Co-Culture of Murine Small Intestine Epithelial Organoids with Innate Lymphoid Cells

Summary

This protocol offers detailed instructions for establishing murine small intestine organoids, isolating type-1 innate lymphoid cells from the murine small intestine lamina propria, and establishing 3-dimensional (3D) co-cultures between both cell types to study bi-directional interactions between intestinal epithelial cells and type-1 innate lymphoid cells.

Abstract

Complex co-cultures of organoids with immune cells provide a versatile tool for interrogating the bi-directional interactions that underpin the delicate balance of mucosal homeostasis. These 3D, multi-cellular systems offer a reductionist model for addressing multi-factorial diseases and resolving technical difficulties that arise when studying rare cell types such as tissue-resident innate lymphoid cells (ILCs). This article describes a murine system that combines small intestine organoids and small intestine lamina propria derived helper-like type-1 ILCs (ILC1s), which can be readily extended to other ILC or immune populations. ILCs are a tissue-resident population that is particularly enriched in the mucosa, where they promote homeostasis and rapidly respond to damage or infection. Organoid co-cultures with ILCs have already begun shedding light on new epithelial-immune signaling modules in the gut, revealing how different ILC subsets impact intestinal epithelial barrier integrity and regeneration. This protocol will enable further investigations into reciprocal interactions between epithelial and immune cells, which hold the potential to provide new insights into the mechanisms of mucosal homeostasis and inflammation.

Introduction

Communication between the intestinal epithelium and gut-resident immune system is central to the maintenance of intestinal homeostasis¹. Disruptions to these interactions are associated with both local and systemic diseases, including Inflammatory Bowel Disease (IBD) and gastrointestinal cancers². A notable example of one more recently described critical regulator of homeostasis comes from the study of innate lymphoid cells (ILCs), which have emerged as key players in the intestinal immune landscape³. ILCs are a group of heterogenous innate immune cells that regulate intestinal homeostasis and orchestrate inflammation largely through cytokine-mediated signalling⁴.

Murine ILCs are broadly divided into subtypes based on transcription factor, receptor, and cytokine expression profiles⁵. Type-1 ILCs, which include cytotoxic Natural Killer (NK) cells and helper-like type-1 ILCs (ILC1s), are defined by expression of the transcription factor (eomesodermin) Eomes and T-box protein expressed in T cells (T-bet)⁶, respectively, and secrete cytokines associated with T helper type-1 (T_H1) immunity: interferon-γ (IFNγ) and tumor necrosis factor (TNF), in response to interleukin (IL)-12, IL-15 and IL-18⁷. During homeostasis, tissue-resident ILC1s secrete Transforming Growth Factor β (TGF-β) to drive epithelial proliferation and matrix remodelling⁸. Type-2 ILCs (ILC2s) primarily respond to helminth infection via secretion of T helper type-2 (T_H2) associated cytokines: IL-4, IL-5, and IL-13, and are characterized by the expression of retinoic acid-related orphan receptor (ROR) α (ROR-α)⁹ and GATA Binding Protein 3 (GATA-3)^10,11,12. In mice, intestinal "inflammatory" ILC2s are further characterized by expression of Killer cell lectin-like receptor (subfamily G member 1, KLRG)¹³ where they respond to epithelial tuft-cell derived IL-25^14,15. Finally, type-3 ILCs, which include lymphoid tissue inducer cells and helper-like type-3 ILCs (ILC3s), are dependent on the transcription factor ROR-γt¹⁶, and cluster into groups that secrete either Granulocyte Macrophage Colony Stimulating Factor (GM-CSF), IL-17, or IL-22 in response to local IL-1β and IL-23 signals¹⁷. Lymphoid tissue inducer cells cluster in Peyer's patches and are crucial for the development of these secondary lymphoid organs during development¹⁸, whereas ILC3s are the most abundant ILC subtype in the adult murine small intestine lamina propria. One of the earliest murine intestinal organoid co-culture systems with ILC3s was harnessed to tease apart the impact of the cytokine IL-22 on Signal Transducer And Activator Of Transcription 3 (STAT-3) mediated Leucine-Rich Repeat Containing G Protein Coupled Receptor 5 (Lgr5)⁺ intestinal stem cell proliferation¹⁹, a powerful example of a regenerative ILC-epithelial interaction. ILCs exhibit imprint-heterogeneity between organs^20,21 and exhibit plasticity between subsets in response to polarizing cytokines²². What drives these tissue-specific imprints and plasticity differences, and what role they play in chronic diseases such as IBD²³, remain exciting topics that could be addressed using organoid co-cultures.

Intestinal organoids have emerged as a successful and reliable model to study the intestinal epithelium^24,25. These are generated by culturing intestinal epithelial Lgr5⁺ stem cells, or entire isolated crypts, which include Paneth cells as an endogenous source of Wnt Family Member 3A (Wnt3a). These 3D structures are maintained either in synthetic hydrogels²⁶ or in biomaterials that mimic the basal lamina propria, for instance, Thermal-crosslinking Basal Extracellular Matrix (TBEM), and are further supplemented with growth factors that mimic the surrounding niche, most notably Epithelial Growth Factor (EGF), the Bone Morphogenetic Protein (BMP)-inhibitor Noggin, and an Lgr5-ligand and Wnt-agonist R-Spondin1²⁷. Under these conditions, organoids maintain epithelial apico-basal polarity and recapitulate the crypt-villi structure of the intestinal epithelium with budding stem cell crypts that terminally differentiate into absorptive and secretory cells in the center of the organoid, which then shed into the internal pseudolumen by anoikis²⁸. Although intestinal organoids alone have been hugely advantageous as reductionist models of epithelial development and dynamics in isolation^29,30, they hold tremendous future potential for understanding how these behaviors are regulated, influenced, or even disrupted by the immune compartment.

In the following protocol, a method of co-culture between murine small intestinal organoids and lamina propria derived ILC1s is described, which was recently used to identify how this population unexpectedly decreases intestinal signatures of inflammation and instead contributes to increased epithelial proliferation via TGF-β in this system⁸.

Protocol

All experiments must be completed in accordance and compliance with all relevant regulatory and institutional guidelines for animal use. Ethical approval for the study described in the following article and video was acquired in accordance and compliance with all relevant regulatory and institutional guidelines for animal use.

