Patent Application
20150219592
Sensitive and selective detection of redox active substances, including those produced by pathogenic microorganisms, has important implications for both medical and environmental research, and in microbial fuel cells (Zhang et al., 2005, Curr Opin Microbiol 8, 276-281; Jacob et al., 2011, Current Opinion in Chem Biol 15, 149-155; Dietrich et al., 2008, Science 321, 1203-1206; and Ren et al., 2012, Microfluid Nanofluid 13, 353-381). Further, the growing rate of hospital infections and the emergence of antibiotic resistant bacteria, such as antibiotic resistant strains of Pseudomonas aeruginosa (PA) highlight the need for small, robust sensors that detect bacterial contamination. (Paramythiotou et al.,2004, Clin Infect Dis 38, 670-677). A variety of amperometric and potentiometric techniques can be utilized to obtain information about the chemical composition of a solution (Pihel et al., 1995 Anal Chem, 67, 4514-4521; Buck et al., 2001, Anal Chem, 73, 88-97; Hengstenberg et al., 2001, Angew Chem Int Ed Engl, 40, 905-908; Müller et al., 1981, J Neuro Meth, 1981, 4, 39-52), and may be used for development of the sensors for detecting redox active substances.
Electrochemical detection offers several advantages over other sensing schemes, such as fast analysis time, ease of use, and low limits of detection (Cheng et al., 2007, Electrophoresis 28, 1579-1586; Zou et al., 2008, IEEE Sens J 8, 527-535; Zevenbergen et al., 2007, Nano Lett 7, 384-388). Low-fabrication cost microscale electrochemical systems which have smaller sample volume requirements are attractive for the detection of molecules by this method. (Ino et al., 2011, Lab Chip 11, 385-388; Hwang et al., 2009, IEEE Sens J 9, 609-615).
Each of the amperometric and potentiometric techniques requires a stable reference electrode to provide accurate measurements. With the emergence of microfabrication techniques, miniaturized electrochemical sensors are now being developed and integrated inside fluidic systems (Lewis et al., 2010, Anal Chem, 82, 1659-1668; Kwakye et al., 2006 Biosens Bioelectron, 21, 2217-2223; Swensen, et al., 2009, J Am Chem Soc, 131, 4262-4266; Wang et al., 2008, Sensors, 8, 2043-2081; Strayer, et al., 2012, Lab Chip, 12, 1548-1553). However, many of these systems rely on macroscale reference electrodes that are not directly inserted into the microfluidic environment or pseudo reference electrodes that are not stable over multiple days (Dam et al., 2007 Analyst, 132, 365-370; Niwa et al., 1991, Electroanal, 3, 163-168; Goluch et al., 2009, Anal Bioanal Chem, 394, 447-456; Kätelhön et al., 2010, Anal Chem, 82, 8502-8509).
Thus, there remains a need for electrochemical sensors having reliable and stable miniaturized reference electrodes.
Devices for electrochemical measurement of a concentration of a redox active substance, methods of fabricating the devices, and methods of measuring a concentration of a redox active substance using the devices are provided.
As used herein, the term “nanoscale” refers to an object or a feature whose size is in the range from about 1 nm to about 999 nm, or less than 1 μm. The term “microscale” refers to an object or feature whose size is in the range from about 1 μm to about 999 μm, or less than 1 mm. A “nanofluidic channel” as used herein is a space generally in the form of a channel and having a cross-sectional dimension of nanoscale size. Other dimensions of a nanofluidic channel, such as its length can be of microscale size or greater, or can also be of nanoscale size. A “microfluidic channel” as used herein is a channel having all cross-sectional dimensions in the microscale range or greater.
An embodiment of the invention is a device for measurement of a concentration of a redox active substance in a sample, the device including: a nanofluidic electrode assembly having a nanofluidic channel disposed in a substrate, a working electrode disposed in the nanofluidic channel, and a reference electrode disposed in the nanofluidic channel; a microfluidic channel in said substrate, the microfluidic channel in fluid communication with the nanofluidic channel; such that the working electrode is suitable for applying a potential sufficient to alternately oxidize and reduce the redox active substance; and such that the concentration of the redox active substance present in the nanofluidic channel is measured as a current flow through the working electrode. In related embodiments, the device further includes a counter electrode. For example, the counter electrode monitors the current flowing through the reference electrode and is useful to keep the current flow through the reference electrode to a low level, such as less than about 1 nA.
Another embodiment of the invention is a device for measurement of a concentration of a redox active substance in a sample, the device including: a nanofluidic electrode assembly having a nanofluidic channel disposed in a substrate, an oxidizing working electrode disposed in the nanofluidic channel, a reducing working electrode disposed in the nanofluidic channel, and a reference electrode disposed in the nanofluidic channel; a microfluidic channel disposed in said substrate, the microfluidic channel in fluid communication with the nanofluidic channel; such that the oxidizing working electrode is suitable for applying a potential sufficient to oxidize the redox active substance and the reducing working electrode is suitable for applying a potential sufficient to reduce the redox active substance; such that the concentration of the redox active substance present in the nanofluidic channel is measured as a current flow through the working electrode(s). In related embodiments, the oxidizing and the reducing working electrodes are separated by a distance of about 20-100 nm across the lumen of the nanofluidic channel.
In other related embodiments of the invention, the substrate of the devices includes a glass having a refractive index greater than 1.5. For example, the refractive index is in the range of 1.6-1.9. For example, the glass has a thickness of about 100 nm.
Related embodiments of the invention include devices that further include integrated Complementary Metal-Oxide Semiconductor (CMOS) electronics and a wireless transmitter.
The devices, in related embodiments, are suitable for measurement of a concentration of a redox active substance in a sample produced by a microorganism. In other related embodiments, the redox active substance is pyocyanin produced by Pseudomonas aeruginosa.
In various embodiments of the invention, the working electrode(s) of the devices include a material selected from the group consisting of: gold, electrically conductive diamond, platinum, and glassy carbon. In related embodiments, the reference electrode includes palladium.
Embodiments of the invention include devices that are configured for use as a medical device or a component of a medical device. In related embodiments, devices are disposable. In other related embodiments, a device is configured for implantation in a patient.
