A Coculture System for Modeling Intestinal Epithelial-Fibroblast Crosstalk

Abstract

Epithelial organoid monoculture is a powerful tool to model stem cell dynamics in vitro. However, extensive efforts have recently revealed various niche players and their significant roles in regulating epithelial stem cells. Among these niche components, fibroblasts have been heavily recognized in the field as a critical niche signal secretor. Thus, understanding the roles of fibroblasts in epithelial dynamics has become increasingly relevant and crucial. This propels the development of approaches to coculture epithelial 3D organoids with fibroblasts to model epithelial-fibroblast crosstalk in vitro. Here, we describe a stepwise coculture method to isolate and culture primary intestinal fibroblasts and epithelial organoids together. Aligned with the recent literature, our coculture protocol allows for primary intestinal fibroblast support of epithelial organoid growth.

1. Introduction

The mammalian intestine has become an ideal model system to study stem cell dynamics, cell fate decisions, and stem cell-niche interactions. This is in part due to the rapid cellular turnover in the epithelium, the distinct morphology and compartmentalization of the stem cell niche (the crypt), and the relationship between active and reserve stem cells [1–5]. The robust intestinal organoid system [6, 7] further allows the modeling of cell-cell interactions and cellular responses in a 3D environment. Apart from the simple protocol for generating organoids from isolated mouse crypts, the organoid culture system has become more cost-effective through the use of conditioned media from cell lines engineered to secrete essential growth factors for organoid culture, instead of relying on expensive recombinant proteins [7–9]. Importantly, the field has grown around the identification and characterization of distinct niche cell types that regulate stem cells through secretion of various factors. This stimulates the development of numerous in vitro coculture models of intestinal epithelial cells with various niche players, including fibroblasts, smooth muscles, enteric neurons, lymphatic endothelial cells, and immune cells [10–15]. Here, with step-by-step instructions, we describe the isolation and culture of intestinal primary fibroblasts, the isolation and maintenance of intestinal epithelial crypts and organoids, as well as the coculture of the two cellular components together to model epithelial-fibroblast crosstalk [6, 8, 10, 11]. While we offer a coculture method of epithelial cells and bulk fibroblast populations, our method can be adapted to culture distinct fibroblast subpopulations after cell purification and enrichment. Performing coculture of epithelial cells and distinct fibroblast subpopulations further defines the necessity of these unique fibroblast subpopulations in regulating epithelial cell behaviors. This system enables the study of live dynamics between fibroblasts and epithelial cells that have been previously difficult to visualize and parse apart.

2. Materials

2.1. Small Intestine Isolation

1. 10X Phosphate-buffered saline (PBS): 1.37 M NaCl, 27 mM KCl, 100 mM Na₂HPO₄, 18 mM KH₂PO₄ in water, pH 7.4. Dilute to 1X in water for experimental use.

2. Dissection tools: Two Dumont #5 forceps and a pair of dissection scissors. Autoclave tools before isolation.

3. 10 mL syringe.

4. 21 gauge needle.

2.2. Mesenchymal Cell Isolation

1. 20 mM Ethylenediaminetetraacetic acid (EDTA) in Hanks Balanced Salt Solution (HBSS, calcium ion free, magnesium ion free; Gibco #14170-112): Add 400 μL of 0.5 M EDTA (pH 8.0) to 9.6 mL HBSS into a 15 mL conical tube for adult tissue. Warm at 37 °C. For postnatal mesenchymal cell isolation, prepare 10 mM EDTA in HBSS.

2. DMEM complete media: 10% fetal bovine serum (FBS) and 1X penicillin-streptomycin in DMEM, high glucose, pyruvate (Gibco #11995-065).

3. Enzyme mixture: 1.5 mg/mL collagenase, type 2, and 1 mg/mL dispase II in DMEM complete media. Prepare 15 mL enzyme mixture per animal. Split it into three 15 mL conical tubes each containing 5 mL enzyme mixture. Warm at 37 °C.

4. ACK buffer.

5. Nutating mixer.

6. 60 mm petri dish.

7. Vortex.

8. 10 mL serological pipettes.

9. 37 °C incubator.

10. 40 μm cell strainer.

11. 15 and 50 mL conical tubes.

