Journal Logo

Original Basic Science—General

Indirectly Activated Treg Allow Dominant Tolerance to Murine Skin-grafts Across an MHC Class I Mismatch After a Single Donor-specific Transfusion

Zhang, Geoff Yu MD1; Hu, Min MD, PhD1,2; Watson, Debbie PhD1,3,4; Wang, Yuan Min MD, PhD1; Knight, John F. MBBS, MBA, FRACP1,5; Alexander, Stephen I. MD, MPH, FRACP1

Author Information
doi: 10.1097/TP.0000000000003173
  • Free
  • Visual Abstract



The induction of donor-specific tolerance with minimal immunosuppression remains the ultimate goal for organ transplantation. Pretransplantation donor-specific transfusion (DST) is a potential strategy for achieving such a state of tolerance.1,2 DST has been associated with tolerance inducing mechanisms of clonal deletion,3 anergy,4,5 and induction of regulatory T cells (Treg).4,6-11 In major histocompatibility (MHC)-mismatched DST models, additional immunological manipulation of costimulatory3,4,7,9,10,12 or coreceptor9,11 blockade is required to reduce the alloreactive effector T-cell pool and allows Treg to exert their suppressive functions.1,4,8

Transplantation tolerance is associated with alloantigen-specific Treg cell expansion, where the Treg is defined as CD4+CD25+ Foxp3+ T cells.1,13 In the transplantation setting, Treg are activated by 2 different pathways; in the direct pathway, the T-cell receptors (TCRs) of recipient CD4 T cells directly bind to intact MHC class II on the cell surface of donor antigen-presenting cells (APCs); in the indirect pathway, recipient CD4 T cells recognize processed allo MHC-derived peptide presented on recipient MHC class II molecules through their TCR. While Treg can be activated both directly and indirectly by alloantigen presenting cells, studies suggest that long-term transplantation tolerance predominantly uses indirect activation of Treg.7,14-19

Survival and proliferation of antigen-specific Treg requires continuous alloantigen recognition in a danger-free environment.20-22 Donor cells in the circulation after DST can provide a source of allostimulation, and the absence of danger promotes Treg survival and expansion. However, rapid removal of donor cells by the host immune system limits their capacity for continuous alloactivation.6,23,24 Therefore, strategies to improve donor cell persistence may assist in augmenting DST-induced tolerance resulting in a reduced need for nonspecific immunological suppression.6

In this study, the use of F1 mice from mice with a mutated class I MHC Kb called bm1 mice with wild-type B6 mice produces mice with both mutant H-2kbm1 and wild-type H-2Kb expressed on all cells. The expression of the normal H-2kb limits the activation of natural killer (NK) cells through inhibitory self-receptors. CD4 T cells can only see this difference indirectly with the Kbm1 molecule fragments presented through class II molecules.

Here, we report dominant skin-allograft tolerance induced by a single DST using this semiallogeneic MHC class I mismatched model without additional immunological manipulation. This DST model shows prolonged donor cell persistence due to less NK cytotoxicity and natural Treg (nTreg) cell expansion primarily through indirect recognition of antigen.



C57BL/6 (H-2b) (B6), B6.C.H-2 bm1 (H-2bm1) (bm1), B10.BR (H-2k), and BALB/c (H-2d) mice were purchased from WEHI (Melbourne, Australia). Rag1−/− mice and B6.Ly5.1 were purchased from the ARC (Perth, Australia). Foxp3GFP and depletion of Treg (DEREG) mice (B6 background) were provided by Professor Rudensky (University of Washington) and Professor Tim Sparwasser (Hannover Medical School, Germany) and were bred at Westmead Animal Care Facility Sydney, Australia. F1 (B6 × bm1) and Ly5.1 Foxp3GFP mice were bred at Westmead Animal Care Facility (Sydney, Australia). All animals were kept under specific pathogen-free conditions. The Westmead AEC approved use of animals in this project.

DST and Skin Grafting

Mouse splenocytes from donor origin (bm1 or F1) were resuspended in PBS and 3 × 107 cells were transfused into B6 recipient mice through the tail vein. Full-thickness tail skin from the donor strain (bm1 or F1) or third party (BALB/c) was transplanted onto a prepared site on the dorsum of the recipients 7 days after DST. Rejection occurred when 90% of the graft had been lost or become necrotic.25 A subset of mice received 2 further DSTs at the time skin grafting with bm1 skin grafts and 1-week later.

In Vivo Tracking of 5- (and 6-) Carboxyfluorescein Succinimidyl Ester Stained Donor Cells

Splenocytes from the donor strain of bm1, F1 (B6 × bm1), or BALB/c mice were stained with high-dose 5- (and 6-) carboxyfluorescein succinimidyl ester (CFSE) (2 µmol/L) and recipient strain splenocytes were stained with low-dose CFSE (0.6 µmol/L), respectively. Equal numbers of stained cells (3 × 107) from donor and recipient origin were cotransfused intravenously (i.v.) into recipient mice. Tail blood was collected from day 1 after DST. Flow cytometry was used to evaluate the relative reduction of donor cells (CFSE high) in comparison with the recipient (CFSE low) cells.

In Vivo Cytotoxicity and NK-cell Depletion

B6 mice received the NK-cell depletion antibody NK1.1 (BioXcell, West Lebanon) or an isotype control antibody (250 µg/20 g body weight) intraperitoneally (i.p.) 2 times over 48 hours. Efficiency of NK-cell depletion was confirmed by flow cytometry using an antimouse NK antibody CD49b (DX5). CFSE high (2 µmol/L) labeled splenocytes (2 × 107) from bm1 or BALB/c mice were cotransferred with equal numbers of CFSE low (0.6 µmol/L) labeled host cells into B6 mice 24 hours after the first antibody injection. Tail blood was collected for 3 consecutive days, and the ratio of CFSE labeled donor to host cells was assessed by flow cytometry. B6 mice were depleted of NK cells and received bm1 grafts as controls using NK1.1 i.v. before skin grafting.

Tolerant Skin-graft Transfer

F1 (B6 × bm1) skin grafts made tolerant by DST at day100 and control syngeneic skin grafts were dissected from the host and retransplanted onto Rag1−/− mice for 45 days. Rag1−/− mice were injected i.v. with 0.5 × 106 splenocytes from naive B6 mice and grafted with fresh F1 and control BALB/c skin grafts.

