Metabonomic characterization of microorganisms

Abstract

The invention relates to metabonomic characterization of microorganisms, in particular to a method for identifying cell types of cells in a sample, comprising: culturing the sample in a growth medium comprising nicotinamide to obtain a cultured growth medium; analyzing the cultured growth medium by chemical analysis; and identifying the cell type as at least one of the genus Escherichia, the genus Klebsiella, the genus Pseudomonas, the genus Enterococcus, the genus Staphylococcus or the genus Streptococcus when the cultured growth medium contains a higher concentration of nicotinate than the growth medium; wherein the growth culture medium comprises the following components in a buffer culture medium: 0.02 to 2 g / l of D-glucose, 0.0001 to 0.1 g / l of nicotinamide, 0.0001 to 0.1 g / l of pyridoxin.HCl, 0.02 to 1 g / l of hypoxanthine, 0.2 to 5 g / l of L-arginine, 0.2 to 5 g / l of L-threonine, 0.1 to 5 g / l of L-histidine, 0.02 to 2 g / l of spermine and catalase. According to the method, identification and toxin sensitivity detection can be completed within about half of time compared with that of an existing method.

Description

[0001] This application is a divisional application of a Chinese patent application with application number 201980098971.5, application date September 20, 2019, and invention title "Metabolomic Characterization of Microorganisms". Technical Field

[0002] The present invention relates to devices, methods, and systems for detecting metabolites consumed and produced in order to classify organisms and measure the sensitivity of organisms to toxic substances. Background Art

[0003] The timely identification of cells is useful in many applications. For example, rapid microbial identification is of great value when considering food safety, genetic engineering research, recombinant verification, microbial detection, and disease treatment.

[0004] Considering disease treatment, especially bloodstream infections, for example, the length of time between the onset of symptoms and the initiation of effective antibiotic treatment for a patient is a major factor in the morbidity and mortality of the infection. In the case of bloodstream infections, within the first 6 hours, the survival rate drops from 80% to 72%, and as the infection worsens, the survival rate continues to decline ( Figure 1 ).

[0005] In current practice, such as sample analysis using chemical detection or spectrometry using, for example, matrix-assisted laser ionization desorption mass spectrometry (MALDI-MS), it takes 2 - 4 days to identify an unknown organism and determine its drug sensitivity level considering the culture time and analysis ( Figure 2 ). Summary of the Invention

[0006] The present invention provides methods and systems for identifying live cells such as microorganisms. By changing the concentration of metabolites consumed and / or produced present in the growth medium, the methods and systems can also be used to measure the sensitivity of microorganisms or other cells to toxic substances. The present invention can complete identification and toxin sensitivity detection in approximately half the time compared to current methods.

[0007] In the present application, the term "metabolite" refers to any substance used in cell metabolism or any substance produced by cell metabolism. Thus, metabolites include nutrients consumed by live cells and waste products produced. Sometimes, the terms nutrients and precursors are used to refer to metabolites consumed in cell metabolism.

[0008] In addition, in the present application, the terms cell, organism, microorganism, microbe, pathogen, and bacterium are used interchangeably. These terms refer to one or more microscopic organisms visible with a microscope, which may include any bacterium, fungus, protozoan, or other isolated living cells such as cell suspensions (i.e., isolated cells, tissue cultures, etc.). Thus, it should be noted that the present invention is capable of being used to identify living cells metabolizing in culture, e.g., bacteria, fungi, protozoa, or isolated cells from ex vivo tissues or tissue cultures, which are collectively referred to herein as cells or microorganisms. For clarity, the present invention is used to analyze living cells, such as: identifying cells causing an infection and their response to toxins, identifying cells contaminating food or cancer cells, e.g., their response to toxins (such as chemotherapy).

[0009] In the present application, a "toxin" is any substance that regulates the metabolic activity of a cell. This can be a substance that kills the cell as well as a substance that impairs or stimulates cell function, such as any one or more metabolic pathways. Thus, cell function can be stopped or modified to cause a detectable change in metabolic consumption or output.

[0010] In addition, in the present application, "medium" (and / or) "media" is any liquid-based or solid-based nutrient for cell growth, maintaining dormant cells, and / or promoting, inhibiting, maintaining cell metabolism. In this context, defined media and natural media are explored. While both defined media and natural media can be complex and rich, defined media have a substantially known composition, and natural media do not have a defined composition and are typically extracts of nutrient sources, such as animal, microbial, or plant extracts. While the composition of natural media may be generally known, it can vary from batch to batch. In this context, defined media specifically exclude animal, microbial, or plant extracts that have not been purified into a single compound.

[0011] According to a broad aspect of the present invention, there is provided a method for identifying the cell type of cells in a sample.

[0012] According to a broad aspect of the present invention, there is provided a method for identifying the cell type of cells in a sample, comprising:

[0013] culturing the sample in a growth medium comprising nicotinamide to obtain a cultured growth medium;

[0014] analyzing the cultured growth medium by chemical analysis; and

[0015] When the cultured growth medium contains a higher concentration of nicotinate than the growth medium, the cell type is identified as at least one of the genus Escherichia, genus Klebsiella, genus Pseudomonas, genus Enterococcus, genus Staphylococcus or genus Streptococcus;

[0016] Wherein, the growth medium comprises 0.02 - 2 g / l D-glucose, 0.0001 - 0.1 g / l nicotinamide, 0.0001 - 0.1 g / l pyridoxine·HCl, 0.02 - 1 g / l hypoxanthine, 0.2 - 5 g / l L-arginine, 0.2 - 5 g / l L-threonine, 0.1 - 5 g / l L-histidine, 0.02 - 2 g / l spermine and catalase in a buffered medium.

[0017] According to another broad aspect of the present invention, there is provided a method for identifying the cell type of cells in a sample, comprising: culturing the sample in a growth medium to obtain a cultured growth medium; analyzing the cultured growth medium by chemical analysis; and when the cultured growth medium contains a higher concentration of N1,N8-diacetylspermidine than the growth medium, identifying the cell type as at least one of Enterococcus faecalis, Staphylococcus saprophyticus or Staphylococcus epidermis.

[0018] According to another broad aspect of the present invention, there is provided a method for identifying the cell type of cells in a sample, comprising: culturing the sample in a Mueller Hinton growth medium to obtain a cultured growth medium; analyzing the cultured growth medium by chemical analysis; and when the cultured growth medium contains a higher concentration of N1,N12-diacetylspermine than the growth medium, identifying the cell type as Enterococcus species.

[0019] According to another broad aspect of the present invention, there is provided a method for identifying the cell type of cells in a sample, comprising: culturing the sample in a growth medium to obtain a cultured growth medium; performing analysis by mass spectrometry to determine whether the cultured growth medium contains N-acetyl leucine, N-acetyl isoleucine or a biomarker with a retention time of 4.3 minutes and a mass of 286.2 in a 15-minute HILIC method; and, (a) if N-acetyl leucine or N-acetyl isoleucine is present in the cultured growth medium, identifying the cell type as Candida freundii, (b) if the biomarker is present in the cultured growth medium, identifying the cell type as Candida albicans.

[0020] According to another broad aspect of the present invention, there is provided a growth medium comprising: 0.5 to 1.5 mM glucose, histidine, nicotinamide, hypoxanthine, threonine, spermine and arginine as metabolic precursors, pyridoxine and catalase, for culturing a sample to identify pathogens in the sample.

[0021] According to another broad aspect of the present invention, there is provided the use of a growth medium for identifying pathogens in a sample, the growth medium comprising 0.02 - 2 g / l D-glucose, 0.0001 - 0.1 g / l nicotinamide, 0.0001 - 0.1 g / l pyridoxine hydrochloride, 0.02 - 1 g / l hypoxanthine, 0.2 - 5 g / l L-arginine, 0.2 - 5 g / l L-threonine, 0.1 - 5 g / l L-histidine, 0.02 - 2 g / l spermine and catalase in a buffered medium, wherein, after culturing the sample, chemical analysis of the growth medium identifies pathogens from at least the following species: Escherichia species, Klebsiella species, Pseudomonas species, Enterococcus species and Candida species.

[0022] According to another broad aspect of the present invention, there is provided a method for identifying the toxin sensitivity of a pathogen in a sample, comprising: culturing the sample in a growth medium to obtain a cultured growth medium; analyzing the cultured growth medium by chemical analysis, and if the cultured growth medium contains mevalonic acid, identifying the pathogen as Staphylococcus aureus; culturing the pathogen in a toxin-containing growth medium, which is known to have an effect on Staphylococcus aureus; and analyzing the cultured toxin-containing growth medium for glucose consumption by chemical analysis in order to determine whether Staphylococcus aureus is resistant to the toxin.

[0023] According to another broad aspect of the present invention, there is provided a method for analyzing a biological sample to identify a pathogen therein, the method comprising: culturing the sample in a first medium to promote metabolism for pathogen identification; simultaneously, culturing the sample in a plurality of toxin-containing media, each medium having a toxin directed against a different pathogen; after culturing, analyzing the first medium to obtain a metabolic result for pathogen identification; and analyzing only the selected toxin-containing medium from the plurality of toxin-containing media, the selected toxin-containing medium being selected as having a relevant toxin directed against the identified pathogen. There can be a non-transitory computer-readable medium for implementing the method, which stores instructions executed by one or more processors.

[0024] It should be understood that other aspects of the present invention will become apparent to those skilled in the art from the following detailed description, in which various embodiments of the present invention are shown and described by way of example. As will be realized, the present invention is capable of other and different embodiments, and several details of its design and implementation can be modified in various other aspects, all of which are included within the scope of the claims of this application. Accordingly, the detailed description and embodiments should be regarded as illustrative rather than restrictive. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] For a better understanding of the present invention, the following drawings are attached:

[0026] Figure 1 Shows the microbiological detection time and the probability of death of a patient caused by infection. Survival data from the onset of symptoms to administration of antibiotics are shown. Resistance refers to the antibiotic sensitivity test time, and ID refers to the microbial identification using MALDI-MS.

[0027] Figure 2, (A) shows the clinical workflow of current medical practice for identifying unknown organisms and their toxin sensitivities. The first 1 - 2 days of culturing await the growth of bacteria to a detectable density. Then, MALDI - MS analysis is completed to identify the unknown organism. A specimen of the culture is inoculated into a new culture, and the antibiotic susceptibility test (“AST”) is completed by culturing the unknown organism in several antibiotics at a series of drug doses. (B) shows a possible timeline for one embodiment of the present invention. The unknown organism is combined with a growth medium containing nutrients and incubated for 4 hours. Metabolite analysis is performed using an MS instrument. The antibiotic susceptibility test is completed during the second stage of incubation (“Inc+anti”). The unknown organism identified in the first stage is combined with a controlled dose of antibiotic in fresh growth medium and incubated. MS is used to evaluate the metabolites present in the growth medium to determine the effective antibiotic dose.

[0028] Figure 3 Shows the detection limits based on MS and based on optics.

[0029] Figure 4A Shows a flowchart of a method for identifying an unknown organism and determining its sensitivity to toxins for an example.

[0030] Figure 4B Schematically illustrates a device according to the present invention.

[0031] Figure 4C Shows another embodiment of the device according to the present invention.

[0032] Figure 5 Shows metabolite - based identification of 7 unknown organisms. Heatmap (plotted as Z - scores) of biomarkers selected from over 250 metabolites observed in the MS spectra. Biomarkers are plotted before and after incubation in rich nutrient medium for 4 hours.

[0033] Figure 6A Shows selected biomarkers, measured by MS, present in the microbial culture medium. Cultures are standardized using 0.5 McFarland dilutions of common pathogens and commercially available organisms. Sample symbol legend: Candida spp., Ca, Cd, Cg, Ck, Cp; Escherichia coli, EC; Klebsiella oxytoca, KO; Klebsiella pneumoniae, KP; Pseudomonas aeruginosa, PA; Pseudomonas putida, Pp; Staphylococcus aureus, SA; Enterococcus faecalis, EF; Streptococcus pneumoniae, SP; Streptococcus group A, SG; Streptococcus viridans, SV; and, coagulase negative Staphylococcus, SN.

[0034] Figure 6B is supportive of Figure 6A heat map of the extended results. Sample symbol description: Candida spp., C.ssp; Candida albicans, C.alb; Escherichia coli, E.col; Klebsiella oxytoca, K.oxy; Klebsiella pneumoniae, K.pne; Pseudomonas aeruginosa, P.aer; Enterococcus faecium, E.fae; Staphylococcus aureus, S.aur; Streptococcus pneumoniae, S.pne; Group A Streptococcus, GAS; Streptococcus viridans, S.vir; Coagulase-negative staphylococcus, C(-)S.

[0035] Figure 7 shows the selected metabolite levels observed in the growth media of 100 cultures of bacterial clinical isolates. Each culture sample was inoculated with 10 8 bacteria / ml and incubated for 4 hours. Then, mass spectrometry was used to identify the metabolite levels of the target organisms.

[0036] Figure 8 shows the selected biomarkers detected in spent blood culture bottles of a clinical diagnostic laboratory measured by MS.

[0037] Figure 8A , (A) shows the original mass spectrometer dataset (extracted ion chromatogram), which shows the signal of succinate observed in Mueller-Hinton medium after incubation for 4 hours under conditions where microorganisms are present. This figure depicts the diagnostic differences between the signals of Escherichia and Klebsiella and the signals found in the other 9 microorganisms. (B) The mass spectrometry intensity of succinate depicted as a box plot. This figure depicts the same data as shown in Figure 8A (A), but in a processed form.

[0038] Figure 9 shows the biomarker groups of sensitive and resistant strains of 3 target organisms. Drug doses are listed in μg / ml. A computer model using these biomarkers successfully distinguished the organisms.

[0039] Figure 10 shows the metabolic detection of antibiotic resistance by mass spectrometry. Biomarker levels were evaluated in sensitive and carbapenem-resistant Klebsiella pneumoniae at a series of meropenem doses. This data shows toxin-induced metabolic changes that distinguish drug-sensitive and resistant isolates. This data is Figure 9 a subset of the larger dataset shown.

[0040] Figure 11 shows the possible contributions of human serum and cells to biomarker signals. 1% human whole blood was added to the samples and incubated for 4 hours.

[0041] Figure 12 Shows computer prediction of drug susceptibility. Changes in drug-induced biomarker levels were recorded in 36 clinically relevant strains (three species with sensitive and resistant isolates, 6 replicates; Escherichia coli + / - extended-spectrum beta-lactamase, Klebsiella pneumoniae + / - carbapenem resistance, and Staphylococcus aureus + / - methicillin resistance). These biomarker levels were used to create a database training set. The computer model predicted drug resistance using the biomarker levels of 18 samples.

[0042] Figure 13 Shows the use of 1 Selected biomarkers detected in waste blood culture bottles of a clinical diagnostic laboratory by 1H NMR. In this example, the patient had a Pseudomonas aeruginosa bloodstream infection. The metabolic action of the pathogen depleted nutrients from the growth medium.

[0043] Figure 13A , using NMR to analyze the sugar monomers (sugars and disaccharides) in two types of growth media after incubating common bacteria. The figure shows the diagnostic metabolite signals observed in the two growth media detected by multidimensional (1H-13C) nuclear magnetic resonance spectroscopy (NMR). The diagnostic NMR regions (100 mM) corresponding to each of the 7 target sugars were shown in the growth media as well as in the pure standard solutions. Bacterial isolates were grown for 4 hours in BacT blood medium (BioMérieux) or Mueller-Hinton medium. The two media contained different carbohydrate nutrients, as shown by the bacteria-free control samples. The microorganisms grown in the two media produced a diagnostic pattern of metabolites that was sufficient to distinguish the microorganisms (e.g., the presence of sucrose after incubation in BacT medium distinguished Escherichia coli from Klebsiella pneumoniae).

[0044] Figure 14 Shows the difference in drug susceptibility (Sen) and drug resistance (Res) isolates of Pseudomonas aeruginosa based on 1 1H NMR in the presence and absence of 60 μg / ml tetracycline (Tet). Metabolite biomarkers of drug efficacy were recorded together with DSS (internal standard).

