Direct versus Indirect Allorecognition: Visualization of Dendritic Cell Distribution and Interactions During Rejection and Tolerization

Transcription

1 American Journal of Transplantation 26; 6: Blackwell Munksgaard C 26 The Authors Journal compilation C 26 The American Society of Transplantation and the American Society of Transplant Surgeons doi: /j x Direct versus Indirect Allorecognition: Visualization of Dendritic Cell Distribution and Interactions During Rejection and Tolerization J. C. Ochando a,, N. R. Krieger b,c and J. S. Bromberg a,b,c, a Department of Gene and Cell Medicine, b Surgery and c Recanati/Miller Transplantation Institute, Mount Sinai School of Medicine, New York, NY 129, USA Corresponding authors: Jonathan S. Bromberg or Jordi C. Ochando, jon.bromberg@msnyuhealth.org or jordi.ochando@mssm.edu Interactions of donor and recipient dendritic cells (DCs) with CD4 + T cells determine the alloantigenic response in organ transplantation, where recipient T cells respond either directly to donor MHC, or indirectly to processed donor MHC allopeptides in the context of recipient MHC molecules. The present study evaluates donor and recipient alloantigen-presenting DC trafficking and their interactions with CD4 + T cells in the lymph nodes (LNs) and the spleen under tolerogenic treatment with anti-cd2 plus anti-cd3 mab compared with untreated rejecting conditions. CX 3 CR1 BALB/c (I-A d ) donor hearts were transplanted into C7BL/6 (I-A b ) mice and quantification of donor DC direct ( + or I-A d+ ) and recipient DC indirect (YAe + ) trafficking and interactions with host CD4 + T cells was performed by fluorescent microscopy. Our data indicate that although both direct and indirect interactions between CD4 + T cells and donor and recipient DCs occur shortly after engraftment, only indirect presentation persists in the LN, but not the spleen, of tolerized recipients. These data suggest that distinct anatomic lymphoid compartments play a critical role in peripheral tolerance induction and maintenance, and persistent indirect presentation to CD4 + T cells within the LNs is an important process during tolerization. Abbreviations: APC, antigen presenting cell; DC, dendritic cell; IHC, immunohistochemistry; LN, lymph node; MLC, mixed leukocyte culture Key words: Direct presentation, indirect presentation, migration, tolerance Received 4 November 2, revised June 26 and accepted for publication 22 June 26 Introduction Induction of tolerance to vascularized allografts in the mature peripheral immune system remains an unsolved goal in transplantation. Although much has been learned about the role of CD4 + T cells in transplant rejection and tolerance, the precise locations where CD4 + T cells interact with professional antigen presenting cells (APC) leading to antigen-specific unresponsiveness under systemic immunosuppression remain unknown. CD4 + T cells are tightly regulated by APC, which process alloantigens and present them in the context of their MHC and costimulatory molecules for T-cell activation to proceed. Donor and recipient APCs present alloantigen to T cells through two distinct pathways of allorecognition, direct and indirect presentations, respectively (reviewed in (1)). In the direct pathway, recipient T cells recognize peptides in context of intact MHC molecules on the surface of donor APCs in the host lymphoid organs, mainly spleen or lymph nodes (LNs). In contrast, in the indirect pathway, recipient T cells recognize donor-derived MHC and minor antigen peptides processed and presented by recipient APCs. The contribution of direct and indirect presentations in the initiation of rejection and the induction and maintenance of tolerance, remain unclear as both types of presentation can take place during the development of these processes. Historically, direct recognition has been linked to acute rejection by the demonstration of strong direct stimulatory responses in allogeneic mixed lymphocyte reactions (1). Additional evidence for the importance of direct presentation in acute rejection is illustrated by prolongation of graft survival by donor APC depletion, and restoration of graft rejection after the injection of small numbers of donor dendritic cells (DCs) in APC-depleted mice (1). Alternatively, the