Method for determining the sensitivity of a bacterial strain to a bacteriophage virus

Description

DOMAINE TECHNIQUE

[0001] The technical field of the invention is the characterization of the ability of a bacterial strain to be infected, then lysed, by a bacteriophage virus. ART ANTERIEUR

[0002] Phage therapy is the use of bacteriophage viruses, commonly referred to as phages, to treat bacterial infections. It involves the use of lytic phages, which are capable of infecting certain bacteria and then lysing them. This is a relatively old treatment, discovered and practiced before the emergence of antibiotics. However, in most countries, this treatment has been somewhat neglected, in favor of the use of antibiotics. Currently, phage therapy is considered a promising therapeutic approach for treating certain bacterial infections, particularly when bacteria are resistant to antibiotics.

[0003] The effectiveness of phage therapy has been demonstrated for the treatment of infections caused by bacteria of the type Pseudomonas aeruginosa, Escherichia coli, Or Staphilococcus Aureus, in skin grafting applications for severe burns or pathologies related to diabetes.

[0004] The phages used in phage therapy applications are lytic phages, so that when a bacterium has been infected by a phage, the latter is lysed. Lysis results from the production of endolysins. During an infection, the metabolism of the bacterium leads to a replication of the number of phages, in the form of virions. These are released following the lysis of the bacterium. An infection is therefore followed by an increase in the quantity of phages capable of infecting and then lysing other bacteria. This results in a reduction or even a halt in bacterial proliferation.

[0005] To test the sensitivity of a bacterium to a phage, tests, usually referred to as "phagograms", are carried out in the laboratory. This involves testing the ability of a bacterial strain to be infected, then lysed, by a phage.

[0006] Documents WO2019 / 0188886, EP2747578 describe observations of bacterial lysis induced by bacteriophages. The observations are carried out using optical methods, such as fluorescence or colorimetry. The same applies to the publications Henry M. et al. “Development of a high throughput assay for indirectly measuring phage growth using the Omnilog TM system” or Estrella Luis et al. “Characterization of novel Staphylococcus aureus lytic phage and defining their combinatorial virulence using the Omnilog system”.

[0007] Laboratory tests require testing different viral strains, with different phage-to-bacteria ratios. The number of viral strains can range from 5 to 100, while 5 to 10 phage-to-bacteria ratios are typically tested for each viral strain. Therefore, the number of tests required is high.

[0008] Document WO2013027146 describes the observation of a sample using a fluorescence method. WO2013027146 is based on the use of an exogenous agent, in this case a fluorescent marker, to characterize the behavior of bacteria in the presence of bacteriophages.

[0009] The inventors propose a simple method for analyzing the sensitivity of a bacterial strain to a viral strain. The method allows for monitoring bacterial development and early detection of bacterial lysis by phages. It can also simultaneously address several combinations between quantities of phages, possibly from different viral strains, and quantities of bacteria, possibly from different bacterial strains. The objective is to reduce the time spent detecting the sensitivity of a bacterial strain to a viral strain. EXPOSE DE L'INVENTION

[0010] An object of the invention is a method for determining the sensitivity of a bacterial strain to a viral strain of bacteriophages, according to claim 1.

[0011] The spectral emission bandwidth is preferably less than 100 nm or 50 nm. The distance between the sample and the image sensor is preferably less than 5 cm or less than 1 cm.

[0012] Step d) may involve determining the attenuation of light emitted by the light source by the sample, such that the bacterial strain is considered to be: slightly or not at all sensitive to the viral strain of bacteriophages when the attenuation increases between two successive measurement times; or sensitive to the viral strain of bacteriophages when the attenuation does not increase between two successive measurement times.

[0013] According to one embodiment: during step a), the bacteria and the bacteriophages are mixed in an aqueous solution; step c) comprises an acquisition of at least two images, at successive measurement times; step d) comprises a determination of a texturing descriptor of each acquired image, such that the bacterial strain is considered to be: not or only slightly sensitive to the viral strain of the bacteriophages when the evolution of the texturing descriptor reflects an increase in light scattering by the sample between two successive measurement times; or sensitive to the viral strain of the bacteriophages when the evolution of the texturing descriptor reflects a decrease or stagnation in light scattering by the sample between two successive measurement times.

[0014] The texturing descriptor can be determined based on one or more regions of interest in the image.

[0015] According to one embodiment: in step a), the bacteria and the bacteriophages are mixed in an agar; step c) comprises an acquisition of at least one image, after step a); step d), comprises an analysis of each acquired image, so as to identify clear regions of interest, each clear region of interest corresponding to an infection of bacteria by bacteriophages, forming a lysis range, each clear region of interest indicating a sensitivity of the bacterial strain with respect to the viral strain of the bacteriophages.

[0016] The process can then include: a count of the number of clear regions of interest on at least one acquired image; an estimate of an infectious titer of bacteriophages in the sample from the number of clear regions of interest counted.

[0017] According to one embodiment: In step a), the bacteria are placed in an agar plate, and the bacteriophages are placed in a solution, step a) comprising the placement of at least one drop of the solution onto the agar plate; step c) comprises the acquisition of at least one image, subsequent to step a); step d) comprises an analysis of each image acquired, so as to identify clear regions of interest, each clear region of interest corresponding to an infection of bacteria by bacteriophages, forming a lysis area, such that the appearance of each clear region of interest indicates a sensitivity of the bacterial strain to the viral strain of the bacteriophages.

[0018] Step a) may involve depositing several drops, the drops being spaced apart from each other, two different drops comprising respectively: bacteriophages of different viral strains; and / or different concentrations of bacteriophages of the same viral strain.

[0019] According to one embodiment: in step a), the bacteriophages are placed in an agar, and the bacteria are placed in a solution, step a) comprising depositing a drop of the solution on the agar; step c) comprises acquiring at least one image, after step a); step d) comprises analyzing each acquired image, so as to identify: light regions of interest, each light region of interest corresponding to an infection of bacteria by bacteriophages, such that each light region of interest indicates a sensitivity of the bacterial strain to the viral strain of the bacteriophages; or dark regions of interest, each dark region of interest corresponding to a development of bacteria in the presence of bacteriophages, such that each dark region of interest indicates an insensitivity or a low sensitivity of the bacterial strain of interest to the viral strain of the bacteriophages.

