Determination of bacterial susceptibility or resistance to antibiotics by mass spectrometry.

Abstract

1. A method for determining the susceptibility or resistance of at least one identified bacterium to at least one antibiotic, the method comprising the steps of: a) contacting a sample containing the bacterium with the at least one antibiotic, which antibiotic induces rupture of the bacterial wall and / or cell membrane and causes release of intracellular compounds of the bacterium, where the bacterium is said to be "susceptible" to the at least one antibiotic; b) incubating the sample with the at least one antibiotic; c) purifying the sample by removing intact bacteria and cellular debris; d) analyzing the purified sample by mass spectrometry; e) detecting the presence or absence of at least one peak of at least one characteristic protein of the bacterium; and f) determining the susceptibility or resistance of the bacterial population to the antibiotic.

Description

[Technical Field]

[0001] This invention relates to the field of microbiology, and more specifically to the mass spectrometry determination of the susceptibility or resistance of bacteria to antibiotics that induce rupture of their walls and / or cell membranes. [Background technology]

[0002] In the 19th century, the discovery of microorganisms and their role in infectious diseases made it possible to envision the treatment of infectious diseases. The fight against bacteria began with the development of arsphenamine in 1907, which was successfully used to treat syphilis. However, it was the discovery of penicillin, followed by many other antibiotics, that led to the widespread use of these drugs. Antibiotics have made it possible to save patients' lives and improve their health, and are undoubtedly one of the greatest successes of modern medicine.

[0003] The remarkable efficacy of antibiotics has led to their repeated and extensive use in the health of humans and animals. Unfortunately, however, antibiotics have sacrificed their own success: repeated and sometimes excessive administration of antibiotics has resulted in the emergence of antibiotic-resistant strains. Thus, within a bacterial population, there may naturally be a proportion that is partially or completely resistant to the action of antibiotics through various mechanisms. Therefore, when bacteria are exposed to antibiotics, those susceptible are destroyed, while those that barely survive may remain and proliferate. The application of antibiotics constitutes selective pressure that favors resistant bacteria. Bacteria may spontaneously acquire the ability to resist a given antibiotic through genetic mutation, but they can also acquire it through the transfer of DNA fragments from one bacterium to another, which accelerates the transmission of resistance.

[0004] Initially isolated, these resistances have grown, becoming a cause for concern. Some strains are multi-resistant, meaning they are resistant to several antibiotics. Others have even become completely resistant, meaning they are resistant to almost all available antibiotics. This phenomenon, while still rare in France, is increasing rapidly, leaving doctors at a treatment deadlock as there are no longer any effective solutions to combat the infection.

[0005] According to the World Health Organization, "antibiotic resistance is now one of the most serious threats to global health." Due to the misuse of these drugs, antibiotic resistance is estimated to cause 12,500 deaths annually in France and may become the leading cause of death globally by 2050.

[0006] Furthermore, it is important to note that antibiotic treatment is not always a minor event for patients. In fact, antibiotics have several undesirable effects, which vary depending on the molecule. Most are minor, but some are serious, and in some cases even severe. Consequently, certain antibiotics may increase patient morbidity or mortality due to undesirable side effects, particularly neurotoxicity and nephrotoxicity (Poirel L et al., 2017). This is especially true for certain antibiotics that act on the bacterial wall and / or cell membrane of certain bacteria. For example, colistin, a polymyxin antibiotic that acts on bacterial cell membranes, adheres itself to the surface of the membrane of nephron tubular cells, which is reflected in proteinuric interstitial nephritis. Colistin also has neuromuscular blocking effects by blocking presynaptic release of acetylcholine and reducing the sensitivity of postsynaptic receptors (Spapen, 2011).

[0007] These effects, particularly neurotoxic effects, must be monitored especially during high-dose treatment. Therefore, these antibiotics are reserved for extreme cases, such as when bacteria are resistant to all other antibiotics, as well as for topical use, such as the local treatment of infections of the skin, mucous membranes, eyes, or ears. In cases of otitis, it should be noted that administering polymyxin to a perforated eardrum may result in irreversible cochlear and vestibular toxicity.

[0008] For all the reasons mentioned above, antibiotics should only be used if the pathogenic bacteria or pathogens are susceptible to them, and for some of these, they should be reserved only as a last resort.

[0009] In this regard, rapid determination of the susceptibility or resistance of an identified bacterium to an antibiotic is essential, preferably before initiating treatment, or at least as soon as possible.

[0010] Currently, both the Clinical Laboratory Standards Institute (CLSI) and the European Commission on Antimicrobial Susceptibility Testing (EUCAST) recommend microdilution in broth or phenotypic methods of microdilution broth (MDB) to determine resistance, particularly to polymyxins, antibiotics that act at the cell membrane level. While this method can be considered a reference method, it is time-consuming, very difficult to implement, and unsuitable for the constraints of clinical microbiology laboratories (Poirel L et al., 2017). Furthermore, possibly due to the phenomenon of heteroresistance to colistin, unreproducible and uninterpretable results have been reported regarding the determination of susceptibility to specific antibiotics that act on the cell membrane, especially colistin (Landman D et al., 2013). Finally, strains carrying plasmids that confer plasmid resistance are not always detectable.

[0011] Other classical methods, including dilution in agar and diffusion from disks or gradients, are not explicitly recommended by CLSI and ECAST for certain antibiotics such as polymyxins. This can be explained by their polycationic properties, which promote their adsorption to surfaces, and their ability to form micelles that are difficult to diffuse properly in agar (Hindler JA et al., 2013). Furthermore, the phenomenon of heterotolerance appears to complicate the use of these methods (Landman D et al., 2013).

[0012] In particular, application WO2015 / 48696 proposes alternative methods to phenotypic methods. This application describes a method for determining susceptibility by contacting an antibiotic at a concentration close to the minimum inhibitory concentration (MIC), followed by mechanical or chemical lysis, and then determining the solubility. Susceptibility is established by comparing this solubility to a threshold. Solubility is determined, in particular, using labeled intracellular compounds present outside the lysed bacterial cells. Thus, it has been proposed to use staining with Coomassie blue, measurement of ATP luminescence, or the use of peptide nucleic acid (PNA) probes with fluorescent labels that hybridize to nucleic acids.

[0013] However, in addition to the use of antibiotics, these methods require conditions to dissolve the bacterial wall, particularly by detergents. Currently, the use of detergents is considered incompatible with MALDI-TOF mass spectrometry. Indeed, detergents generate ions that mask signals from proteins. Furthermore, bacteria have walls with biochemical structures that can vary from species to species. Therefore, different species may have varying degrees of resistance to dissolution conditions, and the selection of these dissolution conditions may require caution when false negatives or false positives must be avoided.

[0014] A method is known for detecting resistance mechanisms by MALDI-TOF without culturing after contact with an antibiotic. In particular, applications WO2018 / 099500 and WO2011 / 154517 filed by Bruker Daltonik detect the hydrolysis of β-lactam antibiotics by the action of β-lactamase. In these applications, the change in antibiotic mass is detected after the reaction of the bacterial sample with β-lactamase. These methods require measuring the levels of active and hydrolyzed antibiotics after a predetermined time of action by the β-lactamase. This time can be 1 to 2 hours for β-lactamases such as KPC, but generally needs to be longer for β-lactamases with low intracellular activity or low abundance (Mirande et al., 2015). Bacterial proteins are not directly observed by these methods. Bacterial lysis under antibiotic action has not been studied.

[0015] Furthermore, methods for detecting resistance by MALDI-TOF after culturing microorganisms in the presence of antibiotics are known, and are described, for example, in applications WO2014 / 187517, US2008 / 009029, and EP2801825.

[0016] In application WO2014 / 187517, the method measures the increase in bacterial biomass when microorganisms are resistant to antibiotics. However, in application US2008 / 009029, the method describes a change in the protein bacterial pattern during culture when microorganisms are resistant to antibiotics. Therefore, these methods require a culture time long enough to ensure that the bacteria respond to the action of antibiotics, either by growing them after an incubation period or after they have stopped dividing (WO2014 / 187517) or by altering or not altering their protein pattern (US2008 / 009029). Furthermore, in certain species and types of resistance, there is no change in the protein pattern. [Overview of the Initiative]

[0017] In this regard, an object of the present invention is to propose a method for determining the susceptibility or resistance of at least one identified bacterium to at least one antibiotic, overcoming the shortcomings of prior art methods; that is, to provide an inexpensive method that does not use detergents, does not require various specific reagents, produces results in a short time of less than one hour, and can be used in routine clinical practice without requiring highly qualified personnel.

[0018] For this purpose, the present invention provides a novel method for determining the susceptibility or resistance of at least one identified bacterium to at least one antibiotic, a) A step of bringing a sample containing the bacteria into contact with at least one antibiotic, which, if the bacteria are said to be "sensitive" to the at least one antibiotic, induces rupture of the bacterial wall and / or cell membrane and causes the release of intracellular compounds of the bacteria. b) The step of incubating the sample with at least one antibiotic, c) A step of purifying the sample by removing intact bacteria and cellular debris, d) A step of analyzing the purified sample by mass spectrometry. e) A step of detecting the presence or absence of at least one peak of at least one characteristic protein of the bacterium, and f) A step of determining the susceptibility or resistance of the bacterial population to the antibiotic. We propose a method that includes this.

