Divergent effects of IL-12 and IL-23 on the production of IL-17 by human T cells

Authors

  • Marieke A. Hoeve,

    1. Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, Leiden, The Netherlands
    2. Institute of Immunology and Infection Research, University of Edinburgh, Edinburgh, UK
    Search for more papers by this author
    • The first two authors contributed equally to this study.

  • Nigel D. L. Savage,

    1. Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
    • The first two authors contributed equally to this study.

  • Tjitske de Boer,

    1. Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
  • Dennis M. L. Langenberg,

    1. Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
  • René de Waal Malefyt,

    1. Department of Experimental Pharmacology and Pathology, DNAX Research Institute, Palo Alto, USA
    Search for more papers by this author
  • Tom H. M. Ottenhoff,

    Corresponding author
    1. Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, Leiden, The Netherlands
    • Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, The Netherlands, Fax: +31-715-216-751
    Search for more papers by this author
    • The last two authors are senior authors on this paper.

  • Frank A. W. Verreck

    1. Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
    • The last two authors are senior authors on this paper.


Abstract

IL-23 is regarded as a major pro-inflammatory mediator in autoimmune disease, a role which until recently was ascribed to its related cytokine IL-12. IL-23, an IL-12p40/p19 heterodimeric protein, binds to IL-12Rβ1/IL-23R receptor complexes. Mice deficient for p19, p40 or IL-12Rβ1 are resistant to experimental autoimmune encephalomyelitis or collagen-induced arthritis. Paradoxically, however, IL-12Rβ2- and IL-12p35-deficient mice show remarkable increases in disease susceptibility, suggesting divergent roles of IL-23 and IL-12 in modulating inflammatory processes. IL-23 induces IL-17, which mediates inflammation and tissue remodeling, but the role of IL-12 in this respect remains unidentified. We investigated the roles of exogenous (recombinant) and endogenous (macrophage-derived) IL-12 and IL-23, on IL-17-induction in human T-cells. IL-23 enhanced IL-17 secretion, as did IL-2, IL-15, IL-18 and IL-21. In contrast, IL-12 mediated specific inhibition of IL-17 production. These data support the role of IL-23 in inflammation through stimulating IL-17 production by T lymphocytes, and importantly indicate a novel regulatory function for IL-12 by specifically suppressing IL-17 secretion. These data therefore extend previous reports that had indicated unique functions for IL-23 and IL-12 due to distinct receptor expression and signal transduction complexes, and provide novel insights into the regulation of immunity, inflammation and immunopathology.

Abbreviations:
CIA:

collagen-induced arthritis

mφ1/2:

macrophage type 1/2

Myc:

Mycobacterium tuberculosis sonicate/lysate

ZymA:

Zymosan A

Introduction

Type-1 cytokines, including IL-12, play an important role in protective immunity to intracellular pathogens and tumors, as well as in inflammatory and organ-specific autoimmune diseases. The biological activities of IL-12 are mediated upon interaction with the IL-12Rβ1/IL-12Rβ2 complex, mainly through STAT4 activation, and include skewing of naïve T cells towards a Th1 phenotype and promotion of IFN-γ production 1, 2. IL-23 has recently been identified as an IL-12-like cytokine, consisting of IL-12p40 coupled to the IL-12p35 homologue p19 3. IL-23 is secreted by activated human macrophages as well as DC 4, and binds a receptor complex that is composed of IL-12Rβ1 as well as a newly identified IL-23-specific receptor chain, IL-23R 3, 5. Like IL-12, IL-23 promotes IFN-γ production and type-1 immunity 3, 5, 6. However, in spite of the structural and functional similarities between IL-12 and IL-23, there are several differences in their biological activities. For instance, although IL-23 activates the same Jak kinases and STAT proteins as IL-12, the resulting DNA binding STAT transcription complexes are different and involve STAT3/4 heterodimers rather than STAT4 homodimers 5, 7. Furthermore, while IL-12 primarily affects naïve T cells, IL-23 predominantly acts on memory T cells 3. In line with this notion, it has been suggested that while IL-12 is involved in the initial phase of the Th1 response, IL-23 sustains proliferation of committed memory T cells, thereby maintaining an adequate memory pool 3, 5, 7.

