Mechanisms of donor-specific transfusion tolerance: preemptive induction of clonal T-cell exhaustion via indirect presentation

Transcription

1 TRANSPLANTATION Mechanisms of donor-specific transfusion tolerance: preemptive induction of clonal T-cell exhaustion via indirect presentation Sergio A. Quezada, Bruce Fuller, Lamis Z. Jarvinen, Mercedes Gonzalez, Bruce R. Blazar, Alexander Y. Rudensky, Terry B. Strom, and Randolph J. Noelle Induction of transplantation tolerance to alloantigens without general immunosuppression remains an enduring challenge. Injecting a donor-specific transfusion (DST) of spleen cells together with blocking CD154 antibody prior to graft transplantation is an effective way to induce long-lived graft acceptance. Using a novel T-cell receptor (TCR) transgenic (Tg) model of CD4 T-cell mediated rejection, Introduction this study sheds new insights into the cellular basis for enhanced graft survival induced by DST and CD154. The study shows that DST and CD154 induce an early, robust, abortive expansion of the Tg T cells that results in profound anergy. This is contrasted with the more delayed, regional, productive response elicited by an allogeneic graft. Studies show that the induction of tolerance to the allograft induced by DST is mediated by indirect presentation by host antigen-presenting cells. Based on these observations, we conclude that DST and CD154 preemptively tolerize the alloreactive T-cell compartment to prohibit subsequent responses to the immunogenic allograft. (Blood. 2003;102: ) 2003 by The American Society of Hematology For more than 2 decades, it has been recognized that the infusion of whole blood from donors (donor-specific transfusion [DST]) into recipients can prolong allograft survival in humans, as reviewed in humans 1 and mice. 2 More recently, it has been shown that the prolonged survival of allografts induced by DST is synergistically enhanced by the coadministration of CD154 ( CD40L). 3 In some cases, permanent graft survival of allogeneic grafts can be observed with CD154 and DST. While some insights into the underlying mechanisms of graft survival have been gained, our understanding of the cellular basis for allograft tolerance is incomplete. A number of recent studies using this means to induce graft survival have shown that active suppression plays an essential role in silencing the effector function of graft-rejecting T cells. While a role of regulatory T cells (Treg s) in this process has been tentatively identified, little is known about the fate and function of alloreactive effector T cells upon exposure to DST and CD154. It is hypothesized that one cellular mechanism of DST/ CD154- induced graft tolerance is the inactivation of alloreactive CD4 and CD8 effector T cells. 2 Based on studies of Buhlmann et al, 4 it was speculated that direct recognition of the infused allogeneic B cells (DST) by alloreactive T cells in the CD154-suppressed environment resulted in inadequate DST activation, with reduced upregulation of costimulatory molecules, and cytokine production by the DST. Under these conditions, it was hypothesized that clonal tolerance of the alloreactive T-cell population was rendered. Whether this tolerance was the result of clonal ignorance, anergy, or apoptosis has not been resolved because of the inherent constraints of the systems used. The use of TCR transgenic (Tg) models to study the basis for T-cell tolerance has permitted great insights into the spectrum of possible defects that can account for the tolerant state. Alloreactive T-cell fate and function have been directly assessed in the studies presented herein through the use of a novel CD4 TCR Tg model wherein alloantigens expressed by the DST, together with CD154 blockade, induce profound unresponsiveness. As such, questions as to the fate and function of normal and tolerant alloreactive T cells have been addressed. Many hypotheses of DST-induced tolerance are based on the proposition that host alloreactive T cells directly recognize alloantigen on the DST. However, recent studies have shed doubt on this premise. Using a spectrum of DST allotypes, matched or mismatched with the host, studies by Niimi et al 5 suggested that presentation of alloantigen-derived peptides in the context of self major histocompatibility complex (MHC) was essential for the beneficial effect of haplotype-shared blood transfusions. If this is true, then processing and presentation of DST-derived allopeptides by the host antigen-presenting cells (APCs) is critical in alloantigen-induced tolerance. Accepting this proposition, one must presume that the synergy observed with CD154 is due to its impact on the host APC machinery. Blockade of CD154 exerts profound effects on the function, longevity, and differentiation of dendritic cells (DCs). 6 As such, the effect of CD154 blockade is to shorten the duration of antigen presentation by the DCs and to limit their capacity to be immunogenic. Preventing DC maturation, and at the same time delivering DST, may be a superlative means for From the Department of Microbiology & Immunology, Dartmouth Medical School, Lebanon, NH; the Division of Bone Marrow Transplantation, University of Minnesota, Minneapolis, MN; the Howard Hughes Institute, University of Washington, Seattle, WA; and the Beth Israel Deaconess Medical Center, Department of Medicine, Harvard, Boston, MA. Submitted February 21, 2003; accepted May 8, Prepublished online as Blood First Edition Paper, May 15, 2003; DOI /blood Supported by grant A (R.J.N.), CA91436 (R.J.N.), AI (B.R.B.), 2R37 HL56067 (B.R.B.), HL63452 (B.R.B.), and the Rosaline Borison fellowship (S.A.Q.). S.A.Q. and B.F. contributed equally to the work presented in this study. Reprints: R. J. Noelle, Department of Microbiology & Immunology, Dartmouth Medical School, 1 Medical Center Dr, Lebanon, NH 03756; rjn@dartmouth.edu. The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked advertisement in accordance with 18 U.S.C. section by The American Society of Hematology 1920 BLOOD, 1 SEPTEMBER 2003 VOLUME 102, NUMBER 5