[0019] In the context of the method of the present invention, samples can be obtained from different sources. By way of example, mention may be made of samples of biological origin, particularly from animals or humans. Such samples may correspond to samples of biological fluids such as whole blood, serum, plasma, urine, cerebrospinal fluid, organic secretions, tissue samples, or isolated cells. The sample can be used as it is, or preferably, before being brought into contact with the antibiotic, it undergoes a step of accumulation or culture-type preparation, concentration, and / or extraction or purification by methods known to a person skilled in the art. However, said preparation cannot correspond to a lysis step that leads to the disruption of microorganisms and the loss of their contents before being brought into contact with the antibiotic. The sample can be used in the form of an inoculum.

[0020] In most cases, the sample may have been pre-cultured in broth or agar to enrich for bacteria. Said media, agar or broth, are well known to those skilled in the art.

[0021] The sample will preferably contain a single bacterial species. However, the use of a sample containing several bacteria is not excluded. In this case, it is known that bacteria are highly likely to express different resistance mechanisms, and it is preferable to know which one expresses the resistance that is identified.

[0022] Bacteria that can be characterized by the method of the present invention are all pathogenic or non-pathogenic bacteria encountered in both industry and clinical practice.

[0023] Within the meaning of the present invention, the term bacteria encompasses Gram-positive bacteria or Gram-negative bacteria.

[0024] Advantageously, the bacteria are Gram-negative and preferably selected from the following species and subspecies: Escherichia coli, Acinetobacter baumannii, Klebsiella pneumoniae, Acinetobacter haemolyticus, Acinetobacter junii, Citrobacter freundii, Enterobacter asburiae, Enterobacter cloacae, and fluorescent bacteria (Pseudomonas fluorescein). Fluorescens, Salmonella enteritica serotype Enteritidis, Salmonella enteritica serotype Paratyphi B variant Java, Salmonella enteritica serotype Agona, Salmonella enteritica serotype Enteritidis, Salmonella enteritica serotype Haifa, Salmonella enteritica serotype Newport, and Pseudomonas aeruginosa.

[0025] "Susceptibility of at least one identified bacterium to at least one antibiotic" means the ability of the antibiotic to kill the bacterium or to sufficiently inhibit its growth.

[0026] "Determining susceptibility" means determining the susceptibility of bacteria to be killed or inhibited by antibiotics.

[0027] "Resistance to at least one antibiotic of at least one identified bacterium" means the phenomenon in which a bacterium retains some or all of its viability, growth, or reproduction when exposed to concentrations of an antibiotic that would otherwise be effective against the bacterium. This resistance may be acquired by one or more strains of a bacterial species that is naturally susceptible to the antibiotic. This resistance may also be innate or natural.

[0028] In the context of the present invention, antibiotic resistance may be chromosomal and / or extrachromosomal resistance, also known as plasmid resistance.

[0029] The bacterial genetics related to the mechanisms of antibiotic resistance are well described in the literature, particularly in the reference ANTIBIOGRAMME (P. Courvalin and R. Leclercq, 2012, 3rd edition, Chapter 3 GENETIQUE DE LA RESISTANCE (Genetics of Resistance)).

[0030] "Determining resistance to at least one type of antibiotic" means determining the ability of bacteria to grow when exposed to an antibiotic to which they are naturally susceptible.

[0031] The method according to the present invention is also applicable to bacterial strains that exhibit heterozygosity, that is, bacterial strains in which only a portion of a population of bacterial clones, whether in small amounts or not, exhibit this resistance.

[0032] The term "antibiotic" refers to any natural or synthetic chemical substance that has a specific effect against bacteria.

[0033] In the context of this invention, antibiotics induce the rupture of the bacterial wall and / or cell membrane, and therefore, if the bacteria are susceptible to this antibiotic, they release intracellular compounds. Antibiotics that induce bacterial wall rupture and antibiotics that induce cell membrane rupture differ in their methods of action on bacteria. In the context of this invention, it is essential that the rupture of the bacterial wall and / or cell membrane is induced by the antibiotic and not by the action of a solvent or acid. In fact, in analysis by mass spectrometry, it is common to use a solvent such as acetonitrile or alcohol (ethanol, methanol, etc.) and / or an acid such as formic acid to induce bacterial lysis. The purpose of this lysis step is to release intracellular proteins and observe them by spectroscopic analysis to identify microorganisms, for example. In the context of this invention, since this lysis step is not performed, it is important that the release of intracellular compounds by bacteria is solely due to the action of the antibiotic if the bacteria are susceptible to this antibiotic, and not due to the action of a solvent or acid, so as not to distort the results.

[0034] Antibiotics that act on bacterial walls include those that induce inhibition of wall precursor synthesis, those that induce inhibition of the movement of wall precursors to lipid carriers while still allowing their transport across the cell membrane, and those that induce inhibition of the insertion of wall precursor glycan units and peptide transfer.

[0035] Blocking cell wall synthesis significantly weakens the bacterial outer envelope, making it highly susceptible to external stresses (osmotic pressure, temperature, mechanical stress) that induce cell lysis, thus triggering the release of intracellular compounds.

[0036] Examples of antibiotics that induce inhibition of wall precursor synthesis include D-cycloserine and fosfomycin.

[0037] Bacitracin is an example of an antibiotic that inhibits the migration of wall precursors to lipid carriers.

[0038] Examples of antibiotics that act on the insertion and peptide transfer of wall precursor glycan units include transpeptidases involved in wall synthesis or β-lactam antibiotics that inhibit glycopeptides that bind to intermediates in peptidoglycan synthesis.

[0039] The presence of an intact cell membrane is essential for bacterial survival. It has two roles: on the one hand, it allows for the sequestration of essential metabolites and ions within the cytoplasm; and on the other hand, it allows for the maintenance of a proton gradient between the inside and outside of the cell, generated by the respiratory chain and the Krebs cycle, which enables the storage of cellular energy. This proton gradient supplies ATP synthase, which produces ATP. Disruption of the membrane's impermeability breaks these trappings, allowing chemiosmotic energy to dissipate and cytoplasmic contents to escape into the extracellular fluid. There are a certain number of antibiotic molecules that act on the cell membrane of cells, either by acting as desiccant agents that break down lipids or by forming pores in the cell membrane that cause the release of cellular compounds.

[0040] Examples of antibiotics that act at the cell membrane level include polymyxins and gramicidins. Polymyxins act as cationic detergents: due to their amphiphilic nature, they penetrate bacterial cells, are inserted into the phospholipids of the cell wall, and disrupt membrane permeability. Gramicidins are peptides that enter the membrane and form cylindrical pores that release cations.

[0041] Advantageously, the antibiotic is selected from polymyxins, β-lactam antibiotics, aminoglycosides, quinolones, and glycopeptides; the antibiotic is preferably a polymyxin, selected from colistin and polymyxin B.

[0042] Advantageously, according to the present invention, the antibiotic has a concentration between 2 times and 10 times the minimum inhibitory concentration of the bacterial population being studied.

[0043] In certain embodiments, the antibiotic is present at a concentration at least 10 times higher than the minimum inhibitory concentration (MDI) of the antibiotic for the bacterial population under study. Preferably, the antibiotic is present at a concentration at least 100 times higher than the MDI for the bacterial population under study. More preferably, the antibiotic is present at a concentration at least 1000 times higher than the MDI for the bacterial population under study. Contrary to all expectations, the inventors have demonstrated that the methods according to the present invention work at these concentrations. Indeed, on the one hand, in the case of analysis by mass spectrometry, the peak corresponding to the antibiotic may mask the peak corresponding to the characteristic bacterial protein, so it is not obvious to use this concentration range for the antibiotic. The ionization step is, in fact, exposed to intermolecular competition, which is generally favorable to the most abundant molecules in the sample, i.e., the antibiotic in this case. Therefore, high concentrations of the antibiotic risk interfering with the detection of the protein peak. On the other hand, such concentrations may be toxic to the patient if they are prescribed.

[0044] In the context of this invention, the minimum inhibitory concentration, or MIC, is the lowest concentration of an antibiotic that can in vitro inhibit a visible culture of a strain being studied over a given temperature and a defined period of time. This value characterizes the bacteriostatic effect of the antibiotic on the bacteria. The MIC is specific to the antibiotic / bacterial pair, and each strain has its own value as a function of innate and / or acquired resistance to the molecule being tested.

[0045] Advantageously, in addition to antibiotics, compounds known to promote enzymatic reactions involved in the resistance mechanisms under consideration can also be brought into contact with the sample. These compounds may, for example, be zinc compounds, particularly in the form of ZnCl2 or zinc sulfate, which are important cofactors for metallo-β-lactamase activity. These compounds can be added in combination with antibiotics or at any other point in the preparation of the sample.

[0046] In the context of the present invention, following the step of bringing a sample into contact with an antibiotic, the method includes an incubation step that allows for interaction between a bacterial population(s) and the antibiotic(s).

[0047] Incubation conditions and time are adapted by those skilled in the art as a function of the population being analyzed. Samples can be left at temperatures ranging from 15 to 100°C, particularly at room temperature (22°C). They can also be transferred to a thermostat-controlled enclosure at, for example, 37°C.

