|Home | About | Journals | Submit | Contact Us | Français|
Reperfusion of organ allografts induces a potent inflammatory response that directs rapid memory T cell, neutrophil and macrophage graft infiltration and their activation to express functions mediating graft tissue injury. The role of cardiac allograft IL-1 receptor signaling in this early inflammation and the downstream primary alloimmune response was investigated. When compared to complete MHC-mismatched wild type cardiac allografts, IL-1R−/− allografts had marked decreases in endogenous memory CD8 T cell and neutrophil infiltration and expression of proinflammatory mediators at early times after transplant whereas endogenous memory CD4 T cell and macrophage infiltration was not decreased. IL-1R−/− allograft recipients also had marked decreases in de novo donor-reactive CD8, but not CD4, T cell development to IFN-γ-producing cells. CD8 T cell-mediated rejection of IL-1R−/− cardiac allografts took 3 weeks longer than wild type allografts. Cardiac allografts from reciprocal bone marrow reconstituted IL-1R−/−/wild type chimeric donors indicated that IL-1R signaling on graft non-hematopoietic-derived, but not bone marrow-derived, cells is required for the potent donor-reactive memory and primary CD8 T cell alloimmune responses observed in response to wild type allografts. These studies implicate IL-1R-mediated signals by allograft parenchymal cells in generating the stimuli provoking development and elicitation of optimal alloimmune responses to the grafts.
Acute T cell mediated rejection remains a major problem in clinical transplantation directly mediating or contributing to early and late failure of organs transplanted to treat end-stage organ disease. For heart and renal grafts, 5–9% are lost in the first year and the average graft survival at 5 years remains only about 80% (1–4). The high frequency of recipient T cells expressing receptors that are cross-reactive with donor allogeneic MHC molecules generates two pools of donor-reactive T cells that undermine successful allogeneic organ transplantation (5, 6). One pool originates from the memory CD4 and CD8 T cells that have developed during immune responses to environmentally encountered antigens and express T cell receptors that cross-react to donor allogeneic MHC molecules (7–9). The endogenous memory CD8 T cells are of the effector memory phenotype and utilize CXCR3 to infiltrate allografts within 8–12 hours after reperfusion and are activated to proliferate within the allograft and to functions that increase inflammation and contribute to graft injury at early times post-transplant (10, 11). A second pool naïve donor-reactive T cells are activated within the allograft recipient’s secondary lymphoid organs to clonally expand and differentiate to primary effector T cells producing IFN-γ and expressing cytolytic function following interaction with graft- and host-derived alloantigen presenting cells (12). These de novo primed donor-reactive T cells are detectable in the spleen 6–8 days after transplantation in recipients not receiving immunosuppression and quickly traffic into the allograft where they are activated to mediate graft tissue injury.
The inflammatory environment within the allograft has a direct influence on the strength of these two donor-reactive T cell responses. Reperfusion of organ allografts, as well as other ischemic tissues, induces the generation of reactive oxygen species (ROS), which amplify the production of acute phase cytokines, TNFα, IL-1βand IL-6 (13–16). The acute phase cytokines activate the graft vascular endothelial cells and other graft cells to upregulate expression of adhesion molecules and to produce components of the coagulation system and the chemoattractants that promote the infiltration of neutrophils, macrophages, activated T cells and other leukocytes into the graft. This reperfusion-induced inflammatory environment within the allograft impacts the strength of effector functions expressed by infiltrating endogenous memory CD8 T cells and their ability to mediate sufficient tissue injury to cause graft failure (17). The reperfusion-induced inflammation also stimulates alloantigen-presenting cell emigration from the allograft to the recipient secondary lymphoid tissues where they activate the naïve donor-reactive CD4 and CD8 T cells. However, the impact of specific proinflammatory cytokine receptor signals generated within the allograft following reperfusion on the infiltration and activation of endogenous memory T cells as well as on the de novo priming of donor-reactive CD4 and CD8 T cells by alloantigen presenting cells remains poorly defined.
