Mechanisms of donor-specific transfusion synergy with anti-CD154 for transplant acceptance

Abstract

Organ transplantation is a life-saving treatment for patients with end-stage organ failure but requires lifelong immunosuppression that can result in significant complications. Achieving stable and durable donor-specific tolerance, whereby immunosuppression can be stopped without precipitating graft rejection, holds the promise to circumvent these problems. While transient inhibition of the CD40/CD154 costimulatory pathway delays transplant rejection in animal models, and antibodies blocking this interaction are currently in clinical trials, the efficacy of anti-CD154 (αCD154) in mouse models is significantly enhanced by the addition of donor splenocyte transfusion (DST). Indeed, αCD154+DST but not αCD154 alone, can successfully induce donorspecific transplantation tolerance to fully mismatched cardiac allografts in mice. Why DST needs to be added to αCD154 is not fully understood. By integrating tracking of graft-specific T cells and of donor cells, we show that systemic, but not subcutaneous, injection of DST enables alloantigen dissemination to secondary lymphoid organs beyond those directly draining the transplant. This wider biodistribution results in a greater number of alloreactive T cells interacting with donor alloantigens in all lymphoid organs, such that more alloreactive T cells can be the target of CD154 blockade. Furthermore, the duration of DST persistence, facilitated by the sharing of MHC alleles between the donor and the recipient, emerges as a critical factor in promoting αCD154-mediated graft acceptance. These results provide insights into rational approaches to improve translation of αCD154 in the clinic.

Graphical Abstract

Introduction

Organ transplantation is the best option for select patients with end-stage organ failure, but usually requires lifelong maintenance immunosuppression to prevent rejection. However, immunosuppressive drugs can induce significant toxicity and because they globally suppress immunity, they are associated with an increased risk of infections and tumors, and ultimately, have limited effectiveness against chronic rejection^1-5. The induction of donor-specific transplantation tolerance could avoid the need for lifelong, global immunosuppression while significantly enhancing the well-being of transplant recipients.

Activation of naïve T cells (including alloreactive T cells) in secondary lymphoid organs depends on engagement of the T cell receptor (TCR) with its cognate antigen and costimulation via binding of CD28 on T cells to B7 ligands (CD80/CD86) on antigen-presenting cells (APCs). Upon activation, T cells upregulate CD154 (CD40L) which, following binding to CD40 and CD11b receptors on APCs, drives further APC expression of MHC and CD80/CD86 molecules, thus amplifying T cell activation⁶. In particular, engagement of CD28 and of CD154 is critical for the ability of T cells to help B cells undergo affinity maturation in germinal centers and produce alloantibodies thought to play a role in chronic antibody-mediated rejection (ABMR)^7-11. Interestingly, transient blockade of CD154/CD40/CD11b interactions with αCD154 has been more efficacious at prolonging graft survival than transient blockade of CD28/B7 interactions using CTLA4-Ig¹². This may be in part because blockade of CD154/CD40/CD11b interactions can expand regulatory T cells (Tregs) in a mechanism that depends on inhibiting IL-1β production by DCs¹³. CTLA4-Ig may be less tolerogenic because Tregs rely on CD28-mediated signals for their development and survival, and because CTLA-4 engagement is important for both Treg suppression and intrinsic inhibition of conventional T cell alloresponses^14-16, all of which would be impaired by CTLA4-Ig¹⁷.

Inhibiting the CD154/CD40/CD11b pathway is seen as promising for the prevention of transplant rejection. However, while transient blockade of CD154 prevents alloantibody production and prolongs cardiac allograft survival in mice¹⁸, it fails as monotherapy to induce permanent graft acceptance in murine recipients of major-mismatched skin and heart transplants^12,18,19. Notably, adding a single transfusion of donor splenocytes (DST) to the short-term αCD154 therapy, either one week prior to transplantation or at the time of transplantation, improves graft acceptance and even induces donor-specific tolerance after cardiac transplantation, defined as spontaneous acceptance of a secondary donor-matched heart allograft in the absence of immunosuppression²⁰. Remarkably, long-term graft acceptance has been achieved through the combined treatment with αCD154+DST in experimental mouse models encompassing skin, heart, and islet transplantation^19,21,22. However, despite decades of observing the benefit of DST on graft survival, why DST needs to be added to αCD154 is not clear.