All mice were culled by cervical dislocation according to the standard ethical procedure, conducted by trained individuals. Before organ and tissue harvesting, slicing of the femoral artery or decapitation was conducted (as appropriate to the protocol in hand) as confirmatory assessments of death. Animals were housed under specific-pathogen-free conditions (unless stated otherwise) at an accredited commercial unit and King's College London animal unit in accordance with the UK Animals (Scientific Procedures) Act 1986 (UK Home Office Project License (PPL:70/7869 to September 2018; P9720273E from September 2018).

1. Establishing murine small intestinal organoids

NOTE: This section of the protocol describes the generation of intestinal organoids from the murine small intestine. Crypts are first isolated from tissue, resuspended in TBEM, and then incubated with media containing EGF, Noggin, and R-Spondin (ENR). Establishing murine small intestinal organoids has also been well-described elsewhere^24,25,27.

1. Place TBEM on ice to thaw (40 µL/well) and put a standard tissue-culture treated (refers to plates with a hydrophilic and negatively charged surface) 24-well plate in a 37 °C incubator to pre-warm.
   NOTE: 500 µL of TBEM will thaw in approximately 2-4 h; do not leave it at room temperature.
2. Prepare ~4 mL or 550 mL per well of ENR medium by adding EGF, Noggin, and R-Spondin to basal medium (Table 1) and place in a 37 °C incubator.
3. Cull a 6-12-week-old animal and dissect out the small intestine using forceps and microdissection scissors. Place it in 15 mL of Phosphate-Buffered Saline (PBS) on ice.
4. Using the stomach and caecum as points of orientation, isolate the desired region of the small intestine (duodenum, jejunum, or ileum). In this protocol, the ileum is isolated.
5. Submerge the intestine in PBS on a 10 cm² Petri dish on ice. Using forceps, remove fat gently, but completely, from the intestine.
6. Cut the tissue longitudinally using microdissection scissors (preferably with rounded tips to avoid perforation). Keeping the intestine submerged in PBS, shake to remove fecal matter, and maintain the mucous side up to track tissue orientation.
7. Transfer tissue to the dry lid of a Petri dish on ice, placing the intestine mucous/villus side up. Holding one end of the intestine with forceps, gently scrape away mucus using the angled edge of a clean glass slide.
8. Fill the plate with ice-cold PBS and rinse the tissue.
9. Pre-coat a 50 mL tube with PBS2 (PBS + 2% FCS; see Table 1) to prevent adhesion of the tissue to the plastic and add 10 mL of ice-cold PBS to this tube. Cut the intestine into small (~2 mm) segments and transfer them into the 50 mL tube pre-coated with PBS2.
10. Using a 25 mL pipette pre-coated with PBS2, pipette the segments up and down 5-10 times to clean the tissue fragments.
11. Allow the segments to settle for 15-30 s, remove and discard the PBS supernatant, and repeat steps 1.10 and 1.11 until the solution is clear (approximately 3-4 times).
12. Allow the segments to settle and discard the PBS supernatant. Add 30 mL of ice-cold crypt isolation buffer (Table 1).
13. Place the tubes on a horizontal roller or rocker at 30-60 rpm (or gentle) for 30 min at 4 °C. Do not incubate at faster speeds, higher temperatures, or longer duration as the crypts will prematurely dislodge and become single cells with lower viability and yield.
    NOTE: From this stage onward, all procedures should be conducted in an aseptic environment using sterile materials and reagents.
14. Allow intestinal segments to settle at the base of the 50 mL tube. Remove crypt isolation buffer without disrupting the settled segments and transfer it into a 50 mL tube on ice.
    NOTE: If the crypt isolation process was too rigorous, crypts may be seen in this fraction. It can therefore be kept as a reserve but will optimally be discarded at the end of the protocol.
15. Add 20 mL of ice-cold PBS to intestinal segments. Using a 25 mL pipette pre-coated with PBS2, pipette intestinal segments up and down 5-8 times.
16. Let the segments settle for 30 s. Pre-coat a 50 mL tube and 25 mL pipette with PBS2. Using this pipette, collect the supernatant (fraction 1) into the pre-coated 50 mL tube and place the tube on ice.
    NOTE: This fraction serves to rinse away the remaining crypt isolation buffer. However, it also serves as an additional quality control step; if pipetting was too vigorous, crypts will be abundant in this rinse fraction, and if it was too gentle, there will be very little debris in this fraction. Optimally, fraction 1 is discarded at the end of the protocol, but it can be used to make organoids if issues arise with fractions 2-4 obtained below.
17. Repeat steps 1.15 and 1.16 but add 10 mL of ice-cold PBS rather than 20 mL and place the supernatant in a fresh 50 mL tube pre-coated with PBS2 (fraction 2).
18. Repeat step 1.17 twice more but pool the resulting supernatant with fraction 2 (into the same 50 mL tube from step 1.17) to obtain fractions 2-4. This single 50 mL tube should contain the dislodged crypts in 30 mL of ice-cold PBS.
19. Remove a 50 µL aliquot from the pooled 2-4 fractions and place it onto a coverslip. Assess for the presence of epithelial crypts using an inverted light microscope.
    NOTE: If crypts are not present, repeat steps 1.17 and 1.18 using more force during pipetting until crypts are released. Ensure that crypts were not prematurely dislodged in the crypt isolation buffer from step 1.14 or in fraction 1 from step 1.16.
20. Pre-coat a 100 µm strainer and 50 mL tube with PBS2. Pass the pooled crypt fractions through this strainer and into the pre-coated 50 mL tube.
21. Spin down strained crypt fractions at 300 x g for 5 min at 4 °C.
22. Discard the supernatant and resuspend the crypts in 10 mL of ice-cold Advanced DMEM/F12. Move the crypts to a 15 mL tube.
23. Spin down crypts at 210 x g for 5 min at 4 °C to remove single cells and lymphocytes.
24. Resuspend crypts in 1 mL of ice-cold Advanced DMEM/F12.
25. Remove a 10 µL aliquot and place it onto a coverslip. Using an inverted light microscope, count the number of crypts to determine crypt concentration.
26. Calculate the volume required for ~400 and ~1,500 crypts. Pre-coat a p1000 tip with PBS2 and use this tip to pipette the required volumes (containing ~400 and ~1,500 crypts) into separate 1.5 mL tubes.
27. Spin down the crypts at 300 x g for 5 min at 4 °C.
28. Remove as much supernatant as possible, and then resuspend the crypts in 80 µL of TBEM placed on ice (pre-cool pipette tips at 4 °C to prevent matrix gelation, which occurs at 12 °C).
29. Remove the 24-well plate, placed in step 1.1, from the incubator. Gently pipette 40 µL of crypts into the center of a well in a slow, circular motion to form a flat but 3D dome structure, or into three separate small domes.
    NOTE: The viability of the organoids is the lowest in the center of the dome where nutrients and gases permeate less effectively³¹; thus, maximizing the surface area exposed to media while maintaining clear 3D structures is critical.
30. Repeat step 1.29 to create a total of two wells with a density of ~200 crypts per well and two more wells with a density of ~750 crypts per well. Directly place the plate in an incubator (37 °C and 5% CO₂) for 15-20 min, taking care to not disrupt the viscous but still liquid matrix domes.
31. Add 550 µL of pre-warmed ENR media per well (from step 1.2) and incubate for 24 h at 37 °C and 5% CO_2. At this stage, it is suggested to supplement ENR medium with 5 µM of Wnt agonist (CHIR 99021) and 5 µM Rho Kinase inhibitor (Y-27632) for the first 24 h of culture post isolation to improve the yield and efficiency of organoid generation of crypts freshly isolated from tissue.
    NOTE: Crypts should close to form rounded structures within 24 h, and crypt buds should appear within 2-4 days (Figure 1A).
32. Replace with fresh ENR medium at the end of the incubation in step 1.31, and then every ~2 days or when the phenol pH indicator in the culture medium turns pale orange but before it turns yellow.

