ANALYSIS INSTRUMENT
The invention provides multi-function instruments for automatically and simultaneously carrying out a variety of tests including identification of targets in specimens and antimicrobial susceptibility testing thereof. Application-specific cartridges are pre-loaded with all required reagents and allow for tests to be performed in a single testing device with no specimen preparation to, for example, rapidly detect infections, identify infectious pathogens, and analyze their susceptibility to various antimicrobial agents. Instruments include various stations for carrying out test steps that can be randomly accessed in any order dictated by the computer-controlled instrument. A carousel stores and incubates cartridges and, with a mechanical conveyor arm, transfers the cartridges between the required stations to carry out the tests. A plurality of different tests may be performed on a plurality of targets within an instrument, and the plurality of targets may be disposed in a single cartridge.
The disclosure relates to systems and methods for automatic testing of specimens.
BACKGROUNDInfectious disease are a leading cause of death and healthcare costs. Although antimicrobial therapy revolutionized treatment of infections and has saved countless lives, overuse of antimicrobials has accelerated the spread of antimicrobial resistance. To insure that these valuable drugs are used only when needed and that the right patients get the right antimicrobials at the right time, there is a need for more rapid and accurate diagnostic methods that are implementable across a wide variety of venues.
Hospital-acquired infections are a significant risk for patients, resulting in serious illness and high mortality and causing significant healthcare costs. A significant factor in the prevalence and seriousness of hospital-acquired infections is the rise of pathogens that are resistant to multiple antimicrobials. Effective treatment of infections requires rapid diagnostics to detect infection, identify the infectious pathogen, and then select appropriate antimicrobials for effective treatment of that pathogen.
Conventional methods are time consuming, requiring two to five days to detect the infection, identify and the pathogen, and determine the effective antimicrobial therapies. To protect the patient at the outset of infection and during the days required to determine the optimum therapy, doctors initially treat empirically by prescribing powerful, broad spectrum antimicrobials. These treatments can be suboptimal and even ineffective. The delay in implementing the appropriate narrow-spectrum therapy targeted to a patient's specific infection can have an enormous impact on morbidity and mortality. Furthermore, empirically treating in the absence of diagnostic results, or when using diagnostics prone to false positive results, leads to widespread administration of powerful antimicrobials to uninfected patients. The indiscriminate use of antimicrobials is a major contributing factor to the spread of antimicrobial resistant target cells. Finally, delay in implementing the optimal therapy incurs significant healthcare costs due increases hospital length of stay and expensive medical complications.
SUMMARYRapidly and accurately identifying patients with infections and rapidly implementing effective therapy to these patients can save lives and attenuate the spread of antimicrobial resistance. The present invention provides systems and methods that can accurately identify the patients that have infections in about 30 minutes and determine targeted therapy for a patient's infection in several hours compared to the days required by today's methods. Determining the effective antimicrobial therapy days earlier can dramatically improve medical outcomes including preventing death.
To detect infections, the invention can detect, quantify, and identify a broad range of pathogens including bacteria, fungi, viruses, and parasites. Also valuable for rapidly and accurately identifying patients with infections is the invention's ability to detect and quantify diagnostically informative toxins, disease-specific biomarkers, human or host cells, and host-response biomarkers. The invention can include any combination of the above capabilities in a single test to most effectively assess a patient specimen for the presence of an infection and to determine the infectious agent.
Diagnostically informative host cells include cells that indicate an inflammatory response to infection (for example, neutrophils), cells infected by pathogens (e.g., virally infected cells), or cells that indicate the quality and anatomical origin of the patient specimen (for example, squamous epithelial cells).
Examples of toxins diagnostic of life-threatening infections include Clostridiodes difficile Toxin B, the presence of which indicates C. difficile infection and Bacillus anthracis Lethal Toxin (or the toxin subunit Lethal Factor) which indicates anthrax infection. indicates disease. Host factors that can help identify infected patients include cytokines such as IL-4 and IL-6.
After detecting an infection and identifying and quantifying the infectious pathogen, methods of the invention can determine which patient therapies will be most effective. This type of analysis is called antimicrobial susceptibility testing. The invention differs from current methods for antimicrobial susceptibility testing, in that it can deliver accurate results directly from the patient specimen in a matter of hours rather than current conventional methods which take days. The conventional methods, unlike the inventive method, require time consuming microbiological culture steps to get millions of purified pathogen cells. The invention's novel antimicrobial susceptibility testing methods, in contrast, can rapidly determine effective therapies directly from patient specimens, without time consuming culture steps, because it does require large numbers of cells or cell purification.
The novel and potentially medically impactful capabilities and practicality of the invention are enabled by inventive systems and methods for enabling single molecule counting and single cell counting using non-magnified digital imaging of informative biological targets directly from patient specimens. Using simple and low-cost cameras without complex and expensive microscopes and optics to digitally count microscopic cells and sub-microscopic molecules allows detection of infections by rapid, sensitive, and automated quantification of disease-causing toxins and disease-specific biomarkers. The inventions systems and methods for identifying and digitally counting pathogen cells underlies the ability to rapidly determine susceptibility or resistance to antimicrobial agents. The inventive method determines if a pathogen is susceptible to an antimicrobial agent by determining if the agent stops the normal pathogen growth (that is, increase in cell number by cell division) when incubated in nutrient microbiological medium. This can be done by the invention by counting the pathogen cells before and after incubation in the medium containing the antimicrobial.
Systems and methods of the invention can automatically and simultaneously run a variety of tests requiring only a specimen input and providing actionable results in a variety of venues ranging from point-of-care to centralized hospital and reference laboratories. Such testing is made possible by a combination of application-specific cartridges pre-loaded with all required reagents, direct specimen input into cartridges and full automation of all processing and analysis to minimize hands-on time for users, a benchtop instrument designed for scalable throughput, instrument.
Steps of inventive methods for counting single molecules and single cells can include fluorescent labeling of the target molecules or cells, magnetically tagging the targets, using magnetic force to deposit the fluorescently labeled magnetic targets on an imaging surface of a device, imaging the targets without (or with minimal) magnification, and counting the targets using image analysis.
The invention allows for simultaneous processing of the steps outlined above in cartridge devices. A single random-access instrument can simultaneously process multiple test cartridges for different diagnostics applications containing different types of patient specimens. The automated nature of the inventive instruments and cartridges allow for operation by medical professionals without significant specialized training. Additionally, the breadth of the platform's potential test menu for the instrument offers the potential for reducing benchtop space allowing for more cost-saving utilization of facilities and enabling near-patient diagnostic testing to provide potentially life-saving diagnostic information to clinicians near the onset of infections when they can have the greatest impact.
Systems and methods of the invention use application-specific cartridges with pre-loaded test reagents. Preferably, cartridges are assembled and packaged with the required test reagents during manufacturing and distributed so that a user need only add a specimen to be tested (e.g., a respiratory specimen from a patient) and insert the cartridge into the instrument. In some instances, a specimen to be tested, such as a blood specimen, may be pre-enriched. For example, blood specimens may undergo pre-enrichment by culture before analysis because many blood infections cannot be tested directly without pre-enrichment due to having too low of a concentration of pathogen cells.
The instruments described herein use a variety of different stations for performing different test steps outlined above. The stations can be positioned around a carousel which is used to receive, store, and transfer application-specific cartridges between the stations according to the test being performed. Stations can include a fluidics station for interfacing with the cartridge and manipulating the specimen and reagents therein, magnetic selection station for magnetically depositing targets on the detection surfaces of a cartridge's imaging wells, imaging stations for detecting the deposited targets in specimens, and waste stations for disposing of used cartridges. In preferred embodiments, tests use a constant temperature throughout all steps or are modified such that the interior of the instrument can be maintained at the required temperature and the carousel can serve as a storage station for incubation steps.
The instrument can use the carousel to access the different stations so that test steps can be performed in the order and with the timing required for various types of tests. Precise computer scheduling and computer-controlled access to the various stations in the instrument are used to automatically carry out all steps of a variety of tests without additional user input. After loading a cartridge into the instrument, a user's next interaction can be receiving or viewing results of the test either at the instrument or remotely. Depending on the test type, the reported results of the platform's automatic analyses may indicate detection of infection; detection, identification, and quantification of pathogens, toxins, biomarkers, or diagnostically informative host cells; or antimicrobial susceptibility results and profiles. Some testing applications perform different kinds of measurements on a single specimen in the same cartridge on the same instrument run. In this case multiple types of results can be reported for the single test.
The instrument can read one or more barcodes or other identifiers on the cartridge for associating patient, test application-specific, or factory information with the cartridge. The instrument can also accept similar input entered by a user. The instrument can also use that input to record and track information associated with the specimen being tested including patient information for reporting results.
Instruments may include a computer comprising a processor and a non-transitory, tangible memory and operable to schedule and control the test being performed within the instrument and track the cartridges therein. The computer can include a user interface for prompting and receiving information from the user and displaying results and status information. The computer can be connected to a network and operable to process test results and send to connected devices over the network.
Instruments of the invention can include a mechanical conveyor arm for moving cartridges between the carousel and the various stations for the performance of required test steps. In preferred embodiments, the carousel and the stations comprise slots sized to accept and position the cartridge within the station. Rotation of the carousel can align the carousel slot with a corresponding slot in the relevant station and the mechanical conveyor arm may be operable to contact a side of the cartridge and slide the cartridge along the aligned slots and into the selected station. The mechanical conveyor arm avoids gripping the cartridges and reduces jams associated with gripping mechanisms. The mechanical conveyor arm can comprise two rotatable prongs operable to flank the cartridge and provide motivating force to one side thereof. The sides of the carousel and stations slots can provide the lateral guidance as the cartridge is slid, avoiding the need for a gripping mechanism for moving the cartridges.
The pre-loaded cartridges allow for control of reagent volumes and distribution on the manufacturing side and the automated instrument controls performance of the test steps and the timing thereof. Accordingly, systems and methods can greatly reduce the potential for user error allowing inexpert staff to conduct a variety of tests without specialized training and to obtain reliable and actionable results without the delay and cost of dedicated off-site testing.
In a preferred embodiment, application-specific cartridges include microbe-specific antimicrobial susceptibility testing cartridges for measuring differential growth of a pathogen in a specimen in the presence of various antimicrobial agents and microbiological growth medium that are selected based on the identity of the pathogen. According to the invention, patient specimens, such as urine, stool, or blood are directly analyzed with minimal or no specimen preparation or culturing. Specimens processed according to the invention are identified and exposed to various antimicrobials or other treatment modalities, allowing the selection of the most-effective treatment. Microbial infections can be identified and the appropriate treatment determined in a matter of hours, greatly reducing the delay in appropriately targeted therapy and avoiding the need for empiric treatment with aggressive broad spectrum antimicrobials. The invention allows health care providers to prescribe effective therapies at the outset to appropriately treat infected patients. Thus, the invention provides an opportunity to improve patient outcomes and reduce the spread of antimicrobial resistance.
Infection detection, target identification, and determination of effective treatment is accomplished directly from patient specimens, such as urine, sputum or other respiratory specimens, blood, stool, wound specimens, or cerebrospinal fluid with little or no specimen preparation steps. For example, a urine specimen is directly pipetted into a testing device for pathogen identification (ID) and antimicrobial susceptibility testing (AST) which is completed in several hours. This contrasts with current culture-based methods which require one or more days of colony purification to produce a large population of pure microbial culture for testing. The invention provides testing devices and instruments capable of receiving and internally processing a patient specimen to identify microbes or cells and/or to determine therapeutic susceptibility and efficacy all within the testing device. Multiple target cells or pathogens in a specimen can be identified and susceptibility to multiple antimicrobials or treatments can be tested in a single device. Testing systems and methods of the invention are robust with respect to specimen matrices, variable inoculum, and the presence of commensal microbes in the specimen. Tests of the invention also deliver accurate results for polymicrobial infections.
Systems and methods of the invention allow for direct processing and imaging of specimens to determine the presence and identity of target cells present in the specimen. As noted above, the processing and imaging steps occur with the specimen in a testing device such as a cartridge and require little to no specimen preparation outside of the testing device. By foregoing time-consuming specimen preparation techniques and using target-specific, distinguishable labels, systems and methods of the invention allow for identification and enumeration of targets in a specimen in as little as thirty minutes or less.
Systems and methods of invention can be used for identifying the pathogen that is causing an infection. For example, the inventive methods include a novel method for identifying and quantifying pathogens directly in a patient specimen without requiring culture-based microbiological pre-enrichment or nucleic acid amplification. The method enumerates the target pathogen(s) in a single reaction mixture by labeling using fluorescent in situ hybridization (FISH)-based method combined with magnetic selection that can be carried out in about 30 minutes in microtiter plates or cartridges in the instrument described herein.
Systems and methods of the invention can be used for diagnostic antimicrobial susceptibility testing (AST), that is, for determining which antimicrobials can prevent the growth of a microbial pathogen in a patient's specimen. This information provides information to clinicians about which antimicrobials should be used to effectively treat that particular patient's infection.
Antimicrobial susceptibility testing can be thought of as stepwise process. The goal is to determine which members of a panel of antimicrobial agents are effective for the particular pathogen strain that is causing a patient's infection. Typically, when an infection is detected, the species of pathogen is first identified. Identifying the species of the pathogen is useful for choosing the antimicrobials and dosing that can generally be used for treating that species. However, since the particular pathogen strain causing the infection may have become resistant to any of the antimicrobials, antimicrobial susceptibility testing must be done to determine to which of the potential treatments the pathogen is actually susceptible.
After species identification the pathogen cells from the patient's specimen are apportioned, or aliquoted, into a series of liquid solutions containing nutrient growth medium various antimicrobials at various concentrations. Then, the aliquots are allowed to incubate at a temperature conducive to microbial replication (generally 35-37° C.). If the pathogen is susceptible to the antimicrobial it can replicate normally, that is, the number of pathogen cells increase as they do in microbiological growth medium the absence of antimicrobials. If the pathogen is susceptible to the antimicrobial, it fails to replicate, replicates to a much lesser extent, or shows morphological or other abnormalities, indicative of effectiveness of the antimicrobial agent. Finally, the replication of pathogen cells is assessed in the various aliquots to determine which antimicrobial agents are effective. We refer to the set of a pathogen's antimicrobial susceptibility/resistance results for a for a series of antimicrobials as its antimicrobial susceptibility profile.
Both conventional methods and the inventive method for antimicrobial susceptibility testing follow the steps above, but the inventive method determines a pathogen's antimicrobial susceptibility profile in several hours while conventional methods require several days. The rapid antimicrobial susceptibility testing results using the inventive method arise from the new method's ability to test patient specimens directly without time-consuming culture-based pre-enrichment growth to achieve high concentrations of pure cells. This enrichment and purification is most commonly done using colony purification on petri dishes.
For conventional methods, the cells recovered after colony purification are first identified using biochemical, microbiological, nucleic acid methods, or Matrix-Assisted Laser Desorption/Ionization-Time Of Flight (MALDI-TOF) mass spectrometry (MS). Once the identity of the pathogen species is known, appropriate antimicrobials and concentrations can be chosen that are appropriate for determining the antimicrobial susceptibility profile for pathogens of that species.
Several novel aspects of the inventive systems and methods enable its ability to rapidly deliver antimicrobial susceptibility results directly from patient specimens.
Firstly, patient specimens generally contain orders of magnitude fewer cells than are required for traditional antimicrobial susceptibility testing. The inventive methods, in contrast to current culture-pre-enrichment dependent methods, can enumerate small numbers of pathogen cells by sensitive single cell counting using non-magnified digital imaging. Furthermore, because the method enumerates small numbers of individual cells, it can very quickly—in only a few bacterial generations—determine whether the cells have increased in number in an aliquot containing an antimicrobial and growth medium.
Secondly, patient specimens contain sample matrix and commensal microbes unrelated to the infectious pathogens. Guidelines for conventional methods (for example, from the Clinical Laboratories Standards Instituter or the European Committee on Antimicrobial Susceptibility Testing) require purified culture cells resulting from clonal growth of colonies on agar-based growth media in petri dishes. These cells contain only a single microbial species and no sample matrix.
As discussed above, the identity of the pathogen species must be known in order to interpret antimicrobial susceptibility testing results correctly for arriving at effective clinical treatment options. This is a key reason underlying why conventional and most emerging antimicrobial susceptibility testing methods require a pure culture of cells.
To determine the antimicrobial susceptibility profile, as described above, the conventional and most emerging methods assess the impact of different antimicrobials at different concentrations on the growth of the target pathogen. The reason why these methods require a pure population of identified cells to interpret the antimicrobial susceptibility testing results is that these methods use non-specific methods, for example light-scattering or microscopy, for assessing growth in the antimicrobial-containing aliquots. Consider the case if there were more than one species present, for example a pathogen and species of normal microbes that are part of the human microbiome—which is the case in most primary patient specimens. If growth were observed in an antimicrobial-containing aliquot, it would be impossible to tell, using a general method for detecting growth, whether the disease-causing pathogen or one or more of the commensal species was resistant and capable of growing.
In contrast, to conventional methods and others that require purified pathogen cells because they use non-specific methods for detecting whether the pathogen grows in the presence of antimicrobials, the inventive methods use pathogen-specific detection to assess growth of the pathogen in antimicrobials. Because only the disease-causing pathogen cells are enumerated after the incubation step (any commensal microbes are not enumerated) the inventive method can be used to determine antimicrobial susceptibility directly in the non-sterile primary specimen containing one or many commensal microbial species.
Systems and methods of the invention can be used to identify target cells or microbes and to, separately or within the same device, test for antimicrobial susceptibility of the target in the specimen. In a preferred embodiment of the invention, testing devices internally divide a specimen into separate portions where some of the portions may be incubated in the presence of various antimicrobial agents before imaging to determine differential growth. One or more of the portions may be directly processed and imaged to provide a baseline reference for determining growth, growth inhibition, or morphology changes in the incubated portions. By quantifying the growth of portions incubated in various antimicrobials, the effectiveness of each antimicrobial agent in reducing or preventing growth of the target is determined. By observing changes in target cell count or cell morphology, the effectiveness of treatments can be determined.
Testing devices and instruments described herein are capable of identifying and testing efficacy of agents on varying classes of targets (e.g., viruses, human cells, bacterial cells, or fungal cells) and also simultaneously performing such identifications and antimicrobial susceptibility testing for multiple different targets in a single device, thereby allowing development of a single instrument that performs tests typically conducted by multiple testing devices designed for different testing applications (e.g., blood, urinary tract, gastrointestinal, and respiratory infections). Accordingly, robust functionality is provided by the instruments and testing devices described herein.
Once a target is identified, the antimicrobial susceptibility testing can include antimicrobial agents or treatments relevant to that target (e.g., those commonly used in treatment or known to inhibit growth of the identified target). As noted earlier, identification of the target cell or microbe can be important for determining the appropriate target-specific therapies. Identification can be performed using the same processing and imaging techniques as used in the antimicrobial susceptibility testing methods described herein and can be performed using the same types of devices, instruments, and methods used for the differential growth or therapeutic efficacy analyses. In certain embodiments, identification and therapeutic susceptibility testing can be performed on the same specimen (divided into separate portions) in the same device. Target identification can also be performed using other techniques, not falling with the Systems and methods of the invention, such as amplification with target-specific primers, immunoassays, mass spectrometry, nucleic acid sequencing, or oligonucleotide probe array analysis. If identification is performed separately, an antimicrobial susceptibility testing device containing the appropriate reagents and antimicrobials for the identified pathogen may be used according to the present invention.
Detecting differential growth in the presence of various antimicrobial agents may require different amounts of time depending on the target pathogen, but is greatly reduced from the days required for standard antimicrobial susceptibility testing techniques. For example, differential growth of microbes commonly associated urinary tract infections can be observed in urine specimens using techniques of the invention after about 4 hours or less. As noted above, identifying and quantifying microbes in a specimen can be accomplished in thirty minutes or less thereby allowing for antimicrobial susceptibility testing results to be obtained within hours after introduction of the specimen to the testing device.
In certain embodiments, a test suite of consumable cartridge devices could be used for identification (ID) and antimicrobial susceptibility testing (AST) for a syndromic infection (for example, pneumonia or urinary tract infection). Such a test suite, referred to as an ID/AST test, could comprises an ID cartridge and a syndromic infection-specific family of AST cartridges each of with contain appropriate antimicrobials for testing individual pathogens or related groups of pathogens.
The ID cartridge in such a syndromic infection test suite could detect infections and identify and quantify the bioburden, or concentration, of the infectious pathogen(s) in the specimen. The same cartridge could also simultaneously test the specimen for diagnostically informative markers including host response biomarkers (e.g., cytokines) and inflammatory cells (eg, neutrophils), or cellular markers of sample quality (e.g., squamous epithelial cells) on the same device. The ability to combine detection and quantification, of pathogens, biomarkers, and diagnostically informative host cells in the same specimen, cartridge, and instrument is a novel and potentially powerful advantage of the inventive System and methods.
If an infection were detected and pathogen identified using the ID cartridge (or an alternative identification method aside from the inventive method) an AST cartridge would be chosen from the family of AST cartridges for AST analysis. The AST cartridge chosen for analysis would contain the appropriate antimicrobials that might be used for treating the particular pathogen and the appropriate FISH reagents to enumerate the particular pathogen.
Antimicrobial susceptibility testing results obtained using systems and methods of the invention are then used to determine the antimicrobial susceptibility profile for the infectious pathogen and to inform treatment decisions of patients so that the patient may be treated with an effective antimicrobial agent.
Detectable labels may include target-specific fluorescent oligonucleotide probe (including probes comprising modified nucleotides or nucleotide analogues) or fluorescent antibodies, specific and nonspecific ligands, lectins, stains, and dyes that bind targets. In certain embodiments, magnetic tags are used in combination with the detectable labels to bind to target microbes before magnetically selecting and imaging the target microbes. Separation can occur within a testing device as described herein and magnetic fields may be used to deposit the labelled microbes on a detection surface in the testing device to be imaged. In certain embodiments, a dye cushion layer, as described in U.S. Pat. No. 9,643,180, incorporated by reference herein, can be used in the separation and imaging steps to minimize or eliminate specimen preparation steps by the user, eliminate wash steps, and reduce background signal. Digital, non-magnified imaging techniques as described in U.S. Pat. Nos. 9,643,180 and 8,021,848, each of which is incorporated herein by reference, can be used to quantify labelled microbes including, for example, single cells.
