Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Nov 20.
Published in final edited form as: ACS Appl Mater Interfaces. 2017 Mar 30;9(14):12832–12840. doi: 10.1021/acsami.6b16571

Rapid Real-Time Antimicrobial Susceptibility Testing with Electrical Sensing on Plastic Microchips with Printed Electrodes

Mohammadali Safavieh †,‡, Hardik J Pandya †,||,‡, Maanasa Venkataraman †, Prudhvi Thirumalaraju †, Manoj Kumar Kanakasabapathy †, Anupriya Singh †, Devbalaji Prabhakar †, Manjyot Kaur Chug †, Hadi Shafiee †,§,*
PMCID: PMC5695042  NIHMSID: NIHMS919782  PMID: 28291334

Abstract

Rapid antimicrobial susceptibility testing is important for efficient and timely therapeutic decision making. Due to globally spread bacterial resistance, the efficacy of antibiotics is increasingly being impeded. Conventional antibiotic tests rely on bacterial culture, which is time-consuming and can lead to potentially inappropriate antibiotic prescription and up-front broad range of antibiotic use. There is an urgent need to develop point-of-care platform technologies to rapidly detect pathogens, identify the right antibiotics, and monitor mutations to help adjust therapy. Here, we report a biosensor for rapid (<90 min), real time, and label-free bacteria isolation from whole blood and antibiotic susceptibility testing. Target bacteria are captured on flexible plastic-based microchips with printed electrodes using antibodies (30 min), and its electrical response is monitored in the presence and absence of antibiotics over an hour of incubation time. We evaluated the microchip with Escherichia coli and methicillin-resistant Staphylococcus aureus (MRSA) as clinical models with ampicillin, ciprofloxacin, erythromycin, daptomycin, gentamicin, and methicillin antibiotics. The results are compared with the current standard methods, i.e. bacteria viability and conventional antibiogram assays. The technology presented here has the potential to provide precise and rapid bacteria screening and guidance in clinical therapies by identifying the correct antibiotics for pathogens.

Keywords: antibiotic susceptibility test, electrical sensing, antibiotic resistant pathogen, flexible electronics, screen-printed electrodes

Graphical abstract

graphic file with name nihms919782u1.jpg

INTRODUCTION

Rapid evolution of bacterial resistance toward antibiotics and the slowdown in progress of developing new antibiotics are critical threats to public health in the foreseeable future.1 Every year, at least 2 million people in the United States acquire serious infections of bacteria that are resistant to one or more of the antibiotics designed to treat those infections. As a direct result of these antibiotic-resistant infections, at least 23 000 Americans die each year.2 Current conventional antibiotic testing involves multiple bacterial culture steps such as disk diffusion, Luria–Bertani (LB) broth serial dilutions, and antimicrobial chemical reactions to detect bacteria and test its antibiotic susceptibility.3 These assays provide vital information on the impact of antibiotic on bacteria and suitable guidance to select the proper antibiotic regimen but are labor-intensive and time-consuming (24 h to 5 days).4,5 Patients are usually treated with empirical antibiotics to cover most probable pathogens, which can result in incorrect initial therapy, leading to further infection and increased mortality. Molecular diagnostic assays that target DNA and genetic mutations associated with drug-resistance require preknowledge with respect to the genetic basis of antibiotic resistance for a target pathogen.6–11 Polymerase chain reaction (PCR)-based assays have limitations, including complex nucleic acid amplification steps, false positive and negative results due to amplification of complex matrix media, and enzymatic inhibition from blood and urine components.12,13 Numerous other techniques have also been developed for antibiotic susceptibility testing such as nucleic acid staining,8 nanoliter optical sensing of stochastic confinement,14 electrochemical bacteria respiration,15–20 motion detector using atomic force microscopy,21 optical detection of bacterial growth,22–25 and mass monitoring of bacterial growth using cantilevers.26 Almost all of these techniques require bulky and complex equipment to perform sensing. The electrochemical sensing-based method requires time-consuming redox concentration optimization for electrochemical respiration monitoring. Electrical sensing modality is a robust, label-free, rapid, and easy-to-use technique that has been employed for various biosensing applications such as bacteria detection27 and monitoring of bacteria growth in various media.28,29 The electronic circuit design for electrical sensing is simple. Here, we report a rapid real-time antimicrobial susceptibility test using electrical sensing of captured bacteria on plastic microchips with printed electrodes in the presence and absence of antibiotics (Figure 1). Figures 1A and B show the schematic and actual photograph of the biosensor. The components of the microchip are shown in detail in Figure S1 (Supporting Information). Components of the microchip for antibiotic testing are (A) plastic substrate, (B) screen-printed carbon electrodes on the back side of the plastic substrate for on-chip sample temperature control and sensing, (C) screen-printed silver interdigitated electrodes on the front side of the plastic substrate, and (D) attached double sided adhesive film (DSA) and poly(methyl methacrylate) (PMMA) sheets on the front side of the plastic substrate to create a closed microchip system. The presented method relies on the significant effect of antibiotics on the growth of captured bacteria on-chip. We first evaluated our hypothesis by monitoring the electrical response of bacteria in the presence and absence of antibiotics on nonfunctionalized chips (Figure 1C). Our results demonstrated that the electrical signal can significantly change due to the presence of effective antibiotics over a period of 1 h.

Figure 1.

