In Vitro Resident Memory CD8 T Cell Differentiation Using Epithelial Organoid-T Cell Co-culture System

CD8 resident memory T cells (TRMs) represent a unique population of memory CD8 T cells that are retained permanently in the peripheral tissues without routinely circulating through blood and lymphoid tissues. These TRM cells are strategically positioned in barrier organs, including skin, lungs, intestines, and the reproductive tract, where they act as sentinels against pathogens and nascent tumors^1,2. Upon pathogen detection, CD8 TRM cells rapidly release cytotoxic granules to directly eliminate infected cells while simultaneously secreting inflammatory cytokines that alert surrounding tissue cells and recruit circulating immune effectors to contain pathogen dissemination^3,4. The effectiveness of CD8 TRMs in antipathogen immunity as well as cancer immunotherapy has already been demonstrated^5,6. Accordingly, positioning an abundant quantity of functionally competent CD8 TRM in barrier mucosal tissues is a crucial goal for vaccines and immunotherapeutic strategies.

The formation and maintenance of CD8 TRM in peripheral organs is critically dependent on local tissue-specific programming. These local environmental cues include unique tissue-derived cytokines and metabolic agents, as well as intercellular interactions that collectively enforce the tissue-specific TRM identity^7,8,9. This tissue programming allows TRM to adapt to tissue-specific demands in the execution of their immunosurveillance duties. However, our understanding of the nature of these interactions is incomplete, owing to the lack of robust model systems that faithfully capture these processes. Small animal models, while being very informative, often suffer from a highly interconnected, complex web of interactions that thwart rapid high-throughput studies.

Here, we describe an in vitro alternative to the mouse model that will allow investigation into the molecular details of epithelial cell-derived cues behind CD8 TRM differentiation. By co-culturing preformed vaginal epithelial organoids (VEOs) with activated CD8 T cells, we generated CD8 TRM in vitro. A detailed description of the CD8 T cell activation and co-culture process will allow wider adoption of this process across tissues, leading to a greater understanding of cellular interactions shaping CD8 T cell memory in frontline tissues.

All procedures involving animals were reviewed and approved by the Institutional Animal Care and Use Committee of Brown University (Protocol number 22-10-0001).

1. Basic setup and preparation of an activation plate

1. One day prior to beginning the experiment, prepare a non-treated tissue culture plate for seeding with naïve CD8 T cells. Prepare a solution of activating antibodies in sterile dPBS at a final concentration of 1 µg/mL for anti-CD3ε (2C11) and 0.167 µg/mL for B7-1 Fc. For a 24-well plate, dispense 600 µL of this solution per well. Incubate overnight at 4 °C until ready to use.
2. Prepare Lymphocyte Enrichment Buffer in sterile dPBS by adding 2% charcoal-stripped FBS and 1 mM EDTA to sterile dPBS. Store at 4 °C until ready to use.
3. Prepare ammonium-chloride-potassium red blood cell (ACK RBC) lysis buffer in ddH₂O by adding 10 mM potassium bicarbonate, 155 mM ammonium chloride, and 0.1 mM EDTA. Filter sterilize and keep at room temperature.
4. Prepare the appropriate amount of harvest media in RPMI by adding 5% charcoal-stripped FBS. Store at 4 °C until ready to use. For each mouse to be harvested, 50 mL of harvest media is recommended.
   NOTE: Penicillin/streptomycin and amphotericin B are optional, but a final concentration of 100 units/mL penicillin, 100 µg/mL streptomycin, and 0.2 µg/mL amphotericin B may help prevent bacterial and fungal contamination of T cell culture during tissue isolation if that presents an issue.
5. Prepare the T cell medium in RPMI by adding 10% charcoal-stripped FBS, 2 mM L-glutamine, 100 units/mL penicillin, 100 µg/mL streptomycin, 1x non-essential amino acids, and 1 mM sodium pyruvate. For a 24-well plate, use 2 mL of T cell medium per well and 50 mL of T cell medium for a full plate. Store at 4 °C until ready to use.
6. Prepare dissection tools (0.4 mm micro dissecting scissors and 0.8 mm fine curved micro dissecting serrated forceps) by autoclaving in a sterilization pouch or soaking in 70% ethanol.

