Journal Logo

Original Basic Science—Liver

Delayed Donor Bone Marrow Infusion Induces Liver Transplant Tolerance

Xie, Yan MD1,2; Wu, Yang MD, PhD1,3; Xin, Kang MD1; Wang, Jiao-Jing MD1; Xu, Hong4; Ildstad, Suzanne T. MD4; Leventhal, Joseph MD, PhD1,5; Yang, Guang-Yu MD, PhD6; Zhang, Zheng MD1,5; Levitsky, Josh MD, MS1,7

Author Information
doi: 10.1097/TP.0000000000001684

Transplantation of solid organs currently relies upon chronic immunosuppressive (IS) agents to prevent graft rejection. These agents have significant cost, side effects, and toxicities, highlighting the need to investigate transformational approaches that promote full IS withdrawal and donor-specific tolerance.1 Interestingly, the liver appears to be the most immunoregulatory organ transplanted. Donor-specific immunoregulatory effects, clonal deletion, alloantibody dilution, circulating and resident regulatory T (Treg) cells, Vδ1 γδ T cells, tolerogenic dendritic cells (DC), and donor-recipient hematopoietic chimerism are components of liver immunoregulation.2-15 This is seen clinically by the lower importance of HLA matching, lower rejection incidence, reduced IS required, and the immunological protection conferred by the liver on other organs. Although the percentage of liver transplantation (LT) recipients able to undergo simple IS withdrawal is the highest of all organ recipients, the rates are still clinically unacceptable and low when attempted early after LT, emphasizing the need for more optimal strategies.16-19

Hematopoietic stem cell (HSC) infusion leading to chimerism is the only tolerance-inducing approach that has been successfully achieved in all species. However, the widespread clinical application of HSC in organ transplantation has been constrained by graft-versus-host disease (GVHD), the need for HLA matching, and the toxicity of recipient conditioning.20-22 Our group recently developed a protocol to induce tolerance and durable chimerism without GVHD using a donor facilitating cell product at the time of mismatched living donor kidney transplantation.23 However, the clinical application of these approaches are not as practical or safe in the LT setting, given the toxicities of conditioning sick liver failure patients at the time of LT and that most LT are performed with deceased donors. Thus, delaying induction of tolerance after recipients have recovered postoperatively, in a period of clinical and immunological quiescence, would be more desirable and could be extended to both living and deceased donor recipients.

Since first reported in the early 1970s, rat LT models have been widely used to investigate mechanisms of liver rejection and tolerance induction.24-26 Liver allografts from PVG (RT1c) or Lewis (RT1l) are spontaneously accepted when transplanted into Dark Agouti (DA, RT1a) and are referred to as low-immune responder combinations.27,28 On the contrary, high-immune responder combinations such as ACI (RT1av) or DA to Lewis transplants develop robust acute rejection,29,30 making them more relevant models for investigations of therapeutic interventions. Several strategies have tested donor-derived cell infusions before or at LT in these preclinical models, such as bone marrow (BM),31,32 mesenchymal stem cells,33,34 blood transfusion,35 Treg cells,36 and immature DC.37,38 While these strategies prolonged graft survival (mostly only up to 100 days post-LT), their effects on longer-term graft protection and the phenotypic immune regulation have not been thoroughly examined. Furthermore, it has not been investigated whether tolerance induction could be achieved in delayed fashion.

Our goal was therefore to adapt our successful kidney transplant protocol to the LT setting in a novel, delayed approach, first in a preclinical high-immune responder animal model (this work) to eventually translate to human early phase clinical trials.



Ten- to 14-week-old (280-350 g) male Lewis (RT1l), ACI (RT1av1) and BN (RT1n) rats were used for this study. The breeder rats were purchased from Harlan Sprague Dawley (Indianapolis, IN) and were bred and maintained in the specific-pathogen-free facility at Center for Comparative Medicine of Northwestern University. All the surgical procedures were approved by the institutional animal care and use committee of Northwestern University.

Orthotopic LT and Postoperative Monitoring

Liver grafts from ACI rats were transplanted orthotopically into allogeneic rats using a modified 2-cuff technique similar to previously described.29,39 Briefly, liver grafts were harvested and perfused thoroughly via the portal vein with cold heparinized (50 U/mL) lactated Ringer solution. The recipient's native liver is removed and followed by the suprahepatic vena cava anastomosis with a 10-0 running suture. The cuff technique was used for anastomosis of both the intrahepatic vena cava and the portal vein. The graft arterial supply was not reconstructed. The common bile duct reconstruction was achieved by using a polyethylene stent. The recipients were monitored daily for clinical signs and symptoms of allograft rejection, which include a gradual weight loss, hunched posture, ruffled fur, reduced mobility, and jaundiced skin. Blood samples were collected periodically for multi-lineage phenotyping, and liver function analysis (by Charles River Laboratories, Wilmington, MA). The day of transplantation was referred to as day 0, where the endpoint of study was defined as greater than 180 postoperative days (POD) of survival, recipient death or when recipients developed clinical signs of liver failure from rejection. Tissues samples were collected at the end of the study for histological examination. Additional transplant recipients were euthanized at preselected earlier time points for sequential functional, immunological, and histological analysis.