[0048] Advantageously, according to the present invention, the incubation temperature is 50°C or about 50°C. Surprisingly, the inventors found that high incubation temperatures (on the order of 50°C), which can be lethal to bacteria, were appropriate in the context of the present invention.

[0049] Under selected conditions, the incubation time must be sufficient to allow bacterial lysis when the bacteria are susceptible, and thus to allow subsequent detection by mass spectrometry of their intracellular compounds, particularly released intracellular proteins. Incubation is generally carried out for a time between 45 and 90 minutes, preferably less than 30 minutes, more preferably less than 15 minutes, and even more preferably less than 10 minutes. This is an advantage over prior art methods, particularly those described in applications US2008 / 009029, WO2014 / 187517, where the incubation time required exceeds 2 hours. This longer time is necessary to ensure that the bacteria have responded to the action of the antibiotic by bacterial growth or changes in protein patterns. Such effects are not required in the methods according to the present invention. Therefore, it is not necessary to wait long enough to allow bacterial cell division or even changes in the expression levels of their proteins.

[0050] In the context of the present invention, it has been found that performing this incubation step does not interfere at all with the subsequent identification of bacteria when it is necessary to perform the aforementioned characterization in addition to determining susceptibility or resistance.

[0051] According to certain embodiments of the present invention, the method of the present invention may include a homogenization step after the incubation step. In the context of the present invention, if the method includes a homogenization step, this step must not induce rupture of bacterial walls and / or cell membranes, particularly by cavitation. This step can be carried out using a thermomixer-type thermal mixer, particularly by techniques such as sonication, mechanical or magnetic vortices, provided that these techniques do not induce rupture of bacterial walls and / or cell membranes.

[0052] This step allows for a dense mixing of the antibiotic and microorganisms being analyzed, and reduces the incubation time required for the antibiotic to diffuse near the walls and / or membrane.

[0053] According to the present invention, the ultrasonic treatment technology consists of incubation of microtubes in an ultrasonic bath.

[0054] According to a preferred embodiment of the present invention, the ultrasonic treatment is performed for a period of time between 5 and 90 minutes.

[0055] According to a preferred embodiment of the present invention, mechanical vortexing is performed for a time between 5 and 90 minutes.

[0056] According to a preferred embodiment of the present invention, magnetic vortexing is performed for a time between 5 and 90 minutes.

[0057] In the context of the present invention, the method includes the step of purifying the sample by removing intact bacteria and cellular debris. This step consists of retaining in the sample only the intracellular compounds released by the bacteria analyzed when the bacteria came into contact with a susceptible antibiotic. Thus, insensitive bacteria are intact, and cellular debris is removed from the sample to be analyzed after lysis.

[0058] Advantageously, this step of purifying the sample can be carried out by techniques such as centrifugation, filtration, chromatography, or electrophoresis.

[0059] These separation techniques can be used individually or in combination to perform multidimensional separation. For example, multidimensional chromatography can be used for the analysis of proteotype protein peptides by combining ion-exchange chromatography with reverse-phase chromatography, as described by T. Fortin et al. (2009) or H. Keshishian et al. (2007). In these studies, the chromatographic medium can be a column or a cartridge (solid-phase extraction). The electrophoretic or chromatographic fractions (or retention times in one-dimensional or multidimensional chromatography) of proteotype peptides are characteristic of each peptide, and therefore, the use of these techniques makes it possible to select the proteotype peptides to be assayed. This fraction of the resulting peptides allows for increased specificity in subsequent measurements by mass spectrometry. It is also possible to work with total protein using so-called "top-down" techniques, particularly as synthesized by Donnelly et al. (Nature Methods, 2019). In these studies, the electrophoretic or chromatographic fractions (or retention times in one-dimensional or multidimensional chromatography) of proteins are characteristic of each protein, and therefore, the use of these techniques allows for the selection of one or more proteins to be assayed.

[0060] According to a preferred embodiment of the present invention, centrifugation is performed for a time between 1 and 60 minutes at a rotational speed between 300 g and 3000 g.

[0061] In the context of the present invention, filtration consists of filtration using a filter having a porosity between 0.02 and 0.22 μm.

[0062] In the context of the present invention, electrophoresis consists of separation techniques under the action of an electric field. Many of these techniques and variations are well known to those skilled in the art. As non-limiting examples, electrophoresis may be selected from isokinetic electrophoresis, polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS-PAGE), isoelectric focusing (IEF), or two-dimensional electrophoresis, which is a combination of IEF and SDS-PAGE.

[0063] In the context of the present invention, chromatography consists of separating molecules present in a mobile phase using a stationary phase. Many techniques and variations thereof are well known to those skilled in the art. In non-limiting examples, chromatography may be selected from thin-layer chromatography, chromatography in a solid-phase extraction (SPE) cartridge, or chromatography in a chromatography column. The cartridge or column may contain a chromatography compound to ensure method selectivity depending on the physicochemical properties of the molecule of interest. In non-limiting examples, this compound may enable reversed-phase chromatography (C18, C8, or C4), serum exclusion chromatography, ion exchange chromatography, or affinity chromatography.

[0064] The mass spectrometry performed in the method of the present invention is generally known to those skilled in the art as a powerful tool for the analysis and detection of different types of molecules. In general, all types of ionizable molecules can be detected as a function of their molecular weight using a mass spectrometer. Depending on the nature of the molecule being detected, such as its protein or metabolic origin, a particular mass spectrometry technique may be more suitable. However, whatever mass spectrometry method is used for detection, it includes an ionization step of the target molecule to so-called molecular ions, in this case an ionization step of at least one type of bacterial protein, and a separation step of the molecular ions obtained as a function of their mass.

[0065] Therefore, all mass spectrometers, - An ionization source intended to ionize molecules present in the sample being analyzed, i.e., to impart a positive or negative charge to these molecules; - A mass spectrometer designed to separate ionized molecules or molecular ions as a function of the mass-to-charge ratio (m / z); - Detectors designed to measure signals generated directly by molecular ions or by ions generated from molecular ions. It is equipped with the following features, which are described in detail below.

[0066] The ionization step required to perform mass spectrometry can be carried out by any method known to those skilled in the art. An ionization source can ionize the molecule being assayed into a gaseous state. The ionization source can be used in either a positive mode for studying cations or a negative mode for studying anions. Several types of sources exist and are used depending on the desired results and the molecule being analyzed. In particular, - Electron ionization (EI), chemical ionization (CI), and desorption chemical ionization (DCI); - Impact by fast atoms (FAB), metastable atoms (MAB), or ions (SIMS, LSIMS) - Inductively coupled plasma (ICP); - Atmospheric pressure chemical ionization (APCI) and atmospheric pressure photoionization (APPI); - Electrospray (ESI); - Matrix-assisted laser desorption / ionization (MALDI) on surface enhancement (SELDI) or silicon (DIOS); - Atmospheric pressure ionization, e.g., desorption electrospray ionization (DESI), nano-desorption electrospray ionization (nDESI), laser ablation electrospray ionization (LAESI), rapid evaporation ionization mass ionization (REIMS), or paper spray ionization; - Ionization / desorption via interaction with metastable species (DART) One could list these:

[0067] The mass spectrometry steps required to perform mass spectrometry can be carried out by any method known to those skilled in the art. A mass spectrometer makes it possible to separate the molecules to be assayed in an ionized gaseous state in relation to their mass-to-charge ratio (m / z). Several types of mass spectrometers exist and are used depending on the desired results and the molecules to be analyzed. Examples include low-resolution analyzers of the quadrupole (Q) type, linear ion trap (LIT) type, also called 3D ion trap (IT) or ion trap, and high-resolution analyzers that enable accurate measurement of the analyte's mass, particularly using magnetic or electrical sectors, time of flight (TOF), cyclotron resonance, or orbitraps.

[0068] In general, any mass spectrometry method suitable for detecting at least one type of bacterial molecule can be used in the context of the present invention.

[0069] According to certain embodiments, the mass spectrometry used is matrix-assisted laser desorption / ionization time-of-flight (MALDI-TOF) mass spectrometry, which offers the advantage of being relatively easy to implement.

[0070] The MALDI ionization source allows for the ionization of molecules starting from a solid sample. It is preferable to bring the sample into contact with the matrix before ionization.

[0071] The matrix to be favorably used includes compounds selected from 3,5-dimethoxy-4-hydroxycinnamic acid (i.e., synapic acid or synapic acid); α-cyano-4-hydroxycinnamic acid (i.e., alpha-cyano, alpha-matrix, or CHCA); ferulic acid; and 2,5-dihydroxybenzoic acid (i.e., DHB).

[0072] Several deposition techniques can be used in the context of this invention to bring a sample into contact with the matrix: deposition of a dry matrix layer, known as "thin layer" deposition; deposition of matrix droplets, known as "dry drop" deposition; deposition of matrix droplets; and addition of matrix drops, known as "sandwich" deposition.

