Preparation method of exosome antibody microsphere conjugate

Abstract

The invention provides a preparation method of an exosome antibody microsphere conjugate, which belongs to the technical field of exosomes, and comprises the following steps: preparing a carboxyl polystyrene microsphere suspension with the particle size of 50 nanometers, and ensuring that the size of the microsphere is stabilized in a range of 50 + / -5 nanometers by optimizing centrifugation and cross-linking conditions. And then coupling the exosome marker antibody with the microspheres, and closing unreacted active groups. A mathematical model is adopted to optimize purification process parameters for purification. A dynamic light scattering and transmission electron microscope is utilized to characterize coupling physics properties, and a conjugate meeting requirements is screened. And finally, obtaining fluffy porous exosome antibody microsphere conjugate freeze-dried powder under the protection of a sucrose solution by adopting a vacuum freeze-drying method. The method solves the technical problems that in an existing exosome microsphere conjugate preparation process, centrifugation and purification parameters are often set according to artificial experience, optimization is difficult to obtain, and the quality of the exosome microsphere conjugate is affected.

Description

Technical Field

[0001] The present invention belongs to the technical field of exosomes, and specifically relates to a method for preparing an exosome-antibody-microsphere conjugate. Background Art

[0002] Exosomes are membrane-like particles with a diameter of 30-150 nanometers secreted by cells. They can carry a variety of bioactive substances, such as proteins, nucleic acids, metabolites, etc., and play an important role in physiological processes such as intercellular signaling, immune regulation, and tumor metastasis. In recent years, exosomes have shown broad application prospects in disease diagnosis and treatment. For example, exosomes secreted by tumor cells carry specific markers and can be used as a new detection method for non-invasive tumor liquid biopsy; exosomes produced during inflammatory response can be used to assess disease progression and prognosis; exosomes secreted by stem cells play a role in tissue repair and regeneration and have potential therapeutic applications.

[0003] However, the current exosome separation and purification technology still faces many challenges. Although the commonly used ultracentrifugation method can efficiently separate exosomes, it is complicated to operate and easily damages biological activity; although immunoaffinity chromatography can improve specific capture, it requires a large amount of antibodies and is difficult to achieve continuous production; although microfluidic chips can achieve the separation of trace samples, the processing is complex and difficult to scale up. In addition, the separated exosomes often contain more impurities, which are difficult to meet the purity requirements of clinical applications and require further purification steps. In the existing exosome-microsphere conjugate preparation process, the centrifugation and purification parameters are often set according to manual experience, which is difficult to obtain the optimal one, affecting the quality of the exosome-microsphere conjugate. Therefore, the development of a simple, efficient and scalable exosome separation and purification technology is of great significance for promoting exosome-related research and clinical transformation. Summary of the invention

[0004] In view of this, the present invention provides a method for preparing an exosome-antibody-microsphere conjugate, which can solve the technical problem that in the existing exosome-microsphere conjugate preparation process, the parameters of centrifugation and purification are often set according to manual experience, which is difficult to obtain the optimal value and affects the quality of the exosome-microsphere conjugate.

[0005] The present invention is achieved in that:

[0006] The present invention provides a method for preparing an exosome antibody microsphere conjugate, which includes: preparing a carboxyl polystyrene microsphere suspension; optimizing the centrifugation parameters by the response surface method and performing a centrifugation operation; optimizing the crosslinking conditions by single factor experiments, adding a crosslinking agent to the microsphere suspension for activation; adding an exosome marker antibody to the activated microsphere suspension; stirring and reacting under specific conditions to couple the microspheres with the antibody; adding a glycine solution to the coupling reaction mixture to block unreacted active groups; optimizing the purification process parameters by a mathematical model and performing a purification operation; characterizing the physicochemical properties of the conjugate by dynamic light scattering and transmission electron microscopy; performing antifreeze protection and vacuum freeze-drying treatment on the selected conjugate, and finally obtaining a fluffy and porous freeze-dried powder. Specifically, it includes the following steps:

[0007] S10. Prepare a carboxyl polystyrene microsphere suspension with a particle size of 50 nanometers;

[0008] S20. Optimize the centrifugation parameters by the response surface method, perform a centrifugation operation on the carboxyl polystyrene microsphere suspension according to the optimized centrifugation parameters, and then collect the microsphere precipitate and resuspend it with 10 milliliters of deionized water;

[0009] S30. Optimize the crosslinking conditions by single factor experiments, and according to the optimized crosslinking conditions, add a crosslinking agent to the resuspended carboxyl polystyrene microsphere suspension, activate it at room temperature for 30 minutes, and monitor the change of microsphere size in real time by dynamic light scattering method to ensure that the particle size is stable within the range of 50 ± 5 nanometers;

[0010] S40. Add an exosome marker antibody with a mass concentration of 0.2 milligrams per milliliter to the activated carboxyl polystyrene microsphere suspension;

[0011] S50. Stir and react at 4 degrees Celsius for 4 hours, set the stirring speed to 200 revolutions per minute, and maintain the pH value of the reaction system at 7.2 ± 0.2 to couple the carboxyl polystyrene microspheres with the exosome marker antibody to obtain a coupling reaction mixture;

[0012] S60. Add 5 milliliters of a glycine solution with a concentration of 1 mole per liter to the coupling reaction mixture, react at 25 degrees Celsius for 30 minutes, and monitor the reaction pH value with a pH test paper to keep it within the range of 7.0 ± 0.2 to block unreacted active groups;

[0013] S70. Optimize the purification process parameters by a mathematical model, perform a purification operation according to the optimized purification process parameters, and collect the precipitate;

[0014] S80. Characterize the physicochemical properties of the conjugate using dynamic light scattering and transmission electron microscopy, including dynamic light scattering characterization and transmission electron microscopy characterization, and perform the next freeze-drying treatment on the conjugate that meets the above two characterization indicators;

[0015] S90. Protect the screened conjugate against freezing with 5 ml of a 10% (mass fraction) sucrose solution, and use the vacuum freeze-drying method. The pre-freezing temperature is -40 °C for 4 hours, the primary drying temperature is -20 °C for 12 hours, the secondary drying temperature is 10 °C for 8 hours, and the vacuum is maintained at 10 Pa. Finally, obtain a fluffy and porous freeze-dried powder to complete the preparation of the exosome antibody microsphere conjugate.

