Rapamycin-Treated, Alloantigen-Pulsed Host Dendritic Cells Induce Ag-Specific T Cell Regulation and Prolong Graft Survival


*Corresponding author: Angus W. Thomson, thomsonaw@upmc.edu


Tolerogenic properties of dendritic cells (DC), particularly those in the immature state, and their therapeutic potential are increasingly being recognized. Among several distinct approaches to generate stably immature DC, pharmacologic manipulation stands out as a promising and clinically applicable option. We have shown recently that the immunophilin ligand rapamycin (Rapa) can inhibit DC maturation and their effector functions. Here, we examined the impact of Rapa exposure on subsequent alloantigen (Ag) presentation by myeloid DC via the indirect pathway. Rapa-treated, allogeneic lysate-pulsed host DC (Rapa-DC) were inferior stimulators of syngeneic T cells, compared to lysate-pulsed control DC. Rapa exposure did not block alloAg uptake by DC nor impair their in vivo homing to splenic T cell areas after adoptive transfer. T cells primed by Rapa-treated, alloAg-pulsed DC showed decreased capacity to produce IL-2 and IFNγ, and were hyporesponsive to subsequent challenge via both the direct and indirect pathways, in an Ag-specific manner. When infused 1 week before transplantation, these Rapa-DC significantly prolonged alloAg-specific heart graft survival. This effect was reversed by systemic IL-2 administration but enhanced by either repeated infusion of the cells or a short post-transplant course of FK506. These therapeutic effects, achieved by targeting both major pathways of allorecognition, provide the basis for a clinically applicable strategy to suppress graft rejection.


Dendritic cells (DC) are bone marrow (BM)-derived professional antigen (Ag)-presenting cells (APC) with the unique ability to both initiate and regulate immune responses (1,2). The nature of the immune response elicited by DC depends on their state of maturation and functional differentiation, which is influenced by microenvironmental factors (microbial products, cytokines and cyclooxygenase metabolites) (3) and commonly used immunosuppressive drugs (4). In the immature state, DC are inherently tolerogenic (5) and can suppress T cell responses to self or foreign Ag (2). By contrast, their maturation in response to inflammatory stimuli is associated with the acquisition of potent immunostimulatory function, linked to upregulated expression of cell surface major histocompatibility complex (MHC) and costimulatory molecules. This dichotomous function of DC, combined with their remarkable plasticity, provides a basis for the design of regulatory DC for potential therapeutic applications (4,6).

The immunophilin ligand rapamycin (Rapa) is a macrolide antibiotic pro-drug with potent immunosuppressive properties (7). Intracellularly, it forms a complex with the immunophilin FK506 binding protein (FKBP)-12. This heterocomplex, in turn, inhibits the function of the serine/threonine kinase mammalian target of rapamycin (mTOR), a central protein involved in the signaling pathways that control cell growth, proliferation and protein translation (7,8). The immunosuppressive activity of Rapa, and its efficacy as an anti-rejection agent in organ transplantation (9,10) have been ascribed principally to its anti-proliferative effects on T cells. Recently, we have shown that in vitro or in vivo exposure of DC to Rapa inhibits their maturation, inflammatory cytokine (bioactive interleukin [IL]-12p70 and tumor necrosis factor [TNF]-α) secretion and T cell allostimulatory capacities (11). Thus, Rapa appears to be a good candidate for pharmacological suppression of DC functions and the generation of DC with immunoregulatory activity. Indeed, Rapa-treated DC (Rapa-DC) induce T cell hyporesponsiveness to donor Ag following their injection into allogeneic recipients (11).

In this study, we have further explored the immunoregulatory capacity of Rapa-DC in the context of alloimmune reactivity. More specifically, we have examined the function of Rapa-DC in indirect alloAg presentation and their role in modulation of organ transplant rejection. Our results show that Rapa-DC can be loaded effectively with Ag derived from donor cell lysates to induce alloAg-specific T cell hyporesponsiveness in vivo. Furthermore, infusion of these Rapa-DC prior to transplantation prolongs fully MHC-mismatched heart allograft survival, in some cases indefinitely, in otherwise untreated mice. The regulatory effect of these DC is more marked in animals given a short postoperative course of subtherapeutic FK506. These novel findings provide insight into clinically applicable strategies for immunomodulation using pharmacologically modified DC of host origin for therapy of graft rejection, with implications for use of regulatory DC in other immune-mediated disorders.