According to another embodiment of the invention, the reference electrode of the device forms at least a portion of a wall of the nanochannel. In related embodiments of the invention, a working electrode in the device forms at least a portion of a wall of the nanochannel.
In related embodiments of the invention, in the device having one working electrode, the reference electrode and the working electrode each form at least a portion of a wall of the nanochannel, and are separated by an insulating portion of the nanochannel, the insulating portion extending from about 30 nm to about 50 μm along the length of the nanochannel.
Related embodiments of the invention include devices having an oxidizing and a reducing working electrode such that the oxidizing and reducing working electrodes form portions of opposite facing walls of the nanochannel, the opposite facing walls separated by a distance of about 20-100 nm. In other related embodiments the reference electrode and the working electrodes each form at least a portion of a wall of the nanochannel, and the reference electrode is separated from both working electrodes by an insulating portion of the nanochannel, the insulating portion extending from about 30 nm to about 50 μm along the length of the nanochannel.
In some related embodiments, the working electrode(s) have a greater surface area than that of the reference electrode.
Embodiments of the invention include a device having a working and a reference electrode, such that the working electrode has a surface area of at least 100 nm2. In related embodiments, the area of the working electrode is less than 100000, less than 50000, less than 20000, less than 10000, or less than 500 nm2.
In other embodiments of the invention, the device has an oxidizing working electrode and a reducing working electrode, and the working electrodes each have a surface area of at least 1 μm2. [each or total?] In related embodiments, the area of each working electrode is less than 100, less than 50, less than 20, less than 10, or less than 5 μm2.
In related embodiments of the invention the one or the two working electrode devices above are such that the volume of the nanochannel in which the electrodes are disposed is less than about 50 nanoliters (nL), less than about 10 nL, less than about 1 nL, less than about 100 picoliters (pL), less than about 50 pL less than about 10 pL, less than about 5 pL, or less than about 1 pL.
Related embodiments include devices such that the width of the nanochannel in which the electrodes are disposed is less than 20 μm. In other related embodiments, the device further includes inlet and outlet ports in fluid connection with the microchannel, the ports configured to add and remove the sample to and from the microchannel. Related embodiments also include a device further including a processor that performs data analysis using the current flow through the working electrode(s). Also included as related embodiments of the invention are devices that display the current flow through the working electrode(s), or display the concentration of the redox active substance.
Another related embodiment of the invention is a device for simultaneous measurement of a concentration of each of a plurality of redox active substances in a sample, the device comprising a plurality of devices according to embodiments above, having a working electrode and a reference electrode.
Included as yet another related embodiment of the invention is a device for simultaneous measurement of a concentration of each of a plurality of redox active substances in a sample, the device comprising a plurality of devices according to the embodiments above having an oxidizing working electrode, a reducing working electrode, and a reference electrode.
In related embodiments the invention is a device for simultaneous measurement of a concentration of a redox active substance, and a presence of a biomass, the device including either the one working electrode or the two working electrodes embodiments above, glass having a refractive index greater than 1.5 as the substrate, and further including a window for measuring surface plasmon resonance of one of the working electrodes, such that the surface plasmon resonance indicates the presence of the biomass at said working electrode. In related embodiments, the invention is a device for simultaneous measurement of an amount of each of a plurality of redox active substances, and each of a plurality of biomass, the device comprising a plurality of devices, each including either the one working electrode or the two working electrodes embodiments above, glass having a refractive index greater than 1.5 as the substrate, and further including a window for measuring surface plasmon resonance of one of the working electrodes, such that the surface plasmon resonance indicates the presence of the biomass at said working electrode.
Another embodiment of the invention is a method of fabricating a nanoscale device for measurement of a concentration of a redox active substance in a sample, the device including a nanofluidic electrode assembly in fluid communication with a microfluidic channel, the method including the steps of: depositing a first electrode layer on a substrate; depositing a sacrificial layer on the first electrode layer; depositing a reference electrode layer on the sacrificial layer; depositing an insulating layer on the reference electrode layer; making two or more holes through the insulating layer, the holes providing a fluid connection to the sacrificial layer; applying an etching agent through the holes, whereby the sacrificial layer is etched away to form a nanochannel, the first electrode layer and reference electrode layer each forming a portion of a wall of the nanochannel; and preparing a microfluidic channel in the substrate, the microfluidic channel in fluid connection with the nanochannel. In related embodiments the method further includes applying a resist material over one or more layers and performing lithography to create a patterned mask prior to applying a subsequent layer. Another related embodiment of the method includes bonding a wire to an electrode layer. In various embodiments the microfluidic channels have forms selected from the group of: a channel, a reservoir, an open access port, and an open access port covered with a semipermeable membrane.
According to other related embodiments, in the method for fabricating a nanoscale device the thickness of the first electrode layer is about 20 nm. In other embodiments the thickness of the reference electrode layer is about 120 nm. Related embodiments include a method wherein the thickness of the sacrificial layer is about 20-200 nm. Further, in other related embodiments, the thickness of the insulating layer is about 500 nm. Embodiments include those in which the insulating layer includes silicon dioxide. Embodiments also include those in which the insulating layer includes a first layer of oxide, a layer of nitride, and a second layer of oxide. For example, the first and the second layer of oxide include SiO2, and the layer of nitride includes Si3N4.
In various related embodiments, the method is such that the first electrode layer includes a material selected from the group consisting of: gold, electrically conductive diamond, platinum, and glassy carbon. In other related embodiments the reference electrode layer includes palladium. In some embodiments the electrode layers are deposited by a method selected from the group consisting of: electron beam deposition, physical vapor deposition, and chemical vapor deposition. In related embodiments the method further includes depositing a second electrode layer on the sacrificial layer, the second layer and the reference layer being essentially coplanar and non-adjacent. In various embodiments the substrate is a silicon wafer having a surface layer of SiO2.