RNA Extraction and Quantitative Real-time Polymerase Chain Reaction

RNA was extracted using Trizol reagent (Invitrogen, CA) and cDNA synthesis performed using random primers and SuperScript II reverse transcriptase (Invitrogen, CA). Forkhead box P3 (Foxp3) mRNA levels were quantified by real-time polymerase chain reaction using TaqMan Gene Expression (Applied-Biosystems, CA) on the Rotor-Gene 3000 machine (Corbett Research, NSW, Australia) as described previously.26 The comparative Ct method (or ΔΔCt) was used to assess relative changes in mRNA levels of Foxp3 normalized to the reference gene of mouse TCR beta constant region.26,27

Flow Cytometry

Flow cytometry used the following cell surface marker antibodies (BD Biosciences, CA); antimouse CD25-fluorescein isothiocyanate (7D4), CD4-phycoerythrin (PE) (L3T4), CD45.1-APC (A20), and CD49b-PE (DX5). Foxp3 intracellular staining used an antimouse Foxp3 PE Staining Kit (eBioscience, CA). Analysis used FACScan or LSRII and data were analyzed by FACSDiva software (Becton-Dickinson, NJ).

Mixed Lymphocyte Reaction and Suppression Assay

Cultures were performed in round-bottom 96-well plates in a total volume of 200 µL complete medium (RPMI 1640 supplemented with 10% fetal calf serum, 2 mmol/L L-glutamine, 25 mmol/L HEPES, and 10 µmol/L 2-mercaptoethanol (GibcoBRL, NJ)) 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) cells and an equal number of irradiated stimulators were cocultured at 37°C 5% CO2 for 72 hours. To evaluate CD4 subsets with regulatory capacity in vitro, CD4+CD25+ and CD4+CD25 cells were isolated from splenocytes using a CD4+CD25+ regulatory T-cell isolation kit (Miltenyi-Biotec, Germany). CD4+CD25− responder (R) cells (2 × 105) were cocultured with equal numbers of irradiated allogeneic splenocytes. CD4+CD25+ and CD4+CD25 stimulator (S) cells isolated from tolerant and naive control mice were added to obtain the responder (R): suppressor (S) ratios of 1:1 to 1:0.06. Plates were incubated at 37°C 5% CO2 for 80 hours. Each well was pulsed with 1 µCi3H-thymidine (MP Biomedical, OH) for the last 18 hours. Cells were procured, and 3H-thymidine uptake was measured using a scintillation counter (Microplate2450 PerkinElmer, MA).


Interferon (IFN)-γ ELISpots were performed as described previously.25 Briefly, 3 × 105 responder splenocytes were stimulated with equal numbers of mitomycin C-treated donor splenocytes using IFN-γ antibody precoated 96-well plates. IFN-γ production was assessed at 24 hours using ELISPOT.

Histology and Immunohistochemistry

Dissected skin grafts were fixed in 10% neutral buffered formalin and embedded in paraffin for sectioning. Slides were stained with hematoxylin and eosin (H&E). Skin Treg was assessed using immunoperoxidase staining for Foxp3 (eBioscience, antimouse Foxp3, clone FJK-16s) as previously described.28

Assessment of Treg Conversion and Proliferation

Negatively selected CD4+ cells (Stemcell Technologies) from Ly5.1 Foxp3GFP mice (CD45.1+) were labeled with antimouse CD4-PE and sorted into CD4+ Foxp3GFP + and CD4+Foxp3GFP − cells by flow cytometry (FACSVantage-Diva BD). Sorted cells were stained with 2.5 µmol/L CellTrace Violet proliferation dye (Invitrogen, CA). CellTrace Violet labeled cells (1 × 106/mouse) were transfused together with 2 × 107 F1 (B6 × bm1) cells or syngeneic B6 cells to the B6 hosts. F1 skin grafting was performed 7 days after injection of cells. Mice were sacrificed at day 12 after cell transfer. CD4 selected splenocytes (Miltenyi-Biotec, Germany) were stained with antimouse CD45.1-APC and CD4-PE. Cells were gated on CD45.1+ and CD4+ cells for expression of FoxP3 GFP+ or dilution of CellTrace Violet proliferation dye by flow cytometry.

Depletion of Treg in DEREG Mice

DEREG mice were used for selective depletion of Foxp3+ Treg cells by diphtheria toxin (DT) injection as previously described.26 DEREG mice and their littermates received F1 (B6 × bm1) DST and F1 skin grafts.


Data were analyzed with GraphPad Prism v6.0 software using Student unpaired t tests. The log-rank test was employed for comparison of survival data between groups using GraphPadv6.0. P ≤0.05 were considered statistically significant.


A Single Transfusion of Semiallogeneic F1 (B6 × bm1) Splenocytes Into B6 Mice Leads to Long-term Acceptance of F1 Skin Grafts

bm1 is a mutant B6 strain with a 3 amino-acid variant of the class I Kb molecule resulting in acute skin-graft rejection between the strains.29,30 Survival of bm1 and semiallogeneic F1 (B6 × bm1) and BALB/c control skin grafts on B6 mice was compared with and without pretransplant DST (Figure 1A). Both bm1 (median survival time [MST] 17 d, n = 4) and F1 (MST 18 d, n = 5) skin-grafts transplanted on B6 mice were acutely rejected (P = NS) (Figure 1A). Pretransplantation (7 d before transplant) transfusion of bm1 cells (bm1 DST) into B6 mice prolonged bm1 skin-graft survival (MST 38 d, n = 9) but did not induce long-term tolerance. However, a single transfusion of F1 (B6 × bm1) splenocytes into B6 mice without any additional immune modulation leads to permanent acceptance of F1 skin grafts (MST > 100 d, n = 12), (P < 0.001), while third-party BALB/c control grafts were acutely rejected (MST 12 d, n = 4) (Figure 1A).