Type-1 immunity is essential for protective immunity against microbial pathogens and tumors, but uncontrolled type-1 cellular immune responses results in tissue damage, immunopathology and disease. Recent studies have indicated that IL-23 rather than IL-12 is the critical cytokine in mediating chronic inflammation. The first indication for an important role for IL-23 in inflammatory diseases came from the finding that IL-23p19 overexpressing transgenic mice developed systemic inflammation and died prematurely 8. Subsequently, IL-23, but not IL-12, was shown to be essential for the induction of experimental allergic encephalomyelitis (EAE) (the murine model for multiple sclerosis, MS) and collagen-induced arthritis (CIA), since mice deficient for IL-23p19 or p40 protein did not develop disease 912. Accordingly, Zhang et al. 13 reported that IL-12Rβ1-deficient mice are resistant to EAE. Unexpectedly, however, IL-12p35- and IL-12Rβ2-deficient mice, in which only the IL-12 and not the IL-23 function is affected, are more susceptible and developed more severe pathology than wild-type mice 9, 10, 12, 13, with increased levels of IL-23p19 in spleen cells compared to wild-type littermates 13, 14. It was concluded from these studies that IL-12 is not critical for disease development, and may even play an undefined protective role.

The mechanism(s) by which IL-23 mediates inflammation are not completely identified, but a direct link between IL-23 and the induction of IL-17-producing T cells has been postulated 10, 12, 15. IL-17 is a pro-inflammatory, T cell-derived cytokine 16 that stimulates the production of various inflammatory mediators and catabolic enzymes (including IL-6, IL-8, membrane co-factor protein-1, IL-1β, TNF-α, PGE2, GM-CSF, inducble NO synthase, NO and cyclooxygenase-2) by chondrocytes and endothelial, epithelial, fibroblastic and phagocytic cells (reviewed in 17 and 18). IL-17 expression is significantly increased during allograft rejection and in chronic inflammatory diseases, including rheumatoid arthritis, psoriasis, asthma, inflammatory bowel disease and MS 17, 18. IL-17 is produced by activated human CD4+CD45RO+ T cells 19, 20 and its production does not correlate with Th1 or Th2 phenotypes 2123. In addition, IL-2 and IL-15 also induce IL-17 production in human PBMC 24. Recently, Aggarwal et al.15 showed that IL-23 specifically induces IL-17 production in mice. It was not known whether IL-23 can also induce IL-17 production from human T cells, nor how additional cytokines may interact to orchestrate IL-17 production in man. In the present study, we show that not only IL-2 and IL-15, but also IL-23, IL-18 and IL-21 stimulate IL-17 production by human T cells. Most importantly, while IL-12 promoted IFN-γ production, it specifically inhibited the secretion of IL-17 by activated T cells, revealing a novel anti-inflammatory role for IL-12. Although previous reports already indicated differences in expression of IL-12 and IL-23 receptor chains and signal transductions components, our data demonstrate contrasting in vitro effects of IL-12 and IL-23, structurally related cytokines that share subunits as well as receptor chains, in the regulation of immunity and inflammation in man.

Results

Supernatant of pro-inflammatory macrophages stimulated via TLR enhances IL-17 production by activated T cells