2 BLOOD, 1 SEPTEMBER 2003 VOLUME 102, NUMBER 5 MECHANISM OF CD154/DST-INDUCED T-CELL ANERGY 1921 inducing tolerance to alloantigen-derived epitopes that are presented by host DCs. It is thought that autologous apoptotic cells that are processed and presented by nonmatured host DCs are critical to maintain peripheral tolerance. 7-9 DST and CD154 may, in fact, take advantage of similar mechanisms to those used in maintaining peripheral self-tolerance in order to induce allotolerance. Data presented herein directly address and measure the contribution of indirect presentation to DST-induced graft survival. Using the CD4 TCR Tg system (TEa), in which receptor specificity is directed to an allopeptide derived from I-E and presented in the context of class II MHC (H-2 b ), we have tracked the behavior of CD4 alloreactive T cells in response to DST, CD154, and an allograft. Data show that DST induces a rapid, robust expansion of alloreactive T cells that is abortive and results in profound T-cell anergy. Blocking of CD154 does not qualitatively change the response to DST, but serves to magnify the intensity of the unresponsiveness. In contrast to the alloresponse to DST, the Tg T-cell response to the allograft is regional, robust, and productive, giving rise to highly responsive alloreactive T cells that ultimately infiltrate the graft and mediate rejection. Therefore, the preemptive and near complete unresponsiveness induced by DST/ CD154 serves to extinguish the subsequent response to the allograft. Furthermore, the data clearly show that DST does not directly present alloantigen to the host, and serves only as an alloantigen depot, providing allopeptides to be presented by host APCs. In light of these findings, a novel and cohesive model of DST/ CD154-induced allotolerance is presented. Materials and methods Mice C57BL/6 MHC class II deficient mice and recombination activating gene (RAG) knock-out (KO) mice were purchased from Taconic (Germantown, NY). CB6F 1 mice (hybrid of C57BL/6 and Balb/c), C57BL/6, C57BL/6 CD45.1, and Balb/c mice were purchased from the National Cancer Institute (Frederick, MD). The TEa CD4 TCR Tg mice 10 were kindly provided by Dr Alexander D. Rudensky (University of Washington, Seattle). CD4 Tg cells express a TCR that recognizes the peptide ASFEAQGLANIAVDKA in the context of I-A b. This peptide corresponds to the positions 52 to 68 from the -chain of I-E class II molecules and is expressed in all APCs from H-2 b /I-E strains (CB6F 1 ). C57BL/6 mice are H-2 b but I-E, whereas BALB/c are H-2 d and I-E, and therefore their F 1 hybrids are H-2 b /I-E and able to directly present antigen. The TEa Tg mice were bred to C57BL/6, congenic C57BL/6 CD45.1 (Ly5.2) mice at the animal facility at Dartmouth Medical School. All mice were bred and housed in microisolator cages in a pathogen-free facility. Skin grafting Skin grafting was performed as a modification of the technique used by Markees et al. 11 Briefly, age-matched male CB6F 1 mice were used as donors of both spleen cells (DST) and skin grafts. In some groups, age-matched C57BL/6 grafts were used as negative control skin donors. CD45.1 and CD45.2 were used as additional congenic markers for tracking the TEa Tg T cells. Age-matched CD45.1 TEa Tg mice were used as donors of alloreactive T cells. Quantification of the TEa Tg T cells was done by staining with anti V 2-TCR phycoerythrin (PE) and anti CD4 fluorescein isothiocyanate (FITC). Recipients were age-matched RAG KO mice. On day 1, donor CB6F 1 mice were killed and tail-skin grafts (0.5 cm 0.5 cm) 12 or mechanically disaggregated spleen cell suspensions were prepared from them. Recipient mice were injected with or without DST cells and TEa Tg T cells in 500 L Hanks balanced salt solution by tail vein injection (intravenously) and 500 g of CD154 (clone MR-1) or control hamster immunoglobulin (H-Ig) in phosphatebuffered saline (PBS) intraperitoneally. Mice were injected with CD154 or H-Ig 3 times per week for the duration of the experiment. On day 0, recipient mice were anesthetized with 50 g per gram body weight of each of ketamine and xylazine injected intraperitoneally (15 mg/ml in PBS), and CB6F 1 or C57BL/6 skin grafts were prepared using established methods. 