2. Harvesting and dissociation of adult mouse secondary lymphoid organs

1. Fast cool a centrifuge to 4 °C with appropriate adapters to hold 15 mL conical tubes.
2. Prepare ice for transporting tissues and lymphocytes during harvest and incubation periods.
3. Prepare a sterile 15 mL conical tube with at least 5 mL of harvest media per mouse to be dissected.
4. Anesthetize and euthanize an adult CD8 T cell transgenic mouse (~8 weeks old) according to institutional protocol in a laminar flow hood.
5. Pin the limbs of the mouse down onto a clean polystyrene foam surface using any size needles, and spray the torso with 70% ethanol to prevent fur from obstructing the incision and sticking to tools.
6. Using clean micro dissecting scissors and forceps, make a vertical abdominal incision and separate the skin from the thin layer of connective tissue underneath. Pin the skin onto the polystyrene foam surface using any size needles, and cut through any remaining connective tissue to expose the organs within the torso.
7. Carefully locate and excise the spleen, removing excess fat using the curved forceps, and transfer it into the 15 mL conical tube containing harvest media. Leave the spleen on ice and isolate as many lymph nodes as possible (e.g., iliac and inguinal) to combine with the spleen in the same conical tube on ice.
   NOTE: The iliac lymph nodes are located on either side of the caudal vena cava in the lower abdominal region underneath the gastrointestinal tract, and the inguinal lymph nodes are located near the branching of the superficial epigastric vein on either side of the torso's lower abdominal skin. Isolate more lymph nodes to improve the yield of CD8 T cells following enrichment; however, isolating the spleen alone is generally sufficient to provide several million CD8 T cells.
8. Transfer the spleen and lymph nodes to a scored 60 mm dish and, using the flat end of a 3 mL syringe plunger, mechanically disassociate the spleen and lymph nodes by grinding the tissue until no large fragments remain.
9. Collect the suspension by passing it through a sterile 70 µm cell strainer into a fresh sterile 15 mL conical tube. Wash the 60 mm dish with 5 mL of harvest media to ensure complete transfer of disaggregated cells and filter into the same 15 mL conical tube.
10. Centrifuge the suspension at 584 × g for 5 min at 4 °C. Discard the supernatant and rack the tube by running the bottom quickly and forcefully across the grid of a centrifuge tube holder to loosen the pellet.
11. Lyse erythrocytes by resuspending the loosened pellet in 2 mL of ACK RBC lysis buffer. Vortex and incubate at room temperature for 2 min.
12. After 2 min, neutralize the lysis buffer by adding 2 mL of harvest media. Aspirate the suspension using a 5 mL serological pipette and filter through a sterile 70 µm filter back into the same 15 mL conical tube.
    NOTE: This filtration is important to ensure the removal of all clumps.
13. Centrifuge the suspension at 584 × g for 5 min at 4 °C. Discard the supernatant and resuspend the pellet in 950 µL of Lymphocyte Enrichment Buffer.

3. CD8 T cell enrichment and activation

1. Proceed to enrich naïve CD8 T cells from lymphocytes using a commercially available isolation kit. Follow the manufacturer's protocol.
2. Once CD8 T cells have been enriched, transfer the enriched suspension to a new sterile 15 mL conical tube and fill the volume up to 10 mL with harvest media.
3. Count using a hemocytometer.
   NOTE: From one adult mouse, it is possible to isolate at least 5-10 million naïve CD8 T cells at >95% purity.
4. Plate at least 500,000 cells in 2 mL of T cell medium per well of a 24-well plate. Mix T cells thoroughly prior to plating to ensure a homogeneous single-cell suspension. Supplement the T cell medium with a final concentration of 100 units/mL IL-2, 2.5 ng/mL IL-12, and 55 µM beta-mercaptoethanol.
5. Remove the previously antibody-coated tissue culture plate from the 4 °C and carefully aspirate the coating solution without touching the bottom of the wells.
6. Immediately plate T cells. Incubate at 37 °C with 5% CO₂ for 2 days.

