Subscriber access provided by Lulea University of Technology

Article
Proximal Bacterial Lysis and Detection in
Nanoliter Wells Using Electrochemistry
Justin D. Besant, Jagotamoy Das, Edward H. Sargent, and Shana O. Kelley
ACS Nano, Just Accepted Manuscript • DOI: 10.1021/nn4035298 • Publication Date (Web): 09 Aug 2013
Downloaded from http://pubs.acs.org on August 16, 2013

Just Accepted

“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Nano is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 22 ACS Nano

1
2
3
4 Proximal Bacterial Lysis and Detection in Nanoliter Wells
5
6
7 Using Electrochemistry
8
9
10
11 Justin D. Besant,a Jagotamoy Das,b Edward H. Sargentc and Shana O. Kelleya,b,d
12
13
14 a
Institute for Biomaterials and Biomedical Engineering, University of Toronto,
15
16 b
17
Department of Pharmaceutical Science, Leslie Dan Faculty of Pharmacy, University of
18
19 Toronto, cDepartment of Electrical and Computer Engineering, Faculty of Engineering,
20
21 University of Toronto, dDepartment of Biochemistry, Faculty of Medicine, University of
22
23
24 Toronto, Toronto, ON, Canada M5S 3M2
25
26
27
28 ABSTRACT: Rapid and direct genetic analysis of low numbers of bacteria using chip-
29
30 based sensors is limited by the slow diffusion of mRNA molecules. Long incubation
31
32 times are required in dilute solutions in order to collect a sufficient number of molecules
33
34
35 at the sensor surface to generate a detectable signal. Here we present an integrated
36
37 device that leverages electrochemistry-driven lysis less than 50 microns away from
38
39 electrochemical nucleic acid sensors to overcome this barrier. Released intracellular
40
41
42 mRNA can diffuse the short distance to the sensors within minutes enabling rapid and
43
44 sensitive detection. We validate this strategy through direct lysis and detection of E. coli
45
46
47
mRNA at concentrations as low as 0.4 CFU/µL in 2 minutes, a clinically-relevant
48
49 combination of speed and sensitivity for a sample-to-answer molecular analysis
50
51 approach.
52
53
54
55 Keywords: bacterial detection, cell lysis, electrochemical detection,
56
57
58
microelectrodes
59
60
1
ACS Paragon Plus Environment
 ACS Nano Page 2 of 22

1
2
3
New strategies for rapidly detecting low levels of bacteria are urgently needed to
4
5
6 control and manage infectious disease.1 No existing method simultaneously satisfies
7
8 the speed and sensitivity requirements to detect sufficiently low levels of bacteria in a
9
10
11
clinically relevant time period.2 Culture, the gold standard for diagnosis for most types
12
13 of bacterial infection diagnosis, requires hours to days to amplify the bacteria to visibly
14
15 detectable levels. Enzymatic amplification methods such as the polymerase chain
16
17
18 reaction (PCR) are complex to automate and often require sample purification that
19
20 slows analysis.3,4
21
22 A wide variety of molecular sensors have been developed to address the
23
24
25 limitations of culture and PCR.5–8 The direct detection of nucleic acids sequences using
26
27 chip-based sensors has been pursued for some time as an attractive solution.9–14 The
28
29
30
use of mRNA sequences such as that corresponding to the RNA polymerase β (rpoβ)
31
32 subunit can provide species-level identification of bacteria15 and many copies of mRNA
33
34 may exist in a single cell which provides an inherent signal enhancement. However,
35
36
37 using this type of biomarker to rapidly detect bacteria at clinically-relevant levels
38
39 remains a challenge; the long diffusion times of large mRNA molecules impose an
40
41
inherent trade-off between speed and sensitivity of detection.16,17 In a typical sample-to-
42
43
44 answer detection scheme, bacteria are lysed in a lysis chamber and the homogenous
45
46 lysate is then transported to a detection chamber (Figure 1A).18 As each bacterium may
47
48
49
harbour multiple copies of the target mRNA, prior to lysis the molecules are present at a
50
51 locally high concentration inside each bacterium. However, after lysis, intracellular
52
53 mRNA is released into bulk solution and thus the overall concentration of mRNA is very
54
55
56 low. When sensing low concentrations of mRNA, long incubations times are required in
57
58
59
60
2
ACS Paragon Plus Environment
Page 3 of 22 ACS Nano