indirect pathway has been reported to initiate both acute and chronic rejections in nonvascularized grafts, where MHC class II deficient donor skin transplants were rejected by MHC class I deficient hosts (1); and for vascularized grafts, where MHC class I mismatched allografts transplanted into class I allopeptide immunized recipients were rejected after treatment with cyclosporine (2). Further evidence for the role of indirect presentation during rejection is supported by recent reports which demonstrate that recipient APC can present antigen indirectly in the draining LN and systemically in peripheral distant LNs with resultant CD

2 Allorecognition During Tolerization and Rejection T-cell activation and clonal expansion (3,4). More recently, it has been proposed that there is a sequential allorecognition pathway, where direct allorecognition initiates indirect allorecognition in culture, which may explain the contribution of both pathways of allorecognition during rejection (). Although rejection can occur through both the direct and indirect pathways of allorecognition, it appears that peripheral tolerance induction through direct presentation may not be possible (6). Indirect allorecognition therefore may play a more significant role in the establishment of peripheral tolerance, at least after vascularized heart transplantation (7). The crucial role for indirect allorecognition in peripheral tolerization has also been demonstrated using T-cell receptor transgenic CD4 + T cells, in which T cells responding only to indirectly presented alloantigen became anergic (8). Long-term tolerant recipients have also been reported to rely on mechanisms regulating the indirect pathway of unresponsiveness (9). The environment where alloantigen is recognized plays an essential role in the development of either rejection or tolerization (1), although the role that each secondary lymphoid organ plays in rejection or tolerization remains controversial. The initial priming of alloreactive CD4 + cells occurs mainly in the secondary lymphoid organs (4), although it can also occur within the allograft at later time points (11). For tolerance, the lymphoid organs or compartments where lymphocytes encounter and interact with alloantigen, under the influence of systemic immunosuppression, are currently not known. We have previously shown that T lymphocyte LN homing and localization are required for anti-cd2 plus anti-cd3 mab-induced tolerance to vascularized heart allografts (12,13). It is possible that distinct anatomic sites promote preferential pathways of antigen presentation, resulting in either rejection or tolerization to occur, as a consequence of specific alloantigen presenting cell migration to secondary lymphoid organs (14). Therefore, defining both direct and indirect alloantigen presentations and DC-CD4 + T-cell interactions in distinct anatomic compartments during leading to allorejection, or the induction and maintenance of tolerance. We studied donor and recipient DC cell numbers and their interactions with CD4 + T cells in the LNs and the spleen of tolerized and rejecting mice. Major differences in DC trafficking were associated with the initiation of the immune response to cardiac allografts compared to the induction of specific immune unresponsiveness. The precise kinetics of donor and recipient DC migration into lymphoid organs, and their interactions with CD4 + T cells, during the induction of either acute rejection or tolerance following transplantation, suggest an essential role for LN occupancy during the induction of tolerance to vascularized allografts. Materials and Methods Mice BALB/c (H-2 d ), C7BL/6 (H-2 b ) and CBA (H-2 k ) mice 8 1 weeks of age were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). CX3CR1 (1) mice on a BALB/c background were a gift from D. Littman (New York University). All mice were housed in a specific pathogen-free facility in microisolator cages. All experiments were performed with age- and sexmatched mice in accordance with Institutional Animal Care and Veterinary Committee-approved criteria. Reagents The 14-2C11 hamster antimurine CD3ξ hybridoma and the 12 1 rat IgG1 antimurine CD2 hybridoma were obtained as previously described (13); the Y-Ae hybridoma (16) was a gift from