Here, we demonstrate how epithelial cells cultured with fibroblasts in the nanofibrous membrane-based two-layer system can stably adhere to and grow on the membrane.
Method Article
Here, we demonstrate how epithelial cells cultured with fibroblasts in the nanofibrous membrane-based two-layer system can stably adhere to and grow on the membrane.
Technical hurdles in a culture of epithelial cells include dedifferentiation and loss of function. Biomimetic three-dimensional (3D) cell culture methods can enhance cell culture efficiency. This study introduces an advanced two-layered culture system intended to cultivate epithelial cells as tissue-like layers with the culture of fibroblasts within a 3D environment. Polyvinyl alcohol (PVA) and poly(ε-caprolactone) (PCL) nanofibrous membranes (NMs) were fabricated via electrospinning and utilized as a physiologically relevant extracellular matrix for the culture of epithelial cells and fibroblasts, respectively. In the upper insert wells, lung epithelial cells were cultivated on the PVA NM, and in the lower chambers, fibroblasts were cultured on the PCL NM. This configuration eliminates direct cell-cell contact and facilitates the examination of paracrine signaling mediated by soluble factors. Confocal microscopy was employed to analyze the distribution, growth pattern, and expression of intracellular proteins, including zona occludens in epithelial cells. Z-stacking techniques enabled detailed 3D reconstructions, providing precise insights into the integrity of tight junctions and spatial organization within the epithelial layer. Scanning electron microscopy (SEM) assessed the morphological characteristics of cell types on the nanofibrous membranes. SEM imaging revealed intricate cell surface structures and interactions with the nanofibers, offering a comprehensive perspective on cellular architecture and cell interaction with nanofibrous structure. The Cell Counting Kit-8 (CCK-8) assay is a simple method for measuring epithelial cell and fibroblast growth rates over time. It provides the proliferative behaviors and potential synergistic effects of coculturing these cells. These findings highlight the effectiveness of a simple insert co-culture system for simultaneous culture of fibroblasts and epithelial cells, which is crucial in various physiological and pharmacological contexts, including epithelial tissue regeneration, tumor microenvironment with endothelial, immune, and other stroma cells, toxicity assay, and drug activity test.
Over the decades, the classic two-dimensional (2D) monolayer culture has been essential for understanding infections, pharmacology, and toxicology1,2. However, it is important to consider the complexity, dynamic interactions between multi-compositions, and three-dimensional (3D) architecture of the tissue microenvironment. Thus, the 2D cell culture has limitations in mimicking tissue-like structures3,4. For example, there is a lack of cell-to-cell and cell-to-extracellular matrix (ECM) signaling, which is essential in cell differentiation, proliferation, and cellular functions in the vivo microenvironment. For these reasons, the systems of 3D cell cultures have been developed5. The cells in 3D culture grow and interact with the surrounding extracellular matrix in three dimensions. Furthermore, a 3D coculture approach, composed of more than one cell type, has been developed to bridge the gap between simplistic single-cell type in vitro models and the dynamic cellular interactions that occur in vivo6. 3D coculture systems have gained popularity in the study of cell-cell communication and cell-extracellular matrix (ECM) interactions7,8,9,10. Current 3D coculture systems have several forms and can be separated into direct and indirect coculture8,9. In direct coculture, various cell functions in cultured cells can be mainly affected by direct contact rather than paracrine effects. In comparison, indirect coculture has a greater advantage in investigating paracrine interaction by using barriers or layers to separate different cell types.
Various materials and methods are utilized to fabricate biomimetic scaffolds mimicking the ECM using biocompatible and biodegradable substances10. The nanofibrous scaffold structure mimics the configuration of native ECM in biological tissues, providing a 3D architecture for cell culture11,12. Electrospun nanofibrous membranes (NMs) have been used in direct and indirect coculture of different types of cells13,14,15.
