Ready-to-use calibration composition and method for calibrating a mass spectrometer

Abstract

The invention relates to a ready-to-use calibration composition for a mass spectrometer, comprising whole E. coli bacteria suspended in a liquid which is based on alcohol (s) and / or acetonitrile and is sanitizing, volatile and does not denature bacterial proteins.

Description

[0001] READY-TO-USE CALIBRATION COMPOSITION AND METHOD FOR CALIBRATING A MASS SPECTROMETER

[0002] The present invention relates to the field of mass spectrometry, which finds various applications in the fields of biomedical research, medicine, diagnostics and biotechnology. More specifically, the invention relates to a calibration composition and a method for calibrating a mass spectrometer using the matrix-assisted desorption-ionization technique, known as MALDI. In particular, the invention is suitable for the matrix-assisted desorption-ionization and time-of-flight technique, known as MALDI-TOF. The calibration composition and the calibration method according to the invention are intended to ensure reliable calibration of the spectrometer used and thus to obtain sufficiently precise measurements of the masses of the ions generated from a sample to be analyzed.

[0003] The invention also relates to the use of a calibration composition according to the invention, as an external standard for the calibration of a mass spectrometer and a method for characterizing a sample containing at least one microorganism.

[0004] The analysis of microorganisms and, more generally, that of a sample likely to contain a microorganism, by mass spectrometry and, in particular, by MALDI or MALDI-TOF has been used for a number of years to carry out rapid identification of microorganisms, and also to determine the resistance of microorganisms to certain antimicrobial agents, as illustrated, for example, in application WO 2016 / 0166580, in the name of the applicant. The identification of a microorganism is carried out from the mass spectrum of the most abundant proteins in the microorganism, by comparison with reference data allowing, in particular, the identification of the family, the genus and most often the species of the microorganism.Routinely, the protocol implemented includes the deposit of a sample to be analyzed, likely to contain a microorganism, on a MALDI analysis plate, the addition of a matrix adapted to the MALDI technique, the acquisition of the mass spectrum using a MALDI-TOF mass spectrometer, and the identification of the species by comparison with reference data stored in a database. The MALDI-TOF technique is also used to detect the resistance of a microorganism to an antibiotic, and in particular to identify a phenotype, responsible for the hydrolysis of beta-lactam antibiotics, due to the secretion of beta-lactamase enzymes, and in particular, carbapenemase enzymes. Various mass spectrometers, suitable for such characterization, are marketed in particular by the applicant (in particular, under the references VITEK® MS and VITEK® MS Prime), as well as by the companies Bruker Daltonics and Autobio in particular.These spectrometers are of the MALDI-TOF type and include a laser ionization source and a time-of-flight mass analyzer. They are designed to operate with an analysis plate, also called a MALDI plate or MALDI target, on which a sample to be analyzed is deposited in combination with a matrix adapted to the MALDI technique, the latter being able to be deposited at the same time or after the sample.

[0005] Then, the analysis plate is introduced into an analysis chamber of the spectrometer which is brought to a relatively high vacuum level, with a pressure which is for example less than 10' 5 mbar (10' 3 Pa), typically in the range of 10' 6 at 10' 9 mbar (10' 4 at 10' 7Pa). Under these vacuum conditions, the sample placed within the MALDI matrix is ​​subjected to gentle laser ionization. The matrix then absorbs the photon energy and the release of this energy leads to the sublimation of the matrix, the desorption of the molecules present in the sample and the appearance of matter in a state called plasma. Within this plasma, charge exchanges occur between molecules from the matrix and molecules from the sample, and in particular from microorganisms. For example, protons can be removed from the matrix and transferred to proteins, peptides and organic compounds present in the sample.

[0006] This step allows for gentle ionization of the molecules present without inducing their destruction. The sample thus releases ions of different sizes. In MALDI-TOF spectrometers, the generated ions are then accelerated by an electric field and fly freely in a tube under reduced pressure, called a flight tube. The smallest ions will then "travel" faster than the larger ions, thus allowing their separation. At the terminal end of the flight tube is located a detector. Thus, a mass spectrum is obtained, representing the intensity of the signal corresponding to the number of ionized molecules of the same mass to charge [m / z], as a function of the time of flight (or TOF for "Time Of Flight" in English) of the molecules that strike the detector. The time of flight taken by the ions is used to calculate their mass to charge ratio [m / z], expressed in Thomson (Th).

[0007] Calibrating a mass spectrometer increases the reliability of the measurements taken and therefore the use of the mass spectra generated, particularly for the identification of microorganisms in a sample. Calibration consists of adjusting the experimental mass-to-charge value obtained to a so-called calibrated value which will then appear on the generated mass spectrum and be used for the characterization of the sample to be analyzed. Such calibrations can be carried out using an internal calibrating agent integrated into the sample, or an external calibrating agent, called an external standard, deposited on at least one deposition zone of a MALDI analysis plate.

[0008] As illustrated in Figure 1, a MALDI I analysis plate typically includes:

[0009] - a series of deposition zones 1 (also called spots), called analysis zones, for the deposition of sample(s) to be characterized; after drying, these zones 1 form so-called characterization zones, and

[0010] - one or more deposition zones 2, called reference zones, for the deposition of an external standard; after drying, these zones 2 form so-called control zones; these zones are in the form of wells, most often circular in shape.

[0011] MALDI plates are typically made from materials such as stainless steel, gold, silicon, a metal-coated plastic polymer, a conductive agent (such as carbon black), or coated glass. They typically have a textured surface to facilitate the deposition of analytical samples and external standards.

[0012] To promote subsequent ionization, the surface of the plate is generally conductive, at least at the analytical and reference deposition areas. For example, such an analytical plate is formed from a polymer such as polypropylene, said polymer being covered with a layer of stainless steel. The polymer may contain a conductive material such as carbon black. Various MALDI plates are commercially available, such as bioMérieux's VITEK® MS plates (disposable) and Bruker Daltonics' MALDI Biotarget plates (reusable). Such plates I most often comprise 48 to 96 analytical areas 1, and at least one, or even two or three reference areas 2, the size of which may be different from that of the analytical areas 1.

[0013] Currently, for the calibration of its devices, the applicant recommends using as an external standard, a freshly cultured Escherichia coli (E. coli) ATCC 8739 strain, as described in the VITEK® MS user manual. The bacteria are cultured on agar medium for 18 to 24 hours at 37°C. A quantity of bacteria is then recovered and placed on the MALDI plate, at a reference zone. This calibration method has the disadvantage of requiring freshly cultured bacteria.

[0014] In addition, take bacteria from an agar culture to place them on a MALDI analysis plate, in a specifically dedicated area of ​​a few mm 2 (especially from 5 to 15 mm 2), and forming a thin and homogeneous layer of cells, turns out to be a particularly delicate manual operation. The results can vary greatly in quality from one operator to another, from one manipulation to another. However, if the external standard deposit is of poor quality, the identification of a microorganism on a MALDI analysis plate which includes a large number of spots for the deposit of a sample of interest (most often from 16 to 96 analysis spots) can fail. The operator is then forced to completely redo the preparation of the plate.

[0015] Other types of external standard for calibrating a mass spectrometer are proposed in the prior art. For example, BTS (Bacterial Test Standard), marketed by Bruker, is an E. coli extract enriched with two high molecular weight proteins. BTS was developed to serve as quality control for Bruker's MALDI-TOF Biotyper spectrometer. The BTS data sheet indicates that its specific composition covers the entire protein mass range used by the Biotyper for the accurate identification of microorganisms, and in particular, the range from 3.6 to 17 kDa. With BTS, the Biotyper performs an automatic quality control before each identification run. The quality control process includes calibrating the mass spectrometer, checking the laser setting, and evaluating the quality of the spectrum.The system performance is finally confirmed by the identification of E coli, which must meet minimum performance criteria. Before use, a dose of BTS is placed in approximately 50 μL of a solvent composed of 50% v / v (volume over total solvent volume) acetonitrile, 47.5% v / v water, and 2.5% v / v trifluoroacetic acid. The mixture is incubated for 5 minutes at room temperature, and various mixing operations by pipetting and centrifugation are recommended by the supplier Bruker in document Ref. 8290190, September 2017, IVD Bacterial Test Standard. 5 μL aliquots thus obtained can be stored in capped tubes and frozen at -18°C or lower, and then used to deposit on four different deposition zones of a MALDI plate.