[0073] Generally, the matrix is ​​photosensitive and crystallizes in the presence of the sample while preserving molecular integrity. Well-known matrices of this type, particularly suitable for the MALDI-TOF method, are selected from 3,5-dimethoxy-4-hydroxycinnamic acid; α-cyano-4-hydroxycinnamic acid, ferulic acid, and 2,5-dihydroxybenzoic acid. Many other compounds are known to those skilled in the art. Even liquid matrices exist that do not crystallize at atmospheric pressure or even under vacuum (Tholey and Heinzie 2006). Any other compound that enables the ionization of sample molecules under the action of a laser beam can be used. In particular, the target, i.e., the carrier on which the sample is deposited, can play the role of the matrix, as in the case of "nano-assisted laser desorption / ionization" (NALDI) or "desorption / ionization on silicon" (DIOS) methods. The laser beam can have any type of wavelength favorable for the sublimation or vaporization of the matrix. Preferably, ultraviolet or even infrared wavelengths are used.

[0074] In the matrix, the compound is dissolved in water, preferably "ultrapure" water, or a mixture of water and an organic solvent. Examples of commonly used organic solvents include acetone, acetonitrile, methanol, or ethanol. Trifluoroacetic acid (TFA) is often added. An example of a matrix is, for example, 20 mg / mL of sinapic acid in a 50 / 50 / 0.1 (v / v) acetonitrile / water / TFA mixture. The organic solvent allows hydrophobic molecules present in the sample to dissolve in the solution, while water dissolves hydrophilic molecules. The presence of an acid such as TFA promotes the ionization of sample molecules by proton (H+) capture.

[0075] Next, the solvent present in the matrix is ​​evaporated, for example, by leaving the sample at a temperature in the range of 17 to 30°C, particularly at room temperature (22°C), for several minutes, for example, 1 minute to 2 hours. This evaporation of the solvent allows for the crystallization of the matrix in which the sample is dispersed. Then, the sample placed in the crystallized matrix is ​​subjected to soft ionization. This ionization is preferably carried out using a nitrogen laser emitting a 337.1 nm UV beam.

[0076] During ionization, the sample is subjected to laser excitation. Next, the matrix crystal absorbs photon energy, and the recovery of this energy causes the matrix to sublimate, the sample to desorb, and matter appears in a state called plasma. Within this plasma, charge exchange occurs between the matrix molecules and the sample molecules. For example, protons can be detached from the matrix and transferred to the proteins and peptides in the sample. This step enables the soft ionization of biomolecules without causing their destruction. Thus, the sample releases ions of various sizes. The latter are then accelerated by an electric field and fly freely within a reduced-pressure tube called a flight tube. The pressure applied during ionization and the acceleration of the generated ions is, in most cases, 10⁻¹⁰ -6 from 10 -9 The range is millibars (mbar). Because the smallest ions "fly" faster than larger ions, it becomes possible to separate them. The detector is located at the very end of the flight tube. The ions' time of flight is used to calculate their masses. In this way, a mass spectrum is obtained, representing the signal intensity corresponding to the number of ionized molecules with the same mass-to-charge ratio (m / z) as a function of the m / z ratio of molecules colliding with the detector. The m / z ratio is expressed in Thomson (Th) units. Once introduced into the mass spectrometer, the spectrum of a sample is acquired very quickly, generally in less than a minute.

[0077] Separation of molecular ions as a function of the m / z ratio can be carried out once (single mass spectrometry or MS), or several consecutive MS separations can be carried out. When two consecutive MS separations are carried out, the analysis is MS / MS or MS 2 , which is called. When three consecutive MS separations are carried out, the analysis is MS / MS / MS or MS 3 , which is called, and more generally, when n consecutive MS separations are carried out, the analysis is MS n , which is called.

[0078] According to another specific embodiment, the mass spectrometry used in the method of the present invention is tandem mass spectrometry (MS 2 , MS 3 , MS 4 or MS 5 ), wherein several mass analyzers are coupled together consecutively. For example, a first analyzer separates ions, a collision cell enables fragmentation of the ions, and a second analyzer separates fragment ions. Certain analyzers such as ion traps or FT-ICR incorporate several analyzers in one, enabling fragmentation of ions and direct analysis of fragments. This technique allows consecutive separation in two mass analyzers, which has the advantage of obtaining very high specificity, particularly by selecting ions in the first analyzer, fragmenting them, and analyzing their daughter ions in the second analyzer.

[0079] Advantageously, the separation can be carried out with MALDI-TOF-TOF, which has two time-of-flight analyzers and the same ease of use as MALDI-TOF.

[0080] Alternatively, the ionization source can be any type of source known to those skilled in the art, and the mass analyzer can be any type of mass analyzer known to those skilled in the art.

[0081] According to another specific embodiment, the mass spectrometry is tandem mass spectrometry, which uses a combination of an electrospray source and at least two of the aforementioned analyzers.

[0082] In particular, ionization can be carried out as follows: A sample containing the target molecule is introduced into an electrospray ionization source capable of ionizing the molecule while transitioning it from a liquid state to a gaseous state. Thus, the molecule is converted into a molecular ion corresponding to the initial molecule. The resulting molecular ion corresponds to the molecule initially present in the liquid sample and has one, two, or three or more additional protons in positive mode, and therefore has one, two, or three or more charges. For example, if the target molecule is a protein in the liquid phase, ionization by an electrospray source operating in positive mode produces a gaseous ion with one, two, or three or more additional protons, and therefore has one, two, or three or more charges. This type of source is particularly suitable when the target molecule, such as a protein, has been previously separated by reversed-phase liquid chromatography.

[0083] Among several techniques using sequential separation, Mode SRM (Selective Reaction Monitoring) for the detection or determination of a single target molecule, Mode MRM (Multiple Reaction Monitoring) or Mode PRM (Concurrent Reaction Monitoring) for the detection or determination of several target molecules, are used in MS. 2 It is a specific application of separation. Similarly, MRM 3 The modes are specific applications of MS / MS / MS isolation (WO2010136706). SRM, MRM, PRM, and MRM 3 The technique is targeted mass spectrometry, which means that the ions of the detected molecules are specifically targeted for analysis.

[0084] Furthermore, DDA (Data-Dependent Acquisition) or DIA (Data-Independent Acquisition) techniques utilize several sequential separations. However, they do not specifically target at least one ion. As an example, the DDA approach consists of i) acquiring an MS spectrum, ii) sequentially selecting each precursor ion observed in the MS spectrum with a strong signal, iii) sequentially fragmenting each precursor ion to acquire its MS / MS spectrum, and iv) querying databases such as SWISS-PROT or NCBI using software such as Mascot (Matrix Science, London, United Kingdom) or SEQUEST (Thermo Scientific, Waltham, United States of America) to identify molecules likely to correspond to the observed MS / MS spectrum. This method can lead to the identification of molecules characteristic of microorganisms without prior targeting of its analysis. The DIA method does not include step ii). The set of ions is fragmented and analyzed in step iii) regardless of their relative intensity. In other words, instead of fragmenting the ions one by one as a function of acquisition as performed in i), a mass window containing several precursor ions is selected in the first mass spectrometer. These aforementioned precursor ions are then transferred to the next analyzer(s) for simultaneous fragmentation and analysis. This technique is particularly advantageous when using a high-resolution mass spectrometer that can very accurately identify the set of fragment ions corresponding to each of the different precursor ions selected simultaneously.

[0085] In single-MS detection mode, the mass / charge ratio of the obtained molecular ion correlates with the detected target molecule.

[0086] For detection in MS / MS mode, two steps are essentially added to the MS assay, which is... Fragmentation of molecular ions called precursor ions to generate daughter ions called fragment ions, and This is the separation of daughter ions called fragment ions as a function of the mass / charge ratio (m / z)² and ratio (m / z)¹, which correspond to the precursor ion ratio (m / z).

[0087] In this case, the mass / charge ratio of the fragment ion obtained in this way correlates with the detected target molecule. A fragment ion is an ion derived from a precursor ion after the fragmentation step, and its mass / charge ratio (m / z) is different from that of the precursor ion.

[0088] The (m / z)1 and (m / z)2 pairs represent doubt transitions and are characteristic ions that are detected.

[0089] The selection of characteristic ions to be detected for correlation with the target molecule is routinely performed by those skilled in the art and, advantageously, leads to the most sensitive, most specific, and most robust determination possible in terms of reproducibility and reliability.

[0090] The principle of SRM or MRM mode is to specifically select a precursor ion, fragment it, and then specifically select one of the resulting fragment ions. In such applications, triple quadrupole or hybrid triple quadrupole instruments with ion traps are commonly used (WO2011 / 045544). The principle of PRM mode differs from SRM and MRM modes in the use of the final high-resolution analyzer. In the latter, a set of fragment ions can be detected in parallel with sufficient resolution to ensure the specificity of the method (Peterson AC et al., 2012). In PRM analysis, quadrupole time-of-flight (Q-TOF) or ion trap and orbitrap type instruments, or hybrid quadrupole and orbitrap instruments are commonly used.