Epithelial cells form a physical barrier of the major organs, whereas fibroblasts are predominant stroma cells that mediate ECM remodeling and regulate the neighboring epithelium6,16,17. Epithelial cells and fibroblasts interact with laminin in the basement membrane and fibronectin in the interstitial matrix, respectively, through integrin receptors on the cell surface18. When epithelial cells are cultured on the poly(vinyl alcohol) (PVA) NMs with diameters of 150 - 250 nm and micropores, they form multi-layers instead of cell aggregates and spheroids, and their growth patterns are similar to those of the cells in epithelial tissue19. In comparison, the poly(caprolactone) (PCL) NMs with 400-1,500 nm diameters and 10-50 µm micropores provide a similar spatial dimensionality to the interstitial matrix for the growth of cultured fibroblasts13,14. Due to their biocompatibility, PVA and PCL membranes are commonly used polymers in tissue engineering20,21.
This study introduces an advanced two-layer culture system using porous electrospun NMs to cultivate epithelial cells as tissue-like layers and fibroblasts within a 3D structure. The protocol can be divided into five key stages: nanofiber fabrication, fiber post-crosslink and quality assessment, preparation of membrane-attached culture well, cell seeding and 3D coculture, and assays of 3D cultured cells.
1. Fabrication of water-stable PVA nanofibers
2. Treatment of a PVA nanofibrous membrane with hydrochloric acid (HCl) vapor and assessment of fiber quality
NOTE: The water stability of PVA NMs is well preserved after the post-crosslinking reaction by HCl vapor. This process is optimized for suitable thickness, pore size, and roughness of the NMs. PVA is a very hydrophilic polymer, and electrospun PVA nanofibers are unstable in water contact and humid conditions. PVA NMs should be treated with HCl vapor just after the fabrication of the PVA polymer.
3. Fabrication of PCL nanofibers and nanofiber quality assessment
4. Assembly of two layers of membranes in a trans-well
NOTE: In this protocol, we use a 24-well plate and its respective insert well. The PVA NM is attached to the insert well, while the PCL NM is attached to the bottom well.
5. Cell seeding and assembly of two layers for the coculture model
NOTE: MLE-12 cells were cultured in 1x Dulbecco's Modified Eagle Medium (DMEM)/F12 with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin antibiotic. NIH3T3 cells were cultured in 1x DMEM high glucose with 10% FBS and 1% penicillin/streptomycin antibiotic.
6. Cell distribution assay
7. Confocal observations of 3D cocultured cells
NOTE: Epithelial cell attachment to a PVA NM is weak compared to that of fibroblasts to a PCL NM. Discard old media manually by pipette aspiration and add new media or solution gently to the wall of wells. During the staining process, try not to let the membranes dry.
NOTE: Cell attachment can be improved by coating laminin or incorporating peptides of integrin-binding motifs in PVA nanofibers19.
8. Assay of cell attachment to nanofibers
9. Cell growth assay
NOTE: In this protocol, the CCK-8 assay is used to quantify cell growth rate following the manufacturer's instructions.
This protocol outlines the critical steps for coculturing MLE-12 lung epithelial cells and NIH3T3 fibroblasts by constructing a nanofibrous membrane-based two-layer model (Figure 1). This model suits live cell imaging, immunohistochemistry, and endpoint quantitative analysis of cocultured cells. Other cell types may also be utilized by optimizing cell density and growth conditions.
SEM shows the uniform distribution of fabricated PVA nanofibers without any bead formation (Figure 2A). The diameter of the PVA nanofibers ranged from 150 nm to 230 nm (180 ± 25 nm, mean ± standard deviation (SD)). The nanofibers in electrospun PCL NMs were randomly oriented and structurally resembled collagen (Figure 2B). Most fibers in PCL NMs had a diameter between 300 nm and 7 µm (3.8 ± 2.5 µm). Thus, PVA nanofibrous membrane had smaller pore sizes than PCL nanofibrous membrane.
