Apparatus and method for characterizing cells subjected to physical stress

Abstract

The present invention relates to a microfluidic device and a method for applying at least one hydrodynamic stress of a determined intensity and duration to cells in a suspended state, the cells moving in the device, and the microfluidic devices and methods are used to characterize the morphological and physiological status of these cells during and optionally after the application of hydrodynamic stress.

Description

Technical Field

[0001] The present invention relates to microfluidic devices and methods for applying at least one hydrodynamic stress of a determined intensity and duration to cells in suspension and for characterizing the morphological and physiological states of the cells to which such physical stress is applied.

[0002] Thus, the present invention belongs to the fields of microfluidic devices and analytical cell biology. Prior Art

[0003] Currently known methods for describing and characterizing cell morphology and physiology are particularly based on the use of cell markers. Morphology itself can serve as a marker of the physiological state of the cell. However, on the one hand, these methods require suitable markers for the physiological state, and on the other hand, they only characterize the physiological state of the cells observed when labeled.

[0004] Furthermore, the efficiency of biological processes using animal cells, especially such as bioprinting processes, the bioproduction of cells for cell therapy and cell-based processes for the production of, for example, therapeutic proteins, viral vaccines or viral vectors, and all cell injection or harvesting processes, depends on the quality of the cells used in said biological processes.

[0005] Cell quality is generally related to physiological characteristics such as cell viability, i.e., the ratio of living cells to cells involved in the cell death process (apoptosis, necrosis, lysis), or cell function (the ability to secrete molecules, the ability of stem cells to differentiate into cell subtypes). Thus, the effectiveness of biological processes is strongly influenced by the changes they produce in cell quality and cell physiology. One of the mechanisms having a major impact on quality, particularly on cell viability, in biological processes is the application of hydrodynamic and mechanical stresses generated by the equipment used, such as bioreactors, systems using syringes or injectors, or separation steps. The interaction between the suspension fluid and the wall generates stress on the cells suspended in the fluid. The direct interaction of the cells with the wall generates a mechanical shock on the cells. Depending on the nature of the process and the technology employed, the intensity and duration of the physical stress applied to the cells can vary greatly. For example, cell therapy processes can generate stresses of up to 5000 Pa on the cells, and bioprinting processes can generate stresses with a duration of up to 30 milliseconds.

[0006] Therefore, it is desirable to be able to characterize the physiological state of cells, particularly by characterizing their morphology, in a simple and reproducible manner during and / or after the application of physical stress. In particular, this type of characterization is desired for cells that may be used in biological processes.

[0007] It is also necessary to characterize the physiological state of cells when they are subjected to physical stress that reproduces the physical stress conditions applied to the cells during a specific biological process.

[0008] WO 2015 / 024690 “Apparatus and method for determining the mechanical properties of cells” relates to methods and devices for determining the mechanical properties of cells and discloses a study of the deformation of cells subjected to shear stress. The document discloses a device comprising a single capillary channel in which suspended cells are subjected to shear stress, and a method for measuring the shape of the cells as they flow through the device. The method uses binary images. The measurements are made in real time.

[0009] EP 3 796 212 “Device for image-based cell classification, method therefor and use thereof” describes a real-time, label-free cell sorting device. The device comprises a microfluidic network that aligns each cell along its main axis, and a classification unit that includes a neural network for classifying cells based on cell images.

[0010] WO 2019 / 006188 “Quantitative deformability cytometry: rapid, calibrated measurements of cell mechanical properties” describes a microfluidic device and a quantitative deformability cytometry (q-DC) method for quantitatively evaluating the intrinsic mechanical properties of cells. The measured mechanical parameters are: elastic modulus E, cell mobility P, transit time T T , migration time T C , cell size D 细胞 and maximum stretch ε max , at a flow rate of 100 cells per second. Machine learning tools are used. The cells are subjected to calibrated shear stress.

[0011] Therefore, there is a need for devices and methods that can not only characterize the mechanical properties of cells subjected to physical stress, but also further characterize their morphology and physiological state.

[0012] Accordingly, there is a need for devices and methods capable of generating at least one physical stress in a biological process, the nature, intensity, and duration of which are defined and controlled to characterize cells that may be subjected to such stress. "Physical stress" is understood to particularly refer to any force applied to a cell, especially a force capable of generating cell stress; such physical stress can in particular be mechanical shock or hydrodynamic stress, such as tensile stress, compressive stress, or shear stress. SUMMARY OF THE INVENTION

[0013] The present inventors have developed a device for applying at least one physical stress of defined intensity and duration to cells in suspension, and the device is used to characterize cells that have been subjected to the at least one physical stress.

[0014] More specifically, the device according to the present invention is configured to:

[0015] - Apply at least one hydrodynamic stress of defined intensity and duration to at least one cell suspended and moving in the device,

[0016] - And optionally apply at least one mechanical shock to the cells.

[0017] The device according to the present invention is configured to characterize the cells during the application of the stress and optionally to characterize the cells after the application of the stress.

[0018] The device according to the present invention includes, on the one hand, a microfluidic circuit for applying at least one physical stress and, on the other hand, means for characterizing cells.

[0019] In the device according to the present invention, the at least one physical stress is a stress selected from the following: mechanical stress, especially mechanical shock, and / or hydrodynamic stress, such as hydrodynamic compressive stress, hydrodynamic shear stress, and / or hydrodynamic tensile stress. More specifically, in the device according to the present invention, the at least one physical stress is a stress selected from hydrodynamic stresses.

[0020] In the device according to the present invention, the means for cell characterization are selected from all known technical devices for cell observation, especially imaging devices or cell counting devices. More specifically, in the device according to the present invention, during the application of the stress, the means for cell characterization are selected from all known technical devices for cell observation, especially imaging devices or cell counting devices.

[0021] The device according to the invention comprises a microfluidic circuit which includes a section configured to apply at least one physical stress, more specifically a hydrodynamic stress, to at least one cell in suspension, said section comprising at least: i) a main channel configured to circulate a fluid containing said cell in suspension, ii) a fluid inlet and a fluid outlet, and iii) means for introducing and establishing a flow of said fluid within said channel.

[0022] Thus, a first object of the invention is a device for applying at least one physical stress of a determined intensity and duration, more specifically at least one hydrodynamic stress, to at least one cell in suspension, and said device is for characterizing said cell during and / or after said application, more specifically during said application and optionally after said application.

[0023] A second object of the invention is a method for applying at least one physical stress of a determined intensity and duration, more specifically at least one hydrodynamic stress, to at least one cell in suspension, and said method is for characterizing said cell during and / or after said application, more specifically during said application and optionally after said application.

[0024] Another object of the invention is a classification model pre-trained on a training data set according to the device and method of the invention, which is for characterizing cells and predicting their physiological state during and / or after applying at least one physical stress. Compared with the existing tools on the market, this classification model saves time and money and improves performance (in particular, it is capable of performing a large number of cell characterizations in a short time).

[0025] Finally, the invention relates to the use of the device, method or classification model according to the invention, which is for characterizing cells and potentially predicting their physiological state during and / or after applying at least one physical stress.

[0026] The device and method according to the invention have the advantage of generating at least one physical stress, the nature, intensity and duration of which are precisely determined and controlled, in order to characterize the cells that have been subjected to said stress.

[0027] The device and method according to the invention have the advantage of being able to apply an adjustable sequence of at least one physical stress, in particular the repetition of the same stress and / or the combination of physical stresses of different natures, with the intensity and duration of each stress being precisely defined. Thanks to this simple and precise tool, it is possible to reproduce the stresses applied in biological processes and to monitor the state of at least one cell at high speed.

[0028] The devices and methods according to the invention also have the advantage of being able to predict the physiological state of cells during and / or after being subjected to physical stress. Such prediction can be carried out during the application of these physical stresses and up to several days after the application of these physical stresses.

[0029] Finally, the devices and methods according to the invention have the advantage of defining the ability of cells to withstand at least one determined physical stress while maintaining a morphological and physiological state compatible with the requirements of said biological process. Thanks to the devices and methods according to the invention, for a given cell, a map of their resistance to at least one physical stress can be defined.

[0030] "Capacity" means the ability of a cell to withstand at least one physical stress of a specific intensity and duration without changing its physiological state. "Change in its physiological state" particularly refers to differentiation, or the transition from a viable state to a lysed state, a necrotic state or an apoptotic state.

[0031] In particular, the use of the devices and methods according to the invention can optimize biological processes, particularly the associated physical stresses, in order to determine the optimal operating conditions and adapt said stresses to the cells of interest. For example, these conditions focus on biological process parameters (temperature, flow rate, rotational speed) and the suspension fluid (viscosity, osmotic pressure). It is indeed desirable to optimize the stresses associated with the biological process in order to adapt said stresses to the data obtained during the characterization and prediction of the physiological state of the cells subjected to these physical stresses. Detailed implementation

[0032] According to a first object of the invention, the invention relates to a device for applying at least one physical stress, more specifically a hydrodynamic stress, to at least one cell in a suspended state, and for characterizing said cell during and / or after said application, said device comprising:

[0033] - a microfluidic circuit comprising at least one section, said section comprising at least: i) a main channel configured to circulate a fluid containing said cells in a suspended state, ii) a fluid inlet and a fluid outlet, iii) means for introducing and establishing a flow of said fluid in said channel, and

[0034] - means for characterizing said cells;

[0035] The device according to the invention is characterized in that said at least one section is configured to apply to said cells at least one physical stress selected from the following:

[0036] - mechanical shock,

[0037] - hydrodynamic compression stress

[0038] - hydrodynamic shear stress, and

[0039] - hydrodynamic tensile stress.

