Induction of Kidney Allograft Tolerance by Soluble CD83 Associated With Prevalence of Tolerogenic Dendritic Cells and Indoleamine 2,3-Dioxygenase

Abstract

Background. 

Tolerogenic dendritic cells (Tol-DCs) play a critical role in inducing and maintaining tolerance. Recognizing that both T-cell inactivation and activation are contingent on signals provided by DCs and that graft-specific activated T cells are major mediators of transplant rejection, we aimed to create an environment favoring Tol-DCs with a novel reagent, human soluble CD83 (hsCD83).

Methods. 

Life-supporting orthotopic kidney transplantation was performed in a C57BL/6-to-BALB/c mouse model. The study group was treated with hsCD83 (100 μg/mouse/day, postoperative days −1 to +7, intravenously) and compared with untreated controls.

Results. 

Treatment with hsCD83 achieved kidney allograft tolerance (>100 days), with negligible antidonor antibody detected. In contrast, kidney grafts in untreated recipients demonstrated severe rejection after 35 days, characterized by cellular infiltration, interstitial hemorrhage and edema, and glomerular and tubular necrosis, as well as high antidonor antibody titers. In addition, splenic DCs of tolerant recipients exhibited significantly decreased levels of surface major histocompatibility complex class II, CD40, CD80, and intracellular interleukin-12, as well as reduced allogeneic stimulatory capacity. Adoptive transfer of CD11c⁺ DCs from tolerant hsCD83-treated animals induced kidney allograft tolerance in syngeneic recipients. Blocking indoleamine 2,3-dioxygenase with 1-methyl-tryptophan (15 mg/mouse/day; gavage) prevented the immunosuppressive effect of hsCD83, abrogating hsCD83-induced Tol-DCs and graft tolerance, and leading to acute kidney graft rejection in 22 days.

Conclusion. 

hsCD83 alone was capable of inducing kidney allograft tolerance through a mechanism involving Tol-DC generation and, at least in part, indoleamine 2,3-dioxygenase activity. Because sCD83 is of human origin, the therapeutic approach used in our mouse transplant model holds significant promise for clinical transplantation.

Kidney transplantation is considered the therapy of choice for end-stage kidney failure or congenital kidney abnormalities. However, organ rejection is still the major obstacle to long-term graft survival, with acute transplant rejection occurring in up to 40% of renal graft recipients despite immunosuppressive treatment (¹). Although conventional immunosuppressive drugs have increased graft survival and made kidney transplantation a clinical reality, lifelong immunosuppression with current drug treatments significantly extends the life span of the transplanted organ and notably reduces the incidence of acute rejection; however, prolonged immunosuppression is deleterious because of systemic toxicity, decreased ability to combat infections, increased risk of malignancies, and increased toxicity to the graft itself. Rather than suppressing a transplant patient's entire immune system, an ideal treatment would target the immune system to induce a state of donor-antigen-specific immunologic tolerance. The two cell types primarily responsible for initiating and maintaining immunologic tolerance are tolerogenic dendritic cells (Tol-DCs) and T regulatory cells (Tregs).

DCs are bone marrow–derived professional antigen-presenting cells (²) whose antigen processing and presentation capabilities enable them to signal for T-cell activation when coupled with signals from costimulatory molecules (³). However, T-cell activation in absence of costimulatory signaling leads to T-cell anergy or death through apoptosis (³). The activation of graft-specific T cells represents an important part of the mechanism behind graft rejection. An understanding of events that determine whether DCs will initiate T-cell activation or anergy is critical to achieving success in the clinical practice of transplantation. Consequently, recent studies in transplantation have focused on creating an environment favoring Tol-DCs. Tol-DCs display an immature DC phenotype, expressing low levels of several surface activation markers (CD80, CD86, CD40, and interleukin [IL]-12) and exhibiting reduced antigen presenting abilities (major histocompatibility complex [MHC] class II^low). These properties allow them to be poor mediators of T-cell activation, effectively inducing anergy or death of naïve and effector T cells (⁴). In particular, Tol-DCs are involved in CD4⁺CD25⁺Foxp3⁺ Treg generation and participate in tandem with Tregs to induce tolerance (^4–6).