12 Rejection was defined as the day on which less than 20% of the skin graft remained. All antibodies were obtained from Pharmingen (San Diego, CA). Statistical analysis Survival data were analyzed using the Kaplan-Meier method with the Wilcoxon rank test and the log-rank test used to verify the significance of the difference in survival between groups. P values less than.05 were considered statistically significant. Fluorescence-activated cell analysis Lymph node (LN) cell suspensions were stained to analyze expansion and purity of TEa Tg T cells in all different groups. This was assessed by staining with anti CD45.1-FITC and anti V 2-TCR-PE. All antibodies were obtained from Pharmingen. Purification and adoptive transfer of in vivo stimulated TEa Tg T cells Age-matched RAG KO mice were injected intravenously with DST cells and CD45.1 TEa Tg T cells, and intraperitoneally with 500 g of either CD154 or H-Ig 3 times per week. On day 7, recipients or naive TEa mice were killed, and their LNs were harvested and mechanically disaggregated. Cells were positively selected for CD45.1 using anti CD45.1- biotin and streptavidin magnetic beads according to the manufacturer s instructions (Miltenyi Biotech, Auburn, CA), stained with anti V 2- TCR-PE and anti CD45.1-FITC to determine percentage of selected T cells (recipients, 70%; naive TEa, 90%), and counted. Age-matched male or female RAG KO recipients were injected with TEa Tg T cells harvested from DST / CD154 treated RAG KO mice. Skin grafting of CB6F 1 grafts was performed, and grafts were monitored as described in Skin grafting. Antibodies were obtained from Pharmingen. In vivo expansion of TEa Tg T cells On day 0, recipient RAG KO mice received CD45.1 TEa Tg T cells and either (1) DST cells (from CB6F 1 or Balb/c donors) with or without CD154, (2) a CB6F 1 skin graft with or without CD154, (3) DST cells (CB6F 1 or Balb/c) and a CB6F 1 skin graft with or without CD154, or (4) a C57BL/6 skin graft. Mice treated with CD154 received 500 g 3 times per week, as described in Skin grafting, and were compared with mice receiving the same dose of H-Ig. Mice were killed either at day 7 for in vitro recall assays or at day 9 for determination of in vivo expansion. For comparison of local versus nonlocal expansion, lymph node cells were harvested, and mechanically disaggregated. Total number of cells per lymph node was determined and standardized to the percentage of TEa Tg positive cells determined by flow cytometry. In vitro recall responses Media used were RPMI (Bio-Whitaker, Walkersville, MD) containing 10% fetal bovine serum, 2 mm L-glutamine, M 2-mercaptoethanol, 100 U/mL penicillin, and 100 g/ml streptomycin. Splenocytes from CB6F 1 or C57BL/6 mice were irradiated with 30 Gy (3000 rad), and cells per well were added to separate wells of a 96-well plate in 100 L media. At day 7, LN cells from the different groups were harvested and selected for CD45.1 expression as described in Purification and adoptive transfer of in vivo stimulated TEa Tg T cells, stained with anti CD45.1- FITC and anti V 2-TCR-PE, counted, and plated in 96-well plates. TEa Tg T cells (20 000) in 100 L media were added to wells containing 100 L irradiated CB6F 1 cells, 100 L irradiated C57BL/6 cells, or 100 L media. Some wells containing irradiated CB6F 1 cells or irradiated C57BL/6 cells received 100 L media and no TEa Tg T cells. Replicate plates were

3 1922 QUEZADA et al BLOOD, 1 SEPTEMBER 2003 VOLUME 102, NUMBER 5 Figure 1. CD154 DST delays skin graft rejection by alloreactive Tg CD4 T cells. On day 1, C57BL/6 RAG KO mice were injected with TEa Tg T cells with or without CB6F 1 spleen cells followed by a CB6F 1 skin graft on day 0. Mice were treated with CD154 or control H-Ig and scored for graft rejection 3 times per week. Grafts were considered rejected when completely necrotic, or when less than 20% of the graft remained. Data were pooled from 3 separate experiments. ( represents H-Ig [n 13]; f, DST [n 16];, CD154 [n 18]; and F, DST CD154 [n 17].) incubated to assess cytokines and proliferation. Proliferation and cytokines were assessed as previously described. 13 Assessment of cell proliferation by measuring cytoplasmic dye dilution To follow in vivo kinetics of division, TEa Tg T cells were labeled with the intracellular fluorescent dye 5- (and 6-) carboxyfluorescein diacetate succinimidyl ester (CFSE) obtained from Molecular Probes (Eugene, OR) prior to adoptive transfer into naive RAG KO recipients. At days 4 and 8, local and nonlocal LN cells were recovered and assayed by multicolor flow cytometry, gating in the TEa Tg cell population (as previously described) to detect the dilution of the dye caused by cell proliferation. Each successive cellular generation exhibits half of the intensity of CFSE fluorescence of its parental population. Assessment of graft infiltration On day 0, recipient RAG KO mice received CD45.1 TEa Tg T cells and either (1) DST cells with or without CD154, (2) a CB6F 1 skin graft with or without CD154, (3) DST cells and a CB6F 1 skin graft with or without CD154, or (4) a C57BL/6 skin graft. Mice treated with CD154 received 500 g 3 times per week, as described throughout Materials and methods, and were compared with mice receiving the same dose of H-Ig. On days 4, 7, and 14 skin grafts were removed, formalin fixed, and processed for hematoxylin and eosin (H&E) staining by established protocols. Results in polyclonal systems (MST 51 days). 