[0016] Patent CN104931572B can also be cited. This provides a calibration composition that, like Bruker's BTS, is of the bacterial extract type, but with the advantage for its user of being ready to use. This bacterial lysate / extract is prepared according to the following method according to the available machine translation:

[0017] (1) Select E. coli strains ATCC 8739, MG 1655, JM 109, cultivate them to their logarithmic growth phase, mix them according to the mass ratio of 3-8:1-5:1-3;

[0018] (2) Protein washing solution A which is deionized water or ultrapure water is added to the strain mixture, according to a mass / volume ratio of 5-10 mg:0.1-0.5 mL, and then protein washing solution B, which is anhydrous ethanol, is added according to a volume ratio of 5-2:1; after mixing and centrifugation, the supernatant is discarded;

[0019] (3) Add to the precipitate obtained in step (2) a protein dissolution solution A, according to a mass / volume ratio of (5-10 mg):(0.025-0.125 mL), mix well, and then add the same volume of a protein dissolution solution B as for the protein dissolution solution A; centrifuge and take the supernatant as a calibration composition; the protein dissolution solution A is a 50-80% formic acid solution of HPLC grade and the protein dissolution solution B is acetonitrile of HPLC grade.

[0020] It therefore appears that this calibration composition is complex to manufacture; it involves multiple and delicate preparation operations.

[0021] In view of the calibration compositions of the prior art and the deficiencies observed, the present invention proposes to provide new calibration compositions which combine the following advantages:

[0022] - ease of use and handling, particularly in terms of the operation of collecting and depositing an appropriate quantity of bacteria in a thin and homogeneous layer, which does not require any particular experience or skill on the part of the user,

[0023] - ready-to-use formulations with a sufficiently long shelf life,

[0024] - simple and inexpensive industrialization.

[0025] In this context, the present invention provides a calibration composition for a mass spectrometer, comprising whole E. coli bacteria, suspended in a liquid based on alcohol(s) and / or acetonitrile.

[0026] For the proposed application, the inventors have thus observed and demonstrated that alcohols of the “unsaturated aliphatic alcohol” type (such as methanol, ethanol, propanol and butanol) and acetonitrile were particularly interesting due to their good volatility, their non-denaturing antiseptic properties for bacteria, and their chemical / reactive neutrality with respect to proteins. Therefore, a calibration composition according to the invention is formulated in the form of an intrinsically volatile, antiseptic liquid, which allows good preservation over time of bacteria and their proteins.

[0027] For the purpose of simplifying writing, in this description, the term "alcohol" designates an unsaturated aliphatic alcohol, in particular an unsaturated C1-C4 aliphatic alcohol, preferably ethanol and isopropanol.

[0028] A calibration composition according to the invention thus contains whole bacteria, preserved in this state until the time of undertaking the calibration operations. Maintained in such an environment, the bacterial proteins, in particular those chosen as references for the calibration, benefit from protection against biological and / or chemical alterations, ensured both by the very structure of the bacteria remaining whole / intact and by the alcohol(s) and / or acetonitrile. While exerting a sanitizing effect on the entire composition, the alcohol(s) and / or acetonitrile of a calibration composition according to the invention, as such, have an advantageous chemical / reaction neutrality with respect to the proteins.

[0029] In addition to ensuring good preservation over time of the bacteria and their proteins, such a liquid and volatile formulation offers the calibration compositions according to the invention the advantage of being able to be used directly and as such, and of greatly facilitating the calibration operations of spectrometers. With a calibration composition according to the invention, the operator will simply have to take the prescribed volume (which corresponds to a pre-established quantity of cells), and deposit it on the analysis plate. To do this, a pipette or any other suitable utensil may be used. The high volatility of the alcohol(s) and / or acetonitrile which compose(s) the liquid phase of a calibration composition according to the invention, allows the liquid phase to evaporate quickly to leave only a thin and homogeneous layer of cells on the deposition zone.Once the matrix has been added and the analysis plate loaded into the spectrometer, data acquisition can begin. Similarly, it is possible to mix the calibration composition with the matrix, then deposit a volume of this mixture onto the analysis plate, and load it into the spectrometer once the mixture is sufficiently dry.

[0030] Advantageously, the liquid phase of a calibration composition according to the invention is based on alcohol(s). According to a preferred embodiment, it is an aqueous liquid with an alcohol(s) content of at least 20% v / v, typically 70% v / v.

[0031] Advantageously and according to the invention, the alcohol(s) included in the calibration composition according to the invention is / are ethanol and / or isopropanol. According to a particularly preferred embodiment, the alcohol-based liquid of a calibration composition according to the invention consists of an aqueous water / ethanol mixture, with an ethanol content of 70%.

[0032] According to another preferred embodiment, the liquid phase of a calibration composition according to the invention is based on acetonitrile, i.e. a 100% v / v acetonitrile solution or a water / acetonitrile mixture, with an acetonitrile content of at least 20% v / v, typically 70% v / v.

[0033] Advantageously, to have the simplest possible composition, the E. coli bacteria of a calibration composition according to the invention are derived from a single strain. In particular, these bacteria are derived from the BL21 strain or the ATCC 8739 strain. The E. coli BL21 strain was described in 1986 (Studier F.W., Moffatt B A., J Mol Biol. 1986 May 5; 189(1): 113-30. “Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes”), and was fully characterized by sequencing in 2015 (Jeong H., Ju Kim H., Jun Leeb S., Genome Announc. 2015 Mar 19;3(2). e00134-15. Complete Genome Sequence of Escherichia coli Strain BL21). This E. coli strain is commercially available and can be obtained from Agilent (catalog number #200133). Strain ATCC 8739 is also commercially available, notably from TATCC. Its genetic sequence is available in GenBank under the number GCA 016864475.1.Although not preferred, instead of strain BL21 or ATCC 8739, it is also possible to use any other strain of E coli known to the person skilled in the art, in particular strain E coli K-12 (ATCC 10798).

[0034] Advantageously and according to the invention, the following calibration compositions are particularly preferred:

[0035] - a calibration composition consisting of a suspension of E coli bacteria of the BL21 strain in an aqueous liquid based on alcohol and / or acetonitrile.

[0036] - a calibration composition consisting of a suspension of E coli bacteria of the ATCC 8739 strain in an aqueous liquid based on alcohol and / or acetonitrile.

[0037] In the calibration compositions according to the invention, the composition of alcohol(s) and / or acetonitrile is chosen for structural preservation of the E coli bacteria present in the composition, at a temperature of 0 to 37°C, in particular of 0 to 25°C.

[0038] In particular, advantageously, this aqueous liquid based on alcohol(s) and / or acetonitrile, or more generally the calibration composition, does not include any acid or other chemical substance likely to react with the proteins, directly or indirectly. Preferably, any chemical substances (such as detergents, bases) likely to extract proteins that are not usually detected with conventional deposition will be avoided.

[0039] In particular, in the calibration compositions according to the invention, the E. coli bacteria are present at a concentration of 2 to 7 McF (McFarland), in particular 3 to 6 McF, for example 5 McF. Such a concentration can be determined by a densitometer (for example, the DENSIMAT marketed by the company bioMérieux), which makes it possible to measure the optical density of a liquid composition, in particular under incident light of a wavelength between 500 and 600 nm, and preferably at 550 nm. The concentration of bacteria in a liquid composition can also be expressed in colony-forming units per milliliter (cfu / mL). Also, advantageously, in the calibration compositions according to the invention, the E. coli bacteria are present at a concentration of 6.10 8 at 9:10 p.m. 8 cfu / mL.