[0040] In the context of the present invention, the term "for" in "for applying physical stress" means "configured to apply physical stress". When indicating an interval or range of numerical values, the cited bounds are considered to be part of the said numerical range.

[0041] The term "microfluidic circuit" refers to a circuit designed to handle small volumes of fluid (from 10 -18 to 10 -3 liters) in channels with a diameter ranging from 5 micrometers to 3000 micrometers.

[0042] "Fluid" is understood to mean a deformable medium suitable for circulation in the device according to the invention and suitable for cell suspension.

[0043] Preferably, in the device according to the invention, the fluid is selected from Newtonian fluids, i.e., those whose viscosity does not vary with shear stress. The fluid has at least one of the following properties:

[0044] - Its Newtonian viscosity is from 1 mPa.s to 1000 mPa.s, preferably from 1 mPa.s to 100 mPa.s,

[0045] - Its pseudoplasticity index is from 0 to 1, preferably 1 mPa.s,

[0046] - It is compatible with cell viability under stress-free conditions and has an osmotic pressure of 250 mOsmol to 410 mOsmol,

[0047] - Its pH value is physiological under ambient air treatment conditions, so the pH value is from 6.5 to 7.8, and

[0048] - It contains nutrients that can be absorbed by cells in suspension, such as glucose, glutamine, or a nutrient composition such as a culture medium.

[0049] More specifically, in the device and method according to the invention, the fluid for cell suspension is preferably selected from Newtonian fluids, especially from:

[0050] - Physiological buffers, commonly used for cell suspension, such as PBS, HBSS; these buffers can be supplemented with nutrient compounds, such as a glucose solution from 0.5 g / L to 6 g / L and a glutamine solution from 2 mM to 4 mM,

[0051] - Cell culture media, such as DMEM, EMEM, α-MEM,

[0052] - Cell separation solutions, such as Ficoll and sucrose solutions,

[0053] - Low molecular weight polymer solutions.

[0054] When applying the at least one physical stress, the cell concentration is preferably less than 100 million cells / ml, preferably from 10,000 cells / ml to 10 million cells / ml. That is, the cell volume preferably occupies at most 30% of the total volume of the suspension fluid.

[0055] In the device according to the invention, the reference fluid and the fluid in which the cells are suspended circulate in a stable and pulsation-free flow manner. Thus, the device is configured to apply at least one stress to at least one cell suspended and moving in the device. The device is preferably designed to direct the cells along defined flow lines to a microfluidic chip.

[0056] In the device and method according to the invention, the intensity of the stress, also referred to in the figures as "stress level" or "stress", and the residence time under the stress, also referred to in the figures as "stress duration" or "time", are carried out in a controlled manner. The capture and release of the cells in the suspended state are also carried out in a controlled manner.

[0057] One or more pumps control the flow rate and the flow pattern. Preferably a pump with an ultra-high pressure of up to 1.37×10 5 kPa.

[0058] The circulation of the fluid within the device according to the invention is initiated and maintained by any suitable means known to those skilled in the art, such as a pump.

[0059] In the device according to the invention, the cross-section of the channel can be constant or variable, or conical, for example in the form of a nozzle. The tapered portions at the end of the channel that increase or decrease gradually exert additional stress on the cells in terms of capture and controlled release.

[0060] When the fluid containing the cells in the suspended state circulates in the microfluidic circuit, current lines are generated to control the trajectory of the cells. The cells to be characterized are located on the current lines capable of controlling the stress intensity.

[0061] The preferred diameter of the main channel is from 10 μm to 3000 μm.

[0062] The channel is made of a material suitable for fluid circulation that can withstand a stress of 10 -3 kPa to 10 6 kPa, and preferably from 1 kPa to 10 4 kPa, and / or made of a material suitable for fluid circulation with a viscosity of 1 mPa·s and 2000 mPa·s.

[0063] The capacity of cells depends on their origin (clone, species, organ), their culture mode (number of doublings during culture, culture medium, environmental conditions of the culture, i.e., whether agitated, temperature, pH), and their collection mode (trypsinization, mechanical harvesting). For example, the type of cell culture, especially 2D or 3D, their adherent or suspension culture, and the type of culture medium used affect cell capacity.

[0064] In a particular embodiment of the device according to the invention, the microfluidic circuit is composed of a set of channels made of a suitable material, said material being especially selected from the following: PEEK (polyetheretherketone), PVC (polyvinyl chloride), PTFE (polytetrafluoroethylene), FEP (fluorinated ethylene propylene copolymer), PDMS (polydimethylsiloxane) or steel. In another embodiment, the microfluidic circuit consists of a microfluidic chip made of a material selected from the following: PDMS, polyacrylate, SEBS (styrene-ethylene-butene-styrene), glass, polycarbonate or ceramic.

[0065] In a particular embodiment, the device according to the invention may include channels arranged in series and / or channels arranged in parallel.

[0066] The fluid circulation in the device according to the invention is characterized by at least one of the following parameters:

[0067] - The fluid flow rate in the channel considered for characterizing the physical stress applied to the cells is from 0 ml / min to 5 ml / min, preferably from 0 ml / min to 1 ml / min, and / or

[0068] - A fluid flow rate of from 5 µm / s to 300 m / s, and / or

[0069] - When analyzed using software, the flux of the cells to be analyzed can exceed 1000 cells / minute.

[0070] The total duration of the cells present in the channel is preferably from 1 microsecond to 10000 seconds, preferably from 1 microsecond to 100 seconds, preferably from 1 microsecond to 1 second, preferably from 10 microseconds to 100 milliseconds.

[0071] "Cells in suspension" refers to any type of cell, animal or plant cell, prokaryotic or eukaryotic cell. Depending on the diameter of the channel and the magnification of the lens used for image analysis, the device according to the invention can be used to characterize, for example, bacterial cells, algal cells or fungal cells.

[0072] The present invention particularly relates to a device for applying at least one physical stress and characterizing at least one cell in suspension, which is used for at least one of the following aspects:

[0073] - The size of the cell,

[0074] - The shape of the cell, in particular the sphericity of the cell

[0075] - The appearance of the outer membrane,

[0076] - The appearance of the cytoplasm, especially its granularity

[0077] - The presence of at least one marker on the cell surface,

[0078] - The protein content of the cell, and

[0079] - The nucleic acid content of the cell, in particular the amount of DNA, the amount of RNA, the nucleotide sequence of one or more nucleic acids, whether DNA or RNA.

[0080] "Cell characterization" should be understood to mean the definition of at least one characteristic of said cell. If more than one cell characteristic is defined, the characterization of the cell includes the definition of the combination of said characteristics.

[0081] Cell characterization enables the determination of the physiological state of the cell during or after the application of said at least one stress. Physiology studies the role, function, and mechanical, physical, and biochemical organization of cells and their components, especially organelles. Physiology also studies the interaction between cells and their environment. Determining the physiological state of a cell particularly includes the determination of the differentiation state and / or nutrition, proliferation, and related functions, such as motility and sensory functions.

[0082] This physiological state is preferably selected from the following cells: living cells, dead cells, lysed cells, necrotic cells, apoptotic cells, differentiated or undifferentiated cells, pathological or healthy cells.

[0083] The relationship between the physiological state and the various aspects of the characterization is known to the person skilled in the art.

[0084] The physiological state of the cell after the application of at least one physical stress can be further compared with the physiological state of the cell before the application of said physical stress.

[0085] For the purposes of the present invention, the ability of a cell to withstand at least one physical stress of a specific intensity and duration while maintaining a physiological state suitable for subsequent use is defined as the "capacity" of the cell.

[0086] According to a particular aspect, the present invention particularly relates to a device configured to apply at least one physical compressive stress to said cell, said device comprising at least one section of type (A), said section comprising or consisting of a first channel connected at the same level to two channels, each channel forming an angle of 30 degrees to 150 degrees with said first channel, said angle also being referred to as the "flow focusing angle".

[0087] "Hydrodynamic compressive stress" means applying a balanced force to the interior of the cell, also known as "flow focusing". In the devices and methods according to the present invention, the intensity of the compressive stress applied to the cell is 10 -3 kPa to 10 3 kPa, preferably 10 -3 kPa to 10 2 kPa, preferably 10 -3 kPa to 10 kPa. Figure 1 An example of the A-type section is schematically shown.

[0088] According to another specific aspect, the present invention particularly relates to a device configured to apply at least one hydrodynamic shear stress to the cell, the device comprising at least one B-type section, the B-type section comprising or consisting of a channel having a diameter of 10 microns to 2000 microns, preferably 20 microns to 200 microns.

