Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809 – 818
Original Article
nanomedjournal.com
An integrated microfluidic device utilizing vancomycin conjugated magnetic beads and nanogold-labeled specific nucleotide probes for rapid pathogen diagnosis Chih-Hung Wang, PhD a , Chia-Jung Chang a , Jiunn-Jong Wu, PhD b , Gwo-Bin Lee, PhD a, c, d,⁎ a Department of Power Mechanical Engineering, National Tsing Hua University, Hsinchu, Taiwan Department of Medical Laboratory Science and Biotechnology, National Cheng Kung University, Tainan, Taiwan c Institute of Biomedical Engineering, National Tsing Hua University, Hsinchu, Taiwan d Institute of Nano Engineering and Microsystems, National Tsing Hua University, Hsinchu, Taiwan Received 30 May 2013; accepted 31 October 2013
b
Abstract A PCR-free assay for rapid pathogen diagnosis was implemented on an integrated microfluidic system in this study. Vancomycin-conjugated magnetic beads were used to capture multiple strains of bacteria and nucleotide probes labeled gold nanoparticles were used to specify and detect a specific strain by hybridization-induced color change. The assay was entirely automated within an integrated microfluidic device that was composed of suction-type micropumps, microvalves, microchannels, and microchambers that fabricated by microfluidic technology. Multiple strains of bacteria could be captured simultaneously by vancomycin-conjugated magnetic beads, with capturing efficiency exceeding 80%. Subsequently, sensitive and strain-specific detection against target bacteria could be achieved by using nanogold labeled specific nucleotide probes. The limit of detection of 10 2 CFU bacteria was achieved. Importantly, nucleic acid amplification was not involved in the diagnostic procedures; the entire analytic process required only 25 min. The developed platform may provide a promising tool for rapid diagnosis of bacterial infections. From the Clinical Editor: In this novel study, a PCR-free pathogen detection method is demonstrated. After vancomycin-conjugated magnetic beads captured bacteria, nucleotide probes-labeled gold nanoparticles were employed to specify and detect specific strains via hybridization-induced color change. Multiple strains of bacteria could be captured simultaneously with an efficiency exceeding 80%, enabling the detection of as low as 10 2 CFU of bacteria. © 2014 Elsevier Inc. All rights reserved. Key words: Vancomycin; Nanogold; Microfluidic; Bacteria
Abbreviations: Bp, base-pair; CFU, colony forming unit; DNA, deoxyribonucleic acid; dNTP, deoxyribonucleotide triphosphate; E. coli, Escherichia coli; EDAC, (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide; EMVs, electromagnetic valves; ddH2O, deionic distilled water; G(−), Gram-negative bacteria; G(+), Gram-positive bacteria; HCl, hydrogen chloride; LB, LuriaBetani; LOC, lab-on-a-chip; LOD, limit of detection; MEMS, micro-electromechanical-systems; MRSA, methicillin-resistant Staphylococcus aureus; PCR, polymerase chain reaction; PDMS, polydimethylsiloxane; SDS, sodium dodecyl sulfate; TBE, Tris-Bornate-EDTA; Tween 20, (polyoxyethylene (20) sorbitan monolaurate; Xcv, Xanthomonas campestris pv. vesicatoria. The preliminary results from this paper were presented at the 2013 IEEE Micro Electro Mechanical Systems (IEEE MEMS 2013), Taipei, Taiwan, 2013. The authors would like to thank the National Science Council in Taiwan (NSC 101-2120-M-007-014) and the “Toward a World-class University” Project for financial support of this study. There is no any conflict of interest. ⁎Corresponding author at: Department of Power Mechanical Engineering, National Tsing Hua University, Hsinchu 300, Taiwan. E-mail addresses: [email protected] (C.-H. Wang), [email protected] (C.-J. Chang), [email protected] (J.-J. Wu), [email protected] (G.-B. Lee).
Infectious diseases have become a growing threat to the public health worldwide. 1 The emerging bacterial diseases and antibiotic-resistance bacteria are crucial issues since the mortality of the bacterial infections has gradually increased. 2 Conventionally, these bacteria can be detected and identified through pure culture of bacterium and several existing biological assays. 3,4 However, bacterial growth in cell culture may take several days to even weeks until a sufficient amount of bacteria can be obtained for biological and chemical analysis. In addition, excessive experimental procedures can easily cause artificial errors, leading to false diagnostic results. Therefore, rapid and accurate diagnosis plays an important role for infectious disease prevention and control. Recently, protein-based immunoassays (e.g. enzyme-linked immunosorbent assay) and nucleic acid amplification based methods (e.g. polymerase chain reaction (PCR), real-time PCR) have been widely applied for pathogen
1549-9634/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nano.2013.10.013 Please cite this article as: Wang C-H, et al, An integrated microfluidic device utilizing vancomycin conjugated magnetic beads and nanogold-labeled specific nucleotide probes for rapid pathogen diagnosis. Nanomedicine: NBM 2014;10:809-818, http://dx.doi.org/10.1016/j.nano.2013.10.013
810
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818 5
detection. These molecular methods can achieve highly sensitive and specific diagnostic results. However, these methods often require expensive reagents and equipments and welltrained operators for performing relatively complicated assays. In particular, nucleic acid amplification based techniques require extracted nucleic acid products from bacterial by performing sample pretreatment for differentiating pathogen DNA. 6–8 In other words, it was a labour-intensive, time-consuming and costly pretreatment in bacteria detection. Vancomycin can be used as a capturing agent for bacteria because it can bind to the peptidoglycans on the bacteria cell wall for both Gram-positive and -negative bacteria. 9–11 Specifically, Gram-negative bacteria were considered to contain a thinner peptidoglycan layer than Gram-positive bacteria. 12 Thus these antibiotics could reversibly bind onto D-alanyl-D-alanine, terminal residues of mucopeptides on the cell wall of the Gram-positive bacteria since specific hydrogen bonds could be formed between the alanine methyl group of the cell wall and vancomycin. 13,14 In addition to Gram-positive bacteria, the vancomycin also had affinity to be bonded with L-lysine-D-alanine residues of peptidoglycans that was expressed on the outer membrane of Gram-negative bacteria. 15 Therefore, vancomycin could be used to capture various species of bacterial pathogens if they may be properly surface-conjugated on the magnetic beads. A thiol chemically modified oligonucleotide linked with gold nanoparticles was reported for DNA selectivity detection. 16,17 The method was used for pathogen identification that provided simply detection for specific target DNA in the mixture of oligonucleotides by colorimetric changed observation. 18,19 The optical properties relied on the light absorption that depended on the distance of gold nanoparticles. When compared with the aggregated nanogold particles, the nanogold particle-target DNA complex had different distance in the tested dispersion that could cause color change by light spectra shift. The distinct optical properties of gold nanoparticles were demonstrated as sensitive and specific DNA biosensors that caused by red-shift effect in a gold nanoparticle aggregation process. 20,21 Nanogold was relatively biocompatible and enable facile surface immobilization chemistries for ready conjugation of oligosaccharides, nucleic acids, peptides or small biofunctional molecules. 22,23 In this study, an integrated microfluidic device for performing the new protocol for rapid pathogen diagnosis was developed. The diagnostic system consisted several functional components in the developed devices, including a recognition element for binding pathogens by using vancomycin-conjugated magnetic beads and a sensing element by using gold nanoparticle labeled with specific nucleotide probes binding to target genes. Advantages of the micro-total-analysis-systems (μ-TAS) or lab-on-a-chip (LOC) platform include reduced consumption of samples and reagents, fast analysis speed, and integrated analytic and sensing techniques without the need of skilled operators. Micro-electro-mechanical-systems (MEMS)-based biomedical systems have recently been widely applied for molecular diagnosis. 24 For instance, an integrated microfluidic system performing sample pretreatment and rapid diagnosis of Staphylococcus aureus was reported. 25 Furthermore, suction-type micropumps, microchannels, microchambers and microvalves fabricated by polydimethylsiloxane (PDMS)-based processes
have been developed to realize on-chip fluidic transportation and sample mixing, which greatly simplifies experimental processes and reduces the size of microfluidic chips. 24 Regulated by electromagnetic valves (EMV), fluid sample could be efficiently transported forward and backward to realize reagent exchange and incubation processes. In this study, these microfluidic devices were integrated on a single chip to automate the entire process for fast diagnosis of bacterial infections with minimal human intervention by using the new protocol described above.
Methods Tested bacteria and primers Gram-positive (G(+)) bacteria, including Streptococcus agalactiae, Streptococcus pyogenes, methicillin-resistant S. aureus (MRSA), and Bacillus subtilis and Gram-negative (G(−)) bacteria, including Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa were provided from National Cheng Kung University Hospital, Tainan, Taiwan. Xanthomonas campestris pv. vesicatoria, a G(−) plant pathogen, was provided by Dr. HuiLiang Wang, from Department of Biotechnology, National Kaohsiung Normal University, Kaohsiung, Taiwan. All of the tested bacteria were freshly incubated in LB broth (Sigma-Aldrich Co. USA) at 37 °C for 16 hours. The sequences of primers for the multiplex PCR were selected from nucleotide database of NCBI website (National Center for Biotechnology Information, http:// www.ncbi.nlm.nih.gov/nucleotide), designed by FastPCR software (ver 6.1, PrimerDigital Co. USA), and listed in Supporting Information Table 1. Chip design and fabrication A schematic layout of the integrated microfluidic device was shown in Figure 1. The device composed of suction-type micropumps, microvalves, microchannels, and loading chambers, is capable of performing eight independent diagnoses simultaneously. The complete microfluidic chip functions as a “cartridge” that integrates seamlessly in a custom-designed microfluidic machine that includes a temperature controller, electromagnetic valves, vacuum pumps, and waste chambers. 26 Samples and other reagents were loaded into respective chambers before testing. The micropumps could deliver all processing buffer and nanogold probes into the sample loading chambers to perform incubation, hybridization and nanogold aggregation processes in an automatic format. The device was fabricated by polydimethyl siloxane (PDMS) molding and bonding processes as previously reported. 26,27 Briefly, PDMS layers were made with PDMS polymer and curing agent (Sylgard 184A/B, Sil-More Industrial Ltd., USA) at a weight ratio of 10:1. The two PDMS layers and a glass substrate were treated with oxygen plasma, aligned, and bonded together to form the integrated microfluidic chip. Method of vancomycin-conjugated with magnetic beads The protocol for coating vancomycin on magnetic beads was modified from a previous study. 28 Briefly, 100 μL of carboxylic modified magnetic beads (Dynabeads® MyOne TM carboxylic
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
811
Figure 1. Overview of the microfluidic chip. (A) Exploded schematic diagram of the integrated microfluidic chip, which was composed of a thick-film PDMS layer, a thin-film PDMS layer, and a glass substrate. (B) Schematic layout of the integrated microfluidic device capable of carrying out four parallel diagnosis processes simultaneously. Micropumps, microvalves, washing buffer chambers, nanogold chambers, HCl chambers, sample loading chambers, and waste outlet were fully integrated. (C) Photograph of the integrated microfluidic chip. The device was measured to be 59 mm (length) × 75 mm (width).
