A method for producing an in vitro lung model using extracellular matrix proteins

Abstract

The present invention relates to a method for preparing an in vitro lung model that mimics the histological and functional characteristics of living lungs using lung-derived cell lines. An in vitro lung model mimicking living lung tissue was prepared by coating an extracellular matrix protein (laminin, collagen, or fibronectin) onto a Transwell insert membrane and then culturing lung-derived cells. It was confirmed that the in vitro lung model prepared in this manner possesses characteristics of living lung tissue well, such as maintaining cell barrier function, having a ciliary structure, expressing goblet cells, and triggering an immune response. Furthermore, by changing the type of cells used for culture to cells derived from lung diseases, it was possible to prepare a lung disease model that mimics lung tissue suffering from lung disease. The lung disease model prepared in this manner exhibits characteristics of lung disease well, such as increased secretion of inflammatory cytokines, and thus can be utilized for studying the mechanisms of lung diseases and screening for lung disease treatments.

Description

Technology Field

[0001] The present invention relates to an in vitro ( in vitro This is about a method for manufacturing a waste model. Background Technology

[0002] Lung diseases are a major cause of death and morbidity worldwide. Among them, Chronic Obstructive Pulmonary Disease (COPD) was the cause of 3.23 million deaths according to 2019 statistics and is the most commonly observed complication affecting the lungs. According to the World Health Organization (WHO), due to deteriorating air quality and the outbreak of COVID-19, the incidence of COPD is gradually increasing and is predicted to reach double the current level by 2030. COPD causes airway obstruction and emphysema, leading to lung tissue damage that necessitates continuous medication and reduces the quality of life. Currently, since treatments to regenerate damaged lung tissue are not available, there is no way to fundamentally treat COPD and lung damage. Therefore, there is a need to research new treatments for lung diseases.

[0003] To research treatments for lung diseases, a model for evaluating safety and efficacy must be established before conducting clinical trials. Traditionally, animal models have been utilized for this research; however, because animal lungs differ from human lungs in anatomical structure, immune systems, and inflammatory responses, there is no certainty that efficacy observed in animal models can be reproduced in humans. Consequently, there is a need to establish a research model to replace animal models.

[0004] Recently, research on new models such as organ-on-a-chip, organoids, and organoid formation using 3D bioprinting has been ongoing in academia and industry, but there are concerns that it is difficult to perform various analyses on a single model due to its small size (less than a millimeter), which may lead to lower reliability of evaluations.

[0005] Existing studies include a study on a three-dimensional respiratory tissue model using human primary airway epithelial cells (Tissue Engineering Part A 26.7-8 (2020): 432-440) and a patent on a method for producing lung organoids by co-culturing extracellular matrix (ECM) extracted from lung tissue and type 2 alveolar cells (Korean Published Patent 10-2022-0071898). However, existing studies differ from the present invention in that they use a method of three-dimensional culture using substrate substitutes such as Matrigel or collagen gel.

[0006] The inventors intended to manufacture an in vitro lung model capable of mimicking living lung tissue by immobilizing an ECM on a Transwell membrane, and confirmed that both healthy lung models and lung disease models could be manufactured by varying the types of cells cultured. Such lung models could be used to develop new treatments for lung diseases, particularly COPD. The problem to be solved

[0007] The object of the present invention is to provide a method for manufacturing an in vitro lung model that mimics the histological and functional characteristics of the lung. means of solving the problem

[0008] In order to achieve the above objective,

[0009] The present invention comprises the steps of: 1) coating an insert membrane of a transwell with an extracellular matrix protein; 2) seeding lung-derived cells onto the insert membrane of the transwell; and 3) culturing the cells in a dynamic culture that maintains a flow of culture medium in a lung-mimicking test tube ( in vitro ) Provides a method for manufacturing a waste model.

[0010] The present invention relates to a waste-mimicking test tube manufactured by the above manufacturing method ( in vitro ) Provides a closed model system.

[0011] The present invention provides a screening method for a lung disease treatment agent comprising: 1) a step of preparing an in vitro lung disease model prepared by the above preparation method using lung disease-derived epithelial cells; 2) a step of treating the in vitro lung disease model prepared through step 1) with a test substance; and 3) a step of determining whether symptoms recover by comparing a treatment group model treated with the test substance of step 2) with an untreated control group model. Effects of the invention

[0012] The present invention involves preparing an in vitro lung model that mimics living lung tissue by coating an extracellular matrix protein (laminin, collagen, or fibronectin) onto a Transwell insert membrane and then culturing lung-derived cells. It was confirmed that the in vitro lung model prepared in this manner possesses characteristics of living lung tissue well, such as maintaining cell barrier function, having a ciliary structure, expressing goblet cells, and exhibiting an immune response. Furthermore, by changing the type of cells used for culture to cells derived from lung disease, it was possible to prepare a lung disease model that mimics lung tissue suffering from lung disease. Since the lung disease model prepared in this manner exhibits characteristics of lung disease such as increased secretion of inflammatory cytokines, it can be utilized for studying the mechanisms of lung disease and screening for lung disease treatments. Brief explanation of the drawing