[0020] Step a) may involve depositing several drops, the drops being spaced apart from each other, two different drops comprising respectively: bacteria of different bacterial strains and / or different concentrations of bacteria of the same bacterial strain.

[0021] Regardless of the embodiment, the sample can be divided into different spatial zones, separated from each other in such a way that: at least two different spatial zones respectively comprise the same bacterial strain and a different concentration of bacteriophages of the same viral strain; and / or at least two different spatial zones respectively comprise the same bacterial strain and bacteriophages of different viral strains; and / or at least two different spatial zones respectively comprise bacteriophages of the same viral strain and different bacterial strains.

[0022] The process is such that each spatial zone is associated with a region of interest in each acquired image, with two different spatial zones being associated with two different regions of interest, so that the analysis of the same acquired image provides information on the sensitivity of a bacterial strain to a viral strain of bacteriophages, and this in different spatial zones.

[0023] The sample may include more than 10, or even more than 100, different spatial zones, separated from each other, each spatial zone being parameterized with three parameters corresponding respectively to the bacterial strain of interest, the viral strain of bacteriophages and the concentration of bacteriophages, at least one parameter of two different spatial zones being different.

[0024] The sample can be distributed in different fluidic chambers, of a fluidic map, each fluidic chamber corresponding to a spatial area of ​​the sample.

[0025] The sample may be formed by depositing drops on the surface of an agar plate at different positions, the drops being spaced apart from each other, such that a spatial area is defined by each drop position. The agar plate may contain bacteria and the drops may contain bacteriophages. Alternatively, the agar plate may contain bacteriophages and the drops may contain bacteria.

[0026] According to one embodiment, the method comprises, prior to step a), an enrichment step comprising: a mixture of bacteriophages of the same viral strain, or of different viral strains, in an aqueous solution comprising at least one strain of bacteria; incubation; filtration of the mixture, so as to retain the bacteria and obtain a solution enriched in bacteriophages; such that during step a), the contacting of the bacteriophages with the bacteria is carried out using the solution enriched in bacteriophages.

[0027] The image defocus distance and / or the object defocus distance is preferably less than 5 mm, or even 1 mm, or even less than 500 µm. The plane of the sample is a plane extending into the sample. It is preferably the surface of the sample closest to the image sensor. According to this embodiment, during step c), the image is acquired in a defocused configuration.

[0028] The invention will be better understood by reading the description of the exemplary embodiments presented in the remainder of the description, in conjunction with the figures listed below. FIGURES

[0029] There figure 1 represents a device allowing implementation of a method according to the invention. The figure 2A schematizes a first embodiment of the invention. The figures 2B et 2C are images obtained by implementing the first embodiment of the invention. The figure 2D is a curve obtained experimentally by implementing the first embodiment. It represents an evolution of an optical density as a function of time. The figures 2E And 2F illustrate examples of fluid maps that can be implemented. The figures 3A et 3B are images obtained by implementing a variant of the first embodiment. The figure 3C shows a curve obtained experimentally by implementing the variant of the first embodiment. It represents an evolution of a texture indicator as a function of time. The figures 4A And 4B schematize a second embodiment of the invention. The figures 4C, 4D And 4E are images obtained by implementing the second embodiment. They show a formation of lysis plaques as a function of time. The figures 5A And 5B schematize a third embodiment of the invention. The figures 5C And 5D are images obtained by implementing the third embodiment. The figures 6A And 6B schematize a fourth embodiment of the invention. The figure 7 shows the main steps of implementing the invention. The figure 8 shows another device allowing an implementation of the invention. EXPOSE DE MODES DE REALISATION PARTICULIERS

[0030] There figure 1 shows a device for implementing the invention. A sample 20 is placed between a light source 10 and an image sensor 30. The light source 10 produces an incident light wave 11, the latter propagating to the sample. Preferably, the incident light wave 11 reaches the sample in the form of a plane wave, or considered to be plane.

[0031] The sample comprises bacteria of one or more bacterial strains, as well as phages (or bacteriophages), of one or more viral strains. The objective of the invention is to visualize an optical effect, in the sample, of a lysis induced by the infection of bacteria by certain phages. Preferably, the bacteria are not labeled with a colorimetric or fluorescent marker. The invention is based on the observation of the interactions of the bacteria with the incident light wave. More precisely, the presence of bacteria in the sample induces a diffusion of the incident light wave. The higher the number of bacteria, the more the incident light wave is diffused. The image sensor comprises pixels, generally distributed in a matrix arrangement in a detection plane P. The image sensor collects the exposure wave 14 resulting from the interactions of the incident light wave 11 in the sample.The exposure light wave 14, which emerges from the sample, is thus constituted: . of a component 12 of the incident light wave, not absorbed and not scattered by the sample. This component is made up of photons usually referred to as ballistic photons. of a component 13 resulting from the scattering of the incident light wave by the sample.

[0032] The greater the number of bacteria, the greater the diffusion component 13 relative to the non-diffusion component 12.

[0033] Preferably, the incident light wave 11 extends along an illumination spectral band Δλ extending into the visible range. It is preferably between 450 nm and 650 nm, and more preferably between 500 nm and 600 nm. It is in fact between 500 nm and 600 nm that the scattering of the incident light wave 11 by the bacteria is maximum. Preferably, the spectral width of the illumination spectral band is less than 50 nm, or even less than 40 nm. By spectral width is meant the width at half-maximum, usually designated by the acronym FWHM (Full Width at Half Maximum).

[0034] The light source 10 is for example a light-emitting diode or a laser diode. It can also be a liquid crystal screen, the advantage being to have an extended luminous surface, from a few cm 2 < to a few tens of cm 2 < , producing uniform illumination. A band-pass filter 16 can be interposed between the light source 10 and the sample 20, so as to adjust the spectral band of the light wave reaching the sample.

[0035] The distance D between the light source 10 and the sample 20 is adjusted such that the light source is considered to be point-like. It may be between 5 cm and 20 or 30 cm. A diaphragm 17, defining an opening with a diameter of less than 100 µm or 200 µm, may be interposed between the light source 10 and the sample 20. Alternatively, an optical fiber may be interposed between the light source and the sample, so as to form a point-like light source.