[0040] Table 1 below gives the correspondence between optical density, value in McFarland units (McF) and bacterial concentration in 108 cfu / mL.

[0041] Table 1: Correspondence between McFarland units and cfu / mL

[0042] The calibration compositions according to the invention are stable. This stability can be demonstrated by mass spectrometry, in particular by MALDI-TOF. No degradation of the proteins present is thus observed, in particular the proteins chosen as references for the calibration. Their masses, apart from the analytical error, do not change. The stability of the calibration compositions according to the invention can also be verified through their capacity to provide calibrations of relatively constant high quality over time.

[0043] Due to their good stability, the calibration compositions according to the invention can be stored in tubes, in particular plastic tubes, closed with a cap, or in any other type of suitable and hermetically sealed container. They can then be stored in this way for several months, in particular for at least 1 month, or at least 2 months, and preferably for at least 6 months, at a temperature of 0 to 37°C, typically 0 to 25°C. Advantageously, it is not necessary to store the compositions according to the invention in frozen form, at negative temperatures, in particular at -18°C.

[0044] Thus, without taking any particular storage precautions, the calibration compositions according to the invention lead to deposits on MALDI plates which are reliable and reproducible and therefore make it possible to ensure calibrations, and therefore mass spectrometry measurements on samples or microorganisms to be characterized, which are just as reliable and reproducible.

[0045] The calibration compositions according to the invention are ready-to-use solutions that can easily be implemented during mass spectrometry measurements and / or calibration procedures. Due to their liquid formulation, they are easy to sample and deposit. Evaporation of the liquid phase then leaves room for a localized deposit, in the form of a thin and uniform cellular layer. In particular, a precise volume of 1 to 2 μL, and typically 1 μL, of a calibration composition according to the invention can be easily and quickly sampled and then deposited on the analysis plates, precisely on the deposition areas. The calibration compositions of the invention simplify the calibration workflow to be carried out during mass spectrometry measurements, by avoiding the need to carry out new cultures of E. coli every day, since the calibration compositions are easily stored in a ready-to-use form.

[0046] According to another of its objects, the invention relates to the use of a calibration composition according to the invention, as an external standard for the calibration of a mass spectrometer, according to the MALDI technique, and in particular according to the MALDI-TOF technique.

[0047] The calibration compositions according to the invention could also be used in internal calibration, for example to adjust the acquisition parameters of mass spectrometers, according to the MALDI technique, and in particular according to the MALDI-TOF technique used, or as quality control.

[0048] Within the scope of the invention, a method for calibrating a mass spectrometer adapted to the MALDI technique, and in particular to the MALDI-TOF technique, is also proposed, comprising the following steps: a) providing a calibration composition according to the invention, b) depositing said calibration composition and a matrix adapted to the MALDI technique, on at least one deposition zone of a MALDI analysis plate, to obtain a control zone after drying, c) placing said MALDI analysis plate in the mass spectrometer and obtaining a mass spectrum for the control zone according to the MALDI technique, and in particular according to the MALDI-TOF technique, over at least one determined mass-to-charge range and using the experimental values ​​obtained from the mass-to-charge peaks of said spectrum corresponding to so-called reference protein ions of the calibration composition, as calibration values,d) calculate a calibration relationship based on the theoretical masses on charge of the reference protein ions and the calibration values.,

[0049] Advantageously, the determined load mass range covers at least the range from 2000 Th to 20000 Th, preferably from 2000 Th to 15000 Th.

[0050] As a specific embodiment of the method according to the invention, the theoretical masses on charge used in step d) may be at least 6 in number, preferably 8 to 50 in number, and very particularly 8 to 32. According to a particular embodiment, when the strain of the calibration composition is an E. coli ATCC 8729 strain, the ions of the reference proteins and therefore the theoretical masses on charge in Th used are chosen from those listed in Table 2.

[0051] According to another particular embodiment, when the strain of the calibration composition is an E. coli BL21 strain, the ions of the reference proteins and therefore the theoretical masses on charge in Th used are chosen from those listed in Table 5.

[0052] According to specific embodiments, the theoretical masses on charge used in step d) are at least 6 in number, preferably 8 to 50 in number and are, in particular, chosen from the following theoretical masses on charge m / z in Th:

[0053] - when the E. coli strain is BL21: 3,128.18; 3,158.57; 3,179.59; 3,206.28;

[0054] 3,579.84; 3,637.70; 3,936.51; 4,163.59; 4,185.35; 4,365.31; 4,438.61; 4,449.60; 4,497.60;

[0055] 4,613.75; 4,768.45; 4,777.56; 5,069.74; 5,096.78; 5,150.50; 5,381.35; 5,461.10; 5,612.32;

[0056] 5,725.60; 6,255.36; 6,298.04; 6,316.14; 6,411.55; 6,508.45; 6,547.83; 6,857.81; 7,158.68;

[0057] 7,274.39; 7,707.56; 7,872.02; 8,326.18; 8,369.69; 8,876.22; 8,898.19; 8,994.20; 9,060.27;

[0058] 9,226.49; 9,535.89; 9,554.12; 9,741.83; 10,138.48; 10,300.00; 11,223.63; 11,450.19 and 12,227.19;

[0059] - when the E coli strain is ATCC 8739: 3128.18; 3158.57; 3206.28; 3579.84;

[0060] 3,637.70; 4,163.59; 4,185.35; 4,365.31; 4,438.61; 4,613.75; 4,768.45; 4,777.56; 5,069.74;

[0061] 5,096.78; 5,150.50; 5,381.35; 5,612.32; 6,255.36; 6,316.14; 6,411.55; 7,158.68; 7,274.39;

[0062] 8,369.69; 8,876.22; 8,994.20; 9,226.49; 9,535.89; 9,554.12; 10,138.48; 10,300.00;

[0063] 11,223.63 and 11,450.19.

[0064] Even more specifically, the theoretical masses on charge used in step d) are at least 6 in number, preferably 8 to 32 in number, and in particular 32 in number, and are, in particular, chosen from the following theoretical masses on charge m / z in Th, when the strain of E. coli is BL21: 3128.18; 3158.57; 3206.28; 3579.84; 3637.70; 4163.59; 4185.35; 4365.31; 4438.61; 4613.75; 4768.45; 4777.56; 5069.74; 5096.78; 5150.50; 5,381.35; 5,612.32; 6,255.36; 6,316.14; 6,411.55; 7,158.68; 7,274.39; 8,369.69; 8,876.22; 8,994.20; 9,226.49; 9,535.89; 9,554.12; 10,138.48; 10,300.00; 11,223.63 and 11,450.19.

[0065] The MALDI matrix may be any matrix suitable for the MALDI technique. In particular, such a matrix comprises one of the following compounds: 3,5-dimethoxy-4-hydroxycinnamic acid, α-cyano-4-hydroxycinnamic acid, ferulic acid, and 2,5-dihydroxybenzoic acid.Finally, according to another of its aspects, the invention relates to a method for characterizing a sample containing at least one microorganism from the acquisition of a mass spectrum, by the MALDI technique, and in particular by the MALDI-TOF technique, implementing the following steps: i) depositing said sample and a matrix adapted to the MALDI technique, on at least one deposition zone of an analysis plate, and carrying out drying, to obtain a characterization zone, ii) placing said analysis plate in a mass spectrometer adapted to the MALDI technique, and in particular to the MALDI-TOF technique and generating a mass spectrum for the characterization zone, iii) applying the calibration relationship calculated according to the calibration method of the invention for said mass spectrometer, and determining the mass on charge of the peaks in the generated mass spectrum.

[0066] In particular, the control analysis area and the characterization area belong to the same MALDI analysis plate. This has the advantage of making calibration operations simpler and faster by reducing the handling steps.