Historical observational studies described the phenomenon of transfusion-related immunomodulation (TRIM). For example, renal transplant studies in the 1980s reported that patients who received pre-transplant DST had improved graft survival compared to those that did not^23,24. The effect was ascribed to leukocytes in the transfusions, but leukoreduced MHC class I-expressing fresh platelets have also been shown to mediate enhanced skin transplant survival in mice²⁵. However, concerns with sensitization risks led to abandoning DST in the clinic, when better immunosuppressive therapies were introduced²⁶.

αCD154 was tested in humans over 2 decades ago, but initial clinical trials were halted because of thromboembolism events due to crosslinking of CD154 on human platelets and of Fcγ receptors^27,28. In the intervening years, αCD154 was fully humanized and its Fc portion mutated to prevent FcγR cross-linking²⁹ to avoid thrombotic complications^30,31. Current modified αCD154 antibodies have re-entered clinical trials in settings of kidney transplantation and autoimmunity [ClinicalTrials.gov Identifier: NCT05027906; NCT06305286]. However, in the absence of DST, this humanized non-FcγR-binding antibody may not reach its full potential. It is therefore critical to understand the mechanisms by which DST synergizes with αCD154 to induce transplantation tolerance, to devise an acceptable translational alternative.

Our study identifies DST’s pivotal role in enabling systemic dissemination of alloantigen beyond the secondary lymphoid organs draining the transplant, thus enhancing alloantigen presentation such that a greater fraction of alloreactive T cells encounter their cognate antigen, upregulate CD154, and therefore can be targeted by αCD154. In addition, we report that longer persistence of DST plays a role in its ability to promote αCD154-mediated graft acceptance. These results help conceptualize approaches to translate the systemic dispersal of alloantigen under cover of αCD154 to position αCD154 for even greater benefit in the clinic.

Materials and Methods

Mice

Male and female mice were utilized throughout the study, with efforts made to ensure age matching whenever feasible. Six- to eight-week-old C57BL/6 (B6) and BALB/c (B/c) mice were obtained from Harlan Envigo TMS. 2W-mOVA-transgenic (Tg) mice on the B6 background (B6.2W-OVA), which express the fused transmembrane model antigens 2W and ovalbumin (OVA) protein under the control of the actin promoter, were obtained from Dr. James Moon (Harvard Medical School). 2W-mOVA-Tg males were crossed to B/c females to generate B/c x B6 2W-mOVA-Tg F1 mice (2W-OVA F1). OT-II-Tg mice, comprising CD4⁺ T cells recognizing the OVA323-339 peptide presented on I-A^b were obtained from Jackson Laboratories (JAX Stock #004194) and crossed to CD45.1⁺Rag1^−/− mice to generate congenically marked CD45.1⁺ OT-II Rag^−/− mice (OT-II). CD45.1⁺ mice (#002014) were also purchased from Jackson laboratories. All mice were housed in a specific pathogen-free facility and utilized in compliance with the guidelines set forth by the University of Chicago Institutional Animal Care and Use Committee (protocol #71095), adhering to the National Institutes of Health guidelines for animal use.

Skin transplantation

Skin transplantation was conducted as previously described³². Briefly, donor mice were euthanized by cervical dislocation under ketamine anesthesia, and their tail skin was transplanted onto the shaved flank of recipient mice. The grafts were covered with Vaseline-coated gauze for 7 days, then monitored three times weekly for size, hair presence, pigmentation, and signs of inflammation. Rejection was recorded if less than 20% viable skin tissue remained. Experiments were terminated at d50 if no signs of graft damage were observed in the αCD154-treated groups. Unless otherwise indicated, immunosuppression therapy for mouse skin grafts consisted of the combination of DST administered intravenously (i.v.) on the day of transplantation (day 0), along with 500μg of αCD154 (MR1) blocking mAb (Bio-X-cell) administered i.v. on day 0, and intraperitoneally (i.p.) on days 7 and 14 post-transplantation.

Preparation of DST and TCR-Tg OT-II T cells for adoptive transfer

DST was prepared by homogenizing splenocytes into a single-cell suspension using a 40μm filter. Most DST injections comprised splenocytes derived from approximately one-quarter to one-sixth of a spleen (suspended in 200μL of PBS for i.v. and 60μL for intrasplenic injections).

OT-II cells were obtained from spleens of naive OT-II CD45.1⁺Rag1^−/− mice and CD4⁺ T cells were isolated using a negative selection CD4⁺ T cell enrichment kit (Miltenyi). Cell count was determined using CountBright counting beads (Invitrogen) and a Fortessa flow cytometer (BD Biosciences). A small proportion of these cells were stained with anti-CD4, anti-CD45.1, and anti-Vβ5.1/5.2 antibodies to confirm the genotype before the adoptive transfer.

