are conductive. The manufacturing process of the PDMS chip is described in the Supplementary Information. Inlet/outlet holes were drilled through a Harris Uni-Core Punch (Pukai Rui Biotechnology Co., Ltd. Beijing, China) with a diameter of 0.5mm on the PDMS chip. It was then irreversibly sealed with the etched ITO glass using air plasma to form a leak-proof integral chip. After that, the whole chip was heated at 121 ºC for 30 min to get rid of contaminants. Subsequently, wires were connected to the respective electrodes by conductive silver paste plus (SPI, USA) to make it convenient to link with the electrochemical workstation. And the 6   

all-purpose adhesive was used to consolidate the bond. The complete chip measured 4 cm × 4 cm, with eight microchambers; the volume of each michorchamber was approximately 20 μL.

2.3. LAMP amplification Serial dilutions (10-fold) of standard samples of MTB, HIN, and KPN ranging from 105 to 100 copies μL-1 were used as templates to evaluate the dynamics of LAMP amplification in microfluidic chips and establish standard curves for quantitative analysis. The Loopamp® reaction mixture contained the following: 20 mM Tris-HCl (pH 8.8), 10 mM KCl, 8 mM MgSO4, 10 mM (NH4)2SO4, 0.1% Tween 20, 0.8 M Betaine, 25 μM Calcein, 0.5 mM MnCl2, 1.4 mM dNTPs, 8 U Bst Polymerase, 0.2 μM each of the outer primer (F3/B3), 0.8 μM each of the loop primer (LF/LB), 1.6 μM each of the inner primer (FIP/BIP), 10μM MB and 2 μL nucleic acid sample. The total volume of the mixture is 25 μL, which is a little more than that of the microchamber (20 μL). This aims to guarantee that the chambers are completely full with the reaction mixture. The mixture was subsequently pipetted into the chambers from the inlets. The inlet/outlet was tightly sealed with the transparent hot melt adhesive to form an integral device, so as to prevent the evaporation. The adhesive is tacky at ~180 ºC and solidifies in a few seconds at room temperature. The amplification was then incubated at 63 ºC (the hot melt adhesive is stable at 63 ºC) on a ZNJR-B smart thermostat electric heating plate.

7   

2.4. Setup for real-time multiplex quantitative and qualitative analysis The whole chip was connected to the multichannel electrochemical workstation (CHI 660C, CH Instruments Co., Ltd. Shanghai, China) by the corresponding wires for electrochemical measurements. Once the reaction system reached 63 ºC, we immediately initiated the time-course electrochemical measurements to measure the redox currents every minute for 60 minutes. Square wave voltammetry (SWV) was selected for monitoring the MB redox currents during LAMP reactions due to its fast measurement time (~4s). To reduce manual inspection and save time, we developed a macro command (Supplementary Information Table S8) to store SWV test results automatically (the details of electrochemical measurements and data processing are described in Supplementary Information). The result of the quantitative detection was obtained and then confirmed by agarose gel electrophoresis. Our novel device was further validated using bacterial culture samples as qualitative analysis. The DNAs of MTB, HIN, and KPN from the bacterial culture were extracted by the bacterial lysis buffer. All microchambers were precoated with 2 μL LAMP probe sets (primers) using the flow patterning technique. Microchambers 1/1’, 2/2’, and 3/3’ were coated with MTB-probes, HIN-probes, and KPN-probes, respectively. These functionalized microchambers could recognize the specific nucleic acid fragments of bacteria in situ, permitting the rapid amplification and simultaneously triggering the LAMP signals. We also functionalized microchambers 4/4’ contained no probe as negative controls. To start the LAMP reaction, DNAs of different bacterial samples were mixed in different sets, and then mixed with LAMP reaction buffer. Dye SYBR Green I was added after the amplification, and the microfluidic chip was observed under the UV light. 8   