[0091] In the case of the quadrupole and orbitrap device (Q1 q2 Orbitrap) used in PRM mode, the first quadrupole (Q1) allows filtering of molecular ions characteristic of the protein, determined in relation to its mass-to-charge ratio (m / z), for the purpose of determining or detecting the target protein. (m / z) i1 Only ions with the required mass / charge ratio of the protein, called the ratio, are transported to the second quadrupole (q2), where they act as precursor ions for subsequent fragmentation. The analyzer q2 determines the mass / charge ratio (m / z). i1 This makes it possible to fragment the ion into fragment ions. Fragmentation is generally achieved in q2 by collisions between the precursor ion and an inert gas such as nitrogen or argon. The fragment ions are transported to an orbit trap, which determines their m / z ratio. Next, the mass / charge ratio (m / z) of the fragment characteristic of the i-th protein is determined. i2 Fragment ions containing [the specified element] are detected or even quantified.

[0092] This operating mode exhibits dual selectivity, on the one hand, with respect to the selection of precursor ions, and on the other hand, with respect to the selection of at least one fragment ion.

[0093] Mass spectrometry in SRM, MRM, or PRM modes is advantageous for quantification insofar as it quantitatively detects fragment ions characteristic of the molecule being detected or quantified.

[0094] The use of detection methods by MS is advantageous because it can be performed in minutes and requires a mass spectrometer with a single analyzer, i.e., an instrument that is less complex than the tandem mass spectrometers used in MS / MS.

[0095] The use of detection methods by MS / MS is also advantageous because it allows for the generation of specific fragments of the molecules to be detected, thereby providing greater specificity to the method according to the present invention.

[0096] According to one embodiment of the present invention, MS / MS spectroscopic analysis is MRM, which has the advantage of using an analysis cycle time of several tens of milliseconds in the mass spectrometer, thereby enabling the detection of a large number of different molecules with high sensitivity and in multiple modes.

[0097] According to another embodiment of the present invention, MS / MS spectroscopy is PRM, which has the advantage of using several fragment ions to characterize the detection of the target molecule.

[0098] Advantageously, the method according to the present invention can be carried out using MALDI-TOF, as described by Claydon et al. and by T. Krishnamurthy and P. Ross. The analysis combines the acquisition of mass spectra with interpretation by expert software. It is extremely simple and can be performed in minutes.

[0099] The method according to the present invention can also be carried out after chromatographic separation using an electrospray source on the crude sample, as described by S. Vaidyanathan et al. or R. Everley et al. In this case, it is possible to identify proteins characteristic of the microorganism being analyzed using different m / z ranges. S. Vaidyanathan et al. used a window between 200 and 2000 Th, while R. Everley et al. used a window between 620 and 2450 Th. The mass spectrum can also be deconvoluted to find the mass of the protein regardless of its charge state. R. Everley et al. thus utilized a mass range of approximately 5000 to 50000 Da.

[0100] Identification of bacteria by detecting their proteins in a sample using MRM in targeted mode was described by the applicant in application WO2011 / 045544. Identification in non-targeted mode is also widely applied by many teams. Examples include the studies of Manes N. et al., who studied the peptidome of Salmonella enterica, or the studies of R. Nandakumar et al. or L. Hernychova et al., who studied the proteome of bacteria after protein digestion with trypsin.

[0101] Therefore, in the context of the present invention, if a bacterial population is susceptible to an antibiotic that induces rupture of its bacterial wall and / or its cell membrane, thereby releasing its intracellular compounds, the spectrum produced by mass spectrometry will have at least one peak of at least one protein characteristic of this bacterial population.

[0102] In the context of this invention, "protein peaks characteristic of bacterial populations" means peaks that enable the distinction of bacterial populations from any other type of molecular sample.

[0103] Conversely, if a bacterial population is resistant to antibiotics, the antibiotics do not cause the bacterial wall or cell membrane to rupture, and as a result, do not release intracellular compounds. Therefore, the spectrum produced by mass spectrometry does not have the protein peak characteristics characteristic of this bacterial population.

[0104] For a given bacterium corresponding to its dominant and characteristic protein, a reference spectrum obtained by mass spectrometry, particularly MALDI-TOF, is available and recorded in databases accessible with commercially available instruments, making it possible to determine, for comparison, the susceptibility of this bacterium to the antibiotic it has been exposed to.

[0105] Advantageously, according to the present invention, the at least one characteristic protein is selected from ribosomal proteins and DNA-binding proteins.

[0106] More advantageously, if the bacterium is Escherichia coli, the at least one characteristic protein is selected from station-phase induced ribosome-associated protein (SPIRAP), acid stress chaperone protein (HdeB), and 50S ribosomal proteins L29, L31, L32, L33, and L35.

[0107] More advantageously, if the bacterium is Klebsiella pneumoniae, the at least one characteristic protein is selected from the DNA-binding protein H-NS and the ribosomal proteins L29, L31, L34, and US9.

[0108] In a particular embodiment of the present invention, following step e) detecting the presence or absence of a protein peak characteristic of the bacterium, the method includes the step of calculating the ratio of the intensity of at least one protein peak characteristic of the bacterium to the intensity of at least one peak characteristic of the antibiotic used.

[0109] In the context of the embodiments described in the previous paragraph, if several characteristic protein peaks are present, the ratio calculation is performed using the sum of the intensities of the protein peaks characteristic of the bacteria. In the same manner, if several peaks characteristic of colistin are present, the ratio calculation is performed using the sum of the intensities of the peaks characteristic of the antibiotic used.

[0110] In the context of the embodiments described in the preceding two paragraphs, the method includes, after the step of calculating the ratio, the step of determining the susceptibility or resistance of a bacterial population to an antibiotic as a function of the obtained ratio and a fixed threshold for each. Those skilled in the art will be able to use their knowledge to determine how to determine the threshold that makes it possible to distinguish resistant bacteria from susceptible bacteria.

[0111] Advantageously, the method according to the present invention further includes the step of identifying the family, genus, or preferably the species of the bacterial population. [Brief explanation of the drawing]

[0112] The method and advantages of the present invention will become apparent from the remainder of this specification, which presents various non-limiting embodiments for carrying out the method of the present invention. Other objects, features and advantages of the present invention will become apparent from the appended figures, the description given below, and the embodiments presented below. [Figure 1] The MALDI-TOF spectrum of colistin sulfate at 10 μg / ml between 100 and 4000 Th is shown. Horizontal axis: mass-to-charge ratio (m / z or Th). Vertical axis: relative intensity %Int, expressed as the percentage of the peak with the highest intensity. [Figure 2] The MALDI-TOF spectrum of colistin sulfate at 10 μg / ml between 2000 and 4500 Th is shown. Horizontal axis: mass-to-charge ratio (m / z or Th). Vertical axis: relative intensity %Int, expressed as the percentage of the peak with the highest intensity. [Figure 3] The MALDI-TOF spectra between 4500 and 10000 Th of the EC_S strain treated according to Example II are shown. [Figure 4] The MALDI-TOF spectra between 4500 and 10000 Th of the EC_R strain treated according to Example II are shown. [Figure 5] The MALDI-TOF spectrum between 4500 and 10000 Th of strain S treated according to Example III is shown. [Figure 6] The MALDI-TOF spectrum between 4500 and 10000 Th of the R16 strain treated according to Example III is shown. [Figure 7] The MALDI-TOF spectra between 4500 and 10000 Th of strain S treated according to Example IV are shown. [Figure 8] The MALDI-TOF spectrum between 4500 and 10000 Th of the R16 strain treated according to Example IV is shown. [Figure 9] The following E. coli strains treated according to Example V—EC_R16, EC_R17, EC_S10, and EC_S15—are shown as MALDI-TOF spectra between 4500 and 10000 Th. [Figure 10]The following Klebsiella pneumoniae strains treated according to Example V are shown as MALDI-TOF spectra between 4500 and 10000 Th. [Figure 11] The following MALDI-TOF spectra between 4500 and 10000 Th are shown for the following Acinetobacter baumannii strains: AB_S044, AB_S045, AB_S046, and AB_R-E105, which were treated according to Example V. [Figure 12] The following Pseudomonas aeruginosa strains treated according to Example V are shown as MALDI-TOF spectra between 4500 and 10000 Th.

[0113] These examples are provided to facilitate a clearer understanding of the present invention, its implementation, and its use. These examples are given for illustrative purposes only and do not limit the scope of the invention. [Examples]

[0114] I. MALDI-TOF Profile of Pure Colistin Colistin (or polymyxin E) is a polymyxin-class antibiotic naturally produced by Paenibacillus polymyxa subsp. colistinus (Benedict RG et al., 1947). Five classes of polymyxins (A, B, C, D, and E) are known, but only polymyxins B and E (or colistin) are used for treatment (Dortet L et al., 2016).

[0115] It is important to note that colistins are polymyxin antibiotics that act on bacterial cell membranes. Due to their amphiphilic nature, like cationic detergents, they penetrate into bacterial cells, are inserted into the phospholipids of the cell wall, and disrupt membrane permeability.