We have previously shown that pro-inflammatory mφ1 cells secrete IL-23 but not IL-12p70 upon (myco)bacterial stimulation, while anti-inflammatory mφ2 produce neither IL-23 nor IL-12p70 4. Since macrophages play a key role in inflammation, we first examined whether supernatants of stimulated macrophages (mφ1 and mφ2) affected the production of IL-17 by human T cells. T cell blasts were cultured with or without mitogenic antibodies (combination of anti-CD3 and anti-CD28) in the presence or absence of supernatants of resting or stimulated mφ1 or mφ2 cells. As reported previously, we found that IL-17 was produced by mitogenically stimulated T cells but not by resting T cells 16, 19, 22. Concentrations of IL-17 varied between individual donors (23 and Fig. 14). However, supernatant of mφ1 stimulated with Mycobacterium tuberculosis sonicate (Myc) or Zymosan A (ZymA; a yeast-derived, TLR-2 stimulating agent) significantly increased IL-17 secretion by these T cells (5–12-fold and 3–6-fold, respectively) (Fig. 1). In contrast, supernatant of similarly stimulated mφ2 poorly induced IL-17 secretion (Fig. 1), in line with their inability to secrete IL-23. None of the macrophage supernatants used for T cell co-culture contained detectable levels of IL-17 protein (data not shown). Together, our results indicate that supernatants of stimulated pro-inflammatory macrophages (mφ1), capable of producing IL-23 4, enhance IL-17 production by activated human T cells.

Figure 1.

Supernatants of stimulated mφ1 but not mφ2 increase IL-17 production by activated T cells. IL-17 concentration determined by ELISA in supernatants of T cell blasts treated with anti-CD3/CD28 antibodies in the presence or absence of conditioned medium obtained from mφ1 or mφ2 cells previously activated with mycobacterial cell lysate (Myc), purified ZymA or unstimulated control (medium). IL-17 production is significantly enhanced in T cells incubated with supernatants of activated mφ1 cells only (p<0.05). Graph represents mean of three experiments, with error bars representing ± SEM.

Figure 4.

Supernatant of mφ1 stimulated with TLR antagonist M. tuberculosis lysate in the presence of IFN-γ induces an inhibition of IL-17 production by activated T cells, which can be prevented by blocking IL-12. (A and C) IL-17 concentration determined by ELISA in supernatants of T cell blasts treated with anti-CD3/CD28 antibodies in the presence of conditioned medium obtained from mφ1 previously exposed to medium (-), IFN-γ (500 U/mL) (IFN-γ), Myc (10 μg/mL) or a combination of Myc with IFN-γ (Myc+IFN-γ). (B) IFN-γ does not modify mitogen-induced IL-17 secretion by T cell blasts or IL-12p70-mediated (200 pg/mL) down-regulation of IL-17 release. (C) The IL-12p70-neutralizing antibody C8.6 (1.6–10 μg/mL), but not a control antibody (10 μg/mL) (Ab), partially blocks the inhibition of IL-17 secretion induced by conditioned medium from mφ1 previously exposed to a combination of Myc+IFN-γ. Neither antibody interferes with IL-17 secretion mediated by mitogen stimulation alone or by mitogens plus conditioned medium from untreated or IFN-γ-treated mφ1. Graphs are representative for at least three independent experiments. Error bars represent triplicate measurements ± SEM.

IL-12 inhibits, whereas IL-23 promotes, IL-17 production in activated T cells

To compare the effect of human IL-23 with the structurally related cytokine IL-12 on IL-17 production by T cells, we cultured T cell blasts from four healthy donors with or without anti-CD3/CD28 in the presence or absence of recombinant IL-23 and/or IL-12. Again, non-stimulated T cells did not produce detectable amounts of IL-17, regardless of the presence of cytokines (not shown). IL-23 significantly enhanced IL-17 secretion from stimulated T cells in all donors tested (Fig. 2A), in accordance with previous murine studies 15. A human recombinant IL-23 protein that recently has become available through R&D Systems (Abingdon, UK) enhanced IL-17 secretion from activated T cell blasts to the same extent as our covalently linked IL-23p19/p40 heterodimer (data not shown). The increase in IL-17 was not due to IL-23-mediated cell expansion, since IL-23 had only an insignificant effect on cell proliferation and relative cell numbers, as determined by measuring the uptake of [3H]thymidine or the intensity of a fluorescent viability stain, respectively (data not shown). In contrast to IL-23, IL-12 significantly inhibited the induction of IL-17 from T cells activated with anti-CD3/CD28 in all donors tested (Fig. 2A). IL-12 and IL-23, however, both induced the production of IFN-γ (Fig. 2B) by activated T cells, as described 7, confirming the biological activity of IL-12, and the specificity of IL-12-mediated inhibition of IL-17 secretion. We analyzed the effect of IL-12 and IL-23 on mitogen-induced IL-17 production at earlier and later time points (days 1–4) and found the opposing effects of the two cytokines at all time points (data not shown). Since no effective IL-23p19 blocking antibody was available at the time of the experiments, we were unable to perform IL-23 inhibition studies.