2. Maintenance of murine small intestinal organoids

NOTE: This section of the protocol describes the maintenance and passaging of murine small intestinal organoids. Organoids are first harvested, and then mechanically disrupted using a bent p1000 tip. This process breaks large organoids consisting of numerous crypts into multiple smaller fragments for expansion and releases dead cells that have accumulated in the pseudolumen. Murine small intestinal organoid maintenance has also been well-described elsewhere^24,25,27. All procedures should be conducted in an aseptic environment using sterile materials and reagents. Passage or expand organoids once every 4-5 days, before bursting of the organoids occurs from substantial accumulation of debris in the organoid lumen. Organoids can be passaged at a ratio of 1:2-1:3 depending on organoid density, which will optimally be between 100-200 organoids per well.

1. As in steps 1.1 and 1.2, thaw TBEM on ice (40 µL/well) and prepare ENR medium (550 µL per well). Place the medium and standard tissue culture treated 24-well plate in a 37 °C incubator.
2. Remove the plate containing organoids from the incubator. Discard the media from the well to be passaged.
3. Add 500 µL of ice-cold Advanced DMEM/F12 to the well. Using a p1000 tip (pre-coated with media to avoid organoid sticking to the tip interior), harvest the organoids into a 15 mL tube.
4. Rinse the bottom of the well with 250-300 µL of ice-cold Advanced DMEM/F12 ensuring no organoids remain in the well, and pool into the 15 mL tube containing harvested organoids from step 2.3. Repeat steps 2.3 and 2.4 when passaging multiple wells and pool the organoids from multiple wells into the same 15 mL tube.
5. Spin down organoids at 300 x g for 3 min at 4 °C.
6. Upon centrifugation, ensure that there are four visible fractions: a base fraction with healthy organoids, a central and clear matrix fraction, a matrix fraction containing single and dead cells, and an upper supernatant layer containing single and dead cells. Remove the supernatant, the debris fraction, and as much of the clear matrix fragment as possible without disrupting the pellet of organoids.
7. Gently bend the tip of a p1000 pipette tip (approximately 2-5 mm bend). Pre-coat this tip in PBS2 or Advanced DMEM/F12. Using this tip, pipette the organoids up and down 10-20 times to mechanically disrupt the organoids and the remaining matrix.
8. Spin down at 210 x g for 3 min at 4 °C.
9. As in step 2.6, remove the supernatant, the debris fraction, and as much of the clear matrix fragment as possible without disrupting the pellet of dissociated crypts. If healthy crypts or small organoids have not formed a clear pellet, centrifuge again at 300 x g for 3 min at 4 °C.
10. Calculate the required volume of TBEM (80-120 µL per well of harvested organoids depending on whether a 1:2 or 1:3 passage ratio is used).
11. Resuspend the pellet in the calculated volume of TBEM and apply 40 µL per well to the pre-warmed 24-well plate to form a dome. Directly place the plate in the incubator (37 °C; 5% CO₂) for 15-20 min.
12. Add 550 µL of ENR medium per well and incubate at 37 °C and 5% CO₂. Check cultures for organoid formation after 24 h_. Replace with fresh ENR medium every 2-3 days post passage.

3. Isolation of small intestinal lamina propria innate lymphoid cells

NOTE: This section of the protocol describes the isolation of ILC1 from the murine small intestine lamina propria of the RORγt^GFP reporter mice. This involves epithelial cell removal, tissue digestion, density gradient separation of lymphocytes, and ILC1 isolation via fluorescence-activated cell sorting (FACS). FACS isolation following the gating strategy in Figure 2 requires the extracellular staining mastermix (Table 4), with the additional staining controls described in Table 2 and Table 3 for the machine (Table 2) and gating (Table 3) set up. The RORγt^GFP reporter mice are used to isolate live, pure ILC1 and gate out RORγt^GFP+ ILC3. Tissue processing for isolation of lamina propria lymphocytes has also been well-described elsewhere³².

