Xenogeneic-free culture of human intestinal stem cells on functional polymer-coated substrates for scalable, clinical-grade stem cell therapy

Exploration of xenogeneic-free culture surfaces

ISCs^3D-hIO differentiated from human induced pluripotent stem cells (hiPSCs) on Matrigel-coated cell culture surfaces could facilitate massive culture and transplantation¹⁸. Nonetheless, their potential application as cellular remedies for humans is inherently limited because of concerns regarding acute immune responses in human recipients²⁷. The development of an ISC^3D-hIO culture platform without the use of xenogeneic ECM could overcome the intrinsic limitations of cellular treatment. Therefore, we aimed to discover an optimal surface for the xenogeneic-free and highly expandable culture of ISCs^3D-hIO using cell culture surface modifications. Leveraging the benefits of the iCVD process²⁸, which enables damage-free polymer film synthesis from various monomers, we conducted a comprehensive polymer-based screening (Fig. 1a). In more detail, the iCVD process allows the formation of compositionally uniform copolymers with a systematically controlled composition of each monomer^{28,29,30,31,32,33}. Notably, the vapor phase characteristics enabled the synthesis of copolymers without phase separation, rendering this process optimal for generating diverse types of uniform copolymer compositions. Additionally, the solvent-free and additive-free nature of the process inherently eliminates the issues related to cell toxicity arising from residual solvents or additives^34,35. The ambient temperature further provided the advantage of coating the polymer film without damaging the conventional cell culture plates³⁶. Moreover, conformal thin films can be deposited directly onto nonflat culture surfaces, such as multiwell plates and membrane inserts, while preserving their original forms³⁷. The benefits of the iCVD process ensure reproducible and scalable surface modifications of cell culture plates with various form factors. By utilizing such advantageous characteristics, we synthesized diverse types of polymers using the iCVD process, thereby adjusting the possibility of xenogeneic-free culture of ISCs^3D-hIO.

a Schematic illustration of polymer synthesis in the vapor phase using the iCVD process and ISCs^3D-hIO cultivation without xenogeneic ECM on polymers. b Chemical structures of 26 types of polymer candidates for the culture of ISCs^3D-hIO. (X_Y denotes the copolymer composed of X and Y, and +p indicates a surface that has undergone postplasma treatment.) c Morphologies of ISCs^3D-hIO cultured on a Matrigel-coated surface, various polymer-coated surfaces and tissue culture polystyrene (TCPS) (n = 2 sample/group). Optical microscope images of each culture surface were captured on Day 4. The scale bar indicates 500 µm. d Heatmap of polymers with surface energy, colony areas at Days 2 and 4, and growth rates.

Full size image

In this study, we deposited 32 distinct types of polymers with diverse functional characteristics, such as cationic/anionic functionalities, crosslinking capabilities, and hydrophobic and hydrophilic properties (Fig. 1b). The names, specific synthesis conditions, and Fourier transform infrared spectroscopy (FT-IR) analysis results of each polymer are summarized in Supplementary Fig. 1 and Supplementary Tables 1 and 2. Polymers such as poly(2-(dimethyl amino) ethyl methacrylate) (pDMAEMA), poly(N,N-dimethyl aminopropyl methacrylamide) (pDMAPMA), poly(2-(diethyl amino) ethyl methacrylate) (pDEAEMA), poly(2-(tert-butyl amino) ethyl methacrylate) (pTBAEMA) and poly(2-(dimethyl amino) ethyl acrylate) (pDMAEA) include tertiary amine moieties, conferring positive charges to the polymer surface, thereby facilitating electrostatic interactions with cell surfaces³⁸. The noncytotoxic epoxy functionality in poly(glycidyl methacrylate) (pGMA) readily captures amine groups via a ring-opening reaction and is thus capable of immobilizing bioactive molecules in culture medium to provide favorable surfaces for cell culture^39,40,41,42. The copolymerization of pGMA with amine functional polymers results in enhanced formation of quaternary ammonium groups via the amine‒epoxy reaction, further stabilizing the culture surface^31,43,44. Poly(ethylene glycol dimethacrylate) (pEGDMA) is a crosslinked polymer with biocompatible ethylene glycol functionality^45,46. When copolymers are formed, the EGDMA crosslinker moiety enhances the chemical and mechanical stability⁴⁷. Halide-containing poly(4-vinylbenzyl chloride) (pVBC), which undergoes ionic crosslinking through the Menshutkin reaction when copolymerized with amine-containing monomers that enhance electrostatic interactions, was also adapted^34,48,49,50. Poly(cyclohexyl methacrylate) (pCHMA), poly(benzyl methacrylate) (pBMA), and poly(2,4,6,8-tetravinyl-2,4,6,8-tetramethylcyclotetrasiloxane) (pV4D4) are characterized by their hydrophobic properties^35,51,52, offering unique hydrophobic protein-surface interactions^32,38,53. Some polymer surfaces are exposed to plasma surface treatment, increasing surface energy and cell adhesion, thereby providing surfaces with hydrophilic properties^54,55.