Embodiments of the invention include a method of measuring a concentration of a redox active substance present in a sample liquid including: providing a device having a one working electrode, or two working electrodes, as described in embodiments above, including the sample liquid loaded into the nanofluidic channel of the device; (ii) charging the reference electrode to stabilize its potential; (ii) recording the current flow through the working electrode(s); and (iv) determining a concentration of the substance producing the current flow by using a previously determined correlation between known concentrations of the substance and the current flow through the working electrode(s), recorded according to steps (i)-(iii). In related embodiments the device includes both oxidizing and reducing working electrodes, the method including: applying a potential suitable for oxidizing the substance to the oxidizing electrode, and a potential suitable for reducing the substance to the reducing electrode; and such that the substance has a half way potential value between the potential of the oxidizing electrode and the potential of the reducing electrode. For example, the sample is a bodily fluid selected from the group consisting of: bronchial lavage, sputum, urine, saliva, spinal fluid, and blood.
According to various related embodiments, the method further includes determining pH of the sample using a pH electrode disposed within the nanochannel, and a previously determined correlation between known amounts of the substance, each in a medium of known pH, and currents produced by the known amounts of the substance, recorded according to the steps (i)-(iii). In other related embodiments, the redox active substance is produced by a microorganism, the method further including: determining the effect of an agent on the microorganism by measuring amounts of the redox active substance produced by the microorganism upon treating the microorganism with defined amounts of the agent. For example, the agent is selected from the group consisting of an antibiotic, an antibody, an siRNA, a nanoparticle, and a signaling molecule.
Another embodiment of the invention is a method of measuring a concentration of a redox active substance present in a sample and a presence of a biomass including: (i) providing a device for simultaneous measurement of a concentration of a redox active substance and a presence of a biomass as described above, including the sample loaded into the microfluidic channel of the device; (ii) charging the reference electrode to stabilize its potential; (ii) recording the current flow through the working electrode(s); and (iv) determining a concentration of the substance producing the current flow by using a previously determined correlation between known concentrations of the substance and the current flow through the working electrode(s), recorded according to steps (i)-(iii), and (v) determining the presence of the biomass by measuring surface plasmon resonance changes at the working electrode proximal to the substrate. For example, the biomass is selected from the group consisting of: microorganisms in a medium, aggregates of microorganisms in a medium, and a biofilm of microorganisms. In related embodiments the working electrode is functionalized with a nucleic acid, a peptide, or an antibody.
The present invention provides devices for highly sensitive electrochemical detection of redox active substances in picoliter volumes of liquid sample. The devices use a nanofluidic electrode assembly having all required electrodes (i.e., reference electrode, one or more working electrodes, and optional counter electrode) integrated within a nanofluidic channel. The invention further provides sensor devices that incorporate the nanofluidic electrode assembly, combinations of such sensor devices, medical devices incorporating the sensor devices, as well as methods for fabricating and using the devices.
Methods of the invention include methods for the fabrication of a microscale or nanoscale palladium hydride reference electrode and its integration inside of a nanoscale constriction (nanofluidic channel) along with a working electrode, such as a gold working electrode, to create a complete electrochemical sensor device. After charging the palladium electrode with hydrogen, the device can be used to detect redox active agents (e.g., pyocyanin produced by Pseudomonas aeruginosa) at concentrations from 1-100 μM, with a detection limit of less than 1 μM. This is the first example of a palladium hydride reference electrode integrated with a microfabricated electrochemical sensor in a nanofluidic setup. In methods of using the device, a redox active agent can be measured over the course of several days, e.g., to monitor bacterial growth. By utilizing suitable voltammetry protocols, a redox active molecule can be selectively detected in a complex solution without the use of any electrode surface modification. In alternative embodiments of the invention, the working electrode is functionalized in order to increase the selectivity of detection.
Microfabricated nanofluidic electrode assemblies having an integrated palladium reference and two closely spaced working electrodes (an oxidizing working electrode and a reducing working electrode) inside the device nanochannel are also described herein. One such device has been used to monitor changes in pyocyanin concentration amperometric ally in real time. A linear response in faradaic current (R2=0.96) was observed over a biomedically relevant range of pyocyanin concentrations (0-100 μM) while continuously measuring the current for 2 hours. Measurement of the current that results from the repeated oxidation and reduction of pyocyanin at two closely spaced electrodes inside the device nanochannel yielded a 1.07 μM limit of detection without electrical isolation of the electrochemical cell. Since a reference electrode is integrated inside the nanofluidic channel of these sensors, they can potentially be employed to detect pyocyanin and other redox-active molecules in wide range of medical and environmental settings where space is limited, and without the risk of a biofilm forming on the reference or other electrodes. Nanofluidic electrode assemblies can also be used with an external Ag/AgCl reference electrode, e.g., to determine the concentration of a redox active agent in a solution, such as a patient sample or a culture medium. This type of analysis is completed in less than two minutes and also has a sub micromolar detection limit.
Amperometric detection provides a simple means of measuring the concentration of an electro-active species in solution. This is done by applying to an electrode a potential that oxidizes or reduces the molecule of interest. As the molecule of interest interacts with the electrode surface, one or more electrons are transferred between the two, resulting in the flow of a measurable current. Numerous compounds, including histamine, glutathione and dopamine, have been successfully detected using this approach (Lacher et al., 2001, Electrophoresis 22, 2526-2536; Dam et al., 2007, Analyst 132, 365-370; Pihel et al., 1995, Anal Chem 67, 4514-4521). However, often, the molecule of interest is in a complex medium that contains multiple electro-active species that react at the same potential, thus requiring that electrochemical detection be coupled with a separation technique, such as liquid chromatography (LC) or capillary electrophoresis (CE). While improving selectivity and sensitivity, these approaches also require additional processing time and reagents, and increase the device footprint. In certain situations, this can be avoided. If the molecule of interest is stable in both its oxidized and reduced form, a technique known as redox cycling can be employed to amplify the measured current over unstable molecules.