BM Cell Isolation and Infusion

Hind leg bones were harvested from anesthetized and euthanized ACI rats and BMCs were collected based on standard procedures.40 The BMCs were manipulated by T cell depletion by using pan T cell MicroBeads according the manufacturer's instruction (Miltenyi Biotec, San Diego, CA). For BM infusion (BMI), recipients received an intravenous infusion of BMCs (100 × 106) at POD24 through penile vein injection under anesthesia with a 25-gauge needle.

Immunosuppression Regimens and Experimental Design

Immunosuppression (IS) regimens included (1) Tacrolimus (TAC) (Astellas Pharma US, Inc.), 1 mg/kg, daily subcutaneously from day 0 to POD21, then followed by once every other day, (2) T cell depletion with anti-rat T cell receptor (TCRαβ) monoclonal antibody (TCR mAb, Clone: R73, BD Pharmingen, San Diego, CA),41-43 500 μg/each, intravenously at POD22 and (3) total body irradiation (TBI) in which rats were exposed to a 60Co[γ]-ray source at a single dose of 300 cGy at POD23.

Recipients were randomly selected into the groups as listed in Table 1. The overall treatment protocol and follow-up is shown in Figure 1A. All rats, except syngeneic transplanted (group 1) and nonimmunosuppressed rejection (group 2), received TAC. At 3 weeks post-LT, 3 groups underwent: (group 3) TAC withdrawal alone at 4 weeks, once every other day until the endpoint; (group 4) nonmyeloablative conditioning (mAb + TBI) followed by TAC withdrawal; (group 5) nonmyeloablative conditioning + donor BMI (100 × 106 T cell–depleted BM cells) followed by TAC withdrawal.

Experimental groups and treatment strategies
Effect of delayed donor bone marrow cell infusion (BMI) in prolongation of LT survival in rats. ACI to Lewis LT were performed at POD0. TAC treatment was initiated from POD0, for 4 weeks. For nonmyeloablative conditioning, a single dose of 0.2 mL anti-αβTCR mAbs was administered at POD22, followed by 300-cGy TBI at POD23, and then donor BMCs were infused at POD24. A, Outline of the treatment protocol. B, Kaplan-Meier curve of survival proportions of LT recipients (n = 10 syn; 8 untreated; 10 TACwd; 9 TCR/TBI; 11 BMI) C, Liver functional panel analysis at the endpoint of the experiment (n > 4/groups). ALT, AST, and total bilirubin. D, Representative histologic sections (H&E staining) of liver grafts (×100).

Skin Transplantation

In long term surviving recipients, skin grafts from either liver donors (ACI) or third-party Brown Norway strains were performed at 20 weeks post-LT to confirm donor-specific unresponsiveness. For skin transplantation, 2 square graft beds (approximately 1 cm2) were prepared by removing shaved rat skin. Full thickness skin grafts of similar size were placed onto the rat graft beds. The animals were monitored for skin survival for an additional 6 weeks after skin transplantation. The endpoint of skin graft rejection was defined as three fourths of the skin graft exhibited dark discoloration, scabbing, necrotic degeneration, or detachment.

Donor-Specific Antibody Analysis

Donor specific antibodies (DSAs) were measured as described previously.44 Splenic lymphocytes from naïve ACI donors were isolated and the cells (0.5 × 106/tube) were first blocked with 0.5 μg Fc blocker (BD Biosciences, San Jose, CA) followed by incubation with plasma samples from recipients (1:10 dilution) for 1 hour at 4°C. Subsequently, cells were washed and stained with PE-conjugated mouse antirat CD45R, FITC-conjugated goat antirat IgG (BioLegend, San Diego, CA), and FITC-conjugated mouse antirat IgM, IgG1, IgG2a, IgG2b (BD Biosciences) Abs for 35 minutes in dark at 4 °C, respectively, and then analyzed by FACS. Levels of DSAs were represented by mean fluorescence intensity (MFI). Negative controls were provided by incubation with naïve rat sera.

Cell Isolation and Immunophenotyping

Leukocytes were isolated from serially collected peripheral blood samples, spleens, or liver grafts. Spleen cells were also isolated using standard methods. To obtain single cell suspensions from liver allografts, livers were cut into small pieces (2 mm) and digested with collagenase IV (Worthington Biochemical Corporation, Lakewood, NJ). The resulting cell suspension was run through a 70-μm filter and washed with PBS. After centrifugation, the leukocytes were purified using lymphocyte separation medium (Fisher Scientific, Pittsburgh, PA). Phenotypical analysis was performed by using several panels of fluorescein-labeled antirat monoclonal antibodies: panel 1: anti-RT1A[a,b]-FITC, anti-CD4-PE, anti-CD3-antigen-presenting cell (APC) (BD Biosciences); panel 2: anti-RT1A[a,b]-FITC, anti-CD8-PE, anti-CD3-APC (BD Biosciences); panel 3: anti-RT1A[a,b]-FITC, anti-CD45R-PE (BD Biosciences), anti-CD11b/c-APC (Biolegend); panel 4: anti-RT1A[a,b]-FITC, anti-CD161-APC (Biolegend). To evaluate the level of mixed chimerism, the antirat RT1A[a, b] antibody was used to distinguish donor cells (ACI) from recipient cells (Lewis) (Figure S1, SDC, Rat IgG2b, κ was used as an isotype control. Frequencies of Treg cells were determined by using APC conjugated CD4, FITC conjugated-CD25, and PE conjugated-Foxp3 (Biolegend) monoclonal antibodies according to the manufacturer's instructions. Cell staining was performed with the antibodies above (1 μg/106 cells) at 4°C for 30 minutes, then washed and analyzed by FACS (BD Biosciences).