Figure 2.

IL-23, IL-2, IL-15, IL-18 and IL-21 promote, whereas IL-12 inhibits IL-17 production by activated T cells. IL-17 (A) and IFN-γ (B) concentrations in supernatants of T cell blasts treated with anti-CD3/CD28 antibodies in the presence of recombinant IL-12 (200 pg/mL) or IL-23 (100 ng/mL). (C) Flow cytometric analysis of intracellular expression of IFN-γ (top) and IL-17 (bottom) by sorted CD4+ T cells cultured for 5 days in medium alone or in the presence of IL-12 or IL-23. Numbers in the upper right corner represent percentages of double positive cells in the upper right quadrant. IL-12 increases, while IL-23 decreases the number of IFN-γ-producing CD4+ cells. In contrast, IL-12 decreases, while IL-23 increases the number of IL-17-producing CD4+ cells. (D) IL-17 concentration in supernatants of T cell blasts treated with anti-CD3/CD28 antibodies in the presence of IL-2, IL-15, IL-18 and IL-21. IL-12 inhibits the production of IL-17, whereas all other cytokines tested significantly increase secretion of IL-17 in T cell blasts (p<0.05). IL-12 induces IFN-γ production to a greater extent than IL-23. Graphs represent mean of four experiments performed on different donors each time. Error bars represent ± SEM.

To analyze a possible function for IL-23 as a growth factor to increase the number of IL-17-producing T cells, naïve CD4+ T cells were stimulated with anti-CD3/CD28 antibodies and cultured with medium, IL-12 or IL-23. After 5 days, cells were restimulated with PMA/ionomycin and stained for intracellular IL-17 and IFN-γ. Fig. 2C shows that, compared to the number of cells measured in cultures without added IL-12 or IL23, the number of IL-17-producing CD4+ cells was increased in the presence of IL-23 (1.8% vs 0.87%, which is a 52% increase), and decreased in the presence of IL-12 (0.65% vs 0.87%, which is a 25% decrease). Measurement of the number of IFN-γ-producing CD4+ cells showed reciprocal data: a decrease of IFN-γ+ cells in the presence of IL-23 (13.6% vs 16.3%, a 17% decrease), and an increase of IFN-γ+ cells in the presence of IL-12 (21.8% vs 16.3%, a 25% increase). These data suggest that anti-CD3/CD28 activation of human naïve CD4+ cells in the presence of IL-23 favors the outgrowth of an IL-17+/IFN-γ cell population, while IL-12 promotes the development of IL-17/IFN-γ+ cells.

By examining the effect of the type-1 cytokines IL-2, IL-15, IL-18 and IL-21 on the production of IL-17 by human T cells, we determined that the IL-17 enhancing capacity of IL-23 was not unique. In accordance with previous observations in both murine and human models 24, 25, we found that IL-2 and IL-15 also promoted IL-17 secretion from mitogen-activated T cells (Fig. 2D). Moreover, we show here that IL-21, a cytokine structurally related to IL-2 and IL-15, and IL-18 can enhance IL-17 production (Fig. 2D).