ISCs^3D-hIO differentiated from hiPSCs were seeded onto these predesigned biocompatible polymers possessing diverse functional properties to evaluate their suitability for xenogeneic-free ISC culture (Fig. 1c and Supplementary Fig. 2). For comparative analysis, we evaluated the colony sizes and the growth rate of ISCs^3D-hIO cultured on each polymer substrate on Days 2 and 4. Additionally, we obtained the surface energy values of the polymer surfaces with contact angles of water and glycerol, which is a simple but important surface property that influences cell–surface interactions^{29,32,35,56,57}. The surface energy values, colony areas, and growth rates were normalized on the basis of the maximum and minimum values for each variable and are depicted in a heatmap (Fig. 1d). The surface energy values of the polymer ranged between the minimum value of 27.7 mJ/m² for pCHMA and the maximum value of 97.1 mJ/m² for pDMAEMA. The three surfaces that demonstrated the largest colony areas on Day 2 were pEGDMA+p, Matrigel-coated surface, and p(DMAEMA_VBC), with colony areas of 286,000 µm², 250,000 µm², and 230,000 µm², respectively. On Day 4, pEGDMA+p, Matrigel-coated surface, and pCHMA+p groups presented the largest colony areas of 641,000 µm², 552,000 µm² and 403,000 µm², respectively. The growth rates were calculated on the basis of colony areas on Days 2 and 4, revealing that p(DMAEA_EGDMA) presented the highest growth rate at 134%, followed by pEGDMA+p at 124% and the Matrigel-coated surface at 120%. Through comprehensive analysis of the heatmap results, pEGDMA+p surface exhibited high initial ISC adhesion, leading to a large colony area and a high growth rate, supporting ISC proliferation. Consequently, we designated this surface XF-DISC, considering it the most suitable for ISC^3D-hIO culture. Our subsequent research efforts focused on exploring the XF-DISC surface for further investigation to ensure that XF-DISC can provide a Matrigel-free, stable surface for the long-term culture of ISCs^3D-hIO.

Evaluation of XF-DISC as a robust culture platform for ISCs^3D-hIO

To assess the potential of XF-DISC as a culture platform for ISCs^3D-hIO, we conducted a rigorous analysis with tissue culture polystyrene plates (TCPS) and Matrigel-coated TCPS as negative and positive control culture surfaces, respectively. XF-DISC exhibited no discernable differences in appearance compared with TCPS even after the iCVD process (Fig. 2a). However, a clear difference in wettability was observed, with the XF-DISC surface being more hydrophilic than the TCPS surface. This phenomenon was evidenced by the presence of red ink droplets, highlighting the advantageous characteristics of vapor‒phase conformal coverage at room temperature (Fig. 2b). ISC^3D-hIO colonies grew rapidly on both the Matrigel-coated surface and XF-DISC, whereas impeded growth was identified on the TCPS (Fig. 2c). ISCs^3D-hIO demonstrated stable attachment and mobility after cell seeding, and colonies exhibited good adherence and dynamic movement on XF-DISC (Supplementary Movies 1–6). On Day 6, direct cell counting revealed 38,600 ± 2,800 cells on the Matrigel-coated surface, 37,000 ± 3,000 cells on XF-DISC, and 27,600 ± 3,000 cells on TCPS (Fig. 2d). Using the water-soluble tetrazolium-1 (WST-1) proliferation assay, a significant decrease in performance was observed on TCPS, whereas Matrigel-coated surfaces and XF-DISC exhibited increased proliferation rates (Fig. 2e). The quantitative analysis results were fully consistent with the observations of colony morphology. In addition to TCPS, we cultured ISCs^3D-hIO on glass surfaces. Exceptional increases in colony areas were evident on the Matrigel- and XF-DISC-coated glass surfaces compared with the bare glass surface (Supplementary Fig. 3a). The colony areas quantified using crystal violet (CV) staining were 1.39 ± 0.23 cm² and 1.50 ± 0.05 cm² on the Matrigel-coated and XF-DISC surfaces, respectively (Supplementary Fig. 3b, c). In contrast, the bare glass surface had an area of only 0.51 ± 0.21 cm², which was less than half of the colony area observed on XF-DISC. These results confirmed that XF-DISC promotes the attachment and growth of ISCs^3D-hIO independent of the substrate material. This finding strongly indicates the possibility of achieving consistent effects when these methods are applied to various cell culture substrates.