Liver Histology

All the recipients were euthanized at the endpoint of liver failure/rejection or POD180 for histopathology. At the endpoint of study, liver samples were preserved in phosphate buffered 10% formalin for 24 hours, then transferred into 70% ethanol for further procession including paraffin embedding and section. Serial sections (4 μm thickness/each section) were stained with hematoxylin-eosin (H&E) at the Northwestern mouse histology core. H&E-stained liver sections were examined under light microscopy by a hepatopathologist (G-Y.Y) who was blinded to all other animal data. Histopathological features of abnormal graft histology and acute rejection were assessed with established criteria.45

Mixed Lymphocyte Reactions

Mixed lymphocyte reactions (MLRs) were set up using 1 × 105 T cells (responders, purity > 85%) enriched from peripheral blood naïve or transplanted Lewis recipients. Pan T cell MicroBeads (Miltenyi Biotec, San Diego, CA) was used for T cell purification. 3 × 105 irradiated ACI (donor) or BN (third party) APCs per well from peripheral blood cells were assed to responder T cells and cocultured for 5 days in complete DMEM medium with 10% heat-inactivated FBS. Proliferation was assessed by 5,6-carboxyfluoresceine diacetate succinimidyl ester (CFSE) (Molecular Probes, Life Sciences) dilution. The proliferating rate of responder T cells were determined by FACSCalibur, and reported as percentage of CD4+ or CD8+ cells with decreased fluorescence intensity.

Statistical Analysis

All data are expressed as the mean ± SD. Comparisons between graft survival times were accomplished by using Kaplan-Meier survival curves with the log-rank test. Statistical significances between 2 groups were determined by Wilcoxon nonparametric tests or by an unpaired Student t test. A P value less than 0.05 was considered statistically significant.


Delayed Donor BMI Facilitates Long-Term Liver Allograft Survival

Recipients were untreated or treated with the IS regimens as indicated in Figure 1A or Table 1. While syngeneic grafts (Lewis to Lewis transplants, n = 10) survived indefinitely with normal liver function and histology, untreated ACI to Lewis allografts (n = 8) succumbed to fulminant liver dysfunction within 12 PODs indicated by significantly elevated levels of alanine transaminase (ALT), aspartate transaminase (AST), bilirubin, and typical features of acute rejection including massive mixed mononuclear cell infiltration, diffuse hepatocyte necrosis and lobular disorganization (Figures 1B-D). As expected, treatment with TAC alone prolonged the allograft survival, but after TAC withdrawal (n = 10), the recipients experienced weight loss, jaundice, significantly elevated ALT, AST, and bilirubin, acute rejection on histology, and 60% graft loss occurred by POD180 (Figures 1B-D). Conditioning with TCR/TBI alone resulted in 78% long-term survival (POD180); however, all recipients (n = 9) displayed impaired liver function and somewhat milder degrees of acute rejection on histology (Figures 1B-D). In contrast, all recipients treated with the conditioning + BMI (n = 11) had long-term TAC-free survival with preserved liver function and graft histology (Figures 1B-D).

Parallel to the histological findings, phenotypical analysis revealed that BMI had significantly reduced numbers of recipient inflammatory cells in the spleen and peripheral blood as well as intragraft infiltrates including T cells (CD3+), myeloid cells (CD11b/c), and NK cells (Figure S2, SDC,

Donor-Specific Tolerance Is Achieved By Delayed Donor BMI

To determine whether the tolerance achieved by delayed BMI was donor specific, we performed 2 sets of experiments. We first performed ex vivo allogenic MLR analyses using APCs isolated from either donor or third-party spleens as stimulators, and T cells enriched from PBMCs of long-term surviving recipients as responders. As shown in Figures 2A and B, CD4+ and CD8+ T cells from BMI-treated recipients exhibited significantly lower levels of proliferation in response to APCs, compared with T cells from TAC withdrawn and TCR/TBI conditioned groups (P < 0.001 for both vs the other groups). However, the response of BMI treated T cells to third-party APCs was preserved. These data confirm that the delayed BMI compromised T cell recall responses only to the donor.

Delayed BMI induces donor-specific tolerance. A, Representative dot-plots showing CD4 and CD8 T cells proliferations measured by CFSE staining and MLR. T cells from Lewis recipients (n = 3-4/groups). were stained with CFSE and incubated in triplicate with 3 × 105 of irradiated (at 3-Gy) APCs from donors (ACI) or third-party BN. 5 days later, cells were collected and staining with T cell markers and analyzed by FACSCalibur. Proliferation rates were expressed as percentages of CD3+CD4+ or CD3+CD8+ cells with decreased fluorescence intensity. B, Quantification of CD4 and CD8 T cells proliferations showing that BMI decreased T cell proliferation to donor antigen stimulation but not to third-party antigens. C, Survival of skin grafts in the long term surviving liver transplants (n = 3/groups). D, Images of skin grafts in long-surviving liver transplants treated with conditioning alone (TCR/TBI) or conditioning plus BMI. BMI led to acceptance of donor type skin grafts but not third-party skin grafts at 3 weeks post skin transplant, while recipients with TCR/TBI rejected both skin grafts.

