CD40/CD154 Blockade Inhibits Dendritic Cell Expression of Inflammatory Cytokines but not Costimulatory Molecules

Abstract

Blockade of the CD40/CD154 pathway remains one of the most effective means of promoting graft survival following transplantation. However, the effects of CD40/CD154 antagonsim on dendritic cell (DC) phenotype and functionality following transplantation remain incompletely understood. To dissect the effects of CD154/CD40 blockade on DC activation in vivo, we generated hematopoietic chimeras in mice that expressed a surrogate minor antigen (OVA). Adoptive transfer of OVA-specific CD4⁺ and CD8⁺ T cells led to chimerism rejection, which was inhibited by treatment with CD154 blockade. Surprisingly, CD154 antagonism did not alter the expression of MHC and costimulatory molecules on CD11c⁺ DC compared to untreated controls. However, DCs isolated from anti-CD154 treated animals exhibited a significant reduction in inflammatory cytokine secretion. Combined blockade of inflammatory cytokines IL-6 and IL-12p40 attenuated the expansion of antigen-specific CD4⁺ and CD8⁺ T cells and transiently inhibited the rejection of OVA-expressing cells. These results suggest that a major effect of CD154 antagonism in vivo is an impairment in the provision of signal three during donor-reactive T cell programming, as opposed to an impact on the provision of signal two. We conclude that therapies designed to target inflammatory cytokines during donor-reactive T cell activation may be beneficial in attenuating these responses and prolonging graft survival.

Introduction

In both bone marrow and solid organ transplantation models, specifically targeting graft-reactive T cell responses to prevent transplant rejection remains an important goal. Many studies in murine and non-human primate models have shown that blockade of costimulatory signals promotes survival of bone marrow, skin, kidney, heart, and islet transplants (1–5). Blockade of the CD40/CD154 costimulatory pathway remains one of the most effective means of inhibiting alloreactive T cell responses and inducing long-term graft survival following transplantation. However, monoclonal antibodies designed to target CD154 resulted in thromboembolic events in early pilot studies in humans (6). Renewed interest in blockade of this pathway for the prevention of graft rejection has been sparked by promising results from several recent studies in non-human primate transplant models using monoclonals directed against CD40 (7–9), and clinical trials using CD40 blockers in renal transplant recipients are now underway (10). Thus, the therapeutic potential of targeting this pathway is high, and understanding the effects of CD154/CD40 blockade in transplant models may uncover other novel downstream targets for therapeutic intervention.

Despite the incontrovertible efficacy of blockade of this pathway, the mechanisms underlying its effect are incompletely understood. While one report indicated that anti-CD154 monoclonal antibodies may impact the outcome of graft rejection by specifically binding to and depleting antigen-specific CD4⁺ T cells which express CD154 following activation (11), subsequent studies using anti-CD40 monoclonal antibodies showed similar efficacy in both bone marrow and solid organ transplant models in mouse and non-human primates (7–9, 12–14). Thus, it is likely that blockade of the CD40/CD154 pathway, rather than antibody-mediated depletion of antigen-specific cells, plays a major role in the observed attenuation of graft rejection. In dendritic cells (DCs), ligation of CD40 by CD154 expressed on activated CD4⁺ T cells leads to the activation of downstream signaling pathways, resulting in several key events that promote the generation of effective T cell responses. These include 1) increasing MHC expression which would enhance the strength of TCR signals (15, 16), 2) inducing costimulatory molecule expression (e.g., CD80, CD86, OX40L) thus enhancing the strength of “second signals” (17, 18), 3) increased production of pro-inflammatory cytokines (IL-12, IL-6, IL-1) sometimes referred to as “signal three” (19), and 4) increasing DC longevity (20, 21), thereby enhancing T cell priming via of all of the above mechanisms. However, the impact of CD154/CD40 blockade on these aspects of DC biology remains incompletely characterized.

Here, we hypothesized that blockade of the CD40/CD154 pathway in vivo would result in altered DC phenotype and/or function, which could in turn result in sub-optimal T cell priming and thus lead to protection of hematopoietic chimerism following bone marrow transplantation. To test this hypothesis, we generated hematopoietic chimeras that expressed membrane-bound chicken ovalbumin (OVA) on all hematopoietically-derived cells. Subsequent adoptive transfer of OVA-specific CD8⁺ OT-I and CD4⁺ OT-II T cells led to a GVHD- like response as measured by a loss of OVA-expressing cells, a process that was attenuated by treatment with anti-CD154 mAb (MR-1). In order to dissect the effects of CD154/CD40 blockade on DC activation in vivo, antigen-bearing splenic DCs were isolated from recipients that had received host-reactive CD4⁺ and CD8⁺ T cells in the presence or absence of anti-CD154 mAb. Results demonstrated that DCs derived from anti-CD154-treated recipients did not differ with regard to their expression levels of MHC or costimulatory molecules, but instead exhibited impaired secretion of inflammatory cytokines.

Materials and Methods

Mice

BALB/c (CD45.2, H2-K^d), C57BL/6 (CD45.2, H2-K^b) and B6-Ly5.2/Cr (CD45.1, H2-K^b) mice were obtained from the National Cancer Institute (Charles River, Frederick, MD). OT-I and OT-II TCR transgenic mice (C57BL/6 background) were bred to C57BL/6 Thy1.1 congenic animals at Emory University. OT-II×RAG^−/− TCR transgenic mice were purchased from Taconic and bred to Thy1.1 congenic animals at Emory University. mOVA mice on a C57BL/6 background (22) were a gift from Dr. Marc Jenkins (University of Minnesota, Minneapolis, MN) and were maintained at Emory University. All animals were housed in pathogen-free animal facilities at Emory University. All studies were approved by the Institutional Animal Care and Use Committee of Emory University.

Bone marrow isolation and establishment and screening of mOVA BM chimera

Recipient B6-Ly5.2/CR (CD45.1) (NCI) mice were treated one day prior to bone marrow adoptive transfer with 500µg of Busulfan (Busulfex, Otsuka America Pharmaceutical, Inc.) intraperitoneally. Bone marrow was flushed from femurs and tibias of mOVA (CD45.2⁺) mice with saline using a 27g needle and was disrupted through the needle. BM cells were subsequently resuspended in saline and adoptively transferred i.v. at a dose of 20×10⁶ cells per mouse. On the day of bone marrow infusion, recipients were treated with 500µg of both CTLA-4 Ig (Bristol-Myers Squibb) and anti-CD154 (MR1, BioExpress), followed by the same dose on days 2, 4, 6, to prevent an immune response against the OVA antigen. mOVA bone marrow chimeras were used in experiments at approximately 8–12 weeks post-transplant. Hematopoietic chimerism was determined by staining peripheral blood with B220-PerCP, CD45.1-PE, and CD45.2-FITC (all from BD Pharmingen). Data were acquired on a LSR II flow cytometer (Becton Dickinson) and analyzed using FlowJo Software (Treestar, San Carlos, CA).

Adoptive transfers and antibody treatment

Spleens of OT-I and OT-II mice were disrupted with frosted glass slides and processed into a single cell suspension. Splenocytes were stained with CD4-APC, CD8-FITC, Vα2-PE and Thy1.1-PerCP monoclonal antibodies (BD Pharmingen) for flow cytometric analysis. Using TruCount beads (BD Pharmingen), absolute numbers of each cell population were obtained and 5×10⁶ OT-I and 10⁶ OT-II T cells were adoptively transferred i.v. into recipient mice. Following adoptive transfer, animals were treated with 250µg of anti-CD154 (MR1, BioXCell, West Lebanon, NJ) or a combination of anti-IL-12/23p40 (clone C17.8, BioXCell) and anti-IL-6R (clone 15A7, BioXCell) on days 0, 2, 4, and 6, in the indicated groups, or left untreated in control groups.

Dendritic cells isolation and flow cytometric analysis

Spleens were removed from mice. One millilitre of 2mg/ml collagenase type 3 (Worthington Bio. Corp, NJ) in HBSS (with Ca⁺²/Mg⁺²) was injected into the spleen, which was then incubated with 2ml collagenase solution at 37°C /5%CO₂ for 30 min. Single cell suspensions were prepared after incubation by mashing the spleen using 3ml syringe plunger on a cell strainer (70µM) and washing cells with PBS. Single cell suspensions were stained for flow cytometric analysis with anti-CD11c-APC, anti-CD11b-PerCP, anti-CD8α-APC-Cy7, anti-H-2K^b-FITC, anti-I-A^b-FITC, anti-ICAM-1-APC, anti-CD80-PE, anti-CD86-PE, and anti-CD40-FITC (all from BD Pharmingen). Data were acquired on a LSR II flow cytometer (Becton Dickinson) and analyzed using FlowJo Software (Treestar, San Carlos, CA).

Assessment of ex vivo cytokine production by DCs

DCs were isolated as described above and single cell suspensions were enriched by negative selection using magnetic beads coated with anti-CD19, CD90.1, and CD90.2 monoclonal antibodies (Miltenyi Biotec Inc.) according to the manufacturer's instructions. Purified DCs (purity >70%) were resuspended in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum, 50µM 2-mercaptoethanol, streptomycin (100µg/ml), and penicillin (100 units/ml). DCs were then distributed in 200-µl aliquots (6×10⁵ cells/well) to a 96-well plate and cultured for 24 hours at 37°C in 5% CO₂ in duplicate. The cultured cell supernatants were measured for the levels of inflammatory cytokines by a cytometric bead array (CBA) (BD Pharmingen) according to the manufacturer's instructions.

T cell intracellular cytokine staining

To measure cytokine production by antigen specific T cells, surface and intracellular stains were performed with monoclonal antibodies to CD8-Pacific Orange (Invitrogen), CD4-Pacific Blue (Invitrogen), Thy1.1-PerCP (BD Pharmingen), TNF-PE-Cy (BD Pharmingen), and IFN-γ-FITC (BD Pharmingen). Spleens of chimeric mice were processed into single cells suspensions and plated onto 96-well flat-bottom plates at 10⁶ cells per well. Cells were stimulated with 10nM OVA_257–264 (Genscript, Inc.) and 10µM OVA_323–339 (Genscript, Inc.) in the presence of 10µg/mL of Brefeldin A for 5 hours. Cells were processed with an intracellular staining kit (BD Biosciences) according to manufacturer’s instructions. As a positive control, cells were stimulated with 10ng/mL phorbol 12-myristate 13-acetate (PMA) and 1µg/mL ionomycin (Sigma). Data were acquired on a LSR II flow cytometer (Becton Dickinson) and analyzed using FlowJo Software (Treestar, San Carlos, CA).

DC purification, RNA isolation and real time PCR analysis

For DC isolation and purification, single cell suspensions were first enriched by negative selection using magnetic beads coated with anti-CD19, CD90.1, and CD90.2 monoclonal antibodies as described above (yielding ~70% purity of CD11c⁺ cells). Following enrichment, CD11c⁺ Thy 1.1⁻ cells were further purified by FACS sorting on a BD FACS Aria. Following FACS-sorting, DC populations were >90% CD11c⁺ cells.

RNAs from the sorted DCs were isolated using RNeasy Isolation kit (Qiagen). Reverse transcription of the RNA into cDNA was performed using Taqman reverse transcription kit (Roche). Mouse immune array cards (Applied Biosystems) were used to obtain a real-time PCR analysis of the selected immune related genes. Arrays were run on a 7900HT Real-Time PCR System from ABI.

Results

CD154/CD40 pathway blockade prevents antigen-specific T cell-mediated host cell destruction

To evaluate the impact of CD154/CD40 blockade on the induction and/or abortion of donor-reactive T cell responses, we developed a modified GVHD model in which antigen-specific CD4⁺ and CD8⁺ T cells recognized and rejected cognate antigen-bearing hematopoietic cells following bone marrow transplantation. Using an ovalbumin expressing transgenic mouse (mOVA, B6 background), we generated hematopoietic chimeras in which busulfan-treated B6 host mice (CD45.1⁺) received a CD45.2⁺ bone marrow transplant expressing a single defined alloantigen (OVA). Adoptive transfer of 5×10⁶ antigen-specific CD8⁺ (OT-I) and 10⁶ antigen-specific CD4⁺ (OT-II) T cells were sufficient to induce rejection of the CD45.2⁺ mOVA-expressing bone marrow cells such that by day 6, the number of donor-derived (CD45.2⁺) peripheral B cells was significantly reduced (Figure 1A). Both CD4⁺ and CD8⁺ T cells were required to mediate this effect, as adoptive transfer of either OT-I or OT-II alone failed to result in rejection (Supplemental Figure 1 A, B). Importantly, treatment of animals with a short course of CD154 blockade (MR-1) on days 0, 2, 4, and 6 post-transfer resulted in protection of the bone marrow from rejection (Figure 1A, B). These results indicated that blockade of the CD154/CD40 pathway alone was sufficient to prevent T cell-mediated destruction of OVA-expressing peripheral leukocytes and preserve hematopoietic chimerism.

Figure 1. CD40/CD154 pathway blockade protects against T cell-mediated rejection of OVA-expressing host cells.

On Day 0, 5×10⁵ OT-I and 10⁶ OT-II T cells were adoptively transferred into mOVA bone marrow chimeric mice. At the time of transfer, mice were treated with 500µg anti-CD154 mAb (MR-1), with continued treatments on days 2, 4 and 6, in indicated groups. A. Peripheral blood B cell chimerism was measured. Representative flow plots of remaining B220⁺ cells, gating on the remaining mOVA⁺ (CD45.2⁺) cells, over time. B. Total number of CD45.2⁺ B220⁺ mOVA⁺ B cells in the blood over time. Data are representative of three experiments with 4–5 mice per group. Statistics shown are mean±s.e.m. *p≤0.05., ***p≤0.0001.

CD154/CD40 blockade attenuates antigen-specific T cell expansion and effector function

To address the potential mechanisms by which this protection occurred, we analyzed the host-reactive CD4⁺ (OT-II) and CD8⁺ (OT-I) T cell responses in anti-CD154-treated animals. We observed that while host-reactive CD8⁺ T cells expanded dramatically in untreated animals by day 4 post-transfer (Figure 2A–C), treatment with anti-CD154 resulted in a significant diminution in the antigen-specific CD8⁺ T cell response (p<0.0001). A similar result was observed for antigen-specific CD4⁺ T cell responses, which exhibited a more than two-fold reduction in the presence of CD154/CD40 blockade (Figure 2D–F). This was true for both T cell frequencies and absolute numbers (Figure 2A–F).

Figure 2. CD40/CD154 pathway blockade impairs host- specific CD8⁺ T cell accumulation but not entry into cell division.

mOVA chimeric mice were adoptively transferred with 5×10⁶ CFSE-labeled Thy1.1⁺ OT-I and 10⁶ CFSE-labeled Thy1.1⁺ OT-II, and treated with four doses of MR-1 on days 0, 2, 4, and 6, where indicated. A, D. Representative flow plots of antigen-specific Thy1.1⁺ T cell responses in the spleens of treated mice. Data displayed are gated on Thy1.1⁺ CD8⁺ or CD4⁺ T cells on day 4 post-transfer. B, E. Frequency of accumulated antigen-specific CD8⁺ (B) and CD4⁺ (D) T cells at day 4 post-transfer. C, F. Absolute numbers of accumulated antigen-specific CD8⁺ (C) and CD4⁺ (F) T cells at day 4 post-transfer. G, H. CFSE analysis of cell division of antigen-specific CD8⁺ T cells at days 2 and 3 post transfer in both untreated and anti-CD154 treated recipients. Data shown are gated on Thy1.1⁺ CD8⁺ T cells in spleens of mice. Data are representative of 2–3 experiments with a total of 8–12 mice per group. Statistics shown are mean±s.e.m. *p≤0.05.

To address whether reduced frequencies of host-reactive T cells were due to reduced activation and proliferation or increased cell death, host-specific CD8⁺ T cells were CFSE labeled prior to transfer into OVA.B6 chimeras. Results demonstrated similar levels of CFSE dilution in anti-CD154 treated animals as compared to PBS-treated animals (Figure 2E, F), suggesting that the reduced accumulation of host-reactive CD8⁺ T cells was likely due to increased apoptosis, rather than impaired expansion, in the presence of CD40/CD154 pathway blockade.

The functionality of the remaining graft-specific CD8⁺ T cells was further assessed by intracellular cytokine staining following ex vivo restimulation with cognate antigen peptide over time. Results revealed that anti-CD154 treatment delayed the differentiation of antigen-specific CD8⁺ T cells into cytokine-producing cells compared to untreated controls. For example, frequencies of IFN-γ⁺ producing cells at day 4 were significantly reduced, and frequencies of IFN-γ⁺ TNF⁺ dual cytokine producers at day 10 were significantly reduced (Figure 3A, 3B). Absolute numbers of IFN-γ-producing cells were significantly reduced in anti-CD154 treated animals at day 4, but not at subsequent timepoints (Figure 3C). Taken together, these results demonstrate that blockade of the CD40/CD154 pathway attenuates and delays both expansion and differentiation of host-reactive T cells.

Figure 3. CD154/CD40 blockade attenuates antigen-specific T cell expansion and effector function.

mOVA chimeric mice were adoptively transferred with 5×10⁶ OT-I and 10⁶ OT-II, and treated with four doses of MR-1, where indicated. A. Representative flow plots of longitudinal analysis of intracellular TNF and IFN-γ cytokine staining of antigen-specific Thy1.1⁺ CD8⁺ T cells following 4-hour ex vivo peptide stimulation. B. Summary data of frequencies of IFN-γ-producing cells (day 4) and IFN-γ+ TNF+ dual producers (days 10, 17, 24) as a percentage of antigen-specific Thy1.1⁺ CD8⁺ T cells. C. Absolute numbers of total IFN-γ producing OT-I T cells over time following adoptive transfer. Data are representative of 2–3 experiments with a total of 8–12 mice per group. Statistics shown are mean±s.e.m. *p≤0.05.

CD154 blockade inhibits provision of signal three from DC, but not provision of signal one or signal two

The DC licensing model of T cell activation suggests that CD154 expressed on activated CD4⁺ T cells binds to CD40 expressed on DCs and functions to initiate the upregulation of class I and class II MHC and several costimulatory molecules on these DCs, thus enhancing the priming of antigen-specific T cell responses. Interestingly, however, in this GVHD model wherein both CD4⁺ and CD8⁺ antigen-reactive T cells are required to precipitate rejection of antigen-bearing cells (Supplemental Figure 1A, B), we observed that both cell types were required in order to elicit optimum costimulatory molecule expression (Supplemental Figure 1C) and cytokine secretion (Supplemental Figure 1D) from dendritic cells. Because the observed effects of CD40/CD154 blockade on host-reactive T cell responses were likely a direct result of alterations in DC phenotype and functionality, we endeavored to determine which aspects of DC activation were altered in the presence of CD40/CD154 blockade. We first characterized the frequency and phenotype of splenic DCs in mOVA bone marrow chimeras following the adoptive transfer of host-reactive T cells. mOVA bone marrow chimeras contained 1.75±0.77×10⁶ total DC per spleen (Figure 4A), and > 90% of those were CD45.2⁺, expressing the OVA antigen (data not shown). Adoptive transfer of host-reactive CD4⁺ and CD8⁺ T cells did not result in a change in the total number of splenic DCs at day 3 post-transfer, either in the presence or absence of CD154/CD40 blockade (1.59±0.44×10⁶ and 2.04±1.04×10⁶, respectively, Figure 4B). In addition, the relative proportions of CD11c⁺ CD11b⁺ CD8α⁻ vs. CD11c⁺ CD11b⁻ CD8α⁺DC were not altered in the presence of CD154/CD40 blockade (Figure 4C–E). Thus, these data demonstrate that blockade of the CD154/CD40 pathway did not impact the overall quantity or myeloid/lymphoid phenotype of DCs in GHVD recipients.

Figure 4. CD40/CD154 pathway blockade does not impact the frequency and number of CD8α⁺ or CD11^b+ dendritic cells.

mOVA chimeric mice were adoptively transferred with 5×10⁶ OT-I and 10⁶ OT-II, and treated with four doses of MR-1, where indicated. A. Representative flow plot of cells gated on CD11c⁺ dendritic cells, excluding B cells (CD19), T cells (CD3) and NK cells (NK1.1). B. Total DCs in spleen of mice prior to and post-T cell transfer in the presence or absence of MR-1, where indicated. C. Representative flow plots of dendritic cell populations, gated on CD11c⁺ cells. D. Frequency of CD11c⁺ CD11b⁺ CD8α⁻ dendritic cells in spleens. E. Frequency of CD11c⁺ CD11b⁻ CD8α⁺ dendritic cells in spleens. Data are representative of 5 experiments with 3 mice per group. Statistics shown are mean±s.e.m.

We next examined the impact of CD154 blockade on the expression of MHC (signal one) and costimulatory (signal two) molecules on DCs. The DC licensing model predicts that CD40 ligation results in upregulation of MHC and costimulatory molecules. This was confirmed in our model with experiments in which an agonistic anti-CD40 monoclonal antibody (FGK4.5) was injected into mice and splenic DCs were observed to express high levels of MHC and costimulatory molecules (data not shown). Next, we confirmed that CD4⁺ T cells in our system could provide the CD40 stimulus necessary for DC licensing. First, as predicted, CD154 was upregulated on antigen-specific CD4⁺ T cells (data not shown). Second, following adoptive transfer of host-reactive CD4⁺ and CD8⁺ T cells into mOVA chimeric recipients, provision of T cell-derived CD154 signals resulted in a statistically significant increase in class I (H2-K^b) expression on the surface of splenic DCs at day 3 post-transfer (Figure 5). Likewise, expression of CD86 and CD40 were also increased (Figure 5). Surprisingly, however, treatment with CD40/CD154 blockade failed to attenuate the expression of class I or class II MHC, or the expression of costimulatory molecules on total CD11c⁺ splenic DCs during graft-specific T cell priming (Figure 5). We also assessed the expression of costimulatory molecules on individual CD11c⁺ DC subsets (data not shown), and on CD11b⁺ CD11c⁻ myeloid cells (Supplemental Figure 2) in animals left untreated or treated with anti-CD154 and again observed no difference in the expression of class I or class II MHC or CD40, CD80, or CD86. These results therefore indicate that diminution in the provision of either signal one or signal two is not the mechanism by which CD154 blockade attenuates graft-specific T cell responses and promotes long term graft survival.

Figure 5. CD40/CD154 pathway blockade does not impact expression of MHC and costimulatory molecules on dendritic cells.

mOVA chimeric mice were adoptively transferred with 5×10⁶ OT-I and 10⁶ OT-II, and treated with four doses of MR-1 on days 0, 2, 4, and 6, where indicated. Mice were sacrificed on day 3 and spleens were analyzed. A. Representative flow plots of CD11c⁺ DCs expression of H-2K^b, CD86, and CD40 (black lines). Surface molecules on CD11c⁺ DCs in the absence of T cell transfer (shaded grey histograms). B. Summary data of relative expression (MFI) of surface molecules on CD11c⁺ dendritic cells isolated from spleens of mice. Data are representative of 5 experiments with 3 mice per group. Statistics shown are mean±s.e.m. **p≤0.01, ***p≤0.0001.

Next, in order to assess the impact of CD154 blockade on the provision of signal three in this model, we conducted similar experiments in which host-reactive Thy1.1⁺ CD4⁺ and CD8⁺ T cells were adoptively transferred into mOVA chimeric recipients. On day 3 post-transfer, CD11c⁺ splenic DCs were FACS-sorted and either immediately processed for real-time PCR analysis of mRNA or cultured in vitro for 24h for supernatant assessment of cytokine production. Results demonstrated that the adoptive transfer of antigen-specific T cells resulted in a profound increase in the ability of DC to produce pro-inflammatory cytokines. Specifically, DC isolated from T cell adoptive transfer recipients exhibited an increase in IL-6 and TNF at the protein level, and in IL-12 and IL-1β at the mRNA level (Figure 6). Importantly, DCs isolated from mice that had been treated with CD154 blockade exhibited significantly reduced IL-6, TNF, IL-12p35 and IL-1β production on day 3 post-T cell transfer (Figure 6). Neither IL-12p35 nor IL-1β were detected at the protein level in any of the groups, suggesting that the expression of these cytokines in this in vitro assay was below the limit of detection. These data suggest that a predominant effect of CD154 blockade in this system may be to inhibit the provision of inflammatory cytokines (signal three) to antigen-specific T cells during priming.

Figure 6. CD40/CD154 pathway blockade impairs dendritic cell production of inflammatory cytokines.

mOVA chimeric mice were adoptively transferred with 5×10⁶ OT-I and 10⁶ OT-II T cells and were treated with anti-CD154 on days 0, 2, 4, and 6, where indicated. Mice were sacrificed on day 3 post transfer. Splenic dendritic cells were isolated, FACS-sorted and cultured in vitro for 24 hours. Secretion of IL-6 and TNF by isolated DCs was measured in supernatants via cytokine bead array analysis. Production of IL-1β and IL-12p35 transcripts were measured by real time PCR in sorted CD11c⁺ DC. Data are representative of two experiments with 3 mice per group. Statistics shown are mean±s.e.m. *p≤0.05.

Combined IL-6/IL-12p40 blockade attenuates antigen-specific T cell expansion and transiently protects from GVHD

In order to assess the contribution of reduced provision of signal three on the observed prolongation in graft survival associated with treatment with anti-CD154, we utilized a similar experimental approach in which adoptive transfer recipients of host-reactive CD4⁺ and CD8⁺ T cells were left untreated, treated with anti-CD154 as a positive control, or treated with a combination of anti-IL-6R and anti-IL-12p40, two inflammatory cytokines which were diminished in DC isolated from anti-CD154 –treated recipients. We chose to interrogate the impact of blockade of IL-6 and IL-12 in these experiments based on previous evidence in the literature indicating that these cytokines can profoundly impact both CD4⁺ and CD8⁺ differentiation programs (23–27). Results indicated that while untreated animals exhibited a significant loss in OVA-expressing CD45.2⁺ B cells by day 4 post-transfer (Figure 7A, p=0.045), animals treated with either anti-CD154 or the combination of anti-IL-6R and anti-IL-12p40 did not exhibit a significant loss in chimerism. Importantly, combined anti-IL-6R/anti-IL-12p40 protected the animals from the host-reactive immune response to a similar degree as anti-CD154 at this time-point (Figure 7A). We further evaluated the host-reactive CD4⁺ and CD8⁺ T cell responses in these animals and observed that the reduction in the expansion of host-reactive CD4⁺ and CD8⁺ T cells in animals treated with combined IL-6/IL-12p40 blockade was comparable to that observed in anti-CD154 treated animals (Figure 7B and 7C, p<0.05 as compared to untreated controls). However, analysis of both survival of OVA-expressing CD45.2⁺ B cells and expansion of host-reactive CD4⁺ and CD8⁺ T cells at later time points (days 7, 10, and 14 post-transplant) revealed that the protection afforded by IL-6/IL-12p40 blockade was not durable, in that OVA-expressing cells were eventually rejected and host-reactive CD4⁺ and CD8⁺ T cells expanded to frequencies comparable to those observed in untreated animals (data not shown).

Figure 7. Combined IL-6/IL-12p40 antagonism attenuates antigen-specific T cell expansion and transiently protects from GVHD.

mOVA chimeric mice were adoptively transferred with 5×10⁶ OT-I and 10⁶ OT-II T cells and were treated with anti-IL-6R and anti-IL12p40 on days 0, 2, 4, and 6, where indicated. A. Total number of CD45.2⁺ mOVA⁺ B cells in the blood prior to T cell transfer (day 0) and on day 4 post transfer. B. Representative flow plots of antigen-specific Thy1.1⁺ T cell responses in the spleens of treated mice. Data displayed are gated on Thy1.1⁺ CD8⁺ or CD4⁺ T cells on day 4 post-transfer. C. Frequency of antigen-specific CD8⁺ and CD4⁺ T cells at day 4 post-transfer. Data are representative of 2 experiments with a total of 6 mice per group. Statistics shown are mean±s.e.m. *p≤0.05.

Discussion

We observed that although blockade of CD40/CD154 did not alter the level of expression of class I or class II MHC or costimulatory molecules on the surface of dendritic cells, it did significantly alter the differentiation of these cells, specifically with regard to their ability to secrete the inflammatory cytokines IL-6, IL-12p40 and TNF. These results suggest that blockade of CD40 ligation on DC during the course of graft rejection critically impacts the provision of signal three to developing donor-reactive T cell populations, with less of an impact on the provision of costimulation (signal two). These data provide a mechanistic basis for the observed synergy between blockade of the CD40 and CD28 pathways, and suggest that therapies designed to target the provision of inflammatory cytokines during the generation of donor-reactive T cell responses may be beneficial in attenuating these responses and prolonging graft survival.

Our results demonstrating that CD154 antagonism (MR-1) does not inhibit CD4⁺ T cell-induced upregulation of class I or class II MHC on DCs are surprising, given the known role for CD154 in DC licensing. Indeed, our results confirmed reports by many other groups demonstrating that activated CD154⁺ CD4⁺ T cells led to the upregulation of MHC and costimulatory molecules on DCs (28–30). There are two potential reasons that anti-CD154 failed to inhibit this activity. First, it is possible that the antibody does not completely block cell-associated CD154 binding to CD40 expressed on the surface of DCs. Altered CD154 binding could result in a partial signal delivered to the APC, resulting in the upregulation of MHC and costimulatory molecules but not the elaboration of inflammatory cytokines. CD40 signaling in DCs is mediated by binding of individual TRAFs to the intracellular domain, and previous studies have revealed that the CD40 TRAF2/3 binding site is critical for costimulatory molecule expression, while the TRAF6 binding site is required for production of inflammatory cytokines (31). Thus, it is possible that binding of anti-CD154 mAb inhibits the ability of TRAF6 but not TRAF2/3 to be recruited to the CD40 intracellular domain, and experiments to test this hypothesis are ongoing. Our results are also consistent with the alternate possibility that there are CD154-dependent mechanisms by which cognate T cells can upregulate costimulatory molecule expression, but not cytokine production, on DC.

Our results also suggest that this diminution in inflammatory cytokines likely contributes to the observed ability of anti-CD154 to mitigate anti-host T cell responses, in that treatment of animals with a combination of anti-IL-6 and IL-12p40 antibodies partially recapitulated the effects of anti-CD154. Specifically, antagonism of these cytokine transiently attenuated the expansion of both antigen-specific CD4⁺ and CD8⁺ T cell populations and impaired their ability to reject host cells. These results are supported by evidence in the literature suggesting that both IL-12 and IL-6 mediated signals can function to enhance the expansion and differentiation of antigen-specific T cells. With regard to IL-12, seminal studies showed that this cytokine was as effective as CFA for inducing clonal expansion, differentiation into competent effectors, and generation of long-lived memory cells in antigen-specific T cell populations (23–25). Importantly, the absence of IL-12 in these studies rendered the T cells tolerant (24), again suggesting that inhibition of inflammatory cytokine signaling may be an important facet of CD154 antagonism. Likewise, IL-6 has been identified as a predominant inflammatory cytokine associated with the enhancement of alloimmune responses, insofar as neutralization of IL-6 was shown to significantly delay graft rejection, diminish differentiation of alloreactive effectors and impact the Th1/Th2 balance during allograft rejection (32, 33). Furthermore, the inhibition of IL-6 and IL-17 was shown to protect cardiac allografts from rapid rejection (34), and the combined reduction of TNF and IL-6 has been shown in an in vitro system to reduce allograft specific T cell proliferation and differentiation (35).

Another interesting aspect of our study was the finding that adoptive transfer of both CD4⁺ and CD8⁺ antigen-specific T cells was required for the optimal upregulation of MHC and costimulatory molecules and cytokines in this model. Traditional DC licensing models would suggest that binding of activated CD4⁺ T cells would be sufficient to license DC (28–30). However, our data are consistent with other reports in models of viral infection showing that CD8⁺ T cells can “license” resting DC (36), and this effect was later shown to be mediated specifically through the elaboration of GM-CSF (37). Thus, future studies on the effect of CD154 antagonism of CD8⁺ T cell-derived GM-CSF in this model are warranted.

Taken together, this study demonstrates that while CD40/CD154 pathway blockade does not significantly alter the costimulatory receptor profile of dendritic cells, it does alter their cytokine secretion profile. While blockade of two such inflammatory cytokines was insufficient to recapitulate the effects of CD40 blockade, the observed attenuation in host-reactive T cell expansion and delay in GVHD progression suggests that decreased inflammatory cytokine production likely contributes to the observed effects of CD40/CD154 blockade. Thus, targeting DC-derived inflammatory cytokines in clinical transplantation could lead to attenuated alloreactivity and improved outcomes.