12. Paper towels

2.3. Primary Fibroblast Culture

1. Fibroblast plating media: 10% FBS, 1X GlutaMAX, 10 mM HEPES, 1X penicillin-streptomycin in advanced DMEM/F12 media (Gibco #12634-010). Supplement with 1X amphotericin B for fresh primary cell isolation to avoid fungal contamination from intestine isolation. Remove amphotericin B in subsequent cell culture.

2.4. Fresh Crypt Isolation

1. 3 mM EDTA in PBS: Add 30 μL of 0.5 M EDTA (pH 8.0) to 5 mL of 1X PBS in a 15 mL conical tube. Keep on ice.

2. 15 mL conical tube.

3. Paper towel.

4. 22 mm × 22 mm glass coverslip.

5. Dumont SS long fine forceps.

6. 60 mm petri dish.

7. 3 mL transfer pipette.

8. 1X PBS.

9. Tube rotator.

2.5. Primary Intestinal Organoid Culture

1. Mouse EGF Recombinant Protein (Gibco): Prepare 0.1 mg/mL stock by dissolving 100 μg EGF in 1 mL sterile filtered 0.1% BSA in 1X PBS. Make 10 μL aliquots. Store at −80 °C.

2. Human Noggin Recombinant Protein (R&D Systems) (see Note 1): Prepare a 250 μg/mL stock by dissolving 25 μg lyophilized Noggin protein in 100 μL 0.1% BSA dissolved in PBS. Store at −20 °C.

3. Recombinant Mouse R-spondin 1 Protein (R&D Systems) (see Notes 1 and 2): Prepare a 250 μg/mL stock by dissolving 50 μg R-spondin 1 in 200 μL of 0.1% BSA in 1X PBS.

4. N-acetyl-L-cysteine (NAC) (Sigma): Prepare a 100 mM stock solution by dissolving 0.0816 g NAC in 5 mL dH₂O. Sterile filter using a 0.2 um syringe filter and make 1 mL aliquots. Store at −20 °C.

5. Basal organoid media base: To advanced DMEM/F12 media, add HEPES to a final concentration of 10 mM, GlutaMAX to 1X final concentration, and penicillin-streptomycin to 1X final concentration. Store at 4 °C.

6. Basal organoid media: For 50 mL, add 48 mL basal organoid media base to a 50 mL conical tube. Add 500 μL N-2 supplement (1X), 1 mL B-27 supplement (1X), and 500 μL of 100 mM N-acetyl-L-cysteine (1 mM). Mix thoroughly. Store at 4 °C for up to 1 month.

7. ENR media: Add 50 ng/mL mouse recombinant EGF, 50 ng/mL human recombinant Noggin, 1 μg/mL mouse recombinant R-spondin 1, and 1X amphotericin into basal organoid media (see Note 3).

8. Growth factor reduced Matrigel (see Note 5): Store at −80 °C.

9. Sterile 1X PBS.

10. 35 mm petri dish.

11. 15 mL conical tube.

2.6. Organoid Passaging

1. ENR media (see Subheading 2.5, step 7).

2. Gentle Cell Dissociation Reagent (see Note 4): Store at 4 °C.

3. Growth factor reduced Matrigel (see Note 5): Store at −80 °C.

4. 15 mL conical tube.

5. P1000 pipette.

6. 35 mm petri dish.

2.7. Epithelial-Fibroblast Coculture

1. Coculture media: Add 10% FBS and 50 ng/mL recombinant mouse EGF protein in basal organoid media (see Subheading 2.5, step 6). Supplement with 10 μM Y-27632 and 1X amphotericin B for coculture with epithelial crypts freshly isolated from intestine. Remove Y-27632 and amphotericin B in subsequent media changes. Prewarm before media changes.

3. Methods

All described animal work was approved by the Institutional Animal Care and Use Committee (IACUC) of Yale University. Permissions for live vertebrate studies must be obtained from relevant institutions prior to performing the following experiments. Proper aseptic technique is required for the described cell culture procedures using a biosafety laminar flow hood.

3.1. Small Intestine Isolation

1. Euthanize an adult mouse per institutional IACUC animal protocol. In this method, adult mice were euthanized by CO₂ inhalation followed by cervical dislocation as a secondary euthanasia method.