To further examine donor-specificity, we transplanted skin grafts from donor-type (ACI) or third-party (BN) rats onto long-term surviving Lewis recipients (n = 3) at the 20th week posttransplant. Although all third-party skin allografts were rejected by day 21, donor-matched skin allografts remained normal in appearance until the end of point of study (6 weeks post skin transplant) in the BMI-treated recipients, supporting a state of donor-specific tolerance (Figures 2C and D).

Delayed Donor BMI Inhibits De Novo Production of Donor-Specific Antibodies

To test our hypothesis that delayed BMI down-regulates DSA production, we analyzed DSA levels in serial plasma samples collected from the allografts using flow cytometry. Our results revealed that DSAs including IgM, total IgG (Figure 3A) and subtypes IgG1, IgG2a, IgG2b (Figure 3B) were significantly increased as early as POD7, compared with the naïve rats and isografts, suggesting that liver allografts trigger an early, robust DSA response. Treatment with TAC significantly reduced DSA, both IgG and IgM production of the untreated allograft to the levels comparable to the isografts at POD7. However, on POD90 and 150, DSA levels (both IgG and IgM) increased significantly in TAC withdrawn groups. Unexpectedly, TCR/TBI conditioning triggered the most robust production of DSAs as indicated by the highest levels of IgM, IgG, and its subtypes IgG1, IgG2a, IgG2b on POD90. In contrast, delayed BMI profoundly decreased level of DSA at both POD90 and POD150 and significantly attenuated the impact of TCR/TBI induced DSA (Figures 3A and B). Collectively, these data suggest that delayed BMI significantly downregulates B cell and plasma cell responses.

Delayed donor BMI inhibits de novo production of donor-specific antibodies. Lymphocytes from ACI donors were incubated with plasma samples from LT recipients (n > 4/group/time)at predetermined dates, followed by staining with anti-rat IgM, IgG and subtypes and analyzed by flow cytometry. DSA levels were represented by MFI and CD45R used to gate out B cells. A, Levels of total IgG and IgM. B, IgG subtypes including IgG1, IgG2a, and IgG2b. MFI, mean-channel fluorescence intensity.

Delayed Donor BMI Increases the Frequency of CD4 + CD25 + FoxP3+ Treg Cells

To determine whether our delayed tolerance induction approach promotes Treg cell generation, we performed a serial analysis of peripheral blood Treg cell frequencies after delayed BMI. As in Figures 4A and B, a significantly higher frequency of peripheral circulating CD4+CD25+FoxP3+ Treg cells was observed in BMI recipients compared to that in TAC withdrawn or TCR + TBI recipients (P < 0.05), starting from POD90; this persisted throughout the study. We then further examined whether delayed BMI influenced the accumulation of Treg cells in spleens and liver grafts on POD180. The results shown in Figure 4C revealed that recipients from BMI group exhibited high levels of CD4+CD25+FoxP3+ populations in liver grafts (P < 0.05 vs. TAC withdrawn and TCR/TBI) as well as spleen tissue (P < 0.01 vs. TAC withdrawn and TCR/TBI). These data suggest that up-regulation of CD4+CD25+FoxP3+ Treg cells may have a role in the tolerance induced by the delayed BMI.

Delayed BMI increases the frequency of CD4 + CD25 + FoxP3+ Treg cells. A, Representative dot plots showing gating strategy and percentages of Treg cells (CD4 + CD25 + FoxP3+) in peripheral blood at POD45 (n > 4/group). B, Kinetic changes of frequency of Treg cells in peripheral blood over the predetermined time-points (n > 4/group/time). C, Representative dot plots and bar graphs showing the percentages of Treg cells in spleens and liver grafts at POD180 (n > 4/group). (* < 0.05, ** < 0.01, *** < 0.001).

Increased Mixed Chimerism Is Associated With Delayed Donor BMI-Induced Tolerance

Donor-derived cells regularly persist in liver-grafted patients for some time and may exert beneficial immunomodulatory properties.46-49 Studies have reported that maintenance of immune tolerance after organ transplantation may be due to mixed chimerism after the infusion of allogeneic BM or HSC.50,51 In our experiments, using an anti-donor MHC class I antibody (RT1A[a,b]), we found that all recipients developed early mixed chimerism (POD14) after LT (Figure 5A), likely related to migration of donor leukocytes out of the liver into the circulation under TAC treatment.52-54 The delayed BMI significantly increased the percentage of this donor-derived population, which peaked on POD45 (3 weeks after BMI) (38.47% ± 2.25%, P < 0.001), compared with recipients of TAC withdrawn (21.08% ± 2.11%) or TCR + TBI conditioning (26.54% ± 0.94%). On POD60, recipients in BMI, TCR + TBI and TACwd groups exhibited mean donor chimerism of 30.67% ± 1.22% (P < 0.001, vs. TACwd; P < 0.01, vs. TCR + TBI), 24.02% ± 1.73% (P < 0.05, vs. TACwd) and 17.54% ± 1.06%, respectively. Although no statistical difference was found in chimerism between BMI group and TCR + TBI group on POD120, POD150 and POD180, recipients with BMI still had significant higher chimerism than TAC withdrawal alone recipients on POD150 (P < 0.05) and POD180 (P < 0.05) (Figure 5A).