[0036] The image sensor 30 may be a CCD or CMOS type sensor. Preferably, the detection surface is greater than 10 mm 2< , or even greater than 1 cm 2< . The inventors used a sensor whose detection surface is 14.9 mm x 22.3 mm, or of the order of 3 cm 2< . This makes it possible to obtain an image with a high field of observation. In order to maximize the field of observation, the distance d between the sample 20 and the image sensor 30 is as small as possible. It is preferably less than 1 cm. Preferably, the sample 20 is placed in contact with the image sensor, that is to say in contact with a protective glass of the image sensor, or a few millimeters from the latter. We note the absence of magnification or image formation optics between sample 20 and image sensor 30. This does not preclude the presence of focusing microlenses at the pixel level of image sensor 30.Thus, the image sensor 30 is arranged in a lensless imaging configuration. This makes it possible to maximize the field of observation of the image sensor. Each image then makes it possible to address a large volume of sample. The image sensor can be a monochrome sensor or a color sensor. It allows formation of an image representative of the exposure wave 14.

[0037] The image sensor 30 is connected to a processing unit 40, which receives the images acquired by the image sensor. The processing unit 40 is connected to a memory 42 in which instructions for implementing certain image processing steps, described below, are stored.

[0038] The thickness ε of the sample 20 can vary between 100 µm and 1 cm or 2 cm. The configuration of the sample, and its thickness, can vary according to the embodiments described below. Generally, the sample comprises a support, in which bacteria, from at least one bacterial source of interest, are brought into contact with phages, of a viral strain. The invention is thus used to determine the viral strain and / or the concentration of phages allowing lysis of the bacteria of the bacterial strain of interest.

[0039] A first embodiment is presented in connection with the figures 2A à 2F . According to this first embodiment, the sample comprises a mixture of bacteria and phages mixed in a solution, preferably an aqueous solution. The aqueous solution comprises nutrients, allowing the bacteria to develop. Thus, the aqueous solution comprises a liquid nutrient medium, preferably non-specific, usually referred to as "broth". This may, for example, be trypticase soy, a Columbia medium, or a Luria-Bertani type medium.

[0040] According to this embodiment, the sample is compartmentalized according to fluidic chambers 22 1 , 22 2 , 22 i isolated from each other. The index i is an integer designating the rank of a fluidic map, with 1 ≤ i ≤ N i where N i corresponds to the total number of fluidic chambers. The surface area of ​​each fluidic chamber, parallel to the detection plane P, may for example be between 1 mm 2 < and 10 mm 2 < . The fluidic chambers are arranged in a fluidic map 22. Preferably, each fluidic chamber is delimited by an opaque and / or reflective wall 23 1 , 23 2 , 23 i . This prevents light, diffused inside a fluidic chamber 22 i , from propagating into another fluidic chamber, forming stray light in the latter. Each fluidic chamber 22 i comprises an elementary sample 20 i which may be different from another elementary sample of another fluidic chamber.For a bacterial strain of interest, each elementary sample 20 i can thus contain phages of the same viral strain, according to a predefined concentration.

[0041] Thus, this embodiment makes it possible to divide the sample into different spatial zones, each spatial zone corresponding to a fluidic chamber 22 i , containing an elementary sample 20 i . Each elementary sample can be parameterized by three parameters: the bacterial strain; the viral strain; the concentration of phages of the viral strain.

[0042] Between two different elementary samples, corresponding to two different spatial zones, at least one parameter is different, except in the case of replicates.

[0043] The number of fluidic chambers 22 i is preferably greater than 10, and even more preferably greater than 100. The use of a large-surface image sensor, as previously described, makes it possible to acquire an image simultaneously addressing a large number of fluidic chambers. This avoids, or limits, the use of a mechanism allowing a translation of the sample 20 relative to the image sensor. It is understood that this embodiment makes it possible to carry out a phagogram. The fluidic card 22 thus takes the form of a well plate, allowing simultaneous analysis of a multitude of elementary samples different from one another.

[0044] The fluidic card 22 is for example made, at least in part, from a transparent plastic material, preferably biocompatible, for example COC (Cyclic Olefin Copolymer) or PMMA (Polymethylmetacrylate). It comprises transparent transverse walls 22 t, which confine each elementary chamber 22 i . The transverse walls preferably extend perpendicular to an axis Z along which the light wave 11 emitted by the light source 10 propagates.

[0045] Each elementary sample contains an initial quantity of bacteria 2. In the absence of phages, or in the absence of a significant effect of phages on the bacterial strain, the bacteria proliferate. On the figure 2A , this case corresponds to the fluidic chamber 22 1 . When the phages are able to infect the bacteria of interest, and to cause their lysis, the number of bacteria of interest decreases, which corresponds to the fluidic chamber 22 2 . The increase or decrease in bacteria leads to an evolution in the transport of light through each elementary sample. The images acquired by the image sensor, over time, make it possible to assess and quantify this evolution.

[0046] The greater the quantity of bacteria, the greater the scattering of light through a fluidic chamber, which reduces the intensity of the exposure light wave 14 propagating towards the image sensor 30. In the example shown on the figure 2A , the intensity of the light wave 14 2 is greater than the intensity of the light wave 14 1 . The image sensor comprises pixels 31. Groups of pixels 31 i can be defined, each pixel of the same group being exposed to an exposure light wave 14 i propagating from a fluidic chamber 22 i . Thus, each group of pixels 31 i is associated with a fluidic chamber 22 i and with an elementary sample 20 i . By acquiring an initial image, or an image of a control sample, then forming an image after a certain duration, it is possible to evaluate the evolution of the bacterial population of an elementary sample 20 i , by determining the temporal evolution of the intensity of the light wave 14 i detected by the pixels of the same group of pixels 31 i . Each group of pixels 31 i corresponds to a region of interest ROI i of the image acquired by the image sensor 30.Thus, the analysis of each region of interest ROI i of the image makes it possible to visualize the development or inhibition of the development of bacteria in an elementary sample 20 i . By region of interest, we mean parts of the image whose intensity is considered homogeneous or representative of the sample, or of a part of the sample.