1. Tissue processing
   NOTE: Tissue from individual biological animal replicates is kept separate in this protocol. These samples should be appropriately labeled and kept on ice whenever possible. All reagents except for digestion enzymes should be allowed to reach room temperature to ensure rapid tissue digestion.
   1. Place TBEM on ice to thaw (40 µL/well). Keep a standard tissue-culture treated 24-well plate in a 37 °C incubator to pre-warm.
   2. Prepare fresh epithelial removal buffer and digestion buffer (see Table 1). Prepare isotonic low viscosity density gradient medium (LVDGM) (90% LVDGM and 10% 10x PBS), 40% isotonic LVDGM, and 80% isotonic LVDGM in neutralizing buffer (Table 1).
   3. Cull a 6-12-week-old animal, dissect out the small intestine using forceps and microdissection scissors, and place the intestine in 15 mL of PBS on ice.
   4. Submerge the intestine in PBS on a 10 cm² Petri dish on ice and use forceps to remove fat gently but completely from the intestine.
   5. Remove Peyer's patches (~5-10; running in a line opposite to the line of fat tissue) using microdissection scissors to deplete lymphoid tissue inducer cells and B-cells.
      NOTE: Peyer's patches are absent in Rag^-/- and other lineage-depleted animals. Unlike air bubbles, Peyer's patches will remain in place if nudged.
   6. Cut the tissue longitudinally using microdissection scissors (preferably with rounded tips to avoid perforation). Keeping the intestine submerged in PBS, shake to remove fecal matter.
   7. Cut the tissue into 2-4 cm length pieces and transfer them into a 50 mL tube.
   8. Add approximately 20-40 mL of ice-cold PBS and shake vigorously for 5-15 s to remove mucous and debris.
   9. Discard the contents of the tube onto a fresh Petri dish and replace intestinal fragments into the same 50 mL tube using forceps.
   10. Repeat steps 3.1.8 and 3.1.9 3-4 times, or until the PBS is clear.
2. Epithelium removal
   1. Place intestinal fragments into a fresh Petri dish on ice. Using forceps, pick up intestinal segments and remove excess liquid by tapping on a dry surface.
   2. Cut the tissue into ~0.75-1 cm pieces and transfer them into a fresh 50 mL tube. Add 5-7 mL of epithelial removal buffer (Table 1). Vortex the tube and ensure that all the intestinal segments are submerged in the buffer.
   3. Incubate the samples at 37 °C with rocking (100-150 rpm) for 12-15 min. Vortex the tube for 20-30 s.
   4. Repeat steps 3.2.1-3.2.3 once more, discarding the cloudy supernatant (fraction contains epithelial and intra-epithelial cells) and replacing it with fresh epithelial removal buffer.
3. Digestion of the tissue
   1. Tip contents onto a fresh Petri dish. Using forceps, pick up the intestinal segments and tap on a dry surface to discard excess liquid. Place the segments into a fresh Petri dish on ice.
   2. Cluster the segments into the center of the Petri dish into a dense mass. Using either two scalpels or sharp scissors, finely shred the tissue until it reaches a viscous consistency that could pass through a p1000 pipette tip.
   3. Using tweezers, place the minced tissue into a clean 50 mL tube. Rinse the Petri dish with 1-2 mL of digestion buffer (Table 1) to collect any remaining tissue and pool it into the 50 mL tube with the shredded tissue.
   4. Add 5-7 mL of digestion buffer to the shredded tissue and ensure that all tissue is collected at the bottom of the tube.
   5. Incubate the samples at 37 °C, with medium rocking (~100-150 rpm) for 15 min. Vortex the tube for 20-30 s every 5 min to aid digestion.
   6. While the samples are incubating, place a 40 µm cell strainer on top of a fresh 50 mL tube for each sample. Coat the filters with 1-2 mL of neutralizing buffer (Table 1).
   7. Vortex the samples for 20-30 s after the incubation has finished.
   8. Filter the partially digested tissue through the coated 40 µm filter into the 50 mL tube containing neutralizing buffer.
   9. Use tweezers to collect undigested tissue from the 40 µm filter and place it back into the 50 mL tube that the digestion was performed in, for the second round of digestion. Remove filters and rinse them with digestion buffer to dislodge any undigested tissue adhering to the filter.
   10. Once as much undigested tissue as possible has been removed from the filter and placed back into the 50 mL tube that digestion was performed in, rinse the filter with 1-2 mL neutralizing buffer over the 50 mL tube containing the filtered supernatant.
   11. Add 20-25 mL of neutralizing buffer to the filtered supernatant and place it on ice to maintain the viability of isolated cells during the second round of tissue digestion.
   12. Add another 5-10 mL of digestion buffer to the undigested tissue and repeat steps 3.3.5-3.3.10, ensuring to filter the digested tissue into the same tube as used in step 3.3.8 (the same filter can also be reused). Rinse the filter with neutralizing buffer until the 50 mL line is reached, and then discard any remaining undigested tissue.
4. Lymphocyte isolation by density gradient
   1. Spin the collected supernatants for each biological replicate at 500 x g for 10 min.
   2. During step 3.4.1, add 5 mL of 80% isotonic LVDGM to a 15 mL tube for each biological replicate.
   3. Following centrifugation, discard the supernatant from the filtered, digested tissue. Resuspend the pellet in 10 mL of 40% isotonic LVDGM, ensuring that the pellet is well homogenized, and no large chunks remain.
   4. Setting the pipette aid to its slowest setting, tilt the 15 mL tube containing 80% isotonic LVDGM, and very gently overlay the 10 mL of cell suspension in 40% isotonic LVDGM ensuring that the cell suspension does not mix with the 80% fraction.
      NOTE: If the fractions should ever accidentally mix, rapidly distribute lymphocytes that reached the 80% fraction between two 50 mL tubes filled with PBS, and centrifuge for 10 min at 500 x g. Then, discard the supernatant, and repeat steps 3.4.3-3.4.4.