Materials and Methods


Eight- to 12-week-old C57BL/10 (B10; H2Kb), C3H/HeJ (C3H; H2Kk) and BALB/c (H2Kd) mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and maintained in the specific pathogen-free Central Animal Facility of the University of Pittsburgh Medical Center. Experiments were conducted under an institutional animal care and use committee-approved protocol and in accordance with National Institutes of Health-approved guidelines.

Generation of BM-derived DC

The BM-derived myeloid DC were propagated as described (12). Briefly, the BM cells were removed from femurs and tibias of C3H mice and depleted of erythrocytes by hypotonic lysis. Erythroid precursors, B lymphocytes and granulocytes were removed by complement depletion using a cocktail of monoclonal antibodies (mAbs) (anti-TER-119 [TER-119], anti-B220 [RA3–6B2] and anti-Gr1 [RB6–8C5]; BD PharMingen, San Diego, CA) followed by incubation (45 min; 37°C) with low-toxicity rabbit complement (Cedarlane, Hornby, ON, Canada). The cells were cultured for 7 days in RPMI-1640 with 10% v/v heat-inactivated fetal calf serum, l-glutamine, non-essential amino acids, sodium pyruvate, penicillin–streptomycin, HEPES (N-2-hydroxyethylpiperazine-N′-2-ethane-sulfonic acid), 2-mercaptoethanol (all from Life Technologies, Gaithersburg, MD), 1000 U/mL recombinant (r) murine granulocyte-macrophage colony-stimulating factor (GM-CSF; Schering-Plough, Kenilworth, NJ) and 1000 U/mL r murine IL-4 (R&D Systems, Minneapolis, MN). On day 2, 10 ng/mL Rapa (Sigma, St Louis, MO) was added. Every 2 days, 75% of the culture supernatant was replaced with fresh cytokine-containing medium (with or without Rapa). On day 4, non-adherent cells were removed; on day 7, 50% or more of the non-adherent cells expressed CD11c.

Phenotypic analysis of DC

The DC surface Ag expression was analyzed by flow cytometry on day 8 of BM culture, 24 h after alloAg pulsing. Stimulation was performed with lipopolysaccharide (LPS; 1μg/mL, Escherichia coli serotype 026:B6; Sigma) in the presence of low-dose GM-CSF (50 U/mL) in RPMI-1640 culture medium for 16 h at 37°C. Fluorescein isothiocyanate (FITC)-, phycoerythrin (PE)- or CyChrome-conjugated or biotinylated mAbs used to detect the expression of CD11c (HL3), CD80 (16–10A1), CD86 (GL1), IAkα chain (11–5.2), H2Kb (AF6–88.5) as well as isotype-matched control mAbs and streptavidin-CyChrome, were purchased from BD PharMingen, unless otherwise noted. Cells (5 × 105) were blocked with 10% v/v normal goat serum (Vector, Burlingame, CA) (10 min; 4°C) then stained with mAb (30 min; 4°C). Appropriate isotype-matched IgGs were used as negative controls. The cells were analyzed using an EPICS Elite flow cytometer (Beckman Coulter, Hialeah, FL).

Pulsing of DC and autologous mixed leukocyte reaction (MLR)

The CD11c immunomagnetic bead (Miltenyi Biotec, Auburn, CA)—purified DC were incubated with allogeneic splenocyte lysates at a DC: splenocyte equivalent ratio of 1:10 for 24 h at 37°C. Normal B10 splenocyte lysates were obtained by three cycles of rapid freeze/thaw exposure in PBS. To disrupt DC:lysate clumps, the cells were washed extensively (3x, 700g, 5 min) with PBS containing 5 mM ethylenediamine tetraacetic acid following pulsing. Graded numbers of γ-irradiated (20 Gy) DC were then used as stimulators in 72 h MLRs with nylon-wool column-enriched syngeneic (C3H) splenic T cells as responders (2 × 105/mL) in 96-well, round-bottom plates, as described (11). For the final 18 h, individual wells were pulse-labeled with 1 μCi 3[H] thymidine. The amount of radioisotope incorporated was determined using a β scintillation counter. Recombinant murine IL-2 (R&D Systems; 100 U/mL) was added at the beginning of cocultures, where indicated. Results are expressed as mean c.p.m. ± 1 SD of triplicates.

