Patent Application
20250066704
This disclosure relates to a microfluidic device and methods for performing a microfluidic assay.
Antibiotic resistance (AR) is a rising crisis worldwide. The rise of AR has largely been attributed to the overuse of antibiotics in human and veterinary medicine, and animal husbandry (Schoepp et al., “Rapid Pathogen-Specific Phenotypic Antibiotic Susceptibility Testing using Digital LAMP Quantification in Clinical Samples,” Sci. Transl. Med. 9, eaal3693 (2017) and Hong et al., “Assessing Antibiotic Permeability of Gram-Negative Bacteria via Nanofluidics,” ACS Nano 11:6959-6967 (2017)). AR can be caused by a number of factors including over-prescribing of antibiotics, a lack of rapid laboratory tests to help identify antibiotic susceptibility, and prescribing either an ineffective or only marginally effective antibiotic (US Food and Drug Administration. Antimicrobial Resistance Information from FDA (FDA, 2022)). Therefore, advancing the development and use of rapid diagnostic tests to identify and characterize resistant bacteria, leading to targeted antibiotic administration is a key in combating antibiotic resistance.
Currently, the workflow of a bacterial infection diagnosis takes 2-3 days and includes three steps: isolation, identification, and sensitivity (Reller et al., “Antimicrobial Susceptibility Testing: A Review of General Principles and Contemporary Practices,” Clin. Infect. Dis. 49:1749-1755 (2009); Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 32nd. CLSI Supplement (Clinical and Laboratory Standards Institute, 2022); and US Food and Drug Administration. Class II Special Controls Guidance Document: Antimicrobial Susceptibility Test (AST) Systems (U.S. Department of Health and Human Services, 2009)). It starts with sample collection, and the sample is either processed on location or shipped to an accredited diagnostic lab for sensitivity test. Upon arrival, the sample is plated onto non-selective and/or selective agar to enrich or isolate the bacteria in the sample. Then the infective agent is identified by morphological, biochemical, or molecular assays. This process can take up to 2 days depending on the type of sample, bacteria, and techniques used at each clinical microbiology lab. Then, after the bacteria are isolated, antibiotic susceptibility testing (AST) is conducted, which takes another 16-20 hours.
The following two phenotypic cultures are the current standard methods for AST: agar disk diffusion methods (Alagumaruthanayagam et al., “Evaluation of Solid (Disc Diffusion)—and Liquid (Turbidity)—Phase Antibiogram Methods for Clinical Isolates of Diarrheagenic E. coli and Correlation with Efflux,” J. Antibiot. (Tokyo) 62:377-384 (2009) and Gefen et al., “TDtest: Easy Detection of Bacterial Tolerance and Persistence in Clinical Isolates by a Modified Disk-Diffusion Assay,” Sci. Rep. 7:41284 (2017)), which provide susceptibility results based on clearing-zone diameter, and Broth microdilution (BMD) methods (Wiegand et al., “Agar and Broth Dilution Methods to Determine the Minimal Inhibitory Concentration (MIC) of Antimicrobial Substances,” Nat. Protoc. 3:163 (2008)), which provide results based on culture turbidity or metabolic activity. Interpretive standard breakpoints are used to evaluate the growth of bacteria in the presence of antibiotics and to determine the pathogenic resistance. Because of the long time from collection of samples to receipt of the results of the culturing, there is often a lack of timely information about antibacterial resistance at the decision making stage of veterinary treatments, especially in emergency cases (Rodriguez & Taban, “A State-of-Art Review on Multi-Drug Resistant (MDR) Pathogens in Foods of Animal Origin: Risk Factors and Mitigation Strategies,” Front. Microbiol. 10:2091 (2019); Verraes et al., “Antimicrobial Resistance in the Food Chain: A Review,” Int. J. Environ. Res. Public. Health 10:2643-2669 (2013); and NAHMS. National Animal Health Monitoring System (NAHMS) Dairy 2007: Salmonella and Campylobacter on US Dairy Operations (https://allie.dbcls.jp/pair/NAHMS; National+Animal+Health+Monitoring+System.html) (2009)). Physicians and veterinarians treat based on clinical presentation of the infection using antibiotics with a broad spectrum of activity rather than wait for AST results to choose an antibiotic specific to the infection (Mohan et al., “A Multiplexed Microfluidic Platform for Rapid Antibiotic Susceptibility Testing,” Biosens. Bioelectron. 49:118-125 (2013) and Nguyen et al., “Diffusion-Convection Hybrid Microfluidic Platform for Rapid Antibiotic Susceptibility Testing,” Anal. Chem. 93:5789-5796 (2021)). Thus, there is a need to develop new platforms for rapid, inexpensive, and easily implemented AST approaches to reduce the assay time without compromising accuracy so that more targeted antibiotics can be used.
Recently, microfluidic platforms have been shown to improve the sensitivity and speed of AST. These platforms use techniques such as single cell confinement (Choi et al., “Direct, Rapid Antimicrobial Susceptibility Test from Positive Blood Cultures Based on Microscopic Imaging Analysis,” Sci. Rep. 7:1-13. (2017); Choi et al., “Rapid Antibiotic Susceptibility Testing by Tracking Single Cell Growth in a Microfluidic Agarose Channel System,” Lab. Chip 13:280-287 (2013); Baltekin et al., “Antibiotic Susceptibility Testing in less than 30 min Using Direct Single-Cell Imaging,” Proc. Natl. Acad. Sci. USA 114:9170-9175 (2017); Choi et al., “A Rapid Antimicrobial Susceptibility Test Based on Single-Cell Morphological Analysis,” Sci. Transl. Med. 6:267ra174-267ra174 (2014); Syal et al., “Antimicrobial Susceptibility Test with Plasmonic Imaging and Tracking of Single Bacterial Motions on Nanometer Scale,” ACS Nano 10:845-852 (2016); Li et al., “Gradient Microfluidics Enables Rapid Bacterial Growth Inhibition Testing,” Anal. Chem. 86:3131-3137 (2014); Li et al., “Adaptable Microfluidic System for Single-Cell Pathogen Classification and Antimicrobial Susceptibility Testing,” Proc. Natl. Acad. Sci. USA 116, 10270-10279 (2019)), microchamber arrays (Nguyen et al., “Diffusion-Convection Hybrid Microfluidic Platform for Rapid Antibiotic Susceptibility Testing,” Anal. Chem. 93:5789-5796 (2021); Weibull et al., “Bacterial Nanoscale Cultures for Phenotypic Multiplexed Antibiotic Susceptibility Testing,” J. Clin. Microbiol. 52:3310-3317 (2014); Azizi et al., “Nanoliter-Sized Microchamber/Microarray Microfluidic Platform for Antibiotic Susceptibility Testing,” Anal. Chem. 90:14137-14144 (2018); Azizi et al., “Gradient-Based Microfluidic Platform for One Single Rapid Antimicrobial Susceptibility Testing,” ACS Sens 6:1560-1571 (2021); Azizi et al., “Biological Small-Molecule Assays Using Gradient-Based Microfluidics,” Biosens. Bioelectron. 178:113038 (2021); Azizi et al., “Antimicrobial Susceptibility Testing in a Rapid Single Test via an Egg-Like Multivolume Microchamber-Based Microfluidic Platform,” ACS Appl. Mater. Interfaces 13:19581-19592 (2021); Lin et al., “An Antibiotic Concentration Gradient Microfluidic Device Integrating Surface-Enhanced Raman Spectroscopy for Multiplex Antimicrobial Susceptibility Testing,” Lab. Chip 22:1805-1814 (2022); and Yi et al., “Direct Antimicrobial Susceptibility Testing of Bloodstream Infection on SlipChip,” Biosens. Bioelectron. 