Mucosal Mixed Lymphocyte Reaction Assay Using Intestinal Lymphocytes as a Biomarker for Intestinal Transplant Tolerance Development

Abstract

Background.

Intestinal transplantation (ITx) has the highest rate of rejection among solid organ grafts. We aimed to study the pathophysiology of rejection after ITx but lacked a tool for assessing cellular responses within the graft. Therefore, we developed a novel mixed lymphocyte reaction (MLR) assay to investigate immune responses in the graft.

Methods.

Intestinal samples were collected, decontaminated, and processed into single-cell suspensions from 9 swine and 2 patients that underwent ITx. Debris was removed using gradient centrifugation. The cells were plated with corresponding stimulator cells and incubated for six days before data acquisition and analysis.

Results.

Tolerant animals showed no anti-donor or anti-recipient responses in their graft mucosa but maintained strong anti-third-party responses, even after weaning immunosuppression. An animal with graft-versus-host disease displayed robust anti-recipient and anti-third-party responses but no anti-donor response. The animals with graft rejection maintained anti-donor responses at all timepoints. Finally, some tolerant animals developed “split tolerance,” with anti-donor responses in the peripheral blood but donor-specific hyporesponsiveness in the mucosal MLR, which Treg depletion suggested was due to local regulatory tolerance. When applied to human sample, this mucosal MLR reliably demonstrated self-tolerance with normal anti-third-party responsiveness.

Conclusion.

The novel mucosal MLR assay presented herein +/− CD25 depletion serves as a useful adjunct for assessing immune responses within the intestinal graft mucosa. This could help elucidate immune responses after ITx in future studies, including our own, and could represent a promising tool for studying ITx tolerance development, guiding immunosuppression strategies, and advancing personalized transplant medicine.

INTRODUCTION

Intestinal transplantation (ITx) is the only remaining definitive treatment for patients with intestinal failure who have failed parenteral nutrition. However, its clinical practice has been limited by high rates of rejection and graft-versus-host disease (GvHD), which requires more intense immunosuppression (ISP) regiments, ultimately leading to ISP-related complications.^1–3 Developing better therapies or achieving immune tolerance would make ITx a more acceptable and beneficial therapy.^4,5 Our laboratory is conducting research on this subject with promising results that will be reported in detail in a separate manuscript (Gunes et al., manuscript under revision in Transplantation, TPA-2024–1496) and have been reported elsewhere.⁶ As part of that work, we developed a novel mucosal mixed lymphocyte reaction (MLR) assay, which we describe herein, that allowed us to demonstrate the possibility of a previously undescribed form of “split tolerance” after ITx mediated by regulatory T cells (Tregs).

For ITx, as for other solid organ grafts, the development of transplant tolerance is evaluated in vitro using either MLR or cell-mediated lympholysis (CML) assays.^5,7,8 While these assays provide insights into the development of transplant tolerance and can be modified to investigate various types of responses (e.g., CD25 depletion to investigate Treg-mediated transplant tolerance)^9,10, they have been limited mostly to the evaluation of peripheral blood mononuclear cells (PBMCs). Therefore, it has not previously been possible to use these tools to assess immune responses within the graft tissue itself. In the case of ITx, developing a mucosal MLR assay with which we can characterize host-versus-graft and graft-versus-host responses +/− Treg depletion, has changed our understanding of how tolerance is induced in our preclinical model of ITx in ways that might have clinical significance.

We herein describe our novel modification of the MLR assay using intestinal mucosal lymphocytes to verify the development and investigate the various mechanisms of intestinal transplant tolerance. Since our assay can be performed using mucosa obtained by endoscopic biopsies, it can be performed non-invasively or without extra anesthesia or surgery events. Therefore, it permits direct immunological assessment of an intestinal allograft allows pre-clinical researchers to understand the local mucosal interactions and guide animal studies of ITx tolerance development, and clinical researchers might use these techniques in the future for making informed decisions on ISP weaning trials for ITx patients.

MATERIALS AND METHODS

Collection of Samples

All animal procedures were approved by the Columbia University Institutional Animal Care and Use Committee (IACUC) under the protocol numbers AC-AABI0555 and AC-AABU2660. All ethical guidelines were followed for the studies performed on animals. All procedures on humans were approved by the Columbia University Institutional Review Board (IRB) under the protocol numbers AAAV0919 and AAAJ5056. Written consents were obtained from each patient.