4. Removing activation signals and resting activated CD8 T cells

1. Observe activated wells under a light microscope. After 2 days of activation, look for small clusters of T cells throughout the wells. Larger clusters are indicative of more activation. If there are no or few clusters, incubate the plate at 37 °C with 5% CO₂ for 1 more day before proceeding with resting.
2. Prepare ice for storing T cell suspension during incubation periods and when not in use.
3. Resuspend T cells using a P1000 pipette by pipetting up and down several times in each corner of the well. When T cells have lifted, transfer the contents of the well to a sterile 50 mL conical tube and count.
   NOTE: Washing each well with 1 mL of dPBS is optional, but pooling dPBS wash with the T cell suspension may increase T cell yield.
4. Re-plate the T cells at the same concentration as in Step 3.4, 250,000 cells/mL at 2 mL per well. Prepare the appropriate amount of T cell medium supplemented with 100 units/mL IL-2 and 55 µM beta-mercaptoethanol.
5. Centrifuge the suspension at 584 × g for 5 min at 4 °C, discard supernatant, and resuspend the pellet in the appropriate volume of T cell medium prepared in the previous step for the number of wells to be plated. Mix thoroughly to ensure a homogeneous single-cell suspension.
6. Plate 2 mL of T cell suspension (500,000 cells/well) in a new non-treated tissue culture plate without antibody coating. Incubate at 37 °C with 5% CO₂ for 2 days.

5. Infecting preformed murine vaginal epithelial organoids with virus

NOTE: We have used both Herpes Simplex virus-2 (HSV-2) and Lymphocytic choriomeningitis virus (LCMV) to perform viral infection of the organoids. Both are considered Biosafety Level-2 (BSL-2) agents. Carry out all procedures involving infectious agents inside a certified BSL-2 biosafety cabinet. Clean work areas with 70% ethanol both prior to and following use, and discard all materials according to institutional biosafety guidelines. For this protocol, we established VEOs from vaginal epithelial cells harvested from the vaginal epithelium of C57BL/6 female mice as described before¹⁰.

1. Plate VEOs at a density of 10,000 cells in 20 µL of basement membrane extract (BME) per well of a 24-well plate 7-10 days prior to infection.
2. One day prior to co-culture, set aside one well of VEOs to be trypsinized and counted to obtain an accurate multiplicity of infection (MOI). Prewarm 0.25% Trypsin-EDTA in a 37 °C water bath for 5 min.
   NOTE: 0.25% Trypsin-EDTA will not be used during infection; it is only necessary for disassociating the VEOs during MOI calculations. Coating plasticware (i.e., pipette tips, conical tubes) with 2.5% BSA in dPBS is optional at this step but may help reduce organoid and cell loss for more accurate counting.
3. Aspirate cell culture media from the designated well and wash with dPBS.
4. Dispense 600 µL of prewarmed 0.25% Trypsin-EDTA into the well and resuspend thoroughly with a BSA-coated P1000 pipette to mechanically disrupt the BME. Transfer the contents of the well into a BSA-coated sterile 15 mL conical tube.
5. Incubate in a 37 °C water bath for 5 min. After 5 min, add 10 mL 10% FBS in RPMI to quench the trypsinization.
6. Centrifuge the suspension at 500 x g for 5 min at 4 °C. Discard the supernatant and resuspend using a BSA-coated pipette tip in 100-300 µL of T cell-organoid culture medium or plain DMEM/F12.
   NOTE: The volume depends on the size of the pellet. If the pellet is barely visible, resuspend in a lower volume.
7. Count using a hemocytometer. Calculate the appropriate amount of virus needed to achieve 0.1 MOI using the known viral titer and cell density per well of VEOs according to the following equation:
   