[0016] Based on the above technical solutions, the method for preparing an exosome antibody microsphere conjugate of the present invention can be further improved as follows:

[0017] Among them, the monomer ratio of the carboxyl polystyrene microsphere suspension is specifically: the amount of styrene monomer is 80 parts by mass, the amount of acrylic acid monomer is 20 parts by mass, the amount of potassium persulfate initiator is 2 parts by mass, and the amount of sodium dodecyl sulfate dispersant is 5 parts by mass; the reaction temperature during the preparation process of the carboxyl polystyrene microsphere suspension is controlled at 75 °C, and the reaction time is 4 hours.

[0018] Further, the dynamic light scattering characterization is specifically: resuspend the collected precipitate in a phosphate buffer solution with a pH value of 7.4 and a concentration of 0.01 mol / L, with a solution volume of 5 ml, and measure the particle size distribution under the conditions of a 633 nm laser wavelength and a 90° scattering angle. Screen the conjugate with a particle size distribution in the range of 80 to 120 nm, and the polydispersity index should be less than 0.2.

[0019] Further, the transmission electron microscopy characterization is specifically: take 2 μl of the resuspended solution after the purification operation and drop it on a copper grid, negatively stain with phosphotungstic acid for 2 minutes, and observe the morphology of the conjugate after natural drying under an acceleration voltage of 80 kV. The conjugate should be in a regular spherical shape and evenly dispersed, with a particle size distribution conforming to a log-normal distribution and no obvious agglomeration phenomenon.

[0020] Among them, the exosome marker antibody specifically includes a mouse-derived IgG1 subtype CD63 monoclonal antibody, a mouse-derived IgG2b subtype CD9 monoclonal antibody, and a rat-derived IgG2a subtype CD81 monoclonal antibody.

[0021] Further, the mass ratio of the three antibodies is 1:1:1.

[0022] Among them, the centrifugation parameters include centrifugation speed, centrifugation time, centrifugation temperature, and centrifugation acceleration.

[0023] The steps for optimizing the centrifugation parameters by the response surface method include:

[0024] First step, establish a quadratic polynomial regression equation;

[0025] Second step, obtain experimental data through Box - Behnken experimental design;

[0026] Third step, fit the response surface by the least - squares method;

[0027] Fourth step, analyze the influence degree of each factor on the target response value;

[0028] Fifth step, obtain the optimal combination of centrifugation parameters.

[0029] Furthermore, the cross - linking conditions include cross - linker concentration, reaction temperature, reaction time, pH value, and ionic strength.

[0030] The steps for optimizing the cross - linking conditions by single - factor experiments include:

[0031] First step, set the cross - linker concentration gradient to be from 0.1 to 1.0 mol / L;

[0032] Second step, set the reaction temperature gradient to be from 15 to 35 °C;

[0033] Third step, set the reaction time gradient to be from 10 to 60 minutes;

[0034] Fourth step, set the pH value gradient to be from 5 to 9;

[0035] Fifth step, set the ionic strength gradient to be from 0.01 to 0.5 mol / L.

[0036] Furthermore, the mathematical model for predicting the purification efficiency includes a mass transfer equation, a hydrodynamics equation, and a heat transfer equation.

[0037] Furthermore, the mass transfer equation is used to calculate the mass transfer rate of substances. The inputs are the concentration gradient and the diffusion coefficient, and the output is the mass transfer flux;

[0038] The hydrodynamics equation is used to calculate the fluid motion state. The inputs are the fluid density, viscosity, and pressure, and the output is the fluid velocity field;

[0039] The heat transfer equation is used to calculate the temperature distribution. The inputs are the thermal conductivity, specific heat capacity, and temperature gradient, and the output is the heat flux.

[0040] Compared with the prior art, the beneficial effects of a method for preparing an exosome - antibody microsphere conjugate provided by the present invention are:

[0041] 1. Optimize the synthesis conditions of microspheres by statistical design to obtain polymer microspheres with uniform particle size and good dispersibility. By using the response surface method to optimize process parameters such as monomer ratio, initiator dosage, and dispersant dosage, carboxyl polystyrene microspheres with an average particle size of 50 nanometers can be stably prepared. Compared with the existing manual preparation method, this method can more precisely control the properties of microspheres and greatly improve the product consistency.

[0042] 2. Stabilize the microsphere structure by centrifugal separation and cross-linking treatment. Optimize the centrifugation parameters by the response surface method to achieve efficient separation and collection of microspheres; use 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride as a cross-linking agent and activate it at room temperature for 30 minutes, which can effectively cross-link the carboxyl groups on the surface of microspheres and improve the stability of microspheres in subsequent coupling processes. These pretreatment methods ensure that microspheres will not agglomerate or disintegrate under complex reaction conditions.

[0043] 3. Optimize the antibody binding ratio and reaction conditions to achieve efficient coupling of microspheres with exosome markers. Select CD63 antibody of mouse-derived IgG1 subtype, CD9 antibody of mouse-derived IgG2b subtype, and CD81 antibody of rat-derived IgG2a subtype. The mass ratio of the three antibodies is 1:1:1, which can make full use of the active groups on the surface of microspheres and capture exosome markers to the greatest extent. At the same time, optimize parameters such as reaction temperature, pH value, and stirring speed through orthogonal experiments, which greatly improves the coupling efficiency.

[0044] 4. Establish a mathematical model to optimize the purification process and improve the purity of the conjugate. Use stepwise centrifugation for purification. First, centrifuge at 3000 revolutions per minute at 25 °C for 10 minutes to remove macromolecular impurities, and then centrifuge at 12000 revolutions per minute at 4 °C for 20 minutes to collect the precipitate. To further optimize the centrifugation process, establish mathematical models of mass transfer equation, hydrodynamic equation, and heat transfer equation, and introduce neural network for correction, which can more accurately predict the purification efficiency.