[0067] According to a specific embodiment, the characterization method according to the invention comprises an identification of a microorganism present in the sample, by comparison of the mass-on-charge values ​​of at least certain peaks of the generated mass spectrum obtained in step iii), with reference values ​​characteristic of the microorganism.

[0068] Other aims, characteristics and advantages of the invention will appear in view of the description which follows and the examples developed below, which refer to the appended figures and in which:

[0069] - Figure 1 represents a schematic top view of a MALDI plate, notably marketed by the applicant;

[0070] - Figure 2, in connection with Example 2, shows the evolution over time of the MALDI-TOF mass spectrum obtained with a VITEK® MS, for a composition of E. coli outside the invention; the bacteria being suspended whole in a commercial mixture of solvents (in this case, the HCC A matrix from bioMérieux; ref. 411071);

[0071] - Figure 3, in connection with Example 3, shows the spectra obtained with a VITEK® MS, for two calibration compositions according to the invention, at 70% v / v ethanol, after a storage period of five months at room temperature; these compositions having been prepared according to 2 different protocols (A: from E. coli ATCC 8739 cultivated on agar medium; B: from E. coli ATCC 8739 cultivated in culture broth); - Figure 4, in connection with Example 4, shows different MALDI-TOF spectra obtained with a VITEK® MS, corresponding to the analysis of the contents of a tube initially comprising 100 μL of a calibration composition according to the invention (E coli ATCC 8739 suspended in an aqueous liquid at 70% v / v of ethanol), after evaporation of 20%, 50% or 80% of the initial volume;

[0072] - Figure 5, in connection with Example 6, shows a MALDI-TOF mass spectrum obtained with a VITEK® MS, before calibration, for a calibration composition of Example 3 (E coli BL21 suspended in an aqueous liquid at 70% v / v ethanol);

[0073] - Figure 6, in connection with Example 7, are two graphical representations of the errors attributed to the peaks (m / z) chosen as references for the calibration, obtained for a calibration composition according to the invention (E coli BL21 suspended in an aqueous liquid at 70% v / v of ethanol), freshly prepared (T0), or after 9 or 12 months of storage at 2-8°C (T9, T12); the associated MS spectra having been acquired with a VITEK® MS before calibration (A) and after calibration (B);

[0074] - Figure 7 shows the first super calibration graph used in Example 8;

[0075] - Figure 8 shows the second final super calibration graph from the 49 theoretical masses selected from the 6,563 MALDI-TOF spectra obtained with a VITEK® MS, used in Example 8.

[0076] EXAMPLE 1: Preparation of a calibration composition according to the invention, deposition on a MALDI analysis plate, and calibration

[0077] A calibration composition according to the invention can be prepared by simply placing a selected quantity of whole E. coli bacteria of the selected strain, in a liquid based on alcohol(s) and / or acetonitrile. The calibration composition thus comprises a suspension of E. coli bacteria, preserved whole. Depending on the quantity of calibration composition to be prepared, the E. coli bacteria used can be those directly purchased or derived from bacteria purchased and cultured, using any suitable technique: on a Petri dish, in broth, etc. These culture techniques are well known to those skilled in the art. As demonstrated in the examples, the culture method has no influence on the stability of the calibration composition obtained.

[0078] As an example of a method for preparing a calibration composition according to the invention, E. coli bacteria are cultured in broth, in a fermenter, until the desired quantity of biomass is reached. The biomass is then pelleted by centrifugation and then diluted, in 70% ethanol solution, to obtain the desired concentration. To facilitate homogenization of the solution and correctly quantify the bacterial concentration, a mechanical dispersion operation may be useful. This can be carried out, for example, by creating a vortex or by dispersing the bacteria with an instrument such as an Ultra-Turrax from IKA (Staufen, Germany) equipped with a high-performance rotor-stator. A sonication step could also be carried out. In all cases, care must be taken to minimize alteration of the bacterial cells to avoid breaking the membrane walls and forming cellular debris.Despite the precautions taken, some bacteria may be altered and cellular debris may be found in the calibration compositions according to the invention. Once the solution has been homogenized and adjusted to the desired concentration, for example 5 McF, all that remains is to distribute it in a suitable container to facilitate its use.

[0079] A calibration composition according to the invention can be used directly or can be stored in any suitable container and / or can be marketed as a ready-to-use composition in any suitable container. Suitable containers include, in particular, a tube, for example made of plastic, closed by a cap, or any type of suitable hermetically sealed container. The container will have a volume selected to be able to contain the desired quantity of calibration composition, in particular from 10 to 300 μL and typically 200 μL. Such volumes make it possible to carry out a large number of deposits, which are typically deposits of 1 μL, and therefore a large number of calibrations.

[0080] The preliminary stability studies presented in the examples have shown that the calibration compositions according to the invention are stable, in particular for at least one year at 2-8°C, at least two months at room temperature (typically at 18-25°C), when stored in a closed container, and for at least one day at room temperature, when the container is opened. The calibration compositions according to the invention are therefore stable over time and make it possible to obtain good calibration.

[0081] When used, a calibration composition according to the invention is deposited at the control zone(s) of a MALDI plate. This deposition can be carried out using a pipetting technique, or possibly a spraying technique. A drop, for example of approximately 1 to 2 μL of the calibration composition can be deposited, so as to cover the entire deposition zone. It is easy to form a thin, uniform layer of calibration composition on the plate, due to the liquid formulation.

[0082] The use of a calibration composition according to the invention does not require any particular expertise or skill. It is compatible with manual and automated preparation (for example using the Colibri® technique proposed by the applicant) of MALDI plates, and in particular of the VITEK® MS-DS type. The operation consisting of depositing a calibration composition according to the invention and a matrix adapted to the MALDI technique, on the MALDI plate can be carried out sequentially or by depositing a mixture of said calibration composition and the matrix.

[0083] Advantageously, it will be operated as follows: deposition of the calibration composition, deposition of the MALDI matrix, then drying. The calibration composition according to the invention can be simply deposited by pipetting 1 μL onto a control area of ​​the MALDI plate, followed by the addition of 1 μL of MALDI matrix, in particular HCC A matrix, and drying. The matrices used in the MALDI technique are generally organic molecules which have specific characteristics. They must be capable of absorbing the energy of the laser used for irradiation, which results in the vaporization of the matrix and the formation of ions. In addition, the matrices must be compatible with the samples analyzed and not interfere with the ions of interest. Generally, the matrices used in the MALDI technique are photosensitive and crystallize in the presence of the population of microorganism(s), while preserving the integrity of the molecules present.Such matrices, particularly suitable for the MALDI-TOF mass spectrometry technique, are well known. They are made, for example, from a compound chosen from: 3,5-dimethoxy-4-hydroxycinnamic acid, α-cyano-4-hydroxycinnamic acid (present in the HCCA matrix), ferulic acid and 2,5-dihydroxybenzoic acid. Many other compounds are known to those skilled in the art. There are also liquid matrices that do not crystallize at atmospheric pressure, or even under reduced pressure. Any other compound that will ionize the molecules present in the characterization zone under the effect of a laser beam can be used. Each matrix has specific properties that make it suitable for different types of samples and analyses.

[0084] To constitute the matrix, such a compound is dissolved, most often in water, preferably of “ultra-pure” quality, or in a mixture of water and organic solvent(s). Examples of organic solvents conventionally used include acetone, acetonitrile, methanol or ethanol. The matrix may contain trifluoroacetic acid which facilitates ionization. The matrix is ​​directly deposited on the control zone and covers or is mixed with the calibration composition.

[0085] The drying of the calibration composition deposited, before or after deposition of the matrix, on the MALDI plate is carried out at a temperature and for a time appropriate to ensure the formation of a thin uniform layer of the latter. Mild conditions well known to those skilled in the art are preferred. The drying of the calibration composition, alone and / or with the addition of a matrix, can be carried out by evaporation by exposing the assembly to ambient air for a few minutes. This evaporation also ensures the crystallization of the matrix present.