Figure 1

Open in a new tab

Experimental design. (A) Schematic diagram of the biosensor with a microfluidic channel. (B) Photograph of a biosensor with interdigitated electrodes, PMMA microfluidic channel, integrated heater, and temperature sensor. (C) Antibiotic susceptibility measurement using label-free electrical sensing on a biosensor without surface modification. (D) On-chip bacteria capture and isolation from whole blood using affinity-based sensing and measurement of bacterial antibiotic susceptibility using electrical sensing.

We evaluated the microchip using E. coli and MRSA as two clinical models. We also evaluated the ability of the microchip to selectively isolate MRSA from whole blood using antibody and tested its electrical response against gentamicin antibiotic (Figure 1D).

2. MATERIALS AND METHODS

2.1. Materials and Cell Culture

LB agar, ampicillin (A9393), erythromycin (E6376), ciprofloxacin (17850), methicillin (51387), daptomycin (D2446), gentamicin (G1914), Tris (2-carboxyethyl) phosphine (TCEP), and LB agar plates (L5542) were purchased from Sigma-Aldrich (St. Luis, MO). LB broth (BP09723) was bought from Fischer Scientific (Pittsburgh, PA). Escherichia coli K12 (ATCC 700891) and methicillin-resistant SRM551 were cultured on a shaker overnight (12 h) at 37 °C in 2% LB broth media. The bacteria pellet was gently separated using a centrifuge for 5 min at 3000 rpm, and the pellet was resuspended in fresh LB broth media. Concentrations of bacteria were set to a constant optical density (OD) value, which was measured by UV–vis from NanoDrop, ND-1000 (Thermo Scientific, Wilmington, DE). The OD of the samples (E. coli and MRSA in LB broth media) was set to 0.125, which corresponds to 108 CFU/ml.

2.2. Affinity-Based Bacteria Capturing

An affinity-based approach was used to confirm specific and efficient capturing of MRSA on-chip. Anti-MRSA antibody (ab62742, Abcam, Cambridge, MA) with 10 μg/mL concentration was injected into the microchannel (30 μL) and incubated for 30 min. The channel was then washed three times using 30 μL 1 × PBS. Then, 30 μL of MRSA-spiked whole blood was added into the chip and incubated for 30 min to allow bacteria capture on-chip. The microchannel was then washed 3 times using 30 μL PBS.

2.3. Antibiotic Susceptibility Assay Viability

Cultured MRSA and E. coli were diluted to 103 CFU/μL and incubated with different antibiotics of clinically relevant concentrations ranging from 0.1 to 100 μg/mL.4 The optical density (OD) value at 600 nm was measured using plate-reader Synergy2 (BioTek, Winooski, VT). After 24 h, the OD was measured, and viability of the bacteria was calculated based on differences between the ODs before and after incubation.4

2.4. Impedance Analyzer

The impedance of the bacterial pathogen (with and without antibiotics) placed on the interdigitated electrodes was measured using an impedance analyzer (LCR8110G, GW Instek, CA) at 1 V for frequencies between 100 Hz and 1 MHz. From our previous studies,9,30,31 we found that the change in the impedance is maximum at 1 kHz. Additionally, we found a high signal-to-noise ratio for 1 kHz. Thus, we selected 1 kHz as the frequency to study the response of bacterial pathogens. For on-chip antibiotic susceptibility utilizing impedance monitoring, the impedance of the bacteria sample in presence and absence of antibiotics was monitored at 1 kHz for 60 min with scanning interval of 10 min and was normalized with respect to the initial impedance magnitude (t = 0).

2.5. Biosensor Fabrication on a Plastic Substrate

A microchip involving interdigitated electrodes (IDEs), a heater, a temperature sensor, and PMMA/DSA microfluidic channels was fabricated using screen printing32 and laser cutting (VLS 2.3 laser printer, Universal Laser System Inc.). The size of the device was 20 × 20 mm (length × width). The active area of the device was 10 × 4 mm (length × width) with pattern interdigitated electrodes (0.6 mm width and 0.8 mm spacing). The schematic diagram of the heater and biosensor are shown in Figures 1A and B. The screen was designed using CorelDRAW©, and windows were created for IDEs and the heater on the DSA film using a laser cutter. One-hundred micrometer clear flexible plastic sheets (3M, Maplewood, Minnesota, United States) were used as substrates for electrode printing. On one side of the substrate, the screen for IDEs was fixed, and silver (Engineered Materials System, Inc., < 0.015 Ω/m) and silicone (DAP Products Inc., Baltimore, MD) paste (4:1) was screen printed to form the IDEs. The plastic substrate with IDEs was dried at 90 °C for 30 min. On the other side of the substrate, a heater and temperature sensor were developed in the same way using a low resistive conducting carbon ink (<10 Ω/m, Engineered Materials System, Inc.) (Figure 1B). The channel dimensions for loading the biological fluid with the bacterial pathogen were 13.85 × 3.43 × 0.7 mm (length × width × depth). To create the microfluidic channel, one of the protective layers of DSA was peeled off and attached to PMMA. PMMA/DSA sheets were cut using a laser cutter with power, scan speed, and pulse per inch rate of 24 W, 5 mm/s, and 1000 pulses/inch, respectively. The protective layer from one side of DSA was peeled off before attaching it to the sensor. The chip reservoir was covered with a plastic tape during the experiments.