[0045] 5. Use characterization means to screen out conjugate products that meet the requirements. Analyze the particle size distribution and polydispersity index by dynamic light scattering, and observe the morphology and agglomeration state by transmission electron microscopy to ensure that the physicochemical properties of the conjugate meet the application requirements. At the same time, optimize the preparation of dry preparations by freeze-drying process to obtain stable final products.

[0046] In summary, the present invention designs a preparation method of exosome microsphere conjugate, realizes the full-process optimization of microsphere synthesis, coupling, and purification, greatly improves the performance and consistency of the product, and solves the technical problem that in the existing preparation process of exosome microsphere conjugate, the parameters of centrifugation and purification are often set according to manual experience, it is difficult to obtain the optimum, which affects the quality of exosome microsphere conjugate. Description of the Drawings

[0047] Figure 1 It is a flowchart of the method provided by the present invention;

[0048] Figure 2 It is a three-dimensional response surface diagram for optimizing the microsphere synthesis conditions;

[0049] Figure 3 It is a kinetic curve diagram of the cross-linking reaction;

[0050] Figure 4 It is a diagram of the influence of temperature on the coupling efficiency;

[0051] Figure 5 It is a histogram of the particle size distribution of the conjugate. Detailed implementation manners

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0053] As Figure 1 shown, it is a flowchart of a method for preparing an exosome antibody microsphere conjugate provided by the present invention. This method includes the following steps:

[0054] S10. Prepare a carboxyl polystyrene microsphere suspension with a particle size of 50 nanometers;

[0055] S20. Optimize the centrifugation parameters by the response surface method, and perform centrifugation on the carboxyl polystyrene microsphere suspension according to the optimized centrifugation parameters. Then collect the microsphere precipitate and resuspend it with 10 milliliters of deionized water;

[0056] S30. Optimize the cross-linking conditions by single-factor experiments, and according to the optimized cross-linking conditions, add a cross-linking agent to the resuspended carboxyl polystyrene microsphere suspension, activate it at room temperature for 30 minutes, and monitor the change of microsphere size in real time by dynamic light scattering method to ensure that the particle size is stably within the range of 50 ± 5 nanometers;

[0057] S40. Add an exosome marker antibody with a mass concentration of 0.2 milligrams per milliliter to the activated carboxyl polystyrene microsphere suspension;

[0058] S50. Stir and react at 4 degrees Celsius for 4 hours, set the stirring speed to 200 revolutions per minute, and maintain the pH value of the reaction system at 7.2 ± 0.2 to make the carboxyl polystyrene microspheres react with the exosome marker antibody to obtain a coupling reaction mixture;

[0059] S60. Add 5 mL of glycine solution with a concentration of 1 mol / L to the coupling reaction mixture, react at 25 °C for 30 minutes, monitor the reaction pH value with pH test paper and keep it within the range of 7.0 ± 0.2 to block the unreacted active groups;

[0060] S70. Optimize the purification process parameters using a mathematical model, and perform purification operations according to the optimized purification process parameters, and collect the precipitate;

[0061] S80. Characterize the physicochemical properties of the conjugate using dynamic light scattering and transmission electron microscopy, including dynamic light scattering characterization and transmission electron microscopy characterization, and perform the next freeze-drying treatment on the conjugate that meets the above two characterization indicators;

[0062] S90. Protect the selected conjugate against freezing with 5 mL of 10% (mass fraction) sucrose solution, use the vacuum freeze-drying method, with a pre-freezing temperature of -40 °C for 4 hours, a primary drying temperature of -20 °C for 12 hours, a secondary drying temperature of 10 °C for 8 hours, and the vacuum degree maintained at 10 Pa, and finally obtain a fluffy and porous freeze-dried powder to complete the preparation of the exosome antibody microsphere conjugate.

[0063] The following is a detailed description of the specific implementation manners of the above steps:

[0064] In step S10, the statistical design optimization method is used to determine the monomer ratio to obtain stable and uniform polymer particles. First, by adjusting the addition amounts of styrene monomer, acrylic acid monomer, initiator, and dispersant, the preparation conditions of the microspheres are optimized. Among them, 80 parts by mass of styrene monomer and 20 parts by mass of acrylic acid monomer can form a copolymerized hydrophobic-hydrophilic structure; the amount of potassium persulfate initiator is 2 parts by mass, which can effectively initiate the free radical polymerization reaction; the amount of sodium dodecyl sulfate dispersant is 5 parts by mass, which can stably form a microsphere suspension. The reaction temperature is controlled at 75 °C and the reaction time is 4 hours. These process parameters help to obtain microspheres with an average particle size of 50 nm.

[0065] In step S20, the response surface method is used to optimize the centrifugation parameters. The prepared microsphere suspension is centrifuged at 4000 revolutions per minute and 25 °C for 15 minutes, the precipitated microspheres are collected and resuspended with 10 mL of deionized water, and the pH value of the resuspended solution is measured to be 6.5. The response surface method is used to optimize the centrifugation parameters to effectively separate and collect the microspheres. To obtain the optimal centrifugation separation effect, the following quadratic polynomial regression model is first established:

[0066]

[0067] where x1 is the centrifugation speed, x2 is the centrifugation time, x3 is the centrifugation temperature, b iis the regression coefficient, and ε is the random error term. Then, the Box-Behnken experimental design is used to obtain experimental data, the least squares method is applied to fit the response surface, and the influence degree of each factor on the target response (precipitation yield) is analyzed. Finally, the optimal centrifugation parameter combination of 4000 revolutions per minute, 25 degrees Celsius, and centrifugation for 15 minutes is determined.

[0068] In step S30, the cross-linking conditions are optimized by single-factor experiments. Add a cross-linking agent of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride with a concentration of 0.5 mg / mL to the resuspended carboxyl polystyrene microsphere suspension, activate it at room temperature for 30 minutes, and monitor the change in microsphere size in real time by dynamic light scattering. To determine the optimal cross-linking conditions, first set the cross-linking agent concentration gradient to 0.1 to 1.0 mol / L, the reaction temperature gradient to 15 to 35 degrees Celsius, the reaction time gradient to 10 to 60 minutes, the pH value gradient to 5 to 9, and the ionic strength gradient to 0.01 to 0.5 mol / L, and investigate the influence of each factor on the cross-linking effect one by one.