[0047] THE figures 2B et 2C illustrate an example of implementation of the first embodiment. The experimental components and parameters are as follows: fluidic card 22: card machined from an aluminum plate, comprising eight circular chambers of diameter 3 mm (i.e. a cross-section of 7.1 mm²); volume of each fluidic chamber: 35 µL; image sensor: Canon 1200D RGB CMOS sensor in APS-C format, dimensions 14.9 mm x 22.3 mm - 18 million pixels; light source 10: LCD screen (Liquid Crystal Display) associated with a bandpass filter (Thorlabs FB 560-10) centered on a wavelength of 560 nm and defining a full width at half maximum of 10 nm; temperature: 25 °C; distance light source - sample: 1 cm; thickness of each fluidic chamber: 5 mm.

[0048] Each sample was taken in 9 mL of Tryptone Soy Broth nutrient medium (usually referred to by the acronym TSB - Trypcase Soy Broth), into which 250 µL of a culture solution had been introduced. Pseudomonas putida ATCC12633, and 100 µL of a phage suspension Pseudomonas virus gh1. One control chamber did not have phages added. One reference chamber was filled with the liquid nutrient medium, without bacteriophages or bacteria. The bacteriophage suspensions had infectivity titers of 6 x 10⁹ < ufp / mL, 6 x 10⁸ < ufp / mL, and 6 x 10⁷ < ufp / mL, respectively. The unit ufp is a unit known to those skilled in the art, meaning plaque-forming unit. The titer of each phage suspension was determined according to a reference method, using an agar medium. Each chamber was duplicated, except for the control and reference chambers.

[0049] The device was placed in a thermostatically controlled chamber at 25°C. Images were acquired every 2 minutes for a period of 6 hours and 4 minutes. figures 2B et 2C represent images acquired respectively at t = 30 minutes (initial image) and t = 6 hours and 4 minutes. The initial time t = 0 corresponds to the inoculation of the bacteria in the liquid medium. The images were formed by taking into account only the green pixels of the sensor. On the figures 2B et 2C , we distinguish different regions of interest ROI i , each region of interest corresponding to an elementary sample 20 i . In this example, the regions of interest of columns C1, C2 and C3 correspond to an identical elementary sample: same bacterial strain - same concentration of phages.

[0050] On the figures 2B et 2C , the wells of columns C1, C2 and C3 correspond respectively to the concentrations of 6.10 7< pfu / mL (i.e. MOI = 0.01), 6.10 8< pfu / mL (i.e. MOI = 0.1) and 6.10 9< pfu / mL (i.e. MOI = 1). The acronym MOI stands for Multiplicity of Infection, a term known to those skilled in the art designating a ratio between a quantity of phages (in pfu / mL) on a quantity of bacteria (in cfu / mL) in each sample. The term cfu stands for colony forming unit, usually used in microbiology.

[0051] The fourth column C4 includes the reference chamber C4-1 and the control chamber C4-2 which does not contain phage.

[0052] In the C4-2 control chamber, a darkening of the image is observed. This is due to bacterial proliferation in the absence of phage.

[0053] On each column, an average of the intensity (i.e. the gray levels) of the images acquired every 2 minutes was taken. An equivalent optical density DO(t) was then calculated, the latter being obtained by the expression: DO t = − log 10 I t I t = 0 + 0.055

[0054] Or I ( t ) is the intensity at a measurement instant t, corresponding to an average of gray levels and I ( t = 0) is the intensity at the initial instant t = 0. The constant 0.055 was added so that the optical density at the initial time corresponds to that measured by a spectrophotometer.

[0055] There figure 2D shows the evolution of the equivalent optical density DO(t) (y-axis) as a function of time (x-axis - unit: hour).

[0056] From these curves, it can be seen that under the action of the phage, and for all the phage concentrations considered, the optical density increases for a few hours after the initial time, but then stabilizes. In the control chamber, the optical density increases continuously. Comparison of the optical densities measured in the control chamber and in the other chambers respectively shows the inhibition of bacterial development by the phage. The action of the phage becomes perceptible two hours after the initial time.

[0057] It is noted that the invention allows real-time monitoring of the action of a phage on bacteria, and this simultaneously on several elementary samples 20 i . It is possible to define, for each sample, a detection threshold, the crossing of which means that bacterial proliferation is suspected in the sample. Such a threshold can be defined from the average optical density measured on the control sample, as well as from a dispersion indicator σ, for example the standard deviation, at a measurement time or at different measurement times. The detection threshold can be equal to µ + kσ, k being a strictly positive real number, preferably greater than 1 or 2, and for example equal to 5. The establishment of such a detection threshold is easily automated and makes it possible to determine the presence of bacterial proliferation early, for example between 5 a.m. and 10 a.m.Other thresholds, based on a comparison between the control sample and samples containing phages, can be defined.

[0058] There figure 2E Figure 22 illustrates a fluidic map enabling implementation of the invention. The fluidic map comprises fluidic chambers 22i, as previously described, each closed by two transverse walls 22t. One transverse wall may be in the form of a transparent adhesive film. The other transverse wall 22t may be made of a transparent material, preferably permeable to oxygen, to allow oxygenation of the fluidic chambers 22i, which is necessary for the growth of certain bacteria. Therefore, it is preferable that at least one transverse wall delimiting each fluidic chamber be permeable to oxygen. This could, for example, be PDMS (Polydimethylsiloxane) or TPX (Polymethylpentene), the latter being described in document EP0745667.

[0059] The nutrient medium and / or the phages may be previously placed in each fluidic chamber 22 i, for example in a freeze-dried state. In the example shown in the figure 2E Each fluidic chamber 22i is connected, via a fluidic channel, to an inlet 22i and vent 22e. Each fluidic chamber is filled using a syringe, by introducing the solution containing the bacteria through inlet 22in. Vent 22e allows air to be expelled during the filling of the fluidic chamber. A liquid-blocking, gas-permeable filter can be placed at vent 22e.

[0060] Alternatively, as shown in the figure 2F , each fluidic chamber is initially under vacuum. The filling of each fluidic chamber is ensured by a common inlet. When the fluidic card 22 is placed under vacuum, the aqueous solution containing the bacteria migrates to each fluidic chamber. The bacteriophages are previously placed in each fluidic chamber, in a lyophilized state. Contact between the bacteria and the bacteriophages therefore takes place during the filling of each fluidic chamber.