[0045] Figure 15Shows metabolic detection of antibiotic resistance by light absorption method. A culture of Pseudomonas aeruginosa was prepared in optically neutral M9 medium. The microbial production of optically active Pseudomonas aeruginosa siderophores was detected by absorbance at 400 nM in the medium from which cells had been removed by centrifugation. The optical discrimination between drug-sensitive (Sen) and drug-resistant (Res) isolates of Pseudomonas aeruginosa in the presence of 0, 60, and 600 μg / ml tetracycline (Tet) is shown.

[0046] Figure 16 , a decision tree developed in Example IX and further refined in subsequent examples, for discriminating blinded patient blood samples incubated in Mueller-Hinton medium. The minimum fold change in metabolite concentration in italics represents the major decision branch points. The metabolites in gray are additional metabolites used to confirm the species. The metabolites shown in white in the black boxes are precursors of the metabolites shown. The gray species metabolic patterns are not yet confirmed. All metabolites used are generated with respect to Mueller-Hinton plus blood control unless specifically noted with a downward arrow. Solid arrows indicate that the designated biomarker meets the designated threshold change, while dashed arrows (thin struck through arrows) indicate that the change in the designated biomarker does not meet the minimum threshold listed in the figure. Considering the background noise to be zero, all thresholds are fold changes except for N1,N12-diacetylspermine (DAS).

[0047] Figure 17A and 17B , from Example IX, Figure 17A is a heatmap showing biomarkers validated using a total of 596 clinical isolates; and, in Figure 17B , the signals of the top 7 biomarkers are shown, which can strongly distinguish the 7 species studied.

[0048] Figures 18A (i) and (ii), from Example IX, data from a blinded clinical trial where 11 key biomarkers were used to differentiate species by metabolomics. The heatmap depicts the fold change in metabolite intensity in positive cultures (monocultures only) versus negative blood cultures relative to an uninfected Mueller-Hinton blood control (ii). In the performance and right hand infection columns, each bar shows the concordance between metabolomics and culture-based assignment, strong concordance shown as black, some concordance shown as dark grey, and discordance shown as light grey. The species in the training set portion are at the top and are separated from the species first observed in the blinded experiment.Abbreviations: Candida albicans, CA; Escherichia coli, EC; Klebsiella pneumoniae, KP; Pseudomonas aeruginosa, PA; Staphylococcus aureus, SA; Enterococcus faecium, EF; SP, Streptococcus pneumoniae; CANLUS, Candida lustiniae; CANGLA, Candida glabratta; BACIL, Bacillus; ODOSPL, Odoribacter splanchnicus; ACIN, Acinetobacter; PROMIR, Proteus mirabilis; SALPARa, Salmonella Paratypi A; KLEOXY, Klebsiella oxytoca; ENTCLOc, Enterobacter cloacae complex; MICRC, Micrococcus; PROP, Propionibacterium; BACFRA, Bacteroides fragilis; CORbac, Coryneformbacilli; STRBOVg, Streptococcus bovis group; STRANGg, Streptococcus anginosus group; STRVIRg, Streptococcus viridans group; LACTB, Lactobacillus species; CLOS, Clostridium species; GEMMOR, Gemella morbillorum; CNS, Coagulase-negative staphylococci, including (STAHOM, Staphylococcus hominis; STAWAR, Staphylococcus warneri; STAEPI, Staphylococcus epidermidis; STACAPT, Staphylococcus capitis); GAS, Group A Streptococcus; GGS, Group G Streptococcus; GBS, Group B Streptococcus.

[0049] Figures 18B(i) to (v), from Example IX, biomarker and pathogen data obtained by MS analysis in positive ion mode. Abbreviations: MHB, control; Candida albicans, CA; Candida glabrata, CGLA; Citrobacter freundii, CFRE; Citrobacter koseri, CKOS; Escherichia coli, EC; Enterobacter cloacae, ECLO; Enterobacter aerogenes, EAER; Klebsiella oxytoca, KOXY; Klebsiella pneumoniae, KP; Proteus mirabilis, PMIR; Proteus vulgaris, PVUL; Pseudomonas aeruginosa, PA; Stenotrophomonas maltophilia, SMAL; Streptococcus pneumoniae, SP; Streptococcus viridans, SVIR; Streptococcus pyogenes, GAS; Streptococcus group G, GGS; Streptococcus group B, GBS; Streptococcus group C, GCS; Aerococcus urinae, AURI; Aerococcus viridans, AVIR; Burkholderia cepacia, BECP; Enterococcus faecalis, EFAS; Enterococcus faecium, EFAM; Staphylococcus aureus, SA; Staphylococcus epidermidis, SEPI; Staphylococcus saprophyticus, SSAP.

[0050] Figure 19, from Example IX - Metabolomics-based Antimicrobial Susceptibility Testing (MAST). Changes in the levels of selected biomarkers for each strain after a 4-hour incubation period and corresponding liquid growth analysis. Biomarker production was observed in the absence of an antimicrobial agent or when resistant strains were incubated with a sub-inhibitory dose of an antimicrobial agent. The upper left graph shows that, relative to the stable response of resistant isolates, biomarker (hypoxanthine) production in sensitive strains decreased depending on the antimicrobial agent concentration. The remaining graphs show the susceptibility of each species to the most commonly used antimicrobial agents at the minimum inhibitory concentration (μg / mL). FLC, fluconazole (2); AMB, amphotericin B (2); 5FC, 5-fluorocytosine (0.5); CRO, ceftriaxone (0.5 for Streptococcus pneumoniae and 1 for Klebsiella pneumoniae / Escherichia coli respectively); CIP, ciprofloxacin (1); MEM, meropenem (0.25, 1, and 2 for Streptococcus pneumoniae, Klebsiella pneumoniae / Escherichia coli, and Pseudomonas aeruginosa respectively); GEN, gentamicin (4; 500 only for Enterococcus faecalis); AMP, ampicillin (8); SXT, trimethoprim-sulfamethoxazole (2 / 38); CAZ, ceftazidime (8); PIP, piperacillin (16); LVX, levofloxacin (2); PEN, penicillin (0.06); ERY, erythromycin (0.25 and 0.5 for Streptococcus pneumoniae and Staphylococcus aureus respectively); OXA, oxacillin (2); VAN, vancomycin (1, 2, and 4 for Streptococcus pneumoniae, Staphylococcus aureus, and Enterococcus faecalis respectively); CFX, cefazolin (4); TET, tetracycline (4). In KP, for the third graph AMP, block CRO and EF, and for the first graph CIP MEM GEN, block MH, non-black borders indicate a negative correlation (2% false discovery rate) between biomarker production and overnight liquid growth analysis.

[0051] Figure 20A and 20B , from Example XI, RPMI component drop out experiments, where individual precursors (abscissa) were omitted from the medium to determine whether the omission of a specific precursor would eliminate biomarker production.

[0052] Figure 21 , production of agmatine from glucose or arginine. When Enterobacteriaceae grow in M9 medium without arginine, agmatine is produced from glucose (A). However, when 15 N-labeled arginine is added to the M9 medium (B), it is consumed by Enterobacteriaceae and converted to 15 N-agmatine (C).

[0053] Figure 22A and22B , Heatmap - Blood in culture does not affect the biomarker.

[0054] Figures 23A to 23C , Generation of biomarkers in the restricted custom media in Table 1A.

[0055] Figure 24A and 24B , Comparison of the generation of selected biomarkers in Mueller - Hinton medium and RPMI ( Figure 24A ) and in the custom media of Table 1C ( Figure 24B ).

[0056] Figure 25 , Data - related sampling process diagram. DETAILED DESCRIPTION

[0057] The detailed description and examples given below are intended to describe various embodiments of the present invention and are not intended to represent the only embodiments accomplished by the inventors. The detailed description includes specific details which are aimed at providing a comprehensive understanding of the present invention. However, it will be apparent to those skilled in the art that the present invention can be practiced without these specific details.

[0058] The current methods for identifying microorganisms and determining their sensitivity to toxic substances (such as antibiotics) are as follows: (1) Collect a sample of a biological fluid or a tissue swab from a patient; (2) Combine the sample with a growth medium containing nutrients (solid or liquid); (3) Incubate the sample to allow the microorganisms to grow to a detectable level (about 18 - 48 hours); (4) Identify the microorganisms based on protein profiles using chemical detection methods or spectroscopic methods [e.g., matrix - assisted laser ionization desorption (MALDI) mass spectrometry (MS)]; (5) Place the microorganism specimen in a growth medium (solid or liquid) containing the toxin; (6) After about 18 - 48 hours, measure the growth rates of the microorganisms with and without the toxin; and (7) Use the data of reference microorganisms to determine the growth rate of the toxin - sensitive relative to the antitoxin.

[0059] Living cells, such as microorganisms, continuously metabolize: taking in nutrients and secreting waste products. Living cells, such as microorganisms, use the metabolic nutrients in their environment to supply their energy, redox, and biosynthetic needs. These general life requirements can be met through various metabolic pathways. To obtain and process their nutrients, microorganisms have developed diverse metabolic strategies. These metabolic activities are restricted by the genetic makeup of the organism and environmental conditions. Each molecule taken in by a microorganism constitutes a complex network of chemical reactions. The metabolic waste products from these networks are substances that do not affect the viability of the organism and are secreted into the environment. These metabolic end points depend on the structure of the metabolic pathways of the organism, which is constrained by genetic and environmental factors. Thus, if the environmental conditions are controlled, then the metabolites can be biomarkers of the genetic makeup of the microorganism, where these metabolites are the molecules taken in by the organism and the substances that it secretes back into its environment. The present invention detects these metabolites and uses these metabolites to identify the cell type of the organism. The cell type of the organism can be the general category of the organism (i.e., Gram-negative, Gram-positive, etc.), the species or species origin (i.e., bacterial species, human, etc.), or, the strain or distinguishing characteristics (i.e., human blood cells, resistance or sensitivity to toxins such as antibiotics or chemotherapy, stationary or active growth, successful genetic transformation, etc.).

[0060] The biological consumption of nutrients and the secretion of waste products are fundamental components of living cells. Environmental conditions, such as the presence of toxins, can regulate cell metabolism by killing cells or by stimulating or severely disrupting the flow of metabolites into or out of cells. The present invention can also detect cell types related to subspecies, strain characteristics such as changes in its toxin-induced nutrient uptake and waste secretion. Toxins can include, for example, antibiotics, inorganic substances, cancer chemotherapeutic agents, etc.

[0061] The growth medium in which the cells grow supplies nutrients and accumulates waste products. Thus, over time, the metabolic signal of the microorganism is amplified by the changes that accumulate in the composition of the medium. As a result, the amount of metabolites in the growth medium is more than 500,000 times higher than the amount of peptides and proteins currently used for the MALDI-MS classification of microorganisms. Thus, metabolism-based experiments can detect microorganisms at low concentrations. For example, the present invention can identify microorganisms based on the analysis of samples with less than 100 bacteria per milliliter ( Figure 3 ). Compared to current methods, such sensitivity shortens the incubation time.

[0062] In addition, in the present invention, the most interesting metabolites are small molecules, e.g., small molecules less than 600 Daltons or even less than 400 Daltons. Such metabolites are mainly monomers. These metabolites are consumed and appear very rapidly in the growth medium, especially much faster than the macromolecules such as peptides and proteins currently used for microbial MALDI-MS classification.

[0063] The devices, methods, and systems of the present invention identify cell types of organisms. It can identify the general category, species, or specific cell characteristics such as toxin sensitivity of an unknown organism.

[0064] In one embodiment, the method ( Figure 4A ) includes: incubating a sample 10 in a growth medium, and performing chemical analysis 11 of metabolite biomarkers in the growth medium after incubation; and identifying the microorganism in the sample 12 by comparing the levels of metabolite biomarkers in the growth medium with a reference metabolite profile.

[0065] The incubation period allows the cells in the sample to metabolize: consume their preferred nutrients and secrete metabolic waste products. The metabolites present in the medium are analyzed after the analysis to obtain the metabolic data of the organisms in the growth medium. As described above, the reference metabolite profile is the known metabolite results of a microbiota or a single species or subspecies and strain, and the reference metabolite profile is compared with the metabolic data obtained from the sample analysis, thereby classifying the unknown organism.

[0066] Samples that can be added to the growth medium include, but are not limited to: food, tissue, biological fluids (such as any one of excreta, blood, urine, or cerebrospinal fluid), or swabs such as from a living or non-living surface (i.e., tissue swab or clinical surface swab). The sample can be unprocessed or can be pre-treated. For example, in one embodiment, the method includes pre-treating the sample to separate the microorganism from the remaining sample content. For example, a swab can be soaked to collect the microorganisms therein. As another embodiment, the method can include separating the microorganism from other sample components (such as other cells). For example, a blood sample can be processed to separate the microorganism from patient cells (such as blood cells). This processing can employ size exclusion, such as filtration or centrifugation. However, it is worth noting that experimental data have shown that the present invention can accurately identify microorganisms from samples even if the samples contain other living cells, such as a biological community or blood cells.

[0067] To make the separation easier, the method can further include sample dilution. Thus, the method can include: pre-treating the sample, which includes diluting the sample and separating the microbial cells from the diluted sample. The diluent can be a growth medium or a nutrient-free wash solution, such as sterile saline.

[0068] Whether or not separation is performed, in another embodiment, the pretreatment step includes concentrating the microorganisms in the sample into a concentrated assay solution. For example, concentration can be achieved by filtration or density separation (i.e., centrifugation). Concentration can reduce the sample volume by a factor of 1 / 10 to 1 / 100,000. For example, a 1 - 10 ml sample can be reduced to less than 50 μl. For example, a sample of approximately 5 - 25 μl is useful. Concentration can result in a microbial concentrate with less than 500 cells per milliliter or possibly less than 100 cells per milliliter. In one embodiment, the concentration step brings the microorganisms to a known or desired cell count (concentration) per unit volume.

[0069] The growth medium contains nutrients that support cell metabolism. Since the present invention is based on the analysis of normal metabolism, the growth medium does not need to contain any atypical biomarkers or macromolecules, but rather, it can be a typical growth medium, for example, liquid or solid formulations of M9, Mueller - Hinton (MH) medium, LB medium (Lysogeny broth), tryptic soy broth, YEPD medium (yeast extract peptone dextrose), BacT TM , BacT / Alert TM , Vitec TM , modified Eagle medium TM (Dulbecco Modified Eagle Medium TM ) or Roswell Park Memorial Institute TM (Roswell Park Memorial Institute TM )(RPMI) medium. In one embodiment, a growth medium with a customized composition is used to support the growth of one or more selected microorganisms and produce biomarkers. The growth medium is designed and manufactured to ensure the growth of the selected pathogen according to nutritional requirements and to ensure biomarker production. This can control the specific cells being cultured and / or can simplify the analysis because there will be fewer metabolites to identify. In one embodiment, control chemicals that do not participate in metabolism can be added to the growth medium for tracking purposes during spectroscopic analysis. In one embodiment, isotope labeling can be used for tracking. For example, known nutrients can be labeled so that their metabolism and modification / secretion can be tracked.

[0070] Enriched culture media containing a variety of nutrients, also known as complex media, can promote cell growth, thereby increasing the analysis speed and the speed of cell type identification. On the other hand, for the growth of fewer cell types, it is easier to select a simpler medium with less nutrients, but the simpler medium is easier to analyze, which is beneficial for cell type identification. In particular, the enriched medium or the simpler medium can be a defined medium or a natural medium.

[0071] In one embodiment, the metabolic profile of the growth medium that has not been altered before adding the sample is worth comparing and can be obtained by chemical analysis. In some methods, a growth medium control can be collected shortly after adding the sample. Any microorganisms in the control can be killed to prevent metabolism. In other embodiments, if the spectral map of the selected growth medium has been obtained or the spectral map of the starting growth medium is not required, then a control is not needed.

