CD11b is a Novel Alternate Receptor for CD154 during Alloimmunity

Abstract

Antagonism of the CD154/CD40 pathway is a highly effective means of inducing long-term graft survival in pre-clinical models. Using a fully allogeneic murine transplant model, we found that CD154 blockade was more effective in prolonging graft survival than was CD40 blockade, raising the possibility that CD154 binds a second receptor. To test this, we queried the impact of CD154 antagonism in the absence of CD40. Data indicated that anti-CD154 functioned to reduce graft-infiltrating CD8⁺ T cells in both WT and CD40^−/− hosts. Because it has recently been reported that CD154 can ligate CD11b, we addressed the impact of blocking CD154-CD11b interactions during transplantation. We utilized a specific peptide antagonist that prevents CD154 binding of CD11b but has no effect on CD154-CD40 interactions. CD154:CD11b antagonism significantly increased the efficacy of anti-CD40 in prolonging allograft survival as compared to anti-CD40+control peptide. Mechanistically, CD154:CD11b antagonism functioned to reduce the frequency of graft-infiltrating CD8⁺ T cells and innate immune cells. These data therefore demonstrate that blocking CD154 interactions with both CD40 and CD11b is required for optimal inhibition of alloimmunity, and provide an explanation for why CD40 blockers may be less efficacious than anti-CD154 reagents for the inhibition of allograft rejection.

Introduction

For decades, blockade of CD40-CD154 interactions following transplantation has been shown to be a highly effective means of inhibiting alloreactive T cell responses and inducing long-term graft survival and even tolerance in both murine and non-human primate models (1, 2). For example, anti-CD154 (MR-1) has been the cornerstone of several tolerance induction protocols in mice (3–5). In NHP, the anti-CD154 mAbs have been shown to promote renal and islet allograft survival and facilitate the establishment of mixed chimerism-based tolerance (6). Mechanistically, anti-CD154 has been shown to induce deletion of alloreactive T cell populations and the conversion of convention CD4⁺ T cells to Foxp3⁺ induced Treg (iTreg) (7, 8). Moreover, alloantigen-presenting PDCA-1⁺ plasmacytoid DC (pDC) are essential for the generation of anti-CD154-induced transplantation tolerance (9).

However, the potential of this therapeutic strategy to have a transformative impact on transplantation outcomes has yet to be realized. Specifically, the clinical development of CD154-specific therapies has been circuitous and challenged by an association with thromboembolic events in preclinical and clinical studies (10). The recognition that anti-CD154 mAbs may cause thromboembolism by binding and cross-linking CD154 on platelets (10) gave rise to the possibility that therapeutic targeting of CD40 may achieve the immunosuppressive effects of inhibiting this pathway without disrupting hemostatic mechanisms. For instance, Chi220 was effective at inhibiting rejection in both allogeneic kidney and islet transplant models, although it caused the peripheral deletion of CD40-expressing cells including B cells (11, 12). Clones 3A8 and 2C10 were also reasonably effective in both kidney and islet models when used in combination with CTLA-4Ig (13–15). While these reagents certainly have an ability to significantly prolong allograft survival, none has achieved the remarkable results observed with anti-CD154 mAbs.

Although it is impossible to compare these reagents head-to-head because they have different affinities and are specific for different epitopes, taken together these data raised the possibility that blockade of CD154 vs. blockade of CD40 are actually not mechanistically equivalent. This issue is highly clinically relevant because there has been a dearth of clinically viable anti-CD154 reagents in the pipeline for transplantation over the last 20 years and safety concerns regarding thromboembolism continued to loom large. Recently, two Fc-silent anti-CD154 reagents were developed—one by Bristol Myers Squibb (16) and one by Noelle’s group (Patent Publication Number: 20190309081)-- and tested in autoimmunity, but there are no active clinical organ transplant trials registered with BMS’ letolizumab (a trial in GvHD is recruiting (ClinicalTrials.gov Identifier: NCT03605927)). On the other hand, an anti-CD40 monoclonal antibody ASKP1240 (17–20) has recently been tested in early phase clinical trials in kidney transplantation (21–23). In addition, CFZ533, an Fc-silent anti-CD40 antibody is currently being tested in transplant clinical trials after demonstrating reasonable efficacy and safety in preclinical primate studies (24–26). Thus, as these anti-CD40 reagents make their way through the pipeline for clinical translation in transplantation, it is imperative to determine if there is a biological explanation underlying the observed inferiority of blocking CD40 as compared to blocking CD154, in order to then devise ways to overcome it.

One possible mechanism to support this idea has been the concept that anti-CD154 might function by targeting activated allo-reactive CD154⁺ CD4⁺ T cells for cytolysis via ADCC, a theory supported by a 2003 study by Monk et al (27). However, our group recently reported that Fc-silent anti-CD154 domain antibodies, which lack the ability to bind to Fc receptors and mediate ADCC, are equivalently effective at inhibiting alloimmunity and prolonging graft survival as compared to Fc-intact anti-CD154 reagents (16, 28) (although whether Fc-silent anti-CD154 can induce donor-specific tolerance as has been described for Fc-intact anti-CD154 remains to be determined). These findings therefore debunked the prevailing idea that anti-CD154 functioned primarily by inducing ADCC in CD154⁺ alloreactive T cells. A second potential mechanism to explain the superiority of blocking CD154 compared to blocking CD40 is the possibility that CD154 binds to a second receptor. Importantly, a 2018 Nature Communications paper reported that CD154 is also a ligand for CD11b (29), an integrin molecule that together with CD18 comprises the myeloid cell lineage marker Mac-1 (30). Expressed primarily on myeloid cells, CD11b mediates inflammation by enhancing leukocyte adhesion and migration (30). CD154 binds to CD11b in a concentration-dependent manner in solid phase binding assays (31), and surface-plasmon-resonance analysis revealed a high-affinity interaction between CD154 and CD11b. Given these compelling new findings, here we sought to understand the contribution of CD154-CD11b interactions to allograft rejection.

Materials and Methods

Mice

C57BL/6 (H-2^b) and BALB/c (H-2^d) mice were obtained from the National Cancer Institute (Frederick, MD). CD40^–/–, RAG^−/− mice on a B6 background, and RAG^−/− mice on a BALB/c background mice were obtained from Jackson Laboratories (Bar Harbor, ME) and were maintained at Emory University. All animals were maintained in accordance with Emory University Institutional Animal Case and Use Committee guidelines (Atlanta, GA). All animals were housed in pathogen-free animal facilities at Emory University.

Skin transplantation and in vivo antibody and peptide treatment

Full thickness tail and ear skins were transplanted onto the dorsal thorax of recipient mice and secured with adhesive bandages as previously described (32). In some experiments, recipients received an adoptive transfer of 10⁶–10⁷ donor splenocytes (DST). Mice were treated with an i.p. injection of 250 μg CTLA4-Ig, anti-CD154 dAb (28), anti-CD40 (7-E1G2b) (33) (all kindly provided by Bristol-Myers Squibb, Princeton, NJ), anti-CD154 (clone MR-1), anti-CD11b (clone M1/70) (both from BioXcell) on days 0, 2, 4, and 6, and then once per week continuously until day 50 for skin graft survival experiments. In some experiments, mice received an i.p. injection of 100 μg of either the CD154:CD11b peptide antagonist cM7 (C-EQLKKSKTL-C), or the scrambled control peptide scM7 (C-KLSLEKQTK-C) (31) (all from GenScript, Inc. Piscataway, NJ) on days 0, 2, 4, and 6 and then weekly thereafter until day 50 post-transplantation for skin graft survival experiments.

Flow cytometry

At days 7, 10, or 14 after transplant as indicated, B6 recipients of BALB/c skin grafts were euthanized. Spleens or graft-draining axillary and brachial LNs were harvested. For phenotypic analysis the cells were surface-stained with anti-CD8 (Invitrogen), anti-CD3, anti-CD44, and anti-CD69 (all from BioLegend). Skin grafts were also harvested, chopped into fragments of 0.5–2mm³ and then digested with 2mg/ml collagenase P (Sigma, 11213865001) for 30 min, at 37°C. Isolated graft-infiltrating cells were then stained with anti-CD11b, anti-F4/80, anti-CD45.1, anti-H2K^b, and anti-Gr-1 (all from BioLegend).

Allostimulation assay for intracellular cytokine staining

To assess for donor-reactive T cells, 10⁶ recipient splenocytes were incubated with 2 × 10⁶ BALB/c splenocytes per well in flat-bottom 96-well plates in the presence of 10 μg/ml Brefeldin A for 5 h at 37°C. Subsequently, cells were stained with anti-CD4, anti-CD8, and anti-K^d (to exclude stimulator cells), and then fixed, permeabilized, and stained with anti-TNF, anti-IL-2, and anti–IFN-γ (all from BD) according to the manufacturer’s instructions (BD).

All cells were acquired on an LSR-II flow cytometer (BD), and flow data were analyzed using FlowJo software (Tree Star).

CD154 binding assay

CD11b MicroBeads (130–049-601, Miltenyi Biotec) were used for positive selection enrichment of CD11b⁺ cells from CD40^−/− mouse splenocytes (60% purity). 2×10⁵ CD11b⁺ -enriched cells per well were incubated with 20 μg /ml of mouse sCD154 (Thermo Fisher, 34–8512-80), or PBS as a control, in flat-bottom 96-well plates, for one hour at 37°C. Subsequently, cells were stained with anti-CD154 and anti-CD11b and analyzed by flow cytometry. The ability of sCD154 to stain CD11b⁺ cells in the absence of CD40 was assessed. In some experiments, 50 μg/ml of the CD154:CD11b peptide antagonist cM7 (C-EQLKKSKTL-C), or the scrambled control peptide scM7 (C-KLSLEKQTK-C) (31) (all from GenScript, Inc. Piscataway, NJ) were added during the 1 hour incubation.

Statistical Analysis

Survival data were plotted on Kaplan-Meier curves and log-rank tests were performed. For analysis of T cell responses, non-parametric Mann-Whitney U-tests were performed. When three or more groups were present, data were compared by one-way ANOVA followed by Kruzkal-Wallis post-test. Results were considered significant if p<0.05. All analyses were done using GraphPad Prism software (GraphPad Software Inc).

Results

CD154 blockade inhibits CD8⁺ T cell infiltration into allografts in CD40^−/− mice

To assess the relative efficacy of CD154 vs. CD40 blockade, naïve B6 animals were grafted with BALB/c skin allografts and treated with 1×10⁷ BALB/c splenocytes (donor-specific transfusion, DST) and either no further treatment, anti-CD154 (clone MR-1) or anti-CD40 (clone 7E1-G2b) as described in Methods. Results indicated that while both costimulation blockers conferred significantly increased graft survival over DST alone, administration of the anti-CD154 mAb conferred a significant survival advantage over anti-CD40 (Fig. 1A).

Figure 1. CD154 blockade inhibits CD8⁺ T cell infiltration into allografts in CD40^−/− mice.

A, WT B6 recipients received BALB/c skin grafts and were treated with 1×10⁷ BALB/c splenocytes (donor-specific transfusion, DST), and either no further treatment, anti-CD154 (clone MR-1) or anti-CD40 (clone 7E1-G2b) on days 0, 2, 4, 6. p<0.0001 by log-rank test; n=9–12/group. B-D, WT B6 or CD40^−/− recipients received BALB/c skin grafts and either no further treatment or anti-CD154 on days 0, 2, 4,6. Animals were sacrificed on day 10 and draining lymph nodes (DLN) and skin graft (SG) were analyzed. C, Representative flow cytometry plots and summary data of frequencies of IFN-γ-producing cells in CD8⁺ cell following ex vivo re-stimulation with BALB/c stimulators. D. Representative flow cytometry plots and summary data of frequencies of infiltrating H-2K^b+ CD3⁺ CD8⁺ T cells isolated from skin allografts. Data are representative of 3 independent experiments with a total 10–15 mice/group. *p<0.05, **p<0.005 by one way ANOVA with Kruskal-Wallis post-test. ns, not significant.

Because we have previously published that the efficacy of CD154 blockade is not due to Fc-mediated depletion of activated T cells (28), we next queried whether CD40 is the sole binding partner for CD154. To test this, naïve WT or CD40^−/− were grafted with BALB/c skin allografts and treated with anti-CD154 mAb (MR-1) (Fig. 1B). On day 10, mice were sacrificed and draining LN and skin grafts were analyzed. DLN cells were restimulated ex vivo with BALB/c stimulators. Results indicated that while anti-CD154 mAb treatment effectively reduced the frequency of alloreactive IFN-γ-producing CD8⁺ T cells in the DLN in WT, it failed to do so in CD40^−/− hosts (Fig. 1C). These data indicate that with regard to the priming of the alloreactive CD8⁺ T cell response in the DLN, there are no CD40-independent effects of CD154 blockade. In contrast, analysis of the skin graft infiltrating cells revealed that anti-CD154 mAb functioned to reduce the frequency of infiltrating (recipient-type) H-2K^b+ CD3⁺ CD8⁺ T cells in both WT and CD40^−/− hosts (Fig. 1D). Taken together, these results suggest that CD154 may function in a CD40-independent manner to locally influence CD8⁺ T cell infiltration into allografts.

Blockade of CD154-CD11b interaction synergizes with anti-CD40 mAb to prolong graft survival

A recent study reported that CD154 can form a functional interaction with CD11b, an integrin molecule that together with CD18 comprises the myeloid cell lineage marker Mac-1 (29). To determine whether CD154 can bind murine splenocytes in a CD40-independent manner, we established an in vitro culture system in which CD40^−/− murine splenocytes were cultured with a fluorescently-tagged soluble CD154 (sCD154) or media alone control. Following the incubation period, cells were stained with anti-CD11b and analyzed by flow cytometry. In these experiments, the frequencies of CD11b⁺ cells were very similar between the cultures (67.8% in media alone vs. 68.2% in sCD154), suggesting that sCD154 does not mask the epitope bound by the anti-CD11b antibody. Data indicated that the CD11b⁺ cells in the media alone culture did not stain positive for CD154 (Fig. 2A, top row). In contrast, in the cultures that were incubated with fluorescent sCD154, ~17% of CD40^−/− CD11b⁺ cells stained CD154⁺ (Fig. 2A, middle row), while only ~0.3% of the CD40^−/− CD11b⁻ cells stained CD154⁺ (Fig. 2A, bottom row). These data suggest that CD11b-expressing, CD40^−/− splenocytes can bind to CD154, while CD11b⁻ CD40^−/− splenocytes cannot.

Figure 2. Blockade of CD154-CD11b interaction synergizes with anti-CD40 mAb to prolong graft survival.

A, MACS-enriched CD11b⁺ cells (purity 60%) from CD40^−/− mouse splenocytes were incubated with 20 μg /ml of mouse soluble CD154 or media alone as a control. Subsequently, cells were stained with anti-CD154, and anti-CD11b. The ability of sCD154 to bind to both CD11b⁺ cells (middle row) and CD11b⁻ cells (bottom row) in the absence of CD40 was assessed. B, CD40^−/− MACS-enriched CD11b⁺ cells were incubated with 20 μg /ml sCD154 in the presence of 50 μg of either the CD154/CD11b antagonist peptide cM7 (C-EQLKKSKTL-C), or scrambled control peptide scM7 (C-KLSLEKQTK-C) for 1h as described in Material and Methods. Surface CD154 was detected by flow cytometry. C-D, WT B6 recipients received BALB/c skin grafts plus 10⁶ DST, and either no further treatment, anti-CD154, or anti-CD40 plus either the CD154/CD11b antagonist peptide cM7 (C-EQLKKSKTL-C), or scrambled control peptide scM7 (C-KLSLEKQTK-C) as described in Material and Methods. D. Graft survival was assessed and compared by log-rank test. n=10–15/group.

We next addressed the in vivo impact of blocking the CD154-CD11b interaction. Wolf et al. identified a peptide (sequence EQLKKSKTL, termed M7) that inhibits binding of CD154 to Mac-1 in a dose-dependent manner in a solid phase binding assay and in SPR-analysis (IC₅₀ of 200–900 nmol/L) (31). To improve plasma stability and resistance to degradation by peptidases, M7 was modified by addition of 2 flanking cysteine residues C- and N-terminal and subsequent cyclization by disulfide-bonds (termed cM7). Pharmacokinetic analysis revealed that cM7 coupled to FITC persisted in plasma between 30 minutes and 4 hours with the highest plasma availability at 30 minutes after i.p. injection (31). Importantly, they also showed that cM7 did not alter binding of CD40 to CD154. We tested the ability of this CD154/CD11b blocking peptide to synergize with anti-CD40 blockade (clone 7E1-G2b) (33) in promoting allograft survival in vivo. We first confirmed that in our hands cM7 blocked binding of sCD154 to CD11b by incubating CD40^−/− CD11b-enriched cells with sCD154 for 1h in the presence of either cM7 or the scrambled control peptide scM7. Results demonstrated that the frequency of CD154⁺ CD11b⁺ cells was diminished in the cultures incubated with the CD154:CD11b blocker cM7 relative to scrambled control (Fig. 2B). In addition, the CD154 MFI was reduced from 25887 in control cultures to 2399 in the presence of the CD154:CD11b blocker (Fig. 2B). Naïve B6 recipients received BALB/c allografts and were treated with either anti-CD154 (MR-1) as a positive control, anti-CD40 + a scrambled control peptide (CKLSLEKQTKC, termed scM7), or anti-CD40 + cM7 as described in Methods (Fig. 2C). Results indicated that while anti-CD154 was more efficacious at prolonging graft survival compared to the anti-CD40 + scM7 scrambled peptide (Fig. 2D) (p<0.05), mice treated with anti-CD40 + cM7 peptide exhibited graft survival that was comparable to that of recipients treated with anti-CD154 (Fig. 2D) (p=ns). Likewise, graft survival in mice treated with anti-CD40+ cM7 was significantly prolonged relative to mice treated with anti-CD40+scrambled peptide (Fig. 2D) (p<0.05).

cM7 synergizes with anti-CD40 to inhibit recipient CD8⁺ T cell infiltration into allografts

To assess the impact of cM7 on the alloimmune response following transplantation, naïve B6 CD45.1⁺ animals received BALB/c skin grafts and were left untreated or treated with either anti-CD40 + the scrambled control peptide scM7, or anti-CD40 + cM7 as above. Animals were sacrificed on day 10 post transplant, and draining LN and skin grafts were harvested. Results indicated that animals treated with anti-CD40 + cM7 peptide failed to demonstrate a difference in the number of IFN-γ-producing donor-reactive CD8⁺ T cells in the DLN relative to animals treated with anti-CD40 + scrambled control peptide (scM7) (Fig. 3A). In contrast, analysis of the skin graft infiltrate revealed that cM7 functioned to reduce the number of graft-infiltrating (recipient-type) CD45.1⁺ CD3⁺ CD8⁺ T cells (Fig. 3B). In addition, treatment with cM7 also functioned to reduce the number of CD45.1⁺ CD11b⁺ myeloid (Fig. 3C) and CD45.1⁺ F4/80⁺ macrophage (Fig. 3D) graft-filtrating cells. Taken together, these results suggest that blockade of the CD154:CD11b interaction synergizes with blockade of CD40 to reduce recipient-derived CD8⁺ T cell and innate cell infiltration into allografts.

Figure 3. cM7 synergizes with anti-CD40 to inhibit recipient CD8⁺ T cell infiltration into allografts.

WT B6 recipients received BALB/c skin grafts plus 10⁶ DST, and either no further treatment or anti-CD40 plus either the CD154/CD11b antagonist peptide cM7, or scM7 control peptide as described in Methods. Animals were sacrificed on day 10 and draining lymph nodes and skin grafts were analyzed. A, Representative flow cytometry plots summary data of absolute numbers of IFN-γ-producing CD8⁺ cells following ex vivo re-stimulation. B-D. Representative flow cytometry plots and summary data of absolute numbers of skin graft-infiltrating H-2K^b+ CD3⁺ CD8⁺ T cells (B), H-2K^b+ CD11b⁺cells (C) and H-2K^b+ F4/80⁺ cells (D). Data are representative of 3 independent experiments with a total 10–15 mice/group. *p<0.05 by one-way ANOVA with Kruskal-Wallis post-test. ns, not significant.

Anti-CD11b prolongs allograft survival in the context of CD40 antagonism

Because we identified an important role of CD154:CD11b interaction in inhibiting CD8⁺ immune responses in allografts and prolonging graft survival, we next endeavored to determine whether blocking CD11b itself has a beneficial effect on graft survival in the setting of CD40 blockade. To test this, naïve B6 mice received BALB/c skin grafts and were treated with either anti-CD40 alone or anti-CD40 + anti-CD11b as described in Methods (Fig. 4A). Data indicated that the addition of anti-CD11b significantly improved allograft survival relative to blockade of CD40 alone (Fig. 4B) (p<0.05).

Figure 4. Anti-CD11b prolongs allograft survival in the context of CD40 antagonism.

WT B6 recipients received BALB/c skin grafts and 10⁶ DST, and either no further treatment, anti-CD40 alone, or anti-CD40 plus anti-CD11b (clone M1/70) on days 0, 2, 4, 6 as described in Materials and Methods. Grafts were followed for survival. Data indicate that the addition of anti-CD11b significantly improved allograft survival relative to blockade of CD40 alone (B) (p<0.05). n=10–15/group.

Anti-CD154 inhibits CD11b⁺ innate immune cell infiltration into allografts in RAG^−/− mice

Because our data suggest that CD154 interacts with CD11b to promote immune cell infiltration into allografts, we next queried whether the CD154 that is functioning in this system is coming from a T cell (likely a CD4⁺ T cell) or an innate immune cell. While CD4⁺ T cells are known to be the major expressors of CD154 in the immune system, recent data suggests that under certain conditions innate immune cells such as DC can also express CD154. B6 RAG^−/− mice received skin allografts from BALB/c RAG^−/− mice and were treated with either PBS control or anti-CD154 as described in Materials and Methods (Fig. 5A). First, using this model we observed significant infiltration of CD154-expressing H-2K^b+ cells in grafted RAG^−/− recipients of RAG^−/− allografts as compared to ungrafted RAG^−/− control skin on day 5 post-transplant (Fig. 5B), demonstrating that CD154 can be expressed on non-T cell, recipient-type graft infiltrating cells following transplantation. Moreover, treatment of B6 RAG^−/− recipients of BALB/c RAG^−/− allografts with anti-CD154 resulted in a diminution in the number of recipient-derived H-2K^b+ non-T cells that infiltrated the grafts (Fig. 5C). These findings demonstrate that CD154 derived from a non-T cell source may play a role in promoting innate immune cell infiltration during allograft rejection.

Figure 5. Anti-CD154 inhibits CD11b⁺ innate immune cell infiltration into allografts in RAG^−/− mice.

B6 RAG^−/− mice received BALB/c RAG^−/− mice skin grafts and were treated with either PBS control or anti-CD154 as described in Methods. Skin grafts were harvested on day 5 post-transplant. A. Representative flow cytometry plots and summary data of frequencies of H-2K^b+ CD154⁺ cells among total live cells in naïve (ungrafted) vs. day 5 transplanted RAG^−/− BALB/c skin. Data are representative of 3 independent experiments with a total 9–12 mice/group. C. Representative flow cytometry plots and summary data of absolute numbers of infiltrating H-2K^b+ cells in allograft in untreated vs. anti-CD154 treated mice. Summary graph depicts the fold increase in number of H-2K^b+ cells in both the “No Rx” and “Anti-CD154” groups over naïve skin (set at 1). Data are cumulative of n=11 mice from 3 independent experiments. *p<0.05, ***p<0.001 by Mann-Whitney U-test.

Discussion

These data demonstrate that therapeutic antagonism of CD154 is not equivalent to therapeutic antagonism of CD40, due in part to the potential for binding of CD154 to CD11b, an interaction that could potentially result in APC activation in a CD40-independent manner. We further show that this CD154-CD11b interaction results in increased CD8⁺ T cell and innate immune cell infiltration into recipient allografts, because when CD154-CD11b interactions were blocked with a specific peptide inhibitor, CD8⁺ T cell and myeloid cell infiltration into allografts was reduced. Our data suggest that this CD154-CD11b interaction may function locally at the site of the allograft, because no differences in alloreactive T cell responses were observed in the draining LN or the spleen. Thus taken together, these studies reveal an as-yet unappreciated contribution of CD154-CD11b interaction to the increase in alloimmunity and acceleration in graft rejection. The data presented here offer an explanation for a myriad of preclinical studies in both mouse and non-human primates suggesting superiority of anti-CD154 over anti-CD40 (1, 2, 15, 33). However, making definitive conclusions about the efficacy of therapeutically antagonizing one pathway over another has in the past been difficult, because differences in observed therapeutic efficacy could be ascribed to individual isotypes, affinities, and/or glycosylation patterns of specific antibody reagents. The data presented here reveal that a second receptor for CD154 may underlie the increased efficacy of anti-CD154 reagents relative to anti-CD40 reagents.

This CD154-CD11b interaction has been previously described in the setting of leukocyte recruitment during atherosclerosis. Wolf et al. found that the cM7 peptide decreased peritoneal inflammation and inflammatory cell recruitment in vivo (31). In addition, they reported that atherosclerosis-susceptible mice treated with intraperitoneal injections of cM7 peptide exhibited a reduction in the development of atherosclerotic lesions as compared to controls treated with the scrambled peptide. They concluded that specific inhibition of CD154-CD11b binding could represent an appealing therapeutic approach for the treatment of atherosclerosis and other inflammatory conditions, potentially avoiding the unwanted immunologic and thrombotic effects of global inhibition of CD154. Their data also suggest that specific inhibition of the CD11b I-domain has the advantage of inhibiting its interaction with CD154, but preserving its interaction with other CD11b ligands (29). In contrast, anti-CD11b antibody blocks interaction with other important ligands, including ICAM-1 and RAGE, and thus significantly impairs cell trafficking. This was demonstrated in a model of sepsis in which treatment with anti-CD11b blocked both adhesion and rolling of leukocytes, whereas specific CD11b/CD154 blockade inhibited only cellular adhesion. These findings suggest that the interaction of CD11b with one or more of its other ligands, but not with CD154, may mediate leukocyte rolling in TNF-stimulated mesenteric venules. Thus, in this model of acute inflammation, the CD11b/CD154 interaction selectively contributes to aberrant inflammation by promoting firm adhesion and transmigration of leukocytes. In line with these findings, blockade of CD11b/CD154 interactions limited aberrant inflammation without negatively affecting host defense. These findings suggest that blocking CD11b/CD154 interactions may confer superior suppression of alloimmunity while maintaining protective immunity to pathogens.

Of note, our data showed that blocking CD154 in RAG^−/− mice resulted in a reduction in the infiltration of CD11b⁺ cells into allografts. Thus, while CD4⁺ cells are traditionally thought to be the major source of CD154 expression for APC activation (34), these results suggest that CD154 derived from a non-T cell source may be important for the influx of CD11b⁺ cells into allografts. Of note, we observed that blocking CD154 reduced CD11b⁺ infiltration in B6.RAG^−/− recipients of both allogeneic BALB/c.RAG^−/− and syngeneic B6.RAG^−/− grafts, suggesting that this pathway is engaged as a result of surgery-induced damage and not necessarily innate allorecognition. Consistent with these findings, a published study demonstrated that CD11c⁺ dendritic cells can express CD154 following exposure to inflammatory cytokines and/or TLR ligation (35). Thus, we posit that DC-derived CD154 may interact with CD11b⁺ monocytes/ macrophages to increase CD8⁺ T cell recruitment into allografts.

Whatever the cellular source of CD154 in this system, the mechanisms by which CD154-CD11b interactions function to increase CD8⁺ T cell infiltration are also unknown. It is likely that this effect is mediated via CD11b-derived signals into the monocyte or macrophage, as no intracellular signaling domain is known to exist on CD154 (34). In contrast, CD11b has been shown to promote MyD88-dependent and MyD88-independent signaling pathways in myeloid DC. In particular, Ling et al. showed that CD11b facilitates LPS-induced TLR4 endocytosis and is required for subsequent signaling in the endosomes (36). They went on to show that CD11b deficiency inhibited dendritic cell-mediated TLR4-triggered responses in vivo, culminating in impaired T-cell activation. The authors concluded that CD11b modulated TLR4 signaling and thereby served as a bridge from innate signals that could further fine-tune the adaptive immune response. However, it is currently unclear if in this system CD11b (upon CD154 ligation) is functioning to modify TLR4 signals (potentially introduced through surgery) or some other receptor-ligand interaction. Future experiments to elucidate these pathways are warranted.

There are clear therapeutic implications of this work. Several CD154 and CD40 blockers currently exist in the development pipeline; for example, Bristol Myers Squibb has developed an anti-CD154 domain antibody (16) and Astellas’ anti-CD40 (ASKP1240) is currently in clinical trials for renal transplantation (22). Whether anti-CD40 vs. anti-CD154 reagents are biologically equivalent has been a topic of intense debate, and the data presented here provide a biological basis for the pursuit of reagents to block CD154 instead of those designed to block CD40. However, the high amounts of bioavailable CD154 present on platelets has meant that large and frequent doses of anti-CD154 are required in order to achieve therapeutic efficacy, presenting possible logistic and financial challenges in the therapeutic use of CD154 blockade. The data presented here provide evidence in a pre-clinical model that an alternative approach could be to use anti-CD40 mAb in combination with a specific antagonist of CD154/CD11b interactions, in order to target residual CD154-mediated effects that occur despite CD40 blockade.

Abbreviations

DLN

draining lymph nodes

DST

donor-specific transfusion

iTreg

induced Treg

pDC

plasmacytoid DC