Analysis of Ag uptake

The uptake of cell lysates by C3H DC was analyzed by flow cytometry. Thus, B10 splenocytes (H2Kb+) were labeled with CFSE (Molecular Probes, Eugene, OR) according to manufacturer's protocol, prior to lysis. Following pulsing, the percentage of CD11c+CFSE+H2Kb− cells was quantified as Ag-loaded DC.

Analysis of T cell apoptosis and intracellular cytokine production

T cells cocultured with DC at a 10:1 ratio were harvested on day 3 of MLR. Apoptosis was analyzed over time by staining externalized phosphatidylserine with FITC-annexin-V, in combination with the vital dye 7-amino-actinomycin D (7-AAD; BD PharMingen) according to the manufacturer's instructions. The cells were costained for CD3 (anti-CD3, 17A2; BD PharMingen) to allow specific analysis of DC by flow cytometry. For intracellular cytokine analysis, harvested T cells were restimulated with plate-bound anti-CD3 and soluble anti-CD28 (37.51; BD PharMingen) in the presence of Brefeldin A (Sigma), the latter to block cytokine secretion, for 5 h. After extracellular staining with fluorochrome-conjugated anti-CD3, CD4 or CD8 mAbs, cells were permeabilized with 1% saponin and stained for IL-2, IL-4, IL-5, IL-10 or IFNγ (BD PharMingen). Appropriate isotype-matched IgGs were used as negative controls.

In vivo imaging of labeled DC and immunohistochemical staining of tissue sections

The DC were labeled green with PKH-67 (Sigma), according to the manufacturer's protocol and infused i.v. (1.5 × 106 in 0.1 mL PBS) via the lateral tail vein. Spleen blocks were embedded in Tissue-Tek OCT (Miles Laboratories, Elkhart, IN), snap frozen in isopentane/liquid nitrogen, and stored at −80°C. Cryostat sections (8 μm) were fixed in 96% ethanol (10 min), blocked with 10% v/v normal goat serum, and incubated overnight (4°C) with biotinylated-anti-CD3 mAb. As a second step, slides were incubated with 1:3000 Cy3-streptavidin (Jackson Immunoresearch Lab, West Grove, PA), for 30 min at room temperature. Cell nuclei were stained with DAPI (4,6-diamidino-2-phenylindole; Molecular Probes, Eugene, OR). Slides were fixed in 2% paraformaldehyde, mounted in glycerol/PBS, and examined with a Zeiss Axiovert 135 microscope equipped with appropriate filters and a cooled CCD camera (Photometrics CH250, Tucson, AZ). Signals from different fluorochromes were acquired independently, and montages edited using the Adobe Photoshop software program (Adobe Systems, Mountain View, CA).

Vascularized heart transplantation

Heterotopic (intra-abdominal) heart transplantation was performed from B10 to C3H mice, as described (13). Animals received either no treatment or were injected i.v. with 1.5 × 106 immunobead-sorted alloAg-pulsed DC (day −7, or days −10, −3 and 0). A subtherapeutic dose of 1 mg/kg/day FK506 (Prograf® for i.v. use; Fujisawa Healthcare, Deerfield, IL) was administered i.m. for 10 consecutive days (days 0–9) in two groups. For the IL-2 treatment protocol, recipient mice were treated with 60 000 U rhIL-2 (Proleukin®, Chiron Therapeutics, Emeryville, CA) i.p. every 8 h for 3 days, starting 8 h post-transplant (14). Graft survival was assessed by daily transabdominal palpation. The rejection was defined by the complete cessation of cardiac contraction, and was confirmed histologically.

Statistical analyses

Statistical analysis was performed using the 2-tailed Student's t and Mann-Whitney tests, and a p value of <0.05 was considered significant. Graft survival data were compared by Kaplan-Meier analysis and the log-rank test. Results are expressed as means ± 1 SD.