[0072] Incubation allows the microorganisms in the sample to metabolize in order to consume and produce metabolites. Incubation can be carried out at a high temperature, such as approximately body temperature, for example 35 - 40 °C. Incubation should be maintained for an appropriate period of time so that a detectable amount of waste is produced by metabolism. In one embodiment, the method includes incubating for 1 - 6 hours, such as 2 - 4.5 or 3.5 - 4.5 hours. In one embodiment, the incubation time is set with a variation limit of + / −30 minutes or less, such as + / −10 minutes. In particular, microorganisms experience various stages of metabolism during their life cycle. During the first stage of metabolism, certain first chemicals are produced, and over time, these first chemicals decompose due to further metabolism or natural decay. In cases such as the acidogenesis to solventogenesis transition, the metabolic profile changes over time. Thus, it is important to control the incubation period in order to establish the metabolic profile of the sample at a specific metabolic stage. Control during incubation can ensure the reproducibility of the method and the accuracy of the sample profile, which is used for comparison with a reference profile obtained from incubation at the same time.

[0073] As described above, it is desirable to reduce the organism identification time. Incubation may take the most time in this method. To reduce the total time for analysis, the method can include any pretreatment steps in the growth medium, and possibly heating during the pretreatment steps. For example, ambient temperature or heated growth medium can be used for dilution, separation, and concentration so that incubation can begin and the microorganisms start to metabolize during these steps. In addition, the equipment used for pretreatment can be heated. For example, equipment such as that used for dilution, filtration, centrifugation, etc. can be heated to 35 - 40 °C.

[0074] After incubation, the growth medium is analyzed to determine its metabolite content, i.e., its metabolic profile. In one embodiment, after a selected incubation time, the growth medium is quenched to stop metabolism. In other words, any living cells in the growth medium are killed. The quenching method can be selected to reduce chemical modification so as to preserve the metabolites. In one embodiment, methanol is added to the growth medium to stop metabolism.

[0075] The method includes chemically analyzing the growth medium after incubation to identify metabolites in the growth medium. In particular, the metabolites are used as biomarkers, and the levels of various metabolites are determined, which may include consumed and produced metabolites.

[0076] The chemical analysis can be simple, such as using pH evaluation, analyzing with a glucometer, simple chromatography, or simple optical analysis. These methods are particularly used for media with a simple profile ( Figure 14 ) or when only a few metabolic biomarkers are of interest. Although straightforward, simple forms of chemical analysis methods can produce a suitable signal indication of the level of one or more metabolic biomarkers in the growth medium.

[0077] However, for more complex analyses, such as when the medium is more complex or there is more than one cell type present in the sample, more complex chemical analysis methods can be used, such as spectroscopic analysis. Of course, the level of any particular metabolite is represented by the spectroscopic signal intensity. The actual concentration of the biomarker is not necessarily determined, but as understood, the spectroscopic data is collected as a signal with intensity, and this intensity may be related to the concentration. All concentrations are referred to as metabolite levels in this article. When spectroscopic analysis is performed on the growth medium, signals representing the intensity of some biomarkers are generated. Since the metabolic system of each type of microorganism is unique, each microorganism's growth medium will produce a unique signal when considering data from one or more metabolic biomarkers. The resulting spectroscopic data on the level of one or more biomarkers is called a metabolic profile. Spectroscopic analysis can use mass spectrometry (MS), nuclear magnetic resonance spectroscopy (NMR), or spectrophotometry such as optical analysis. Some useful MS platforms are liquid chromatography MS (LC-MS), triple quadrupole MS, or high-resolution MS.

[0078] It should be noted that metabolite signals in spectral analysis can be complex and in fact each molecule may include more than one signal. Generally, a set of signals of the molecule can be decomposed and regarded as a single metabolite signal. For example, a mass spectrometer will detect 10 - 50 signals for each molecule, which are caused by the original molecule (parent molecule) as well as various fragments, adducts (chemical combinations occurring in the instrument), and isotopologues (naturally occurring forms of molecules with one or more additional neutrons). Detecting any of these signals can represent a molecule and can represent a metabolite of interest in the present application.

[0079] Once the analysis is performed, the microorganism can be identified by comparing the metabolite profile generated from the sample (i.e., the spectral data of the biomarker levels in the growth medium after incubation) with the reference metabolite profile of known microorganisms grown in a similar medium during a similar incubation period. As is evident from the examples, subsequently, this comparison can be done manually. However, to increase the analysis speed, the data can be analyzed using an analog processor, an electronic processor, or a computer processor based on, for example, a stored database of reference metabolite profiles. For example, using a computer system with software, a large number of metabolite reference profiles can be quickly compared with the metabolite data obtained from the sample to identify the cell type in the sample.

[0080] If the analysis data is compared with a reference metabolite profile obtained from a growth medium, cell concentration, and incubation time under similar conditions, then there is no need to identify the metabolites used as biomarkers. Reproducible spectral features can be obtained from molecules related by structure or metabolic function, such as amino acids, nucleosides, carbohydrates, tricarboxylic acid cycle intermediates, and fatty acids.

[0081] As mentioned above, the most interesting metabolic molecules are those that are easily consumed or formed by cell metabolism, such as simple carbohydrates, amino acids, nucleic acid bases, and their derivatives. Such molecules are generally less than 600 or 400 daltons and are monomers or simple complexes of two or three monomers.

[0082] The diagnostic metabolites observed in a microbial culture are usually secreted together with closely related molecules from the same metabolic pathway. As Figure 13AAs shown, culturing Escherichia coli in the presence of lactose will produce glucose and galactose, which are the breakdown products of lactose. Thus, any metabolite present in this pathway can be used to diagnose the presence of Escherichia coli under the conditions used in this study. Similarly, inosine, hypoxanthine, xanthine, guanine, inosine monophosphate, xanthosine monophosphate, and uric acid are all metabolites that are derived from guanosine monophosphate and reflect the action of a common metabolic pathway. More generally, the presence of diagnostic nucleotides or their breakdown products in a growth medium represents specific microbial metabolic activities that can be used as diagnostic indicators of the type of microorganism. Similarly, amino acid breakdown products from specific metabolic activities (such as arginine catabolism) convey overlapping diagnostic information. For example, agmatine, putrescine, and ornithine, all derived from arginine breakdown, can be used as diagnostic indicators of microorganisms. Molecular classes of metabolic pathway activities that produce closely related groups of diagnostic metabolites include carbohydrate metabolism (e.g., Figure 13A , glucose and galactose from lactose), nucleotide metabolism, amino acid metabolism, tricarboxylic acid cycle metabolism, and fatty acid metabolism.

[0083] Specific metabolites that have been identified as useful for differentiating 85% or more of the pathogens of clinical interest include adenine, adenosine, arginine, 4-aminobutyric acid, cytidine, glucose, glutaric acid, glycine, guanine, guanosine, hypoxanthine, inosine, N-acetyl-phenylalanine, ornithine, sn-glycerol-3-phosphate, succinate, taurine, uridine, urocanate, and xanthine, or derivatives thereof. There are also 9 metabolites that have not been identified but have been seen in spectroscopic analysis and used to differentiate microorganisms. Only these 30 molecules are required for the correct identification of the following 11 microorganisms: Escherichia coli; Klebsiella pneumoniae; Klebsiella oxytoca; Pseudomonas aeruginosa; Staphylococcus aureus; Enterococcus faecalis; Enterococcus faecium; Streptococcus pneumoniae; Group A Streptococcus; Candida albicans; and Candida parapsilosis, which cause more than 85% of human blood infections. Other pathogens of interest include Citrobacter, Enterobacter, Proteus, Acinitobacter species, and Streptococcus and Staphylococcus species, which are different from the microorganisms listed above and can be differentiated using the biomarkers described above.

[0084] By analyzing the growth medium after incubation, metabolic differences can be detected, enabling the identification of the cell types incubated. For example, see Figure 5 , Figure 6A and Figure 6B, Gram-negative bacteria such as Escherichia coli (EC) and Klebsiella spp. (KO and KP) secrete diagnostic amounts of succinate and consume glucose when grown in Mueller-Hinton medium. However, under the same conditions, Gram-negative Pseudomonas aeruginosa (PA) hardly affects succinate and glucose, but consumes diagnostic amounts of ornithine. Similarly, Gram-positive Staphylococcus aureus (SA) consumes diagnostic amounts of taurine and secretes N-acetyl-phenylalanine when grown in Mueller-Hinton medium, while Gram-positive Enterococcus faecalis (EF) consumes diagnostic amounts of arginine. Likewise, Gram-positive Streptococcus viridans (SV) and Streptococcus pyogenes (SP) produce diagnostic levels of glutaric acid when incubated in Mueller-Hinton medium, while group A Streptococcus (SG), Enterococcus faecalis (EF), Staphylococcus aureus (SA), and coagulase-negative Staphylococcus (SN) do not.

[0085] Further testing of pathogen metabolism in Mueller-Hinton medium revealed that:

[0086] · Succinate levels distinguish Escherichia coli and Klebsiella from all other organisms in the panel;

[0087] · Urocanate levels distinguish Escherichia coli and Klebsiella;

[0088] · 10-Hydroxydecanoate distinguishes Pseudomonas aeruginosa from all other organisms in the panel;

[0089] · Arbitol levels distinguish Candida from all other organisms in the panel;

[0090] · Glucose levels distinguish Pseudomonas aeruginosa from all other Gram-negative organisms;

[0091] · N-acetyl-aspartate levels distinguish Enterococcus from all other organisms in the panel;

[0092] · Xanthine distinguishes Streptococcus viridans from all other organisms in the panel;

[0093] · The presence of galactose is diagnostic for Escherichia coli; and

[0094] · The glucose-to-lactose ratio distinguishes Enterococcus faecalis from other Gram-positive organisms.

[0095] After incubation in BacT medium:

[0096] · Sucrose levels distinguish Escherichia coli, Klebsiella subsp., and Staphylococcus aureus; and

[0097] · Trehalose levels distinguish CN-Staphylococcus from Staphylococcus aureus and Escherichia coli.

[0098] Further metabolic analysis revealed that a large number of organisms could be reliably identified as follows:

[0099] · If, after metabolism, the growth medium contains nicotinamide (also known as niacinamide), then the production of nicotinate in the growth medium indicates the presence of Escherichia, Klebsiella, Pseudomonas, Enterococcus, Staphylococcus, and / or Streptococcus in the sample. In other words, if, after culturing an unknown microbial sample in a growth medium containing nicotinamide, the concentration of nicotinate is higher compared to the original medium, then it can be concluded that the sample contains at least one of Escherichia, Klebsiella, Pseudomonas, Enterococcus, Staphylococcus, or Streptococcus. Pyridoxine is a cofactor of the enzyme that converts nicotinamide into nicotinate and can be added to facilitate metabolism. However, some microorganisms do not require this cofactor because they can synthesize it themselves;

[0100] · When the sample grows in a medium containing arginine, the metabolism that causes the production of citrulline indicates the presence of Gram-positive Enterococcus or Streptococcus in the sample.

[0101] · When the sample grows in a medium containing carbohydrates (such as glucose, sucrose, fructose, etc.): a) Culturing that causes the production of arabitol indicates the presence of Candida, such as Candida albicans, in the sample; b) Culturing that causes the production of succinate indicates the presence of Escherichia or Klebsiella in the sample. In other words, after culturing, if the concentrations of arabitol and succinate are produced - then the sample contains microorganisms of all these genera.

[0102] · Culturing that causes the production of xanthine indicates the presence of Pseudomonas in the sample. For Pseudomonas, there are also other biomarkers for pathogen identification and verification, as described above. It should be noted that the production of xanthine is associated with Streptococcus viridans ( Figure 6A ), but further tests have shown that, probably due to retention time, this association is not as clear as the characterization of Pseudomonas in the sample.

[0103] · When a sample grows in a culture medium containing histidine and carbohydrates (such as glucose, sucrose, fructose, etc.), the metabolism that causes the production of urocanate indicates the presence of Klebsiella and / or Streptococcus group A in the sample. It should be noted that the production of urocanate is associated with Streptococcus viridans, but further tests have shown that, possibly due to the complexity of the retention time, this association may not be conclusive. Therefore, as described above, the analysis shows a culture profile that consumes carbohydrates and produces succinate and thus indicates the presence of Escherichia or Klebsiella. Further analysis of urocanate will clearly distinguish the presence of Escherichia or Klebsiella. In addition, referring to the above data, the analysis shows a culture profile that produces nicotinate, indicating the presence of Escherichia, Klebsiella, Pseudomonas, Enterococcus, Staphylococcus, or Streptococcus in the sample. Further analysis finds that the concentration of urocanate can distinguish Streptococcus group A, Streptococcus viridans, and Klebsiella present from all other organisms. To distinguish Streptococcus group A from Klebsiella, the presence of urocanate and the non-production of succinate indicate the presence of Streptococcus group A and Streptococcus viridans. Optionally or additionally, culturing can be carried out in a culture medium containing arginine, where, as described above, the metabolism that causes the production of citrulline indicates Streptococcus group A, Enterococcus, or Streptococcus pneumoniae. Streptococcus viridans and Streptococcus group A can be further distinguished by the production of glutaric acid.

[0104] · In a culture medium containing threonine, the metabolism that causes the production of N-acetylthreonine seems to indicate the presence of Enterococcus in the sample.

[0105] · In a culture medium containing carbohydrates, the metabolism that causes the production of mevalonic acid indicates the presence of Staphylococcus or Enterococcus in the sample.

[0106] · In a culture medium with arginine, a cultured sample containing the metabolite agmatine indicates the presence of a species of Enterobacteriaceae.

[0107] · In the culture medium, an increase in the concentration of methylbutylamine indicates the presence of a species of Proteus in the cultured sample.

[0108] · In the culture medium, the presence of N 1 ,N 8 -diacetylspermidine increase indicates the presence of Enterococcus faecalis, Staphylococcus saprophyticus, and Staphylococcus epidermidis in the cultured sample. This marker can be used to distinguish Enterococcus faecalis from Enterococcus faecium, where Enterococcus faecalis produces N 1 ,N 8 -diacetylspermidine while Enterococcus faecium does not.

[0109] · After culturing, tyramine indicates Enterococcus.

[0110] · There are many species of Candida, and arabitol has been identified as an indicator of yeast and especially Candida albicans. Other biomarkers have been identified for differentiating other species. For example, N-acetyl leucine / N-acetyl isoleucine indicates Candida freundii and can be used to distinguish it from Candida albicans. In addition, a biomarker with a retention time of 4.3 minutes and a mass of 286.2366 in the 15-minute HILIC method indicates Candida albicans, which can distinguish Candida albicans from Candida freundii.

[0111] · Although all Enterobacteriaceae tested produce agmatine, the specific organisms listed below produce only cadaverine and putrescine, thus enabling the differentiation of some of these Enterobacteriaceae. In particular, after cultivation, the presence of cadaverine indicates Escherichia coli, Enterobacter aerogenes, Klebsiella, and Stenotrophomonas maltophilia.

[0112] aerogenes), Klebsiella, and Stenotrophomonas

[0113] maltophilia). After cultivation, the presence of putrescine indicates Citrobacter, Escherichia coli, Enterobacter, Klebsiella, and Proteus mirabilis.

[0114] Some metabolomic profiles are more specific for the growth medium used. For example:

[0115] · When the sample is grown in RPMI medium containing hypoxanthine, nicotinamide, and pyridoxine, the production of xanthine and 6-hydroxy nicotinate is detected, indicating the presence of Pseudomonas in the sample. It is believed that xanthine is produced from hypoxanthine, and 6-hydroxy nicotinate is produced as a metabolite of nicotinamide and pyridoxine.

[0116] · The presence of agmatine in a cultured sample indicates the presence of a species of Enterobacteriaceae. In a medium such as M9 medium, where there is a single carbon source, for example, the medium contains only carbohydrates as the carbon source, the metabolite agmatine indicates the presence of a species of Enterobacteriaceae. Alternatively, in other more complex media, such as MH with arginine, a cultured sample containing the metabolite agmatine indicates the presence of a species of Enterobacteriaceae.