   5. Spin the tubes at 900 x g and 20 °C, with the acceleration and deacceleration set to the lowest setting to ensure that the fractions are not disrupted. Centrifuge for 20 min (not including break time).
      NOTE: All the procedures from this step onward should be conducted in an aseptic tissue culture hood using sterile materials and reagents.
   6. Pre-coat a 50 mL tube for each sample with PBS2 and add 45 mL of ice-cold PBS.
   7. After centrifugation, gently remove the upper debris layer from the 15 mL tube. Coat a p1000 tip with PBS2 and use it to collect the lymphocyte-laden interphase between the 40%-80% gradients into the 50 mL tube containing 45 mL of ice-cold PBS.
   8. Spin the tubes at 300 x g for 5 min at 4 °C.
   9. Discard the supernatant and resuspend the pellet in 500 µL of FACS buffer (PBS, 2% FCS, 0.5 mM EDTA, and 10 mM HEPES; see Table 1). Pre-coat a 40 µm filter with FACS buffer and use it to filter the cell suspension into a flow tube. Rinse the 50 mL tube with an additional 500 µL of FACS buffer and filter into the same flow tube.
   10. Remove a 10 µL aliquot from the resulting 1 mL of cell suspension. Count the lymphocyte yield using a cell counter or hemocytometer and calculate the cell concentration for each sample.
5. Sample preparation for sorting - extracellular staining
   NOTE: The antibodies used in the extracellular staining mastermix, as well as the reagents to prepare the fixable viability dye and Fc block solution, are used in concentrations enough for up to 5 x 10⁶ cells per 100 µL. Volumes should be adjusted accordingly. For FACS machine compensation to sort ILC1, single-color controls for each of the fluorophores in the extracellular staining mastermix are required. For the antibodies, compensation beads that contain a positive antibody binding bead population and a negative antibody non-binding bead population can be used. For the ultraviolet (UV) fixable viability dye single color control, compensation beads that contain amine-reactive (for positive signal) and non-reactive (for negative signal) populations can be used. It is not recommended to use cells from the RORγt^GFP reporter mice for the UV fixable viability dye single color control, as these cells will contain a GFP⁺ signal.
   1. Remove a small aliquot of ~10 µL from each sample and pool it into a separate flow tube containing 250 µL of FACS buffer. Place the tube on ice. This will be used for the unstained control.
   2. Add 1 mL of PBS to the remaining volume in the sample tubes. Centrifuge at 300-400 x g for 3-5 min.
   3. Prepare 200 µL of UV fixable viability dye solution per sample. Dilute UV fixable viability dye (resuspended in DMSO following the manufacturer's instructions) 1:500 in PBS (or following the manufacturer's instructions).
   4. Discard the supernatant and resuspend the samples in 200 µL of UV fixable viability dye solution. Vortex the tubes and incubate in the dark at 4 °C for 10-15 min.
   5. Add 2 mL of FACS buffer to the tubes to quench the UV fixable viability dye, vortex for 10 s, and centrifuge the tubes at 300-400 x g for 3-5 min at 4 °C. Discard the supernatant from the centrifuged tubes.
   6. Prepare 200 µL of Fc block solution per sample (0.25 mg/mL of Fc block in the FACS buffer).
   7. Add 200 µL of Fc block solution to each of the samples from step 3.5.5, vortex for 10 s, and incubate in the dark at 4 °C for 10 min.
   8. Add 500 µL of FACS buffer to a fresh flow tube. Remove a 2-5 µL aliquot from each sample and pool it into the tube for the fluorescence minus one (FMO) controls.
   9. Vortex the FMO tube for 10 s and distribute evenly (100 µL per tube) into fresh flow tubes labeled for each of the FMO controls (Lineage cocktail FMO, CD127 FMO, KLRG1 FMO, NKp46 FMO, and NK1.1 FMO; see Table 3). Add all the antibodies except the antibody of interest to each tube (as described in Table 3) and vortex for 10 s. Place the FMO controls aside in the dark at 4 °C.
   10. Centrifuge the samples (not FMO controls) at 300-400 x g for 3-5 min at 4 °C.
   11. Prepare extracellular staining mastermix (200 µL per sample) with the final antibody dilutions as described in Table 4. Adjust the staining volume for the appropriate cell concentration (up to 5 x 10⁶ cells per 100 µL) depending on the cell counts from step 3.4.10.
   12. Add extracellular staining mastermix to the samples, vortex for 10 s, and place them aside in the dark at 4 °C.
   13. Prepare a fresh single color control for each antibody intended to be used in the gating strategy (Figure 2). Vortex antibody compensation beads for 20 s, add 1 drop of beads to a flow tube, stain with 0.5 µL of antibody (as shown in Table 2, or following the manufacturer's instructions), and vortex for 10 s.
   14. Prepare a UV fixable viability dye single color control using an amine-reactive compensation bead kit. Vortex amine-reactive compensation beads for 20 s, add 1 µL of Live/Dead UV dye (as shown in Table 2, or following the manufacturer's instructions) and vortex for 10 s.
   15. Incubate the samples, FMOs, and single color controls in the dark for 30 min at 4 °C.
   16. During this time, prepare tubes to collect the sorted cells. Add 400-500 µL of 10% FBS in Advanced DMEM/F12 to 1.5 mL tubes for each sample.
   17. After the incubation, add 2 mL of PBS2 to each of the samples, FMO controls, and single color controls.
   18. Centrifuge the samples, FMO controls, and single color controls at 300-400 x g for 3-5 min at 4 °C.
   19. Discard the supernatant from all the centrifuged tubes. Resuspend the samples and FMO controls in 250 µL of FACS buffer and vortex for 10 s.
   20. Resuspend the single color controls in 250 µL of PBS2. Add 1 drop of amine non-reactive beads to UV fixable viability dye single color control only. Vortex all controls for 10 s.
   21. The cells are now ready to be sorted. Keep samples, FMO controls, and single color controls on ice in the dark whenever possible to improve cell viability and prevent loss of signal from photobleaching.