[0086] Advantageously, the amount of E. coli bacteria of the calibration composition deposited per control area is generally at least 10 5 cfu, and in particular 6.10 5 at 42.10 5 cfu.

[0087] Once the MALDI plate is prepared with a layer of calibration composition, in the form of a control zone as described previously, it is placed in a mass spectrometer, notably MALDI or MALDI-TOF, and a corresponding mass spectrum is acquired. This involves irradiating the MALDI plate with a MALDI laser, which results in the desorption and ionization of the molecules present in the irradiated spot. The desorbed and ionized proteins are then detected to generate the mass spectrum. The acquisition of the mass spectrum is therefore carried out by irradiating the MALDI plate with a MALDI laser and detecting the desorbed and ionized proteins.

[0088] In particular, any type of MALDI-TOF mass spectrometer can be used for the development of the mass spectrum. Such devices include: i) an ionization source (usually a UV laser) intended to ionize the control zone and therefore the calibration composition and / or a sample present on a characterization zone; ii) an accelerator of the ionized molecules by application of a potential difference; iii) a tube under reduced pressure in which the ionized and accelerated molecules move; iv) a mass analyzer intended to separate the molecular ions formed, according to their mass on charge (m / z); v) a detector intended to measure the signal produced directly by the molecular ions.

[0089] The laser beam used for ionization can have any type of wavelength favorable to the sublimation or vaporization of the matrix. Preferably, an ultraviolet or even infrared wavelength will be used. This ionization can, for example, be carried out with a nitrogen laser emitting a UV ray at 337 nm (for example used in the VITEK® MS), or with a neodymium-doped yttrium lithium fluoride (Nd:YLF) laser emitting at 349 nm (for example used in the VITEK® MS PRIME).

[0090] The time of flight taken by the ions to reach the detector is used to calculate their mass. Thus, a mass spectrum is obtained, representing the intensity of the signal corresponding to the number of ionized molecules of the same mass to charge [m / z], as a function of the m / z ratio of the molecules hitting the detector. The mass to charge ratio [m / z] is expressed in Thomson (Th).

[0091] The mass spectrum is generated in particular by detecting at least a portion of the ionized and accelerated molecules then allowed to move freely in a tube under reduced pressure, so as to measure the time they take to travel through the tube under reduced pressure and to obtain a signal corresponding to the number of ionized molecules reaching the spectrometer detector at a given time. An uncalibrated spectrum, with the flight time of the ions on the abscissa and the intensity of the observed signal on the ordinate, is thus obtained. It is transformed into a mass spectrum by calculating the mass-to-charge ratios (m / z) corresponding to the flight times of the detected molecules. To do this, a flight time calibration equation is established using reference molecules. This results in a calibrated mass spectrum corresponding to the signal of the number of ionized molecules of the same mass-to-charge m / z, as a function of the m / z ratio of the detected molecules.

[0092] During the calibration step, the peaks corresponding to the so-called reference molecules, in ionized form (called ion of a reference molecule) of the calibration composition, called reference peaks, are identified and their theoretical m / z masses will be used as calibration values. This identification can be carried out by comparing the measured, so-called experimental, flight times of the peaks of the uncalibrated spectrum with the theoretical masses of said ions of the reference proteins, or with the measured, so-called experimental, m / z masses of the mass spectrum if a previous calibration is already available in the instrument and applied. Once these calibration values ​​are identified, a calibration relationship and an error are calculated. Different linear or non-linear regression methods can be used to calculate such a calibration relationship.For example, the least squares method or any method well known to the person skilled in the art can be used to perform a calibration. In general, mass spectrometers are equipped with suitable software to implement such methods. The calculation of an error in parts per million (ppm) is carried out for each reference charge mass value (or reference protein ion), considering the corresponding experimental value and theoretical value. To do this, the following formula can be: (theoretical mass - experimental mass) / theoretical mass * 10. 6. For all the ions of the selected reference proteins, it is then possible to calculate an average error which corresponds to the average value of the errors expressed in absolute values. The lower the average error, the better the calibration. The calibration relationship thus established is then applied to determine the masses on charge of the peaks of protein or molecule ions present in the analyzed samples, on characterization zones of the MALDI plate. These corrected masses, also called calibrated, are obtained in step iii) of the sample characterization procedure, as previously defined.

[0093] In summary, the calibration compositions and the method for calibrating a mass spectrometer according to the invention make it possible to obtain precise measurements of the masses of the ions analyzed, and therefore a reliable identification of microorganisms. The calibration compositions and the calibration method according to the invention are simple to implement and can be used in various analysis applications, and in particular for the identification of microorganisms, by mass spectrometry using the MALDI technique, and in particular MALDI-TOF. In the context of the invention, the MALDI-TOF analysis can be a simple MALDI-TOF analysis, or a MALDI-TOF-TOF analysis.

[0094] The methods according to the invention (calibration method and method for characterizing a sample) are computerized or automated methods. In the methods according to the invention, the steps of acquisition, calculation, comparison, determination, etc. are carried out by a computer or an electronic device programmed to execute said steps, integrated into the spectrometer used. Specific software may be developed and installed on the computer or electronic device that will be used to execute the calibration method and, in particular, to obtain the calibration relationship. In particular, the software is programmed to support the required functionalities, such as the acquisition of mass spectrometry data, the calculation of the masses on charge of the mass peaks of the mass spectrum corresponding to the ions of unknown proteins or molecules.

[0095] For the identification on the mass spectrum of the calibration composition of peaks corresponding to the calibration values, the calibration software can be programmed to analyze the mass spectrometry data and identify the peaks corresponding to the calibration values. This step may involve adaptive peak detection, mass-to-charge comparison, and thresholding algorithms to accurately identify relevant peaks.

[0096] For the calculation of the calibration relationship, once the peaks corresponding to the calibration values ​​have been identified, the calibration software performs a calculation to determine the calibration relationship. This calculation can use different statistical methods, such as linear or non-linear regression, to establish a mathematical equation obtained from the theoretical masses on charge of the ions of the reference proteins and the masses on charge measured for the peaks corresponding to said ionized reference proteins, used as calibration values.

[0097] For the application of the calibration relationship, once the latter has been determined, the software applies it to convert the measured flight times or masses on charge (also called experimental) of the peaks corresponding to an unknown entity on the mass spectra obtained for the samples to be analyzed, into calibrated masses on charge which are then reported on the generated mass spectrum. This step makes it possible to obtain precise measurements of the masses on charge of the ions present in the analyzed samples.

[0098] The samples analyzed may be of different origins. For example, samples of biological origin, in particular animal or human, may be cited. Such a sample may correspond to a sample of biological fluid, such as whole blood, serum, plasma, urine, cerebrospinal fluid, organic secretion, a tissue sample or isolated cells. This sample may be deposited as is, or will preferably be subjected, prior to depositing on the MALDI plate, to a preparation such as enrichment or culture, to a concentration and / or to an extraction or purification step according to methods known to the person skilled in the art. The sample may also be a food product such as meat, milk, yogurt and any other consumable product likely to be contaminated, or even a cosmetic or pharmaceutical product.Here again, such a product may be subjected to an enrichment or culture type preparation, a concentration and / or an extraction or purification step, before being deposited on the MALDI plate. Most often, the sample to be analyzed will come from a culture in a broth or on an agar in order to enrich it with microorganisms to be searched for. It is, for example, possible to proceed directly to the deposit of a biomass, a drop of a suspension of microorganisms, in ultra-pure water or a buffer.

[0099] The following examples, with reference to certain of the appended Figures, illustrate the invention, but are not limiting in nature.

[0100] In the examples below, optical density is measured with the DENSIMAT densitometer (bioMérieux, France) following the recommendations of the user manual (reference 99535, version C, published in 2007).