3. Results and discussion 3.1. Fabrication of μME-LAMP chip We constructed an ITO-PDMS hybrid microfluidic chip with eight microchambers (1/1’, 2/2’, 3/3’, and 4/4’) (Fig. 1A). The microchambers (Radius, 2.1 mm; Depth, 1.4 mm) are isolated with each other to prevent contamination and to avoid interference of electrochemical process and LAMP reaction. The inlet hole (Fig. 1A, 5) and outlet hole (Fig. 1A, 6) is necessary for the convenient addition of the DNA sample and at the same time making the capillary force available to transport the LAMP reaction mixture into the microchambers (Fang et al. 2010). Each of the microchamber contains a three-electrode system, including a working electrode, a counter electrode and a reference electrode (Fig. 1B). The eight working electrodes are isolated (Fig. 1B, 7). The reference electrode (Fig. 1B, 8) is converged in the middle as a circle, which is deposited with Ag/AgCl. The counter electrode (Fig. 1B, 9) is like a great ring, which is led out by a line so as to connect wire conveniently. The μME-LAMP chip is easy to fabricate without using any precise valves or pumps. And the electrodes were etched precisely, easily, and cheaply by laser on an ITO glass. The μME-LAMP chip is disposable since both the PDMS chip and etched ITO electrodes are easy to fabricate with high reproducibility and low cost. It also could be easily expanded to integrate as many microchambers as possible and thus provides the possibility of ultra-high-throughput analysis of genes.

9   

Fig. 1. (A) Photograph of the μME-LAMP chip. It contains eight isolated electrochemical microchambers (1/1’, 2/2’, 3/3’, and 4/4’). An inlet hole (5) and an outlet hole (6) are drilled on each microchamber. (B) Structure of the ITO electrodes. It contains eight isolated working electrodes (7), a reference electrode (8), and a counter electrode (9). (C-D) Schematic of the principle of the μME-LAMP reaction. The beginning (C) and the end (D) of the reaction. (E) The corresponding changes of electrochemical signals. (Black, 0 min) represents the beginning of the reaction; (Red, 60 min) represents the end of the reaction.

10   

3.2. Basic principle of μME-LAMP microfluidic chip The schematic of the principle of the μME-LAMP reaction is shown in Fig. 1(C-E). At the beginning of the reaction, MB molecules are free in solution and generate a measurable redox current (Fig. 1C), and a relatively large peak current is obtained through SWV method (Fig. 1E, black, 0 min). As the reaction progresses, dumbbell-shaped intermediate structures and stem-loop amplicons are produced in succession (Fig. 1D). The intercalation of MB with amplified dsDNA leads to the rapid reduction of redox currents (Fig. 1E, red, 60 min). Time-course current traces of the μME-LAMP reactions acquired from real-time monitoring provide valuable information that end-point detection cannot provide, which makes the quantitative analysis possible. Methylene blue (MB) was applied as the electrochemically active indicator for the double stranded DNA (dsDNA) (Ju et al., 1995; Liu et al., 2013). As a tricyclic heteroaromatic compound, MB could bind to dsDNA via intercalation, and be used as the indicator for the electrochemical detection (Hsieh et al., 2012; Ju et al. 1995). Other electrochemical active molecules, such as Hoechst 33258(Chang et al., 2013; Nagatani et al., 2011), metalointercalators (Storari et al., 2013), and the dye acridine orange probes (Hashimoto et al., 1994) also gained good results. However, these molecules either require laborious immobilization of the sensing layer or may inhibit the polymerase enzyme activity. In contrast, our method does not require any immobilization, and exhibits stable electrochemical performance.

11   

3.3. The fidelity, sensitivity and specificity of μME-LAMP system In order to prove the fidelity of our μME-LAMP system in the presence of MB and in the microfluidic chip, we performed standard (MB-free) LAMP and MB-LAMP reaction in PCR tubes while also performing equivalent reactions in our μME-LAMP chip. We used same target DNA concentrations (approximately 2.8×105 copies μL-1 of MTB) in all cases and compared the results using agarose gel electrophoresis (Fig. 2A). Similar distributions of bands and band intensities were observed. The high fidelity may be due to that no temperature alternation was demanded in our μME-LAMP system.