Flow cytometry

Lymph nodes (LNs) and spleens were recovered from mice, homogenized, and resuspended as single cells. Blood was heparinized and red blood cells lysed in ACK lysis buffer. Cells were counted and incubated in a fixable Live Aqua live/dead stain (Invitrogen) at a dilution of 1:1000 for 20 minutes at room temperature (RT) in the dark. When indicated, pMHC-tetramer staining was performed by incubating 5x10⁶ unenriched splenocytes with both phycoerythrin- and allophycoerytherin-coupled pK^d (PEYWEEQTQRAKSD):I-A^b tetramers (National Institutes of Health Core Facility) for 60 minutes at RT in a dark water bath. Cells were labeled with fluorophore-conjugated antibodies (1:200) targeting the following surface markers: CD4 (GK1.5), CD8α (53-6.7), B220 (RA3-6B2), CD44 (IM7), CD127 (A7R34), CD73 (TY/11.8), PD-1 (29F.1A12), CD45.1 (A20), CD45.2 (104) F4/80 (BM8), CD11b (M1/70), I-A/I-E (M5/114.15.2), CD103 (M290), B220 (RA3-6B2), YAe-IA^b (Eα-peptide) (eBioY-Ae), CD3 (17A2), NK1.1 (PK136). Cells were then fixed and permeabilized with an intracellular Foxp3 staining kit (Invitrogen) and intracellularly stained with anti-Foxp3 (FJK-16s) for 30 minutes at RT in the dark. Events were collected using a Penteon 5-30, Fortessa 4-15, or Fortessa 4-15_HTS flow cytometer at the University of Chicago's CAT Facility (RRID: SCR_017760) and analyzed using FlowJo software (Treestar).

Statistical analyses

Statistical analyses were conducted using GraphPad Prism (GraphPad, La Jolla, California). The statistical tests and the statistical significance are indicated in each figure legend. *p<0.05, **p<0.01, ***p<0.001.

Results

Systemic DST promotes long-term graft acceptance

To determine the conditions and route of administration under which donor-specific transfusion (DST) promotes long-term graft acceptance, we utilized a murine skin transplantation model. Recipient B6 mice received F1 (B6 × B/c).2W-OVA skin grafts and were treated with αCD154 (500μg on days 0, 7, and 14). DST was administered either systemically (i.v. or intrasplenic, i.s.) or locally (s.c.) on day 0 (Figure 1A).

Figure 1. Systemic DST is necessary for αCD154-mediated long-term skin graft acceptance.

(A) Experimental model: Recipient B6 mice underwent skin transplantation with 2W-OVA F1 skin grafts. Recipients were treated with a combination of αCD154 (d0 i.v., d7 and 14 i.p.) and DST (d0) via the indicated administration routes (s.c., i.v., or i.s.). One group was subjected to a splenectomy (ΔSpl) 5 days prior to transplantation. An untreated group (No Rx) served as a control. (B) Kaplan–Meier survival curves were analyzed using the log-rank test. Combination therapy with systemic DST (i.v., i.s., or i.v. + ΔSpl) and αCD154 significantly prolonged graft survival compared to all other groups (p < 0.01). Combination therapy with s.c. DST and αCD154 modestly improved outcomes compared to monotherapy groups (p < 0.05). Monotherapy with αCD154 also outperformed other monotherapies (p < 0.05). The number of mice is indicated in the legend.

As illustrated in Figure 1B, only the combination of αCD154 with systemic administration of DST, either i.v. or i.s., resulted in long-term skin graft survival. These findings highlight not only the importance of adding DST to αCD154, but also the key role of a systemic DST delivery, as localized DST administration via s.c. injection did not drive long-term graft acceptance in most animals. These results suggest that a systemic route of DST administration is critical for its synergy with αCD154.

Given the spleen’s proposed role in immune tolerance induction and antigen presentation^33-35, we tested whether its presence was required for graft acceptance induced by aCD154 + DST. However, i.v. DST was similarly effective at synergizing with αCD154 in intact and splenectomized skin-transplanted mice, indicating that the spleen is not essential for achieving long-term graft acceptance in this model (Figure 1B). Of note and as expected, αCD154 abrogated donor-specific alloantibody production in all cases (data not shown).