[0101] EXAMPLE 2: Evaluation of the stability of an E. coli suspension in a commercial solvent mixture (bioMérieux; ref. 411071).

[0102] Commercially available E. coli strain ATCC 8739, purchased from ATCC, was added to a concentration of 5 McF (corresponding to an optical density at 550 nm of 1.0-1.1) in an α-cyano-4-hydroxycinnamic acid matrix (HCCA matrix) composed of a mixture of 333 ml of ethanol, 333 ml of acetonitrile, 333 ml of water, 30 ml of trifluoroacetic acid, and 31 g of HCCA matrix (bioMérieux; ref. 411071). HCCA matrix is ​​commonly used on MALDI-TOF mass spectrometers for the measurement of peptides and proteins.

[0103] The stability of the suspension was then analyzed with a MALDI-TOF mass spectrometer (VITEK® MS, bioMérieux), following the evolution of the calibration peaks between T0 and T7 days (7 days after T0). A progressive modification of the peak shape was observed on all the spectra.

[0104] For example, the doubly charged peak characteristic of the DNA-binding protein HU-alpha was clearly visible at T0 (m / z 4767). However, a third peak appeared at T24h (1 day after T0), a fourth at T48h (2 days after T0) and at T7d (7 days after T0), 8 peaks were detected (Figure 2).

[0105] Degradation was thus observed over time and given the mixture of solvents used (ethanol, acetonitrile and trifluoroacetic acid), the degradation being of chemical rather than biological origin. Considering that the observed delta of the masses over charge of the ionized proteins between the different peaks over time was approximately 28 Da and that the presence of ethanol and trifluoroacetic acid can protonate the proteins, the inventors believe that it is likely that the degradation corresponds to ethylation.

[0106] Therefore, the HCCA matrix solvent mixture commonly implemented in MALDI-TOF instruments does not provide sufficient stability of the calibration peaks of an E. coli strain over time and therefore cannot be used as a solvent in the calibration solution.

[0107] EXAMPLE 3: Absence of impact of the strain culture method on the performance of a calibration composition according to the invention

[0108] Two calibration compositions according to the invention were prepared from the E. coli ATCC 8739 strain used in Example 2, but cultured according to two different methods.

[0109] A. Protocol for preparing a calibration composition from the E. coli ATCC 8739 strain obtained by growth on a Petri dish in solid agar medium:

[0110] - culture of the strain on Petri dishes in solid agar medium (bioMérieux; ref. 43039) at 37°C for 18 to 24 hours, - dilution of the strain to 5 McF (i.e. to obtain an optical density at 550 nm of 1.0-1.1) in 70% (v / v) ethanol (i.e. 70% by volume of ethanol and 30% by volume of water), so as to obtain a calibration composition,

[0111] - filling tubes (50 μL, 100 μL and 200 μL) with said composition,

[0112] - storage of tubes: room temperature (18-24°C) and 2-8°C.

[0113] When using the calibration composition, 1 μL of said composition and 1 μL of HCCA matrix are deposited on the MALDI plate.

[0114] B. Protocol for preparing a calibration composition from the same E. coli ATCC 8739 strain obtained by broth culture:

[0115] - culture of the strain in culture broth based on soy trypticase (bioMérieux; ref. 42100) at 37°C for 18 to 24 hours,

[0116] - centrifugation for 10 minutes at 4500 rpm in order to concentrate the cells in a pellet,

[0117] - double rinsing of the base with water,

[0118] - resuspension of the pellet in a 70% ethanol solution and dilution to obtain a calibration composition according to the invention containing E. coli bacteria concentrated at 5 McF (which is equivalent to an optical density at 550 nm of 1.0-1.1) in 70% ethanol,

[0119] - filling tubes (50 μL, 100 μL and 200 μL) with said composition,

[0120] - storage of tubes: at room temperature (18-24°C) or at 2-8°C.

[0121] When using the calibration composition, 1 μL of said composition and 1 μL of HCCA matrix are deposited on the MALDI plate.

[0122] The calibration compositions prepared according to the two previous protocols were then used to calibrate a MALDI-TOF mass spectrometer (VITEK® MS, bioMérieux). Both compositions allowed to obtain a very good calibration of the mass spectrometer, because the reference peaks (including in particular, the calibration values ​​used during the calibration) were still present and not degraded (data not shown) thus demonstrating that the strain production method (culture on Petri dish or culture broth) does not affect the quality of the calibration.

[0123] Furthermore, after 5 months of storage at room temperature, no degradation was observed on the peaks of the different spectra (Figure 3). In other words, the peaks observed in Figure 3, present the expected masses for the strain ATCC 8739, as they appear in Table 2. Unlike the formulation of Example 2, no degradation is observed, the masses of the peaks remain constant, to the nearest analytical error. This analytical error is inherent to the MALDI-TOF technology. The results obtained with storage between 2-8°C were also similar.

[0124] Therefore, the method of culturing the E. coli strains used to obtain the calibration composition does not impact either the quality of the calibration on the MALDI-TOF mass spectrometer, or even the stability of the said composition over time.

[0125] EXAMPLE 4: Evaluation of the impact of evaporation on the quality of a calibration composition according to the invention

[0126] Calibration composition used: A. coli ATCC 8739 strain suspended at a concentration of 5 McF (i.e. to obtain an optical density at 550 nm of 1.0-1.1) in a 70% ethanol solution, and stored in 50 μL, 100 μL, 200 μL and 300 μL tubes.

[0127] The tubes were left open at room temperature (24°C±1°C) until 20, 50 or 80% of the initial volume had evaporated. The different partially evaporated calibration compositions were then used as calibration solution on a MALDTI-TOF mass spectrometer (VITEK® MS, bioMérieux).

[0128] After analysis, all tested compositions allowed to obtain a very good calibration of the mass spectrometer independently of the initial volume and the level of solvent evaporation. The different MALDI-TOF spectra for the 100 μL tube with 20%, 50% or 80% evaporation are presented in Figure 4. Apart from the analytical error inherent in MALDI-TOF technology, the spectra are very similar: the peaks described in Table 2 are observed with their expected mass; they therefore allow to calibrate the spectrum.

[0129] EXAMPLE 5: Preparation of a calibration composition according to the invention of E. coli BL21 in 70% ethanol

[0130] In a similar manner to Example 3 and implementing preparation protocol A, a calibration composition according to the invention was prepared with the E. coli BL21 strain (E. coli BL21 suspended in 70% v / v ethanol).

[0131] As demonstrated in Example 6 below, this solution provided excellent calibration results on a MALDI-TOF mass spectrometer.

[0132] EXAMPLE 6: Calibration of the MALDI-TOF instrument using a calibration composition according to the invention Generally speaking, a MALDI-TOF mass spectrometer measures the time of flight of ions in a vacuum, after their acceleration by an electric field. The time of flight is related to the mass of the ions according to the law of kinetic energy:

[0133] E c =½ mV 2 with E c kinetic energy, m mass and V velocity of the ions.

[0134] It is therefore possible to determine the mass of the ions from the measurement of their flight time by establishing the relationship linking these two physical quantities, generally by means of a quadratic equation, that is to say a polynomial of degrees 2.

[0135] In some cases, a higher degree polynomial can be used to account for physical and electronic imperfections of the mass spectrometer. In practice, the equation is calibrated using molecules of known masses to account for experimental variations inherent to the instrument, the MALDI plate on which the sample was deposited or the acquisition conditions (temperature, pressure in the vacuum tube, etc.).

[0136] The instrument can be delivered by the manufacturer with a pre-recorded factory calibration. This allows the user to acquire and analyze the first spectra directly using Thomson (Th) values. The Thomson is a unit corresponding to the mass divided by the charge of the ion. Subsequently, the instrument can be regularly recalibrated using Th spectra without ever returning to time-of-flight values. This allows the user to get used to adjusting the instrument and viewing spectra using masses and not times-of-flight.