Induction of stable skin-graft tolerance with donor-specific transfusion (DST) and long-term persistence of F1 (B6 × bm1) donor cells in vivo. A, Long-term F1 skin-graft survival after a single pretransplantation DST. B6 mice (H-2b) received 3 × 107 donor cells i.v. 7 d before skin grafting. F1 (B6 × bm1) skin grafts on B6 mice after F1 (B6 × bm1) DST (▀ n = 12); F1 (B6 × bm1) skin grafts on B6 mice without DST (♦ n = 5), bm1 (H-2bm1) skin grafts on B6 mice after bm1 DST (• n = 9); bm1 skin grafts on B6 without DST (▴n = 4), BALB/c (H-2d) grafts on B6 mice with F1 (B6 × bm1) DST (▾ n = 4). B, Hematoxylin and eosin staining of F1 (B6 × bm1) skin grafts at d 14 (a) or d 100 posttransplant (c) from mice receiving F1 (B6 × bm1) DST; F1 (B6 × bm1) skin graft at d 14 posttransplant (b) following no DST, and a third party BALB/c skin graft at d 12 posttransplant (d) following an F1 DST. Syngeneic graft control (e). F1 graft at 100 d after DST (f).

Histological analysis shows accepted F1 grafts have minimal cell infiltrate with intact skin structure (Figure 1Ba) compared to rejection with massive cellular infiltration and total destruction of epidermis in no DST control mice at day 14 (Figure 1Bb). At day 100, accepted F1 grafts following F1 DST showed no signs of macroscopic (Figure 1Bf) similar to syngeneic controls (Figure 1Be). However, third-party BALB/c skin grafts following F1 DST show rejection with cellular infiltrates at day 12 (Figure 1Bd).

Graft Acceptance Is Associated With Donor Cell Persistence In Vivo

We further investigated the survival of transfused cells in vivo in relation to graft survival. To perform this, splenocytes from donor and recipient strains were stained with low-dose CFSE or high-dose CFSE, respectively. Equal numbers of stained cells from donor and recipient origin were cotransferred into hosts and the ratio of the 2 cell populations was determined (CFSE high to CFSE low); this ratio reflects the rate of donor cell removal (Figure 2A). We found graft acceptance is associated with donor cell persistence in vivo (Figure 2A and B). At day 6 after DST, 68% of F1 splenocytes compared with 39% of bm1 splenocytes (relative to equal numbers of CFSE labeled host cells) persisted in the recipient mice. At day 21, 60% of F1 splenocytes were maintained while bm1 splenocytes were undetectable. At day 97, 25% of F1 splenocytes were still present. However, controls consisting of fully mismatched BALB/c splenocytes were undetectable after day 3 (Figure 2B).

Graft acceptance is associated with donor cell persistence in vivo. A and B, 3 × 107 of 5- (and 6-) carboxyfluorescein succinimidyl ester (CFSE) low labeled host splenocytes and equal numbers of CFSE high labeled donor splenocytes from F1 (B6 × bm1), bm1 or BALB/c mice were cotransferred i.v. respectively into 3 groups of recipient B6 mice. Tail blood was collected and ratio of labeled donor/host cells was assessed by flow cytometry from d 1 through to d 97 (B), ▀ F1/B6 ratio, n = 3, mean ± SD; □ bm1/B6 ratio, n = 3, mean ± SD; and BALB/c/B6 n = 1. natural killer (NK)-cell inhibition prolongs in vivo persistence of transfused allogeneic donor cells. B6 mice received NK-cell depletion antibody NK 1.1 or isotype control ▀ antibody intraperitoneally 2 times over 48 h. Twenty-four h after first NK 1.1 antibody injection, 2 × 107 CFSE high labeled splenocytes from (C) bm1 or (D) BALB/c mice were cotransferred i.v, respectively with equal numbers of CFSE low labeled host cells into B6 mice. Tail blood was collected at 24, 48, and 72 h after cell transfer and the ratio of labeled donor: host cells were assessed by flow cytometry. To assess whether NK cells playing a role in skin-graft rejection, B6 mice received NK-cell depletion antibody NK 1.1 (n = 5) or isotype control antibody (n = 5) intraperitoneally 2 times over 48 h. Twenty-four h after first NK 1.1 antibody injection, bm1 skin grafting was performed (E). To assess the effect of multiple donor-specific transfusion (DST) on skin-graft survival (F), 3 × 107 donor cells from bm1 mice were transfused i.v. into 2 groups of B6 recipients, bm1 skin and third-party skin grafting was performed 7 d after. Once group of B6 mice received a single DST (n=5) bm1 •, Balb/c ♦. A second B6 group (n=4) was given 2 additional DSTs at time of skin grafting and 7 d after, bm1 ▀, Balb/c ▾ skin graft survival shown.

NK-cell Depletion Prolongs Persistence of Transfused Allogeneic Cells In Vivo

Next, we investigated whether the rapid removal of bm1 and BALB/c cells is due to removal by NK cells which is limited in semiallogeneic F1 DST which express inhibitory host MHC.31 B6 mice were injected with an NK1.1 depletion antibody, leading to 90% of NK-cell depletion at 24 hours after injection. CFSE high labeled splenocytes from bm1 or BALB/c mice were cotransferred with CFSE low labeled B6 host cells into B6 mice, and the ratio of CFSE stained donor and host cells were evaluated.

Without NK depletion (black bars), the transferred bm1 cells were reduced to 67% at 24 hours, 61% at 48 hours, and 48% at 72 hours (Figure 2C). However, the rapid removal of bm1 cells was inhibited after NK-cell depletion (gray bars) with 93%, 87%, and 63% of bm1 cells detectable at 24, 48, and 72 hours, respectively (Figure 2C).

The same effect was found when using fully mismatched BALB/c cells. In the presence of NK cells (black bars), BALB/c cells were removed faster with a decrease in circulating BALB/c cells to 62% at 24 hours, 26% at 48 hours, and elimination after 72 hours (Figure 2D). However, after NK-cell depletion (gray bars) BALB/c cells showed prolonged survival with a retention rate of 82%, 47%, and 12% at 24, 48, and 72 hours, respectively (Figure 2D). Further, depletion of NK cells showed no increased survival for bm1 grafts on B6 mice, suggesting its role was primarily due to prolonging the DST survival (Figure 2E). Additional bm1 DSTs at the time of skin transplantation and 1 week after did not lead to enhanced skin-graft survival (Figure 2F).