Aspects of the invention provide methods for determining antimicrobial susceptibility. Such methods preferably include the steps of obtaining a specimen from a patient suspected of having a syndromic infection wherein the specimen type would potentially contain the infectious pathogen(s). The specimen is then introduced into a testing device, divided into a plurality of aliquots. One aliquot is analyzed immediately to determine the baseline concentration of pathogen cells before incubation. For this aliquot the pathogen cells are fluorescently labeled, magnetically tagged, drawn through the dye-cushion, and deposited on the imaging surface of an imaging well in the cartridge, imaged, and quantified using image analysis. The other aliquots are incubated at 35° C. in the presence of growth media and various antimicrobial agents, all within the testing device. After differential growth in the presence of the various antimicrobials the pathogen cells are fluorescently labeled, magnetically tagged, drawn through the dye-cushion, and deposited on the imaging surface of an imaging well in the cartridge, imaged, and quantified using image analysis. The number of pathogen cells enumerated in the aliquots containing antimicrobials are compared to the number of pathogen cells enumerate initially (before incubation) to determine the antimicrobial susceptibility profile and which antimicrobials would be effective for treating the patient.
Methods of the invention may include detecting infections and detecting and identifying the infectious pathogen cell before introducing the specimen into the testing device and selecting the plurality of different antimicrobial agents based on the identity of the target cell or microbe. Identification of a target pathogen preferably includes exposing a first specimen from the patient to magnetic tags and fluorescent labels that can bind to the first targets such that complexes comprising magnetically tagged and fluorescently labeled targets are specifically formed. Applying a magnetic field to the testing device to attract the complex to a detection surface; and imaging the detection surface to detect and quantify the detectable label where presence and concentration of the detectable label indicates whether there is target pathogen present and how much of that target pathogen is present. The detection step may take less than about 30 minutes.
For the inventive antimicrobial susceptibility testing applications, the target pathogen cells can be detected after differential growth as described above.
The systems and methods of the invention can use a similar strategy for detecting and quantifying subcellular targets (for example, toxins, biomarkers, host-response factors, viral-specific molecules, or virus particles). Target-specific magnetic tags and fluorescent labels are preferably used to bind to such targets to form complexes. The systems and methods of the invention are used to deposit these complexes on the imaging surface of for enumeration by imaging and image analysis. The target-specific magnetic tags and fluorescent labels for detecting subcellular targets are preferably magnetic and fluorescent particles that are conjugated to target-specific binding agents (e.g., antibodies, aptamers, receptors, ligands). Specific formation of the magnetically tagged and fluorescently labeled target complexes occurs in various ways. Either the magnetic tag or the fluorescent label may be designed to specifically bind to the target. Alternatively, both the magnetic tag and fluorescent label may be designed to specifically bind to the target. In either case, magnetic selection combined with imaging of the magnetically selected complexes results in detection and enumeration of the specific targets in a specimen. There are various mechanisms by which that magnetic tags and fluorescent labels can associate with the targets.
There are various ways that magnetic tags or fluorescent labels may bind non-specifically to targets. For example, binding magnetic tags or fluorescent labels that bind to a conserved site across various categories of targets can be achieved by conjugating the magnetic tags or fluorescent labels to moieties that bind to those sites (e.g., antibodies or other protein binding partners, lectins, or ligands). Magnetic tags or fluorescent labels may also bind non-specifically due to general chemical or colloidal attributes. For example, positively-charged magnetic particles or fluorescent labels can bind non-specifically to bacterial cells, which are generally negatively charged. Cells can be labeled non-specifically by various dyes (e.g., calcofluor) or fluorogenic dyes (e.g., propidium iodide, fluorescein diacetate). Dyed fluorescent particles can be used as fluorescent labels that bind non-specifically to target cells by virtue of their chemical or colloidal attributes or by conjugating them to non-specific binding molecules such as those described above.
Magnetic tags or fluorescent labels can also be chosen in various ways so that they bind to targets specifically. For example, magnetic tags (or fluorophores) can be conjugated to antibodies that bind to target-specific antigens. To similar effect, magnetic tags or fluorophores could be conjugated to a molecule (e.g., avidin) that binds specifically to a ligand (e.g., biotin) that is bound to (or can bind to) such a target specific antibody. Cells can be also labeled specifically by reassociation or hybridization with target-specific nucleic acid probes (or nucleic acid analog probes) that are themselves labeled with fluorophores. Dyed fluorescent particles can be used as fluorescent labels that bind specifically to target cells by conjugating them to target specific binding molecules such as those described for certain embodiments, the labelling and imaging steps include exposing the specimen portions to a fluorophore-labelled target-specific binding molecule and a magnetic particle wherein the fluorophore-labeled target-specific binding molecule and the magnetic particle bind to the target forming a complex; applying a magnetic field to the testing device to attract the complex to an detection surface; and imaging the detection surface. The fluorophore-labeled target-specific binding molecule may include an oligonucleotide probe that binds specifically to the target cell. The exposing and imaging steps can include fluorescent in situ hybridization (FISH) methodology and analysis.
Methods of the invention may include determining a recommended antimicrobial for the patient, usually comprising the antimicrobial or other treatment determined to inhibit growth of the target. Determination of the treatment that inhibits growth of the target can occur about 4 hours after introducing the specimen into the testing device. The bodily specimen can be a tissue specimen (e.g., a wound or biopsy specimen) or a bodily fluid specimen. Preferred bodily fluids for use with the invention include, but are not limited to, respiratory (for example, sputum endotracheal aspirate, protected specimen brush, broncho-alveolar lavage), blood, urine, stool, swabs (for example, nasal, oral/pharyngeal, surgical site, skin and soft tissue, rectal), and cerebrospinal fluid.
In certain aspects, the invention provides systems for determining therapeutic susceptibility of a target cell or microbe in a specimen. Preferred systems include a testing device operable to receive a specimen comprising a bodily fluid from a patient and a target cell or microbe as well as an instrument or instrument. The instrument is preferably operable to manipulate the testing device to divide the specimen into a plurality of portions within the testing device; incubate the portions in the presence of a plurality of different therapeutic agents within the testing device; fluorescently label and magnetically tag the target cell within the incubated portions within the testing device; separate the magnetically tagged and fluorescently labeled target complexes from the unbound fluorescent labels; and image the portions within the testing device to quantify the target complexes so as to determine which treatment(s) inhibit replication of the target cells.
Systems may comprise a testing device that, for microbiological applications, can be used to detect infections and identify pathogens. A patient specimen can be added to the device where it can be split into multiple aliquots, each of which can be contacted with magnetic tags and multiple types of target-specific detectable labels (e.g., binding molecules with distinct fluorophore labels) such that labeled, magnetically tagged target complexes are formed; apply a magnetic field to the testing device to attract the complex to an detection surface; and image the detection surface to detect the labeled complexes, wherein detection of complexes labeled with a particular detectable label in a particular aliquot indicate the presence of a particular target.
The instrument may be operable to contact the specimen aliquotes with various antimicrobial agents for antimicrobial susceptibility applications; to contact aliquots or portions containing specimen to magnetic tags and detectable labels chosen so that complexes are specifically formed with a target, the magnetic tags, and the detectable label; apply a magnetic field to the testing device to attract the complex to a detection surface; image the detection surface; and perform image analysis to determine the test results.
Detecting the number of identified target cells or microbes in each of the incubated specimens may include contacting the incubated specimens magnetic tags and detectable labels chosen so that complexes are specifically formed with a target, the magnetic tags, and the detectable label; applying a magnetic field to the complex to attract the complex to an detection surface; and imaging the detection surface to determine an effect of each of the therapeutic agents on growth of each of the identified target.
Multiple aliquots may be combined with the same antimicrobial agent present in different concentrations, preferably corresponding to 2-fold serial dilutions of the antimicrobial agent or to concentrations corresponding to CLSI breakpoints for antimicrobial susceptibility testing. The number of such portions and concentrations of antimicrobials can be chosen so as to deliver a Susceptible/Resistant result, a categorical (Susceptible, Intermediate Resistant, Resistant or SIR result), or a Minimum Inhibitory Concentration (MIC) result. The relevant concentrations of antimicrobials for specific microbial pathogen species are documented by the Clinical Laboratory Standards Institute (CLSI).
The invention provides systems and methods for simultaneous and automated performance of different tests using a single instrument. Contemplated tests to be executed by the instruments include identification and antimicrobial susceptibility testing analysis of target cells, molecules, viruses, or microbes in specimens with little or no specimen preparation required by the user. The invention provides systems and methods for detecting, quantifying, and identifying target pathogens, diagnostically informative host cells, and diagnostically informative molecular biomarkers in a specimen. Embodiments of the invention perform different types of assays, or assessments of a single analyte. Some embodiments of the invention perform a test, or set of assays that are done on a single specimen in a single cartridge. Assays according to the invention detect specific molecular targets and cellular targets and determine effective therapy.
Target microbes, pathogens, and cells contemplated for testing using systems and methods of the invention include viruses, bacteria or fungal cells, human cells (including but not limited to leukocytes, squamous epithelial cells, or virally infected human cells), or other eukaryotic cells (including parasite, animal, and plant cells). Target molecules contemplated for testing include target-specific antibodies, nucleic acid probes, ligands, aptamers, and statins. Target cells, molecules, viruses, or microbes in specimens can be analyzed for differential growth in the presence of various antimicrobial agents or the effects of therapeutic agents on cell viability, morphology, viral load, cell functionality, or other measures of therapeutic efficacy can be determined using systems and methods described herein.
Specimen manipulation, incubation, processing, and analysis steps are performed within a single testing device in under about thirty minutes for target cell, molecule, virus, or microbe identification and under about four and a half hours for antimicrobial susceptibility testing in the case of urine analysis for common causes of urinary tract infection. Analysis time may vary based on the targets being analyzed (e.g., depending on the growth rate of the target) and the therapeutic agent being tested. Systems and methods detect specific targets using non-magnified digital imaging and image analysis to accurately and quickly quantify targets in a specimen.
Embodiments of the invention allow for testing with minimal or no specimen preparation by the user. By reducing extraneous steps, and using imaging methods and image analysis capable of quantifying single cells or colony forming units (CFU) in certain instances, actionable results directly from patient specimens for target cell, molecule, virus, or microbe identification and antimicrobial susceptibility testing are obtained in a matter of hours as compared to multiple days with conventional techniques. Variations that may be required across various target cells, molecules, viruses, and microbes may include the target specific binding molecules (e.g., target-specific antibodies, nucleic acid probes, ligands, aptamers, and statins), the treatments being tested (e.g., antivirals, antimicrobials, or other therapies), incubation times, characteristics being analyzed, image analysis algorithms, parameters, and results reported.
Testing devices can be automatically manipulated by an instrument to carry out each of the steps of dividing a specimen, culturing in the presence of different antimicrobial agents or other treatments, and processing and imaging the resulting specimen portions to measure differential growth and determine effectiveness of the tested antimicrobial agents.
Application-specific cartridges can include a series of interconnected wells preloaded with the reagents required for a particular test. A user need only add a specimen to the cartridge for the desired test and specimen type and then load the cartridge into an instrument for automatic processing. The instrument can scan labels containing both information about the application-specific cartridge itself and patient specimen information. Alternatively, information can be input via the user interface by the operator. Instruments as described herein can include stations for performing various steps that may be required for the different tests and may include a carousel or other mechanism for storing multiple in-process cartridges and transferring them between the stations as required for performance of the test steps.
The cartridge can include wells and various valves and channels for connecting different wells therein as needed to perform the steps of the desired test. For example, a cartridge for an antimicrobial susceptibility testing test may include a series of division wells pre-loaded with growth media and different antimicrobials to be analyzed. Cartridges may also include wells for processing the post-growth specimens and labeling targets therein as well as imaging wells providing a detection surface for imaging the labelled targets. Instruments may include a fluidics module to interface with the cartridge to allow for external manipulation of the valves and pressures in the cartridge to connect different wells and move the specimen volumes therebetween as required for the various tests.
The instrument may be an automated benchtop instrument designed to accommodate a menu of application-specific consumable cartridges for major infectious diseases. Due to unique non-magnifying digital imaging technology, the instrument allows for high performance. The instrument allows for elimination of user steps, use of inexpensive reagents, and streamlined instrumentation, resulting in low cost and ease-of-use advantages.
The instrument achieves single molecule counting and counts individual cells without magnification. The instrument is unique in its ability to rapidly and sensitively detect target toxin molecules, biomarkers, cells, and antimicrobial resistance.
The unique dye cushion layer allows the instrument technology to rapidly and specifically count target molecules in a complex specimen without specimen preparation (
In an example, Clostridium difficile is the target molecule. The magnetic nanoparticles are coated with antibodies specific to one antigenic site on the C. difficile toxin molecule and the fluorescent nanoparticles are coated with complementary antibodies binding to a distinct antigenic site on the same molecule. Upon hydration of the dried reagents by the specimen, two layers form: the dense dye-cushion layer and the assay layer. In the assay layer, the target molecules bind to the magnetic and fluorescent nanoparticles, tethering them together. The high concentration (˜109/ml) and small size (200-500 nm) of the particles drive rapid binding kinetics with diffusive mixing only, which simplifies the instrumentation by eliminating the need for mechanical mixing functionality.
Placing the imaging well over permanent magnets for 3 minutes draws the magnetic particles—and any fluorescent particles that are tethered to them via target molecules—through the dye-cushion layer, depositing them on the bottom imaging surface. The captured fluorescent particles are imaged and counted instantaneously with non-magnified digital imaging as above. Because a single molecule can tether a fluorescent particle to a magnetic particle at low target concentrations, counting the number of magnetically deposited particles corresponds to the number of captured target molecules—and this is done without using magnification.
The dye-cushion eliminates specimen preparation and wash steps.
The invention can be used to detect a broad range of analytes. In addition to detecting molecular targets such as toxins and biomarkers, the instrument can also detect and count cellular pathogens (e.g., bacterial, fungi, and parasites), viruses, and diagnostically important human cells. For example, the technology can detect infections and identify a bacterial pathogen in just 30 minutes. Plus, antimicrobial susceptibility tests allows for selection of target antimicrobial therapy to treat an infection in just 4 hours—as compared to the 2-4 days required using traditional culture tests.
In some embodiments, specimens 104 may be introduced 107 directly into the testing device 109 and generally do not require any pre-processing steps. For example, a urine specimen, direct from a patient may be pipetted into a testing device 104 without the need for prior specimen preparation including mixing with reagents, target purification, treatment to remove components of the specimen, centrifugation, biochemical enrichment, colony purification, or other treatments. Once obtained 103, the specimen 104 is introduced 107 into a testing device 109.
In some embodiments, steps include specimen processing 106. The patient specimen 104 may be obtained 103. The patient specimen 104 may then undergo specimen processing 106 before being introduced 107 to the testing device 109. In certain embodiments, the specimen processing may be up-front specimen processing. Some embodiments may include onboard specimen processing including filtration, cell purification, specific removal, enrichment, or any combination thereof for specific classes of cells or analytes.
The cartridge 201 can interface with the fluidics module of an instrument as described herein at, for example, a pneumatic port 217. The valves 207 can control the movement of fluid between the division wells 205 and the processing 209 and imaging wells 211 by opening and closing connections therebetween. The valves 207 may be configured to allow one or more portions of a specimen to bypass the division wells 205 and proceed directly to a processing 209 and imaging well 211 to provide an un-incubated reference (nt0) for identifying the presence of a target cell or microbe, for antimicrobial susceptibility testing, or both. The valves 207 may be in a sliding bar 223 design, as shown in
The division wells 205 can be pre-filled with growth media and/or one or more antimicrobial agents. The valves 207 can hold the specimen portions in the division wells 205 to be incubated in the presence of the various antimicrobial agents for any period of time.
The specimen may be mixed with an amount of growth media within the testing device 109 at any point between introduction 107 and incubation 115 of the specimen 104. The incubation 115 step may last less than about 30 minutes, 1 hour, 2 hours, 3, hours, 4 hours, 5 hours, 6 hours, 12 hours, or 24 hours or more but is preferably about 4 hours to allow for measurable differential growth of the target cells or microbes in the specimen 104. The left-most path shows an un-incubated portion that is imaged to determine the initial number of target cells (nt0) present in the absence of incubation.
After incubation 115, the incubated portions 117 can be exposed to, for example, magnetic tags and detectable labels that can form complexes with 119 the specific target cell in the specimen portions for imaging 127. The labelled target complex 121 can then be magnetically deposited 123 on the detection surface. The deposited target complex 125 can be imaged 127 in the portions, and the images 129 can be analyzed by the instrument's image analysis software to quantify the amount of target 121 detected in each portion.
By comparing the quantity of target cells or microbes in portions that had been incubated in the presence of various antimicrobial agents, one can determine which of those agents inhibited growth. Comparison preferably includes a target cell quantity (nt0) from processing and imaging an un-incubated reference portion of the specimen.
Systems and methods of the invention can be used to identify appropriate therapeutic treatment for pathogens (including bacteria, fungi, parasites, and viruses). For these applications, target pathogens are incubated with various therapeutic agents under conditions that in the absence of the therapeutic agents would result in pathogen decreased viability. Potentially effective agents for treating infections caused by target pathogens are identified by determining which agents negatively impact target viability as evidenced by reproduction, replication, growth, or phenotypes indicative of these qualities and behaviors. Target bacterial pathogens, for example, can be incubated in nutrient bacterial growth medium in the presence of various antimicrobial agents and the effects of those agents on target bacterial growth in the specimen can be analyzed to determine the effectiveness of the antimicrobials.
Differential growth of microbial pathogens following incubation in the division wells 205 containing growth medium and antimicrobials is useful for determining antimicrobial susceptibility testing. In various embodiments, specimen portions may be incubated in the division wells 205 for less than about 1 hour, 2 hours, 3, hours, 4 hours, 5 hours, 6 hours, 12 hours, or 24 hours before processing and imaging in the imaging wells. The growth media and/or antimicrobial agents may be selected and included in the testing device 201 based on the target cell or microbe to be analyzed in the specimen. For example, growth media known to support growth of an identified target cell or microbe and therapeutic or antimicrobial agents commonly used to treat the identified target may be selected. In various embodiments, a testing device 201, may be pre-set with growth media and therapeutic agents for a certain target such that a user, having identified a target in a specimen (e.g., determined the source of a patient's infection) can select the appropriate pre-packaged testing device for antimicrobial susceptibility testing of the target microbe or therapeutic efficacy analysis of the target cell.
After an appropriate incubation period, the valves 207 can be manipulated to transfer the incubated specimen portions from the division wells 205 to the reagent wells 209. The processing 209 wells may be pre-filled with processing reagents for labelling the target cell or microbe for imaging. The processing reagents may be lyophilized or air-dried for storage and activated upon contact with the fluid specimen portion and growth media. The reagent wells 209, for example, may contain target-specific detectable labels and magnetic tags. As the incubated specimen portion passes through the reagent well 209, the target-specific complexes are formed comprising the target cells, detectable labels, and magnetic tags.
The testing device 201 can then be subjected to a magnetic field (e.g., placed on a permanent magnet) to pull the complexes to the detection surface on the bottom of the imaging well 211. The imaging wells 211 may contain a dye cushion that forms a dense opaque aqueous layer lying below the upper assay layer in the imaging well as described herein. All of the magnetic particles including target complexes, that is magnetically-tagged fluorescently-labeled target complexes, are drawn through the lower dye cushion layer and deposited on the detection surface 215 of the imaging well 211 by the magnetic field bringing the labelled target cells or microbes through the dye cushion layer to a detection surface 215 on the imaging well 211. The detection surface 215 may be optically clear to allow for optical detection of labelled target cells, molecules, viruses, or microbes. After processing and magnetic selection, the testing device 201 can be placed on an imaging stage for image processing as described below.
Testing devices can have any number of division wells and corresponding processing and imaging wells. While systems and methods of the invention are primarily described herein in the context of a testing device having interconnected wells for automated processing, the techniques can also be performed in any known fluidic platform including manual processing in the wells of microtiter plates or on substrates with droplets functioning as wells. In some embodiments, the number of wells is preferably large (e.g., one hundred or more) and is limited only by the size constraints of the testing device and instrument.
By quantifying the amount of target cells, molecules, viruses, or microbes present in each specimen portion after incubation in the presence of various antimicrobial agents and comparison to an un-incubated growth reference (nt0) taken from the specimen pre-incubation, the effect of each therapeutic agent on target cell, molecule, virus, or microbe growth can be determined. By determining which antimicrobial agents best inhibited growth, an effective therapy can be determined to treat the patient's infection.
Cartridges are used to process assays for detecting specific molecular targets, specific cellular targets, and to determine effective therapy. When detecting specific microscopic or sub-microscopic cellular or molecular targets, target-specific optical probes, such as fluorescent probes, are designed to bind to the specific cellular or molecular targets. Target-specific magnetic particles are also designed to bind to the cellular or molecular targets. A specimen potentially containing targets is first mixed with the labeling probes and particles in the cartridge. The cartridge is incubated for a period of time in order to form complexes, each complex comprising a target, optical probe, and magnetic particle. A magnetic field is applied to pull the magnetic particles, and the rest of the complex, down through a dye cushion. Optical probes bound to the target will fluoresce, and imaging is used to identify the presence of the target cell or molecule. Imaging technology detects targets labeled with optical probes without using magnification. Illuminating the labeled targets causes them to fluoresce. The emitted photons impinge on a digital camera's CMOS chip containing an array of independent photosensitive pixel elements. The pixel elements lying directly above the individual targets “light up” as white spots in the resulting image. A computer instantaneously enumerates the illuminated pixels indicating the number of targets present. At low analyte concentrations, digitally counting individual labeled targets generates a better signal to noise ratio compared to the more common method of integrating the signal across the detection area. Non-magnified imaging allows a large field to be imaged instantaneously—allowing for a small number of cellular targets to be rapidly detected in large volume of specimen.