Earlier studies focused on determining the underlying mechanisms involved between the interplay between Tol-DCs and Tregs revealed that the tryptophan-catabolizing enzyme indoleamine 2,3-dioxygenase (IDO) is an important mediator between the two cell types in inducing tolerance (⁷). The immunosuppressive function of IDO was shown in a study, where wild-type mice up-regulated IDO expression and were able to efficiently suppress potentially lethal allogenic T-cell responses on injection of allogenic CD8⁺ cells, in contrast to IDO-deficient mice that were unable to suppress T-cell responses, even after treatment with the tolerance- inducing agent CTLA-4-Ig (⁸). In another report, it was found that IDO activation in graft recipient DCs occurred in response to the interaction of CTLA-4-Ig with the costimulatory molecule B7 (CD80/CD86). This discovery suggested that Tregs and specific DC subsets are involved in a pathway that expresses functional IDO (⁸).

In addition to IDO being involved in the induction of tolerance, our recent and other previous results also suggest a potential role for sCD83 in this process. CD83 exists as two different protein forms of CD83 in vivo: a membrane bound form (mCD83) and a soluble form (sCD83) (^{9, 10}). This gene is coregulated by a limited number of genes in murine and human DCs (¹¹) and is one of the most characteristic cell surface markers for mature DCs. Studies have shown that mCD83 possesses immunostimulatory capacity (^{9, 12}). On the other hand, sCD83 has been correlated with an immunosuppressive capacity, because although it is found in healthy donors, increased levels have been detected in a number of patients suffering from hematologic malignancies, including patients with chronic lymphatic leukemia and mantle cell lymphoma, as well as patients affected by rheumatoid arthritis. In addition, recombinant sCD83 has been found to inhibit both DC maturation and DC-dependent T-cell responses (⁸). Moreover, in vivo experiments showed that sCD83 is capable of inhibiting the development of paralysis in the mouse experimental autoimmune encephalomyelitis model (¹³). In addition, sCD83 was discovered to delay skin allograft rejection in mouse skin transplant (¹⁴) Although there has been increasing support for the existence of tolerance-inducing properties of sCD83, its mechanism of action is still unknown.

In this study, we investigated the immunosuppressive effects of human soluble CD83 (hsCD83) in a life-supporting kidney allograft model. We proposed that hsCD83 interference with DC maturation was relevant to attaining graft tolerance and that IDO activity may be associated with this process. Because tolerance-inducing effects of IDO can be reversed using 1-methyl-tryptophan (1-MT), use of this IDO-inhibitor enabled examination of the interplay among hsCD83, DC maturation, and immunosuppressive functions with respect to IDO. The goal of this investigation was to determine the potential of hsCD83 as a therapeutic intervention strategy in the field of clinical transplantation.

RESULTS

hsCD83 Induces Kidney Allograft Tolerance

To determine the role of hsCD83 in inducing long-term renal allograft acceptance, we injected hsCD83 into BALB/c mice (3 mg/kg/day from postoperative day (POD) −1 to 7) receiving C57BL/6 kidney grafts and compared their transplant survival with allograft recipients that were untreated. It was found that untreated recipients rapidly rejected kidney allografts with a median survival time (MST) of 35 days. In contrast, hsCD83 treatment achieved indefinite kidney allograft survival for more than 100 days (Table 1) in this fully MHC-mismatched transplantation model.

TABLE 1: Median survival of C57BL/6-to-BALB/c mouse kidney allografts

To further determine whether long-term graft survival was induced through donor-specific tolerance as a result of hsCD83 treatment, we transplanted donor C57BL/6 or third-party C3H mouse skin grafts to 100-day surviving BALB/c recipients. We found that the third-party skin survived only 11 days (range, 9–12 days), whereas the donor C57BL/6 skin grafts survived for more than 50 days (range, 45–56 days) (data not shown). These results indicated that hsCD83 monotherapy was capable of inducing donor-specific tolerance in this renal allograft model.

