Biological ink, in-vitro humanized neural model and application of biological ink and in-vitro humanized neural model in construction of opioid drug classification prediction model and dose-effect relationship

Abstract

The invention belongs to the technical field of in-vitro drug evaluation model construction, and particularly relates to bio-ink, an in-vitro humanized neural model and application of the bio-ink in construction of an opioid drug classification prediction model and a dose-effect relationship. The bio-ink comprises a neuroma cell line, methacrylated gelatin, a photoinitiator and a basement membrane matrix. The in-vitro humanized neural model prepared on the basis of the bio-ink provided by the invention has cell-cell and cell-matrix microenvironment interaction characteristics, the complexity of in-vivo tissues can be more truly reduced, and the structure and function of human neurons can be highly biomimetic simulated on the cellular level; the biomimetic property is obviously superior to that of traditional two-dimensional culture and animal experiments, and the accuracy and reliability of drug toxicity and dependency evaluation are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of constructing in vitro drug evaluation models, and particularly relates to a bioink, an in vitro humanized nerve model, and their applications in constructing an opioid drug classification prediction model and a dose-effect relationship. Background Art

[0002] With the development of social economy, the demand and consumption of psychotropic drugs are on the rise, and the types of drug abuse are being updated and iterated at an accelerating pace, resulting in a lag in legal control. The situation of abuse of narcotic and psychotropic drugs remains severe and complex.

[0003] To address the issue of the lag in the control of new narcotic and psychotropic drugs in judicial practice, prevent and combat the substitution and abuse of psychotropic drugs, and ensure public safety and people's health, judicial organs have actively carried out relevant work but face many difficulties and challenges. On the one hand, there are numerous types of psychotropic drugs, with complex chemical structures and unclear mechanisms of action, making it difficult to accurately identify and classify them. On the other hand, there is a lack of sufficient experimental data and scientific basis, making it difficult to determine their specific degree of damage to human functions, the strength of addiction, and the relevant criminal liability and penalty standards.

[0004] Identifying the psychic toxicity or physiological / psychological dependence of psychotropic drugs is of great significance for judicial work and related research. Currently, the methods for evaluating the psychic toxicity or physiological / psychological dependence of drugs mainly include animal experiments and in vitro experiments. Among them, although animal experiments have relatively high biological relevance and reliability, there are problems such as animal welfare, ethics, cost, time, and scale. Although in vitro experiments have the advantages of flexibility and high efficiency, the current in vitro models are mainly two-dimensional cell culture models, which cannot simulate the in vivo environment, resulting in inaccurate evaluation results of drug absorption, metabolism, and toxicity. Summary of the Invention

[0005] To achieve the above object, the present invention can adopt the following technical solutions: On the one hand, the present invention provides a bioink, which includes a neuroma cell line, methacrylated gelatin, a photoinitiator, and a basement membrane matrix.

[0006] Preferably, the above bioink can meet one or more of the following conditions: (i) The neuroma cell line is selected from one or more of the SH-SY5Y cell line, SK-N-SH cell line, or SK-N-BE cell line; (ii) The photoinitiator is selected from one or more of LAP, Irgacure 2959, or Irgacure 819; (iii) The neuroma cell line is a single-cell suspension; (iv) In the bioink, the mass percentage of methacrylated gelatin is 4% - 6%, and the mass percentage of basement membrane matrix is 20% - 30%; (v) In the bioink, the cell density of the neuroma cell line is 1.2x10^6 / ml - 1.2x10^7 / ml.

[0007] Preferably, the above bioink can be any one of the following: 1) In the bioink, the mass percentage of methacrylated gelatin can be 4.2%, and the mass percentage of basement membrane matrix can be 30%; 2) In the bioink, the mass percentage of methacrylated gelatin can be 5.6%, and the mass percentage of basement membrane matrix can be 20%.

[0008] On the other hand, the present invention provides an in vitro humanized nerve model, which is prepared by 3D printing using the bioink in the present invention as a raw material.

[0009] On yet another aspect, the present invention provides a preparation method of the in vitro humanized nerve model in the present invention, which may include: printing the bioink into at least two droplets by microextrusion printing as an in vitro humanized nerve model for high-throughput drug toxicological evaluation; or printing the bioink into a three-dimensional grid by microextrusion printing as an in vitro humanized nerve model for drug dependence evaluation.

[0010] Preferably, in the above preparation method, the cell amount in each droplet can be 1.0x10^ 5 cells - 1.0x10^ 7 cells.

[0011] On yet another aspect, the present invention provides an application of the in vitro humanized nerve model in the present invention in evaluating the psychotoxicity or physiological / psychological dependence of drugs.

[0012] On yet another aspect, the present invention provides an opioid drug classification prediction model based on the in vitro humanized nerve model in the present invention. The opioid drug classification prediction model includes: Step 1, obtain the data features of the neurotoxicity target data, randomly select candidate features from the data features as the root node of the classification tree, and select the best splitting point of the data features based on Gini importance for node splitting to obtain the child nodes of the classification tree. Recursively select the best splitting point of the data features until the node splitting meets the stopping condition to obtain the classification tree model; Step 2, repeat Step 1 to obtain multiple classification tree models; Step 3, perform sampling processing on the neurotoxicity target data to obtain multiple neurotoxicity target sub-datasets. Among them, each neurotoxicity target sub-dataset is used as a classification training sample, and the number of neurotoxicity target sub-datasets is the same as the number of classification tree models; Step 4: Use a classification training sample to train a classification tree model to obtain classification prediction labels, calculate the accuracy rate between the classification prediction labels and the test labels, and complete the training process of a single classification tree model when the accuracy rate reaches the preset accuracy threshold; Step 5: Repeat Step 4, and perform parallel training using a classification training sample corresponding to a classification tree model, and integrate all the trained classification tree models to obtain an opioid classification prediction model; In Step 1, the data characteristics of the neurotoxicity target data are obtained through the in vitro humanized nerve model in the present invention.