a External appearance of TCPS (left) and XF-DISC (right). Scale bar: 10 cm. b Wettability differences between TCPS and XF-DISC were observed by dropping 30 µL of red ink dissolved in water on each surface. Scale bar: 5 cm. c Morphologies of ISCs^3D-hIO cultured on a Matrigel-coated surface (top), XF-DISC (middle), and TCPS (bottom) at Days 2, 4, and 6. The images in the right column are the zoomed-in images taken on Day 6. Scale bar: 500 µm. d Quantification of ISC^3D-hIO cell numbers on Day 6 after the seeding of cells on the Matrigel-coated surface, XF-DISC and TCPS. Data represent the mean ± SD (n = 3 biological samples), and a two-tailed t-test was applied to measure p values. e WST-1 assay of ISCs^3D-hIO cells cultured for 7 days on the Matrigel-coated surface, XF-DISC and bare TCPS. Data represent the mean ± SD ((n) = 3 biological samples), and a two-tailed t-test was applied to measure p values. f AFM images of XF-DISC surfaces from 3 different batches. Scale bar: 1 µm. g Crystal violet staining images of ISCs^3D-hIO cultured on XF-DISC from 3 different batches. Scale bar: 500 µm. h Quantification of colony areas on XF-DISC from 3 different batches. Data represent the mean ± SD (n = 3 biological samples), and a two-tailed t-test was applied to measure p values. i Morphologies of ISCs^3D-hIO cultured for 7 days on a Matrigel-coated surface (left), XF-DISC (middle) and XF-DISC stored for 2 years (right), passage 1. Scale bar: 500 µm. j, WST-1 assay of ISCs^3D-hIO cultured for 7 days on a Matrigel-coated surface, XF-DISC and XF-DISC stored for 2 years. Data represent the mean ± SD (n = 5 biological samples), and a two-tailed t-test was applied to measure p values. k Fluorescence microscopy images of ISCs^3D-hIO cultured on Matrigel-coated surfaces and XF-DISC. ISC^3D-hIO viability was assessed on Day 6 after seeding. Scale bar: 500 µm. l Quantification of the viability of ISCs^3D-hIO cultured on a Matrigel-coated surface and XF-DISC. Data represent the mean ± SD (n = 3 biological samples), and a two-tailed t-test was applied to measure p values.

Full size image

The XF-DISC surface was smooth, with a root mean square (RMS) roughness (R_q) less than 1 nm, as confirmed by atomic force microscopy (AFM) (Fig. 2f). The results of the quantitative analysis of colony areas derived from CV staining verified the excellent reproducibility of the XF-DISC culture system (Fig. 2g, h). In addition, XF-DISC showed exceptional long-term stability. Even after a storage period exceeding 2 years in ambient temperature air, XF-DISC demonstrated no apparent changes in its chemical composition and morphology or proliferation of ISCs^3D-hIO (Fig. 2i, j and Supplementary Fig. 4), which is highly desirable for reliable long-term use and storage of XF-DISC. The reproducibility and long-term stability characteristics of XF-DISC emphasize the potential for broader utilization of xenogeneic-free surfaces. Upon culture on the XF-DISC surface, ISCs^3D-hIO demonstrated a cell viability exceeding 99%, indicating that XF-DISC is suitable for cell proliferation (Fig. 2k, l). In the immunogenicity test using peripheral blood mononuclear cells (PBMCs)⁵⁸, the percentages of proliferating CD3-positive T cells were 1.86% and 1.93% in XF-DISC and TCPS, respectively, which were far lower than those in the phytohemagglutinin-L-stimulated group (5.73%) (Supplementary Fig. 5a). The inflammatory cytokine, tumor necrosis factor-α (TNF-α) secretion test was also performed to assess the immunogenicity of XF-DISC (Supplementary Fig. 5b, c). The levels of TNF-α secreted from RAW 264.7 mouse macrophages and THP-1 human monocytes after exposure to XF-DISC for 12 h and 48 h, respectively, were remarkably low and comparable to the level observed from cells on TCPS^59,60. To validate the clinical application potential, we conducted an endotoxin level test on XF-DISC (Supplementary Fig. 5d). Considering that the acceptable endotoxin level for implantable medical devices is 0.5 EU/mL according to the Food and Drug Administration (FDA) guidelines, both XF-DISC and TCPS are acceptable for implantable devices with endotoxin levels less than 0.12 EU/mL⁶¹.