Tolerant Mice Show Reduced Proliferation and IFN-γ Production In Vitro

Proliferation of responder cells from F1 DST tolerant mice was compared with non-DST and naive B6 controls in a standard mixed lymphocyte reaction in triplicate in 96-well plates (Figure 3A). Cells from F1 tolerant mice (DST+) demonstrated significantly reduced proliferation in response to F1 splenocytes (black bars), (cpm = 3860 ± 196.3) in contrast to proliferation of naive B6 responders to F1 stimulators (cpm = 19 340 ± 4579.7), (P = 0.002) and non-DST controls to F1 stimulators (cpm = 39 063 ± 2349.5), (P < 0.001) (Figure 3A). Despite the significantly reduced response to F1 stimulators, splenocytes from F1 tolerant mice maintained their proliferative response to allogeneic third-party B10.Br stimulators (striped bars), (cpm = 29 719 ± 1106), similar to responders from naive B6 mice (cpm = 35 343 ± 3032) and non-DST controls (cpm = 41 734 ± 6411) (Figure 3A).

Splenocytes of tolerant mice exhibit reduced proliferation and decreased interferon (IFN)-γ production specific to the donor strain in an mixed lymphocyte reaction (MLR). A, MLR assay. 3 × 105 of splenocytes from B6 mice (naive), B6 mice tolerant to F1 (B6 × bm1) graft after donor-specific transfusion (DST) (DST+), and B6 mice that rejected F1 (B6 × bm1) graft without DST (DST−) were stimulated with equal numbers of irradiated splenocytes from naive B6 □, F1 (B6 × bm1) ▀, and third-party B10.Br mice in triplicate wells of the 96-well plate. Proliferation was assessed at 72 h by 3H-Thymidine incorporation. Results were expressed as the average of the 3 wells in counts per min (cpm). B, IFN-γ secretion assay. 3 × 105 of splenocytes from B6 (naive), B6 tolerant to F1 (B6 × bm1) graft after (DST+), and B6 rejected F1 (B6 × bm1) without DST (DST−) stimulated with equal numbers of mitomycin C-treated splenocytes from naive B6 □, F1 (B6 × bm1) ▀, and third-party B10.Br mice in triplicated wells of the IFN- γ antibody precoated 96-well plate. IFN-γ production was assessed at 24 h using ELISPOT (enzyme-linked immunospot). Results were expressed as the average spot-forming cells /2.5 × 105 splenocytes in triplicated wells.

In vitro, IFN-γ production in response to allostimulation was evaluated using ELISPOT (Figure 3B). Responder cells from F1 DST tolerant mice had a significantly reduced number of IFN-γ producing cells in response to F1 stimulators (black bars), (158 ± 47.6 spots/2 × 105 cells) compared to non-DST controls (689 ± 74.2 spots/2 × 105 cells), (P < 0.001) (Figure 3B) and naive B6 controls (137 ± 26 spots/2 × 105 cells). Responder cells from F! DST mice maintained their response to third-party stimulation with B10.Br cells (striped bars), (874 ± 147.6 spots/2 × 105 cells) similar to non-DST controls (895 ± 175.8 spots/2 × 105 cells) (P = 0.88) and naive B6 controls (680 ± 84 spots/2 × 105 cells) (P = 0.12) (Figure 3B).

Both CD4+CD25+ and CD4+CD25 Cells From Tolerant Mice Are Suppressive In Vitro in a Mixed Lymphocyte Reaction

To evaluate the suppressive capacity of CD4+ subsets in a standard tolerance suppression assays, CD4+CD25+ and CD4+CD25 cells were isolated from tolerant F1 DST mice and naive B6 control mice. These cells were added to a mixed lymphocyte reaction consisting of purified B6 CD4+CD25 cells as responders and F1 splenocytes as stimulators.

CD4+CD25+ cells derived from tolerant mice (black bars) exhibited more potent functional suppression than CD25+ cells from naive B6 mice (open bars) on a cell per cell basis, showing 88% inhibition of proliferation at a 1:1 suppressor/responder ratio, 73% at a 1:2 ratio and 60% suppression at a 1:4 ratio (Figure 4A). This was compared with naive CD25+ cells (white bars) showing 62% suppression at 1:1, 67% at 1:2 and 33% at 1:4 suppressor/responder ratio (Figure 4A). Furthermore, CD4+CD25 cells from tolerant mice (black bars) contained cells with regulatory capacity, showing inhibition of 65%, 48%, and 44% at suppression/responder ratio of 1:1, 1:2, and 1:4, respectively (Figure 4B), while no inhibitory effects were observed from the naive control (white bars) (Figure 4B). This dose-dependent response suggests a regulatory component in the CD4+CD25 fraction of the CD4+ T cells from tolerant mice.

In vitro regulatory capacity of CD4+CD25+ and CD4+CD25 cells from tolerant mice. 2 × 105 of CD4+CD25 responder cells from a naive B6 spleens were cocultured with an equal number of irradiated F1 (B6 × bm1) splenocytes in triplicated wells in the 96-well plate. To assess regulatory capacity of CD4 cells, (A) beads isolated CD4+CD25+ and (B) CD4+CD25 cells from B6 mice tolerant to F1graft after donor-specific transfusion (DST) ▀ and naive B6 control mice □ were added to obtain the suppressor: responder ratios of 1:1, 1:2, 1:4, and 1:16. Cell proliferation was assessed at 72 h by 3H-Thymidine incorporation. Results were expressed as the average of the triplicate wells in counts per min (cpm). Background proliferation of CD4+CD25+ cells, cpm = 146 ± 4.6 and background proliferation of CD4+CD25 cells, cpm = 1235 ± 84. (Representative of 2 independent assays.)

Tolerance Is Dominant and Is Transferable by Cells That Have Infiltrated the Graft

To assess whether the active tolerance occurred in vivo, mice that had received DSTs and had F1 skin-grafts intact (MST > 100 d) were transfused with naive B6 splenocytes and regrafted with a second F1 and a third-party B10.Br skin graft. Both the original and the retransplanted F1 grafts demonstrated long-term survival (MST > 60 d, n = 6) while third-party B10.Br grafts were rapidly rejected (MST 10 d, n = 5), (P < 0.001) (Figure 5A). Control B6 mice that received naive B6 splenocytes also rapidly rejected F1 grafts (MST 17 d, n = 5), (P < 0.001) (Figure 5A).