[0089] "Hydrodynamic shear stress" means a mechanical stress applied parallel to or tangent to the surface of the material. In the devices and methods according to the present invention, the intensity of the compressive stress applied to the cell is 10 -3 kPa to 10 5 kPa, preferably 10 - 3 kPa to 10 4 kPa, preferably 0.1 kPa to 10 3 kPa.

[0090] Particularly, the hydrodynamic shear stress is applied during the cell circulation in a capillary channel, the diameter of which is preferably between 10 microns and 2000 microns, more preferably 20 microns to 200 microns. Figure 1 An example of the B-type section is schematically shown.

[0091] According to another specific aspect, the present invention particularly relates to a device configured to apply at least one physical tensile stress to the cell, the device comprising at least one C-type section, the C-type section comprising or consisting of: i) a channel with an increasing or decreasing cross-section, or ii) a first channel connected to a second channel, in which the fluid flows in a different direction from the first channel, preferably in the opposite direction.

[0092] "Hydrodynamic tensile stress" means applying a balanced force to the outside of the cell, or applying a tensile stress. In the devices and methods according to the present invention, the intensity of the tensile stress applied to the cell is 10 -3 kPa to 10 3 kPa, preferably 10 -3 kPa to 102 kPa, preferably 10 -3 kPa to 10 kPa. Figure 1 An example of a C-type section is schematically shown.

[0093] According to another specific aspect, the present invention particularly relates to a device configured to apply at least one mechanical shock to the cells, the device comprising at least one D-type section, the D-type section comprising or consisting of a first channel in which the cell path encounters an obstacle, such as in particular the wall of a second channel. Figure 1 An example of a D-type section is schematically shown.

[0094] "Mechanical shock" refers to any type of mechanical shock, such as a shock against a wall or an inertial collision with a surface. In the device and method according to the present invention, the intensity of the mechanical shock applied to the cells is from 1 m / s to 300 m / s, from 1 m / s to 100 m / s, preferably from 1 m / s to 10 m / s. The duration of the shock time is preferably less than 1 microsecond.

[0095] More specifically, the device according to the present invention for applying at least one physical stress and characterizing at least one cell in suspension comprises a microfluidic circuit, the microfluidic circuit comprising or consisting of:

[0096] - at least one A-type section, and / or

[0097] - at least one B-type section, and / or

[0098] - at least one C-type section, and / or

[0099] - at least one D-type section,

[0100] The sections are combined with each other.

[0101] More specifically, the device according to the present invention for applying at least one physical stress and characterizing at least one cell in suspension comprises a microfluidic circuit, the microfluidic circuit comprising or consisting of:

[0102] - at least one A-type section, and / or

[0103] - at least one B-type section, and / or

[0104] - at least one C-type section,

[0105] - and optionally at least one D-type section,

[0106] The sections are combined with each other.

[0107] The device according to the invention is specifically designed to apply a sequence of the same type of physical stress, including one or more than one iteration, at a determined frequency and intensity, and / or to apply a sequence of several different types of physical stress at a determined frequency and intensity.

[0108] "Applying at least one physical stress of a determined intensity and duration" means:

[0109] - applying at least once a specific physical stress of a determined intensity and duration, optionally followed by one, two, three, four or more than four repetitions of said specific physical stress, and / or

[0110] - applying at least once a sequence of at least two specific physical stresses of a determined intensity and duration, optionally followed by one, two, three, four or more than four sequences of at least two physical stresses, and / or

[0111] - applying any type of physical stress sequence described in the present application.

[0112] According to another specific aspect, the invention particularly relates to a device configured to apply to said cells at least one sequence comprising or consisting of at least two successive physical stresses of different natures. This sequence may be repeated once, twice, three times or more than three times.

[0113] According to another specific aspect, the invention particularly relates to a device configured to apply to said cells at least one sequence comprising or consisting of at least two successive physical stresses of the same nature. According to this specific aspect, the device is configured to apply to said cells a physical stress repeated at least 1 time, at least 2 times, at least 3 times, at least 4 times, at least 5 times, at least 6 times, at least 7 times, at least 8 times, at least 9 times, at least 10 times or even more than 10 times.

[0114] Figure 2 An example of this type of device according to the invention is shown, which shows a device designed to apply a large amount of compressive stress to cells, the device being separated by sections designed to ensure that the cells are not subjected to stress. These physical stress sequences can be considered equivalent to applying dynamic cell deformation. These sequences are repetitions of the same stress sequence.

[0115] More specifically, the device according to the invention is characterized in that the total intensity of said at least one physical stress applied to the cells is 10 -3 kPa to 10 5 kPa, preferably 10 -3 kPa to 10 4 kPa, preferably 0.1 kPa to 10 3 kPa, preferably 1 kPa to 102 kPa.

[0116] Furthermore, more specifically, the device according to the present invention is characterized in that the total duration of applying the at least one physical stress is from 1 microsecond to 10,000 seconds, preferably from 1 microsecond to 100 seconds, preferably from 1 microsecond to 1 second, preferably from 10 microseconds to 100 milliseconds.

[0117] In the microfluidic device according to the present invention, during and / or after applying at least one physical stress to the cells, the at least one cell in suspension is characterized. According to a first embodiment, during applying at least one physical stress to the cells, the at least one cell in suspension is characterized.

[0118] According to another embodiment, the device according to the present invention is characterized in that after applying the at least one stress, the cells are characterized, and this characterization is carried out within a time period of 0 days to 120 days, preferably 0 days to 30 days, preferably 0 days to 1 day after applying the at least one physical stress.

[0119] More specifically, the device according to the present invention is characterized in that after applying the at least one stress, the cells are characterized, and the characterization comprises or consists of: at least one discrete characterization or at least one characterization with a duration of from 1 microsecond to 10,000 seconds, preferably from 1 microsecond to 100 seconds, preferably from 1 microsecond to 1 second, preferably from 10 microseconds to 100 milliseconds.

[0120] More specifically, the device according to the present invention is characterized in that the device for characterizing the at least one cell is selected from: a cell counter, a microscope, a device for analyzing the protein content of cells, a device for analyzing and sequencing the nucleic acids of cells, and an image and / or electrical signal capture device combined with a signal analysis device. For example, the device according to the present invention may include microelectrodes or photodiodes.

[0121] More specifically, the device according to the present invention is characterized in that the signal and / or image analysis device comprises or consists of a central computer unit, and the central computer unit includes software devices suitable for signal and / or image analysis.

[0122] Even more specifically, the device according to the present invention including a signal and / or image analysis device further includes a first classification model and / or a second classification model. The presence of at least the first classification model and / or the second classification model has the advantages of accelerating the image analysis process and achieving real-time analysis.

[0123] "Classification model" refers to a pre-trained machine learning algorithm, particularly a supervised learning algorithm, as well as the training data set and evaluation data set used to train said algorithm. The classification model may consist of a computer program, which can be written in any suitable computer language known to those skilled in the art. The computer program is capable of being implemented on a computer to produce technical results. Examples of these technical results are described below.

[0124] The training data set may include a training set and a model test set. In this way, the model can be tested on a test basis, and the test set can be used to determine whether the model learning is satisfactory. The training set and the test set may be different. Alternatively, the test set may correspond to a part of the training set.

[0125] Even more specifically, the device according to the present invention including an image analysis device further includes a first classification model, which is pre-trained with a training data set and includes a supervised machine learning algorithm, an unsupervised machine learning algorithm or a semi-supervised machine learning algorithm. The first classification model is applicable to predicting the physiological state of a given cell based on at least one feature of the shown cell.

[0126] In a specific embodiment of the first classification model, the input data is an image of an object. The output data is the object with labels: the percentage of members in a specific classification. The training algorithm includes at least 10 epochs, the loss calculation is cross-entropy, and the optimizer is Adam (improved gradient descent). The model is a transfer learning with YOLO, and the nature of the network is a supervised model (1733 cell images / 1733 annotations). The model preferably includes 106 convolutional layers. The functions performed in the neurons are: convolution, addition, softmax, up sampling. Neuron / layer connections are formed. The training data set may include a plurality of data pairs, each data pair including a first data item representing at least one feature of the cell and a second data item representing the physiological state of the cell.

[0127] By analyzing the characteristics of cells whose physiological states have been determined, a training data set can be pre-established from the data obtained in the laboratory.

[0128] In particular, the first classification model can be implemented on a computer to generate technical results, which consist of, for example, the classification of cells according to their characteristics.

[0129] The first classification model is used to generate a three-dimensional map, for example, for a given unit, represented by:

[0130] - The intensity of the physical stress applied to the cell, in Pa,

[0131] - The duration of the physical stress applied to the cells, or the dwell time, or "time", expressed in seconds,

[0132] - The physiological state of the cells, namely cell survival, necrosis, apoptosis or lysis, expressed as a percentage, or the nature of differentiation or undifferentiation.

[0133] In particular, if the classification reaches a minimum F1 score of 70%, the first classification model is considered to have achieved a satisfactory learning level on all profiles in the test set.

[0134] Even more specifically, when the device according to the invention includes an image analysis device, the image analysis device further includes a second classification model, which is pre-trained with a training data set and includes a supervised automatic learning algorithm, an unsupervised automatic learning algorithm or a semi-supervised automatic learning algorithm, and the second classification model is suitable for detecting and monitoring the deformation of a given cell in response to at least one physical stress.