Preferably, the target cell, molecule, virus, or microbe has already been identified and the various antimicrobials have been selected accordingly as, for example, antimicrobials commonly used to treat the identified target. Each division well may contain a single antimicrobial agent or a combination of antimicrobial agents to assess combined effects on target cell, molecule, virus, or microbe growth. Different antimicrobial agents at different concentrations may be combined with the specimen portions or aliquots in different division wells and some division wells may be used for replicating treatment with the identical antimicrobial agents to strengthen results. Specimen portions or aliquots in some division wells may not be combined with antimicrobial agents to quantify uninhibited growth and/or to detect growth inhibition due to either non-antimicrobial inferences or reagent instability.
After incubation 507, the incubated specimen aliquots can be processed and imaged in a manner similar to the t0 specimen aliquot 505 and the counts of bacteria or other microbes can be compared to each other, the t0 specimen aliquot, or to other control counts 509 to determine effectiveness for the antimicrobial agents or combinations thereof. For example, the antimicrobial agent in Abx well 2 inhibited growth, while the others did not. Accordingly, the antimicrobial agent from Abx well 2 would be recommended for treatment of the patient from whom the specimen was obtained.
Various instruments or instruments may be used to interact with testing devices to carry out target identification and antimicrobial susceptibility testing of specimens among other tests. An instrument may include an input mechanism for accepting and cataloging testing devices. Testing devices may include one or more identifier tags readable by the instrument (e.g., a barcode) to determine the type of application-specific testing device and specimen and patient information.
The instrument preferably is controlled by a computer to automate manipulation of the testing device, performance of target cell, molecule, virus, or microbe identification and antimicrobial susceptibility testing analyses, and generate and analyze imaging results. The instrument may be linked to a user interface such as a touch screen to display prompts and results and receive commands. An instrument may comprise a conveyor or robotic arm to move testing devices within the instrument. To perform the required steps for target identification, the instrument preferably comprises a magnetics module having, for example, a permanent magnet or an electro-magnet to provide a magnetic field to deposit complexes of magnetic particles and labeled targets on the detection surface to be imaged. The instrument can also include an imaging module such as those described in U.S. Pat. Nos. 9,643,180 and 8,021,848, the contents of each of which are incorporated herein by reference, to capture images of the labelled target cells and a stage to manipulate the detection surface of the testing device relative to the imaging module of the instrument. The imaging module can be operably associated with the computer to provide image processing, analysis, and display capabilities. The instrument can also comprise a waste module for disposal of testing devices after use.
For antimicrobial susceptibility testing analysis, the instrument may include one or more incubation areas for storage of testing devices during incubation for growth and/or test incubation. The incubation area may include a heating and/or cooling element and a thermostat to control that element to maintain the incubation area at a desired temperature for growth of the target cells or microbes (e.g., 35° C.) or for carrying out test incubation. The instrument may include a fluidics module to drive movement of the specimen and reagents within the testing device through, for example, by manipulation of valves, plungers, and actuators using functionality provided by hydraulic, pneumatic pressure or vacuum, or mechanical means incorporated into the fluidics module. In some embodiments a mechanical conveyor arm may be operable to manipulate the testing device amongst the various functional modules including the incubation, fluidics, magnetics, imaging, and waste modules.
Detectable labels may include any suitable type of optically detectable label. Non-limiting examples of detectable labels include resonance light scattering particles and quantum dots. The detectable label may comprise a target-specific portion to preferentially bind to a target cell, molecule, microbe, virus, host cell, or other non-bacterial molecule. The target-specific binding molecule may include for example, an antibody that binds to a target-specific antigen or a nucleic acid (or nucleic acid analog) probe complementary to a target-specific nucleic acid sequence.
Fluorophores on the distinct fluorescent probes can have distinct photonic signals so that the fluorescent signal for different categories of target cells or microbes in the test can be differentiated by distinct photonic signals. Imaging methods to distinguish multiple distinct photonic signals are known to those familiar with the art. For example, multiple images can be acquired using distinct pairs of excitation and emission optical filters that correspond to the action spectrum of the distinct fluorophores. Accordingly, a single specimen portion can be tested for the presence of multiple specific target cells, molecules, viruses, or microbes in a single processing and imaging well.
In certain embodiments, detectable labels and magnetic tags, and target-specific binding molecules may be separate or combined/linked. In certain embodiments, magnetic particles and detectable labels may be non-specific such as avidin-coated mag particles and SYBR-green dye. For example, a biotin-labeled target-specific antibody (i.e., a target-specific binding molecule) could be used to target a specific cell or microbe which would then be tagged by the avidin magnetic particle for specific magnetic selection. All cells would be labeled by SYBR-green, but only the magnetically separated (i.e., biotin tagged) targets would be deposited on the detection surface for imaging.
In a non-limiting embodiment, detectable labelling of microbes comprises fluorescent in situ hybridization (FISH). FISH analysis uses fluorescent probes comprising nucleic acid (or nucleic acid analog) probe moieties and fluorophore moieties to bind via reassociation with target-specific nucleic acid sequences so that the targets can then be detected optically. For example, probes targeting target-specific 16S rRNA can be used to selectively label and detect microorganisms. See, Volkhard, et al., 2000, Fluorescent In Situ Hybridization Allows Rapid Identification of Microorganisms in Blood Cultures, J Clin Microbiol., 38(2):830-838, incorporated herein by reference. For identification tests, a number of distinct target-specific fluorescent probes may be used to independently identify multiple distinct categories of targets in a single specimen. The distinct fluorescent probes can comprise distinct nucleic acid probe moieties designed such that under test reassociation conditions, the nucleic acid probe moieties of the fluorescent probes preferentially reassociate with target-specific cellular nucleic acid sequences for distinct categories of target cells or microbes. Detailed discussion of probe reassociation can be found in United States Patent Publication 2003/0228599, incorporated herein by reference.
In certain embodiments, FISH analysis for identification and/or quantification of target cells or microbes in a specimen can be performed isothermally, without reagent changes, and without cell fixation allowing for automatic, on-device processing in about 30 minutes or less between reaction initiations and obtaining imaging results.
Magnetic particles and applied magnetic fields can be used to physically separate bound detectable labels from unbound detectable labels, in solution, without a wash step. As described in U.S. Pat. No. 9,643,180, the contents of which are incorporated herein by reference, image background can be minimized or eliminated by using a dye-cushion layer to optically sequester the specimen matrix and the unbound optical labels from the detection surface of the imaging well. The dye in the dye-cushion layer is preferably chosen to absorb the excitation and emitted light used by the instrument for imaging. Thus, the signal from unbound labeling moieties in the assay layer does not significantly interfere with detecting the signal from the labeled target cell, molecule, virus, or microbe complexes magnetically deposited on the detection surface. Similarly, the use of the dye-cushion prevents any auto-fluorescence from the specimen matrix, also contained in the assay layer, from significantly interfering with detection of the signal from the deposited labeled target-cell complexes. These attributes of the dye-cushion can make it possible to detect the target-cells or microbes without specimen preparation by the user and without wash steps to remove the unbound label from the test device.
Digital imaging of labelled target cells, molecules, viruses, or microbes can be accomplished using digital imagers. In the preferred case of fluorescent labelling, various lenses, illumination sources, excitation light sources, and filters may be used. Imaging modules may include any device capable of producing a digital image of the detectably labeled target cells, molecules, viruses, or microbes in a solution or pulled to a detection surface in a well or testing device. Imaging modules may include, for example, CCD cameras, CMOS cameras, line scan cameras, CMOS avalanche photodiodes (APD's), photodiode arrays, photomultiplier tube arrays, or other types of digital imaging detectors.
Imaging can be carried out under a single set of conditions or light sources, filters, and/or lenses may be changed between images to detect different optically distinguishable labels (e.g., different fluorescent probes corresponding to different target cells or microbes). The imaging techniques and instruments described in U.S. Pat. Nos. 9,643,180 and 8,021,848, the contents of each of which are incorporated by reference herein, may allow for observation and enumeration of individual cellular, molecular, or viral targets.
Subsystems of the instrument 601 may include a pneumatic subsystem 709, a magnetic subsystem 711, an imaging subsystem 713 and a trash subsystem 715. The magnetic subsystem 711 may include, for example, a permanent magnet or an electro-magnet to provide a magnetic field to deposit complexes of magnetic particles and targets on the detection surface of the analytical cartridge for imaging. The imaging subsystem 713 may be such as those described in U.S. Pat. Nos. 9,643,180 and 8,021,848, the contents of each of which are incorporated herein by reference, as discussed above and a stage to manipulate the detection surface of the analytical cartridge 109 relative to the imaging module of the instrument 601. The imaging subsystem 713 can be operably associated with the computer to provide image processing, analysis, and display capabilities. The pneumatic subsystem 709 may be operable to drive movement of the specimen 104 and reagents within the analytical cartridge 109 through, for example, manipulation of valves 207 (e.g., plungers and actuators) using functionality provided by pneumatic pressure or vacuum. In some embodiments, hydraulic or mechanical means may also be incorporated into the pneumatic subsystem 709 to effectuate movement of the specimen 104. The waste subsystem 715 may include a receptacle (e.g., a removable bin) for disposal of analytical cartridges 109 after use.
The instrument 601 may also include one or more incubation areas for holding (or storing) analytical cartridges during incubation for growth and/or test incubation. The incubation area may include a heating and/or cooling element and a thermostat to control that element to maintain the incubation area at a desired temperature for growth of the target cells or microbes (e.g., 35° C.) or for carrying out test incubation. In various embodiments, tests to be run on the instrument may be designed to be isothermal such that a single temperature is required and the interior of the instrument can be maintained at that temperature. In such instances, simple storage slots adjacent to the carousel or the carousel itself can serve as incubation storage for the various incubation steps of the tests being performed.
In some embodiments the mechanical conveyor arm mechanism 707 may be operable to manipulate the analytical cartridge 109 amongst the various subsystems within the instrument 601. In some embodiments of the invention, the mechanical conveyor arm 707 transfers each of the analytical cartridges 109 between the carousel 605 and the various subsystems of the instrument. The mechanical conveyor arm 707 applies a pulling or pushing force to transfer analytical cartridges 109 onto and off of the carousel 605. The carousel 605 rotates to position an analytical cartridge 109 adjacent another one of the subsystems and the mechanical conveyor arm 707 may then apply force to slide the analytical cartridge 109 onto the subsystem. Analytical cartridges 109 are never grabbed by the mechanical conveyor arm 707 or any other elements of the instrument 601. Sliding, or pushing the analytical cartridge 109 within the instrument 601 reduces exposure to debris. The various stations or subsystems within the instrument as well as the carousel 605 comprise slots 717, 719 sized to accept and guide the cartridge 109 as it is slid between the carousel 605 and the various stations. Rotation of the carousel 605 functions to align the slots 717 thereon with a corresponding slot 719 on a station such that a guide or track is formed along which the cartridge 109 can be slid by the mechanical conveyor arm 707.
In some embodiments, the instrument includes a task scheduler for managing the analytical cartridges 109 within the instrument 601. The task scheduler is operable to control the movement, such as the transport and transfer of each of the analytical cartridges 109 amongst the plurality of subsystems. In some embodiments, the time each analytical cartridge 109 spends in a subsystem may also be managed by the task scheduler. The task scheduler may reserve time on various subsystems as needed for analysis of each of the analytical cartridges 109. In some embodiments of the invention, the task scheduler may manage the movement of an analytical cartridge 109 (i.e., the steps/parameters of the analysis to be performed) by identifying the contents of the cartridge.
The scheduler can comprise software stored on a tangible, non-transitory memory 1305 and operated by a processor 1303 as shown in
In some embodiments, the instrument 601 may also include a reader operable to analyze one or more identifiers (e.g., barcodes) 219 located on an analytical cartridge 109. The contents of an analytical cartridge 109 and the required processing may be associated with an identifier 219 on the analytical cartridge 109. Each of the analytical cartridges 109 may include an identifier 109 readable by the instrument 601. The instrument 601 may read the identifier 219 via a reader and associate the identifier 219 with a particular set of instructions for the task scheduler to execute. The reader may be located on the exterior of the instrument 601 and may read the identifier 219 before the analytical cartridge 109 is transferred from the loading rack 703 to the carousel 605. The identifier 219 may be a barcode and may be specific/unique to the specimen in the analytical cartridge 109. The one or more identifiers may specify specimen information, patient information, testing information, or a combination thereof.
Upon reading of the identifier, the computer processor can access a test associated with that identifier (e.g., antimicrobial susceptibility testing for E. coli). The processor can then determine a schedule for perform the required steps of the test and determine, upon commencement, when each station or subsystem will be needed and for how long to complete the test. The processor can access the schedules of other cartridges currently running in the instrument from its memory and compare the availability of the various stations at the required time. Certain steps may be flexible (e.g., incubation) and the schedule may offer a range of lengths that can be altered in order to accommodate other scheduled operations on other cartridges. If beginning a test at a certain time would result in irreconcilable conflicts for any of the subsystems or stations, the instrument may reject the cartridge and notify the user of an acceptable later time at which to start and run the test conflict-free. In certain embodiments, instruments may comprise two or more of any of the subsystems or stations to avoid such conflicts. For example, high-traffic stations such as the fluidics module may warrant the inclusion of two or more subsystems or stations, depending on the desired capacity of the instrument.
Systems and methods of the invention may include a computer operable to control the instrument and testing device and/or to process imaging results. Computers can comprise a processor coupled to a non-transitory memory device. The memory preferably stores instructions executable by the processor to cause the system to manipulate the testing device within the instrument and to obtain and process images of labelled target cells.
Processor refers to any device or system of devices that performs processing operations. A processor will generally include a chip, such as a single core or multi-core chip, to provide a central processing unit (CPU). A process may be provided by a chip from Intel or AMD. A processor may be any suitable processor such as the microprocessor sold under the trademark XEON E7 by Intel (Santa Clara, Calif.) or the microprocessor sold under the trademark OPTERON 6200 by AMD (Sunnyvale, Calif.).
Memory refers a device or system of devices that store data or instructions in a machine-readable format. Memory may include one or more sets of instructions (e.g., software) which, when executed by one or more of the processors of the disclosed computers can accomplish some or all of the methods or functions described herein. Preferably, the computer includes a non-transitory memory such as a solid state drive, flash drive, disk drive, hard drive, subscriber identity module (SIM) card, secure digital card (SD card), micro SD card, or solid-state drive (SSD), optical and magnetic media, others, or a combination thereof.
An input/output device is a mechanism or system for transferring data into or out of a computer. Exemplary input/output devices include a video display unit (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device (e.g., a keyboard), a cursor control device (e.g., a mouse), a disk drive unit, a signal generation device (e.g., a speaker), a touchscreen, an accelerometer, a microphone, a cellular radio frequency antenna, and a network interface device, which can be, for example, a network interface card (NIC), Wi-Fi card, or cellular modem. Input/output devices may be used to allow a user to control the instrument and receive data obtained from the testing devices by the instrument.
EXAMPLES Example 0: OverviewMicrotiter Plate for Pathogen Identification and Antimicrobial Susceptibility Testing
The systems and methods of the invention allow for pathogen identification (ID) and antimicrobial susceptibility testing (AST) for a full range of infections including healthcare associated infections (HAIs) such as sepsis, Catheter-Associated Urinary Tract Infection (CAUTI), ventilator-acquired pneumonia, and surgical site infections. The invention achieves rapid results by its ability to enumerate low levels of pathogens directly from patient specimens, avoiding the days required for current culture methods. The invention achieves high sensitivity and quantification by using digital non-magnified imaging to identify and count pathogen cells labeled with specific fluorescent nucleic acid probes. Following identification, the invention uses a robust phenotypic AST method to determine the appropriate targeted therapy. The tests detect infections and identify pathogens in about 30 minutes and deliver susceptibility profiles in about 4 hours. The testing workflow is similar to automated identification and antimicrobial susceptibility testing tests except that patient specimens are tested directly, eliminating the time-consuming culture-based colony purification steps. The invention eliminates cell lysis, amplification, biochemical purification, and wash steps. The method requires little or no specimen preparation. Other advantages include the ability to deliver test results for polymicrobial infections, high throughput (>80 tests/shift), and costs that are a fraction of competing rapid methods. Thus, there is a need for diagnostic tests that can rapidly detect infections, identify the pathogens, and determine the appropriate targeted antimicrobial therapy.
The ID/AST platform for rapidly detecting syndromic infections and determining targeted therapy. The ID/AST platform addresses the gap in current culture-based technologies by providing tests that rapidly and cost-effectively identify patients with serious syndromic infections, identify the pathogen, and determine the appropriate targeted antimicrobial therapy. ID tests of the invention are inherently quantitative, and AST tests of the invention use a robust phenotypic method of assessing growth in a series of antimicrobial dilutions. In contrast with culture, however, methods of the invention detect infections and identify pathogens in about 30 minutes and can deliver AST results in about 4 hours. The speed of the tests results from: (1) the ability to test specimens directly avoiding time-consuming colony purification steps; (2) the technology's ability to rapidly enumerate small numbers of pathogen cells using non-magnified digital imaging technology; and (3) the ability to assess a pathogen's susceptibility to an antimicrobial in just a few bacterial generations.
Platform overview. The ID/AST platform provides rapid ID/AST results so that patients with serious syndromic infections receive the appropriate targeted therapy at the onset of infections. The automated ID/AST instrument accommodates a menu of application-specific microtiter plate-based ID and AST consumables.
To detect infections and identify pathogens, patient specimens are analyzed in an ID consumable to first determine if there is an infection and, if so, to identify the pathogen using digital imaging of the fluorescently labeled cells. Once infection is detected and the pathogen identified, another aliquot of the same specimen is incubated in an AST consumable in nutrient medium containing various dilutions of the relevant antimicrobials. Susceptibility profiles are established by determining in which concentrations of the antimicrobials the pathogen grows.
Technical advantages. Advantages of the invention for syndromic infection ID/AST testing include rapid infection detection and pathogen ID: about 30 min; rapid, robust, comprehensive phenotypic AST: about 4 hr; ID/AST results for non-sterile specimen types; ID and AST results for polymicrobial infections; low-cost consumable and reagents; simple affordable instrument; no or minimal specimen prep, robust to specimen matrix; ultra-sensitive pathogen detection and enumeration; high-throughput (>100 specimens/shift/instrument); fully automated specimen tracking (no typing required); and internal controls for accuracy.
The invention detects individual microscopic or sub-microscopic targets without magnification. The imaging technology of the invention detects target cells labeled with fluorescent oligonucleotide probes without using magnification. Illuminating the labeled cells causes them to fluoresce. The emitted photons impinge on a digital camera's CMOS chip containing an array of independent photosensitive pixel elements. The pixel elements lying directly above the individual targets “light up” as white spots in the resulting image. A computer instantaneously enumerates the illuminated pixels indicating the number of targets present. At low analyte concentrations, digitally counting individual labeled cells generates a better signal to noise ratio compared to the more common method of integrating the signal across the detection area. Non-magnified imaging allows a large field to be imaged instantaneously—allowing for a small number of cellular targets to be rapidly detected in large volume of specimen.
A microtiter plate format allows for a large number of wells to allow Minimum Inhibitory Concentration (MIC) determination for numerous antimicrobials, as well as a consumable pipeline of tests for a broad range of syndromic infections, pathogens, and antimicrobials. FISH-based ID is used to achieve specificity, efficiency of probe design, and cost-effectiveness. A simple, efficient, and cost-effective approach for magnetic selection uses cationic paramagnetic particles which bind to negatively charged bacterial cells. Simple and proven instrumentation streamlines product development and lowers costs.
The CAUTI ID test. The CAUTI ID test determines whether a patient has a UTI caused by one or more of the common CAUTI pathogens. Table 1 lists the pathogens that will be identified by the test. Together these pathogens are responsible for 95% of CAUTI cases in the U.S. Specimens with pathogen counts greater than 10,000 CFU/ml will be scored as positive.
To assess total bacterial bioburden, the CAUTI ID test also enumerates microbial cells using a DNA stain. The test includes internal positive and negative controls demonstrating assay effectiveness using both Gram positive and Gram negative bacterial targets.
The CAUTI AST test. The AST method assesses the ability of pathogens identified by the CAUTI ID test to grow in the presence of antimicrobials.
After selective growth at about 35° C., the number of target cells in the various wells is enumerated to determine the minimum inhibitory concentration (MIC) for each antimicrobial. To assess differential growth the fluorescent pathogen-specific probe corresponding to the identified pathogen and the FISH reagents are combined with the contents of each well which are transferred to an imaging plate containing dried dye-cushion. Following incubation and magnetic selection, the ID/AST hospital instrument determines the MIC for each antimicrobial after enumerating the number of target cells in each well.
Instrumentation. The platform is a semi-automated system comprising the automated ID/AST hospital instrument, a magnetic station for microtiter plates, and an off-the-shelf 96-well pipettor for liquid transfers. The ID/AST instrument and magnetic station will be based on current automated imaging instruments and magnetic stations for analyzing microtiter plates.
Feasibility data. Developing a rapid, simple, high-performance FISH assay for pathogens. A 30-minute FISH-based assay for pathogens was developed. All steps proceed in a single assay mix and the method requires no wash steps. Concerted cell-permeabilization, hybridization/labeling, and binding to magnetic particles is followed by magnetic selection and imaging. Methods use inexpensive and readily available reagents, such as unmodified magnetic particles, oligonucleotides, and detergents.
Labeling methods were developed that greatly improved FISH signal intensity by optimizing probe design permeabilization and hybridization conditions. Methods use multiple labeled target-specific oligonucleotide probes and helper probes that help unwind rRNA secondary structure. Detection of 8 target CAUTI pathogens at a required threshold (10,000 CFU/ml) was demonstrated in just 30 minutes directly in in urine specimens (Escherichia coli, Enterococcus faecalis, Klebsiella pneumoniae, Pseudomonas aeruginosa, Proteus mirabilis, Staphylococcus aureus, Streptococcus agalactiae, and Enterobacter aerogenes). An approach for enumerating the total bacterial count in urine specimens using a DNA stain was developed, which is valuable for ruling out infection and identifying polymicrobial specimens and asymptomatic UTI.