A.Y. Rudensky (University of Washington). The hybridomas were grown in culture, and supernatants were purified over protein G or A columns (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Vascularized cardiac transplantation BALB/c CX3CR1 hearts were transplanted as fully vascularized heterotopic grafts into allogeneic C7BL/6 or syngeneic BALB/c recipient mice, receiving either tolerogenic treatment with anti-cd2 plus anti-cd3 mabs, or control Ig mab as previously described (13). Graft function was monitored every other day by abdominal palpation. Rejection was defined as complete cessation of a palpable beat and was confirmed by direct visualization at laparotomy. Cell preparations Mice were sacrificed and the LNs (axillary, brachial, cervical, para-aortic and hepatic) and spleens were removed and gently dissociated into single-cell suspensions. RBCs were removed by tris-nh 4 CL lysis. Cells were placed in complete RPMI medium (RPMI 164 supplemented with 1% FCS, 1 mm sodium pyruvate, 2 mm L-glutamine, 1 IU/mL penicillin, 1 lg/ml streptomycin, 1 nonessential amino acids and 2 1 M 2-ME). BALB/c hearts were homogenized with % collagenase type II (Worthington, Lakewood, NJ, USA) for 3 min at 37 C before leukocyte isolation. Mixed leukocyte culture (MLC) A total of 2 1 responder splenocytes were cocultured in triplicate with 1 1 rad c -irradiated stimulator cells for 3 days. Eighteen hours before the termination of the culture, the wells were pulsed with. lci [ 3 H]-thymidine and incorporation was quantified with a scintillation counter. Results are expressed as stimulation index (SI), determined from mean of triplicate determinations ± SEM (n = 3 mice). Flow cytometry Fluorochrome-conjugated antibodies specific for CD11c, CD11b, NK1.1 and CD3 were purchased from PharMingen (San Jose, CA, USA). PE-conjugated anti-mouse PDCA-1 was purchased from Miltenyi (Auburn, CA, USA). An LSR II (BD, San Jose, CA, USA) was used for flow cytometry, and data were analyzed with FlowJo software (Tree Star, Ashland, OR, USA). Immunofluorescent microscopy Mice were sacrificed and two inguinal LNs/mouse were recovered at indicated time points. Biotinylated mouse anti-mouse I-A d ; biotinylated IgG2b, j; biotinylated rat anti-mouse CD4 IgG2a, j (H129.19); and biotinylated Rat IgG2a, j were purchased from PharMingen. Cy2-streptavidin and AMCAstreptavidin were purchased from Jackson Immunoresearch (Baltimore, PA, USA). The pea:i-ab complexes were detected in situ by staining sections sequentially with biotin-labeled Y-Ae antibody, streptavidin-hrp, followed American Journal of Transplantation 26; 6:

3 Ochando et al. by biotinyl tyramide (NEN), and then streptavidin-cy3 (PerkinElmer, Boston, MA, USA). All slides were mounted with Vectashield (Vector Laboratories, Burlingame, CA, USA) to preserve fluorescence. Images were acquired using a Leica DMRA2 fluorescence microscope (Wetzlar, Germany) and a digital Hamamatsu CCD camera (Hamamatsu-city, Japan). Separate green, red and blue images were collected and analyzed with Openlab software (Improvision, Coventry, UK). Captured image layers were sliced to standard density values and total cells number per mm 2 tissue were measured as single independent intensity objects with the Openlab cell measurement module. Twelve bit grayscale objects with an area less than.1 units were ignored. Final image processing was performed using Volocity software (Improvision, Coventry, UK). Statistical analysis For cell proliferation and cell counts, one-way analysis of variance (ANOVA) was performed at each time point. The group effects were all significant at p.. To examine individual differences, comparison between every pair of groups was performed. Results Donor and recipient alloantigen-presenting DC distribution during rejection and tolerance induction To assess donor DC migration to lymphoid tissues, we used BALB/c mice in which the gene was inserted into the CX3CR1 locus (CX3CR1 mice) (1). Heterozygous mice are immunologically, phenotypically and functionally normal, and the only + cells in the donor hearts are CD11c + CD11b + myeloid DCs and CD11c + PDCA-1 + plasmacytoid DCs (pdcs) (Figure 1). Hearts from these CX3CR1 donors were transplanted into allogeneic C7BL/6 or syngeneic BALB/c wild-type recipients. During