[0061] Alternatively, different fluid chambers contain different bacterial strains. These may be present in a lyophilized state. In this case, each fluid chamber can be filled via a common inlet, through which a solution containing phages flows. The objective is then to identify viral strains that may be relevant for certain bacterial strains. According to this possibility, a preliminary phage enrichment step can be provided. During this step, the different bacterial strains are placed in the same chamber, in which a phage solution, comprising a single viral strain or several viral strains, is mixed. If, among the bacterial strains present, a strain is sensitive to a phage, the latter is replicated, which leads to an increase in its concentration. This results in non-specific amplification of phages.The mixture is then filtered to retain the bacteria, with a filtration size of around 0.2 µm. The bacterial strains are then distributed into different fluidic chambers, so as to end up with a single bacterial strain per well. The filtered solution is injected into each fluidic chamber to identify the bacterial strain sensitive to the phage.

[0062] This embodiment can be adapted to test the effect of different viral strains on the same bacterial strain. For this purpose, different fluid chambers can respectively contain phages of different phage strains. Within a fluid chamber containing a bacterial strain sensitive to the phage, no bacterial proliferation is observed, and the phage is specifically amplified. When the bacterial strain in a chamber is not sensitive to the phage, bacterial proliferation is observed.

[0063] In this embodiment, an optical density or, more generally, a change in a light intensity transmitted by each elementary sample 20 i and detected by the image sensor is measured. In order to obtain uniform illumination of each elementary sample 20 i , that is to say of each elementary fluidic chamber 22 i , it is preferable for the light source 10 to be an extended source, for example a liquid crystal screen. This promotes the compactness of the device. Alternatively, a point light source 10 can be used, provided that it is sufficiently distant from the sample.

[0064] THE figures 3A et 3B illustrate a variant of the first embodiment, in which each elementary sample is present, in the liquid state, in a fluidic chamber. According to this variant, the presence of microorganisms is quantified by an analysis of the texture of the image. Under the effect of the development of bacteria, it has been observed that the texturing of the images is more marked. On each region of interest ROI i of an acquired image, a texturing indicator is established, so as to characterize the elementary sample 20 i corresponding to the region of interest. The texturing indicator can be a vector or a scalar quantity. This embodiment is mainly intended for samples of low thickness, for example between 100 µm and 1 mm, and preferably between 100 µm and 250 µm. On the figures 3A et 3B , the regions of interest considered have been enlarged.

[0065] The texturing of the image results from the formation of interference patterns on the image sensor, between the light wave 12 transmitted by the sample, and the light wave 13 resulting from the scattering of the incident light wave 11 by the sample. According to this embodiment, it is preferable for the light source to be spatially coherent (point-like), and sufficiently far from the detector so that the incident light wave 11 reaches the sample in the form of plane waves, or considered as such.

[0066] THE figures 3A à 3C illustrate an example of implementation of the variant of the first embodiment. The experimental components and parameters are: fluidic card 22: microscope slide on which a Geneframe fluidic chamber with a thickness of 250 µm is deposited, the fluidic chamber being closed by another microscope slide; volume of each fluidic chamber: 25 µL; image sensor: monochrome 10.5 Mpixel sensor with a side size of 1.67 µm, forming a detection surface of 29.5 mm 2< (6.4 mm x 4.6 mm); light source 10: light-emitting diode centered on the wavelength of 650 nm with a width at half-maximum of 40 nm; temperature: 30 °C; light source - sample distance: 5 cm; thickness of each fluidic chamber: 250 µm.

[0067] In this example, the fluidic chamber is not compartmentalized. There is only one sample analyzed. The sample was collected in 9 mL of Tryptone Soy Broth nutrient medium (usually referred to by the acronym TSB - Trypcase Soy Broth), into which 250 µL of a culture solution of Pseudomonas putida ATCC12633, diluted by a factor of 10², was used. 100 µL of a phage suspension was added. Pseudomonas virus gh1, known to infect bacteria Pseudomonas putida, whose concentration was estimated at 7 10 5< ufp / mL. 30 µL of sample were taken and introduced into the fluidic card 22.

[0068] THE figures 3A et 3B These images show an image of the fluidic chamber at an initial time and 10 hours later. These images were obtained using a control sample without phages. They illustrate the evolution of the image texture as the bacteria grow. For each image, a texture descriptor was determined by calculating the standard deviation of three reduced 80 x 80 pixel areas of the image. figures 3A et 3B A reduced area of ​​this type was represented. The texture descriptor corresponds to the average of the three standard deviations thus determined. figure 3C shows an evolution of the determined texturing indicator: for a control sample, without phage (curve a) for a sample with phages Pseudomonas virus gh1 - 7 10 5< ufp / mL (curve b)

[0069] This figure was obtained by performing an image acquisition per 10-minute period over a period of 18 hours. We observe that: for the control sample, the texture descriptor increases, under the effect of bacterial proliferation; for the sample containing the phages, the texture descriptor stagnates over time, reflecting an inhibition of bacterial development.

[0070] It is possible to define a detection threshold, the crossing of which indicates that bacterial proliferation is suspected in the sample. Such a threshold is determined from a mean (or median) value µ and a dispersion indicator, for example the standard deviation σ, of the texture descriptor, over a short period of time after the initial time, for example 1 or 2 hours. The detection threshold can be equal to µ + kσ, where k is a strictly positive real number, preferably greater than 1 or 2, and for example equal to 5. Establishing such a detection threshold is easily automated and allows for the early determination of the presence of bacterial proliferation, for example between 5 and 10 hours.

[0071] THE figures 4A à 4E illustrate a second embodiment, in which the bacteria and phages are arranged in an agar plate, comprising a nutrient medium. The nutrient medium is transparent and is preferably non-specific. The agar plate comprises phages, in the form of virions. The latter infect the bacteria and multiply following the lysis of the latter. The lysis of bacteria thus spreads locally, in the vicinity of the locations of virions in the initial state. The propagation allows the formation of a lysis plaque, corresponding to a part of the agar plate depleted in bacteria. As a result, at the level of the lysis plaques, the transmission of light is increased compared to the other parts of the sample, in which the bacteria proliferate.