8. Aspirate cell culture media from the VEO wells to be infected and wash with 500 µL of prewarmed (37 °C) dPBS. Incubate the wells with dPBS for 5 min at 37 °C.
9. After 5 min, discard the dPBS and add 500 µL of organoid media per well containing enough viral particles to achieve a MOI of 0.1. Incubate at 37 °C with 5% CO₂ for 1 h with gentle manual swirling of the tissue culture plate every 20-30 min.
10. After the 1 h incubation, aspirate the viral organoid media and wash with 500 µL of prewarmed (37 °C) dPBS. Aspirate dPBS and replace with fresh organoid media.

6. Co-culturing CD8 T cells with virus-infected, preformed murine vaginal epithelial organoids

1. Prior to co-culture (ideally 4-24 h in advance), transfer the BME from the -80 °C to 4 °C for thawing. Anticipate the appropriate amount of BME to thaw based on the co-culture experimental design: plan to plate 8 μL of BME per well of a 96-well plate and 20 μL of BME per well of a 24-well plate.
   NOTE: BME is viscous, and a longer thawing period at 4 °C will reduce the viscosity and improve maneuverability.
2. Prepare 2.5% BSA in dPBS and filter sterilize. Use this BSA solution to coat every plastic surface that comes into contact with the VEOs (i.e., pipette tips, conical tubes).
   NOTE: This coating will prevent VEOs from sticking to the plastic surface and being lost during handling.
3. Check VEOs under a light microscope for size and overall health. Confirm that VEOs are 7-10 days post culture and on average ~100 microns in diameter. Healthy VEOs will present few or no single cells and feature a uniform, round morphology.
   NOTE: Replate VEOs 1:1 in co-culture if grown in a 24-well plate and replating in a 24-well plate (i.e., one well of VEOs will reseed one well of T cell-organoid co-culture). If replating in a 96-well plate from a 24-well plate, replate VEOs 1:2 (i.e., one well of VEOs will reseed two wells of T cell-organoid co-culture).
4. Place the necessary number of sterile 96- or 24-well tissue culture plates to be used for the co-culture in the 37 °C incubator.
   NOTE: A warmed plate will expedite BME adhesion and solidification.
5. Prepare T cell-organoid culture medium in DMEM/F12 by adding 2% B-27 supplement, 2 mM L-glutamine, 100 units/mL penicillin, 100 µg/mL streptomycin, 1x Non-essential amino acids, 1 mM sodium pyruvate, and 0.2 µg/mL amphotericin B. Store in 4 °C until ready to use.
6. Aspirate cell culture media from VEO wells and wash each well with 500 µL of cold (2-8 °C) sterile dPBS.
7. Aspirate dPBS and add 200-500 µL of cold (2-8 °C) organoid harvesting solution (at least 10x volume of organoid BME droplet). Pipette up and down with a BSA-coated P200 pipette to mechanically disrupt the BME.
8. Incubate plate with gentle shaking (~100 rpm) on an orbital shaker at 4 °C for 20 min. If a cold orbital shaker is unavailable, incubate the plate on ice with manual shaking every 5 min. If BME has not depolymerized after 20 min, lengthen the incubation time as necessary.
9. Transfer the contents of the wells using a BSA-coated pipette tip into a BSA-coated sterile 15 mL or 50 mL conical tube on ice. Centrifuge the suspension at 500 × g for 5 min at 4 °C.
10. Discard the supernatant and wash VEOs with 5-10 mL of cold (2-8 °C) dPBS (at least 10x volume of combined organoid BME droplets). Centrifuge the suspension at 500 × g for 5 min at 4 °C.
11. Discard the dPBS and resuspend the pellet in 1 mL of T cell-organoid culture medium using a BSA-coated P1000 pipette tip. Leave the VEO suspension on ice until T cells are ready to be co-cultured.
12. Begin collecting the T cells for co-culture. Check resting T cells under a light microscope for size and overall health. Without activating signals, T cells will slow their proliferation but should still have formed some clusters. Confirm that there are no cell debris or fragments, and T cells have an enlarged morphology. Resuspend T cells using a P1000 pipette by pipetting up and down several times in each corner of the well and count the total number of live CD8 T cells.
    NOTE: Per well of a 96-well plate, seed 100,000 T cells for each co-culture in 8 µL of BME. Per well of a 24-well plate, seed 200,000 T cells for each co-culture in 20 µL of BME.
13. Centrifuge the suspension at 584 × g for 5 min at 4 °C. Discard the supernatant and resuspend the pellet in 10 mL of T cell-organoid culture medium. Combine T cells with VEOs by transferring an appropriate number of T cells for the wells to be plated into the conical tube containing the VEOs.
14. Centrifuge at 584 × g for 5 min at 4 °C. Discard the supernatant and resuspend VEOs and T cells in the appropriate amount of BME. Mix thoroughly using a BSA-coated pipette tip, and plate onto prewarmed tissue culture plate, 20 µL per well of a 24-well plate or 8 µL per well of a 96-well plate.
    NOTE: Keep BME on ice or in a cooling block at 4 °C during plating. Placing pipette tips in the 4 °C fridge for 30 min prior to plating may help prevent premature solidification of the BME.
15. Place the plate upside down in the 37 °C incubator and wait 30 min for complete BME adhesion and solidification.
16. During this incubation period, prepare the complete T cell-organoid culture medium by adding 100 ng/mL murine EGF, 55 µM beta-mercaptoethanol, and 10 units/mL murine IL-2 to the premade T cell-organoid culture medium.
17. After the incubation period, carefully add 500 µL of complete T cell-organoid culture medium per well of a 24-well plate or 250 µL of complete T cell-organoid culture medium per well of a 96-well plate. Dispense media against the sides of the well to avoid disrupting the BME droplet. Change the media every 2 days until harvested for analysis.