Inhibition of CMR and AMR Using hsCD83 Monotherapy

Pathologic examination of the various kidney allografts revealed that without treatment renal grafts at endpoint (POD35) showed typical features of acute vascular rejection characterized by intravascular thrombosis and interstitial hemorrhage, as well as acute cellular rejection (Fig. 1A-a) involving massive intragraft infiltration of CD4⁺ (Fig. 1A-c, B) and CD8⁺ (Fig. 1A-e, B) cells. In contrast, hsCD83 monotherapy resulted in normal graft histology on both POD35 (data not shown) and POD100 (Fig. 1A-b), with significant inhibition of CD4⁺ (Fig. 1A-d, B) and CD8⁺ (Fig. 1A-f, B) cell infiltration on POD100. The maintenance of normal kidney graft histology and function after hsCD83 treatment not only suggests immunosuppressive properties of hsCD83 but also indicates its lack of renal toxicity.

FIGURE 1.: Histology and immunohistology of human soluble CD83 (hsCD83)-treated and untreated kidney allograft recipients. (A) Kidney grafts were harvested at the time of rejection or at postoperative day (POD) 100 (for the therapy group) and evaluated by hematoxylin-eosin (H&E; a, b) and immunohistochemistry (c–f) staining (n=5) to enable comparison between untreated (a, c, e) and hsCD83-treated recipients (b, d, f). Magnification, ×400. (B) Immunoperoxidase staining for intragraft CD4⁺ and CD8⁺ cells in untreated and hsCD83-treated mice (n=5) at the time of rejection or at study endpoint (POD100) was analyzed by counting all positively stained cells in the section, divided by the section area examined (cells/mm²; Empix Imaging, Mississauga, ON). Statistical significance, as determined by one-way analysis of variance, is indicated (*P<0.01, hsCD83-treated renal allograft recipients vs. untreated allograft recipients). Results are representative of three independently performed experiments.

Low Levels of Circulating Antidonor Antibodies in hsCD83-Treated Kidney Allograft Recipients

To determine the inhibitory effect of hsCD83 on the humoral response, serum levels of antidonor antibodies were measured in BALB/c allograft recipients. Flow cytometric evaluation revealed that high levels of circulating antidonor IgG and IgM were detected in untreated recipients (Fig. 2). In contrast, antidonor antibodies were found to be undetectable in hsCD83-treated recipients at POD100 (Fig. 2).

FIGURE 2.: Evaluation of circulating antidonor antibody levels. Sera harvested at postoperative day (POD) 30 or at the study endpoint (POD100) were assessed for circulating antidonor IgG and IgM levels among untreated and human soluble CD83 (hsCD83)-treated allograft recipients, respectively (n=5), by monitoring the binding to donor allogeneic CD3⁺T cells. Statistical significance of the differences in serum antidonor antibody levels, measured as mean fluorescence intensity, were compared between the recipient groups using one-way analysis of variance (*P<0.01, antidonor antibody levels in hsCD83-treated allograft recipients at POD100 vs. antidonor antibody levels of untreated renal allograft recipients at POD30). hsCD83, human soluble CD83.

hsCD83 Induces Tol-DCs in Tolerant Recipients

Investigations into the potential of hsCD83 inducing Tol-DCs in long-term surviving recipients were carried out by isolating splenic CD11c⁺DCs from tolerant recipients (POD100) and studying their phenotypic and functional properties in comparison with DCs from untreated/rejecting recipients (POD30). Phenotypic analysis of the various CD11c⁺DC subpopulations by flow cytometry revealed that tolerant recipients possessed a significantly greater proportion of splenic CD11c⁺DCs that displayed an immature phenotype with low expression of MHC-II, CD40, CD80, and IL-12, compared with the CD11c⁺DC subpopulations of untreated/rejecting mice (Fig. 3A).