[0069] According to the experimental results, it is found that the cross-linking agent concentration is the most critical factor. The cross-linking reaction can be described by the following first-order Arrhenius kinetic model:

[0070]

[0071] where c is the cross-linking agent concentration (mol / L), t is the reaction time (seconds), k0 is the frequency factor (per second), E a is the activation energy (joules per mole), R is the gas constant (8.314 joules per mole-Kelvin), and T is the absolute temperature (Kelvin). By fitting the experimental data, k0 = 5.2×10 8 per second and E a = 42 kJ / mol are obtained. On this basis, the cross-linking conditions are optimized by single-factor experiments, and the specific steps include:

[0072] The first step: Set the cross-linking agent concentration gradient to 0.1 to 1.0 mol / L;

[0073] The second step: Set the reaction temperature gradient to 15 to 35 degrees Celsius;

[0074] The third step: Set the reaction time gradient to 10 to 60 minutes;

[0075] The fourth step: Set the pH value gradient to 5 to 9;

[0076] The fifth step: Set the ionic strength gradient to 0.01 to 0.5 mol / L.

[0077] After optimization by single-factor experiments, the optimal cross-linking conditions were determined to be a concentration of 0.5 mg / mL, a reaction temperature of room temperature, and a reaction time of 30 minutes, which could effectively stabilize the microsphere structure, and the particle size was stable within the range of 50 ± 5 nm.

[0078] In step S40, the antibody binding ratio was optimized. An exosome marker antibody with a concentration of 0.2 mg / mL was added to the activated microsphere suspension, including a mouse-derived IgG1 subtype CD63 monoclonal antibody, a mouse-derived IgG2b subtype CD9 monoclonal antibody, and a rat-derived IgG2a subtype CD81 monoclonal antibody. The mass ratio of the three antibodies was 1:1:1. This antibody ratio combination could make full use of the active groups on the microsphere surface, maximize the capture of exosome markers, and improve the recognition performance of the microspheres.

[0079] In step S50, the reaction conditions were optimized by orthogonal experiments. The carboxyl polystyrene microspheres were coupled with the exosome marker antibody at 4 °C, pH 7.2 ± 0.2, and a stirring speed of 200 rpm for 4 hours. These reaction parameters helped to improve the coupling efficiency and avoid the denaturation of microspheres and antibodies at higher temperatures or under strong stirring.

[0080] According to the reaction kinetics analysis, the coupling reaction can be described by the following second-order kinetic model:

[0081]

[0082] where c A and c B are the concentrations of microspheres and antibodies (mol / L), respectively, and k f is the forward reaction rate constant (per mol-s). By fitting the experimental data, k f = 2.8×10 3 per mol-s. On this basis, it was determined that the above reaction conditions could maximize the coupling efficiency.

[0083] In step S60, a glycine solution with a concentration of 1 mol / L was added to the coupling reaction mixture, and the reaction was carried out at 25 °C for 30 minutes, and the pH value was maintained within the range of 7.0 ± 0.2. Glycine could effectively block the residual active carboxyl or amino groups on the microsphere surface and avoid non-specific adsorption during subsequent purification.

[0084] In step S70, a mathematical model is established to optimize the purification process, and the coupled reaction mixture is purified by stepwise centrifugation. First, centrifuge at 3000 revolutions per minute for 10 minutes at 25 degrees Celsius, and collect the supernatant; then centrifuge at 12000 revolutions per minute for 20 minutes at 4 degrees Celsius, and collect the precipitate. To further optimize the centrifugation process, the following mathematical models of mass transfer equation, hydrodynamic equation and heat transfer equation are established:

[0085] The mass transfer equation is specifically expressed as follows:

[0086]

[0087] In the formula, c is the solute concentration (mol / L), and the value range is 0.001 - 1.0 mol / L; t is the time (s); v is the fluid velocity (m / s); x is the spatial coordinate (m); D is the diffusion coefficient (m 2 / s); α is the mass transfer coefficient (1 / s); c s is the saturation concentration (mol / L); R(c,T) is the reaction rate term; ξ(t) is the random perturbation term; w i is the neural network weight; u i is the input layer weight; b i is the bias term; σ is the activation function, and the ReLU function is adopted: σ(x) = max(0, x); n is the number of neurons, and the value is 3 - 10.

[0088] Among them, the parameter acquisition method is:

[0089] 1) The diffusion coefficient D is calculated based on the improved Stokes-Einstein equation:

[0090]

[0091] In the formula, k B is the Boltzmann constant, 1.38×10 -23 J / K; T is the absolute temperature (K); η is the fluid viscosity (Pa·s); r is the particle radius (m); λ is the shape factor, and 1 is taken for spherical particles; φ is the volume fraction.

[0092] 2) The mass transfer coefficient α is fitted by combining the ink diffusion model with experiments:

[0093]

[0094] In the formula, k0 is the pre-exponential factor (1 / s); E a is the activation energy (J / mol); R is the gas constant, 8.314 J / (mol·K); β is the concentration-dependent coefficient; Δc is the concentration difference (mol / L); γ is the flow intensity coefficient; Pe is the Péclet number, where L is the characteristic length (m).

[0095] The hydrodynamic equation is specifically expressed as follows:

[0096]

[0097] In the formula, ρ is the fluid density (kg / m 3 ); v is the fluid velocity vector (m / s); p is the pressure (Pa); μ is the dynamic viscosity (Pa·s); F is the external force term (N / m 3 ); η(t) is the turbulent perturbation term; w j is the neural network weight; K j is the convolution kernel matrix; b j is the bias term; τ is the Tanh activation function: m is the number of convolution kernels, and the value ranges from 4 to 8.

[0098] Parameter acquisition method:

[0099] 1) The fluid density ρ is measured using a vibrating densitometer:

[0100] ρ = ρ0 + A(T - T0) + B(T - T0) 2 ;

[0101] In the formula, ρ0 is the density at the reference temperature T0; A and B are the temperature coefficients, which are obtained by fitting multiple temperature points.

[0102] 2) The dynamic viscosity μ is described by a non-Newtonian fluid model:

[0103]

[0104] In the formula, μ ∞ is the infinite shear viscosity; μ0 is the zero shear viscosity; λ is the characteristic time; is the shear rate; n is the flow index.