4. Co-culture of small intestinal organoids with innate lymphoid cells

NOTE: This section describes the co-culture of sorted murine small intestinal ILC1 (isolated following the protocol in section 3) with murine small intestinal organoids (described in sections 1 and 2). Organoids should optimally be used 1-2 days following passage. Co-culture involves harvesting the organoids, adding the appropriate number of ILC1, centrifuging to pellet organoids and ILC1 together, and resuspending in TBEM. Complete this section as soon as possible once ILC1 have been isolated. All the procedures should be conducted in an aseptic environment using sterile materials and reagents.

1. Ensure that TBEM from step 3.1.1 has been thawed.
2. Harvest 1-2-day old organoids as described in steps 2.3 and 2.4.
3. Resuspend the organoid pellet in 1 mL of cold ice-cold Advanced DMEM/F12. Do not bend the p1000 tip as done when passaging the organoids, as the intention here is to maintain the organoid structure for co-culture. If required, place the organoids in separate PBS2 coated 1.5 mL tubes on ice for a short period to be distributed between different co-culture conditions.
   NOTE: Only begin preparation of organoids once ILC samples are ready for co-culture; work on ice and rapidly as soon as organoids have been harvested to minimize epithelial cell death.
4. Pre-coat one 1.5 mL tube with PBS2 per ILC replicate sample. Distribute ~100-200 organoids (approximately 1 well of a 24-well plate) into each tube.
5. Pre-coat a p1000 tip in PBS2 and use it to assess the volume of ILC1s after FACS purification (250-300 µL + sorted volume). Determine the concentration of ILC1 per mL using the number of cells recorded from the sort and determine the required volume.
6. Using the same PBS2 coated p1000 tip, add no fewer than 500 ILC1s to the PBS2 coated 1.5 mL tube containing the organoids. If the number of ILC1s yielded from the sort is less than 500, add the organoid directly to the tube containing the original sort to minimize the loss of cells from the transfer.
7. Spin down ILC1 and organoids at 300 x g for 5 min at 4 °C. If possible, avoid small diameter tabletop centrifuges, as these will pool the cells along the edge of the tube interior as opposed to creating a pellet at the tip of the tube. Since cell numbers are very low, it is imperative to minimize the loss of cells during these steps.
8. Slowly and gently remove as much of the supernatant as possible without disturbing the pellet (which may not be visible by eye, especially in no-organoid ILC-only controls). Place the sample on ice.
9. Resuspend the cold pellet in 30 µL per well of TBEM. While holding the tube on a cold surface (e.g., a small box of ice placed in the tissue culture hood), mix the organoids and ILC at least 10-15 times to ensure even distribution. Triturate frequently but gently, to avoid damaging the organoids and the formation of bubbles.
10. Apply 30 µL per well of ILC-organoids in TBEM to a pre-warmed 24- or 48-well plate to form a single dome. Directly place the plate in the incubator (37 °C and 5% CO₂) for 10-20 min.
    NOTE: It is strongly recommended that ILC-only and organoid-only controls are set up for downstream analysis. ILC1 are rare, but GFP^- ILC2 can be substituted for immunofluorescence or FSC/SSC flow cytometry controls where scientifically appropriate.
11. Add 550 µL (per well) of complete ILC1 medium (ENR + IL-2 + IL-7 + IL-15 + 2-Mercaptoethanol; see Table 1) with any desired experimental cytokines or blocking antibodies and incubate at 37 °C and 5% CO₂ for 24 h.
12. After 24 h, gently remove the plate from the incubator and allow the plate to sit in a tissue culture hood for 1 min to ensure the lymphocytes are settled.
13. Remove 200-250 µL of media and place it into an empty well of a 24-well plate. Check this supernatant with an inverted microscope to ensure that no lymphocytes were removed.
    1. If the supernatant is clear, add 300 µL of fresh ILC1 medium to the co-culture in the original well. If lymphocytes are present in the supernatant, centrifuge at 300-400 x g for 3-5 min at 4 °C and resuspend in 300 µL of fresh ILC1 medium and add this to the remaining 200-250 µL of media in the original well. Ensure to replenish media every 1-2 days or when media becomes pale-orange/yellow.
       NOTE: Do not allow media to evaporate sufficiently that the tip of the matrix dome breaks the surface of the media. Always ensure that matrix is completely submerged. Excess evaporation can be avoided by using central wells in the tissue culture plates and adding ~600 µL of PBS to the surrounding wells. Co-cultures with adult ILC are stable for 1-4 days, after which point the organoids that were seeded without major disruption will rupture and reseed as new crypts. If analyzing the epithelium, it is recommended that cultures are analyzed within 1-4 days of establishing co-cultures.
    2. Perform downstream analysis using immunofluorescence, flow cytometry, or FACS purification of target populations into lysis buffer for gene expression analysis by single cell or bulk RNA-seq or RT-qPCR, as described in reference⁸.

Representative Results

When successfully completed, freshly isolated crypts should form budding crypt structures within 2-4 days (Figure 1A). Healthy and robust organoid cultures should be actively growing and can be passaged and expanded as detailed in the protocol.

This protocol describes the isolation of small intestinal ILC1 from the RORγt^GFP murine transgenic reporter line, which allows the isolation of live ILC1 by FACS (Figure 2). Using the protocol outlined here, the expected ILC1 count range is 350-3,500 isolated cells.

After being seeded with organoids, co-cultures can be visualized by immunocytochemistry (Figure 3A-B). ILCs and epithelial cells can also be analyzed by flow cytometry, as demonstrated in Figure 3C. ILC1 upregulate epithelial CD44, as characterized by flow cytometry (Figure 4A-B) and immunocytochemistry (Figure 4C). Specifically, ILC1 induce expression of the CD44 v6 splice variant in organoids, as demonstrated by RT-qPCR (Figure 4D).

Table 1: Media and buffer compositions. Please click here to download this Table.

Table 2: Single color controls. Composition of single color controls to isolate small intestine lamina propria ILC1 using the gating strategy defined in Figure 2. Details of antibodies used can be found in the Table of Materials. Please click here to download this Table.

Table 3: Fluorescence minus one (FMO) mastermixes. Composition of FMO mastermixes for Lineage cocktail FMO, CD127 FMO, KLRG1 FMO, NKp46 FMO, and NK1.1 FMO. FMO mastermixes contain all of the antibodies used except the antibody of interest and are used to stain a sample aliquot. Lineage cocktail is defined as CD19, CD3e, CD5, Ly-6G/Ly-6C. Details of antibodies used can be found in the Table of Materials. Please click here to download this Table.

Table 4: Extracellular staining mastermix. The concentrations are adjusted for staining up to 5 x 10⁶ cells in 200 µL of FACS buffer. Details of antibodies used can be found in the Table of Materials. Please click here to download this Table.

Figure 1: Murine small intestinal organoids. Representative image of (A) successfully generated small intestinal organoids 2-3 days post passage and (B) unsuccessful culture. Scale bar: 100 µm. Please click here to view a larger version of this figure.

Figure 2: Gating strategy to isolate ILC1s from the small intestine lamina propria of the transgenic RORγt^GFP reporter mice. Representative flow cytometric plot of ILC1 isolation from the small intestine lamina propria of the transgenic RORγt^GFP reporter mice by FACS. ILC1s are defined as live, CD45⁺, Lin^- (CD3, CD5, CD19, Ly6C), CD127⁺, KLRG1^-, RORγt^-, NKp46⁺, and NK1.1⁺. Representative from N = >50 mice. Please click here to view a larger version of this figure.