Systemic antagonism of TNFα at the time of graft reperfusion very effectively attenuates the early inflammatory events in allografts and results in substantial prolongation of vascularized renal and cardiac allograft survival in rodent transplant models (18–21). Although recent studies have implicated IL-1 receptor (IL-1R) signaling on dendritic cell function, including in the generation of CD8 T cell responses to viruses (22–24), the role of graft- or recipient-derived IL-1R signals in alloimmune T cell responses to organ allografts has not been well investigated. We hypothesized that IL-1R signaling on allograft dendritic cells would be required to provoke optimal donor alloantigen-reactive endogenous memory T cell and de novo T cell responses. Therefore, we tested the impact of cardiac allografts with an IL-1 receptor deficiency on the activation of the two pools of donor-reactive T cells during the early and late responses to the allograft. The results indicate that allograft IL-1R deficiency has little direct effect on the donor-reactive CD4 T cell response whereas the donor-reactive endogenous memory and donor-reactive naïve CD8 T cell compartments are severely compromised. Furthermore, the expression of functional IL-1 receptor on graft parenchymal and not bone marrow-derived cells plays a key role in evoking these alloreactive CD8 T cell responses. The results implicate IL-1 receptor signaling on graft parenchymal cells in programming the function of the alloantigen-presenting dendritic cells that generate the donor-reactive CD8 T cell response.
C57/BL6 (H-2b), A/J (H-2a) mice and IL1R−/− mice on the C57BL/6 background (B6.IL1R−/−) were obtained from Jackson Labs (Bar Harbor, ME). Male mice, 8–10 weeks of age, were used throughout these studies and all procedures involving animals were approved by the Institutional Animal Care and Use Committee at the Cleveland Clinic.
Heterotopic, intra-abdominal cardiac transplantation was performed as previously reported (11, 21). Total operative times averaged 35 min and hearts resumed spontaneous contraction immediately upon reperfusion. Graft survival was monitored daily by abdominal palpation of the graft and rejection confirmed by laparotomy. At the time of graft harvest, the circulatory system was drained and the graft was immediately snap-frozen in liquid nitrogen or placed in media for analysis of graft-infiltrating cells.
Reciprocal bone marrow chimeras of wild type C57Bl/6 and B6.IL-1R−/− mice for use as cardiac allograft donors were prepared as previously detailed (25). Briefly, wild type and IL-1R−/− mice to be used as bone marrow recipients were lethally irradiated with 1100 Rad using a Cs137 source and 6 hours later received 20 × 106 bone marrow cells intravenously. Irradiated wild type C57Bl/6 mice received bone marrow flushed from tibiae and femurs of B6.IL-1R−/− donors and vice versa. The reconstituted recipients received antibiotics (0.2 mg/ml sulfamethoxazole and 0.4 mg/ml trimethoprim) in the drinking water from days −1 to 7 for microbial prophylaxis. Sentinel chimeras using CD90.1 and CD90.2 mice as donors and recipients indicated ≥ 85% donor bone marrow reconstitution in irradiated recipients. Approximately 6 weeks after the bone marrow transplant, hearts from the chimeras were transplanted to groups of A/J mice.
Antibodies used for in vivo depletion of CD8 or CD4 T cells were purchased from BioXCell (West Lebanon, NH). CD8 T cells were depleted by injecting mice with 100 µg of each anti-CD8 mAb, YTS 169 and TIB-150, i.p. on three consecutive days before transplantation. CD4 T cells were depleted on days 4, 5, 6, and 8 after transplantation using 100 µg of each anti-CD4 mAb, GK1.5 and YTS 191. In each experiment utilizing mAb depletion of T cells, treated sentinel mice were used to evaluate the efficiency of cell depletion by antibody staining and flow cytometry analysis of spleen and lymph node cells (LNC) and was always >95% for CD4 and CD8 T cells when compared to cells from control, rat IgG treated mice. For costimulation blockade, 250 µg CTLA4Ig (BioXCell) was given i.p. on days 0 and +1 and 200 µg anti-LFA-1 mAb (clone FD441.8 from BioXCell) was given i.p. on days −1 and 0. Agonist anti-CD40 mAb FGK4.5 (BioXCell) was administered on days 3 and 4 post-transplant, 100 µg i.p. as previously reported (26). Control rat IgG was purchased from Sigma-Aldrich (St. Louis, MO).
Flow cytometry detection of graft-infiltrating cells was performed using a modification of the approach developed by Afanasyev and colleagues (11, 17, 27). Briefly, harvested tissues were weighed prior to 1 h incubation at 37°C in RPMI with Type II collagenase (Sigma-Aldrich). The tissue was then pressed through a 40 µm filter and the collected cells were washed twice in RPMI, counted using a hemocytometer, and stained for phenotypic markers (CD45, CD4, CD8, Gr1, or F4/80) using commercially available antibodies (BD Bioscience, San Jose, CA; eBioscience, San Diego, CA). Flow cytometry was performed using a FACSCalibur (BD Biosciences) cytometer and FlowJo analysis software (Tree Star Inc., Ashland OR). The side scatter and FL1 (CD45+) channels were used to gate the leukocytes in the graft tissue followed by analysis of the specific leukocyte populations and 200,000 events were accumulated for each sample. Total numbers of each leukocyte population were calculated by: (the total number of leukocytes counted) × (% of the leukocyte population counted in the CD45+ cells)/100. The data are reported as number of each leukocyte population/mg graft tissue ± SEM for 4–5 grafts per group.
Snap-frozen grafts were crushed, homogenized, and RNA was isolated using RNeasy Fibrous Tissue Kits (Qiagen, Valencia, CA). Reverse transcription and real-time PCR were performed using commercially available reagents, primers, and a 7500 Fast Real-Time Thermocycler, all from Applied Biosystems (Foster City, CA).
ELISPOT assays to enumerate donor-reactive T cells in the spleens of cardiac graft recipients were performed as previously detailed (11, 21). Briefly, splenic responder cells and mitomycin C-treated self, donor and third-party stimulator cell populations were cocultured for 24 h at 37°C in serum-free HL-1 media in 96 well plates coated with anti-IFN-γ or anti-IL-2 capture mAb (BD Biosciences). To compare alloreactive CD4 and CD8 T cell priming, each cell population was purified from spleen cell suspensions using magnetic bead chromatography (Stem Cell Technologies, Vancouver, Canada) and then aliquots of the purified responder cells were stimulated with T cell-depleted splenocytes. After development with the chromagen, the total number of spots per well was quantified using an ImmunoSpot Series 4 Analyzer (Cellular Technology Ltd., Shaker Heights, OH).
One-way mixed lymphocyte reactions (MLRs) were performed testing isolated splenic CD4 and CD8 T cells labeled with carboxyfluorescein succinimidyl ester (CFSE) (Invitrogen, Carlsbad, CA), 0.5 µM per 107 cells at 37 °C for 10 minutes. For stimulator cells, CD11c+ cells were purified from spleen cell suspensions using magnetic bead chromatography (Stem Cell Technologies, Seattle, WA) and were treated with mitomycin C. CD4 or CD8 responder T cells (1 – 4 × 105) were co-cultured with 7 × 105 allogeneic stimulator cells for 96 h in 200 µL of complete RPMI 1640 medium supplemented with 10% fetal calf serum in 96-well flat-bottomed plates (Becton Dickinson Labware, Franklin Lakes, NJ). After 72–96 hours, cells were harvested, stained with CD4- or CD8-specific antibodies, and the dilution of CFSE was assessed by flow cytometry.
Following cardiac allograft excision from recipients, a midventricular portion was prepared and fixed with acetic methanol and 6 µm sections were cut and stained with hematoxylin and eosin (H&E). Images were captured and analyzed with Image-Pro Plus (Media Cybernetics, Silver Springs, MD).
Data analysis was performed using GraphPad Prism Pro (GraphPad Software Inc, San Diego CA). Graft survival between experimental groups was compared using Kaplan-Meier survival curves and Log-rank statistics. The Mann–Whitney nonparametric test was used to analyze differences between experimental groups. p < 0.05 was considered a significant difference between groups. Error bars reflect SEM for each group of 4–5 individual mice, sample grafts, or test T cell populations.
To determine the role of allograft expressed IL-1 receptor signaling on early acute phase cytokine expression and recipient leukocyte infiltration into the allograft, groups of A/J mice received complete MHC-mismatched cardiac allografts from wild type C57BL/6 or B6.IL-1R−/− donors. The allografts were harvested 48 hours after reperfusion and the numbers of graft infiltrating leukocytes were compared. The absence of graft IL-1 receptor signaling did not affect the numbers of memory CD4 T cells infiltrating the allografts but the infiltration of neutrophils and memory CD8 T cells was significantly reduced in the B6.IL-1R−/− allografts (Figure 1A). In addition there was a modest increase in the infiltration of macrophages into IL-1R−/− allografts vs. wild type allografts. As previously observed (10, 11, 17), infiltration of memory CD4 and CD8 T cells into isografts at 48 hrs post-transplant was near the numbers of T cells observed in the native (non-transplanted) heart (data not shown). In order to determine if the CD4 T cells infiltrating IL-1R−/− cardiac allografts might express a suppressive activity inhibiting endogenous memory CD8 T cell accumulation in the allografts, A/J mice were depleted of CD4 T cells with mAb prior to receiving IL-1R−/− allografts. Allograft recipient CD4 T cell depletion did not increase memory CD8 T cell numbers in the IL-1R−/− heart allografts (Figure 1).
Gene expression of key inflammatory mediators was compared in wild type and IL-1R–deficient allografts. At 48 hrs post-transplant expression of genes associated with classically activated macrophages (TNFα and iNOS) was significantly increased in wild type C57Bl/6 allografts vs. isografts and was significantly decreased in IL-1R−/− allografts (Figure 2A). In contrast, expression of genes associated with wound healing/regulatory macrophages (NOX2 and IRF5) was reversed, with levels significantly higher in IL-1R−/− vs. wild type allografts. Consistent with the decreased infiltration of neutrophils and CD8 T cells into the IL-1R−/− allografts, expression levels of neutrophil chemoattractants (CXCL1 and CXCL2), chemokines directing memory CD8 T cells into the allograft (CXCL9 and CXCL10), and effector mediators expressed by activated memory CD8 T cell (granzyme B, perforin and IFN-γ) in IL-1R−/− allografts were at the low-absent levels observed in isografts, significantly lower than the levels expressed in the wild type cardiac allografts (Figure 2B). In addition, expression of the adhesion molecules ICAM-1 and VCAM-1 was significantly higher in wild type C57BL/6 allografts when compared to the lower and equivalent expression observed in A/J isografts and B6.IL-1R−/− allografts. Donor class I (H-2Kb) and class II (H-2I–Abα) MHC were expressed at similar levels in B6.IL-1R−/− and wild type C57BL/6 allografts (data not shown).
The absence of IL-1 receptor signaling in complete MHC-mismatched allografts resulted in a modest but significant prolongation of survival when compared to wild type allografts, with median survival times of 12.0 vs. 7.5 days, respectively (Figure 3A). The prolonged survival of IL-1R−/− allografts correlated with decreased leukocyte infiltration and myocyte injury when examined on day 7 post-transplant, the time of wild type allograft rejection (Figure 3B).
The prolongation in IL-1R−/− allograft survival was accompanied by marked decreases in CD8 T cell and neutrophil infiltration into allografts when examined at day 7 post-transplant, the time when most wild type allografts were rejected in the A/J recipients (Figure 3C). However, CD4 T cell infiltration into IL-1R−/− allografts was not decreased and the modest increased macrophage infiltration into IL-1R−/− allografts on day 2 post-transplant was also observed on day 7 post-transplant when compared to wild type allografts on day 7 post-transplant. By day 11 post-transplant, the numbers of CD8 T cells and neutrophils infiltrating IL-1R−/− allografts increased to the levels observed in wild type allografts at day 7 post-transplant whereas the numbers of macrophages infiltrating into the IL-1R−/− allografts remained higher than the level of infiltration into wild type allografts on day 7 post-transplant (data not shown).
To investigate potential mechanisms underlying the decreased CD8 T cell infiltration into IL-1 receptor-deficient allografts at day 7, donor-reactive T cell priming in recipients of wild type vs. IL-1R−/− allografts was compared. Separated CD4 and CD8 T cell populations from individual mice in each of the two recipients groups were tested for numbers of donor-reactive T cells producing IFN-γ by ELISPOT (Figure 4). Whereas high numbers of donor-reactive CD8 T cells producing IFN-γ were observed in the spleens of recipients of wild type allografts on day 7 post-transplant, these cells were markedly and significantly lower at that time point in the recipients of IL-1R−/− allografts. In contrast, priming of donor-reactive CD4 T cells was similar in the spleens of recipients of wild type and IL-1R−/− allografts when assessed on day 7 post-transplant. When compared at the time of allograft rejection, donor-reactive CD4 and CD8 T cell numbers were similar in the spleens of wild type and IL-1R−/− allograft recipients.
The delayed priming of donor-reactive CD8 T cells in recipients of IL-1R−/− allografts raised the possibility of an antigen-presenting cell defect. To investigate this further, dendritic cells were purified from the spleens of wild type C57BL/6 and B6.IL-1R−/− mice, or from A/J mice as a negative control, and were tested for the ability to stimulate the proliferation of purified CD4 and CD8 T cells from A/J mice in MLR cultures. The wild type and IL-1 receptor deficient dendritic cells were equally potent in stimulating the proliferation of A/J CD8 T cells (Figure 5). However, IL-1R−/− dendritic cells were clearly more potent stimulators of alloreactive CD4 T cell proliferation.
The ability of the individual CD4 and CD8 T cell populations in A/J mice to reject wild type C57BL/6 and B6.IL-1R−/− allografts was compared. First, groups of A/J mice were depleted of CD8 T cells prior to transplantation (Figure 6). In contrast to the delayed rejection of the IL-1R−/− allografts observed in A/J mice, CD4 T cell mediated rejection of wild type and IL-1 receptor-deficient allografts occurred at similar times (median time survival, day 12.5 vs. 14 post-transplant). Groups of A/J mice were also depleted of CD4 T cells by administering depleting antibody beginning on day 4 post-transplant to allow sufficient CD4 T cell mediated help to occur for the de novo priming of alloreactive CD8 T cells. CD8 T cell mediated rejection of the IL-1R−/− allografts was markedly delayed in the absence of CD4 T cells when compared to rejection of the wild type allografts (median time of survival, day 35 vs. 18 post-transplant).
Induction of the primary donor-reactive CD8 T cell response in cardiac allograft recipients is dependent on the delivery of CD4 T cell mediated help (28, 29). The defective CD8 T cell priming in response to IL-1R−/− allografts occurred in the presence of CD4 T cells, but the possibility remained that the CD4 T cells were not delivering sufficient help to induce optimal donor-reactive effector CD8 T cell responses. Although priming of donor-reactive CD4 T cells to IFN-γ producing cells was equivalent when assessed on day 7 post-transplant, priming of donor-reactive CD4 T cells to IL-2 producing cells was significantly decreased in response to the IL-1R–deficient allograft (Figure 7A). To extend investigation of potential defects in CD4 help to induce optimal CD8 T cells responses to IL-1R−/− cardiac allografts, the ability of agonist CD40 antibody to provide this help for donor-reactive CD8 T cells during the response to IL-1R−/− allografts was tested. Groups of A/J mice were given wild type C57BL/6 or B6.IL-1R−/− heart allografts and groups of the A/J recipients were injected with control IgG or agonist anti-CD40 mAb on days 3 and 4 post-transplant to provide helper signals and the priming of donor-reactive CD4 and CD8 T cells was assessed on day 7 post-transplant. Donor-reactive CD4 T cells producing IL-2 or IFN-γ in response to both wild type and IL-1R−/− allografts were equally increased by the agonist anti-CD40 mAb (Figure 7A). Administration of the anti-CD40 mAb significantly enhanced donor-reactive CD8 T cell development to IFN-γ producing cells in response to wild type and IL-1R−/− allografts. Whereas the anti-CD40 antibody accelerated rejection of wild type allografts by only a day, this treatment significantly accelerated rejection of the IL-1R−/− allografts from day 12 to day 9 (Figure 7B).
The early post-transplant inflammation and delayed primary donor-reactive CD8 T cell response to IL-1R–deficient allografts suggested that strategies employed to increase cardiac allograft survival might have greater efficacy for prolonging IL-1R−/− allograft outcomes. Groups of A/J mice received either wild type C567BL/6 or B6.IL-1R−/− cardiac allografts and were treated with anti-LFA-1 mAb on days −1 and 0 or with CTLA-4Ig on days 0 and +1 (Figure 8). The efficacy of CTLA-4Ig was significantly increased in prolonging the survival of the IL-1R−/− vs. the wild type cardiac allografts (median allograft survival, day 65 vs. 44.5 post-transplant). Similarly, peri-transplant treatment with anti-LFA-1 mAb significantly prolonged the survival of the IL-1R−/− allografts when compared to wild-type allografts (median allografts survival, day 63.5 vs. 44 post-transplant).
The impact of functional IL-1 receptor expression on bone marrow-derived vs. parenchymal cells in the allograft on the early inflammatory events and on the priming of donor-reactive CD4 and CD8 T cells was investigated by generating reciprocal bone marrow chimeras of IL-1R−/− and wild type C57BL/6 mice for use as cardiac allograft donors.
First the allografts were harvested from recipients 48 hours after transplant and graft infiltrating leukocyte populations were evaluated. When compared to allografts from wild type bone marrow to irradiated wild type chimeras (WT → WT), allografts from WT → IL-1R−/− chimeras had decreased infiltration with CD4 and CD8 T cells and macrophages but neutrophil infiltration was slightly and insignificantly lower than observed in control, wild-type chimeric allografts (Figure 9A). In contrast, allografts from IL-1R−/− bone marrow → wild type chimeras had marked increases in memory CD4 T cell and macrophage infiltration and equivalent memory CD8 T cell infiltration when compared to allografts from control, wild type chimeras.
Then the priming of donor-reactive CD4 and CD8 effector T cells to IFN-γ producing cells in response to the chimeric allografts was tested (Figure 9B). Allografts from control chimeras (wild type C57BL/6 bone marrow → irradiated wild type C57BL/6 hosts) induced donor-reactive CD4 T cells producing IFN-γ and this response was modestly but significantly decreased in response to allografts where IL-1R expression was restricted to hematopoietic-derived (wild type C57BL/6 bone marrow → irradiated B6.IL-1R−/− hosts) or to non-hematopoietic-derived cells (B6.IL-1R−/− bone marrow → irradiated wild type C57BL/6 hosts). The number of donor-reactive CD8 T cells producing IFN-γ induced in response to the control chimeric heart allografts was decreased in recipients of heart allografts from the wild type C57BL/6 bone marrow → irradiated B6.IL-1R−/− chimeric donors. However, the donor-reactive CD8 T cell response was increased more than 2-fold in recipients of the B6.IL-1R−/− bone marrow → irradiated wild type C57BL/6 chimeric allograft donors when compared to the wild type control allografts.
Finally, the CD8 T cell mediated rejection of the chimeric allografts was assessed in A/J recipients of the two different chimeric allografts treated with anti-CD4 mAb beginning on day 4 post-transplant (Figure 9C). CD8 T cell mediated rejection of the chimeric wild type C57BL/6 bone marrow → irradiated B6.IL-1R−/− host donor allografts was markedly delayed in the absence of CD4 T cells when compared to rejection of the chimeric B6.IL-1R−/− bone marrow → irradiated wild type C57BL/6 host donor allografts (median time of survival, day 29 vs. 15 post-transplant).
The production of a triad of acute phase cytokines, IL-1β, IL-6 and TNFα, is rapidly induced by reactive oxygen species generated during the influx of oxygen when ischemic organs are revascularized (15, 30). We had previously reported rapid and high levels of IL-1β, IL-6 and TNFα following reperfusion of cardiac iso- and allo-grafts followed by the infiltration of neutrophils and donor-reactive memory CD4 and CD8 T cells into the grafts within 12–24 hours of reperfusion (11, 17, 21). Consistent with the key roles of these cytokines promoting inflammation, neutralization of each of the acute phase cytokines has a marked effect in attenuating ischemic tissue injury following reperfusion (16, 31–34). The roles of IL-6 and TNFα in the development of acute and chronic allograft injury have been well established in studies where their absence or neutralization not only decreases early post-reperfusion inflammation within grafts in experimental models, but also attenuates de novo donor-reactive T cell responses and prolongs allograft survival (21, 35–37). Systemic blocking of IL-1 signaling with soluble IL-1 receptor antagonist results in a modest decrease in donor-reactive T cell priming with a slight extension in cardiac allograft survival (38, 39). Whether IL-1 receptor signaling to cells in the allograft vs. cells of the recipient has different consequences on the donor-reactive T cell response and on allograft outcome remains unknown and was the focus of the current studies. The results indicate that the absence of IL-1R signaling on cardiac allograft cells has profound effects on both the early post-transplant inflammation that occurs shortly following reperfusion of the graft and on the down-stream primary/de novo alloreactive T cell response. It is worth noting that transplant of wild type heart allografts to IL-1R−/− recipients did not have a noticeable effect on the early inflammatory events in the graft or on the de novo priming of alloreactive T cell responses (S. Iida, data not shown) indicating the important role of IL-1R signaling by cells in the allograft on the development of the donor-reactive response.
IL-1R-deficient cardiac allografts had a marked attenuation of reperfusion-induced inflammation including decreased infiltration of neutrophils and endogenous memory CD8 T cells when assessed at early times post-transplant, whereas memory CD4 T cell infiltration into the allograft was not affected. These results are consistent with our previous results indicating that it is the activities of the endogenous memory CD8 T cells and not the endogenous memory CD4 T cells that have a major impact on the intensity of this early post-transplant inflammation and allograft outcome (10, 11, 17). These results further suggest that allograft expressed IL-1R induces signals through IL-1β and/or IL-18 that are required to direct the recruitment of endogenous memory CD8 T cells, but not memory CD4 T cells, into the allografts shortly after reperfusion. IL-1 is rapidly produced by endothelial cells (ECs) following reperfusion and induces the vasculature to express the adhesion molecules and produce the chemokines directing the infiltration of T cells and other leukocyte populations into tissue inflammation sites (40–43). The decreased expression of CXCL9/Mig and CXCL10/IP-10 observed in the allografts at day 2 post-transplant is likely to account for the decreased memory CD8 T cell infiltration into the allografts and suggests endogenous memory CD4 T cell infiltration being directed through other mechanisms. It is also worth noting that expression of the neutrophil chemoattractants CXCL1 and CXCL2 was markedly decreased in IL-1R−/− allografts and studies from this and other laboratories have reported that inhibition of neutrophil infiltration and activation within allografts decreases donor-reactive memory and primary effector CD8 T cell infiltration into allografts (44, 45). In addition, expression of ICAM-1 and VCAM-1 was markedly reduced in IL-1R−/− allografts and is likely to contribute to the decreased neutrophil and endogenous memory CD8 T cell graft infiltration early after reperfusion but raises questions about the role of these adhesion molecules in the infiltration of macrophages and memory CD4 T cells into the IL-1R–deficient allografts. Chimeric allografts were used to show that IL-1R-mediated signals on parenchymal/non-bone marrow-derived, rather than on hematopoietic-derived, cells in the allograft are key components promoting this early endogenous memory CD8 T cell infiltration into the allografts.
Macrophage infiltration was increased 2–3 fold into allogeneic, but not syngeneic, IL-1R−/− grafts (data not shown), suggesting that reactivity to graft alloantigens, possibly by the low numbers of memory CD8 T cells or NK cells within the graft, induces the increased macrophage infiltration. Alternatively, recent studies have demonstrated macrophage recognition of alloantigen as a potential contributor to allograft rejection (46), raising the possible direct alloantigen-mediated activation of macrophages as a potential mechanism underlying the accumulation of the high numbers of macrophages within the IL-1R−/− allografts. Similar to our previous studies indicating a skewing of macrophage development to a regulatory/wound healing functional phenotype in the absence of graft CCL2-mediated inflammation (47), the increased macrophage infiltration into IL-1R−/− heart allografts was associated with their development to a regulatory/wound healing functional phenotype including increased expression of NOX2 and IRF5 and decreased iNOS and IL-6 expression. The IL-1R-mediated regulation of macrophage infiltration and activation to an inflammatory phenotype could be mediated directly during the transition from blood monocytes to infiltrating macrophages or indirectly through the activities of the low numbers of endogenous memory CD8 T cells and/or neutrophils within the allograft. We had also considered that the observed effects of allograft IL-1R–deficiency on macrophage numbers and development early after reperfusion might be imposed on cardiac graft resident macrophages. However, the allografts in the current studies were harvested for analysis 48 hours after reperfusion and this time period is likely to be insufficient for the marked increases in macrophage numbers observed in the IL-1R−/− allografts being of graft origin. Nevertheless, these results are important for indicating the role of the allograft IL-1R in separating inflammatory signals that direct recruitment of different leukocyte populations into the allograft and provoking their activities at early times after graft reperfusion.
Similar to the different effects that the absence of allograft IL-1R signaling had on early post-transplant endogenous memory CD8 vs. CD4 T cell infiltration into the allografts, allograft IL-1R deficiency compromised the de novo priming of donor-reactive CD8, but not CD4, T cells to IFN-γ producing cells from naïve precursors in the recipient spleen. Several studies have reported the direct effects of IL-1 on antigen-reactive CD4 and CD8 T cells in enhancing the magnitude and differentiation of T cell responses to protein antigens as well as to allogeneic cells using in vivo and/or in vitro approaches (43, 48–50). More relevant to the current studies, IL-1R signaling on antigen-presenting dendritic cells has been shown to enhance dendritic cell function in activating viral antigen-reactive CD8 T cells (23), suggesting the absence of IL-1R on graft-derived dendritic cells as a potential mechanism decreasing the donor-reactive CD8 T cell responses to IL-1R−/− allografts. However, cardiac allografts from reciprocal bone marrow reconstituted IL-1R−/−/wild type chimeric donors indicated that IL-1R signaling on graft non-hematopoietic-derived, but not bone marrow-derived, cells was required for the donor-reactive memory and primary CD8 T cell alloimmune responses. Allograft parenchymal cells are activated in response to the inflammatory signals generated during graft reperfusion to express functions regulating leukocyte infiltration into the graft and the activation of antigen presenting cells to emigrate from the allograft to the recipient lymphoid tissue where naïve donor-reactive CD4 and CD8 T cells are primed (12, 30, 51). The current results indicate that signals generated through graft parenchymal cell IL-1R instruct the function of the graft-derived dendritic cells to activate donor-reactive CD4 and CD8 T cells. Moreover, the ability of the dendritic cells from IL-1R−/− allografts to stimulate development of donor-reactive CD4 but not CD8 T cell responses suggests that at least two distinct IL-1R-mediated signals are delivered from the graft parenchymal cells to the graft interstitial dendritic cells.
During primary responses to allografts CD4 T cells deliver help for the generation of donor-reactive antibody and CD8 T cells through CD40-CD154 mediated signals during interaction with alloantigen-presenting cells in the spleen (51–53). Although isolated dendritic cells from the spleens of wild type and IL-1R−/− mice were equally stimulatory for alloreactive CD8 T cell proliferation, the in vivo results suggested a defect in the ability of the IL-1R–deficient allograft derived dendritic cells to induce CD4 T cell delivery of the helper signals required for donor-reactive CD8 T cell responses. CD8 T cell-mediated rejection of IL-1R−/− cardiac allografts was markedly delayed when compared to rejection of wild type allografts, but provision of agonist anti-CD40 mAb corrected the decreased de novo CD8 T cell priming and the ability of the CD8 T cells to reject the IL-1R−/− allografts. A potential defect in CD4 T cell mediated help for CD8 T cell responses to IL-1R−/− allografts is further suggested by the decreased donor-reactive CD4 T cell development to IL-2 producing cells. Collectively, the decreased primary donor-reactive CD8, but not CD4, T cell responses evoked in the spleen in response to cardiac allografts with IL-1R signaling defects restricted to graft parenchymal cells indicate that IL-1R-mediated signals from the graft parenchymal cells differentially influence the function of the graft antigen-presenting cells that emigrate from the graft to the recipient spleen for activation of donor-reactive CD4 vs. CD8 T cells. Identification of these IL-1R–induced parenchymal derived signals should be important in developing strategies to optimize or attenuate antigen-specific CD8 T cell responses as warranted for patient benefit.
IL-1R antagonists have been used effectively to attenuate ongoing autoinflammatory and autoimmune disease in the clinic (54–57). Although blockade of IL-1 signaling with soluble IL-1 receptor attenuates T cell responses to allografts and modestly improves allograft outcome in experimental models (38), IL-1 neutralizing strategies have not been incorporated into immunosuppression regimens used in transplant patients. The absence of IL-1R signaling on graft parenchymal cells attenuates many key early and late components of the alloimmune response that lead to acute graft tissue injury and promote graft failure. Furthermore, the efficacy of costimulatory blockade in prolonging cardiac allograft survival was increased in recipients of IL-1R−/− allografts suggesting an effective synergy of targeting IL-1R plus costimulatory blockade in attenuating early inflammatory components and donor-reactive T cell responses. Overall, these results indicate a key role for IL-1R signaling in provoking the early and downstream alloimmune T cell responses to cardiac allografts and suggest that incorporation of IL-1R antagonism into immunosuppression regimens, particularly at the time of transplant, is likely to have a beneficial effect in attenuating the incidence of delayed graft function and in promoting better graft outcomes.
1This work was supported by National Institutes of Health grants RO1 AI40459 to R.L. Fairchild and PO1 AI087506 to R.L. Fairchild, A. Valujskikh and W.M. Baldwin
The authors have no conflict of interest to declare.