[0105] The heat transfer equation is specifically expressed as follows:

[0106]

[0107] In the formula, ρ is the density (kg / m 3 ); c p is the specific heat capacity (J / (kg·K)); T is the temperature (K); k is the thermal conductivity (W / (m·K)); Q is the heat source term (W / m 3 ); φ(t,x) is the heat loss function; w k is the neural network weight; M k is the feature matrix; b kis the bias term; ψ is the Sigmoid activation function: l is the number of feature matrices, and its value ranges from 5 to 12.

[0108] Parameter acquisition method:

[0109] 1) Specific heat capacity c p It is measured by an improved differential scanning calorimetry:

[0110]

[0111] In the formula, Q DSC is the heat flow signal (W); m is the sample mass (kg); β is the heating rate (K / s); δ, ∈ are temperature correction coefficients; ΔT is the temperature deviation (K).

[0112] 2) Thermal conductivity k is measured by the transient hot-wire method:

[0113]

[0114] In the formula, q is the line heat source power (W); L is the hot-wire length (m); t is the time (s); ω is the convection correction coefficient; Re is the Reynolds number; Pr is the Prandtl number.

[0115] 3) Neural network parameters (w i , u i , b i , w j , K j , b j , w k , M k , b k ) are obtained through the following steps:

[0116] Step 1: Collect experimental data to construct a training set, including spatio-temporal distribution data such as concentration, velocity, and temperature;

[0117] Step 2: Design a three-layer lightweight neural network structure, with the number of nodes in the input layer, hidden layer, and output layer being [4, 8, 1] respectively;

[0118] Step 3: Use the Adam optimizer to minimize the mean square error loss function:

[0119]

[0120] In the formula, N is the number of samples; y i is the true value; is the predicted value; θ is the regularization coefficient; ||w i || 2 is the weight norm.

[0121] By optimizing the parameters and neural network structure of the above three models, the flow field, temperature field, and concentration field during the centrifugation process can be predicted more accurately, thereby determining the optimal stepwise centrifugation process parameters.

[0122] In step S80, the physicochemical properties of the conjugate are characterized by dynamic light scattering and transmission electron microscopy. First, the collected precipitate is resuspended in a phosphate buffer solution with a pH of 7.4 and a concentration of 0.01 mol / L, and the particle size distribution is measured under the conditions of a laser wavelength of 633 nm and a scattering angle of 90 degrees. According to Mie scattering theory, the relationship between the scattering intensity I of the particle and the particle radius r is as follows:

[0123]

[0124] where, n m is the refractive index of the solvent, n p is the refractive index of the particle, P is the incident light power, and λ is the laser wavelength. By fitting the measurement data, conjugates with a particle size distribution ranging from 80 to 120 nm and a polydispersity index less than 0.2 are selected.

[0125] Then, the morphology of the conjugate is observed by transmission electron microscopy. According to electron diffraction theory, when an electron beam is incident on a crystal sample, a characteristic diffraction pattern will be generated, and the structural information of the sample can be judged. It is observed that the conjugate presents a regular spherical shape and is evenly dispersed, the particle size conforms to a log-normal distribution, and there is no obvious agglomeration phenomenon. Only conjugates that meet the above two indicators can be subjected to subsequent lyophilization treatment.

[0126] In step S90, the selected conjugate is cryoprotected with 5 mL of 10% sucrose solution, and dry powder is prepared by vacuum lyophilization. The glass transition temperature T g of the sucrose solution can be estimated by the Gordon-Taylor equation:

[0127]

[0128] where, w s and w w are the mass fractions of the solid (sucrose) and water, respectively, T g,s and T g,w are the glass transition temperatures of the solid and water, respectively, and k is the Gordon-Taylor constant. According to the experimental measurement data, T g,s = 338 Kelvin, T g,w= 136 Kelvin, k = 3.31. On this basis, the following stepwise freeze-drying process is adopted: the pre-freezing temperature is -40 degrees Celsius for 4 hours, the primary drying temperature is -20 degrees Celsius for 12 hours, the secondary drying temperature is 10 degrees Celsius for 8 hours, and the vacuum degree is maintained at 10 Pa. This stepwise freeze-drying process can effectively protect the microsphere structure and obtain a fluffy and porous dry powder.

[0129] The steps of collecting experimental data to construct a training set for the three-layer lightweight neural network involved in S70 are described in detail below:

[0130] I. Experimental data collection steps:

[0131] 1. Concentration field data acquisition:

[0132] (1) Arrange 12 fiber optic chemical sensor arrays in the reactor with a spatial interval of 10 mm to form a three-dimensional monitoring grid;

[0133] (2) Adopt the fluorescence tracer method and select sodium fluorescein as the tracer;

[0134] (3) Use a 405 nm blue light LED as the excitation light source and set the sampling frequency to 10 Hz;

[0135] (4) Record the fluorescence intensity of each measurement point through a spectral analyzer and convert it to a concentration value according to the standard curve;

[0136] (5) Repeat each group of experiments 3 times and record the 30-minute dynamic change process.

[0137] 2. Velocity field data acquisition:

[0138] (1) Use a particle image velocimetry (PIV) system and select tracer particles with a particle size of 1 - 5 μm;

[0139] (2) Use a double-pulse Nd:YAG laser to generate a light sheet with a pulse interval set to 100 μs;

[0140] (3) Capture the particle motion images through a high-speed camera with a resolution of 2048×2048 pixels;

[0141] (4) Use the cross-correlation algorithm to process the image sequence to obtain a two-dimensional velocity vector field;

[0142] (5) Repeat the measurement at different cross-sections to construct a three-dimensional velocity field distribution.

[0143] 3. Temperature field data acquisition:

[0144] (1) Adopt a platinum resistance temperature sensor array and set a total of 16 measurement points;

[0145] (2) The sensor accuracy is ±0.1 °C and the response time is less than 1 s;

[0146] (3) Synchronously collect the temperature data of each point through the data acquisition module;

[0147] (4) Set the sampling frequency to 1 Hz and continuously record for 2 hours;

[0148] (5) At the same time, use an infrared thermal imager to obtain the surface temperature distribution.