Taken together, the results suggest that XF-DISC fully supported the culture of ISCs^3D-hIO at levels comparable to those of the Matrigel coating. Furthermore, a substantial amount of cell production would be possible with the advantageous characteristics of XF-DISC, such as minimal batch variation, high long-term stability, biocompatibility, and the absence of biological safety concerns.

Xenogeneic-free culture system for scalability, long-term expansion, and stock banking of ISCs^3D-hIO

To determine the feasibility of the xenogeneic-free ISC^3D-hIO culture system, we investigated the scalability, long-term expansion, and applicability of the stock banking system with ^ISC3D-hIO cultured on XF-DISC (Fig. 3a). Theoretical growth was extrapolated over a 30-day culture period using the increased cell count number observed in the first passage to assess the scalability of XF-DISC (Figs. 2d, 3b). The results indicated that the expected number of cells was 24-fold greater on both the XF-DISC surface and the Matrigel-coated surface. However, on the TCPS surface, only a 5-fold increase was observed. We then plotted experimental growth curves determined directly by counting the cell numbers during four consecutive passages and compared them with the theoretical growth curves obtained (Fig. 3c, d). Notably, a strong alignment between the experimental growth curve and the theoretical growth curve was observed. In addition, our consecutive passaging of ISCs^3D-hIO on XF-DISC yielded an impressive average expansion efficiency of greater than 93% over four passages compared with that of Matrigel-coated surfaces. Regarding the regenerative transplantation method, the ability to obtain a consistent, reliable supply of ISCs, especially a large number of ISCs, that are suitable for transplantation without xenogeneic issues is of utmost importance. In this study, ISCs^3D-hIO cultured on XF-DISC exhibited a high expansion rate, resulting in a 24-fold increase in cell number within 30 days, suggesting that XF-DISC is not only appropriate for the culture of ISCs^3D-hIO but also holds great promise as a stem cell culture platform for the large-scale production of ISCs^3D-hIO for extended cell therapeutics. Moreover, XF-DISC was capable of supporting long-term expansion, retaining the continuous formation of ISC^3D-hIO colonies for at least 30 consecutive passages without any morphological abnormalities (Fig. 3e). The demonstration of 30 successive passages every 7 to 10 days provided robust evidence that ISC^3D-hIO can be sustained for a minimum of 210 days, underscoring the remarkable stability and reliability of XF-DISC.

a Schematic illustration of the long-term expansion capability and implementation of a stock banking system for ISCs^3D-hIO cultured on XF-DISC. b Theoretical growth curve of ISCs^3D-hIO cultured on a Matrigel-coated surface; XF-DISC and TCPS were estimated, and the cell numbers were quantified. c d Theoretical growth rate and experimentally determined growth rate after ISCs^3D-hIO were passaged for four passages on a Matrigel-coated surface and XF-DISC, respectively. e Morphologies of ISCs^3D-hIO on a Matrigel-coated surface and XF-DISC over the course of long-term expansion culture across 30 sequential passages. Scale bar: 1 mm (n = 3 samples/group). f Morphologies of ISCs^3D-hIO on a Matrigel-coated surface and XF-DISC after freezing and thawing cycles on Days 2 and 5 (n = 3 samples/group). Before freezing, the number of passages was 10. Scale bar: 500 µm. g Quantitative efficiency of viable cells after freezing and thawing. Data represent the mean ± SD (n = 3 biological samples), and a two-tailed t-test was applied to measure p values. h WST-1 assay of thawed ISCs^3D-hIO cultured for 5 days on a Matrigel-coated surface and XF-DISC. Data represent the mean ± SD (n = 3 biological samples), and a two-tailed t-test was applied to measure p values. i Quantitative efficiency of viable cells after the freezing and thawing of stocks stored for more than 3 years. Data represent the mean ± SD (n = 3 biological samples), and a two-tailed t-test was applied to measure p values. j WST-1 assay of thawed ISCs^3D-hIO cultured for 5 days on a Matrigel-coated surface and XF-DISC. Stocks were stored for more than 3 years. Data represent the mean ± SD (n = 3 biological samples), and a two-tailed t-test was applied to measure p values.

Full size image

Next, we subjected the cell stocks to freezing and thawing to validate the feasibility of the stock banking system. Upon thawing the cell stock, we verified the formation of ISC^3D-hIO colonies on both the Matrigel-coated surface and XF-DISC (Fig. 3f). Quantification of the number of attached cells revealed efficiencies of greater than 82% on both the Matrigel-coated surface and the XF-DISC (Fig. 3g). In the proliferation assay, normalization to the values of the Matrigel-coated surface did not significantly differ from those of XF-DISC (Fig. 3h). Subsequently, ISCs^3D-hIO demonstrated normal colony formation in successive passages (Supplementary Fig. 6a). CV staining also revealed that more than 80% of both the Matrigel-coated surface and the XF-DISC covered the surface (Supplementary Fig. 6b, c). In addition, upon thawing stocks that had been stored for more than 3 years, the efficiency values were greater than 35%, and there was no significant difference between the Matrigel-coated surface and XF-DISC in supporting ISC^3D-hIO proliferation, illustrating long-term cell stock stability (Fig. 3i, j).