Rapa inhibits DC maturation and their subsequent capacity to stimulate T cells through the indirect pathway

DC generated from C3H BM in the presence or absence of a clinically relevant concentration of Rapa (10 ng/mL) were incubated overnight on day 7 with freeze-thaw lysates of B10 splenocytes. Preliminary experiments revealed that a DC:splenocyte equivalent ratio of 1:10 was optimal for pulsing DC, and gave the most consistent results (data not shown). After pulsing, DC were analyzed for both spontaneous and LPS-induced maturation by flow cytometry. AlloAg-pulsed, gated CD11c+ Rapa-DC showed decreased surface expression of CD80, CD86 and MHC class II (IAk) molecules compared with alloAg-pulsed untreated control DC (Figure 1A). These DC were then washed extensively, purified by CD11c magnetic beading and then used as stimulators of naïve C3H splenic T cells. Rapa-treated, alloAg-pulsed DC were poorer stimulators of T cells compared to alloAg-pulsed control DC (Figure 1B). Lysates alone did not exhibit a significant stimulatory capacity for allogeneic naïve T cells, demonstrating that the direct pathway, a possible contributing factor that may arise from contamination of lysates with intact splenocytes or membrane fragments, was not a significant contributor to the induction of T cell proliferative responses.

Figure 1.

Inhibition of DC maturation and T cell allostimulatory function via the indirect pathway. (A) C3H BM-derived DC were generated as described in section ‘Materials and Methods’, in the presence or absence of 10 ng/mL Rapa. Rapa inhibited the cell surface expression of CD80, CD86 and MHC-II on gated CD11c+ cells on day 8, following alloAg pulsing. Upregulation of these molecules in response to LPS (1μg/mL) was also suppressed by exposure of DC to Rapa. The mean fluorescence intensity of CD11c+ cells expressing the marker of interest is indicated. (B) B10 alloAg-pulsed, Rapa-DC were inferior stimulators of naïve syngeneic C3H T cells, compared to alloAg-pulsed control (Ctr) DC. The ratio of DC:T cells is indicated in parentheses. Data are representative of results obtained from 3 (A) and 4 (B) similar experiments.

Rapa treatment does not interfere with lysate uptake by DC

To ensure that the suppressed T cell stimulatory activity observed was due to active regulation and not simply to decreased uptake of lysates by Rapa-DC, we pre-labeled B10 splenocytes with CFSE, a fluorescent reagent that binds stably to intracellular proteins, prior to cell lysis. Using the standard overnight pulsing protocol, we examined the uptake of lysates by CD11c+ cells using flow cytometry. To eliminate false-positive costaining due to DC-lysate cell surface adherence, we used a fluorochrome-conjugated mAb against MHC class I expressed by the allogeneic lysates (H2Kb) and compared CD11c+CFSE+H2Kb− populations of control and Rapa-DC (H2Kb is expressed by the B10-derived lysates, but is absent on the surface of C3H DC). As seen in Figure 2, Rapa treatment did not significantly block lysate uptake by DC when the DC were incubated with allogeneic cell lysates for an extended time period.

Figure 2.

Exposure to Rapa does not block alloAg uptake by DC. C3H BM-derived DC were cultured with or without Rapa. On day 7, they were harvested and incubated overnight with CFSE pre-labeled B10 splenocyte lysates. At the end of the pulsing period, the DC were stained for CD11c and colabeling with CFSE was analyzed to determine alloAg uptake. Additional H2Kb staining was performed to quantify DC-lysate clusters. The incidence of positive cells is indicated in each quadrant. Results are representative of three similar experiments.

