Since several years ago, interleukin(IL)-12 is known to be responsible for the differentiation of naive CD4+ T cells into type 1 helper T cells producing interferon-γ. Recently, two other cytokines of the IL-12 family, IL-23 and IL-27, were shown to play key roles in experimental autoimmune disorders mediated by Th17 cells, a novel pro-inflammatory CD4+ T-cell subset secreting IL-17. As our knowledge of IL-12 family members is rapidly growing and changing, it will be important to specify their involvement in the induction and regulation of allograft rejection in animal models as well as in clinical settings. Herein, we review key features of cytokines belonging to the IL-12 family and discuss their potential relevance to transplantation immunity.
There is increasing evidence that acute transplant rejection might be triggered by different types of helper T cells acting via redundant or synergistic pathways. Indeed, type 1 helper T-cell (Th1) responses initiate allograft rejection by promoting cytotoxic T-cell activities and interferon-γ (IFN-γ)-mediated delayed-type hypersensitivity reactions, whereas type 2 helper T-cell (Th2) responses cause allograft damage through the recruitment of eosinophils induced by interleukin (IL)-4 and IL-5. The newly identified Th17 population, which preferentially produces IL-17, is also endowed with potent pro-inflammatory properties, although its involvement in transplant rejection requires further documentation (1). On the other hand, different types of regulatory T cells able to inhibit the rejection process can also emerge as consequence of the interaction of T lymphocytes with antigen-presenting cells. Cytokines produced in the early phase after organ or cell transplantation represent critical factors governing the fate of helper T-cell responses and thereby tipping the balance toward graft rejection or acceptance. Indeed, dendritic cells (DC) presenting alloantigens either by the direct pathway (donor DC) or the indirect pathway (host DC) elicit different types of effector or regulatory T cells depending on their status of maturation and the profile of cytokines they secrete. In the early phase after transplantation, engagement of Toll-like receptors (TLR) by endogenous ligands released during ischemia reperfusion injury or by microbial products is thought to play an important role in the process of DC activation leading to the production of cytokines responsible for the polarization of T-cell responses (2).
IL-12 is a prototypical DC-derived cytokine, which was shown to be a key factor for the development of Th1 responses. In recent years, IL-12 appeared to be part of a family of heterodimeric cytokines which comprises IL-12p70(p35/p40), IL-23(p19/p40) and IL-27(EBI3/p28) (3). While these cytokines share structural and functional similarities, they display rather distinct roles in shaping helper T-cell responses. Furthermore, their production is differentially regulated upon TLR-mediated DC activation. Herein, we review the main properties and activities of IL-12 family members, their potential influence on the outcome of allografts, as well as the TLR-mediated signaling pathways, which regulate IL-12p70 and IL-23 production by DC.
Overview of Interleukin-12 Family Members and Their Receptors
The structure of IL-12 family members has been the subject of several reviews and will only briefly be depicted here (Figure 1) (3). IL-12p70 is a heterodimeric cytokine composed of the p35 and p40 subunits. While p35 is structurally related to type I cytokines, the p40 subunit is homologous to the α-chain of the soluble IL-6 receptor. The IL-12 receptor is formed by the association of the IL12Rβ1 and β2 chains. IL-12p40 is also secreted as monomers or as homodimers (IL-12p80) that signal through IL-12Rβ1. A new protein, termed p19, was identified on the basis of its homology with IL-6 and IL-12p35. This protein was shown to associate with the p40 subunit to form another heterodimeric cytokine known as IL-23. The receptor for this novel cytokine is composed of the IL-12Rβ1 chain and another subunit termed IL-23R. Epstein-Barr virus-induced molecule 3 (EBI3) has been identified as an IL-12p40 homologue. It can associate with IL-12p35 but the function of this heterodimer remains unclear. More recently, EBI3 was found to associate noncovalently with p28, another IL-12p35 homologue, to form IL-27. The IL-27 receptor is composed of WSX1 and gp130 (the latter is also part of the IL-6 receptor complex). This promiscuous interaction between subunits and receptors within the IL-12 family implies that interpretation of experiments performed with mice deficient for a single cytokine or receptor subunit should be interpreted cautiously. As a matter of fact, results obtained with p40−/− or IL-12Rβ1−/− mice were initially attributed to lack of IL-12p70 activity and it is now clear that deficient IL-23 production or signaling is also affected in these mice.