12 DST or CD154 alone significantly delayed graft rejection (MST 35.5 days, P.05; MST 41.7 days, P.0002, respectively), but not to as great a degree as with DST and CD154 together (MST 72.4 days, P.0001). All groups treated with CD154 or DST have some rare long-term surviving grafts ( 100 days), the largest percentage of these being in the DST CD154 treated group (DST 13%, CD154 11%, DST CD154 24% longterm surviving grafts). Control groups received TEa Tg T cells and a syngeneic graft or no TEa Tg T cells and CB6F 1 grafts with or without DST. None of these groups ever rejects its grafts (data not shown). CD154 reduces the in vivo expansion of alloreactive Tg T cells induced by DST or an allogeneic skin graft in vivo The in vivo response profile of the TEa Tg T cells in response to DST, CD154, and/or a skin allograft was measured. Briefly, on day 0, mice received TEa Tg T cells and either (1) DST cells with or without CD154, (2) an allogeneic CB6F 1 skin graft with or without CD154, (3) DST cells and a CB6F 1 skin graft with or without CD154, or (4) a syngeneic C57BL/6 skin graft. CD154 or H-Ig was administered 3 times per week. As shown in Figure 2, at day 9, CB6F 1 skin grafts induced extensive expansion of TEa Tg T cells in draining LNs compared with expansion induced in nondraining lymph nodes. On the other hand, DST induced systemic expansion of TEa Tg T cells with equivalent numbers in all LNs. In all groups, treatment with CD154 blocked TEa Tg T-cell expansion by approximately 50%. CD154 accentuates the alloreactive T-cell unresponsiveness induced by DST The combined administration of CD154 and DST enhanced graft survival more effectively than either agent alone. To address the cellular basis for enhanced survival, the functional activity of TEa Tg T cells following in vivo tolerization was measured. Tg T cells were purified from treated and untreated mice and restimulated in vivo and in vitro. Briefly, CD45.1 TEa Tg T cells were adoptively transferred into RAG / recipients together with DST, CB6F 1 skin, and with or without CD154 treatment. On day 7, mice were killed, LN cells were harvested, and the TEa Tg T cells were purified by positive selection for CD45.1 expression. For in vivo restimulation, equivalent numbers of tolerized (DST CD154 exposed TEa Skin graft rejection and prolonged graft acceptance induced by DST and CD154 blockade in a CD4 TCR transgenic model Initial experiments demonstrated that TEa Tg T cells could mount an effective graft rejection response in vivo. As shown in Figure 1, transfer of CD4 TEa Tg T cells into RAG / C57BL/6 mice alone mediates graft rejection with similar kinetics to those seen in non-tg rejection systems (mean survival time [MST] 18.6 days in the Tg system; MST 10 days in the non-tg system). 12 To evaluate if treatment with CD154 with or without DST could interfere with TEa Tg T-cell mediated graft rejection, recipient RAG / mice were injected with or without CB6F 1 spleen cells and 500 g CD154 or control H-Ig. One day after injection, treated mice received a CB6F 1 skin graft. CD154 or control antibody was subsequently administered 3 times per week. Treatment with DST CD154 significantly delayed graft rejection by the CD4 TEa Tg T cells (MST 72.4 days, P.0001), and the degree to which it delayed the rejection is also similar to that seen Figure 2. CD154 inhibits in vivo expansion of alloreactive Tg CD4 T cells. C57BL/6 RAG KO mice were injected with TEa Tg T cells in the presence or absence of CB6F 1 spleen cells and a CB6F 1 skin graft. Mice were injected with CD154 or control H-Ig 3 times per week. Then, 9 days later, local and nonlocal LNs were harvested and cells were counted and analyzed by fluorescence-activated cell sorter for the percentage of TEa Tg T cells in order to determine the total number of TEa Tg T cells in each group.

4 BLOOD, 1 SEPTEMBER 2003 VOLUME 102, NUMBER 5 MECHANISM OF CD154/DST-INDUCED T-CELL ANERGY 1923 to allogeneic skin were greatly reduced by the coadministration of CD154/DST. In addition, tolerized TEa Tg T cells were incapable of producing the Th2 cytokine IL-4 or to suppress skin rejection by naive TEa Tg T cells (data not shown). Finally, no overt phenotypic differences in cell surface profiles (CD25, CD44, and CD62L) were apparent between the tolerized and immune Tg T cells. In summary, on a per cell basis, DST and, moreover, CD154/DST induced profound T-cell anergy. DST induces an early, abortive T-cell response that preempts the productive response to an allograft Figure 3. CD154/DST treatment reduces the functional activity of TEa Tg T cells in vivo and in vitro. C57BL/6 RAG KO mice were injected with TEa Tg T cells with or without CB6F 1 spleen cells, and in the presence or absence of a CB6F 1 skin graft on day 0. Anti-CD154 or control H-Ig was injected 3 times per week. On day 7, LNs were harvested and TEa Tg T cells were positively sorted. (A) Purified TEa Tg T cells were transferred to naive C57BL/6 RAG KO recipients together with a CB6F 1 skin graft. Graft rejection was monitored as described previously ( represents naive TEa Tg T cells [n 10]; f, DST-exposed TEa Tg T cells [n 9]; F, DST CD154 exposed TEa Tg T cells [n 8]). Data were pooled from 2 separate experiments. Also, purified effector TEa Tg T cells were plated in vitro with irradiated CB6F 1 stimulator cells as described in Materials and methods. Then, 72 hours later, cells and supernatants were