2. Using autoclaved dissection tools, make a large vertical incision on the skin layers along the ventral midline. Make horizontal cuts to create skin flaps as necessary to expose the abdominal cavity. Then, make a similar incision on the peritoneal membrane to expose the intestine.

3. Pull out the small intestine while separating it from the mesentery with forceps. Remove as much connective tissue as possible. To isolate the small intestine, locate the stomach proximal to the small intestine and the cecum distal to the small intestine. Using scissors, cut the intestine at the stomach and just proximal to the cecum to remove.

4. Cut out 5–10 cm of the desired small intestinal region (see Note 6). Place the tissue segment in ice-cold PBS in a 60 mm dish. Flush out the luminal content with ice-cold PBS using a 10 mL syringe and 21-gauge needle.

5. Follow Subheading 3.2 for mesenchymal cell isolation or Subheading 3.4 for epithelial crypt isolation.

3.2. Mesenchymal Cell Isolation

1. Cut the intestinal tissue into ~2 cm pieces. Then, on a paper towel, cut longitudinally to open the gut tube and expose the epithelial surface (Fig. 1a). Transfer the flat tissue to EDTA solution (see Note 7) prewarmed at 37 °C. Incubate the tissue at 37 °C for 20 min on a nutating mixer.

2. After 20 min, vortex the tissue vigorously at room temperature (RT). To remove epithelium, vortex the tissue in EDTA solution at max speed for 1 min. Massive epithelial sheets, including both crypts and villi, should be released during this initial vortex (Fig. 1b, c). Replace the EDTA solution with ice-cold 1X PBS. Repeat vortex and replacement of ice-cold PBS twice or until epithelial sheets are not released upon vortex.

3. Under a dissection microscope, use fine forceps to peel off muscle layers from one corner of the mesenchymal tissue. Then hold down the mesenchymal tissue while stripping the rest of the muscle layers from the mesenchymal tissue. This also removes any remaining connective tissues from the mesentery during intestinal isolation in Subheading 3.1, step 3 (Fig. 1d) (see Note 8). Vortex the mesenchymal tissue without muscle layers in ice-cold PBS again at max speed for 45 s at RT. The resulting mesenchymal tissue should be transparent (Fig. 1e).

4. Place the thin, transparent mesenchymal tissue on top of a 60 mm dish lid. Mince the tissue as finely as possible with dissection scissors and a razor blade. A well-minced tissue is critical for a good fibroblast yield.

5. Transfer the minced tissue to 5 mL enzyme mixture with collagenase, type 2, and dispase II in a 15 mL conical tube for digestion, prewarmed at 37 °C. Pipette up and down with a 10 mL serological pipette to declump the minced tissue.

6. Incubate at 37 °C for 10 min on a nutating mixer for tissue digestion.

7. After 10 min, pipette up and down with a serological pipette to disrupt tissue clumps and vortex (see Note 9). Then, incubate for an additional 10 min at 37 °C on a nutating mixer.

8. Pass the tissue suspension through a 40 μm cell strainer to collect isolated mesenchymal cells on ice. Collect the undigested tissue from the cell strainer and place it in the second tube of 5 mL enzyme mixture for longer digestion.

9. Repeat step 6–8 for two more rounds of digestion and cell collection, with a total of 60 min digestion/incubation time. The resulting 15 mL of cell suspension in enzyme mixture can be collected into a single 50 mL conical tube on ice.

10. Spin down the mesenchymal cell suspension at 350 G for 5 min at RT. Cell pellet may appear slightly reddish due to the presence of substantial red blood cells (RBCs). Carefully remove supernatant. Flick the tube bottom gently to loosen the cell pellet.

11. To remove RBCs, add 1 mL of ACK buffer to the loosened mesenchymal cell pellet on ice for 1 min sharp. After 1 min, immediately quench ACK buffer with 10 mL of ice-cold DMEM complete media (see Note 10).

12. Spin down the mesenchymal cell suspension at 350 G for 5 min at RT. The resulting cell pellet should be white after RBC removal. In this method, a 5–10 cm intestinal fragment generally yields above six million isolated single mesenchymal cells with over 90% viability.

Figure 1: De-epithelialization of intestinal tissue during mesenchymal cell isolation.