[0117] · In addition, in MH, gamma-aminobutyric acid is used to differentiate Staphylococcus aureus and coagulase-negative staphylococci because Staphylococcus aureus produces gamma-aminobutyric acid while coagulase-negative staphylococci do not.

[0118] · In some media containing spermine, causing N 1 ,N 12- The metabolism resulting in N1,N12-diacetylspermine indicates the presence of Enterococcus, Klebsiella, and Escherichia coli in the sample. In RPMI without spermine, Enterococcus does not produce N 1 ,N 12 - 1,N12-diacetylspermine, however, in other media such as MH, only Enterococcus produces N 1 ,N 12 - 1,N12-diacetylspermine.

[0119] An increase in the concentration of any of these biomarkers in the cultured medium compared to the control medium indicates pathogen-related metabolic activity in the cultured medium. However, fold changes in the concentration of at least some precursors and biomarkers have been determined. For example, preferred fold changes are shown in the Figure 16 decision tree, but it should be understood that a positive concentration change can be 1 / 2 of the indicated value. For example, referring to Figure 16 , from the control to the cultured medium, a 2.5-fold increase in arabitol and more clearly a 5-fold increase indicates the presence of Candida albicans in the culture. As another example, an increase in the concentration of mevalonic acid in the cultured medium indicates Staphylococcus and Enterococcus. It has been determined that under noisy conditions, an increase in mevalonic acid of more than 5-fold and more clearly an increase of 10-fold or more from the control to the cultured medium (as shown in Figure 16 ) indicates the presence of Staphylococcus and / or Enterococcus in the culture. Of course, these can be distinguished using the biomarker N1,N12-diacetylspermine (intensity increase > 2.5E3 or possibly > 5E3) or gamma-aminobutyric acid (> 2.5-fold or possibly > 5-fold increase).

[0120] If quantitative variability is found to be a problem, then the analysis of the cultured sample can include the addition of isotopes (precursors or biomarkers) of the target analyte at known concentrations. The signal intensity of the isotopically labeled target at known concentration can be used to standardize the quantitative variability and thus calculate the concentration of the target analyte. The method includes: detecting the target analyte and simultaneously eluting the isotopically labeled form of the detected target analyte, and comparing their peak intensities. Using isocratic continuous elution and isotope dilution, it is possible to accurately analyze the sample even if ion suppression changes. This isotope dilution strategy enables accurate quantification of the target analyte in multiple samples being analyzed, as described in co-pending application PCT / CA2019 / 050763 filed by the applicant on May 31, 2019, which is incorporated herein by reference.

[0121] These aforementioned target metabolites, while useful for pathogen identification, are not affected by blood cell metabolism. Thus, even though the blood cells in the sample to be analyzed may also be metabolizing, such metabolism does not interfere with the production of the aforementioned target metabolites ( Figure 11 and Figure 24).

[0122] While previously known culture media such as MH or RPMI can be used, it may be desirable to employ a customized culture medium that includes minimal essential precursor nutrients and buffers, salts, enzymes, etc. for the general support of some or all of the microorganisms identified above and that is useful for differentiating and identifying common microorganisms causing infections such as blood-based infections. Thus, while the above results and the subsequent examples show BACT, M9, MH, and / or RPMI, the observed phenotypes can be recreated in more restrictive culture media provided that (a) sufficient nutrients are provided to allow the cells to grow for a sufficient length of time to obtain metabolic signals, and (b) appropriate precursors are present to produce the target biomarkers. Growth requires only a short time, e.g., less than 6 hours or about hours.

[0123] In one embodiment, for example, an engineered medium for differentiating microorganisms in a biological sample may include at least the following precursors: 0.5 to 1.5 mM carbohydrates such as glucose, histidine, pyridoxine, nicotinamide (niacinamide), and arginine. The amount of carbohydrates is much lower than in previous media but is sufficient to support the short-term growth of microorganisms in the culture, such as at least 3 to 8 hours, which is within the range of 3.5 to 5 hours found to be suitable for generating good metabolomic signals. In another embodiment, for example, an engineered medium for differentiating microorganisms in a biological sample may include at least the following precursors: glucose such as 0.5 to 1.5 mM glucose, histidine, pyridoxine, nicotinamide, hypoxanthine, threonine, spermine, and arginine, as well as catalase.

[0124] In another embodiment, an engineered defined composition culture medium includes the components listed in Table 1A, with the actual composition for testing in the first column and given wide and narrow concentration ranges.

[0125] Table 1A: Engineered Defined Composition Culture Medium for Metabolomics Research

[0126] TM Limited Medium (MLM) Range (g / L)

[0127]

[0128] Reference Figure 16 to the decision tree and Table 1B, for example, these precursors and nutrients are selected to support metabolism and some of them can be used for differentiation as follows:

[0129] · Glucose (precursor) is converted to the following:

[0130] - Arabitol (for differentiating Candida species such as Candida albicans)

[0131] - Mevalonic acid (for differentiating Staphylococcus and Enterococcus)

[0132] - Succinate (for differentiating Klebsiella and Escherichia);

[0133] · Conversion of histidine (precursor) to urocanate (for differentiating Klebsiella and Group A Streptococcus); · Conversion of nicotinamide (precursor) to nicotinate (for differentiating Pseudomonas aeruginosa, Escherichia, Klebsiella, Group A Streptococcus, Streptococcus pneumoniae and Enterococcus). Pyridoxine is a cofactor for the enzyme that converts nicotinamide to nicotinate, but some microorganisms do not require this cofactor because they can synthesize it themselves.

[0134] · Conversion of arginine (precursor) to citrulline (for differentiating Streptococcus pneumoniae, Group A Streptococcus and Enterococcus). Arginine (precursor) is also converted to agmatine (for differentiating Enterobacter).

[0135] · Spermine (precursor) causes the production of N 1 ,N 12 - Diacetylspermine (for differentiating the presence of at least Enterococcus). · Metabolism of hypoxanthine to xanthine (for differentiating Pseudomonas aeruginosa).

[0136] · Metabolism of threonine to n-acetylthreonine (for differentiating Enterococcus).

[0137] Catalase reduces free radicals to assist the growth of Streptococcus pneumoniae.

[0138] Table 1B: Biomarkers and Some Identified Pathogens

[0139]

[0140]

[0141] In addition to the above precursors, the culture medium may include:

[0142] a. Leucine to further promote the growth of Staphylococcus species; and / or

[0143] b. Leucine and glutamine, which, in addition to histidine, also promote the growth of Streptococcus species

[0144] (Streptococcus species).

[0145] In one embodiment, a complex but defined medium can include the chemicals listed in Table 1C at various concentrations. It will be understood that the concentration of each of the various chemicals can be varied and still support metabolism useful for metabolomics discrimination of some pathogens. However, in one embodiment, as indicated in Table 1A, the chemicals in the defined MPA / MIA medium in Table 1C (also listed in Table 1A) have a wide or narrow concentration range, with other chemical concentrations being different. For the examples listed below, the medium is prepared according to the formulation in Table 1C below.

[0146] Table 1C: Defined MPA / MIA Medium

[0147]

[0148]

[0149]

[0150] Optionally or additionally, a custom medium can be used according to one of the above compositions, but it contains labeled discriminative nutrients or labeled components for ratio normalization. Labeling can be carried out, for example, using isotope labeling. Thus, another aspect of the present invention relates to designing a medium with specific isotope results. Since the data herein show that the target metabolites are derived from known precursors, we are able to substitute stable, labeled (such as isotope-labeled) precursors into the designed medium to ensure that the detected biomarkers are produced by the specific precursors we choose. When combined with mass spectrometry detection, this strategy ensures that any possible background signals or biomarkers produced by pathways other than the target pathway are not detected as false positive signals. Thus, in one example, a specific labeled precursor can be introduced into the medium so that the labeled metabolite can be easily linked to the labeled precursor. For example, succinate can come from various carbon sources (such as glucose and glutamine) and through various pathways, but introducing labeled glucose can easily identify the labeled succinate as being produced by the metabolism of the microorganism of interest, as described above. In one embodiment, a medium can be prepared and used in which one or more of the following precursors are isotope-labeled: glucose, histidine, nicotinamide, and arginine.

[0151] In some embodiments, the culture medium can include an isotope-coded standard gradient. Typically, metabolites are quantified relative to a standard curve, and these metabolite standards are added to samples at multiple concentrations to establish the relationship between concentration and intensity. The drawback of this strategy is that multiple samples must be analyzed to establish the standard curve. In one embodiment, multiple isotopic forms (e.g., 1C-glucose, 2C-glucose, 3C-glucose, 4C-glucose, 5C-glucose, and / or 6C-glucose) can be added to a sample to encode the metabolite concentration in that single sample. This method can be used by isotope dilution (where standards and unlabeled target molecules are present in the same sample) or by external calibration (where only standards are present to compare relative intensities between samples).

[0152] Analyzing whether only the metabolites arabitol, xanthine, succinate, urocanate, nicotinate, mevalonic acid, and citrulline are present in the cultured medium can correctly identify the pathogen in the cultured medium as belonging to the genus Candida, Escherichia, Klebsiella, Pseudomonas, Enterococcus, Staphylococcus, or Streptococcus at least at the genus level. Analysis of agmatine can further assist in the identification of Enterobacter. The above analysis can reach a conclusion after only about 4 hours of culture time.

[0153] Thus, metabolite biomarkers are useful for identifying cell types, such as broad categories, species, or strains of cells. One or more biomarkers are specific for a single species, and thus, when considering the presence of one or more biomarkers in the growth medium of cells of an unknown cell type during incubation, this method can accurately identify the cell type.

[0154] An interesting cell type is the characterization of cell sensitivity to toxins. Thus, this method can also be used to analyze the toxin sensitivity of cells in a sample. A similar method is employed, but the growth medium includes a certain amount of toxin, such as an antibiotic or a chemotherapeutic agent. Referring to the above options and Figure 4A, the method includes combining a sample that may contain microorganisms with a growth medium and a certain amount of a toxin, such as an antibiotic or a chemotherapeutic agent, for example, any one or more of amoxicillin, penicillin, tetracycline, vancomycin, streptomycin, cephalexin, erythromycin, clarithromycin, azithromycin, ciprofloxacin, levofloxacin, ofloxacin, chloramphenicol, co-trimoxazole, bacitracin, linezolid, cefepime, cefoperazone, cephalexin, meropenem, clotrimazole, econazole, azide, rotenone, antimycin A, chloroquine, nitazoxanide, melarsoprol, eflornithine, tinidazole, miltefosine, metronidazole, 5-fluorouracil, Gen, SXT, ceftriaxone, ampicillin, ceftazidime, or deoxycytidine. The mixture of the cell-containing sample, the growth medium, and the toxin is incubated for a period of time. During incubation, organisms that are insensitive to the toxin will consume their preferred nutrients and secrete metabolic wastes, while organisms that are sensitive to the toxin will become metabolically disordered or inactive. After incubation, the growth medium is chemically analyzed 11, and the metabolite levels in the growth medium are compared with a reference metabolite profile of a cell culture in which the toxin induces metabolic changes in order to identify the toxin sensitivity 12 of the microorganisms in the sample. The comparison can be carried out using various methods, such as a manual method or an automated method. For example, using computer software, it is possible to easily compare the metabolite reference profile with the metabolic data obtained by analyzing the incubated growth medium in order to classify the toxin sensitivity of the organisms. This method can be used alone to classify the toxin sensitivity of organisms. Optionally, this method can be applied sequentially or simultaneously with a method for identifying microorganisms in the sample to identify toxin sensitivity.

[0155] While microorganism identification requires more complex analysis, antimicrobial susceptibility testing may only require broad methods. In addition, there may be independent susceptibility markers that are different from the markers used for identification. Therefore, different strategies can be used to identify toxin sensitivity. For antimicrobial susceptibility testing, the growth medium of one method is MH and contains glucose, nicotinamide (niacinamide), and pyridoxine, and the following results identified drug-resistant microorganisms after cultivation:

[0156] · Glucose consumption, for example, indicates Escherichia, Klebsiella, Enterococcus, Staphylococcus, and Streptococcus with antimicrobial resistance;

[0157] · Succinate production clearly indicates the presence of Escherichia or Klebsiella with antimicrobial resistance; and

[0158] · Nicotinate production clearly indicates that the culture contains Streptococcus and other species with antimicrobial resistance.

[0159] Table 7 shows some toxin sensitivity indicators.

[0160] Accordingly, the method of the present invention can be used to detect a single cell type in a sample, such as the drug sensitivity of bacterial species and / or microorganisms present in the sample. The method is useful for distinguishing between two or more microorganisms. The present invention can also be used to identify cell classes in a cell mixture or the sensitivity of one or more toxins of one or more organisms present in the sample.

[0161] The present invention can be used to analyze samples from a single sample or a single patient. The present invention can also be used to obtain data on multiplexed samples from multiple samples or multiple patients.

[0162] The method can include operating an analysis device to perform one or more steps of the method. For example, the growth medium after incubation can be loaded into the analysis device for analysis and comparison. Optionally, the method can include loading the sample into the device and the method can be implemented entirely within the device.

[0163] As Figure 4B Schematically illustrated, the device 100 according to the present invention includes at least two components: an analysis data acquisition tool 102, and a computer system 104, which compares the output of the analysis data acquisition tool 102 with a reference standard, a reference metabolite profile, in order to identify cell types or cell characteristics, such as its major class, its species, and cell characteristics such as strain, toxin sensitivity or verification of recombination.

[0164] The analysis data acquisition tool 102 can be a tool based on simple chemical analysis, such as pH, conductivity or the presence of glucose, or, more complex techniques, such as techniques based on spectroscopic analysis, such as mass spectrometry ("MS"), nuclear magnetic resonance spectroscopy ("NMR") or spectrophotometry such as using optical analysis. The tool 102 includes an inlet 102a for receiving an amount of growth medium for analysis. The device can be configured to process raw or processed growth medium. When considering processed growth medium, the device can be configured to receive and process packaged growth medium 106, such as on a cassette or strips or in a test tube, gel, etc.

[0165] The device 100 further includes a computer system 104, which communicates with the tool 102 and is configured to receive the results from the tool 102. The system 104 further includes a processor configured to analyze the data from the tool 102 and identify cell types, such as major classes, species, and / or, identify strain / cell characteristics, such as toxin resistance. In one embodiment, the data is a metabolite profile and the computer system includes a computer storage element for storing a database of reference metabolite profiles.

[0166] Multiple reference metabolite profiles are used to populate a database and enable a computer system. A computer model compares information received from sample testing with the reference metabolite profiles to determine the identity of an unknown microorganism and / or its sensitivity to toxins. A computer system, such as software, compares data obtained from a sample with the reference metabolite profiles. The sample is classified by a single biomarker level or multiple biomarker levels, where the levels are, for example, presence, absence, or a quantity above or below a threshold or within a specified range. The biomarker level (i.e., presence, absence, or level) is represented by a spectral signal intensity. The reference metabolite profiles can include reference data for one or more selected biomarkers in a specified growth medium and the general cell concentration of a class of microorganisms or a specified microorganism after a specified incubation period. Optionally, the reference metabolite profile is the cumulative spectral signal or spectral pattern of a class of microorganisms or a specified microorganism in a specified growth medium after a specified incubation period. Data for the analyzed growth medium can be matched to the reference metabolite profiles using simple processing or algorithms (e.g., support vector machines, principal component analysis, singular value decomposition).

[0167] In one embodiment of the complex analysis, the device includes a support vector machine algorithm that can be used to automatically classify microorganisms from clinical samples and distinguish between drug-sensitive and drug-resistant strains of the microorganisms.