Dominant tolerance was demonstrated and this was transferrable by the graft onto immunodeficent mice. A, Dominant tolerance was demonstrated in F1 donor-specific transfusion (DST) tolerant mice as F1 grafts were not rejected following challenge with naive B6 splenocytes. F1 DST tolerant B6 hosts (n = 6) were regrafted with a fresh F1 graft (▀n = 6) and third-party grafts B10.Br (H-2k) (•n = 5) and then challenged with 3 × 107 fresh naive B6 splenocytes through tail vein. Control naive B6 mice were grafted with a F1 grafts (▴n = 5) and injected with 3 × 107 fresh naive B6 splenocytes via the tail vein. B, Homeostatic proliferation of graft infiltrates in Rag1−/− mice and transfer of tolerance through graft infiltrates. F1 grafts dissected from tolerant B6 hosts were transplanted onto Rag-1−/− mice (n = 4). Forty d after skin-graft transplantation, flow cytometric analysis was conducted on cells from tail blood showed the majority of expanded T cells are CD4+, of which 4.1% are forkhead box P3 (Foxp3) positive. C, At 45 d after tolerant skin-graft transplantation, the Rag-1−/− hosts were regrafted with a fresh F1 graft (▀n = 4) and third-party Balb/c graft (•n = 4) and were reconstituted i.v, with 0.5 × 106 naive B6 splenocytes. A control group of Rag-1−/− mice (n = 4) were grafted with a syngeneic B6 graft and regrafted with a fresh F1 graft (▴n = 4) and third-party BALB/c graft (♦n = 4) and reconstituted with the same number of naive B6 splenocytes.

We further assessed if immune regulatory cells in the tolerant grafts can be expanded in secondary hosts and exert dominant tolerance as previously shown from tolerant skin and kidney grafts.26,32,33 Tolerant F1 skin grafts, as well as control B6 skin grafts, were regrafted on Rag1−/− mice. T-cell engraftment was detected in Rag1−/− mice 40 days after the tolerant F1 graft transplantation, with almost all expanded cells being CD4+ cells (Figure 5B) and 4.12% of which were Foxp3 positive (Figure 5C). No T cells were found in Rag1−/− mice with B6 grafts (data not shown). On day 45, all mice were reconstituted with naive B6 splenocytes and challenged with a fresh F1 and third-party BALB/c grafts. F1 grafts on Rag1−/− mice with CD4 cell expansion from tolerant F1 skin grafts exhibited prolonged graft survival (MST 77 d, n = 4) (Figure 5D). F1 grafts on control Rag1−/− mice with B6 grafts were rejected significantly earlier (MST 40 d, n = 3) (P < 0.05) (Figure 5D). Both groups rejected their third-party BALB/c grafts at a similar time (MST 11 d, n = 3 and MST 12 d, n = 4) (P = NS) (Figure 5D).

Graft Acceptance Is Associated With Increased CD4+Foxp3+ Cells

Tolerant mice challenged with naive splenocytes maintained tolerance. Further transfer studies also demonstrated tolerance. Given the dominant nature of the tolerance in this model, we then quantified Treg using Foxp3.

As shown in Figure 5, mice tolerant to F1 grafts a number of CD4+Foxp3+ cells in the spleen (23.3 ± 2.7%, n = 3) compared to mice that received F1 grafts without DST (18.7 ± 0.14%, n = 3), (P = 0.025) and naive mice (14.3 ± 1.7%, n = 3), (P = 0.0012). Interestingly, a CD25 subset of CD4+Foxp3+ T cells were expanded in tolerant mice (Figure 6A), suggesting that expanded Treg did not express CD25 or had downregulated CD25 expression.26

Increased forkhead box P3 (Foxp3) expressing cells in tolerant mice. A, Flow cytometric analysis of Foxp3 expression on splenocytes from tolerant mice (DST+), F1 grafted B6 mice without donor-specific transfusion (DST) (DST−) and naive B6 mice. Values are the mean of 3 mice in each group with a representative dot-plot. B, Foxp3 expression using real-time quantitative polymerase chain reaction (PCR) on F1 graft □, third-party B10.Br graft and syngeneic B6 graft ▀ on B6 mice with F1DST (DST+) and F1 graft □ on B6 mice without DST (DST−) on d 12 after transplantation. Foxp3 expression is relative to T-cell receptor (TCR) beta constant region expression to reflect the Treg proportion of the total T-cell infiltrate. C, Foxp3 expression in draining (DLN), nondraining lymph node (NLN), and spleen (SP) from tolerant mice. Foxp3 expression is relative to TCR beta constant region expression to reflect Treg proportion out of total T cells in the lymphatic organ. *P < 0.05, **P < 0.001. D, Foxp3 immunohistochemistry staining of tolerant F1 skin graft.

Real-time polymerase chain reaction was used to quantitate Foxp3 relative to the expression of the TCR constant region, which represents total T-cell infiltrates. Foxp3 expression in tolerant F1 grafts was increased 4-fold compared to third-party B10.Br grafts and F1 grafts on non-DST control mice (Figure 6B), reflecting the increased Treg/eff ratio of T-cell infiltrates in tolerant grafts.12,34 We compared Foxp3 expression in the spleen, nondraining lymph-nodes, and draining lymph-nodes in tolerant mice (Figure 6C). Expression of FoxP3 in the draining lymph-nodes of tolerant mice was increased 4-fold, compared to nondraining lymph-nodes and spleens in the same mice at day 12 after F1 skin grafting (Figure 6C). Foxp3 cells were found in the tolerant grafts by immunohistochemistry for Foxp3 (Figure 6D).