Porcine samples were collected at various time points from swine that underwent orthotopic ITx (n=9) across haplotype-match, partial MHC class II-match, and fully-mismatch disparities (Figure 1A, Table 1) using ISP based on our clinical regimen.¹¹ All of the animals had varying degrees of donor leukocyte replacement (52.15% - 99.9%) within the grafts at the time of the assays. Third-party blood was collected from an outbred Yucatan swine. Human samples were collected at the time of stoma closures from patients who previously underwent either intestinal (n=1) or multivisceral transplantation (n=1). Third-party blood for humans was acquired from the New York Blood Center (New York, NY).

Figure 1.

A) Representative images of the Sachs Miniature Swine Herd haplotypes (left) and the MHC disparity of experimental groups (right) are shown. Note that A and D haplotypes share a part of their MHC Class II alleles. B) Representative images of an intestinal sample before and after decontaminating and removing muscular and serosal layers. C) FSC-A and SSC-H plots of mucosal cells before and after debris removal using different methods. Notice that while straining through a 20μm filter and using a density-gradient medium removes some of the debris, the Debris Removal Solution successfully removes debris and has a very clear leukocyte population. D) Representative image of the cell culture plate using cells that underwent clean-up and did not undergo clean-up after six days of incubation. Notice the change in the color of the media in the not cleaned-up wells compared to cleaned-up wells.

Table 1.

Overview and Outcomes of Intestinal Transplants in Swine

		MHC Disparity Between Donor	Days Survived	Cause of Death	Weaned Off ISP	Peripheral Blood Tolerance	Mucosal Graft Tolerance	Split Tolerance
Group 1	Animal 1	Haplotype-Mismatched	23	Graft vs Host Disease	NO	YES	YES	NO
	Animal 2	Haplotype-Mismatched	34	>30% Weight Loss Recovering from a Rejection Episode	NO	NO	YES	YES
	Animal 3	Haplotype-Mismatched	56	Broken Femur	NO	YES	YES	NO
	Animal 4	Haplotype-Mismatched	63	>30% Weight Loss Due to High Stoma Output	NO	NO	YES	YES
	Animal 5	Class II Partial Match	174	End point	YES	NO	YES	YES
	Animal 6	Full Mismatch	176	Rejection	YES	NO	NO	N/A
	Animal 7	Haplotype-Mismatched	551	End-point	YES	YES	YES	NO
Group 2	Animal 8	Class I Single Allele Mismatch	252	End-point	YES	YES	YES	NO
	Animal 9	Class I Single Allele Mismatch	203	End-point	YES	YES	YES	NO

Peripheral blood was collected in BD Vacutainer^® Heparin Tubes (McKesson, Irving, TX, Cat: 770811) or syringes with 100U heparin (McKesson, Irving, TX, Cat: 916433) per ml of blood. Graft specimens were collected in RPMI-1640 (Thermo Fisher, Waltham, MA, Cat: MT15040CV) supplemented with 50,000U/50mg of Pen-Strep (Sigma-Aldrich, St. Louis, MO, Cat: P0781) in 500ml centrifuge bottles (Sigma-Aldrich, St. Louis, MO, Cat: CLS431123). For surgical specimens, the distal and proximal ends were ligated using 2–0 silk ties (McKesson, Irving, TX, Cat: 220264) to prevent leakage of intestinal contents and contamination of the media during storage. The intestinal specimens were stored at 4°C for up to 24 hours before processing into single-cell suspension of lymphocytes.

Isolation of PBMC

PBMCs were isolated from swine and human whole blood to be used as responders and stimulators in the MLR assays. We performed PBMC isolation using our standard technique.¹²

Preparation and Decontamination of Intestinal Sample for Processing

The intestinal mucosa was first separated from the muscularis and serosal layers and washed thoroughly to remove stool and microbial contents (Fig. 1B). To do this, four 500ml centrifuge bottles were labeled, placed on a rack, and each filled with 250ml of HBSS. The intestinal sample was removed from the RPMI-1640 and placed in a 150mm petri dish (Fisher Scientific, Waltham, MA, Cat: 08-757-148) filled with 10ml of HBSS. The specimen was cut longitudinally on its mesenteric axis. The large pieces of stool were gently washed away with HBSS, and the intestinal graft was moved into another 150mm petri dish. Using scissors and forceps, the mucosa of the graft was separated from the muscular and serosal layer, cut into 5cm long strips, and placed in the first 500ml centrifuge bottle. The bottle was shaken vigorously for 30 seconds to maximize the gross removal of stool and microbial contents. The mucosa was removed from the first bottle and placed in the second one, and the process was repeated until it was washed 4 times (Fig. 1B).