[0168] To establish a database of reference metabolite profiles, known organisms can be incubated under known conditions of nutrient source, time, and toxin concentration. After such incubation, the growth medium can be analyzed, and the resulting data for one or more natural metabolite biomarkers (such as up to 21 or 30 biomarkers as described above) can be recorded, or the complete signal of the entire spectrum can be recorded, and this data can be recorded as a reference metabolite profile. When reference metabolite profiles are obtained from multiple cells (such as microorganisms of interest), the data can be stored in a computer system to create a database and support a fully automated computer program that helps identify unknown organisms and classify toxin sensitivity based on the metabolite profiles obtained from clinical samples. The reference data can be readily applied to resolve signals from samples containing one or more types of microorganisms.

[0169] The computer system is configured to output information regarding the identification of microorganisms in the sample and any possible toxin resistance.

[0170] In another embodiment, the device may include a sample pretreatment chamber and / or an incubation chamber 108. Then, the inlet 102a will provide a sample input to these chambers. These chambers may contain supplies of growth medium, diluents, etc., so that the device is self - sufficient and configured to perform a method from receiving an untreated sample to microbial identification. The sample pretreatment chamber can be configured to process the raw sample into a form suitable for incubation. For example, the sample pretreatment chamber may include dilution and size exclusion, such as a filtration instrument 108a. The incubation chamber and possibly the sample pretreatment chamber include a heater 108b.

[0171] Figure 4C Another embodiment of the device 100 according to the present invention is shown. Figure 4C The device 100 includes a first microbial incubation chamber 112, a second microbial growth chamber 114, and a sample preparation chamber 122. The sampling port 116 provides a connection between chambers 112, 114 and the sample transfer tube 120, and the sample transfer tube 120 leads to the chamber 122. The timing device 118 controls the operation of the sampling port 116, so it only opens when permitted by the timing device. The incubation time of chambers 112, 114 can be controlled by the timing device.

[0172] The device further includes an analytical metabolite data acquisition device 124. The tube 120 leads from the chamber 122 to the device 124. The analytical device 124 is connected to the data analysis device 128 through physical or wireless communication 126 and operates together with the database of the data analysis device 128 of the reference metabolite map. The data analysis device 128 outputs to a display device 132 on the device, such as a printer, a screen, a computer terminal, etc., or is coupled to the data analysis device of the device 100 wirelessly or through a physical connection.

[0173] Components such as the port 116, the timing device, the tube 120, etc. can be components of an automatic sampler.

[0174] In use, a microbial sample and a growth medium are added to the microbial incubation chamber 112. Synchronously or sequentially, the sample in chamber 112 is transferred to the second microbial growth chamber 114, and the growth medium containing the sample in the second microbial growth chamber 114 can be mixed with a toxin 114. The cultures in chambers 112 and / or 114 are incubated for a fixed period of time using the timing device 118. The device 118 controls the operation of the sampling port 116.

[0175] The culture samples of chambers 112 and / or 114 are transferred through the sampling port and the sample transfer tube 120 and delivered to the sample preparation chamber 122.

[0176] After preparation, the sample is transferred from the preparation chamber to the metabolite analysis device 124 and a signal of metabolite level is generated. The observed metabolite signal is transmitted to the data analysis device 128, which uses the reference data set 130 and a processor to identify the cell type or cell types present in the microbial sample. Then, the organism type is reported by the display device 132.

[0177] There can be software for controlling the operation of the device components. In one embodiment, there are some processing strategies, such as data-dependent feedback of automatic sampling. For example, there can be data-dependent communication software for identification and then indicating which sampling operations to perform, such as which culture on the automatic sampler is performing well.

[0178] After loading the sample, the device can output the organism type within 2 to 8 hours. As described herein, the metabolic information can be automatically analyzed by the device and the control software. In another embodiment, there can be a method, for example, in a specific software, that matches the changes in the biomarkers listed herein for inhibitor identification or sensitivity, such as thresholds. For example, processing strategies such as data-dependent feedback in process control and data-dependent communication software can be employed. In another embodiment, there can be a method, for example, in software, that uses the thresholds for identification or toxin sensitivity to communicate (connect) and possibly control operations between the biomarker analysis machine (i.e., mass spectrometer) and the sample input machine (i.e., automatic sampler).

[0179] This method will greatly shorten the antibiotic testing time.

[0180] In one embodiment, for example, as Figure 25Schematically, for ease of identification and toxin sensitivity, the system / process may include: (i) a culturing method of culturing a sample in a first culture medium to promote metabolism for identification, and simultaneously culturing in a plurality of toxin-containing culture media, each medium having a toxin for a different pathogen, and (ii) an analysis method of, after culturing, analyzing the metabolic results of the first culture medium, such as biomarkers, to identify the pathogen, and then only analyzing those toxin-containing culture media that have a relevant toxin for the identified pathogen. All culturing can be performed on a common instrument, such as a single multi-well plate, so that the culturing for identification and the culturing for toxin sensitivity can all be performed at the same time and can be processed as a unit using an analytical instrument. The analysis can be guided by software configured to sample the first culture (upper left), the first culture having a culture medium for general pathogen identification and sending it for analysis (step 1), in step 2, the analysis identifies the pathogen, and then based on the identification, in step 3, only samples those cultures that have a toxin relevant to the identified pathogen (four samples 1). Thus, based on the culturing performed simultaneously with the sample processing as the pathogen identification culture, a sensitivity analysis is performed. In addition, the use of the sensitivity analysis instrument is only applicable when the pathogen grows in a toxin-containing culture medium that contains a toxin relevant to the identified pathogen. This helps to simplify the culturing and analysis process to achieve a diagnosis and treatment strategy more quickly.

[0181] These methods can be implemented by a computer and thus stored on a non-transitory computer-readable medium having instructions executable by one or more processors, such as in an autosampler.

[0182] In another embodiment, the present invention relates to a treatment regimen for treating an infection, comprising one or more aspects of the above methods and identifying a toxin for acting on the identified cell type. In another embodiment, a method for treating an infection includes one or more aspects of the above methods, and further administering to an infected patient an effective amount and type of antibiotic based on the identification of the microorganism and its antibiotic sensitivity using the obtained metabolite data. Thus, one or more aspects of the above methods and the identification of the microorganism and its antibiotic sensitivity based on the use of the obtained metabolite data are further used to select the type of antibiotic for treating an infected patient.

[0183] In another embodiment, the method according to the present invention includes screening the effectiveness of chemotherapy on cancer cells using one or more aspects of the above methods, wherein a chemotherapeutic agent is added to the growth medium.

[0184] In another embodiment, the method according to the present invention includes screening the effectiveness of gene recombination on cells using one or more aspects of the above method, wherein the recombinant-regulated metabolism and this method rapidly confirm that the recombination is successful.

[0185] In another embodiment, the present invention is a method for determining whether a patient is infected, including: obtaining a patient specimen (e.g., blood, urine, swab, stool) or a clinical specimen (i.e., a hospital equipment swab); combining the specimen with a growth medium; obtaining data on the metabolic activity of the organism; diagnosing an infected patient based on changes in metabolite biomarker concentrations or the presence relative to a reference metabolite profile. The method may further include recommending treatment or administering appropriate antibiotics and antibiotic doses related to the concentration of metabolites present.

[0186] In another embodiment, as described above, the method of the present invention determines whether food is contaminated with microorganisms.

[0187] The following examples are for illustrative purposes only and are not intended to limit the scope of the present invention or the claims.

[0188] Example:

[0189] Example I: MS Detection of Organisms

[0190] The growth medium used in this example, Mueller-Hinton, enables metabolites to be absorbed by microorganisms. This medium was prepared by dissolving 2 g of beef extract, 17.5 g of casein hydrolysate, and 1.5 g of starch in 1 liter of deionized water. A sample containing microorganisms was mixed with the growth medium, and a diagnostic data collection tool was used to evaluate the biomarkers consumed and produced. The microbial sample was prepared by making a 0.5 McFarland standard dilution (approximately 1x 10 8)To prepare: Escherichia coli, EC; Klebsiella pneumoniae, KP; Pseudomonas aeruginosa, PA; Staphylococcus aureus, SA; Enterococcus faecalis, EF; Streptococcus pneumoniae, SP; and Candida albicans, CA. These 7 target microorganisms were selected because they cause more than 85% of human blood infections. At 0 hours, standardized microbial samples were mixed with Mueller-Hinton growth medium at a 1:1 ratio. An aliquot (100 μL) of each sample was harvested immediately, quenched metabolically by mixing the aliquot with an equal volume of methanol, and stored at 4 °C until data collection for metabolite profiling was available. Another microbial aliquot was incubated at 37 °C in its growth medium for 4 hours. At 4 hours, the sample (100 μL) was harvested, mixed with an equal volume of methanol to quench metabolism, and transferred to 4 °C until data collection for metabolite profiling was obtained. The microbial culture experiments were completed three times to generate three independent biological replicates. Then, all samples were analyzed by liquid chromatography-mass spectrometry (LC-MS) using a hydrophobic interaction liquid chromatography column and a high-resolution mass spectrometer in positive and negative ionization modes. More than 250 metabolites were analyzed by LC-MS, and metabolite standards were used to analyze on the same LC-MS platform to verify retention times and ionization characteristics. Known metabolites (defined by co-elution of observed signals with reference standards using extracted ion chromatograms with 5 ppm mass windows) and unknown metabolites were identified and metabolite intensities were determined. Metabolite levels before and after 4-hour incubation were analyzed, and 30 metabolites were confirmed to be sufficient to clearly distinguish the 7 different target microorganisms. These metabolites were adenine, adenosine, arginine, 4-aminobutyric acid, cytidine, glucose, glutaric acid, glycine, guanine, guanosine, hypoxanthine, inosine, N-acetyl-phenylalanine, ornithine, sn-glycerol-3-phosphate, succinate, taurine, uridine, urocanate, xanthine, and 9 signals from unknown metabolites. From this preliminary experiment, it was concluded that metabolite patterns in microbial growth media can be detected by LC-MS and used to distinguish common microbial pathogens.

[0191] Example II: Automatic Detection of Clinical Isolates

[0192] To determine the diagnostic feasibility of the present invention, 100 microbial cultures of clinical isolates were prepared and analyzed according to the method of Example 1. These 100 isolates were prepared from 9 groups of organisms ( Figure 6A ) representing common pathogens and commensal organisms observed in clinical diagnostic laboratories. Data were obtained for 250 known metabolites and all unknown signals. One-way analysis of variance ("ANOVA") was used to determine the 60 most statistically significant signals observed after 4-hour incubation. Hierarchical clustering of these selected biomarkers showed distinct species-related clustering at the metabolite level (Figure 7 )。All 60 of these diagnostic signals were used to build a support vector machine (SVM) model for microbial species. A total of 21 samples were used to create the SVM computer system to predict the presence of microbes in the remaining 79 clinical samples using metabolite levels. The SVM computer system correctly identified the organisms in a total of 79 blinded clinical samples. Additionally, no samples were misidentified in this analysis, indicating a sensitivity > 99% and a false discovery rate < 1%. This experiment demonstrated that an automated computer system based on metabolite levels in the analytical medium can correctly identify pathogens and common commensals from a representative transect of clinical isolates.

[0193] Table 2 shows the clinical isolates identified by automated computer analysis of metabolite levels. Classification was completed according to the steps of Example 2. The number of organisms correctly identified is shown - all organisms appropriately identified by the SVM computer system of metabolite levels observed after 4 hours of incubation in Mueller-Hinton growth medium in this study. Candida represents different species from the genus Candida; VRE Enterococcus faecium represents vancomycin-resistant Enterococcus faecium. It can be concluded from this study that automated data collection of metabolite levels in the growth medium is a feasible mechanism for the diagnostic evaluation of clinical microbiology samples.

[0194] Table 2

[0195]

[0196]

[0197] Example III: Sensitivity Limit Analysis

[0198] To ensure the compatibility of the present invention with clinical implementation, the analytical sensitivity of the device was tested. A culture of Pseudomonas aeruginosa was grown to approximately 0.5 McFarland. Then, the culture was diluted 5 orders of magnitude using serial 1:10 dilutions in metabolite-free phosphate-buffered saline. Then, the limit of detection of the metabolite-based analysis was determined using LC-MS according to the steps of Example 1 and compared with the traditional optical analysis of bacterial cell light scattering at 600 nm. Compared with the optical method, the LC-MS-based analysis showed a wider dynamic range and a lower limit of detection (< 100 cells / ml). This example demonstrated that the device and LC-MS analysis have sufficient sensitivity for the clinical application of the device.

[0199] Example IV: Automatic Detection of Antibiotic Sensitivity

[0200] Differentiating drug-induced metabolite level changes enables faster diagnostic analysis. To determine the feasibility of this method, in addition to supplementing the Mueller-Hinton growth medium of Example 1 with a series of antibiotic concentrations ( Figures 9 - 10 ), microbial cultures were prepared and analyzed and classified according to the steps of Example 1. Standard drug doses used in diagnostic laboratories were used to clinically screen for drug-sensitive and resistant strains. The metabolite patterns absorbed and secreted into the medium within the antibiotic dose range matched the minimum inhibitory concentration (“MIC”) ( Figure 9 ) determined for each bacterium. Additionally, despite 1% blood being deliberately added to the samples to simulate blood culture applications, the background metabolite signal of the blood did not overlap with the diagnostic signal of microbial metabolism ( Figure 11 ). This indicates that the metabolite-based drug-sensitivity system is compatible with the clinical application of this technology.

[0201] To determine whether automated metabolite analysis can be used to detect microbial drug sensitivity, changes in drug-induced biomarker levels were recorded for 36 clinically relevant strains (three species with sensitive and resistant isolates, six replicates each; Escherichia coli + / - extended-spectrum beta-lactamase, Klebsiella pneumoniae + / - carbapinem resistance, and Staphylococcus aureus + / - methicillin resistance). A computer model (SVM) was constructed to use biomarker levels to predict drug resistance. Then, resistance levels were predicted in 18 blinded test samples. The computer model correctly differentiated all drug-resistant strains ( Figure 12 ). This study shows that changes in toxin-induced metabolite levels can be detected and used to automatically classify antibiotic sensitivity for clinically relevant species.

[0202] Example V: Clinical Evaluation of Metabolite - Based Microbial Detection

[0203] To determine the compatibility of the metabolite-based microbial detection system, human blood cultures were collected directly from a clinical diagnostic laboratory and analyzed using MS. The existing VITEK (BioMerieux) culture system for high-volume bloodborne pathogen detection platforms functions by combining clinical blood specimens with microbial growth media. This is done to enable the growth of microorganisms for downstream protein analysis used in current technologies ( Figure 8)。Once the bacterial density reaches a detectable level (about 1,000 cells / ml), harvest the specimen and discard the culture flask. Collect these discarded blood cultures directly from the clinical sample stream and analyze them using existing metabolite-based detection platforms. Quench microbial metabolism by mixing 100 μl of the specimen with an equal volume of methanol, remove insoluble components by centrifugation, and analyze the soluble extract by LC-MS. LC-MS analysis revealed general biomarkers of infection that distinguish positive and negative cultures, as well as species-specific biomarkers similar to those in Examples 1, 2, and 3( Figures 5 - 7 )。This study demonstrated that metabolite detection technology can be directly integrated into existing clinical workflows.

[0204] This study also showed that blood metabolites and pathogen metabolites can be detected, such that it is even possible to analyze samples of cell mixtures and identify the cells therein. The identification of cell mixtures is also shown in Figure 8A 。

[0205] Example VI: Detection of Organisms and Determination of Drug Sensitivity Using NMR

[0206] Although LC-MS is a powerful method for detecting changes in the metabolic composition of growth media, it is just one example of the various possible analytical techniques that can be used to detect microbial metabolism. To determine the feasibility of an NMR-based microbial detection system, obtain and prepare clinical blood samples according to the procedure of Example V. Then, dry the extract to remove the methanol component and resuspend it in 100% D2O containing 500 μM 4,4-dimethyl-4-silapentane-1-sulfonic acid (DSS; an internal standard for referencing chemical shifts). Obtain the 1 1H NMR spectrum of the medium on an instrument for Pseudomonas aeruginosa positive cultures at 600 MHz. These data showed the metabolic changes caused by Pseudomonas aeruginosa.