1
2
3
order to accumulate enough target molecules at the sensor surface to generate a
4
5
6 detectable signal.
7
8 Here, we propose a novel approach to rapid genetic analysis that uses an
9
10
11
electrochemical approach to lyse bacteria in close proximity to a microelectrode where
12
13 their mRNA can be analysed. Proximal electrochemical lysis shortens the required
14
15 distance over which the target mRNA molecules must diffuse. This approach provides
16
17
18 a ten-fold improvement in performance and permits rapid analysis of bacteria at
19
20 clinically-relevant levels.
21
22
23
24
25 RESULTS AND DISCUSSION
26
27 Summary of Approach. Detection of mRNA using chip-based sensors is limited by
28
29
30
the rate of molecular diffusion to the sensor. Molecules of bacterial mRNA, which can
31
32 be up to thousands of base pairs long, have low diffusion coefficients compared to short
33
34 synthetic oligomers and long timescales are therefore required to collect enough
35
36
37 molecules to generate a robust response.19 To study the diffusional flux of analytes at
38
39 the sensor after lysis, we created a model of a single E. coli inside a well that contains a
40
41 20 µm hemispherical sensor. We assumed each E. coli contained multiple copies of
42
43
44 RNA polymerase β subunit (rpoβ), a 4000 bp transcript. We assumed a copy number of
45
46 1400 as calculated previously.17 Figure 1B compares the diffusional flux of the analytes
47
48
49 at the sensor as a function of time for the homogenous and local lysis approaches. In
50
51 the homogenous lysis case, the flux is low and nearly constant. In the local lysis
52
53
approach, the flux rapidly increases in the first minutes as the analytes reach the sensor
54
55
56 and slowly decays as the molecules diffuse away. Figure 1C shows the number of
57
58
59
60
3
ACS Paragon Plus Environment
 ACS Nano Page 4 of 22

1
2
3
analyte molecules captured at the sensor surface over time for both lysis scenarios.
4
5
6 Rapidly (< 30 minutes) detecting less than 1 CFU/µL is not possible with a homogenous
7
8 lysis approach as no analyte molecules are captured after 30 minutes. On the other
9
10
11
hand, if just a single bacterium is lysed 50 µm from the sensor, many target analytes will
12
13 accumulate on the sensor surface within minutes. A device was therefore designed that
14
15 would minimize the distance between the site of cell lysis and mRNA detection.
16
17
18 In light of the findings from our diffusional model, we designed a device that
19
20 would minimize the distance between the site of cell lysis and mRNA detection. The
21
22 device featured electrodes for electrochemical lysis that surrounded a detection sensor
23
24
25 (Figure 1D). Previous studies have shown that hydroxide ions can be produced locally
26
27 to initiate cellular lysis, but this approach has primarily been used in bulk solution as a
28
29
30
means to prepare samples for analysis using PCR or other detection approaches.20-22
31
32 Here, lysis electrodes were placed within 50 microns of a microelectrode sensor, which
33
34 allows large mRNA molecules to diffuse to the sensor within 10 minutes. The reaction
35
36
37 that initiates cell lysis is based on the production of hydroxide ions from water at a
38
39 cathode at a potential of 20 V. Hydroxide ions can break down bacterial membranes,
40
41 causing the contents of the bacteria, including the nucleic acids, to be released into
42
43
44 solution.
45
46 The placement of a chip-based sensor in close proximity to the lysis electrodes
47
48 provides the potential for very sensitive analysis of bacterial cells because limited
49
50
51 diffusion is required of the molecules that provide information on cellular identity. The
52
53 functionalization of a nearby sensor with a thiolated probe molecule that is
54
55
56
complementary to a unique portion of a bacterial mRNA allows capture of the marker if
57
58
59
60
4
ACS Paragon Plus Environment
Page 5 of 22 ACS Nano

1
2
3
bacteria is present within a sample undergoing analysis. The presence of bound target
4
5
6 mRNA can be readout using an electrochemical reporter system that senses the
7
8 change in electrostatics at the sensor surface (Figure 1E).23 Ru(NH3)63+ ions
9
10
11
accumulate at the sensor surface, and serve as electron acceptors. Fe(CN)64- ions
12
13 serve to reoxidize Ru(II) as it is made electrochemically, and regenerate the electron
14
15 acceptor to make the reaction electrocatalytic. Differential pulse voltammetry provides
16
17
18 an effective sampling method that can be used to investigate whether current levels rise
19
20 above a threshold that indicates that a sample is positive for a particular pathogen.
21
22 Design and validation of lysis electrodes for electrochemical hydroxide
23
24
25 generation. Given that large applied potentials could disrupt the bond between a probe
26
27 sequence and the sensor, thereby compromising the integrity of the surface assembled
28
29
30
monolayer, we simulated the electric fields generated by different lysis electrode
31
32 geometries (Figure 2A). Lysis electrodes which sandwich the sensor induce a high
33
34 potential at the sensing electrode which could reduce probe coverage by dissociating
35
36
37 the gold-thiol bond. Figure 2B shows an alternative layout in which both lysis electrodes
38
39 wrap around the sensing electrode. The electrodes act as an on-chip faraday cage to
40
41 shield the sensor from the electric fields. We chose this layout to minimize the electric
42
43
44 fields experienced by the sensor.
45
46 Another feature of the device we fabricated is patterned wells that contain the
47
48 lysis and sensing electrodes to enable the contents of the bacterial cells to remain close
49
50
51 to the sensor. The wells introduced on the surface of the chip hold 1 nanoliter of liquid,
52
53 and were generated using SU-8 and patterned using photolithography. NME sensors
54
55
56
57
58
59
60
5
ACS Paragon Plus Environment
 ACS Nano Page 6 of 22