[0013] Preferably, in the above-mentioned opioid classification prediction model, after performing parallel training using a classification training sample corresponding to a classification tree model, it further includes: performing a feature importance ranking on the data characteristics of the neurotoxicity target data to obtain the contribution degree of the data characteristics for classification prediction, and constructing a dose-effect relationship of opioids according to the contribution degree of the data characteristics.

[0014] More preferably, in the above-mentioned opioid classification prediction model, constructing a dose-effect relationship of opioids according to the contribution degree of the data characteristics includes: Use the univariate toxicity calculation formula to calculate the single toxicity ratio of opioids on a single biological characteristic; Take the contribution degree of the data characteristics as the toxicity weight, calculate the toxicity ratio of opioids on each biological characteristic, and perform a summation operation on the toxicity ratios of opioids on each biological characteristic to obtain the total weighted toxicity ratio score; Use the total weighted toxicity ratio calculation formula to calculate the total weighted toxicity ratio of opioids at multiple concentrations; Perform a sensitivity analysis on the single toxicity ratio, the total weighted toxicity ratio score, and the total weighted toxicity ratio, and obtain the target weight scheme according to the sensitivity analysis results, and calculate the comprehensive toxicity ratio of opioids.

[0015] The beneficial effects of the present invention at least include: (1) The in vitro humanized nerve model prepared based on the bioink provided by the present invention has the characteristics of cell-cell and cell-matrix microenvironment interaction, can more truly restore the complexity of in vivo tissues, can highly biomimic the structure and function of human neurons at the cell level, and exhibits physiological characteristics similar to natural nerve tissues. This high degree of biomimesis is significantly better than traditional two-dimensional culture, and improves the accuracy and reliability of drug toxicity and dependence assessment.

[0016] (2) The method for constructing an in vitro humanized neural model provided by the present invention can complete the self-assembly and printing of cell clusters in a short time, effectively saving time and cost compared with animal experiments (verifying drug toxicity in animal experiments requires three processes: model establishment, sampling, and analysis, and the drug dosage used is much higher than that of in vitro models, and the model establishment period is long). Moreover, this in vitro humanized neural model can replace animal experiments, reduce the species differences in animal experiments (there are large individual differences among animals in animal experiments and species gene differences from humans), and improve the experimental accuracy.

[0017] (3) Based on the bioink provided by the present invention, a large number of in vitro humanized neural models can be rapidly prepared on one well plate, and toxicity and dependence screening under multiple experimental conditions can be completed simultaneously on one well plate, meeting the simultaneous evaluation requirements of different drugs and greatly improving the experimental efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic flow chart for rapidly constructing SH-SY5Y neural models (SH-SY5Y cell clusters) based on a low-adhesion microporous array; among them, (a) is a schematic diagram of a PDMS membrane; (b) is a schematic diagram of a 24-well plate device with a selectively breathable PDMS membrane at the bottom; (c) is that the PDMS membrane with a microporous array can be placed into the culture well for cell culture; Figure 2 It is the statistical chart of cell viability; Figure 3 It is the statistical chart of the influence of different concentrations of fentanyl on dopamine release; Figure 4 It is the statistical chart of the influence of different concentrations of fentanyl on the expression levels of four opioid receptors; Figure 5 It is the verification of the changes in the expression levels of genes related to neuron differentiation by qPCR; among them, (a) is the expression level of the undifferentiated neuron marker Tuj1 in each system; (b) is the expression level of the mature differentiated neuron marker MAP2 in each system; (c) is the expression level of the neuron differentiation marker neurofilament light chain protein NEFL in each system; (d) is the expression level of the neurofilament middle protein NEFM, a marker for the formation and maintenance of nerve fibers, in each system; one-way ANOVA followed by Holm-Šídák post hoc test, *p<0.05, **p<0.01, ***p<0.001; the data are from 3 independent experiments; Figure 6 It is the verification of the expression of the gene TH related to neuron differentiation by qPCR; among them, one-way ANOVA followed by Holm-Šídák post hoc test, *p<0.05, **p<0.01, ***p<0.001; the data are from 3 independent experiments; Figure 7It is the confusion matrix of the random forest model, where F is fentanyl, H is diacetylmorphine, M is morphine, F-1 is the treatment condition of 1 uM fentanyl, and so on; Figure 8 It is about the dopamine release amount, cell survival rate, and the contribution of different gene expression levels to classification prediction. Specific implementation manners

[0019] The examples given are for better illustration of the present invention, but the content of the present invention is not limited only to the examples given. Therefore, those skilled in the art who make non-essential improvements and adjustments to the implementation manners according to the above invention content still fall within the protection scope of the present invention.

[0020] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. Unless having significantly different meanings in the context, the expressions in singular form include those in plural form. As used herein, it should be understood that terms such as "including", "having", "containing" are intended to indicate the existence of features, numbers, operations, components, parts, elements, materials or combinations. Terms of the present invention are disclosed in the specification, and are not intended to exclude the possibility of the existence or addition of one or more other features, numbers, operations, components, parts, elements, materials or their combinations. As used herein, according to circumstances, " / " can be interpreted as "and" or "or".

[0021] An embodiment of the present invention provides a bioink, which includes a neuroblastoma cell line, methacrylated gelatin (hereinafter also referred to as GelMA), a photoinitiator, and a basement membrane matrix (hereinafter also referred to as Matrigel).