I.v. DST enhances the dissemination and accumulation of donor cells

We proposed an “antigen dispersal hypothesis” to explain the importance of systemic DST administration for long-term graft acceptance, positing that i.v. DST delivers circulating donor antigens to all secondary lymphoid organs, while skin graft-derived antigens primarily drain to local lymph nodes. We attempted to track antigen dispersal following transplantation of a B/c skin graft into a B6 host, in the absence of DST administration. However, as previously reported^36,37, the number of donor cells reaching the draining LNs (dLNs) or spleen, which was evaluated via a congenic marker or via donor MHC Class I (CD45.2⁺ K^d+ B/c graft into a CD45.1⁺ B6 recipient), was just at or below our level of detection when analyzed at 2h to 3 days post-skin transplantation (data not shown). Thus, to determine more directly whether skin drainage limited antigen dissemination compared to i.v. injection, subsequent experiments were performed comparing antigen dispersal post-s.c. DST to mimic the dermal positioning of a skin graft to post-i.v. DST, using the same number and identity of donor cells injected s.c. versus i.v.

B/c DST was labeled with CFSE, and cellular distribution was tracked in recipient mice 2h and 24h post-injection (Figure 2A, gating in Figure 2B). I.v. administration resulted in greater accumulation of donor cells at both time points in both lymph nodes (Figure 2C) and spleen (Figure 2D), supporting greater alloantigen dispersal following i.v. than s.c. administration.

Figure 2. I.v. administration of DST enhances the accumulation and dissemination of donor cells to lymphoid organs compared to s.c. DST.

(A) Experimental design: recipient B6 mice were administered CFSE-labeled B/c DST via either s.c. or i.v. routes, and subsequent cell recovery was assessed at 2h and 24h time points. (B) Gating strategy of the infused donor cells and (C, D) donor cell recovery in skin-draining lymph nodes (brachial, axillary, and inguinal) (C) and spleen (D). Dotted lines represent levels of detection. Comparisons were made by unpaired t-test. Symbols represent biological replicates and results are representative of two independent experiments.

I.v. DST results in greater alloantigen presentation potential than s.c. DST

Systemic alloantigen dispersal by donor splenocytes may result not only in more direct presentation of intact donor MHC than subcutaneous dispersal, but also in more indirect presentation of donor peptides by host MHC, whether on MHC molecules shared by donor and recipient cells, or presented only by host cells. To determine if i.v. administration of DST led to enhanced presentation of alloantigens compared to s.c. administration, F1 DST was injected s.c or i.v. and alloantigen presentation was assessed 1d post-administration (Figure 3A). The presentation of alloantigen was tracked using the Y-Ae antibody that recognizes a peptide from I-E^d presented on I-A^b (gating in Figure 3B). Injecting F1 rather than B/c DST enabled us to visualize total alloantigen indirect presentation, a combination of donor F1 cells presenting I-E^d-derived peptide processed from their intact I-E^d molecules and presented on their intact I-A^b molecules, and host APCs indirectly displaying donor I-E^d-derived peptide on I-A^b. I.v. F1 DST resulted in greater Y-Ae staining in both lymph nodes and spleen (Figure 3C) and on both B220⁺ and CD11c⁺ cells (Figure 3D) when compared to s.c. DST, consistent with enhanced combined potential of donor and host APCs to present alloantigen to alloreactive T cells in the lymphoid organs following i.v. DST.

Figure 3: I.v. DST results in greater alloantigen presentation potential than s.c. DST.

(A) Experimental model: recipient B6 mice received i.v. or s.c. B6 × B/c F1 DST and lymph nodes and spleen cells were analyzed 1d later following staining with Y-Ae. (B) Representative Y-Ae staining of splenocytes. (C) Total Y-Ae staining in lymph nodes and spleen of host mice. (D) Y-Ae staining in B220⁺ (B cells) and in CD11c⁺ (dendritic cells) within the hosts’ spleen. Comparisons were made by unpaired t-test. Symbols represent biological replicates and results are representative of two independent experiments.