FIGURE 3.: Phenotypic and functional characterization of splenic CD11c⁺ dendritic cells (DCs) in renal allograft recipients. (A) Comparison of splenic CD11c⁺ DC maturation levels in human soluble CD83 (hsCD83)-treated and untreated allograft recipients. Expression levels of surface CD40, CD80, major histocompatibility complex (MHC) class II, as well as intracellular interleukin (IL)-12 on CD11c⁺-gated DCs (n=5) were analyzed by flow cytometry and expressed in percentages of positively stained cells, with values compared for statistical significance using one-way analysis of variance (*P<0.01, DCs of hsCD83-treated group vs. DCs of untreated group). The results are representative of four independently performed experiments. (B) Determining ability of allograft recipient DCs to activate allogeneic T cells. Splenic CD11c⁺ DCs were isolated from long-term surviving recipients or untreated BALB/c recipients (postoperative day 30) (n=5). Irradiated DCs were used as stimulators, whereas CD3⁺ T cells from C57BL/6 donor mice were used as responders in mixed lymphocyte reaction, as described in Materials and Methods section. Results are expressed as counts per minute±standard error of the mean (*P<0.01, stimulatory capacity of hsCD83-treated recipient DCs vs. stimulatory capacity of untreated recipient DCs).

The role of hsCD83 in affecting DC function was then explored by assessing the ability of tolerant recipient DCs in stimulating allogeneic T-cell responses. Splenic CD11c⁺DCs isolated from hsCD83-treated tolerant recipients (POD100) failed to stimulate allogeneic T-cell responses in mixed lymphocyte reaction in comparison with the same cell subpopulation isolated from untreated allograft recipients (POD30; Fig. 3B). Together, these data suggest that CD11c⁺DCs isolated from hsCD83-treated long-term surviving recipients are phenotypically immature and display low responsiveness to maturation stimuli on POD100, suggesting that hsCD83 plays a critical role in the generation of Tol-DCs.

Tol-DCs Generated in hsCD83-Treated Recipients Can Induce Allograft Tolerance Through Adoptive Transfer

To further investigate the potential tolerogenic properties of the DCs present in hsCD83-treated recipients, CD11c⁺DCs were isolated from the spleens of 100-day surviving BALB/c recipients and compared with the same cell subpopulation of untreated allograft recipients (POD30). Adoptive transfer of the isolated CD11c⁺DCs into syngeneic mice 1 day before their subsequent receipt of a C57BL/6 kidney revealed that more than 100-day kidney graft survival was achieved through adoptive transfer of the DCs from hsCD83-treated tolerant recipients. In contrast, adoptive transfer of DCs from untreated recipients (POD30) resulted in rejection of kidney allografts within 38 days (Table 1). These results indicate that adoptive transfer of Tol-DCs can transfer tolerogenic properties from primary tolerant recipients to secondary naïve recipients, providing protection to a subsequent kidney allograft and, therefore, indicate that the Tol-DCs generated in the presence of hsCD83 have the ability to induce allograft tolerance.

Functionally Blocking IDO Abrogates hsCD83-Induced Kidney Allograft Tolerance

Having determined that hsCD83 was a major contributor to allograft protection, we were interested in identifying the putative mechanisms by which the Tol-DCs rendered their immunosuppressive effects to confer kidney allograft protection. To explore the potential mechanism, we investigated the consequence of functionally blocking IDO with a potent specific inhibitor 1-MT (¹⁵) on allograft survival, while simultaneously treating with hsCD83. Interestingly, we found that 1-MT abrogated hsCD83-induced kidney allograft tolerance and induced acute rejection in 100% of hsCD83-treated renal allograft recipients (MST=22 days [range, 19–25 days]; Table 1), which was characterized by interstitial hemorrhage and glomerular/tubular necrosis (Fig. 4A). In addition, IDO-blockade eliminated the inhibitory effects of hsCD83 on DC maturation in kidney allograft recipients. As shown in Figure 4(B), expression levels of MHC-II, CD40, CD80, and IL-12 on DCs from hsCD83 plus 1-MT–treated recipients were much higher than those of hsCD83-treated recipients in the absence of 1-MT, with phenotypes indistinguishable from those of untreated recipients. These data suggest that immunosuppressive effects of hsCD83 in inducing kidney allograft tolerance are inhibited when IDO is blocked with 1-MT, indicating that at least one of the mechanisms involved in hsCD83-induced allograft tolerance involves generation of Tol-DCs and the activity of IDO.