Distinct Functions of IL-12 Family Members
The most important functions of IL-12p70, IL-23 and IL-27 as well as their interrelations in the regulation of inflammation are depicted schematically in Figure 1. IL-12p70 acts on many different cell types. As it is produced by APCs in response to microbial or inflammatory signals and subsequently directs helper T-cell differentiation, it is thought to play a major role in linking innate and adaptive immunity as reviewed by Trinchieri (3). Through activation of STAT4, the primary action of IL-12 is to support differentiation of CD4+ T cells in Th1 effectors and also to promote IFN-γ production by other cell types including CD8+ T cells and natural killer (NK) cells (4,5). As far as humoral immunity is concerned, IL-12 induces a switch in immunoglobulin isotypes by acting on B cells both directly and indirectly via T-cell-derived IFN-γ, resulting in enhanced production of IgG2a antibodies and inhibition of IgE and IgG1 synthesis (6). IL-23 displays structural similarity with IL-12 and was initially described as an alternative Th1-driving cytokine. As a matter of fact, it increases IFN-γ production by CD4+ T cells, but this effect is far less potent than that exerted by IL-12p70 (7,8). The generation of IL-23p19−/− mice indicated that IL-23 but not IL-12 plays a crucial role in the development of experimental allergic encephalomyelitis (EAE), collagen-induced arthritis or T-cell-mediated colitis, as recently reviewed (9). Furthermore, IL-12p35−/− and IFN-γ−/− mice develop more severe EAE, suggesting a protective rather than a causative role of the IL-12p70/IFN-γ axis in this type of inflammatory disorders (10). Although, IL-23 might act directly on monocytes, macrophages and DC (11), IL-17 produced by T cells mediates most pro-inflammatory properties of IL-23. Indeed, analysis of the cytokine profile of pathogenic effector CD4+ cells in autoimmune models where IL-23 plays a dominant role indicated that these cells belong to a new helper T-cell subset named Th17 because it produces IL-17 (1). Recent mouse studies delineated the role of IL-23 in promoting Th17-mediated responses (1,12,13). It appears that IL-23 acts by amplifying and/or stabilizing the responses of differentiated Th17 cells, which express the IL23R chain. Contrary to the initial assumption, IL-23 is not involved in the early differentiation of naïve CD4+ T cells into Th17 cells, a process which, at least in the mouse, is dependent on the combined action of transforming growth factor (TGF)-β and IL-6 (1,12,13). This is an unexpected finding as TGF-β is critically involved in the extrathymic development of Foxp3+ Treg cells (14). IL-6, produced upon infection or inflammation, suppresses the generation of adaptive Treg cells and shifts the balance toward the pro-inflammatory Th17 response, which relies on expression of a novel transcription factor, RORγt, rather than T-bet, GATA-3 or Foxp3 (15). Other mediators, such as IL-1β or TNF-α amplify this Th17 phenotype but cannot substitute for TGF-β and IL-6. Hence, this dual role of TGF-β should be taken into account for the development of strategies aimed at inducing allograft tolerance through the generation of Treg cells.