Limit-of-detection for pathogens in urine.
The FISH assay must be inclusive, that is it must detect any strain of a given target pathogen.
The CAUTI ID results in
Overview: The following example demonstrates that very low concentrations of cells can be detected using the novel isothermal fluorescence in situ hybridization method. The limit of detection for three common human urinary tract infection (UTI) pathogens are shown.
Experimental MethodsBacterial cell preparation: Bacterial cultures for E. coli ATCC 19138, K. pneumoniae ATCC 700603 and P. aeruginosa ATCC 9721 were obtained by inoculating Trypticase Soy Broth (TSB, Hardy Diagnostics cat. U65) with 3 to 5 colonies from fresh tryptic soy agar plates (TSA, BD cat. 221185) and growing for 1.5 to 3 hours at 35° C. to achieve log-phase growth. After the cells had reached an optical density reading at 600 nm of 0.15-0.30, cells were placed on ice for at least 15 minutes before dilution. After cooling, the cells were diluted in 1× cation-adjusted Mueller-Hinton broth (MHBII, Teknova cat. M5860) to the concentrations to be assayed (approximately 19200, 9600, 4800, 2400, 1200, 600, 300, and 150 colony-forming units (CFU)/reaction). For more accurate cellular concentrations, these estimated bacterial inputs were adjusted using colony counts. Plate counts were determined by diluting the log-phase cultures to about 500 CFU/mL in MHBII, plating 100 μL on TSA plates and counting colonies after growth at 35° C. for 16 to 24 hours. Using the average plate counts, the actual CFU present in each concentration tested was computed.
Preparation of Magnetic Particles: Polyaspartic acid-conjugated magnetic particles (Fluidmag-PAA, Chemicell, cat. 4108) and carboxyl-coated magnetic particles with high iron (Carboxyl Magnetic Particles, Spherotech, cat. CM-025-10H) were used to non-specifically capture bacterial cells. Each particle was diluted 1:40 into 50 mM Epps buffer, pH 8.2, with final concentrations of approximately 1.38×109 particles per reaction for the polyaspartic acid particles and 3.46×109 particles per reaction for the carboxyl particles. Fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. These particles enable the optical system to focus on the correct plane. The magnetic particle mixture was sonicated for 1 minute immediately prior to use to minimize aggregation.
Preparation of FISH probes: Two species-specific DNA oligomer sets for E. coli and K. pnuemoniae and one for P. aeruginosa was heated in a water bath between 80-85° C. for 10 minutes and then placed on ice to reduce aggregation. A DNA oligomer set contained a species-specific DNA oligonucleotide labeled with a fluorescent dye (Alexa647N, Thermo Fischer) on either the 5′ end, or on both the 5′ and 3′ ends of the oligonucleotide, and 2-6 helper oligonucleotides that bind adjacent or near the specific probe and are designed to disrupt the local secondary structure of the ribosomal subunit, and allow the labeled, specific probe greater hybridization efficiency to the target rRNA. Probe sequences used in this example are shown in Table A in
Preparation of the dried hybridization buffer plates: A mixture of 10×SSC (1.5M NaCl, 0.15M Sodium citrate, Sigma, cat. S6639), 2.6% w/v CHAPSO (Sigma cat. C3649), 2.4% w/v SB3-12 (Sigma cat. D0431), 0.43M Guanidine thiocyanate (Sigma cat. G9277) and 0.6% w/v Cetrimide (Sigma cat. M7365) was prepared. 30 uL of this mixture was added to each well of a 96 well plate. The plates were placed into a convection oven at 50° C. and allowed to dry overnight. When 100 uL of liquid is added to these wells, the correct hybridization buffer concentrations of 3×SSC (0.45M NaCl, 0.045M Sodium citrate), 0.77% w/v CHAPSO (Sigma cat. C3649), 0.72% w/v SB3-12 (Sigma cat. D0431), 0.13M Guanidine thiocyanate (Sigma cat. G9277) and 0.18% w/v Cetrimide (Sigma cat. M7365) are achieved.
Limit of Detection (LoD) Assay procedure: A mixture of DNA oligonucleotide sets appropriate for the bacteria of interest was combined with urine and a concentrated cation-adjusted Mueller Hinton Stock (MHBII) to make a final solution containing 1×MHBII and 30% pooled human urine (Innovative Research, cat. IRHUURE500ML). Probe concentrations varied between different bacterial species but ranged from 0.2-0.6 μM for the labeled oligonucleotide and 1.5-6 μM for the corresponding helper probes. 90 uL of this mixture was placed into the appropriate dried hybridization buffer plate. 10 uL of the magnetic particle mixture was added, followed by 10 uL of the appropriate cell dilution. Twelve replicates of each cell concentration and 24 replicates of the blank (media containing no bacteria) were assessed for each target bacteria tested. 100 μL of the final reaction mixture was transferred to a microtiter plate containing 50 μL per well (previously dried) of “dye-cushion” (50 mM TRIS pH 7.5 (Sigma cat. T1075), 7.5% v/v Optiprep (Sigma cat. D1556), 50 mg/mL Direct Black 19 (Orient cat. 191L) and incubated at 35° C. for 30 minutes to allow for the simultaneous rehydration of the “dye-cushion”, labeling of bacterial cells, and binding of magnetic particles to bacterial cell surfaces. After incubation, microtiter plates were placed onto a magnetic field (Dexter magnetic technologies, cat. 54170260) for 4 minutes to bring magnetic particles, a fraction containing labelled cells, through the “dye-cushion” and into proximity to the imaging surface at the bottom of the wells.
Imaging of labeled cells: The MultiPath™ laboratory imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a microtiter plate. It uses a high precision linear stage from Prior Scientific (Rockland, Mass.) to position each well over a fluorescence-based image acquisition subsystem. The instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire microtiter plate well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 680/40 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Data analysis: At each cell concentration, the number of fluorescent objects detected was determined. The data from all eight cell concentrations was fit to a linear regression line, and the slope, intercept and standard deviation of the lowest 3 cell inputs was used to determine the limit of the blank (LoB) and limit of detection (LoD) for each bacterium tested.
Results:
All three bacteria tested showed low limits of detection. The figures show the data generated for E. coli, K. pneumoniae, and P. aeruginosa with the linear fit used to calculate the LoB and LoD. The LoB and LoD are indicated in CFUs detectable in a single reaction well.
Conclusions. The novel and rapid FISH method described in this example is shown to be a sensitive method with limits of detection of about 500 CFU or less per reaction, using minimally processed urine matrix.
Variations. This example is illustrative of the performance of this novel FISH method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations), concentration of urine and urine processing procedures. This methodology can also clearly be extended to other biological specimens and to other bacterial and non-bacterial pathogens.
Overview. This example demonstrates the use of the invention to detect different strains for a targeted bacterial species. Raw data for 11 different E. coli strains are presented and data for K. pneumoniae, P. aeruginosa, P. mirabilis and Enterococcus spp. are summarized. Bacterial cell targets were labeled in 30 minutes using isothermal fluorescence in situ hybridization (FISH) and detected on the MultiPath™ CCD-camera-based detection system.
Experimental MethodsBacterial cell preparation: Bacterial cultures for different strains were obtained by inoculating Trypticase Soy Broth (TSB, Hardy Diagnostics Cat. U65) with 3 to 5 colonies from fresh tryptic soy agar plates (TSA, BD cat. 221185) and growing for 1.5 to 3 hours at 35° C. to achieve log-phase growth. Using optical density at 600 nm to estimate cell concentration, cells were diluted to approximately 600 CFU and 3000 CFU per reaction in 1× cation-adjusted Mueller-Hinton broth (MHBII, Teknova cat. M5860). For more accurate percent cellular detection calculations, these estimated bacterial inputs were adjusted using colony counts. Plate counts were determined by diluting the log-phase cultures to about 500 CFU/mL in MHBII, plating 100 μL on TSA plates and counting colonies after growth at 35° C. for 16 to 24 hours.
Preparation of Magnetic Particles: Polyaspartic acid-conjugated magnetic particles used to non-specifically capture bacterial cells (Fluidmag-PAA, Chemicell, cat. 4108) were diluted 1:20 into 50 mM EPPS buffer, pH 8.2 to 2.75×1012 particles/mL. Fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. These particles enable the optical system to focus on the correct plane. The magnetic particle mixture was sonicated for 1 minute immediately prior to use to minimize aggregation. Separate magnetic particle suspensions were prepared for the time zero and time four-hour assays described below.
Labeling of Bacterial Cells: 100 μL labeling reactions were prepared by combining diluted cells, isothermal hybridization buffer (0.9×MHBII, 3×SSC (1.5M NaCl, 0.15M Sodium citrate, Sigma, cat. S6639), 0.77% w/v CHAPSO (Sigma cat. C3649), 0.72% w/v SB3-12 (Sigma cat. D0431), 0.13M Guanidine thiocyanate (Sigma cat. G9277), 0.18% w/v Cetrimide (Sigma cat. M7365)), species-specific Alexa647N-labelled DNA or LNA-containing DNA probes (Integrated DNA Technologies, IDT) targeted to the 16S or 23S bacterial rRNA, helper probes to facilitate effective hybridization (IDT) and 30 μL of pooled human urine (Innovative Research, cat. IRHUURE500ML). Probe sequences are shown in the Table in
The urine was first processed through a Zeba 7K MWCO spin column (Thermo Fisher, Cat. 89893 or 89892 depending on urine volume) according to the manufacturer's instructions. 10 μL of the magnetic particle preparation was then added to this mixture. The final reaction mixture was transferred to a microtiter plate containing 50 μL (previously dried) “dye cushion” (50 mM TRIS pH 7.5 (Teknova cat. T1075), 7.5% v/v Optiprep (Sigma cat. D1556), 50 mg/mL Direct Black 19 (Orient cat. 191L) incubated at 35° C. for 30 minutes to allow for the simultaneous rehydration of the “dye cushion”, labeling of bacterial cells, and binding of magnetic particles to bacterial cell surfaces. After incubation, microtiter plates were placed onto a magnetic field (Dexter magnetic technologies, cat. 54170260) for 4 minutes to bring magnetic particles, a fraction containing labelled cells, through the “dye cushion” and into proximity to the imaging surface at the bottom of the wells.
Imaging of labeled cells: Labeled bacterial cells on the MultiPath laboratory imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a microtiter plate. It uses a high precision linear stage from Prior Scientific (Rockland, Mass.) to position each well over a fluorescence-based image acquisition subsystem. The instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire microtiter plate well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 680/40 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Data analysis: For each bacterium, the number of fluorescent objects was determined (assay signal). A bacterial strain was considered detected if signal was detected above three standard deviations of the signal in the no cell condition.
Results.
The table in
Conclusions. The novel FISH method described in this example detected all strains tested for 5 different bacterial species that are among the major pathogens leading to clinical symptoms in patients with UTI.
Variations. This example is illustrative of the performance of this novel FISH method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations), concentration of urine and urine processing procedures. This methodology can also clearly be extended to other biological specimens and to other bacterial and non-bacterial pathogens.
Overview. This example demonstrates that the novel isothermal FISH method specifically detects a target bacterium while not detecting related non-target bacteria, even at very high concentrations. This example presents assay conditions that specifically detect E. coli yet do not detect 16 other bacteria that also cause urinary tract infections (UTI), have similar rRNA sequences or are commensal organisms.
Experimental MethodsBacterial cell preparation: Bacterial cultures for 16 off-target bacteria (listed in Table 1) and E. coli strain ATCC 25922 were grown from a single colony selected from a fresh tryptic soy agar plates (TSA, BD cat. 221185), inoculated into Trypticase Soy Broth (TSB, Hardy Diagnostics cat. U65) and grown with shaking overnight at 35° C. 50-80 μL of the overnight culture was added into fresh TSB and grown for 1.5-2 hours, until the optical density at 600 nm reached 0.15-0.3. Each bacterium was then diluted to approximately 1×108 cells per mL in cation-adjusted Mueller Hinton (MHBII, Teknova cat. M5860).
Selection of bacterial targets to evaluate: Bacterial pathogens to test for specificity were selected for their rRNA sequence similarity to the target bacteria's rRNA sequence or because they are pathogens that are commonly found in urinary tract infections (the disease target) and therefore, cross-reactivity to these organisms would be most problematic. The table in
Preparation of FISH probes: A DNA probe set for E. coli was heated in a water bath between 80-85° C. for 10 minutes and then placed on ice to reduce aggregation. This DNA probe set is shown in Table in
Preparation of Magnetic Particles: Polyaspartic acid-conjugated magnetic particles used to non-specifically capture bacterial cells (Fluidmag-PAA, Chemicell, cat. 4108) were diluted 1:20 into 50 mM EPPS buffer, pH 8.2 to 2.75×1012 particles/mL. Fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. These particles enable the optical system to focus on the correct plane. The magnetic particle mixture was sonicated for 1 minute immediately prior to use to minimize aggregation.
Labeling of Bacterial Cells: 100 μL labeling reactions were prepared by combining diluted cells, isothermal hybridization buffer (0.9×MHBII (Teknova cat. M5860), 3×SSC (0.45M NaCl, 0.045M Sodium citrate, Sigma, cat. cat. S6639), 0.77% w/v CHAPSO (Sigma cat. C3649), 0.72% w/v SB3-12 (Sigma cat. D0431), 0.13M Guanidine thiocyanate (Sigma cat. G9277), 0.18% w/v Cetrimide (Sigma cat. M7365)), species-specific Alexa647N-labelled probes (Integrated DNA Technologies, IDT) targeted to the 16S or 23S bacterial rRNA, helper probes to facilitate effective hybridization (IDT) and 30 μL of pooled human urine (Innovative Research, cat. IRHUURE500ML). The specific probe sets tested are shown in Table in
Imaging of labeled cells: The MultiPath™ laboratory imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a microtiter plate. It uses a high precision linear stage from Prior Scientific (Rockland, Mass.) to position each well over a fluorescence-based image acquisition subsystem. The instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire microtiter plate well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 680/40 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Data analysis: For each bacterium, the number of fluorescent objects was determined (assay signal). A bacterium was considered cross-reactive if signal was detected within three standard deviations of the signal in the blank (no bacteria added).
Results.
The images show the rapid novel FISH method only detects E. coli and not the other 16 different challenge bacteria, and each show that very high concentrations (1×106 cells per reaction) of 8 clinically relevant challenge bacteria are not detected under the same assay conditions that generate high assay signal for the E. coli targeted bacteria. The two bars represent two different probe sets designed to be specific for E. coli (see Table in
Conclusions. The novel rapid FISH method described in this example specifically, by design, detects E. coli but does not detect 16 clinically relevant potential cross-reactive bacteria. This demonstrates the method has high specificity for the identification of a target UTI pathogen which is of critical importance for the clinical treatment of the infection.
Variations. This example is illustrative of the performance of this novel FISH method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations), concentration of urine and urine processing procedures. This methodology can also clearly be extended to other biological specimens and to other bacterial and non-bacterial pathogens. Assays have also been designed that demonstrate high specificity for K. pneumoniae, K. oxytoca, P. aeruginosa, P. mirabilis and E. faecalis.
Overview. This example demonstrates the use of the invention to simultaneously detect, in a single reaction, E. coli, K. pneumoniae, P. aeruginosa, and K. oxytoca using fluorescently labeled probes specific for each bacteria's rRNA. Each pathogen was specifically detected in the mixture through the use of 4 distinct fluorophores—one for each bacterial species—that have different excitation/emission spectral properties.
Experimental Method. Bacterial cell growth: Bacterial cultures for Escherichia coli ATCC 25922, Klebsiella pneumoniae ATCC 13883, Pseudomonas aeruginosa ATCC 27853, and Klebsiella oxytoca ATCC 8724 were obtained by inoculating Trypticase Soy Broth (TSB, Hardy Diagnostics cat. U65) with 3 to 5 colonies from fresh tryptic soy agar plates (TSA, BD cat. 221185) and growing for 1.5 to 3 hours at 37° C. to achieve log-phase growth. Each culture was then diluted in cation-adjusted Mueller-Hinton Broth (MHBII, Teknova, cat. M5860) to an optical density at 600 nm of 0.15, which is approximately 1.0×108 colony-forming units (CFUs) per mL.
Preparation of Magnetic Particles: Polyaspartic acid-conjugated magnetic particles (Fluidmag-PAA, Chemicell, cat. 4108) and carboxyl-coated magnetic particles (Carboxyl Magnetic Particles, Spherotech, cat. CM-025-10H) were used to non-specifically capture bacterial cells. Each particle was diluted 1:40 into 50 mM Epps buffer, pH 8.2, with final concentrations of approximately 1.38×109 particles per reaction for the polyaspartic acid particles and 3.46×109 for the carboxyl particles.
Labeling of Bacterial Cells: 100 μL labeling reactions were prepared by combining diluted cells of all four bacteria, isothermal hybridization buffer (0.9×MHBII, 3×SSC (1.5M NaCl, 0.15M Sodium citrate, Sigma, cat. S6639), 0.77% w/v CHAPSO (Sigma cat. C3649), 0.72% w/v SB3-12 (Sigma cat. D0431), 0.13M Guanidine thiocyanate (Sigma cat. G9277), 0.18% w/v Cetrimide (Sigma cat. M7365)), species-specific DNA probes (Integrated DNA Technologies, IDT) targeted to the 16S or 23S bacterial rRNA, helper probes to facilitate effective hybridization (IDT) and 30 μL of pooled human urine (Innovative Research, cat. IRHUURE500ML). 10 μL of the magnetic particle preparation was then added to this mixture. Probe sequences and the location of their dye modifications are shown in Table in
The cells/hybridization mixture (1 mL) was transferred into the cartridge. The cartridge was placed onto the analyzer (as described below) which automated the remaining assay steps and image acquisition and analysis. Briefly, the fluidic system of the analyzer moved the reaction mixture into the optical window containing 46 μL per well (previously dried) of “dye-cushion” (50 mM TRIS pH 7.5 (Sigma cat. T1075), 7.5% v/v Optiprep (Sigma cat. D1556), 50 mg/mL Direct Black 19 (Orient cat. 191L). The cartridge was incubated within the analyzer at 35° C. for 30 minutes. Following this incubation, the cartridge was moved for 4 minutes onto the magnet station (Dexter magnetic technologies, cat. 54170260) to bring magnetic particles, a fraction containing labeled cells, through the rehydrated “dye-cushion” and into proximity to the imaging surface at the bottom of the wells. After the magnet station, the cartridge was moved to the imaging station within the analyzer and a series of images taken in each of the four color channels (red (excitation 635/25 nm, emission 680/40 nm), yellow (excitation 530/20 nm, emission 572/23 nm), green (excitation 470/40 nm, emission 520/40 nm), orange (excitation 569/25 nm, emission 609/34 nm)).
Imaging of labeled cells: The MultiPath Analyzer imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a MultiPath Cartridge as part of a fully automated test. It uses a custom designed precision 3 axis positioning system to locate each well over a fluorescence-based image acquisition subsystem. The Analyzer can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire Cartridge Imaging Well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. For the red channel, 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 680/40 nm emission filters. For the orange channel, 24 frames were captured at a 100 msec exposure using 569/25 nm excitation and 609/34 nm emission filters. For the Yellow channel, 48 frames were captured at a 100 msec exposure using 530/20 nm excitation and 572/23 nm emission filters. For the Green channel, 32 frames were captured at a 100 msec exposure using 470/40 nm excitation and 520/40 nm emission filters. The focusing plane for imaging the labeled cells was determined experimentally in this example.
Results.
Conclusions. This method allows for a single rapid FISH method to simultaneous detect and quantify four different bacteria in a single well of a cartridge.
Variations:
This example is illustrative of the multiplex capability of this novel FISH method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.) and alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations). This methodology can also clearly be extended to other biological specimens and to other bacterial and non-bacterial pathogens for which specific probes can be designed.
Overview. The presence and number of polymorphonuclear neutrophils (PMNs) can be diagnostically informative for detecting infections. For example, a low number of neutrophils in urine samples helps rule out urinary tract infections. In this example, non-magnified digital imaging was used to enumerate polymorphonuclear neutrophil (PMN) targets that were stained with fluorescent labeled antibodies. In this example, we have used anti-CD15 and anti-CD16 antibodies which is specific molecules present on the surface of polymorphonuclear neutrophils (PMNs) for capture and detection. In one embodiment, anti-CD15 antibodies were conjugated to magnetic particles which were used for PMN capture and fluorescently labeled anti-CD16 antibody for detection using non-magnified digital imaging.
Experimental MethodsBlood Samples
Fresh blood sample from healthy donor were obtained from Research Blood Components (Boston, Mass.) and used as the source of PMN.
Preparation of Magnetic Particles:
Antibody conjugated magnetic particles were made by coupling magnetic particles (Ademtec, 292 nm) to mouse anti-CD15 antibodies (Biolegend) using standard EDAC (1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide) coupling chemistry.
Detection Antibody
Alexa Fluor® 488 anti-human CD16 Antibody (Biolegend) was used as detection antibody.
Labeling and Capture of PMNs:
The assays were carried out in a 96-well microtiter plate. Each well included 38 ul of Phosphate-buffered saline (PBS), 2 uL of fresh blood sample or 2 uL of PBS (no PMN control), 5 uL of Alexa-488 labeled anti CD-16 antibody (1 ug) and 5 ul of antibody-conjugated magnetic particles (2e10/mL). The reactions were incubated at room temperature for 15 min. After the incubation, 40 uL of reaction mixture was carefully overlaid on 75 ul of “dye cushion” (15% Optiprep with 5 mg/mL Chromotrope 2R) which was pre-aliquoted in black, clear bottom half area microtiter plate (Greiner 675096, VWR part #82050-056). The microtiter plate was placed onto a magnetic field for 4 minutes to bring magnetic particles, a fraction containing labelled cells, through the “dye cushion” and into proximity to the imaging surface at the bottom of the wells.