IL-12 inhibits IL-23- and IL-18-mediated IL-17 production by activated T cells

IL-12 is known to co-operate with IL-23 and IL-18 to augment IFN-γ production 1, 2, 6. We therefore examined the combined effect of IL-12 plus IL-23, and of IL-12 plus IL-18, on IL-17 production by human T cells. Anti-CD3/CD28-stimulated T cell blasts were cultured with increasing concentrations of IL-12 in the absence or presence of IL-23 (Fig. 3A, B) or IL-18 (Fig. 3C, D). As shown in Fig. 3A, IL-12 significantly inhibited mitogen-induced, as well as IL-23-enhanced, IL-17 secretion in a dose-dependent fashion. Real-time RT-PCR quantification of IL-17 expression confirmed this finding, indicating that the IL-12-mediated inhibition of IL-17 production is apparent at the transcriptional level (data not shown). In contrast, IL-12 significantly enhanced the production of IFN-γ, which was measured in parallel, and IL-23 moderately but consistently increased this IL-12-mediated IFN-γ production (this increase, however, never reached statistical significance) (Fig. 3B). Similarly, while IL-12 cooperated with IL-18 for the induction of IFN-γ (Fig. 3D), it significantly reduced in a dose-dependent fashion IL-18-mediated IL-17 production by activated T cells (Fig. 3C). Taken together, our data show that IL-23 and IL-12 have diametrically opposed effects on the production of IL-17. While IL-12 enhances IL-23- and IL-18-mediated IFN-γ secretion, it suppresses IL-23- and IL-18-mediated IL-17 production by activated T cells. This unique anti-inflammatory potential for IL-12 may be relevant for balancing protective immunity versus immune-mediated pathology.

Figure 3.

IL-12 inhibits IL-23- and IL-18-mediated IL-17 production by activated T cells. IL-17 (A, C) and IFN-γ (B, D) concentrations in supernatants of T cell blasts treated with anti-CD3/CD28 antibodies in the presence or absence of IL-23 (A, B) and IL-18 (C, D), with varying concentrations of recombinant IL-12 depicted on the x-axis. In contrast to IL-23 and IL-18, IL-12 dose-dependently inhibits mitogen-induced IL-17 production as well as IL-23- and IL-18-mediated IL-17 production (A and C, respectively). In contrast, IL-12 dose-dependently induces IFN-γ production from anti-CD3/CD28 stimulated T cell blasts (B and D), and cooperates with IL-23 (B) and IL-18 (D) to enhance the secretion of IFN-γ. Data are representative for at least four independent experiments. Error bars represent values of triplicate measurements ± SEM.

Supernatant of mφ1 stimulated in the presence of IFN-γ inhibits IL-17 production by activated T cells

We have previously shown that mφ1 cells secrete high levels of IL-23 but no IL-12p70 after stimulation with TLR antagonists, and that IFN-γ is required as a secondary signal to induce IL-12p70 secretion 4. Since we found that recombinant IL-12p70 inhibits IL-17 production by activated T cells, we next examined whether stimulation of mφ1 in the presence of IFN-γ would yield supernatants with IL-17 inhibitory capacity. Supernatants of unstimulated or stimulated mφ1 cultured with or without IFN-γ were added to mitogen-activated T cells. Indeed, IL-17 production induced by supernatants of mφ1 stimulated in the presence of IFN-γ (which contained IL-12p70; data not shown) was significantly lower compared to induction by supernatants of mφ1 stimulated with mycobacterial lysate in the absence of IFN-γ (which did not contain IL-12p70; data not shown) (Fig. 4A, C). The reduction of IL-17 secretion was not due to IFN-γ itself, present in the macrophage supernatant, since medium from macrophages cultured in the presence of IFN-γ without Myc did not inhibit IL-17 production (Fig. 4A). Moreover, recombinant IFN-γ did not modify mitogen-induced IL-17 secretion or IL-12-mediated inhibition of mitogen-induced IL-17 production by T cells (Fig. 4B). Addition of the IL-12p70-neutralizing antibody C8.6 blocked, in a dose-dependent fashion, the reduction of IL-17 secretion by activated T cells exposed to supernatants of mφ1 stimulated in the presence of IFN-γ, while a control antibody had no such effect (Fig. 4C). This indicates that IL-12 is indeed mediating the suppression of IL-17. The fact that the reduction of IL-17 could only be partially reverted is likely due to the expected cross-reactivity of C8.6 with the IL-23p40/p19 heterodimer. Thus, addition of C8.6 will not only block IL-12-mediated suppression of IL-17 secretion, but will also interfere with IL-23-mediated enhancement of IL-17 production, preventing complete reversal of the supernatant-induced IL-17 inhibition (Fig. 4C).