FIGURE 4.: Effects of indoleamine 2,3-dioxygenase antagonist 1-methyl-tryptophan (1-MT) on renal allograft survival and splenic CD11c⁺ DC phenotype. (A) Histology of kidney allograft of human soluble CD83 (hsCD83)-treated BALB/c recipients receiving 1-MT. Kidney grafts were harvested at the time of rejection (n=5) and evaluated by hematoxylin-eosin staining. Photograph is representative of kidney graft rejection observed in all hsCD83 plus 1-MT–treated recipients. Magnification, ×400. (B) Comparison of splenic CD11c⁺ DC maturation levels in untreated, hsCD83-treated, and hsCD83 plus 1-MT–treated allograft recipients. Expression levels of surface CD40, CD80, major histocompatibility complex (MHC) class II, and intracellular interleukin (IL)-12 on CD11c⁺-gated DCs (n=5) were analyzed by flow cytometry and expressed in percentages of positively stained cells, with values compared for statistical significance using one-way analysis of variance (*P<0.01, DCs of hsCD83 plus 1-MT–treated group vs. DCs of hsCD83-treated group). Results are representative of three independently performed experiments.

DISCUSSION

In this study, we have demonstrated that a novel immunosuppressive agent, hsCD83, can induce donor-specific tolerance in a fully MHC-mismatched mouse kidney allograft model. Undetectable antidonor antibody levels, normal kidney graft histology, and healthy creatinine levels demonstrated by the hsCD83-treated allograft recipients on POD100 strongly suggested the involvement of hsCD83 in establishment of tolerance and also revealed its lack of renal toxicity.

Although the molecular mechanisms behind the hsCD83-mediated graft protection remain unclear, our study revealed that generation of Tol-DCs and IDO activity may play an important role. Previously documented observations that hsCD83 markedly altered DC morphology and phenotype, significantly inhibited DC maturation and synapse formation with T cells (¹⁶), and reduced the allogenic T-cell stimulatory capabilities of hsCD83-treated DCs in vitro (^17–19), suggested that hsCD83 may have the potential to induce the generation of immature DCs or Tol-DCs or both in vivo; however, no previous studies have demonstrated this. These properties of hsCD83 promised to be relevant and advantageous to success in transplantation, particularly because immature DCs are highly phagocytic but are poor stimulators of T-cell activation (²⁰). Indeed, in this study, we found that hsCD83 therapy in vivo induced a significant prevalence of CD11c⁺DCs that were ineffective at both priming and activating naive and memory T cells in vitro and were capable of inducing allograft tolerance upon their adoptive transfer in vivo. Considering that the role of immature DCs in healthy individuals is to continuously present “nondangerous” antigens to naïve T cells for deletion and anergy of autoreactive T cells, as well as for generation of self antigen-specific Treg cells (²¹), we hypothesize that a key factor in the ability of hsCD83 to induce graft tolerance in treated transplant recipients was its direct or indirect interference with DC maturation and function. We also hypothesize that the resultant hsCD83-treated recipient DCs, expressing suboptimal costimulatory or adhesion molecules or both, would become tolerogenic and would process apoptotic donor (C57BL/6)-derived cells and cross-present donor-specific peptides to recipient T cells (^22–25). Our hypotheses are supported from witnessing a similar phenomenon of cross-presentation and infectious tolerance by recipient DCs in another one of our recent studies involving hsCD83 as a component of a triple drug treatment of mouse cardiac allografts (Ge W, Arp J, Lian D, et al. unpublished data). In addition, this phenomenon of cross-presentation has been well documented in other transplant models describing the adoptive transfer of recipient DCs loaded with donor allopeptides down-regulating antidonor responses and prolonging allograft survival (^{26, 27}). Of greater relevance are reports of recipient mouse DCs acquiring and processing alloantigen from heterotopic cardiac allografts that produced large amounts of IDO (^{28, 29}) and generated donor-specific Tregs (³⁰).