135:200-207 (2019)), droplet microfluidics (Boedicker et al., “Detecting Bacteria and Determining their Susceptibility to Antibiotics by Stochastic Confinement in Nanoliter Droplets Using Plug-Based Microfluidics,” Lab. Chip 8:1265-1272 (2008); Churski et al., “Rapid Screening of Antibiotic Toxicity in an Automated Microdroplet System,” Lab. Chip 12:1629-1637 (2012); and Svensson et al., “Coding of Experimental Conditions in Microfluidic Droplet Assays Using Colored Beads and Machine Learning Supported Image Analysis,” Small 15:1802384 (2019)), and asynchronous magnetic bead rotation (Sinn et al., “Asynchronous Magnetic Bead Rotation (AMBR) Biosensor in Microfluidic Droplets for Rapid Bacterial Growth and Susceptibility Measurements,” Lab. Chip 11:2604-2611 (2011)) to immobilize or contain the bacteria and conduct AST. Microchamber-based platforms allow integration of microfluidic concentration gradient generators, thus offering high-throughput screening of multiple antibiotic concentrations and experimental controls on the same chip. Previous platforms using simple diffusion have produced continuous concentration gradients of antibiotics (Kim et al., “Microfluidic-Based Observation of Local Bacterial Density Under Antimicrobial Concentration Gradient for Rapid Antibiotic Susceptibility Testing,” Biomicrofluidics 13:014108 (2019) and Kim et al., “On-Chip Phenotypic Investigation of Combinatory Antibiotic Effects by Generating Orthogonal Concentration Gradients,” Lab. Chip 19:959-973 (2019)), but these approaches present a problem in determining the minimum inhibitory concentration (MIC) precisely. More recent platforms were successful at generating concentration gradients that are linear (Kim et al., “Microfluidic-Based Observation of Local Bacterial Density Under Antimicrobial Concentration Gradient for Rapid Antibiotic Susceptibility Testing,” Biomicrofluidics 13:014108 (2019); Kim et al., “On-Chip Phenotypic Investigation of Combinatory Antibiotic Effects by Generating Orthogonal Concentration Gradients,” Lab. Chip 19:959-973 (2019); Chen et al., “High-Throughput Generation of a Concentration Gradient on open Arrays by Serial and Parallel Dilution for Drug Testing and Screening,” Sens. Actuators B Chem 305:127487 (2020); Tang et al., “A Linear Concentration Gradient Generator based on Multi-Layered Centrifugal Microfluidics and its Application in Antimicrobial Susceptibility Testing,” Lab. Chip 18:1452-1460 (2018); Nguyen et al., “A High-Throughput Integrated Biofilm-on-a-Chip Platform for the Investigation of Combinatory Physicochemical Responses to Chemical and Fluid Shear Stress,” PloS One 17:e0272294 (2022)), logarithmic (Chen et al., “High-Throughput Generation of a Concentration Gradient on open Arrays by Serial and Parallel Dilution for Drug Testing and Screening,” Sens. Actuators B Chem 305:127487 (2020)), or sigmoidal (Avesar et al., “Nanoliter Cell Culture Array with Tunable Chemical Gradients,” Anal. Chem. 90:7480-7488 (2018)). Nevertheless, these techniques employ unstandardized concentration gradients. They deviate from the recommended standard quantitative AST technique that is based on 2-fold serial dilutions of an antibiotic to produce an exponential gradient (Clinical and Laboratory Standards Institute, “Performance Standards for Antimicrobial Susceptibility Testing,” 32nd. CLSI Supplement (Clinical and Laboratory Standards Institute, 2022)). Note that the susceptibility/resistance determination is based on the MIC interpretive breakpoints; the lack of harmonization between these methods and the current standards may present a challenge in interpreting results and consequently the utilization of these platforms.
The present disclosure is directed to overcoming these and other deficiencies in the art.
One aspect of the present disclosure relates to a microfluidic circuit comprising a drug inlet port; an outlet port; a drug inlet main channel fluidically connecting the drug inlet port and the outlet port, where said drug inlet main channel comprises (i) a plurality of serpentine mixers and (ii) a plurality of dead-end first microchamber sets; a negative inlet port; a negative inlet main channel fluidically connecting the negative inlet port and the drug inlet main channel; a plurality (n) of ladder channels, where each of the plurality (n) of the ladder channels is fluidically connected to both the drug inlet main channel and the negative inlet main channel; and an outlet channel fluidically connected to the drug inlet main channel between the drug inlet port and the outlet port.
Another aspect of the present disclosure relates to a microfluidic device comprising a support layer; a substrate layer disposed on the support layer; and a microfluidic circuit of the present disclosure, where the microfluidic circuit is disposed within the substrate layer.
A further aspect of the present disclosure relates to a method for performing an assay. This method involves loading a test solution into a microfluidic device of the present disclosure. A culture medium containing an antibiotic solution is loaded into the drug inlet port, a culture medium is loaded into the negative medium inlet port, and the device is then washed with oil to isolate each of the plurality of microchambers. The method further involves detecting a fluorescent signal in each of the plurality of microchambers.
The Examples of the present disclosure (infra) demonstrate the design and development of certain embodiments of a microfluidic system with an optimized ladder shaped network that combines and distributes culture medium and antibiotic in a 2-fold serial dilution. Based on previously established protocols for microscale culture-based AST (Nguyen et al., “Diffusion-Convection Hybrid Microfluidic Platform for Rapid Antibiotic Susceptibility Testing,” Anal. Chem. 93:5789-5796 (2021); Azizi et al., “Nanoliter-Sized Microchamber/Microarray Microfluidic Platform for Antibiotic Susceptibility Testing,” Anal. Chem. 90:14137-14144 (2018); Azizi et al., “Gradient-Based Microfluidic Platform for One Single Rapid Antimicrobial Susceptibility Testing,” ACS Sens 6:1560-1571 (2021); Azizi et al., “Biological Small-Molecule Assays Using Gradient-Based Microfluidics,” Biosens. Bioelectron. 178:113038 (2021); and Azizi et al., “Antimicrobial Susceptibility Testing in a Rapid Single Test via an Egg-Like Multivolume Microchamber-Based Microfluidic Platform,” ACS Appl. Mater. Interfaces 13:19581-19592 (2021), which are hereby incorporated by reference in their entirety), antibiotic susceptibility can be determined in less than 5 hours using the disclosed ladder system. The ability of this platform to perform AST of bacterial isolates as called for by the standard method was evaluated and the feasibility of its use on bacteria retrieved directly from canine urine samples without prior isolation or enrichment was investigated. This new platform provides an adaptable diagnostic tool and maintains relevance with current national and international interpretive standards.
Unless otherwise indicated, the definitions and embodiments described in this and other sections are intended to be applicable to all embodiments and aspects of the present application herein described for which they are suitable as would be understood by a person skilled in the art.
Singular forms “a”, “an”, and “the” include plural references unless the context clearly dictates otherwise. Thus, for example, a reference to “a method” includes one or more methods, and/or steps of the type described herein and/or which will become apparent to those persons skilled in the art upon reading this disclosure. In another example, reference to “a compound” includes both a single compound and a plurality of different compounds.
The term “about” includes being within a statistically meaningful range of a value. Such a range can be within an order of magnitude, preferably within 50%, more preferably within 20%, still more preferably within 10%, and even more preferably within 5% of a given value or range. The allowable variation encompassed by the term “about” depends on the particular system under study, and can be readily appreciated by one of ordinary skill in the art.
The term “and/or” as used herein means that the listed items are present, or used, individually or in combination. In effect, this term means that “at least one of” or “one or more” of the listed items is used or present.
In understanding the scope of the present disclosure, the term “comprising” and its derivatives, as used herein, are intended to be open ended terms that specify the presence of the stated features, elements, components, groups, integers, and/or steps, but do not exclude the presence of other unstated features, elements, components, groups, integers and/or steps. The foregoing also applies to words having similar meanings such as the terms, “including”, “involving”, “having”, and their derivatives. The term “consisting” and its derivatives, as used herein, are intended to be closed terms that specify the presence of the stated features, elements, components, groups, integers, and/or steps, but exclude the presence of other unstated features, elements, components, groups, integers and/or steps. The term “consisting essentially of”, as used herein, is intended to specify the presence of the stated features, elements, components, groups, integers, and/or steps as well as those that do not materially affect the basic and novel characteristic(s) of features, elements, components, groups, integers, and/or steps. In embodiments or claims where the term comprising (or the like) is used as the transition phrase, such embodiments can also be envisioned with replacement of the term “comprising” with the terms “consisting of” or “consisting essentially of.” The methods, kits, systems, and/or compositions of the present disclosure can comprise, consist essentially of, or consist of, the components disclosed.