[0013] Figure 1 shows the results of confirming the characteristics of a model prepared using an immortalized lung cell line. Figure 1a shows the results of observing the external appearance of the model, and Figure 1b shows the results of measuring the transepithelial electrical resistance (TEER) of the epithelial cells. Figure 1c is a graph confirming the permeability of the cell barrier, and Figure 1d shows the results of performing immunofluorescence staining using ZO-1, a marker for tight junctions. Figure 2 shows the results of confirming the characteristics of a model prepared using lung-derived primary cells. Figure 2a shows the results of observing the cell morphology with an optical or electron microscope and performing immunofluorescence staining for ZO-1, Figure 2b is a graph summarizing TEER values ​​according to culture conditions, and Figure 2c is a graph confirming the permeability of the cell barrier. Figure 2d shows the results of performing hematoxylin & eosin (H&E) staining and immunofluorescence staining for acetylated α-tubulin or MUC5B on a cross-section of the prepared model. Figure 3 shows the results of comparing the CNBio NHBE-NHLF model, in which primary cells were co-cultured under dynamic culture conditions, with a commercially available airway model. Figure 3a shows the results of immunofluorescence staining for ZO-1, acetylated alpha-tubulin, or MUC5B and observations via electron microscopy, and Figure 3b is a graph summarizing the TEER values. Figure 4 shows the results of changes according to the type of transwell insert and membrane coating material in the manufacture of a closed imitation model. Figures 4a to 4d show the impedance values ​​under each condition, and Figure 4e shows the results of observing the characteristics according to the type of insert. Figure 5 shows the results of preparing a lung disease model using COPD-derived primary cells. Figure 5a shows the results of observing the cell morphology with an optical or electron microscope and the results of immunofluorescence staining for ZO-1, and Figure 5b is a graph summarizing the TEER values. Figure 5c shows the results of confirming the permeability of the cell barrier, and Figure 5d shows the results of H&E staining or immunofluorescence staining for acetylated alpha-tubulin or MUC5B on cross-sections of the model prepared under static and dynamic culture conditions. Figure 6 shows the results of comparing the lung disease model of the present invention with a commercially available lung disease model. Figure 6a shows the results of immunofluorescence staining and electron microscopy for ZO-1, acetylated alpha-tubulin, or MUC5B, and Figure 6b is a graph summarizing the TEER values. Figure 7 shows the results of measuring cytokine secretion levels using the healthy lung model and lung disease model of the present invention. Specific details for implementing the invention

[0014] The present invention will be described in detail below.

[0015] The present invention comprises the steps of: 1) coating an insert membrane of a transwell with an extracellular matrix protein; 2) seeding lung-derived cells onto the insert membrane of the transwell; and 3) culturing the cells in a dynamic culture that maintains a flow of culture medium in a lung-mimicking test tube ( in vitro ) Provides a method for manufacturing a waste model.

[0016] The extracellular matrix (ECM) is a cellular structure formed by various substances secreted by cells outside the cell. It not only provides structural support and intercellular connections but also plays an important role in cell differentiation and intercellular communication, including signal transduction. The extracellular matrix proteins may be one or more selected from the group consisting of laminin, fibronectin, and collagen.

[0017] The above lung-derived cells may be lung-derived immortalized cell lines or primary cells.

[0018] The above-mentioned lung-derived immortalized cell line may be BEAS-2B, a bronchial epithelial cell line, and / or A549, a lung cancer cell line.

[0019] The primary cells may be epithelial cells derived from the lungs and / or fibroblasts derived from the lungs. The epithelial cells may be derived from a healthy person or from a diseased lung. This can be selected depending on the nature of the model to be manufactured. Specifically, the primary cells may be, but are not limited to, normal human bronchial epithelial cells (NHBE), diseased human bronchial / tracheal epithelial cells (DHBE), and / or normal human lung fibroblasts (NHLF).

[0020] The cells used in the above manufacturing method may include two or more types of cells that are different from each other.

[0021] The flow of the above medium may be a flow rate of 0.1 μL / s to 1.0 μL / s, and preferably a flow rate of 0.3 μL / s to 0.7 μL / s.

[0023] The present invention relates to a waste-mimicking test tube manufactured by the above manufacturing method ( in vitro ) Provides a closed model system.

[0025] The present invention provides a screening method for a lung disease treatment agent comprising: 1) a step of preparing an in vitro lung disease model prepared by the above preparation method using lung disease-derived epithelial cells; 2) a step of treating the in vitro lung disease model prepared through step 1) with a test substance; and 3) a step of determining whether symptoms recover by comparing a treatment group model treated with the test substance of step 2) with an untreated control group model.

[0026] The above-mentioned test substance may be any one selected from the group consisting of natural compounds, synthetic compounds, RNA, DNA, polypeptides, enzymes, proteins, ligands, antibodies, antigens, metabolites of bacteria or fungi, and live molecules.

[0027] The above symptoms may be one or more selected from the group consisting of increased expression of goblet cells, increased expression of inflammatory cytokines, decreased tight junction formation, and decreased cell layer thickness.

[0028] If the expression of goblet cells in the treatment group decreases by more than 1.2 times, preferably more than 1.5 times, and more preferably more than 2 times compared to the untreated group, it can be determined that the substance under test has the potential to be effective as a therapeutic agent.

[0029] If the expression of inflammatory cytokines in the treatment group decreases by 1.2 times or more, preferably 1.5 times or more, and more preferably 2 times or more compared to the untreated group, it can be determined that the substance under test has the potential to be effective as a therapeutic agent. The inflammatory cytokines may be IP-10 and / or IL-6.