Increased mixed chimerism associated with BMI-induced tolerance. Serial peripheral blood samples were collected from the liver transplants (n > 4-11/group/time) at the indicated timepoints, and stained with surface markers and analyzed by flow cytometry. A, Representative dot plots showing percentages of circulating donor cells (RT1A[a,b]+) in peripheral blood at POD45, and line graph showing kinetic changes in the absolute number of donor cells. Rat IgG2b, κ was used as an isotype control. B, Phenotypes and quantifications of cellular compositions PBMCs at POD45. C, Representative dot plots showing phenotypes of donor cells and bar graphs quantifying the numbers of CD161lowCD11b/c+ and CD161highCD11b/c+ sub-populations. (* < 0.05, ** < 0.01, *** < 0.001).

Further analysis revealed that donor cell populations primarily consisted of 3 cell subsets: CD11b/c+, CD161low and CD161high (Figure 5B). Recipients with BMI had significant higher percentages of chimeric CD11b/c+ (88.63% ± 1.03%) and CD161low (87.28% ± 1.24%) than the TAC withdrawal (78.70% ± 1.67%, P < 0.001; 79.90% ± 1.92%, P < 0.05) and TCR + TBI (56.58% ± 8.42%, P < 0.01; 56.58% ± 8.42%, P < 0.01) groups. We then calculated absolute cell numbers of chimeric-CD11b/c+ and CD161low and found that recipients with BMI had the highest level in the peripheral blood (5.91 ± 0.22 × 106 and 5.86 ± 0.13 × 106 cells per mL, respectively) (Figure 5C). We further found that CD161lowCD11b/c+ cells represented the major population (>90%) in the context of donor-recipient chimerism.


Immunosuppression weaning could address the long term toxicities and cost of IS therapy, although this is mainly only achievable in older patients further out from LT in which the clinical benefit may be negligible.16,17 Therefore, novel immune manipulation approaches are needed to induce tolerance earlier after LT.55 Such strategies have been attempted previously at the time of kidney transplantation and include conditioning regimens followed by donor hematopoietic stem cell transplantation or BMI.20-23 However, they might have even more success and benefit in LT recipients given the tolerogenic environment, less IS required, and the high degree of IS toxicity in this population. That said, even the lower intensity conditioning regimens may be too toxic at the time of LT in patients with liver failure. With this rationale, we have herein demonstrated the efficacy of a reduced intensity conditioning/BMI approach in inducing delayed donor-specific tolerance after LT. If our delayed, lower conditioning approach could be translated to future clinical trials, it might be safer and more applicable to a larger pool of LT recipients.

We also analyzed for differences in chimerism, cell populations, and donor-specific antibodies that may provide clues as to why tolerance was only achieved after BMI, expanding extensively on previous donor-cell infusion studies.31-38 Consistent with the known phenomenon of donor hematopoietic cells entering the periphery after LT,52-54 all recipients displayed an early mixed chimerism (POD14) after LT, while non–BMI-infused recipients undergoing TAC withdrawal with or without conditioning had significantly lower percentages of chimerism that ultimately was associated with graft rejection. In contrast, after delayed BMI, donor chimerism was augmented and remained higher compared to control groups throughout the follow up course, correlating with the lack of rejection or GVHD with IS withdrawal. This indicates that some degree of mixed chimerism may be important in achieving a balance between rejection and GVH responses with early IS withdrawal in LT, as opposed to other organs in which higher degrees of chimerism are needed.23

We also found that the majority of peripheral circulating chimeric cells were myeloid (CD11c/b+ CD161+) and less CD3+ T cell derived. While the origins, phenotypes and precise roles of these chimeric myeloid cells remains to be further elucidated, we speculate that these chimeric cells were derived from both infused donor BM stem cells and liver resident progenitors. Myeloid-derived suppressor cells have been shown to induce Treg cell expansion and inhibit T cell proliferation by down-regulating the TCRζ chain.56,57 Treg cells inhibit activation and proliferation of alloreactive donor and recipient T cells, thereby reducing GVHD and rejection.58-60 Not surprisingly, we found that CD4+CD25+FoxP3+ Treg cells expanded significantly in the blood, liver graft and spleen in BMI versus controls. Our next step is to specifically test the importance of Tregs and chimeric myeloid cells in delayed BMT tolerance, such as breaking tolerance with cell depletion or testing their suppressive capacity in vitro and in vivo.

Humeral responses involving B cells, plasma cells and DSA have also been associated with graft rejection in organ recipients but only recently implicated in LT graft injury.61,62 One concern in delaying tolerance induction is that donor alloantigens may have already presensitized the recipient.63 However, flow cytometry performed immediately post-LT demonstrated reduced B cell responses and low level of DSA in all recipients on TAC. Thus, standard IS may promote immunological quiescence that sets the stage for delaying tolerance induction. However, later TAC withdrawal led to DSA resurgence and rejection. In addition, TCR/TBI alone increased DSA generation which has been demonstrated in previous studies. Khiong et al64 demonstrated that lymphopenia-induced expansion of CD4+ T cells can induce systemic autoimmunity coinciding with autoantibody production. Iida et al65 found that transient lymphopenia induced by TBI breaks costimulatory blockade-based tolerance and correlates with the accumulation of B cells in lymph nodes and increased DSA titers. Thus, the increased DSA observed in TCR/TBI alone in our study may be due to transient lymphopenia-induced alloreactive T cell activation and B cell expansion. Interestingly, this was not seen in BMI-treated recipients undergoing TAC withdrawal who had low DSA levels and excellent graft function, histology and survival, similar to isografts. This underscores the importance of minimizing DSA in establishing tolerance even in LT, and approaches such as BMI may promote these mechanisms.