[0149] II. Application of network parameters in three equations:

[0150] 1. Neural network term in the mass transfer equation:

[0151] Used to compensate for the errors of the traditional diffusion model, where:

[0152] (1) The input x is the local concentration gradient and the flow field characteristics;

[0153] (2) Capture the non-linear diffusion behavior through the activation function σ;

[0154] (3) The weights w i and u i Reflect the importance of each influencing factor;

[0155] (4) The bias term b i Is used to adjust the network output range.

[0156] 2. Neural network term in the hydrodynamic equation:

[0157] Used to describe complex flow characteristics, where:

[0158] (1) The convolution kernel K j Extracts the local flow field characteristics;

[0159] (2) The weights w j Balance the contributions of different scale flow structures;

[0160] (3) The Tanh activation function τ introduces a non-linear mapping;

[0161] (4) This term can characterize phenomena such as turbulent pulsations that are difficult to directly model.

[0162] 3. Neural network term in the heat transfer equation:

[0163] Used to correct the heat transfer model, where:

[0164] (1) The feature matrix M k Extracts the spatial characteristics of the temperature field;

[0165] (2) The sigmoid activation function ψ processes the temperature jump;

[0166] (3) The weight w k Adjusts the influence of each order of heat transfer effect;

[0167] (4) This term can describe complex heat transfer processes such as phase change.

[0168] III. Network parameter training strategy:

[0169] 1. Adopt a three-stage training method:

[0170] (1) The first stage: Use experimental data to pre-train each sub-network;

[0171] (2) The second stage: Integrate the sub-networks into the differential equation for joint training;

[0172] (3) The third stage: Use new data to fine-tune the network parameters.

[0173] In summary, the present invention designs a preparation method of exosome microsphere conjugate, mainly adopting the following key technologies: 1) Statistically design and optimize the microsphere synthesis conditions to obtain stable and uniform polymer microspheres; 2) Stabilize the microsphere structure through centrifugal separation and cross-linking treatment; 3) Optimize the antibody binding ratio and reaction conditions to achieve efficient conjugation of microspheres with exosome markers; 4) Establish a mathematical model to optimize the purification process and improve the purity of the conjugate; 5) Use characterization means to screen out the conjugate products that meet the requirements; 6) Optimize the freeze-drying process to obtain a stable dry preparation.

[0174] Specifically, the principle of the present invention is:

[0175] 1. As a solid-phase carrier, microspheres can greatly increase the contact opportunity between exosomes and antibodies and improve the capture efficiency. Compared with free antibodies, the antibodies immobilized on the microsphere surface can make full use of their specific binding sites to efficiently capture target exosomes. At the same time, multiple antibodies are co-modified on the microsphere surface, enabling simultaneous recognition of multiple exosome markers.

[0176] 2. The nano-scale of microspheres and their rich surface groups are conducive to the efficient adsorption and rapid separation of exosomes. Nano-scale microspheres have a large specific surface area, which can provide more binding sites; the surface carboxyl groups can bind to exosome membrane proteins through electrostatic and hydrogen bond forces. This affinity capture mechanism greatly improves the separation efficiency of exosomes.

[0177] 3. The mechanical stability and modifiability of the microspheres provide a good foundation for subsequent purification and characterization. By optimizing the cross-linking conditions, the structure of the microspheres can be effectively stabilized, avoiding deformation or damage under complex reaction conditions. At the same time, the abundant active groups on the surface of the microspheres provide a good reaction platform for antibody conjugation. This pretreatment method ensures that the microspheres maintain good performance throughout the preparation process.

[0178] 4. A mathematical model was established to optimize the separation and purification process, achieving efficient separation and high-purity enrichment. A corresponding mathematical model was established for the hydrodynamic, heat transfer, and mass transfer phenomena during centrifugation. By optimizing the model parameters and neural network structure, the velocity field, temperature field, and concentration field distributions during the separation process can be predicted more accurately, and the optimal stepwise centrifugation process parameters can be determined. This model-driven optimization method significantly improves the efficiency and reliability of separation and purification.

[0179] 5. Characterization means were used to screen out the coupling products that meet the requirements. Through dynamic light scattering and transmission electron microscopy analysis, the physicochemical properties such as the particle size distribution, morphology, and aggregation state of the conjugate can be comprehensively evaluated, and screening can be carried out accordingly. This ensures the quality stability of the final product and lays the foundation for subsequent freeze-drying preparation.

[0180] In summary, the present invention uses polymeric microspheres as the solid-phase capture carrier for exosomes, and through comprehensive optimization of the key processes such as the synthesis, coupling, and purification of the microspheres, the efficiency and reliability of exosome separation and enrichment are achieved.

[0181] An embodiment of a specific application scenario of the present invention is provided below: A research and development team plans to develop a tumor diagnosis product based on exosomes and needs to establish a stable and efficient exosome separation and enrichment technology. According to the preparation method provided by the present invention, the research and development team designed the following specific implementation plan.

[0182] First, the synthesis conditions of the microspheres were optimized according to step S10. In the experiment, styrene and acrylic acid were selected as the copolymerization monomers, potassium persulfate as the radical initiator, and sodium dodecyl sulfate as the dispersant. Optimization experiments were carried out on factors such as the monomer dosage, initiator dosage, and dispersant dosage by the response surface method, and the results are shown in Table 1:

[0183] Table 1 Optimization of microsphere synthesis process parameters

[0184] Experiment number Styrene (g) Acrylic acid (g) Initiator (g) Dispersant (g) Average particle size (nm) 1 70 30 1.5 4 58.2 2 75 25 2 5 52.4 3 80 20 2 5 50.1 4 85 15 2.5 6 55.9 5 90 10 2.5 6 63.7

[0185] Through statistical analysis of the experimental data, it was found that the styrene dosage and acrylic acid dosage are the key factors affecting the particle size. Figure 2 The three-dimensional response surface showing the optimization of the microsphere synthesis conditions demonstrates the influence of the styrene dosage and acrylic acid dosage on the average particle size of the microspheres.