Foxp3 Treg Expansion Is Due to nTreg Proliferation

Increased Foxp3 Treg could potentially come from proliferation of nTreg or from conversion of CD4+Foxp3 cells. To test this, CD4+ cells from a Ly5.1 (CD45.1) Foxp3GFP congenic strain were flow-sorted intoCD4+Foxp3GFP+ and CD4+Foxp3GFP− cell populations. Sorted cells were stained with CellTrack Violet proliferation dye and transferred into separate B6 hosts followed by F1 DST at day 0 and skin grafting at day 7. At day 12, transferred CD45.1+CD4+Foxp3GFP− cells and CD45.1+CD4+Foxp3GFP+ were gated on CD45.1 and analyzed for green fluorescence protein (GFP) expression to assess the occurrence of Treg conversion by flow cytometry (Figure 7A). After donor cell stimulation, there was no GFP expression observed in transferred CD45.1+CD4+Foxp3GFP− cells (Figure 7A), indicating de novo conversion of Foxp3+ cells from naive CD4 cells has not occurred. In contrast, there is significant proliferation of nTreg after in vivo F1 donor cell stimulation (Figure 7B), representing a significant expansion (43%) of alloreactive Treg. The lower proliferation rate (22%) in the syngeneic cell transfer control group may reflect proliferation of Treg driven by the skin allograft.34

Forkhead box P3 (Foxp3) cell expansion is due to nTreg proliferation, and deletion of Foxp3 Treg abrogates skin-graft tolerance. Flow cytometry sorted CD45.1+ CD4+Foxp3GFP+ (nTreg) and CD45.1+CD4+Foxp3GFP− cells were stained with a proliferation dye (Celltrace Violet) and transfused separately into different recipient B6 mice. Mice were then transfused i.v. with F1 (B6 × bm1) or B6 splenocytes (syngeneic control) on the following d and F1 skin grafting on d 7 after cell transfusion. Flow cytometric analysis was performed at d 12 after donor-specific transfusion (DST) on splenocytes. A, Transferred CD45.1+CD4+Foxp3GFP− cells were gated on green fluorescence protein (GFP) expression to assess the occurrence of Treg conversion. There is no GFP expression on CD45.1+CD4+Foxp3GFP− after F1 donor cell transfusion (left). GFP expression in transferred CD45.1+CD4+Foxp3GFP+ cells was readily detected (right). B, Transferred CD45.1+CD4+Foxp3GFP+ cells were gated for assessment of proliferation in reaction to F1 and syngeneic B6 cell transfusions. C and D, Foxp3 Treg depletion abrogates DST-induced skin-graft tolerance. Depletion of Treg (DEREG) mice (♦ n = 5) and its wild-type littermate control (▀n = 4) received F1 (B6 × bm1) DST. To deplete Foxp3 Treg weekly diptheria toxin (DT) treatment by i.p injection was started 4 d after DST. Tail blood collected 3 d after DT treatment showed (C) >99% of Foxp3 Treg were deleted in DEREG mice. D, DEREG mice and littermate controls were then transplanted with F1 skin grafts 7 d after DST and weekly DT treatment continued until skin-graft rejection was observed.

Depletion of Foxp3 Treg Abrogates Skin-graft Tolerance

To assess the role of Treg in maintaining tolerance in this model, DEREG mice were used for specific depletion of Foxp3 Treg. DT was given to DEREG mice and their littermate controls with 99% of the Foxp3 GFP+ cells depleted after DT injection (Figure 7C). Foxp3 Treg depleted DEREG mice rapidly rejected their F1 skin grafts (MST 22 d, n = 4). In contrast, the littermate control group showed long-term F1 graft survival (MST > 100 d, n = 5) (P = 0.0027) (Figure 7D).


The bm1 model has been well studied as a potential way to induce delayed rejection using DSTs with a role for CD47 in protecting against donor cell loss and prolonging grafts and a potential role for a regulatory double negative T-cell subset also in prolonging bm1 graft survival.35-37 We have used this model with the F1 mice to develop a robust model of tolerance. In this study, we demonstrate that the persistence of transfused semiallogeneic donor cells mismatched at MHC class I can enhance tolerance to subsequent skin allografts through indirectly expanded nTreg leading to dominant tolerance without additional immunological manipulation. Achieving Treg driven tolerance may have importance for the current development of human Treg and mixed chimerism trials of organ transplant tolerance.

The importance of persistence of donor cells after DST in tolerance induction has been shown in several ways. Donor cells are required to persist for at least 3 days to limit host reactions38 and multiple DSTs, which extend the presence of donor cells in circulation, assist in promoting graft tolerance.6 In this study, we observed, using long-term donor cell tracking that donor cell persistence correlated with allografts with long-term tolerance. The rapid removal of transfused cells appears due to host innate immunity. NK cells recognize and eliminate cells that fail to express self-class I MHC molecules (ie, missing-self hypothesis). Secondary inflammation from early NK-cell activation may also mature antigen-presenting cells and enhance Th1-like responses promoting allograft injury.39 The level of NK-cell activation has previously been shown to be inversely correlated with graft acceptance.23,38 In our in vivo donor cell-tracking study, the finding of long-term persistence of semiallogeneic F1 cells and reduced elimination of bm1 and BALB/c donor cells after NK-cell depletion is consistent with the need for expression of self-antigens to limit NK cytotoxicity and promote donor cell survival. Therefore, the persistence of low-level mixed chimerism resulting from F1 donor cell persistence may be an additional element enhancing tolerance through Treg expansion.32,40 The presence of donor cells in circulation and lymphatic organs6,23 provides continuous antigen stimulation in an environment without inflammation, which can expand allospecific Treg.20-23,32

Allospecific Treg are thought to be mainly derived from expanded crossreactive clones arising from naturally occurring Treg (nTreg) arising from the thymus. However, Foxp3+ Treg arising from de novo conversion of naive CD4+ T cells (iTreg) has also been reported in models using costimulatory blockade.41-44 In the present study, using a model without immunosuppression, we observed significant proliferation of nTreg after in vivo donor cell stimulation, while de novo conversion of Foxp3+ cells from naive CD4 cells was not observed. The expansion of allospecific nTreg predominating over Treg conversion is consistent with that previously shown in an islet transplant model using fate-mapping with fluorescent tagging of Treg.34