1
2
3
were plated within the wells, and were shown to exhibit similar morphologies to what
4
5
6 has been demonstrated previously (Figures 2C and 2D).17,18, 24
7
8 As an initial assessment of lysis, we measured E. coli growth on agar plates after
9
10
11
they were subjected to on-chip electrochemical lysis (Figure 3A). We applied pulse
12
13 voltages from 0 to 20 V at 1 Hz for 1 minute to E. coli and allowed the lysate to incubate
14
15 on agar plates overnight. E. coli viability dropped after applying 5 V and no growth was
16
17
18 observed after applying 20 V. While this approach is an indirect measure of whether
19
20 lysis is occurring, it provides a means to identify an interesting potential range to look for
21
22 membrane permeability.
23
24
25 To study whether lysis made the E. coli membrane permeable, we measured
26
27 cellular uptake of propidium iodide (PI), a fluorescent dye which intercalates with DNA
28
29
30
(Figure 3B). Cellular uptake of PI is used as an indicator of lysis as PI cannot cross
31
32 intact cell membranes.20 Using flow cytometry, we measured propidium iodide uptake
33
34 as function of pulse voltage. Increasing voltage caused greater uptake of propidium
35
36
37 iodide, with the largest uptake at 20 V.
38
39 Using fluorescent microscopy, we visualized PI uptake in real time (Figure 3C).
40
41 Voltage pulses (20 V) were applied to E. coli expressing green fluorescent protein
42
43
44 (GFP) in the presence of PI. We observed greater PI uptake as increasing number of
45
46 voltage pulses were applied. E. coli lyse first at the positive outer electrode and then
47
48 near the grounded inner electrode.
49
50
51 Identification of compatible solution conditions for electrochemical lysis
52
53 and detection. Initiating lysis close to a chip-based sensor has the potential to improve
54
55
56
detection limits, but may also introduce issues that would interfere with the function of
57
58
59
60
6
ACS Paragon Plus Environment
Page 7 of 22 ACS Nano

1
2
3
the sensors. As discussed above, the presence of a strong field could interfere with the
4
5
6 attachment of the probe to the sensor. In addition, the ions generated electrochemically
7
8 could degrade the probe molecules.25
9
10
11
We tested different lysis conditions and investigated whether probe molecules
12
13 were dissociating from the sensor surface. An assay based on the blocking of the
14
15 surface in the presence of probe was used, where signals generated by Fe(CN)64- were
16
17
18 analysed. The iron reporter group is repelled by the anionic probe, and if stripping
19
20 occurred, the signal it generates would increase.
21
22 As shown in Figure 4, differing levels of probe stripping were observed when
23
24
25 buffer conditions were varied. Lysis performed in water with short (30 µs) or long (10
26
27 ms) pulses did not cause probe dissociation. However, lysis in 1X phosphate-buffered
28
29
30 saline, did cause damage with large signal changes occurring for both pulse times.
31
32 Diluting the PBS, or the use of phosphate buffer attenuated this effect. In PBS, the
33
34 presence of chloride ions likely contribute to the production of hypochlorite ions after the
35
36
37 production of Cl2 at the anode. Hypochlorite is highly reactive, and persists in buffered
38
39 solutions and is expected to cause significant damage both to immobilized probes and
40
41
the molecules being liberated within a sample. It is therefore desirable to avoid
42
43
44 generating this species.
45
46 Lysis in phosphate buffer, especially with short pulses, resulted in minimal probe
47
48
49
loss, and provides buffering capacity to ensure that samples only transiently experience
50
51 elevated hydroxide levels, and provides ionic strength to promote hybridization on the
52
53 sensor surface. This buffer system was therefore identified as the best set of solution
54
55
56 conditions that was compatible with both electrochemical lysis and detection.
57
58
59
60
7
ACS Paragon Plus Environment
 ACS Nano Page 8 of 22