7. Collecting co-cultured CD8 T cells for flow cytometry analysis

1. Once co-cultures have reached at least day 7 in culture, harvest for analysis. If T cells do not express TRM-like markers in co-culture by day 7, optimize the harvest date between day 7-14, depending on phenotypic acquisition and VEO and T cell viability.
2. Prepare fluorescent activated cell sorting (FACS) buffer for flow cytometry staining and analysis by adding 0.2% w/v bovine serum albumin and 0.1% w/v sodium azide to dPBS.
   NOTE: Sodium azide is an acutely toxic chemical. Follow institutional recommendations in handling and disposal of all waste in compliance with institutional biosafety regulations.
3. Prepare a fluorescent antibody cocktail in FACS buffer for surface staining, depending on which markers are desired for analysis. For each well, account for at least 50 µL of antibody cocktail.
4. Prepare a fluorescent viability dye in dPBS. For each well, account for at least 50 µL of viability dye.
5. Carefully remove the media in each well and thoroughly resuspend in cold (2-8 °C) FACS buffer, pipetting up and down 10-20x to mechanically disrupt the BME. Transfer the contents of each well to a labelled 96-well round-bottom plate.
6. Centrifuge at 859 × g for 2 min at 4 °C. Lower the deceleration ramp for all centrifugation steps onward to prevent sample loss from pellet remixing.
7. Remove the supernatant by inverting the plate and expelling the liquid with a single, firm downward motion. Resuspend in 50 µL of antibody cocktail. Pipette up and down at least 5x using a multichannel pipette to ensure thorough resuspension of the cell pellet.
8. Incubate for 30 min at 4 °C or room temperature as necessary, protected from light. At the end of the incubation period, add 150 µL of dPBS to each well using a multichannel pipette and centrifuge at 859 × g for 2 min at 4 °C.
9. Decant the supernatant by turning the plate upside down and flicking it downward once in a continuous motion. Resuspend in 50 µL of viability dye. Pipette up and down at least 5x using a multichannel pipette to ensure thorough resuspension of the cell pellet.
10. Incubate for 15 min at room temperature, protected from light. At the end of the incubation period, add 150 µL of FACS buffer to each well using a multichannel pipette and centrifuge at 859 × g for 2 min at 4 °C.
11. Decant the supernatant as before, and resuspend in 150 µL of FACS buffer if samples will be run on a flow cytometer immediately or fix in 150 µL of 0.5% paraformaldehyde and leave at 4 °C. Pipette up and down at least 5x using a multichannel pipette to ensure thorough resuspension of the cell pellet. Filter fixed samples to prevent instrument clogging and run on a flow cytometer within 12-36 h to prevent loss of fluorescence.
    NOTE: Paraformaldehyde is considered a hazardous chemical. Follow institutional biosafety regulations in handling and disposal.