The critical role of regulatory cells has been reported in DST models with mismatch at both MHC class I and class II,4,6-10 where alloreactive Treg has expanded through direct and indirect pathways. Due to the much higher precursor frequency of Treg activated directly as compared with indirect activation, expanded Treg are likely to occur predominantly through the direct pathway of allorecognition in those models.40,41 However, studies have shown that transplantation tolerance depends predominantly on indirectly activated Treg in long-term graft acceptance.7,14-19,45 Indeed, modified semiallogeneic DCs, which can indirectly activate Treg, have been shown to prevent experimental kidney rejection,17 while eliminating the indirect pathway has been shown to block DST-induced long-term cardiac allograft survival.7 Further, patient data on heart and kidney transplants from the era of donor-specific infusions shows a benefit when HLA-DR is matched allowing indirect antigen presentation.2 In the current model using an MHC class I disparity, the Kbm1 allopeptide can only be indirectly recognized in the context of self-class II MHC by CD4 cells.29 It is likely that the persistent indirect presentation of the Kbm1 derived peptide by both host APCs as well as the self-peptide on donor APCs leads to the proliferation of allospecific nTreg preventing chronic rejection in this model. Previously multiple transfusions before skin transplant have been shown to prolong graft survival.6 We tested whether additional DSTs at the time of grafting and after grafting could also enhance tolerance but found no effect possibly because of activation of effector cells posttransplant. As expected in this model dominant tolerance occurs through indirectly expanded nTreg alone. Dominant tolerance was demonstrated by resistance to challenge with naive splenocytes and transfer of tolerance by tolerant skin-graft infiltrates.32,33 The critical role of Foxp3+ Treg in the allograft tolerance was further demonstrated by loss of skin-graft tolerance after depletion of Foxp3+ Treg using DEREG mice. Further, it is possible that F1 DSTs may improve graft outcome for full bm1 grafts as well as F1 grafts. The presence of CD25 Treg in this model is suggested in our suppression assay and was the reason for using DEREG mice rather than CD25 depletion with PC61. This reflects the more robust data found with DEREG Treg depletion as compared to anti-CD25 (PC61) CD25+ Treg depletion which may miss antigen-specific CD25 Treg in our previous kidney study.26

In conclusion, this study outlines a stringent model of skin-graft tolerance using a single DST without additional immunological manipulation. The induced tolerance is associated with prolonged donor cell persistence leading to and dependent on indirectly expanded allospecific nTreg.


We thank the staff of the Westmead Hospital Animal House Facility for the care of the animals. We thank Dr Xin Maggie Wang and Mr Suat Dervish from Flow Cytometry Facility of Westmead Millennium Institute for their help with cell sorting and flow cytometry analysis and Ms Virginia James of Westmead Millennium Institute for processing HE sections. We thank Dr Grant Logan of Children’s Medical Research Institute and Steve Schibeci of Westmead Millennium Institute for their assistance with mixed lymphocyte reaction tests. We thank Eric Au for his visual abstract.