Certain terms employed in the specification, examples, and claims are collected herein. Unless defined otherwise, all technical and scientific terms used in this disclosure have the same meanings as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
Preferences and options for a given aspect, feature, embodiment, or parameter of the disclosure should, unless the context indicates otherwise, be regarded as having been disclosed in combination with any and all preferences and options for all other aspects, features, embodiments, and parameters of the disclosure.
One aspect of the present disclosure relates to a microfluidic circuit comprising a drug inlet port; an outlet port; a drug inlet main channel fluidically connecting the drug inlet port and the outlet port, where said drug inlet main channel comprises (i) a plurality of serpentine mixers and (ii) a plurality of dead-end first microchamber sets; a negative inlet port; a negative inlet main channel fluidically connecting the negative inlet port and the drug inlet main channel; a plurality (n) of ladder channels, where each of the plurality (n) of the ladder channels is fluidically connected to both the drug inlet main channel and the negative inlet main channel; and an outlet channel fluidically connected to the drug inlet main channel between the drug inlet port and the outlet port.
One embodiment of a microfluidic circuit of the present disclosure is illustrated in
Positioned in drug inlet main channel 16 are serpentine mixers 18 and dead-end first microchamber sets 20. In the embodiment illustrated in
With further reference to
A plurality (n) of ladder channels, illustrated in
Microfluidic circuit 10 of
In some embodiments, the microfluidic circuit of the present disclosure comprises a plurality of hydraulic resistor pinches. In the embodiment illustrated in
Examples of resistor pinches 30 that may be used in microfluidic circuit 10 of the present disclosure is illustrated in
With reference again to
The inset of
Each of dead-end first microchamber set 20 and dead-end second microchamber set 36 is illustrated in greater detail in
In some embodiments of the microfluidic circuit according to the present disclosure, the drug inlet main channel 16 and the negative inlet main channel 24 are fixedly positioned parallel to each other.
In some embodiments of the microfluidic circuit according to the present disclosure, each of the ladder channels 26 is positioned perpendicular to each of the main channels (i.e., the drug inlet main channel 16 and the negative inlet main channel 24). In some embodiments, the ladder channels 26 may be fixedly positioned parallel to each other between the drug inlet main channel and the negative inlet main channel.
In some embodiments of the microfluidic circuit according to the present disclosure, n−1 of the plurality (n) of ladder channels 26 each comprises a hydraulic resistor pinch 30. In some embodiments, the ladder channel positioned most proximate to the drug inlet port does not possess a hydraulic resistor pinch 30. In some embodiments, the hydraulic resistor pinch 30 in each of the ladder channels 26 is of a different length than the other ladder channels 26.
In some embodiments where the hydraulic resistor pinch 30 in each of the ladder channels 26 is of a different length than the other ladder channels 26, the length of each hydraulic resistor pinch 30 increases in the direction of the drug inlet port 12 to the outlet port 14. In some embodiments, the length of each hydraulic resistor pinch 30 is between 200 μm-10,000 μm or 2000 μm-10,000 μm.
In some embodiments of the microfluidic circuit according to the present disclosure, each of the microchambers 44 in the plurality of dead-end first microchamber sets 20 is individually fluidically connected to the drug inlet main channel 16 by a side channel 32.
The plurality of dead-end first microchamber sets 20 may comprise any suitable shape or size. In the embodiments illustrated in the drawings referenced herein, the microchamber sets have a circular shape, although other shapes may also be contemplated. In some embodiments, microchambers 44 are circular in shape. In some embodiments the dead-end first microchamber sets 20 comprise microchambers 44 that have a volume capacity of about 50 μm, about 100 μm, about 200 μm, about 300 μm, about 400 μm, or about 500 μm. In some embodiments, side channels 32 have a length between the main channel they are connected to and the microchamber of about 20 μm, about 30 μm, about 40 μm, about 50 μm, about 60 μm, about 70 μm, about 80 μm, about 90 μm, about 100 μm, about 110 μm, or about 120 μm.
In some embodiments of the microfluidic circuit according to the present disclosure, a pair of ladder channels 26 in sequence between the drug inlet main channel 16 and the negative inlet main channel 24 comprises a subunit 34 of the microfluidic circuit 10. In accordance with such embodiments, each subunit may comprise only one of the dead-end first microchamber sets 20. In some embodiments, at least one of the subunits 34 comprises a plurality of dead-end second microchamber sets 36. Each of the dead-end second microchambers 46 of the dead-end second microchamber sets 36 may be individually fluidically connected to the negative channel by a side channel 38. In some embodiments, the plurality of dead-end second microchamber sets 36 is comprised of three distinct microchambers 46.
In some embodiments of the microfluidic circuit according to the present disclosure, the plurality of dead-end first microchamber sets 20 is comprised of three distinct microchambers 44.
In some embodiments of the microfluidic circuit according to the present disclosure, the outlet channel 28 is positioned parallel to the drug inlet main channel 16. The outlet channel 28 may be fluidically connected to the drug inlet main channel 16 by one or more outlet side channels 40. In some embodiments, only a single outlet side channel 40 is positioned in the drug inlet main channel 16 between proximate pairs of the ladder channels.
In some embodiments of the microfluidic circuit according to the present disclosure, the outlet channel 28 comprises a plurality of serpentine mixers 42. In some embodiments, only a single serpentine mixer 42 is positioned in the outlet channel 28 between proximate pairs of the ladder channels 26.
The microfluidic circuit of the present disclosure pertains to a fluidic system of ports, channels, and microchambers, all fluidically connected. Since, as discussed in the Examples below, the particular structure of this fluidically connected system enables assays to be performed based on the unique structure of the circuit, the circuit may be used in a variety of contexts or on a variety of platforms. One particular platform where the microfluidic circuit of the present disclosure is useful is a chip-like platform. For example, the microfluidic circuit may be formed into a planar material to create the system of ports, channels, and microchambers, accessible for input or output of fluid only through the ports.
Thus, another aspect of the present disclosure relates to a microfluidic device comprising a support layer; a substrate layer disposed on the support layer; and a microfluidic circuit of the present disclosure, where the microfluidic circuit is disposed within the substrate layer.
With reference now to
In some embodiments of microfluidic device 50 according to the present disclosure, drug inlet main channel 16 and negative inlet main channel 24 may be fixedly positioned parallel to each other.
In some embodiments of the microfluidic device according to the present disclosure, each of ladder channels 26 may be positioned perpendicular to each of the main channels (i.e., drug inlet main channel 16 and negative inlet main channel 24). For example, ladder channels 26 may be fixedly positioned parallel to each other between the drug inlet main channel and the negative inlet main channel.
In some embodiments of microfluidic device 50 according to the present disclosure, n−1 of the plurality (n) of ladder channels 26 each comprises a hydraulic resistor pinch 30. In accordance with such embodiments, ladder channel 26 positioned most proximate to the drug inlet port may not possess a hydraulic resistor pinch 30 and/or hydraulic resistor pinch 30 in each of ladder channels 26 may be of a different length than the other ladder channels. In some embodiments, the length of each hydraulic resistor pinch 30 increases in the direction of drug inlet port 12 to outlet port 14.
In some embodiments of microfluidic device 50 according to the present disclosure, each of the microchambers in the plurality of dead-end first microchamber sets is individually fluidically connected to the drug inlet main channel by a side channel.
In some embodiments of microfluidic device 50 according to the present disclosure, a pair of ladder channels 26 in sequence between drug inlet main channel 16 and negative inlet main channel 24 comprises a subunit of microfluidic circuit 10. In some embodiments, each subunit 34 comprises only one of dead-end first microchamber sets 20. In other embodiments, at least one of subunits 34 comprises a plurality of dead-end second microchamber sets 36. In some embodiments, each of dead-end second microchambers 46 of dead-end second microchamber sets 36 is individually fluidically connected to negative inlet main 24 channel by side channel 38. In some embodiments, plurality of dead-end second microchamber sets 36 is comprised of three distinct microchambers 46.