I.v. DST results in activation of a greater fraction of alloreactive T cells

To test whether the increased antigen presentation potential achieved by i.v. versus s.c. DST led to activation of a greater number of alloreactive T cells, we tested the response of CD45.1⁺ OVA-reactive OT-II CD4⁺ T cells to 2W-OVA F1 DST (Figure 4A). CD45.1⁺ OT-II cells (on a Rag^−/− background) were injected into congenic CD45.2⁺ B6 mice 1d prior to administration of 2W-OVA F1 DST by different routes (s.c., i.s. and i.v., Figure 4A). Lymph node cells and splenocytes were analyzed 3 days later (gating of OT-II T cells in Figure 4B) for upregulation of CD44 (gating in Figure 4C), which denotes antigen experience, and expression of Ki67 as a marker of T cell proliferation. Consistent with the superior antigen dispersal and antigen presentation potential observed following systemic DST, a greater proportion of OT-II cells expressed CD44^hi and Ki67 in both lymph nodes (Figure 4D) and spleen (Figure 4E) following injection of i.v. or i.s. DST compared to s.c. DST. These data demonstrate that systemic administration of donor DST more quickly and effectively drives activation of a greater proportion of the donor-reactive T cells present in secondary lymphoid tissues than s.c. DST.

Figure 4. I.v. DST targets a higher proportion of alloreactive T cells than s.c. DST.

(A) Schematic representation of the experimental model: CD45.1⁺CD4⁺ OT-II T cells (on a Rag1^−/− background) were transferred into congenic CD45.2⁺ B6 mice on day -1, followed by administration of 2W-OVA F1 DST on day 0. Three days later, OT-II T cells were assessed by flow cytometry in the spleen and lymph nodes. (B) FACS gating strategy of OT-II cells in a recipient mouse. (C) Representative flow plot of CD44 expression on CD4⁺ OT-II cells (No RX = no treatment). (D,E) Assessment of percent CD44^hi and Ki67⁺ within OT-II cells in lymph nodes (D) and spleen (E). Comparisons were made using one-way ANOVA followed by Tukey's multiple comparisons test. Symbols represent biological replicates and results are representative of two independent experiments.

I.v. but not s.c. DST promotes αCD154-mediated induction of anergy markers and Tregs in graft-reactive T cells

Our results above demonstrate that i.v. DST results in more alloantigen dissemination and presentation than s.c. DST and in activation of a greater proportion of alloreactive T cells. αCD154 functionally hampers only those T cells whose TCR has been engaged and thus have upregulated CD154, providing a target for αCD154. We hypothesized that administration of αCD154 would impair more alloreactive T cells in hosts receiving with i.v. DST than s.c. DST or no DST. To explore this, we transplanted 2W-OVA F1 skin grafts onto B6 recipients and administered αCD154 +/− DST on day 0. Graft-reactive T cells were analyzed in the spleen 7 days post-transplantation (Figure 5A). The expression of the 2W-OVA model antigen in the graft and on DST cells enabled us to track polyclonal endogenous donor-reactive CD4⁺ cells using fluorescently labeled 2W:I-A^b tetramers, as we previously described³⁸. Furthermore, we have reported that αCD154 + i.v. DST increases the proportion of graft-reactive Tregs and results in functional hyporesponsiveness of graft-reactive CD4⁺ conventional T cells that express increased levels of the anergy markers CD73 and FR4, and of the co-inhibitor PD-1³⁹. Indeed, αCD154 prevented expansion of 2W:I-A^b in all cases (Figure 5C) and the addition of i.v. DST to αCD154, resulted in the highest proportion of Foxp3⁺ 2W:I-A^b-reactive Tregs (Figure 5D), and in the greatest percentage of CD73^hiFR4^hi (Figure 5E) and PD-1⁺ (Figure 5F) 2W:I-A^b-reactive conventional CD4⁺ T cells, when compared to no DST, or s.c. DST.

Figure 5. I.v. but not s.c. DST promotes αCD154-mediated induction of anergy markers and Tregs in graft-reactive 2W:I-A^b CD4⁺ T cells.

(A) Experimental design: B6 mice received 2W-OVA F1 skin grafts and were treated with DST administered either i.v. or s.c. One-week post-transplantation, splenocytes were isolated, stained with 2W:I-A^b tetramers, and analyzed by flow cytometry. (B) Representative gating strategy for the identification of 2W:I-A^b–reactive CD4⁺ conventional T cells and Tregs. (C) Total 2W:I-A^b–reactive T cells in splenocytes across treatment groups. (D) Frequency of Foxp3⁺ 2W:I-A^b–reactive Tregs among 2W:I-A^b–reactive CD4⁺ T cells. (E,F) Expression of anergy/inhibitory markers by 2W:I-A^b–reactive conventional (Foxp3⁻) CD4⁺ T cells under distinct tolerogenic conditions. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test. Each symbol denotes a biological replicate. Data are representative of two independent experiments.