[0116] Colistin is used as a last resort in cases of multi-resistance infections. This is especially true for strains of Enterobacteriaceae, Pseudomonas spp., or Acinetobacter spp. that are simultaneously resistant to carbapenems, aminoglycosides, and fluoroquinolones (Hancock RE, 1997). This phenomenon of multi-resistance is showing a worrying increase in certain countries, such as Greece and Italy, necessitating physicians prescribing colistin. Unfortunately, this more frequent use has led to the emergence of colistin-resistant pathogens (Dortet L et al., 2016). This trend is observed in both bacteria that are naturally resistant to colistin, such as Serratia, Morganella, Proteus, or Providencia, and species that are normally susceptible but express chromosomal or plasmid mechanisms for resistance (e.g., Klebsiella, Escherichia, Salmonella, Enterobacter, Sigella). A plasmid gene (mcr-1) that confers resistance to colistin via horizontal transmission was recently first reported in China (Liu YY et al., 2016). Since then, its transmission has been documented, along with the existence of several variants in both humans and animals (Chen L et al., 2018). These unsettling observations are reason to fear an acceleration in the emergence of colistin resistance (Dortet L et al., 2016).

[0117] Experimental formula C 52 H 98 N 16 O 13 Colistin has a theoretical monoisotopic mass of 1154.750 Da and a theoretical mean chemical mass of 1155.434 Da. After ionization with protons, MALDI-TOF mass spectrometry can detect its first peak (monoisotopic peak) as an isotopic distribution with a mass of 1155.758 Da, or as an undefined distribution with a peak at 1156.442 Da. The possibility of detecting only the monoisotopic peak or only the peak of the isotopic distribution depends on the resolution of the mass spectrometer. High-resolution instruments can generally detect the monoisotopic peak, while low-resolution instruments can only detect the peak of the isotopic distribution.

[0118] Previous experiments were conducted to detect colistin using the VITEK® MS Plus instrument, a MALDI-TOF type instrument commercially available from bioMérieux. These experiments were performed using the following steps: - 1 μl of colistin sulfate (Sigma reference C4461), diluted to 10 μg / ml in water, was deposited onto a disposable target (biomérieux reference 410893). - 1 μl of HCCA matrix (see bioMérieux 411071) was deposited on colistin. - Drying of the target, - Placement of the target on the equipment, and - Analysis by positive mode ionization in the measurement range of 2 to 20000 Th.

[0119] Delayed extraction was optimized for a mass of 2000 Th, and the laser power was fixed at 85. 100 profiles were accumulated from 10 firings to form a mass spectrum. After baseline subtraction and signal smoothing, peaks were detected.

[0120] Figure 1 shows the mass spectra obtained in the mass range of 100 to 4000 Th, showing a cluster at 1156.46 m / z and another cluster at 1170.56 m / z. The first corresponds to unmodified colistin, and the second corresponds to methylated colistin (+14Da). Indeed, colistin contains four amine functional groups that can be methylated. The peak at 1178.68 m / z corresponds to a monovalent colistin ion containing a sodium adduct (+22Da), and the peak at 1192.70 m / z corresponds to a monovalent methylated colistin ion containing a sodium adduct (14+22=+36Da). This research will subsequently make it possible to distinguish between peaks characteristic of colistin and protein peaks characteristic of the bacteria being studied.

[0121] Figure 2 shows spectral observations over the mass range of 2000–4500 m / z. Four clearly defined peaks are observed at approximately 2283;2987, 3439, and 4143 m / z. These peaks correspond to colistin polymer. The mass difference between the peaks at 2283 m / z and 3439 m / z is 1156 m / z, which is the mass of colistin. The same is true for the difference between the peaks at 4143 m / z and 2987 m / z.

[0122] This same spectral profile was obtained at all colistin concentrations tested (10, 5, and 2.5 mg / ml), regardless of incubation time (10 minutes to 4 hours) and the presence or absence of susceptible or resistant strains (not shown). Therefore, these peaks correspond to molecules originating from the colistin sulfate solution used (Sigma reference C4461) and are not characteristic of the bacterial sample.

[0123] II. Determination of susceptibility or resistance to colistin in E. coli after 4 hours This test was performed using the method according to the present invention to determine the susceptibility or resistance of the Gram-negative bacterium Escherichia coli to colistin. This bacterium, also known as Colibacillus and abbreviated as E. coli, is a mammalian intestinal bacterium (Gram-negative) and a symbiotic bacterium of humans. Certain strains of E. coli can be pathogenic, causing gastroenteritis, urinary tract infections, meningitis, or sepsis. While E. coli is generally susceptible to colistin, certain strains have been known to express chromosomal or plasmid resistance mechanisms.

[0124] In the context of this study, two bacterial strains were analyzed: a colistin-susceptible strain of E. coli called EC_S and a resistant strain called EC_R.

[0125] This test was conducted using the following steps: - Prepare suspensions of E. coli EC_S and EC_R strains in 1 McFarland (McF) water (suspension medium, see bioMérieux 70700). - Prepare a solution of colistin sulfate (Sigma, reference C4461) at a concentration of 5 mg / ml in pure water (taking into account the colistin titer indicated by the manufacturer). - Mix 250 μl of each suspension with 250 μl of colistin to obtain a solution of 0.5 McF microorganisms and colistin at a concentration of 2.5 mg / ml. - Homogenize using a 5-second vortex. - Incubate the mixture at 37°C for 4 hours. - Homogenize using a 5-second vortex. - Centrifuge at 4700 revolutions / minute (1500g) for 5 minutes. - Deposit 1 μl of supernatant onto a disposable target (see bioMérieux 410893). - Deposit 1 μl of HCCA matrix (biomérieux, see 411071) onto the supernatant droplet. - Analyze samples using a method suitable for microbiology, namely positive ionization mode and MALDI-TOF spectroscopy in the mass range of 2000 to 20000 Th, with VITEK-MS Plus (biomérieux). - Accumulate 100 profiles from 10 firings and compare that data with data contained in the E. coli database. - Observe the presence or absence of protein peaks characteristic of E. coli. - If a protein characteristic of E. coli is detected, the bacterium's sensitivity to colistin is determined; if no E. coli protein is detected, resistance is determined.

[0126] Figure 3 shows the mass spectrum obtained for the EC_S strain. Within the mass window between 4500 and 10000 Th, the presence of low-intensity peaks distinct from the colistin peaks is observed. These peaks are mainly located at 4650.01, 4666.55, 6258.69, 6318.33, 6414.85, 7176.22, 7276.56, and 7871.90 m / z. A poorly resolved cluster can also be seen at approximately 5370 m / z.

[0127] Masses 6318.33, 7162.22, 7276.56, and 7871.90 correspond to peaks 6316.14, 7158.68, 7274.39, and 7872.02 in E. coli, which were identified as corresponding to ribosomal proteins L32, L35, L29, and L31, respectively (Arnold RJ and Reilly JP. 1999; Wilcox SK et al., 2001; Ryzhov V and Fenselau C, 2001; Jones JJ et al., 2003; Kallow W et al., 2010; Welker M and Moore ERB, 2011; Momo RA et al., 2013). Therefore, these proteins are characteristic of E. coli. This observation indicates that bacterial cells undergo rupture of their cell membranes, and ribosomal proteins are released into the supernatant during incubation at 37°C for 4 hours in the presence of 2.5 mg / ml colistin. Therefore, the method according to the present invention makes it possible to demonstrate that the EC_S strain is sensitive to colistin. Note that a colistin concentration of 2.5 mg / ml corresponds to 1250 times the minimum inhibitory concentration (MIC) above which the strain is considered resistant.

[0128] Figure 4 shows the mass spectrum obtained for the EC_R strain. While the cluster at 5313 m / z and the peaks at 4649 and 4666 m / z are still detectable, the proteins of E. coli L32, L35, L29, and L31, characterized by peaks at 6316.14, 7158.68, 7274.39, and 7872.02 m / z, respectively, are not visible. Therefore, bacterial proteins are not detected. Thus, the bacteria do not appear to have suffered cell membrane rupture during incubation at 37°C for 4 hours in the presence of colistin at a concentration of 2.5 mg / ml, which is well above the MIC. Therefore, this method makes it possible to demonstrate that the EC_R strain is resistant to colistin.

[0129] III. Determination of susceptibility or resistance to colistin in E. coli at 37°C for 30 minutes. This test was conducted to demonstrate that the method of the present invention allows for the determination of susceptibility or tolerance in 30 minutes.

[0130] E. coli strains S and R16 were analyzed using the same protocol as in Example II, except for a 30-minute incubation period at 37°C. Strain S was susceptible to colistin, while strain R16 was resistant to colistin.

[0131] Figure 5 shows the mass spectrum obtained for strain S. This spectrum has strong peaks at 6257.2, 6318.7, 7275.4, and 7871.6 m / z. These peaks are characteristic of E. coli ribosomal proteins L33, L32, L29, and L31, respectively. Therefore, strain S underwent cell membrane rupture during incubation at 37°C for 30 minutes in the presence of 2.5 mg / ml colistin. Thus, its characteristic sensitivity to colistin was confirmed.

[0132] Figure 6 shows the mass spectrum obtained for strain R16. In contrast to the spectrum of strain S (Figure 5), this spectrum does not have a strong peak between 6000 and 10000 m / z. Therefore, it did not undergo lysis during incubation at 37°C for 30 minutes in the presence of 2.5 mg / ml colistin. This method confirms the resistance of this strain to colistin.

[0133] IV. Determination of susceptibility or resistance to colistin in E. coli at 50°C for 30 minutes. The same E. coli strains R16 and S were analyzed using the same protocol as in Example III, except for incubation at 50°C.