Rapa treatment does not affect homing of DC to the spleen after their adoptive transfer, but confers capacity to suppress alloAg-specific responses

Next, to explore their potential for delivery of tolerogenic signals in vivo, we examined the in vivo migratory capacity of Rapa-DC. Following i.v. infusion (1.5–2 × 106) into naïve C3H recipients, C3H Rapa-DC labeled (green) with the lipophilic marker PKH-67 localized to T cell areas in spleen (Figure 3B) as efficiently as control DC (Figure 3A), as we have shown previously for genetically modified immature DC (15). To investigate the in vivo T cell priming ability of Rapa-DC, we isolated T cells from the spleens of recipient mice, 7 days after the adoptive transfer of these cells. The T cells were challenged ex vivo by either DC of B10 origin (Figure 4A) or B10 lysate-pulsed, C3H-derived DC (Figure 4B). The T cells that had been primed in vivo by alloAg-pulsed Rapa-DC showed markedly decreased proliferative responses to secondary stimulation, via either the direct or indirect pathways of allorecognition. This hyporesponsiveness was alloAg specific, as T cells responded strongly to third party (BALB/c; H2Kd) DC (Figure 4A). The addition of 100 U/mL of exogenous rIL-2 at the start of cultures abrogated this hyporesponsiveness and restored the T cell proliferative responses to levels similar to those of T cells primed by alloAg-pulsed, control DC.

Figure 3.

Rapa treatment does not affect the homing of DC to splenic T cell areas after adoptive transfer. Immunobead-purified, C3H BM-derived DC were labeled with PKH-67 (green) and injected i.v. into naïve syngeneic animals (1.5 × 106/animal). Twenty-four h later, recipient animals were killed and the spleens harvested. Frozen sections were stained with anti-CD3 mAb (red). Control DC (A; arrowhead) and Rapa-DC (B; arrowheads and inset) both localized to CD3+ T cell areas of the spleen. Nuclei were counterstained with DAPI (blue). Original magnification for both panels, 400×; inset B, 1000×. Images are representative of multiple sections at different levels for two animals/group analyzed.

Figure 4.

AlloAg-pulsed Rapa-DC induce Ag-specific T cell hyporesponsivenessin vivo. AlloAg-pulsed (B10), immunobead-purified control (Ctr)- or Rapa-DC were adoptively transferred (i.v.) to syngeneic C3H mice (1.5 × 106/animal). Seven days later, recipient animals were killed and splenic T cells challenged ex vivo with graded numbers of γ-irradiated allogeneic (B10) DC or third party (BALB/c) DC (A), or alloAg-pulsed syngeneic (C3H) DC (B). Mean proliferation of responder T cells in 72 h MLRs is shown (±1 SD). At the beginning of the cocultures, 100 U/mL rmIL-2 was added to the groups indicated. T cells from non-immunized animals (naïve) were used as controls. Results are from a single experiment representative of four (each) experiments performed.

Rapa-DC do not increase the incidence of T cell death but markedly inhibit T cell IL-2 and IFNγ production

To investigate other possible mechanisms underlying the regulatory influence of Rapa-DC on alloreactive T cell responses, we tested their potential to promote T cell apoptosis in vitro. We harvested T cells from DC:T cell cultures on day 3 and assessed the percentage of early apoptotic (Annexin-V+/7-AAD) and late apoptotic/necrotic (Annexin-V+/7-AAD+) T cells. As seen in Figure 5A, there was no difference in the incidence of T cell death between cultures in which control or Rapa-treated, alloAg-pulsed DC were used as stimulators. We also performed intracellular cytokine staining of T cells from these MLR cultures. Whereas a significant proportion of T cells primed indirectly by alloAg-pulsed control DC expressed IL-2 and IFNγ, production of these cytokines was markedly reduced in T cells stimulated by alloAg-pulsed Rapa-DC (Figure 5B,C). The expression of the Th2 signature cytokines IL-4, IL-5 and IL-10 was also decreased, but to a lesser extent, in these allostimulated T cells (Figure 5D).

Figure 5.

Rapa-DC suppress IL-2 and IFN (production by T cells, without affecting the incidence of T cell death. Naïve T cells were cultured with alloAg-pulsed control (Ctr)- and Rapa-DC at 10:1 ratio for 3 days. (A) Cells were then analyzed after staining of phosphatidylserine externalization with FITC-annexin-V in combination with the vital dye 7-AAD. The incidence of CD3+ cells expressing the marker of interest is indicated. (B–D) T cells were also permeabilized and analyzed for intracellular cytokine production. T cells primed in vitro by alloAg-pulsed Rapa-DC produced lower amounts of IL-2 and IFNγ (B, C); nevertheless this was not selective for these Th1 cytokines as IL-4, IL-5 and IL-10 production (D) were also decreased compared to T cells primed with Ctr-DC. Numbers in upper right quadrants indicate percent positive CD3+ cells for the cytokines indicated. (C) Results of four different experiments are shown. Results are representative of 3 (A), 4 (B, C) and 3 (D) experiments. (*) p < 0.05 versus Ctr-DC (Mann-Whitney test).