The route of alloantigen delivery determines its tolerogenic potential

The inability of s.c. DST to synergize with αCD154 may not be due to the s.c. route of administration, but rather to the fact that the s.c. route delivers a lower amount of alloantigen to the skin-dLNs than that attained post-i.v. injection. To address this hypothesis, we tested whether increasing the amount of alloantigen injected s.c. would synergize better with αCD154. To this end, we administered four s.c. injections of DST (4x s.c. DST), targeting bilateral inguinal and axillary dLNs (Figure 6A). While the 4x s.c. DST administration increased the total number of donor CD11b⁺F4/80⁺ macrophages infiltrating the dLNs to similar levels as post-i.v. DST (Supplemental Figure 1A-C) and activated a similar number of 2W:I-A^b-reactive endogenous T cells as i.v. DST (Figure 6B,C) demonstrating the successful increase in dose of alloantigen achieved by 4x DST, donor DST cells found in skin dLNs displayed a more activated state as denoted by higher expression levels of CD86 (Supplemental Figure 1C). Indeed, αCD154 + 4x s.c DST did not improve graft outcomes and was even detrimental compared to αCD154 + 1x s.c. DST (p<0.05, Figure 6C). Consistent with the worse graft outcome following αCD154 + 4x DST, 2W:I-A^b-reactive T cells in lymph nodes expressed lower levels of PD-1 than after αCD154 + i.v. DST (Figure 6E). Thus, although 4x s.c. DST distributed more transplant antigen to secondary lymphoid organs than 1x s.c DST, these experiments suggest that the s.c. route carries less tolerogenic potential than the i.v. route. These data highlight the importance of the systemic route of DST administration in promoting αCD154-dependent T cell modulation.

Figure 6. Systemic antigen delivery is uniquely effective at promoting graft acceptance despite increased antigen dosing via alternative routes.

(A) Experimental design: B6 mice received 2W-OVA F1 skin grafts on day 0, along with αCD154 and DST. DST was administered either i.v., s.c., or as four separate s.c injections (4x s.c.), each near the dorsal area of inguinal and axillary lymph nodes to enhance local antigen availability. Draining lymph nodes were harvested on day 7 and stained with 2W:I-A^b tetramers. (B) Representative flow cytometry plots showing identification of CD44^hi expression as a marker of antigen experience among 2W:I-A^b–reactive T cells. (C) Frequency of antigen-experienced 2W:I-A^b–reactive CD4⁺ cells in skin-dLN across treatment groups. (D) Skin graft survival comparing the efficacy of different DST administration routes. Survival data for the No Rx, αCD154 +s.c. DST, and αCD154 +i.v. DST groups are reproduced from Figure 1 to facilitate comparisons. Combination therapy of i.v. DST + αCD154 significantly improved graft survival compared to αCD154 + 4x s.c. DST (p < 0.01); 4x s.c. DST accelerated rejection compared with 1x s.c. DST (p<0.05). (E) Expression of anergy/inhibitory markers in skin-draining lymph nodes among 2W:I-A^b–reactive CD4⁺ cells. Statistical comparisons were performed using one-way ANOVA followed by Tukey’s multiple comparisons test. Each symbol denotes a biological replicate. Data are representative of two independent experiments.

Semi-allogeneic DST induces better graft outcomes in αCD154-treated mice than fully allogeneic DST

While αCD154 combined with DST can induce donor-specific cardiac allograft tolerance with spontaneous acceptance of a second donor-matched heart, this regimen fails to induce tolerance of skin grafts^20,40. We have previously shown that skin is more immunogenic than heart because of its greater content in dendritic/Langerhans cells, and because of immunogenic signals from the colonizing microbiota that are not present in sterile hearts^41,42. We reasoned that an optimized protocol may involve pre-tolerizing alloreactive T cells prior to skin transplantation. Moreover, it was unclear whether DST should be fully or semi-allogeneic. Indeed, clinical settings have suggested that when recipients share HLA alleles with the donor, pre-transplantation blood transfusion may improve allograft survival³⁴. To address these questions, we administered F1 or B/c DST on either d-7 or d0 before B/c skin graft transplantation, with αCD154 being injected on days −7, 0 and 7, or on days 0, 7, 14, respectively (Figure 7A). No statistical difference was observed in the survival of these skin grafts on d36 post-transplantation whether DST was B/c or F1, or whether it was administered on d-7 or d0 (Figure 7B before the second skin transplant).