Before seeding, MLE-12 and NIH3T3 cells were stained with Cell-Tracker red and green, respectively, and were monitored throughout the coculture process. As shown in Figure 3, the spatial distribution of MLE-12 cells in the insert chamber and NIH3T3 cells in the lower chamber could be distinctly observed. Focus-stacking analysis revealed that NIH3T3 cells were dispersed and detected at varying depths (102 ± 20 µm), whereas MLE-12 cells formed layers with a thickness of 80 ± 10 µm. SEM shows cell attachment to the membranes. NIH3T3 cells infiltrate the PCL nanofibrous membrane and extend along the nanofiber axes, while MLE-12 cells adhered to the surface of the PVA nanofibrous membrane and exhibited an aggregation (Figure 4). Next, we examined the cytoskeletal organization of NIH3T3 cells and MLE-12 cells by actin staining. NIH3T3 cells cultured on PCL nanofibrous membrane show a projectile distribution of actin filaments resembling cells in tissue, and actin staining in MLE-12 cells cultured on PVA nanofibrous membrane shows round aggregates with no cellular spreading (Figure 5).
The distribution and growth pattern of MLE-12 cells in monoculture and coculture were compared. Cell-Tracker red-stained cells in monoculture on the PVA NMs formed aggregates, but the cells in coculture condition maintained an even distribution on the membrane (Figure 6A). In a 3D analysis of confocal microscopy images, the growth pattern of cocultured MLE-12cells on the PVA NMs shows layered adhesion of cultured cells to the membrane with more confluent cell density, compared to cells cultured in the absence of fibroblasts (Figure 6B). In addition, cell-to-cell adhesion and expression of zonal occludin (ZO)-1 on the cell surfaces (arrow indicated), but not of cell aggregates, were more evident in the coculture condition than in culture without fibroblasts (Figure 6C).
The proliferation of cells on the membranes was measured using the CCK-8 assay. NIH3T3 and MLE-12 cell proliferation increased with time (Figure 7). However, lower growth rates of cells were observed on the membranes than on a culture plate (data not shown). Cocultured NIH3T3 and MLE-12 cells on NMs show a stable growth rate compared to the monocultured cells in a culture plate.

Figure 1: Schematic to illustrate the NIH3T3 fibroblast and MLE-12 epithelial cell co-culture method in the scaffold-based two-layer system. Abbreviations: PVA = polyvinyl alcohol; PCL = poly(ε-caprolactone). Please click here to view a larger version of this figure.

Figure 2: Structure of PVA and PCL nanofibers. (A) PVA and (B) PCL nanofibrous membranes were produced by electrospinning. The structure and diameter of the nanofibers are measured by SEM of the membrane surfaces. Scale bars = 1 µm (A), 10 µm (B). Abbreviations: PVA = polyvinyl alcohol; PCL = poly(ε-caprolactone). Please click here to view a larger version of this figure.

Figure 3: In a coculture system, the spatial distribution of NIH3T3 cells on PCL nanofibrous membrane and MLE-12 cells on PVA nanofibrous membrane. NIH3T3 cells (4 × 105) were stained with Cell-Tracker Green, and MLE-12 cells (3 × 105) were stained with Cell-Tracker Red. The seeded cells were cultured for 24 h and observed using a Z-stacking image of a confocal microscope. The numbers are indicated in µm; arrows show dimensions with x (width), y (length), z (depth). Abbreviations: PVA = polyvinyl alcohol; PCL = poly(ε-caprolactone). Please click here to view a larger version of this figure.

Figure 4: Adhesion and morphology of NIH3T3 cells on PCL nanofibrous membrane and MLE-12 cells on PVA nanofibrous membrane. The cells were cultured on the membranes for 24 h and observed using SEM at 2,000x magnification. Scale bars = 15 µm. Abbreviations: PVA = polyvinyl alcohol; PCL = poly(ε-caprolactone). Please click here to view a larger version of this figure.

Figure 5: Actin stress fibers in NIH3T3 cells on PCL nanofibrous membrane and MLE-12 cells on PVA nanofibrous membrane. NIH3T3 cells (4 × 105) and MLE-12 cells (3 × 105 cells) were seeded on the PCL and PVA nanofibrous membranes, respectively. After 48 h culture, cells were stained with phalloidin-iFlour 488 and observed using confocal microscopy. Scale bars = 50 µm. Abbreviations: PVA = polyvinyl alcohol; PCL = poly(ε-caprolactone); FITC = fluorescein isothiocyanate. Please click here to view a larger version of this figure.