Processing Intestinal Samples into Single-Cell Suspensions

The intestinal specimens were processed using a 3-step digestion protocol into intraepithelial lymphocytes (IEL) and lamina propria lymphocytes (LPL), the concept and procedures of which have been previously described for human intestinal graft tissues.¹³

Eliminating Debris

IELs and LPLs were separated from debris, which makes up around 90–95% of the pellet (data not shown), using Debris Removal Solution (Miltenyi Biotec, Gaithersburg, MD), as the debris decreases cell-to-cell contact during incubation and blocks the magnetic sorting columns and flow cytometry-based cell sorters (Fig. 1C).

The manufacturer’s guidelines were modified to allow for debris removal from larger volumes of tissues for our application. To remove the debris, the pellets were resuspended in 25ml of cold HBSS. 8ml of Debris Removal Solution was added, and the tubes were gently inverted. 15ml of cold HBSS was gently overlaid. The tubes were spun in a 4°C swing bucket centrifuge at 3000G for 10 minutes with minimal acceleration and no brake. After the spin, the floating debris and the debris removal solution were aspirated and discarded. The pellet was quenched and spun with HBSS to remove the entirety of the Debris Removal Solution. After the supernatant was discarded, the pellet was resuspended with HBSS and stored at 4°C until further use.

Magnetic CD25 Depletion and Sorting of Porcine Intestinal Lymphocytes and PBMC

CD25 depletion was performed on the porcine PBMC and mucosal responders to investigate the development of Treg-mediated tolerance.

To deplete CD25+ cells, anti-pig-CD25 (Clone: K231.3B2, Bio-Rad, Hercules, CA, Cat: MCA1736GA) antibody was conjugated to biotin using Biotin Conjugation Kit (Fast, Type A) - Lightning-Link^® (Abcam, Waltham, MA, Cat: ab201795) and titrated to optimal dilution. The cells were pelleted, incubated with the biotinylated anti-pig-CD25 antibody for 30 minutes, and quenched with MACS running buffer (Miltenyi Biotec, Gaithersburg, MD, Cat: 130-091-221). The cells were pelleted, incubated with Anti-Biotin MicroBeads UltraPure for 15 minutes (Miltenyi Biotec, Gaithersburg, MD, Cat: 130-105-637), and washed with MACS running buffer. The cells were applied to an LS Column (Miltenyi Biotec, Gaithersburg, MD, Cat: 130-042-401) and placed in a MidiMACS^™ Magnetic Separator (Miltenyi Biotec, Gaithersburg, MD, Cat: 130-042-301) according to manufacturer guidelines. Positively selected CD25+ cells were discarded, and negatively selected cells were used for CD25 depletion MLRs.

Mixed Lymphocyte Reaction (MLR) Assay

MLR assays using non-CD25-depleted and CD25-depleted peripheral and mucosal cells were simultaneously performed to check for systemic and local development of tolerance.

MLR media was prepared by adding 6% heat-inactivated porcine or human serum to AIM-V solution (Fisher Scientific, Waltham, MA, Cat: 12-055-091) for respective responder cell origin. The media was sterilely filtered using a 0.2μm vacuum filter (Fisher Scientific, Waltham, MA, Cat: 09-741-05).

Stimulator cells were stained with Violet Proliferation Dye 450 (VPD450, BD, Franklin Lakes, NJ, Cat: 562168) to distinguish them from responder cells during analysis. To stain the stimulator cells, the cells were resuspended at 30×10⁶ cells/ml. To prepare the working concentration of the VPD450 dye, 2.63 ml of DMSO (Sigma-Aldrich, St. Louis, MO, Cat: 34869) was added to the stock dye to create a 1mM dye solution. 1μl of the dye solution was added for every 30×10⁶ cell/ml at a working concentration of 0.1μM. The cells were incubated at 37°C for 12 minutes, quenched with MLR media, pelleted, and resuspended at 2×10⁶ cells/ml. Stimulator cells were irradiated at 35Gy (X-RAD320) to prevent proliferation before plating.