acute rejection, donor DCs were present in the LNs during the first week after transplantation and decreased in number over that time (Figure 2A). On the contrary, the number of donor DCs present in the spleen during the first week following transplantation increased slightly during rejection (Figure 2C). During the first week after tolerance induction, the number of donor DCs in the LNs also decreased (Figure 2B) in a fashion similar to rejecting animals. In contrast to rejecting animals, the number of donor DCs declined in the spleen of tolerized recipients (Figure 2D). Low numbers of donor DCs were also present in the syngeneic controls. To assess recipient DC presenting donor alloantigen in lymphoid tissues, we took advantage of the YAe mab, which recognizes the epitope defined by donor MHC class II I-E d 2-68 peptide presented in the context of recipient MHC II I-A b (16). During rejection, recipient alloantigen-presenting, YAe +, DCs initially increased in the LNs shortly after transplantation, followed by a decrease by day 7 (Figure 2A), whereas in the spleen, they increased over the first week (Figure 2C). In contrast, during the first week after tolerance induction, recipient alloantigen-presenting DCs increased progressively in the LNs, whereas they were almost absent in the spleen (Figures 2 B,D). Overall, in the first week after transplantation, the most striking findings were the presence of recipient DCs carrying donor alloanti- SSC CD11b NK Untransplanted BALB/c CX3CR1 heart FSC FSC PDCA-1 CD Figure 1: Myeloid and plasmacytoid DCs constitute the majority of donor MHC Class II + cells in CX3CR1 mice. Donor cells from CX3CR1 untransplanted BALB/c hearts were gated on SSC and FSC and CD11c (not shown), and subsequently gated on and the indicated cell surface markers. Representative results of three independent experiments are shown (n = 3). gen in the spleen of rejecting recipients (Figure 2C) and in the LNs of tolerized recipients (Figure 2B). Direct and indirect DC-CD4 interactions during rejection and tolerance induction We next characterized direct and indirect DC-CD4 + T- cell interactions during acute rejection and following tolerance induction. Direct presentation was visualized by immunostaining for donor DC (CX3CR1 ) and CD4 + T cells (Figure 3E). Indirect presentation was visualized by double immunostaining with the Y-Ae mab (YAe-Cy3) and CD4 + T cells (Figure 3E). Binary color layers representing digital images were merged together to visualize and monitor cell interactions. The total number of CD4 + T-cell interactions between donor and recipient DCs was enumerated per mm 2 of tissue from the double-stained images of the distinct anatomic compartments. 249 American Journal of Transplantation 26; 6:

4 Allorecognition During Tolerization and Rejection LN a Rejection b Tolerization Spleen DC cell number/mm 2 DC cell number/mm d 2d 3d 4d d 6d 7d Syn2d Synd c d 2d 3d 4d d 6d 7d Syn2d Synd d Recipient Donor Syngeneic Recipient Donor Syngeneic 1 1 1d 2d 3d 4d d 6d 7d Syn2d Synd 1d 2d 3d 4d d 6d 7d Syn2d Synd days after transplantation Figure 2: Donor and recipient alloantigen-presenting DCs in the LN and the spleen of rejecting and tolerant recipients during the first week following transplantation. (A) Distribution of donor (CX3CR1 ) and recipient (YAe) alloantigen-presenting DCs during rejection in the LN. (B) Distribution of donor (CX3CR1 ) and recipient (YAe) alloantigen-presenting DCs during tolerance induction in the LN. (C) Distribution of donor (CX3CR1 ) and recipient (YAe) alloantigen-presenting DCs during rejection in the spleen. (D) Distribution of donor (CX3CR1 ) and recipient (YAe) alloantigen-presenting DCs during tolerance induction in the spleen. Four mice per group were sacrificed on days. 7 after vascularized cardiac transplantation in the anti-cd2 plus anti-cd3 mab-treated group, and from untreated control recipients. LNs and spleen were recovered, and cell numbers were measured using binary color layers after fluorescent immunohistochemistry. A total of 1 sections per organ were performed from each mouse. Results are representative of three independent experiments. During the first week following transplantation, there was an initial increase followed by a steady decrease