(A) Longitudinally cut whole intestine after EDTA treatment before vortexing. (B) Massive epithelial sheets and fragments detached from the whole intestine after initial vortexing. (C) Higher magnification image of the detached epithelial sheets after initial vortexing. Note the successful removal of both crypts (yellow arrowhead) and villi (red arrow) from the intestinal tissue. (D) Separation of muscle layers from mesenchymal tissue. Red arrowhead marks the muscle layers. (E) Transparent mesenchymal tissue resulted from the removal of epithelial sheets and muscle layers after vigorous vortexing. Note the difference in tissue transparency relative to A. Scale bars, 0.5 mm.

3.3. Primary Fibroblast Culture

The mesenchymal cell isolation protocol described in Subheading 3.2 results in a mixture of fibroblasts and other mesenchymal cell types. This fibroblast culture protocol is designed to enrich for fibroblasts without performing fluorescence-activated cell sorting (FACS) to reduce cell stress, experimental cost, time and effort.

1. Resuspend cell pellet from mesenchymal cell isolation in fibroblast plating media, supplemented with 1X amphotericin B.

2. Seed cells derived from the adult animal into a tissue culture dish in the range of 0.47 million to 0.67 million cells per cm² area. In this protocol, 0.4 million cells were seeded per 48 well (0.75 cm² cell growth area) (see Note 11). For immunofluorescence staining and fibroblast imaging, place a glass coverslip in the tissue culture dish before seeding cells. Culture cells in a 37 °C, 5% CO₂ tissue culture incubator.

3. After 24 h, cells will have attached to the dish bottom, and some fibroblasts will have begun spreading (Fig. 2a). Wash once with sterile 1X PBS to remove debris. Replace with prewarmed fibroblast plating media.

4. Over the next several days, fibroblasts will continue to elongate (Fig. 2b, c). Change media every 3–4 days as needed.

5. Within 5–8 days, fibroblasts should have reached confluency and acquired the stereotypical spindle-like morphology. By day 8 in culture, the majority of the cells are fibroblasts (Fig. 2d–f).

Figure 2: Confluent monolayer culture consisting of primarily fibroblasts as a result of mesenchymal cell isolation without fluorescence-activated cell sorting.

The spreading and elongation of primary fibroblasts 1 day (A), 2 days (B), and 5 days (C) after mesenchymal cell isolation (red arrowhead). Note the round-up and detachment of other mesenchymal cell types (yellow arrows) as fibroblasts continue to spread and elongate in the dish. (D-F) After 8 days in culture, fibroblasts reach confluency. Note that fibroblasts with nuclei labeled with PDGFRα-H2B-GFP (D) are the major mesenchymal cell type in the monolayer culture, relative to all cells with nuclei labeled with DAPI (E). Overlaid image of PDGFRα-H2B-GFP fibroblasts and all cell nuclei (F). Scale bars, 0.1 mm.

3.4. Fresh Crypt Isolation

1. Thaw a 250 μL aliquot of Matrigel on ice.

2. Immediately following small intestine isolation, cut the 5–10 cm tissue segment into multiple 1–2 cm pieces (Fig. 3a). Open the 1–2 cm tissue segments longitudinally on a paper towel to expose the epithelium and flatten the intestine, with the muscle layer facing the paper towel (Fig. 3b).

3. Using a 22 mm × 22 mm coverslip, scrape the intestinal surface to remove intestinal villi (Fig. 3b) (see Note 12). The scraped intestinal segment will appear translucent (Fig. 3c).

4. Place the scraped intestinal segment into a 15 mL conical tube containing ice-cold 3 mM EDTA. Incubate for 25 min at 4 °C rotating.

5. Under a dissecting microscope, pour the contents of the conical tube into a 60 mm petri dish. Prepare another 60 mm petri dish with 1X PBS. Using long fine forceps, pick up the intestinal segment and quickly shake it in the clean PBS petri dish to release villi (Fig. 4a, c). After a few shakes, discard the PBS into a waste container and replace it with fresh 1X PBS. Repeat shaking the intestinal segment in PBS, disposing the PBS with villar fragments, and adding fresh 1X PBS twice more for a total of three times to wash off the EDTA and deplete villi from the sample.

6. With fresh 1X PBS in the 60 mm petri dish containing the intestinal segment, shake the intestinal segment vigorously and observe closely for individual and/or sheets of intestinal crypts. The intestinal segment will become progressively translucent as epithelium is released from the tissue (Fig. 4b). If there are still numerous villi, continue to dispose and replace the PBS until minimal villi remain and there is an enrichment of crypts (Fig. 4d).