Responder cells were stained with CFSE (Fisher Scientific, Waltham, MA, Cat: 50-591-407) to track response and proliferation. To stain the responder cells, the responder cells were resuspended in HBSS at 1×10⁶ cells/ml. To prepare the working concentration of the CFSE dye, a vial of CFSE dye was warmed to room temperature, and 18μl of DMSO was added to reconstitute the dye. 15μl of the reconstituted CFSE dye was removed from the vial and added to 1500μl of HBSS to create a 50μM concentration of the CFSE dye. 10μl of diluted CFSE dye was added for every 1×10⁶ cell/ml at a working concentration of 0.5μM. The cells were incubated at room temperature for 7 minutes, quenched with MLR media, and placed on ice for 5 minutes to stop the CFSE reaction. The cells were pelleted and resuspended with MLR media at 1.5×10⁶ cells/ml for porcine PBMC responders and 2×10⁶ cells/ml for porcine and human mucosal responders. The reason for using fewer porcine PBMC responders per ml is due to the high background proliferation seen at crowded wells.

The assays were plated in triplicate on a 96-well round bottom plate (Fisher Scientific, Waltham, MA, Cat: 07-200-95) using 125μL of each stimulator and responder cell suspension (250μL of total well volume). Media and self-stimulator wells were used as negative controls, and third-party stimulators and PHA wells (0.02μg/μL) (InvivoGen, San Diego, CA, Cat: inh-phap-2) were used as positive controls. The plates were wrapped with Scotch Magic Tape (Staples, Framingham, MA, Cat: 810K12) to prevent evaporation and placed in an incubator with 6% CO₂ at 37°C.

On day 6, the contents of each well were collected into 5ml Round-Bottom Polystyrene Test Tubes (Thermo Fischer, Waltham, MA, Cat: 14-959-2A) and washed with HBSS. The cells were incubated with previously titrated anti-CD3 (Pig Clone: BB23–8E6–8C8, Human Clone: SP34–2, BD, Franklin Lakes, NJ, Cat: 561477, 562406), anti-CD4 (Pig Clone: 74-12-4, Human Clone: L200, BD, Franklin Lakes, NJ, Cat: 561474, 560836), and anti-CD8 (Pig Clone: 76-2-11, Human Clone: RPA-T8, BD, Franklin Lakes, NJ, Cat: 561475, 563823) mAbs for 30 minutes, then washed with wash buffer. Wash buffer was made by combining 1000ml of HBSS with 1000μg of bovine serum albumin (Sigma Aldrich, St. Louis, MO, USA, Cat: 12306C), 1000μg of 99% ultrapure Sodium Azide (Thermo-Fischer, Waltham, MA, USA, Cat: 190380050) and 210μl of 1N Sodium Hydroxide (Thermo-Fischer, Waltham, MA, USA, Cat: 124260250). 5μL of DAPI (10mM) diluted 1:10000 with wash buffer (Abcam, Waltham, MA, Cat: ab228549) was added immediately before acquisition to distinguish dead cells.

Flow Cytometry Acquisition

The data was acquired using a 5-Laser Cytek Aurora Flow Cytometer (Cytek, Fremont, MD) and analyzed using FCS Express (Version 7.1, De Novo Software, Pasadena, CA). To analyze, non- γδ CD3+, CD3+CD4+, and CD3+CD8+ cells were selected from the live singlet population, and the %CFSElow of each subset was calculated (Fig 2).

Figure 2.

Demonstrative analysis of CD3, CD4, and CD8 gates with CFSElow and CFSEhigh populations are shown.

RESULTS

Overcoming Stool and Microbial Contamination

One of the biggest challenges in developing the mucosal MLR assay was overcoming the stool and microbial contamination present during sample collection. We approached the decontamination process in two steps: mechanical gross decontamination and antimicrobial sterilization. To achieve the first mechanical gross decontamination step, we devised a serial wash of the intestinal graft in 500ml bottles (explained above). When performed in 4 steps, this allowed for the maximal gross decontamination of the intestinal mucosa (Fig. 1B). For antimicrobial sterilization, we developed a regimen that would concurrently sterilize the cells as they were dissociating from the mucosa and also allow immediate downstream applications. Accordingly, we opted to use an antibacterial and antifungal regimen (described above), which allowed the tissue to be incubated for 3–4 hours with high-dose antibiotics and antifungals during the processing (Table 2). Using this two-step regimen, we successfully avoided complications from stool and microbial contamination in any of our subsequent mucosal MLR assays (n=14).