2.6. Statistical Analysis

Statistical analysis on all results was performed using one-way ANOVA with the test for single column comparison with respect to the control sample. Statistical signiflcance threshold was set at 0.05 (p < 0.05). Error bars represent a standard deviation of the mean (n = 3). An asterisk shows statistically significant impedance magnitude change compared to all other data sets.

3. RESULTS AND DISCUSSION

3.1. On-Chip Microheater

To achieve 37 °C for on-chip bacteria incubation, the electrical characterization of the screen-printed heater was performed (Figure 2). We observed a linear response for change in temperature of the heater with respect to increase in voltage (R2 = 0.95) (Figure 2A). The change in resistance was converted to temperature, which was measured by an on-chip temperature sensor. The heater required 9 V to reach a stable temperature of 37 °C. To evaluate the response of the heater, a constant voltage was applied until the heater’s temperature reached an asymptotic value. After waiting for 60 s, the voltage was brought down to 0 V, which resulted in the heater reaching room temperature (Figure 2B). We observed that the turn on and off times for the heater to reach 37 °C and room temperature (21 °C) were 100 and 200 s, respectively. The heater was stable at 37 °C for 14 h at 9 V (Figure 2C) (n = 3). The microchannels were empty during these experiments.

Figure 2.

Figure 2

Open in a new tab

Electrical characterization of a screen printed heater on a flexible plastic substrate. (A) Plot shows a linear relationship between applied voltage and temperature with R2 of 0.954 in an empty microchannel. (B) Plot shows the response of the heater for different voltages. (C) Temperature stability (37 °C) of the heater for 14 h at constant voltage of 9 V in an empty channel. Error bars represent standard error of the mean (n = 3), p < 0.005.

3.2. Electrical Response of Bacteria On-Chip in the Presence and Absence of Antibiotics

The electrical response of the bacteria in the presence and absence of antibiotics was first evaluated using nonfunctionalized chips. In these proof-of-concept experiments, we wanted to monitor the electrical response of the bacteria in the presence and absence of antibiotics over time without considering the complexity of on-chip bacteria capture and isolation. Therefore, bacteria-spiked LB broth samples were injected into nonfunctionalized chips, and their impedance response was monitored in the presence and absence of antibiotics. The antibiotic susceptibility assay was performed using six antibiotics: ampicillin, ciprofloxacin, and erythromycin on E. coli and daptomycin, gentamicin, and methicillin on MRSA. For urinary tract infections (UTI), the concentration of bacteria >105 CFU/ml is considered clinically relevant.5 The impedance magnitudes of samples (30 μL) at any time during the experiments were normalized with respect to the initial impedance magnitudes of the samples (t = 0). To be certain that the antibiotic is fully effective, we selected a higher concentration of the bacteria (106 CFU/mL). Figures 3A, E, and I show the real-time change in impedance magnitude from t = 0 to 60 min at 1 kHz and 1 V. We observed a significant difference between the electrical response of spiked and control samples at t = 30 min and afterward. We chose t = 30 min for statistical analysis between the normalized impedance magnitudes because we intended to develop a rapid diagnostics for antimicrobial susceptibility testing. Figures 3B, F, and J show the normalized impedance magnitude of E. coli-spiked samples at t = 30 min when ampicillin, ciprofloxacin, and erythromycin were used, respectively. The control samples were antibiotic-free E. coli-spiked samples (106 CFU/ml). These results showed that the normalized impedance magnitudes of E. coli-spiked bacteria samples were significantly different than those of control samples at ampicillin concentrations of 10 and 100 μg/mL (Figure 3B, n = 3, p < 0.001), while 1 and 0.1 μg/mL did not show any statistical significance with respect to control (Figure 3B, n = 3, p > 0.05). The cell viabilities in the presence of 100 μg/mL, 10 μg/mL, 1 μg/mL, and 0.1 μg/mL ampicillin concentration were 4.5% ± 1, 20% ± 17, 40% ± 8, and 95% ± 4, respectively which was significantly different than the cell viability of control samples (Figure 3C, p < 0.001, n = 3) and is in agreement with minimum inhibitory growth (MIG) of ampicillin.4 The normalized impedance magnitudes of E. coli-spiked samples, when ciprofloxacin at concentrations ranging from 0.1 μg/mL to 100 μg/mL was used, were significantly different than control samples (Figure 3E, F, n = 3, p < 0.001). The cell viability in the presence of ciprofloxacin at concentrations of 100, 10, 1, and 0.1 μg/mL were 7.4 ± 4.6, 0.9 ± 0.9, 8.9 ± 8.9, and 6.8 ± 2.2%, respectively, which was significantly different than the cell viability of control samples (Figure 3G, n = 3, p < 0.001).4 Similarly, we observed significant electrical response for E. coli samples in the presence of erythromycin with concentrations ranging from 0.1 to 100 μg/mL compared to that in antibiotic-free control samples (Figure 3J, n = 3, p < 0.001). The real-time change in impedance magnitude of E. coli for erythromycin shows an increase in normalized impedance magnitude for 100 and 10 μg/mL, which could be due to the low pH ~ 6 of erythromycin that provides a higher positive charge. The interaction between positively charged erythromycin with the negatively charged bacteria cell membrane may neutralize the charge balance of the solution, which consequently increases the impedance, as shown in Figure 3I. The minimum inhibitory concentration (MIC) against erythromycin was less than 0.1 μg/mL, as E. coli viability in the presence of erythromycin with concentrations ranging from 0.1 to 100 μg/mL was negligible and significantly different from the viability of E. coli without erythromycin (Figure 3K, n = 3, p < 0.001). The viability of E. coli bacteria in the presence of erythromycin with concentrations of 100, 10, 1, and 0.1 μg/mL were 4.3 ± 1.1, 10.3 ± 8.9, 38.9 ± 15.1, and 34 ± 5.3%, respectively (Figure 3K). The antibiotic susceptibility of E. coli against ampicillin, ciprofloxacin, and erythromycin using the disc diffusion method (24 h) are in agreement with the cell viability and electrical sensing results (Figures 3D, H, and L). We also performed similar experiments with samples spiked with MRSA and evaluated its antibiotic susceptibility in the presence of daptomycin, gentamicin, and methicillin (Figure 4). When the antibiotic concentration in the pathogen sample was increased, the rate of bacteria production decreases and there will therefore be less charge in the sample solution. The reduction of the charge in the sample yields a small change in normalized impedance of samples with antibiotics compared with samples without antibiotics within the culturing time, as shown in Figures 4A, E, and I. The impedance magnitude of the MRSA-spiked samples in the presence of daptomycin with the concentration of 0.1 μg/mL was not significantly different than control samples (Figure 4B, n = 3, p > 0.05), however, the impedance magnitudes of the MRSA-spiked samples in the presence of daptomycin with the concentrations of 1, 10, and 100 μg/mL were significantly different than those of the control at t = 30 min (Figure 4B, n = 3, p < 0.05). The cell viability of MRSA in the presence of 100, 10, and 1 μg/mL daptomycin were 8.7 ± 5.6,10.2 ± 8.5, and 14.7 ± 12.5%, respectively, which were significantly different than the viability of MRSA-spiked samples with 0.1 μg/mL daptomycin concentration and control samples (Figure 4C, n = 3, p < 0.001). The impedance magnitudes of the MRSA-spiked samples in the presence of gentamicin with the concentrations of 0.1 and 10 μg/mL were not significantly different than control samples (Figure 4F, n = 3, p > 0.05). The cell viabilities of MRSA in the presence of 1 and 0.1 μg/mL gentamicin were 60 ± 13 and 82.7 ± 4.9%, respectively, which reflect the fact that gentamicin is not effective for concentrations below 10 μg/mL (Figure 4G). These results also showed that the impedance magnitudes of MRSA-spiked samples in the presence of methicillin with concentrations ranging from 0.1 to 100 μg/mL were not significantly different than those of control samples (Figure 4J, n = 3, p > 0.05). The cell viabilities of MRSA-spiked samples in the presence of 100, 10, 1, and 0.1 μg/mL methicillin were 95.7 ± 4.2, 93.2 ± 1.7, 92.9 ± 3.6, and 92.5 ± 0.5%, respectively, which were not significantly different than those of antibiotic-free control samples (Figure 4K, n = 3, p > 0.05). This observation may be attributed to the fact that the methicillin MIC is >400 μg/mL for MRSA SRM551 samples.33 These electrical sensing readouts and cell viability results were in agreement with the disc diffusion-based results (Figures 4D, H, and L).