1. Kingsley CI, Nadig SN, Wood KJ. Transplantation tolerance: lessons from experimental rodent models. Transpl Int. 2007; 20:828–841
2. Lagaaij EL, Hennemann IP, Ruigrok M, et al. Effect of one-HLA-DR-antigen-matched and completely HLA-DR-mismatched blood transfusions on survival of heart and kidney allografts. N Engl J Med. 1989; 321:701–705
3. Iwakoshi NN, Mordes JP, Markees TG, et al. Treatment of allograft recipients with donor-specific transfusion and anti-CD154 antibody leads to deletion of alloreactive CD8+ T cells and prolonged graft survival in a CTLA4-dependent manner. J Immunol. 2000; 164:512–521
4. Quezada SA, Bennett K, Blazar BR, et al. Analysis of the underlying cellular mechanisms of anti-CD154-induced graft tolerance: the interplay of clonal anergy and immune regulation. J Immunol. 2005; 175:771–779
5. Quezada SA, Fuller B, Jarvinen LZ, et al. Mechanisms of donor-specific transfusion tolerance: preemptive induction of clonal T-cell exhaustion via indirect presentation. Blood. 2003; 102:1920–1926
6. Bushell A, Karim M, Kingsley CI, et al. Pretransplant blood transfusion without additional immunotherapy generates CD25+CD4+ regulatory T cells: a potential explanation for the blood-transfusion effect. Transplantation. 2003; 76:449–455
7. Kishimoto K, Yuan X, Auchincloss H Jr, et al. Mechanism of action of donor-specific transfusion in inducing tolerance: role of donor MHC molecules, donor co-stimulatory molecules, and indirect antigen presentation. J Am Soc Nephrol. 2004; 15:2423–2428
8. Lee K, Nguyen V, Lee KM, et al. Attenuation of donor-reactive T cells allows effective control of allograft rejection using regulatory T cell therapy. Am J Transplant. 2014; 14:27–38
9. Ochando JC, Homma C, Yang Y, et al. Alloantigen-presenting plasmacytoid dendritic cells mediate tolerance to vascularized grafts. Nat Immunol. 2006; 7:652–662
10. Sho M, Kishimoto K, Harada H, et al. Requirements for induction and maintenance of peripheral tolerance in stringent allograft models. Proc Natl Acad Sci U S A. 2005; 102:13230–13235
11. Bushell A, Morris PJ, Wood KJ. Transplantation tolerance induced by antigen pretreatment and depleting anti-CD4 antibody depends on CD4+ T cell regulation during the induction phase of the response. Eur J Immunol. 1995; 25:2643–2649
12. Lee I, Wang L, Wells AD, et al. Recruitment of Foxp3+ T regulatory cells mediating allograft tolerance depends on the CCR4 chemokine receptor. J Exp Med. 2005; 201:1037–1044
13. Waldmann H, Chen TC, Graca L, et al. Regulatory T cells in transplantation. Semin Immunol. 2006; 18:111–119
14. Golshayan D, Jiang S, Tsang J, et al. In vitro-expanded donor alloantigen-specific CD4+CD25+ regulatory T cells promote experimental transplantation tolerance. Blood. 2007; 109:827–835
15. Hori S, Kitagawa S, Iwata H, et al. Cell-cell interaction in graft rejection responses: induction of anti-allo-class I H-2 tolerance is prevented by immune responses against allo-class II H-2 antigens coexpressed on tolerogen. J Exp Med. 1992; 175:99–109
16. Joffre O, Santolaria T, Calise D, et al. Prevention of acute and chronic allograft rejection with CD4+CD25+Foxp3+ regulatory T lymphocytes. Nat Med. 2008; 14:88–92
17. Mirenda V, Berton I, Read J, et al. Modified dendritic cells coexpressing self and allogeneic major histocompatability complex molecules: an efficient way to induce indirect pathway regulation. J Am Soc Nephrol. 2004; 15:987–997
18. Niimi M, Roelen DL, Witzke O, et al. The importance of H2 haplotype sharing in the induction of specific unresponsiveness by pretransplant blood transfusions. Transplantation. 2000; 69:411–417
19. Sánchez-Fueyo A, Domenig CM, Mariat C, et al. Influence of direct and indirect allorecognition pathways on CD4+CD25+ regulatory T-cell function in transplantation. Transpl Int. 2007; 20:534–541
20. Scully R, Qin S, Cobbold S, et al. Mechanisms in CD4 antibody-mediated transplantation tolerance: kinetics of induction, antigen dependency and role of regulatory T cells. Eur J Immunol. 1994; 24:2383–2392
21. Sayegh MH, Zheng XG, Magee C, et al. Donor antigen is necessary for the prevention of chronic rejection in CTLA4IG-treated murine cardiac allograft recipients. Transplantation. 1997; 64:1646–1650
22. Yates SF, Paterson AM, Nolan KF, et al. Induction of regulatory T cells and dominant tolerance by dendritic cells incapable of full activation. J Immunol. 2007; 179:967–976
23. Sheng-Tanner X, Miller RG. Correlation between lymphocyte-induced donor-specific tolerance and donor cell recirculation. J Exp Med. 1992; 176:407–413
24. Westerhuis G, Maas WG, Willemze R, et al. Long-term mixed chimerism after immunologic conditioning and MHC-mismatched stem-cell transplantation is dependent on NK-cell tolerance. Blood. 2005; 106:2215–2220
25. Watson D, Zhang GY, Sartor M, et al. “Pruning” of alloreactive CD4+ T cells using 5- (and 6-)carboxyfluorescein diacetate succinimidyl ester prolongs skin allograft survival. J Immunol. 2004; 173:6574–6582
26. Hu M, Wang C, Zhang GY, et al. Infiltrating Foxp3(+) regulatory T cells from spontaneously tolerant kidney allografts demonstrate donor-specific tolerance. Am J Transplant. 2013; 13:2819–2830
27. Wang YM, Zhang GY, Wang Y, et al. Foxp3-transduced polyclonal regulatory T cells protect against chronic renal injury from adriamycin. J Am Soc Nephrol. 2006; 17:697–706
28. Polhill T, Zhang GY, Hu M, et al. IL-2/IL-2ab complexes induce regulatory T cell expansion and protect against proteinuric CKD. J Am Soc Nephrol. 2012; 23:1303–1308
29. Ossevoort MA, De Bruijn ML, Van Veen KJ, et al. Peptide specificity of alloreactive CD4 positive T lymphocytes directed against a major histocompatibility complex class I disparity. Transplantation. 1996; 62:1485–1491
30. Kitagawa S, Iwata H, Sato S, et al. Heterogenous graft rejection pathways in class I major histocompatibility complex-disparate combinations and their differential susceptibility to immunomodulation induced by intravenous presensitization with relevant alloantigens. J Exp Med. 1991; 174:571–581
31. Yagisawa T, Tanaka T, Miyairi S, et al. In the absence of natural killer cell activation donor-specific antibody mediates chronic, but not acute, kidney allograft rejection. Kidney Int. 2019; 95:350–362
32. Domenig C, Sanchez-Fueyo A, Kurtz J, et al. Roles of deletion and regulation in creating mixed chimerism and allograft tolerance using a nonlymphoablative irradiation-free protocol. J Immunol. 2005; 175:51–60
33. Graca L, Cobbold SP, Waldmann H. Identification of regulatory T cells in tolerated allografts. J Exp Med. 2002; 195:1641–1646
34. Fan Z, Spencer JA, Lu Y, et al. In vivo tracking of ‘color-coded’ effector, natural and induced regulatory T cells in the allograft response. Nat Med. 2010; 16:718–722
35. Young KJ, Yang L, Phillips MJ, et al. Donor-lymphocyte infusion induces transplantation tolerance by activating systemic and graft-infiltrating double-negative regulatory T cells. Blood. 2002; 100:3408–3414
36. Wang H, Wu X, Wang Y, et al. CD47 is required for suppression of allograft rejection by donor-specific transfusion. J Immunol. 2010; 184:3401–3407
37. Hu Y, Zhou H, Gao B, et al. Role of regulatory T cells in CD47/donor-specific transfusion-induced immune tolerance in skin-heart transplantation mice. Transpl Infect Dis. 2019; 21:e13012
38. Sheng-Tanner XF, Miller RG. Correlation between lymphocyte-induced donor-specific tolerance and donor cell recirculation. Role of class I and class II major histocompatibility complex. Transplantation. 1994; 57:1081–1087
39. Gill RG. NK cells: elusive participants in transplantation immunity and tolerance. Curr Opin Immunol. 2010; 22:649–654
40. Zuber J, Sykes M. Mechanisms of mixed chimerism-based transplant tolerance. Trends Immunol. 2017; 38:829–843
41. Burrell BE, Bromberg JS. Fates of CD4+ T cells in a tolerant environment depend on timing and place of antigen exposure. Am J Transplant. 2012; 12:576–589
42. Ferrer IR, Wagener ME, Song M, et al. Antigen-specific induced Foxp3+ regulatory T cells are generated following CD40/CD154 blockade. Proc Natl Acad Sci U S A. 2011; 108:20701–20706
43. Gao W, Lu Y, El Essawy B, et al. Contrasting effects of cyclosporine and rapamycin in de novo generation of alloantigen-specific regulatory T cells. Am J Transplant. 2007; 7:1722–1732
44. Vu MD, Xiao X, Gao W, et al. OX40 costimulation turns off Foxp3+ Tregs. Blood. 2007; 110:2501–2510
45. Tsang JY, Tanriver Y, Jiang S, et al. Conferring indirect allospecificity on CD4+CD25+ Tregs by TCR gene transfer favors transplantation tolerance in mice. J Clin Invest. 2008; 118:3619–3628
Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.