[0135] In particular, the first classification model can be implemented on a computer to generate a technical result, which lies in, for example, monitoring the morphological evolution of cells according to different applications of physical stress.

[0136] Preferably, the second classification model uses at least one neural network, and its functions are as follows:

[0137] i) Locate and isolate cells from the said image

[0138] ii) Verify the presence of single cells

[0139] iii) Improve the image quality, and then process the image by applying a mask to the cells and determining their contours.

[0140] iv) Locate and measure the major axis and minor axis of the ellipse depicting the cell contour. The deformation is defined as the ratio of the major axis to the minor axis.

[0141] According to a specific embodiment, the second classification model includes: input data: cell images of 440 gray-scale adjusted slices, output data: a segmented binary mask. The training data includes 420 training images. For the training algorithm, at least 10 epochs are required, the loss calculation is cross-entropy, and the optimizer is Adam (improved gradient descent). The nature of the network is U-net. The number of layers is: 5 contracting layers, 5 expanding layers, a total of 10 layers. The functions executed in the neurons are: 2D convolution between the image and the filter, that is, compressing the image, extracting the feature vector containing the object of interest, and decompressing the image. The characteristics of the neuron / layer connection are that the layer consists of two convolutions, both of which follow the activation function (ReLU).

[0142] Examples of cell location and isolation areFigure 3 as shown. Examples of how to locate and measure the major and minor axes of an ellipse are Figure 4 as shown.

[0143] If the classification reaches a minimum F1 score of 65%, preferably at least 80%, the second classification model is considered to have achieved a satisfactory learning level on all profiles in the test set.

[0144] According to a second object, the present invention relates to a method for applying at least one physical stress of a determined intensity and duration to at least one cell in a suspended state, and for characterizing the cell after said application, the method comprising the following steps:

[0145] Depositing and circulating a fluid containing said at least one cell in a suspended state in a microfluidic circuit comprising at least one section, said section comprising at least: i) a main channel configured to circulate the fluid containing the cells, ii) a fluid inlet and a fluid outlet, iii) a means for introducing and establishing a flow of the fluid within said channel, and

[0146] Characterizing the cell after applying said at least one stress,

[0147] The device according to the invention is characterized in that said at least one section is configured to apply to the cells at least one physical stress selected from the following:

[0148] - Mechanical shock,

[0149] - Hydrodynamic compression stress,

[0150] - Hydrodynamic shear stress, and

[0151] - Hydrodynamic tensile stress.

[0152] The method according to the invention particularly relates to the application of a physical stress and the characterization of at least one cell in a suspended state, said characterization relating to at least one of the following aspects: size, shape, appearance of the outer membrane, appearance of the cytoplasm, presence of at least one marker on the cell surface, protein content of the cell and nucleic acid content of the cell.

[0153] More specifically, the present invention relates to a method according to the invention, the method comprising applying at least one physical stress to at least one cell in a suspended state, said physical stress being characterized in that the cell is subjected to:

[0154] - At least one mechanical shock having an intensity of from 1 m / s to 300 m / s, preferably from 1 m / s to 100 m / s, preferably from 1 m / s to 10 m / s and / or having a duration of less than 1 microsecond, and / or

[0155] - At least one shear stress with a strength of 10 -3 kPa to 10 5 kPa, preferably 10 -3 kPa to 10 4 kPa, preferably 0.1 kPa to 10 3 kPa, preferably 1 kPa to 10 2 kPa and / or its duration is from 1 microsecond to 10,000 seconds, preferably from 1 microsecond to 100 seconds, preferably from 10 microseconds to 1 second, and / or

[0156] - At least one compressive stress with a strength of 10 -3 kPa to 10 5 kPa, preferably 10 -3 kPa to 10 4 kPa, preferably 0.1 kPa to 10 3 kPa, preferably 1 kPa to 10 2 kPa and / or its duration is from 1 microsecond to 10 seconds, preferably from 1 microsecond to 1 second, preferably from 10 microseconds to 10 milliseconds, and / or

[0157] - At least one tensile stress with a strength of 10 -3 kPa to 10 5 kPa, preferably 10 -3 kPa to 10 4 kPa, preferably 0.1 kPa to 10 3 kPa, preferably 1 kPa to 10 2 kPa and / or its duration is from 1 microsecond to 10 seconds, preferably from 1 microsecond to 1 second, preferably from 10 microseconds to 10 milliseconds.

[0158] More specifically, in the method according to the present invention, the parameters of i) fluid flow rate and ii) fluid viscosity are selected to reproduce the stress intensity and residence time, reproducing the stress experienced by cells in a specific biological process. When the fluid flows through the device, streamlines are generated. When the fluid flows through the device, the cells follow the streamlines formed by the fluid motion and its interaction with the channel geometry. By calculating or measuring the hydrodynamic stress and the movement speed of the cells along this line, the stress and residence time of the cells can be deduced.

[0159] Preferably, in the method according to the present invention, the fluid flow rate is 10 -3 ml / min to 10 ml / min, preferably at 10 -3 ml / min to 1 ml / min. Preferably, in the method according to the present invention, the residence time of the cells is from 1 second to 10,000 seconds, preferably from 1 microsecond to 1 second, preferably from 10 seconds to 100 milliseconds.

[0160] More specifically, the present invention relates to a method according to the present invention, the method comprising applying at least one physical stress to at least one cell in a suspended state, subsequently characterizing the cell, and the method further comprising predicting the physiological state of the cell by means of a first pre-trained classification model based on the characteristics determined in step b).

[0161] More specifically, the present invention further relates to a method according to the present invention, the method comprising applying at least one physical stress to at least one cell in a suspended state, subsequently characterizing the cell, and the method further comprising predicting the physiological state of the cell by means of a first pre-trained classification model based on the characteristics determined in step b), the first classification model comprising: a machine learning algorithm pre-trained with a training data set, a supervised learning neural network, a semi-supervised learning neural network or an unsupervised learning neural network.

[0162] More specifically, the present invention further relates to a method according to the present invention, the method further comprising monitoring the deformation of the cell at different times during its residence in the microfluidic channel by means of a second pre-trained classification model based on the characteristics determined in step b).

[0163] According to a third aspect, the present invention further relates to a classification model which is pre-trained on a training data set to predict the physiological state of a cell after applying at least one physical stress in a method according to the present invention.

[0164] According to a fourth aspect, the present invention relates to the use of a device or method according to the present invention, or a classification model according to the present invention, to characterize the following types of cells: prokaryotic cells, eukaryotic cells, animal cells, plant cells, human cells, stem cells, epithelial cells, fibroblasts, blood cells, genetically modified cells or synthetic cell mimics.

[0165] More specifically, according to this fourth aspect, the present invention relates to the use of a device or method according to the present invention, or a classification model according to the present invention, to determine the capacity of a cell.

[0166] Even more specifically, according to this fourth aspect, the present invention relates to the use of a device or method according to the present invention, or a classification model according to the present invention, to define at least one parameter of a biological process.

[0167] Even more specifically, the present invention relates to the use of a device or method according to the present invention, or a classification model according to the present invention, to define at least one parameter of a biological process selected from: bioprinting, cell therapy and cell production.

[0168] The present invention can be better understood by reading the following embodiments, which are provided to illustrate the invention and not to limit its scope. In particular, if the selection of features disclosed below is sufficient to confer a technical benefit or to distinguish the invention from the prior art, it is contemplated that variants of the invention will include only the selection of features disclosed below and not the other features disclosed. BRIEF DESCRIPTION OF THE DRAWINGS

[0169] Figure 1 is a schematic diagram of an example of section A, section B, section C, and section D.

[0170] Figure 2 is a schematic diagram of an example of a device in which several types of physical stress are applied to cells, in this case a series of shear stress and elongation stress. The following figure shows an enlarged view of the above figure. The cells are subjected to repeated stress.

[0171] This type of stress can also be defined as dynamic deformation, or oscillatory deformation.

[0172] Figure 3 shows the steps involved in separating cells.

[0173] Figure 4 shows the measurement of the minor axis and major axis of an ellipse.

[0174] Figure 5 , in Example 2, shows the viability state of AD-MSC human mesenchymal stem cells after applying shear stress of different intensities and durations. The percentage of cells in a specific physiological state is given as a function of the duration (time) in seconds and the intensity (stress) in Pa of the applied shear stress. Figure 5 shows the percentages of viable cells (box A), lysed cells (box B), necrotic cells (box C), and apoptotic cells (box D).

[0175] Figure 6 , in Example 3, shows the viability state of fibroblasts after applying shear stress. Fibroblasts are characterized after applying shear stress of different intensities and durations. The percentage of cells in a specific physiological state is a function of the duration (time) in seconds and the intensity (stress) in Pa of the applied shear stress. Figure 6 shows the percentages of viable cells (box A), lysed cells (box B), necrotic cells (box C), and apoptotic cells (box D).

[0176] Figure 7, in Example 4, the proportions of undifferentiated AD-MSC stem cells (white circles) and differentiated cells (black circles) are shown after the application of shear stress, as a function of the duration of application (time) expressed in seconds and the intensity of shear stress (stress) expressed in Pa.