1
2
3
Bacterial detection with proximal lysis. With effective lysis demonstrated, the
4
5
6 detection of bacteria using the devices was tested. A series of studies were conducted
7
8 that focused on the effect of lysis pulse length on detection of E. coli (Figure 5A).
9
10
11 Increasing the pulse time from 30 to 300 µs increased the amount of current change
12
13 generated when a solution of 400 CFU/µL E. coli cells per microliter, indicating greater
14
15
lysis efficiency with the longer pulses. The optimal current at 300 µs is likely due to the
16
17
18 competing effects of hydroxide generation. High hydroxide concentrations lead to
19
20 efficient lysis, but also to degradation of the target molecules. Below 300 µs, the lysis is
21
22
23
not as efficient, while above 300 µs, the target likely degrades due to excessive
24
25 hydroxide concentration causing a lower current change.
26
27 To evaluate the detection limit of the sensors with proximal lysis, E. coli were
28
29
30 serially diluted from 400 CFU/µL to 0.4 CFU/µL (Figure 5B). E. coli were lysed on chip
31
32 by applying the optimal 300 µs pulses for 1 minute and the lysate was allowed to
33
34 incubate at 37˚C for 30 min. 5000 CFU/µL S. aureus was used as a negative control. E.
35
36
37 coli were detected with a limit of detection of 0.4 CFU/uL and high specificity as no
38
39 signal was observed from S. aureus.
40
41
The sensitivity enhancement provided by the combination lysis/detection device
42
43
44 was tested by challenging the sensors with serial dilutions of E. coli lysed off-chip. The
45
46 limit of detection of the device when challenged with E. coli lysate prepared off chip was
47
48
49
only 4 CFU/µL, which is 10 times higher than when E. coli were lysed on chip. This
50
51 indicates that lysis in close proximity to the sensors provided a 10-fold sensitivity
52
53 advantage.
54
55
56
57
58
59
60
8
ACS Paragon Plus Environment
Page 9 of 22 ACS Nano

1
2
3
In addition to providing enhanced detection limits, proximal lysis should also
4
5
6 speed the progress of hybridization and produce fast results. To determine the time
7
8 dependence of the detection approach, we challenged the sensor with E. coli at 0.4
9
10
11
CFU/µL using 2 minute and 5 minute hybridization times (Figure 5C). We observed a
12
13 positive signal from 0.4 CFU/µL after both 2 and 5 minute hybridization periods
14
15 indicating the sensor has a rapid response time. This is a record-breaking level of
16
17
18 speed and sensitivity combined in a single sensor system. The previous sensitivity
19
20 record for direct electrochemical detection of mRNA in crude lysate is 1 CFU/µL after a
21
22
23
20 minute incubation time.18 Here we show detection of 0.4 CFU/µL within 2 minutes.
24
25 Real sample matrices have a wide range of pH and salt concentrations which
26
27 need to be controlled in proximal electrochemical lysis. To use this device with complex
28
29
30 matrices the sample could be pre-diluted in the appropriate buffer. Alternatively, this
31
32 device could be coupled to a pre-concentration step to separate the bacteria of interest
33
34 from the sample and allow for buffer exchange.
35
36
37
38
39 CONCLUSIONS
40
41
42 We developed an integrated device capable of electrochemical lysis and detection of E.
43
44
45 coli at concentrations as low as 0.4 CFU/µL in 2 minutes. Lysing in the vicinity of the
46
47 sensors allowed high concentrations of released intracellular mRNA to reach the sensor
48
49 rapidly. Our experiments highlight the importance of optimizing the buffer and electrode
50
51
52 geometry when lysing nearby surface modified electrodes in order to maintain the
53
54 integrity of the probe monolayer, and provide a powerful sample-to-answer approach for
55
56
bacterial detection.
57
58
59
60
9
ACS Paragon Plus Environment
 ACS Nano Page 10 of 22

1
2
3
METHODS
4
5
6
7 Simulations. Simulations of electric fields for various lysis electrode designs were
8
9 conducted using COMSOL Multiphysics. The applied voltages were set to keep the
10
11 average field constant across both geometries.
12
13
14
Simulations of the flux of analyte molecules at the sensor surface were calculated using
15
16
17 a COMSOL model. These calculations are described in the supporting information.
18
19
20 Device fabrication. Devices were fabricated by the Canadian Photonics Fabrication
21
22 Centre (Ottawa, ON) on 300 µm thick 6” silicon wafers coated with 450 nm of thermally
23
24
25 grown SiO2. To pattern the electrodes 300 nm Au was deposited on a 25 nm Ti adhesion
26
27 layer. After patterning the electrodes using standard photolithography and wet etching,
28
29
the electrodes were passivated with 500 nm SiO2 using plasma enhanced chemical
30
31
32 vapour deposition. The lysis electrodes, contacts, and apertures were exposed using
33
34 reactive ion etching. Wells were patterned using SU-8 30-25 (Microchem, Newton, MA)
35
36
37
using photolithography. The devices diced and fixed to the bottom of a custom-designed
38
39 PMMA reservoir (QuickCUTCNC, Atlanta, Georgia).
40
41
42 NME electrodeposition. Nanostructured microelectrodes (NMEs) were
43
44 electrodeposited using a two-step process using a three-electrode setup with a Pt
45
46
47 counter electrode and a Ag/AgCl reference electrode. To grow a sensor with a large
48
49 footprint, we applied 0 mV with respect to the Ag/AgCl reference electrode for 20 s in a
50
51
52
solution of 50 mM HAuCl4 and 0.5 M HCl. Sensors were decorated with nanostructured
53
54 features by applying -250 mV in a solution of 5 mM PdCl2 and 0.5 M HClO4 for 5 s.
55
56
57
58
59
60
10
ACS Paragon Plus Environment
Page 11 of 22 ACS Nano