Figure 7. Semi-allogeneic DST promotes superior allograft outcomes in αCD154-treated mice compared to fully allogeneic DST.

(A) Experimental design: B6 recipient mice received B/c skin grafts and were treated with αCD154 in combination with DST derived from either B/c or F1 donors, administered either one week prior to transplantation or on the day of transplantation. The second B/c skin graft was placed between days 36 and 50 with no additional treatment. (B) Primary skin graft survival in recipient mice before and after placement of the second skin graft. (C) Second skin graft survival in mice that retained their primary graft for >36-50 days and subsequently received a second B/c graft. (D) Experimental model to track the persistence of B/c or F1 CD45.2⁺ DST in congenic CD45.1⁺ allogeneic B6 recipient mice. The αCD154 antibody was administered on the same day as DST, and donor cell persistence was monitored in peripheral blood starting 20 minutes after DST infusion (day 0). (E) Representative flow cytometry gating strategy to track the transferred DST cells (CD45.2⁺). (F) DST persistence over time in host CD45.1⁺ B6 recipient mice. The area under the curves was calculated for each mouse, and a comparison of both curves followed by unpaired T test for each time point was performed.

To evaluate whether any of the mice that had retained the B/c skin graft to day 36 had developed donor-specific tolerance, mice with surviving first grafts received a second donor-matched B/c skin graft on the opposite flank on d36 post-1st skin transplantation, in the absence of immunosuppression. Transplantation of the second graft triggered the rejection of all first grafts, indicating a lack of true donor-specific tolerance irrespective of whether the animals had been pre-tolerized with αCD154 and DST on d −7 pre-first transplantation or not (Figure 7B). However, mice that had received F1 DST rejected their second B/c grafts more slowly compared to those that had received B/c DST (Figure 7C), suggesting a superior effect of semi-allogeneic DST compared to fully allogeneic DST. Thus, the timing of DST (day −7 vs. day 0) was less influential than the semi-allogeneic versus fully allogeneic donor status of the DST, even though the skin grafts were fully allogeneic.

Given our published findings that alloantigen persistence is essential for αCD154-mediated tolerance induction to heart transplants^43,44, we speculated that alloantigen may persist longer following F1 than B/c DST because F1 cells express both donor and host MHC class I molecules, protecting them from NK cell-mediated clearance. To test this hypothesis, CD45.1⁺ B6 hosts were injected with CD45.2⁺ B/c or F1 DST, in combination with αCD154, both on d0 (Figure 7D). DST persistence was tracked in the blood over time (Figure 7E). Consistent with our hypothesis, F1 cells persisted detectably for at least 12 days whereas B/c cells were rapidly eliminated (Figure 7F). Thus, the prolonged graft acceptance observed in F1 DST-treated mice was associated with increased in vivo persistence of donor cells.

Overall, our findings underscore the critical role of systemic DST in promoting long-term graft acceptance, with some advantage to matching some alleles of the DST to those of the host, and provide insight into the mechanisms underlying the immunomodulatory effects of i.v. DST.

Discussion

Our study investigates the mechanisms by which DST augments αCD154-mediated long-term graft acceptance. Our findings indicate that systemic administration of DST improves dissemination of alloantigens and their presentation to alloreactive T cells across secondary lymphoid organs beyond those that drain a transplant, thus ensuring that more alloreactive T cells undergo TCR engagement during CD154 blockade therapy than following transplantation and treatment with αCD154 alone. Moreover, enabling DST’s longer persistence by matching donor and host MHC alleles also contributes to its synergy with αCD154. While αCD154 is showing promising results in its translation to clinical transplantation, developing strategies to pair it with systemic alloantigen delivery could further enhance transplant success.

The clinical use of pre-transplant donor-specific transfusions has undergone a significant evolution since the early years of transplantation. Historically, transfusions were employed with the intent to promote TRIM, or immunologic tolerance, despite their known risk of sensitizing recipients to HLA antigens^26,45,46. Transfusions with shared HLA antigens were once believed to confer immunomodulatory benefits^24,47-49 especially before the introduction of calcineurin inhibitors such as cyclosporine A, which shifted clinical priorities away from transfusion-based strategies⁵⁰ to discouraging their use⁵¹. Additionally, the reliance on red blood cell transfusions for anemic patients awaiting kidney transplants decreased with the introduction of erythropoiesis-stimulating agents, which are now routinely used in non-emergency situations⁵². However, with renewed interest in cell-based tolerance induction and non-thrombogenic αCD154, donor cell transfusion has re-emerged as an approach of potential interest. In mice, combining i.v. DST with αCD154 monoclonal antibody therapy prevents post-DST MHC sensitization, which was one of the most significant drawbacks of traditional cell transfer therapy. Indeed, αCD154 has been shown to prevent the production of allogeneic antibodies⁴⁰, inhibit the differentiation of alloreactive B cells into germinal center B cells, and induce a B cell-intrinsic tolerant state⁵³. Our findings now extend these mechanistic insights, help contextualize historical challenges, and support re-evaluation of possible DST-based strategies under modern immunomodulatory protocols.