Figure 3: Organoid and ILC1 co-cultures. Bright-field images (A), confocal microscopy images (B), and FACS plots (C) of small intestinal organoids (SIO) cultured alone (top) or with ILC1s (bottom) (representative of experiments with ILC1s from N = 3 mice). (B) Staining with CD45 illustrates ILC1s and Zonula occludens protein 1 (ZO-1) marks epithelial cells in organoids. Scale bars: 50 µm. (C) Previously gated on single, live cells. Epithelial cellular adhesion molecule (EpCAM) marks intestinal epithelial cells (IEC) in organoids and CD45 marks ILC1s. The figure is adapted from reference⁸. Please click here to view a larger version of this figure.

Figure 4: ILC1s in co-culture with small intestinal organoids drive upregulation of CD44 in intestinal epithelial cells. (A) A representative cytometric plot of CD44 expression in epithelial cellular adhesion molecule (EpCAM) positive epithelial cells (live, CD45^-, EpCAM⁺) from small intestinal organoids (SIO) cultured alone (left) or with ILC1s for 4 days (right). (B) Flow quantification of CD44 expression in intestinal epithelial cells (IEC) (ILC1s from N = 5 mice). (C) Representative confocal microscopy image of CD44 localization in d4 SIO alone (left) or co-cultured with ILC1s (right) (representative of N = 3 mice). Scale bars: 50 µm. (D) RT-qPCR with exon-specific primers for CD44 splice variants v4, and v6 (N = 3). This figure is adapted from reference⁸. Please click here to view a larger version of this figure.

Discussion

This protocol describes the methods for establishing murine small intestine organoids, isolating rare ILC1 by minimizing the loss of lymphocytes during the intestinal dissociation protocol, and establishing co-cultures between these two compartments. There are many steps to this protocol, and while some are specific to ILC1s, this approach can be applied to other intestinal immune cell types, and co-culture setups can be modularly adapted to suit individual research questions. Several critical steps (that are recommended to be not deviated from), as well as troubleshooting guidelines for the more technically challenging elements of this protocol, are highlighted here.

The use of murine small intestinal organoids from single Lgr5⁺-eGFP intestinal stem cells is becoming increasingly well established^33,34; however, in this protocol, it is suggested to isolate intact, entire crypts of Lieberkuehn from CD45.1 animals. Not only do intact crypts recover more rapidly than single Lgr5⁺ cells, but the use of CD45.1 animals without a GFP reporter ensures that no cross-contaminating CD45.2⁺ ILC are analyzed from the organoid co-cultures and is compatible with the use of immune cells containing a GFP-based reporter. In the authors' experience, no mesenchymal or immune cells carry over after 1-3 passages of the organoids. The use of CD45.2 or other animals for establishing organoids is therefore entirely acceptable. During organoid establishment, if crypts are not present at step 1.1.19, more rigorous manual shaking may be necessary to dislodge the intact crypts. Environmental factors such as ambient room temperature (e.g., whether the procedure is carried out in the summer or the winter) may add some variability to incubation timings during dissociation. The seeding density of crypts will impact initial organoid formation yield; it is therefore recommended to seed a minimum of two different densities to ensure success (e.g., 200-750 suggested here, but this range can be adapted based on individual needs).

Once established, intestinal organoid cultures are heterogenous both between lines established from the same mouse strain, along the gastrointestinal tract (e.g., duodenum versus ileum), and even within the same well of organoid cultures^35,36. Although this protocol was found to be robust over many different batches of organoids, this heterogeneity could contribute to data variability. It is good practice to be consistent with organoid maintenance (passaging and media changes) to reduce technical noise from phenotypically irrelevant data. This includes being consistent with the pre-seeding passaging timeline and with the force used to dissociate organoids. It is also recommended to use the same basal matrix for experiments being compared, and for experiments to be performed using biological replicates of organoids derived from different animals (when financially and technically feasible) to ensure that results are robustly reproducible.

In establishing co-cultures, the ratio of immune to epithelial cells is a critical consideration that will require optimization based on the research questions. If the impact of epithelial cells on ILC is being interrogated, the number of organoids seeded will need to be sufficient to saturate all ILC. Conversely, when assessing the impact of ILC on the epithelium, different ILC/epithelial ratios may result in different phenotypic outputs, reflecting differential states of ILC subset enrichment in the mucosa. ILC1 viability is well maintained in culture, and the population will undergo mild expansion, with ~500 ILC1 and 100 small intestinal organoids (SIO) undergoing a 2-3-fold expansion on average. However, this yield will be impacted by additional treatments, with TGF-neutralization improving and p38-inhibition decreasing the absolute number of ILC1 after co-culture⁸. Any unanticipated loss of more than 50% of the seeded ILC1 numbers may be the result of either an imbalanced ratio of ILC1 to SIO (increase number of seeded crypts), SIO contamination (ensure antibiotic cocktail is functioning and test the supernatant for mycoplasma), or of quality issues in the cytokine stocks, with ILC1 being particularly sensitive to a lack of IL-2 or IL-15. Co-culture supernatants from concentrated 96- or 48-well plates have been successfully used for ELISAs. When dissociating co-cultures, it is recommended to incubate cells with DNase after a 20 min gentle trypsin replacement or perform EDTA-based dissociation to single cells to prevent cell clumping from damaged epithelial cells.

A strength of this protocol is that it balances reductionist culture conditions with complex cell types. However, the behaviors of other ILC subsets in these cultures may be dependent on factors not present in this particular protocol. For example, in the Lindemans' protocol used for ILC3 co-cultures, IL-23 was additionally supplemented into co-culture media to support ILC3 maintenance and activation⁸. IL-15 was found to be particularly important in the maintenance of ILC1 in the co-culture system described in this protocol, which was congruent with previous reports of ILC1 requiring this cytokine for homeostasis, though not development⁶. To activate ILC, or to maintain ILC2s, the growth medium may require further optimization. Moreover, other cellular compartments in the intestine, aside from the epithelium, regulate ILCs. For example, intestinal neurons are known to modulate ILC2s partly through the activity of secreted neuropeptides³⁷. Microbial factors also influence ILC phenotype³⁸. This limitation could be overcome through the addition of these elements, e.g., cytokines, peptides, or microbial factors, into the co-culture system. This could even allow for the interrogation of the interaction between ILCs and multiple cellular compartments in a reductionist setting. Following this logic, it is critical that anti-biotic/anti-mycotic reagents are added and frequently replenished to organoid media prior to establishing co-cultures. It is also critical that all cultures are performed in aseptic environments because any culture contamination (e.g., fungal growth or mycoplasma) would likely activate the antigen non-specific ILCs, creating significant phenotypes that may not be present in non-contaminated cultures. For this reason, withdrawal of anti-biotic/-mycotic reagents is not recommended, even in the co-cultures, as they were not found to cause an adverse impact on the epithelium or ILCs.