Microscopic visualization of CD8 T cell-infected organoid co-culture
Preformed mouse VEOs were infected with LCMV and co-cultured with LCMV-specific CD8 T cells. We have also performed these co-cultures with HSV-2 infection model to establish the versatility of the protocol but are showing microscopic data for LCMV model and have included T cell differentiation data in HSV-2 model below. The LCMV expresses yellow fluorescent protein (YFP), so the infected cells can be visualized using routine epifluorescence microscopy. For this, 7-day-old VEOs were infected with LCMV-YFP, 24 h before CD8 T cell incubation (Figure 1A). Separately, LCMV-specific transgenic P14 CD8 T cells were activated using the activation and resting protocol described above. On the day of co-culture, the infected organoids were released from the BME matrix and mixed with the appropriate number of activated CD8 T cells mentioned earlier. This mixed population was resuspended in a new BME matrix and was plated in a 96-well plate for imaging (Figure 1B). Abundant viral replication in organoids can be noted as early as 4 h after co-culture, marked by green fluorescence. However, the CD8 T cell accumulation near infected organoids became prominent later. We also noted loss of organoid integrity at later time points presumably caused by CD8 T cell-induced cell death.

Phenotypic characterization of co-cultured CD8 T cells
We performed flow cytometry-based phenotyping of CD8 T cells that were incubated with infected VEOs at different times post-co-culture; the gating strategy used for analysis is shown in Supplemental Figure S1. For this, HSV-2 infected VEOs were co-cultured with HSV-2-specific transgenic gBT-I CD8 T cells for 7-30 days. A group of activated gBT-I CD8 T cells embedded in BME matrix in the absence of VEOs was also maintained as a negative control (Figure 2A, top row). CD8 T cells isolated from these conditions were assessed for their adoption of phenotypic features of TRM including upregulation of CD69+ CD103+ (double positive) markers, suggestive of an epithelial CD8 TRM population (Figure 2A,B). The co-cultured cells also exhibited downregulation of CD62L and upregulation of PD-1 and CD38, markers associated with a residency signature11. Importantly, the residency features were more pronounced on day 13 and 30 compared to day 7 post co-culture, indicating a gradual differentiation process under the influence of epithelial signals. CD8 T cells co-cultured with uninfected VEOs failed to acquire the canonical CD69⁺CD103⁺ TRM phenotype, indicating that epithelial co-culture alone is insufficient to drive TRM differentiation in the absence of antigenic stimulation (Supplemental Figure S2). Comparatively, CD8 T cells maintained alone were very few on day 7 and showed an aberrant phenotype bearing markers CD103+, CD69-, CD62L- and low for PD-1 and CD38 (Figure 2A). To ensure these in vitro-generated TRM are functional, we also stimulated these cells with 1 µg/mL antigenic peptide for 3 h and tested their ability to secrete cytokines. The CD69+ CD103+ CD8 T cells significantly upregulated intracellular interferon-gamma and IL-2 upon stimulation compared to untreated CD8 T cells (Figure 3A,B).