in direct interactions in the LNs of rejecting animals (Figure 3A), in contrast to the spleen where there was a steady increase in direct interactions during this time (Figure 3C). The number of CD4 + T-cell indirect DC interactions in the rejecting LN and spleen was similar to the number of direct interactions (Figures 3 A,C), with more direct and indirect interactions in the spleen compared to the LN (Figure 3A vs. 3C). During tolerance induction, there was a decrease in direct interactions in both the LN and the spleen over the first week (Figures 3 B,D). In contrast, there were increasing numbers of indirect interactions in the LNs following treatment with anti-cd2 plus anti-cd3 mabs (Figure 3B), whereas there were decreasing numbers of indirect interactions in the spleen (Figure 3D). Together, the data suggest that during the initial week following vascularized transplantation there are more direct and indirect DC interactions and alloantigen presentation in the spleen during rejection, but greater indirect presentation in the LNs during tolerization. Evolution of donor and recipient alloantigen-presenting DC distribution We next characterized donor and recipient alloantigenpresenting DC numbers beyond the first week, at the time of acute rejection (day 9 11), or at 1 month after the induction of tolerance to vascularized heart transplants. At the time of acute rejection, there were a few donor DCs in the LNs, whereas there were greater numbers in the spleen (Figure 4 A,B). At the time of rejection, the number of recipient YAe + DCs presenting donor alloantigen was also few in number in the LNs, whereas they were 3-fold greater in the spleen (Figure 4 A,B). In tolerogen-treated animals, by day 3 there were no donor DCs present within either the LNs or the spleen (Figure 4 A,B), consistent with the trend noted in the first week after transplant (Figure 2 A,B). On the contrary, the density of alloantigen-presenting, YAe +, recipient DCs in the LNs was greater in the anti-cd2 plus anti-cd3 mab treated compared to the rejecting untreated recipients, a difference also consistent with trend in the first week after transplantation (Figures 2B and 4A). In the American Journal of Transplantation 26; 6:

5 Ochando et al. Figure 3: Direct and indirect DC-CD4 T cell interactions in the LN and the spleen of rejecting and tolerant recipients during the first week following transplantation. (A) Distribution of direct and indirect DC-CD4 + T cells interactions during rejection in the LN. (B) Distribution of direct and indirect DC-CD4 + T-cells interactions during tolerance induction in the LN. (C) Distribution of direct and indirect 2492 American Journal of Transplantation 26; 6:

6 Allorecognition During Tolerization and Rejection spleen, alloantigen-presenting recipient DCs were almost absent by day 3 (Figure 4B). Evolution of direct and indirect DC-CD4 interactions Direct and indirect interactions between CD4 + T cells and DCs were also measured at the time of acute rejection and 3 days following tolerogen treatment (Figure 4C E). During rejection, the number of direct interactions was fewer in the LNs compared to the spleen. During rejection, the number of indirect interactions was also substantially less in the LNs compared to the spleen. After anti-cd2 plus anti-cd3 mab treatment, while no direct interactions were observed in the LNs or spleen by day 3, indirect interactions were dramatically increased in the LNs (Figure 4C), yet indirect interactions were almost absent in the spleen (Figure 4D). Figure 4E shows representative images with lower numbers of direct and indirect interactions in the rejecting LNs compared to the rejecting spleen, and substantial numbers of indirect interactions in LNs of tolerized recipients. Differential alloreactivity of T cells from the LN and the spleen Since we demonstrated differences in DC distribution and interactions with CD4 + T cells in the secondary lymphoid tissue compartments of the anti-cd2 plus anti-cd3 mabtolerogen treated and acutely rejecting untreated recipients, we next assessed the cellular function of lymphocytes recovered from the LNs and the spleen by evaluating MLC responses. Figure shows that the lymphocytes from the LNs and the spleen of untreated rejecting mice