Imaging of PMNs:
Following magnetic capture of labeled PMNs:magnetic particle complexes, the microtiter plate was placed on a stage above a CCD digital camera (IDS, model UI-2250SE-M) and illuminated with light from LEDs passing through an optical filter (469 nm, 35 nm FWHM). Fluorescent signal passing through an emission filter (520-35 nm) was detected by the camera to create an image of the fluorescent complexes. The images were analyzed using FLimage software (First Light Biosciences) that enumerates the individual cells.
Results.
Conclusions. This large area, non-magnified imaging system is capable of detecting and enumerating PMNs from blood sample that are labeled with fluorescently labeled antibodies against the cell-surface marker. The results show the potential of the inventive systems and methods for enumerating diagnostically informative human or host cells.
Variations. Using other cell-specific antibodies, different cells can be detection in various biological samples. For example, other cell-surface marker antibodies can recognize different diagnostically informative cells. For example, quantifying squamous epithelial cells is important for assessing respiratory sample quality in pneumonia diagnostics. Multiple antibodies/cell surface markers can be used and labeled cells can be distinguished by using multiple excitation and emission wavelengths for fluorescence detection. The spectrum of fluorescence associated with an object can be used to determine whether a signal of specific type of cell or not.
Example 6. Automated Rapid AST of E. coli in Clinical Urine Specimens in a Cartridge on an InstrumentOverview: This example demonstrates the use of the systems, devices, and methods of invention to determine the antimicrobial susceptibility of a targeted bacterial pathogen (E. coli in this example) in urine in 4 hours without requiring cell purification. The example using a concerted FISH method for labeling and magnetic selection and quantifies specific target cells after differential growth using non-magnified digital imaging. This new method has comparable performance to the gold standard CLSI broth microdilution (BMD) method.
Experimental MethodsUrine Specimens: 48 remnant de-identified urine specimens collected from patients with a urinary tract infection (UTI) and known to contain E. coli were received from Dr. Kirby's lab at Beth Israel Hospital (Boston, Mass.). Samples were received 1-5 days post collection and contained a urine preservative to limit loss of cell viability. For each sample, color of urine, pH, and presence of particulates were noted. Upon receipt, conventional urine culture was performed to determine the approximate CFU/mL of bacteria present, and to confirm single or mixed bacterial morphology as reported by Dr. Kirby's lab. Briefly, a calibrated 1 μL loop was placed into a well-mixed urine sample and the 1 μL was evenly spread over a Tryptic soy agar (TSA, BD cat. 221185) plate and incubated in a 35° C. incubator for 18-24 hours. The remainder of the urine samples were processed and assayed as described below.
Urine Processing: Prior to testing, urine preservative and other potentially interfering compounds were removed using size exclusion chromatography. 2.5 mL of each clinically positive urine sample was applied to a pre-washed Zeba™ 7K MWCO spin column (ThermoFisher, cat. #89893) and centrifuged according to the manufacturer's instructions. Urine culture was repeated on this processed sample as described above, to examine bacterial loss following processing.
Preparation of Magnetic Particles: Polyaspartic acid-conjugated magnetic particles used to non-specifically capture bacterial cells (Fluidmag-PAA, Chemicell, cat. 4108) were diluted 1:20 into 50 mM EPPS buffer, pH 8.2 to 2.75×1012 particles/mL. Fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. These particles enable the optical system to focus on the correct plane. The magnetic particle mixture was sonicated for 1 minute immediately prior to use to minimize aggregation. Separate magnetic particle suspensions were prepared for the time zero and time four-hour assays described below.
Bacterial Cell Labeling at AST Time Zero: Assay signal at time zero (TO) prior to the initiation of bacterial growth in the presence or absence of antibiotics was determined for each clinical urine specimen. 30 μL of each processed urine was added to 70 μL of 1X cation-adjusted Mueller-Hinton Broth (MHBII) containing species-specific Alexa647N-labeled DNA oligonucleotide FISH probes and unlabeled DNA helper probes. Probe sequences used are shown in Table A. The 100 μL mixture was then added to a well of a microtiter plate containing dehydrated hybridization buffer (3×SSC (0.45 M NaCl, 0.045 M Na citrate) buffer (Sigma, cat. #S6639), 0.18% cetrimide (Sigma, cat. #H9151), 0.77% CHAPSO (Sigma cat. #C3649), 0.72% SB3-12 (Sigma cat. #D0431) 0.13M guanidine thiocyanate (Sigma, cat. #G9277)). 10 μL of the prepared magnetic particle mixture was then added to the well. 100 μL of this reaction mixture was transferred to a microtiter plate containing 50 μL per well (previously dried) of “dye-cushion” (50 mM TRIS pH 7.5 (Sigma cat. T1075), 7.5% v/v Optiprep (Sigma cat. D1556), 50 mg/mL Direct Black 19 (Orient cat. 191L) and incubated at 35° C. for 30 minutes. After incubation, microtiter plates were placed onto a magnetic field (Dexter magnetic technologies, cat. 54170260) for 4 minutes to bring magnetic particles, a fraction containing labeled cells, through the “dye-cushion” and into proximity to the imaging surface at the bottom of the wells.
Imaging of labeled cells: The MultiPath™ laboratory imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a microtiter plate. It uses a high precision linear stage from Prior Scientific (Rockland, Mass.) to position each well over a fluorescence-based image acquisition subsystem. The instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire microtiter plate well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 667/30 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Preparation of Antibiotic Plates: Microtiter plates containing six concentrations of each antibiotic in a 2-fold serial dilution series were prepared, starting at a 10-fold higher concentration than the expected minimum inhibitory concentration (MIC). Antibiotics used were Cefazolin, Ciprofloxacin, Nitrofurantoin, and Trimethoprim-Sulfamethoxazole. Antibiotic dilutions were verified to be within the appropriate tolerance by confirming that the MIC for at least two CLSI QC strains fell within the QC range reported in CLSI document M100Ed29E-2019. The concentrations selected for testing of each antibiotic straddled the CLSI-reported breakpoints for the antibiotic for E. coli. In addition to the wells containing the antimicrobial dilution series, eight wells containing water or diluent were included in the plates to allow for a no antibiotic positive and negative growth control.
Four Hour Growth: While the time zero cell quantification was occurring, 32.4 μL of processed clinical urine and 75.6 μL of 1.43×MHB II (Teknova, cat. #M5860) was added to each well of the antibiotic plate (already containing 12 μL of antibiotics). The samples were allowed to grow in a standard incubator at 35° C. for 4 hours.
Bacterial Cell Labeling at AST time four-hour growth: After samples had incubated in the presence and absence of antibiotics for four hours (T4), cells were labeled and quantified to determine how much growth, if any, occurred. 100 μL of each well of the incubated sample-antibiotic plate was transferred to a corresponding well of a dehydrated buffer plate and combined with FISH probes, helper probes, magnetic particles, and focus particles in the same manner as described above for assay time zero.
Comparison Method: Results for the MulitPath™ Assay were compared to broth microdilutions (BMD) performed according to the CLSI method M07-Ed13E 2018.
Data Analysis and Threshold Generation: Using the image captured by the CCD camera, detected cells were estimated by an algorithm that looked at both number of objects in the field of view and the intensity of the objects. Number of cells based on this detection algorithm were generated at time zero, and at time four hours without antibiotic and with all 6 concentrations of each antibiotic. For each urine sample/drug concentration, fold growth was calculated as the signal in the well containing antibiotic after growth (time four) to the signal in the urine sample prior to growth (time zero). Using fold growth and the observation of growth in the corresponding well in the CLSI-compliant broth microdilution, a logistic regression model was used to generate thresholds for determining the fold growth cutoff above which cells are growing in the presence of the antibiotic (and thus, resistant at that concentration) and below which, cells are in the process of dying (and thus, sensitive at that concentration). The point where the fold growth number falls below the determined threshold is the MIC value generated by the assay. Results were correspondingly assigned to categories of susceptible, intermediate, or resistant to each antibiotic. All data was then compared to CLSI standard BMD. Four-hour growth in the absence of antibiotic is a control condition to ensure viable bacterial are present in the processed urine sample.
Results.
Conclusions. This novel method shows that accurate AST results (MIC determinations) may be made with only 4 hours of differential growth of minimally processed urine clinical specimens, notably without lengthy colony purification steps. The AST results, whether reported as MIC categorical antibiotic susceptibility results, compare favorably to the gold standard, broth microdilution method.
Variations. This example is illustrative of the performance of this novel AST method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations, etc.), concentration of urine and urine processing procedures. This methodology can also clearly be extended to other antibiotics, biological specimens and to other bacterial and non-bacterial pathogens.
Overview. This example demonstrates the use of the invention to accurately determine the antimicrobial susceptibility of pathogens with known antibiotic susceptibility profiles added into bacteria-free urine. Differential growth in microbiological media containing antimicrobial agents followed by assessment of growth using the inventive concerted FISH method for target specific cell quantification required just 4.5 hours. This new method has comparable performance to the gold standard CLSI broth microdilution (BMD) method.
Experimental MethodsBacterial cell preparation: 50 bacterial strains with known resistance profiles were collected from either the ATCC or from the CDC antibiotic resistance bank (AR bank) and are shown in Table A. Bacterial cultures for each of these were obtained by inoculating Trypticase Soy Broth (TSB, Hardy Diagnostics cat. U65) with 3 to 5 colonies from fresh tryptic soy agar plates (TSA, BD cat. 221185) and growing for 1.5 to 3 hours at 35° C. to achieve log-phase growth. Using optical density at 600 nm to estimate cell concentration, each culture was diluted to approximately 5×106 colony-forming units (CFU)/mL in cation-adjusted Mueller Hinton II (MHBII, Teknova cat. M5860).
Urine Processing: Prior to testing, pooled human urine (Innovative Research, cat. IRHUURE500ML) was applied to a pre-washed Zeba™ 7K MWCO spin column in a ratio of 4 mL urine to one pre-washed 10 mL spin column (ThermoFisher, cat. #89893) and centrifuged according to the manufacturer's instructions.
Preparation of Magnetic Particles: Polyaspartic acid-conjugated magnetic particles used to non-specifically capture bacterial cells (Fluidmag-PAA, Chemicell, cat. 4108) were diluted 1:20 into 50 mM EPPS buffer, pH 8.2 to 2.75×1012 particles/mL. Fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. These particles enable the optical system to focus on the correct plane. The magnetic particle mixture was sonicated for 1 minute immediately prior to use to minimize aggregation. Separate magnetic particle suspensions were prepared for the time zero and time four-hour assays described below.
Bacterial Cell Labeling at AST Time Zero: Assay signal at time zero (TO) prior to the initiation of bacterial growth in the presence or absence of antibiotics was determined for each bacterium. A reaction mixture was prepared consisting of 30 μL processed urine, 10 μL of the 5×106 CFU/mL bacterial dilution, 60 μL MHBII (1× final concentration in 100 μL) and the appropriate species-specific Alexa647N-labeled DNA oligonucleotide FISH probe and its associated unlabeled DNA helper probes for the target bacterial species. Probe sequences used are shown in Table in
Imaging of labeled cells: The MultiPath laboratory imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a microtiter plate. It uses a high precision linear stage from Prior Scientific (Rockland, Mass.) to position each well over a fluorescence-based image acquisition subsystem. The instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire microtiter plate well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 667/30 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Preparation of Antibiotic Plates: Microtiter plates containing six concentrations of each antibiotic in a 2-fold serial dilution series were prepared. The 2-fold dilution series was prepared at a 10-fold higher concentration than the desired concentration in the final broth microdilution, such that addition cells/urine/media mixture would yield the correct antibiotic range. 12 uL of each antibiotic dilution was then aliquoted into the appropriate wells of a 96 well plate. Different antibiotics were tested for different bacteria. Antibiotic dilutions were verified to be within the appropriate tolerance by confirming that the MIC for at least two CLSI QC strains fell within the QC range reported in CLSI document M100Ed29E-2019. The concentrations selected for testing of each antibiotic straddled the CLSI-reported breakpoints for the antibiotic for the appropriate bacterial species such that categorical determinations (sensitive/intermediate/resistant) could be made from this data. In addition to the wells containing the antimicrobial dilution series, several wells containing water or other diluent were included for a no antibiotic positive growth and negative growth (no cell) control. Antibiotic plates were frozen at −80° C. and thawed completely before use.
Four Hour Growth: While the time zero cell quantification was occurring, 12 μL of prepared bacterial culture, 36 uL pooled human urine processed as done for the assay time zero, 60 uL of 2×MHB II (Teknova, cat. #M5860) and 2 uL water was added to each well of the prepared antibiotic plate. The samples were allowed to grow in a standard incubator at 35° C. for 4 hours.
Bacterial Cell Labeling at AST time four-hour growth: After samples had incubated in the presence and absence of antibiotics for four hours (T4), cells were labeled and quantified to determine how much growth, if any, occurred. 100 μL of each well of the incubated sample-antibiotic plate was transferred to a corresponding well of a dehydrated buffer plate and combined with FISH probes, helper probes, magnetic particles, and focus particles in the same manner as described above for assay time zero.
Comparison Method: Results for the MulitPath™ Assay were compared to broth microdilutions (BMD) performed according to the CLSI method M07-Ed13E 2018.
Data Analysis and Threshold Generation: Using the image captured by the CCD camera, detected cells were estimated by an algorithm that looked at both number of objects in the field of view and the intensity of the objects. Number of cells based on this detection algorithm were generated at time zero, and at time four hours without antibiotic and with all concentrations of each antibiotic. For each bacteria sample/drug concentration, fold growth was calculated as the signal in the well containing antibiotic after growth (time four) to the signal in the urine sample prior to growth (time zero). Using fold growth and the observation of growth in the corresponding well in the CLSI-compliant broth microdilution, a logistic regression model was used to generate thresholds for determining the fold growth cutoff above which cells are growing in the presence of the antibiotic (and thus, resistant at that concentration) and below which, cells are in the process of dying (and thus, sensitive at that concentration). The point where the fold growth number falls below the determined threshold is the MIC value generated by the assay. Results were correspondingly assigned to categories of susceptible, intermediate, or resistant to each antibiotic. Results were then compared to the MIC values and categorical calls reported by ATCC or the CDC. Four-hour growth in the absence of antibiotic is a control condition to ensure viable bacterium were added to each sample or for use when calculating fold inhibition.
Results.
The table in
Conclusions. This novel method shows that MIC determinations that match the published values for highly characterized strains of bacteria with multiple drug resistance mechanism may be made with only 4 hours of growth in the context of sample matrix.
Variations. This example is illustrative of the performance of this novel AST method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations, etc.), concentration of urine and urine processing procedures. This methodology can also clearly be extended to other antibiotics, biological specimens and to other bacteria for which specific probes can be designed.
Overview. This example demonstrates the use of the systems and methods of the invention to automatically determine AST results for a pathogen in a clinical urine sample in 4 hours without requiring lengthy cell purification steps. The automated instrument performs the steps required in the reagent-containing cartridge to determine antimicrobial susceptibility at a constant physiological temperature. The temperature is compatible with both microbial growth and the inventive method for detecting and quantifying target cells. The latter method is performed on the inventive system using FISH-based labeling, magnetic selection, and non-magnified digital imaging.
The instrument's pneumatics subsystem is used to automatically distribute the specimen in the cartridge into portions or aliquots containing various antimicrobial agents in various concentrations plus microbiological medium. One of the portions is used to quantify the pathogen cells before growth incubation. The system incubates the cartridge for 4 hours and then quantifies the number of target cells in the wells containing antimicrobial agents. Comparison of the number of cells in the incubated portions containing antimicrobial agents to the number of cells measured before incubation is used to determine the antimicrobial susceptibility of the pathogen in the various antibiotics.
The example shows the results using the inventive automated systems, devices, and methods for rapid and automated antimicrobial susceptibility testing directly in clinical specimens from hospital patients that had E. coli in their urine. The invention delivered in just 4 hours accurate performance compared to the gold standard CLSI broth microdilution (BMD) method.
Experimental MethodsUrine Specimens: Remnant de-identified urine specimens collected from patients with a urinary tract infection (UTI) and known to contain E. coli were received from Dr. Kirby's lab at Beth Israel Hospital (Boston, Mass.). Samples were received 1-5 days post collection and contained a urine preservative to limit loss of cell viability. For each sample, color of urine, pH, and presence of particulates were noted. Upon receipt, conventional urine culture was performed to determine the approximate CFU/mL of bacteria present, and to confirm single or mixed bacterial morphology as reported by Dr. Kirby's lab. Briefly, a calibrated 1 μL loop was placed into a well-mixed urine sample and the 1 μL was evenly spread over a Tryptic soy agar (TSA) plate and incubated in a 35° C. incubator for 18-24 hours. The remainder of the urine samples were processed and assayed as described below.
Preparation of the AST cartridge—Media and Antimicrobials Days prior the cartridge was prepared by distributing 25 uL of 4×MHB II (Teknova, Cat. #101320-356) into each of the 8 individual growth wells (see the figure for a diagram of the cartridge) Growth wells 1 and 2 are for the time zero measurement (see description below), so only growth media is contained in the growth wells. Growth wells 3 and 4 also only contained media. These wells serve as a positive control to make sure growth is observed over four hours. Into growth wells 5 and 6, and 7 and 8, 2 concentrations antibiotic was added. To do this 4.5 μL of a 22.2-fold more concentrated antibiotic than the target concentration in micrograms per mL was deposited into appropriate growth wells. Cartridges either contained 2 concentrations of both Ciprofloxacin (CIP) and Nitrofurantoin (NIT) or Cefazolin (CFZ) and Trimethoprim/Sulfamethoxazole (TMP/SXT). For final concentrations of each antibiotic in the cartridge, see Table 1. The media and antibiotics were then dried for 16-20 hours in a 40° C. convection oven.
Preparation of the AST Cartridge—Hybridization Reagents.
Hybridization buffer containing 3×SSC (0.45 M NaCl, 0.045 M sodium citrate, pH 7.5) (Sigma, cat. #S6639), 0.18% w/v cetrimide, 0.77% CHAPSO (Sigma cat. #C3649), 0.72% SB3-12 (Sigma cat. #D0431), and 0.13M guanidine thiocyanate (Sigma, cat. #G9277) was prepared. Trehalose (Sigma, cat. #T9449) was dissolved in this mixture to a final concentration of 10% w/v. This hybridization buffer-trehalose mixture was lyophilized in 8.3 μL volume beads. Two 8.3 uL beads were placed into each of 8 reagent wells (see figure for location on cartridges)
Preparation of the AST cartridge—Magnetic Particles Poly-aspartic acid-conjugated magnetic particles (Fluidmag-PAA, Chemicell, cat. 4108) were diluted 1:20 into 50 mM Epps buffer, pH 8.2 to a concentration of 2.75×1014 particles/mL with a final concentration of 10% w/v Trehalose (Sigma, cat. #T9449). To this dilution, fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added were added to the suspension at a final concentration of 3×106 particles/mL. The magnetic particle mixture was sonicated for 1 minute prior to immediately use to minimize aggregation. The mixture was then lyophilized in 10 μL volume beads (2.64×1012 particles per reaction). One magnetic particle lyophilized bead was placed in each of the 8 reagent wells along with the 2 hybridization mix beads.
Procedure for placing samples into the cartridge—Urine processing Prior to testing, urine preservative and other potentially interfering compounds were removed using size exclusion chromatography. 2.5 mL of each clinically positive urine sample was applied to a pre-washed Zeba™ 7K MWCO spin column (ThermoFisher, cat. #89893) and centrifuged according to the manufacturer's instructions. Urine culture was repeated on this processed sample as described above, to examine bacterial loss following processing.
Procedure for Placing Samples into the Cartridge—Putting Samples on Cartridges
750 μL of each processed urine sample was combine with 1705 μL of water and 45 μL of species-specific DNA oligonucleotide fluorescence in situ hybridization (FISH) probes and unlabeled DNA helper probes to make solution containing 30% urine v/v final concentration. Oligonucleotides used for each bacterium, their concentrations and dye labels can be found in Table 2. 1 mL of the mixture was added to the sample pot of the cartridge and the cartridge placed onto the analyzer.
Running the AST Cartridges on an Automated Analyzer
After the cartridge was then placed on the instrument, all subsequent actions other than data analysis were automatically performed. The Urine/water/FISH probe mixture (sample) was first directed under vacuum into the 8 growth wells at the top of the cartridge. Sample in the first two growth wells was then immediately relocated to reaction wells, rehydrating the hybridization buffer/FISH probe mix and lyophilized magnetic particles. Sample then continued to the imaging windows containing 46 μL of dehydrated “dye-cushion” (50 mM TRIS pH 7.5 (Teknova, cat. T5075), 7.5% v/v Optiprep (Sigma, cat. D1556), 5 mg/mL Direct Black-19 (Orient, cat. #3222), dried for 60° C. for 3 hours in a convection oven) and incubated at 35° C. for 30 minutes on the analyzer. After this incubation, the cartridges were then relocated to the magnet station, and placed atop a strong permanent magnet (Dexter magnetic technologies, cat. 54170260) for 4 minutes to bring the labeled and magnetic-particle-interacting bacterial cells into close proximity to the imaging surface. Finally, the cartridge was moved to the imaging station and imaging taken using non-magnified CCD imager described below.
Sample in the remaining six growth wells were held in that location, and the bacteria allowed to grow for 4 hours at 35° C. in the rehydrated media, either in the presence or absence of antibiotics. Following growth, the cell suspensions were relocated to the reagent wells as was done for the time zero assay, and the exact same hybridization reaction, magnetic pull-down, and imaging was performed as described above.