[0177] Figure 8 , in Example 5, the viability status of HEK293T cells is shown after the application of tensile stress. HEK293T cells after the application of tensile stress of different intensities and durations are characterized. The percentage of cells in a specific physiological state is shown as a function of the duration of application of tensile stress (time) expressed in seconds and the intensity (stress) expressed in Pa. Figure 8 The percentages of apoptotic cells (box A), viable cells (box B), necrotic cells (box C), and lysed cells (box D) are shown.

[0178] Figure 9 , in Example 6, the viability status of fibroblasts after the application of shear stress is shown, and the viability is measured by a cell counter (histogram) or trypan blue staining (black circles).

[0179] Figure 10 , in Example 7, the viability status of HEK293T cells is shown after the application of shear stress of different intensities and durations. The percentage of viable cells measured by trypan blue staining is shown as a function of the duration of application of shear stress (time) expressed in seconds and the intensity (stress) expressed in Pa. The percentage of viable cells is shown after a small number of passages (HEK293T P07, white triangles) or a large number of passages (HEK293T P18, black circles).

[0180] Figure 11 , in Example 8, the viability status of fibroblasts after the application of shear stress of different intensities and durations is shown according to two series of measurements on the same cell sample (P06), represented by white triangles and black circles respectively. P06 represents the 6th passage of cell culture.

[0181] Figure 12 , in Example 9, the percentage of cell viability as a function of hydrodynamic stress intensity and residence time is shown in three dimensions and on a logarithmic scale.

[0182] Figure 13 , in Example 9, the percentage viability of fibroblasts is shown on the y-axis in two dimensions as a function of hydrodynamic shear stress intensity on the x-axis, and the residence time is depicted according to a point pattern.

[0183] Figure 14, in Example 9, the percentage of cell viability as a function of hydrodynamic stress intensity and residence time is shown in two dimensions.

[0184] Figure 15 , in Example 10, a histogram depicting cell viability as a function of shear stress intensity in kPa is shown. For each stress intensity, the viability values were determined after several experiments.

[0185] Figure 16 , in Example 10, a plot of the standard deviation (light bars), standard error (black bars), and coefficient of variation (gray bars) of various measurements as a function of shear stress intensity in kPa is shown.

[0186] Figure 17 , in Example 10, a histogram depicting cell viability as a function of shear stress intensity in kPa is shown.

[0187] Figure 18 , in Example 11, the following are shown: i) on the left, the function of the applied hydrodynamic stress intensity and the percentage of cell viability of different cell types, in kPa, and ii) on the right, for each cell studied, the function of the applied hydrodynamic stress and the percentage of cell viability.

[0188] Figure 19 , in Example 12, the function of the intensity of the applied shear stress and the percentage of cell viability, in kPa, is shown. The symbols P07 (circles), P08 (squares), P09 (triangles), and P10 (stars) represent the percentage of cell viability defined according to the number of cell passages in culture.

[0189] Figure 20 , in Example 12, the function of the intensity of the applied shear stress and the percentage of cell viability, in kPa, is shown. P09, P10, and P11 represent the percentage of cell viability defined according to the number of cell passages in culture.

[0190] Figure 30A , in Example 13, the same sequence of hydrodynamic stresses is shown.

[0191] Figure 30B , in Example 13, the function of the intensity of the applied hydrodynamic stress and the percentage of cell viability, in kPa, is shown.

[0192] Figure 31 , in Example 13, the function of the number of stress application cycles and the percentage of cell viability for a stress of 0.2 kPa is shown.

[0193] Survival, in Example 14, a diagram showing the steps of localizing and separating cells from the original image.

[0194] A = -4.009e-12, B = 4.046, C = 90.26, D = 30.6 , in Example 15, the steps of detecting the cell axis from the original image are shown.

[0195] Lysis , in Example 15, the deformability of cells under shear stress based on their initial diameter is shown.

[0196] A = 9.183e-13, B = 4.215, C = 4.172, D = 30.6 , in Example 15, the deformability of cells under the influence of mechanical shock stress based on their initial diameter is shown.

[0197] Necrosis , in Example 16, a "convolutional autoencoder" neural network is schematically shown, along with an example of converting an initial image (input) into a final image (output) after processing by this network.

[0198] A = 9.872e-08, B = 2.474, C = 2.918, D = 30.6 , in Example 16, the differences in image processing of the initial image by a supervised network (U-net) with preprocessing or an unsupervised network (autoencoder) without preprocessing are shown; the first row shows the result without applying the autoencoder, the second row shows the result of applying the "variable autoencoder", and the third row shows the result of applying the "denoising autoencoder".

[0199] Apoptosis and A = 43.1, B = -0.6166, C = 0.8814, D = 30.6 , in Example 17, the contour results after applying different neural networks and related metrics are shown, and then the differences in deformability are calculated relative to the manually completed contour (with 0 as the control).

[0200] Figure 5 , in Example 18, the classification results of three types of cells are shown, from left to right: necrotic cells, apoptotic cells, and live cells.

[0201] Example

[0202] Example 1: Materials and Methods for Characterizing Cells after Applying Physical Stress

[0203] AD-MSC cells were cultured in MSC-growth medium at an inoculation concentration of approximately 2500 cells / cm 2 . The medium was changed every two days. To achieve 70% confluence, the cells were cultured in a T175 culture flask for seven days.

[0204] Fibroblasts were cultured in DMEM GLUTAMAX - Gibco medium at an inoculation concentration of approximately 5500 cells / cm 2。To achieve 80% confluence, the cells were incubated in a T175 culture flask for seven days.

[0205] HEK293T cells were cultured in DMEM GLUTAMAX - Gibco medium at an inoculation density of approximately 12,000 cells / cm 2 。To achieve 80% confluence, the cells were cultured in a T175 culture flask for four days. During the culture, the temperature was maintained at 37 °C and the CO2 concentration was approximately 5%.

[0206] To detach the cells from the culture flask surface, the medium was first removed. The cells attached to the flask were rinsed with 15 mL of PBS. Subsequently, 5 mL of 0.5% trypsin - EDTA was added to the flask. At 37 °C, the trypsin was applied for two minutes, and then 10 mL of medium containing fetal bovine serum was added to the flask to terminate the reaction. The suspension was placed in a test tube and centrifuged at 1200 rpm / 210 g for 5 minutes. The liquid was removed, and the cells were resuspended in a solution of PBS and Ficoll at a concentration of one million cells per milliliter.

[0207] The microfluidic shear stress device consists of a PDMS microfluidic channel with a diameter of 50 μm and a length of 10 cm. The cells were suspended in a fluid (solution of PBS and Ficoll) with a viscosity of 1.91 mPa·s.

[0208] The microfluidic device for elongation stress consists of a main microfluidic channel with a diameter of 162 μm and a length of 1 cm. The cross - section of this channel first decreases from 162 μm to 30 μm and then increases from 30 μm to 162 μm. This change in cross - section occurs over a distance of 360 μm. The viscosity of the suspending fluid (solution of PBS and Ficoll, 60% Ficoll by volume concentration) is 1.91 mPa·s.

[0209] The cell suspension and cell - free liquid were injected into the device using an injection pump and a 3 - mm - diameter PEEK tube.

[0210] The cells were injected into the device at a flow rate ranging from 25 μL per minute to 800 μL per minute. After achieving flow stability within the system, measurements and cell harvesting were performed.

[0211] After passing through the stress region, the cells were harvested. The harvested cell suspension fluid was centrifuged to remove the suspension fluid (Ficoll solution and PBS) and replaced with labeling buffer.

[0212] For the viability test, cells were stained with Annexin V, a marker of apoptotic cells, or propidium iodide, a marker of necrotic cells. To this end, a population of 100,000 cells corresponding to each injection rate was suspended in 100 μl of labeling buffer. Subsequently, 2 μl of Annexin V marker and 2 μl of propidium iodide were added to the buffer. The cell suspension with viability markers was incubated in the dark for 15 minutes. Then the cells were centrifuged and washed with labeling buffer.

[0213] To detect the stemness and differentiation status of AD-MSC stem cells, cells were labeled using the BD Human Mesenchymal Stem Cell Analysis Kit (BDB562245). This kit contains hMSC positive markers CD90, CD105, CD73, and CD44, as well as hMSC negative markers CD34, CD11b, CD19, CD45, and HLA-DR.

[0214] For the staining procedure, 100,000 cells were suspended in 100 μl of BD staining buffer (FBS), and then 5 μl of each positive marker and 20 μl of each negative marker were added to the buffer. The cell suspension with markers was incubated in the dark for 15 minutes. Then the cells were centrifuged and washed with labeling buffer.

[0215] Cells were studied using a cell sorter (FACS Canto II) and characterized according to their status: viable cells, lysed, necrotic, apoptotic, or stem. Each measurement point corresponded to the analysis of at least 50,000 cells.