Figure 3.

Figure 3

Open in a new tab

Electrical response of E. coli in the presence of ampicillin, ciprofloxacin, and erythromycin over a period of 1 h. (A) Real time impedance measurement of E. coli in the presence of ampicillin. (B) Statistical analysis of the effect of ampicillin on E. coli (p = 0.0055, n = 3). (C) Viability test results of E. coli in the presence of ampicillin after an overnight incubation by measuring their OD at 600 nm wavelength, (p < 0.0001, n = 3). (E) Real time impedance measurement of E. coli in the presence of ciprofloxacin. (F) Statistical analysis of the effect of ciprofloxacin on E. coli (p = 0.1915, ns, n = 3). (G) Viability test results of E. coli in the presence of ciprofloxacin (p = 0.0850, n = 3). (I) Real time impedance measurement of E. coli in the presence of erythromycin. (J) Statistical analysis of the effect of erythromycin on E. coli (p = 0.0055, n = 3). (K) Viability test results of E. coli in the presence of erythromycin (p = 0.0850, n = 3). (D, H, and L) Antibiotic susceptibility of E. coli against ampicillin, ciprofloxacin, and erythromycin, respectively, using the disc diffusion method (24 h).

Figure 4.

Figure 4

Open in a new tab

Electrical response of MRSA in the presence of daptomycin, gentamicin, and methicillin over a period of 1 h. (A) Real time impedance measurement of MRSA in the presence of daptomycin. (B) Normalized impedance magnitude of MRSA-spiked samples in the presence of daptomycin at t = 30 min (p = 0.05, n = 3). (C) Viability test results of MRSA in the presence of daptomycin after an overnight incubation by measuring their OD at 600 nm wavelength, (p < 0.0001, n = 3). (E) Real time impedance measurement of MRSA in the presence of gentamicin. (F) Normalized impedance magnitude of MRSA-spiked samples in the presence of gentamicin at t = 30 min (p = 0.1915, ns, n = 3). (G) Viability test results of MRSA in the presence of gentamicin (p = 0.0850, n = 3). (I) Real time impedance measurement of MRSA in the presence of methicillin. (J) Normalized impedance magnitude of MRSA-spiked samples in the presence of methicillin at t = 30 min (p = 0.0055, n = 3). (K) Viability test results of MRSA in the presence of methicillin (p = 0.0850, n = 3). (D, H, and L) Antibiotic susceptibility of MRSA against daptomycin, gentamicin, and methicillin, respectively, using the disc diffusion method (24 h).