[0072] Such an agar can be made by mixing a conventional agar nutrient medium, brought above its supercooling temperature, for example between 40°C and 60°C, with an aqueous solution containing bacteria and phages. Thus, the sample contains a soft agar, comprising bacteria and phages. The thickness of the sample can be between 2 mm and 5 mm. Having bacteria and phages at such a depth allows better demonstration of bacterial lysis. The mass fraction of agar is for example between 0.5% and 1.4%, which is usually referred to as "soft agar".

[0073] Alternatively, part of the sample thickness is formed from a standard nutrient medium, without bacteria or phage. For example, it may comprise agar-agar in a mass fraction of 1.5%. The function of this nutrient medium is to form a nutrient reservoir for the bacteria. However, because the sample analysis process is rapid, in the order of a few hours or a few dozen hours, the presence of such a nutrient reservoir is not necessary. The absence of such a reservoir is considered advantageous, as it prevents light from being scattered by the reservoir.

[0074] THE figures 4A à 4B schematize a sample 20 respectively at an initial time, considered as corresponding to the formation of the surface layer 20s, and at a time subsequent to the initial time. In the vicinity of the phages initially present, bacteria are lysed, and the lysis of the bacteria spreads from one place to another, under the effect of the replication of the phages. Depleted zones 23, or lysis plaques, form in the surface layer 20s of the sample. The transmission of light through the lysis plaques is greater than through the rest of the sample. Thus, when the sample is illuminated by a light source, groups of pixels 31, located opposite the depleted zones 23, collect a quantity of light greater than the quantity of light captured by the other pixels. On the image acquired by the image sensor, this results in the appearance of clear regions of interest. Each clear region of interest corresponds to a lysis plaque.Counting clear regions of interest allows an estimation of an infectious titer of the solution containing the bacteriophages. This infectious titer can be expressed in number of plaque-forming units (pfu).

[0075] THE figures 4C, 4D And 4E are experimental images acquired respectively 1h30, 1h50 and 2h10 after sample preparation. 250 µl of a bacterial culture was mixed in 5 ml of soft agar (50% TSA - Trypticase Soy Agar - 50% TSB Trypticase Soy Broth). Pseudomonas putida ATCC12633 (approximately 10 9< cfu / mL) in broth, as well as 100 µl of an aqueous solution containing phages Pseudomonas virus gh1. The preparation was homogenized by vortexing. 2 mL of the preparation was poured into a 3.5 mm diameter Petri dish, forming a fluidic chamber. The sample was placed on a 30 Mpixel image sensor and the whole was placed in a thermostatically controlled chamber at 30 °C. The experimental parameters are as follows: Image sensor: Canon 1200D RGB CMOS sensor with dimensions of 14.9 mm x 22.3 mm - 18 million pixels; Light source 10: Light-emitting diode centered on the wavelength of 560 nm; Light source - sample distance: approximately 20 cm; Thickness of each fluidic chamber: 2 mm.

[0076] The initial time t = 0 is considered to be confounded with the deposition of the preparation in the Petri dish. The distribution of biomass in the dish is considered homogeneous. Over time, lysis patches form, which correspond to light spots in images 4C, 4D and 4E. In the figure 4D , some bright spots are indicated by arrows. These images were formed from the green pixels of the image sensor. The gradual formation of bright regions of interest can be observed in the images.

[0077] According to another approach, a compartmentalized fluidic card is used, as described in connection with the first embodiment. The fluidic card comprises fluidic chambers 22 i separated from each other. Each fluidic chamber has a mixture of phages and a freeze-dried cold gelling agent. Each fluidic chamber can be supplied with a solution comprising bacteria. A gel is then formed, comprising a mixture of bacteria and phages. The cold gelling agent is water-soluble. It is configured to gel when brought into contact with an aqueous solution, at room temperature, thus forming a hydrogel. It can be a water-soluble polysaccharide. The cold gelling agent can be chosen from carboxymethylcellulose, guar gum, gum arabic, gellan gum, xanthan gum, or a gelling agent obtained from animal bones, for example pork, beef or chicken.

[0078] Such an approach allows the sample to be divided into different spatial zones, each spatial zone corresponding to a fluidic chamber.

[0079] Alternatively, the fluidic card is brought, during its filling, to a supercooling temperature of the agar containing the bacteria. Thus, the agar containing the bacteria can fill each fluidic chamber, so as to mix, in each chamber, with the phages previously placed in a lyophilized state. This avoids the use of a cold gelling agent, the latter being able to diffuse.

[0080] A third embodiment is shown in the figures 5A à 5D According to this embodiment, the bacteria are initially present in an agar plate, or in a surface layer of a solidified agar plate. Drops of a solution containing phages are deposited onto the surface of the agar plate. The drops are separated from each other. When the bacteria in the agar plate are sensitive to the bacteriophage, plaques of lysis form in the sample, as described in the second embodiment. An advantage of this embodiment is that the position of each deposited drop can be known, making it possible to search for the development of plaques of lysis at each of these positions. This facilitates the image processing step.

[0081] THE figures 5A And 5BThis embodiment is schematically illustrated, respectively, during the application of the drops to the agar, and following the formation of plaques of lysis when the bacteria are sensitive to the viral strain of the phages present in the drops. The applied drops may contain different viral strains, and / or different concentrations of phages of the same viral strain. Similar to the first embodiment, this embodiment allows for parallel testing of different phages or different concentrations. On the figure 5B , a formation of two lysis areas has been schematized. As in the second embodiment, under the effect of the lysis of the bacteria, the lysis areas appear, through which the transmission of light is higher than in the rest of the sample. The pixels 31 located opposite the lysis areas collect a signal whose intensity is higher. This results in the appearance of light spots on the image formed by the image sensor.

[0082] A sample was prepared by mixing 120 µL of a culture medium, comprising Pseudomonas putida ATCC12633 and 6 mL of soft agar (7.5 g / L of agar). This preparation was poured into a petri dish, resulting in an agar approximately 2 mm thick. 5 µL drops containing: either a suspension of phages Pseudomonas virus gh1 whose titer was estimated at 3.6 10 6< pfu / mL; or a pharmaceutical diluent to simulate the presence of a non-active phage on the bacterial strain used.