7. Collect the PBS containing the crypts using a transfer pipette and transfer the crypts back into the 15 mL conical tube on ice. Continue adding PBS and shaking the intestinal segment until no crypts come off (see Note 13). Discard the remaining intestinal segment.

8. Centrifuge the 15 mL conical tube containing the crypt suspension at 250 G for 5 min.

Figure 3. Removal of villi during intestinal epithelial crypt isolation.

(A) A 2 cm intestinal tissue segment obtained for fresh crypt isolation on a paper towel. Arrowhead points to mesentery that is removed before longitudinally cutting open the tissue segment to expose the epithelium and scraping the villi using a glass coverslip (outlined by the dotted white line) (B). (C) The same intestinal segment after being scraped; resulting in a more translucent appearance. Scale bars, 1 cm.

Figure 4. Isolation of intestinal epithelial crypts for primary organoid culture.

(A, B) Scraped intestinal tissue segments at the beginning and end of shaking for intestinal crypts, respectively. Intestinal tissue at the beginning of shaking looks fluffy due to most epithelial cells still being attached to the mesenchyme. By the completion of shaking, the tissue acquires a more transparent appearance, as most epithelial cells have been shaken off (arrowheads). Scale bars, 1 cm. (C, D) Shaken epithelial crypts and villi at the beginning and end of shaking, respectively. Magenta arrows point to villi fragments and sheets. The number of crypts and the ratio of crypts:villi increase during shaking (shown in the progression from (C) to (D)), which will indicate when to collect crypts for epithelial-fibroblast co-culture or stock organoid culture. Scale bars, 1 mm.

3.5. Primary Intestinal Organoid Culture

1. Prepare ENR media and warm in 37 °C water bath.

2. Following fresh crypt isolation, take the conical tube into a tissue culture hood and aspirate the PBS, making sure to not perturb the crypt pellet.

3. Pipette the crypt pellet (~50–100 μL) and resuspend in a thawed 250 μL aliquot of Matrigel on ice (see Note 14).

4. Pipette 40 μL domes of the crypt-Matrigel mixture onto a 35 mm petri dish.

5. Place the 35 mm petri dish in a 37 °C, 5% CO₂ tissue culture incubator for 20 min to allow the Matrigel domes to polymerize.

6. Add 2 mL of prewarmed ENR media and maintain in a tissue culture incubator. Epithelial crypts will be intact and evident within the Matrigel (Fig. 5a).

7. The next day, small spheroids should have formed (Fig. 5b). Change media every 2–3 days, as needed. Over the course of 5–7 days, organoids will begin to generate extensive buds (Fig. 5c) (see Note 15). Once organoids have budded, they are ready to be passaged, generally every 5–7 days (see Subheading 3.6 and Note 16).

Figure 5. Primary organoid formation from fresh isolated crypts.

(A) Intestinal crypts plated in a Matrigel dome (magenta arrowheads). (B) Spheroids grown from the isolated intestinal crypts one day after plating the crypts. (C) Budding organoids seven days after plating (blue arrows pointing to buds). These organoids are ready to be passaged. Scale bars, 0.25 mm.

3.6. Organoid Passaging

1. Prepare ENR media and warm in a 37 °C water bath.

2. Aspirate media from 35 mm petri dish of organoids.

3. Wash with 2 mL sterile 1X PBS.

4. Add 1 mL of Gentle Cell Dissociation Reagent and incubate for 1 min.

5. Using a P1000 pipette, pipette the Gentle Cell Dissociation Reagent numerous times up and down to detach the Matrigel domes from the plate and to break apart the Matrigel domes. After thoroughly breaking apart the domes, transfer to a 15 mL conical tube. Wash the plate with an additional 1 mL Gentle Cell Dissociation Reagent and add to the 15 mL conical tube.

6. Incubate the 15 mL conical tube with dissociated organoids on a nutating mixer at room temperature for 10 min.

7. Centrifuge the dissociated organoids at 250 G for 5 min.

8. Aspirate the Gentle Cell Dissociation Reagent, careful not to disturb the organoid pellet.

9. Resuspend the organoid pellet with 2–5 mL of sterile PBS and pipette up and down to further disrupt the organoids, with the goal of obtaining dissociated crypt buds.