In some embodiments of microfluidic device 50 according to the present disclosure, plurality of dead-end first microchamber sets 20 is comprised of three distinct microchambers 44.
In some embodiments of microfluidic device 50 according to the present disclosure, outlet channel 28 is positioned parallel to drug inlet main channel 16.
Outlet channel 28 may be fluidically connected to drug inlet main channel 16 by one or more outlet side channels 40. In some embodiments, only a single outlet side channel 40 is positioned in drug inlet main channel 16 between proximate pairs of the ladder channels.
In some embodiments of microfluidic device 50 according to the present disclosure, outlet channel 28 comprises a plurality of serpentine mixers 42. In some embodiments, only a single serpentine mixer 42 is positioned in outlet channel 28 between proximate pairs of ladder channels 26.
A variety of materials may be used to form the solid support and/or a substrate, which itself is formed on or connected to the solid support. The support and/or substrate materials should be selected based upon their compatibility with known microfabrication techniques, for example, photolithography, 3-D printing, wet chemical etching, laser ablation, air abrasion techniques, injection molding, embossing, and other techniques, or based on the application being used. The support and/or substrate materials are also generally selected for their compatibility with the full range of conditions to which the microfluidic devices may be exposed, including extremes of pH, temperature, salt concentration, and/or application of electric fields, should these be relevant in performing assays using the microfluidic device of the present disclosure.
In some embodiments, suitable support and/or substrate materials include, without limitation, glass, pyrex, glass ceramic, polymer materials, semiconductor materials, and combinations thereof. In some embodiments, the support and/or substrate material may include materials normally associated with the semiconductor industry in which microfabrication techniques are regularly employed, including, e.g., silica based substrates such as glass, quartz, silicon, or polysilicon, as well as other substrate materials, such as gallium arsenide and the like. In some embodiments, the support layer comprises glass.
Exemplary polymeric materials include, without limitation, plastics such as polymethylmethacrylate (PMMA), polycarbonate, polytetrafluoroethylene (TEFLON™), polyvinylchloride (PVC), polydimethylsiloxane (PDMS), and polysulfone. Other plastics can also be used. Such materials are readily manufactured from microfabricated masters, using well known molding techniques, such as injection molding, embossing or stamping, or by polymerizing the polymeric precursor material within a mold. Such polymeric substrate materials are known for their case of manufacture, low cost and disposability, as well as their general inertness to most extreme reaction conditions. These polymeric materials may include treated surfaces, for example, derivatized or coated surfaces, to enhance their utility in the microfluidic device or, for example, to provide enhanced fluid direction should that be a needed factor in carrying out assays using microfluidic devices described herein. In some embodiments, the substrate layer comprises PDMS.
As described in more detail supra, there is a need to develop new platforms for rapid, inexpensive, and easily implemented antibiotic resistance testing. Accordingly, a further aspect of the present disclosure relates to a method for performing an assay. This method involves loading a test solution into a microfluidic device of the present disclosure. A culture medium containing an antibiotic solution is loaded into the drug inlet port, a culture medium is loaded into the negative medium inlet port, and the device is then washed with oil to isolate each of the plurality of microchambers. The method further involves detecting a fluorescent signal in each of the plurality of microchambers.
In some embodiments, the test solution comprises a metabolic dye. Suitable metabolic dyes include, without limitation, resazurin.
In some embodiments, the test solution comprises a biological sample. The biological sample may be in solution, suspension, or solid form. In some embodiments, the test solution comprises a bacterial suspension.
Suitable biological samples include, without limitation, blood, urine, saliva, sputum, and cerebral spinal fluid.
The test solution and/or biological sample may comprise one or more bacterial species. In some embodiments, the test solution and/or biological sample comprises a single bacterial species.
The bacterial species may be a pathogenic bacterial species. Suitable exemplary pathogenic bacterium include, without limitation, Escherichia coli, Proteus mirabilis, Enterococcus faecalis, Staphylococcus pseudintermedius, Salmonella, Staphylococcus aureus, Streptococcus pyogenes, Mycobacterium tuberculosis, Clostridium botulinum, Clostridium difficile, Neisseria meningitidis, Neisseria gonorrhoeae, Helicobacter pylori, Vibrio cholerae, Bacillus anthracis, Yersinia pestis, Listeria monocytogenes, Legionella pneumophila, Shigella, Campylobacter jejuni, Haemophilus influenzae, Bordetella pertussis, Brucella, Francisella tularensis, Rickettsia rickettsii, Chlamydia trachomatis, Treponema pallidum, Borrelia burgdorferi, Vibrio vulnificus, Acinetobacter baumannii, Pseudomonas aeruginosa, and Klebsiella pneumoniae.
In some embodiments, the test solution and/or biological sample contains an unknown bacterial species.
In some embodiments, the test solution comprises at least 104 CFU/mL, 105 CFU/mL, or 106 CFU/mL. For example, the test solution may comprise at least 105 CFU/mL of a bacterial species.
Loading the test solution into the microfluidic device may be carried out by loading the test solution through the drug inlet port and/or the negative inlet port. In some embodiments, loading the test solution into the microfluidic device is carried out through the drug inlet port while the remaining openings (i.e., the negative inlet port and the outlet port) are blocked. Alternatively, loading the test solution into the microfluidic device is carried out through the negative inlet port while the remaining openings (i.e., the drug inlet port and the outlet port) are blocked; or through the outlet port while the remaining openings (i.e., the negative inlet port and the drug inlet port) are blocked. Loading the test solution is carried out to completely fill the device and load the microchambers.
Loading the culture medium comprising the antibiotic solution is carried out through the drug inlet port. In some embodiments, the culture medium comprising the antibiotic solution is carried out at a flow rate of 1-1000 μL/hour, or any amount therebetween. In some embodiments, the culture medium comprising the antibiotic solution is carried out at, e.g., 35 μL/hour.
Loading the culture medium is carried out through the negative medium inlet port at a flow rate of 1-1000 μL/hour, or any amount therebetween. In some embodiments, the culture medium comprising the antibiotic solution is carried out at, e.g., 262 μL/hour.
The culture medium comprising the antibiotic solution and the culture medium may be loaded simultaneously or sequentially. In some embodiments, the culture medium comprising the antibiotic solution and the culture medium are loaded simultaneously.
Loading the culture medium comprising the antibiotic solution and the culture medium may be carried out for 1-10 minutes or more. In some embodiments, the culture medium comprising the antibiotic solution and the culture medium are loaded for about 1 minute, about 2 minutes, about 3 minutes, about 4 minutes, about 5 minutes, about 6 minutes, about 7 minutes, about 8 minutes, or about 10 minutes.
Loading the culture medium comprising the antibiotic solution through the drug inlet port and loading the culture medium through the negative inlet port as described in the present disclosure enables the generation of serial 2-fold concentration gradients in the dead-end first microchamber sets of the microfluidic device of the present disclosure.
Suitable antibiotics for use in the method according to the present disclosure include, without limitation, penicillin (e.g., amoxicillin, ampicillin, and penicillin V), cephalosporin (e.g., cefalexin, cefuroxime, ceftriaxone, ceftiofur), macrolides (e.g., azithromycin, clarithromycin, and crythromycin), fluoroquinolones (e.g., ciprofloxacin, enrofloxacin, levofloxacin, and moxifloxacin), tetracyclines (e.g., doxycycline and tetracycline), aminoglycosides (e.g., gentamicin and tobramycin), and sulfonamides (e.g., sulfamethoxazole/trimethoprim). In some embodiments, the antibiotic solution comprises a single antibiotic. In other embodiments, the antibiotic solution comprises a combination of at least two antibiotics, at least three antibiotics, or at least four antibiotics. For example, the antibiotic solution may comprise Amoxicillin and Clavulanic acid or Trimethoprim and Sulfamethoxazole.
Washing the device with oil may be carried out through the drug inlet port and/or negative inlet port.