Figure 6: Growth pattern of cocultured MLE-12 cells on PVA nanofibrous membrane. (A) Cell-Tracker red-stained MLE-12 cells (3 × 105 cells) were seeded on the PVA nanofibrous membrane and cultured alone (-NIH3T3 cells) and with NIH3T3 cells in a two-layer system (+NIH3T3 cells) for 48 h. (B) Monocultured and cocultured MLE-12 cells for 48 h were stained with Hoechst 33342 (blue) and Phalloidin-iFlour 488 and observed using a confocal microscope. The 3D images are shown using the surface function of Imaris software. (C) MLE-12 cells were cultured for 48 h and fluorescently labeled with Phalloidin-iFlour 488 and anti-ZO-1 antibody (red). Scale bars = 50 µm. Abbreviations: PVA = polyvinyl alcohol. Please click here to view a larger version of this figure.

Figure 7: Growth rate of the cells in a two-layer coculture system. NIH3T3 cells (4 × 105) and MLE-12 cells (3 × 105) are cocultured for the indicated time. Cell growth in the insert and bottom well is measured using the CCK-8 assay. Data are presented as mean ± SD (n = 3). Please click here to view a larger version of this figure.
The transition from 2D culture to 3D culture models represents a significant advancement in developing in vitro coculture models. A 3D culture model must mimic the tissue microenvironment in which cells can proliferate, aggregate, and differentiate22. The use of scaffold-based 3D coculture models enhances the replication of tissue architecture. In this study, an NM-based trans-well system provides indirect 3D coculture of two types of cells. The PCL NM with large pores in the lower chamber permits cellular infiltration and mimics the interstitial matrix. This scaffold supports the growth and spatial organization of fibroblasts. Using a PVA nanofibrous membrane in the upper chamber allows the formation of MLE-12 epithelial cells in the layer. The diameter and pore size of the PVA nanofibrous membrane make it an ideal substrate for supporting the adhesion and growth of epithelial cells, which mimics epithelial cells on the basement membrane in epithelial tissues19.
Cell-adhesive proteins such as fibronectin and laminin are key proteins in the ECM, which connect cells with collagen fibrils23. Epithelial cells interact with laminin through integrin receptors on their surface18. MLE-12 cells seeded on the PVA NMs adhered to and grew three-dimensionally on the membrane. However, epithelial cell binding to the porous PVA NM was low, resulting in aggregation because they lack specific biological motifs interacting with cultured cells. In our coculture system, direct cell-cell contact between NIH3T3 and MLE-12 cells is absent, and communication occurs exclusively through paracrine signaling. The presence of cocultured NIH3T3 cells may stabilize the attachment of MLE-12 cells to the membrane throughout the culture period. This stabilization can be attributed to the secretion of soluble factors by the fibroblasts, including collagen, fibronectin, laminin, and various growth factors24. In comparison, in a tumor model with direct coculture of fibroblasts and cancer cells in a PCL nanofibrous scaffold, fibroblasts alter the properties of the ECM through matrix remodeling25.
Epithelial cells rest on a basement membrane that acts as a growth support and selectively permeable layer. Several technological strategies, including organoid technology, are currently being developed to create 3D complex models of epithelial tissues26. However, organoid structure hampers the use of conventional assays27. In addition, conventional microscopy for experimental data collection is complicated by the fact that organoids are cultured while embedded in a 3D hydrogel matrix27. In this study, the insert chamber was separated from the low chamber to facilitate precise observation of epithelial cells cultured on PVA NM. Moreover, 3D organoids are produced by the differentiation of stem cells. Thus, culturing stem cells takes a long time and is a complex procedure compared to a coculturing system, and the heterogeneity of organoids is reported28.