Discussion

While IL-12 and IL-23 share the capacity to promote cell-mediated immunity by inducing the production of IFN-γ from Th1 cells 3, 5, 6, 26, it has been postulated that IL-12 and IL-23 also have unique properties. Studies in gene-deficient mice lacking either IL-23-specific, IL-12-specific, or shared IL-23/12 signaling components, have indicated that IL-23 rather than IL-12 is responsible for the chronic inflammation observed in EAE or CIA 914. The molecular mechanism underlying the differential effect of IL-23 versus IL-12 on inflammation, however, remains elusive. To better characterize the role of IL-23 and IL-12 in human T cell-mediated pro-inflammatory signaling, we have studied their ability to modulate the production of IL-17. IL-17 is a pro-inflammatory cytokine which has been associated with active autoimmune disease, e.g., arthritic inflammation of the joint 24, 27, 28 and inflammation of the central nervous system in patients with MS 29, 30. We show here that IL-23 enhances the secretion of IL-17, while, in surprising contrast, IL-12 inhibits the secretion of IL-17 by activated human T cells. Our data thus describe opposing functional properties of IL-12 and IL-23 in human T cells and, importantly, are the first to reveal a unique, direct anti-inflammatory potential for IL-12, which may be relevant for balancing protective immunity versus immune-mediated pathology. Further studies are required to unravel the underlying mechanism(s) behind the differential regulation of IL-17 production by IL-12 and IL-23 in humans.

Two recent papers by Harrington et al.31 and Park et al.32 provide new insights into IL-17 regulation in mice. Both groups show that IL-17-producing CD4+ T cells are a distinct and early lineage of Th effector cells, supporting recent findings by Cua and co-workers 33, 34. The development of these IL-17+ effector cells requires IL-23, CD28 and ICOS, but is independent of the cytokines and transcription factors required for Th1 and Th2 differentiation, and is suppressed by IFN-γ and IL-4. These data suggest a model in which murine IL-17-producing Th cells differentiate directly from naïve CD4+ T cells along a distinct developmental program that diverges early from Th1 and Th2 lineages, and is antagonized by proteins and pathways that dictate the development of Th1 and Th2 cells 35. Our data showing inhibition of (IL-23-mediated) IL-17 production by the Th1 cytokine IL-12 in activated human T cells are in agreement with this view. Following Harrington's and Park's publications 31, 32, we analyzed a possible function for IL-23 as a growth factor to increase the number of IL-17-producing T cells in our system. Mitogen-simulated CD4+ T cells were cultured with medium, IL-12 or IL-23. After 5 days the cells were re-stimulated and stained for intracellular IL-17 and IFN-γ. We found that, compared to the number of cells in cultures with no added IL-12 or IL23, the number of IL-17-producing CD4+ cells was increased in the presence of IL-23, and decreased in the presence of IL-12. Measurement of the number of IFN-γ-producing CD4+ cells showed reciprocal results: a decrease of IFN-γ+ cells in the presence of IL-23, and an increase of IFN-γ+ cells in the presence of IL-12. These data suggest that anti-CD3/CD28 activation of human naïve CD4+ cells in the presence of IL-23 induces the outgrowth of an IL-17+/IFN-γ cell population, while IL-12 favors the development of IL-17/IFN-γ+ cells. However, the differences in cell numbers detected with FACS are too small to fully account for the effect of IL-23 and IL-12 on IL-17 production in activated T cells, especially taking into account that these effects were noticeable already after short-term culture (1–4 days) of the stimulated T cells with the respective cytokines.