It has been shown that antigen-presenting cells such as DCs, vary in the degree of their IDO expression to regulate immune responses (³¹). IDO is known to act as a bridge between DCs and Tregs, influencing the T cells' full effector or regulatory functions or both (³²), with high levels of IDO expression by Tol-DCs associated with preventing allogeneic T-cell proliferation (²³) and inducing T-cell apoptosis or anergy or both. Indeed, a previous study revealed that induced overexpression of IDO within allografts was associated with prolonged graft survival (³¹). Interestingly, in this study, it was found that the known IDO antagonist, 1-MT, prevented the immunosuppressive effects of hsCD83, abrogating hsCD83-induced graft tolerance and leading to acute kidney graft rejection. The negative effect of 1-MT on allograft survival was accompanied by concomitant reduction in the prevalence of Tol-DCs.

In conclusion, we report that hsCD83 therapy leads to indefinite kidney allograft survival and establishment of donor-specific tolerance. Induction of tolerance by hsCD83 in this allograft model was associated with a prevalence of splenic Tol-DCs and was dependent on IDO activity. Considering that human and mouse sCD83 are homologous, and that human sCD83 currently proved to be effective and safe in mice, suggests that hsCD83 holds significant promise for its therapeutic use in future clinical transplantation.

MATERIALS AND METHODS

Animals and Immunosuppressant

Male adult C57BL/6 (H-2^b), BALB/c (H-2^d), and C3H (H-2^k) mice (Jackson Laboratory, Bar Harbor, ME) were used as donors, recipients, and third-party controls, respectively. Animals were housed in the Animal Care Facility at The University of Western Ontario and handled according to guidelines of the Canadian Council on Animal Care (³³). sCD83, the extracellular domain of human CD83 (amino acids 20–145), was purified from bacteria, and the glutathione S-transferase purification tag was removed before intravenous (IV). injection (<50 endotoxin units/mg protein; Argos Therapeutics, Durham, NC) (¹²).

Orthotopic Kidney Transplantation

Left renal transplantations were performed as previously described (³⁴). After bilateral recipient nephrectomies, the harvested C57BL/6 donor graft was revascularized with end-to-side anastomoses between the donor renal artery and BALB/c recipient abdominal aorta, as well as the donor renal vein and recipient inferior vena cava. Subsequently, end-to-end ureteric anastomosis was fashioned. All surgeries involved in this life-supporting kidney transplant model were performed by the same microsurgeon to ensure consistency. Graft rejection leading to death was used as an indicator for the endpoint of rejection, while mice with long-term surviving grafts were euthanized at POD100.

Experimental Groups

Kidney allograft recipients were randomly assigned to 5 groups (n=5–7): group 1, untreated; group 2, BALB/c recipients receiving hsCD83 at 3 mg/kg/day, IV, POD −1 to +7; Groups 3 and 4, adoptive transfer of CD11c⁺ DCs from 100-day surviving BALB/c allograft recipients or untreated rejecting BALB/c allograft recipients (POD30) to syngeneic mice 1 day before C57BL/6 kidney transplantation; and group 5, BALB/c recipients receiving hsCD83 (same regimen as group 2) and 1-MT (15 mg/mouse/day; gavage, on day POD −2).

Skin Transplantation for Testing Antigen-Specific Tolerance

When hsCD83-treated kidney allograft recipients survived to 100 days, full-thickness skin grafts (1×1 cm²) from C57BL/6-donor or third-party-C3H mouse strains were transplanted to these recipients (n=5). Graft rejection was defined as scar formation or epidermis sloughing or both.

Graft Histology

Renal graft tissue samples collected at POD100 or at rejection endpoint, as appropriate, were fixed in 10% formaldehyde, embedded in paraffin, and sectioned at 5 μm for hematoxylin-eosin staining (n=5) (³⁵). The sections were examined and scored in a blinded fashion for severity of rejection by a pathologist (B.G.) (^{36, 37}). Criteria for graft rejection included vasculitis, hemorrhage, edema, thrombosis, glomerular and tubular necrosis, and lymphocyte infiltration. These features were scored as 0, no change; 1, minimal; 2, mild; 3, moderate; or 4, marked changes relative to normal tissues.