harvested for quantification of (B) 3 H-thymidine incorporation, (C) IL-2 production, and (D) IFN- production. Tg T cells [1 10 5, 70% pure]) or control (DST H-Ig exposed TEa Tg T cells) TEa Tg T cells were transferred into naive RAG / recipients. A third group of mice received only naive TEa Tg T cells ( cells). A CB6F 1 skin graft was placed on each mouse, and rejection was followed over time. As shown in Figure 3A, TEa Tg T cells exposed to DST CD154 showed a reduced capacity to reject CB6F 1 skin grafts (MST days; P.0001), compared with naive controls (MST 25.8 days). TEa Tg T cells exposed to DST hamster Ig (H-Ig) also showed a reduced capacity to reject CB6F 1 skin but to a less significant degree (MST days; P.0001). Thus, DST and CD154 treatment induced substantial T-cell unresponsiveness to alloantigen restimulation in vivo. Seeking an explanation to the hyporesponsiveness seen in Figure 3A, we assessed the in vitro functional responsiveness of purified TEa Tg T cells. TEa Tg T cells were purified from treated mice (as described in Materials and methods ), and their proliferative and cytokine production profiles in response to alloantigen were determined. All results presented used purified TEa Tg T cells and represent responses on a per cell basis. As shown in Figure 3B-D, TEa Tg T cells from mice that were grafted with allogeneic skin had markedly higher in vitro proliferative and cytokine production responses (interleukin-2 [IL-2] and interferon [IFN ]) than all other groups. DST, while inducing substantial in vivo expansion (Figure 2), did not appear to prime the TEa Tg T cells to respond upon recall to alloantigen in vitro. Thus, alloantigen provided by allogeneic skin compared with that provided by allogeneic spleen cells produced markedly distinct effects on the recall response of the TEa Tg T cells in vitro. Most interestingly, administration of DST to mice that received skin grafts markedly reduced the recall responses to challenge with alloantigen in vitro. Even more, concomitant administration of CD154/DST led to a more profound inhibition of recall responses. The recall proliferative response to allogeneic skin ( cpm/culture) was reduced to less than 5% (2000 cpm/culture) by the coadministration of CD154/DST. Similarly, cytokine production profiles in response The data thus far indicated that heightened TEa Tg T-cell recall responses were induced by an allograft and that the coadministration of DST prior to an allograft could block allograft-induced T-cell priming. Furthermore, the T-cell unresponsiveness elicited by DST was further accentuated by CD154. One way in which DST could exert a dominant impact over priming by an allograft was if the Tg T cells were anergized by the DST prior to encountering the immunogenic allograft. The tempo of the response of TEa Tg T cells to DST versus an allograft was studied in vivo by evaluating the extent of in vivo Tg T-cell division over time. This was achieved by following the dilution of the intracellular fluorescent dye, CFSE. CFSE-labeled Tg TEa T cells were adoptively transferred into RAG / recipients together with CB6F 1 DST, CB6F 1 skin, or syngeneic skin. At days 4 and 8, cells from draining and nondraining LNs were harvested and CFSE dilution was analyzed by flow cytometry. Figure 4 shows that at day 4, all the TEa Tg T cells in draining or nondraining LNs from the recipient that received DST had divided, with no remaining cells within the high-staining population. The same effect was observed in the presence of DST and CD154, but with a reduction in overall expansion (data not shown). At the same time point, allogeneic skin induced limited division of the Tg cells in the draining LNs and no division in nondraining LNs. This is demonstrated by a small number of total TEa Tg T cells with reduced dye dilution, yet with a sizable high-staining population still present on day 4. However, after 8 days, skin alone was able to strongly stimulate T-cell division in draining LNs with some evident migration of dividing cells to the nondraining LNs. As a control, syngeneic skin did not induce TEa Tg T-cell division in draining or nondraining LNs. Therefore, DST induces an early, systemic expansion of alloreactive T cells leading to T-cell anergy, which may preempt the ensuing productive response to the allograft. Figure 4. The allogeneic response of TEa Tg T cells to DST and skin is temporally and spatially separated. CFSE-labeled Tg TEa T cells were adoptively transferred into RAG / recipients together with CB6F 1 DST, CB6F 1 skin, or syngeneic skin as a negative control for proliferation. At days 4 and 8, cells from draining and nondraining LNs were harvested and CFSE dilution of the TEa Tg T cells was analyzed by flow cytometry. Total number of TEa Tg T cells/lns is shown for each group.