[0022] It should be noted that GelMA is a photocrosslinkable hydrogel. Based on the dual characteristics of temperature sensitivity and photocrosslinking, it has good biocompatibility and plasticity. Using GelMA hydrogel as a bioink to print a neuroblastoma cell line can manufacture a nerve tissue scaffold with complex shapes and porous structures, providing a suitable growth environment for neuroblastoma cells; moreover, the RGD peptide in GelMA specifically enhances the adhesion of SH-SY5Y cells and promotes the expression of the immature neuron marker tuj1; the matrix can induce neuron differentiation and promote the expression of the mature neuron marker map2; in addition, GelMA and the basement membrane matrix together can prepare a biological model that can simulate the in vivo environment and simulate human nerve functions (such as the characteristics and functions of human brain neurons), promoting the expression of the dopaminergic neuron marker TH; when the two matrix gels are combined, the expression levels of the three markers TUJ1, MAP2, and TH are all induced to the highest.

[0023] It should also be noted that in the present invention, the cell clusters of the biological model constructed based on GelMA and basement membrane matrix have better stability than the existing cell clusters constructed based on low-adhesion microporous arrays, and at the same time have the advantages of cell clusters (such as having a dense extracellular matrix that reduces the efficiency of exogenous drug uptake and transport and having biomimicry).

[0024] In some specific examples, in the above biological ink, the neuroblastoma cell line can be selected from one or more of the SH-SY5Y cell line, SK-N-SH cell line or SK-N-BE cell line.

[0025] It should be noted that the neuroblastoma cell line in the present invention refers to a cell line with neuronal characteristics or expressing neural markers, such as the SH-SY5Y cell line, SK-N-SH cell line or SK-N-BE cell line; among them, the SH-SY5Y cell line is preferably selected. The SH-SY5Y cell line is a human neuroblastoma cell line with various neuronal characteristics and can be used to study the protective or damaging effects of drugs on human brain neurons. Moreover, within 0-6 days of culturing the biological model constructed with the SH-SY5Y cell line, the cell clusters maintain high activity and the proliferation ratio is significantly down-regulated, indicating that the cells exhibit a differentiated phenotype, which is a cell line suitable for constructing a biological model.

[0026] In some specific examples, the above biological ink can meet one or more of the following conditions: (i) The neuroblastoma cell line is selected from one or more of the SH-SY5Y cell line, SK-N-SH cell line or SK-N-BE cell line; (ii) The photoinitiator is selected from one or more of LAP, Irgacure 2959 or Irgacure 819; (iii) The neuroblastoma cell line is a single-cell suspension; (iv) In the biological ink, the mass percentage of methacrylated gelatin is 4% - 6%, and the mass percentage of basement membrane matrix is 20% - 30%; specifically, when the mass percentage of methacrylated gelatin is higher than 8%, the Young's modulus is too large and not suitable for cell culture. In the present invention, 4% - 6% is preferably selected, such as 4.2%, 4.5%, 4.7%, 4.9%, 5.2%, 5.5%, 5.7% or 5.9%, etc.; in addition, when the mass percentage of basement membrane matrix is too large, it cannot be in a semi-gel state during printing, resulting in printing difficulties. In addition, exceeding 30% will not have a positive guidance on the expression of neuronal markers and will increase the economic cost. In the present invention, 20% - 30% is preferably selected, such as 23%, 25%, 27% or 29%, etc.; (v) In the biological ink, the cell density of the neuroblastoma cell line is 1.2x10^6 / ml - 1.2x10^7 / ml.

[0027] In some specific examples, any one of the following can be preferably selected from the above bioinks: 1) In the bioink, the mass percentage of GelMA is 4.2%, and the mass percentage of the basement membrane matrix is 30%; 2) In the bioink, the mass percentage of GelMA is 5.6%, and the mass percentage of the basement membrane matrix is 20%.

[0028] It should be noted that the bio - models prepared from the above two groups of bioinks are more accurate than the bio - models prepared from bioinks with other ratios in terms of evaluating the damage or addiction of psychotropic drugs to human body functions and evaluating the effects of psychotropic drugs causing physiological / psychological dependence.

[0029] In some specific examples, the neuroma cell line in the above bioink can be in the form of a single - cell suspension.

[0030] It should be noted that the neuroma cell line in the bioink of the present invention can be a single - cell suspension, and the preparation of the single - cell suspension is well - known in the art. For example, it can be obtained by trypsin digestion.

[0031] The embodiment of the present invention also provides an in - vitro humanized nerve model, which is prepared by 3D printing using the bioink in the present invention as a raw material.

[0032] It should be noted that the three - dimensional bioprinting technology based on 3D printing is a technology that precisely locates and stacks biological materials, cells, etc. according to a preset three - dimensional structure to generate artificial tissues or organs with specific shapes and functions. The bio - models constructed based on this technology can be used to evaluate the effects of drugs on different tissues or organs, and have high customizability, complexity, flexibility, and precision. They can realize the spatial distribution and interaction of multiple cell types, multiple biological materials, and multiple biological signals, so as to simulate the microscopic structure and macroscopic functions of natural tissues.

[0033] The embodiment of the present invention also provides a preparation method of the in - vitro humanized nerve model in the present invention, which includes: printing the bioink into at least two droplets by micro - extrusion as an in - vitro humanized nerve model for high - throughput drug toxicology evaluation; or printing the bioink into a three - dimensional grid by micro - extrusion as an in - vitro humanized nerve model for drug dependence evaluation.

[0034] In some specific examples, in the above preparation method, the cell amount in each droplet is 1.0x10^ 5 cells - 1.0x10^ 7 cells.

[0035] It should be noted that the amount of cells in each droplet in the present invention is the limit carrying capacity of two-dimensional cell culture. However, in three-dimensional culture, the cells can exceed this carrying capacity after proliferation during culture.