Immunohistochemistry

Cryosections embedded in Tissue-Tek O.C.T. (Skura Finetek, Torrance, CA) and mounted on gelatin-coated slides were stained using an avidin-biotin immunoperoxidase method (Cedarlane Laboratories, Hornby, ON) (³⁸) (n=5). Intragraft T-cell infiltration was detected using anti-mouse CD4 and CD8 monoclonal antibodies (clone YTS 191.1.2; Cedarlane Laboratories and clone 53-6.7; BD Biosciences, Franklin Lakes, NJ). Negative controls were performed by omitting biotin-conjugated primary antibodies. Antibody reactivity was evaluated on five randomly selected high-powered bright-phase microscope fields of each tissue section (n=5) obtained from eight animals/treatment group. Intragraft CD4⁺/CD8⁺ cells were quantified by counting positively stained cells and dividing by the specific section area (cells/mm²; Empix Imaging, Mississauga, ON) (³⁸).

Phenotypic Analysis CD11c⁺DCs

Splenic CD11c⁺-enriched DC subpopulations from the various treatment groups (n=5) were isolated (75%–80% purity; Miltenyi Biotech, Auburn, CA) and phenotypically characterized by double-staining with anti-mouse CD11c-fluorescein isothiocyanate, in combination with phycoerythrin- labeled anti-mouse MHC-II, CD40, CD80, and IL-12 (p70) (BD Biosciences). Intracellular IL-12 (p70) detection required a cell permeabilization kit (Cedarlane Laboratories). Stained samples were acquired and analyzed using a FACS-Calibur flow cytometer (CellQuest software; Becton Dickinson, Mountain View, CA) and reported as percentages. Appropriate isotype-matched negative controls were used for all flow cytometry screening.

Mixed Lymphocyte Reaction

CD11c⁺DC subpopulations (>85%–90% pure; Miltenyi Biotech) were isolated from spleens of long-term surviving or rejecting BALB/c recipient mice (n=5). CD11c⁺ cells were irradiated at 3000 rad and used as stimulators. Splenic CD3⁺ T cells isolated by positive selection (>94%–97% pure; StemCell Technologies, Vancouver, BC) from naïve C57BL/6, BALB/c, or C3H mice were used as responders (2×10⁵/well), and co-cultured with DCs (DC:T-cell ratio of 1:4), or incubated alone (³⁹) After 96 h, T-cell proliferation was assessed by [³H]-Thymidine (GE Healthcare, Baie d'Urfé, Quebec, Canada) incorporation, and reported in counts per minute of triplicate cultures (±standard error of the mean) using a Wallac-Betaplate counter (Turku, Finland).

Serum Antidonor Antibody Levels

Circulating antidonor (C57BL/6) and anti-self (BALB/c) IgG/IgM antibodies were evaluated in recipient sera by flow cytometry, as previously described, after incubation with allogeneic donor (n=5) or syngeneic splenocytes (n=5) and reported in corrected mean fluorescence intensity(MFI) values (MFI observed with donor cells−MFI observed with syngeneic cells [³⁹]).

Adoptive Transfer

Highly purified CD11c⁺DCs (>97% pure; CD11c-positive selection; Miltenyi Biotech) were isolated on POD30 from spleens of both hsCD83-treated and untreated BALB/c mice (n=5). After confirmation of phenotype by flow cytometry, splenic CD11c⁺DCs purified from the two groups were adoptively transferred IV (2×10⁶) into syngeneic naïve BALB/c mice (POD −1). The following day, DC-recipients were transplanted with C57BL/6 kidneys.

Statistical Analysis

Data were reported as means±standard deviation. Allograft survival was reported as MST and compared using log-rank testing. Histologic findings were analyzed using analysis of variance on rank. Flow cytometry and immunohistologic data were analyzed using one-way analysis of variance. Differences with P less than or equal to 0.05 were considered significant.

ACKNOWLEDGMENTS

The authors thank Dr. Gill Strejan for his critical review of the manuscript and Dr. Miren Baroja for preparing the reagent.