3.3. On-Chip Bacteria Capture and Antibiotic Susceptibility Testing

We also evaluated the ability of the microchip to selectively capture and isolate bacteria from whole blood and measured its electrical response in the presence of antibiotics. We selected MRSA-spiked whole blood samples with bacteria concentrations of 105, 106, and 107 CFU/mL. Figure 5A shows the actual photograph of the surface-modified microchip with immobilized anti-MRSA antibody loaded with MRSA-spiked whole blood (i) and the actual device after washing with PBS (30 μL) three times (ii). Figure 5B shows bacteria recovery (43 ± 2%) over bacterial loads (MRSA) (107 and 108 CFU/mL) spiked into 1 mL of whole blood. To calculate the bacteria recovery rate using our functionalized microchips, we measured bacteria concentration in (i) MRSA-spiked whole blood sample used for this study and (ii) a washed sample after releasing the captured bacteria on-chip by measuring the OD values of the sample. Captured bacteria were released after the washing step by cleaving the antibody using TCEP (10 μg/mL, 30 μL), a thiol-free compound that is highly effective at reducing protein and disulfide bonds in antibody structure.34,35 For performing on-chip bacteria capture and antibiotic susceptibility testing, MRSA-spiked blood was loaded into the microchip, and the electrical response of the captured bacteria was monitored at different concentrations of gentamicin (Figure 5C). It should be noted that the electrical response of MRSA in non-functionalized microchips (Figure 4F) and microchips functionalized with anti-MRSA antibody (Figure 5C) demonstrated a similar result that 100 and 10 μg/mL gentamicin concentration significantly affected MRSA viability and its electrical response compared to antibiotic-free MRSA-spiked samples. A same approach can be implemented for efficient capturing of E. coli in urine.

Figure 5.

Figure 5

Open in a new tab

On-chip bacteria capture and antibiotic susceptibility testing. (A) Photograph of the fabricated microchip: (i) surface modified microchip loaded with MRSA-spiked whole blood (106 CFU/mL) (incubated for 30 min) and (ii) microchip washed with PBS. (B) Histogram plot indicating the bacterial load and recovery (>43%) for the two spiked blood samples used in the study (sample 1, 108 CFU/mL; sample 2, 107 CFU/mL) (n = 3). Ten micrograms per microliter anti-MRSA antibody was used for chip functionalization. (C) Electrical response of the captured MRSA (105 to 107 CFU/mL) in the presence of gentamicin (0.1–100 μg/mL). Error bars represents standard error of mean (n = 3). MRSA-spiked whole blood without antibiotic was used as control.

These results demonstrated the ability of the microchip to detect antibiotic susceptibility of bacteria spiked in LB broth and whole blood samples. This method can also be used for bacteria susceptibility testing in urine samples. We observed that the impedance magnitude changes of purified and unpurified urine samples spiked with E. coli and MRSA were lower compared to those in spiked LB broth and PBS samples (Figure 6, n = 3, P < 0.001). The normalized impedance magnitude of E. coli-spiked samples of bacteria was 70 ± 2% in LB broth, 30 ± 3% in PBS, 16 ± 8% in unpurified urine, and 15 ± 8% in purified urine (Figure 6A). The normalized impedance magnitude of MRSA-spiked samples was 58 ± 7% in LB broth, 49 ± 1% in PBS, 37 ± 6% in unpurified urine, and 22 ± 4% in purified urine (Figure 6B).

Figure 6.

Figure 6

Open in a new tab

Growth of microorganisms in different media was monitored by measuring their impedance magnitude over a period of 1 h. Electrical response of E. coli (A) and MRSA (B) during bacterial growth in LB broth, 1× PBS, and purified and unpurified urine.

We also developed a portable impedance analyzer, as shown in Figure S2A. Figures S2B, C, and D show the electronic module with a chip interfacing area, a sensor with silver interdigitated electrodes, a screen-printed carbon heater, and a conceptual plot showing the effect of antibiotics on the bacteria, respectively (Supporting Information). The block diagram and circuit of the portable impedance analyzer are shown in Figures S3A and B, respectively. The AD5933 is a high precision impedance converter IC integrated with an on-board frequency generator, 12-bit analog-to-digital converter (ADC), and an onboard DSP engine. The on-board frequency generator allows an external complex impedance to be excited with a known frequency. (i) The sample in the microfluidic device was excited, and the impedance magnitude was measured. The desired excitation frequency or a series of excitation frequencies was programmed into the IC. The response signal was sampled by the on-board ADC, and a discrete Fourier transform (DFT) was performed by the on-board DSP. The magnitude and relative phase of the impedance at each frequency point along the sweep were calculated, which was done off-chip using the real and imaginary register contents, which were read from the serial I2C interface. (ii) Peripheral module impedance analyzer (PmodIA) was interfaced with ATMEGA328 (a low-power CMOS 8-bit microcontroller). (iii) The ATmega328 was used to send commands and receive data from the PmodIA using I2C communication. The received data was sent to a PC by Universal Asynchronous Receiver-Transmitter (UART). (iv) To establish communication between the PC and ATmega328, a single-chip USB to UART bridge IC (CP2102) was used. (v) We developed a Python code that receives data at the corresponding communication (COM) port. The user can set up the initial parameters such as starting frequency, number of settling cycles, step size, number of increments gain, etc. in the graphical user interface (GUI) of the Java program. The data were stored in a CSV file format for further analysis. Our portable impedance analyzer has the ability to plot the data by pressing the plot button in the GUI. We evaluated the performance of the portable impedance analyzer by measuring the data obtained from PBS solution (1, 0.1, and 0.05%), E. coli, and MRSA using the portable impedance analyzer and comparing it with the data obtained from the impedance analyzer (Figure S4, Supporting Information).