1
2
3
Preparation of bacterial samples. E. coli (Invitrogen, Carslbad, CA) and S. aureus
4
5
6 (ATCC) were cultured in an incubating shaker at 33˚C in LB Miller and Tryptic Soy Agar
7
8 broth respectively. Concentrations were measured using optical density measurements
9
10
11
at 600 nm with a UV-vis spectrometer (Agilent, Santa Clara, CA) and by counting the
12
13 number of colonies on agar plates incubated overnight at 37 ˚C. Using centrifugation,
14
15
the growth medium was replaced with the appropriate buffer before lysis. To assess cell
16
17
18 viability, 100 µL of lysate were spread on LB agar plates and incubated overnight at 37
19
20 ˚C. E. coli were lysed off chip using OmniLyse (Claremont Bio, Upland, CA).
21
22
23 Flow cytometry. E. coli were diluted to 1x107 CFU/mL in various buffers. Potentials
24
25
26 from 0 to 20 V were applied to lysis electrodes using a 1 Hz repetition rate for 60 s. After
27
28 lysis, samples were incubated with 2 µg/mL propidium iodide for 30 minutes in the dark.
29
30 Measurements were made using a BD FACS Canto flow cytometer and plotted as
31
32
33 histograms of fluorescence intensity.
34
35
36 Fluorescent microscopy. E. coli expressing GFP were diluted to 1x107 CFU/mL in the
37
38 appropriate buffer and lysed with varying numbers of 20 V pulses (30 µs pulse time).
39
40
41 Samples were incubated with 2 µg/mL propidium iodide and red and green fluorescent
42
43 images were acquired with a Nikon Eclipse LV150 microscope.
44
45
46 Synthesis and purification of PNA probes. Probes complementary to E. coli rpoβ
47
48
49
mRNA (4029 bp) were designed with the following sequence: NH2-Cys-Gly-Asp-ATC
50
51 TGC TCT GTG GTG TAG TT-Asp-CONH2. Probes were synthesized using a protein
52
53 Technologies Prelude peptide synthesizer and purified using reverse phase high
54
55
56
57
58
59
60
11
ACS Paragon Plus Environment
 ACS Nano Page 12 of 22

1
2
3
performance liquid chromatography. Probe concentration was calculated using the
4
5
6 extinction coefficient and the absorbance at 260 nm.
7
8
9 Sensor functionalization, lysis, and hybridization. Sensors were functionalized with
10
11 1 µM PNA probe and 9 µM mercaptohexanol for 30 min at room temperature. Chips
12
13
14 were washed twice for 5 min with 1 x PBS buffer after probe deposition. E. coli and S.
15
16 aureus were diluted to the appropriate concentration in 50 mM phosphate buffer and 50
17
18
µL was added to the chip. After washing, potentials were applied to lysis electrodes
19
20
21 using 1 Hz pulses with varying voltages and pulse times. The chip was incubated in a
22
23 humidity chamber for 2 to 30 minutes at 37˚C followed by washing twice with PBS for 5
24
25
26
min. After washing, DPV measurements were performed following probe deposition and
27
28 sample hybridization in the electrocatalytic solution.
29
30
31 Electrochemical measurements. For all electrochemical measurements we used a
32
33 three-electrode setup with a Ag/AgCl reference electrode and Pt counter electrode
34
35
36 connected to a potentiostat (BASi, West Lafayette, IN). To measure the effect of lysis
37
38 on probe detachment, NMEs modified with PNA probe were scanned from 0 to 0.5 V
39
40
41
using differential pulse voltammetry in a solution of 2.5 mM Fe(CN)64– and 0.1x PBS.
42
43 Oxidation currents were measured before and after applying 30 µs and 10 ms 20 V
44
45 pulses to the lysis electrodes at 1 Hz for 60 s.
46
47
48 To measure the amount of hybridized nucleic acid, electrochemical signals were
49
50
51 measured in 0.1 x PBS with 10 µM [Ru(NH3)6]Cl3 and 4 mM K3[Fe(CN)6]. DPV signals
52
53 were obtained with a potential step of 5 mV, pulse amplitude of 50 mV, pulse width of 50
54
55
ms and a pulse period of 100 ms. Signal changes that corresponded to target
56
57
58 hybridization were calculated with background-subtracted currents: ∆I = (Iafter – Ibefore)
59
60
12
ACS Paragon Plus Environment
Page 13 of 22 ACS Nano