IL-17 induces inflammation primarily by inducing the release of an array of chemokines and cytokines, which promote recruitment of neutrophils (16). Indeed, IL-17 produced under the influence of IL-23 by tissue-resident T cells (17)—including γ−δ and NK T cells—and antigen-specific Th17 cells stimulates synthesis of IL-1, IL-6, tumor necrosis factor, CXCL8 and CXCL1 by stromal, endothelial cells, epithelial cells and monocytes. It is assumed that a critical role of the IL-23/IL-17 axis is to provide a rapid influx of neutrophils at sites of acute infection or aseptic injury (16). There is an evidence that activated neutrophils can participate to allograft rejection both in the induction and effector phases. In experimental ischemia-reperfusion models, influx of neutrophils is responsible for early graft damage (18,19). Their role in acute rejection has also been demonstrated in a model of fully incompatible cardiac graft (20). Interestingly, absence of IFNγ in these experiments accelerated rejection and uncovered a role for neutrophils in allograft rejection (20). Conversely, rejection of MHC class II incompatible skin graft in IL-4−/− mice was also shown to be mediated by neutrophils (21). As both Th1 and Th2 cytokines inhibit the differentiation in Th17 cells, it is tempting to speculate that the IL-23/IL-17 axis is implicated in neutrophil-mediated rejection when either Th1 or Th2 responses are down-regulated. Importantly, peritubular infiltration of neutrophils was also found in biopsies of kidney allografts during antibody-mediated rejection in humans, indicating a possible role of these cells in clinical settings (22).
Apart from their implication in the development of tissue damage during autoimmune diseases, IL-23 and IL-17 actually participate in host defenses against microbial pathogens (9,12,23). Furthermore, in contrast with IL-12p70 and IFN-γ, which are known to exert anti-tumor activity, IL-23 and IL-17 were recently found to promote cancer incidence and growth by stimulating tumor angiogenesis and inhibiting tumor infiltration by CD8+ cytotoxic T cells (24).
IL-12p80 homodimers were shown to be produced in mice where they might antagonize IL-12p70 action and attract macrophages through binding to IL-12Rβ1. In human monocytes or DC, no formation of IL-12p40 homodimer is observed, leading to the notion that IL-12p40 homodimer might not play any physiological role in humans. However, in both mouse and human airway epithelial cells, IL-12p40 expression induced in allograft results in the preferential formation of IL-12p80. Importantly, concurrent viral infection is associated with synergistic increase in the production of IL-12p80 and exaggerated accumulation of macrophages, thereby providing a possible molecular link between epithelial injury caused by infections and chronic lung allograft dysfunction associated with bronchiolitis obliterans (25).
IL-27 appears to play quite complex roles in the control of innate and adaptive immune responses, as recently reviewed (9). IL-27 was first suggested to promote Th1 responses by inducing STAT-1-dependent expression of the transcription factor T-bet in naïve CD4+ T cells, leading to up-regulation of IL-12Rβ2 and IFN-γ synthesis (26); however, IL-27 was subsequently found to be most often dispensable for the induction of IFN-γ responses in vivo. Whereas IL-27 was initially shown to enhance autoimmune pathology by yet unknown mechanisms, both Th1-type and Th2-type inflammatory responses were found to be amplified in EBI3- or WSX-1-deficient mice infected with parasites or bacteria, indicating a regulatory role for IL-27 on effector T cells (27–29). Recently, IL-27 was shown to suppress pathogenic Th17 cell activities in experimental autoimmune encephalomyelitis as well (30). The duality of IL-27 effects was also apparent at the level of T-cell proliferation. On one hand, IL-27 was shown to stimulate STAT3-dependent T-cell proliferation (31); on the other hand, absence of the IL-27 receptor resulted in enhanced CD4+ T-cell proliferation, which is consistent with the observation that IL-27 is a potent suppressor of IL-2 production (32). Finally, a key function of IL-27 might be to regulate early inflammatory events during acute infections. Indeed, in a series of elegant experiments, Wirtz et al. demonstrated that IL-27 neutralization protects mice against lethal septic peritonitis by enhancing the influx and oxidative burst capacity of neutrophils (33). Altogether, IL-27 now emerges as a cytokine, which suppresses effector functions of several types of immune cells involved in inflammation. It would be of great interest to determine whether IL-27 influences regulatory T-cells responses as these cells were shown to express the IL-27R complex (34). Allograft rejection results from a complex process involving multiple mechanisms operating in synergy. Therefore, the ability of IL-27 to down-regulate different effector pathways, including Th1, Th2 and Th17 responses should prompt us to investigate its therapeutic potential in allograft transplantation.