[0207] These studies indicate that NMR has sufficient sensitivity to detect microbial metabolic activity in samples collected directly from existing clinical routes.

[0208] To determine whether NMR can be used to distinguish microorganisms, cultures of 8 different microorganisms (Candida albicans, Escherichia coli, Klebsiella pneumoniae, Enterococcus faecalis, Pseudomonas aeruginosa, coagulase-negative Staphylococcus, Streptococcus pneumoniae, and Staphylococcus aureus) were inoculated into BacT blood medium (BioMérieux) or Mueller-Hinton medium and grown for 4 hours. Metabolites were extracted as described above (Example VI) and analyzed by multidimensional 1H-13C heteronuclear single quantum coherence (HSQC) NMR. The diagnostic regions of interest in the NMR spectra corresponding to 7 target sugars were extracted and compared with the reference signals of metabolite standards prepared at 100 mM. The patterns of sugars observed in the growth medium were sufficient to distinguish all target pathogens. This study demonstrated that NMR can distinguish types of microorganisms.

[0209] To determine whether NMR can be used to detect antibiotic-induced microbial metabolic disorders, cultures of tetracycline-sensitive and -resistant isolates of Pseudomonas aeruginosa were prepared according to the procedure of Example 1. M9 medium was used instead of Mueller-Hinton growth medium (M9 medium was prepared according to a known formula, such as disclosed at http: / / cshprotocols.cshlp.org / content / 2010 / 8 / pdb.rec12295.short, which is 47.7 mM Na2HPO4, 22 mM KH2PO4, 8.6 mM NaCl, 18.7 mM NH4Cl, 22 mM glucose, 2 μM MgSO4, and 100 nM CaCl2). Growth media with and without 60 μg / ml tetracycline were prepared, and each strain was inoculated and incubated for 12 hours. NMR samples of the media were prepared and analyzed as described above. The NMR data showed that all resistant isolates and drug-sensitive isolates incubated in the antibiotic-free medium were metabolically active; each active strain consumed glucose and produced acetate and pyoverdine of Pseudomonas aeruginosa ( Figure 14 ). In contrast, NMR analysis of the drug-sensitive cell lines incubated with tetracycline showed metabolic inactivation (i.e., minimal glucose consumption and minimal production of acetate and pyoverdine of Pseudomonas aeruginosa). This study demonstrated that NMR can detect the inhibition of drug-induced microbial metabolism.

[0210] Although Figure 14 the analysis was by NMR, the simplicity of the system, including the use of a simple growth medium containing only limited nutrients, also allows for simpler chemical analysis. For example, using a blood glucose meter can yield results for differentiating resistant isolates. When simpler medium options are used, growth may be slower, but it may be beneficial for growth medium analysis and identification of control reference spectra.

[0211] Example VII: Determination of Drug Sensitivity Using Spectrophotometry

[0212] While NMR and MS provide opportunities for decoding complex metabolite mixtures, the optically engineered growth medium also provides a mechanism for differentiating microorganisms and measuring their sensitivity to toxins. To determine the feasibility of using an optical-based method to detect drug-induced microbial metabolic disorders, Pseudomonas aeruginosa cultures were prepared as described in Example V. M9 growth medium, as described in Example VI, was used in this example because of its minimal optical properties, which enable sensitive detection of optically active metabolic waste. Drug-sensitive and resistant strains were incubated in M9 medium containing 0, 60, and 600 μg / ml tetracycline for 4 or 8 hours. The cells were then removed by centrifugation, and the cell-free medium composition was analyzed by spectrophotometric absorbance measurements at 400 nM. The Pseudomonas aeruginosa siderophore secreted by Pseudomonas aeruginosa absorbs at this wavelength and can be used as a marker of metabolic activity ( Figure 15 ). The drug-sensitive isolates showed impaired production of the Pseudomonas aeruginosa siderophore when incubated with tetracycline, while the drug-resistant cell lines did not. In addition, the degree of impairment of Pseudomonas aeruginosa siderophore secretion was proportional to the tetracycline concentration ( Figure 15 ). This study shows that spectrophotometric analysis can be used to detect microorganisms and measure drug sensitivity.

[0213] Example VIII: Metabolite Signatures of Common Pathogens

[0214] A group of the following organisms were analyzed by mass spectrometry after incubation in Mueller-Hinton medium:

[0215] Candida albicans, Candida subspecies (other species of Candida), Escherichia coli, Klebsiella oxytoca, Klebsiella pneumoniae, Pseudomonas aeruginosa, Enterococcus faecium, Staphylococcus aureus, Streptococcus pneumoniae, Group A Streptococcus, Streptococcus viridans, Coagulase-negative staphylococci (CN-staphylococci) (CN-Staph).

[0216] The following metabolites that can be used for identification were identified and saved as reference metabolite profiles:

[0217] · Succinate levels distinguish Escherichia coli and Klebsiella subspecies from all other organisms in the group;

[0218] · Urocanate levels distinguish Escherichia coli and Klebsiella subspecies;

[0219] · Hydroxydecanoate distinguishes Pseudomonas aeruginosa from all other organisms in the group;

[0220] · Arabitol levels distinguish yeast from all other organisms in the group;

[0221] · Glucose levels distinguish Pseudomonas aeruginosa from all other Gram-negative organisms;

[0222] · N-acetyl-aspartate levels distinguish Enterococcus from all other organisms in the group;

[0223] · Xanthine distinguishes Streptococcus viridans from all other organisms in the group;

[0224] · The presence of galactose diagnoses Escherichia coli; and

[0225] · The glucose-to-lactose ratio distinguishes Enterococcus faecalis from other Gram-positive organisms.

[0226] A group of organisms was analyzed by NMR after incubation in BacT medium:

[0227] · Sucrose levels distinguish Escherichia coli, Klebsiella spp., and Staphylococcus aureus; and

[0228] · Trehalose levels distinguish CN-Staphylococcus from Staphylococcus aureus and Escherichia coli.

[0229] Example IX: Further Metabolomics Testing

[0230] In a large study, metabolic preference analysis (MPA) for measuring supernatant biomarker production and consumption disclosed herein was used to distinguish 7 different species (Candida albicans, Klebsiella pneumoniae, Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, Enterococcus faecalis, and Streptococcus pneumoniae) that account for 85% of bloodstream infections. Strains were grown in Mueller-Hinton medium for 4 hours and the supernatant was analyzed by ultra-high performance liquid chromatography-mass spectrometry (UHPLC MS) to identify potential biomarkers. For initial biomarker discovery, 3 clinical isolates per species (2 for PA) were analyzed in duplicate (n = 9). Untargeted MS analysis was used to discover biomarkers that were common to all 3 isolates per species and that could distinguish different species. Subsequently, the previously identified biomarkers were evaluated for stability in a large number of clinical isolates (n = 596) representing these 7 species. The molecular properties of the selected biomarkers were verified by comparing standards and samples using standard addition and MS / MS fragmentation patterns.

[0231] To evaluate the specificity and selectivity of MPA for discriminating pathogens in blood, a further blinded study was performed on patient samples (n = 809) collected within 10 days. In addition, to test the reliability of MPA for antibiotic susceptibility testing (AST), metabolomics-based antibiotic susceptibility testing (MAST) was performed to study the metabolite concentration changes of each original strain (n = 3; n = 2 for Staphylococcus aureus and Pseudomonas aeruginosa) grown in the presence of antibiotics for biomarker discovery.

[0232] Strains, growth, and sample preparation

[0233] Unless otherwise stated, all chemicals were obtained from the following companies: Sigma-Aldrich (St. Louis, Missouri, USA), VWR (Radnor, Pennsylvania, USA), or Fisher Scientific (Waltham, Massachusetts, USA). Clinical isolates used in this study were obtained from patients with bloodstream infections. All strains were first identified and tested for antibiotic susceptibility using the clinical laboratory testing route (single colony isolates were streaked on selective media, Gram stained, identified by MALDI-TOF-MS species identification, subcultured to pure cultures and inoculated into VITEK 2 (bioMerieux)), and then refrigerated. All strains except Streptococcus pneumoniae grew normally on Mueller-Hinton medium (BD Difco, Mississauga, Ontario, Canada). Streptococcus pneumoniae isolates were first revived on tryptic soy agar plates containing sheep blood (BD BBL, Mississauga, Ontario, Canada), and then subcultured on Mueller-Hinton medium supplemented with catalase (1000 U / mL). For biomarker discovery and biomarker validation using MPA, exponential-phase cultures were used to inoculate 96-well culture plates (Corning, New York, NY, USA) containing Mueller-Hinton medium including 10% donated human blood to 0.5 McFarland (OD 600 ~0.07 or ~1.5×10 8CFU / mL). The cultures were incubated for 4 hours in a humidified incubator (Heracell VIOS250i Tri-Gas Incubator, Thermo Scientific, Waltham, Massachusetts, USA) in an atmosphere of 5% CO2 and 21% O2. After incubation, the samples were transferred to a 96-well PCR plate (VWR) and centrifuged at 4000 g for 10 minutes at 4 °C to remove the cells. The supernatant was removed, mixed 1:1 with 100% LC-MS grade methanol, frozen at -80 °C for further processing or centrifuged again at 4000 g for 10 minutes at 4 °C to remove any protein precipitate. Then, the supernatant was diluted 1:10 with 50% LC-MS grade methanol and analyzed using UHPLC-MS. However, AST MPA was performed as described above, inoculating the cultures with 0.05 McFarland, without adding blood to avoid possible antibiotics or antibodies carried from the donated blood, and measuring the periodic growth (Mutiskan GO, Thermo Fisher Scientific, Waltham, Massachusetts, USA). The antibiotics used for each species were based on the prevalence of use for treatment. The published strain-specific minimum inhibitory concentrations (MICs) of each antibiotic were used (C.L.S.I. (CLSI). (CLSI, Wayne, PA, USA, 2018), pp. 1 - 296.). 600 The antibiotics used for each species were based on the prevalence of use for treatment. The published strain-specific minimum inhibitory concentrations (MICs) of each antibiotic were used (C.L.S.I. (CLSI). (CLSI, Wayne, PA, USA, 2018), pp. 1 - 296.).

[0234] UHPLC-MS

[0235] The metabolite samples were resolved using hydrophilic interaction liquid chromatography (HILIC) by the Thermo Fisher Scientific Vanquish UHPLC platform. A binary solvent mixture of a solution of LC-MS grade water with 20 mM ammonium formate pH 3.0 (solvent A) and a solution of LC-MS grade acetonitrile with 0.1% formic acid (% v / v) (solvent B) was used and a 100 mm × 2.1 mm Syncronis with a particle size of 2.1 μm was used TMChromatographic separation was carried out using a HILIC LC column (Thermo Fisher Scientific). For general metabolic profiling runs (15 minutes), the following gradient was used: 0 - 2 minutes, 100% B; 2 - 7 minutes, 100 - 80% B; 7 - 10 minutes, 80 - 5% B; 10 - 12 minutes, 5% B; 12 - 13 minutes, 5 - 100% B; 13 - 15 minutes, 100% B. For the accelerated runs (5 minutes) used in the ID and ASTrace experiments, the gradient was as follows: 0 - 0.5 minutes, 100% B; 0.5 - 1.75 minutes, 100 - 80% B; 1.75 - 3 minutes, 80 - 5% B; 3 - 3.5 minutes, 5% B; 3.5 - 4 minutes, 5 - 100% B; 4 - 5 minutes, 100% B. The flow rate used for all analyses was 600 μL / min and the sample injection volume was 2 μL. Samples were ionized by electrospray using the following conditions: spray voltage was -2000 V, sheath gas was 35 (arbitrary units), auxiliary gas was 15 (arbitrary units), sweep gas was 2 (arbitrary units), capillary temperature was 275 °C, and auxiliary gas temperature was 300 °C. The positive ion mode source conditions were the same except that the spray voltage was +3000 V. On a Thermo Scientific Q Exactive TM HF (Thermo Scientific) mass spectrometer, data was acquired using full scan acquisition (50 - 750 m / z) with a resolution of 240,000, an automatic gain control target of 3e 6 and a maximum injection time of 200 ms. All data was acquired in the negative ion mode except for MS / MS fragmentation analysis and confirmation of N 1 ,N 12 -diacetylspermine (which ionizes more efficiently in the positive ion mode). MS / MS analysis was used to confirm selected biomarkers under conditions where the collision energy range for the previously observed parent ions was 10 - 50 eV, the resolution was 30,000, the automatic gain control target was 5e 4 , and the isolation window was 4 m / z. Fragmentation spectra and retention times were used to match biomarkers to standards. N 1 ,N 12-Diacetylspermine was purchased from Cayman Chemical Company (Ann Arbor, Michigan, USA), and all other standards were purchased from Sigma-Aldrich. The cleavage data were analyzed using Xcalibur 4.0.27.19 software (ThermoScientific). All other MS analyses were performed using MAVEN (H. Li et al., Adaptable microfluidic system for single-cell pathogen classification and antimicrobialsusceptibility testing. Proc Natl Acad Sci USA, (2019)).

[0236] Because some differences were noted in the negative ion mode (such as regarding N 1 ,N 12 -diacetylspermine), so some data were acquired in positive ion mode. The test was repeated as described above, but the data were acquired in positive ion mode. This identified additional biomarkers of interest, as shown in Figures 18B (i) to (v).

[0237] Clinical Microbiology and Test Methods

[0238] All isolates used in this study were obtained from patient blood specimens. Blood cultures were collected aseptically by trained phlebotomists. Two sets of blood cultures were drawn from adults, each from a separate venipuncture with a total blood volume of 40 mL (i.e., each bottle set included both aerobic (FA) and anaerobic (FN) blood culture resin media). Children had only a single bottle (PF) drawn. All blood specimens were processed identically using an automated blood culture microbiology testing system (bioMérieux Inc., Saint-Laurence, Quebec, Canada). Specimens were monitored continuously for growth, and blood culture bottles were immediately subjected to Gram staining and pelleting when positive. Blood culture pellets were subcultured on appropriate solid agar media and subsequently identified using a combination of the following methods: matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS) (bioMérieux Inc., Saint-Laurence, Quebec, Canada), the VITEK 2 automated system (bioMérieux Inc., Durham, NC, USA), and rapid phenotypic testing as needed. Isolates that could not be identified using conventional phenotypes and / or MALDI-TOF MS were subsequently identified using DNA sequencing of the 16S rRNA gene and a comprehensive database network system. Identify using the (https: / / www.SmartGene.com, Lausanne, Switzerland) database to analyze the variability of the V1-V3 region (Clinical Laboratory Standards Institute (CLSI). Interpretive criteria for identification of bacteria and yeast using DNA sequencing. M18-2 nd Edition, c2018, pp. 1-296, CLSI, Wayne, PA USA). Obtain the antimicrobial spectrum simultaneously using the VITEK 2 automated system. Verify the antibiotic resistance of each isolate according to the previously cited guidelines. Freeze and archive the patient isolates obtained through this process.

[0239] Blinded ID testing of patient samples

[0240] To evaluate the specificity and selectivity of MPA to identify pathogens in blood samples using biomarkers that can distinguish our 7 target organisms, a blinded study was conducted on patient samples (n = 809) collected within 10 days. Blood extracted from patients suspected of having bloodstream infection (BSI) was incubated in BacT bottles until they showed positive, or incubated for 5 days, at which time the samples were considered negative. Positive samples were subjected to pathogen identification through the clinical workflow. Specimens (10%) of positive and negative samples were also incubated in MH medium for 4 hours and then quenched with methanol according to the MPA protocol. The samples were then subjected to MS analysis. Identification was performed using the 20 biomarkers depicted in the Figure 16 decision tree. Set the minimum fold change threshold for each biomarker (compared to the MHB control) to classify biomarker changes. Conduct MPA identification calls (∼80 per day) every day, and then provide clinical identification before each new sample group to refine the biomarker patterns of species that have never been evaluated by MPA before. This can discover other biomarkers ( Figure 16 MS4 in and identify new patterns to help distinguish selected organisms that have not been evaluated (Figure 18A).