1
2
3
(where Iafter = current after target hybridization and Ibefore = current before target
4
5
6 hybridization i.e., current with only probe).
7
8
9 AUTHOR INFORMATION
10
11 Corresponding Author
12 * E-mail: shana.kelley@utoronto.ca
13
14
15
16
AUTHOR CONTRIBUTIONS
17 The manuscript was written through contributions of all authors. All authors have given
18
19 approval to the final version of the manuscript.
20
21
22
ACKNOWLEDGMENTS
23
24 This work was funded by CMC Microsystems, Ontario Research Fund, the Canadian
25
26
27 Institute for Health Research, and the Natural Sciences and Engineering Research
28
29 Council of Canada. We would like to extend our appreciation to Brian Lam for his
30
31
32
advice throughout the project. We thank the ECTI facility at University of Toronto for
33
34 their cleanroom facilities. We wish to acknowledge CMC Microsystems for their help
35
36 with CAD design and fabrication services.
37
38
39
40
41
42 SUPPORTING INFORMATION
43
44
45 Additional information on the diffusional model and supporting experiments are included
46
47
48 as supporting information. This material is available free of charge via the Internet at
49
50 http://pubs.acs.org.
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment
 ACS Nano Page 14 of 22

1
2
3
ABBREVIATIONS
4
5
6
CFU: colony-forming unit; mRNA: messenger RNA; PCR: polymerase chain reaction;
7
8 NME: nanostructured microelectrode.
9
10
11 REFERENCES
12
13
14
1. Mancini, N.; Carletti, S.; Ghidoli, N.; Cichero, P.; Burioni, R.; Clementi, M. The Era
15 of Molecular and Other Non-Culture-Based Methods in Diagnosis of Sepsis. Clin.
16 Microbiol. Rev. 2010, 23, 235–251.
17
18 2. Ecker, D. J.; Sampath, R.; Li, H.; Massire, C.; Matthews, H. E.; Toleno, D.; Hall, T.
19 A.; Blyn, L. B.; Eshoo, M. W.; Ranken, R.; et al. New Technology for Rapid
20
21 Molecular Diagnosis of Bloodstream Infections. Expert Rev. Mol. Diag. 2010, 10,
22 399–415.
23
24 3. Bloos, F.; Hinder, F.; Becker, K.; Sachse, S.; Mekontso Dessap, A.; Straube, E.;
25 Cattoir, V.; Brun-Buisson, C.; Reinhart, K.; Peters, G.; et al. A Multicenter Trial to
26
27
Compare Blood Culture with Polymerase Chain Reaction in Severe Human
28 Sepsis. Intens. Care. Med. 2010, 36, 241–247.
29
30 4. Ferguson, B. S.; Buchsbaum, S. F.; Wu, T.-T.; Hsieh, K.; Xiao, Y.; Sun, R.; Soh,
31 H. T. Genetic Analysis of H1N1 Influenza Virus from Throat Swab Samples in a
32
Microfluidic System for Point-of-Care Diagnostics. J. Amer. Chem. Soc. 2011,
33
34 133, 9129–9135.
35
36 5. Lee, H.; Yoon, T.-J.; Weissleder, R. Ultrasensitive Detection of Bacteria Using
37 Core-Shell Nanoparticles and an NMR-Filter System. Angew. Chem. Int. Ed.
38 2009, 48, 5657–5660.
39
40
41 6. Anker, J. N.; Hall, W. P.; Nilam, C.; Zhao, J.; Van, P. Biosensing with Plasmonic
42 Nanosensors. Nat. Materials. 2008, 7, 8–10.
43
44 7. Chung, H. J.; Castro, C. M.; Im, H.; Lee, H.; Weissleder, R. A Magneto-DNA
45
46
Nanoparticle System for Rapid Detection and Phenotyping of Bacteria. Nat. Nano.
47 2013, 8, 369–375.
48
49 8. Cederquist, K. B.; Kelley, S. O. Nanostructured Biomolecular Detectors: Pushing
50 Performance at the Nanoscale. Biotechnol. Adv. 2012, 16, 415–421.
51
52
53 9. Foudeh, A. M.; Fatanat Didar, T.; Veres, T.; Tabrizian, M. Microfluidic Designs
54 and Techniques Using Lab-on-a-Chip Devices for Pathogen Detection for Point-
55 of-care Diagnostics. Lab Chip 2012, 12, 3249–3266.
56
57
58
59
60
14
ACS Paragon Plus Environment
Page 15 of 22 ACS Nano