The culture of iPSC-derived ISCs^3D-hIO in a xenogeneic-free environment is of prime importance given the advantages of the potential use for universal regenerative therapies and the development of personalized therapies because hiPSCs can differentiate into all body cell types, enabling personalized medicine. Given that expanding the application scope of XF-DISC is also important, we present work in progress toward extending the application of XF-DISC to verify its adaptability across different stem cell types. First, greater than 99% H9-ISC^3D-hIO viability was observed when cultured on XF-DISC (Supplementary Fig. 7a, b). Additionally, the exceptional potential of our platform was confirmed again, supporting 10 consecutive passages of H9-ISCs^3D-hIO (Supplementary Fig. 7c). This observation strongly suggested broader expansibility of XF-DISC and its applicability to ISCs from diverse origins.

Together, these results clearly indicate that XF-DISC can effectively support xenogeneic-free culture of ISCs^3D-hIO while allowing for high scalability and long-term expansion. Moreover, the scalable XF-DISC enabled the establishment of a prolonged stock-banking system for future clinical applications.

Cellular trait evaluation of ISCs^3D-hIO on XF-DISC

To further elucidate the feasibility of XF-DISC as an ISCs^3D-hIO culture platform, we conducted a thorough characterization of ISCs^3D-hIO cultured on XF-DISC and Matrigel-coated surfaces. We investigated whether ISCs^3D-hIO cultured on XF-DISC exhibited messenger ribonucleic acid (mRNA) expression related to intestinal epithelial markers with a lack of the expression of markers associated with human pluripotent stem cell (hPSC) stemness. We conducted quantitative polymerase chain reaction (qPCR) analysis of the ISCs^3D-hIO cultured on Days 4, 6, and 8 and visualized the results through a heatmap (Fig. 4a). Upon examination of the heatmap of mRNA expression, it was apparent that markers associated with intestinal epithelial features, such as LGR5, EPHB3, SOX9, CD24, CD44, CD133, and CD166, were expressed at higher levels in ISCs^3D-hIO than in hPSCs. In contrast, the expression of the stemness markers for hPSCs, OCT4, and NANOG, and the proliferation marker MKI67 was much lower in ISCs^3D-hIO than in hPSCs, confirming the successful establishment of ISCs^3D-hIO from hPSCs. During long-term culture, to verify whether ISCs^3D-hIO maintains epithelial features, a comparison of marker expression between cells cultured on XF-DISC and those cultured on Matrigel-coated surfaces was conducted (Fig. 4b). Following passaging, the expression levels of ISC markers (LGR5, EPHB3, CD44 and SOX9) and the proliferation marker MKI67 remained comparable between ISCs^3D-hIO cultured on the two surfaces. In addition, maintenance of epithelial features was also observed in H9-ISCs^3D-hIO (Supplementary Fig. 7d). These results indicated that XF-DISC supports mass culture of ISCs^3D-hIO through successive passages while preserving the unique ISC characteristics.

a Heatmap of gene expression in hPSCs and ISCs^3D-hIO cultured on Matrigel-coated surfaces and XF-DISC for 4, 6, and 8 days at passage 3, as quantified by qRT‒PCR (n = 3). b Gene expression of intestinal markers (LGR5, EPHB3, CD44 and SOX9) and a proliferation marker (MKI67) of ISCs^3D-hIO cultured on a Matrigel-coated surface and XF-DISC for 7 days at passages 3, 8 and 18 quantified by qRT‒PCR. Data represent the mean ± SD (n = 3 biological samples), and a two-tailed t-test was applied to measure p values. c Cell population of markers (CD44, SOX9 and CD34) of ISCs^3D-hIO cultured on a Matrigel-coated surface and XF-DISC for 7 days, as determined by FACS analysis (Red: negative control cells in each sample, black: specific antibody stained-ISCs^3D-hIO cultured on Matrigel-coated surface and blue: specific antibody stained-ISCs^3D-hIO cultured on XF-DISC). d Representative images showing the immunostaining of SOX9, Ki67, EIF3E, FABP1, LDHB, CHGA and MUC2 expression in ISCs cultured on a Matrigel-coated surface and XF-DISC for 7 days at passage 2 (n = 3 samples/group). Scale bar: 100 µm. e Volcano plot depicting gene abundance changes in ISCs cultured on a Matrigel-coated surface and XF-DISC for 7 days at passage 17. The vertical red lines indicate ± 4-fold changes, and the horizontal red line indicates a minimum significance of 20 (p value = 0.01). Multidimensional scaling analysis of the pairwise distances of the sample was conducted. f MDS plot of the pairwise distances between samples. Six homogeneous sample groups were observed; Matrigel (black, n = 3), XF-DISC (red, n = 3), ALI (blue, n = 3), hSI (green, n = 2), iPSC (CRL, pink, n = 1), and ESC (H9, yellow, n = 1). g Volcano plot depicting protein abundance changes in ISCs cultured on a Matrigel-coated surface and XF-DISC for 3 days at passage 12. The vertical red lines indicate ± 4-fold changes, and the horizontal red line indicates a minimum significance of 20 (p value = 0.01). Label-free quantification-based proteomic analysis was performed via ANOVA.