Consistent with other models^19,54, i.v. DST alone in our study was insufficient to induce prolonged graft survival, emphasizing the necessity of combining DST with costimulation blockade. One of DST’s primary roles appears to be to facilitate systemic alloantigen dissemination and presentation, thereby enhancing the effectiveness of costimulation blockade on alloreactive T cells. Previous studies have indirectly linked donor cell presence with improved graft outcomes^55,56, suggesting a potential role for antigen persistence in tolerance induction. Our findings directly support and expand this correlation, demonstrating that prolonging DST persistence significantly enhances graft survival. Consistent with this, we and others have shown that repeated i.v. DST administrations every other day, along with αCD154 (d0,7,14), promote improved graft acceptance over 1x i.v. DST+αCD154, even for hard-to-tolerize T cell clones, further supporting the beneficial impact of sustained donor cell presence^44,57.

However, increasing antigen dispersal, even under cover of αCD154, is only effective when using the systemic route of DST injection, as multiple s.c. injections in several sites to increase antigen dispersal were incapable of synergizing with αCD154. This is compatible with the long held notion that systemic injection of soluble antigen is more tolerogenic⁵⁸ while s.c. injections are more immunogenic⁵⁹. Our results support the conclusion that the ability of the systemic route to transport alloantigens to alloreactive T cells everywhere in the body combined with the tolerogenic potential of this route allow it to synergize with αCD154. The success of DST in combination with αCD154, mediated through systemic alloantigen dissemination, may help explain why vascularized skin allografts outperform conventional non-vascularized ones⁶⁰, as one could speculate that vascularized skin grafts might be more efficient at alloantigens’ systemic release and presentation. Indeed, vascularized skin allografts have been associated with a degree of donor-specific tolerance and enable subsequent acceptance of donor-matched heart grafts⁶⁰. DST may also synergize with other costimulatory blockade approaches, as was shown with CTLA4-Ig in a rat heart graft model⁶¹.

By integrating graft-specific T cell tracking and phenotypic analyses, our work provides a better understanding of why DST synergizes with αCD154. Our results highlight how alloantigen biodistribution, presentation, and persistence influence the effectiveness of αCD154-based regimens and provide insights for improving graft survival and patient outcomes in clinical settings of αCD154-based therapies. Several strategies have been explored using DST to prolong graft survival such as DST with extracorporeal photochemotherapy (ECP)⁶², administration of various donor cell types such as mitomycin C-treated cells^63,64, ethylene carbodiimide-treated cells⁶⁵, donor-derived mononuclear cells⁶⁶, regulatory macrophages (Mreg)⁶⁷ or hematopoietic stem cells⁶⁸. To extrapolate our results to the clinic, one could conceive of DST being generated from donor spleens harvested during organ procurement, from peripheral blood mononuclear cells obtained from living donors, from donor bone-marrow-derived APCs, or from cell lines engineered to express MHC alleles from the donor and the host. Additionally, matching MHC class I alleles to prevent rapid elimination by NK cells will also be important. Although considerably more work will be needed to identify the optimal translation of DST, the concept of alloantigen dissemination and biodistribution to improve αCD154 efficacy is one that transcends the skin graft model in the mouse and should be applicable to transplantation at large.

Abbreviations:

αCD154

anti-CD154

ABMR

antibody-mediated rejection

APC

antigen-presenting cell

B6

C57Bl/6

B/c

BALB/c

dLN

draining lymph node

DST

donor-specific transfusion

ECP

extracorporeal photochemotherapy

EV

extracellular vesicles

i.p.

intraperitoneally

i.v.

intravenously

LN

lymph node

RT

room temperature

Rx

treatment

s.c.

subcutaneously

TCR

T cell receptor

Tg

transgenic

Treg

regulatory T cell

TRIM

transplant-related immunomodulation