Finally, our BMI-treated recipients developed donor-specific hyporesponsiveness, which may be the most reliable marker of long-standing tolerance.66 After co-culture with donor APCs, recipient T cells in BMI-treated recipients had significantly lower proliferation compared to the non-BMI recipients, while all displayed responses to third-party APCs. These results support previous studies showing lymphocytes from long-term acceptors of kidney or vascularized composite allografts were tolerant to donor APCs but responsive to third-party cells.63 We also demonstrated that BMI-treated LT recipients accepted donor skin grafts but immediately rejected third-party skin, compared to the other groups who rejected all skin grafts.

Our approach is not the first to examine the use of hematopoietic cells in inducing LT tolerance in animal and clinical models but has several aspects that are distinct. First, the delayed approach of inducing tolerance after donor-recipient recognition has not been previously demonstrated and BMI may be required to achieve robust tolerance, as simple TAC withdrawal with or without conditioning led to rejection. Other small clinical series have utilized bone marrow or stem cell infusion immediately or several years after LT, rather than an earlier delayed approach.67,68 Second, tolerance with chimerism and absence of GVHD was seen with our T cell–depleted BMI. Similar findings were seen in our center’s living donor kidney transplant facilitating cell tolerance study.23 Whether this will be translatable to human LT is not known, but these preliminary data will hopefully lead to future clinical investigation given the greater applicability of this approach in LT. Finally, we analyzed cell populations, chimerism, antibodies, and donor-specific alloreactivity together at key study time points. That said, we do not know which of these cells or components are most important to the development of delayed tolerance.

Therefore, the next critical step is to focus in on the key mechanisms of tolerance in our model and more specifically determine if Tregs and chimeric myeloid cells are indeed necessary components. We speculate that augmentation of chimerism generates a tolerogenic environment that promotes host hypo-responsiveness via expansion of regulatory cells and reduction of donor-specific antibodies. Future studies need to test the hypothesis that depleting regulatory cells after conditioning + BMI will break tolerance and lead to rejection or GVHD. Clarifying this might have applications to tolerance approaches in clinical LT, given the increasing utilization of immunoregulatory cells for immune monitoring and as cell therapy.


The authors recognize the Northwestern University Microsurgery Core for performing the surgeries and also thank the Northwestern University Mouse Histology and Phenotyping Laboratory for tissue processing and histology.