Redox cycling typically uses two working electrodes, with the potential of one electrode set to oxidize the target molecule, while the second electrode is set to a reducing potential. When the target molecule interacts with the oxidizing electrode it donates an electron, and when it interacts with the reducing electrode it gains the electron back and is able to repeat the process (
The theoretical limiting current for a given concentration of an electrochemical molecule undergoing redox cycling in a confined channel is determined by
Where i is the limiting current, n is the number of electrons being transferred, F is Faraday's constant, S is the surface area that overlaps between the two electrodes, D is the diffusion constant of the molecule being studied, C is the concentration of the redox species, and z is the distance between the two electrodes (Goluch et al. 2009, Anal Bioanal Chem 394, 447-456). Equation 1 shows that for a given molecule, the net current measured at an electrode can be increased by either increasing the area that overlaps between electrodes or decreasing the spacing between electrodes. Decreasing the vertical spacing between the two electrodes (i.e. the two electrodes are placed on top of one another), increases the net current response of these devices. The benefit of decreasing the spacing between electrodes over increasing the area of overlap is that it allows for smaller device footprints.
Electrochemical detection systems utilizing redox cycling in nanofluidic channels have been shown to have increased sensitivity over their large scale counterparts, and have even been shown to detect single molecules (Zevenbergen et al., 2007, Nano Lett 7, 384-388); Zevenbergen et al., 2011, Nano Lett 11, 2881-2886). The ease of use of this method lends itself to the detection of biologically relevant molecules (Goluch et al., 2009, Anal Bioanal Chem 394, 447-456; Wolfrum, et al., 2008 Anal Chem 80, 972-977; Webster and Goluch, 2012 Lab Chip 12, 5195-5201). However, none of these earlier studies has utilized an integrated reference electrode. In contrast, the present invention provides a microfabricated electrochemical sensor with an integrated reference electrode. The entire footprint of the sensor, including the entire electrode assembly, can be, for example, as small as approximately 15 μm by 250 μm.
An electrochemical cell used for potentiometric analysis typically consists of three electrodes: a working electrode, a reference electrode, and a counter electrode, which are used to measure the current at a given potential. Measurements are made at the surface of the working electrode and the resulting potential is compared against the half-reaction taking place at the reference electrode. An ideal reference electrode is constructed from a material that has a high current exchange density, is nonpolarizable, and is not destroyed by the reaction taking place. The counter electrode is employed to ensure that the current passing through the reference electrode is never large enough to damage the electrode. When small currents are measured at the working electrode, such as less than 1 nA, the reference electrode can also serve as a counter electrode, allowing a two electrode setup. The current flowing through the working electrode, which can be measured using an amperometry circuit that is part of the device or a separate component, can be used as a measure of the concentration or amount of the selected analyte.
By decreasing the size of the electrode assembly, many individually addressable electrode assemblies can be packaged onto a single detection platform. By constructing electrode assemblies spaced less than 100 nm apart, it has been shown that the sensitivity of electrochemical devices can be increased several-fold as a result of the short molecular transport times between electrodes in these confined spaces (Wolfrum et al., 2008, Anal Chem, 80, 972-977; Zevenbergen et al., 2007, Nano Lett, 7, 384-388; Zevenbergen et al., 2011, Nano Lett, 11, 2881-2886). However, miniaturization, stability, and reliability of the reference electrode has not been addressed prior to the present invention.
Macroscale reference electrodes can be used in tandem with micro and nanofluidic electrodes. However, IR drop, which is based on the distance between the reference electrode and working electrode, can lead to errors in the measured currents, which negatively impacts device sensitivity. Scale down is further impeded by the dissolution of the electrolyte solution that many reference electrodes employ to increase their stability. For instance, a silver/silver chloride (Ag/AgCl) reference electrode consists of a silver wire that is coated with silver chloride. The wire is usually surrounded by a solution saturated with chloride ions to maintain a constant potential at the reference electrode. When placed into a test solution small amounts of the chemicals can diffuse through the barrier, but since the solution is saturated, the Ag/AgCl reference electrode is not affected by this small change. However when a Ag/AgCl electrode is scaled down to the microscale, then the diffusion of even a small amount of the test solution into the reference electrode solution is more pronounced and quickly causes degradation of the reference electrode. Thus, by employing the nanoscale reference electrode of the present invention, which fits within a nanofluidic channel, a considerably more stable reference electrode is achieved, providing improved accuracy and durability of the sensor device.
To avoid dissolution issues involved with microscale AgCl surfaces, a different reference material is required. Palladium (Pd), a noble transition metal, is capable of holding up to 900 times its volume of dissolved hydrogen within its crystal structure. It functions in a manner similar to the standard hydrogen reference electrode in that the potential is set by the hydrogen ion concentration versus the amount of absorbed hydrogen molecules this type of reference electrode is also referred to as a palladium hydride (PdH) reference electrode. The potential remains constant as the amount of absorbed hydrogen changes over a wide range of values. By storing the hydrogen within the metal, the electrode stability greatly increases and is only affected very slightly by changes in the solution pH, making it an attractive option for biosensors. The increased stability makes PdH a useful reference electrode material and researchers have focused on optimizing PdH REs for bulk electrochemical measurements (Czerwiński et al., 1991 J Electroanal Chem Interfacial Electrochem, 316, 211-221; Czerwinski and Marassi, 1992, J Electroanaly Chem Interfacial Electrochem, 322, 373-381; Czerwiński et al., 1995, J Electroanal Chem Interfacial Electrochem, 386, 207-211), but not for microfluidic or nanofluidic environments, prior to the present invention. Microfabricated Pd structures are often used in sensors to detect hydrogen gas at very low concentrations where the electrode potential is affected by the hydrogen concentration (Zeng et al., 2011, Nano Lett, 11, 262-268). Odijk et al (2009) integrated a microfabricated Pd reference electrode into a microfluidic device that is employed in oxidizing metabolites in aqueous solution (Odijk et al., 2009, Lab Chip, 9, 1687-1693). However, in the microfluidic device the electrodes are accessible to large particles such as a bacterial cell which can interact with and attach to the electrodes. In the devices of the present invention, however, access of bacterial cells and larger cells to the reference electrode is prevented by placing the reference electrode surface within a nanofluidic channel too small to permit the entry of bacteria or other cells.