812
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
acid, Invitrogen Dynal AS, Oslo, Norway) were dispersed by a vortex mixer and aliquotted into a micro-centrifuge tube. Magnetic beads were washed twice with 1000-μL ddH2O and collected by a magnetic particle concentrator (DynaMag TM-2, Invitrogen Dynal AS, Oslo, Norway). For testing the optimal concentration of vancomycin, 950-μL ddH2O and 30-μL vancomycin (Sigma-Aldrich Co. USA) at various concentrations were added and vortexed for 5 seconds. Subsequently, 20-μL of 120 mg/ml EDAC (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide, Sigma-Aldrich Co. USA) was added to the mixture and vortexed for 5 seconds. The mixture was incubated on a wheeling rotator at 20 rpm at room temperature in the dark for 18 hours. After this incubation, the mixture was washed twice by using 1000-μL of 0.02% Tween 20 solution (polyoxyethylene (20) sorbitan monolaurate, Sigma-Aldrich Co. USA) and collected by the magnetic particle concentrator. Next, 1000-μL of 0.1% SDS (sodium dodecyl sulfate, Sigma-Aldrich Co. USA) was used to further wash the vancomycin-bead complexes. Finally, 1000-μL of 0.1 M ethanol amine (Sigma-Aldrich Co. USA) was used to block the free uncoupling site on magnetic beads at room temperature for 1 hour. The beads were finally washed by using 1000-μL ddH2O, collected by the magnetic particle concentrator, re-suspended in 1 × PBS, and stored at 4 °C until use. Bacterial binding efficiency testing, microscopic observation, and multiplex PCR setting Three methods – bacterial binding efficiency test, microscopic observation, and multiplex PCR – against various types of bacteria were used to confirm the capturing ability and specificity of vancomycin-conjugated magnetic beads in this study. The capturing efficiency of vancomycin-conjugated magnetic beads was determined by counting the colony number of vancomycin-captured bacteria. Specifically, 1 × 10 5 magnetic beads surface-conjugated with 100 nM vancomycin were incubated with 1 × 10 5 CFU (colony forming unit) of various types of bacteria and incubated at 20 rpm at room temperature for 10 min. Magnetic beads and captured bacteria were collected by a magnetic particle concentrator and washed twice by ddH2O. Then, the collected pellet was plated on LB plates and incubated at 37 °C for 16 hours. In parallel, bacteria from the original sample were also plated to serve as a comparison. Colonies of grown bacteria in both cases were counted and the capture rate of vancomycin-conjugated magnetic beads was calculated as followed. capture rate ð%Þ total number of collected bacteria 100% ð1Þ ¼ original number of tested bateria 105 CFU In this study, the mean ± SD was used to indicate the capture rate of vancomycin conjugated magnetic beads for pathogen detection. Each data in the table was determined from three independent experiments. Each data was calculated by one-way ANOVA assay for significant difference determination. The capturing specificity of vancomycin-conjugated magnetic beads was directly observed by microscopy. Following a
previously mentioned protocol, 10 3 of vancomycin-conjugated magnetic beads was mixed with various types of bacteria and collected by a magnetic particle concentrator. Then, each collected pellet was dropped on a microscope slide and observed under microscope at 500 × magnification. Following the protocol, the total number of captured bacteria was collected for multiplex PCR tests. Total DNA of vancomycin-bead-captured bacteria was extracted by boiling at 95 °C for 10 min and washed twice by ddH2O. The 10-μL of ddH2O was added in treated vancomycin-conjugated beadsbacteria complex. Extracted bacterial DNA was added to 20-μL PCR mix which contained 1.5-μL of deoxyribonucleotide triphosphate (dNTP, 10 mM, Promega, USA), 3-μL of 10 × PCR buffer (20 mM Tris–HCl, pH 8.0, 100 mM KCl, 0.1 mM 2-[2-(Bis (carboxymethyl) amino) ethyl-(carboxymethyl) amino] acetic acid (EDTA), 1 mM Dithiothreitol (DTT), 0.5% Tween, 0.5% Nonidet and 50% (v/v) glycerol, JMR Holdings, UK), 1-μL of mecA specific primer pairs (0.5-μL of each primer for the forward/reverse primers), 1-μL of Superthermo Gold Taq DNA polymerase (5 U/μL, JMR Holdings, UK), 1-μL of 10 mM MgCl2 and 12.5-μL of ddH2O. The thermocycling process for PCR was performed under the following conditions: 95 °C for 5 min (initial denaturation), 35 cycles of 95 °C for 30 seconds (denaturation), 58 °C for 30 seconds (annealing), and 72 °C for 30 seconds (extension), and finally 72 °C for 10 min. Amplified PCR products were confirmed by 2% TBE (Tris-BornateEDTA)-agarose electrophoresis. Preparation and detection of gold nanoparticles labeled with specific nucleotide probes Gold nanoparticles labeled with specific nucleotide probes were prepared by linking thiol-modified oligonucleotide probes to gold nanoparticles via gold-thiol bond. The synthesis of these probes could be referred to a previous study. 29 First, thiolmodified oligonucleotide probe that was specific to E. coli lamB was dissolved to a final concentration of 100 μM in 100 mM dithiothreitol (DTT)/sodium phosphate buffer (pH7.0, SigmaAldrich Co. USA). The removal of DTT was essential for the gold nanoparticle labeled with E. coli lamB oligonucleotide probe. The NAP-10 column (GE Healthcare, UK) was used for DTT elimination, which equilibrated 15 mL of 50 mM sodium phosphate buffer (pH 7.0, Sigma-Aldrich Co., USA). The oligonucleotide sample was eluted with 1000-μL sodium phosphate buffer from the NAP-10 column. Four nmol of thiol modified oligonucleotide was incubated overnight with 1000-μL gold nanoparticles (20 nm, Sigma-Aldrich Co. USA) in a wheeling rotator at 20 rpm at room temperature in the dark. After the overnight incubation, 9 mM phosphate buffer (pH 7.0) and 0.1% of sodium dodecyl sulphate (SDS) were added. The mix was then shaken on an orbital shaker for 30 min. The salting solution was added at various times to a final concentration of 0.3 M NaCl. (2 M NaCl in 10 mM PBS pH7) and equilibrated overnight at room temperature. The pellet was precipitated at 13000 g for 20 min and washed with 500-μL resuspension buffer (10 mM PBS, pH 7.4, 150 mM NaCl, 0.1% SDS) and resuspended in 50-μL of the same buffer. The gold nanoparticles labeled with specific nucleotide probes were stored in light-tight
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
813
Figure 2. Schematic diagram of the experimental procedure in the developed microfluidic chip. (A) Tested bacteria were incubated with vancomycin-conjugated magnetic beads. (B) Bacteria were captured by beads and collected by external magnets. The chip was heated to 95 °C to thermally lyse captured bacteria and denature their genomic DNA. (C) Strain-specific nanogold probes were injected and allowed to hybridize with denatured bacterial genomic DNA. (D) Finally, 0.01 N HCl was added to induce color change in the sample.
containers at room temperature until use. For optical detection with nanogold labeled with the E. coli lamB probe, the vancomycin bead-captured complexes were washed twice by ddH2O and boiling at 95 °C for 5 min for lysing bacteria and denaturing genomic DNA. Next, 5-μL of 10 nM nanogold probes was added and hybridized with the denatured genomic DNA at 65 °C for 10 min. The color exchange of reactant was applied to optical observation under final 0.01 N HCl treatment. Results Characterization of microfluidic devices The developed integrated microfluidic device was used for rapid bacterial diagnosis. The working principle was schematically shown in Figure 2. Target bacteria in culture medium were first incubated and captured by vancomycin-conjugated magnetic beads and subsequently collected onto the bottom of the microchamber by a permanent magnet. Unwanted substances and debris in the sample were then washed twice by deionic distilled water (ddH2O). After washing, the chip was heated at 95 °C for 10 min to thermally lyse captured bacteria and denature the target bacterial DNA. Next, nanogold labeled with specific nucleotide probes were added and allowed to hybridize with the complementary region of denatured target DNA at 58 °C for 10 min. Finally, 0.01 N hydrogen chloride was added to the mix. After a brief incubation (approximately 5 min) at ambient temperature, nanogold probes that were hybridized to the targeted gene remained dispersed and thus appeared dark-red
in color. On the other hand, unhybridized gold nanoparticles were aggregated under the acidic condition and thus presented a purple-gray color. The color differences could be clearly differentiated by the naked-eye. As a result, the diagnostic results were determined directly by visual observation. Alternatively, the color change could be quantified by a spectrophotometer (NanoPhotometer Pearl, Implen GmBH, München, Germany). From the addition of reagents and samples to the visual detection of results, the entire analytic process required only 25 min and was completely realized within the integrated chip in an automated fashion. An exploded view of the integrated microfluidic chip, which consisted of two PDMS layers and a glass substrate, is shown in Figure 1, A. The PDMS reservoir layer harbored microchambers for storing samples and required reagents. The PDMS transport layer was constituted with micropumps, microvalves and microchannels for the transportation and processing of samples and reagents. Following oxygen plasma treatments and manual alignment, the PDMS layers and the glass substrate were bonded to form the completed chip. The width of the air channel and the fluidic channel were 350 and 200 μm, respectively. The radius of microchambers was 2000 and 2250 μm, respectively. The related position and detailed size of modules were shown in Supporting Information Figure 1. Figure 1, B depicts the schematic diagram of the integrated microfluidic chip, which possesses eight identical sets of analytic unit for parallel analyses. Each analytic unit is composed of four microchambers for storing the sample, the washing buffer, the nanogold probes, and the hydrochloric acid solution as well as a
814
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
micropump, microchannels, and microvalves for transporting the sample and reagents between chambers. Fluids are transported through microvalves by the negative gauge pressure (suction force) filling up the cavities to deform the PDMS membranes. The relationship between the pumping rate of the micropump and the driving frequency at different operating air pressures is shown in Supporting Information Figure 2. The results revealed that the pumping rate increases as the driving frequency increases and then drops as the driving frequency reaches an optimal value. A flow rate as high as 50 μL/s was achieved (at a driving frequency of 8 Hz) at an air pressure of − 60 kPa. In addition, the floating-block structure of the normally-closed microvalve capable of increasing the flow pumping rate has also been demonstrated to block liquids in the microchannel successfully. 30 Hence, the micropump and the microvalve are capable of transporting samples and reagents or blocking the fluids inside the microchannel. The procedure of the entire assay in the microfluidic device was first demonstrated with dyed solutions and shown in the Supporting Information Video. Prior to the assaying steps, the test sample and vancomycin-conjugated beads (represented as “brown” color solution), the washing buffer (transparent solution), nanogold probes (red color), and 0.01 N hydrochloric acid solution (blue color) were loaded in their respective chambers. When the incubation step for the test sample and vancomycin-conjugated beads was finished and the magnetic bead-complex was collected by applying an external magnetic field, the wash buffer was transported into the sample chamber to remove the unbound waste. Note that all of wastes in the analytic processes were removed from the waste chamber by activating the suction-type micropump under negative gauge pressure. After the cell lysis process, nanogold probes were transported into the sample chamber and then incubated with target bacterial DNA. Similarly, the waster was transported into the waste chamber afterwards. Finally, hydrochloric acid was added into the sample chamber to induce color change. The simulated assay indicated that the eight samples could all work in the microfluidic chip in parallel. The photograph of the microfluidic device is shown in Figure 1, C. The dimensions of the integrated microfluidic device were measured to be 59 mm (length) × 75 mm (width) × 1.25 mm (height). Optimization of bacterial capturing by using vancomycin-conjugated magnetic beads The optimal condition of the bacterial capturing assay was determined by adjusting the vancomycin concentration, the number of magnetic beads, and the number of bacteria. First, different concentrations of vancomycin were conjugated to magnetic beads and subsequently incubated with 1 × 10 8 CFU of methicillin-resistant S. aureus (MRSA). To determine the capture rate of vancomycin-conjugated beads, the mecA gene of captured MRSA bacteria was further amplified by a PCR protocol. 24 Supporting Information Table 2 shows that, at vancomycin concentrations of 100 mM, 100 μM, 100 nM and 100 pM, the capture rates were 20.71, 23.73, 23.72 and 11.75%, respectively, with 1 × 10 8 CFU of MRSA. The results demonstrated that the capacity of vancomycin-conjugated beads was
Table 1 The capture rate of vancomycin-conjugated magnetic beads at different bacteria concentrations. Concentration of tested bacteria⁎ (CFU)
Capture rate †(%)
10 7 10 6 10 5 10 4 10 3
23.94 34.27 66.07 92.96 91.82
± ± ± ± ±
3.98 14.91 3.57 2.09 2.35
⁎ 1 × 10 5 magnetic beads surface-conjugated with 100 nM vancomycin and serially diluted methicillin-resistant S. aureus (MRSA) were used in this experiment. † All of results were measured by three independent experiments.