Cross-Regulation of IL-12p70/IFN-γ and IL-23/IL-17 Pathways: Potential Relevance to Transplant Rejection
Until recently, IL-12p70 was considered as a master switch of T-cell-mediated immunopathology. The discovery of the Th17 CD4+ T-cell subset and the demonstration of the key role of IL-23 in experimental autoimmune diseases led to drastically reconsider this paradigm (1,23,35). It now appears that the Th17 subset producing IL-17 and not the Th1 subset secreting IFN-γ is the major T-cell population responsible for tissue damage in experimental autoimmune encephalomyelitis or collagen-induced arthritis. Whether this will be relevant to human pathology still awaits confirmation. There is evidence that the IL-12p70/IFN-γ and IL-23/IL-17 pathways are cross-regulated. Indeed, IFN-γ is a suppressor of IL-23 synthesis, Th17 development and IL-17 production whereas TGF-β inhibits IFN-γ responses and promotes Th17 differentiation (1,35–37). With the recent demonstration that IFN-γ produced by regulatory T cells contributes to their suppressive function, the cross-regulation between IL-12p70/IFN-γ and IL-23/IL-17 pathways provides a mechanistic basis for enhanced autoimmune inflammation in mice deficient in IFN-γ, IL-12p70 or IL-12Rβ2 chain expression (1,10).
What is the impact of these new concepts on transplantation immunity? Answer to this question requires revisiting classical schemes about the pathophysiology of acute cellular transplant rejection in which Th1 cells and cytotoxic CD8+ T cells induced by IL-12p70 were thought to play a dominant role. There is a multitude of nonexclusive pathways leading to acute or chronic allograft rejection. As the Th1/Th17 dichotomy is presently solely based on data obtained from animal models of autoimmune diseases, generalization of these concepts to transplantation immunity should be approached cautiously. Nevertheless, in several skin or heart allograft models, IL-12p70 was found either to be dispensable or even to delay acute rejection (38,39). This unexpected regulatory role of IL-12p70 was also clearly demonstrated in a model of graft-versus-host disease (GVHD) (40). In this model, donor-derived IFN-γ induced by IL-12 treatment up-regulates Fas expression and sensitivity to FasL-mediated killing, leading to apoptosis of host-reactive T cells by a Fas-dependent pathway, and abortion of the GVH response (41). Furthermore, it is now assumed that the immunosuppressive action of IL-12p70 is mediated by IFN-γ and involves induction of nitric oxide synthase and indoleamine-2,3-dioxygenase in antigen-presenting cells (10) (Figure1). Interestingly, IL-12p70 was also shown to regulate acute vascular xenogeneic rejection by suppressing the production of xenoreactive antibodies (42). On the other hand, the pathogenic role of the IL-23/IL-17 pathway in transplantation immunity remains to be specified but there is already evidence that IL-17 participates to allograft rejection. Indeed, inhibition of IL-17 activity reduced inflammatory infiltrates and delayed acute allograft rejection in rodent models of heart transplantation (43). In a mouse aortic allograft model, IL-17 antagonism decreased acute vascular rejection but did not prevent chronic graft vascular disease (44). The clinical relevance of these findings is suggested by increased expression of IL-17 by graft-infiltrating cells during kidney graft rejection (45) and increased IL-17 levels in bronchoalveolar lavage during acute lung transplant rejection in humans (46). Interestingly, an experimental study recently published in this journal revealed that lung ischemia-reperfusion injury elicits anti-type V collagen pathogenic T cells that express IL-17, suggesting that early inflammatory events post-transplantation might promote IL-17-mediated autoimmune pathology (47).