[0134] Figure 7 shows the mass spectrum obtained for strain S. Strain S clearly shows strong peaks at 6256.4, 6316.9, 7274.1, and 7870.2 m / z. These peaks are characteristic of ribosomal proteins L33, L32, L29, and L31, respectively. Therefore, the cell membrane of strain S was lysed during incubation at 50°C for 30 minutes in the presence of 2.5 mg / ml colistin. Thus, the sensitivity of strain S to colistin was confirmed.

[0135] Figure 8 shows the mass spectrum obtained for strain R16. This spectrum does not show a strong peak between 6000 and 10000 m / z. Therefore, this strain was not lysated during incubation at 50°C for 30 minutes in the presence of 2.5 mg / ml colistin. Thus, resistance to colistin of this strain is confirmed.

[0136] Unexpectedly, protein peaks L33, L32, L29, and L31 were stronger in susceptible strains after incubation at 50°C than after incubation at 37°C, but they remained almost undetectable in resistant strains.

[0137] Determination of susceptibility or resistance of various bacteria to V. colistin These tests were conducted using the method of the present invention, and the susceptibility or resistance of several bacteria of several different strains was determined to demonstrate the reproducibility of this method in different bacteria and different strains. These tests were also conducted using incubation at 50°C for 10 minutes, demonstrating that a very short incubation time could be used. The tests were conducted using the following steps: - Prepare a suspension of microorganisms at 2 McF in water (suspension medium, see bioMérieux 70700). - Prepare a solution of colistin sulfate (Sigma reference C4461) in pure water at a concentration of 2X (taking into account the colistin titer indicated by the manufacturer). X = 2.5 mg / ml for Escherichia coli and Klebsiella pneumoniae, and X = 20 μg / ml for A. baumannii and Pseudomonas aeruginosa. - Mix 200 μl of microorganisms in suspension with 200 μl of colistin 2X to obtain a solution of 1 McF of microorganisms and colistin at concentration X. - Perform a negative control in parallel by diluting the microbial solution from Step 3 with a solution of pure water (without colistin). - Homogenize using a 5-second vortex. - Incubate the mixture at 50°C for 10 minutes while stirring at 1400 rpm using a thermomixer. - Homogenize using a 5-second vortex. - Filtration using a filter with a porosity of 0.22 μm (Centricon, Merck-Millipore), - Deposit 1 μl of filtrate onto a disposable target (see bioMérieux 410893). - Deposit 1 μl of HCCA matrix (biomérieux, see reference 411071) into the filtrate. - The sample is analyzed using a method commonly used in microbiology, namely, positive ionization mode and MALDI-TOF spectroscopy in the mass range of 2000 to 20000 Th. - Accumulate 100 profiles from 10 launches and compare that data with data contained in a database of the species being investigated. - Observe the presence or absence of protein peaks characteristic of the species being studied. - If a protein characteristic of the species being studied is detected, it determines sensitivity to colistin; if the protein of the species being studied is not detected, it determines tolerance. - Verify the results of the determination to check whether or not there are microbial protein peaks in the negative control.

[0138] As shown in Table 1 below, different strains of Escherichia coli (EC_S10, EC_S15, EC_R16, EC_R17) and Klebsiella pneumoniae (KP_S10, KP_S15, KP_R9, KP_R16) were analyzed using the protocol described above. To evaluate the effectiveness of the method according to the present invention, different strains were pre-analyzed by microdilution in broth to determine their state (susceptibility or resistance). This method is considered a reference method by CLSI and EUCAST. [Table 1]

[0139] Therefore, among the analyzed bacterial strains, some are susceptible to colistin and others are resistant, and among the resistant strains, some exhibit chromosomal resistance and others plasmid resistance.

[0140] The proteins characteristic of Klebsiella pneumoniae species are the DNA-binding protein H-NS, ribosomal proteins L29, L31, L34, and US9, a cold shock protein containing a CsbD domain (CsbD domain-containing protein), and uncharacterized proteins with m / z 6290 and 8308. These proteins are characterized by peaks of 7705 (DNA-binding protein H-NS), 7274 (L29), 7743 (L31), 5381 (L34), 7384 (US9), and 8309 (CsbD), respectively, at more or less parts per million (ppm). For those parts, the hypothetical proteins are characterized by peaks of 6290 and 7678, at more or less parts per million (ppm).

[0141] Figures 9 and 10 show the mass spectra obtained for the Escherichia coli and Klebsiella pneumoniae strains shown in Table 1, respectively. The spectra of strains EC_S10, EC_S15, KP_S10, and KP_S15 have peaks characteristic of their respective species and are therefore susceptible to colistin. Conversely, strains EC_R16, EC_R17, KP_R9, and KP_R16 do not have peaks characteristic of their respective species and are therefore resistant to colistin.

[0142] Therefore, the results obtained by the reference method (trace dilution with broth) were confirmed by the method according to the present invention. In fact, the same state was obtained for each strain by trace dilution with broth using the method of the present invention.

[0143] To further demonstrate the reproducibility of the method according to the present invention, Table 2 below shows different strains of Acinetobacter baumannii (AB_S044, AB_S045, AB_S046, AB_R-E105) and Pseudomonas aeruginosa (PA_S062, PA_SE64, PA_RE66, PA_RE68) analyzed according to the protocol described above. As before, different strains were pre-analyzed by microdilution in broth to determine their status (susceptibility or resistance). [Table 2]

[0144] Strains of A. baumannii (AB_S044, AB_S045, AB_S046, AB_R-E105) and Pseudomonas aeruginosa (PA_S062, PA_SE64, PA_RE66, PA_RE68) were analyzed using the protocol described in Example V. Accordingly, they were incubated in the presence of 20 μg / ml colistin.

[0145] Figure 11 shows the mass spectra obtained for the A. baumannii strains listed in Table 2. In these spectra, strains AB_S044, AB_S045, and AB_S046 have characteristic protein peaks for their species, particularly at approximately 5748 and 5770 Th, and are therefore susceptible to colistin. Conversely, strain AB_R-E105 does not have a characteristic protein peak for its species and is therefore resistant to colistin.

[0146] Figure 12 shows the mass spectra obtained for the Pseudomonas aeruginosa strains listed in Table 2. In these spectra, strains PA_S062 and PA_SE64 have characteristic peaks for their species, particularly at approximately 5449, 5469, 5793, and 6975Th, and are therefore susceptible to colistin. Conversely, strains PA_RE66 and PA_RE68 do not have characteristic peaks for their species and are therefore resistant to colistin.

[0147] As previously stated, these results demonstrate that the method according to the present invention makes it possible to determine the state of each strain. These results are confirmed by the so-called reference method.

[0148] VI. Calculation of ratios for determining bacterial susceptibility or resistance to antibiotics In the context of this embodiment, as in Example V, the present invention was used to perform tests to determine the susceptibility or resistance of several bacteria in several strains, including the following steps: - Measure the intensity of each observed peak. - Find the ratio between the sum of the intensities of microbial protein peaks and at least one colistin peak. - The state of tolerance or sensitivity to colistin is determined as a function of the sum of the obtained ratios and fixed thresholds for each. - Resistant microorganisms will have a low proportion, while susceptible microorganisms will have a high proportion.

[0149] In the context of this invention, the threshold is empirically determined by observation of the results. Those skilled in the art will know how to determine this threshold without difficulty using their own knowledge. A method for determining this threshold is illustrated in the following examples.

[0150] [Analysis of E. coli strains] First, we will study the same E. coli strains used previously (EC_S10, EC_S15, EC_R16, EC_R17) and analyze proteins characteristic of these E. coli species. These proteins are: stationary-phase induced ribosome-associated protein (SPIRAP), acid-stress chaperone HdeB, and 50S ribosomal proteins L29, L31, and L33.

[0151] Table 3 below shows the peak intensities observed for colistin and other proteins characteristic of E. coli. [Table 3]

[0152] Table 4 below shows the ratio of the peak intensity observed for each protein to the peak intensity observed for colistin, as well as the sum of these ratios for each protein. [Table 4]

[0153] The results shown in Table 4 indicate that strains EC-S10 and EC-S15 have a combined ratio greater than 0.3, while strains EC-R16 and EC-R17 have a combined ratio of zero. Based on these observations, the inventors fixed the threshold at 0.30. Therefore, in this case, strains with a combined ratio strictly less than 0.30 are considered resistant, while strains with a combined ratio of 0.30 or more are considered susceptible. Consequently, strains EC-S10 and EC-S15 are classified as susceptible, while strains EC-R16 and EC-R17 are classified as resistant.

[0154] [Analysis of Klebsiella pneumoniae strains] We also studied KP S10-1, KP S10-2, KP S15-1, KP S15-2, KP R9-1, KP R9-2, KP R16-1, and KP R16-2 strains of Klebsiella pneumoniae and analyzed proteins characteristic of these species. These characteristic proteins include the DNA-binding protein H-NS, ribosomal proteins L29, L31, L34, and US9, a cold shock protein containing a CsbD domain (a protein containing a CsbD domain), and uncharacterized proteins with m / z 6290 and 8308.