have an increasing allospecific response from the fifth posttransplant day to the day of rejection. Interestingly, the lymphocytes from LNs and spleen of the tolerogen-treated recipients also have a strong response to alloantigen on the fifth posttransplant day, similar in magnitude to that of the untreated rejecting recipients. However, the lymphocytes from the tolerogen-treated mice are unresponsive to specific alloantigen at later times (days 1 3). Responsiveness of cells from tolerogen-treated mice to third party CBA (H-2 k ) stimulators indicated antigen-specific tolerance (data not shown). These results show that unresponsiveness takes time to develop and is distributed throughout all compartments. Discussion DCs are key mediators of the immune response, and continuously recirculate through lymphoid and nonlymphoid tissues in response to a variety of stimuli. Differences in donor and recipient DC trafficking and distribution have been reported to play a major role in the initiation of the immune response to cardiac allografts (17), and in the induction of specific immune unresponsiveness (18,19). The anatomic domains where lymphocytes interact with alloantigen, and whether the interactions for immunity and rejection are anatomically separate from those important for tolerance, are likely critical determinants in graft survival. Uncovering APC trafficking, as well as DC-CD4 + T-cell interactions, is therefore essential for a complete understanding of immune function and regulation in transplantation. We examined whether the particular pathway by which allopeptide was presented to CD4 + T cells in separate anatomic compartments could be indicative of the development of either tolerance or rejection. During rejection indirect and direct presentations occur with similar kinetics (11). Here, we extend these studies by analyzing the kinetics in mice treated with a tolerogenic treatment. Our results reveal that recipient indirect alloantigenpresenting DC homing and localization within the LNs are associated with tolerization, suggesting that the LN domain provides a unique environment for peripheral tolerance. On the contrary, the localization of donor and recipient alloantigen-presenting DCs to other compartments, such as the spleen, is predominant during rejection. The quantitative and temporal interactions of DC-CD4 + T cells in the LNs compared to other compartments also indicate further the development of tolerance instead of rejection. Table 1 summarizes the relative magnitude of direct and indirect presentation of alloantigen during rejection and tolerization in the LNs and spleen. While these results suggest that the LNs provide a preferential milieu for tolerogenic DC-CD4 + T-cell interactions, it is important to define if other secondary lymphoid organs, such as the spleen, influence tolerance. Whereas priming can take place in the spleen, further experiments should be directed to investigate whether the spleen can also provide a tolerogenic environment. It is possible that the spleen may play a major role during rejection, since we observed persistent donor alloantigen-presenting DCs interacting with recipient CD4 + T cells in the spleen of untreated recipients. Even when donor DCs represent only a small fraction of the DCs in the compartments under study, they can present antigen to several T cells and induce their proliferation leading to graft rejection, similar DC-CD4 + T-cells interactions during rejection in the spleen. (D) Distribution of direct and indirect DC-CD4 + T-cells interactions during tolerance induction in the spleen. (E) Representative images of direct allopresentation showing interactions between donor DC (green) and recipient CD4 + T cells (red). Indirect allopresentation shows interactions between alloantigen-presenting recipient DC stained with YAe (blue) and recipient CD4 + T cells (red). Images are representative of LNs and spleens collected from four individual animals. A total of 1 sections per organ were performed from each mouse. Original magnification 2. Inset magnification, 1. Experimental procedure as in Figure 2. American Journal of Transplantation 26; 6:

7 Ochando et al. Figure 4: Direct and indirect alloantigen-presenting DC and interactions with CD4 + T cells in the LN and spleen of rejecting and tolerant recipients at the time of rejection and 3 days following transplantation. (A) Donor DC (I-A d ), and recipient DC (YAe-blue) cell numbers at the time of rejection and during tolerization in the LN. (B) Donor DC (I-A d ) and recipient DC (YAe) cell numbers at the time of rejection and during tolerization in the spleen. (C) Direct and indirect DC-CD4 + T-cell interactions at the time of rejection and during tolerization in the LN. (D) Direct and indirect DC-CD4 + T-cell interactions at the time of rejection and during tolerization in the spleen. Cells were enumerated by fluorescent microscopy as in Figures 1 and 2. Results from untreated control (day1) or tolerant (day 3) recipient mice are shown. Results are mean of two independent experiments ± SEM. (E) Representative images of direct allopresentation showing distribution of interactions between donor DC stained with I-A d (yellow) and recipient CD4 + T cells (red). Indirect allopresentation shows distribution of interactions between alloantigen-presenting recipient DC stained with YAe (blue) and recipient CD4 + T cells (red). Images represent LN and spleen collected from four different animals. A total of 1 sections per organ were performed from each mouse. Original magnification 2. Inset magnification, American Journal of Transplantation 26; 6:

8 Allorecognition During Tolerization and Rejection a [ 3 H]-TdR uptake (stimulation index) b [ 3 H]-TdR uptake (stimulation index) LN days after transplantation Spleen days after transplantation Tolerization Rejecting Tolerization Rejecting Figure : Anti-CD2 plus anti-cd3 mab therapy alters T-cell alloreactivity. LNs and spleens and were recovered from anti-cd2 plus anti-cd3 mab-treated recipients (n = ) on days, 1, 2 and 3 following transplantation; and untreated control recipients on days and 1 (time of acute rejection). Lymphocytes were isolated from each anatomic compartment, and samples were pooled into single-cell suspensions. Results are expressed as stimulation index (SI), determined from mean of three independent experiments ± SEM. Table 1: Relative magnitude of direct and indirect alloantigen presentation in the LN and spleen during rejection and tolerization LN Spleen Rejection D + ++ I Tolerization D + + I to recipient DCs (2). Consistent with this, Matsuno and colleagues (17) showed that DCs from untreated rejecting donor cardiac transplants migrate preferentially to the spleen, whereas they are absent in distant LNs. In this report, we extend this finding to indirect presentation, in that high levels of indirect interactions were observed in the spleen of untreated acutely rejecting recipients. In addressing the trafficking of recipient DCs and their indirect interactions with CD4 + T cells during tolerance, we observed donor alloantigen-presenting recipient DCs in the LNs, but not the spleen. Yet the MLC data (Figure ) show unresponsiveness in both the LN and the spleen. This suggests that the unresponsiveness to alloantigen observed in tolerant mice may be due in part to active suppression, perhaps as the result of the expansion/and or migration of regulatory T cells through the indirect pathway of allorecognition. Consistent with this idea, we recently reported that regulatory T cells develop in the LNs of tolerant animals under the same tolerogenic conditions (13). Likewise, Lechler s group recently showed that unresponsiveness to alloantigens was established through indirect, but not direct, pathways of allorecognition in DC coexpressing donor and recipient alloantigen, suggesting that expansion of allospecific regulatory T cells occurs through indirect recognition (21). It is also possible that the cells in the spleen 3 days after transplant may have been tolerized in the LN and are in the spleen as this is part of the normal migratory route of these cells. We conclude that the specific anatomic domains where lymphocytes interact with alloantigen presenting cells, through the direct or indirect pathway of allorecognition under the influence of systemic immunosuppression, are critical determinants of the outcome of vascularized cardiac allografts. Further study of the receptors and ligands involved during trafficking of DCs, and indirect allorecognition between DC and