In some embodiments of the methods of the present disclosure, the method further involves incubating the microfluidic device. The microfluidic device may be incubated at, e.g., 37° C. Said incubating may be carried out for about 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 11 hours, 12 hours, 13 hours, 14 hours, 15 hours, 16 hours, 17 hours, 18 hours, 19 hours, 20 hours, 21 hours, 22 hours, 23 hours, 24 hours, or more. In some embodiments, the microfluidic device is incubated for about 4 hours to about 5 hours.
In some embodiments, detecting a fluorescent signal in each of the plurality of microchambers is carried out at least 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 11 hours, 12 hours, 13 hours, 14 hours, 15 hours, 16 hours, 17 hours, 18 hours, 19 hours, 20 hours, 21 hours, 22 hours, 23 hours, 24 hours, or more after said washing step. For example, a fluorescent signal in each of the plurality of microchambers may be carried out after about 4 to 5 hours following the washing step.
In operation, the test solution is loaded into the microfluidic device using a syringe. With reference to
In some embodiments, the method further involves calculating a minimum inhibitory concentration based on said detecting.
Bacterial isolates (30) from canine urinary tract infections were obtained from the Cornell Animal Health Diagnostic Center (AHDC). The isolates were identified to species level, with antibiotic susceptibilities examined using a Sensititre™ system (SVU) with a urinary plate for isolates of veterinary origins. The 30 isolates include Escherichia coli (n=14), Proteus mirabilis (n=8), Staphylococcus pseudintermedius (n=5), Enterococcus faecalis (n=3), and were selected to reflect prevalence of these organisms as observed at Cornell AHDC in canine urine. Isolates were subcultured onto fresh blood agar plate prior to testing and kept at refrigeration temperature for no longer than 5 days.
For direct-from-sample testing, de-identified discarded canine urine from AHDC was obtained. Samples were refrigerated overnight until confirmed for the presence of bacterial growth by the diagnostic center. Among the samples, four were growth negative and seven were growth positive. The inclusion criteria for growth positive samples were single organism growth with a count of at least 105 CFU mL−1. With the exception of swabs (as this method does not yield fluid samples), no sample collection methods were excluded. The antibiotic susceptibilities of these pathogens bacteria were examined by the AHDC using the Sensititre™ system (SVU) and the microfluidic system according to the present disclosure, and the results were compared.
Seven different antibiotics and antibiotic combinations were tested, following those included in the SVU plate. They were Enrofloxacin, Ceftiofur, Tetracycline, Cefalexin, Ampicillin, and two antibiotic combinations Amoxicillin/Clavulanic acid 2:1 ratio, and Trimethoprim/Sulfamethoxazole 8:152 ratio. Antibiotic solutions were prepared from powdered stocks then aliquoted and stored at −20° C.
The microfluidic device was made from polydimethylsiloxane (PDMS) (Silgard 184; Dow Corning) patterned with standard soft lithography technique and permanently bonded with a glass substrate (Xia & Whitesides, “Soft Lithography,” Angew. Chem. Int. Ed. 37:550-575 (1998), which is hereby incorporated by reference in its entirety). SU-8 2075 was used to make the silicon mold. PDMS was mixed at 1:10 ratio of curing agent to polymer and cured at 70° C. for 2 hours. The PDMS and glass substrate was treated with plasma for 1 minute to make a permanent bond, and then the device was incubated at 70° C. for at least another 30 minutes to promote the bonding.
Resazurin was used to characterize the diffusion of molecules into the microchambers. Fluorescent images were taken using a fluorescent microscope (ZOE™ Fluorescent Cell Imager system; Bio-rad, Hercules, California) every 30 seconds up to 5 minutes, then analyzed using ImageJ. To measure the concentration gradient experimentally, resazurin solutions were loaded for a resazurin-specific loading time, then the main channel was washed with mineral oil to isolate the microchambers. The fluorescent intensities of the resazurin in the microchambers were analyzed using ImageJ software. Because the concentration profile on the device expands across more than two orders of magnitude (100%-0.3%), three resazurin solutions (i.e., 10 μg mL−1, 1 μg mL−1, 0.1 μg mL−1) were used to characterize the concentration gradient in three zones: high (M1-3), medium (M3-6), and low (M6-9). At each microchamber, the relative fluorescent signal compared to the previous microchamber was determined, then the concentration at each was calculated. To examine the ability to generate the designed concentration gradient for different molecules, fluorescein and Calcein were loaded using their molecule-specific loading time, then calculated the concentration profiles generated.
Comsol multiphysics 5.4a software was used for Finite Element Analysis of the equivalent constriction length to the hydraulic resistance of the mixer. For the detailed characterization of the network, a generalized computational method was developed to calculate the flow rates and concentrations. The algorithm resembled the Hardy Cross method of momentum distribution through pipe networks (Cross, H., “Analysis of Flow in Networks of Conduits or Conductors,” https://www.ideals.ilinois.edu/items/4876 (1936), which is hereby incorporated by reference in its entirety). It was unrealistic to hand-calculate the flow rates and concentrations for all 65 nodes on the chip.
All nodes, including both ends of the features in the chip and the channel junctions, are assigned a unique number between one and the total number of the nodes (n). An Excel sheet should be generated with the first column starting node (i) and second column ending node (j) for each feature. Each row contains width and length of the feature and in the starting node (i) and ending node (j) the numbers should be placed in increasing order. The subroutine then reads the excel sheet and creates a resistor matrix (R) in which the row index would be the number of the starting node and the column index is the ending node and the value is the resistance based on the following formula.
L, W and H are the length, width, and height of the feature respectively.
A second sheet in the Excel file was named “PQC” that contains the boundary conditions of pressure, flow rate and concentration (Table 1).
Using these two sheets, a matrix will be created containing C(n,2)+n rows and columns, in which C(n,2) is the binomial coefficient of 2 from n for each combination of (i,j) from 1 to n. This is the coefficient matrix for the flow rates between each two nodes plus the pressure at each node which are solved for. The first C(n,2) elements are related to flow rates between each two nodes (zero if there is no connection between the nodes) and the remaining n would be the pressures. The following equation regarding the balance of pressure for the connection between i and j can be written:
for all the combinations of i and j or if nothing is known then
Pi−Pj−Ri,jqi,j=0.
If there is an external flow source attached to a node, it can be assumed that there is a ghost node outside of the network that is connected to the node via a resistor with resistance of 1. Therefore, the right-hand side of the above equation becomes
in which qe is the external flow source amount. Another set of n equations is the mass balance at each node.
Σk=1nqk=0 because all incoming flows rates to a node are positive and the outgoing flow rates are negative such that at each node the flow rate sums to zero. With these C(n,2)+n equations the set would be closed and can be solved for pressures and flow rates. Once these are calculated, assuming that all the incoming flows mix thoroughly, concentrations at each node can be solved for.
In which Σjq>0qj,k means summation over all the neighbor nodes that have incoming flow rates into the node; that is qj,k>0. To understand how the indexes are managed refer to the code itself.