10. Centrifuge the dissociated organoids at 250 G for 5 min.

11. Aspirate the PBS.

12. Pipette ~50 μL of organoid pellet and resuspend in an aliquot of thawed Matrigel on ice. Pipette up and down to uniformly distribute organoid fragments through the Matrigel, careful to avoid introducing air bubbles.

13. Pipette 40 μL domes of the crypt-Matrigel mixture onto a 35 mm petri dish.

14. Place the 35 mm petri dish in a 37 °C, 5% CO₂ tissue culture incubator for 20 min to allow the Matrigel domes to polymerize.

15. Add 2 mL of prewarmed ENR media and leave to grow in a tissue culture incubator. Change media every 2–3 days, as needed. Passage every 5–7 days by repeating this section (Fig. 6).

Figure 6. Progression of growth in passaged stock organoids.

(A) Organoid fragments immediately after passaging and plating in new Matrigel (magenta arrowheads). (B) Spheroids grown from the organoid fragments one day after passaging (yellow arrowheads). (C) Budding organoids seven days after passaging. These organoids are ready to be passaged again. Scale bars, 1 mm.

3.7. Epithelial-Fibroblast Coculture

This epithelial-fibroblast coculture system serves to model epithelial-fibroblast interactions. Without supplementing in culture media, fibroblasts secrete and provide essential growth factors to support epithelial cell growth. This is particularly useful when defining the role of fibroblasts as niche components.

• To prepare epithelial cells, follow Subheading 3.4 to freshly isolate crypts from the small intestine or follow Subheadings 3.5 and 3.6 to culture and passage intestinal organoids into smaller spheroids prior to coculture.

• To prepare fibroblasts, follow Subheadings 3.2 and 3.3 to isolate and culture primary intestinal fibroblasts. Begin coculture experiment when fibroblast monolayer reaches 80–90% confluency (see Note 17).

  1. Thaw an aliquot of Matrigel on ice.
  2. Prepare coculture media and warm in 37 °C water bath.
  3. Resuspend fresh epithelial crypts (from Subheading 3.4) or passaged epithelial spheroids (from Subheading 3.6) in Matrigel. Here, freshly isolated epithelial crypts were resuspended in Matrigel.
  4. Aspirate media from fibroblast monolayers. Here, fibroblast monolayers were grown in 48-well format and reached 80–90% confluency 5 days after isolation and plating (Fig. 2c).
  5. Add 40 μL of Matrigel containing epithelium on top of the fibroblast monolayer cultured in a 48 well. A thick layer of Matrigel is expected to form on top of the fibroblasts instead of a dome as the bottom fibroblast monolayer is moist (see Notes 18 and 19).
  6. Allow Matrigel to solidify for 20 min in a 37 °C, 5% CO₂ tissue culture incubator.
  7. Add prewarmed coculture media with 10 μM Y-27632. 1X amphotericin B is recommended for coculture when using freshly isolated epithelial crypts to minimize fungal contamination from intestine isolation (Fig. 7a).
  8. Over the next several days, epithelial spheroids in coculture continue to grow in the presence of fibroblasts without supplementing Noggin or R-spondin 1 in culture media. Change media every 2 days with fresh coculture media (see Fig. 7).

Figure 7. Fibroblasts support intestinal epithelial growth in co-culture.

(A) Graphic outlining the assembly of epithelial-fibroblast co-culture. Isolated fibroblasts are plated first to spread and reach 80-90% confluency, then freshly isolated crypts or passaged epithelial spheroids in Matrigel are added on top of the fibroblast monolayer. (B) Co-culture immediately after plating freshly isolated crypts (magenta arrowheads) in Matrigel, on top of a fibroblast monolayer (red arrows pointing to individual fibroblasts that make up the monolayer). Co-culture two days (C) and four days (D) after plating, tracking the same epithelial spheroids by blue and yellow arrowheads, respectively. Epithelial spheroids have grown with the influence of fibroblasts (red arrows) without supplementing essential growth factors in the culture media. Note the drastic increase in spheroid size as an indicator of epithelial cell growth and a successful co-culture. Scale bars, 0.5 mm.