1. Ojo AO, Held PJ, Port FK, et al. Chronic renal failure after transplantation of a nonrenal organ. N Engl J Med. 2003;349:931–940.
2. Balan V, Ruppert K, Demetris AJ, et al. Long-term outcome of human leukocyte antigen mismatching in liver transplantation: results of the National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database. Hepatology. 2008;48:878–888.
3. Calne RY, Sells RA, Pena JR, et al. Induction of immunological tolerance by porcine liver allografts. Nature. 1969;223:472–476.
4. Kamada N, Wight DG. Antigen-specific immunosuppression induced by liver transplantation in the rat. Transplantation. 1984;38:217–221.
5. Starzl TE. The “privileged” liver and hepatic tolerogenicity. Liver Transpl. 2001;7:918–920.
6. Starzl TE, Trucco M, Zeevi A, et al. Systemic chimerism in human female recipients of male livers. Lancet. 1992;340:876–877.
7. Thomson AW, Lu L, Wan Y, et al. Identification of donor-derived dendritic cell progenitors in bone marrow of spontaneously tolerant liver allograft recipients. Transplantation. 1995;60:1555–1559.
8. Ayala R, Grande S, Albizua E, et al. Long-term follow-up of donor chimerism and tolerance after human liver transplantation. Liver Transpl. 2009;15:581–591.
9. Hove WR, van Hoek B, Bajema IM, et al. Extensive chimerism in liver transplants: vascular endothelium, bile duct epithelium, and hepatocytes. Liver Transpl. 2003;9:552–556.
10. Starzl TE, Demetris AJ, Trucco M, et al. Cell migration and chimerism after whole-organ transplantation: the basis of graft acceptance. Hepatology. 1993;17:1127–1152.
11. Zhao X, Li Y, Ohe H, et al. Intragraft Vδ1 γδ T cells with a unique T-cell receptor are closely associated with pediatric semiallogeneic liver transplant tolerance. Transplantation. 2013;95:192–202.
12. Verdonk RC, Haagsma EB, Jongsma T, et al. A prospective analysis of the natural course of donor chimerism including the natural killer cell fraction after liver transplantation. Transplantation. 2011;92:e22–e24.
13. Castellaneta A, Mazariegos GV, Nayyar N, et al. HLA-G level on monocytoid dendritic cells correlates with regulatory T-cell Foxp3 expression in liver transplant tolerance. Transplantation. 2011;91:1132–1140.
14. Li Y, Zhao X, Cheng D, et al. The presence of Foxp3 expressing T cells within grafts of tolerant human liver transplant recipients. Transplantation. 2008;86:1837–1843.
15. Pons JA, Revilla-Nuin B, Baroja-Mazo A, et al. FoxP3 in peripheral blood is associated with operational tolerance in liver transplant patients during immunosuppression withdrawal. Transplantation. 2008;86:1370–1378.
16. Benítez C, Londoño MC, Miquel R, et al. Prospective multicenter clinical trial of immunosuppressive drug withdrawal in stable adult liver transplant recipients. Hepatology. 2013;58:1824–1835.
17. Feng S, Ekong UD, Lobritto SJ, et al. Complete immunosuppression withdrawal and subsequent allograft function among pediatric recipients of parental living donor liver transplants. JAMA. 2012;307:283–293.
18. Levitsky J. Operational tolerance: past lessons and future prospects. Liver Transpl. 2011;17:222–232.
19. Takatsuki M, Uemoto S, Inomata Y, et al. Weaning of immunosuppression in living donor liver transplant recipients. Transplantation. 2001;72:449–454.
20. Kawai T, Cosimi AB, Sachs DH. Preclinical and clinical studies on the induction of renal allograft tolerance through transient mixed chimerism. Curr Opin Organ Transplant. 2011;16:366–371.
21. Scandling JD, Busque S, Dejbakhsh-Jones S, et al. Tolerance and chimerism after renal and hematopoietic-cell transplantation. N Engl J Med. 2008;358:362–368.
22. Scandling JD, Busque S, Shizuru JA, et al. Induced immune tolerance for kidney transplantation. N Engl J Med. 2011;365:1359–1360.
23. Leventhal J, Abecassis M, Miller J, et al. Chimerism and tolerance without GVHD or engraftment syndrome in HLA-mismatched combined kidney and hematopoietic stem cell transplantation. Sci Transl Med. 2012;4:124ra28.
24. Lee S, Charters AC, Chandler JG, et al. A technique for orthotopic liver transplantation in the rat. Transplantation. 1973;16:664–669.
25. Kamada N, Davies HS, Wight D, et al. Liver transplantation in the rat. Biochemical and histological evidence of complete tolerance induction in non-rejector strains. Transplantation. 1983;35:304–311.
26. Houssin D, Charpentier B, Gugenheim J, et al. Spontaneous long-term acceptance of RT-1-incompatible liver allografts in inbred rats. Analysis of the immune status. Transplantation. 1983;36:615–620.
27. Zimmermann FA, Davies HS, Knoll PP, et al. Orthotopic liver allografts in the rat. The influence of strain combination on the fate of the graft. Transplantation. 1984;37:406–410.
28. Dresske B, Lin X, Huang DS, et al. Spontaneous tolerance: experience with the rat liver transplant model. Hum Immunol. 2002;63:853–861.
29. Zhong R, He G, Sakai Y, et al. The effect of donor-recipient strain combination on rejection and graft-versus-host disease after small bowel/liver transplantation in the rat. Transplantation. 1993;56:381–385.
30. Redaelli CA, Wagner M, Tien YH, et al. 1 alpha,25-Dihydroxycholecalciferol reduces rejection and improves survival in rat liver allografts. Hepatology. 2001;34:926–934.
31. Wu B, Song HL, Yang Y, et al. Improvement of Liver Transplantation Outcome by Heme Oxygenase-1-Transduced Bone Marrow Mesenchymal Stem Cells in Rats. Stem Cells Int. 2016;2016:9235073.
32. Campos L, Alfrey EJ, Posselt AM, et al. Prolonged survival of rat orthotopic liver allografts after intrathymic inoculation of donor-strain cells. Transplantation. 1993;55:866–870.
33. Qi H, Chen G, Huang Y, et al. Foxp3-modified bone marrow mesenchymal stem cells promotes liver allograft tolerance through the generation of regulatory T cells in rats. J Transl Med. 2015;13:274.
34. Niu J, Yue W, Song Y, et al. Prevention of acute liver allograft rejection by IL-10-engineered mesenchymal stem cells. Clin Exp Immunol. 2014;176:473–484.
35. Abe Y, Urakami H, Ostanin D, et al. Induction of Foxp3-expressing regulatory T-cells by donor blood transfusion is required for tolerance to rat liver allografts. PLoS One. 2009;4:e7840.