Full size image

Fluorescence-activated cell sorting (FACS) analysis confirmed the presence of stem cells expressing CD44 and SOX9 within the ISC^3D-hIO population, whereas cells expressing the endothelial marker CD34 were notably scarce (Fig. 4c). In more detail, FACS analysis revealed that 91.4% and 93.6% of the cells from ISCs^3D-hIO cultured on the Matrigel-coated surface and XF-DISC, respectively, were CD44 positive. Regarding the SOX9-positive cell population, the results revealed that 97.0% of cells on the Matrigel-coated surfaces and 95.0% of cells on XF-DISC were positive for SOX9 expression, demonstrating that over 90% of the cells were in a stemness-enriched state. Furthermore, ISC^3D-hIO colonies stained positive for the cell proliferation marker Ki67 and exhibited an appropriate expression pattern of the intestinal stem cell marker SOX9. The expression patterns of the fetal stem cell markers EIF3E and LDHB were also similar on the two surfaces (Fig. 4d). FABP1, a marker of enterocytes, was also expressed in ISC colonies, whereas specific markers of goblet cells (MUC2) and enteroendocrine cells (CHGA) were not detected in ISC^3D-hIO colonies. Importantly, the staining patterns were similar when comparing ISCs^3D-hIO cultured on XF-DISC with those cultured on the Matrigel-coated surface.

To analyze ISC^3D-hIO characteristics at a comprehensive level, we performed RNA sequencing (RNA-seq) and proteomics analyses using ISCs^3D-hIO cultured on XF-DISC and Matrigel-coated culture dishes. A volcano plot of the RNA-seq data revealed no significant differences in global gene expression patterns between ISCs^3D-hIO cultured on XF-DISC and those cultured on Matrigel-coated surfaces (Fig. 4e). The results of multidimensional scaling (MDS) using our original and publicly available RNA-seq datasets of hiPSCs, hESCs, air‒liquid interface (ALI)-differentiated intestinal tissue and human small intestine (hSI) samples demonstrated that ISCs^3D-hIO cultured on XF-DISC were most closely related to ISCs^3D-hIO cultured on Matrigel-coated surface, whereas ISCs^3D-hIO exhibited distinct patterns compared with the hPSC, ALI or hSI samples (Fig. 4f). In addition, at the protein level, no significant differences were observed between ISCs^3D-hIO cultured on XF-DISC and those cultured on the Matrigel-coated surface (Fig. 4g).

Taken together, these data showed that the gene and protein expression of ISCs^3D-hIO and the cellular patterns within colonies exhibited negligible differences between those cultured on the Matrigel and XF-DISC surfaces, and these characteristics remained consistent even following consecutive passaging.

Xenogeneic-free formation of functional intestinal tissue on XF-DISCs through ALI differentiation

We investigated the ability of XF-DISC to support the formation of intestinal tissue through the ALI differentiation method¹⁷. XF-DISC-coated inserts were used to culture ISCs^3D-hIO in the absence of xenogeneic ECM, and the ALI differentiation process was induced to promote the formation of intestinal tissue-like structures (Fig. 5a). Owing to the advantageous characteristic of the iCVD process, which enables fully conformal coating in a solvent-free manner without any apparent damage, the XF-DISC-coated inserts revealed no observable differences from the bare inserts (Fig. 5b). Following 8 days of ALI differentiation, the intestinal epithelium derived from ISCs^3D-hIO cultured on XF-DISC-coated inserts exhibited a characteristic serpentine morphology resembling native intestinal structures (Fig. 5c). We performed qPCR analysis on ALI-differentiated intestinal tissues and compared them with human small intestine tissues. The expression profiles of markers associated with ISCs (LGR5, CD44, SOX9, AXIN2, and CTNNB), proliferation (MKI67), differentiation (VIL1, FABP1, KRT20, LYZ, and MUC2), intestinal maturation (OLFM4), and epithelial cells (CDH1) were examined in the intestinal epithelium formed on both the XF-DISC and Matrigel-coated surfaces (Fig. 5d). The presence of gene expression related to stem cells, proliferation, and differentiated cells clearly confirmed the successful development of functional intestinal tissues.