[0036] The embodiments of the present invention also provide an application of the in vitro humanized neural model in the present invention in evaluating the psychotoxicity or physiological / psychological dependence of drugs.

[0037] In some specific examples, in the above application, the drugs include psychoactive drugs, anti-tumor drugs or central nervous system drugs.

[0038] It should be noted that psychoactive drugs, anti-tumor drugs or central nervous system drugs are all well-known drugs in the art; for example, psychoactive drugs can be caffeine-based, hallucinogens (such as lysergic acid diethylamide, mescaline, phencyclidine or ketamine, etc.), opioids (such as diacetylmorphine, morphine, methadone, dihydroetorphine, pethidine, buprenorphine or fentanyl, etc.); for another example, anti-tumor drugs can be carboplatin, vincristine sulfate or cisplatin, etc.; for another example, central nervous system drugs can be antipsychotic drugs (such as risperidone, aripiprazole, chlorpromazine, haloperidol, etc., these drugs relieve symptoms by blocking dopamine receptors in the brain or regulating other neurotransmitter systems), anti-epileptic drugs (such as carbamazepine, phenytoin sodium or sodium valproate, etc., these drugs prevent abnormal discharges by stabilizing the nerve cell membrane or regulating neurotransmitters), anti-depressant drugs (such as tricyclic antidepressants (such as amitriptyline, imipramine or clomipramine, etc.), selective serotonin reuptake inhibitors (such as paroxetine, sertraline, fluoxetine), 5-HT1A receptor partial agonists, etc., these drugs play an anti-depressant role by increasing the level of serotonin in the brain, activating 5-HT1A receptors or other mechanisms), anti-anxiety drugs (such as diazepam, lorazepam, etc., these drugs play a role by enhancing the effect of gamma-aminobutyric acid in the brain, which is an inhibitory neurotransmitter), ganglionic blocking drugs (such as mecamylamine, trimethaphan or hexamethonium, etc., these drugs can block the transmission of nerve impulses), nerve ending blocking drugs (such as oryzanol, adenosylcobalamin or methylcobalamin, etc.), adrenergic receptor blocking drugs (such as metoprolol tartrate, metoprolol succinate or bisoprolol, etc.), etc.

[0039] The embodiments of the present invention also provide an opioid drug classification prediction model based on the in vitro humanized neural model in the present invention. The opioid drug classification prediction model includes: Step 1: Obtain the data characteristics of the neurotoxicity target data, randomly select candidate features from the data characteristics as the root node of the classification tree, and select the best splitting point of the data characteristics based on Gini importance to perform node splitting to obtain the child nodes of the classification tree. Recursively select the best splitting point of the data characteristics until the node splitting meets the stopping condition to obtain the classification tree model; Step 2: Repeat Step 1 to obtain multiple classification tree models; Step 3: Sample the neurotoxicity target data to obtain multiple neurotoxicity target sub-datasets. Each neurotoxicity target sub-dataset serves as a classification training sample, and the number of neurotoxicity target sub-datasets is the same as the number of classification tree models; Step 4: Use a classification training sample to train a classification tree model to obtain classification prediction labels, calculate the accuracy rate of the classification prediction labels and the test labels. When the accuracy rate reaches the preset accuracy threshold, the training process of a single classification tree model is completed; Step 5: Repeat Step 4, and use one classification training sample corresponding to one classification tree model for parallel training. Integrate all the trained classification tree models to obtain an opioid classification prediction model; In Step 1, the data features of the neurotoxicity target data are obtained through the in vitro humanized neural model in the present invention.

[0040] In some specific examples, in the above opioid classification prediction model, after using one classification training sample corresponding to one classification tree model for parallel training, it further includes: ranking the importance of the data features of the neurotoxicity target data to obtain the contribution degree of the data features for classification prediction, and constructing the dose-effect relationship of opioids according to the contribution degree of the data features.

[0041] In some specific examples, in the above opioid classification prediction model, constructing the dose-effect relationship of opioids according to the contribution degree of the data features includes: Using the univariate toxicity calculation formula to calculate the single toxicity ratio of opioids on a single biological feature; Taking the contribution degree of the data features as the toxicity weight, calculating the toxicity ratio of opioids on each biological feature, and performing a summation operation on the toxicity ratios of opioids on each biological feature to obtain the total weighted toxicity ratio score; Using the total weighted toxicity ratio calculation formula to calculate the total weighted toxicity ratio of opioids at multiple concentrations; Performing sensitivity analysis on the single toxicity ratio, the total weighted toxicity ratio score, and the total weighted toxicity ratio, and obtaining the target weight scheme according to the sensitivity analysis result, and calculating the comprehensive toxicity ratio of opioids.

[0042] To better understand the present invention, the content of the present invention will be further clarified below with specific examples, but the content of the present invention is not limited to the following examples.

[0043] I. Construction of Humanized Neural Model In the following examples, GelMA and photoinitiator LAP are from EFL Company (Engineering For Life); Matrigel is from Corning®, and the product number is 356237.