[0186] When the amount of styrene is 80 g and the amount of acrylic acid is 20 g, microspheres with an average particle size of 50.1 nm can be prepared, meeting the expected requirements. After further optimizing the amounts of initiator and dispersant, the finally determined microsphere synthesis process is: 80 g of styrene, 20 g of acrylic acid, 2 g of potassium persulfate, 5 g of sodium dodecyl sulfate, reaction temperature 75 °C, and reaction time 4 h.

[0187] Then, the centrifugation separation parameters of the microspheres are optimized according to step S20. Using the Box-Behnken experimental design, three factors of centrifugation speed, centrifugation time, and centrifugation temperature are set. Through quadratic polynomial regression analysis of the experimental data, the following prediction model is established:

[0188]

[0189] Among them, x1 is the centrifugation speed (thousand revolutions / min), x2 is the centrifugation time (min), x3 is the centrifugation temperature (°C), and Y is the precipitation yield (%). Through optimization analysis of this model, the optimal centrifugation parameters are obtained: speed 4000 revolutions / min, time 15 min, and temperature 25 °C. Experimental verification is carried out under these conditions, and the actual precipitation yield reaches 95.3%, which is in good agreement with the model prediction value.

[0190] Next, the cross-linking conditions of the microspheres are optimized according to step S30. First, set the cross-linking agent concentration gradient to 0.1 - 1.0 mol / L, the reaction temperature gradient to 15 - 35 °C, the reaction time gradient to 10 - 60 min, the pH gradient to 5 - 9, and the ionic strength gradient to 0.01 - 0.5 mol / L, and the influence of each factor is investigated one by one through single-factor experiments. The experimental results show that the cross-linking agent concentration is the most critical factor.

[0191] The first-order Arrhenius kinetic model is used to describe the cross-linking reaction kinetics:

[0192]

[0193] By fitting the experimental data, the kinetic parameters are obtained as: k0 = 5.2×10 8 s -1 , E a = 42 kJ / mol. Based on this, the optimal cross-linking conditions are determined as: cross-linking agent concentration 0.5 mg / mL, reaction temperature room temperature, and reaction time 30 min. Verification experiments are carried out under these conditions, and the particle size of the microspheres is stable at 50 ± 5 nm, meeting the requirements. As Figure 3 shown, the kinetic curve of the cross-linking reaction is presented, including experimental data points and the first-order kinetic model fitting curve.

[0194] In step S40, the antibody binding ratio was optimized. CD63 antibody of mouse-derived IgG1 subtype, CD9 antibody of mouse-derived IgG2b subtype, and CD81 antibody of rat-derived IgG2a subtype were selected, and the concentration of all three antibodies was 0.2 mg / mL, with a mass ratio of 1:1:1. This antibody ratio can maximize the capture of exosome markers.

[0195] Then, the coupling reaction conditions were optimized according to step S50. Through orthogonal experiments, the effects of factors such as temperature, pH value, and stirring speed were investigated, and the experimental results are shown in Table 2:

[0196] Table 2 Optimization of coupling reaction conditions

[0197] Experiment number Temperature (°C) pH value Stirring speed (r / min) Coupling efficiency (%) 1 0 6.8 150 72.4 2 4 7.2 200 88.6 3 4 7.4 250 84.1 4 10 7.2 200 79.5 5 15 7.6 150 71.8

[0198] Figure 4 The effect of temperature on the coupling efficiency is shown, including experimental data points and a quadratic polynomial fitting curve. It can be seen from the experimental results that under the conditions of 4 °C, pH 7.2, and a stirring speed of 200 r / min, the coupling efficiency reaches 88.6%, which is much higher than other combinations of process parameters. This is because a lower reaction temperature helps to maintain the activity of the microspheres and antibodies, a moderate pH value can optimize the charge state to improve the binding force, and moderate stirring is conducive to the full contact of the reactants. Therefore, the above process parameters were determined as the optimal coupling conditions.

[0199] Next, the coupling reaction mixture was blocked according to step S60. The blocking conditions were optimized by the response surface method, and the following quadratic polynomial model was obtained:

[0200]

[0201] Among them, z1 is the glycine concentration (mol / L), z2 is the reaction temperature (°C), and Ψ is the blocking rate of unreacted groups (%). Model analysis shows that when the glycine concentration is 1 mol / L and the reaction temperature is 25 °C, the blocking rate can reach 97.5%. Verification experiments were carried out under these conditions, and the results were in good agreement with the model prediction.

[0202] Subsequently, according to step S70, a mathematical model was established to optimize the purification process of the conjugate. First, the flow field, temperature field, and concentration field during centrifugation were modeled:

[0203] Hydrodynamic equation:

[0204]

[0205] Heat transfer equation:

[0206]

[0207] Mass transfer equation:

[0208]

[0209] By optimizing the above model parameters and neural network structure, the optimal stepwise centrifugation process was obtained: First, centrifuge at 25 °C and 3000 r / min for 10 min to remove macromolecular impurities; then centrifuge at 4 °C and 12000 r / min for 20 min to collect the precipitated conjugate. This optimized separation process can significantly improve the purity of the product.

[0210] Finally, the obtained conjugate was characterized according to step S80. Through dynamic light scattering analysis, samples with a particle size distribution in the range of 80 - 120 nm and a polydispersity index less than 0.2 were screened out. The results of transmission electron microscopy observation showed that these conjugates presented regular spherical shapes and were evenly dispersed, with the particle size conforming to a log-normal distribution and no obvious agglomeration phenomenon. The conjugates meeting the above index requirements entered the freeze-drying process optimization in step S90. As Figure 5 shown, it is the particle size distribution histogram of the conjugate, with a normal distribution fitting curve attached.

[0211] According to the glass transition temperature of sucrose, the following Gordon-Taylor equation was used to estimate the T of a 10% sucrose solution g :

[0212]

[0213] where, w s = 0.1, w w = 0.9, T g,s = 338 K, T g,w = 136 K, k = 3.31, and by calculation, T g = 253 K.

[0214] On this basis, a stepwise freeze-drying process was designed: pre-freezing temperature -40 °C, 4 h; primary drying temperature -20 °C, 12 h; secondary drying temperature 10 °C, 8 h; vacuum degree 10 Pa. This stepwise freeze-drying process can effectively protect the microsphere structure and obtain a fluffy and porous dry powder.

[0215] In summary, the R & D team successfully prepared exosome microsphere conjugates with excellent performance according to the preparation method of the present invention by optimizing key steps such as microsphere synthesis, conjugation, and purification. The specific situation is as follows:

[0216] 1. Microsphere synthesis: The monomer ratio, initiator dosage, and dispersant dosage were optimized by the response surface method to obtain carboxyl polystyrene microspheres with an average particle size of 50.1 nm and good dispersibility.

[0217] 2. Centrifugal separation: By applying the Box-Behnken experimental design and the quadratic polynomial regression model, the optimal centrifugation parameters were determined as: 4000 r / min, 15 min, 25 °C, and the precipitation yield reached 95.3%.

[0218] 3. Microsphere crosslinking: According to the first-order Arrhenius kinetic model, the optimal crosslinking conditions were determined as: 0.5 mg / mL crosslinker concentration, room temperature, 30 min reaction time, and the microsphere particle size was stabilized at 50 ± 5 nm.

[0219] 4. Antibody conjugation: Mouse-derived IgG1 CD63 antibody, mouse-derived IgG2b CD9 antibody, and rat-derived IgG2a CD81 antibody were selected, with a mass ratio of 1:1:1, and the conjugation efficiency reached 88.6%.

[0220] 5. Blocking treatment: Optimized by the response surface method, when the glycine concentration was 1 mol / L and the reaction temperature was 25 °C, the blocking rate of unreacted groups could reach 97.5%.

[0221] 6. Purification process: Mathematical models of hydrodynamics, heat transfer, and mass transfer were established, and neural networks were introduced for correction to determine the optimal stepwise centrifugation process parameters, resulting in a significant improvement in product purity.

[0222] 7. Product characterization: Conjugates with a particle size distribution of 80 - 120 nm and a polydispersity index < 0.2 were screened out, with a regular spherical morphology and no aggregation. Lyophilization was carried out using a 10% sucrose solution to obtain a stable dry powder preparation.

[0223] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A method for preparing an exosome-antibody-microsphere conjugate, comprising the following steps: preparing a suspension of carboxylated polystyrene microspheres; The response surface methodology was used to optimize the centrifugation parameters and perform centrifugation operations. The single-factor experiment was used to optimize the cross-linking conditions, and a cross-linking agent was added to the microsphere suspension for activation. The exosome marker antibody was added to the activated microsphere suspension. The reaction was stirred under specific conditions to allow the microspheres and antibodies to undergo a coupling reaction. A glycine solution was added to the coupling reaction mixture to block the unreacted active groups. A mathematical model was used to optimize the purification process parameters and perform purification operations. The dynamic light scattering method and transmission electron microscopy were used to characterize the physical and chemical properties of the coupling. The screened conjugates were subjected to antifreeze protection and vacuum freeze-drying to obtain a fluffy and porous freeze-dried powder.

2. The method for preparing an exosome-antibody-microsphere conjugate according to claim 1, characterized in that: The monomer ratio for preparing the carboxyl polystyrene microsphere suspension is 80 parts by mass of styrene monomer, 20 parts by mass of acrylic acid monomer, 2 parts by mass of potassium persulfate initiator, and 5 parts by mass of sodium dodecyl sulfate dispersant; the reaction temperature during the preparation process is controlled at 75 degrees Celsius and the reaction time is 4 hours.

3. The method for preparing an exosome-antibody-microsphere conjugate according to claim 2, characterized in that: The dynamic light scattering method is used to characterize the physical and chemical properties of the conjugate. The collected precipitate is resuspended in a phosphate buffer with a pH value of 7.4 and a concentration of 0.01 mol / L. The solution volume is 5 ml. The particle size distribution is measured under the conditions of a laser wavelength of 633 nm and a scattering angle of 90 degrees. The conjugates with a particle size distribution in the range of 80 to 120 nm are screened, and the polydispersity index must be less than 0.

2.

4. The method for preparing an exosome-antibody-microsphere conjugate according to claim 3, characterized in that: Transmission electron microscopy was used to characterize the physical and chemical properties of the conjugate. Two microliters of the resuspension after the purification operation was completed was dropped on a copper grid, negatively stained with phosphotungstic acid for 2 minutes, and then naturally dried. The morphology of the conjugate was observed at an accelerating voltage of 80 kilovolts. The conjugate should be regular spherical and evenly dispersed, with a particle size distribution that conforms to the lognormal distribution and no obvious agglomeration.

5. The method for preparing an exosome-antibody-microsphere conjugate according to claim 4, characterized in that: The exosome marker antibodies added to the activated microsphere suspension include mouse IgG1 subtype CD63 monoclonal antibody, mouse IgG2b subtype CD9 monoclonal antibody, and rat IgG2a subtype CD81 monoclonal antibody.

6. The method for preparing an exosome-antibody-microsphere conjugate according to claim 5, characterized in that: The mass ratio of the three antibodies in the exosome marker antibody is 1:1:

1.

7. The method for preparing an exosome-antibody-microsphere conjugate according to claim 6, characterized in that: The centrifugal parameters optimized by response surface methodology include centrifugal speed, centrifugal time, centrifugal temperature and centrifugal acceleration.

8. The method for preparing an exosome-antibody-microsphere conjugate according to claim 7, characterized in that: The cross-linking conditions optimized by single factor experiment included cross-linking agent concentration, reaction temperature, reaction time, pH value and ionic strength.

9. The method for preparing an exosome-antibody-microsphere conjugate according to claim 8, characterized in that: The mathematical models used to predict purification efficiency include mass transfer equation, fluid dynamics equation, and heat transfer equation.

10. The method for preparing an exosome-antibody-microsphere conjugate according to claim 9, characterized in that: In the mathematical model, the mass transfer equation is used to calculate the migration rate of matter. The input is the concentration gradient and diffusion coefficient, and the output is the mass transfer flux. The fluid dynamics equation is used to calculate the motion state of the fluid. The input is the fluid density, viscosity, and pressure, and the output is the fluid velocity field. The heat transfer equation is used to calculate the temperature distribution. The input is thermal conductivity, specific heat capacity, and temperature gradient, and the output is heat flux.

Images