Differential Regulation of IL-12p70 and IL-23 Synthesis
IL-12p70 and IL-23 are secreted by antigen-presenting cells (macrophages, DC) in response to danger signals. There is evidence that the nature of the danger signal determines whether IL-12p70 or IL-23 will be preferentially produced by DC, which represents an additional level of cross-regulation between IL12p70/IFN-γ and IL-23/IL-17 pathways (23). For example, both TLR2 ligands and ATP are strong inducers of IL-23 but not of IL-12p70 (48,49). This might especially be relevant to kidney transplantation as TLR2 was shown to be involved in renal ischemia-reperfusion injury (50). Since IL-12p40 chain is shared by IL-12p70 and IL-23 whereas IL-12p35 is unique to IL-12p70 and p19 to IL-23, these findings suggest that signaling events controlling IL-12p35 gene expression are distinct from those involved in IL-12p40 and p19 expression. Indeed, IL-12p35 is not synthesized upon TLR2 ligation whereas IL-12p40 and IL-12p19 are secreted in the same setting (48). Furthermore, IL-4 strongly increases bioactive IL-12p35 expression while inhibiting IL-12p40 synthesis (51), whereas IFN-γ up-regulates IL-12p40 and IL-12p35 but inhibits IL-23 synthesis (36), presumably by affecting p19 expression.
Since the myeloid differentiation factor 88 (MyD88)-dependent signaling pathway coupled to most TLR was found to be critically involved in certain models of transplant rejection (2), it is relevant to mention here that TLR-mediated induction of both IL-12p70 and IL-23 requires MyD88. Downstream of TLR4, a major receptor for bacterial endotoxin and endogenous stress ligands, IL-12p70 synthesis also critically depends on the MyD88-independent pathway (52) (Figure 2). This MyD88-independent pathway involves the recruitment of Toll/IL-1 receptor domain-containing adaptor inducing IFN-β (TRIF) and results in activation and nuclear translocation of IRF-3, a key transcription factor for type 1 IFN-dependent genes. Interestingly, p35 but neither p40 nor p19 activation is affected in DC from IRF-3 deficient mice. IRF-3 activation controls IL-12p35 gene expression by two mechanisms. Firstly, IRF-3 directly binds and transactivates the IL-12p35 gene promoter (52). Secondly, this transcription factor induces the expression of IFN-β which was shown to promote IL-12p35 gene expression in an autocrine manner (53). At the IFNβ promoter, IRF-3 is responsible for the recruitment of transcriptional coactivators and chromatin remodeling complexes. The latter mechanism could also be implicated in IL-12p35 gene transcription since disruption of a strategically positioned nucleosome is required for the interaction of Sp1 with a critical binding site in the IL-12p35 gene promoter (54). Deficient IRF-3 activation can thus lead to selective impairment in IL-12p70 synthesis in response to bacterial endotoxin, as we observed in studies on cord blood-derived DC (55). These findings might be relevant to cord blood stem cell transplantation since responsiveness of donor antigen-presenting cells to endotoxin represents an important risk factor for acute graft-versus-host disease (56).
Elucidating the role of the Th17 CD4+ T-cell subset in graft rejection emerges as a priority for future transplantation research. Indeed, Th17 cells might represent important players in certain forms of rejection, especially those characterized by neutrophil infiltrates. However, current available data on the role of Th17 cells stems from a few experimental models of auto-immunity. There could be major differences in the Th1/Th17 paradigm between mice and men, especially in the context of complex situations such as transplant rejection. Nevertheless, as the balance between IL-23 and IL-12p70 appears to be critical for the differentiation of Th17 cells, it will be interesting to determine the impact of existing and novel immunosuppressive drugs on the production of IL-23 and IL-12p70 by antigen-presenting cells as well as on IL-17 synthesis by T lymphocytes. Special attention should be paid to the consequences of ischemia-reperfusion injury and infections on the IL-23/IL-17 pathway as this might have a major influence on graft outcome. The relation between regulatory T cells and Th17 cells will also deserve in-depth investigations in view of the role of TGF-β in Th17 cell development. Finally, beside IL-23 and IL-12p70, IL-27 is the third IL-12 family member that might be important to consider in transplantation immunity in view of its regulatory role during T-cell responses.
The Institute for Medical Immunology is supported by the government of the Walloon Region, GlaxoSmithKline Biologicals and the Fonds National de la Recherche Scientifique (FNRS, Belgium). S.G. is a postdoctoral researcher of the FNRS.