Described herein are methods of fabricating miniaturized Pd metal electrodes in a single metal patterning step such that the electrode is integrated inside a nanofluidic sensor assembly. The fabrication can be carried out using standard photolithography and physical deposition techniques. The half-reaction in a PdH reference electrode evades the common problem of electrolyte dissolution in buffered solutions. Miniaturizing the reference electrode allows it to be placed near the working electrode, thus avoiding IR drop. For these reasons a PdH reference electrode was microfabricated inside a nanofluidic cavity along with a working electrode (e.g., a gold working electrode) to obtain a nanofluidic electrode assembly (
An exemplary use of the nanofluidic electrode assembly described herein is detection of a microbe, such as a human pathogen, by detection of a redox active compound released from the microbe. Ventilator associated pneumonia is a prevalent affliction that plagues many immune compromised patients in hospitals, and high mortality rates have been observed. A significant contributor to ventilator associated pneumonia as well as wound and burn victim infections is the opportunistic pathogen Pseudomonas aeruginosa (PA). One of the many toxins produced and secreted by Pseudomonas aeruginosa is pyocyanin. It has the ability to oxidize and reduce other molecules and has been shown to induce neutrophil apoptosis (programmed cell death), decreased glutathione levels in epithelial cells, cause oxidative stress to cells, and inhibit the beating of lung cilia, all of which allow bacterial infections to thrive, and increases the prevalence of chronic infections.
Currently, the most common method for identification of microbial pathogens is culturing colonies, which takes 24 hours or more. Polymerase chain reaction (PCR) identification is faster, but it still takes over an hour and requires expensive reagents. In the case of PA pulmonary infections, the sputum of patients can contain up to 130 μM concentrations of pyocyanin, and concentrations in the millimolar range have been reported in ear secretions obtained from patients with PA ear infections. These concentrations make it possible to use pyocyanin as a diagnostic marker in electrochemical sensing techniques. Since the pyocyanin molecule is redox active, it can be detected electrochemically. Carbon electrodes were used by Sharp et al (2010) to selectively detect pyocyanin in bulk PA cultures; however the large size of the electrodes does not lend itself to microfluidic systems (Sharp et al., 2010, Bioelectrochem, 77, 114-119). On the other hand, embodiments of the microfabricated nanoelectrode assembly described herein can be used for monitoring the production of pyocyanin by several strains of PA using only a few hundred microliters of sample. Sensor devices employing such nanoelectrode assemblies can be incorporated into medical devices such as respirators, or disposable components thereof, such as tubes or masks. The sensors can be connected to control or monitoring electronic devices either by direct electrical connections or by wireless transmitter, and can provide important patient monitoring information in real time, or at selected times or time intervals.
Another embodiment of the invention is a device having a nanofluidic electrochemical assembly with two working electrodes and an integrated palladium reference electrodes. Such a device makes use of redox cycling to provide greater sensitivity. The exemplary redox active substance pyocyanin produced by PA was detected in these devices using redox cycling, which allows continuous real time monitoring of samples containing the bacteria. The devices have been used to measure changes in pyocyanin concentration in a buffered solution in a microwell that is open to the environment and not electrically insulated. Unlike many biological compounds that react at oxidizing potentials, pyocyanin reacts at a fairly negative reducing potential. By judiciously selecting the potentials that are applied to the two working electrodes, it is possible to avoid the effects of other phenazines and redox-active molecules found in bodily fluids (Imokawa, et al., 2006, Anal Chem, 78, 265-271; Zeng et al., 2011, Nano Lett,11, 262-268). Using the device of this embodiment, pyocyanin concentration was measured also in trypticase soy broth. For these measurements an external Ag/AgCl electrode was used in conjunction with the two working electrode nanofluidic electrode assembly device. The device setup was electrically isolated from the environment, and the potentials turned off between samples. In contrast to methods involving cell culture and genetic identification, concentration measurements for detection of P. aeruginosa can be carried out using the nanoelectrode assembly device described herein in less than 2 minutes.
Solutions were prepared with ultra-purified water from a Milli-Q system (Millipore resistance>18 Mohms). Phosphate buffer (PB), for use as supporting electrolyte, was prepared as a 1 M stock concentration from sodium phosphate dibasic heptahydrate and sodium phosphate monobasic anhydrous (Fisher Cat. BP331-1 and Cat. BP329-1, respectively) at pH 6.8. Pyocyanin was purchased as a solid from Cayman Chemicals (Cat. 10009594) and dissolved in Milli-Q water to make 150 μM stock solutions. The stock pyocyanin solutions were then diluted with phosphate buffer to make samples for testing. As a confirmation that the constructed devices were functioning properly, ferrocene dimethanol (Sigma-Aldrich Cat. 372625) was dissolved in 1 M KCl (Fisher Cat. P330-500) at concentrations ranging from 1 to 1000 μM (see, e.g.,
A scheme for the fabrication of a nanoelectrode assembly sensor with integrated reference electrode is shown in
Metal deposition was accomplished using an electron beam physical vapor deposition system that operated at 9 kV and currents of 15, 45, 50, and 120 mA for chromium (Cr), titanium (Ti), palladium (Pd), and gold (Au) deposition, respectively. The bottom electrode was constructed by depositing 5 nm of Ti, 20 nm of Au (120), and 2 nm of Cr. The sacrificial Cr layer (130) was constructed by depositing 56 nm of Cr on top of the bottom electrode (120). The Pd electrode (140) was aligned over the sacrificial Cr layer and bottom electrode, and was fabricated by depositing 2 nm of Cr, followed by 120 nm of Pd and 5 nm of Ti. 500 nm of silicon dioxide (110) was plasma sputtered to insulate the devices. Access holes (150) were etched down to the sacrificial Cr layer using inductively coupled plasma (Unaxis ICP Etch PlasmaTherm 790) at an RF1 and RF2 power of 250 and 300 watts respectively with an argon and sulfur hexafluoride flow rate of 4.0 and 6.0 sccm, respectively, for 3 minutes. In
Electrical connections were made by wire bonding the finished microfabricated devices to a USB connector (West Bond). Removal of the sacrificial Cr layer was accomplished using chromium etchant (Transene 1020AC) at room temperature. Upon removal of the sacrificial layer an open space (160) was created between the bottom electrode and the Pd electrode, forming the lumen of the nanofluidic channel Removal of the Cr layer was monitored by using the Au electrode as the working electrode and the Pd electrode as the reference/counter electrode and measuring current as a function of time at a potential of 0.005 V. As the chromium metal is removed the resistance between the circuit increases, leading to a decrease in current. The remaining chromium etchant was washed out with deionized water before testing the device further. This fabrication process created the entire nanoelectrode assembly.