[0137] The quality of the calibration can easily be assessed by calculating the average of the errors, for the masses on charge of the ions of the reference proteins. As a reminder, this average of the errors is calculated using the absolute values ​​of said errors.

[0138] As an example, a MALDI-TOF mass spectrum obtained with a VITEK® MS spectrometer from bioMérieux was calibrated using the calibration composition prepared according to Example 5.

[0139] The instrument acquisition range was set between 2,000 and 20,000 Th, a common mass-to-charge range for identifying microorganisms in microbiology laboratories. The resulting calibration composition spectrum showing the experimental m / z values ​​without calibration is shown in Figure 5.

[0140] This spectrum was then calibrated using a set of mass-to-charge ratios of protein ions present in E. coli extracted from Table 2. In Table 2, in the column "ion charge", 5 denotes an ion carrying one charge, d denotes an ion carrying two charges. The proteins present post-translational modifications compared to the protein whose name and sequence are given. In the column "post-translational modifications", c corresponds to the cleavage of the methionine at the N-terminus and m corresponds, after cleavage of the methionine at the N-terminus, to a methylation of the 2 eme amino acid (then terminal). The name of the reference protein ion is given in such a way as to combine the name of the protein corresponding to the NCBI sequence, its charge and the post-translational modification(s).

[0141] Table 2: 32 selected theoretical m / z for E. coli ATCC 8739

[0142] The reference protein ions are searched with a tolerance of 800 ppm to calibrate the spectrum by the least squares method using a quadratic equation. The average error before calibration is 372.69 ppm. This average value is then greatly reduced to 68.50 ppm after the first calibration, thus demonstrating that the masses on charge recalibrated by the composition according to the invention (columns "After 1st calibration", Table 3) are more accurate than before calibration.

[0143] To further improve the calibration, it is even possible to perform a second calibration after reducing the tolerance to 300 ppm. This operation eliminates some erroneous associations that may slightly distort the first calibration. These erroneous associations correspond to peaks mistakenly assigned to a given theoretical mass, taking into account a significant error tolerance (800 ppm). This second calibration slightly improves the measurement accuracy by only considering peaks with an error of less than 300 ppm. The average error is then equal to 63.18 ppm.

[0144] Interestingly, it is not necessary to find all the masses on charge sought on the mass spectrum. Thus, peak 28 is not observed in the spectrum before recalibration. After the first recalibration, peaks 23 and 27 become out of tolerance and are no longer used for the 2 emecalibration. It is therefore not necessary for all peaks to be present to ensure good calibration. The method is robust to the absence of one or more peaks. This property is particularly important for calibrating a MALDI-TOF spectrum. MALDI-TOF technology has stochastic behavior that can randomly cause certain peaks to disappear or shift in mass. For a given sample, some peaks may be visible or invisible from one acquisition to another depending on the crystals sublimated by the laser beam. Some peaks may also have a more or less noisy shape, which can lead to a more or less significant error in the detected mass.

[0145] Table 3: m / z values ​​recorded for a calibration composition according to the invention (E. coli

[0146] BL21; ethanol 70% v / v), before and after calibration, using the 32 theoretical m / z of E. coli ATCC 8739 shown in Table 2.

[0147] EXAMPLE 7: Stability study of a calibration composition according to Example 5

[0148] The possibility of calibration using the calibration composition according to Example 5, which is stable over time, was established by verifying that it was possible to calibrate a MALDI-TOF mass spectrometer of the VITEK® MS type with a batch of this calibration composition freshly prepared according to Example 5 (condition T0), then with the same batch of calibration composition stored for 9 months (condition T9) or 12 months (condition T12) at 2-8°C.

[0149] To simulate intensive use conditions, the batches stored at 2-8°C for 9 or 12 months were then left for 1h30 with the cap open to allow part of the solvent to evaporate before being then closed for 180h at room temperature (18-25°C). The masses on charge before and after calibration obtained with the list of 32 reference peaks are shown in Tables 4, 4bis and 4ter below. The measurement errors of the calibration peaks have been reported in Figure 6 (A: before calibration; B: after calibration).

[0150] Excellent calibration is obtained at TO, T9 and T12, characterized by a significant reduction in the average error, which goes from 157.69 to 17.64 ppm, from 186.45 to 19.45 ppm and from 287.29 to 29.01 ppm respectively.

[0151] These results demonstrate excellent calibration with a liquid solution from T0 to 12 months despite the simulation of intensive use with an evaporation time of 1h30 and a duration of 7.5 days at room temperature. Interestingly, the improvement in calibration quality is obtained even if the presence of some peaks / values ​​(noted NA) is not observed.

[0152] Table 4: m / z values ​​recorded for a calibration composition according to the invention (E. coli BL21; ethanol 70% v / v), freshly prepared, before and after calibration with reference to the 32 theoretical m / z of E. coli ATCC 8739.

[0153] Table 4bis: m / z values ​​recorded for a calibration composition according to the invention (E. coli BL21; ethanol 70% v / v), stored for 9 months at 2-8°C, before and after calibration

[0154] Table 4ter: m / z values ​​recorded for a calibration composition according to the invention (E. coli BL21; ethanol 70% v / v), stored for 12 months at 2-8°C, before and after calibration

[0155] EXAMPLE 8: Selection of calibration values ​​depending on the strain used As demonstrated in Example 6, the calibration suspension prepared with the BL21 strain of E. coli allows for improved calibration when acquiring MALDI-TOF spectra. However, the quality of the calibration depends on the reference proteins selected to perform the calibration.

[0156] The purpose of this example is to present the methodology for selecting, for a given strain (in this case, the E. coli BL21 strain), the reference proteins whose theoretical charge masses are known and which will be used for calibration. This methodology is applicable to any other strain of E. coli.

[0157] Generally speaking, the interpretation of a MALDI-TOF spectrum consists of searching for a list of theoretical masses on charge expected among the experimental values, that is to say the masses on charge measured and obtained from the mass spectrum before calibration. This association is carried out under a certain threshold of precision, a tolerance or an error noted in "ppm" for "parts per million".

[0158] A) Selection of theoretical masses on charge for the E. coli BL21 strain A selection of theoretical masses on charge was carried out over the m / z range from 3000 to 12500 Th (trivially fixed in relation to the capabilities of the mass spectrometer) with the aim of selecting proteins and associated ions so that the corresponding theoretical masses on charge are distributed uniformly over said range.

[0159] Table 5 below shows the theoretical masses on charge (m / z) of a number of pre-selected proteins as well as the corresponding names, according to the same convention as Table 2. Not encountered in this Table 2, post-translational modifications can also be of the formylation type, or acetylation of 2 eme amino acid after cleavage of methionine at the N-terminus, these modifications are noted f or a.

[0160] Table 5: 49 theoretical m / z of E. coli BL21

[0161] B) Experimental MALDI-TOF spectra

[0162] In its databases, the applicant has 14,332 MALDI- spectra

[0163] Experimental TOFs of E. coli BL 21 strain, acquired over time (6,614 were obtained with the VITEK® MS mass spectrometer and 7,718 with the VITEK® MS PRIME mass spectrometer).

[0164] Among these spectra, 52 of them admit less than half of the theoretical masses on charge sought and are discarded to finally keep 14,280 experimental MALDI-TOF spectra.

[0165] C) Validation of the theoretical load masses used for calibration

[0166] Cl. First validation of theoretical load masses

[0167] Before exploiting the interpretation results, the quality of the calibration was checked. Indeed, a poor calibration can distort the interpretation of the experimental peaks, particularly through irregular precision.

[0168] For example, it is possible to have "offset" theoretical masses on charge, systematically adding a bias to the calibration, or "unstable" theoretical masses on charge, blurring the overall accuracy of the calibration. In practice, both situations occur, hence the need to exclude theoretical masses that can be disruptive to the calibration.