[0044] Example 1 (1) Cell seeding: The SH-SY5Y cell line (human neuroblastoma cell line) (cell concentration of 1×10^4 cells / mL) was seeded in DMEM / F12 medium containing 10% fetal bovine serum and 1% penicillin-streptomycin double antibody (medium model: 11320033, Thermo, a 1:1 (V / V) mixture of DMEM and Ham's F-12, and this formulation contains the glucose, amino acids, and vitamins of DMEM and F-12) (hereinafter also referred to as complete medium), and cultured in an incubator at 37°C with a volume fraction of 5% carbon dioxide. When the cells grew to 90% confluence, stable passage was carried out (i.e., 2D cultured cells were obtained); (2) Fabrication of neural microspheres (hereinafter also referred to as cell microspheres): When the cells proliferated to 90% confluence in a two-dimensional culture dish, the proliferated SH-SY5Y cells were digested into a single-cell suspension with 0.25% trypsin; then the single-cell suspension was seeded into an oxygen-permeable device with a regular hexagonal honeycomb micropore array (by pouring polydimethylsiloxane (PDMS) onto an etched silicon plate mold, an oxygen-permeable device with a regular hexagonal honeycomb micropore array was prepared, with a single pore inner diameter of 126 μm and a PDMS membrane thickness of about 0.5 mm) and cultured for 48 h (2 days) to aggregate into neural microspheres; (3) Bioink mixing: The neural microspheres were centrifuged at 1000 rpm for 3 min, and then the supernatant was discarded. After resuspending the cells, bioink was prepared (the photoinitiator LAP powder was dissolved in D-PBS buffer to form a 0.25% mass fraction solution, and then GelMA was dissolved in this solution to form a GelMA solution with a GelMA mass concentration of 6%. The GelMA solution and Matrigel were mixed at a volume ratio of 7:3 to obtain a mixture, and the mixture and the single-cell suspension were mixed to obtain bioink; in the bioink, the final GelMA mass fraction was 4.2%, the Matrigel mass fraction was 30%, and the cell density was 1.0x10^ 7 / mL); (4)Fabrication of droplet-shaped neural model: The prepared bioink was aspirated into a 3 mL syringe, assembled with a dispensing needle with an inner diameter of 0.34 mm onto a 3D printer (SunP 3D BioPrinter SunP BioMaker 4) (photocured for 20 s after printing); in the high-throughput toxicological characterization study experiment in a 96-well plate, 10 μl of droplets were printed in each well, and the cell amount in each well was 1.0x10^5, which played a better role in subsequent experimental measurements; among them, the printing parameters included: the size of the dispensing needle used for printing was: 23G (inner diameter 0.34 mm), printing speed: 4 mm / s; extrusion speed: 1 mm³ / s, ensuring that the droplet extrusion amount was 10 uL. In addition, a three-dimensional grid was printed in a 60 mm culture dish by micro-extrusion as a drug dependence evaluation model (printing parameters: 23G (inner diameter 0.34 mm), printing speed: 4 mm / s; extrusion speed: 1 mm³ / s, extrusion morphology: 15×15×1.2 mm grid-like printing layer height 0.4, printing layer 3 layers), and the specification of the grid structure was 15 mm x 15 mm x 1.2 mm, and about 100 μL of bioink was required for each structure.

[0045] Example 2 The difference between Example 2 and Example 1 is that the bioink mixture in step (3) is different, and the others are the same as Example 1; in Example 2, the bioink mixture includes: centrifuging the neural microspheres at 1000 rpm for 3 min and then discarding the supernatant, resuspending the cells and preparing the bioink (dissolving the LAP powder of the photoinitiator in D-PBS buffer solution to form a 0.25% solution, and then dissolving GelMA with this solution to form a GelMA solution with a GelMA mass concentration of 6%, mixing the GelMA solution and Matrigel to obtain a mixture, and mixing the mixture and the single-cell suspension to obtain the bioink; in the bioink, the final GelMA mass fraction is 5.6%, the Matrigel mass fraction is 20%, and the cell density is 1.0x10^ 7 / mL).

[0046] Comparative Example 1 According to Figure 1 The rapid construction of the SH-SY5Y neural model (SH-SY5Y cell clusters) was achieved based on the low-adhesion microporous array as shown in the process schematic diagram as follows: By pouring polydimethylsiloxane (PDMS) into an etched silicon plate mold, an oxygen-permeable device with a regular hexagonal honeycomb microporous array was prepared, with a single-hole inner diameter of 126 μm and a PDMS membrane thickness of about 0.5 mm; this low-adhesion microenvironment effectively guided the self-assembly of SH-SY5Y cells. After 48 hours of culture, microscopic imaging confirmed that a large number of cell clusters with uniform morphology, high stability and good activity were formed in the device, providing a reliable in vitro model for the efficient construction of the neural model.

[0047] However, SH-SY5Y are semi-adherent cells. Preliminary experiments have confirmed that they will show a suspended state under environmental stress. The SH-SY5Y cell clusters planted in the PDMS oxygen-permeable microporous array device cannot meet the subsequent neurotoxicity evaluation. In the presence of drugs, the morphology and structure of the cell clusters are unstable and are prone to deformation or rupture.

[0048] Comparative Example 2 Comparative Example 2 is the 2D cultured cells prepared in step (1) of Example 1.

[0049] Comparative Example 3 The difference between Comparative Example 3 and Example 2 is that the mixing of the bioink in step (3) is different, and the others are the same as Example 2; in Example 2, the mixing of the bioink includes: centrifuging the neural microspheres at 1000 rpm for 3 min and then discarding the supernatant, resuspending the cells and then preparing the bioink. Specifically, the photoinitiator LAP powder is dissolved in D-PBS buffer to form a 0.25% solution, and then GelMA is dissolved in this solution to form a GelMA solution with a GelMA mass concentration of 6%. The GelMA solution and the single-cell suspension are mixed to obtain the bioink; in the bioink, the cell density is 1.0x10^ 7 / mL.

[0050] Comparative Example 4 The difference between Comparative Example 3 and Example 2 is that the mixing of the bioink in step (3) is different, and the others are the same as Example 2; in Example 2, the mixing of the bioink includes: centrifuging the neural microspheres at 1000 rpm for 3 min and then discarding the supernatant, resuspending the cells and then preparing the bioink. Specifically, the Matrigel solution and the single-cell suspension are mixed to obtain the bioink; in the bioink, the cell density is 1.0x10^ 7 / mL.