The device loading procedure is similar to protocols in previous studies (Nguyen et al., “Diffusion-Convection Hybrid Microfluidic Platform for Rapid Antibiotic Susceptibility Testing,” Anal. Chem. 93:5789-5796 (2021); Azizi et al., “Biological Small-Molecule Assays Using Gradient-Based Microfluidics,” Biosens. Bioelectron. 178:113038 (2021); and Azizi et al., “Antimicrobial Susceptibility Testing in a Rapid Single Test via an Egg-Like Multivolume Microchamber-Based Microfluidic Platform,” ACS Appl. Mater. Interfaces 13:19581-19592 (2021), which are hereby incorporated by reference in their entirety). In brief, bacterial inoculum was loaded into an empty device from one of the openings (e.g., the C0 inlet), while the remaining openings were blocked, to remove air bubbles and to completely fill the device specifically in the microchambers. Then, C0 solution containing a maximum concentration of an antibiotic, and culture medium alone were loaded into the device from two different inlets at flow rates 35 μL h−1 and 262 μL h−1, respectively (flow rate ratio of ˜7.5). The two solutions were loaded for a specific time depending on the antibiotic, to make the designed concentration gradient which diffuses into the microchambers. For most of the antibiotics tested, the loading time was under 6 minutes, however, the loading time for the Amoxicillin/Clavulanic acid and the Trimethoprim/Sulfamethoxazole antibiotic combinations were set at 10 minutes. A longer loading time for antibiotic combinations ensures that both antibiotics have saturated the microchambers, as each has a different diffusion rate. Finally mineral oil was loaded from the inlet to wash out the solutions, create an oil/water cap on the microchambers, and prevent contamination. Prior to incubation, water was added to the surrounding water bath through one of its openings, which helps to reduce evaporation of the reaction inside the microchambers by water vapor exchanging through the porous PDMS walls. All AST testing using the presently disclosed microfluidic system follows the following procedures. First, the bacterial suspension was prepared and calibrated to a McFarland 0.5 standard using a spectrophotometer, which correlates to a concentration of 1.5×108 CFU mL−1. For the study using bacterial isolates, the suspension was prepared by resuspending 2-3 colonies in sterile PBS. Then, 30 μL of the bacterial suspension was added to 1 mL of culture medium (final inoculum concentration ˜3×106 CFU mL−1). In all AST experiments, Mueller-Hinton broth supplemented with 2 v/v % PrestoBlue Cell Viability Reagent was used as a culture medium. Each C0 antibiotic solution was prepared by diluting frozen stocks in culture medium to a concentration 2 times that of the maximum concentration on the SVU plate, except for Amoxicillin/Clavulanic acid and Trimethoprim/Sulfamethoxazole antibiotic combinations were diluted to the same as the maximum concentrations on the plate. One device was used for each antibiotic/bacteria combination, nine 2-fold diluted antibiotic concentrations were examined, and one positive control where bacteria were not exposed to the antibiotic was included. The bacteria was loaded, the antibiotic concentration gradient was generated, and the oil was loaded as described supra. Extra oil was added to the top of the inlet/outlet to make an oil droplet and help prevent air from entering the system and disrupting the oil/culture medium interface. The devices were placed into a petri dish with enough water to cover the glass substrate and incubated at 37° C. in an incubator. After 4 hours, the red fluorescent intensities of the microchambers on each device were examined to determine the MIC using a fluorescent microscope (ZOE™ Fluorescent Cell Imager system; Bio-rad, Hercules, California). For slow growth samples where the MIC was not clear after 4 hours incubation, samples were incubated for an additional 30 minutes to 1 hour.
Spiked urine samples were prepared by spiking overnight MHB-culture of bacteria into pooled negative growth urines at 106 CFU mL−1. The spiked samples were refrigerated for approximately the same duration that the clinical urine samples would be kept prior to AST testing in our laboratory. Seven of the 30 bacterial isolates were randomly chosen for the spiking study. The bacteria from both spiked and positive growth urine samples were retrieved by a two-step centrifugation/filtration method. First, 2 mL samples were centrifuged at 13,000 g for 1 minute, and the pellet was resuspended in 1 mL fresh MHB. Then, the solution was filtered through a 5 μm filter to remove debris and non-bacterial cells. Afterward, the bacterial solution was incubated at 37° C. with shaking at 120 rpm for 2 hours. Subsequent inoculum preparation and AST were performed as described.
JMP version 16 was used to fit a generalized linear model to assess the probability of the two methods matching and how that depended on the drug and the bacteria tested. A binomial distribution with a logit link was used, and effects of antibiotic, bacteria, and their interaction were fixed. Because certain drug-bacteria combinations had a 100% match rate, Firth Bias-Adjusted Estimates was used. Other experiments on fluorescent dyes kinetic and concentration gradients were conducted with independently prepared replicates. Data are presented as means and standard deviations.
The ladder microfluidic system for antibiotic susceptibility testing (AST) combines a method of nanoliter microchamber-based AST (Nguyen et al., “Diffusion-Convection Hybrid Microfluidic Platform for Rapid Antibiotic Susceptibility Testing,” Anal. Chem. 93:5789-5796 (2021) and Azizi et al., “Nanoliter-Sized Microchamber/Microarray Microfluidic Platform for Antibiotic Susceptibility Testing,” Anal. Chem. 90:14137-14144 (2018), which are hereby incorporated by reference in their entirety) with a ladder-shape concentration gradient generator (LCGG). This combination provides a standardized and tunable antibiotic concentration profile for rapid phenotypic AST (
The device has three openings: a drug inlet (culture medium containing antibiotic), a negative inlet (pure culture medium), and an outlet (detailed dimensions in
A critical engineering challenge of the device was matching the flow rate of the diluter (pure culture media) to the incoming antibiotic stream and to create the 2-fold dilution at each loop. Without any control feature, the diluter would reach only half of the loops as the path of least resistance is a close-by outlet where the fluid can flow out of the system. This was solved by increasing the hydraulic resistances on the side channels, thus driving the diluter to reach all the loops. The approximate value of the resistance which results in desired concentration profile in the designed ladder layout can be calculated using circuit logic modeling Nguyen et al., “Diffusion-Convection Hybrid Microfluidic Platform for Rapid Antibiotic Susceptibility Testing,” Anal. Chem. 93:5789-5796 (2021) and Oh et al., “Design of Pressure-Driven Microfluidic Networks Using Electric Circuit Analogy,” Lab. Chip 12:515-545 (2012), which are hereby incorporated by reference in their entirety). Initially the resistance of the main channels was ignored due to their larger dimensions with respect to the rest of the chip features. A more detailed model was later used to evaluate the performance of the design, which included the resistance of the main channels.
At the side channels the flow rates were designed to be a half that of the drug inlet flow rate. For simplicity flow rates are referred to as their ratio to the drug inlet. At each loop, a relative flow rate, 0.5 of the drug, is mixed with a flow rate of 0.5 of the diluent coming from the top main channel. After this mixture passes through the mixer it flows into a series of 3 microchambers which are attached to the main channel. Right after these microchambers, 0.5 of this flow rate will be removed through a small added section.
To find the resistance and flow within the mixer RM (
where R is the resistance, M is the mixer, q is the flow rate, S is the side channels and i is the number of the loop. Then for each of the loops:
where qM=1; qd
such that Rd
Once the required resistances are determined, the length and width of each side channel is calculated to provide the desired resistance. A constriction (
Computational fluid dynamics (CFD) simulations were used to determine the hydraulic resistance of the constriction as a function of its length (
After the network layout and the dimensions of all resistors were determined, the concentration profile resulting from the network was examined by considering all the features in the chip. Since there are multiple features in the chip, a generic Matlab code that takes the pressure nodes (the black dots in
The ladder microfluidic system is an improvement over previous designs (Nguyen et al., “Diffusion-Convection Hybrid Microfluidic Platform for Rapid Antibiotic Susceptibility Testing,” Anal. Chem. 93:5789-5796 (2021), which is hereby incorporated by reference in its entirety) as it allows for the generation of a broader range of concentrations with dilutions over two orders of magnitude. Further, the standardized concentration profile (2-fold dilution) ensures relevance and translatability of the results when interpreted with standard breakpoints set by national and international organizations such as the Clinical and Laboratory Standards Institute or the FDA.