5 1924 QUEZADA et al BLOOD, 1 SEPTEMBER 2003 VOLUME 102, NUMBER 5 Figure 5. Indirect presentation of DST antigens by host APCs mediates hyporesponsiveness of TEa Tg T cells and long-term allograft survival. TEa Tg T cells ( ) were adoptively transferred into C57BL/6 RAG / recipients (n 15 per group) in the presence of CB6F 1 skin alone or together with CB6F 1 DST or T-cell depleted Balb/c DST to assess indirect presentation. A separate group of mice received TEa cells and syngeneic skin as a negative control. All the groups received injections of CD154 or control H-Ig 3 times per week. (A) After 7 days, 4 mice per group were killed and cells were harvest from LNs. The same number of TEa Tg T cells were plated with irradiated CB6F 1 spleen cells to verify their ability to respond to in vitro restimulation by measuring 3 H-thymidine incorporation. The remainder of the mice were followed for transplant rejection kinetics over time in presence of (B) control H-Ig or (C) CD154. DST-induced hyporesponsiveness and increased graft survival is mediated by indirect presentation of alloantigen and CD154 blockade While the prevailing hypothesis proposes that DST-induced tolerance is due to direct presentation of donor alloantigens to host alloreactive T cells, this has not been rigorously tested. We evaluated the potential contribution of direct versus indirect antigen presentation to T-cell hyporesponsiveness. Cells from Balb/c mice (H-2 d ) provide the antigen (I-E ), but not the appropriate MHC-restricting element (I-A b ). Thus Balb/c DST cannot be directly recognized by the TEa Tg T cells. If Balb/c DST results in reduced TEa Tg T-cell response to in vitro restimulation or enhanced graft survival in B6 RAG / recipients, it can be due only to the fact that the Balb/c-derived I-E was presented indirectly by host (H-2 b ) APCs. As shown in Figure 5A, the response to an F 1 skin graft was reduced equivalently by either Balb/c or CB6F 1 DST / CD154. In concordance with this observation, BALB/c, like CB6F 1 DST, was also able to induce enhanced graft survival with CD154 treatment (Figure 5B-C). Thus, in the absence of direct presentation, indirect presentation of alloantigen is capable of inducing Tg TEa systemic expansion (data not shown), unresponsiveness to in vitro restimulation, and enhanced allograft survival. DST is incapable of inducing TEa Tg T cell systemic expansion via direct presentation of alloantigen While the aforementioned data demonstrated that indirect presentation of alloantigen was sufficient for Tg T-cell tolerance induction (Figure 5), it did not address the potential contribution of direct alloantigen presentation by DST. To evaluate the contribution of direct presentation, the responsiveness of TEa Tg T cells to CB6F 1 DST in class II / mice was determined. In contrast to RAG / recipients, the class II / recipients are unable to indirectly present the E peptide provided by the F 1 DST. As shown in Figure 6, while F 1 DST induced vigorous expansion of Tg cells in RAG / mice, in the class II / mice there was no expansion of TEa Tg T cells in response to CB6F 1 DST. These data strongly indicate that the CB6F 1 DST is not directly seen by the TEa Tg T cells but provides allopeptides for host APC presentation. Reduced graft infiltration by TEa Tg T cells by DST and CD154 treatment To critically evaluate the impact of CD154/DST on T-cell infiltration and pathology, skin grafts were histologically evaluated. Skin grafts were removed from the mice on days 4, 7, and 14, and examined by H&E staining for infiltration and inflammation of the grafts. As can be seen in Figure 7, mice receiving a syngeneic graft have no inflammation or infiltration of the graft until day 14, at which point there is very minor spongiosis of the stratum basale of the epidermis and hair follicles, and a very minor lymphocytic infiltrate. This is most likely due to the normal process of wound healing, as these grafts are never rejected. In mice receiving a CB6F 1 graft, on day 4 the grafts appear the same as those seen in mice receiving syngeneic grafts, but by day 7 there is wellestablished interface dermatitis. The interface dermatitis consists of vacuolar changes in the stratum basale of the epidermis, necrotic keratinocytes, and lymphocytic infiltrates apparent in the dermis. By day 14, the grafts on these mice are necrosing. Mice receiving a CB6F 1 graft and DST cells show a similar, although not quite as pronounced, pattern of graft infiltration and necrosis as seen in mice receiving CB6F 1 grafts alone. Mice receiving a CB6F 1 graft together with CD154 treatment show a different pattern, with no inflammation on day 7 with the exception of rare dead keratinocytes (typically one dead cell per section). By day 14, the grafts on these mice are necrosing. Skin graft sections from mice receiving a CB6F 1 graft together with DST/ CD154 treatment demonstrate a few scattered dead keratinocytes on days 4 and 7. On day 14, one can observe the beginnings of an interface reaction, milder than that seen on day 7 of the untreated or DST-treated mice. Thus CD154 treatment delays skin graft infiltration and together with DST treatment can greatly delay infiltration and prolong graft survival. Further characterization of the infiltrating lymphocytes by confocal microscopy showed clear staining of CD4 CD45.1 Figure 6. Direct presentation of alloantigens in class II KO recipients is unable to induce systemic expansion of Tg TEa cells. TEa Tg T cells ( ) were adoptively transferred into C57BL/6 RAG / or class II KO recipients (n 3 per group) in the presence of CB6F 1 skin alone or together with CB6F 1 DST. After 7 days in vivo, lymph nodes were collected and the total number of Tg TEa cells per LN were quantified as previously described.