A single infusion of alloAg-pulsed, Rapa-DC prolongs alloAg-specific heart graft survival; multiple infusion leads to long-term survival

Finally, to explore the therapeutic potential of alloAg-pulsed, Rapa-DC in vivo, we employed the fully MHC-mismatched (B10 to C3H) heterotopic vascularized heart transplant model. In this model, B10 heart grafts are rejected by C3H recipients within 7–11 days (MST = 9.1 days) without immunosuppressive treatment. Graft survival was not affected significantly when alloAg-pulsed, control syngeneic DC were infused i.v. on day −7 (Figure 6). By contrast, a single infusion of donor alloAg-pulsed, Rapa-treated DC prolonged graft survival significantly (MST = 23.8 days; p < 0.005). This effect was Ag specific, as no graft prolongation was observed in the group that received Rapa-DC pulsed with third party (BALB/c) alloAg, and was reversed by the systemic administration of rIL-2 for 3 days, commencing 8 h post-transplant. In addition, the beneficial effect of donor alloAg-pulsed Rapa-DC was improved to a MST of 46.8 days (p = 0.0005 vs. untreated) by short-term administration of a subtherapeutic dose of FK506 (1 mg/kg/day, i.m., days 0–9), which alone did not prolong graft survival. Strikingly, repeated infusion of donor alloAg-pulsed Rapa-DC, but not donor alloAg-pulsed control DC (×3; days −10, −3 and 0) led to indefinite graft survival (>100 days) in 40% of otherwise unmodified graft recipients (Figure 6).

Figure 6.

AlloAg-pulsed Rapa-DC prolong heart graft survival. AlloAg-pulsed, control (Ctr)- or Rapa-DC of host strain origin were injected i.v. into syngeneic C3H mice, 7 days before the transplantation with B10 heart grafts. FK506 (1 mg/kg/day) was administered i.m. for 10 days and IL-2 (60 000 U/animal/8 h) was administered i.p. for 3 days from the time of transplant, where indicated. Untreated, third party (BALB/c) lysate-pulsed Rapa-DC-treated (Rapa-DC:third party) and FK506-treated mice were used as control recipients. Alternatively, the alloAg-pulsed Rapa-DC or alloAg-pulsed Ctr-DC were infused i.v. ×3 (days −10, −3 and 0) into otherwise unmodified graft recipients. ap < 0.0001, bp = 0.0005 and dp = 0.0014 versus untreated mice; cp = 0.0029 and ep = 0.0464 versus Rapa-DC-treated mice, analyzed by the log-rank test.


DC, once known almost exclusively for their unrivalled capacity to stimulate immune responses, are now recognized increasingly for their roles in immune regulation and the induction/maintenance of tolerance (2,5,6). In the immature state, DC exhibit inherent tolerogenic properties, as they fail to provide adequate costimulation for T cell activation. They can increase pancreatic islet (16) and vascularized organ transplant survival (13,17). DC can also delay the onset or regulate the severity of autoimmune diseases in animal models (18–21). In the transplantation setting, both donor and recipient DC contribute to immune responses that lead to allograft rejection (22–26). Direct alloAg presentation, mediated by donor APC, results in the activation of host T cells by allogeneic MHC and costimulatory molecules expressed on these donor APC. This in turn, leads to vigorous T cell proliferation and anti-graft effector immune responses. As donor DC undergo attrition, their role as presenters of alloAg subsides, and recipient DC, that can traffic to the graft become the predominant APC (27). Emerging evidence suggests that, once recipient DC populate the graft, they convey and present alloAg indirectly to T cells, in much the same way as other Ags in the periphery are presented in the normal steady state (28).

Reports of the induction by DC of peripheral tolerance to self or MHC class II Ags encountered in the absence of inflammatory stimuli (29,30), coupled to evidence that direct alloAg presentation diminishes with time after transplantation, provides a strong impetus for targeting indirect allorecognition via DC-induced cross-tolerance. In human volunteers, subcutaneous injection of Ag (influenza matrix peptide and keyhole limpet hemocyanin)-pulsed autologous immature myeloid DC led to specific inhibition of Ag-specific effector T cell functions (31). Moreover, IL-10-treated human monocyte-derived DC induce T cell anergy via cross-presentation of phagocytosed necrotic fragments (32). Similarly, immature DC pretreated with N-acetyl-l-cysteine induce alloAg-specific T cell hyporesponsiveness after loading with apoptotic cells (33). Furthermore, while mice infused i.v. with alloAg-pulsed peritoneal exudate cells (PEC) develop a delayed-type hypersensitivity (DTH) response similar to that of allograft recipients after challenge with the alloAg, i.v. infusion of PEC pulsed with alloAg in the presence of IL-10/transforming growth factor-β results in an anti-inflammatory, ‘allograft acceptor-like’ DTH (34). Herein, we show for the first time that DC exposed to Rapa become resistant to maturation and induce Ag-specific T cell hyporesponsiveness via indirect Ag presentation. In this model, T cells cross-primed by alloAg-pulsed, Rapa-DC become hyporesponsive to subsequent challenge with the same Ag, not only through the indirect pathway, but also the direct pathway, providing evidence for the regulation of both these major pathways of allorecognition by pharmacologically modified DC.

The Ag-specific T cell hyporesponsiveness induced by our approach appeared to be the result of Rapa-DC trafficking to secondary lymphoid tissue and active regulation of T cell function that was reversible by exogenous IL-2. Freshly isolated immature DC differ from mature DC in that they cannot provide adequate costimulation (signal 2) and inflammatory cytokine (e.g. IL-12p70) support (signal 3) in conjunction with their ability to ligate TCR through Ag presentation (signal 1). Whereas stimulation of T cells by mature DC results in their activation, priming by immature DC causes T cell anergy (35). Different approaches to generation of either stably immature or ‘alternatively-activated’ DC such as those conditioned with IL-10 (36,37) have been reported over the past few years. The IL-10-conditioned DC (38) and Vitamin D3-treated DC (39) both induce allogeneic T cell anergy. Repetitive stimulation of human T cells by allogeneic immature DC generates anergic T cells with limited resistance to IL-2 stimulation (40). In a similar manner, Rapa-DC become stably immature, and as we have reported previously (11), lose their capacity to produce IL-12p70 and TNF-α upon LPS stimulation. Rapa treatment also suppresses IL-12 signaling in DC via the inhibition of Janus kinase 2/Stat4 activation (41). Although our results suggest maturation blockade as the main mechanism of the Rapa-DC-induced T cell hyporesponsiveness, we cannot rule out the possibility of semi-mature or ‘alternative’ activation of DC exposed to Rapa. Importantly, after exposure to Rapa, DC retain their capacity to take up cell lysate Ag in extended overnight culture (16–24 h), in contrast to previous reports by us (12) and others (42) that concern inhibition of Ag uptake over much shorter periods (≤2h). Moreover, it should be noted that the overnight pulsing with cell lysates was performed in the absence of Rapa.

The T cell hyporesponsiveness induced by Rapa-DC was not the sole result of the cells' inability to prime T cells effectively. It was rather due to immunomodulatory activity of these cells since T cells primed by Rapa-DC later responded strongly to third party DC challenge, but responded to DC from the same donor comparatively poorly. These in vitro and ex vivo observations were confirmed by Ag-specific prolongation of organ graft survival following systemic infusion of alloAg-pulsed Rapa-DC. We also observed a striking inhibition of the two main mediators of the type I T cell response, i.e. IL-2 and IFNγ production by T cells stimulated with Rapa-DC. Although there was a clear balance shift in type I versus type II T cell biosignatures in the MLR cultures, there was no increase per se in the expression of the latter (IL-4, IL-5 and IL-10). This universal T cell cytokine downregulation is likely a contributing mechanism to the effects of Rapa-DC in transplant models, as Th2 cells play a role in alloimmune responses (25) and can be sufficient for graft rejection (43–45). Furthermore, IL-4 is recognized for its stimulatory effect on IL-12p70 production by DC (46). Thus it is possible that by reducing IL-4 production, activation of ‘bystander’ APC in the recipient is also inhibited by Rapa-DC. In our studies, we were not able to demonstrate an increase in the incidence or activity of T cells with previously defined regulatory T cell (Treg) phenotype such as CD4+CD25+, in response to either in vitro or in vivo stimulation by Rapa-DC (Taner T. et al., unpublished observations).