Devices fabricated according to Example 2 were reversibly bonded to the bottom of a polydimethysiloxane (PDMS) well, which contained a 500 μm diameter hole in the bottom, forming a water-tight seal. The sample volume of the wells was approximately 200 μL. An Ag/AgCl reference electrode was placed in the solution through the top of the well during the characterization of the PdH reference electrode.
Electrochemical measurements were performed using the devices. To control the potentials applied to the nanoelectrode assembly a bipotentiostat (CHI842C) was utilized. For palladium reference electrode characterization a Ag/AgCl electrode (1 M KCl) (CHI111P) was used. To complete the electrochemical cell a platinum (Pt) counter electrode (CHI115) was employed. To ensure that the devices were functioning as expected, cyclic voltammograms of 1-100 μM ferrocene dimethanol versus Ag/AgCl reference electrode were obtained (see Example 4 and
Square wave and differential pulse voltammetry, which are more sensitive measurement techniques than cyclic voltammetry, were used for pyocyanin measurements. A counter electrode was not included in the microfabricated nanoelectrode assembly. However, the construction of microfabricated Au and Pt electrodes has been well documented and could be added with one extra photolithography step Goluch et al., 2009, Anal Bioanal Chem, 2009, 394, 447-456; Silva et al., 2009, J Brazil Chem Soc, 20, 1235-1241; Chung-Chiun, 1993, Appl Biochem Biotech, 41, 99-107.
Testing of the completed devices was performed using ferrocene dimethanol in concentrations ranging from 1-100 μM dissolved in 1 M KCl solution (
To determine the stability of the constructed reference electrode with time, differential pulse voltammograms of ferrocene dimethanol were measured at different time intervals. Over a 4 day period of measurements the oxidation potential of ferrocene dimethanol versus the PdH reference electrode was found to be approximately 28 mV with a standard error of only 2.2 mV. (
Once the chromium in the sacrificial layer was removed, the PdH electrode was used as a working electrode and charged at −0.375 V vs. a Ag/AgCl reference electrode in 2 N HCl for 20 minutes. This was done to load the Pd electrode with hydrogen to stabilize its potential allowing it to be used as a reference electrode. After charging, the open circuit potential (OCP) versus time was monitored to determine the stability of the PdH reference electrode in 2 N HCl. As can be seen in
The fully charged PdH reference converted to a semi-charged state after a few hours, even without use, resulting in a significant shift of its open cell potential versus Ag/AgCl (
The pH and NaCl concentration dependence were measured and are shown in
The nanoelectrode assembly was used to detect pyocyanin in phosphate buffer.
Different potential ranges were used in the two sets of experiments to account for the potential difference between the two reference half-reactions.
Differential pulse voltammetry was used to scan pyocyanin solutions (1-100 μM) in a voltage interval of −0.7 to −0.1 V at an amplitude voltage of 0.05 V and a frequency of 15 Hz. Three samples were measured at each concentration. Current measured with 100 mM phosphate buffer alone was subtracted from each scan. The average maximum current and standard deviation for three measurements at each concentration are shown
Two different nanoelectrode assembly devices were compared (
The oxidation potential of pyocyanin in phosphate buffer was compared to that in trypticase soy broth. A significant shift in the oxidation potential was observed when pyocyanin was dissolved in phosphate buffer versus pyocyanin produced by wild type P. aeruginosa in trypticase soy broth measured using a Pd reference electrode. The shift in oxidation potential was observed was believed to be associated with components in the trypticase soy broth. Scans of 100 μM pyocyanin in phosphate buffer and trypticase soy broth were performed. Systematically changing the composition of the solution revealed a gradual shift in oxidation potential for pyocyanin as shown in
In order to selectively detect the production of pyocyanin by P. aeruginosa, mutant strains (e.g., phzS) that are incapable of producing pyocyanin were used as controls. A peak was observed in the selected scan range for all strains. In the case of phzS, the size of the peak was significantly smaller than the pyocyanin peak produced by the wild type strain throughout the duration of the experiments and its maximum was shifted approximately 0.075 V. From
Differential pulse voltammograms of PA14 wild type supernatants were recorded over a six day period. Scans were performed from −0.5 to −0.1 V at an amplitude voltage of 0.05 V and a frequency of 15 Hz. A shift of only 0.020 V was observed in the oxidation peak of pyocyanin produced by wild type bacteria over the course of 6 days. The results are shown in
The microfabricated nanoelectrode assembly devices were used to detect pyocyanin produced from P. aeruginosa strains PA14 (wild type), pelA, phzS, and phzM. The strains were grown at 37° C. for a period of 8 days in trypticase soy broth. A single nanoelectrode assembly was charged prior to the start of these tests. For each measurement 400 μL of solution was removed from each of the liquid cultures and filtered using a 0.2 μm sterile nylon filter (Fisher Cat. 09-719C). The filtered samples were scanned from −0.5 to −0.1 V using differential pulsed voltammetry to detect the presence of pyocyanin. It was observed that the pyocyanin oxidation peak shifted to more positive potentials in trypticase soy broth compared to phosphate buffer (
The currents measured in
As shown in
To confirm that pyocyanin was responsible for the increase in current, four strains of PA were studied, in two of which pyocyanin production was inhibited. Strain pelA is capable of producing pyocyanin but does not contain the genes necessary to produce a robust biofilm. It is believed that a biofilm deficient strain would reach a lower cell density than the wild type, thus leading to a lower concentration of pyocyanin. The data in
In the phzS and phzM strains the necessary genes required to produce pyocyanin have been removed, and little to no current was expected to be detected from the supernatant of these samples. However, this was not the case. Most likely the current detected in these strains is from another molecule that is produced by these strains which has a redox potential close to pyocyanin. Several pyocyanin precursor molecules can still be produced by these strains. If the pyocyanin deficient strains produced another phenazine with a very similar redox potential compared to pyocyanin additional analysis would be needed to distinguish between the molecules (Wang and Newman, 2008, Environ Sci Tech, 42, 2380-2386). It was observed that the peak produced by phzS was at a different potential than the peak generated by the wild type strain, indicating that an electroactive molecule other than pyocyanin was being produced by phzS (