[0155] Table 5 below shows the peak intensities observed for colistin and characteristic proteins of Klebsiella pneumoniae. [Table 5]

[0156] Table 6 below shows the ratio of the peak intensity observed for each protein to the peak intensity observed for colistin, as well as the sum of these ratios for each protein. [Table 6]

[0157] The results shown in Table 6 indicate that the combined ratio of strains KP-S15 and KP-S10 exceeds 0.45, while the combined ratio of strains KP-R16 and KP-R9 is less than 0.11. Therefore, based on the observation of the results, the inventors fixed the threshold at 0.45. Consequently, in this case, strains whose combined ratio is strictly less than 0.45 are considered resistant, while strains whose combined ratio is 0.45 or greater are considered susceptible. Therefore, strains KP-S15 and KP-S10 are classified as susceptible, and strains KP-R16 and KP-R9 are classified as resistant.

[0158] [List of References] - Arnold RJ and Reilly JP, Observation of Escherichia coli ribosomal proteins and their posttranslational modifications by mass spectrometry. Analytical Biochemistry, 269:105-112, 1999 - Benedict RG et al., Antibiotic activity of Bacillus polymyxa. Journal of Bacteriology, 1947 - Chen L et al., Newly identified colistin resistance genes, mcr-4 and mcr-5, from upper and lower alimentary tract of pigs and poultry in China. PLoS One, 13(3):e0193957, 2018 - Claydon M et al., The rapid identification of intact microorganisms using mass spectrometry, Nature Biotechnology. 14 :1584-1586, 1996 - Donnelly et al., Best practices and benchmarks for intact protein analysis for top-down mass spectrometry, Nature Methods, 16(7):587-594, 2019 - Dortet L et al., Emergence of colistin resistance in enterobacteria: a breach in the last bastion against pan-resistance! Journal des Anti-infectieux, 18 :139-159, 2016 - Everley R, et al., Characterization of Clostridium species utilizing liquid chromatography / mass spectrometry of intact proteins, Journal of Microbiological Methods, 77:152-158, 2009 - Fortin T, et al., Clinical Quantitation of Prostate-specific Antigen Biomarker in the Low Nanogram / Milliliter Range by Conventional Bore Liquid Chromatography-Tandem Mass Spectrometry (Multiple Reaction Monitoring) Coupling and Correlation with ELISA Tests, Molecular & Cellular Proteomics, 8(5) : 1006-1015, 2009 - Hancock RE. Peptide antibiotics. The Lancet, 349(9049) :418-422, 1997 - Hernychova L, et al., Detection and Identification of Coxiella burnetii Based on the Mass Spectrometric Analyses of the Extracted Proteins, Analytical Chemistry, 80 :7097-7104, 2008 - Hindler JA, et al. Colistin MIC variability by method for contemporary clinical isolates of multidrug-resistant Gram-negative bacilli. Journal of Clinical Microbiology, 51(6) :1678-1684, 2013 - Jones JJ et al., Investigation of MALDI-TOF and FT-MS techniques for analysis of Escherichia coli whole cells. Analytical Chemistry, 75:1340-1347, 2003 - Kallow W et al., MALDI-TOF MS and microbial identification: years of experimental development to an established protocol, p. 255-276. In: H. N. Shah, S. E. Gharbia, and V. Encheva (eds.), Mass spectrometry for microbial proteomics. Wiley, Chichester, 2010 - Keshishian H et al., Quantitative, Multiplexed Assays for Low Abundance Proteins in Plasma by Targeted Mass Spectrometry and Stable Isotope Dilution, Molecular & Cellular Proteomics, 6(12):2212-2229, 2007 - Krishnamurthy T and Ross P, Rapid Identification of Bacteria by Direct Matrix-Assisted Laser Desorption / Ionization Mass Spectrometric Analysis of Whole Cells, Rapid Communication in Mass Spectrometry, 10 :1992-1996, 1996 - Landman D et al., Irreproducible and uninterpretable Polymyxin B MICs for Enterobacter cloacae and Enterobacter aerogenes. Journal of Clinical Microbiology, 51(12):4106-4111, 2013 - Liu et al., Emergence of plasmid-mediated colistin resistance mechanism MCR-1 in animals and human beings in China: a microbiological and molecular biological study. The Lancet Infectious Diseases, 16(2):161-168, 2016 - Manes NP et al., Targeted Protein Degradation by Salmonella under Phagosome-mimicking Culture Conditions Investigated Using Comparative Peptidomics, Molecular & Cellular Proteomics, 6(4): 717-727, 2007 - Mirande C. et al. Rapid detection of carbapenemase activity: benefits and weaknesses of MALDI-TOF MS, European Journal of Clinical Microbiology & Infectious Diseases, 34(11):2225-2234, 2015 - Momo RA et al., MALDI-ToF mass spectrometry coupled with multivariate pattern recognition analysis for the rapid biomarker profiling of Escherichia coli in different growth phases. Analytical and Bioanalytical Chemistry, 405:8251-8265, 2013 - Nandakumar R et al., Proteomic analysis of endodontic infections by liquid chromatography-tandem mass spectrometry, Oral Microbiology Immunology, 24 :347-352, 2009 - Peterson AC et al., 2012, Parallel Reaction Monitoring for High Resolution and High Mass Accuracy Quantitative, Targeted Proteomics, Molecular & Cellular Proteomics 11: 10.1074 / mcp.O112.020131, 1475-1488, 2012 - Poirel L et al., Polymyxins: Antibacterial Activity, Susceptibility Testing, and Resistance Mechanisms Encoded by Plasmids or Chromosomes, Clinical Microbiology Reviews, 30(2):557-596, 2017 - Ryzhov V and Fenselau C, Characterization of the protein subset desorbed by MALDI from whole bacterial cells. Analytical Chemistry 73:746-750, 2001 - Spapen H et al., Renal and neurological side effects of colistin in critically ill patients. Annals of intensive care, 1(1)?:1-14, 2011 - Vaidyanathan S et al., Discrimination of Aerobic Endospore-forming Bacteria via Electrospray-Ionization Mass Spectrometry of Whole Cell Suspensions, Analytical Chemistry, 73 :4134-4144, 2001 - Welker M and Moore ERB, Applications of whole-cell matrix-assisted laser-desorption / ionization time-of-flight mass spectrometry in systematic microbiology. Systematic and Applied Microbiology, 34:2-11, 2011 - Wilcox SK et al., Single ribosomal protein mutations in antibiotic-resistant bacteria analyzed by mass spectrometry. Antimicrobial Agents and Chemotherapy, 45:3046-3055, 2001

Claims

1. A method for determining the susceptibility or resistance of at least one identified bacterium to at least one antibiotic, a) A step of bringing a sample containing the bacteria into contact with at least one antibiotic, which, if the bacteria are said to be "sensitive" to the at least one antibiotic, induces rupture of the bacterial wall and / or cell membrane and causes the release of intracellular compounds from the bacteria. b) The step of incubating the sample with at least one antibiotic, c) A step of purifying a sample containing only intracellular compounds released by the rupture of bacterial walls and / or cell membranes by at least one antibiotic in steps a) and b), by removing intact bacteria and cellular debris. d) A step of analyzing the purified sample by mass spectrometry. e) A step of detecting the presence or absence of at least one peak of at least one characteristic protein of the bacterium, and f) The step of determining that the bacterial population is susceptible to the antibiotic if at least one peak of at least one characteristic protein is detected, and determining that the bacterial population is resistant to the antibiotic if at least one peak of at least one characteristic protein is not detected. Includes, The mass spectrometry is of the MS type, MS / MS type, or MS type followed by MS / MS type spectroscopy, and the MS type mass spectrometry is of the MALDI-TOF type mass spectrometry. method.

2. The method according to claim 1, wherein the MS / MS type mass spectrometry is PRM, SRM, MRM, DDA (data-dependent acquisition), or DIA (data-independent acquisition) type mass spectrometry.

3. The method according to claim 1 or 2, wherein the step of purifying the sample is carried out by at least one of the following techniques: centrifugation, filtration, electrophoresis, or chromatography.

4. The method according to any one of claims 1 to 3, wherein the antibiotic is selected from polymyxin, β-lactam antibiotics, aminoglycosides, quinolones, and glycopeptides.

5. The method according to claim 4, wherein the antibiotic is polymyxin, selected from colistin and polymyxin B.

6. The method according to any one of claims 1 to 5, wherein the antibiotic is concentrated at a concentration at least 10 times higher than the minimum inhibitory concentration of the antibiotic for the bacterial population.

7. The method according to any one of claims 1 to 6, wherein the bacterial population is Gram-negative bacteria.

8. The method according to claim 7, wherein the Gram-negative bacteria are selected from Escherichia coli, Acinetobacter baumannii, Klebsiella pneumoniae, Acinetobacter haemolicus, Acinetobacter junii, Cytrobacter freundii, Enterobacter asbriae, Enterobacter cloacae, fluorescent bacteria, Salmonella enteritidis, Salmonella paratifida subspecies Java, Salmonella serotype agona, Salmonella serotype enteritidis, Salmonella serotype haifa, Salmonella serotype newport, and Pseudomonas aeruginosa.

9. The method according to any one of claims 1 to 8, wherein the incubation step takes less than 30 minutes.

Images