[0216] Based on the obtained results, a mathematical model was developed to predict the physiological state of cells based on stress intensity and duration. The model is a mathematical expression related to parameters, such as polynomials, power laws, sums of sines, or other 2D or 3D mathematical expressions. For example, we can use a power law of the form y = ax k + c, which establishes a relationship between x and y. In this model, a is the proportionality constant, k is the exponent, and c is the error term. Using this method, we modeled the physiological state of cells before and / or after passing through the stress region based on stress intensity and stress residence time. To this end, the model was established as follows: physiological state = A × stress B + C; residence time = D / stress. The parameters A, parameter B, parameter C, and parameter D are different for each cell type, stress type, and passage number.

[0217] Example 2: Characterization of the viability state of AD-MSC stem cells after applying shear stress

[0218] AD-MDC stem cells were cultured and subjected to shear stress of different intensities and durations. After applying these stresses, the viability of the cells was characterized based on the intensity of the shear stress and the residence time of the cells under the stress. The physiological state (survival, lysis, necrosis, or apoptosis) of the cells was characterized as in Example 1.

[0219] The mathematical model defined in this case is as follows:

[0220] State = A × stress B + C; Time = D / stress.

[0221] The values of A, B, C, and D defined for each possible physiological state are as follows:

[0222] Table 1

[0223] Survival Stress ≤ 800 Pa A = -3.105e-12, B = 5,288, C = 86.53, D = 30.6 Survival Stress > 800 Pa A = 1.342e+16, B = -4.928, C = 2.308, D = 30.6 Lysis Stress < 800 Pa

[0224] The results obtained are shown in A = 1.003e-15, B = 5,288, C = 7.159, D = 30.6 These results indicate that, under stress, human AD-MSC stem cells are sensitive to the intensity of shear stress and the residence time. We observed that the mechanical capacity of AD-MSC cells to stress is approximately 1000 Pa, and the residence time is approximately 0.25 seconds. Beyond these values, AD-MSC stem cells can no longer withstand the stress and die. Death by lysis or necrosis is the most common pathway, and the level of apoptosis is negligible within this range of stress and residence time.

[0225] Example 3: Physiological state of fibroblasts after applying shear stress

[0226] Fibroblasts were cultured and then subjected to shear stress of different intensities and durations. After applying these constraints, the viability, lysis, necrosis, or apoptosis of the cells was determined as described in Example 1.

[0227] The mathematical model defined in this case is as follows:

[0228] State = A × stress B + C; Time = D / stress.

[0229] The values of A, B, C, and D defined for each possible physiological state are as follows:

[0230] Table 2

[0231] Lysis Stress > 800 Pa A = -1.304e+16, B = -4.928, C = 84.04, D = 30.6 Necrosis A = 1.054e-14, B = 4.52, C = 5.476, D = 30.6 Apoptosis Figure 6 Figure 7 Figure 8 Figure 6 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 <![CDATA[A = -1.355 e +05, B = -1.881, C = 4.821, D = 30.6]]>

[0232] The results obtained are shown in Figures 15 to 17Among them. These results indicate that under stress, fibroblasts are sensitive to shear stress intensity and residence time. We can see that fibroblasts can withstand a shear stress of up to 900 Pa and a duration of 0.04 seconds without being damaged, that is, without showing lysis, apoptosis or necrosis. These values represent the mechanical capacity of fibroblasts to withstand shear stress. However, in addition to this mechanical capacity, the physiological response of fibroblasts, namely the loss of viability, is quite significant. In fact, fibroblasts cannot withstand stress, and the viability level of 85% at 900 Pa drops suddenly to 30% at 1000 Pa and reaches a viability level of less than 5% at 2000 Pa. Fibroblast death is more caused by the lysis pathway, and other physiological states can be ignored.

[0233] Example 4: Differentiation state of AD-MSC human mesenchymal stem cells after applying shear stress

[0234] Culture AD-MSC stem cells as shown in Example 1. As a prerequisite, before applying shear stress of different intensities and durations, characterize the AD-MSC stem cells by flow cytometry to confirm their stemness. After passing through the stress area, harvest the cells and resuspend them in DMEM(+) medium. Culture the cells again to characterize the differentiation state. Inoculate at 80% confluence. After incubating for three weeks, harvest the cells, label them with the BD Human Mesenchymal Stem Cell Analysis Kit (BDB562245), and analyze by flow cytometry (FACS Canto II). Each measurement point corresponds to the analysis of at least 30,000 cells. Use flow cytometry for characterization, and the cell counter identifies the state of the cells after applying stress in a binary manner: differentiated or undifferentiated. Perform the flow cytometry characterization procedure as shown in Example 1.

[0235] The mathematical model defined in this case is as follows: Undifferentiated state: time = D / stress, where D = 30.6, 0.147 s < time < 0.39 s, 78 Pa < stress < 2082 Pa.

[0236] The results obtained are shown in Figure 18 Among them. These results indicate that there are no differentiated cells. This shows that after undergoing shear stress within the range shown in the figure, AD-MSC human mesenchymal stem cells maintain their stemness. Therefore, the physiological state of AD-MSC cells, here the maintenance of stemness, is not affected by stresses from 0 Pa to 2100 Pa and residence times from 0.015 s to 0.4 s.

[0237] Example 5: Viability state of HEK293T cells after applying elongation stress

[0238] HEK293T cells were cultured and then subjected to elongation stresses of different intensities and durations. After applying these stresses, the viability, lysis, necrosis or apoptosis of the cells was determined as described in Example 1.

[0239] The results obtained are shown in Figure 19 . These results indicate that HEK293T cells have high viability when facing elongation stresses of 10 Pa to 110 Pa and dwell times of 0.1 milliseconds to 0.8 milliseconds. Lysis, necrosis and apoptosis are negligible compared to the viable state.

[0240] Example 6: Determination of the viability state of fibroblasts by two protocols after applying shear stress

[0241] Fibroblasts were cultured and then subjected to elongation stresses of different intensities and durations. After applying these stresses, two different protocols were used to determine cell viability: cell counting analysis and trypan blue counting. Viability was shown based on the shear stress intensity. The dwell time under stress corresponds to the following equation:

[0242] Time (seconds) = 30.6 / Shear stress intensity (Pa)

[0243] The cells were harvested after passing through the stress zone. Then they were divided into two batches for cell counting and trypan blue analysis respectively. For cell counting analysis, this batch was stained with annexin V and propidium iodide. The labeled cells were analyzed by a cell counter. Each measurement point corresponded to the analysis of at least 50,000 cells. For trypan blue analysis, the cells in this batch were mixed with trypan blue. Three counts were performed over the entire surface of a Malassez counting chamber. Error bars corresponded to these three counts. The results obtained are shown in Figure 20 .

[0244] Example 7: Viability state of HEK293T cells after applying shear stress

[0245] HEK293T cells were cultured and then subjected to shear stresses of different intensities and durations as described in Example 1. The HEK293T cells were harvested after passing through the stress zone and mixed with trypan blue. Three counts were performed over the entire surface of a Malassez counting chamber. The values corresponded to the average of these three counts.

[0246] The results obtained are shown in Figure 21Among these results, it is shown that the number of cell passages in cell cultures significantly affects the mechanical capacity of HEK293T cells. These cells with passage P07 (7 passages) are completely stress-tolerant and have a 100% survival rate for stresses below 200 Pa. However, beyond this mechanical capacity, the viability of HEK293T cells of the same passage P07 decreases. However, the same cell line with passage P18 (18 passages) loses this mechanical capacity and cannot withstand the slightest stress.

[0247] Example 8: Demonstration of measurement reproducibility

[0248] Characterization of the physiological state of fibroblasts after application of shear stress was performed twice. Both measurements were performed on the same cell sample (P06). The average error of the reproducibility of the experiment was 3.45% viability. Thus, the analysis performed by the system is reproducible. The results obtained are shown in Figure 22 Among.

[0249] Example 9: Mechanical capacity of cells based on stress intensity and stress dwell time

[0250] The dwell time is defined as the time during which the cells are exposed to hydrodynamic stress. In the microfluidic device and method according to the present invention, the stress is maintained at a constant intensity, while the dwell time depends on the flow rate and viscosity of the cell suspension. This method makes it possible to separate these two parameters and study their effects independently. Fibroblasts are suspended in fluids with viscosities of 1 mPa·s, 5 mPa·s, 10 mPa·s, 15 mPa·s, or 20 mPa·s and injected through a microfluidic capillary at flow rates from 117 μl / min to 942 μl / min. This configuration covers a range of shear stress intensities from 0.16 kPa to 25.59 kPa and dwell times from 12.5 ms to 100.7 ms.

[0251] Figure 23 And Figure 24 show plots of fibroblast viability as a function of stress intensity and dwell time. These two plots show the same data in 3D and 2D views and in linear and logarithmic scales. The white dots are experimental data and the surface represents the mathematical model fit. Within the range shown, for a given dwell time, increasing the stress intensity decreases cell viability. However, for a constant stress intensity, cell viability is independent of the dwell time. Figure 24 show fibroblast viability as a function of stress intensity only, with the gray shading indicating the dwell time. Due to the three representations of the same data, it can be concluded that the viability and mechanical capacity of fibroblasts do not depend on the dwell time from 12.5 ms to 100.7 ms.