1
2
3
10. Lam, B.; Das, J.; Holmes, R. D.; Live, L.; Sage, A.; Sargent, E. H.; Kelley, S. O.
4
5 Solution-Based Circuits Enable Rapid and Multiplexed Pathogen Detection. Nat.
6 Comm. 2013, 4, 1–8.
7
8 11. Drummond, T. G.; Hill, M. G.; Barton, J. K. Electrochemical DNA Sensors. Nat.
9 Biotechnology. 2003, 21, 1192–1199.
10
11
12 12. Sorgenfrei, S.; Chiu, C.; Gonzalez, R. L.; Yu, Y.-J.; Kim, P.; Nuckolls, C.;
13 Shepard, K. L. Label-Free Single-Molecule Detection of DNA-Hybridization
14 Kinetics with a Carbon Nanotube Field-effect Transistor. Nat. Nano. 2011, 6, 126–
15 132.
16
17
18 13. Hsieh, K.; Patterson, A. S.; Ferguson, B. S.; Plaxco, K. W.; Soh, H. T. Rapid,
19 Sensitive, and Quantitative Detection of Pathogenic DNA at the Point of Care
20 through Microfluidic Electrochemical Quantitative Loop-Mediated Isothermal
21 Amplification. Angew. Chem. Int. Ed. 2012, 51, 4896–4900.
22
23
24 14. Zhang, J.; Song, S.; Zhang, L.; Wang, L.; Wu, H.; Pan, D.; Fan, C. Sequence-
25 Specific Detection of Femtomolar DNA via a Chronocoulometric DNA Sensor
26 (CDS): Effects of Nanoparticle-mediated Amplification and Nanoscale Control of
27 DNA Assembly at Electrodes. J. Amer. Chem. Soc. 2006, 128, 8575–8580.
28
29
30 15. Adékambi, T.; Drancourt, M.; Raoult, D. The rpoB Gene as a Tool for Clinical
31 Microbiologists. Trends Microbiol. 2009, 17, 37–45.
32
33 16. Sheehan, P. E.; Whitman, L. J. Detection Limits for Nanoscale Biosensors. Nano.
34 Lett. 2005, 5, 803–807.
35
36
37 17. Soleymani, L.; Fang, Z.; Lam, B.; Bin, X.; Vasilyeva, E.; Ross, A.; Sargent, E. H.;
38 Kelley, S. O. Hierarchical Nanotextured Microelectrodes Overcome the Molecular
39 Transport Barrier to Achieve Rapid, Direct Bacterial Detection. ACS Nano 2011,
40 5, 3360-3366.
41
42
43 18. Lam, B.; Fang, Z.; Sargent, E. H.; Kelley, S. O. Polymerase Chain Reaction-Free,
44 Sample-to-Answer Bacterial Detection in 30 Minutes with Integrated Cell Lysis.
45 Anal. Chem. 2011, 84, 21–25.
46
47 19. Smith, D. E.; Perkins, T. T.; Chu, S. Dynamical Scaling of DNA Diffusion
48
49
Coefficients. Macromolecules 1996, 29, 1372–1373.
50
51 20. Lee, H. J.; Kim, J.-H.; Lim, H. K.; Cho, E. C.; Huh, N.; Ko, C.; Park, J. C.; Choi, J.-
52 W.; Lee, S. S. Electrochemical Cell Lysis Device for DNA Extraction. Lab Chip
53 2010, 10, 626–633.
54
55
56 21. Jha, S. K.; Chand, R.; Han, D.; Jang, Y.-C.; Ra, G.-S.; Kim, J. S.; Nahm, B.-H.;
57 Kim, Y.-S. An Integrated PCR Microfluidic Chip Incorporating Aseptic
58
59
60
15
ACS Paragon Plus Environment
 ACS Nano Page 16 of 22