a Workflow of intestinal tissue formation on XF-DISC. b External appearance of bare (left) and XF-DISC (right) inserts. Scale bar: 5 cm. c Surface views of 3D intestinal tissues subjected to ALI differentiation on a Matrigel-coated surface and XF-DISC for 0, 2, 4, 6, and 8 days at passage 17 (n = 3 samples/group). Scale bar: 500 µm. d Gene expression of markers of human small intestine and ALI differentiated intestinal tissues on a Matrigel-coated surface and XF-DISC for 0, 4 and 8 days at passage 6, as quantified by qRT‒PCR. Data represent the mean ± SD (n = 3 biological samples). Statistical significance between Matrigel D0 cells and other cells was assessed by one-way ANOVA followed by Tukey’s post hoc test. e Representative images showing immunostained sections of ALI differentiated intestinal tissues on a Matrigel-coated surface and XF-DISC for 0, 4 and 8 days at passage 6 (n = 3 samples/group). Scale bar: 100 µm. f Representative images showing histological sections of ALI differentiated intestinal tissues on a Matrigel-coated surface and XF-DISC for 0, 4 and 8 days at passage 6. Scale bar: 20 µm (n = 3 samples/group). g TEER values of ALI-differentiated intestinal tissues on a Matrigel-coated surface and XF-DISC for 2, 4, 6, and 8 days at passage 5. Data represent the mean ± SD (n = 3 biological samples), and a two-tailed t-test was applied to measure p values.

Full size image

To further assess the expression patterns of differentiated intestinal tissues, we stained cryosectioned samples of the intestinal epithelium on Days 0, 4, and 8 (Fig. 5e). Over time, the increased structure became more prominent and intricate with increased expression of differentiation markers (KRT20 and VIL1). Based on the expression of a stem cell marker (CD44) and a proliferation marker (Ki67), we confirmed the ability to maintain homeostasis within a xenogeneic-free environment, underscoring the similarity to the native intestinal structure. The hematoxylin and eosin (H&E) staining results clearly revealed a similar trend of increasing height and complexity of the tissue structure (Fig. 5f). Even with the use of a different cell line, namely, H9-ISCs^3D-hIO, for ALI differentiation, we confirmed a distinctively differentiated morphology, illustrating the adaptability of XF-DISC for generating intestinal tissue across diverse cell sources (Supplementary Fig. 8a). No significant differences in the gene expression or expression patterns of differentiated H9-ISC^3D-hIO were observed between the XF-DISC surface and the Matrigel-coated surface (Supplementary Fig. 8b, c). The TEER values, a representative indicator of epithelial integrity⁶², were 295 ± 25 ohm × cm² for the Matrigel-coated surface and 293 ± 4 ohm × cm² for XF-DISC on Day 8 (Fig. 5g). During the ALI differentiation process, the TEER values remained constant for the two surfaces, indicating superior epithelial integrity and well-formed functional characteristics of the intestinal tissues.

Together, these results suggest that XF-DISC can fully support the formation of the intestinal epithelium, facilitating its application in various in vitro intestinal models without the need for xenogeneic ECM.

Therapeutic effect of xenogeneic-free cultured ISCs^3D-hIO

We transplanted ISCs^3D-hIO grown on XF-DISC for 4–6 days into the ethylenediamine tetraacetic acid (EDTA)-injured colonic epithelium of immunodeficient (NIG) mice and examined the progress of injury recovery via colonoscopy at Days 0 and 14 posttransplantation (PT) (Fig. 6a). On Day 0, EDTA-induced injuries exhibited the same degree of epithelial damage in both the fibrin-only and ISC^3D-hIO on XF-DISC groups (3 mice in each group). After orthotopic transplantation of ISCs^3D-hIO, enhanced thickening of the colon and no colonic bleeding were observed on Day 14 (Fig. 6b). Histological analysis with H&E and alcian blue/periodic acid-Schiff (AB-PAS) staining demonstrated that ISCs^3D-hIO successfully engrafted and induced regeneration of the rectal epithelium. The ISC^3D-hIO xenografts shaped the whole crypt and secreted mucin, as shown in Fig. 6c. A previous study indicated that the xenografted human colon formed larger crypt structures than the mouse colon¹². Notably, ISC^3D-hIO on XF-DISC (n = 246 crypts from 3 mice) exhibited markedly greater crypt depth than the fibrin-only group (n = 303 crypts from 3 mice) (Fig. 6d). Immunofluorescence staining further verified that the cells expressing human-specific E-cadherin (hECAD)⁺ were detected only in the ISCs^3D-hIO in the XF-DISC group (Fig. 6e). Thus, these data confirm that ISCs^3D-hIO on XF-DISC are potent cell sources for the treatment of intestinal injury. Additionally, visual observation confirmed normalcy in major adjacent organs (liver, kidney, spleen, stomach, and small intestine) (Fig. 6f).