This method provides a unique way to characterize signaling modules between ILCs and the intestinal epithelium, allowing for the biology of both compartments to be investigated. In comparison to other in vitro methods consisting of a single cell type, the system presented here is more comparable to in vivo physiology and enables multiple potential signaling mechanisms between epithelial cells and ILCs to be interrogated. Other methods of in vitro ILC culture predominantly rely on stromal feeder cell lines, such as OP9 or OP9-DL1³⁹. This line is derived from newborn mouse calvaria, which is not representative of the intestinal environment. While these have provided the gold standard for maintaining ILCs in vitro to date, they suffer substantial limitations in their application to understanding the impact of ILC on the epithelium.

The co-culture protocol described here between murine small intestinal organoids and lamina propria derived ILCs has significant research applications. This system of co-culture has already been used to determine the role of ILC1 derived TGF-β in the expansion of CD44⁺ epithelial crypts⁸, which contributes to the understanding of epithelial dynamics in inflammatory bowel disease. These studies contribute to an increasing body of literature that underpins the critical importance of epithelial-ILC signaling in intestinal homeostasis and inflammation³.

Materials

Name	Company	Catalog Number	Comments
Reagents
2-Mercaptoethanol	Gibco	21985023
Anti-mouse CD45 (BV510)	BioLegend	103137
Anti-mouse NK1.1 (PE)	Thermo Fisher Scientific	12-5941-83
B-27 Supplement (50X), serum free	Gibco	17504044
CD127 Monoclonal Antibody (APC)	Thermo Fisher Scientific	17-1271-82
CD19 Monoclonal Antibody (eFluor 450)	Thermo Fisher Scientific	48-0193-82
CD3e Monoclonal Antibody (eFluor 450)	Thermo Fisher Scientific	48-0051-82
CD5 Monoclonal Antibody (eFluor 450)	Thermo Fisher Scientific	48-0031-82
CHIR99021	Tocris	4423/10
COLLAGENASE D, 500MG	Merck	11088866001
Cultrex HA- RSpondin1-Fc HEK293T Cells	Cell line was used to harvest conditioned RSpondin1 supernatant, the cell line and Materials Transfer Agreement was provided by the Board of Trustees of the Lelands Stanford Junior University (Calvin Kuo, MD,PhD, Stanford University)
DISPASE II (NEUTRAL PROTEASE, GRADE II)	Merck	4942078001
DMEM/F12 (1:1) (1X) Dulbecco's Modified Eagle Medium Nutrient Mixture F-12 (Advanced DMEM/F12)	Gibco	11320033
DNASE I, GRADE II	Merck	10104159001
Dulbecco's Modified Eagle Medium (1X)	Gibco	21969-035
Ethilenediamine Tetraacetate Acid	Thermo Fisher Scientific	BP2482-100
FC block	2B Scientific	BE0307
Fetal Bovine Serum, qualified, hear inactivated	Gibco	10500064
GlutaMAX (100X)	Gibco	3050-038
Hanks' Balanced Salt Solution (10X)	Gibco	14065056
HBSS (1X)	Gibco	12549069
HEK-293T- mNoggin-Fc Cells	Cell line was used to harvest conditioned Noggin supernatant, cell line acquired through Materials Transfer Agreement with the Hubrecth Institute, Uppsalalaan8, 3584 CT Utrecht, The Netherlands, and is based on the publication by Farin, Van Es, and Clevers Gastroenterology (2012).
HEPES Buffer Solution (1M)	Gibco	15630-056
KLRG1 Monoclonal Antibody (PerCP eFluor-710)	Thermo Fisher Scientific	46-5893-82
Live/Dead Fixable Blue Dead Cell Stain Kit, for UV excitation	Thermo Fisher Scientific	L23105
Ly-6G/Ly-6C Monoclonal Antibody (eFluor 450)	Thermo Fisher Scientific	48-5931-82
Matrigel Growth Factor Reduced Basement Membrane Matrix, Phenol Red-free, LDEV-free	Corning	356231
N-2 Supplement (100X)	Gibco	17502048
N-acetylcysteine (500mM)	Merck	A9165
NKp46 Monoclonal Antibody (PE Cyanine7)	Thermo Fisher	25-3351-82
PBS (1 X) 7.2 pH	Thermo Fisher Scientific	12549079
PBS (10X)	Gibco	70013032
Percoll	Cytiva	17089101
Recombinant Human EGF, Animal-Free Protein	R&D Systems	AFL236
Recombinant Human IL-15 GMP Protein, CF	R&D Systems	247-GMP
Recombinant Human IL-2 (carrier free)	BioLegend	589106
Recombinant Mouse IL-7 (carrier free)	R&D Systems	407-ML-005/CF
UltraComp eBeads	Thermo Fisher Scientific	01-2222-42
Y-27632 dihydrochloride (ROCK inhibitor)	Bio-techne	1254
Plastics
50 mL tube	Falcon	10788561
1.5 mL tube	Eppendorf	30121023
10 mL pippette	StarLab	E4860-0010
15 mL tube	Falcon	11507411
25 mL pippette	StarLab	E4860-0025
p10 pippette tips	StarLab	S1121-3810-C
p1000 pippette tips	StarLab	I1026-7810
p200 pippette tips	StarLab	E1011-0921
Standard tissue culture treated 24-well plate	Falcon	353047
Equipment
Centrifuge	Eppendorf	5810 R
CO2 and temperature controled incubator	Eppendorf	Galaxy 170 R/S
Flow Assisted Cellular Sorter	BD equipment	FACS Aria II
Heated shaker	Stuart Equipment	SI500
Ice box	-	-
Inverted light microscope	Thermo Fisher Scientific	EVOS XL Core Imaging System (AMEX1000)
p10 pippette	Eppendorf	3124000016
p1000 pippette	Eppendorf	3124000063
p200 pippette	Eppendorf	3124000032
Pippette gun	Eppendorf	4430000018
Wet ice	-	-