1
2
3
Electrochemical Cell Lysis and Capillary Electrophoresis Amperometric DNA
4
5 Detection for Rapid and Quantitative Genetic Analysis. Lab Chip 2012, 12, 4455–
6 4464.
7
8 22. Nevill, J. T.; Cooper, R.; Dueck, M.; Breslauer, D. N.; Lee, L. P. Integrated
9 Microfluidic Cell Culture and Lysis on a Chip. Lab Chip 2007, 7, 1689-1695.
10
11
12 23. Lapierre, M. a; O’Keefe, M.; Taft, B. J.; Kelley, S. O. Electrocatalytic Detection of
13 Pathogenic DNA Sequences and Antibiotic Resistance Markers. Anal. Chem.
14 2003, 75, 6327–6333.
15
16
17
24. Lam, B.; Holmes, R.D.; Das, J.; Poudineh, M.; Sage, A.; Sargent, E.H.; Kelley,
18 S.O. Optimized Templates for Bottom-Up Growth of High-Performance Integrated
19 Biomolecular Detectors. Lab on a Chip, 2013, 13, 2569-2575
20
21 25. Bhatt, N.; Huang, P.-J. J.; Dave, N.; Liu, J. Dissociation and Degradation of Thiol-
22
modified DNA on Gold Nanoparticles in Aqueous and Organic Solvents. Langmuir
23
24 2011, 27, 6132–6137.
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16
ACS Paragon Plus Environment
Page 17 of 22 ACS Nano

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38 Figure 1. Schematic of lysis and electrochemical readout. (A) (i) Typically, bacteria are
39 lysed in bulk solution and the homogenous lysate is transferred to the sensor. (ii) In an
40 alternative proximal lysis approach, bacteria are lysed in the vicinity of the sensor and
41 released mRNA can rapidly diffuse the short distance to the sensor. (B) Diffusional flux
42
43 of mRNA released from a single bacterium harboring 1400 transcripts. The flux from
44 lysis at varying distances from the sensor is compared to the flux from homogenous
45 lysate from 4 CFU/µL. (C) Molecules accumulated at the sensor over time after local
46 lysis (50 µm away) of a single bacterium compared to homogenous lysate at varying
47 concentrations. Detecting less than ~1 CFU/µL in 5 minutes is not possible in a
48
49
homogenous lysate. In contrast, intracellular RNA from just a single bacterium lysed
50 locally accumulates within minutes at the sensor surface. The dashed black line
51 represents a typical threshold for detection of 10 molecules. (D) Each well contains two
52 lysis electrodes and an NME sensor. Bacteria are lysed electrochemically by an applied
53 potential. Released intracellular mRNA rapidly hybridizes to the complementary PNA
54 probe molecules functionalized on the NME surface. (E) The amount of hybridized
55
56 mRNA is readout using an electrocatalytic reporter pair and DPV to measure the peak
57 current before and after hybridization.
58
59
60
17
ACS Paragon Plus Environment
 ACS Nano Page 18 of 22

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39 Figure 2. Device design. Simulations of the electric fields induced by lysis electrodes
40
which span the sensor (A) and wrap around the sensor (B). (C) Optical microscopy
41
42 image of NMEs electrodeposited into the well array. (D) Optical microscopy and
43 scanning electron microscopy images an electrodeposited NME. Scale bars represent
44 100 µm.
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
18
ACS Paragon Plus Environment
Page 19 of 22 ACS Nano

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51 Figure 3. Validation of electrochemical lysis. (A) Effect of applied potentials on E. coli
52
53 viability after incubating on agar plates. IPA was used as a positive control. (B) Uptake
54 of PI measured by flow cytometry as a function of voltage. IPA is used as a positive
55 control. (C) Uptake of PI measured with optical microscopy as function of the number of
56 pulses applied. E. coli expressing GFP are green and E. coli that uptake PI are red.
57 Scale bars represent 100 µm.
58
59
60
19
ACS Paragon Plus Environment
 ACS Nano Page 20 of 22

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41 Figure 4. Effect of lysis on probe integrity. Percent change in magnitude of peak
42 oxidation currents of Fe(CN)64– measured before and after on-chip lysis. The percent
43 increase in current measured after applying 30 µs and 10 ms pulses in various buffers.
44
45 High percentage changes indicate probe is removed from the surface. Error bars
46 represent standard error.
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20
ACS Paragon Plus Environment
Page 21 of 22 ACS Nano

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49 Figure 5. Lysis and electrochemical detection of E. coli. (A) The effect of electrical lysis
50 pulse length on the electrochemical signal change. (B) Concentration-dependent signal
51
change of E. coli lysed on-chip as compared to E. coli lysed off-chip. The data were
52
53 normalized to the signal from 400 CFU/µL. The dashed line corresponds to the average
54 signal-change in the absence of E. coli. (C) Hybridization time-dependence of the signal
55 from E. coli lysed on chip. The dashed line corresponds to the average signal-change in
56 the absence of E. coli. Error bars represent standard error.
57
58
59
60
21
ACS Paragon Plus Environment
 ACS Nano Page 22 of 22

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
TOC graphic
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22
ACS Paragon Plus Environment