Although the IL-17-inducing capacity of IL-23 may help to explain the resistant phenotype of IL-23-signaling-defective mice to EAE and CIA 913, and may contribute to the inflammation observed in IL-23p19-overexpressing mice 8, it is relevant to note that the IL-17-inducing capacity is not unique for IL-23. In accordance with previous observations in both murine and human models 24, 25, we found that IL-2 and IL-15 promote IL-17 secretion by activated T cells. Furthermore, we identified the structurally related cytokine IL-21 as another potent enhancer of IL-17 production. Besides these lymphoproliferative cytokines, IL-18 was also able to enhance IL-17 release, and co-operated with IL-23 to enhance IL-17 production from T cells (M.A.H., unpublished observation). The IL-12-mediated inhibition of IL-17 production, however, was not modified by the presence of either IL-18 or IL-23.

The aforementioned murine EAE and CIA studies elucidated the role of IL-23 rather than IL-12 in these chronic, autoimmune inflammatory processes. However, they provided no explanation for why abolishing IL-12(R)-mediated signaling in p35- or IL-12Rβ2-deficient mice yielded a phenotype of enhanced disease susceptibility/severity 912, 14. Our finding of the exclusive capacity of IL-12 to suppress the release of IL-17 from activated human T cells provides a possible molecular mechanism behind this apparent anti-inflammatory activity of IL-12. Elevated levels of IL-17, resulting from enhanced production by cytokines, including IL-23, IL-18 and IL-21, or from impaired down-modulation by IL-12, may thus contribute to the production of various inflammatory mediators and catabolic enzymes and lead to associated pathology.

While IL-23 and IL-17 play a critical role in autoimmune inflammation, we postulate that IL-12 will accordingly suppress the inflammatory response by inhibiting the secretion of IL-17 from activated T cells. Since IL-23 promotes the production of IFN-γ, which in turn co-stimulates IL-12 production 2, 4, IL-23 indirectly contributes to the inhibition of IL-17 production. Thus, we postulate a self-regulating model in which the inflammatory response initiated by IL-23-producing macrophages at the site of challenge is attenuated by subsequent IFN-γ and IL-12 release (Fig. 5). This implies that the ratio between IL-12 and IL-23, which is critically modulated by macrophage- and DC-derived IFN-γ and dictated by the mode of APC priming and activation 36, 37, may be a relevant determinant in the balance between protective immunity and inflammation. Previously, we and others have identified patients with genetic defects in the IFN-γ/IL-12 signaling cascade who present with serious pathology due to infection with opportunistic mycobacteria and /or salmonella (reviewed in 1). According to the model presented here, these patients may suffer not only from a lack of IFN-γ/IL-12-induced protective cell-mediated immunity, but also from abrogated down-modulation of IL-17-mediated immune responses. Ongoing studies address the possible role of IL-17 in the pathology observed in these patients. Moreover, future studies should aim at elucidating how the regulation of the type-1 cytokines IL-12 and IL-23 controls the balance between protective immunity and immunopathology in patients with various inflammatory and infectious diseases.

Figure 5.

Diagram representing differential effects of IL-12 and IL-23 on T cell-derived IL-17. Stimulation with TLR antagonists induces IL-23 expression by pro-inflammatory mφ1 cells. IL-23 (as well as other cytokines) induces T cells to secrete IL-17 and IFN-γ. IL-17 mediates recruitment of additional inflammatory cells, while IFN-γ costimulates IL-12 secretion from mφ1 cells. IL-12, in turn, inhibits T cell-mediated IL-17 secretion while further increasing IFN-γ production, thus enhancing the loop that down-regulates IL-17 production in T cells.

In summary, our results demonstrate diametrically opposed roles of the structurally similar cytokines IL-12 and IL-23. While IL-23 induces IL-17 expression in human T cells, IL-12 efficiently antagonizes the production of IL-17.