6 BLOOD, 1 SEPTEMBER 2003 VOLUME 102, NUMBER 5 MECHANISM OF CD154/DST-INDUCED T-CELL ANERGY 1925 Figure 7. Reduced graft infiltration with DST and CD154 treatment. TEa Tg T cells ( ) were adoptively transferred into C57BL/6 RAG / recipients that received transplants of a CB6F 1 skin graft and were treated with or without CB6F 1 DST and/or CD154 or H-Ig. Another group of mice received a syngeneic C57Bl/6 skin transplant as a negative control. Skin grafts were removed on days 4, 7, and 14, formalin fixed, and processed for H&E staining. Original magnification, 100. donor TEa Tg T cells correlating to the levels of lymphocyte infiltration observed with H&E staining (data not shown). Discussion Insights into the behavior of the CD4 alloreactive T-cell compartment in graft rejection and acceptance have been hampered by the lack of the appropriate in vivo systems to visualize T-cell responses. Here we present a novel model using alloreactive CD4 TCR Tg T cells to follow the fate and function of T cells specificto a major donor antigen, and at the same time define the contribution of direct and indirect pathways of antigen presentation to the tolerogenic process. The validity of this model for studying graft rejection and tolerance is shown by the fact that CD4 T cells from a TCR transgenic mouse (TEa) demonstrate the capacity to reject allogeneic (H-2 bxd ) but not syngeneic skin at a tempo consistent with that observed with polyclonal, alloreactive T-cell populations. Furthermore, we show that immune intervention ( CD154/DST) can enhance allograft longevity. Unique to this system is the ability to directly measure, on a per cell basis, the response profiles of alloreactive T cells that have been immunized or tolerized in vivo. Compared with naive TEa Tg T cells, TEa Tg T cells that have been primed to an allograft in vivo have higher proliferative indices and higher levels of cytokine production on a per cell basis. Other studies of in vivo primed TCR Tg T cells have also shown that primed or memory CD4 TCR Tg T cells are hyperresponsive. 14 Using this system, we were able to address the cellular basis of the synergy between CD154 and DST. First, expansion of alloreactive Tg T cells induced by the allograft in the regional nodes is reduced by 60% by CD154. Hence, the total number of allospecific T cells in the host is reduced. Second, while DST alone induces profound anergy, the anergy induced by DST/ CD154 is more pronounced. That is, the proliferative responsiveness of the residual T cells following DST/ CD154 is about 30% of that observed with DST alone (on a per cell basis). Furthermore, when equivalent numbers of DST-treated versus DST/ CD154-treated TEa Tg T cells are transferred into naive mice (Figure 3A), the latter retain grafts for more than 50 days longer. Thus, the intrinsic functional capacity of the DST/ CD154-treated T cells is impaired compared with DST-treated TEa Tg T cells. Remarkable is the fact that the major qualitative difference between tolerized and nontolerized TEa Tg T cells resides in their capacity to proliferate and produce proinflammatory cytokines upon restimulation; meanwhile other parameters, such as their activation phenotype, remain unchanged. DST/ CD154 induces a profound systemic T-cell expansion that is rapid and results in the generation of a population of anergic T cells. The finding that indirect presentation mediates DSTinduced T-cell anergy alters the prevailing paradigm and strongly suggests that DST-induced tolerance may use mechanisms usually operative in inducing cross-tolerance to self-antigens. 15 In models of peripheral T-cell tolerance to self-antigens, it has been shown that specific subsets of host, immature dendritic cells constitutively present self-antigens to autoreactive T cells resulting in T-cell anergy or deletion. 15,19 It is also known that tissue destruction and apoptotic cells are superb sources of self-antigen for delivery to these immature APCs. 7,20-22 We would propose that shortly after infusion, cells of the DST apoptose and efficiently deliver alloantigen to host APCs for the indirect presentation of donor allopeptides. The overwhelming tolerogenic impact of alloantigen delivered by DST can be seen by the anergy induced by DST alone. The synergistic actions of CD154 with DST can be explained by this same model. First, CD154 is likely proapoptotic for the infused leukocytes, as we know that CD154 expressed by the host would lead to activation and longevity of the donor-derived B cells and DCs in the DST. Second, and more important, blocking CD154 blocks the potential activation of those immature DCs that are presenting newly acquired alloantigen from the DST. 6 It has been shown that triggering via CD40 activates DCs and breaks peripheral tolerance, and thus in the context of DST, it is important to block endogenous CD154 function. 23,24 The contrast in scope and tempo of T-cell responsiveness to DST versus the allograft helps to explain the effectiveness of DST-induced tolerance. Studies have shown that the timing of DST administration relative to allografting is an important parameter in long-lived graft survival. Longer periods of time between DST (within limits) and grafting, and multiple administrations of DST prior to grafting can enhance graft survival. 25 All of these parameters likely manifest as more effective measures to preemptively induce anergy in the alloreactive T-cell compartment prior to transferring the immunogenic allograft. The second feature of DST is its systemic impact on alloreactive T-cell responses. As would be predicted, DST administered intravenously systemically anergizes the alloreactive T-cell pool. In contrast, the predominant impact of the allograft is regional, eliciting profound local expansion and T-cell activities. It is becoming increasingly clear that Treg s play a central role in long-lived graft acceptance induced by DST/ CD However, in this Tg system, there is a lack of CD4 CD25 regulatory T cells, allowing us to study the independent impact of DST/ CD154 on the effector T-cell compartment. In the absence of