saturated in this testing. Excessive free MRSA may lead to the low capture rate. Furthermore, the capture rate against different number of bacteria was determined and listed in Table 1. Using 100-nM vancomycin-conjugated beads and ten-fold serially diluted MRSA, the capture rate increased as the number of bacteria decreased; the capture rate reached as high as 91.02 and 92.96% for 1 × 10 3 and 1 × 10 4 CFU MRSA, respectively. The results revealed that the capacity of capture rate for 100 nM vancomycin-conjugated beads was saturated at 10 3 CFU of MRSA. Moreover, the limit of detection (LOD) of MRSA that was determined by electrophoretic photograph has similar results with the capturing testing (Figure 3, A). It indicated the highest capture rate of vancomycin-conjugated beads for 10 4 CFU of bacteria. However, as the number of bacteria increased, the capture rate of vancomycin-conjugated beads was reduced. Note that according to the equation of capture rate for vancomycinconjugated beads, the capture rate will decrease with excessive bacteria. In addition, the electrophoretic photograph in Figure 3, B reveals that no amplified product was observed by uncoated magnetic beads and 100 pM vancomycin-conjugated beads (lanes 1 and 5). On the other hand, a 267-bp PCR product band appeared in samples with 100 mM, 100 μM, and 100 nM vancomycin-conjugated beads (lanes 2, 3 and 4), indicating successful capture of MRSA at these vancomycin concentrations. In order to minimize the use of antibiotics, the lowest concentration of vancomycin, 100 nM was selected for bacteria capturing in this study. For exploring optimal number of beads, 10 4, 10 5 and 10 6 vancomycin-conjugated beads were used for the capturing assay and the results were shown in Supporting Information Table 3. There were no significant differences in the capture rate among the three numbers of vancomycin-conjugated beads. Summarizing the experimental results above, the optimal assay condition was achieved with 100-nM vancomycin-conjugated 1 × 10 4 beads for the bacteria capturing test. Specificity testing for vancomycin-conjugated magnetic beads Vancomycin-conjugated magnetic beads proposed in the present study were capable of capturing various species of bacteria, thus facilitating rapid detection. To demonstrate this capability, both Gram-positive and Gram-negative bacteria from human and agriculture origins were tested with the vancomycin-
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
815
Table 2 The capture rate of vancomycin-conjugated magnetic beads for different types of bacteria. Tested bacteria⁎
Capture rate (%) †
S. pyogenes B. subtilis E. coli K. pneumonia P. aeruginosa
93.74 83.35 91.01 90.26 65.35
± ± ± ± ±
2.26 9.37 1.45 1.25 2.08
⁎ Bacteria sample with a concentration of 1 × 10 4 CFU and 1 × 10 5 magnetic beads surface-conjugated with 100 nM vancomycin was used in this experiment. † All of results were measured by three independent experiments.
Figure 3. Characterization of vancomycin-conjugated magnetic beads in bacteria capture via PCR and gel electrophoresis. (A) The PCR assay for testing the limit of detection by vancomycin-conjugated magnetic beads. 100 nM of vancomycin-conjugated magnetic beads was used in the capturing assay. The 267-bp product was indicated as amplified products of MRSA mecA gene. Lane NC is the negative control with ddH2O; Lanes 1 through 6 are 10 7, 10 6, 10 5, 10 4, 10 3 and 10 2 CFU of MRSA, respectively. (B) The optimal concentration of vancomycin for conjugating on magnetic beads was determined by capturing 10 3 CFU of MRSA and performing PCR. The 267-bp product bands on the gel image were the appropriate amplified products of MRSA mecA gene. In this and subsequent gel images, lane L represents 100-bp DNA ladder. Lane NC is the negative control with ddH2O; lane 1 indicates bare magnetic beads; lanes 2 3, 4, and 5 indicate magnetic beads surface-conjugated with 10 mM, 100 μM, 100 nM and 100 pM of vancomycin, respectively.
based capturing method. Specifically, 1 × 10 4 CFU Grampositive (S. pyogenes, S. agalactiae, and MRSA) or Gramnegative (E. coli, K. pneumonia, and Xanthomonas campestris pv. vesicatoria, Xcv) bacteria were used as targets and incubated with 1 × 10 4 magnetic beads conjugated with 100 nM vancomycin. After incubation, bacteria-bead complexes were washed twice and observed under AV microscope at 500 × magnification (Supporting Information Figure 3). Microscopic observation showed that bacteria of all tested strains (transparent in color in the micrographs) indeed adhered to the surface of vancomycin-conjugated beads (red-brown in color in the micrographs) — a direct evidence for the general capture capability of vancomycin-conjugated beads against various strains of bacteria. Furthermore, capture rates of vancomycinconjugated beads against these species of bacteria were
accurately measured by bacterial counting (Table 2). After capturing and washing steps, collected bacteria-bead complexes were plated on Luria-Betani (LB) plates for 16 hours at 37 °C. The capture rate was then calculated as the ratio between the colony number formed on the plate and the original number of tested bacteria. Overall, the capture rate for most of tested bacteria was greater than 80%. Strains of E. coli, K. pneumonia and S. pyogenes even exceeded 90%. One strain of Gramnegative bacteria (P. aeruginosa), however, was captured at only 65%. Since the composition of bacterial cell wall varies from species to species, it was speculated that the thinner peptidoglycan layer – and thus less bond peptide components with vancomycin – on the cell wall of this Gram-negative bacteria may have led to its reduced capture rate. As additional verification for the capturing process, a multiplex PCR assay was used to amplify specific genes of captured bacteria and confirmed by slab-gel electrophoresis (Supporting Information Figure 4). Electrophoretic results validated that all of the five species of bacteria that were tested could be captured by vancomycin-conjugated magnetic beads and simultaneously amplified by the multiplex PCR method. Not only could these results support that vancomycin-based bacteria capture method could capture a broad range of bacteria in parallel, but it also demonstrated that it could be coupled with different detection methods to achieve the detection of multiple species of bacteria at one time. Optimal condition testing of specificity for nanogold labeled with E. coli specific nucleotide probe As a complement to the general bacteria capturing capability of the vancomycin-based capture method, a strain-specific detection method based on nanogold labeled with specific nucleotide probes was developed in the present work. Using E. coli as the sample target, a nucleotide probe that can specifically hybridize to the lamB gene of E. coli was designed. In this study, all nanogold colorimetric tests were processed in the microfluidic system. All tested results were then removed to new chamber wells of a PDMS layer with a white background to enhance the color contrast for better observation. The optimal hybridization temperature between nanogold probes and genomic DNA of E. coli was first tested; the results are shown in Supporting Information Figure 5. The results revealed that, under temperatures between 50 °C and 65 °C and against 10 2 to10 4
816
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
After the bacteria capturing process by vancomycin-conjugated beads, the specificity of lamB nanogold probe was tested against three strains of E. coli (Topo10, ER2738, 119) and other non-E. coli strains (MRSA, K. pneumoniae) and the results are shown in Figure 4, A. The dark-red color was presented for all of the three tested E. coli strains, demonstrating that the nanogold probe can detect multiple strains of E. coli. Furthermore, the negative control, MRSA, and K. pneumoniae samples yielded the blue-gray color that indicated as no probe-target hybridization, thus verifying the specificity of the lamB nanogold probe. Importantly, this nanogold probe-based detection obviates any nucleic acid amplification (i.e. PCR-free), which significantly improves the assay speed. Furthermore, the results could be simply examined by naked-eye, precluding expensive optical equipment for detection. Therefore, the nanogold probe-based detection method may present as a powerful tool for simple and rapid diagnosis of pathogens. Limit of detection for nanogold labeled with E. coli specific nucleotide probe Figure 4, B shows the LOD by utilizing the E. coli lamB nanogold-labeled probe. After nanogold probe hybridization, the negative control and samples containing 1 × 10 8 CFU of non-E. coli strains (MRSA, B. subtilis and K. pneumoniae) appear grayblue in color. However, samples containing two strains of E. coli (ER2738, 119) turned dark-red color. Here, the LOD of E. coli diagnosis was determined by the optical intensity of red color that compared with the negative control after nanogold probe hybridization. The intensity of the red color decreased along with decreasing dose of E. coli ranging from 10 8 to 10 2 CFU/ reaction. According to the color change, the limit of detection using the nanogold probe method was measured to be approximately 10 2-10 3 CFU of bacteria. Note that there is no significant difference in terms of LOD when comparing the nanogold-labeled probe assay with the on-chip PCR assay. 24 However, no expensive equipment was needed and total operation time was only 30 min for the vancomycin-conjugated beads/nanogold-labeled probe assay. The developed microfluidic device is therefore relatively simple and rapid and could be promising for diagnosis of bacteria. Discussion Figure 4. The aggregation assay by using nanogold probes specific E. coli after 0.01 N HCl treatments. (A) The dark red color for the three strains of E. coli indicated that the specific nanogold-labeled probes were hybridized with target E. coli DNA. The NC was the negative control by ddH2O; MRSA was methicillin resistance S. aureus and Kp was K. pneumonia. All of non-E. coli treatments still present blue color. (B) The limit of detection for E. coli reached as low as 10 2-10 3 CFU.