CONCLUSION

Here, we report a rapid (<90 min), label-free antimicrobial susceptibility testing through real-time monitoring of bacterial growth in the presence and absence of antibiotics. This device can further be used to provide proper antibiotics in <90 min to patients suffering from UTI,36,37 where the main causes are E. coli and Staphylococcus aureus pathogens. The advantage of this technique over commonly used electrochemical monitoring for bacterial metabolic activity is that it does not require any redox molecule such as ferrocyanide or resazurin or redox concentration optimization, which significantly reduces the effort on assay optimization. We also performed an on-chip isolation of the target bacteria from the spiked whole blood with clinically relevant bacterial concentrations using antibodies and monitored the captured bacteria’s electrical response in the presence and absence of antibiotics for antimicrobial susceptibility testing. We did not observe any sample evaporation-related issues in our study, as the chip reservoirs were covered with plastic tapes during the experiments. In this study, we chose E. coli and MRSA as two clinical models to evaluate the performance of the reported microchip technology. This electrical sensing-based mechanism for antibiotic testing can be potentially used for a broad range of bacteria, including Psudomonas aeruginosa,38 vancomycin-resistant S. aureus (VRSA),39 extended spectrum beta-lactamase (ESBL), vancomycin-resistant Enterococcus (VRE), and multidrug-resistant Acinetobacter baumannii (MRAB).40,41 This method can be potentially used for detecting antimicrobial susceptibility testing for different strains of bacteria as long as there are different specific antibodies available for different bacterial strains. Further studies are needed to evaluate the specificity of electrical response of different strains of bacteria in case there is a universal antibody available for on-chip bacteria capture.

Supplementary Material

SI

Acknowledgments

Research reported in this publication was partially supported by the National Institute of Allergy and Infectious Disease of the National Institute of Health under Award R01AI118502, the Brigham and Women’s Hospital and Harvard Medical School through Bright Futures Prize and Innovation Evergreen Award, and the National Engineering Research Council of Canada (NSERC) through NSERC Postdoctoral fellowship (M.S.). The authors also extend their gratitude to Karan Dhingra for performing preliminary experiments. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

Notes: The authors declare no competing financial interest.

Supporting Information

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b16571.

Microchip fabrication details, details of the portable impedance analyzer and its electronic circuit diagram, and evaluation of the portable impedance analyzer by comparing the results with the results obtained by an LCR meter (PDF)