Figure 1: Localization of antiviral CD8 T cells near infected organoids. (A) Schematics of the co-culture process. (B) Seven-day-old mouse VEOs were infected with LCMV-YFP and were cultured with P14 CD8 T cells. Imaging was performed at 4 h, 41 h, 75 h, and 114 h post co-culture. Representative images from three separate experimental repeats are shown. The rectangular box indicates the region shown in the enlarged image beneath. White arrow indicates clustering of CD8 T cells next to an infected group of VEOs, and black arrows show further aggregation of CD8 T cells and consequent reduction in YFP signal (in green). Scale bars = 500 µm. Please click here to view a larger version of this figure.

Figure 2: Acquisition of CD8 TRM phenotypes among co-cultured CD8 T cells. (A) Flow cytometry plots showing expression of indicated markers on CD8 T cells isolated from day 7, day 13, and day 30 post co-culture. Top row shows CD8 T cells cultured in the absence of organoids. The cells are gated on live CD45.1+ CD8 T cells (CD45.1 marks gBT-I HSV-2 specific CD8 T cells). (B) Bar graph showing percentage of CD69+ CD103+ TRM cells at different days post-co-culture indicating gradual upregulation of TRM markers. Data are representative of three repeats with n= 3-4/condition. Bars indicate mean ± SEM. Ordinary one-way ANOVA with Tukey's multiple comparison test. **p < 0.01, ***p < 0.001 and ****p < 0.0001. Please click here to view a larger version of this figure.

Figure 3: In vitro differentiated CD8 TRM functionality. Flow cytometry plots (A) and bar graph (B) showing upregulation of IFN-gamma and IL-2 on CD8 T cells upon antigenic stimulation for 3 h with 1 µg/mL antigenic peptide. The cells are gated on live CD69+ CD103+ CD8 T cells. Data are representative of two repeats with n= 4-10/condition. Bars indicate mean ± SEM. Student's T test. ****p < 0.0001. Please click here to view a larger version of this figure.

Supplemental Figure S1: Gating scheme to identify CD8 T cells in coculture. Representative flow cytometry gating strategy used to identify CD8 T cells from organoid-T cell cocultures. Lymphocytes were first gated based on forward and side scatter, followed by singlet discrimination (FSC-A vs FSC-H), live cell gating, and selection of CD45.1⁺ cells. Expression of CD69 and CD103 was assessed on gated live CD45.1⁺ CD8 T cells. Please click here to download this File.

Supplemental Figure S2: Uninfected VEOs failed to generate CD69+ CD103+ CD8 TRM. Representative flow cytometry plots showing CD69 and CD103 expression on CD8 T cells isolated from day 7 co-cultures with either LCMV-infected or uninfected VEOs (in the absence of supplemental antigenic peptide). Cells are gated on live CD45.1⁺ CD8 T cells. Please click here to download this File.

CD8 TRMs rely on local environmental cues for differentiation and maintenance. These cues instill tissue-specific programming that allows CD8 TRMs to adapt to their surroundings and perform their immunosurveillance duties as needed. Unfortunately, TRMs failed to survive in vitro once extracted from the tissue, and this has limited their detailed analysis^12,13. In this protocol, we have provided a method by which CD8 T cells can be co-cultured with epithelial cells to develop into TRMs with the phenotypic features similar to those of in vivo TRMs. This method can be used to study TRMs in the long term, as well as be easily manipulated to accommodate other tissues, viruses, and genetic modifications.