a Workflow of ISC^3D-hIO transplantation in NOD/SCID mice. b Representative mouse colonoscopic observations over time. c Top, histopathology of the xenograft colon tissues based on H&E staining (n = 5 samples/group). Bottom, mucin staining of the xenograft colon tissues based on AB‒PAS staining (n = 2 samples/group). Scale bar: 500 µm. d Box and scatterplots of the crypt depth of fibrin-only (n = 246 crypts from three mice; 87, 80, and 79 crypts respectively) and ISCs^3D-hIO on XF-DISC (n = 303 crypts from three mice; 89, 117, and 97 crypts respectively) treated xenograft colon tissues following H&E staining. The quartiles of the boxplot are mean ± SD, and a Welch’s unpaired t-test was applied to measure p value. e Human E-cadherin staining of xenograft tissues, indicating xenotransplanted ISCs^3D-hIO cultured on XF-DISC (n = 2 from each group). Scale bar: 100 µm. f, Bright-field images of the main organs after xenotransplantation (n = 4).

Full size image

Next, we established a DSS-induced colitis model in NIG mice and xenotransplanted ISCs^3D-hIO into these mice (Fig. 7a). Although a large area of ulcers was noted on Day 0 and the severity was quite high in the ISC^3D-hIO on XF-DISC group, symptoms were quickly alleviated on Day 14 PT, and ulceration was reduced considerably along with mucosal changes according to colonoscopic analysis (Fig. 7b, Supplementary Movies 7, 8). On Day 28 PT with ISCs^3D-hIO grown on XF-DISC, the colonic epithelium was apparently thickened, with low vasculature and practically no ulceration. Histopathological analysis with H&E staining also revealed that xenografted ISCs^3D-hIO regenerated the colonic epithelium, and functional goblet cells were detected via AB-PAS staining, as evidenced by mucin secretion (Fig. 7c). Notably, the crypt depth in the ISC^3D-hIO on XF-DISC group (n = 714 crypts from 6 mice) was also far greater than that in the fibrin-only group (n = 228 crypts from 3 mice) (Fig. 7d). In the immunofluorescence analysis, positive detection of hECAD was exclusively observed in the ISC^3D-hIO on the XF-DISC xenograft group (Fig. 7e).

a Workflow of ISC^3D-hIO transplantation in NOD/SCID mice. b Representative mouse colonoscopic observations over time. c Top, histopathology of the xenograft colon tissues by H&E staining (n = 10 samples/group). Bottom, mucin staining of the xenograft colon tissues by AB‒PAS staining. Scale bar: 500 µm (n = 2 samples/group). d Box and scatterplots of the crypt depth of fibrin-only (n = 228 crypts from three mice; 76, 81, and 71 crypts respectively) and ISCs^3D-hIO on XF-DISC (n = 714 crypts from six mice; 115, 131, 110, 118, 65, and 175 crypts respectively) treated xenograft colon tissues following H&E staining. The quartiles of the boxplot are mean ± SD, and a Welch’s unpaired t–test was applied to measure p-value. e Human E-cadherin staining of xenograft tissues, indicating xenotransplanted ISCs cultured on XF-DISC (n = 2 from each group). Scale bar: 100 µm.

Full size image

Collectively, the in vivo data revealed that ISCs^3D-hIO grown on XF-DISC were successfully implanted into EDTA-injured mice, and these cells were integrated with and restored the colon tissue in the DSS-induced colitis model, clearly confirming that the ISCs^3D-hIO grown on XF-DISC represent promising therapeutics for intestinal diseases.

• SEO Powered Content & PR Distribution. Get Amplified Today.
• PlatoData.Network Vertical Generative Ai. Empower Yourself. Access Here.
• PlatoAiStream. Web3 Intelligence. Knowledge Amplified. Access Here.
• PlatoESG. Carbon, CleanTech, Energy, Environment, Solar, Waste Management. Access Here.
• PlatoHealth. Biotech and Clinical Trials Intelligence. Access Here.
• Source: https://www.nature.com/articles/s41467-024-54653-9