Table 2.

Contents of Buffers Used to Digest the Intestinal Mucosa

	Buffer 1	Buffer 2	Buffer 3
HBSS-CMF	50ml	50ml	-
RPMI-1640	-	-	50ml
DTT, 1M	100μl	-	-
EDTA, 0.5M	50μl	50μl	-
Collagenase D	-	-	50mg
DNAse	0.5mg	0.5mg	0.5mg
Penicillin/Streptomycin	5000U/5mg	5000U/5mg	5000U/5mg
Amphotericin B	50μg	50μg	50μg

Optimization of Debris Removal and Sorting

We observed that when intestinal mucosa is processed, 90–95% of the pellet consists of debris (data not shown). In our preliminary experiments, the mucosal cells could maintain cell-to-cell interactions and mount anti-third party responses, even in the presence of debris (data not shown). However, the acquisition data showed that when the debris was not removed, almost 85–90% of the acquired events were of debris, greatly reducing the number of events in the live cell gate and clogging the sorting columns or cytometers, making it impossible to perform downstream procedures (Fig. 1C). To overcome these issues, we tried straining the cells through serially smaller filters (smallest 20μm). However, the debris could easily pass through the strainer (Fig. 1C). We experimented with density-gradient separation, but the debris was not dense enough to pass through the density-gradient medium (Fig. 1C). We further investigated the Dead Cell Removal Kit (Miltenyi Biotec, Gaithersburg, MD, Cat: 130-090-101), which is advertised to bind and remove debris. However, using this required a magnetic sorting column, which would continuously clog due to the column collecting debris from the large amount of tissue processed. Lastly, we investigated the Debris Removal Solution. In our preliminary tests, it was successful in removing 90–95% of the debris. By scaling up and modifying the manufacturer’s guidelines (explained above), we successfully overcame and optimized the debris removal process, consistently allowing successful downstream depletion or sorting of the mucosal cells (Fig. 1C). Additionally, the percentage of live cells in the acquisition greatly improved from around 5% to 50% (Data not shown).

Self-Tolerance and Allo-Recognition in Swine Mucosal MLR Assays

We next sought to establish whether the mucosal cells immediately after enzymatic and mechanical dissociation could survive up to 6 days when cultured and maintain hyporesponsiveness against self-antigens while recognizing and responding against allo-antigens. Regardless of the addition of the debris removal step, cells plated immediately after enzymatic and mechanical dissociation survived in sufficient numbers for analysis for up to 6 days. These mucosal cells showed unresponsiveness to self-antigens with robust alloresponses against third-party antigens (Fig. 3, Naïve)

Figure 3.

The mucosal MLR assay results of a naïve animal and all 9 experimental animals at the end-point against self, donor, and third-party PBMC stimulators are shown. The assays were performed using surgically collected intestinal grafts. Mucosal lymphocytes were resuspended at 2×10⁶ cells/ml and plated at a 1:1 ratio with stimulators resuspended at 2×10⁶ cells/ml. The y-axis represents the proliferated CFSElow cell population of CD3+ cells on day 6 of incubation. Anti-self response indicates background proliferation, and anti-third party acts as a positive control, showing the proliferative capability of the cells. Lack of anti-donor response with an anti-third party response indicates tolerance development. High anti-self response indicates GvHD. High anti-donor response indicates lack of tolerance development.

Mucosal MLR Accurately Predicts Intestinal Tolerance, Rejection, GvHD, and Treg Mediated Tolerance in Swine

Having validated and optimized our assay, we performed mucosal MLR assays on swine that underwent orthotopic ITx with a tolerogenic regimen (n=9). Seven animals received orthotopic ITx (Group 1), and two received bone marrow transplantation and delayed ITx (Group 2) (Table 1). These experiments will be reported in detail in a separate manuscript (Gunes et al., manuscript under revision in Transplantation, TPA-2024–1496) and have been reported elsewhere.⁶ Briefly, five animals were weaned off ISP, one was euthanized due to severe GvHD, and three were euthanized due to non-immunological complications (Table 1). Of the five animals that were weaned, four developed tolerance and did not reject their graft, and one developed chronic cellular and antibody-mediated rejection (Table 1). When we performed a mucosal MLR, all animals that developed tolerance showed no anti-donor or anti-recipient responses while maintaining robust anti-third-party responses before or after weaning of ISP (Fig. 3, Table 1). In the case of the animal with GvHD (Animal 1), there was a robust anti-recipient response, no anti-donor responses, and a robust anti-third-party response (Fig. 3). In the animal that rejected its graft (Animal 6), mucosal MLR showed a robust anti-donor response at all timepoints, including before and during weaning of ISP (Fig. 3, end-point data shown). Furthermore, in one of the animals that was completely weaned of ISP and showed robust anti-donor responses in peripheral blood MLR but tolerance to its graft in the mucosal MLR (Animal 5, Figs. 3 and 4), we performed a CD25 depletion mucosal MLR (Fig. 4A). When we depleted the CD25+ Tregs from the mucosa (Figs. 4A), the previously hyporesponsive mucosal cells recovered a strong anti-donor response, indicating the development of Treg-mediated tolerance in the mucosa (Figs. 3 and 4B). In further investigation, the percentage of Tregs among CD4+ cells in this graft (6.43%) did not vary significantly compared to the graft that was rejected (5.21%), but the percentage of donor Tregs did. Donor Treg chimerism in the rejected graft was 0.51% in contrast to 26.30% on day 112 and 49.64% on days 112 and 174 after ITx in the graft of the tolerant animal.

Figure 4.

A) Demonstrative example and flow cytometric analysis of responder cell phenotype before and after CD25 depletion using magnetic beads is shown. B) Peripheral and Mucosal MLR responses with and without CD25 depletion against self, donor, and-third party PBMC stimulators of Animal 5 are shown. The mucosal MLR assays were performed using surgically collected intestinal grafts. PBMC and mucosal lymphocytes were resuspended at 1.5×10⁶/ml and 2×10⁶/ml, respectively, and plated at a 1:1 ratio with stimulators resuspended at 2×10⁶/ml. The y-axis represents the proliferated CFSElow cell population of CD3+ cells on day 6 of incubation. Anti-self response indicates background proliferation, and anti-third party acts as a positive control, showing the proliferative capability of the cells. Lack of anti-donor response with an anti-third party response indicates tolerance development. High anti-donor response indicates lack of tolerance development. An increase in response after CD25 depletion indicates the role of Tregs in the tolerance.

Auto- and Allo-Responses in Human Mucosal MLR Assay

Having successfully validated and utilized our mucosal MLR assay in swine after ITx, we investigated the feasibility of performing mucosal MLR assays in human patients after ITx. Our initial experiments included establishing the auto- and allo-responses of the mucosal cells immediately after enzymatic and mechanical dissociation, as previously described above for swine experiments. To do so, graft stoma specimens were collected from two patients who underwent routine stoma closure after intestinal transplantation. Both were in good health without rejection for several months before the stoma closure. One patient had undergone isolated intestinal transplantation approximately three years previously, and the other patient had received a multivisceral graft approximately three and a half years previously. Both patients shared only 1 MHC class I allele with their donor. At the time of stoma closure, the first patient had approximately 82% recipient replacement of the donor lymphoid cells in the mucosa, while no chimerism data was available for patient 2. Both patients had clinically been doing well without rejection. After processing the samples, an adequate amount of IEL and LPL were collected, and mucosal MLR assays were plated against self and third-party stimulators. In both cases, the mucosal cells were unresponsive to self-antigens while demonstrating robust third-party responses (viable donor cells were not available to evaluate for donor-specific hyporesponsiveness). (Fig. 5)

Figure 5.

The preliminary mucosal MLR assay results of two patients who underwent ITx and MvTX against self, donor, and third-party PBMC stimulators are shown. The assays were performed using surgically collected intestinal grafts during stoma closure. Mucosal lymphocytes were resuspended at 2×10⁶ cells/ml and plated at a 1:1 ratio with stimulators resuspended at 2×10⁶ cells/ml. The y-axis represents the proliferated CFSElow cell population of CD3+ cells on day 6 of incubation. Anti-self response indicates background proliferation, and anti-third party acts as a positive control, showing the proliferative capability of the cells.

DISCUSSION

Evaluation of immune responses in the graft is a critical part of understanding the pathophysiology of rejection and the mechanism of tolerance. We herein present our novel modification of the MLR assay using intestinal mucosal lymphocytes to characterize the immune responses within the intestinal graft. We demonstrate how this assay facilitates direct investigation of the local immune responses and mechanisms, including the development of tolerance, within the graft mucosa. We have successfully established a decontamination and debris-removal protocol that has consistently allowed us to overcome stool and microbial contamination and minimize debris in both swine and human samples to analyze mucosal T cell responses to self, donor, and third-party antigens. Our assay was validated by consistently showing unresponsiveness to self-antigens and strong responses to third-party antigens.

Furthermore, we demonstrate the ability of our assay to elucidate previously unknown immune responses that would otherwise be undetected. In this case, it is the development of a previously undescribed form of split tolerance. Moreover, CD25 depletion of the mucosal cells also allowed us to demonstrate the role of mucosal Tregs in mediating intestinal transplant tolerance. We are currently investigating the sorting of the donor and recipient mucosal cells to investigate donor and recipient-specific responses within the graft mucosa and potential downstream assays such as TCR-seq and multiomic studies, demonstrating how the assay can easily be tailored to specific mechanistic questions and to categorize immune rejection and tolerant features better.^14,15 We note that, if validated in human ITx recipients, this assay might potentially serve as a biomarker or predictor of rejection or quiescence, clarifying pathophysiology while also guiding prognosis and treatment.

The clinical usefulness of the assay described herein is limited by the need for a larger portion of the graft than that usually sampled by endoscopy. We therefore validated our model using surgical resection specimens, which easily supplied enough tissue. However, we used our experimental animals to demonstrate that enough graft mucosa for immunological assays could be obtained from multiple mucosal biopsies (>15–20). For the purposes of developing this assay, we used only surgical specimens since we could not justify the risk of more invasive biopsies for non-clinical reasons. However, with the utility of this method having been demonstrated, multiple endocopic biopsy specimens could perhaps be justified in the future if it were to aid in management, and we plan to test this technique in future patients. Additionally, we intend to save pre-transplant donor cells to be used as stimulators for this assay, which would allow us to validate the clinical relevance of this assay in humans, as we have in swine. Performing this technique in humans would also provide the opportunity to track the TCR sequences of graft lymphocytes. This technique has not yet been developed for swine and would add depth to the analysis by including information about both donor-reactive T cell clones as well as host-versus-graft-reactive clones. Additionally, this study can always be performed at the time of stoma closure or any other procedure in which a portion of graft is resected since a mucosal sample of ~2cm² is enough to perform the assay. Since stoma closures of transplant patients usually occur only after a period of stable intestinal graft function, stoma closure is often an optimal timepoint to investigate the question of tolerance.¹⁶

To conclude, we have successfully modified, optimized, and validated an MLR assay using mucosal lymphocytes. We have also created a decontamination and sterilization regimen that has consistently minimized microbial contamination during culturing. Thus far, in all our experimental animals after ITx, mucosal MLR has successfully predicted the development of tolerance and rejection and has demonstrated a novel tolerance mechanism, even in the absence of tolerance in the peripheral blood. With these promising initial results and the possible utilization in human ITx patients, we believe this assay will help guide clinicians and researchers in future studies of immune responses after ITx.

ABBREVIATIONS

ACK

Ammonium-Chloride-Potassium

CFSE

Carboxyfluorescein succinimidyl ester

CML

Cell-mediated lympholysis

DAPI

4’,6-diamidino-2-phenylindole

DMSO

Dimethyl sulfoxide

DNAse

Deoxyribonuclease

DTT

Dithiothreitol

EDTA

Ethylenediaminetetraacetic acid

GvHD

Graft versus host disease

HBSS

Hanks’ Balanced Salt Solution

IEL

Intraepithelial lymphocytes

ISP

Immunosuppression

ITx

Intestinal transplantation

LPL

Lamina propria lymphocytes

MLR

Mixed lymphocyte reaction

MP

Methylprednisolone

PBMC

Peripheral blood mononuclear cells

PHA

Phytohemagglutinin

RBC

Red blood cells