Stability tests for the PdH reference electrode showed that it was not stable for longer than an hour of continuous measurement. Therefore, after an initial charge, measurements, lasting only a few minutes each were made at discrete times during experiments without recharging the electrode. Scans of the wild type samples on different days showed a shift in the oxidation peak potential of only approximately 0.020 V (
Electrodes were microfabricated on 3-inch silicon wafer substrates, on which 500 nm of silicon dioxide (SiO2) was grown thermally to provide electrical insulation. Metal layers were patterned using standard photolithography techniques and a lift-off process. Gold, chromium, gold, and palladium metal layers were deposited in this order using electron beam deposition. The thicknesses of the gold bottom electrode (
Electrochemical measurements were made with a CHI 1040C multi-potentiostat. A potential of −1.5 V and 0 V was applied to the top and bottom working electrodes respectively, and current was measured every 0.1 second for 5000 seconds. The palladium metal served as both the counter and reference electrodes for these experiments. Since only currents up to a few nanoamps were measured in these experiments, a separate counter electrode was not necessary. The large difference between the potentials at the two working electrodes was chosen to ensure that the pyocyanin, which has a half-wave potential of −0.48 V vs. a semi-charged PdH reference, was both oxidized and reduced inside the nanochannel. Previous studies have shown palladium reference electrodes to be stable, and the large potential difference between the potentials at the two working electrodes, while unnecessary, demonstrated the ability of these devices to monitor pyocyanin production by redox cycling. For solitary analysis of sample concentration, an external Ag/AgCl reference electrode (CHI 111P) was used and 0 V was applied to the top working electrode while −0.35 V was applied to the bottom working electrode in the nanoelectrode assembly.
The performance of nanoelectrode assemblies according to Example 14 with integrated palladium reference electrodes was characterized using amperometric detection. Pyocyanin was chosen as it is a stable redox-active molecule of significant biological importance.
A microwell made of PDMS was placed on top of a nanoelectrode assembly, aligned with, and attached to the assembly. The microwell was filled with 60 μL of a 100 mM phosphate buffer test sample. To achieve a specific concentration of pyocyanin in the microwell, 20 μL of the buffer was removed and replaced with 20 μL of a phosphate buffer solution containing the required concentration of pyocyanin. During amperometric detection, the top electrode was held at −1.5 V while the bottom electrode was held at 0V versus the palladium reference electrode.
The detections were performed in the manner above because the sensor must remain filled with a solution containing ions during throughout the period of the detections. The sodium and phosphate ions present in the solution screen the charge at the electrode surface. If the nanoelectrode assembly were allowed to run dry, there would be no ions to screen the charge at the electrodes, resulting in an electric field and attractive force that may collapse the nanochannel.
Initially, the microwell was filled with 60 μL of 100 mM phosphate buffer and the response to the buffer alone control sample (blank) was monitored as a function of time. To show the response of the fabricated devices to different concentrations of pyocyanin, at different time points 20 μL of sample solution was removed from the well and replaced by 20 μL of 100 μM pyocyanin in 100 mM phosphate buffer, followed by agitation. As seen in
A total of six additions were made, resulting in sample concentrations of 33.3, 55.6, 70.4, 80.2, 86.8, and 91.2 μM. As expected, the current response increased linearly as the concentration of pyocyanin in the microwell increased (
To determine the response of the device to decreasing pyocyanin concentration the opposite experiment was performed. Starting with a concentration of 100 μM pyocyanin in 100 mM PB in the microwell, 20 μL of solution was removed and replaced with 20 μL of 100 mM PB followed by agitation. This procedure diluted the pyocyanin concentration in the sample. A total of six dilutions were performed (
Using Equation 1 the calculated current for 100 μM pyocyanin in the constructed device is approximately 36.1 nA. The calculated current is almost 4 times larger than that observed and shown in
The average current for each concentration of pyocyanin was calculated from 25 seconds of steady-state current obtained after sample injection. The error bars shown in
Differences between continuous environmental and single sample laboratory measurements using the nanoelectrode assembly device having two working electrodes was analyzed. Response of the nanoelectrode assembly device sensor to five samples containing 0, 1, 10, 50, and 100 μM pyocyanin in trypticase soy broth is shown in
Under controlled experimental conditions, the devices performed very well. A single device was used to obtain the data in
The average current for each concentration was calculated from 25 seconds of steady-state current that is obtained two minutes after the start of the measurement. The average response of the blank control was subtracted from each sample and the resulting current versus concentration was plotted for the bottom electrode in
An integrated micro/nanofluidic device that includes electrochemical and plasmonic sensors integrated inside the channels is constructed. These sensors permit in situ and in vitro analysis of biomass, such as formation of a biofilm by bacteria. Micro/nanofluidic structures are integrated with a 2-dimensional array of electrochemical sensors based on the design shown in
Measurement of biofilm growth in microchannels using surface plasmon resonance imaging is shown in
Polymer nanochannels are constructed that allow efficient transfer of fluid to and from the sensor area while preventing the escape of cells into the nanochannels. Nanochannels are patterned using electron-beam lithography.
Bacterial cells are exposed to different chemical environments and gradients and the phenotypic response is monitored. Initially, P. aeruginosa wild-type and mutant strains with biofilm production suppressed, such as pelA, are used. Cultures of each strain are grown and diluted so that each microchamber has only one or a few cells initially. The effects of growth phase on the measurements recorded is taken into account. The number of cells that are present in the microchamber at the beginning of each experiment is recorded. Control experiments are performed to determine if there is any observable difference in the development of biofilms with small variations in starting concentrations ranging from 1 to about 20 cells per chamber. The cells are transported through the microfluidic channels using pressure driven flow. After cells are trapped and fluid exchange takes place between the microfluidic and nanofluidic channels due to diffusion, a redox active chemical of interest produced by the cells reaches a nanochannel, where it is detected electrochemically due to their redox active nature and interaction with a nanofluidic electrode assembly located within the nanochannel.