CFU/μL of E. coli, nanogold probes hybridized efficiently to E. coli genomic DNA, as indicated by the dark-red color of the sample. In the absence of targets, the sample appeared blue-gray as expected. The highest hybridization temperature, 65 °C was therefore used to minimize non-specific hybridization in this study.
A new microfluidic chip was developed that integrated two novel techniques for bacteria detection: vancomycin-conjugated magnetic beads for general capture of multiple strains of bacteria and nanogold labeled with specific nucleotide probes for sensitive and strain-specific bacteria detection. 31 The vancomycin-conjugated beads demonstrated high capture rate for various species of both Gram-positive and Gram-negative bacteria. The nanogold labeled with specific nucleotide probes displayed high specificity for targeted bacteria while demonstrating a limit of detection as low as 10 2 CFU of tested bacteria per reaction. In addition, specific nucleotide probe was designed from a highly conserved region within a specific gene of a particular bacteria strain and labeled with 20-nm gold nanoparticle as a strainspecific detection probes. The gold nanoparticle-DNA
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
complexes were synthesized when the nanogold probe hybridized with target DNA and diffused in solution. The welldistributed complexes were presented as red color by 0.01 N hydrogen chloride (HCl) treatment. Otherwise, the un-hybridized gold nanoparticles aggregated together by surface charge interaction and presented as purple-gray color at the same acidic condition. The dark red color was then used as an indicator for target pathogen diagnosis. Moreover, some E. coli spiked biosamples, such as serum, urine and sputum were tested by using the developed microfluidic system and shown in Supporting Information Figure 6. The dark red color could be successfully observed in E. coli spiked samples while pure biosamples without spiked E. coli as negative controls still kept blue-gray color. The results demonstrated that the developed microfluidic chip could be applied in future clinical diagnosis. Furthermore, the gold nanoparticles can be used for colorimetric sensing specifically that they have different colors depending on the size of gold nanoparticles. The color of gold nanoparticles less than 100 nm represented as dark red. Alternatively, blue or purple colors were observed for gold nanoparticles above 100 nm. Therefore, the multiple pathogen diagnosis could be applied in the microfluidic device by using different sizes of nanogold-labeled specific nucleotide probes. When compared to conventional methods, the entire diagnosis process, including bacteria capturing, nanogold probe hybridization, and visual observation can be automatically performed in the microfluidic chip within 30 min. By obviating time-consuming and expensive nucleic acid amplification and optics-based detection, the developed microfluidic chip should provide a powerful tool for simple, rapid, and yet reliable diagnosis of bacteria. Note that the device can be mass-produced by using injection molding technique. Similarly, the nanoparticles can be mass-produced as well. Therefore the scalability of device/functional particle fabrication and synthesis should be feasible.
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.nano.2013.10.013.
References 1. Hay SI, Battle KE, Pigott DM, Smith DL, Moyes CL, Bhatt S, et al. Global mapping of infectious disease. Phil Trans Royal Soc B-Biol Sci 2013:368 http://dx.doi.org/10.1098/rstb.2012.0250. 2. Mintzer MA, Dane EL, O’Toole GA, Grinstaff MW. Exploiting dendrimer multivalency to combat emerging and re-emerging infectious diseases. Mol Pharm 2012;9:342-54. 3. Sarkonen N, Könönen E, Summanen P, Könönen M, Jousimies-Somer H. Phenotypic identification of Actinomyces and related species isolated from human sources. J Clin Microbiol 2001;39:3955-61. 4. Miñana-Galbis D, Farfán M, Lorén JG, Fusté MC. Biochemical identification and numerical taxonomy of Aeromonas spp. isolated from environmental and clinical samples in Spain. J Appl Microbiol 2002;93:420-30. 5. Muldrew KL. Molecular diagnostics of infectious diseases. Curr Opin Pediatr 2009;21:102-11.
817
6. Aitichou M, Henkens R, Sultana AM, Ulrich RG, Ibrahim MS. Detection of Staphylococcus aureus enterotoxin A and B genes with PCR-EIA and a hand-held electrochemical sensor. Mol Cell Probes 2004;18:373-7. 7. Liao JC, Mastali M, Gau V, Suchard MA, Moller AK, Bruckner DA, et al. Use of electrochemical DNA biosensors for rapid molecular identification of uropathogens in clinical urine specimens. J Clin Microbiol 2006;44: 561-70. 8. Yean CY, Kamarudin B, Ozkan DA, Yin LS, Lalitha P, Ismail A, et al. Enzyme-linked amperometric electrochemical genosensor assay for the detection of PCR amplicons on a streptavidin-treated screen-printed carbon electrode. Anal Chem 2008;80:2774-9. 9. de Tejada GM, Sanchez-Gomez S, Razquin-Olazaran I, Kowalski I, Kaconis Y, Heinbockel L, et al. Bacterial cell wall compounds as promising targets of antimicrobial agents I. Antimicrobial peptides and lipopolyamines. Curr Drug Targets 2012;13:1121-30. 10. Vollmer W, Blanot D, de Pedro MA. Peptidoglycan structure and architecture. FEMS Microbiol Rev 2008;32:149-67. 11. Lin YS, Tsai PJ, Weng MF, Chen YC. Affinity capture using vancomycin-bound magnetic nanoparticles for the MALDI-MS analysis of bacteria. Anal Chem 2005;77:1753-60. 12. Beveridge TJ, Graham LL. Surface layers of bacteria. Microbiol Rev 1991;55:684-705. 13. William DH, Bardsley B. The Vancomycin Group of Antibiotics and the Fight against Resistant Bacteria. Angew Chem Int Ed 1999;38:1172-93. 14. Batchelor M, Zhou D, Cooper MA, Abell C, Rayment T. Vancomycin dimer formation between analogues of bacterial peptidoglycan surfaces probed by force spectroscopy. Org Biomol Chem 2010;8:1142-8. 15. Kell AJ, Stewart G, Ryan S, Peytavi R, Boissinot M, Huletsky A, et al. Vancomycin-modified nanoparticles for efficient targeting and preconcentration of Gram-positive and Gram-negative bacteria. ACS NANO 2008;2:1777-88. 16. Merkoci A. Nanoparticle-based strategies for DNA, protein and cell sensors. Biosens Bioelectron 2010;26:1164-77. 17. Khanna VK. Nanoparticle-based sensors. Def Sci J 2008;58:608-16. 18. Suebsing R, Prombun P, Srisala J, Klatpathomchal W. Loop-mediated isothermal amplification combined with colorimetric nanogold for detection of microsporidian Enterocytozoon hepatopenael in penaeid shrimp. J Appl Microbiol 2013;114:1254-63. 19. Bakthavathsalam P, Rajendran VK, Mohammed JAB. A direction of Escherichia coli genomic DNA using gold nanoprobes. J Nanobiotechnol 2012;10:8-17. 20. Asian K, Lakowicz JR, Geddes CD. Nanogold-plasmon-resonancebased glucose sensing. Anal Biochem 2004;330:145-55. 21. Ploschner M, Cizmar T, Mazilu M, Falco AD, Dholakia K. Bidirectional optical sorting of gold nanoparticles. Nano Lett 2012;12:1923-7. 22. Tan WB, Zhang Y. Surface modification of gold and quantum dot nanoparticles with chitosan for bioapplications. J Biomed Mater Res A 2005;75:56-62. 23. Tsutsumi H, Ohkusa H, Park H, Takahashi T, Yuasa H, Mihara H. Gold nanoparticles conjugated with monosaccharide-modified peptide for lectin detection. Bioorg Med Chem Lett 2012;22:6825-7. 24. Wang CH, Lien KY, Wu JJ, Lee GB. A magnetic bead-based assay for the rapid detection of methicillin-resistant Staphylococcus aureus by using a microfluidic system with integrated loop-mediated isothermal amplification. Lab Chip 2011;11:1521-31. 25. Lagally ET, Scherer JR, Blazej RG, Toriello NM, Diep BA, Ramchandni M, et al. Integrated portable genetic analysis microsystem for pathogen/ infectious disease detection. Anal Chem 2004;76:3162-70. 26. Chang WH, Yang SY, Wang CH, Tsai MA, Wang PC, Chen TY, et al. Rapid isolation and detection of aquaculture pathogens in an integrated microfluidic system using loop-mediated isothermal amplification. Sensor Actuator B Chem 2013;180:96-106. 27. Lien KY, Lee SH, Tsai TJ, Chen TY, Lee GB. A microfluidic-based system using reverse transcription polymerase chain reactions for rapid detection of aquaculture diseases. Microfluid Nanofluid 2009;7: 795-806.
818
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
28. Wang CH, Lien KY, Wang TY, Chen TY, Lee GB. An integrated microfluidic loop-mediated-isothermal-amplification system for rapid sample pre-treatment and detection of viruses. Biosens Bioelectron 2011;26:2045-52. 29. Hill HD, Mirkin CA. The bio-barcode assay for the detection of protein and nucleic acid targets using DTT-induced ligand exchange. Nat Protoc 2006;1:324-36.
30. Weng CH, Lien KY, Yang SY, Lee GB. A suction-type, pneumatic microfluidic device for liquid transport and mixing. Microfluid Nanofluid 2011;10:301-10. 31. Wang CH, Chang CJ, Wu JJ, Lee GB. A new pathogen detection system by utilizing nanogold modified specific probe and vancomycin coated magnetic beads on an integrated microfluidic device. IEEE MEMS 2013:94-7.