[0241] Real-time ID and AST testing

[0242] Exponential-phase Escherichia coli and Staphylococcus aureus were used to inoculate bottles supplemented with 10 mL of human blood, and the final CFU / mL was 25 (corresponding to 100 CFU / mL of patient blood). At Incubation bottles in an automated blood culture microbial detection system. Once the bottles are labeled, one specimen is used for the clinical trial route and the second specimen is used for MPA. A 10% blood-bacteria-BacT / Alert medium mixture is inoculated into a medium containing the most commonly used antibiotics for each strain (Staphylococcus aureus, CIP, OXA, SXT, CFZ, and AMP; Escherichia coli, CIP, SXT, AMP, and GEN) in a concentration range consistent with the VITEK 2 automated system. After a 4-hour incubation period, the samples are processed and analyzed by UHPLC-MS using a 5-minute HILIC-MS method. The data is analyzed in real-time using the MAVEN software package. The positive control (medium without antibiotics) is analyzed first to enable species identification. Subsequently, the samples containing the lowest concentration of antibiotics are analyzed to evaluate sensitivity. Samples incubated with higher concentrations of antibiotics are only analyzed if the strain shows resistance to the lower concentration of antibiotics, in order to minimize the time for MS analysis.

[0243] Statistical analysis

[0244] Non-target biomarkers in the preliminary dataset (7 species, 3 isolates, 9 replicates) were identified by peak picking data in Maven with a 10 ppm m / z window and a minimum peak intensity set to 50,000. All subsequent tests were performed using in-house software tools with the R statistical software platform (R Core Team, R: A language and environment for statistical computing. R foundation for Statistical Computing, Vienna, Austria. http: / / www.R-project.org / ). Non-target analysis identified 4,372 signals in the mass spectrum. This list was sorted according to p-values (calculated by ANOVA), and after Bonferroni correction (α = 0.05), 1,758 signals were identified as significant signals at 4 hours. These signals were further thresholded (peak apex area > 20,000), and fold change (interspecies pairwise mean difference > 2-fold) resulted in 799 peaks. Then, these signals were clustered into 104 groups using a weighted probability function, considering retention time, common adducts / fragments / isotope masses, and covariance of signal intensities among all replicates (equal weights) using custom R software. Then, the most likely parent ion was selected from each group based on signal intensity and evaluated by manual inspection of the original MS data. Then, informatics tools (a combination of the Madison Metabolomics Consortium Database (Q. Cui et al., Metabolite identification via the Madison Metabolomics Consortium Database. Nat Biotechnol 26, 162-164 (2008)) and the Human Metabolome Database (D. S. Wishart et al., HMDB 4.0: the human metabolome database for 2018. Nucleic Acids Res 46, D608-D617 (2018))) were used to assign the parent ion of each biomarker. Then, the putative metabolite assignments were verified by purchasing standards and performing MS / MS fragmentation and standard addition experiments.

[0245] Results: Metabolomics Identification

[0246] As described above, the metabolic boundary fluxes of 7 common blood pathogens [Candida albicans (CA), Klebsiella pneumoniae (KP), Escherichia coli (EC), Pseudomonas aeruginosa (PA), Staphylococcus aureus (SA), Enterococcus faecalis (EF), and Streptococcus pneumoniae (SP)] were determined. For the initial biomarker discovery, replicate analyses (n = 9) were performed on 3 clinical isolates for each target species (n = 7) (2 for PA). Microbial cultures were inoculated into Mueller-Hinton broth (MHB) containing 10% human blood at 0.5 McFarland (OD 600 ~0.07 or 1.5×10 8 CFU / mL). At 0 and 4 hours, metabolite levels present in the cultures were analyzed in negative ion mode on a Thermo Q Exactive TM HF MS. Non-target analysis identified 799 peaks with significant fold changes between species (calculated by ANOVA with Bonferroni-corrected α = 0.05). These signals were clustered into 104 groups based on retention time, known adduct / fragment masses, and covariance of signal intensities between replicates using R software. Then, the most likely parent ion of each biomarker was identified from each group of the 104 groups based on signal intensity. Putative metabolite assignments were made through the Madison Metabolomics Consortium Database (MMCD) and the Human Metabolome Database, and selected metabolites were verified by MS / MS fragmentation and standard addition.

[0247] Species-dependent consumption or production of the 104 selected biomarkers differed significantly among the 7 target species, as Figure 17A and 17B shown. Although the overall patterns of biomarkers were similar between closely related microorganisms (i.e., Klebsiella pneumoniae and Escherichia coli), they could still be distinguished by the selected biomarkers.

[0248] To determine the stability of these biomarkers in a larger population, we performed a validation study of metabolic preference analysis (MPA) using 596 clinical isolates. Changes in the top 104 biomarkers were consistent with those observed in the biomarker discovery dataset. Significantly, only 7 production biomarkers were sufficient to distinguish the target pathogens and served as binary predictors for each species ( Figure 17A and 17B ). In particular, D-arabitol, xanthine, and N 1 ,N 12-Diacetylspermine is produced only by Candida albicans, Pseudomonas aeruginosa, and Enterococcus faecalis, respectively. Both Klebsiella pneumoniae and Escherichia coli produce succinate, but the latter does not produce urocanate. Mevalonic acid is produced by Staphylococcus aureus and to a lesser extent by Enterococcus faecalis, but unlike Enterococcus faecalis, Staphylococcus aureus does not produce N 1 ,N 12 -Diacetylspermine. Lactate is produced by Streptococcus pneumoniae and to a lesser extent by Enterococcus faecalis. Equally interesting, as Figure 22A and 22B shown, regardless of the donor (n = 20), when compared to the inoculated culture on the right, the addition of 10% blood to the medium had a negligible effect on the metabolite profile, indicating that these biomarkers are pathogen-specific and do not reflect blood metabolism or donor-specific metabolite carriage.

[0249] Results: Blind performance testing of MPA in the clinical population

[0250] As described above, the data herein established a set of MPA-based biomarkers for differentiating the most common BSI pathogens. To evaluate the practical clinical application of MPA, we conducted a blind performance trial of our new method and scored it according to the results obtained from standard clinical trial practices. The trial was based on 809 blood cultures.

[0251] Interestingly, the MPA-based classification algorithm was calibrated for only 7 organisms, but this blind trial encountered many species (N = 131) that were not in the original training set. This highlights the advantage of the metabolomics-based MPA method as it is able to capture both targeted and non-targeted data. This enables the new classification algorithm to be trained or improved on the fly. Therefore, we were able to use these non-targeted metabolomics datasets, along with the results of the daily assigned batches, to establish preliminary prediction models for each new organism encountered. These preliminary microbial predictions were submitted together with the results of the models we established for the 7 target species (Figure 18A(i) and (ii)). The supporting data and species-specific performance data are shown in Table 3 for expanding the decision tree.

[0252] Table 3 Blood culture isolate ID and concordance of MPA with MALDI-TOF

[0253]

[0254]

[0255]

[0256] Blinded trials demonstrated that the MPA-based approach is an effective clinical diagnostic strategy. When considering only those organisms present in the training set, the MPA-based classification correctly differentiated every infected sample (N = 169) from non-infected samples (N = 477) without missing any. Additionally, the MPA-based analysis correctly classified each target organism in 88% of the samples (148 out of 169 at the species level). In particular, most misclassifications (10 out of 21) were due to the ambiguity between Klebsiella and Escherichia, which are closely related organisms with similar treatment needs. Considering all data, including organisms not included in the training set, the MPA-based approach correctly labeled infected samples in 96.1% of the samples with a specificity of 99.8% (N = 809, including 332 positives). Moreover, despite the dataset containing a large number of species not present in the training set, the MPA method was still able to identify pathogens at the genus level in 94% of the cases. This was achieved through our metabolomics approach, which enabled our classification model to be recalibrated in real time to handle newly observed organisms. Two examples of this real-time modeling were Streptococcus pyogenes (GAS) and coagulase-negative staphylococci (CNS), which were relatively abundant in the dataset (N = 9 and N = 48, respectively). Our dynamic model allowed us to correctly predict these organisms in 89% and 60% of the samples, despite never having seen them before and having only a few replicates available for training the model. A small number of samples (N = 34) grew multiple microorganisms. These mixed cultures were labeled as metabolically abnormal and did not conform to the classification scheme. In summary, MPA is an effective tool for pathogen identification, supporting dynamic recalibration of the classification scheme to accommodate unexpected new microorganisms.

[0257] Results: Rapid antibiotic susceptibility testing using MPA

[0258] An attractive aspect of using metabolomics for microbial diagnosis is that metabolism is a sensitive reporter of cellular physiology. Nutrient precursors are converted to waste products at rates many orders of magnitude faster than microbial growth. Additionally, these processes are significantly altered or completely halted when cells are exposed to toxic substances. Thus, metabolomics approaches provide a unique opportunity to empirically assess antibiotic susceptibility in a fraction of the time required by current growth-based antibiotic susceptibility testing (AST) methods. In this article, we evaluated the utility of an MPA-based AST (MAST) workflow.

[0259] The MAST test is completed by monitoring changes in the metabolic composition of microbial cultures after a 4-hour incubation period with and without an antimicrobial agent. Microorganisms are inoculated into MHB medium (to a final OD of ∼0.007) at 10% of 0.5 McFarland 600 ), and metabolomics analysis is performed using the same method as used for the general MPA test. Figure 19 The upper left panel of Figure 19 illustrates MAST, which relies on metabolic concepts: drug-sensitive strains of KP showed a progressive decrease in inosine production proportional to meropenem concentration, while resistant strains remained unaffected within clinically relevant antimicrobial concentrations. Similar antibiotic-induced metabolic perturbations were observed in all target pathogens when isolates were exposed to commonly used antibiotics at their minimum inhibitory concentrations ( Figure 19 ).

[0260] To evaluate MAST as a potential clinical tool, 3 patient isolates (2 patient isolates each of Staphylococcus aureus and Pseudomonas aeruginosa) of each target pathogen were analyzed. Antifungal drugs (azoles, polyenes, and antimetabolites) were tested against Candida albicans. Bactericidal antibiotic classes (penicillins, cephalosporins, carbapenems, glycopeptides, aminoglycosides, and fluoroquinolones) and bacteriostatic antibiotic classes (macrolides, tetracyclines, and trimethoprim / sulfamethoxazole) were evaluated. The antimicrobial susceptibility curves of the MAST assay were consistent with 98% of the curves observed in traditional microbial growth analyses. The test results were consistent across all antimicrobial drug mechanisms of action. For example, succinate production of ampicillin (AMP)- and trimethoprim / sulfamethoxazole (SXT)-resistant Escherichia coli was compared when the strains were incubated in the presence of AMP, SXT, or without an antimicrobial agent. However, succinate production was significantly lower when the strains were grown in the presence of their sensitive antibiotic (all paired comparisons, p < 0.01). In most cases, the biomarkers used to identify microorganisms were also useful for differentiating drug-sensitive and resistant strains (e.g., arabitol for Candida albicans, succinate for Klebsiella pneumoniae and Escherichia coli, N 1 ,N 12 -diacetylspermine for Enterococcus faecalis, xanthine for Pseudomonas aeruginosa, and lactate for Streptococcus pneumoniae). An exception to this trend was mevalonic acid, which was a good biomarker for Staphylococcus aureus but an unreliable biomarker for drug resistance. Instead, an alternative compound with an m / z of 204.069 was identified as a more stable metric for differentiating resistant and sensitive strains of Staphylococcus aureus. Subsequently, we evaluated MAST in a larger cohort (N = 300) and found that MAST is an excellent predictor of antibiotic susceptibility. These data confirm that MPA, together with MAST, can provide a sound mechanism for differentiating and characterizing pathogens.

[0261] Result: Real-time test result

[0262] One of the main motivations for this project was the urgent need for rapid diagnostic testing technologies. To use our MPA diagnostic workflow to evaluate potential time savings, we conducted a competition between an academic laboratory and the standard clinical testing route. 10 mL of blood containing 100 CFU / mL of exponential-phase bacteria (Staphylococcus aureus and Escherichia coli) was inoculated into aerobic BacT / Alert bottles and incubated in a BacT / Alert 3D (bioMérieux) microbiology detection system until the bottles were marked positive. One specimen was tested using the standard clinical testing route (spreading single colony isolates on selective media, Gram staining, MALDI-TOF-MS species identification, and subculturing of pure cultures and inoculation onto VITEK 2 (bioMerieux) antibiotic susceptibility panels), and a second specimen was used for diagnosis by MPA. A 10% blood-bacteria-BacT / Alert media mixture was inoculated into media containing the most common antibiotics for each strain (CIP, OXA, AMP, CFZ, SXT, and CIP for S. aureus; AMP, GEN, SXT, and CIP for E. coli) in a concentration range consistent with MicroScan Panels. After a 4-hour incubation period, the samples were processed and analyzed by LC-MS using a 5-minute HILIC method. Data were analyzed in real time using the MAVEN software package. Positive controls (media without antibiotics) were analyzed first to identify the species. Subsequently, samples containing the lowest concentrations of antibiotics were analyzed to assess sensitivity. To minimize MS analysis time, samples incubated in higher concentrations of antibiotics were only analyzed if they showed resistance to lower concentrations. Our MPA-based setup and MAST test results were consistent with the standard clinical test results and reduced the total test time, on average 40.6 hours for S. aureus and 44.3 hours for E. coli, corresponding to 3.0- and 3.9-fold reductions in total test time, respectively. The time required for strain identification (ii) and antibiotic susceptibility (iii) was reduced by 8.2- and 9.0-fold for S. aureus and E. coli, respectively.Studies have shown that a one-hour delay in effective antimicrobial therapy in patients with septic shock results in a 7.6% increase in mortality (A. Kumar et al., Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 34, 1589 - 1596 (2006)), and up to 30% of patients are given ineffective antimicrobial therapy (E. H. Ibrahim, G. Sherman, S. Ward, V. J. Fraser, M. H. Kollef, The influence of inadequate antimicrobial treatment of bloodstream infections on patient outcomes in the ICU setting. Chest 118, 146 - 155 (2000); and A. Kumar et al., Initiation of inappropriate antimicrobial therapy results in a fivefold reduction of survival in human septic shock. Chest 136, 1237 - 1248 (2009)). Given this, a reduction in BSI detection time of > 40 hours means a > 5% reduction in mortality in patients with septic shock.

[0263] Example X: Precursors Identification Study

[0264] To identify whether the biomarker is from glucose or from some other nutrient, 13 a [¹³C]-glucose labeling experiment was performed.

[0265] The organism was inoculated into RPMI medium supplemented with uniformly labeled 13 [¹³C]-glucose to determine whether the biomarker is from glucose or from other possible precursors present in RPMI. When the biomarker is detected mainly in the unlabeled form ( 12 [¹²C]), they are from RPMI components other than glucose.

[0266] Table 5 [¹³C]-glucose growth experiment in RPMI medium 13 [¹³C]-glucose growth experiment

[0267]

[0268]

[0269]

[0270] Subsequent metabolic pathway analysis was used to identify putative precursors of the identified biomarkers.

[0271] Example XI: Comparative Study of Common Pathogen Identification

[0272] Known organisms were further tested using Mueller-Hinton (MH) medium, Roswell Park Memorial Institute (RPMI) medium, or M9 medium. The method was the same as Example I above, with a culture time of 4 hours. The RPMI medium was prepared as shown in Table 6.

[0273] Table 6 Composition of Roswell Park Memorial Institute (RPMI) Medium

[0274]

[0275]

[0276]

[0277] Comparison of the production of the foregoing biomarkers of arabitol, xanthine, succinate, urocanate, nicotinate, and citrulline in Mueller-Hinton and RPMI media demonstrated that microorganisms CA, PA, EC, KP, SA, CNS, SP, and EF had similar production patterns in the two media. The data showed that some diagnostic target biomarker metabolites appeared in MH but not in RPMI, such as a) N 1 ,N 12 -diacetylspermine and b) mevalonic acid. The reason for their absence was that the precursors necessary for the overproduction of these molecules were absent or the amount of the molecules in RPMI was sufficient to inhibit the biosynthesis of the target biomarker.