References

  • 1.Chait R, Craney A, Kishony R. Antibiotic interactions that select against resistance. Nature. 2007;446:668–671. doi: 10.1038/nature05685. [DOI] [PubMed] [Google Scholar]
  • 2.Antibiotic Resistance Threat in the United States. Centers for Disease Control and Prevention; [accessed November 12, 2016]. http://www.cdc.gov/drugresistance/pdf/ar-threats-2013-508.pdf. [Google Scholar]
  • 3.Bush K, Courvalin P, Dantas G, Davies J, Eisenstein B, Huovinen P, Jacoby GA, Kishony R, Kreiswirth BN, Kutter E, Lerner SA, Levy S, Lewis K, Lomovskaya O, Miller JH, Mobashery S, Piddock LJ, Projan S, Thomas CM, Tomasz A, Tulkens PM, Walsh TR, Watson JD, Witkowski J, Witte W, Wright G, Yeh P, Zgurskaya HI. Tackling antibiotic resistance. Nat Rev Microbiol. 2011;9:894–896. doi: 10.1038/nrmicro2693. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Besant JD, Sargent EH, Kelley SO. Rapid electrochemical phenotypic profiling of antibiotic-resistant bacteria. Lab Chip. 2015;15:2799–2807. doi: 10.1039/c5lc00375j. [DOI] [PubMed] [Google Scholar]
  • 5.Kothari A, Morgan M, Haake DA. Emerging technologies for rapid identification of bloodstream pathogens. Clin Infect Dis. 2014;59(2):272–278. doi: 10.1093/cid/ciu292. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Safavieh M, Ahmed M, Sokullu E, Ng A, Braescu L, Zourob M. A simple cassette as point-of-care diagnostic device for naked-eye colorimetric bacteria detection. Analyst. 2014;139(2):482–487. doi: 10.1039/c3an01859h. [DOI] [PubMed] [Google Scholar]
  • 7.Safavieh MA, Minhaz Uddin A, Tolba M, Zourob M. Microfluidic electrochemical assay for rapid detection and quantification of Escherichia coli. Biosens Bioelectron. 2012;31(1):523–528. doi: 10.1016/j.bios.2011.11.032. [DOI] [PubMed] [Google Scholar]
  • 8.Safavieh M, Kanakasabapathy MK, Tarlan F, Ahmed MU, Zourob M, Asghar W, Shafiee H. Emerging Loop-mediated Isothermal Amplification-based Microchip and Microdevice Technologies for Nucleic Acid Detection. ACS Biomater Sci Eng. 2016;2(3):278–294. doi: 10.1021/acsbiomaterials.5b00449. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Safavieh M, Kaul V, Khetani S, Singh A, Dhingra K, Kanakasabapathy MK, Draz M, Memic A, Kuritzkes D, Shafiee H. Paper Microchip with Graphene-modified Silver Nanocomposite Electrode for Electrical Sensing of Microbial Pathogens. Nanoscale. 2017;9:1852–1861. doi: 10.1039/c6nr06417e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Wang C, Lien K, Wu J, Lee G. 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(8):1521–1531. doi: 10.1039/c0lc00430h. [DOI] [PubMed] [Google Scholar]
  • 11.Safavieh M, Ahmed MU, Zourob M. Colorimetric assay for urinary track infection disease diagnostic on flexible substrate. Photonics North, International Society for Optics and Photonics; Montreal, QC: 2012. pp. 84120A–84120A-6. [Google Scholar]
  • 12.Ince J, McNally A. Development of rapid, automated diagnostics for infectious disease: advances and challenges. Expert Rev Med Devices. 2009;6(6):641–651. doi: 10.1586/erd.09.46. [DOI] [PubMed] [Google Scholar]
  • 13.Weile J, Knabbe C. Current applications and future trends of molecular diagnostics in clinical bacteriology. Anal Bioanal Chem. 2009;394(3):731–742. doi: 10.1007/s00216-009-2779-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Gosiewski T, Brzychczy-Wloch M, Pietrzyk A, Sroka A, Bulanda M. Evaluation of the activity of thermostable DNA polymerases in the presence of heme, as a key inhibitor in the real time PCR method in diagnostics of sepsis. Acta Biochim Pol. 2013;60(4):603–606. [PubMed] [Google Scholar]
  • 15.Roth BL, Poot M, Yue ST, Millard PJ. Bacterial Viability and Antibiotic Susceptibility Testing with SYTOX Green Nucleic Acid Stain. Appl Environ Microbiol. 1997;63(6):2421–2431. doi: 10.1128/aem.63.6.2421-2431.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Boedicker JQ, Li L, Kline TR, Ismagilov RF. Detecting bacteria and determining their susceptibility to antibiotics by stochastic confinement in nanoliter droplets using plug-based microfluidics. Lab Chip. 2008;8:1265–1272. doi: 10.1039/b804911d. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ertl P, Robello E, Battaglini F, Mikkelsen SR. Rapid Antibiotic Susceptibility Testing via Electrochemical Measurement of Ferricyanide Reduction by Escherichia coli and Clostridium sporogenes. Anal Chem. 2000;72:4957–4964. doi: 10.1021/ac0003596. [DOI] [PubMed] [Google Scholar]
  • 18.Mann TS, Mikkelsen SR. Antibiotic Susceptibility Testing at a Screen-Printed Carbon Electrode Array. Anal Chem. 2008;80(3):843–848. doi: 10.1021/ac701829c. [DOI] [PubMed] [Google Scholar]
  • 19.Ertl P, Wagner M, Corton E, Mikkelsen SR. Rapid identification of viable Escherichia coli subspecies with an electrochemical screen-printed biosensor array. Biosens Bioelectron. 2003;18:907–916. doi: 10.1016/s0956-5663(02)00206-3. [DOI] [PubMed] [Google Scholar]
  • 20.Chotinantakul K, Suginta W, Schulte A. Advanced Amperometric Respiration Assay for Antimicrobial Susceptibility Testing. Anal Chem. 2014;86(20):10315–10322. doi: 10.1021/ac502554s. [DOI] [PubMed] [Google Scholar]