The operation of the ladder microfluidic system follows a previously established protocol (Nguyen et al., “Diffusion-Convection Hybrid Microfluidic Platform for Rapid Antibiotic Susceptibility Testing,” Anal. Chem. 93:5789-5796 (2021); Azizi et al., “Nanoliter-Sized Microchamber/Microarray Microfluidic Platform for Antibiotic Susceptibility Testing,” Anal. Chem. 90:14137-14144 (2018); Azizi et al., “Gradient-Based Microfluidic Platform for One Single Rapid Antimicrobial Susceptibility Testing,” ACS Sens 6:1560-1571 (2021); Azizi et al., “Biological Small-Molecule Assays Using Gradient-Based Microfluidics,” Biosens. Bioelectron. 178:113038 (2021); and Azizi et al., “Antimicrobial Susceptibility Testing in a Rapid Single Test via an Egg-Like Multivolume Microchamber-Based Microfluidic Platform,” ACS Appl. Mater. Interfaces 13:19581-19592 (2021), each of which is hereby incorporated by reference in its entirety). The standardized cation-adjusted Mueller Hinton broth was used to prepare all the solutions, with 2 v/v % PrestoBlue added as a cell metabolism indicator. Loading of the platform consists of the following three steps: (1) a bacterial suspension was loaded into the device using a syringe (
Four bacterial species representing pathogens relevant to canine UTI were selected: Escherichia coli (EC; n=14), Proteus mirabilis (PRM; n=8), Staphylococcus pseudintermedius (SP; n=5), Enterococcus faecalis (EF; n=3). Suspensions of these bacteria were prepared to be used in the platform at ˜3×106 CFU mL−1, which is slightly higher than the standard method which calls for concentrations around 5×105 to 1×106 CFU mL−1. It was found that this concentration improved sample dispersions in all the microchambers without changing the resulting MICs. The result interpretation protocol was followed as previously established based on the difference in fluorescent intensities of resazurin as a cell metabolic indicator (Nguyen et al., “Diffusion-Convection Hybrid Microfluidic Platform for Rapid Antibiotic Susceptibility Testing,” Anal. Chem. 93:5789-5796 (2021); Azizi et al., “Nanoliter-Sized Microchamber/Microarray Microfluidic Platform for Antibiotic Susceptibility Testing,” Anal. Chem. 90:14137-14144 (2018); Azizi et al., “Gradient-Based Microfluidic Platform for One Single Rapid Antimicrobial Susceptibility Testing,” ACS Sens 6:1560-1571 (2021); Azizi et al., “Biological Small-Molecule Assays Using Gradient-Based Microfluidics,” Biosens. Bioelectron. 178:113038 (2021); and Azizi et al., “Antimicrobial Susceptibility Testing in a Rapid Single Test via an Egg-Like Multivolume Microchamber-Based Microfluidic Platform,” ACS Appl. Mater. Interfaces 13:19581-19592 (2021), each of which is hereby incorporated by reference in its entirety). It was hypothesized that there are three patterns of bacterial growth in the ladder microfluidic system: finding the minimum inhibitory concentration (MIC) (
During the incubation period, it was we observed that the background fluorescence of the resazurin increased, however, the fluorescent intensity indicating bacterial metabolism remained distinguishable. Due to differences in growth rate and doubling times between the four species, the signal intensities were different among the Gram-negative and Gram-positive samples. More specifically, Gram-negative bacteria EC and PRM doubled much faster than the other Gram-positive bacteria and therefore had higher overall fluorescent intensities. However, it was observed that after the 4-5 hour incubation period, the changes in fluorescent intensities were sufficient to determine the MICs for all species. For most of the samples, especially the EC and PRM samples, results could be read after 4 hours and as early as 3.5 hours. For some of the Gram-positive samples, however, the signal took longer to develop due to slow growth rate, and was given an additional 30 minutes to 1 hour of incubation time. Prolonged incubation, up to 5 hours, of the Gram-negative bacteria did not change the MICs. Among the three steps, bacterial loading, antibiotic loading and microchamber sealing, only the loading time in step 2 (
The performance characteristics of the ladder microfluidic system in determining MICs were examined with bacterial isolates from canine urinary tract infection samples which were submitted to the Cornell Animal Health Diagnostic Center (AHDC) for AST (
For the purposes of result comparison, the range of concentration and the combination ratios for these antibiotics follow those on the SVU plate. To compare the MICs obtained by AST using the ladder microfluidic system and using the SVU, FDA guidance for characterizing the performance of new AST systems was followed and the data was interpreted based on the number of agreements. First, the on-chip MIC and the MIC obtained from the SVU plate as carried out by the AHDC was paired for each bacteria/antibiotic combination. An agreement designation (match) was assigned if the on-chip MIC was within ±one 2-fold dilution. For many instances, the MIC determined by the SVU was in the form of a range, for example, many of the MICs for AST with Trimethoprim/Sulfamethoxazole were ≤2 μg mL−1. For these pairs, a match was confirmed if the on-chip MIC fulfilled the condition of the SVU plate MIC (for example, on-chip MIC=0.5 μg/mL was a match for SVU MIC ≤2 μg mL−1). There were four occasions where data from the SVU (three for Cephalexin and one for Ceftiofur) was not available, so these were excluded from the statistical evaluation.
In total, 206 bacterial samples were evaluated for MIC on both the ladder microfluidic system and the gold standard SVU plates. An overall 91.75% accuracy of determining the MIC between the two methods was observed; 189 samples matched between the two platforms. Further, the matching rate for each individual bacteria strain was calculated as the number of samples that matched (m) divided by the total number of bacteria samples of that type (n). Among the four bacteria, EC and PRM had the highest matching rate at 92.71% ( 89/96) and 94.54% ( 52/55), respectively, and SP and EF had the lowest matching rate at 88.57% ( 31/35) and 85.00% ( 17/20), respectively.
The data was also examined from the perspective of individual bacterial species, antibiotics, and their combinations (
To further understand the potential sources of errors leading to disagreements in some of the bacteria/antibiotic groups, those samples which had more than three disagreements were repeated. While improvement in accuracy with some samples was observed after repeating, suggesting random handling error in the first experiment, most of them had the same MICs. The same cannot be said for the original SVU data obtained from AHDC, which was conducted once. This observation indicates that the exemplified method is reliable and repeatable, and the mismatch between the MICs could be due to errors from the SVU as run by the AHDC.
While the time-to-result of the AST of clinical isolates was reduced, whether the overall time required for obtaining a susceptibility result can further be reduced for some urine samples, up to 24 hours, by bypassing the bacteria isolation step was next investigated (Baltekin et al., “Antibiotic Susceptibility Testing in less than 30 min Using Direct Single-Cell Imaging,” Proc. Natl. Acad. Sci. USA 114:9170-9175 (2017); Li et al., “Adaptable Microfluidic System for Single-Cell Pathogen Classification and Antimicrobial Susceptibility Testing,” Proc. Natl. Acad. Sci. USA 116, 10270-10279 (2019); Avesar et al., “Nanoliter Cell Culture Array with Tunable Chemical Gradients,” Anal. Chem. 90:7480-7488 (2018); Mo et al., “Rapid Antimicrobial Susceptibility Testing of Patient Urine Samples Using Large Volume Free-Solution Light Scattering Microscopy,” Anal. Chem. 91:10164-10171 (2019); Sabhachandani et al., “Integrated Microfluidic Platform for Rapid Antimicrobial Susceptibility Testing and Bacterial Growth Analysis Using Bead-Based Biosensor via Fluorescence Imaging,” Microchim. Acta 184:4619-4628 (2017); and Dong & Zhao, “Rapid Identification and Susceptibility Testing of Uropathogenic Microbes via Immunosorbent ATP-Bioluminescence Assay on a Microfluidic Simulator for Antibiotic Therapy,” Anal. Chem. 87:2410-2418 (2015), which are hereby incorporated by reference in their entirety). Thus, the ability of the presently claimed device to conduct AST directly with urine samples was evaluated. The study was conducted with seven urine samples that were cultured positive for the presence of a single unknown bacteria. Because these are canine urine samples, the volumes were limited, ranging from 2-3 mL. Each urine sample (2 mL) was first centrifuged at 13,000 g for 1 minute, and the resulting pellet resuspended in 1 mL of fresh MHB. Afterward, the suspension was filtered through a 5 μm syringe filter to remove any debris or eukaryotic cells. It was found that this sequence of bacterial collection was the most consistently effective for the limited volume of samples. The first step collects and concentrates all the bacteria and cells or debris inside the sample, while the second step isolates the targeted bacteria. This centrifuge-filter sequence leads to a decrease in bacteria loss as the larger pellet from the first spinning step helps in recovering bacterial cells. It was found that samples containing >105 CFU mL−1 gave consistent results, and thus excluded the diagnosis low-end of 104−105 CFU mL−1. This limit is considered as the limit of detection of the method when testing directly from urine, however, most of the samples received at the facility are above 105 CFU mL−1. Finally, the isolated bacteria were preincubated for 2 hours to bring the cells out of lag phase before conducting AST. Similar to the original method with bacterial isolates, the MICs of these bacteria from clinical samples was determined in 4-5 hours. It is believed that the preincubation period is necessary when testing bacteria directly isolated from stored patient urine samples, such as the samples in this study, which were refrigerated around 24-36 hours before they were received for testing. This period appears to bring the bacteria out of the lag phase induced by refrigeration.