[0252] These results show that, in this case, the viability of fibroblasts can only be characterized based on stress intensity.

[0253] In the method according to the present invention, the stress intensity and the residence time are completely separated. One of the parameters can be kept constant while the other is changed. In this way, the influence of the residence time and the shear stress intensity as two independent parameters can be understood in detail.

[0254] Example 10: Equipment reproducibility and precision

[0255] To ensure the reproducibility of the experiments and results, the inventors determined the mechanical capacity of Madin-Darby canine kidney (MDCK) cells six times. To eliminate the biological variability that could affect the error assessment of the machine and program performance, a group of cells was divided into six batches. The mechanical capacity of each batch of cells was determined. The cells were exposed to shear stresses of 0.175 kPa, 0.469 kPa, 2.716 kPa, and 6.397 kPa, and then their viability was determined by a cell counter. The standard deviation, standard error, and coefficient of variation were calculated using the following formulas:

[0256]

[0257] The results showed that, in the worst case, the standard error of the six trials was less than 1%. This example perfectly shows the precision and performance of the machine, as well as the process leading to reliable and very precise determinations. This precision allows highly reliable advanced analyses, such as Figure 24 as shown.

[0258] Example 11: Mechanical capacity of different cell lines

[0259] Using the equipment according to the present invention, the mechanical ability of different cells to resist hydrodynamic stress can be measured: primary cell lines, immortalized cell lines, stem cells, etc. The mechanical capacity of the following cell lines was analyzed: adipose tissue-derived mesenchymal stem cells (AD-MSC), Wharton's jelly-derived mesenchymal stem cells (WJ-MSC), bone marrow-derived mesenchymal stem cells (BM-MSC), human dermal fibroblasts, human umbilical vein endothelial cells (HUVEC), cancer-associated fibroblasts (CAF), human embryonic kidney 293T cells (HEK293T), and Madin-Darby canine kidney (MDCK).

[0260] Cells were suspended in two fluids with viscosities of 1 mPa·s and 5 mPa·s, and the cells were injected through a microfluidic channel with a diameter of 50 μm at a flow rate of 85 μL / min to 942 μL / min, and the cells were subjected to a shear stress of 0.11 kPa to 6.4 kPa. The collected cells were stained with propidium iodide and analyzed by flow cytometry to quantify the necrotic cell population caused by shear stress. To confirm that the cells were damaged only by shear stress during injection into the microfluidic channel, rather than by static syringe stress, fibroblasts were injected without passing through the microfluidic channel and analyzed by flow cytometry. We confirmed that there was no change in cell viability compared to the viability of the control cell population of 97%. In addition, for adherent cells, special attention was paid to limiting their death due to suspension conditions during the experiment. In addition, the time between cell harvest and their exposure to shear stress was only 5 minutes, and this was the same for all cell lines.

[0261] Figure 28 The viability state of the cell lines after the application of shear stress is shown. The dots are experimental data and the lines are the fitted curves. This cell viability evolution curve can be described as having two phases. First, the cell viability shows a steady state when reaching a certain shear stress limit, that is, reaching the ability of the cells to withstand shear stress. Then, after the shear stress corresponding to the mechanical capacity of the cells, the cell viability deteriorates and drops to less than 5% of the viable cell viability. Before the shear stress is associated with the mechanical capacity of each cell, the change in viability is less than 5% compared to the control cell population not subjected to shear force. It is also worth noting that although all the cell types studied showed similar trends, each cell line had a specific sensitivity to shear stress as the mechanical capacity, AD-MSC: 0.64 kPa, WJ-MSC: 0.35 kPa, BM-MSC: 0.48 kPa, fibroblasts: 0.64 kPa, HUVEC: 0.32 kPa, and HEK293T: 0.43 KPa.

[0262] Example 12: Evolution of mechanical capacity with cell aging and culture time

[0263] To explore the link between cellular senescence and mechanical capacity, fibroblasts were passaged (subcultured) once a week and subjected to shear stress. The experiment lasted for four generations from P07 to P10. Passaging is the subculture of animal cells. "P07" refers to the 7th passage of the cultured cells. The subculture practice between different passages was strictly the same. The viability of fibroblasts after being subjected to several intensities of shear stress was measured for each passage number by flow cytometry. The cells were suspended in fluids with viscosities of 1 mPa.s and 5 mPa.s and then exposed to shear stresses of 0.159 kPa and 6.14 kPa. For the two groups of suspended fluids, the flow rate was from 117 μL / min to 942 μL / min.

[0264] Figure 29 It is shown that the continuous subculture practice does not affect the mechanical capacity of fibroblasts for shear stress. In fact, regardless of the passage number, the cells have similar mechanical capacities. Up to the shear stress corresponding to the mechanical capacity of the cells of 0.577 kPa, the viability of fibroblasts is not affected (96%), while the viability of the control group is 97%. In the range from 0.639 kPa to 4.78 kPa, the viability drops sharply from 92% to 5%, reaching a steady state below 5%.

[0265] A similar study was conducted on HEK293T cells. Different from fibroblasts, HEK293T cells were passaged twice a week from the P09th generation to the P11th generation. Therefore, between consecutive subcultures, the duration of cell growth alternated between 3 days and 4 days.

[0266] The cells were suspended in fluids with viscosities of 1 mPa.s and 5 mPa.s and then exposed to shear stresses in the range of 0.11 kPa to 6.4 kPa. For the two groups of suspended fluids, the flow rate was from 85 μL / min to 942 μL / min. The cell viability was analyzed by flow cytometry.

[0267] As Figure 29 shown, the mechanical capacity of HEK293T cells is affected by the variation in the culture period of continuous cell subculture. The culture period before stress application and before measurement was 3 days or 4 days. HEK293T cells with a 3-day culture period had a mechanical capacity of 0.516 kPa, while HEK293T cells with a culture period had a reduced mechanical capacity of 0.319 kPa.

[0268] Example 13: Repeated series of stresses

[0269] To study the effect of repeated identical stresses on the mechanical capacity of cells, AD-MSC cells were suspended in a fluid with a viscosity of 1 mPa.s and exposed to repeated hydrodynamic stresses. This type of stress is defined as a cyclic sequence of identical stresses.Figure 30A Such a sequence is shown. The cells are subjected to shear stress, enter a shear-free region, and then are subjected to shear again. At a stress of 0.2 kPa, this cycle is performed on a batch of cells, and the number of cycles is equal to 77, 154, 231, 308, 385, and 462 times. This stress is less than the mechanical capacity of 0.6 kPa of the cells, as U-NET shown. After exposure to different numbers of cycles, the cells are analyzed by a cell counter to determine cell viability. VAE + U-NET The cyclic stress below the mechanical capacity of the AD-MSC cells is shown, which does not affect cell viability.

[0270] Example 14: Cell localization and separation

[0271] Cell localization and separation are performed using an algorithm based on the original images taken by a high-speed camera. First, a mask is created to reduce the size of the original image by superposition. Second, image processing functions such as erosion and dilation are applied to reduce image background noise and display cell localization with greater contrast. Subsequently, separation is performed by defining the region of interest (ROI) ( DAE + U-NET ).

[0272] Example 15: Localization and measurement of the short axis and long axis

[0273] A supervised learning model is used to perform the localization and measurement of the short axis and long axis. The first step is to apply a supervised model (U-Net in this example) to the Precision cell photos generated. This determines the size and shape of the cells. In the second step, an "edge extraction" type image processing function is performed to locate the cell contour in the Recall image obtained. Then, the short axis and long axis are measured, where the long axis is the largest diameter and the short axis is the smallest diameter. These results can be displayed in a graph showing the level of deformability based on cell size. Here, our results demonstrate the accuracy of our image processing deformability measurements for shear and collision tests.

[0274] Example 16: Image quality improvement using an unsupervised model

[0275] The images obtained from the cell localization and separation steps for measuring the long axis and short axis are not always of high quality. This is because the noise in these images is sufficient to prevent the supervised model from segmenting the cells. To improve this image quality, the application of an unsupervised model is necessary. The application of an "autoencoder" neural network is shown here to reconstruct the image generated by ROI extraction with better quality ( F1 Score ). Figure 30B The relevance of this method is shown. In ​Among them, the result of applying no autoencoder is shown in the first row, the result of applying a variational autoencoder is shown in the second row, and the result of applying a denoising autoencoder is shown in the third row.

[0276] Example 17: Improving cell detection and deformability measurement by coupling unsupervised and supervised models

[0277] The coupling of unsupervised and supervised models can be used to improve the detection of cells within an image and the measurement of their deformability. In this example, we illustrate this method by coupling an autoencoder and a U-Net neural network to an image obtained from shear measurements ( ​ ).

[0278] The numerical values are as follows (Table 3):

[0279] ​ ​ ​ ​ 0.59 0.78 0.81 ​ 0.63 0.69 0.71 ​ 0.61 0.75 0.77

[0280] Table 3

[0281] We can observe that applying the supervised network (U-Net) alone cannot detect and track all cell movements, while coupling with an unsupervised model (VAE autoencoder or DAE autoencoder) can improve detection and measurement (higher precision, recall, and F1-score) ( ​ ).