[0169] For this purpose, only a portion of the spectra was used. The association between a sample of 7,724 MALDI-TOF spectra and the 49 selected theoretical masses resulted in 295,829 interpretations, or approximately 38 interpretations per spectrum, i.e. 38 observed theoretical masses on charge, called experimental, out of the 49 sought.

[0170] To get an overview of the calibration, the overall interpretation of each mass on theoretical load was displayed on the same graph, called super calibration graph. The median of the interpretation error of each mass on theoretical load was displayed with a variance interval corresponding to F standard deviation of the error, as shown in Figure 7.

[0171] From the super calibration graph in Figure 7, three criteria were then used to exclude the disturbing theoretical masses on charge for the calibration:

[0172] - Visibility of theoretical load masses: some theoretical load masses are observed very infrequently and are therefore excluded for calibration. This concerns four theoretical load masses.

[0173] - Mass-on-charge shift: some theoretical masses on charge seem abnormally shifted, so a threshold of 100 ppm is applied to the median error. This criterion concerns seven theoretical masses on charge.

[0174] - Mass stability: some theoretical masses on charge have a higher standard deviation and variance interval. Thus, a threshold of 100 ppm is applied to the standard deviation in order to be able to exclude theoretical masses on charge with an error standard deviation greater than 100 ppm. This criterion concerns six theoretical masses on charge.

[0175] By combining the three criteria, twelve distinct theoretical masses on load were discarded for calibration and are listed in Table 6 below.

[0176] Table 6: 12 m / z of a calibration composition according to the invention (E. coli BL21, ethanol 70% v / v), excluded from the calibration process

[0177] A Tissue of the first validation, 37 theoretical masses on charge were preserved.

[0178] C2. Second validation of theoretical load masses

[0179] The analysis was then carried out on a sample of 7,650 spectra to obtain the super calibration graph representing for each mass on theoretical charge the median of interpretation error with a variance interval corresponding to T standard deviation of Terror.

[0180] Then, a threshold of 40 ppm was applied to the median of Terror. The application of this threshold made it possible to discard 5 theoretical masses on charge for the calibration which are listed in Table 7 below.

[0181] Table 7: 5 m / z of a calibration composition according to the invention (E. coli BL21, ethanol 70% v / v), also excluded from the calibration process

[0182] In the second validation, 32 theoretical masses on charge are retained for calibration and the 17 not retained are still sought to improve control. This is summarized in the following section.

[0183] D) Final method of interpretation

[0184] 6,563 spectra acquired with the VITEK® MS spectrometer were studied. Before interpretation, the spectra were calibrated using 32 preselected theoretical masses. The calibration consisted of two successive quadratic regressions at tolerance thresholds of 500 ppm and 300 ppm respectively using the 32 selected theoretical calibration masses.

[0185] Finally, a total of 49 masses on charge (theoretical calibration and additional) were sought with a tolerance threshold of 300 ppm.

[0186] E) Results

[0187] As mentioned earlier, to understand the calibration from a general point of view, the overall interpretation of each mass on theoretical load was displayed on the same graph. The median of the interpretation error of each mass on theoretical load was displayed with a confidence interval corresponding to the standard deviation of the error, as shown in Figure 8.

[0188] Then, the interpretation performances were calculated on all 32 masses on load, used as calibration values ​​and the 17 additional masses on load. The results are presented in Table 8, below.

[0189] Table 8: Calibration performance according to the invention

[0190] It is clear from Table 8 that, firstly, the 32 masses on charge used corresponding to the calibration values ​​are very frequently visible, with an average visibility rate of 95% across all spectra.

[0191] Secondly, the absolute measurement error was also rectified by the calibration with a median decreasing from 119.48 ppm to 19.41 ppm, a mean decreasing from 152.48 ppm to 25.23 ppm and a standard deviation decreasing from 126.39 ppm to 23.50 ppm. In other words, the method of validating the masses on charge corresponding to the calibration values ​​made it possible to provide interpretations 6 times more accurate with a measurement randomness 4 times smaller. The method of selecting the theoretical reference masses on charge thus made it possible to significantly improve the calibration performance. Furthermore, the mass spectrometer thus calibrated could then be used successfully for the identification of different microorganisms.

[0192] EXAMPLE 9: Stability and performance of calibration compositions according to the invention

[0193] Table 9 below provides, for information purposes, the results of a stability test over time carried out on different calibration compositions according to the invention.

[0194] The tested compositions were prepared with E. coli BL 21 bacteria, suspended in a pure alcohol solution, or aqueous solutions of alcohol or acetonitrile, in this case:

[0195] - 100% ethanol solution (v / v),

[0196] - a 20% (v / v) ethanol solution,

[0197] - a 50% (v / v) ethanol solution,

[0198] - a 70% (v / v) ethanol solution,

[0199] - a 70% (v / v) isopropanol solution, and

[0200] - a 70% (v / v) acetonitrile solution.

[0201] The stability of these calibration compositions was evaluated according to the calibration performance obtained. This is expressed as the average of the errors associated with the 32 calibration masses selected in Example 8.

[0202] Measurements were performed with “freshly” prepared calibration compositions (T0) or after 8 days of storage at room temperature (T8).

[0203] Table 9: Stability of calibration compositions, in terms of calibration performance (expressed as average errors).

[0204] The six calibration compositions, the calibration performance of which is reported in Table 9, therefore make it possible to reduce the average errors by an average factor of approximately 6 at T0 and approximately 10 at T8, which demonstrates excellent calibration according to the invention.

[0205] For comparison, calibrations carried out with freshly cultured E. coli ATCC 8739 bacteria, as prescribed in the VITEK® MS user manual, obtain average errors of the order of 30 ppm after calibration (and according to recent measurements carried out under the same conditions as example 8: 109.94 ppm before calibration and 26.9 ppm after calibration).

[0206] It should also be noted that a preparation of E. coli BL 21 bacteria, suspended in a 70% formic acid solution, led to mass spectra with too few peaks to allow calibration.

Claims

Claims 1. Calibration composition for mass spectrometer comprising whole E. coli bacteria, suspended in a liquid based on alcohol(s) and / or acetonitrile [sanitizing, volatile and non-denaturing / non-reactive with respect to bacterial proteins].

2. Calibration composition according to claim 1, characterized in that said liquid is based on alcohol(s).

3. Calibration composition according to claim 2, characterized in that said liquid is an aqueous liquid having an alcohol content of at least 20% v / v.

4. Calibration composition according to claim 2 or 3, wherein said alcohol(s) is / are ethanol and / or isopropanol.

5. Calibration composition according to claim 3 or 4, wherein said liquid is a water / alcohol mixture, preferably a water / ethanol mixture, having an alcohol content of 70% v / v.

6. Calibration composition according to claim 1, wherein said liquid is an aqueous water / acetonitrile mixture.

7. Composition according to any one of the preceding claims, in which the E. coli bacteria are of a single strain.

8. A calibration composition according to any preceding claim, wherein the E. coli bacteria are of strain BL21 or strain ATCC 8739.

9. A calibration composition according to any preceding claim, wherein the E. coli bacteria are present at a concentration of 6.10 8 at 9:10 p.m. 8 cfu / mL of the calibration composition.

10. Method for calibrating a mass spectrometer adapted to the MALDI technique, comprising the following steps: a) providing a calibration composition according to any one of claims 1 to 9, b) depositing said calibration composition and a matrix adapted to the MALDI technique, on at least one deposition zone of a MALDI analysis plate, to obtain a control zone after drying, c) placing said MALDI analysis plate in the mass spectrometer and obtaining a mass spectrum for the control zone according to the MALDI technique, and in particular according to the MALDI-TOF technique, over at least one determined mass-to-charge range and using the experimental values ​​obtained from the mass-to-charge peaks of said spectrum corresponding to so-called reference protein ions of the calibration composition, as calibration values, d) calculating a calibration relationship based on the theoretical mass-to-charge of the reference protein ions and the calibration values.

Images