CD4 + T cells in the LNs, may uncover the detailed mechanisms for tolerance induction and maintenance. Acknowledgments We acknowledge the technical contribution of Minwei Mao, Dan Chen and Yansui Li, and the helpful discussion with Drs. Barbara Murphy, Bernd Schröppel and Sergio Lira. This work was supported by National Institutes of Health Grants R1 AI41428 and AI6276 (to J.S.B.). References 1. Gould DS, Auchincloss H, Jr. Direct and indirect recognition: The role of MHC antigens in graft rejection. Immunol Today 1999; 2: Lee RS, Yamada K, Houser SL et al. Indirect recognition of allopeptides promotes the development of cardiac allograft vasculopathy. Proc Natl Acad Sci U S A 21; 98: Richards DM, Dalheimer SL, Ehst BD et al. Indirect minor histocompatibility antigen presentation by allograft recipient cells in the draining lymph node leads to the activation and clonal expansion of CD4(+) T cells that cause obliterative airways disease. J Immunol 24; 172: Reed AJ, Noorchashm H, Rostami SY et al. Alloreactive CD4 T cell activation in vivo: An autonomous function of the indirect pathway of alloantigen presentation. J Immunol 23; 171: Wallgren AC, Andersson B, Backer A, Karlsson-Parra A. Direct allorecognition promotes activation of bystander dendritic cells and licenses them for Th1 priming: A functional link between direct American Journal of Transplantation 26; 6:

9 Ochando et al. and indirect allosensitization. Scand J Immunol 2; 62: Rulifson IC, Szot GL, Palmer E, Bluestone JA. Inability to induce tolerance through direct antigen presentation. Am J Transplant 22; 2: Yamada A, Chandraker A, Laufer TM, Gerth AJ, Sayegh MH, Auchincloss H, Jr. Recipient MHC class II expression is required to achieve long-term survival of murine cardiac allografts after costimulatory blockade. J Immunol 21; 167: Quezada SA, Fuller B, Jarvinen LZ et al. Mechanisms of donorspecific transfusion tolerance: Preemptive induction of clonal T- cell exhaustion via indirect presentation. Blood 23; 12: Asiedu CK, Dong SS, Lobashevsky A, Jenkins SM, Thomas JM. Tolerance induced by anti-cd3 immunotoxin plus 1- deoxyspergualin associates with donor-specific indirect pathway unresponsiveness. Cell Immunol 23; 223: Lakkis FG, Arakelov A, Konieczny BT, Inoue Y. Immunologic ignorance of vascularized organ transplants in the absence of secondary lymphoid tissue. Nat Med 2; 6: Baratin M, Bonin K, Daniel C. Frontline: Peripheral priming of alloreactive T cells by the direct pathway of allorecognition. Eur J Immunol 24; 34: Bai Y, Liu J, Wang Y et al. L-selectin-dependent lymphoid occupancy is required to induce alloantigen-specific tolerance. J Immunol 22; 168: Ochando JC, Yopp AC, Yang Y et al. Lymph node occupancy is required for the peripheral development of alloantigen-specific Foxp3+ regulatory T cells. J Immunol 2; 174: Ochando JC, Homma C, Yang Y et al. Alloantigen-presenting plasmacytoid dendritic cells mediate tolerance to vascularized grafts. Nat Immunol 26; 7: Jung S, Aliberti J, Graemmel P et al. Analysis of fractalkine receptor CX(3)CR1 function by targeted deletion and green fluorescent protein reporter gene insertion. Mol Cell Biol 2; 2: Murphy DB, Lo D, Rath S et al. A novel MHC class II epitope expressed in thymic medulla but not cortex. Nature 1989; 338: Saiki T, Ezaki T, Ogawa M, Matsuno K. Trafficking of host- and donor-derived dendritic cells in rat cardiac transplantation: Allosensitization in the spleen and hepatic nodes. Transplantation 21; 71: Niimi M, Shirasugi N, Ikeda Y, Wood KJ. Oral antigen induces allograft survival by linked suppression via the indirect pathway. Transplant Proc 21; 33: Steptoe RJ, Fu F, Li W et al. Augmentation of dendritic cells in murine organ donors by Flt3 ligand alters the balance between transplant tolerance and immunity. J Immunol 1997; 19: Nathan MJ, Mold JE, Wood SC et al. Requirement for donor and recipient CD4 expression in cardiac allograft rejection: Induction of Th1 responses and influence of donor-derived dendritic cells. J Immunol 24; 172: Mirenda V, Berton I, Read J et al. Modified dendritic cells coexpressing self and allogeneic major histocompatability complex molecules: An efficient way to induce indirect pathway regulation. J Am Soc Nephrol 24; 1: American Journal of Transplantation 26; 6: