Intragraft memory-like CD127hiCD4+Foxp3+ Tregs maintain transplant tolerance

CD4+Foxp3+ regulatory T cells (Tregs) play an essential role in suppressing transplant rejection, but their role within the graft and heterogeneity in tolerance are poorly understood. Here, we compared phenotypic and transcriptomic characteristics of Treg populations within lymphoid organs and grafts in an islet xenotransplant model of tolerance. We showed Tregs were essential for tolerance induction and maintenance. Tregs demonstrated heterogeneity within the graft and lymphoid organs of tolerant mice. A subpopulation of CD127hi Tregs with memory features were found in lymphoid organs, presented in high proportions within long-surviving islet grafts, and had a transcriptomic and phenotypic profile similar to tissue Tregs. Importantly, these memory-like CD127hi Tregs were better able to prevent rejection by effector T cells, after adoptive transfer into secondary Rag–/– hosts, than naive Tregs or unselected Tregs from tolerant mice. Administration of IL-7 to the CD127hi Treg subset was associated with a strong activation of phosphorylation of STAT5. We proposed that memory-like CD127hi Tregs developed within the draining lymph node and underwent further genetic reprogramming within the graft toward a phenotype that had shared characteristics with other tissue or tumor Tregs. These findings suggested that engineering Tregs with these characteristics either in vivo or for adoptive transfer could enhance transplant tolerance.


Supplemental Methods
The isolation protocol for porcine neonatal islet cell-clusters and culture.Pancreas were extracted from two to seven days old piglets and chopped in 15ml of 1× Hanks Balanced Salt Solution (HBSS) without porcine sera into small pieces (about 2 mm in size) and then were digested with 1 mg/ml Collagenase Type V (Sigma-Aldrich) at 37° for 12-14 minutes and filtered through a metal mesh.Then the islets were washed twice with 1× HBSS solution with 1% heat inactivated porcine sera, and centrifuged at 300g for 1 minutes.
The supernatant was aspirated, and the islets were cultured in 20ml Hyclone Ham's F-10 culture media (GE Healthcare) containing 10% porcine sera, 10 mM glucose, 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, Hepes 80 mM, 10 mM nicotinamide, 50 mM isobutylmethlxanthine, CaCl2 0.236 g/l, and NaHCO3 21.3% at 37C, 4.9% CO2 in 150mm × 20mm Petri dishes.Media were changed on day 1 and 3, where the cells were settled, and the supernatant was removed.The supernatant was then spun down at 300×g for 1 minutes at room temperature, and the supernatant was again removed.The cell pellet was suspended in HAMS F-10 culture medium.On day 5, each dish was topped up by 10ml of HAMS F-10 culture media.Islets were collected on day 6-7 and their diameter measured and categorised under the microscope, with the smallest being 50μm, and increasing incrementally by 50μm to 400μm.To normalise islet volume, the islet equivalent (IEQ) was calculated from the diameters, assuming that 1 IEQ is equal to a spherical islet with a diameter of 150μm.The islets under each diameter category were then converted to IEQ and 4000 IEQ porcine-NICC for each transplant.

Single cell suspensions for flow cytometric analysis and cell sorting.
For preparing single cell suspensions from PB, PB was collected from mice into 4% w/v sodium citrate solution (Sigma Aldrich), which was kept at 4⁰C.The sample tubes were centrifuged at 300g for 10 minutes, and the supernatant was poured off.The samples were lysed using sterile Ammonium-Chloride-Potassium (ACK) lysis buffer (Lonza Bioscience) for 3 minutes at room temperature and centrifuged at 300×g for 10 minutes at 4°C, then were washed using the FACS buffer solution [containing Dulbecco's phosphate-buffered saline (DPBS), 0.5% bovine serum albumin (BSA) and 2mM EDTA)] twice at 4°C.The cells were resuspended in the FACS buffer solution at 1 × 10 6 cells/100 μl.
For preparing single cell preparation from spleen, the spleens were removed from mice and transferred to a tube with RPMI 1640 (Lonza) supplemented with 10% fetal calf serum (FCS), then dissociated in a gentle MACS C Tube (MiltenyiBiotec) using the gentle MACS Dissociator (MiltenyiBiotec) with 3ml of FACS buffer solution as per the manufacturer's instructions.The samples were centrifuged at 300×g for 30 seconds at 4°C, and were washed through 70µm cell strainer (BD Biosciences) with the FACS buffer solution and centrifuged again at 300×g for 10 minutes at 4°C, then incubated in ACK lysis buffer for 2 minutes at room temperature, and washed using the FACS buffer solution twice at 4°C.The cells were resuspended in the FACS buffer solution at 1 × 10 6 cells/100 μl.
For preparing single cell preparation from lymph nodes, the lymph nodes were removed from mice and transferred to a tube with 3ml RPMI 1640 medium containing 10%FCS and 2mM EDTA, the lymph nodes were gently ground through a 70μm cell strainer (BD Biosciences) in a petri dish into a single cell suspension, were transferred and centrifuged at 300×g for 10 minutes at 4°C in the FACS tube, and washed twice using the FACS buffer solution at 4°C.The cells were resuspended in the FACS buffer solution at 1 × 10 6 cells/100 μl.
For preparing single cell preparation from graft infiltrating cells, the islet grafts were removed from mouse kidney and transferred to a petri dish with 3ml RPMI 1640 medium containing 10%FCS and 2mM EDTA, and were gently flushed with 19G needle syringe, then were ground gently through a 70μm cell strainer (BD Biosciences) into a single cell suspension.The samples were washed twice using the FACS buffer solution at 4°C and resuspended in the FACS buffer solution at 1 × 10 6 cells/100 μl.

Staining protocols for flow cytometric analysis and cell sorting.
For the staining protocol to assess CD4 + GFP + Tregs in PB samples, the single cell suspension samples in the FACS buffer solution at 1 × 10 6 cells/100 μl were blocked with 1l of 1:100 dilution purified anti-mouse-CD16/CD32 (2.4G2, BD Biosciences) for 20 minutes at 4⁰C, then the samples were stained with 1l of 1:100 dilution anti-mouse-CD4-Pacific blue for another 30 minutes at 4⁰C.After washing twice in the FACS buffer solution at a speed of 300×g for 10 minutes at 4°C, samples were resuspended in the FACS buffer solution for analysis on a flow cytometry for GFP and CD4 expressions.
For the surface staining protocol of multicolor panels, fluorescent-antibody cocktails were made in the FACS buffer solution prior to staining, and for the panels that contain two BD Horizon Brilliant antibodies were pre-mixed in BD Horizon Brilliant Stain Buffer (BV buffer) (BD Biosciences) at a 1:2 ratio.The antibody concentration in the cocktails was based on antibody titration and product instructions.
For Treg phenotyping panels, the samples of single cell suspension at 1 × 10 6 cells/100 μl were stained with viability dye Zombie yellow (BioLegend) for 30 minutes at 4⁰C based on the manufacturer's instructions with a temperature modification, then washed with the FACS buffer solution and centrifuged at 300×g for 10 minutes at 4°C.The samples were resuspended in FACS buffer solution at 1 × 10 6 cells/100 μl, and were blocked with purified anti-mouse-CD16/CD32 for another 20 minutes at 4⁰C, then stained with surface antibody cocktails for another 30 minutes at 4⁰C.The samples were washed twice with the FACS buffer solution at a speed of 300×g for 10 mins at 4°C, and resuspended for acquisition.
For the Foxp3 intracellular staining protocol, the samples were prepared at 4⁰C during processes according to the Foxp3/transcription Factor Staining Buffer Set protocol (eBioscienceTM).Briefly, after surface staining completion, the sample were fixed and permeabilized with transcription factor buffer set for 50 minutes at 4˚C, then washed with fixation washing buffer twice.Samples were stained with Foxp3 antibody for 40 minutes at 4˚C.Then the samples were washed twice with fixation washing buffer, and resuspended for acquisition.
For the phosphorylation of STAT5 intracellular staining, single-cell suspensions containing 2 × 10 6 cells from DLN and graft, and 10 × 10 6 cells from the spleen in 1 mL PBS/2% FCS were stimulated with murine IL-7 (5 ng/ml; PeproTech) or recombinant human IL-2 (320 ng/ml; Novartis) for 5 minutes at 37°C in a water bath, followed by 25 minutes in a 37°C incubator with 5% CO2.Samples were fixed with 1 mL CytoFix (BD Biosciences, warmed to 37°C before use) for 10 minutes at 37°C.These fixed cell samples were washed with PBS/2% FCS and then permeabilised with Phosflow Perm Buffer III (BD Biosciences) on ice for 30 minutes.Next, samples were washed with PBS/2% FCS, and stained with CD4, CD127, and STAT5 antibodies for 1.5 hours at room temperature in the dark.Finally, the samples were washed with PBS/2% FCS twice and then resuspended for acquisition.

Staining protocol of immunohistochemistry for insulin Staining.
Formalin-fixed paraffin waxembedded samples of kidney containing grafts were sectioned at 6-7μm.Slides were de-waxed in xylene and rehydrated in decreasing concentrations of ethanol, finishing in an H2O wash.Slides were incubated in 3% H2O2/methanol for 10 minutes at room temperature and washed in phosphate-buffered saline solution (PBS) with 0.05% Tween™ 20 (PBSt).Slides were blocked in PBSt with 7% rabbit serum for 20 minutes at room temperature and incubated in the primary antibody for 1 hour at room temperature.Slides were washed in PBSt and incubated in the secondary antibody mix for 30 minutes at room temperature.
Slides were washed in PBSt, incubated in DAB for 3 minutes before rinsing in H2O.Slides were stained with hematoxylin and dehydrated in increasing concentrations of ethanol and xylene before coverslipping.

Staining protocol of immunofluorescence for insulin.
Frozen OCT samples of kidney containing grafts were cryo-sectioned at 6-7μm.Antibodies used are detailed in table S5.Sections were fixed in 4% paraformaldehyde for 10 minutes at room temperature and washed in PBS solution.Slides were blocked in DPBS with 2% BSA for 20 minutes and then incubated in the primary antibody mix (diluted in 2% BSA/PBS) overnight at 4°C.Slides were then washed in PBSt and incubated in secondary antibody mix (diluted in 2% BSA/PBSt) for 1 hour at room temperature, in the dark.Slides were washed in PBSt and then counterstained with DAPI before being mounted and coverslipped.

Staining protocol of imaging mass cytometry.
Frozen OCT NICC graft-kidney samples for imaging mass cytometry (IMC) were cryo-sectioned at 6-7μm two days before antibody incubation.Subsequent slides were stained for hematoxylin and insulin to confirm graft sites.Slide staining was split into two groups with evenly distributed samples from each experimental group to prevent both batch effects and inconsistent staining times.IMC slides were fixed with 100% methanol for 10 minutes at room temperature and washed in PBSt.Slides were blocked using 3% BSA/PBSt, with anti-CD16/32 added to block low-affinity Fc receptors.Slides were incubated with the primary antibody cocktail (Supplementary Table 6) overnight at 4°C.Slides were washed in PBSt and a secondary fixation of 4% paraformaldehyde was carried out at room temperature for 20 minutes.Slides were washed in PBSt and counter-stained with iridium-DNA-intercalator diluted in 3% BSA/PBSt for 30 minutes at room temperature.Slides were washed in PBSt with a final ultra-pure H2O 5 second wash before air-drying.Slides were stored in a sealed container at room temperature prior to acquisition.

IMC data visualisation, cell segmentation and ROI extraction.
IMC data was visualised using MCD Viewer.Colour thresholding was performed on each channel to remove background noise and was consistently applied to the same channel across all samples.Channels of interest were overlayed and processed in ImageJ.Prior to quantitative analysis, cell masks for each sample were created using a process modified from the Bodenmiller Lab protocol by the Sydney Cytometry Facility.MCD files of each sample acquired from the Hyperion™ Imaging System were extracted to produce tiff image files corresponding to each marker channel.CellProfiler was then used to create random cropped sections for each sample.These crops were used in Ilastik to train feature identification using supervised machine learning.Pixels were classified as belonging to either nucleus, cytoplasm, or background, generating a probability map.The probability map was then imported back into CellProfiler to generate an individual cell mask file for each sample, identifying cell nuclei and cytoplasmic regions, for use in all proceeding analysis.A false-positive edge artifact in the B220_176Yb channel was observed in all 24 samples, resulting in a strong signal running the length of the tissue edge, which has commonly been reported in similar methods.Since this artefact was consistent across every sample, an additional step was incorporated to exclude the edge-regions using a gating method common to IMC ROI extraction.IMC data for each sample, including respective cell masks, was loaded into computational histology topography cytometry analysis toolbox (histoCAT).Each sample was then manually gated to exclude the edge-region (distance) and all single-cell information (marker abundance, xy spatial data and neighbourhood data) was extracted as an csv file per sample.
Archsinh transformed expression values were scaled per image and cells were assigned as positive for each marker based on scaled values greater than background and verified by manual image analysis.
Cell types were assigned according to standard type marker designation.Heatmaps were generated using the SPECTRE do.aggregate function (95) and pheatmap (89).Other visualizations were generated using ggplot2 (90).
Error bars indicate the mean ± SEM.