In the beginning of the measurements the bacterium is contained in minimal medium. Food is then introduced through nanochannels that connect to the microfluidic channel for the delivery of chemical substances to the microchannel. Penicillin and kanamycin, antibiotics that do not kill P. aeruginosa, are introduced to check if the antibiotics have any effect on biofilm formation. Next, ampicillin, which is known to kill P. aeruginosa, is used. For each antibiotic, the concentration is varied to create a concentration gradient. The starting time for addition of an antibiotic is also varied to systematically determine any effect the biofilm may have on the development of antibiotic resistance. The antibiotic is then diluted and fresh growth media added to check for persister (dormant) cells. P. aeruginosa cells change their behavior when co-cultured with other cells, indicating that they detect molecules, or signaling factors, produced by other species. The response of P. aeruginosa to acyl homoserine lactones (AHLs) and Pseudomonas quinolone signal (PQS), common signaling factors for other bacterial species, is determined.
An array of nano-electrochemical sensors for providing label-free detection of redox-active small molecules and metabolites in mature biofilms is constructed. The electrochemical sensors are coupled with SERS to obtain additional information about the spatiotemporal chemical changes that take place in the biofilm. The bacteria S. aureus and S. epidermidis are used for producing biofilms as they are primarily responsible for infection in medical implants. A schematic diagram of such an array is shown in
A two layer process is employed to create valves for directing the flow of cells and to trap them once they are near a sensor. The basic fabrication scheme is shown in
Once completed, a custom-built, chip alignment instrument is utilized to arrange the microfluidic channels (1680) so that they sit on top of the sensors (1650). Pneumatic actuation will be used to move and confine the cells in the vicinity of the sensors, thus creating a confined volume of approximately 100 picoliters. For more complicated systems, innovative designs will be required in order to efficiently transport and trap large numbers of individual cells. To simplify the number of control valves involved, a self-assembly approach may be utilized, where only a single cell is able to fit into a chamber containing sensors.
An advantage of using nanofluidic electrochemical detection is that real-time electrochemical information can be gathered passively while permitting integration with microfluidic systems, where the technology exists for increasing the device complexity. The microfluidic sample handling and nanofluidic sensors are developed in tandem, followed by testing of the complete system. Initial electrochemical testing is conducted with known commercially available quorum sensing molecules (e.g. pyocyanin, autoinducer-2), and validation of the microfluidic architecture utilizes microscale beads to simulate bacterial cells. In the final stage of this project, bacterial cells (e.g. E. coli, P. aeruginosa) are transported and trapped next to the nanofluidic sensing elements using a microfluidic platform. The generation of quorum sensing molecules by the cells is monitored electrochemically while the cellular response to the molecules is observed optically. A schematic of a single bacterial cell trapped above an electrochemical sensor is shown in
A nanoelectrode assembly was used for measuring the concentration of pyocyanin mixed into different biological fluids. Graphs of current measured for detecting pyocyanin in biological fluids using a commercial sensor Ag/AgCl reference electrode are shown in
The properties of biofilms produced by the same species can vary dramatically depending on the external conditions to which the cells are exposed. For applications such as testing of antimicrobial materials and coatings, it is useful to determine whether or not a biofilm is forming or being removed. Microfluidic SPRI setup described herein is useful for testing various antimicrobial coatings and materials. GFP E. coli cells were grown on the prism surface inside an incubator for several hours and allowed to dry. The prism covered with the dry biofilm was placed inside an SPRI system (
The sensor surface can be modified with various chemical coatings both nonspecifically and covalently. Since the surface plasmon extends about 200 nm away from the surface, monolayer and thin coatings can be employed without sacrificing sensitivity, although the field decays exponentially with distance. The imaging capability combined with customizable microfluidic geometries is used to perform multiple tests simultaneously. The channel wall opposite the sensor surface can be modified if the modification adversely affects sensor response. Since the channels are only a few micrometers tall, it is possible to monitor biofilm formation as cells accumulate and span the height of the channel. Solutions containing diluted cell concentrations can be flowed through the channels at varying flow rates to monitor adhesion.
Incorporated herein by reference in its entirety is an article authored by Thaddaeus A. Webster and Edgar D. Goluch, titled “Electrochemical Detection of Pyocyanin in Nanochannels with Integrated Palladium Hydride Reference Electrodes” Lab Chip, 2012, 12, 5195-5201; and an article authored by Thaddaeus A. Webster, Hunter J. Sismaet, and Edgar D. Goluch, titled “Amperometric detection of pyocyanin in nanofluidic channels”, Nano Life 2013, 3, 1340011-1 to 1340011-8. While the present invention has been described in conjunction with certain preferred embodiments, one of ordinary skill, after reading the foregoing specification, will be able to effect various changes, substitutions of equivalents, and other alterations to the compositions and methods set forth herein.
As used herein, “consisting essentially of” does not exclude materials or steps that do not materially affect the basic and novel characteristics of the claim. Any recitation herein of the term “comprising”, particularly in a description of components of a composition or in a description of elements of a device, can be exchanged with “consisting essentially of” or “consisting of”.
While the present invention has been described in conjunction with certain preferred embodiments, one of ordinary skill, after reading the foregoing specification, will be able to effect various changes, substitutions of equivalents, and other alterations to the compositions and methods set forth herein.
This application claims the benefit of provisional application Ser. No. 61/673,852 filed Jul. 20, 2012 and entitled “MICROFABRICATED ELECTROCHEMICAL SENSORS FOR P. AERUGINOSA INFECTION”, which is hereby incorporated herein by reference in its entirety.
The invention was developed with financial support from Grant No. 1125535 from the National Science Foundation. The U.S. Government has certain rights in the invention.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US13/51471 | 7/22/2013 | WO | 00 |
| Number | Date | Country | |
|---|---|---|---|
| 61673852 | Jul 2012 | US |