[0278] Growth media enable microorganisms to uptake metabolites. In some cases, the identification of nutrients involved in metabolism was confirmed by drop out tests, in which customized RPMI-based media were established with suspected nutrient precursors omitted. In particular, to determine whether the biomarker was actually produced by the listed substrates, each substrate was omitted one by one from the RPMI medium. It was concluded that when the medium without a component did not produce the biomarker, the biomarker was produced by the omitted substrate ( Figure 20A and 20B)。Individual components (abscissa) were omitted from the culture medium to determine whether the omission of a specific substrate eliminated biomarker production. For example, the removal of histidine from the culture medium caused KP and GAS not to produce urocanate. In contrast, the elimination of any component alone did not affect the production of arabitol in CA, indicating that arabitol is derived from glucose.

[0279] The following conclusions were drawn:

[0280] · When samples are grown in Mueller-Hinton (MH) or Roswell Park Memorial Institute (RPMI) medium (each medium containing nicotinamide (niacinamide) and pyridoxine), the production of nicotinate demonstrates the presence of Escherichia, Klebsiella, Pseudomonas, Enterococcus, Staphylococcus, and Streptococcus in the samples. In other words, after culturing a sample of an unknown microorganism in MH or RPMI, if there is a higher concentration of nicotinate compared to the original medium, then it can be concluded that the sample contains at least one of Escherichia, Klebsiella, Pseudomonas, Enterococcus, Staphylococcus, or Streptococcus.

[0281] · When samples are grown in MH or RPMI medium (each medium containing arginine),

[0282] Metabolism that causes the production of citrulline indicates the presence of Gram-positive Enterococcus or Streptococcus in the sample.

[0283] · When samples are grown in MH or RPMI medium, each medium containing glucose: a) Cultivation that causes the production of arabitol indicates the presence of Candida, such as Candida albicans;

[0284] and b) Cultivation that causes the production of succinate indicates the presence of Escherichia or Klebsiella in the sample.

[0285] · When samples are grown in RPMI medium containing hypoxanthine, nicotinamide, and pyridoxine,

[0286] The production of xanthine and 6-hydroxynicotinate indicates the presence of Pseudomonas in the sample. It is believed that xanthine is produced from hypoxanthine, and 6-hydroxynicotinate is produced as a metabolite of nicotinamide and pyridoxine.

[0287] · When samples are grown in MH or RPMI medium containing histidine, metabolism that causes the production of urocanate indicates the presence of Klebsiella and / or group A Streptococcus in the sample.

[0288] · When samples are grown in MH medium containing glucose and threonine, metabolism that causes the production of mevalonic acid and N-acetylthreonine indicates the presence of Staphylococcus or Enterococcus in the sample.

[0289] · When the sample grows in MH, it causes N 1 , N 12 - The metabolism of diacetylspermine production indicates the presence of Enterococcus in the sample. Spermine in MH is considered to be the precursor.

[0290] · In MH, an increase in the concentration of methylbutylamine indicates the presence of Proteus in the cultured sample.

[0291] · In the culture medium, the presence of N 1 , N 8 - The increase in diacetylspermidine indicates the presence of Enterococcus faecalis, Staphylococcus saprophyticus, and Staphylococcus epidermidis in the cultured sample. This biomarker is useful for differentiating Enterococcus faecalis and Enterococcus faecium, where Enterococcus faecalis produces N 1 , N 8 - diacetylspermidine, while Enterococcus faecium does not.

[0292] · Tyramine indicates Enterococcus.

[0293] · Fumarate is produced by Escherichia coli and Klebsiella.

[0294] · N-acetylornithine is produced by SP.

[0295] · Methylbutylamine is produced by Proteus.

[0296] · Aminopropanol is produced by Proteus.

[0297] · Pyrrolidine is produced by Citrobacter, EC, Enterobacter cloacae, and Proteus mirabilis.

[0298] · There are many species of yeast. Arabitol has been identified as a marker for yeast and especially Candida albicans, but other biomarkers have been identified for the identification of other species. For example, N-acetylleucine / N-acetylisoleucine indicates Candida freundii, and this biomarker can be used to differentiate it from Candida albicans. In addition, the biomarker with a retention time of 4.3 minutes and a mass of 286.2366 in the 15-minute HILIC method indicates Candida albicans and can be used to differentiate it from Candida freundii.

[0299] · Agmatine in the cultured sample indicates the presence of species of Enterobacteriaceae. In a culture medium such as M9 medium, which has a single carbon source, such as where the medium contains only glucose as the carbon source, the metabolite agmatine indicates the presence of species of Enterobacteriaceae. Optionally, in a medium with arginine such as MH or RPMI, a cultured sample containing the metabolite agmatine indicates the presence of species of Enterobacteriaceae. ( Figure 21 )

[0300] · After cultivation, the presence of cadaverine is indicative of Escherichia coli, Enterobacter aerogenes, Klebsiella spp., and Stenotrophomonas maltophilia. This can assist in differentiating these pathogens within the family Enterobactericiae, since all Enterobacteriaceae detected were found to produce agmatine.

[0301] Enterobactericiae), as all Enterobacteriaceae detected were found to produce agmatine.

[0302] · After cultivation, the presence of putrescine can identify Citrobacter spp., Escherichia coli, Enterobacter spp., Klebsiella spp., and Proteus mirabilis, and can assist in differentiating Enterobacteriaceae identified by agmatine.

[0303] · In RPMI, Enterococcus does not produce N 1 ,N 12 -diacetylspermine, but in MH, the production of N 1 ,N 12 -diacetylspermine indicates the presence of Enterococcus in the sample. It is believed that the culture medium must contain spermine.

[0304] · When the sample grows in RPMI medium containing hypoxanthine, nicotinamide, and pyridoxine,

[0305] the production of xanthine and 6-hydroxy-nicotinate is detected, indicating the presence of Pseudomonas spp. in the sample. It is thought that xanthine is produced from hypoxanthine, and 6-hydroxy-nicotinate is produced as a metabolite of nicotinamide and pyridoxine.

[0306] · Additionally, in MH, aminobutyric acid is used to differentiate Staphylococcus aureus and coagulase-negative staphylococci, as Staphylococcus aureus produces aminobutyric acid while coagulase-negative staphylococci do not.

[0307] · When the sample culture of MH medium causes the production of marker 106 (131.0713@2.28,

[0308] using the 5-minute method), the sample contains Enterococcus spp. and / or Streptococcus pneumoniae.

[0309] · Antimicrobial susceptibility testing uses MH medium determined to contain glucose, nicotinamide, and pyridoxine, and it is found that:

[0310] o The consumption of glucose indicates resistant Escherichia, Klebsiella, Enterococcus, Staphylococcus, and Streptococcus;

[0311] o The production of succinate clearly indicates the presence of resistant Escherichia spp. or Klebsiella spp.; and

[0312] o The production of nicotinate clearly indicates that the culture contains resistant Streptococcus spp.

[0313] For further improvement Figure 16 of the decision tree.

[0314] Example XII: Sensitivity Test

[0315] To better evaluate the stability of the present invention in a clinical setting, we subsequently evaluated the workflow for a larger cohort of isolates (n = 273), which included: Escherichia coli (n = 50), Staphylococcus aureus (n = 64), Klebsiella pneumoniae (n = 35), Streptococcus pneumoniae (n = 48), GAS (n = 29), Enterococcus faecalis (n = 23), and Enterococcus faecium (n = 24). To maximize sample throughput, a rapid 5-minute HILIC chromatography was used to analyze the metabolic profiles. Then, the modified classification scheme was used to predict the microbial ID and AST. Regarding the microbial ID, all Enterococcus isolates were accurately identified to the genus level (n = 47), and almost all other isolates were precisely identified to the species level (n = 225 / 226), with the exception of one Escherichia coli isolate, which was classified as Klebsiella pneumoniae due to low production of urocanate and succinate. These data support the refined microbial ID we previously observed by MPA and suggest that these assignments can be made using our higher throughput analysis method.

[0316] We also used this larger cohort to further screen for metabolic markers of susceptibility. From these data, we determined glucose consumption as the most reliable indicator of antibiotic susceptibility for Staphylococcus aureus, GAS, and Enterococcus species; succinate production as the most reliable indicator of susceptibility for Escherichia coli and Klebsiella pneumoniae; and nicotinate production as the most reliable indicator of susceptibility for Streptococcus pneumoniae. Then, these susceptibilities were expressed as a metabolic inhibition index [(C-T) / C]×100, where T is the biomarker signal intensity in the antibiotic-treated sample and C is the signal intensity observed in the antibiotic-free control. Then, metabolite-specific inhibition indices for glucose consumption (> -50), succinate production (< 50), and nicotinate production (< 55) were used to determine classification breakpoints, which were empirically determined to distinguish susceptible isolates from resistant isolates. Using these thresholds, MIA correctly predicted susceptibility access in 93.8% of the instances (Table 7). These data demonstrate that the analysis herein provides a sound mechanism for identifying and characterizing pathogens.

[0317] Table 7: Susceptibility testing with cut-off values

[0318]

[0319] Example XIII: Customized Culture Medium

[0320] Prepare a customized medium formulation based on the analysis of precursor identification and medium-dependent biomarker production. Produce the medium according to the actual composition listed in Table 1A. Using the above method, in which 7 pathogens were cultured separately, the metabolomics analysis is shown in Figures 23A to 23C . Although metabolite production was still observed for most isolates after 4 h of incubation, clearly only 50% of the isolates grew. In particular, SA and SP did not grow properly. Interestingly, here, N 1 ,N 12 -diacetylspermine identified the presence of Enterococcus, Klebsiella, and Escherichia coli in the samples.

[0321] Produce the medium according to the actual composition (with and without spermine) listed in Table 1C. Using the above method, in which the pathogens were cultured separately and metabolomics analysis was performed. Figure 24A The heatmap of shows the top biomarkers of Mueller-Hinton and RPMI, Figure 24B The heatmap of shows the customized MPA / MIA medium with and without spermine in Table 1C. The whitened areas indicate that the designated biomarkers are not suitable for differentiation on the given medium.

[0322] The customized medium, even with a very limited composition as in Table 1A, can contain nutrients that support cell metabolism for a long enough time to enable at least selected pathogens to produce consistent biomarkers. Reliable and reproducible metabolomics analysis can be performed using the customized medium.

[0323] The foregoing description and examples are provided to enable those skilled in the art to better understand the present invention. The present invention is not limited by the description and examples, but is given a broad interpretation based on the following claims.

Claims

1. A method for identifying the cell type of cells in a sample, comprising: Cultivate a sample in a growth medium comprising nicotinamide to obtain a cultivated growth medium; Analyze the cultivated growth medium using chemical analysis; and When the cultivated growth medium contains a higher concentration of nicotinate than the growth medium, identify the cell type as at least one of the genus Escherichia, genus Klebsiella, genus Pseudomonas, genus Enterococcus, genus Staphylococcus, or genus Streptococcus; Wherein, the growth medium comprises 0.02 - 2 g / l D-glucose, 0.0001 - 0.1 g / l nicotinamide, 0.0001 - 0.1 g / l pyridoxine·HCl, 0.02 - 1 g / l hypoxanthine, 0.2 - 5 g / l L-arginine, 0.2 - 5 g / l L-threonine, 0.1 - 5 g / l L-histidine, 0.02 - 2 g / l spermine, and catalase in a buffered medium.

2. The method according to claim 1, wherein, When the cultivated growth medium contains a higher concentration of succinate than the growth medium, identify the cell type as the genus Escherichia or genus Klebsiella.

3. The method according to claim 1, wherein, When the cultivated growth medium contains higher concentrations of succinate and urocanate than the growth medium, identify the cell type as the genus Klebsiella.

4. The method according to claim 1, wherein, When the cultivated growth medium contains a higher concentration of xanthine than the growth medium, identify the cell type as the genus Pseudomonas.

5. The method according to claim 1, wherein, When the cultivated growth medium contains a higher concentration of mevalonic acid than the growth medium, identify the cell type as the genus Enterococcus and genus Staphylococcus.

6. The method according to claim 1, wherein, When the cultured growth medium contains a higher concentration of N than the growth medium 1 , N 12 -diacetylspermidine, the cell type is identified as Escherichia coli or Klebsiella or Enterococcus.

7. A growth medium, comprising: 0.5 to 1.5 mM glucose, histidine, nicotinamide, hypoxanthine, threonine, spermine, and arginine as metabolic precursors, pyridoxine, and catalase, for cultivating a sample to identify pathogens in the sample.

8. The growth medium according to claim 7, wherein, The growth medium comprises 0.02 - 2 g / l D-glucose, 0.0001 - 0.1 g / l nicotinamide, 0.0001 - 0.1 g / l pyridoxine·HCl, 0.02 - 1 g / l hypoxanthine, 0.2 - 5 g / l L-arginine, 0.2 - 5 g / l L-threonine, 0.1 - 5 g / l L-histidine, 0.02 - 2 g / l spermine, and catalase in a buffered medium.

9. Use of a growth medium comprising 0.02 - 2 g / l D - glucose, 0.0001 - 0.1 g / l nicotinamide, 0.0001 - 0.1 g / l pyridoxine hydrochloride, 0.02 - 1 g / l hypoxanthine, 0.2 - 5 g / l L - arginine, 0.2 - 5 g / l L - threonine, 0.1 - 5 g / l L - histidine, 0.02 - 2 g / l spermine and catalase in a buffered medium for identifying pathogens in a sample, wherein, After cultivating the sample, the chemical analysis of the growth medium identifies pathogens from at least the following species: the genus Escherichia, genus Klebsiella, genus Pseudomonas, genus Enterococcus, and genus Candida.

10. The use according to claim 9, wherein, The chemical analysis determines the presence of xanthine in the growth medium and identifies the pathogen as the genus Pseudomonas.

11. The use according to claim 9, wherein, The chemical analysis determines the presence of urocanate in the growth medium and identifies the pathogen as the genus Klebsiella.

12. The use according to claim 9, wherein, The chemical analysis determines the presence of arabitol in the growth medium and identifies the pathogen as the genus Candida.

13. The use according to claim 9, wherein, The chemical analysis determines the presence of mevalonic acid in the growth medium and identifies the pathogen as the genus Enterococcus.

14. The use according to claim 9, wherein, The chemical analysis determines the presence of succinate in the growth medium and identifies the pathogen as Escherichia coli or the genus Klebsiella.

15. The use according to claim 9, wherein, The chemical analysis determines the presence of putrescine in the growth medium and identifies the pathogen as Citrobacter, Escherichia coli, Enterobacter, Klebsiella or Proteus mirabilis.

16. A method for identifying the toxin sensitivity of pathogens in a sample, comprising: Cultivate the sample in a growth medium to obtain a cultured growth medium; Analyze the cultured growth medium by chemical analysis, and if the cultured growth medium contains mevalonic acid, identify the pathogen as Staphylococcus aureus; Cultivate the pathogen in a toxin-containing growth medium known to have an effect on Staphylococcus aureus; and Analyze the cultured toxin-containing growth medium for glucose consumption by chemical analysis to determine whether Staphylococcus aureus is resistant to the toxin.

17. A method for analyzing a biological sample to identify a pathogen therein, the method comprising: Cultivate the sample in a first medium to promote metabolism for pathogen identification; Meanwhile, cultivate the sample in a plurality of toxin-containing media, each medium having a toxin specific to a different pathogen; After cultivation, analyze the first medium to obtain a metabolic result for pathogen identification; and Analyze only the selected toxin-containing medium from the plurality of toxin-containing media, and select the selected toxin-containing medium as having a relevant toxin specific to the identified pathogen.

18. The method according to claim 17, wherein, The first cultivation and the plurality of toxin-containing cultivations are all carried out on a common instrument, and the analysis step includes connecting the common instrument to an automatic sampler.

19. The method according to claim 17 or 18, wherein, At least a part of the process is implemented by a computer.

Images