[0083] The drops were dried (15 minutes), then the sample was placed in a thermostatically controlled enclosure at 28°C. The experimental parameters are as follows: Image sensor: Canon CMOS 1200D RGB sensor with dimensions of 14.9 mm x 22.3 mm - 18 million pixels; Light source 10: Light-emitting diode centered on the wavelength of 560 nm; Light source - sample distance: approximately 20 cm; Fluidic chamber thickness: 2 mm.

[0084] A thin layer (between 0.1 mm and 1 mm) of oil 25 can be placed on the sample, so as to prevent drying. Oil 25 can be silicone oil, allowing diffusion of oxygen to the nutrient medium.

[0085] THE figures 5C And 5DThese are experimental images acquired 45 minutes and 4 hours 30 minutes, respectively, after droplets were deposited on a sample. In each of these figures, each ellipse e1 represents the droplet deposition area, and each ellipse e2 corresponds to an image analysis region. In each of these images, the ellipses e1, shown on the left and right, correspond to the deposition of the phage and the pharmaceutical diluent solution, respectively. By comparing the figures 5C And 5D , we observe an increase in the gray level at the ellipse e2 on the left of the image. This corresponds to the formation of a lysis zone. On the other hand, at the ellipse e2 located on the right of the image, the gray level decreases, following the development of bacteria.

[0086] Such an embodiment makes it possible to divide the sample into different spatial zones, each spatial zone corresponding to a position at which a drop was deposited. The image acquired by the image sensor makes it possible to analyze several spatial zones simultaneously.

[0087] A fourth embodiment is shown diagrammatically on the figures 6A And 6B. According to this embodiment, an agar plate comprising phages 3 is provided. Drops of a solution comprising bacteria are deposited on the surface of the latter. The drops are separated from each other. When the bacteria are not sensitive to the viral strain, they develop on the surface of the sample, forming colonies. When the bacteria contained in the drops are sensitive to the viral strain of the bacteriophages present in the agar plate, no colony formation is observed on the surface of the sample. An advantage of this embodiment is that the position and composition of each deposited drop can be known, which makes it possible to search for the development of colonies at each of these positions. This facilitates the image processing step.

[0088] THE figures 6A And 6Bschematize this embodiment, respectively when the drops are deposited on the agar, and following the formation of colonies when the bacteria present in a drop are little or not sensitive to the phage present in the agar. The drops deposited may respectively contain different strains of bacteria, or different concentrations of the same strain. This embodiment makes it possible to carry out, in parallel, tests of different bacterial strains or different concentrations of the same bacterial strain. On the figure 6B , we have schematized a formation of two colonies. The pixels 31 located opposite the colonies collect a signal whose intensity is lower. This results in the appearance of dark spots on the image formed by the image sensor.

[0089] Such an embodiment makes it possible to divide the sample into different spatial zones, each spatial zone corresponding to a location at which a drop was deposited. The image acquired by the image sensor makes it possible to analyze several spatial zones simultaneously.

[0090] There figure 7 schematizes the main stages of a process according to the previous embodiments. Step 100: bringing bacterial strains into contact with phages. This step can be carried out in a single fluidic chamber (see second, third, or fourth embodiments), or in different fluidic chambers (see first embodiment). During this step, the sample can be divided into different spatial zones. These spatial zones can correspond to: different fluidic chambers (seefirst embodiment); or at different positions where drops of an aqueous phage solution are deposited on an agar containing bacteria (third embodiment); or at different positions where drops of an aqueous bacterial solution are deposited on an agar containing phages (fourth embodiment) Step 110: illumination of the sample using the light source; Step 120: acquisition of an image by the image sensor; Step 130: analysis of the image by the processing unit, so as to evaluate the sensitivity of at least one bacterial strain to at least one viral strain.

[0091] The advantages of the process as previously described are: early detection of bacterial lysis by phages. Indeed, using the methods of the prior art, it is considered that exploitable results are obtained in at least one day; a possibility of implementation so as to simultaneously address different bacterial strains in combination with different phage strains, or different ratios between the quantities of phage strains and the quantity of bacterial strains; simplicity of implementation, without requiring expensive equipment.

[0092] There figure 8 shows another device for implementing the invention. The device includes an optical system 19 disposed between the sample 20 and the image sensor 30. The optical system may be a lens or an objective. The image sensor defines a detection plane P. The sample defines a sample plane P20. The sample plane is a plane, preferably parallel to the detection plane, extending into the sample. It is preferably the surface of the sample closest to the image sensor. The optical system defines an image plane Pi and an object plane Po. According to a defocused configuration: the image plane is offset from the detection plane by an image defocus distance; and / or the object plane is offset from the sample plane by an object defocus distance.

[0093] The image defocus distance and / or the object defocus distance is preferably less than 1 mm, or even less than 500 µm.

[0094] On the figure 8 , a defocused configuration is shown, in which the object plane P o is merged with the sample plane P 20 . The image plane is offset from the detection plane by an image defocusing distance δ. Generally speaking, the defocused configuration is preferred when seeking to quantify a texturing of the image.

[0095] Compared to the configurations shown in connection with the figure 8 , the lensless configuration, described on the figure 1 , is preferred because it allows the use of a compact and inexpensive device.

Claims

1. Method for determining the sensitivity of a bacterial strain to a viral strain of bacteriophages (3), the method comprising: a) preparing a sample (20), this comprising bringing bacteria (2), of the bacterial strain, into contact with bacteriophages (3), each bacteriophage belonging to the same viral strain, the bacteria being either in a liquid medium, or in an agar medium; b) placing the sample between a light source (10) and an image sensor (30), the light source emitting a light wave (11) in an emission spectral band comprised between 500 nm and 600 nm, the image sensor lying in a detection plane; c) illuminating the sample using the light source and acquiring at least one image of the sample, with the image sensor, in the emission spectral band, the image sensor being: - either arranged in a lensless imaging configuration, no magnifying or image-forming optics being placed between the sample and the image sensor; - or arranged in a defocused imaging configuration, an optical system, i.e. a lens or objective, being placed between the sample and the image sensor, the optical system defining an object plane and an image plane, the object plane being offset from the detection plane or the image plane being offset from the sample; d) on the basis of each acquired image, determining a sensitivity of the bacterial strain to the viral strain of the bacteriophages, the method being such that: - in step a), the bacteria and the bacteriophages are mixed in an aqueous solution; - step c) comprises acquiring at least two images, at successive measurement times; - step d) comprises determining a light intensity detected by all or part of the image sensor, in each image respectively acquired at each measurement time (t), such that the bacterial strain is considered to be: • hardly or not sensitive to the viral strain of the bacteriophages when the detected light intensity decreases between two successive measurement times; • or sensitive to the viral strain of the bacteriophages when the detected light intensity does not decrease or increases between two successive measurement times.