12   

RT-HDA=Real-time helicase-dependent isothermal DNA amplification; ELISA= enzyme linked immunosorbent assay

Fig. 2. (A) Fidelity of the μME-LAMP system. We successfully demonstrate that the on-chip MB-LAMP reaction (lane 1), MB-free on-chip LAMP reaction (lane 2), and a MB-LAMP in-tube reaction (lane 3) display equivalent amplification efficiency to a standard MB-free in-tube LAMP reaction (lane 4). Sensitivity evaluation using 10-fold serial dilutions of standards: (B) MTB (lanes 1-6, 2.8×105, 2.8×104,…, 2.8×100 copies μL-1); (C) HIN (lanes 1-6, 1.7×105, 1.7×104,…, 1.7×100 copies μL-1); and (D) KPN (lanes 1-6, 1.6×105, 1.6×104,…, 1.6×100 copies μL-1). (E) Comparison of the characteristic ladder-like bands of the LAMP product of the three standards (lanes 1-2, 2.8×105 copies μL-1 MTB; lanes 3-4, 1.7×105 copies μL-1 HIN; lanes 5-6, 1.6×105 copies μL-1 KPN). Response time of μME-LAMP system for the detection of (F) MTB (2.8×105 copies μL-1), (G) HIN (1.7×105 copies μL-1), and (H)KPN (1.6×105 copies μL-1). Lanes 1-4 are 5, 15, 25, and 35 min. M represents marker and N represents the negative control without target DNA. (I) Merits of μME-LAMP system compared with other detecting methods.

To investigate the sensitivity, we used 10-fold serial dilutions of standard samples for agarose gel electrophoresis (2%) followed by staining with SYBR Green I (Fig. 2, B-D). It exhibits high sensitivity with detection limits of 28, 17, and 16 copies μL-1 for MTB, HIN, and KPN, respectively. The detection limits of this assay of ~10 copies μL-1 appear to exceed that of most other methods and could be compared with RT-qPCR assays (Supplementary Information Fig. S7). The specificity 13   

of this assay was mainly dependent on the LAMP primers used. The primers in used in this study were designed and synthesized according to the references ((Iwamoto et al. 2003) for MTB, (Kim et al. 2011) for HIN, (Zhang et al. 2011) for KPN). The specificity has been evaluated by these publications. As a matter of fact, the high specificity of the LAMP system was insured by the inherent reaction mechanism, in which the positive signal occurred only when all of LAMP primers bind simultaneously with the six distinct sequences of the target gene. We found that 25 min was enough to trigger the LAMP signal for MTB, 15 min for HIN and KPN, which was relatively fast when compared with other nucleic acid amplification-based methods (Fig.2, F-H). The merits of μME-LAMP system are listed and compared with other typical methods including nucleic acid amplification methods like RT-PCR (Karami et al., 2011) and RT-HDA (Ramalingam et al., 2009), and serological methods like ELISA (Fang et al. 2010) in Fig. 2I. We believe that this novel assay has the potential in clinical diagnosis for various bacteria or genetic diseases with high sensitivity and specificity. Meanwhile, the merits of the isothermal temperature make this system suitable for multiplex detection for potential clinically predicting various bacteria at the same time. By combing the advantages of real-time electrochemical readout and LAMP with a multichamber microchip platform, our μME-LAMP method obviates the need for bulky and sophisticated detectors and temperature controls, while also ensuring multiple microfluidic chip DNA amplification by eliminating the potential for high-temperature-induced reaction failures. All the benefits above make it much more acceptable in resource-poor settings.

14   

3.4. μME-LAMP microfluidic chip for quantitative and qualitative analysis of three URTI-related bacteria Fig. 3A shows that the two real-time current traces with and without target DNA differ dramatically. No obvious peak current changes occur in the negative sample, which means electrochemical signal of MB has almost no change in the non-target DNA components. It also demonstrates that enzyme, LAMP reaction buffer, and on-chip reaction influence the concentration of free MB little. While for the positive sample, the current trace decreases in a sigmoid pattern that resembles the reaction kinetics of LAMP reactions (Heid et al., 1996; Mori et al., 2004). The derivative curves of the real-time current traces (Fig.3B) shows that the negative control reaction does not display any threshold (Fig. 3B, black), however, the target-containing reaction shows a clear threshold (Fig. 3B, red), which is the local minimum of the derivative curves. We then obtained values of “time to threshold” (TT, defined as the reaction time required for a particular sample reach sufficiently positive signals above the baseline during the real-time amplification). For this MTB sample, 19 min was determined as the TT.

15   

  Fig. 3. Electrochemical monitoring of the μME-LAMP reaction. (A) The normalized real-time current traces for a negative (no DNA) sample and a positive sample (containing MTB genomic DNA). (B) The derivative curves of the real-time current traces of the samples. All the detections were repeated for three times to enhance the reproducibility of the analysis and provide more stable test results. We applied this μME-LAMP system in quantitative assays with different initial concentrations of each standard (ten-fold serial dilutions from 105 to 101 copies μL-1) and recorded the corresponding TT values simultaneously (Fig. 4, left). The standard curves (Fig. 4, right) reveal a log-linear relationship between the initial target concentration and the TT across the entire range. We obtained standard curves with correlation coefficients of 0.9917, 0.9944, and 0.9943 for MTB, 16   

HIN, and KPN, respectively, which display a high degree of reproducibility. It also exhibits high sensitivity with detection limits of 28, 17, and 16 copies μL-1 for MTB, HIN, and KPN, respectively. In order to prove of the reliability of our methods, we compared it with the commercial Loopamp® turbidimeter system with same initial concentrations of DNA, and similar results were obtained (Supplementary Information Fig. S3 and Table S4). Although there are slight differences of TT values between our μME-LAMP system and the commercial device, our system still shows good consistency with it. We also applied this system in quantitatively detecting the bacterial culture samples and promising results were obtained (the quantitative detection using μME-LAMP system with lysed bacteria of MTB is shown in Supplementary Information Fig. S5 and Table S6). The stable real-time quantitative detection reduces the possibility of false-positive and provides more detailed information that end-point determination could not provide. The multiplex determination of several samples at the same time is also time-saving in clinical diagnosis.

17   

Fig. 4. Dynamic (left) and standard curves (right) of the (A) MTB (a-e, from 2.8×105 to 2.8×101 copies μL-1), (B) HIN (a-e, from 1.7×105 to 1.7×101 copies μL-1), and (C) KPN (a-e, from 1.6×105 to 1.6×101 copies μL-1). Values and error bars represent mean values and standard deviations from three separate measurements.

Apart from the real time monitoring of the process of the LAMP reaction, we also performed a 18   

rapid qualitative analysis of bacteria culture samples (~105 CFU mL-1) within the μME-LAMP chip (Fig. 5). With different mixtures of the three DNAs of bacterial samples, different phenomena occur on the chip. It shows that the DNAs of MTB, HIN, and KPN triggered the green fluorescence signal in chamber 1/1’, 2/2’, and 3/3’, respectively (Fig. 5, B-D). Two of the three genomic DNAs triggered the signal in different two sets of chambers, respectively (Fig. 5, E-G). And the mixture of the three bacteria triggered three sets of chambers (Fig. 5H). None of these samples could induce the green fluorescence in the negative control chamber 4/4’. The different patterns of the μME-LAMP signal presented here permitted the differentiation of different bacteria and identification of URTI-related bacteria effectively.

  Fig. 5. Qualitative analysis of the three bacteria in the μME-LAMP chip. (A) Vertical view of the diagram of the μME-LAMP chip. Microchambers 1/1’, 2/2’, and 3/3’ were coated with LAMP probe sets of MTB, HIN, and KPN, respectively; microchambers 4/4’ were negative controls. The signal pattern indicates: (B) MTB; (C) HIN; (D) KPN; (E) MTB+HIN; (F) HIN+KPN; (G) 19   

MTB+KPN; (H) MTB+HIN+KPN. The detections were repeated for three times to enhance the reproducibility.

Fig.6. Sensitivity of the three bacteria within the μME-LAMP chip. (A-C) Green fluorescence detection (microchambers 1-8 represent the initial concentration of 106, 105, 104, 103, 102, 101, 100, 0 CFU mL-1 for (A) MTB, (B) HIN, and (C) KPN, respectively). Microchambers 1-6(5) show obvious green fluorescence (microchambers appear green) while microchambers 7(6)-8 do not (microchambers appear dark). (D-F) Sensitivity of the LAMP determined by standard agarose gel electrophoresis. (Lane 1-8) Bacterial samples located at 106, 105,…, 100, 0 CFU mL-1 for (D) MTB, (E) HIN, and (F) KPN, respectively; M represents the DNA marker.

20   

We evaluated the detection limits of the bacterial samples using 10-fold series of dilutions of MTB, HIN, and KPN, each of which was subjected to triplicate tests. The detection limits can reach ~10 to 102 CFU mL-1 (Fig.6). The detection limit of ~10 CFU mL-1 observed was not very stable, showing both positive and negative results in three independent tests for the three bacteria, while the detection limit of 102 CFU mL-1 exhibit very reproducible results in triplicate tests. This phenomenon was possibly contributed to the direct application of the crude lysate of the bacteria in this assay. However, this level of detection limit is much lower than most assays, including PCR and ELISA techniques.

4. Conclusions As a step toward developing a rapid, sensitive and multiplex quantitative point-of-care detection

method

of

pathogens,

we

have

shown

that

μME-LAMP

chip,

a

microfluidic-electrochemical-real-time LAMP platform, can differentiate a variety of bacteria in very small quantities qualitatively and quantitatively, with high sensitivity and specificity in a single step. The system has the great potential to apply in diagnosing the URTIs rapidly and accurately in the future, which can effectively reduce the abuse of antibiotics. Most importantly, this technique holds much more significance than the typical RT-PCR and sophisticated optical techniques in developing point-of-care diagnostic assays to combat various epidemics or genetic diseases, especially in some resource limited countries. It opens the door for development of a portable and disposable platform for analyzing nucleic acids, which should be helpful in basic research on 21   

medicine and pharmacy, environmental hygiene, agriculture, food testing and more.

Acknowledgements We are grateful for the kind help from Prof. Wu Wenjuan from the Shanghai Public Health Clinical Center for providing the bacterial culture samples from clinical sputum specimens. We would like to acknowledge the National Natural Science Foundation of China (21335002, 21175029) and the Shanghai Leading Academic Discipline Project (B109) for financial support.

References Branquinho, R., Veigas, B., Pinto, J.V., Martins, R., Fortunato, E., Baptista, P.V., 2011. Biosensors and Bioelectronics 28, 44-49. Chang, C., Chang, W., Wang, C., Wang, J., Mai, J.D., Lee, G., 2013. Lab on a Chip 13, 1225-1242. Eccles, R., 2005. Lancet Infectious Diseases 5, 718-725. Fang, X., Chen, H., Xu, L., Jiang, X., Wu, W., Kong, J., 2011. Analytical Chemistry 83, 690-695. Fang, X., Liu, Y., Kong, J., Jiang, X., 2010. Analytical Chemistry 82, 3002-3006. Furuya, E.Y., Lowy, F.D., 2006. Nature Reviews Microbiology 4, 36-45. Gao, G., Chernock, R.D., Gay, H.A., Thorstad, W.L., Zhang, T.R., Wang, H., Ma, X.J., Luo, Y., Lewis, J.S., Wang, X., 2013. International Jounal of Cancer 132, 882-890. Hara Kudo, Y., Yoshino, M., Kojima, T., Ikedo, M., 2005. FEMS Microbiology Letters 253, 155-161. Hashimoto, K., Ito, K., Ishimori, Y., 1994. Analytical Chemistry 66, 3830-3833. Heid, C.A., Stevens, J., Livak, K.J., Williams, P.M., 1996. Genome Research 6, 986-994. Heikkinen, T., Jarvinen, A., 2003. Lancet 361, 51-59. Hill, J., Beriwal, S., Chandra, I., Paul, V.K., Kapil, A., Singh, T., Wadowsky, R.M., Singh, V., Goyal, A., Jahnukainen, T., 2008. Journal of Clinical Microbiology 46, 2800-2804. Horisaka, T., Fujita, K., Iwata, T., Nakadai, A., Okatani, A.T., Horikita, T., Taniguchi, T., Honda, E., Yokomizo, Y., Hayashidani, H., 2004. Journal of Clinical Microbiology 42, 5349-5352. Hsieh, K., Patterson, A.S., Ferguson, B.S., Plaxco, K.W., Soh, H.T., 2012. Angewandte Chemie-International Edition 51, 4896-4900. Iwamoto, T., Sonobe, T., Hayashi, K., 2003. Journal of Clinical Microbiology 41, 2616-2622. Ju, H.X., Zhou, J., Cai, C.X., Chen, H.Y., 1995. Electroanalysis 7, 1165-1170. Karami, A., Gill, P., Motamedi, M.H.K., Saghafinia, M., 2011. Journal of Global Infectious Diseases 3. 22   

DOI:10.4103/0974-777X.83538. Kenealy, T., Arroll, B., 2013. Cochrane Database of Systematic Reviews. DOI: 10.1002/14651858. Kim, D.W., Kilgore, P.E., Kim, E.J., Kim, S.A., Dang, D.A., Seki, M., 2011. Journal of Clinical Microbiology 49, 3621-3626. Kivlehan, F., Mavré, F., Talini, L., Limoges, B., Marchal, D., 2011. Analyst 136, 3635-3642. Levy, S.B., Marshall, B., 2004. Nature Medicine 10S, S122-S129. Liu, X., Sun, X., Chen, J., Huang, Y., Xie, J., Li, W., Sheng, G., Zhang, Y., Zhao, F., Lu, R., Yu, H., 2013. Scientific Reports 3. DOI: 10.1038/srep01616. McNerney, R., Daley, P., 2011. Nature Reviews Microbiology 9, 204-213. Mori, Y., Kitao, M., Tomita, N., Notomi, T., 2004. Journal of Biochemicsl and Biophysical Methods 59, 145-157. Mori, Y., Notomi, T., 2009. Journal of Infection and Chemotherapy 15, 62-69. Nagatani, N., Yamanaka, K., Saito, M., Koketsu, R., Sasaki, T., Ikuta, K., Miyahara, T., Tamiya, E., 2011. Analyst 136, 5143-5150. Olenec, J.P., Kim, W.K., Lee, W., Vang, F., Pappas, T.E., Salazar, L.E., Evans, M.D., Bork, J., Roberg, K., Lemanske Jr, R.F., 2010. Journal of Allergy and Clinical Immunology 125, 1001-1006. Principi, N., Bosis, S., Esposito, S., 2010. Emerging Infectious Diseases 16, 183. Ramalingam, N., San, T.C., Kai, T.J., Mak, M.Y.M., Gong, H., 2009. Microfluidics and nanofluidics 7, 325-336. Shefrin, A.E., Goldman, R.D., 2009. Canadian Family Physician 55, 1081-1083. Storari, M., Rohr, R., Pertot, I., Gessler, C., Broggini, G.A., 2013. Journal of Applied Microbiology 114, 1193-1200. Tani, H., Teramura, T., Adachi, K., Tsuneda, S., Kurata, S., Nakamura, K., Kanagawa, T., Noda, N., 2007. Analytical Chemistry 79, 5608-5613. Tomita, N., Mori, Y., Kanda, H., Notomi, T., 2008. Nature Protocols 3, 877-882. Toumazou, C., Shepherd, L.M., Reed, S.C., Chen, G.I., Patel, A., Garner, D.M., Wang, C.A., Ou, C., Amin-Desai, K., Athanasiou, P., 2013. Nature methods 10, 641-646. Veigas, B., Branquinho, R., Pinto, J.V., Wojcik, P.J., Martins, R., Fortunato, E., Baptista, P.V., 2014. Biosensors and Bioelectronics 52, 50-55. Wang, K., Chen, Y., Lei, Y., Zhong, G., Liu, A., Zheng, Y., Sun, Z., Lin, X., Chen, Y., 2013. Microchimica Acta 180, 871-878. Zhang, H., Zhang, X., Hou, L., Zhen, W., 2011. Acta Scientiarum Naturalium Universitaties Nankaiensis 44, 38-43 (In Chinese).

23   

Highlights: 1. We create an eight-chamber microfluidic chip for analyzing multiple genes. 2. It is based on an ITO-electrode device for quantitative analysis. 3. High sensitivity and specificity is obtained. 4. The device can be applied to the differentiation of different bacteria. 5. Our device is much cheaper than the commercial equipment.  

24   

A real-time microfluidic multiplex electrochemical loop-mediated isothermal amplification chip for differentiating bacteria.

This report shows that loop-mediated isothermal amplification (LAMP) of nucleic acid can be integrated on a laser etched indium tin oxide (ITO) electr...
2MB Sizes 0 Downloads 3 Views