Some culture negative samples were tested and the lack of false positives was confirmed by observing no growth-indicative changes in the fluorescent intensity in the positive growth chambers which did not get exposed to antibiotics. Additionally, seven clinical isolates from the previously analyzed samples were spiked into pooled canine culture negative urine, and the same isolation/pre-enrichment process was carried out before performing AST. The same overall rate of agreement between the clinical and spiked samples was observed (91.84%) (
When selecting the culture positive urine samples, any sampling method that would generate a liquid volume sample were not excluded (swab samples were excluded), or type of collection tube, some of which contain additives to stabilize bacterial count. Comparing the results from all the urine samples suggests that the sample collection method does not have a effect on the performance of the method in determining MIC. These results collectively support direct-from-sample AST using the ladder microfluidic system.
Examples 1-5 describe the design, optimization, and use of a microfluidic system according to the present disclosure that performs phenotypic AST; incorporates standardized, translatable testing parameters; and tests both isolated bacteria from culture plates or bacteria isolated directly from urine samples. The system utilizes a ladder shaped network to generate a serially 2-fold diluted concentration gradient that follows current standardized methods and regulations and is tunable to accommodate a broad number of clinically relevant antibiotics and antibiotic combinations. The total preparation time, which includes bacteria loading, antibiotic loading, and oil loading, can be as fast as 10 minutes when testing with a single antibiotic and no >15 minutes for an antibiotic combination. An important consideration of the ladder microfluidic system is the two-fold serial dilution of the antibiotics on the microscale and within a small space. Although exponential concentration gradients have been generated in microscale, by adapting traditional gradient generators such as the Christmas tree (Chen et al., “A Review on Microfluidics Manipulation of the Extracellular Chemical Microenvironment and its Emerging Application to Cell Analysis,” Anal. Chim. Acta 1125:94-113 (2020), which is hereby incorporated by reference in its entirety), or H-filter (Selimović et al., “Generating Nonlinear Concentration Gradients in Microfluidic Devices for Cell Studies,” Anal. Chem. 83:2020-2028 (2011), which is hereby incorporated by reference in its entirety), these designs tend to be larger in size. Controlling the fluidic flow in a ladder-shape grid results in a simpler and smaller device, which is also easier to handle. This ladder microfluidic system is the first, to date, to use circuit logic to generate a concentration gradient in microfluidics.
Rapid AST can be performed on both bacterial isolates which is the standard type of sample for AST, or on clinical urine samples that contain the disease-causing pathogen. Compared with standard AST which requires 18-24 hours, the time-to-result using this microfluidic platform is shorter; results can be obtained within a single working day. The Examples of the present disclosure demonstrate that the presently disclosed method is >90% effective at determining the MIC of the most common UTI pathogens in canines against disease relevant antibiotics, when examined with 30 bacterial isolates from clinical samples. Specifically, the method was able to determine the MICs of the Gram-negative bacteria E. coli and P. mirabilis accurately up to 92.71% and 94.54%, respectively. The rate of accuracy for the other two Gram-positive bacteria, S. pseudintermedius and E. faecalis that we tested were slightly lower, at 88.57% and 85.00%, respectively, but no statistical significance was found. Further, the presently disclosed method is capable of testing bacteria recovered directly from clinical samples, eliminating the need to culture the pathogen for 12-48 hours. Because of the small dimensions at the microscale, the presently disclosed method does require a higher inoculum than the standardized method to minimize sampling effects. This requirement is not an issue when testing bacterial isolates plated on culture plates, but poses a challenge when testing clinical samples especially at the lower UTI diagnostic concentration 104-105 CFU mL−1. Bacteria losses during the recovery process is the biggest contributing factor to this challenge. Therefore, a better recovery method is needed, or testing of isolated and enriched bacteria would be more appropriate for these samples. The method is effective for urine with higher bacteria concentration, which was most of the UTI diagnosed samples.
In a preliminary experiment, bacterial loads from 104 to 108 CFU mL−1 were evaluated. A higher frequency of false negative growth results were observed in the 104-105 range, most likely due to sampling error because of the small volume of the microchambers. An increase in MIC with the 107-108 CFU L−1 sample was observed, because there were too many bacterial cells. Out of the 189 clinical MICs from isolates, 64 of them were discrete values (not a <= or > range). Among them, 19 had higher MIC, 28 had lower MIC, and 17 had the same MIC as the standard method. In total, 9 out of the 64 received the no-match designation. This trend was improved in the later experiments with the direct from urine samples where we had 39 discrete MICs, with 2 higher MICs, 11 lower MICs, and 23 exact MICs, with an overall 5 no-match designations. This improvement is most likely due to better familiarity with the operation of the platform and assay. Interestingly, there was an increase in the frequency of MICs that were lower than the standard method, which was not expected with a higher than standard bacterial load. However, most of these decreases in MIC are within the acceptable range for method comparison. Furthermore, the PDMS material of the disclosed device is known to absorb some small molecules, especially hydrophobic molecules (Regehr et al., “Biological Implications of Polydimethylsiloxane-Based Microfluidic Cell Culture,” Lab. Chip 9:2132-2139 (2009), which is hereby incorporated by reference in its entirety), however the data presented herein neither strongly supports, nor fully rejects, the effect of adsorption on the MICs.
The performance of the system was characterized on urine samples containing a single organism and samples with multiple organisms were excluded. This is a limitation for many phenotypic AST methods aiming to test directly from samples where the bacteria species is unknown. Depending on the sample collection process, urine samples can be contaminated with commensal bacteria, and therefore it is difficult to confidently identify the presence of the pathogen(s). Further, false classification of antibiotic resistance might occur if one or more of the contaminants is resistant to an antibiotic, resulting in erroneous exclusion of a potentially effective therapy. Conducting AST on multi-organism samples will require more complex result interpretation and may need the judgment of medical professionals for final diagnostic determinations. For these reasons, multi-organism samples were not included in the scope of this study.
The characterization studies focused on the four most common pathogens of UTIs at a veterinary diagnostic center, and the results from those bacteria suggests that the method could be extended to additional species of Gram-negative and Gram-positive UTI pathogens. The usability of the disclosed method to test bacterial isolates in place of the current SVU plate was demonstrated, as well as the feasibility to test bacteria retrieved directly from samples at a clinically relevant concentration. UTIs are among the most common bacterial infections dealt with at a clinical microbiology laboratory. While bacteria speciation is required for final susceptibility determination, the disclosed method can perform AST without prior information on the bacteria and interpreted after identification of bacteria is confirmed.
A rapid AST method would improve patient outcome and streamline clinical laboratory workflow by 18 hpurs if tested with identified bacterial isolates. Using the rapid urine test, AST can be conducted as soon as a sample is suspected to contain a UTI pathogen, while other identification methods such as MALDI-TOF are carried out simultaneously to identify bacteria, potentially reducing the time to result up to 2 days. Rapid AST methods can be used in place of, or in complement to, the SVU plate method. A Rapid AST test that could be read within a day would allow veterinarians and medical professionals to provide more targeted initial prescription during early diagnosis. This would reduce the use of broad-spectrum antibiotics and decrease the potential for antibiotic resistance. With improvement in testing time and adherence to industry standards, it is believed that this platform can be used at various stages in the clinical workflow, and thus provides another tool for combating antibiotic resistance without compromising patient health or increasing costs.
Although preferred embodiments have been depicted and described in detail herein, it will be apparent to those skilled in the relevant art that various modifications, additions, substitutions, and the like can be made without departing from the spirit of the invention and these are therefore considered to be within the scope of the invention as defined in the claims which follow.
This application claims the benefit of U.S. Provisional Patent Application Ser. No. 63/578,002, filed Aug. 22, 2023, which is hereby incorporated by reference in its entirety.
This invention was made with government support under 2019-38420-28975 awarded by the United States Department of Agriculture's National Institute of Food and Agriculture. The government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 63578002 | Aug 2023 | US |