The analyzer imaging system and imaging process
The MultiPath Analyzer imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a MultiPath Cartridge as part of a fully automated test. It uses a custom designed precision 3 axis positioning system to locate each well over a fluorescence-based image acquisition subsystem. The Analyzer can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire Cartridge Imaging Well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 667/30 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Data analysis: Using the image captured by the CCD camera, detected cells were estimated by an algorithm that looked at both number of objects in the field of view and the intensity of the objects. Number of cells based on this detection algorithm were generated at time zero, and at time four hours without antibiotic and with both concentrations of each antibiotic. For each urine sample/drug concentration, fold growth was calculated as the signal in the well containing antibiotic after growth (time four) to the signal in the urine sample prior to growth (time zero). Comparison of fold growth and the observation of growth in the corresponding well in the CLSI-compliant broth microdilution, thresholds were selected for the fold growth cutoff to maximize agreement with the broth microdilution results. In conditions where cells are growing in the presence of the antibiotic (and thus, resistant at that concentration), the fold growth will be high and in conditions where cells are in the process of dying (and thus, sensitive at that concentration), the fold growth number will be low. In these cartridges, if both concentrations of antibiotic show no growth based on their fold growth numbers, the bacteria in that urine sample is called sensitive. If there is growth in the lower concentration but not the higher concentration, the bacteria in the urine sample is intermediate in the case of Ciprofloxacin, Nitrofurantoin and Trimethoprim/Sulfamethoxazole and resistant in the case of Cefazolin. If both concentrations of antibiotic show growth based on their fold growth thresholds, the bacteria in that urine sample is called resistant. All sensitive/resistant calls data compared to the sensitive/resistance call made by the MIC determination in a CLSI-compliant standard BMD. Four-hour growth in the absence of antibiotic is a control condition to ensure viable bacterial are present in the processed urine sample.
Results.
The figure shows the average fold growth of four replicates in two cartridges containing clinical urine sample BIUR0084, which contained a K. pneumonaie strain. The graph shows the mean fold growth in each of the 2 concentrations each of Cefazolin and Trimethoprim/Sulfamethoxazole across 4 replicates in 2 different cartridges. Using a fold growth value of 2 for both antibiotics, the MulitPath assay calls all the antibiotic concentrations of both Cefazolin and Trimethoprim/Sulfamethoxazole as growth. Therefore, this strain of K. pneumoniae is resistant to both antibiotics. This matches both the CLSI-standard broth microdilution done in house.
Conclusions. The example shows the results using the inventive automated systems, devices, and methods for rapid and automated antimicrobial susceptibility testing directly in clinical specimens from hospital patients that had E. coli in their urine. The invention delivered in just 4 hours accurate performance compared to the gold standard CLSI broth microdilution (BMD) method.
Variations. This example is illustrative of the performance of this novel AST method on a cartridge and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations), concentration of urine and urine processing procedures and alterations to reactant and antimicrobial stabilization, different bacterial targets, different antimicrobial agents etc. This methodology can also clearly be extended to other biological specimens and to other bacterial and non-bacterial pathogens.
Overview. Robustness to variable inoculum concentrations is important for the rapid AST method because when testing specimens directly from specimens the target cell concentration is unknown. This example demonstrates the use of the invention to provide accurate and consistent results directly from a urine specimen when for contrived specimens covering a wide range of target cell concentrations. This example demonstrates that variable cell inputs of E. coli BAA-2469, P. aeruginosa ATCC 27853, K. pneumoniae ATCC 700603 and K. pneumoniae CDC-0043 in the presence of 10% urine deliver accurate AST results compared to the Broth Microdilution (BMD) gold standard for AST.
Experimental ProcedurePreparation of Antibiotic Plates: Antibiotic plates containing either concentrations of three to five antibiotics in a 2-fold serial dilution series were prepared by distributing 10 μL of 10-fold higher concentration than the desired final concentration into the wells of a 96 well plate. The concentrations selected for testing of each antibiotic straddled the CLSI-reported MICs for the bacterial strains tested. Plates were prepared with all or a subset of the following antibiotics: Cefazolin, Ciprofloxacin, Levofloxacin, Nitrofurantoin, and Trimethoprim-Sulfamethoxazole. In addition to the wells containing the antimicrobial dilution series, four wells contained water to allow for positive (bacteria growth in the absence of antibiotic) and negative (no bacterial cells) controls.
Preparation of Cultures: Bacterial cultures for E. coli BAA-2469, P. aeruginosa ATCC 27853, K. pneumoniae ATCC 700603, and K. pneumoniae CDC-0043 were obtained by inoculating Trypticase Soy Broth (TSB, Hardy Diagnostics cat. U65) with 3 to 5 colonies from fresh tryptic soy agar plates (TSA, BD cat. 221185) and growing for 1.5 to 3 hours at 35° C. to achieve log-phase growth. The cells were diluted in 1× cation-adjusted Mueller-Hinton broth (MHBII, Teknova cat. M5860) to various inoculum (2×103 CFU/mL-1×107 CFU/mL). For more accurate cellular concentrations, these estimated bacterial inputs were adjusted using colony counts. Plate counts were determined by diluting the log-phase cultures to about 500 CFU/mL in MHBII, plating 100 μL on TSA plates and counting colonies after growth at 35° C. for 16 to 24 hours. Using the average plate counts, the actual CFU present in each concentration tested was computed.
Preparation of Magnetic Particles: 2 hydroxypropyl trimethylammonium chloride coated silica magnetic particles (SiMag-Q, Chemicell, cat. 1206-5) were diluted 1:20 into 50 mM EPPS buffer, pH 8.2 to 2.75×1012 particles/mL. Fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. These particles enable the optical system to focus on the correct plane. The magnetic particle mixture was sonicated for 1 minute immediately prior to use to minimize aggregation. Separate magnetic particle suspensions were prepared for the time zero and time four-hour assays described below.
Bacterial Cell Labeling at AST time zero: Assay signal prior to the initiation of bacterial growth in the presence or absence of antibiotics (time zero or TO) was determined for each organism and inoculum. 10 μL of each sample was added to 80 μL of hybridization buffer to final concentrations of 3×SSC (0.45 M NaCl, 0.045 M Na citrate, Sigma, cat. #S6639), 1% CHAPS (Sigma, cat. #C3023), 1% NOG (Sigma cat. #08001), 1× cation-adjusted Mueller-Hinton Broth (MHBII), species-specific DNA oligonucleotide FISH probes and unlabeled DNA helper probe. The oligonucleotide probes used are shown in Table B. A final concentration of 10% urine was obtained by adding 10 μL of pooled urine (in-house collected and filtered) directly to the mixture. 10 μL of the magnetic particle mixture prepared as described above was then added. 100 μL of this reaction mixture was transferred to a microtiter plate containing 50 μL per well (previously dried) of “dye cushion” (50 mM TRIS pH 7.5 (Teknova, cat. T5075), 7.5% v/v Optiprep (Sigma, cat. D1556), 5 mg/mL Direct Black-19 (Orient, cat. #3222) (dry-cushion plate) and incubated at 35° C. for 30 minutes. After incubation, microtiter plates were placed onto a magnetic field (Dexter magnetic technologies, cat. 54170260) for 4 minutes to bring magnetic particles, a fraction containing labeled cells, through the “dye cushion” and into proximity to the imaging surface at the bottom of the wells.
Imaging of labeled cells: The MultiPath™ laboratory imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a microtiter plate. It uses a high precision linear stage from Prior Scientific (Rockland, Mass.) to position each well over a fluorescence-based image acquisition subsystem. The instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire microtiter plate well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 667/30 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Four-Hour Growth: While the time zero cell quantification was occurring, 10 μL of each organism inoculum, 10 μL of pooled urine, and 70 μL of 1×MHBII was added to the appropriate wells of the antibiotic plate (already containing 10 μL of antibiotic). The samples were allowed to grow in a standard air incubator at 35° C. for 4 hours.
Bacterial Cell Labeling at AST time four-hour growth: After samples had incubated in the presence and absence of antibiotics for four hours (T4), cells were labeled and quantified to determine how much growth, if any, occurred. 10 μL of the incubated sample-antibiotic plate (10%) was transferred to a microtiter plate and combine with 100 μL hybridization buffer, FISH probes, helper probes, magnetic particles, and focus particles in the same manner as described above for assay time zero.
Comparison Method: Results using the novel AST method described here were compared to broth microdilutions (BMD) performed according to CLSI M07-Ed13E 2018.
Data Analysis and Threshold Generation: Using the image captured by the CCD camera, detected cells were estimated by an algorithm that looked at both number of objects in the field of view and the intensity of the objects. The number of cells based on this detection algorithm were generated at time zero, and at time four hours without antibiotic and with all concentrations of each antibiotic. For each sample inoculum/drug concentration, fold growth was calculated as the signal in the well containing antibiotic after growth (time four) to the signal in the urine sample prior to growth (time zero). Using fold growth and the observation of growth in the corresponding well in the CLSI-compliant broth microdilution, a logistic regression model was used to generate thresholds for determining the fold growth cutoff above which cells are growing in the presence of the antibiotic (and thus, resistant at that concentration) and below which, cells are in the process of dying (and thus, sensitive at that concentration). The point where the fold growth number falls below the determined threshold is the MIC value generated by the assay. Results were correspondingly assigned to categories of susceptible, intermediate, or resistant to each antibiotic. All data was then compared to CLSI standard BMD. Four-hour growth in the absence of antibiotic is a control condition to ensure viable bacterial are present in the processed urine sample.
Results.
The figures below show how this method is robust to varying inoculum levels while matching the gold-standard broth microdilution method.
Conclusion.
The rapid 4-hour AST method presented here is robust to initial cell concentration over a wide range of target cell concentrations. Robustness to variable inoculum concentrations is important for the rapid AST method because when testing specimens directly from specimens the target cell concentration is unknown.
Variations. This example is illustrative of the performance of this novel AST method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations), concentration of urine and urine processing procedures. This methodology can also clearly be extended to other biological specimens and to other bacterial and non-bacterial pathogens for which specific probes can be designed, and for other antimicrobial or chemical agents.
Overview. Current methods for antimicrobial susceptibility testing require lengthy culture-based colony purification to ensure a pure population of the target pathogen cells free of other microbes. The usual method, colony purification requires, 2-5 days to deliver results. In the interim, patients are treated empirically with powerful broad-spectrum antibiotics that may not be optimal for killing the pathogen causing the infection and can even be completely ineffective. Plus, empiric treatment with broad-spectrum antibiotics causes the spread of antibiotic resistance.
Current methods require the lengthy cell purification process because these methods use non-specific detection methods, such as increase in turbidity, to determine which antimicrobial agents inhibit the growth of the target pathogen in microbiological medium. When using non-specific measurement of cellular replication one can only know that the growth seen is due to the target pathogen if the contains only cells of the target pathogen. Cell purification must be undertaken for current antimicrobial susceptibility testing methods because most medical specimens are non-sterile. Specimens generally contain microbes that make up the human microbiome, the benign normal bacterial population that populate our bodies.
In contrast, the inventive method can deliver accurate antimicrobial susceptibility testing results directly from specimens without the colony purification step. The method differs from current methods in that it assesses growth specifically for the target pathogen in microbiological medium containing antimicrobial agents.
This example demonstrates that the rapid antimicrobial susceptibility testing method accurately determines the minimum inhibitory concentration (MIC) for an E. coli strain in contrived samples comprising urine matrix (10%) for 15 different culture-negative urine samples. Here we show that using the new method antimicrobial susceptibility testing results are accurate and not significantly impacted by off-target bacteria in urine samples containing high concentrations of other microbial species.
Experimental ProcedurePreparation of Antibiotic Plates: Prior to initiating experimental procedure, a plate containing five concentrations in a 2-fold serial dilution series were prepared by distributing 10 μL of 10-fold higher concentration than the desired concentration. The concentrations selected for testing of each antibiotic straddled the CLSI-reported breakpoints for the antibiotic for E. coli. In addition to the wells containing the antimicrobial dilution series, four wells containing water were included in the plates to allow for a positive and negative control.
Preparation of Cultures: Three to five colonies of E. coli BAA-2469 as well as eight other off-target species (S. aureus ATCC 25923, C. freundii ATCC 43864, A. baumannii ATCC 19606, S. epidermidis ATCC 12228, M. luteus (environmental isolate), C. minutissmum ATCC 23348-BAA 949, K. pneumoniae CDC 0043, and K. pneumoniae CDC 0141) were each inoculated separately into 5 mL of Tryptic Soy Broth (TSB, Hardy Diagnostics cat. U65) and incubated while shaking for 1-2 hours at 35° C. The optical density was measured by a spectrophotometer and the organisms were diluted in 1× cation-adjusted Mueller-Hinton Broth (MHBII, Teknova cat. M5860). E. coli was diluted to approximately 5×106 CFU/mL (final assay concentration is 5×105 CFU/m) while the other off-target species were diluted to various inoculum (ranging from 1×105 to 5×108 CFU/mL).
Preparation of Magnetic Particles: 2 hydroxypropyl trimethylammonium chloride-coated silica magnetic particles (SiMag-Q, Chemicell, cat. 1206-5) were diluted 1:20 into 50 mM EPPS buffer, pH 8.2 to 2.75×1012 particles/mL. Fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. These particles enable the optical system to focus on the correct plane. The magnetic particle mixture was sonicated for 1 minute immediately prior to use to minimize aggregation. Separate magnetic particle suspensions were prepared for the time zero and time four-hour assays described below.
Bacterial Cell Labeling at AST time zero: Assay signal prior to the initiation of bacterial growth in the presence or absence of antibiotics (time zero or TO) was determined for each species of E. coli. 10 μL of each sample was added to 80 μL of hybridization buffer (3×SSC (0.45 M NaCl, 0.045 M sodium citrate) (Sigma, cat. #S6639), 1% CHAPS (Sigma, cat. #C3023), 1% SB3-12 (Sigma cat. #08001), 1× Cation-adjusted Mueller-Hinton Broth (MHBII), E. coli-specific DNA oligonucleotide FISH probes and unlabeled DNA helper probe)). Probe sequences are shown in Table in
Imaging of labeled cells: The MultiPath™ laboratory imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a microtiter plate. It uses a high precision linear stage from Prior Scientific (Rockland, Mass.) to position each well over a fluorescence-based image acquisition subsystem. The instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire microtiter plate well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 667/30 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Four-Hour Growth: The E. coli BAA 2469, in the presence of Staphylococcus epidermidis, Micrococcus luteus, Corynebacterium minutissimum, Staphylococcus aureus, Acinetobacter baumannii, Citrobacter freundii) were tested for their susceptibility against 3 antimicrobial agents: Ciprofloxacin (CIP), Levofloxacin (LVX), and Nitrofurantoin (NIT). E. coli BAA 2469, in the presence of Klebsiella pneumoniae was tested against 5 antimicrobial agents: Cefazolin (CFZ), Ciprofloxacin (CIP), Levofloxacin (LVX), Nitrofurantoin (NIT), and Trimethoprim-Sulfamethoxazole (TMP/SXT). Antibiotic plates containing these antimicrobial agents were prepared according to the method described above. While the time zero cell quantification was occurring, 10 μL of either E. coli species (5×106 CFU/mL), 10 μL of an off-target species (1×105 to 5×108 CFU/mL), 10 μL of pooled urine, and 60 μL of MHB II (Teknova, cat. #M5860) was added to each well of the antibiotic plate already containing 10 μL of antibiotics. The samples were allowed to grow in a standing air incubator at 35° C. for 4 hours.
Bacterial Cell Labeling at AST time four-hour growth: After samples had incubated in the presence and absence of antibiotics for four hours (T4), cells were labeled and quantified to determine how much growth, if any, occurred. 10 μL of the incubated sample-antibiotic plate (10%) was transferred to a microtiter plate containing dried “dye cushion” and combined with the 100 μL mixture of hybridization buffer, FISH probes, helper probes, magnetic particles, and focus particles as described above for assay time zero.
Comparison Method: Results for the novel assay method described here were compared to broth microdilutions (BMD) performed according the M07-Ed13E 2018.
Data Analysis and Threshold Generation: Using the image captured by the CCD camera, detected cells were estimated by an algorithm that looked at both number of objects in the field of view and the intensity of the objects. Number of cells based on this detection algorithm were generated at time zero, and at time four hours without antibiotic and with all concentrations of each antibiotic. For each bacteria sample/drug concentration, fold growth was calculated as the signal in the well containing antibiotic after growth (time four) to the signal in the urine sample prior to growth (time zero). Using fold growth and the observation of growth in the corresponding well in the CLSI-compliant broth microdilution, a logistic regression model was used to generate thresholds for determining the fold growth cutoff above which cells are growing in the presence of the antibiotic (and thus, resistant at that concentration) and below which, cells are in the process of dying (and thus, sensitive at that concentration). The point where the fold growth number falls below the determined threshold is the MIC value generated by the assay. Results were correspondingly assigned to categories of susceptible, intermediate, or resistant to each antibiotic.
Results.
The data shown demonstrate the 4-hour AST method described above is robust to non-sterile samples while a CLSI BMD method where extra bacteria is present is not.
Conclusions. The example demonstrates that using the invention for antimicrobial susceptibility testing, cell purification is not required for achieving accurate antimicrobial susceptibility testing results for a target pathogen in samples containing even large numbers of other microbes of other species.
Variations: This example is illustrative of the performance of this novel FISH method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations), concentration of urine and urine processing procedures. This methodology can also clearly be extended to other biological specimens and to other bacterial and non-bacterial pathogens.
Overview. Current methods for antimicrobial susceptibility testing require lengthy culture-based colony purification to ensure a pure population of the target pathogen cells free of other microbes. The usual method, colony purification requires, 2-5 days to deliver results. In the interim, patients are treated empirically with powerful broad-spectrum antibiotics that may not be optimal for killing the pathogen causing the infection and can even be completely ineffective. Plus, empiric treatment with broad-spectrum antibiotics causes the spread of antibiotic resistance.
One reason that current methods require the lengthy cell purification process because these methods use non-specific detection methods, such as increase in turbidity, to determine which antimicrobial agents inhibit the growth of the target pathogen in microbiological medium. When using non-specific measurement of cellular replication one can only know that the growth seen is due to the target pathogen if the contains only cells of the target pathogen.
In contrast, the inventive method can deliver accurate antimicrobial susceptibility testing results directly from specimens without the colony purification step. The method differs from current methods in that it assesses growth specifically for the target pathogen in microbiological medium containing antimicrobial agents. We demonstrate in another example, that the inventive method is accurate in the presence of large numbers of cells from off-target species.
In this example, we address another challenge that could arise by performing antimicrobial susceptibility testing for a target pathogen in the presence of off-target species. Here we demonstrate that the inventive method delivers accurate antimicrobial susceptibility testing results for a target pathogen in contrived urine specimens containing large numbers of an off-target species that makes an enzyme known to break down the antimicrobial agent being tested. Theoretically this could potentially change the concentration of the antimicrobial agent significantly enough to alter the antimicrobial susceptibility testing result.
In this example, we demonstrate that the rapid antimicrobial susceptibility testing achieves accurate antimicrobial susceptibility testing results for two carbapenem antibiotics (Meropenem and Imipenem) in the presence of large numbers of an off-target pathogen that produces a enzyme that breaks down this type of antimicrobial agent.
Experimental procedure. Preparation of Antibiotic Plates: Antibiotic plates prepared as described in Impact of Non-Sterile Sample on Target MIC example.
Preparation of Cultures: Three to five colonies of E. coli ATCC 25922, a strain of bacteria sensitive to most antibiotics and K. pneumoniae CDC 0141, a strain that, among many other resistance genes, expresses the beta-lactase OXA-181, were each inoculated separately into 5 mL of Tryptic Soy Broth (TSB, Hardy Diagnostics cat. U65) and incubated while shaking for 1-2 hours at 35° C. The Optical Density was measured by a spectrometer and the organisms were diluted in 1× cation-adjusted Mueller-Hinton Broth (MHBII, Teknova cat. M5860). E. coli was diluted to 5×105 CFU/mL (CLSI standard concentration) while K. pneumoniae was diluted to various inoculum (ranging from 1×106 and 5×108 CFU/mL).
Preparation of Magnetic Particles: 2 hydroxypropyl trimethylammonium chloride-coated silica magnetic particles (SiMag-Q, Chemicell, cat. 1206-5) were diluted 1:20 into 50 mM EPPS buffer, pH 8.2 to 3.75×106 particles/mL. Fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. These particles enable the optical system to focus on the correct plane. The magnetic particle mixture was sonicated for 1 minute immediately prior to use to minimize aggregation. Separate magnetic particle suspensions were prepared for the time zero and time four-hour assays described below.
Bacterial Cell Labeling at AST time zero: Assay signal at time zero (TO) prior to the initiation of bacterial growth in the presence or absence of antibiotics was determined for each clinical urine specimen. 30 μL of each processed urine was added to 70 μL of 1× cation-adjusted Mueller-Hinton Broth (MHBII) containing species-specific Alexa647N-labeled DNA oligonucleotide FISH probes and unlabeled DNA helper probes. Probe sequences used are shown in Table A. The 100 μL mixture was then added to a well of a microtiter plate containing dehydrated hybridization buffer (3×SSC (0.45 M NaCl, 0.045 M Na citrate) buffer (Sigma, cat. #S6639), 0.18% cetrimide (Sigma, cat. #H9151), 0.77% CHAPSO (Sigma cat. #C3649), 0.72% SB3-12 (Sigma cat. #D0431) 0.13M guanidine thiocyanate (Sigma, cat. #G9277)). 10 μL of the prepared magnetic particle mixture was then added to the well. 100 μL of this reaction mixture was transferred to a microtiter plate containing 50 μL per well (previously dried) of “dye-cushion” (50 mM TRIS pH 7.5 (Sigma cat. T1075), 7.5% v/v Optiprep (Sigma cat. D1556), 50 mg/mL Direct Black 19 (Orient cat. 191L) and incubated at 35° C. for 30 minutes. After incubation, microtiter plates were placed onto a magnetic field (Dexter magnetic technologies, cat. 54170260) for 4 minutes to bring magnetic particles, a fraction containing labeled cells, through the “dye-cushion” and into proximity to the imaging surface at the bottom of the wells.