Compared to monoculture, coculture of epithelial cells and fibroblasts improves adhesion to scaffold and maintains stable growth and function of the cultured epithelial cells. In addition, supplementary materials, including growth factors and cytokines, are not necessary in the culture medium due to soluble factors secreted by cocultured fibroblasts. Despite the advantages of using nanofibrous scaffolds in cell coculture by providing structural support, they cannot provide ECM surrounding culturing cells compared to biocompatible hydrogels29. Thus, this limitation can be achieved using cells on NM with over-layered hydrogels.
In conclusion, a simple insert co-culture system for simultaneous culture of fibroblasts and epithelial cells is crucial in various physiological and pharmacological contexts, including epithelial tissue regeneration, tumor microenvironment, toxicity assay, and drug activity test.A nanofibrous membrane-based 3D coculture system will be instrumented and designed for high throughput drug absorption and transport screening studies.
The authors declare no conflicts of interest.
This work was supported by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (HR16C0001); the Basic Science Research Capacity Enhancement Project through the Korea Basic Science Institute (National Research Facilities and Equipment Center) grant funded by the Ministry of Education (2019R1A6C1010003); and National Research Foundation of Korea (NRF) grants funded by the Korean Government (MSIT) (2022R1A4A5032702).
| Name | Company | Catalog Number | Comments |
|---|---|---|---|
| 0.05% Trypsin/EDTA | Welgene | LS015-01 | |
| 100x Penicilin/Streptomycin | Gibco | 10378016 | |
| 20x PBS | LPS solution | CBP007A | |
| 4% Paraformaldehyde | Biosesang | PC2031-050-00 | |
| Alexa Fluor-594 conjugated ZO-1 antibody | Invitrogen | 339194 | |
| Antibody diluent OP Quanto | Epredia | 10129-576 | |
| CCK-8 assay kit | Donginbio | CCK-3000 | |
| Cell culture dish (150x20) | SPL | 20151 | |
| CellTracker Green CMFDA | Invitrogen | C7025 | |
| CellTracker Red CMTPX | Invitrogen | C34552 | |
| Chloroform | Samchun | C0584 | |
| Conical tube | SPL | 50050 | |
| Cover glass | Corning | CLS2980245 | |
| DMEM high glucose | Welgene | LM001-05 | |
| DMEM/F12 | Welgene | LM 002-05 | |
| DURAN Desiccator bases with plane flange, screw thread | DWK Life Sciences | 7.022 260 | |
| DURAN Glass Desiccator Lid & Stopcock | Daihan Science | SM.2444061 | |
| DURAN Stopcock, with PTFE Spindle | DWK Life Sciences | 10322671 | |
| Electrospinning machine | NanoNC | ESR200RD | |
| Ethyl alcohol, Pure | Sigma Aldrich | 459844 | |
| FBS | Sigma Aldrich | TMS-013-BKR | |
| Fluorescence Laser Confocal Scanner Module K1-fluo | Nanoscope Systems | ||
| Gel/mount | Biomeda Corp. | M01 | |
| Glutaraldehyde solution | Sigma Aldrich | 340855 | |
| Hoechst 33342 | Invitrogen | H1399 | |
| Metal nozzle 27G | NanoNC | ||
| Nail polish | Nature republic | ||
| Normal goat serum | Vector Laboratories | S-1000 | |
| Osmium tetroxide | Sigma Aldrich | 201030 | |
| Phalloidin-iFlour 488 | abcam | ab176753 | |
| Plastic Syringe_Lure Lock (10 mL) | HENKE SASS WOLF | AL10 | |
| Poly(acrylic acid) | Sigma Aldrich | 323667 | |
| Poly(vinyl alcohol) | Sigma Aldrich | 341584 | |
| Polycaprolactone | Sigma Aldrich | 440744 | |
| Pure HCL | Duksan | 1129 | |
| Scanning Electron Microscopy | SEC | SNE-4500M | |
| Slide glass | Marienfeld Superior | K15663717 | |
| Sylgard 184 set | omniscience | OMNI.05255 | |
| Synergy H1 Multimode Reader | Biotek | ||
| Triton X-100 | Sigma Aldrich | X100 |
Request permission to reuse the text or figures of this JoVE article
Request Permission