[0030] If the formation of tight junctions in the treatment group is 1.2 times or more, preferably 1.5 times or more, more homogeneously formed compared to the untreated group, it can be determined that the substance under test has the potential to be effective as a therapeutic agent.

[0031] If the thickness of the cell layer in the treatment group increases by more than 1.2 times, preferably more than 1.5 times, compared to the untreated group, it can be determined that the substance under test has the potential to be effective as a therapeutic agent.

[0033] In a specific embodiment of the present invention, a lung-mimicking model was prepared by culturing immortalized cell lines in the insert membrane of a Transwell under various conditions, such as single culture, co-culture, static culture, or dynamic culture. As a result, the highest integrity of the cell barrier was observed when A549 was placed at the apex and BEAS2B at the basal end and co-cultured under dynamic culture, and low cell permeability was confirmed for all models under all conditions (Fig. 1). However, when a lung-mimicking model was prepared using immortalized cell lines, it was observed that tight junctions were not formed normally (Fig. 1).

[0034] To complement this, lung-mimicking models were constructed using lung-derived primary cells instead of immortalized cell lines. It was confirmed that tight junctions were well formed in all lung-mimicking models constructed with primary cells (Fig. 2). Additionally, more cilia were observed in models cultured under dynamic conditions compared to static culture conditions (Fig. 2). Furthermore, when the integrity of the cell barrier was measured, high electrical resistance values ​​were observed in all models, indicating that the function of the cell barrier was maintained normally (Fig. 2).

[0035] As a result of comparing a healthy lung model prepared using primary cells with a commercially available healthy lung model, it was confirmed that tight junctions were better formed and ciliary structures were better developed in the lung model of the present invention (Fig. 3). Through this, it can be seen that the lung model of the present invention is more similar to actual in vivo lung tissue compared to the commercially available lung model.

[0036] In addition, to determine whether there were changes in the growth / differentiation and cell barrier function of the lung model depending on the type of ECM protein coated on the insert membrane or the type of insert of the transwell, real-time impedance was measured by coating with collagen or fibronectin or varying the type of insert, and as a result, it was confirmed that collagen could promote the growth and differentiation of the model more than fibronectin (Fig. 4).

[0037] A COPD model was prepared by seeding lung disease-derived epithelial cells at the apex of a Transwell insert membrane, and when co-cultured under dynamic culture conditions, excellent ciliary structure and cell barrier function were observed, and in particular, a high expression of goblet cells was observed (Fig. 5). The phenomenon of a high number of goblet cells is a characteristic frequently observed in COPD patients, indicating that the COPD model of the present invention can reproduce the characteristics of actual patients.

[0038] In addition, when comparing the COPD model of the present invention with a commercially available COPD model, it was observed that the formation of cilia structure and the integrity of the cell barrier were at similar levels, but the COPD model of the present invention had an incomplete tight junction structure and more expression of goblet cells was observed (Fig. 6).

[0039] It is known that immune responses occur in the lungs in vivo and cytokines are secreted, and that the secretion of inflammatory cytokines increases in lung diseases. Therefore, we confirmed whether cytokines are formed and secreted in the lung disease model of the present invention that mimics the lungs. Cytokine secretion was confirmed in both the healthy lung mimic model and the COPD model, and in particular, we observed that the secretion of inflammatory cytokines IP-10 and IL6 increased in the COPD model compared to the healthy lung model (Fig. 7). Through this, it can be seen that IP-10 and IL6 can be used as markers for COPD.

[0040] Through the present invention, an in vitro lung model mimicking a living lung was manufactured. It was confirmed that this lung model mimics actual lung tissue by forming ciliary structures and tight junctions, expressing goblet cells, and maintaining excellent cell barrier function. Furthermore, since it was confirmed that disease models can be manufactured depending on the type of cell used for production, disease models can be manufactured by varying the cell type as needed to study the pathological mechanisms of lung diseases or to develop therapeutic agents for lung diseases.

[0042] The present invention will be explained in more detail below through specific embodiments.

[0043] However, the following examples are merely illustrative of the present invention, and the content of the present invention is not limited by the following examples.

[0045] <Example 1> Optimization of Model Manufacturing Using Immortalized Cell Lines

[0046] To construct a lung-mimicking model using an immortalized lung cell line, the characteristics of the model were observed under various culture conditions. Culture conditions were classified into dynamic or static conditions, and further divided into single-cell culture or co-culture of two cell lines. Additionally, within the co-culture conditions, evaluations were conducted separately based on which cells were cultured on the apical and basal sides.

[0047] For static conditions, cell cultures were cultured in a standard 24-well plate without medium flow, and for dynamic conditions, CNBio’s PhysiMimix™, a microfluidic 3D system, was used to induce a continuous medium flow at a flow rate of 0.5 μL in the basolateral compartment of the insert.

[0048] To enhance the complexity of intercellular interactions observed in actual lungs, a method for co-culturing A549 and BEAS-2B cells was devised, and conditions were divided by changing the type of cells cultured at the apex or basal region. As a result, 1) a static single-cell culture condition in which 10,000 BEAS-2B cells were statically cultured (hereinafter referred to as 'Static BEAS2B'); 2) a static co-culture condition in which 10,000 BEAS-2B cells were statically cultured at the apex and 10,000 A549 cells at the basal region (hereinafter referred to as 'Static BEAS2B-A549'); 3) a dynamic co-culture condition in which 10,000 BEAS-2B cells were dynamically cultured at the apex and 10,000 A549 cells at the basal end (hereinafter referred to as 'CNBIO 10,000 BEAS2B-A549'); and 4) a dynamic co-culture condition in which 10,000 A549 cells were dynamically cultured at the apex and 10,000 BEAS-2B cells at the basal end (hereinafter referred to as 'CNBIO 10K A549-BEAS2B'), and four conditions were observed.

[0049] Specifically, two immortalized cell lines (BEAS-2B and A549) were cultured in Dulbecco's Modified Eagle Medium (DMEM) and placed in transwells at a rate of 4 x 10 6 Pores / cm 2 A standard Corning (Cat: COR3470, Corning, Life Sciences) insert membrane was used, and the membrane was coated with 2.5 μg / mL of laminin.

[0051] <1-1> Function of Cell Barrier and Confirmation of Cell Morphology

[0052] Transepithelial electrical resistance (TEER) measurements were performed on cultures under each condition to verify the integrity of the cell barrier. Cell barrier integrity is useful for predicting drug permeability, drug transport, and drug interactions. For TEER measurements, the apical and basal compartments were filled with fresh medium, and the TEER values ​​were measured. The TEER value for each model was calculated by subtracting the TEER value from an insert filled only with medium in a cell-free chamber (background value). Then, the membrane surface area of ​​0.33 cm² was added to that value. 2 Multiply by the unit area resistance value (Ω·cm) 2 Secured ).

[0053] To confirm the permeability of the cell barrier for cultures under each condition, FITC-Dextran (70 kDa) was mixed into cell culture medium (FluoroBrite™ DMEM) to a concentration of 1 mg / mL. The cell culture medium present in the apical or basal compartment of the Transwell insert being cultured was removed, and the Transwell insert was transferred to a new 24-well plate. 800 μL of FluoroBrite™ was added to the basal compartment of the 24-well plate, and 250 μL of the 1 mg / mL FITC-Dextran solution was added to the apical compartment. The mixture was then incubated in a dark 37°C incubator for 30-40 minutes. After incubation, the medium from the basal compartment was collected and transferred to a 96-well test plate, and the fluorescence intensity was measured using a Victor X plate reader. The excitation / emission wavelength for FITC-Dextran measurement was 485 / 535 nm, and conditions performed using a cell-free insert were used as a positive control, and a cell culture medium not containing FITC-Dextran was used as a negative control.

[0054] Microscopic observations were performed every two days to confirm cell growth and morphology under each condition. After 11 days of culture, it was confirmed that cultures from all four conditions grew normally and formed a dense cell layer (Fig. 1a).

[0055] As a result of measuring the integrity of the cell barrier, 200 Ω·cm under CNBio 10K A549-BEAS2B conditions 2 It was confirmed to have the highest electrical resistance value at this level, and under the CNBio 10K BEAS2B-A549 condition, 100 Ω·cm 2 It followed at the level, and the two static culture conditions were 50 Ω·cm 2 It showed a value of approximately (Fig. 1b).

[0056] Overall, it can be observed that static culture conditions exhibit lower resistance values ​​compared to dynamic culture conditions, indicating that the flow rate applied under dynamic culture conditions enhances the integrity of the cell barrier.

[0057] When evaluating the permeability of the cell barrier, it was found that even under the same dynamic culture conditions, there were differences in permeability depending on the type of cell cultured at the base or apex of the insert, and that cell permeability was lower under the CNBio 10K A549-BEAS2B condition than under the CNBio 10K BEAS2B-A549 condition (Fig. 1c). This is consistent with the results of observing the integrity of the cell barrier.

[0058] However, regarding cell barrier permeability, since no statistical significance was found compared to the negative control group under all four conditions, it can be seen that permeability is low and the cell barrier functions normally under all four conditions.

[0060] <1-2> Confirmation of tight joint formation

[0061] To determine whether tight junctions observed in human lungs are reproduced in four model conditions, immunofluorescence staining was performed using ZO-1, an intercellular protein present in the tight junction complex, as a marker.

[0062] Specifically, cell cultures under dynamic co-culture conditions (CNBio 10K A549-BEAS2B and CNBio 10K BEAS2B-A549) were washed with PBS and fixed with 4% paraformaldehyde for 15 minutes. Then, they were stained using anti-ZO-1 antibody (ThermoFisher, Cat#: 339100, Clone: ​​1A12) as the primary antibody and goat-anti-mouse IgG (Assay Matrix Pty Ltd, Cat#: 20130-1) as the secondary antibody, and the cell nuclei were stained with DAPI. Afterward, they were observed using a confocal fluorescence microscope.

[0063] As a result, it was observed that the structure of the tight junction was uneven and fragmented in both models, confirming that the tight junction could not be formed normally in this model (Fig. 1d).

[0064] Based on these results, it can be seen that a lung-mimicking model using immortalized cell lines can be produced with excellent cell barrier integrity when two different cell lines, such as A549 and BEAS2B, are co-cultured under dynamic culture conditions with a flow of medium.

[0066] <Example 2> Optimization of Model Manufacturing Using Primary Cells

[0067] Since normal tight junction structures could not be observed in models using immortalized cell lines, we intended to mimic the human lung environment and construct a respiratory mucosal structure using primary cells, namely normal human bronchial epithelial cells (NHBE) and normal human lung fibroblasts (NHLF), instead of immortalized cell lines. Fibroblasts are known to assist the function of epithelial cells by promoting proliferation and differentiation, regulating mucin secretion, and inducing proper spatial distribution. Therefore, it was hypothesized that co-culturing fibroblasts and epithelial cells would allow the bronchial epithelial cells to assemble well due to the fibroblasts and maintain a mucociliary phenotype for an extended period.

[0068] As in Example 1, four condition models were constructed using 10,000 NHBE and NHLF cells each, depending on static or dynamic culture, and single culture or co-culture, etc.: 1) NHBE static single culture (static NHBE); 2) NHBE and NHLF static co-culture (static NHBE-NHLF); 3) NHBE dynamic single culture (CNBio NHBE); and 4) NHBE and NHLF dynamic co-culture (CNBio NHBE-NHLF).

[0069] NHLF was cultured and maintained in fibroblast growth medium-2 (FGM-2), and NHBE and DHBE were cultured and maintained in bronchial epithelial growth medium (BEGM). On day 7, the medium was changed to PneumaCult-ALI for differentiation of the basal lateral compartment, and no separate medium was used for differentiation of the apical compartment. Culture was performed for a total of 25 days.

[0071] <2-1> Confirmation of cell morphology and tight junction formation

[0072] After 25 days of culture were completed, cell morphology was observed. Normally proliferated cell layers were observed under all conditions, but it was confirmed that the cell morphology was non-uniform under dynamic conditions (single culture and co-culture) compared to static conditions (Fig. 2a). The circles in Fig. 2a indicate the non-uniform areas. This may be because the flow of the medium affected cell growth.

[0073] As described in Example 1, immunofluorescence staining for ZO-1 was performed to confirm the formation of tight junctions in the cell culture, and as a result, a continuously well-organized tight junction structure was observed in the periphery of the cells under all conditions (Fig. 2a). This indicates that tight junction structures between cells can be formed by using primary cells.

[0074] In addition, to observe the apical morphology of the cell culture using a surface electrical microscope (SEM), the cell culture was washed twice with PBS and fixed in a 2.5% glutaraldehyde / PBS solution for 2 hours. Afterward, it was washed three times with PBS, the membrane was cut from the insert, and it was progressively dehydrated using ethanol solutions ranging from 30% to 100%, followed by drying in a critical point dryer. Then, it was coated with gold to a thickness of 10 nm, stored in a dryer filled with silica gel, and SEM images were taken.

[0075] Through SEM images, it was found that cilia structures were not observed under static culture conditions, but hair-like cilia were formed under dynamic culture conditions regardless of single culture or co-culture (Fig. 2a).

[0076] This indicates that microcirculation following culture medium flow and the two cell types present in lung tissue are important for forming lung-mimicking structures.

[0078] <2-2> Verification of Cell Barrier Function

[0079] The integrity and permeability of the cell barrier were measured in the same way as described in Example 1.

[0080] As a result, electrical resistance was observed to be higher in all four models using primary cells than in the CNBio A549-BEAS2B model (Fig. 2b). Cell barrier permeability using FITC-Dextran was also observed to be lower in all four models (Fig. 2c). Through this, it can be seen that all four models using primary cells are capable of forming a normal cell barrier.

[0082] <2-3> Histological Analysis

[0083] Cross-sections of models prepared using primary cells were observed by performing hematoxylin & eosin (H&E) staining or immunofluorescence staining. Immunofluorescence staining was performed using antibodies against acetylated alpha-tubulin (α-tubulin), a marker for cilia, or MUC4b, a marker for goblet cells.

[0084] As a result, it can be confirmed that the cross-sections of all models form a multilayer structure, which means that all models contain fully differentiated airway epithelial cells (Fig. 2d).

[0085] In addition, staining results for acetylated alpha-tubulin, a marker for cilia, and MUC4b, a marker for goblet cells, showed that more cilia and goblet cells were observed in the dynamically cultured model compared to the statically cultured model (Fig. 2d). This implies that differentiation and development of the model occur better under dynamic culture conditions. These results are consistent with the SEM results showing more cilia structures in the dynamically cultured model.

[0086] Therefore, it was confirmed that by co-culturing different primary cells but performing dynamic culture, a model that better mimics healthy human lungs can be manufactured.

[0088] <Example 3> Comparison between a primary cell dynamic culture model and a commercially available lung model

[0089] To verify the structure and function of the CNBio NHBE-NHLF (NHBE-NHLF dynamic co-culture conditions) model prepared through Example 2, a commercially available airway model (SmallAir-HF healthy) composed of human airway epithelial cells and fibroblasts was compared with the model of the present invention.

[0090] Immunofluorescence staining for ZO-1, a marker for tight junctions, was performed using the model prepared in Example 2 and a commercially available airway model.

[0091] As a result, both models exhibited robust tight junctions, but it was confirmed that tight junctions (ZO-1) were expressed more continuously in the model prepared through Example 2 compared to the commercially available airway model (Fig. 3a).

[0092] In addition, scanning electron microscope images confirmed that both models possessed ciliated cells (Fig. 3a).

[0093] In addition, as described in Example 2, each model was histologically analyzed by performing H&E staining and immunofluorescence staining on the cross-section of the model.

[0094] As a result, it was confirmed that both models formed a multilayer structure and had a ciliary structure (Fig. 3a). However, compared to the commercially available model, the ciliary structure of the model of the present invention prepared through Example 2 appeared more distinctly and was well positioned at the apex (Fig. 3a).

[0095] The electrical resistance of the two models was measured using the method described in Example 1. As a result, the resistance values ​​of both models ranged from 700 to 1400 Ω·cm. 2 It was observed to be significantly high at the level (Fig. 3b).

[0096] The TEER value can increase as the model's cell layers become multi-layered and also depending on the degree of cell differentiation. The TEER value of healthy lungs is 700 to 1200 Ω·cm.2 It can be seen that both models show TEER values ​​at the level of healthy lungs.

[0097] Through this, it was confirmed that while both models mimic the structural characteristics of the lungs, the model of the present invention better mimics the histological structure of actual in vivo airway epithelial cells.

[0099] <Example 4> Optimization of a Closure Mimicking Model According to the Type of Coating Material and Transwell Membrane

[0100] To further optimize the CNBio NHBE-NHLF model prepared by the method described in Example 2, coating materials such as collagen and fibronectin and transwell membranes having different pore densities were used.

[0101] To evaluate the coating material, 3 mg / mL of collagen type I solution or 2.5 μg / mL of fibronectin was prepared and the surface of the Transwell membrane was activated for 5 minutes using a plasma cleaner (PDC-002-HP), after which the collagen or fibronectin solution was applied and left to stand for 30 minutes. After 30 minutes, the solution was removed and the membrane was primed with cell culture medium for 1 hour, and the Transwell membrane was washed twice with RNase-free PBS before adding cells.

[0102] Since the pore density of Transwell insert membranes is known to affect the proliferation and differentiation of models growing at the air-liquid interface, inserts with different pore densities, 4×10 6 Pores / cm 2 Standard Corning (Cat: COR3470, Corning, Life Sciences) and 100x10 6 Pores / cm 2CellQart (Cat: 9,320,402, Sabeu) was prepared. Both inserts were cultured and maintained under the same conditions (7 days in liquid, 3 weeks in ALI), and the medium was replaced every two days during the process, and the basic model preparation method was the same as described in Example 2.

[0103] An impedance measurement system was used to observe the integrity of the cell barrier. Unlike TEER, which is measured once at a specific point in time, the impedance measurement system has the advantage of enabling real-time measurement. Therefore, TEER values ​​and capacitance can be calculated through the impedance curve, allowing not only the function of the barrier to be verified through TEER values ​​but also the growth and differentiation processes of cells to be verified through capacitance.

[0105] <4-1> Verification of Impedance Differences Based on Coating Material and Insert Type

[0106] Impedance measurements were performed every two days using Locsense, a high-throughput automated monitoring system, with the apical and basal lateral compartments filled with fresh medium, and the impedance of the cell layer was measured at a wide frequency range from 10 to 100,000 Hz. The measured data were collected in Microsoft Excel and analyzed using GraphPad Prism. Impedance measurements for each model were performed for 24 days.

[0107] The TEER value for each model was calculated based on the height difference between the curve of each model and the control group, which is indicated by an arrow in the left panel of Figures 4a to 4d. As the TEER value of each model increases, the plateau of the curve moves upward.

[0108] The capacitance of each model can be determined by the width between the curves of each model. This is indicated by arrows in the right panel of Figures 4a to 4d. As the capacitance of each model increases, the high point of the curve narrows.

[0109] During the first 15 days of the culture period, TEER was the dominant signal, and a gradual increase in this signal was observed. This increase in TEER values ​​indicates that tight junctions and multilayer structures are being formed in the model (left panel of Figures 4a to 4d).

[0110] We observed that the capacitance signal increased and became dominant starting from day 13 of culture, which suggests that cell differentiation is taking place within the model (right panel of Figures 4a to 4d).

[0112] Regardless of the type of membrane coating material, the TEER value was higher when the model was manufactured using the Corning insert than when using the CellQart insert (left panel of Figures 4a to 4d).

[0113] As a result on day 3 of culture, the TEER value for cultures on collagen-coated membranes and Corning inserts was 140 Ω·cm 2 And the TEER value when cultured on fibronectin-coated membranes and Corning inserts was 20 Ω·cm 2 This confirms that collagen aids in cell adhesion and the rapid formation of a cell monolayer; however, the results on day 8 of culture showed that the TEER value for cultures on collagen-coated membranes and Corning inserts was 442 Ω·cm 2 And the TEER value when cultured on fibronectin-coated membranes and Corning inserts was 187 Ω·cm 2For 5 days, the growth rate increased threefold when coated with collagen and ninefold when coated with fibronectin, indicating that the cell growth rate can be increased when coated with fibronectin (left panel of Figures 4a and 4c).

[0114] However, when the model was manufactured on a CellQart insert, there was no difference in TEER values ​​depending on the coating material (left panel of Figs. 4b and 4d).

[0115] On day 10 of culture, TEER values ​​decreased by approximately 25% in all models; this may be due to the change in culture conditions from liquid to the air-liquid interface (ALI) and is also associated with cell differentiation. In models manufactured on Corning inserts, TEER values ​​showed a steady decreasing trend from day 10 to day 15 of culture, and on day 15, they appeared at similar levels regardless of the coating material (198 Ω·cm² for collagen coating). 2 , 172 Ω·cm when coated with fibronectin 2 )(Left panel of FIG. 4a and FIG. 4c).

[0116] These results indicate that the type of coating material does not affect cell growth in the later stages of culture.

[0117] On the other hand, in the model prepared using the CellQart insert, the TEER value increased steadily even on days 10 to 15 of culture, and from these results, it can be inferred that the cells are steadily forming physiological structures such as cell layers and differentiating into a uniform structure including tight junctions (left panel of Figures 4b and 4d).

[0118] Therefore, significant fluctuations in TEER values ​​in the model using the Corning insert indicate non-specific cell differentiation and the formation of incomplete tight junctions as a result, whereas the constant and steady increase in TEER values ​​in the model using the CellQart insert indicates the formation of uniform and complete tight junction structures.

[0120] As a result from days 13 to 24 of culture, regardless of the insert type, when coated with collagen, the cell electrical capacity tended to increase (right panels of Figs. 4a and 4b). On the other hand, when coated with fibronectin, the cell electrical capacity increased from days 13 to 20 of culture, decreased from days 20 to 22, and increased again from days 22 to 24 of culture (right panels of Figs. 4c and 4d).

[0121] These results indicate that collagen can promote the growth and differentiation of the model more effectively than fibronectin, as confirmed by the TEER results.

[0123] <4-2> Verification of Membrane Topographic and Mechanical Properties

[0124] Atomic force microscopy (AFM) and nanoindentation were used to confirm the topography and mechanical properties of the membrane.

[0125] Observation using an atomic force microscope confirmed that the CellQart insert had a uniform structure with pore sizes uniformly distributed at the level of ~0.5 μm (Fig. 5e). On the other hand, the Corning insert had relatively fewer pores and a non-uniform structure (Fig. 5e).

[0126] CellQart is considered to be advantageous for the exchange of nutrients and paracrine signals due to the high presence of pores, whereas Corning inserts are thought to result in less exchange of cell culture media or nutrients.

[0128] As a result of nanoindentation, the average Young's modulus of the wire in the insert was 511 ± 482 kPa for the CellQart insert and 266 ± 315 kPa for the Corning insert. Since this difference was not statistically significant, it was determined that there was no difference in stiffness between the CellQart insert and the Corning insert.

[0130] <Example 5> Preparation of a Chronic Obstructive Pulmonary Disease Model

[0131] Since it was confirmed that a lung-mimicking model can be manufactured using primary cells through the method described in Example 2, we wanted to verify whether a disease model mimicking a lung disease state could be manufactured by applying this method.

[0132] To this end, diseased human bronchial / tracheal epithelial cells (DHBE) were used instead of normal bronchial epithelial cells (NHBE). As described in Example 2, models were prepared under three conditions: static or dynamic culture, single culture or co-culture: 1) static single culture (Static DHBE); 2) static co-culture (Static DHBE-NHLF); and 3) dynamic co-culture (CNBio DHBE-NHLF). For DHBE, BEGM was used as the culture medium, as for NHBE, and 10,000 cells were used to prepare the COPD model, and Corning inserts were used for the Transwell inserts.

[0133] Subsequently, the external appearance of each model was observed using a microscope and an electron microscope, and the level of tight junction formation was confirmed by performing immunofluorescence staining on ZO-1.

[0134] As a result, after 25 days of culture, it was confirmed that a normal cell monolayer was formed in all disease models, but a relatively uneven cell structure was observed in the CNBio DHBE-NHLF model (uneven areas are indicated by circles) (Fig. 5a).

[0135] According to the expression results of ZO-1, the single-culture model differed from the co-culture model in the level of tight junction formation. The Static DHBE model formed robust and continuous tight junctions, whereas the Static DHBE-NHLF and CNBio DHBE-NHLF models were observed to have altered protein positions and breakage (Fig. 5a).

[0136] Observations via SEM images revealed that the Static DHBE and Static DHBE-NHLF models had underdeveloped cilia, whereas the apical cross-section of the CNBio DHBE-NHLF model showed well-developed cilia (Fig. 5a). However, it was confirmed that the cilia in the disease models were significantly reduced compared to the healthy lung model (Fig. 3a).

[0137] In addition, TEER was measured to verify the cell barrier integrity of each model, and the electrical resistance of the Static DHBE-NHLF model was 941 Ω·cm 2 It was the highest, and the electrical resistance values ​​of the CNBio DHBE-NHLF model and Static DHBE model were approximately 250 Ω·cm 2 was (Fig. 5b).

[0138] As a result of checking the permeability of the cell barrier, it was found that FITC-Dextran permeability was low in all models, indicating that all models form a normal cell barrier (Fig. 5c).

[0140] For each model, H&E staining and immunofluorescence staining for acetylated alpha-tubulin and MUC5B were performed for histological analysis.

[0141] As a result, it can be seen that although the thickness of each layer is thin compared to the health model, a multilayer structure has been formed, indicating that the differentiation and growth of the disease model have been successfully achieved (Fig. 5d).

[0142] In addition, when examining the expression pattern of MUC5B, high expression of goblet cells was observed under dynamic culture conditions compared to static culture conditions (Fig. 5d). This implies that goblet cells tend to be highly expressed in actual COPD patients, and that dynamic culture conditions can accurately reproduce the characteristics of this disease.

[0144] <Example 6> Comparison with a commercially available disease model

[0145] To compare the COPD model (CNBio DHBE-NHLF) prepared in Example 5 with the commercially available COPD model MucilAir®-HF (MucilAir®-HF COPD) derived from a COPD patient, the morphology was observed using an electron microscope, histological analysis was performed using H&E staining, and the expression of tight junctions, cilia, and goblet cells was observed using immunofluorescence staining.

[0146] SEM images and acetylated alpha-tubulin staining results showed no significant difference between the two models in that the number of cilia was reduced compared to the healthy model (Fig. 6a).

[0147] However, when observing tight junctions, the CNBio DHBE-NHLF model appears in a broken form, whereas the commercial model appears in a solid and continuous form (Fig. 6a). It is known that the tight junctions of patients with COPD are also observed to have an incomplete structure. Therefore, it can be seen that the CNBio DHBE-NHLF model of the present invention mimics the tissue in a diseased state better than the commercial model.

[0148] Regarding the expression of goblet cells, an increase in the number and size of goblet cells was observed in the CNBio DHBE-NHLF model, but these changes could not be confirmed in the commercial model (Fig. 6a).

[0150] To verify the integrity of the cell barrier, the TEER value of each model was measured, and the commercially available model was approximately 200 Ω·cm 2 At this level, the TEER value of CNBio DHBE-NHLF is 160 Ω·cm 2 It was higher (Fig. 6b).

[0152] <Example 7> Confirmation of inflammatory response using a lung-mimicking model

[0153] In actual lungs, various immune responses, including inflammatory reactions, can occur, and cytokines are secreted during this process. In particular, cytokine levels can be assessed as markers of lung inflammatory responses in disease conditions such as COPD. To verify whether these responses are reproduced in a lung-mimicking model, cytokine levels were measured and compared between a healthy model (CNBio NHBE-NHLF) and a disease model (CNBio DHBE-NHLF). Cytokine levels were measured using a Luminex bead-based assay.

[0154] As a result, the expression levels of 27 cytokines were generally observed to be higher in the disease model than in the health model (Fig. 7). In particular, it was confirmed that the levels of IP-10 and IL-6, cytokines that induce inflammatory responses, increased significantly in the disease model (Fig. 7).

[0155] Through this, it can be seen that the lung-mimicking model of the present invention can mimic functions such as immune responses like cytokine production and secretion, and can even mimic disease states such as the induction of more inflammatory responses in disease models.

[0156] In the present invention, a lung model mimicking the lung can be manufactured by co-culturing immortalized lung cell lines or lung-derived primary cells in a Transwell. The lung model thus manufactured exhibits characteristics of lung tissue well, such as forming a multilayered structure, including ciliated structures and goblet cells, and confirming normal cell barrier function, and it is observed to develop more similarly to lung tissue compared to commercially available models. Furthermore, it can mimic immune responses such as cytokine secretion, and disease models can be manufactured depending on the type of primary cells used; thus, the pathological mechanisms of diseases can be studied through these models, and they can be utilized in the development of therapeutic agents.

Claims

Claim 1 A method for preparing an in vitro lung model of chronic obstructive pulmonary disease, characterized by comprising: 1) a step of coating an insert membrane of a transwell with an extracellular matrix protein; 2) a step of seeding two or more different types of primary cells, including diseased human bronchial / tracheal epithelial cells (DHBE) and normal human lung fibroblasts (NLF), onto the insert membrane of the transwell; and 3) a step of culturing the cells in a dynamic co-culture that maintains a flow of culture medium. Claim 2 A method for preparing an in vitro lung model of chronic obstructive pulmonary disease, characterized in that, in claim 1, the extracellular matrix protein is one or more selected from the group consisting of laminin, fibronectin, and collagen. Claim 3 delete Claim 4 delete Claim 5 delete Claim 6 delete Claim 7 A method for preparing an in vitro lung model of chronic obstructive pulmonary disease, characterized in that, in claim 1, the flow rate of the medium is 0.1 μL / s to 1.0 μL / s. Claim 8 A method for preparing an in vitro lung model of chronic obstructive pulmonary disease, characterized in that, in claim 7, the flow rate of the medium is 0.3 μL / s to 0.7 μL / s. Claim 9 A chronic obstructive pulmonary disease in vitro lung model system characterized by being manufactured by the method of claim 1. Claim 10 A screening method for a chronic obstructive pulmonary disease treatment agent characterized by comprising: 1) a step of preparing an in vitro lung model of chronic obstructive pulmonary disease by the method of claim 1; 2) a step of treating the in vitro lung model of chronic obstructive pulmonary disease prepared through step 1) with a test substance; and 3) a step of comparing a treatment group model treated with the test substance of step 2) with an untreated control group model to determine whether symptoms recover following treatment with the test substance based on whether there is a decrease in the expression of one or more inflammatory cytokines selected from a group consisting of IP-10 and IL-6. Claim 11 A screening method for a treatment for chronic obstructive pulmonary disease according to claim 10, wherein the above-mentioned test substance is selected from the group consisting of natural compounds, synthetic compounds, RNA, DNA, polypeptides, enzymes, proteins, ligands, antibodies, antigens, metabolites of bacteria or fungi, and live molecules. Claim 12 A screening method for a chronic obstructive pulmonary disease treatment according to claim 10, characterized in that the determination of whether symptoms recover following treatment with the above-mentioned test substance is further determined by selecting one or more from the group consisting of a decrease in goblet cell expression, an increase in tight junction formation or more homogeneous tight junction formation, and an increase in cell layer thickness. Claim 13 delete Claim 14 delete

Images