Imaging of labeled cells: The MultiPath™ laboratory imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a microtiter plate. It uses a high precision linear stage from Prior Scientific (Rockland, Mass.) to position each well over a fluorescence-based image acquisition subsystem. The instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire microtiter plate well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 667/30 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Four-Hour Growth: The E. coli was tested in the presence of varying inoculum of K. pneumoniae-OXA for susceptibility against 2 antimicrobial agents: Imipenem and Meropenem. While the time zero cell quantification was occurring, 10 μL of the E. coli species (5×106 CFU/mL), 10 μL of the K. pneumoniae (1×106 to 1×108 CFU/mL) or 10 uL media (control), 10 μL of pooled urine, and 60 μL of MHB II (Teknova, cat. #M5860) was added to each well of the antibiotic plate already containing 10 μL of antibiotics. The samples were allowed to grow in a standing air incubator at 35° C. for 4 hours.
Bacterial Cell Labeling at AST time four-hour growth: After samples had incubated in the presence and absence of antibiotics for four hours (T4), cells were labeled and quantified to determine how much growth, if any, occurred. 100 μL of each well of the incubated sample-antibiotic plate was transferred to a corresponding well of a dehydrated buffer plate and combined with FISH probes, helper probes, magnetic particles, and focus particles in the same manner as described above for assay time zero.
Comparison Method: Results for the MulitPath™ Assay were compared to broth microdilutions (BMD) performed according to the CLSI method M07-Ed13E 2018.
Data Analysis and Threshold Generation: Using the image captured by the CCD camera, detected cells were estimated by an algorithm that looked at both number of objects in the field of view and the intensity of the objects. Number of cells based on this detection algorithm were generated at time zero, and at time four hours without antibiotic and with all 6 concentrations of each antibiotic. For each urine sample/drug concentration, fold growth was calculated as the signal in the well containing antibiotic after growth (time four) to the signal in the urine sample prior to growth (time zero). Using fold growth and the observation of growth in the corresponding well in the CLSI-compliant broth microdilution, a logistic regression model was used to generate thresholds for determining the fold growth cutoff above which cells are growing in the presence of the antibiotic (and thus, resistant at that concentration) and below which, cells are in the process of dying (and thus, sensitive at that concentration). The point where the fold growth number falls below the determined threshold is the MIC value generated by the assay. Results were correspondingly assigned to categories of susceptible, intermediate, or resistant to each antibiotic. All data was then compared to CLSI standard BMD. Four-hour growth in the absence of antibiotic is a control condition to ensure viable bacterial are present in the processed urine sample.
Results.
Conclusions. The novel 4.5-hour AST method of this invention shows accurate MIC determination of bacteria sensitive to carbapenem antimicrobial agents even in the presence of high concentrations of a resistant bacteria expressing a carbapemase enzyme which degrades the antibiotic.
Variations. This example is illustrative of the performance of this novel AST method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations, etc.), concentration of urine and urine processing procedures. This methodology can also be extended to additional pairings of lactam sensitive and beta-lactamase expressing bacteria.
Overview: Current methods for antimicrobial susceptibility testing require lengthy culture-based colony purification to ensure a pure population of just pathogen cells free of the specimen itself. Consequently, antimicrobial susceptibility testing results that indicate which antibiotics are optimal for killing the pathogen causing the infection are not available for 2-5 days. In the interim, patients are treated empirically with powerful broad-spectrum antibiotics that may not be optimal for killing the pathogen causing the infection and can even be completely ineffective. Plus, empiric treatment with broad-spectrum antibiotics causes the spread of antibiotic resistance.
In contrast, the inventive method can deliver accurate antimicrobial susceptibility testing results directly from specimens without the lengthy coloniy purification step. Here we show that the new antimicrobial susceptibility testing results are not significantly impacted when bacteria in urine samples are tested without colony purification. This example demonstrates that the rapid antimicrobial susceptibility testing method accurately determines the minimum inhibitory concentration (MIC) for an E. coli strain in contrived samples comprising urine matrix (10%) for 15 different culture-negative urine samples.
Experimental procedure. Urine specimens: Fifteen culture negative clinical urine samples (remnants) were purchased from Discovery Life Sciences. Samples were received >7 days post collection and stored at −80° C. until use. For each sample, color of urine, pH, and presence of particulates were noted. Upon receipt, conventional urine culture was performed on the urines to determine samples were culture negative. Briefly, a calibrated 1 uL loop was placed into a well-mixed urine sample and evenly spread over a Tryptic soy agar (TSA) plate and incubated in a 35° C. air incubator for 18-24 hours. The remainder of the urine samples were assayed as described below.
Preparation of Antibiotic Plates: Microtiter plates containing six concentrations of each antibiotic in a 2-fold serial dilution series were prepared, starting at a 10-fold higher concentration than the expected minimum inhibitory concentration (MIC). Antibiotics used were Cefazolin, Ciprofloxacin, Nitrofurantoin, and Trimethoprim-Sulfamethoxazole. Antibiotic dilutions were verified to be within the appropriate tolerance by confirming that the MIC for at least two CLSI QC strains fell within the QC range reported in CLSI document M100Ed29E-2019. The concentrations selected for testing of each antibiotic straddled the CLSI-reported breakpoints for the antibiotic for E. coli. In addition to the wells containing the antimicrobial dilution series, eight wells containing water or diluent were included in the plates to allow for a no antibiotic positive and negative growth control.
Preparation of Cultures: A log culture for E. coli (BAA-2469) was grown using three to five colonies inoculated into 5 mL of Tryptic Soy Broth (TSB, Hardy Diagnostics cat. U65) and incubated while shaking for 1-2 hours at 35° C. The Optical Density was measured by a spectrophotometer and the organisms were diluted to 5×106 CFU/mL (for a final concentration of 5×105 CFU/mL in each 100 μL reaction) in 1× Cation-adjusted Mueller-Hinton Broth (MHBII, Teknova cat. M5860).
Preparation of Magnetic Particles: 2 hydroxypropyl trimethylammonium chloride-coated silica magnetic particles (SiMag-Q, Chemicell, cat. 1206-5) were diluted 1:20 into 50 mM EPPS buffer, pH 8.2 to 2.75×1012 particles/mL. Fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. These particles enable the optical system to focus on the correct plane. The magnetic particle mixture was sonicated for 1 minute immediately prior to use to minimize aggregation. Separate magnetic particle suspensions were prepared for the time zero and time four-hour assays described below.
Bacterial Cell Labeling at AST time zero: Assay signal prior to the initiation of bacterial growth in the presence or absence of antibiotics (time zero or TO) was determined for each urine sample. 10 μL of diluted E. coli was added to 70 μL of hybridization buffer: final concentration: 3×SSC (0.45 M NaCl, 0.045 M Na citrate) buffer (Sigma, cat. #S6639), 1% CHAPS (Sigma, cat. #C3023), 1% NOG (Sigma cat. #08001), 1× Cation-adjusted Mueller-Hinton Broth (MHBII) (from a 2× stock) (Teknova, cat. M5866), and non-specific DNA oligonucleotide FISH probes and unlabeled DNA helper probe (see Table A for probe labels, sequences, and concentrations). A final concentration of 10% urine was obtained by adding 10 μL of each individual urine directly to the mixture. 10 μL of the magnetic particle mixture prepared as described above was added directly to this mixture. 100 μL of the sample, now containing the hybridization mixture, urine, and magnetic particles, was transferred to a microtiter plate containing 50 μL per well (previously dried) of “dye cushion” (50 mM TRIS pH 7.5 (Teknova, cat. T5075), 7.5% v/v Optiprep (Sigma, cat. D1556), 5 mg/mL Direct Black-19 (Orient, cat. #3222), dried at 60° C. in a convection oven for 3 hours) and incubated at 35° C. for 30 minutes. After incubation, microtiter plates were placed onto a magnetic field (Dexter magnetic technologies, cat. 54170260) for 4 minutes to bring magnetic particles, a fraction containing labeled cells, through the “dye cushion” and into proximity to the imaging surface at the bottom of the wells.
Imaging of labeled cells: The MultiPath™ laboratory imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a microtiter plate. It uses a high precision linear stage from Prior Scientific (Rockland, Mass.) to position each well over a fluorescence-based image acquisition subsystem. The instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire microtiter plate well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 667/30 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Four-Hour Growth: Spiked culture negative clinical UTI urine samples were tested for their susceptibility against 5 antimicrobial agents: Cefazolin, Ciprofloxacin, Levofloxacin, Nitrofurantoin, and Trimethoprim-Sulfamethoxazole. Antibiotic plates containing these antimicrobial agents were prepared according to the method described above. At the same time as the time zero cell quantification was occurring, 10 μL of E. coli, 10 μL of urine, and 70 μL of 1×MHB II (Teknova, cat. M5860) were added to each well of the antibiotic plate. The samples were allowed to grow in a standing air incubator at 35° C. for 4 hours.
Bacterial Cell Labeling at AST time four-hour growth: After samples had incubated in the presence and absence of antibiotics for four hours (T4), cells were labeled and quantified to determine how much growth, if any, occurred. 100 μL of each well of the incubated sample-antibiotic plate was transferred to a corresponding well of a dehydrated buffer plate and combined with FISH probes, helper probes, magnetic particles, and focus particles in the same manner as described above for assay time zero.
Comparison Method: Results for the MulitPath™ Assay were compared to broth microdilutions (BMD) performed according to the CLSI method M07-Ed13E 2018.
Data Analysis and Threshold Generation: Using the image captured by the CCD camera, detected cells were estimated by an algorithm that looked at both number of objects in the field of view and the intensity of the objects. Number of cells based on this detection algorithm were generated at time zero, and at time four hours without antibiotic and with all 6 concentrations of each antibiotic. For each urine sample/drug concentration, fold growth was calculated as the signal in the well containing antibiotic after growth (time four) to the signal in the urine sample prior to growth (time zero). Using fold growth and the observation of growth in the corresponding well in the CLSI-compliant broth microdilution, a logistic regression model was used to generate thresholds for determining the fold growth cutoff above which cells are growing in the presence of the antibiotic (and thus, resistant at that concentration) and below which, cells are in the process of dying (and thus, sensitive at that concentration). The point where the fold growth number falls below the determined threshold is the MIC value generated by the assay. Results were correspondingly assigned to categories of susceptible, intermediate, or resistant to each antibiotic. All data was then compared to CLSI standard BMD. Four-hour growth in the absence of antibiotic is a control condition to ensure viable bacterial are present in the processed urine sample.
Results. The figures below show there is little to no matrix effect on AST results.
Conclusion. The inventive method accurately determined the MIC (within the essential agreement zone relative to the gold standard BMD method) for a UTI pathogen (E. coli) for all 5 antibiotics tested in all 15 distinct urine matrices. Thus, this novel 4-hour antimicrobial susceptibility test has the capability to provide accurate results directly from urine specimens without the requirement of lengthly growth-based colony purification, saving substantial time. Rapid AST results can improve patient care by allowing the correct, effective antibiotic treatment to be initiated quickly and avoid adding to the spread of antibiotic resistance.
Variations. This example is illustrative of the performance of this novel AST method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations) and concentration of urine. This methodology can also clearly be applied to other bacterial and non-bacterial pathogens and to minimally processed clinical matrices other than urine.
-
- Cefazolin, Ciprofloxacin, Levofloxacin, Nitrofurantoin, and Trimethoprim-Sulfamethoxazole.
Overview. Polymicrobial infections are common in many types of infections including wounds. For such infections, which can be life-threatening it is critical to determine which antimicrobial agents can be effective for each infectious pathogen. Current antimicrobial susceptibility testing methods require 2-5 days to purify large numbers of each infectious pathogen in a polymicrobial infection before they can be analyzed.
This example demonstrates the potential for the inventive systems and methods to generate rapid AST in results in just 4.5 hours directly from a patient specimen without the need for lengthy colony purification. The method achieves accurate AST results (MIC values) for each target species in contrived 2-target polymicrobial mixtures compared to the broth microdilution reference standard result.
Experimental ProcedurePreparation of Antibiotic Plates: Microtiter plates containing 6 Ciprofloxacin concentrations 2-fold serial dilution series were prepared. The 2-fold dilution series was prepared at a 10-fold higher concentration the desired concentration in the final broth microdilution, such that addition cells/urine/media mixture would yield the correct antibiotic range. 10 uL of each antibiotic dilution was then aliquoted into the appropriate wells of a 96 well plate. Antibiotic dilutions were verified to be within the appropriate tolerance by confirming that the MIC for at least two CLSI QC strains fell within the QC range reported in CLSI document M100Ed29E-2019. In addition to the wells containing the antimicrobial dilution series, enough wells containing water or other diluent were included for a no antibiotic positive growth control. Antibiotic plates were frozen at −80° C. and thawed completely before use.
Preparation of Cultures: Both a susceptible and resistant strain were chosen for four different organisms (E. coli ATCC 25922, E. coli BAA-2469, K. pneumoniae CDC 0076, K. pneumoniae CDC 0043, P. aeruginosa CDC 0233, P. aeruginosa CDC 0236, E. faecalis ATCC 29212, and E. faecium ATCC 19434). The strains and their resistances to each antibiotic tested are shown in Table A. Each strain was grown separately with three to five colonies inoculated into 5 mL of Tryptic Soy Broth (TSB) and incubated while shaking for 1-2 hours at 35° C. The Optical Density was measured by a spectrometer and the organisms were diluted to 1×107 CFU/mL in 1× Cation-adjusted Mueller-Hinton Broth (MHB II) (Teknova, cat. #M5860).
Preparation of Magnetic Particles: A solution of Poly-aspartic acid-conjugated magnetic particles (Fluidmag-PAA, Chemicell, cat. 4108) were diluted 1:20 into 50 mM EPPS buffer, pH 8.2 to 2.75×1012 particles/mL. Fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. These particles enable the optical system to focus on the correct plane. The magnetic particle mixture was sonicated for 1 minute immediately prior to use to minimize aggregation. Separate magnetic particle suspensions were prepared for the time zero and time four-hour assays described below). An identical procedure was done with the 2 hydroxypropyl trimethylammonium chloride-coated silica magnetic particles (SiMag-Q, Chemicell, cat. 1206-5).
Bacterial Cell Labeling at AST time zero: Assay signal prior to the initiation of bacterial growth in the presence or absence of antibiotics (time zero or TO) was determined for each species and strain. 5 μL of target A was combined with either 5 μL target B or 5 μL of MHB II for a final concentration of 5×106 CFU/mL per organism was added to 80 μL of hybridization buffer (final concentration: 3×SSC (0.45 M NaCl, 0.045 M sodium citrate pH 7) (Sigma, cat. #S6639), 0.25M Guanidine Thiocyanate (Sigma, cat. #503-84-0), 5% PEG MW 3350 (Sigma, cat. #P-3640), 7.5% Igepal CA-630 (Sigma, cat. #13021), 0.2% cetrimide (Sigma, cat. #H9151), 1× Cation-adjusted Mueller-Hinton Broth (MHBII), species-specific DNA oligonucleotide FISH probes and unlabeled DNA helper probe (sequences and concentrations found in Table B)). A final concentration of 10% urine was obtained by adding 10 μL of pooled urine (Innovative Research, cat. IR100007P-24203) directly to the mixture for a 100 μL total reaction. 10 μL of the either the SiMag-Q magnetic particle mixture (for conditions where E. coli, K. pneumoniae and P. aeruginosa strains were being labeled) or the Fluidmag-PAA magnetic particle mixture (for conditions where Enterococcus spp. were labeled), prepared as described above, was added directly to this mixture. 100 μL of the sample, now containing the hybridization mixture, urine, and magnetic particles, was transferred to a microtiter plate containing 50 μL of dye-cushion (50 mM TRIS pH 7.5 (Teknova, cat. T5075), 7.5% v/v Optiprep (Sigma, cat. D1556), 5 mg/mL Direct Black-19 (Orient, cat. #3222), dried down at 60° C.) (dry-cushion plate) and incubated at 35° C. for 30 minutes. After this incubation, the microtiter plates were placed onto a strong permanent magnet (Dexter magnetic technologies, cat. 54170260) for 4 minutes to bring the labeled and magnetic-particle-interacting bacterial cells into close proximity to the imaging surface.
Imaging of labeled cells: The MultiPath™ laboratory imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a microtiter plate. It uses a high precision linear stage from Prior Scientific (Rockland, Mass.) to position each well over a fluorescence-based image acquisition subsystem. The instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire microtiter plate well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 667/30 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Four-Hour Growth: A polymicrobial sample, containing two species, was tested for susceptibility against 1 antimicrobial agent: Ciprofloxacin. Antibiotic plates containing these antimicrobial agents were prepared according to the method described above. At the same time as the time zero cell quantification was occurring, 5 μL of either the species to be labeled and detected and 5 μL of either a bacterial species the might be present in a polymicrobial UTI infection (but will not label) or MHB II as control, 10 μL of pooled urine, and 70 μL of MHB II were added to each well of the antibiotic plate. The samples were allowed to grow in a standing air incubator at 35° C. for 4 hours. Each strain in this example served in once instance as the labeled target species, and in another instance as the unlabeled member of the polymicrobial pair.
Bacterial Cell Labeling at AST time four-hour growth: After samples had incubated in the presence and absence of antibiotics for four hours (T4), cells were labeled and quantified to determine how much growth, if any, occurred. 10 μL of the incubated sample-antibiotic plate (10%) was transferred to a microtiter plate and combine with 100 μL hybridization buffer, FISH probes, helper probes, magnetic particles, and focus particles in the same manner as described above for assay time zero.
Comparison Method: Results using the novel AST method described here were compared to broth microdilutions (BMD) performed according to CLSI M07-Ed13E 2018.
Data Analysis and Threshold Generation: Using the image captured by the CCD camera, detected cells were estimated by an algorithm that looked at both number of objects in the field of view and the intensity of the objects. The number of cells based on this detection algorithm were generated at time zero, and at time four hours without antibiotic and with all 6 concentrations of Ciprofloxacin. For each sample inoculum/drug concentration, fold growth was calculated as the signal in the well containing antibiotic after growth (time four) to the signal in the urine sample prior to growth (time zero). Using fold growth and the observation of growth in the corresponding well in the CLSI-compliant broth microdilution, a logistic regression model was used to generate thresholds for determining the fold growth cutoff above which cells are growing in the presence of the antibiotic (and thus, resistant at that concentration) and below which, cells are in the process of dying (and thus, sensitive at that concentration). The point where the fold growth number falls below the determined threshold is the MIC value generated by the assay. MIC results were correspondingly assigned to categories of susceptible, intermediate, or resistant based on the CLSI M100Ed28 2018 guidelines. All data was then compared to CLSI standard BMD.
Results.
Conclusions. The inventive AST method can accurately determine antibiotic susceptibility for each species in a polymicrobial sample in 4.5 hours without requiring the time consuming colony purification needed by current methods. The results show the potential for the invention to determine the antimicrobial agents that can effectively treat life-threatening polymicrobial infections in just hours rather than the days required by today's methods.
Variations. This example is illustrative of the performance of this novel AST method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.) and alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations). This methodology can also clearly be extended to other biological specimens, to other bacteria and to other antibiotics.
Overview. Polymicrobial infections, that is infections caused by more than one bacterial species, are common. Current, culture-based and MALDI-TOF based methods for identifying pathogens, require lengthy colony purification steps for separately purifying large number of cells each target species. This example demonstrates the use of the inventive FISH method to detect and identify multiple species of target pathogens present in contrived urine samples in 30 minutes on an automated analyzer inside a single-use consumable cartridge containing all assay reagents. The example shows the potential of the systems and methods of the invention to rapidly and specifically identify multiple target pathogens in polymicrobial infections.
Experimental ProcedureUrine specimens: Ten culture negative clinical urine samples (remnant) were purchased from Discovery Life Sciences. Samples were received >7 days post collection and stored at −80° C. until use. For each sample, color of urine, pH, and presence of particulates were noted. Upon receipt, conventional urine culture was performed on the urines to determine samples were culture negative. Briefly, a calibrated 1 uL loop was placed into a well-mixed urine sample and evenly spread over a Tryptic soy agar (TSA, BD cat. 221185) plate and incubated in a 35° C. air incubator for 18-24 hours. The remainder of the urine samples were processed and assayed as described below.
Urine processing: Prior to performing identification (ID), urine preservative and other potentially interfering compounds were removed using size exclusion chromatography. 2.5 mL of each clinically negative urine sample was applied to a pre-washed Zeba™ Spin Desalting column, 7K MWCO (ThermoFisher, cat. #89893). The sample was passed through the column via centrifugation as described by the manufacturer.
Preparation of Dehydrated Reagents in Cartridge: Prior to performing identification (ID), 45 μL of 2.2× concentrated hybridization buffer (6.7×SSC (1 M NaCl, 0.1 M sodium citrate, (Sigma, cat. #S6639), 0.4% w/v cetrimide (Sigma, cat. #H9151), 1.71% w/v CHAPSO (Sigma cat. #C3649), 1.6% SB3-12 w/v (Sigma cat. #D0431), and 0.29M guanidine thiocyanate (Sigma, cat. #G9277)) was distributed into 6 of the reagent wells of the cartridge. Once rehydrated in the final 100 uL volume after processing by the analyzer, the normal 1× hybridization buffer (3×SSC (0.45 M NaCl, 0.045 M Na citrate), 0.18% cetrimide, 0.77% CHAPSO, 0.72% SB3-12, and 0.13M guanidine thiocyanate) will be achieved. 1.8 μL of each target species-specific DNA oligonucleotide FISH probe and unlabeled DNA helper probe mixture was added to 2 out of 8 of the reagent wells (N=2 for each target in 1 cartridge). E. coli FISH oligonucleotide probe sets were added to reagents wells corresponding to cartridge location A1 and A2, K. pneumoniae probe sets were added to reagents wells corresponding to cartridge location A3 and A4 and P. aeruginosa probe sets were added to reagents wells corresponding to cartridge location A5 and A6. These cartridge wells containing hybridization buffer and specific probes were then incubated in a 50° C. convection oven for 16-20 hours to dehydrate the materials.
Preparation of Magnetic Particles: Poly-aspartic acid-conjugated magnetic particles (Fluidmag-PAA, Chemicell, cat. 4108) were diluted 1:20 into 50 mM Epps buffer, pH 8.2 to a concentration of 2.75×1012 particles/mL with a final concentration of 10% w/v Trehalose (Sigma, cat. #T9449). To this dilution, fluorescent magnetic microspheres containing a green dye (Dragon Green Fluorescent Microspheres, BANGS Laboratories, cat. MEDG001) were added to the suspension at a final concentration of 3×106 particles/mL. The magnetic particle mixture was sonicated for 1 minute prior to immediately use to minimize aggregation. The mixture was then lyophilized in 10 μL volume beads (2.64×1012 PAA particles per reaction) and 1 bead was placed in each of the 6 reagent wells.
Preparation of Cultures: Log cultures of three different target pathogens (E. coli ATCC 25922, K. pneumoniae ATCC 13883, and P. aeruginosa ATCC 27853) were grown separately with three to five colonies inoculated into 5 mL of Tryptic Soy Broth (TSB, Hardy Diagnostics cat. U65) and incubated while shaking for 1-2 hours at 35° C. The Optical Density was measured by a spectrometer and the organisms were diluted to about 5×106 CFU/mL in 1× Cation-adjusted Mueller-Hinton Broth (MHBII, Teknova cat. M5860).
Bacterial Cell Labeling and imaging for Identification: Assay signal was determined for each target pathogen in contrived polymicrobial mixture containing two bacteria of interest (3 total 2-bacteria combinations) in a final concentration of 30% processed urine. Each polymicrobial bacterial combination was tested in 10 unique different culture negative clinical samples (30 urines tested in total). 103.5 μL of bacterial target A (˜5×105 CFU/mL per reaction) 103.5 μL of bacterial target B (˜5×105 CFU/mL per reaction), 360 μL urine, and 633 μL were combined for a total volume of 1.2 mL; 1 mL of that mixture was transferred to the cartridge sample addition port. The cartridge was then placed on the instrument and all subsequent actions were automatically performed. The sample was first directed under vacuum into the 6 growth wells at the top of the cartridge. Sample was then immediately moved to reaction wells, rehydrating the hybridization buffer/FISH probe mix and lyophilized magnetic particles. Sample then continued to the optical windows containing 45 μL of dehydrated “dye cushion” (50 mM TRIS pH 7.5 (Teknova, cat. T5075), 7.5% v/v Optiprep (Sigma, cat. D1556), 5 mg/mL Direct Black-19 (Orient, cat. #3222), dried for 60° C. for 3 hours in a convection oven) and incubated at 35° C. for 30 minutes on the analyzer. After this incubation, the cartridges were relocated to the magnet station, and placed atop a strong permanent magnet (Dexter magnetic technologies, cat. 54170260) for 4 minutes to bring the labeled and magnetic-particle-interacting bacterial cells into proximity to the imaging surface at the bottom of the wells. Finally, the cartridge was moved to the imaging station and imaging taken using non-magnified CCD imager as described below. In brief, focusing on each individual well was performed by taking successive images of the fluorescent magnetic microspheres in the green channel, the plane of focus determined, and a corresponding image at that location taken in the red color channel to image labeled bacterial cells.
Imaging of labeled cells: The MultiPath™ Analyzer imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a MultiPath Cartridge as part of a fully automated test. It uses a custom designed precision 3 axis positioning system to locate each well over a fluorescence-based image acquisition subsystem. The Analyzer can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire Cartridge Imaging Well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. 16 frames were captured at a 100 msec exposure using 635/25 nm excitation and 667/30 nm emission filters. The focus particles are imaged at 470/40 nm excitation and 520/40 nm excitation filters and captured 2 frames at a 20 msec exposure.
Data Analysis: Using the image captured by the CCD camera, detected cells was estimated by an algorithm that looked at both number of objects in the field of view and the intensity of the objects. Signal in a channel was considered detected if assay signal was above 130.
Results. The data demonstrate successful identification of 2 target pathogens in a single sample with no detection of the pathogen that is absent (i.e. no cross reactivity of the FISH probes to the non-target bacteria).
Conclusions. This example demonstrates the inventive isothermal FISH method performed on an automated analyzer with stabilized reagents inside a consumable cartridge can specifically identify multiple target bacterial species in a contrived urine sample. This shows the potential of the method to identify multiple pathogens in polymicrobial infections. The example also demonstrates the specificity of the method, as no cross-species detection was observed.
Variations. This example is illustrative of the performance of this novel FISH method on a cartridge and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different probe sequences and nucleic acid structures (PNA, LNA, etc.), alternative assay chemistries (different detergents, chaotropes, fluorophores, buffers, pH, temperatures, reaction times, component concentrations), concentration of urine and urine processing procedures and alterations to reactant stabilization (lyophilization of components). This methodology can also clearly be extended to other biological specimens and to other bacterial and non-bacterial pathogens.
Overview. In this example, large area imaging was used to detect individual S. aureus bacterial cell targets that were stained with fluorogenic esterase substrates. The substrates can diffuse through cell membranes of intact living cells where they become both fluorescent and charged when acted upon by esterase enzymes found in metabolically active cells. These charged fluorescent products can no longer passively diffuse through cell membranes and become trapped in intact cells. This technique can thus distinguish live cells from dead cells, as only cells with active esterases and intact cell membranes will stain properly. In this example, S. aureus cells are labeled with the fluorogenic substrate carboxy-fluorescein diacetate (cFDA) and imaged using non-magnified digital imaging.
Experimental MethodsBacterial Cell Preparation:
S. aureus ATCC 29213 was grown overnight in Tryptic Soy Broth (TSB, BD cat. #211822). The log culture of S. aureus was made by inoculating 100 uL of overnight culture into 5 ml of fresh TSB media and further incubating for 2.5 hours at 35° C. in a shaking incubator.
Preparation of Magnetic Particles:
Antibody conjugated magnetic particles were made by coupling magnetic particles (Ademtec, 292 nm) to chicken anti-protein A antibodies (Meridian Biosciences) using standard EDAC (1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide) coupling chemistry.
Labeling and Capture of Bacterial Cells:
The assays were carried out in a 96-well microtiter plate and each well included 40 uL of S. aureus (10,000 cells in TSB) or only TSB (no cell control), 5 uL of 10 mM cFDA (Life Technologies) and 5 ul of antibody-conjugated magnetic particles (2e10/mL). The reactions were incubated at room temperature for 15 min. After the incubation, 40 uL of reaction mixture was carefully overlaid on 75 ul of “dye cushion” (15% Optiprep with 5 mg/mL Chromotrope 2R) which was pre-aliquoted in black, clear bottom half area microtiter plate (Greiner 675096, VWR part #82050-056). The microtiter plate was placed onto a magnetic field for 4 minutes to bring magnetic particles, a fraction containing labelled cells, through the “dye cushion” and into proximity to the imaging surface at the bottom of the wells.
Imaging of Bacterial Cells:
Following magnetic capture of labeled cell:magnetic particle complexes, the microtiter plate was placed on a stage above a CCD digital camera (IDS, model UI-2250SE-M) and illuminated with light from LEDs passing through an optical filter (469 nm, 35 nm FWHM). Fluorescent signal passing through an emission filter (520-35 nm) was detected by the camera to create an image of the fluorescent complexes. The images were analyzed using FLimage software (First Light Biosciences) that enumerates the individual cells.
Results.
Conclusions. This example demonstrates a method for enumerating bacteria non-specifically. This approach can be used to count the total number of cells from a mixed population covering a broad range of bacterial species in a specimen. Specifically, the example demonstrated the capability of the inventive non-magnified imaging method to enumerate small numbers bacterial cells labeled with carboxy-fluorescein diacetate, a fluorescence substrate that non-specifically labels viable cells containing the ubiquitous esterase enzymes.
Variations. There are other stains for non-specific cell labeling including nucleic acid stains such as STYO and SYBR family of stains, propidium iodide, and DAPI. Other stains distinguish live or dead cells can be used. For example, other fluorogenic substrates, or DNA stains that can or cannot cross intact cell membranes can be used instead of or in conjunction with cFDA or FDA. Multiple stains and dyes can be distinguished by using multiple excitation and emission wavelengths for fluorescence detection. The spectrum of fluorescence associated with an object can be used to determine whether a cell is counted as live or dead. In addition, fluorogenic substrates that are specific for the biochemical activity of a particular type of bacteria can be used to determine its presence. For example, a fluorogenic □-galactosidase substrate can be cleaved to its fluorescent product by □-galactosidase, which is specific to coliforms. This methodology can also be applied to most bacteria and to specimens that contain multiple bacterial species. The method is suitable to detect bacteria in many different clinical specimen types with minimal processing (e.g. urine, sputum, swabs, spinal fluid, etc.).
Example 15: Non-Specific Detection of Bacteria Using DNA Staining DyesOverview. In this example, large area imaging was used to detect individual live S. aureus bacterial cell targets that were stained with DNA binding stains. The DNA binding dyes can diffuse through the cell membrane of intact living cells where they become highly fluorescent after binding to DNA of the bacteria. Once bound to DNA, these dyes can no longer easily and passively diffuse through intact cell membranes and become trapped in intact cells. This technique can be useful when it is important to distinguish live cells from dead cells, as live cells with intact cell membranes will stain differently with different dyes compared to dead cells or cells with compromised membranes. In this example, S. aureus is labeled with DNA binding fluorescent dye, SyBR Green and imaged using non-magnified large area CCD imaging.
Experimental MethodsBacterial Cell Preparation:
S. aureus ATCC 29213 was grown overnight in Tryptic Soy Broth (TSB, BD cat. #211822). The log culture of S. aureus was made by inoculating 100 uL of overnight culture into 5 ml of fresh TSB media and further incubating for 2.5 hours at 35° C. in a shaking incubator.
Preparation of Magnetic Particles:
Antibody conjugated magnetic particles were made by coupling magnetic particles (Ademtec, 292 nm) to chicken anti-protein A antibodies (Meridian Biosciences) using standard EDAC (1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide) coupling chemistry.
Labeling and Capture of Bacterial Cells:
The assays were carried out in a 96-well microtiter plate and each well included 40 uL of S. aureus (10,000 cells in TSB) or only TSB (no cell control), 5 uL of 500-fold diluted SyBR Green dye (Cat #S7563, Life Technologies) and 5 ul of antibody-conjugated magnetic particles (2e10/mL). The reactions were incubated at room temperature for 15 min. After the incubation, 40 uL of reaction mixture was carefully overlaid on 75 ul of “dye cushion” (15% Optiprep with 5 mg/mL Chromotrope 2R) which was pre-aliquoted in black, clear bottom half area microtiter plate (Greiner 675096, VWR part #82050-056). The microtiter plate was placed onto a magnetic field for 4 minutes to bring magnetic particles, a fraction containing labelled cells, through the “dye cushion” and into proximity to the imaging surface at the bottom of the wells.
Imaging of Bacterial Cells:
Following magnetic capture of labeled cell:magnetic particle complexes, the microtiter plate was placed on a stage above a CCD digital camera (IDS, model UI-2250SE-M) and illuminated with light from LEDs passing through an optical filter (469 nm, 35 nm FWHM). Fluorescent signal passing through an emission filter (520-35 nm) was detected by the camera to create an image of the fluorescent complexes. The images were analyzed using FLimage software (First Light Biosciences) that enumerates the individual cells.
Results.
Conclusions. This large area, non-magnified imaging system is capable of detecting and enumerating bacterial cells that are labeled with SyBR Green, a DNA binding dye that non-specifically labels living cells.
Variations. Other dyes that distinguish live or dead cells can be used. For example, other fluorogenic DNA stains that can or cannot cross intact cell membranes can be used instead of or in conjunction with SyBR Green. Multiple stains and dyes can be distinguished by using multiple excitation and emission wavelengths for fluorescence detection. The spectrum of fluorescence associated with an object can be used to determine whether a cell is counted as live or dead. In addition, fluorogenic substrates that are specific for the biochemical activity of a particular type of bacteria can be used to determine its presence. For example, a fluorogenic □-galactosidase substrate can be cleaved to its fluorescent product by □-galactosidase, which is specific to coliforms. This methodology can also be applied to most bacteria and to specimens that contain multiple bacterial species. The method is suitable to detect bacteria in many different clinical specimen types with minimal processing (e.g. urine, sputum, swabs, spinal fluid, etc.). Other nucleic acid stains can label bacterial cells non-specifically, including the members of the SYTO/SYBR family of stains, propidium iodide, and DAPI.
Example 17. Automated Sensitive Detection of C. difficile Toxin B in Stool Specimens Using the Inventive SystemOverview. C. difficile causes more hospital acquired infections and patient deaths than any other pathogen and is the top pathogen on CDC's urgent threat list. The two current laboratory methods for diagnosing C. difficile infections are inaccurate. Enzyme immunoassay tests for C. difficile infection lack clinical sensitivity, that is they can fail to detect patients that have the disease. Nucleic acid amplification tests, lack clinical specificity—these tests misdiagnose patients that do not have the disease as positive for infection. A more sensitive test for the C. difficile toxin that causes the infection can be both highly sensitive and highly specific. A more accurate test will lead to better patient outcomes. This example demonstrates the use of the invention to detect very low concentrations of C. difficile toxin B in stool specimens.
Experimental MethodsMaterials. 2 monoclonal antibodies that bind complementary epitopes of the C. difficile Toxin B protein were attached to nanoparticles. Fluorescent nanoparticles (Thermo Fisher Scientific, Waltham, Mass.) were conjugated to anti-C. difficile Toxin B monoclonal antibodies (BBI Solutions, Cardiff, UK). Polystyrene carboxylate magnetic particles (Ademtech, Pessac, France) were conjugated to anti-C. difficile Toxin B monoclonal antibodies (Fitzgerald, Acton, Mass.). Both fluorescent and magnetic particles were lyophilized after conjugation. Lyophilized particles are placed into the First Light cartridge during assembly. Native Toxin B protein purified from C. difficile was purchased from List Laboratories (Campbell, Calif.). Casein, Casein hydrosylate, Trizma®-HCl were from Sigma-Aldrich (St. Louis, Mo.). Poly-BSA was from Roche. Protease inhibitor cocktail was from Takara Bio (Mountain View, Calif.). Spin columns were purchased from Pierce/Thermo-Fisher Scientific.
Estimating the limit of detection of the C. difficile toxin B test on MultiPath Instrument. The LoD measurement was performed using the pooled negative stool sample. The limit of detection (LoD) was determined in accordance to approved Clinical & Laboratory Standards Institute (CLSI) guidelines by running 24 replicates of sample with no analyte and 12 replicates each with 5 different toxin B concentrations. A pooled negative stool sample was made from 14 individual stool samples that had been scored as C. difficile negative by real-time PCR. The pooled stool samples were spiked with C. difficile toxin B in a series of two-fold dilutions (0, 31.2, 62.5, 125, 250, 500 pg/mL). 100 μL of each stool sample was added to stool diluent (900 μL) consisting of Tris buffer, Poly-BSA, caseins and protease inhibitor cocktail. 0.95 mL of each diluted sample was transferred to a Pierce spin column and centrifuged at 11,700×g for 5 minutes. After centrifugation, 700 μL of the supernatant was transferred to the sample addition port in the cartridge and the cap was closed. Then cartridge was then placed into the cartridge input rack and inserted into the instrument.
Running cartridges on an automated instrument. After the cartridge was placed into the instrument, all subsequent actions other than data analysis (performed offline using Excel or JMP software) were automatically performed. The diluted stool samples were first directed under vacuum into individual reaction wells within the cartridge and moved to the imaging windows containing 46 uL of dehydrated “dye-cushion” (50 mM TRIS pH 7.5 (Teknova, cat. T5075), 7.5% v/v Optiprep (Sigma, cat. D1556), 5 mg/mL Direct Black-19 (Orient, cat. #3222), dried for 60° C. for 3 hours in a convection oven) and incubated at 35° C. for 30 minutes on the instrument. After this incubation, the cartridges were then relocated to the magnet station, and placed atop a strong permanent magnet (Dexter magnetic technologies, cat. 54170260) for 4 minutes to bring the fluorescent particle:toxin B:magnetic particle complexes through the “dye cushion” and into proximity to the imaging surface at the bottom of the wells. Finally, the cartridge was moved to the imaging station and images taken using the non-magnified CCD imager described as described below.
The instrument imaging system and imaging process. The MultiPath imaging system is a custom-built instrument and software that is capable of automatically capturing image data from selected wells of a MultiPath Cartridge as part of a fully automated test. It uses a custom designed precision 3 axis positioning system to locate each well over a fluorescence-based image acquisition subsystem. The Instrument can image in 4 separate color channels and uses an objective lens, illumination LEDs, fluorescent filter sets, and camera. The objective lens has a field of view designed to capture the image of an entire Cartridge Imaging Well. The illumination module light source consists of 2 high power LEDs per color channel. A series of fluorescent image frames are captured with a camera using a 3.1 MP Sony IMX265 monochrome sensor with 12-bit per pixel quantization. The final image for each well is then formed by summing multiple frames. For the C. difficile toxin tests, the test channel is 470/40 nm excitation and 520/40 nm emission filters and captured 2 frames at a 20 msec exposure. The focus particles are imaged at 569/25 nm excitation and 609/34 nm excitation filters and captured 2 frames at a 10 msec exposure.
Results.
Conclusions. This method is capable of very sensitive and precise detection of C. difficile Toxin B. The LoD was determined to be 58 pg/mL of toxin B. The low signal variability across technical and biological replicates indicates robustness to matrix effects.
Variations. This example is illustrative of the performance of this inventive method and is not limited to the specific details included in the description. One skilled in the art will readily understand that many variations are therefore possible, including using different fluorescent particles, alternative assay chemistries (different buffers, pH, temperatures, reaction times, component concentrations), different amounts of stool and different means of processing stool specimens. In addition, alternative biomarkers specific to C. difficile could be used (e.g. Toxin A). This novel technology can also clearly be extended to other target molecules as well as various bacterial and non-bacterial pathogens for which a specific biomarker can be identified.
INCORPORATION BY REFERENCEReferences and citations to other documents, such as patents, patent applications, patent publications, journals, books, papers, web contents, have been made throughout this disclosure. All such documents are hereby incorporated herein by reference in their entirety for all purposes.
EQUIVALENTSVarious modifications of the invention and many further embodiments thereof, in addition to those shown and described herein, will become apparent to those skilled in the art from the full contents of this document, including references to the scientific and patent literature cited herein. The subject matter herein contains important information, exemplification and guidance that can be adapted to the practice of this invention in its various embodiments and equivalents thereof.
Claims
1. An instrument comprising:
- a conveyance system operable to move a cartridge containing a specimen to modules within the instrument;
- a magnetics module that pulls magnetic particles within the cartridge onto an imaging surface in the cartridge;
- an imaging module with a sensor that receives light from the imaging surface; and
- a control module that detects and counts a number of labelled molecules or cells bound to the magnetic particles.
2. The instrument of claim 1, wherein the control module and imaging module are able to detect and count fluorescently-labeled individual cells or molecules on the imaging surface without magnification.
3. The instrument of claim 1, wherein the imaging module includes an XYZ stage operable to move the cartridge in x, y, and z directions for placement during imaging
4. The instrument of claim 1, wherein the control module operates the XYZ stage to count individual cells in a series of different imaging wells within the cartridge.
5. The instrument of claim 1, wherein the imaging module includes an optics assembly mounted below a deck that supports the cartridge over the optics assembly, wherein the optics assembly includes one or more of the sensor, a light source, a lens, and a filter wheel.
6. The instrument of claim 1, further comprising a fluidics module that uses pressure to divide the specimen from an input well on the cartridge into division wells in the cartridge.
7. The instrument of claim 1, further comprising at least one heater block on the deck that can maintain a cartridge at a controlled temperature.
8. The instrument of claim 1, wherein the magnetics module has a magnet disposed on a bottom surface of a cartridge slot.
9. The instrument of claim 1, wherein the mechanical conveyance system includes a carousel and a mechanical arm that transfers cartridges between the carousel and the modules
10. The instrument of claim 1, wherein the control module reads a code from the cartridge and analyzes the counted labelled molecules or cells to provide a result for a test corresponding to the code.
11. The instrument of claim 10, wherein the test is a microbial identification test.
12. The instrument of claim 10, wherein the test is an antimicrobial susceptibility test and the control module analyzes differential growth among wells of the cartridge.
13. The instrument of claim 12, wherein the cartridge comprises a plurality of wells comprising different antimicrobials or different concentrations of one or more antimicrobials.
14. The instrument of claim 1, further comprising an incubation station operable to maintain the cartridge contents at a desired incubation temperature when the cartridge is positioned therein.
15. The instrument of claim 1, further comprising a waste module for cartridge disposal after performance of the test steps.
16. The instrument of claim 9, wherein the carousel is enclosed and the instrument is operable to maintain the carousel at an incubation temperature required for performing the test steps.
17. The instrument of claim 9, wherein the carousel and the modules comprise slots sized to accommodate the cartridge,
- wherein rotation aligns a slot of the carousel to a module slot, and
- wherein the mechanical arm is operable to push the cartridge between the carousel and the module slot.
18. The instrument of claim 1, wherein the control module comprises a processor coupled to a non-transitory, tangible memory and operable to receive an input designating a test to be performed on the specimen in the cartridge and control the instrument to perform the test.
19. The instrument of claim 18, wherein performing the test includes:
- accessing a routine of test steps from the non-transitory tangible memory corresponding to the test to be performed;
- accessing existing scheduled routines for cartridge tests already being performed within the instrument;
- scheduling the routine of test steps for the test to be performed on the cartridge avoiding conflicts with the existing scheduled routines; and
- executing the routine of test steps on the cartridge within the instrument by rotating the carousel and transferring the cartridge between one or more of the plurality of stations corresponding to the routine of test steps.
20. The instrument of claim 18, wherein the input designation designating the test to be performed is received from a scanning device in communication with the processor and operable to read a tag on the cartridge indicative of the test to be performed.
Type: Application
Filed: Oct 4, 2019
Publication Date: Nov 11, 2021
Inventor: Don Straus (Cambridge, MA)
Application Number: 17/282,627