  • 21.Longo G, Alonso-Sarduy L, Marques Rio L, Bizzini A, Trampuz A, Notz J, Dietler G, Kasas S. Rapid detection of bacterial resistance to antibiotics using AFM cantilevers as nano-mechanical sensors. Nat Nanotechnol. 2013;8:522–526. doi: 10.1038/nnano.2013.120. [DOI] [PubMed] [Google Scholar]
  • 22.Li B, Qiu Y, Glidle A, McIlvenna D, Luo Q, Cooper J, Shi H-C, Yin H. Gradient Microfluidics Enables Rapid Bacterial Growth Inhibition Testing. Anal Chem. 2014;86(6):3131–3137. doi: 10.1021/ac5001306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Deiss F, Funes-Huacca ME, Bal J, Tjhung KF, Derda R. Antimicrobial susceptibility assays in paper-based portable culture devices. Lab Chip. 2014;14:167–171. doi: 10.1039/c3lc50887k. [DOI] [PubMed] [Google Scholar]
  • 24.Mohan R, Mukherjee A, Sevgen SE, Sanpitakseree C, Lee J, Schroeder CM, Kenis PJA. A multiplexed microfluidic platform for rapid antibiotic susceptibility testing. Biosens Bioelectron. 2013;49:118–125. doi: 10.1016/j.bios.2013.04.046. [DOI] [PubMed] [Google Scholar]
  • 25.Chiang Y-L, Lin C-H, Yen M-Y, Sue Y-D, Chene S-J, Chen H-F. Innovative antimicrobial susceptibility testing method using surface plasmon resonance. Biosens Bioelectron. 2009;24(7):1905–1910. doi: 10.1016/j.bios.2008.09.020. [DOI] [PubMed] [Google Scholar]
  • 26.Gfeller KY, Nugaeva N, Hegner M. Rapid Biosensor for Detection of Antibiotic-Selective Growth of Escherichia coli. Appl Environ Microbiol. 2005;71(5):2626–2631. doi: 10.1128/AEM.71.5.2626-2631.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Yang L, Bashir R. Electrical/electrochemical impedance for rapid detection of foodborne pathogenic bacteria. Biotechnol Adv. 2008;26:135–150. doi: 10.1016/j.biotechadv.2007.10.003. [DOI] [PubMed] [Google Scholar]
  • 28.Varshney M, Li Y. Interdigitated array microelectrodes based impedance biosensors for detection of bacterial cells. Biosens Bioelectron. 2009;24:2951–2960. doi: 10.1016/j.bios.2008.10.001. [DOI] [PubMed] [Google Scholar]
  • 29.Couniot NV, Van Overstraeten-Schlogel N, Afzalian A, Francis LA, Flandre D. Capacitive Biosensing of Bacterial Cells: Sensitivity Optimization. IEEE Sens J. 2016;16(3):586–595. [Google Scholar]
  • 30.Shafiee H, Kanakasabapathy MK, Juillard F, Keser M, Sadasivam M, Yuksekkaya M, Hanhauser E, Henrich TJ, Kuritzkes DR, Kaye KM, Demirci U. Printed Flexible Plastic Microchip for Viral Load Measurement through Quantitative Detection of Viruses in Plasma and Saliva. Sci Rep. 2015;5:9919. doi: 10.1038/srep09919. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Shafiee H, Jahangir M, Inci F, Wang S, Willenbrecht R, Giguel FF, Tsibris A, Kuritzkes DR, Demirci U. Acute on-chip hiv detection through label-free electrical sensing of viral nano-lysate. Small. 2013;9(15):2553–2563. doi: 10.1002/smll.201202195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Ahmed MU, Hossain MM, Safavieh M, Wong YL, Rahman IA, Zourob M, Tamiya E. Toward the development of smart and low cost point-of-care biosensors based on screen printed electrodes. Crit Rev Biotechnol. 2016;36(3):495–505. doi: 10.3109/07388551.2014.992387. [DOI] [PubMed] [Google Scholar]
  • 33.Murakami K, Tomasz A. Involvement of Multiple Genetic Determinants in High-Level Methicillin Resistance in Staphylococcus aureus. J Bacteriol. 1989;171(2):874–879. doi: 10.1128/jb.171.2.874-879.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Liu P, O’Mara BW, Warrack BM, Wu W, Huang Y, Zhang Y, Zhao R, Lin M, Ackerman MS, Hocknell PK, Chen G, Tao L, Rieble S, Wang J, Wang-Iverson DB, Tymiak AA, Grace MJ, Russell RJ. A Tris (2-Carboxyethyl) Phosphine (TCEP) Related Cleavage on Cysteine-Containing Proteins. J Am Soc Mass Spectrom. 2010;21(5):837–844. doi: 10.1016/j.jasms.2010.01.016. [DOI] [PubMed] [Google Scholar]
  • 35.Burns JA, Butler JC, Moran J, Whitesides GM. Selective reduction of disulfides by tris (2-carboxyethyl) phosphine. J Org Chem. 1991;56(8):2648–2650. [Google Scholar]
  • 36.Johnson JR. Virulence factors in Escherichia coli urinary tract infection. Clin Microbiol Rev. 1991;4(1):80–128. doi: 10.1128/cmr.4.1.80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Ronald A. The etiology of urinary tract infection: traditional and emerging pathogens. DM, Dis-Mon. 2003;49(2):71–82. doi: 10.1067/mda.2003.8. [DOI] [PubMed] [Google Scholar]
  • 38.Yang L, Haagensen JAJ, Jelsbak L, Johansen HK, Sternberg C, Høiby N, Molin S. In Situ Growth Rates and Biofilm Development of Pseudomonas aeruginosa Populations in Chronic Lung Infections. J Bacteriol. 2008;190(8):2767–2776. doi: 10.1128/JB.01581-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Foucault M-L, Courvalin P, Grillot-Courvalin C. Fitness Cost of VanA-Type Vancomycin Resistance in Methicillin-Resistant Staphylococcus aureus. Antimicrob Agents Chemother. 2009;53(6):2354–2359. doi: 10.1128/AAC.01702-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Cetinkaya Y, Falk P, Mayhall CG. Vancomycin-Resistant Enterococci. Clin Microbiol Rev. 2000;13(4):686–707. doi: 10.1128/cmr.13.4.686-707.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Biggest Threats: Antibiotic/Antimicrobial Resistance. CDC; [accessed November 12, 2016]. http://www.cdc.gov/drugresistance/biggest_threats.html. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SI
NIHMS919782-supplement-SI.pdf (442.3KB, pdf)

RESOURCES