[0282] Therefore, the measurement improvement is described by subtracting the deformability measurement from the standard manual method. That is, the user has processed each image without performing image processing or applying a learning model.

[0283] Example 18: Predicting the physiological state of cells using supervised learning through a classification model.

[0284] Supervised learning models can be used to classify the physiological state of cells through image analysis. This example shows the classification of three types of cells annotated as necrotic (A), apoptotic (B), and viable (C).

Claims

1. A device for applying at least one hydrodynamic stress of a determined intensity and duration to at least one cell suspended and moving in the device, and for characterizing the cell during the application of the stress and optionally after the application of the stress, the device comprising: - a microfluidic circuit comprising at least one section, the section comprising at least: i) a main channel configured to circulate a fluid containing the cells in suspension, ii) a fluid inlet and a fluid outlet, iii) means for introducing and establishing a flow of the fluid within the channel, and - means for characterizing the cell during the application of the stress, characterized in that the at least one section is configured to apply at least one hydrodynamic stress to the cell, the hydrodynamic stress being selected from: hydrodynamic compression stress, hydrodynamic shear stress and hydrodynamic tensile stress, and for optionally applying at least one mechanical shock to the cell.

2. The device according to the preceding claim, wherein the characterization of the cell has as its object at least one criterion selected from: size, shape, appearance of the outer membrane, appearance of the cytoplasm, presence of at least one marker on the cell surface, protein content of the cell and nucleic acid content of the cell.

3. The device according to any one of the preceding claims, characterized in that, At least one section A configured to apply a compression stress to the cell comprises or consists of a first channel connected at the same level to two channels, each channel forming an angle of 30 degrees to 150 degrees with the first channel.

4. The device according to any one of the preceding claims, characterized in that, At least one section B configured to apply a shear stress to the cell comprises or consists of a channel having a diameter of 10 micrometers to 2000 micrometers, preferably a diameter of 20 micrometers to 200 micrometers.

5. The device according to any one of the preceding claims, characterized in that, At least one section C configured to apply a tensile stress to the cell comprises or consists of: i) a channel with a gradually increasing or decreasing cross-section, or ii) a first channel connected to a second channel, wherein the direction of circulation of the fluid in the second channel is different from the direction of circulation in the first channel, preferably opposite to the direction of circulation in the first channel.

6. The device according to any one of the preceding claims, characterized in that, At least one section D configured to apply a mechanical shock to the cell comprises or consists of a first channel in which the streamline of the cell encounters an obstacle, in particular for example the wall of a second channel.

7. The device according to any one of the preceding claims, characterized in that, The circuit comprises: at least one section A and / or at least one section B and / or at least one section C, and optionally at least one section D, combined with each other.

8. The device according to any one of the preceding claims, characterized in that, The circuit is configured to apply to the cell at least one sequence comprising or consisting of at least two consecutive physical stresses of different nature.

9. The device according to any one of claims 1 to 7, characterized in that, The circuit is configured to apply to the cell at least one sequence comprising or consisting of at least two consecutive hydrodynamic stresses of the same nature, and optionally at least two consecutive mechanical stresses of the same nature.

10. The device according to any one of the preceding claims, characterized in that, The intensity of at least one hydrodynamic stress applied to the cells is 10 -3 kPa to 10 5 kPa, preferably 10 -3 kPa to 10 4 kPa, preferably 0.1 kPa to 10 3 kPa, preferably 1 kPa and 10 2 kPa, and / or is characterized in that, The total duration of the application of the at least one hydrodynamic stress is from 1 microsecond to 10000 seconds, preferably from 1 microsecond to 100 seconds, preferably from 1 microsecond to 1 second, preferably from 10 microseconds to 100 milliseconds.

11. The device according to any one of the preceding claims, characterized in that, After applying the at least one stress, the characterization of the cells is carried out within a time period of 0 days to 120 days, preferably 0 days to 30 days, preferably 0 days to 1 day after applying the at least one physical stress.

12. The device according to any one of the preceding claims, characterized in that During the application of the at least one stress, the characterization of the cells comprises or consists of: at least one discrete characterization or at least one characterization having a duration of 1 microsecond to 10,000 seconds, preferably 1 microsecond to 100 seconds, preferably 1 microsecond to 1 second, preferably 10 microseconds to 100 milliseconds.

13. The device according to any one of the preceding claims, characterized in that, The device for characterizing the at least one cell is selected from: a cell counter, a microscope, a device for analyzing the protein content of the cells, a device for analyzing and sequencing the nucleic acids of the cells, and an image capture device combined with an image analysis device.

14. The device according to the preceding claim, characterized in that, The image analysis device comprises or consists of: a central computer unit, which includes software means suitable for image analysis.

15. The device according to any one of claims 13 or 14, characterized in that, The image analysis device further comprises a first classification model, which is pre-trained with a training data set and includes a supervised machine learning algorithm, an unsupervised machine learning algorithm or a semi-supervised machine learning algorithm, and the first classification model is suitable for predicting the physiological state of a given cell based on the characteristics of the cell.

16. The device according to any one of claims 13 to 15, characterized in that, The image analysis device further comprises a second classification model, which is pre-trained with a training data set and includes a supervised automatic learning algorithm, an unsupervised automatic learning algorithm or a semi-supervised automatic learning algorithm, and the second classification model is suitable for detecting and monitoring the deformation of a given cell in response to at least one physical stress.

17. A method for applying at least one physical stress of a determined intensity and duration to at least one cell in suspension and for characterizing the cell during and optionally after the application, the method comprising the following steps: a) depositing and circulating a fluid containing the at least one cell in suspension in a microfluidic circuit comprising at least one section, the section comprising at least: i) a main channel configured to circulate the fluid containing the cells, ii) a fluid inlet and a fluid outlet, iii) means for introducing and establishing a flow of the fluid within the channel, and b) characterizing the cell during the application of the at least one stress; and optionally characterizing the cell after applying the at least one stress, characterized in that the at least one section is configured to apply at least one hydrodynamic stress to the cell, the hydrodynamic stress being selected from: hydrodynamic compression stress and / or hydrodynamic shear stress and / or hydrodynamic tensile stress, and optionally applying at least one mechanical shock.

18. The method according to the preceding claim, wherein the characterization of the cell has at least one criterion as its object, the criterion being selected from: size, shape, appearance of the outer membrane, appearance of the cytoplasm and the presence of at least one surface marker.

19. The method according to any one of claims 17 or 18, wherein the cell is subjected to: - at least one shear stress with a strength of 10 -3 kPa to 10 5 kPa, preferably 10 -3 kPa to 10 4 kPa, preferably from 0.1 kPa to 10 3 kPa, and / or a duration of 1 microsecond to 10,000 seconds, preferably 1 microsecond to 100 seconds, preferably 10 microseconds to 1 second, preferably 10 microseconds to 10 milliseconds, and / or - At least one compressive stress with a strength of 10 -3 kPa to 10 3 kPa, preferably 10 -3 kPa to 10 2 kPa, preferably 10 - 3 kPa to 10 kPa, and / or a duration of 1 microsecond to 10 seconds, preferably 1 microsecond to 1 second, preferably 10 microseconds to 10 milliseconds, and / or - At least one tensile stress with a strength of 10 -3 kPa to 10 3 kPa, preferably 10 -3 kPa to 10 2 kPa, preferably 10 - 3 kPa to 10 kPa, and / or a duration of 1 microsecond to 10 seconds, preferably 1 microsecond to 1 second, preferably 10 microseconds to 10 milliseconds, - and optionally at least one mechanical shock having an intensity of 1 m / s to 300 m / s and / or a duration of less than 1 microsecond.

20. The method according to any one of claims 17 to 19, characterized in that, The method further includes a step of predicting the physiological state of the cell based on the features determined in step b) through a pre-trained first classification model.

21. The method according to any one of claims 17 to 20, characterized in that, The method further includes a step of predicting the physiological state of the cell through a pre-trained first classification model, wherein the first classification model includes: a machine learning algorithm pre-trained with a training data set, a supervised learning neural network, or a multi-class probability classification algorithm.

22. The method according to any one of claims 17 to 21, characterized in that, The method further includes a step of monitoring the deformability of the cell at different times during the cell staying in the microfluidic channel based on the features determined in step b) through a pre-trained second classification model.

23. A classification model pre-trained on a training data set for predicting the physiological state of a cell during and / or after applying at least one hydrodynamic stress in the method according to any one of claims 17 to 22.

24. Use of the device according to any one of claims 1 to 16, or the method according to any one of claims 17 to 22, or the classification model according to claim 23 for characterizing the following types of cells: prokaryotic cells, eukaryotic cells, animal cells, plant cells, human cells, stem cells, epithelial cells, fibroblasts, blood cells, genetically modified cells, or synthetic cell mimics.

25. Use of the device according to any one of claims 1 to 16, the method according to any one of claims 17 to 22, or the classification model according to claim 23 for determining the cell volume.

26. Use of the device according to any one of claims 1 to 16, the method according to any one of claims 17 to 22, or the classification model according to claim 23 for defining at least one parameter of a biological process.

27. The use according to claim 26, wherein The biological process is selected from: bioprinting, cell therapy, and bioproduction.

Images