2. Method according to Claim 1, wherein the width of the emission spectral band is narrower than 50 nm.

3. Method according to either of the preceding claims, wherein the distance between the sample (20) and the image sensor (30) is smaller than 5 cm or smaller than 1 cm.

4. Method according to any of the preceding claims, wherein step d) comprises determining an attenuation of the light, emitted by the light source, by the sample, such that the bacterial strain (2) is considered to be: • hardly or not sensitive to the viral strain of the bacteriophages when the attenuation increases between two successive measurement times; • or sensitive to the viral strain of the bacteriophages when the attenuation does not increase between two successive measurement times.

5. Method according to any of the preceding claims, wherein: - in step a), the bacteria and the bacteriophages are mixed in an aqueous solution; - step c) comprises acquiring at least two images, at successive measurement times; - step d) comprises determining a texture descriptor for each acquired image, such that the bacterial strain is considered to be: • hardly or not sensitive to the viral strain of the bacteriophages when the variation in the texture descriptor is indicative of an increase in scattering of the light by the sample between two successive measurement times; • or sensitive to the viral strain of the bacteriophages when the variation in the texture descriptor is indicative of a decrease or a stagnation in scattering of the light by the sample between two successive measurement times.

6. Method according to any of Claims 1 to 4, wherein: - in step a), the bacteria and the bacteriophages are mixed in an agar medium; - step c) comprises acquiring at least one image, subsequently to step a); - step d) comprises analysing each acquired image, so as to identify light regions of interest, each light region of interest corresponding to an infection of bacteria by bacteriophages, forming a viral plaque, each light region of interest indicating a sensitivity of the bacterial strain to the viral strain of the bacteriophages.

7. Method according to Claim 6, comprising: - counting the number of light regions of interest in at least one acquired image; - estimating a viral load of the bacteriophages in the sample on the basis of the number of light regions of interest counted.

8. Method according to any of Claims 1 to 4, wherein: - in step a), the bacteria are located in an agar medium, and the bacteriophages are located in a solution, step a) comprising depositing at least one droplet of the solution on the agar medium; - step c) comprises acquiring at least one image, subsequently to step a); - step d) comprises analysing each acquired image, so as to identify light regions of interest, each light region of interest corresponding to an infection of bacteria by bacteriophages, forming a viral plaque, such that the appearance of each light region of interest indicates a sensitivity of the bacterial strain to the viral strain of the bacteriophages.

9. Method according to Claim 8, wherein step a) comprises depositing a plurality of droplets, the droplets being spaced apart from one another, two different droplets respectively comprising: - bacteriophages of various viral strains; - and / or various concentrations of bacteriophages of a given viral strain.

10. Method according to any of Claims 1 to 4, wherein: - in step a), the bacteriophages are located in an agar medium, and the bacteria are located in a solution, step a) comprising depositing a droplet of the solution on the agar medium; - step c) comprises acquiring at least one image, subsequently to step a); - step d) comprises analysing each acquired image, so as to identify: • light regions of interest, each light region of interest corresponding to an infection of bacteria by bacteriophages, such that each light region of interest indicates a sensitivity of the bacterial strain to the viral strain of the bacteriophages; • or dark regions of interest, each dark region of interest corresponding to a development of bacteria in the presence of bacteriophages, such that each dark region of interest indicates an insensitivity or a low sensitivity of the bacterial strain of interest to the viral strain of the bacteriophages.

11. Method according to Claim 10, wherein step a) comprises depositing a plurality of droplets, the droplets being spaced apart from one another, two different droplets respectively comprising: - bacteria of various bacterial strains; - and / or various concentrations of bacteria of a given bacterial strain.

12. Method according to any of the preceding claims, wherein the sample is divided into various spatial regions (20i) that are separated from one another such that: - at least two different spatial regions respectively comprise the same bacterial strain and a different concentration of bacteriophages of a given viral strain; - and / or at least two different spatial regions respectively comprise the same bacterial strain and bacteriophages of various viral strains; - and / or at least two different spatial regions respectively comprise bacteriophages of the same viral strain and various bacterial strains; the method being such that each spatial region is associated with one region of interest (ROIi) of each acquired image, two different spatial regions being associated with two different regions of interest, so that analysis of a given acquired image allows information relating to the sensitivity of a bacterial strain to a viral strain of bacteriophages to be obtained in various spatial regions.

13. Method according to Claim 12, wherein the sample comprises more than 10, or even more than 100, different spatial regions that are separated from one another, each spatial region being parameterized by three parameters respectively corresponding to the bacterial strain of interest, to the viral strain of the bacteriophages and to the concentration of the bacteriophages, at least one parameter of two different spatial regions being different.

14. Method according to either of Claims 12 and 13, wherein the sample is distributed between various fluidic chambers (22i) of a fluidic chip, each fluidic chamber corresponding to one spatial region of the sample.

15. Method according to either of Claims 12 and 13, wherein the sample is formed following a deposition of droplets on the surface of an agar medium, in various positions, the droplets being spaced apart from one another, such that one spatial region is defined by each droplet position.

16. Method according to any of the preceding claims, comprising, prior to step a), an enriching step, comprising: - mixing bacteriophages of a given viral strain, or of various viral strains, in an aqueous solution comprising at least one strain of bacteria; - incubating; - filtering the mixture, so as to retain the bacteria and obtain a solution enriched in bacteriophages; such that, in step a), when the bacteriophages are brought into contact with the bacteria the solution enriched in bacteriophages is used.