36. Pu LY, Wang XH, Zhang F, et al. Adoptive transfusion of ex vivo donor alloantigen-stimulated CD4(+)CD25(+) regulatory T cells ameliorates rejection of DA-to-Lewis rat liver transplantation. Surgery. 2007;142:67–73.
37. Wang GY, Yang Y, Li H, et al. Rapamycin combined with donor immature dendritic cells promotes liver allograft survival in association with CD4(+) CD25(+) Foxp3(+) regulatory T cell expansion. Hepatol Res. 2012;42:192–202.
38. Xie J, Wang Y, Bao J, et al. Immune tolerance induced by RelB short-hairpin RNA interference dendritic cells in liver transplantation. J Surg Res. 2013;180:169–175.
39. Kamada N, Calne RY. Orthotopic liver transplantation in the rat. Technique using cuff for portal vein anastomosis and biliary drainage. Transplantation. 1979;28:47–50.
40. Rahhal DN, Xu H, Huang WC, et al. Dissociation between peripheral blood chimerism and tolerance to hindlimb composite tissue transplants: preferential localization of chimerism in donor bone. Transplantation. 2009;88:773–781.
41. Heidecke CD, Hancock WW, Jakobs F, et al. alpha/beta-T cell receptor-directed therapy in rat cardiac allograft recipients. Treatment prior to alloantigen exposure prevents sensitization and abrogates accelerated rejection. Transplantation. 1995;59:78–84.
42. Yoshino S, Cleland LG, Mayrhofer G, et al. Prevention of chronic erosive streptococcal cell wall-induced arthritis in rats by treatment with a monoclonal antibody against the T cell antigen receptor alpha beta. J Immunol. 1991;146:4187–4189.
43. Wang M, Qu X, Stepkowski SM, et al. Beneficial effect of graft perfusion with anti-T cell receptor monoclonal antibodies on survival of small bowel allografts in rat recipients treated with brequinar alone or in combination with cyclosporine and sirolimus. Transplantation. 1996;61:458–464.
44. Chen G, Kheradmand T, Bryant J, et al. Intragraft CD11b(+) IDO(+) cells mediate cardiac allograft tolerance by ECDI-fixed donor splenocyte infusions. Am J Transplant. 2012;12:2920–2929.
45. Klopfleisch R. Multiparametric and semiquantitative scoring systems for the evaluation of mouse model histopathology–a systematic review. BMC Vet Res. 2013;9:123.
46. Schlitt HJ, Raddatz G, Steinhoff G, et al. Passenger lymphocytes in human liver allografts and their potential role after transplantation. Transplantation. 1993;56:951–955.
47. Schlitt HJ, Kanehiro H, Raddatz G, et al. Persistence of donor lymphocytes in liver allograft recipients. Transplantation. 1993;56:1001–1007.
48. Karimi MH, Geramizadeh B, Malek-Hosseini SA. Tolerance induction in liver. Int J Organ Transplant Med. 2015;6:45–54.
49. Knechtle SJ, Kwun J. Unique aspects of rejection and tolerance in liver transplantation. Semin Liver Dis. 2009;29:91–101.
50. Takeuchi Y, Ito H, Kurtz J, et al. Earlier low-dose TBI or DST overcomes CD8+ T-cell–mediated alloresistance to allogeneic marrow in recipients of anti-CD40L. Am J Transplant. 2004;4:31–40.
51. Fehr T, Takeuchi Y, Kurtz J, et al. Early regulation of CD8 T cell alloreactivity by CD4 + CD25- T cells in recipients of anti-CD154 antibody and allogeneic BMT is followed by rapid peripheral deletion of donor-reactive CD8+ T cells, precluding a role for sustained regulation. Eur J Immunol. 2005;35:2679–2690.
52. Qian S, Demetris AJ, Murase N, et al. Murine liver allograft transplantation: tolerance and donor cell chimerism. Hepatology. 1994;19:916–924.
53. Meyer D, Löffeler S, Otto C, et al. Donor-derived alloantigen-presenting cells persist in the liver allograft during tolerance induction. Transpl Int. 2000;13:12–20.
54. Tashiro H, Fukuda Y, Kimura A, et al. Assessment of microchimerism in rat liver transplantation by polymerase chain reaction. Hepatology. 1996;23:828–834.
55. Levitsky J. Immunosuppression withdrawal following liver transplantation: the older, the wiser… but maybe too late. Hepatology. 2013;58:1529–1532.
56. Yang R, Cai Z, Zhang Y, et al. CD80 in immune suppression by mouse ovarian carcinoma-associated Gr-1 + CD11b + myeloid cells. Cancer Res. 2006;66:6807–6815.
57. Wu T, Zhao Y, Zhao Y. The roles of myeloid-derived suppressor cells in transplantation. Expert Rev Clin Immunol. 2014;10:1385–1394.
58. Choi J, Ritchey J, Prior JL, et al. In vivo administration of hypomethylating agents mitigate graft-versus-host disease without sacrificing graft-versus-leukemia. Blood. 2010;116:129–139.
59. Trzonkowski P, Dukat-Mazurek A, Bieniaszewska M, et al. Treatment of graft-versus-host disease with naturally occurring T regulatory cells. BioDrugs. 2013;27:605–614.
60. Rieger K, Loddenkemper C, Maul J, et al. Mucosal FOXP3+ regulatory T cells are numerically deficient in acute and chronic GvHD. Blood. 2006;107:1717–1723.
61. O'Leary JG, Klintmalm GB. Impact of donor-specific antibodies on results of liver transplantation. Curr Opin Organ Transplant. 2013;18:279–284.
62. Levitsky J, Kaneku H, Jie C, et al. Donor-specific HLA antibodies in living versus deceased donor liver transplant recipients. Am J Transplant. 2016;16:2437–2444.
63. Chen B, Xu H, Corbin DR, et al. A clinically feasible approach to induce delayed tolerance in recipients of prior kidney or vascularized composite allotransplants. Transplantation. 2012;94:671–678.
64. Khiong K, Murakami M, Kitabayashi C, et al. Homeostatically proliferating CD4 T cells are involved in the pathogenesis of an Omenn syndrome murine model. J Clin Invest. 2007;117:1270–1281.
65. Iida S, Suzuki T, Tanabe K, et al. Transient lymphopenia breaks costimulatory blockade-based peripheral tolerance and initiates cardiac allograft rejection. Am J Transplant. 2013;13:2268–2279.
66. Matzinger P. The danger model: a renewed sense of self. Science. 2002;296:301–305.
67. Donckier V, Troisi R, Le Moine A, et al. Early immunosuppression withdrawal after living donor liver transplantation and donor stem cell infusion. Liver Transpl. 2006;12:1523–1528.
68. Tryphonopoulos P, Tzakis AG, Weppler D, et al. The role of donor bone marrow infusions in withdrawal of immunosuppression in adult liver allotransplantation. Am J Transplant. 2005;5:608–613.

Supplemental Digital Content

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.