Strategies to utilize DC for the induction of donor-specific tolerance and the prolongation of organ allograft survival have until now been concerned mainly with donor DC (13,17,47–49), whereas studies exploiting recipient DC in therapeutic approaches to graft rejection have been limited. One exception is the use of DC–DC hybrids that express MHC of both parental APC, to delay the onset of alloAg-specific graft-versus-host disease when engineered to overexpress CD95L (50). Recently, in a rat kidney transplant model, preoperative infusion of dexamethasone-treated immature F1 DC, followed by CTLA4Ig, was found to promote the indefinite graft survival and immune regulation via the indirect pathway (51).

It has also been shown that intrathymic or i.v. injection of recipient BM-derived or thymic DC pulsed with immunodominant alloMHC-I-derived peptide prior to transplantation, prolongs cardiac or pancreatic islet allograft survival in anti-lymphocyte serum (ALS)-treated rats (52–55). These latter effects have been attributed, in part, to the induction of central tolerance by the allopeptide-pulsed DC as they were abrogated when the recipient rats were thymectomized prior to cell infusion (56). The clinical application of this strategy, however, is limited by the necessity to identify unrelated donor MHC peptides. The advantage and comparative simplicity of the present model lies in the use of whole donor cell lysates, which constitute the complete library of donor alloAgs to be presented indirectly by Rapa-DC. This approach also permits Ag-specific immunomodulation, as opposed to alternative approaches that target recipient DC via pharmacologic treatment of the host—a method shown to induce tolerogenic DC in graft recipients (57). The incomplete prolongation of graft survival we observed uniformly following a single infusion of the DC was presumably due to the activation of T cells via the direct pathway, since once this early response was suppressed, albeit minimally, by a transient sub-therapeutic dose of FK506 (which alone did not prolong allograft survival), rejection was more markedly and significantly delayed. Calcineurin inhibition might also work in other ways to augment the beneficial effect of pretreatment with alloAg-pulsed Rapa-DC, e.g. by inhibiting donor DC migration (58). Notably, we were able to enhance the tolerogenic potential of Rapa-DC by repeated infusion of the alloAg-pulsed cells, 40% of recipients in this protocol exhibited long-term graft survival, confirming the hyporesponsiveness achieved to both direct and indirect challenge in vitro. The long-term survivors were challenged with donor skin grafts on day 100, which were rejected within 2 weeks. Beating grafts harvested 130–150 days post-transplant exhibited chronic inflammatory changes and mononuclear cell infiltration.

In summary, Rapa-treated, alloAg-pulsed host DC can induce Ag-specific T cell hyporesponsiveness in vivo and prolong the survival of MHC-mismatched vascularized heart allografts. This approach that obviates systemic delivery of the DC-modifying pharmacologic agent constitutes a clinically applicable cell-based ‘negative vaccination’ strategy for suppression of allograft rejection.


The authors thank Mr. William J. Shufesky, Dr. Bridget L. Colvin and Ms. Jaime A. Cavallo for technical support and valuable discussion, Drs. Michael T. Lotze and Rosemary A. Hoffman for supplying rhIL-2 for in vivo use.

This work was supported by National Institutes of Health grants DK49745, AI41011, AI/DK51698 (to A.W.T.), HL69725 and AI55027 (to A.E.M.) and by the German Foundation of Hemotherapy Research (to H.H.). T.T. is the recipient of an American Heart Association Pre-doctoral Fellowship.