[0051] Comparative Example 5 The differences between Comparative Example 3 and Example 1 are as follows: in step (3), the bioink mixtures are different. Additionally, after the printing in Comparative Example 3 is completed, it is soaked in a calcium chloride solution dissolved in physiological saline at a concentration of 2% for 1 minute to complete the curing process, and the rest is the same as in Example 1. In Comparative Example 3, the bioink mixture preparation includes: centrifuging the neural microspheres at 1000 rpm for 3 minutes and discarding the supernatant, then resuspending the cells to prepare the bioink, specifically including: dissolving Sodium alginate (sigma, 180947) in D-PBS buffer solution to form a sodium alginate solution with a sodium alginate mass concentration of 2%; dissolving gelatin (sigma, V900863) in D-PBS buffer solution to form a 10% gelatin solution; mixing the sodium alginate solution and the gelatin solution in a volume ratio of 1:1 to obtain a mixture, and mixing the mixture with the single-cell suspension to obtain the bioink; in the bioink, the final sodium alginate mass fraction is 1%, the gelatin mass fraction is 5%, and the cell density is 1.0x10^ 7 / mL.

[0052] Comparative Example 6 The differences between Comparative Example 4 and Example 1 are as follows: in step (3), the bioink mixtures are different. Additionally, after the printing in Comparative Example 4 is completed, it is soaked in a calcium chloride solution dissolved in physiological saline at a concentration of 2% for 1 minute to complete the curing process, and the rest is the same as in Example 1. In Comparative Example 3, the bioink mixture preparation includes: centrifuging the neural microspheres at 1000 rpm for 3 minutes and discarding the supernatant, then resuspending the cells to prepare the bioink, specifically including: dissolving Sodium alginate (sigma, 180947) in D-PBS buffer solution to form a sodium alginate solution with a sodium alginate mass concentration of 4%; dissolving gelatin (sigma, V900863) in D-PBS buffer solution to form a 10% gelatin solution; mixing the sodium alginate solution and the gelatin solution in a volume ratio of 1:1 to obtain a mixture, and mixing the mixture with the single-cell suspension to obtain the bioink; in the bioink, the final sodium alginate mass fraction is 2%, the gelatin mass fraction is 5%, and the cell density is 1.0x10^ 7 / mL).

[0053] II. Application of the humanized neural model (I) Appearance of differentiated phenotypes in cells The clusters prepared in Example 1 maintained high activity during the culture (37°C, CO2) for 0 - 6 days as a high-throughput drug toxicology evaluation model and as a drug dependence evaluation model ( Figure 2 ), and the significantly down-regulated proliferation ratio indicated the appearance of differentiated phenotypes in the cells.

[0054] (II) Study on the toxicological characterization of fentanyl (1)The high-throughput drug toxicology evaluation models prepared in Example 1, Example 2, and Comparative Examples 2 to 6 were cultured for 24 h (37 °C, CO2) respectively; (2)After the culture was completed, 100 μL of fentanyl solutions with different concentrations (1 μM, 10 μM, 100 μM) (the fentanyl hydrochloride was dissolved in methanol to prepare a stock solution with a concentration of 100 mM, and then diluted to different concentrations using cell culture medium) were added for incubation for 24 h; (3)After the incubation was completed, a CCK8 kit (YEASEN, 40203ES60) was used to detect the effects of fentanyl at different concentrations on cell proliferation activity (the detection method was referred to the kit instruction manual), and thus the fentanyl toxicological characterization data were obtained.

[0055] (III)Fentanyl neurodependence characterization An increase in the expression level of opioid receptors indicates an increase in the neurodependence of cells on fentanyl, which can be used as a reference standard for quantifying fentanyl crimes; opioid receptor desensitization and down-regulation of expression levels are the main mechanisms of opioid tolerance; while an increase in the expression levels of dopamine and dopamine receptors, and a decrease after tolerance are the main mechanisms of opioid addiction. Therefore, the effects of psychotropic drugs on multi-dimensional neurotoxicity indicators such as dopamine release in an in vitro 3D neural model, the expression of four opioid receptors (μ-opioid receptor, κ-opioid receptor, δ-opioid receptor, and nociceptin / orphanin FQ receptor (NOP)), and the expression of dopamine receptors can be evaluated.

[0056] The specific steps in this characterization are as follows: (1)The drug dependence evaluation models prepared in Example 1, Example 2, and Comparative Examples 2 to 6 were cultured in 35 mm dishes respectively, 2 mL of medium was added to each dish, and they were cultured for 24 h (37 °C, CO2); (2)After the culture was completed, fentanyl solution (the fentanyl hydrochloride was dissolved in methanol to prepare a stock solution with a concentration of 100 mM, and then diluted to different concentrations using methanol) was added to make the concentration of fentanyl in the medium 10 μM; then it was incubated for 24 h; (3)After the incubation was completed, the GelMA scaffold was dissolved with GelMA lysate (add the working solution that can submerge the GelMA gel block in the dish (500 μL was added to each structure), and pipette repeatedly with a pipette gun to separate the gel block from the bottom of the plate and break it sufficiently. The smaller the gel block, the faster the lysis rate) (sterile lysis in a 37 °C incubator, observe the lysis situation under a microscope every 15 minutes); (4)After sufficient lysis (generally 20 min is required), centrifuge at 1000 rpm for 5 min, discard the supernatant, add 5 mL of complete medium, repeat the washing and centrifugation once to obtain cell samples; (5) The expression levels of μ-type, δ-type, and σ-type opioid receptors were measured using a qPCR kit (SYBR GREEN I Master Mix (11184ES08, YEASEN), and the test method was referred to the kit instructions). At the same time, the expression level of secreted dopamine was measured using an enzyme-linked immunosorbent assay kit (Boke Biotechnology: Human Dopamine ELISA Detection Kit).

[0057] The statistical results of dopamine release at different concentrations of fentanyl (0, 1 μM, 10 μM, 100 μM) are as Figure 3 shown, and the statistical results of the expression levels of four opioid receptors at different concentrations of fentanyl (0, 1 μM, 10 μM, 100 μM) are as Figure 4 shown.

[0058] In addition, a qPCR kit (SYBR GREEN I Master Mix (11184ES08, YEASEN), and the test method was referred to the kit instructions) was used to detect the expression of neuronal differentiation biomarkers Tuj1, MAP2, NEfL, NEfM, and TH in the cell samples obtained from the models in Examples 1 to 2 and Comparative Examples 2 to 6. The results showed (see Figure 5 and Figure 6 ) that the expression levels of Tuj1, MAP2, NEfL, NEfM, and TH in Examples 1 and 2 were significantly increased (the comprehensive analysis of the expression levels of the neural markers TUJ1, MAP2, and TH in the model of Example 2 was higher than that of Example 1), indicating that the 3D neural model in the present invention has good biological activity and neural-related functionality, can be used for subsequent high-throughput drug screening and neurotoxicity assessment, and the expression levels of neuronal differentiation biomarkers in this mixed system were significantly increased compared with the pure GelMA and Matrigel systems or other mixed systems.

[0059] (IV) Construction of a drug toxicology evaluation model and an opioid drug classification prediction model and dose-effect relationship as a drug dependence evaluation model In the embodiments of the present invention, the data characteristics of the neurotoxicity target data refer to the relevant factors reflecting the characteristics of organisms affected by opioid drugs. For example, dopamine secretion, cell viability, apoptosis index (Bax / bcl2 ratio), and the expression levels of different representative genes (μ-type opioid receptor, κ-type opioid receptor, δ-type opioid receptor, nociceptin receptor (NOP), dopamine receptor, and mitochondrial dynamin).

[0060] As an embodiment of the present invention, the opioid drug classification prediction model includes: Step 1: Obtain the data characteristics of the neurotoxicity target data. Randomly select candidate features from the data characteristics as the root node of the classification tree, and select the best split point of the data characteristics based on Gini importance to perform node splitting to obtain the classification tree child nodes. Recursively select the best split point of the data characteristics until the node splitting meets the stopping condition to obtain the classification tree model; Step 2: Repeat Step 1 to obtain multiple classification tree models; Step 3: Perform sampling processing on the neurotoxicity target data to obtain multiple neurotoxicity target sub-datasets. Among them, each neurotoxicity target sub-dataset is used as a classification training sample, and the number of neurotoxicity target sub-datasets is the same as the number of classification tree models; Step 4: Use a classification training sample to train a classification tree model to obtain a classification prediction label, calculate the accuracy rate of the classification prediction label and the test label. When the accuracy rate reaches the preset accuracy threshold, complete the training process of a single classification tree model; Step 5: Repeat Step 4, and use a classification training sample corresponding to a classification tree model for parallel training. Integrate all the trained classification tree models to obtain an opioid classification prediction model.

[0061] Further, after using a classification training sample corresponding to a classification tree model for parallel training, it further includes: performing feature importance ranking on the data characteristics of the neurotoxicity target data to obtain the contribution degree of the data characteristics for classification prediction, and constructing the dose-effect relationship of opioids according to the contribution degree of the data characteristics.

[0062] In the embodiment of the present invention, using a classification training sample corresponding to a classification tree model for training can improve the training efficiency.

[0063] In the embodiment of the present invention, the node splitting meeting the stopping condition can be that the classification tree node reaches the maximum depth.

[0064] In the embodiment of the present invention, the opioid classification prediction model can adopt a random forest prediction model. Among them, Random Forest is a machine learning model based on ensemble learning, which improves the prediction performance and generalization ability by combining multiple decision trees.

[0065] Among them, the confusion matrix of the random forest model refers to Figure 7, a confusion matrix can be used to evaluate the prediction effect of an opioid classification prediction model, including: analyzing the classification performance of the model on different classes by using the correspondence between the true labels and predicted labels of each class in the confusion matrix; obtaining the correct prediction quantity of each class by analyzing the diagonal elements in the confusion matrix to further evaluate the classification accuracy of the model; evaluating the possible misclassification situations of the model on certain classes by combining the non - diagonal elements in the confusion matrix; based on the results of the confusion matrix, the model's parameter settings can be optimized or data augmentation can be performed to improve the overall classification performance of the model.

[0066] Referring to Figure 8 shown, it is a schematic diagram of the contribution degree of data features for classification prediction in the present invention.

[0067] As an embodiment of the present invention, constructing the dose - effect relationship of opioid drugs according to the contribution degree of data features includes: Calculating the single toxicity ratio of opioid drugs on a single biological feature by using the univariate toxicity calculation formula; Taking the contribution degree of data features as the toxicity weight, calculating the toxicity ratio of opioid drugs on each biological feature, and performing a summation operation on the toxicity ratios of opioid drugs on each biological feature to obtain the total weighted toxicity ratio score; Calculating the total weighted toxicity ratio of opioid drugs at multiple concentrations by using the total weighted toxicity ratio calculation formula; Performing a sensitivity analysis on the single toxicity ratio, the total weighted toxicity ratio score, and the total weighted toxicity ratio, obtaining the target weight scheme according to the sensitivity analysis results, and calculating the comprehensive toxicity ratio of opioid drugs.