The most critical element of this protocol is related to cell health. T cells must be proliferating and alive prior to co-culture. T cells that have not been sufficiently activated and provided sustained IL-2 will still not survive long-term culture unless provided with other survival signals, including IL-7 and IL-15; although we have seen that VEOs themselves can partially support T cell viability¹⁰. If T cells are not above 80% viability at the time of co-culture, the co-culture should not move forward, and freshly isolated T cells should be activated for another attempt. VEOs must also be of sufficient size (~100 microns) and healthy (>90% viable) at the time of co-culture. It is important not to use preformed VEOs that are too old, as this will impact VEO longevity during co-culture, especially for long-term studies. Conversely, it is also important to avoid using preformed VEOs that are too young and small. Small VEOs are indicative of insufficient growth and/or differentiation, which will impact the communication of signals being received by the T cells for their differentiation. It is recommended that preformed VEOs between 7-10 days post culture be used to balance differentiation and size with VEO maintenance during co-culture.

Some of the issues that may arise during this protocol include maintaining VEO integrity, T cell differentiation, and BME plating. VEOs can lose longevity after several passages, depending on the depletion of stem cells. In such cases, it is recommended that earlier passages of VEOs be used. If preformed VEOs are struggling to survive until the experimental endpoint, disaggregated VEOs may be used instead of preformed VEOs. We have previously demonstrated that disaggregated VEOs co-cultured with T cells can also give rise to TRMs¹⁰; however, disaggregated VEOs experience reduced organoid growth due to the removal of the TGF-β signaling inhibitor in the co-culture media that is needed to support TRM differentiation. Viral infection of VEOs may also drastically reduce VEO longevity in co-culture due to T cell killing and effector function. If intact VEOs are desired for end analysis, VEOs may be pulsed with antigenic peptides instead. We have demonstrated that VEOs pulsed with antigen and co-cultured with antigen-specific T cells can last up to 14 days in co-culture¹⁰. Besides cell health, the BME matrix may also be prone to depolymerizing during longer culturing periods. Some of the issues that may contribute to this include compromised integrity during plating and changing of media. It is important that bubbles be minimized to preserve the BME structure while ensuring complete resuspension of the VEOs and T cells within the BME mixture for homogeneous plating. BME should be worked with quickly when exposed to room temperature and should not be disturbed during regular media changes by focusing the media deposition on the outer edges of the well. If the BME dome containing VEOs and T cells detach from the bottom of the wells during co-culture, the co-cultures are still viable, but special care must be taken when changing media to avoid aspirating the dome. In these instances, it is recommended that a light microscope be used during the process to locate the dome prior to removing media.

As a reductionist model, this co-culturing method is inherently limited by the lack of other cell types that would be otherwise present in vivo. Here, we are specifically focused on the interactions between epithelial cells and CD8 T cells; however, CD8 T cells in barrier tissues interact with and are influenced by other immune and non-immune cells, such as dendritic cells, macrophages, and fibroblasts. There have been several recent developments in building assembloid systems that incorporate more stromal and immune cells to better mimic the native tissue^14,15. To fully harness the potential of these in vitro models, several engineering and biological challenges must be systematically addressed and overcome. One such challenge is the static nature of our system as it does not reflect the dynamic processes in vivo, such as continuous nutrient and gas exchange and waste removal. More complex organoid models with continuous perfusion and vascularized compartments are being attempted to address these issues^16,17.

Despite these limitations, this VEO-T cell co-culture model nonetheless provides a high-throughput tool for studying CD8 TRM differentiation that has been hitherto limited by the difficulty of TRM extraction and maintenance in vitro. An improved in vitro model reduces the burden of animal study and provides an ethical and accurate cell culture alternative. In addition, T cells can be genetically modified to study the transcriptional regulation important for TRM development, function, and maintenance. Although this protocol focuses on murine vaginal organoids and T cells, it can potentially be adapted for other mucosal and non-mucosal epithelial organoids as well as human tissues, for many of which organoids have already been developed. Co-culturing organoids and T cells presents a powerful tool with which to study the signals important for CD8 TRM differentiation that can, in the future, enable more mechanistic studies into the immune control of infections and tumorigenesis, as well as the means to eliminate TRM in autoimmune pathologies¹⁸.