7 1926 QUEZADA et al BLOOD, 1 SEPTEMBER 2003 VOLUME 102, NUMBER 5 Treg s, it is shown that DST and CD154 induce hyporesponsiveness of the TEa Tg T effector T cells. The hyporesponsive TEa Tg T cells when cotransferred with naive TEa Tg T cells do not suppress graft rejection, establishing that they are not regulatory in nature (data not shown). We have similarly shown this effect of DST and anti-cd154 using purified polyclonal CD4 CD25 T cells transferred into RAG / mice (data not shown). Future studies using this transgenic system will investigate the dynamic interactions between regulatory and effector allogeneic T cells in response to the tolerogenic stimuli provided by DST/ CD154 therapy. Finally, this novel TCR Tg system offers unique insights into the processes governing immunity and tolerance in graft acceptance. While only CD4 responses were studied in this report, the systematic inclusion of CD4 regulatory T cells and CD8 Tg T-cell populations will allow a more expansive understanding of how these individual alloreactive populations influence the development of graft tolerance. The capacity to track each of these populations independently in vivo will provide a more comprehensive appreciation for how complex interactions between distinct alloreactive T-cell populations result in graft rejection or acceptance. Acknowledgments We thank Anne Perry, MD, for interpretation of H&E sections, Evan F. Lind for help with histology, Kathy Bennett for assistance with animal care, and the Englert Cell Analysis Laboratory. References 1. Brennan DC, Mohanakumar T, Flye MW. Donorspecific transfusion and donor bone marrow infusion in renal transplantation tolerance: a review of efficacy and mechanisms. Am J Kidney Dis. 1995;26: Wood ML, Gottschalk R, Monaco AP. Comparison of immune responsiveness in mice after single or multiple donor-specific transfusions. J Immunol. 1984;132: Parker DC, Greiner DL, Phillips NE, et al. Survival of mouse pancreatic islet allografts in recipients treated with allogeneic small lymphocytes and antibody to CD40 ligand. Proc Natl Acad Sci U S A. 1995;92: Buhlmann JE, Foy TM, Aruffo A, et al. In the absence of a CD40 signal, B cells are tolerogenic. Immunity. 1995;2: Niimi M, Roelen DL, Witzke O, van Rood JJ, Claas FH, Wood KJ. The importance of H2 haplotype sharing in the induction of specific unresponsiveness by pretransplant blood transfusions. Transplantation. 2000;69: Miga AJ, Masters SR, Durell BG, et al. Dendritic cell longevity and T cell persistence is controlled by CD154-CD40 interactions. Eur J Immunol. 2001;31: Steinman RM, Turley S, Mellman I, Inaba K. The induction of tolerance by dendritic cells that have captured apoptotic cells. J Exp Med. 2000;191: Sauter B, Albert ML, Francisco L, Larsson M, Somersan S, Bhardwaj N. Consequences of cell death: exposure to necrotic tumor cells, but not primary tissue cells or apoptotic cells, induces the maturation of immunostimulatory dendritic cells. J Exp Med. 2000;191: Huynh ML, Fadok VA, Henson PM. Phosphatidylserine-dependent ingestion of apoptotic cells promotes TGF-beta1 secretion and the resolution of inflammation. J Clin Invest. 2002;109: Grubin CE, Kovats S, deroos P, Rudensky AY. Deficient positive selection of CD4 T cells in mice displaying altered repertoires of MHC class IIbound self-peptides. Immunity. 1997;7: Markees TG, Phillips NE, Gordon EJ, et al. Prolonged skin allograft survival in mice treated with Flt3-ligand- induced dendritic cells and anti- CD154 monoclonal antibody. Transplant Proc. 1999;31: Markees TG, Phillips NE, Noelle RJ, et al. Prolonged survival of mouse skin allografts in recipients treated with donor splenocytes and antibody to CD40 ligand. Transplantation. 1997;64: Howard LM, Miga AJ, Vanderlugt CL, et al. Mechanisms of immunotherapeutic intervention by anti-cd40l (CD154) antibody in an animal model of multiple sclerosis. J Clin Invest. 1999; 103: Ahmadzadeh M, Farber DL. Functional plasticity of an antigen-specific memory CD4 T cell population. Proc Natl Acad Sci U S A. 2002;99: Liu K, Iyoda T, Saternus M, Kimura Y, Inaba K, Steinman RM. Immune tolerance after delivery of dying cells to dendritic cells in situ. J Exp Med. 2002;196: Kurts C, Cannarile M, Klebba I, Brocker T. Dendritic cells are sufficient to cross-present self-antigens to CD8 T cells in vivo. J Immunol. 2001;166: Belz GT, Behrens GM, Smith CM, et al. The CD8alpha( ) dendritic cell is responsible for inducing peripheral self-tolerance to tissue-associated antigens. J Exp Med. 2002;196: Scheinecker C, McHugh R, Shevach EM, Germain RN. Constitutive presentation of a natural tissue autoantigen exclusively by dendritic cells in the draining lymph node. J Exp Med. 2002;196: Hugues S, Mougneau E, Ferlin W, et al. Tolerance to islet antigens and prevention from diabetes induced by limited apoptosis of pancreatic beta cells. Immunity. 2002;16: Inaba K, Turley S, Yamaide F, et al. Efficient presentation of phagocytosed cellular fragments on the major histocompatibility complex class II products of dendritic cells. J Exp Med. 1998;188: Ferguson TA, Herndon J, Elzey B, Griffith TS, Schoenberger S, Green DR. Uptake of apoptotic antigen-coupled cells by lymphoid dendritic cells and cross-priming of CD8( ) T cells produce active immune unresponsiveness. J Immunol. 2002; 168: Hawiger D, Inaba K, Dorsett Y, et al. Dendritic cells induce peripheral T cell unresponsiveness under steady state conditions in vivo. J Exp Med. 2001;194: Roth E, Schwartzkopff J, Pircher H. CD40 ligation in the presence of self-reactive CD8 T cells leads to severe immunopathology. J Immunol. 2002; 168: Ichikawa HT, Williams LP, Segal BM. Activation of APCs through CD40 or Toll-like receptor 9 overcomes tolerance and precipitates autoimmune disease. J Immunol. 2002;169: Markees T, Phillips N, Gordon E, et al. Improved skin allograft tolerance induced by treatment with donor splenocytes and an extended course of anti-cd154 monoclonal antibody. Transplant Proc. 1998;30: Honey K, Cobbold SP, Waldmann H. CD40 ligand blockade induces CD4 T cell tolerance and linked suppression. J Immunol. 1999;163: Graca L, Honey K, Adams E, Cobbold SP, Waldmann H. Cutting edge: anti-cd154 therapeutic antibodies induce infectious transplantation tolerance. J Immunol. 2000;165: Yamada, Sayegh MH. The CD154-CD40 costimulatory pathway in transplantation. Transplantation. 2002;73:S36-S39.