[0068] In the embodiment of the present invention, calculating the single toxicity ratio of opioid drugs on a single biological feature by using the univariate toxicity calculation formula can adopt the following calculation formula: Wherein, is the single - variable toxicity ratio of the th biological feature, is the morphine concentration, is the diacetylmorphine concentration, is the th measurement function of the biological feature.

[0069] In the embodiment of the present invention, taking the contribution degree of data features as the toxicity weight, calculating the toxicity ratio of opioid drugs on each biological feature, and performing a summation operation on the toxicity ratios of opioid drugs on each biological feature to obtain the total weighted toxicity ratio score , can adopt the following calculation formula: Wherein, is the toxicity score for the th feature of the opioid drug, is the weight of the th biological feature, is the total number of biological features.

[0070] In the embodiments of the present invention, the formula for calculating the total weighted toxicity ratio can adopt the following formula: where is the number of concentrations tested.

[0071] In the embodiments of the present invention, by calculating the univariate toxicity ratio, the weighted toxicity ratio, and the total weighted toxicity ratio, it is possible to gradually analyze the toxicity mechanism and comprehensively compare the drug toxicities.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the purpose and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A bioink, characterized in that, It includes a neuroma cell line, methacrylated gelatin, a photoinitiator, and a basement membrane matrix.

2. The bioink according to claim 1, characterized in that, The bioink meets one or more of the following conditions: (i) The neuroma cell line is selected from one or more of the SH-SY5Y cell line, SK-N-SH cell line, or SK-N-BE cell line; (ii) The photoinitiator is selected from one or more of LAP, Irgacure 2959, or Irgacure 819; (iii) The neuroma cell line is a single-cell suspension; (iv) In the bioink, the mass percentage of methacrylated gelatin is 4% - 6%, and the mass percentage of the basement membrane matrix is 20% - 30%; (v) In the bioink, the cell density of the neuroma cell line is 1.2x10^6 / ml - 1.2x10^7 / ml.

3. The bioink according to claim 2, wherein The bioink is selected from any of the following: 1) In the bioink, the mass percentage of methacrylated gelatin is 4.2%, and the mass percentage of the basement membrane matrix is 30%; 2) In the bioink, the mass percentage of methacrylated gelatin is 5.6%, and the mass percentage of the basement membrane matrix is 20%.

4. An in vitro humanized neural model, characterized in that, It is prepared by 3D printing using the bioink described in any one of claims 1 to 3 as a raw material.

5. The method for preparing the in vitro humanized neural model according to claim 4, characterized in that, It includes: Printing the bioink into at least two droplets by microextrusion as an in vitro humanized nerve model for high-throughput drug toxicology evaluation; or printing the bioink into a three-dimensional grid by microextrusion as an in vitro humanized nerve model for drug dependence evaluation.

6. The preparation method according to claim 5, wherein The cell amount in each droplet is 1.0x10^ 5 cells - 1.0x10^ 7 cells.

7. Use of the in vitro humanized nerve model according to claim 6 in evaluating the psychotoxicity or physiological / psychological dependence of drugs.

8. The opioid classification prediction model based on the in vitro humanized neural model according to claim 4, characterized in that, The opioid drug classification prediction model includes: Step 1: Obtain the data features of the neurotoxicity target data, randomly select candidate features from the data features as the root node of the classification tree, and select the best splitting point of the data features based on Gini importance to perform node splitting to obtain the sub-nodes of the classification tree. Recursively select the best splitting point of the data features until the node splitting meets the stop condition to obtain the classification tree model; Step 2: Repeat Step 1 to obtain multiple classification tree models; Step 3: Perform sampling processing on the neurotoxicity target data to obtain multiple neurotoxicity target sub-datasets. Among them, each neurotoxicity target sub-dataset is used as a classification training sample, and the number of neurotoxicity target sub-datasets is the same as the number of classification tree models; Step 4: Use a classification training sample to train a classification tree model to obtain a classification prediction label, calculate the accuracy rate of the classification prediction label and the test label. When the accuracy rate reaches the preset accuracy rate threshold, complete the training process of a single classification tree model; Step 5: Repeat Step 4, and perform parallel training using one classification training sample corresponding to one classification tree model. Integrate all the trained classification tree models to obtain the opioid drug classification prediction model; In Step 1, the data features of the neurotoxicity target data are obtained through the in vitro humanized nerve model according to claim 4.

9. The opioid drug classification prediction model based on an in vitro humanized neural model according to claim 8, wherein After parallel training using one classification training sample corresponding to one classification tree model, it further includes: ranking the importance of data features of neurotoxicity target data to obtain the contribution degree of data features for classification prediction, and constructing the dose-effect relationship of opioid drugs based on the contribution degree of data features.

10. The opioid drug classification prediction model based on an in vitro humanized neural model according to claim 9, characterized in that, Constructing the dose-effect relationship of opioid drugs based on the contribution degree of data features includes: Calculating the single toxicity ratio of opioid drugs on a single biological feature using the univariate toxicity calculation formula; Taking the contribution degree of data features as the toxicity weight, calculating the toxicity ratio of opioid drugs on each biological feature, and performing a summation operation on the toxicity ratios of opioid drugs on each biological feature to obtain the total weighted toxicity ratio score; Calculating the total weighted toxicity ratio of opioid drugs at multiple concentrations using the total weighted toxicity ratio calculation formula; Performing a sensitivity analysis on the single toxicity ratio, the total weighted toxicity ratio score, and the total weighted toxicity ratio, obtaining the target weight scheme according to the sensitivity analysis results, and calculating the comprehensive toxicity ratio of opioid drugs.

Images