Original Article
nanomedjournal.com
An integrated microfluidic device utilizing vancomycin conjugated magnetic beads and nanogold-labeled specific nucleotide probes for rapid pathogen diagnosis Chih-Hung Wang, PhD a , Chia-Jung Chang a , Jiunn-Jong Wu, PhD b , Gwo-Bin Lee, PhD a, c, d,⁎ a Department of Power Mechanical Engineering, National Tsing Hua University, Hsinchu, Taiwan Department of Medical Laboratory Science and Biotechnology, National Cheng Kung University, Tainan, Taiwan c Institute of Biomedical Engineering, National Tsing Hua University, Hsinchu, Taiwan d Institute of Nano Engineering and Microsystems, National Tsing Hua University, Hsinchu, Taiwan Received 30 May 2013; accepted 31 October 2013
b
Abstract A PCR-free assay for rapid pathogen diagnosis was implemented on an integrated microfluidic system in this study. Vancomycin-conjugated magnetic beads were used to capture multiple strains of bacteria and nucleotide probes labeled gold nanoparticles were used to specify and detect a specific strain by hybridization-induced color change. The assay was entirely automated within an integrated microfluidic device that was composed of suction-type micropumps, microvalves, microchannels, and microchambers that fabricated by microfluidic technology. Multiple strains of bacteria could be captured simultaneously by vancomycin-conjugated magnetic beads, with capturing efficiency exceeding 80%. Subsequently, sensitive and strain-specific detection against target bacteria could be achieved by using nanogold labeled specific nucleotide probes. The limit of detection of 10 2 CFU bacteria was achieved. Importantly, nucleic acid amplification was not involved in the diagnostic procedures; the entire analytic process required only 25 min. The developed platform may provide a promising tool for rapid diagnosis of bacterial infections. From the Clinical Editor: In this novel study, a PCR-free pathogen detection method is demonstrated. After vancomycin-conjugated magnetic beads captured bacteria, nucleotide probes-labeled gold nanoparticles were employed to specify and detect specific strains via hybridization-induced color change. Multiple strains of bacteria could be captured simultaneously with an efficiency exceeding 80%, enabling the detection of as low as 10 2 CFU of bacteria. © 2014 Elsevier Inc. All rights reserved. Key words: Vancomycin; Nanogold; Microfluidic; Bacteria
Abbreviations: Bp, base-pair; CFU, colony forming unit; DNA, deoxyribonucleic acid; dNTP, deoxyribonucleotide triphosphate; E. coli, Escherichia coli; EDAC, (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide; EMVs, electromagnetic valves; ddH2O, deionic distilled water; G(−), Gram-negative bacteria; G(+), Gram-positive bacteria; HCl, hydrogen chloride; LB, LuriaBetani; LOC, lab-on-a-chip; LOD, limit of detection; MEMS, micro-electromechanical-systems; MRSA, methicillin-resistant Staphylococcus aureus; PCR, polymerase chain reaction; PDMS, polydimethylsiloxane; SDS, sodium dodecyl sulfate; TBE, Tris-Bornate-EDTA; Tween 20, (polyoxyethylene (20) sorbitan monolaurate; Xcv, Xanthomonas campestris pv. vesicatoria. The preliminary results from this paper were presented at the 2013 IEEE Micro Electro Mechanical Systems (IEEE MEMS 2013), Taipei, Taiwan, 2013. The authors would like to thank the National Science Council in Taiwan (NSC 101-2120-M-007-014) and the “Toward a World-class University” Project for financial support of this study. There is no any conflict of interest. ⁎Corresponding author at: Department of Power Mechanical Engineering, National Tsing Hua University, Hsinchu 300, Taiwan. E-mail addresses: [email protected] (C.-H. Wang), [email protected] (C.-J. Chang), [email protected] (J.-J. Wu), [email protected] (G.-B. Lee).
Infectious diseases have become a growing threat to the public health worldwide. 1 The emerging bacterial diseases and antibiotic-resistance bacteria are crucial issues since the mortality of the bacterial infections has gradually increased. 2 Conventionally, these bacteria can be detected and identified through pure culture of bacterium and several existing biological assays. 3,4 However, bacterial growth in cell culture may take several days to even weeks until a sufficient amount of bacteria can be obtained for biological and chemical analysis. In addition, excessive experimental procedures can easily cause artificial errors, leading to false diagnostic results. Therefore, rapid and accurate diagnosis plays an important role for infectious disease prevention and control. Recently, protein-based immunoassays (e.g. enzyme-linked immunosorbent assay) and nucleic acid amplification based methods (e.g. polymerase chain reaction (PCR), real-time PCR) have been widely applied for pathogen
1549-9634/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nano.2013.10.013 Please cite this article as: Wang C-H, et al, An integrated microfluidic device utilizing vancomycin conjugated magnetic beads and nanogold-labeled specific nucleotide probes for rapid pathogen diagnosis. Nanomedicine: NBM 2014;10:809-818, http://dx.doi.org/10.1016/j.nano.2013.10.013
810
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818 5
detection. These molecular methods can achieve highly sensitive and specific diagnostic results. However, these methods often require expensive reagents and equipments and welltrained operators for performing relatively complicated assays. In particular, nucleic acid amplification based techniques require extracted nucleic acid products from bacterial by performing sample pretreatment for differentiating pathogen DNA. 6–8 In other words, it was a labour-intensive, time-consuming and costly pretreatment in bacteria detection. Vancomycin can be used as a capturing agent for bacteria because it can bind to the peptidoglycans on the bacteria cell wall for both Gram-positive and -negative bacteria. 9–11 Specifically, Gram-negative bacteria were considered to contain a thinner peptidoglycan layer than Gram-positive bacteria. 12 Thus these antibiotics could reversibly bind onto D-alanyl-D-alanine, terminal residues of mucopeptides on the cell wall of the Gram-positive bacteria since specific hydrogen bonds could be formed between the alanine methyl group of the cell wall and vancomycin. 13,14 In addition to Gram-positive bacteria, the vancomycin also had affinity to be bonded with L-lysine-D-alanine residues of peptidoglycans that was expressed on the outer membrane of Gram-negative bacteria. 15 Therefore, vancomycin could be used to capture various species of bacterial pathogens if they may be properly surface-conjugated on the magnetic beads. A thiol chemically modified oligonucleotide linked with gold nanoparticles was reported for DNA selectivity detection. 16,17 The method was used for pathogen identification that provided simply detection for specific target DNA in the mixture of oligonucleotides by colorimetric changed observation. 18,19 The optical properties relied on the light absorption that depended on the distance of gold nanoparticles. When compared with the aggregated nanogold particles, the nanogold particle-target DNA complex had different distance in the tested dispersion that could cause color change by light spectra shift. The distinct optical properties of gold nanoparticles were demonstrated as sensitive and specific DNA biosensors that caused by red-shift effect in a gold nanoparticle aggregation process. 20,21 Nanogold was relatively biocompatible and enable facile surface immobilization chemistries for ready conjugation of oligosaccharides, nucleic acids, peptides or small biofunctional molecules. 22,23 In this study, an integrated microfluidic device for performing the new protocol for rapid pathogen diagnosis was developed. The diagnostic system consisted several functional components in the developed devices, including a recognition element for binding pathogens by using vancomycin-conjugated magnetic beads and a sensing element by using gold nanoparticle labeled with specific nucleotide probes binding to target genes. Advantages of the micro-total-analysis-systems (μ-TAS) or lab-on-a-chip (LOC) platform include reduced consumption of samples and reagents, fast analysis speed, and integrated analytic and sensing techniques without the need of skilled operators. Micro-electro-mechanical-systems (MEMS)-based biomedical systems have recently been widely applied for molecular diagnosis. 24 For instance, an integrated microfluidic system performing sample pretreatment and rapid diagnosis of Staphylococcus aureus was reported. 25 Furthermore, suction-type micropumps, microchannels, microchambers and microvalves fabricated by polydimethylsiloxane (PDMS)-based processes
have been developed to realize on-chip fluidic transportation and sample mixing, which greatly simplifies experimental processes and reduces the size of microfluidic chips. 24 Regulated by electromagnetic valves (EMV), fluid sample could be efficiently transported forward and backward to realize reagent exchange and incubation processes. In this study, these microfluidic devices were integrated on a single chip to automate the entire process for fast diagnosis of bacterial infections with minimal human intervention by using the new protocol described above.
Methods Tested bacteria and primers Gram-positive (G(+)) bacteria, including Streptococcus agalactiae, Streptococcus pyogenes, methicillin-resistant S. aureus (MRSA), and Bacillus subtilis and Gram-negative (G(−)) bacteria, including Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa were provided from National Cheng Kung University Hospital, Tainan, Taiwan. Xanthomonas campestris pv. vesicatoria, a G(−) plant pathogen, was provided by Dr. HuiLiang Wang, from Department of Biotechnology, National Kaohsiung Normal University, Kaohsiung, Taiwan. All of the tested bacteria were freshly incubated in LB broth (Sigma-Aldrich Co. USA) at 37 °C for 16 hours. The sequences of primers for the multiplex PCR were selected from nucleotide database of NCBI website (National Center for Biotechnology Information, http:// www.ncbi.nlm.nih.gov/nucleotide), designed by FastPCR software (ver 6.1, PrimerDigital Co. USA), and listed in Supporting Information Table 1. Chip design and fabrication A schematic layout of the integrated microfluidic device was shown in Figure 1. The device composed of suction-type micropumps, microvalves, microchannels, and loading chambers, is capable of performing eight independent diagnoses simultaneously. The complete microfluidic chip functions as a “cartridge” that integrates seamlessly in a custom-designed microfluidic machine that includes a temperature controller, electromagnetic valves, vacuum pumps, and waste chambers. 26 Samples and other reagents were loaded into respective chambers before testing. The micropumps could deliver all processing buffer and nanogold probes into the sample loading chambers to perform incubation, hybridization and nanogold aggregation processes in an automatic format. The device was fabricated by polydimethyl siloxane (PDMS) molding and bonding processes as previously reported. 26,27 Briefly, PDMS layers were made with PDMS polymer and curing agent (Sylgard 184A/B, Sil-More Industrial Ltd., USA) at a weight ratio of 10:1. The two PDMS layers and a glass substrate were treated with oxygen plasma, aligned, and bonded together to form the integrated microfluidic chip. Method of vancomycin-conjugated with magnetic beads The protocol for coating vancomycin on magnetic beads was modified from a previous study. 28 Briefly, 100 μL of carboxylic modified magnetic beads (Dynabeads® MyOne TM carboxylic
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
811
Figure 1. Overview of the microfluidic chip. (A) Exploded schematic diagram of the integrated microfluidic chip, which was composed of a thick-film PDMS layer, a thin-film PDMS layer, and a glass substrate. (B) Schematic layout of the integrated microfluidic device capable of carrying out four parallel diagnosis processes simultaneously. Micropumps, microvalves, washing buffer chambers, nanogold chambers, HCl chambers, sample loading chambers, and waste outlet were fully integrated. (C) Photograph of the integrated microfluidic chip. The device was measured to be 59 mm (length) × 75 mm (width).
812
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
acid, Invitrogen Dynal AS, Oslo, Norway) were dispersed by a vortex mixer and aliquotted into a micro-centrifuge tube. Magnetic beads were washed twice with 1000-μL ddH2O and collected by a magnetic particle concentrator (DynaMag TM-2, Invitrogen Dynal AS, Oslo, Norway). For testing the optimal concentration of vancomycin, 950-μL ddH2O and 30-μL vancomycin (Sigma-Aldrich Co. USA) at various concentrations were added and vortexed for 5 seconds. Subsequently, 20-μL of 120 mg/ml EDAC (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide, Sigma-Aldrich Co. USA) was added to the mixture and vortexed for 5 seconds. The mixture was incubated on a wheeling rotator at 20 rpm at room temperature in the dark for 18 hours. After this incubation, the mixture was washed twice by using 1000-μL of 0.02% Tween 20 solution (polyoxyethylene (20) sorbitan monolaurate, Sigma-Aldrich Co. USA) and collected by the magnetic particle concentrator. Next, 1000-μL of 0.1% SDS (sodium dodecyl sulfate, Sigma-Aldrich Co. USA) was used to further wash the vancomycin-bead complexes. Finally, 1000-μL of 0.1 M ethanol amine (Sigma-Aldrich Co. USA) was used to block the free uncoupling site on magnetic beads at room temperature for 1 hour. The beads were finally washed by using 1000-μL ddH2O, collected by the magnetic particle concentrator, re-suspended in 1 × PBS, and stored at 4 °C until use. Bacterial binding efficiency testing, microscopic observation, and multiplex PCR setting Three methods – bacterial binding efficiency test, microscopic observation, and multiplex PCR – against various types of bacteria were used to confirm the capturing ability and specificity of vancomycin-conjugated magnetic beads in this study. The capturing efficiency of vancomycin-conjugated magnetic beads was determined by counting the colony number of vancomycin-captured bacteria. Specifically, 1 × 10 5 magnetic beads surface-conjugated with 100 nM vancomycin were incubated with 1 × 10 5 CFU (colony forming unit) of various types of bacteria and incubated at 20 rpm at room temperature for 10 min. Magnetic beads and captured bacteria were collected by a magnetic particle concentrator and washed twice by ddH2O. Then, the collected pellet was plated on LB plates and incubated at 37 °C for 16 hours. In parallel, bacteria from the original sample were also plated to serve as a comparison. Colonies of grown bacteria in both cases were counted and the capture rate of vancomycin-conjugated magnetic beads was calculated as followed. capture rate ð%Þ total number of collected bacteria 100% ð1Þ ¼ original number of tested bateria 105 CFU In this study, the mean ± SD was used to indicate the capture rate of vancomycin conjugated magnetic beads for pathogen detection. Each data in the table was determined from three independent experiments. Each data was calculated by one-way ANOVA assay for significant difference determination. The capturing specificity of vancomycin-conjugated magnetic beads was directly observed by microscopy. Following a
previously mentioned protocol, 10 3 of vancomycin-conjugated magnetic beads was mixed with various types of bacteria and collected by a magnetic particle concentrator. Then, each collected pellet was dropped on a microscope slide and observed under microscope at 500 × magnification. Following the protocol, the total number of captured bacteria was collected for multiplex PCR tests. Total DNA of vancomycin-bead-captured bacteria was extracted by boiling at 95 °C for 10 min and washed twice by ddH2O. The 10-μL of ddH2O was added in treated vancomycin-conjugated beadsbacteria complex. Extracted bacterial DNA was added to 20-μL PCR mix which contained 1.5-μL of deoxyribonucleotide triphosphate (dNTP, 10 mM, Promega, USA), 3-μL of 10 × PCR buffer (20 mM Tris–HCl, pH 8.0, 100 mM KCl, 0.1 mM 2-[2-(Bis (carboxymethyl) amino) ethyl-(carboxymethyl) amino] acetic acid (EDTA), 1 mM Dithiothreitol (DTT), 0.5% Tween, 0.5% Nonidet and 50% (v/v) glycerol, JMR Holdings, UK), 1-μL of mecA specific primer pairs (0.5-μL of each primer for the forward/reverse primers), 1-μL of Superthermo Gold Taq DNA polymerase (5 U/μL, JMR Holdings, UK), 1-μL of 10 mM MgCl2 and 12.5-μL of ddH2O. The thermocycling process for PCR was performed under the following conditions: 95 °C for 5 min (initial denaturation), 35 cycles of 95 °C for 30 seconds (denaturation), 58 °C for 30 seconds (annealing), and 72 °C for 30 seconds (extension), and finally 72 °C for 10 min. Amplified PCR products were confirmed by 2% TBE (Tris-BornateEDTA)-agarose electrophoresis. Preparation and detection of gold nanoparticles labeled with specific nucleotide probes Gold nanoparticles labeled with specific nucleotide probes were prepared by linking thiol-modified oligonucleotide probes to gold nanoparticles via gold-thiol bond. The synthesis of these probes could be referred to a previous study. 29 First, thiolmodified oligonucleotide probe that was specific to E. coli lamB was dissolved to a final concentration of 100 μM in 100 mM dithiothreitol (DTT)/sodium phosphate buffer (pH7.0, SigmaAldrich Co. USA). The removal of DTT was essential for the gold nanoparticle labeled with E. coli lamB oligonucleotide probe. The NAP-10 column (GE Healthcare, UK) was used for DTT elimination, which equilibrated 15 mL of 50 mM sodium phosphate buffer (pH 7.0, Sigma-Aldrich Co., USA). The oligonucleotide sample was eluted with 1000-μL sodium phosphate buffer from the NAP-10 column. Four nmol of thiol modified oligonucleotide was incubated overnight with 1000-μL gold nanoparticles (20 nm, Sigma-Aldrich Co. USA) in a wheeling rotator at 20 rpm at room temperature in the dark. After the overnight incubation, 9 mM phosphate buffer (pH 7.0) and 0.1% of sodium dodecyl sulphate (SDS) were added. The mix was then shaken on an orbital shaker for 30 min. The salting solution was added at various times to a final concentration of 0.3 M NaCl. (2 M NaCl in 10 mM PBS pH7) and equilibrated overnight at room temperature. The pellet was precipitated at 13000 g for 20 min and washed with 500-μL resuspension buffer (10 mM PBS, pH 7.4, 150 mM NaCl, 0.1% SDS) and resuspended in 50-μL of the same buffer. The gold nanoparticles labeled with specific nucleotide probes were stored in light-tight
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
813
Figure 2. Schematic diagram of the experimental procedure in the developed microfluidic chip. (A) Tested bacteria were incubated with vancomycin-conjugated magnetic beads. (B) Bacteria were captured by beads and collected by external magnets. The chip was heated to 95 °C to thermally lyse captured bacteria and denature their genomic DNA. (C) Strain-specific nanogold probes were injected and allowed to hybridize with denatured bacterial genomic DNA. (D) Finally, 0.01 N HCl was added to induce color change in the sample.
containers at room temperature until use. For optical detection with nanogold labeled with the E. coli lamB probe, the vancomycin bead-captured complexes were washed twice by ddH2O and boiling at 95 °C for 5 min for lysing bacteria and denaturing genomic DNA. Next, 5-μL of 10 nM nanogold probes was added and hybridized with the denatured genomic DNA at 65 °C for 10 min. The color exchange of reactant was applied to optical observation under final 0.01 N HCl treatment. Results Characterization of microfluidic devices The developed integrated microfluidic device was used for rapid bacterial diagnosis. The working principle was schematically shown in Figure 2. Target bacteria in culture medium were first incubated and captured by vancomycin-conjugated magnetic beads and subsequently collected onto the bottom of the microchamber by a permanent magnet. Unwanted substances and debris in the sample were then washed twice by deionic distilled water (ddH2O). After washing, the chip was heated at 95 °C for 10 min to thermally lyse captured bacteria and denature the target bacterial DNA. Next, nanogold labeled with specific nucleotide probes were added and allowed to hybridize with the complementary region of denatured target DNA at 58 °C for 10 min. Finally, 0.01 N hydrogen chloride was added to the mix. After a brief incubation (approximately 5 min) at ambient temperature, nanogold probes that were hybridized to the targeted gene remained dispersed and thus appeared dark-red
in color. On the other hand, unhybridized gold nanoparticles were aggregated under the acidic condition and thus presented a purple-gray color. The color differences could be clearly differentiated by the naked-eye. As a result, the diagnostic results were determined directly by visual observation. Alternatively, the color change could be quantified by a spectrophotometer (NanoPhotometer Pearl, Implen GmBH, München, Germany). From the addition of reagents and samples to the visual detection of results, the entire analytic process required only 25 min and was completely realized within the integrated chip in an automated fashion. An exploded view of the integrated microfluidic chip, which consisted of two PDMS layers and a glass substrate, is shown in Figure 1, A. The PDMS reservoir layer harbored microchambers for storing samples and required reagents. The PDMS transport layer was constituted with micropumps, microvalves and microchannels for the transportation and processing of samples and reagents. Following oxygen plasma treatments and manual alignment, the PDMS layers and the glass substrate were bonded to form the completed chip. The width of the air channel and the fluidic channel were 350 and 200 μm, respectively. The radius of microchambers was 2000 and 2250 μm, respectively. The related position and detailed size of modules were shown in Supporting Information Figure 1. Figure 1, B depicts the schematic diagram of the integrated microfluidic chip, which possesses eight identical sets of analytic unit for parallel analyses. Each analytic unit is composed of four microchambers for storing the sample, the washing buffer, the nanogold probes, and the hydrochloric acid solution as well as a
814
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
micropump, microchannels, and microvalves for transporting the sample and reagents between chambers. Fluids are transported through microvalves by the negative gauge pressure (suction force) filling up the cavities to deform the PDMS membranes. The relationship between the pumping rate of the micropump and the driving frequency at different operating air pressures is shown in Supporting Information Figure 2. The results revealed that the pumping rate increases as the driving frequency increases and then drops as the driving frequency reaches an optimal value. A flow rate as high as 50 μL/s was achieved (at a driving frequency of 8 Hz) at an air pressure of − 60 kPa. In addition, the floating-block structure of the normally-closed microvalve capable of increasing the flow pumping rate has also been demonstrated to block liquids in the microchannel successfully. 30 Hence, the micropump and the microvalve are capable of transporting samples and reagents or blocking the fluids inside the microchannel. The procedure of the entire assay in the microfluidic device was first demonstrated with dyed solutions and shown in the Supporting Information Video. Prior to the assaying steps, the test sample and vancomycin-conjugated beads (represented as “brown” color solution), the washing buffer (transparent solution), nanogold probes (red color), and 0.01 N hydrochloric acid solution (blue color) were loaded in their respective chambers. When the incubation step for the test sample and vancomycin-conjugated beads was finished and the magnetic bead-complex was collected by applying an external magnetic field, the wash buffer was transported into the sample chamber to remove the unbound waste. Note that all of wastes in the analytic processes were removed from the waste chamber by activating the suction-type micropump under negative gauge pressure. After the cell lysis process, nanogold probes were transported into the sample chamber and then incubated with target bacterial DNA. Similarly, the waster was transported into the waste chamber afterwards. Finally, hydrochloric acid was added into the sample chamber to induce color change. The simulated assay indicated that the eight samples could all work in the microfluidic chip in parallel. The photograph of the microfluidic device is shown in Figure 1, C. The dimensions of the integrated microfluidic device were measured to be 59 mm (length) × 75 mm (width) × 1.25 mm (height). Optimization of bacterial capturing by using vancomycin-conjugated magnetic beads The optimal condition of the bacterial capturing assay was determined by adjusting the vancomycin concentration, the number of magnetic beads, and the number of bacteria. First, different concentrations of vancomycin were conjugated to magnetic beads and subsequently incubated with 1 × 10 8 CFU of methicillin-resistant S. aureus (MRSA). To determine the capture rate of vancomycin-conjugated beads, the mecA gene of captured MRSA bacteria was further amplified by a PCR protocol. 24 Supporting Information Table 2 shows that, at vancomycin concentrations of 100 mM, 100 μM, 100 nM and 100 pM, the capture rates were 20.71, 23.73, 23.72 and 11.75%, respectively, with 1 × 10 8 CFU of MRSA. The results demonstrated that the capacity of vancomycin-conjugated beads was
Table 1 The capture rate of vancomycin-conjugated magnetic beads at different bacteria concentrations. Concentration of tested bacteria⁎ (CFU)
Capture rate †(%)
10 7 10 6 10 5 10 4 10 3
23.94 34.27 66.07 92.96 91.82
± ± ± ± ±
3.98 14.91 3.57 2.09 2.35
⁎ 1 × 10 5 magnetic beads surface-conjugated with 100 nM vancomycin and serially diluted methicillin-resistant S. aureus (MRSA) were used in this experiment. † All of results were measured by three independent experiments.
saturated in this testing. Excessive free MRSA may lead to the low capture rate. Furthermore, the capture rate against different number of bacteria was determined and listed in Table 1. Using 100-nM vancomycin-conjugated beads and ten-fold serially diluted MRSA, the capture rate increased as the number of bacteria decreased; the capture rate reached as high as 91.02 and 92.96% for 1 × 10 3 and 1 × 10 4 CFU MRSA, respectively. The results revealed that the capacity of capture rate for 100 nM vancomycin-conjugated beads was saturated at 10 3 CFU of MRSA. Moreover, the limit of detection (LOD) of MRSA that was determined by electrophoretic photograph has similar results with the capturing testing (Figure 3, A). It indicated the highest capture rate of vancomycin-conjugated beads for 10 4 CFU of bacteria. However, as the number of bacteria increased, the capture rate of vancomycin-conjugated beads was reduced. Note that according to the equation of capture rate for vancomycinconjugated beads, the capture rate will decrease with excessive bacteria. In addition, the electrophoretic photograph in Figure 3, B reveals that no amplified product was observed by uncoated magnetic beads and 100 pM vancomycin-conjugated beads (lanes 1 and 5). On the other hand, a 267-bp PCR product band appeared in samples with 100 mM, 100 μM, and 100 nM vancomycin-conjugated beads (lanes 2, 3 and 4), indicating successful capture of MRSA at these vancomycin concentrations. In order to minimize the use of antibiotics, the lowest concentration of vancomycin, 100 nM was selected for bacteria capturing in this study. For exploring optimal number of beads, 10 4, 10 5 and 10 6 vancomycin-conjugated beads were used for the capturing assay and the results were shown in Supporting Information Table 3. There were no significant differences in the capture rate among the three numbers of vancomycin-conjugated beads. Summarizing the experimental results above, the optimal assay condition was achieved with 100-nM vancomycin-conjugated 1 × 10 4 beads for the bacteria capturing test. Specificity testing for vancomycin-conjugated magnetic beads Vancomycin-conjugated magnetic beads proposed in the present study were capable of capturing various species of bacteria, thus facilitating rapid detection. To demonstrate this capability, both Gram-positive and Gram-negative bacteria from human and agriculture origins were tested with the vancomycin-
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
815
Table 2 The capture rate of vancomycin-conjugated magnetic beads for different types of bacteria. Tested bacteria⁎
Capture rate (%) †
S. pyogenes B. subtilis E. coli K. pneumonia P. aeruginosa
93.74 83.35 91.01 90.26 65.35
± ± ± ± ±
2.26 9.37 1.45 1.25 2.08
⁎ Bacteria sample with a concentration of 1 × 10 4 CFU and 1 × 10 5 magnetic beads surface-conjugated with 100 nM vancomycin was used in this experiment. † All of results were measured by three independent experiments.
Figure 3. Characterization of vancomycin-conjugated magnetic beads in bacteria capture via PCR and gel electrophoresis. (A) The PCR assay for testing the limit of detection by vancomycin-conjugated magnetic beads. 100 nM of vancomycin-conjugated magnetic beads was used in the capturing assay. The 267-bp product was indicated as amplified products of MRSA mecA gene. Lane NC is the negative control with ddH2O; Lanes 1 through 6 are 10 7, 10 6, 10 5, 10 4, 10 3 and 10 2 CFU of MRSA, respectively. (B) The optimal concentration of vancomycin for conjugating on magnetic beads was determined by capturing 10 3 CFU of MRSA and performing PCR. The 267-bp product bands on the gel image were the appropriate amplified products of MRSA mecA gene. In this and subsequent gel images, lane L represents 100-bp DNA ladder. Lane NC is the negative control with ddH2O; lane 1 indicates bare magnetic beads; lanes 2 3, 4, and 5 indicate magnetic beads surface-conjugated with 10 mM, 100 μM, 100 nM and 100 pM of vancomycin, respectively.
based capturing method. Specifically, 1 × 10 4 CFU Grampositive (S. pyogenes, S. agalactiae, and MRSA) or Gramnegative (E. coli, K. pneumonia, and Xanthomonas campestris pv. vesicatoria, Xcv) bacteria were used as targets and incubated with 1 × 10 4 magnetic beads conjugated with 100 nM vancomycin. After incubation, bacteria-bead complexes were washed twice and observed under AV microscope at 500 × magnification (Supporting Information Figure 3). Microscopic observation showed that bacteria of all tested strains (transparent in color in the micrographs) indeed adhered to the surface of vancomycin-conjugated beads (red-brown in color in the micrographs) — a direct evidence for the general capture capability of vancomycin-conjugated beads against various strains of bacteria. Furthermore, capture rates of vancomycinconjugated beads against these species of bacteria were
accurately measured by bacterial counting (Table 2). After capturing and washing steps, collected bacteria-bead complexes were plated on Luria-Betani (LB) plates for 16 hours at 37 °C. The capture rate was then calculated as the ratio between the colony number formed on the plate and the original number of tested bacteria. Overall, the capture rate for most of tested bacteria was greater than 80%. Strains of E. coli, K. pneumonia and S. pyogenes even exceeded 90%. One strain of Gramnegative bacteria (P. aeruginosa), however, was captured at only 65%. Since the composition of bacterial cell wall varies from species to species, it was speculated that the thinner peptidoglycan layer – and thus less bond peptide components with vancomycin – on the cell wall of this Gram-negative bacteria may have led to its reduced capture rate. As additional verification for the capturing process, a multiplex PCR assay was used to amplify specific genes of captured bacteria and confirmed by slab-gel electrophoresis (Supporting Information Figure 4). Electrophoretic results validated that all of the five species of bacteria that were tested could be captured by vancomycin-conjugated magnetic beads and simultaneously amplified by the multiplex PCR method. Not only could these results support that vancomycin-based bacteria capture method could capture a broad range of bacteria in parallel, but it also demonstrated that it could be coupled with different detection methods to achieve the detection of multiple species of bacteria at one time. Optimal condition testing of specificity for nanogold labeled with E. coli specific nucleotide probe As a complement to the general bacteria capturing capability of the vancomycin-based capture method, a strain-specific detection method based on nanogold labeled with specific nucleotide probes was developed in the present work. Using E. coli as the sample target, a nucleotide probe that can specifically hybridize to the lamB gene of E. coli was designed. In this study, all nanogold colorimetric tests were processed in the microfluidic system. All tested results were then removed to new chamber wells of a PDMS layer with a white background to enhance the color contrast for better observation. The optimal hybridization temperature between nanogold probes and genomic DNA of E. coli was first tested; the results are shown in Supporting Information Figure 5. The results revealed that, under temperatures between 50 °C and 65 °C and against 10 2 to10 4
816
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
After the bacteria capturing process by vancomycin-conjugated beads, the specificity of lamB nanogold probe was tested against three strains of E. coli (Topo10, ER2738, 119) and other non-E. coli strains (MRSA, K. pneumoniae) and the results are shown in Figure 4, A. The dark-red color was presented for all of the three tested E. coli strains, demonstrating that the nanogold probe can detect multiple strains of E. coli. Furthermore, the negative control, MRSA, and K. pneumoniae samples yielded the blue-gray color that indicated as no probe-target hybridization, thus verifying the specificity of the lamB nanogold probe. Importantly, this nanogold probe-based detection obviates any nucleic acid amplification (i.e. PCR-free), which significantly improves the assay speed. Furthermore, the results could be simply examined by naked-eye, precluding expensive optical equipment for detection. Therefore, the nanogold probe-based detection method may present as a powerful tool for simple and rapid diagnosis of pathogens. Limit of detection for nanogold labeled with E. coli specific nucleotide probe Figure 4, B shows the LOD by utilizing the E. coli lamB nanogold-labeled probe. After nanogold probe hybridization, the negative control and samples containing 1 × 10 8 CFU of non-E. coli strains (MRSA, B. subtilis and K. pneumoniae) appear grayblue in color. However, samples containing two strains of E. coli (ER2738, 119) turned dark-red color. Here, the LOD of E. coli diagnosis was determined by the optical intensity of red color that compared with the negative control after nanogold probe hybridization. The intensity of the red color decreased along with decreasing dose of E. coli ranging from 10 8 to 10 2 CFU/ reaction. According to the color change, the limit of detection using the nanogold probe method was measured to be approximately 10 2-10 3 CFU of bacteria. Note that there is no significant difference in terms of LOD when comparing the nanogold-labeled probe assay with the on-chip PCR assay. 24 However, no expensive equipment was needed and total operation time was only 30 min for the vancomycin-conjugated beads/nanogold-labeled probe assay. The developed microfluidic device is therefore relatively simple and rapid and could be promising for diagnosis of bacteria. Discussion Figure 4. The aggregation assay by using nanogold probes specific E. coli after 0.01 N HCl treatments. (A) The dark red color for the three strains of E. coli indicated that the specific nanogold-labeled probes were hybridized with target E. coli DNA. The NC was the negative control by ddH2O; MRSA was methicillin resistance S. aureus and Kp was K. pneumonia. All of non-E. coli treatments still present blue color. (B) The limit of detection for E. coli reached as low as 10 2-10 3 CFU.
CFU/μL of E. coli, nanogold probes hybridized efficiently to E. coli genomic DNA, as indicated by the dark-red color of the sample. In the absence of targets, the sample appeared blue-gray as expected. The highest hybridization temperature, 65 °C was therefore used to minimize non-specific hybridization in this study.
A new microfluidic chip was developed that integrated two novel techniques for bacteria detection: vancomycin-conjugated magnetic beads for general capture of multiple strains of bacteria and nanogold labeled with specific nucleotide probes for sensitive and strain-specific bacteria detection. 31 The vancomycin-conjugated beads demonstrated high capture rate for various species of both Gram-positive and Gram-negative bacteria. The nanogold labeled with specific nucleotide probes displayed high specificity for targeted bacteria while demonstrating a limit of detection as low as 10 2 CFU of tested bacteria per reaction. In addition, specific nucleotide probe was designed from a highly conserved region within a specific gene of a particular bacteria strain and labeled with 20-nm gold nanoparticle as a strainspecific detection probes. The gold nanoparticle-DNA
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
complexes were synthesized when the nanogold probe hybridized with target DNA and diffused in solution. The welldistributed complexes were presented as red color by 0.01 N hydrogen chloride (HCl) treatment. Otherwise, the un-hybridized gold nanoparticles aggregated together by surface charge interaction and presented as purple-gray color at the same acidic condition. The dark red color was then used as an indicator for target pathogen diagnosis. Moreover, some E. coli spiked biosamples, such as serum, urine and sputum were tested by using the developed microfluidic system and shown in Supporting Information Figure 6. The dark red color could be successfully observed in E. coli spiked samples while pure biosamples without spiked E. coli as negative controls still kept blue-gray color. The results demonstrated that the developed microfluidic chip could be applied in future clinical diagnosis. Furthermore, the gold nanoparticles can be used for colorimetric sensing specifically that they have different colors depending on the size of gold nanoparticles. The color of gold nanoparticles less than 100 nm represented as dark red. Alternatively, blue or purple colors were observed for gold nanoparticles above 100 nm. Therefore, the multiple pathogen diagnosis could be applied in the microfluidic device by using different sizes of nanogold-labeled specific nucleotide probes. When compared to conventional methods, the entire diagnosis process, including bacteria capturing, nanogold probe hybridization, and visual observation can be automatically performed in the microfluidic chip within 30 min. By obviating time-consuming and expensive nucleic acid amplification and optics-based detection, the developed microfluidic chip should provide a powerful tool for simple, rapid, and yet reliable diagnosis of bacteria. Note that the device can be mass-produced by using injection molding technique. Similarly, the nanoparticles can be mass-produced as well. Therefore the scalability of device/functional particle fabrication and synthesis should be feasible.
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.nano.2013.10.013.
References 1. Hay SI, Battle KE, Pigott DM, Smith DL, Moyes CL, Bhatt S, et al. Global mapping of infectious disease. Phil Trans Royal Soc B-Biol Sci 2013:368 http://dx.doi.org/10.1098/rstb.2012.0250. 2. Mintzer MA, Dane EL, O’Toole GA, Grinstaff MW. Exploiting dendrimer multivalency to combat emerging and re-emerging infectious diseases. Mol Pharm 2012;9:342-54. 3. Sarkonen N, Könönen E, Summanen P, Könönen M, Jousimies-Somer H. Phenotypic identification of Actinomyces and related species isolated from human sources. J Clin Microbiol 2001;39:3955-61. 4. Miñana-Galbis D, Farfán M, Lorén JG, Fusté MC. Biochemical identification and numerical taxonomy of Aeromonas spp. isolated from environmental and clinical samples in Spain. J Appl Microbiol 2002;93:420-30. 5. Muldrew KL. Molecular diagnostics of infectious diseases. Curr Opin Pediatr 2009;21:102-11.
817
6. Aitichou M, Henkens R, Sultana AM, Ulrich RG, Ibrahim MS. Detection of Staphylococcus aureus enterotoxin A and B genes with PCR-EIA and a hand-held electrochemical sensor. Mol Cell Probes 2004;18:373-7. 7. Liao JC, Mastali M, Gau V, Suchard MA, Moller AK, Bruckner DA, et al. Use of electrochemical DNA biosensors for rapid molecular identification of uropathogens in clinical urine specimens. J Clin Microbiol 2006;44: 561-70. 8. Yean CY, Kamarudin B, Ozkan DA, Yin LS, Lalitha P, Ismail A, et al. Enzyme-linked amperometric electrochemical genosensor assay for the detection of PCR amplicons on a streptavidin-treated screen-printed carbon electrode. Anal Chem 2008;80:2774-9. 9. de Tejada GM, Sanchez-Gomez S, Razquin-Olazaran I, Kowalski I, Kaconis Y, Heinbockel L, et al. Bacterial cell wall compounds as promising targets of antimicrobial agents I. Antimicrobial peptides and lipopolyamines. Curr Drug Targets 2012;13:1121-30. 10. Vollmer W, Blanot D, de Pedro MA. Peptidoglycan structure and architecture. FEMS Microbiol Rev 2008;32:149-67. 11. Lin YS, Tsai PJ, Weng MF, Chen YC. Affinity capture using vancomycin-bound magnetic nanoparticles for the MALDI-MS analysis of bacteria. Anal Chem 2005;77:1753-60. 12. Beveridge TJ, Graham LL. Surface layers of bacteria. Microbiol Rev 1991;55:684-705. 13. William DH, Bardsley B. The Vancomycin Group of Antibiotics and the Fight against Resistant Bacteria. Angew Chem Int Ed 1999;38:1172-93. 14. Batchelor M, Zhou D, Cooper MA, Abell C, Rayment T. Vancomycin dimer formation between analogues of bacterial peptidoglycan surfaces probed by force spectroscopy. Org Biomol Chem 2010;8:1142-8. 15. Kell AJ, Stewart G, Ryan S, Peytavi R, Boissinot M, Huletsky A, et al. Vancomycin-modified nanoparticles for efficient targeting and preconcentration of Gram-positive and Gram-negative bacteria. ACS NANO 2008;2:1777-88. 16. Merkoci A. Nanoparticle-based strategies for DNA, protein and cell sensors. Biosens Bioelectron 2010;26:1164-77. 17. Khanna VK. Nanoparticle-based sensors. Def Sci J 2008;58:608-16. 18. Suebsing R, Prombun P, Srisala J, Klatpathomchal W. Loop-mediated isothermal amplification combined with colorimetric nanogold for detection of microsporidian Enterocytozoon hepatopenael in penaeid shrimp. J Appl Microbiol 2013;114:1254-63. 19. Bakthavathsalam P, Rajendran VK, Mohammed JAB. A direction of Escherichia coli genomic DNA using gold nanoprobes. J Nanobiotechnol 2012;10:8-17. 20. Asian K, Lakowicz JR, Geddes CD. Nanogold-plasmon-resonancebased glucose sensing. Anal Biochem 2004;330:145-55. 21. Ploschner M, Cizmar T, Mazilu M, Falco AD, Dholakia K. Bidirectional optical sorting of gold nanoparticles. Nano Lett 2012;12:1923-7. 22. Tan WB, Zhang Y. Surface modification of gold and quantum dot nanoparticles with chitosan for bioapplications. J Biomed Mater Res A 2005;75:56-62. 23. Tsutsumi H, Ohkusa H, Park H, Takahashi T, Yuasa H, Mihara H. Gold nanoparticles conjugated with monosaccharide-modified peptide for lectin detection. Bioorg Med Chem Lett 2012;22:6825-7. 24. Wang CH, Lien KY, Wu JJ, Lee GB. A magnetic bead-based assay for the rapid detection of methicillin-resistant Staphylococcus aureus by using a microfluidic system with integrated loop-mediated isothermal amplification. Lab Chip 2011;11:1521-31. 25. Lagally ET, Scherer JR, Blazej RG, Toriello NM, Diep BA, Ramchandni M, et al. Integrated portable genetic analysis microsystem for pathogen/ infectious disease detection. Anal Chem 2004;76:3162-70. 26. Chang WH, Yang SY, Wang CH, Tsai MA, Wang PC, Chen TY, et al. Rapid isolation and detection of aquaculture pathogens in an integrated microfluidic system using loop-mediated isothermal amplification. Sensor Actuator B Chem 2013;180:96-106. 27. Lien KY, Lee SH, Tsai TJ, Chen TY, Lee GB. A microfluidic-based system using reverse transcription polymerase chain reactions for rapid detection of aquaculture diseases. Microfluid Nanofluid 2009;7: 795-806.
818
C.-H. Wang et al / Nanomedicine: Nanotechnology, Biology, and Medicine 10 (2014) 809–818
28. Wang CH, Lien KY, Wang TY, Chen TY, Lee GB. An integrated microfluidic loop-mediated-isothermal-amplification system for rapid sample pre-treatment and detection of viruses. Biosens Bioelectron 2011;26:2045-52. 29. Hill HD, Mirkin CA. The bio-barcode assay for the detection of protein and nucleic acid targets using DTT-induced ligand exchange. Nat Protoc 2006;1:324-36.
30. Weng CH, Lien KY, Yang SY, Lee GB. A suction-type, pneumatic microfluidic device for liquid transport and mixing. Microfluid Nanofluid 2011;10:301-10. 31. Wang CH, Chang CJ, Wu JJ, Lee GB. A new pathogen detection system by utilizing nanogold modified specific probe and vancomycin coated magnetic beads on an integrated microfluidic device. IEEE MEMS 2013:94-7.