Materials and methods

Reagents

T cells blasts were activated using an immobilized mitogenic anti-CD3 antibody (1 μg/mL; Orthoclone OKT3, Janssen-Cilag, Netherlands), in the presence or absence of soluble human anti-CD28 (2 μg/mL; CLB-CD28/1, Sanquin, Leiden, The Netherlands). The following cytokines were used in the lymphocyte stimulation assays: IL-2 (50 U/mL; Cetus, Emeryville, CA, USA), IL-12 (8–1000 pg/mL; R&D Systems), IL-15 (50 ng/mL; R&D Systems), IL-18 (75 ng/mL; MBL, Woburn, MA, USA) and IL-21 (100 ng/mL; R&D Systems). IL-23 (100 ng/mL) was generated as a covalently linked fusion protein containing FLAG epitope tags as described previously 3. The anti-IL-12 antibody C8.6 was a kind gift from Dr. G. Trinchieri 38. Macrophages were stimulated with 10 μg/mL Myc (obtained by ultrasonication of heat-inactivated M. tuberculosis H37Rv, lyophilized and resuspended in PBS; a kind gift from Dr A. Kolk, Royal Tropical Institute, Amsterdam, The Netherlands), 100 μg/mL ZymA (Molecular Probes, Breda, The Netherlands), 500 U/mL IFN-γ (LUMC, The Netherlands) or appropriate combinations.

Macrophage generation and stimulation

Macrophages (mφ) were generated from monocytes isolated from PBMC of healthy donors as described previously 4. Briefly, MACS-enriched CD14+ monocytes (>98% purity) were cultured for 6 days in RPMI 1640 (GIBCO/Invitrogen, Breda, The Netherlands) with 10% FCS and either 50 U/mL GM-CSF (Novartis Pharma, Arnhem, The Netherlands) to generate pro-inflammatory macrophages (mφ1; IL-12p40+) or 50 ng/mL M-CSF (R&D Systems) to generate anti-inflammatory macrophages (mφ2; IL-10+). After harvesting the cells with trypsin-EDTA, macrophages were washed and seeded at 106/mL in 24-well culture plates (Corning Life Sciences, Schiphol-Rijk, The Netherlands). Cells were allowed to adhere and subsequently cultured for 24 h in the presence or absence of Myc, ZymA, IFN-γ or appropriate combinations. Supernatants were collected, centrifuged and used for co-culture with T cells. Levels of secreted IL-12p70 were determined using a specific CBA (BD Biosciences, Amsterdam, The Netherlands) according to the manufacturer's recommendations, and levels of secreted IL-17 were determined by ELISA (Biosource, Etten-Leur, The Netherlands).

Cellular responses to cytokines and macrophage supernatants

PBMC were obtained from heparinized blood of healthy donors by density gradient centrifugation. To generate T cell blasts, PBMC (106/mL) were incubated in IMDM (Cambrex, Verviers, Belgium) with 10% human serum (Sanquin), PHA (2 μg/mL; Abbott Diagnostics, Dartford, UK) and IL-2 (25 U/mL; Cetus). At day 10, PHA blasts were used directly, or frozen for later use. FACS analysis showed that 95% of these blasts were CD45RO positive, with a ratio of CD3/CD4-positive versus CD3/CD8-positive cells of 2:1. The absence of monocytes, macrophages and DC was confirmed using markers for CD19, CD20, CD14 and CD1a.

T cell blasts were stimulated by culturing 5 × 104 cells in 200 μl in 96-well-plates in triplicate wells in the presence or absence of mitogenic antibodies and (combinations of) cytokines or supernatant of macrophages (representing 50% of final culture volume). Levels of secreted IL-17 and IFN-γ were determined after 72 h in supernatants of pooled triplicate wells by ELISA (Biosource for IL-17 and U-CyTech, Utrecht, The Netherlands, for IFN-γ). In control experiments, cell proliferation and viability were determined by measuring incorporation of [3H]thymidine (0.5 μCi/well for the last 16 h of an 88-h culture) or uptake of calcein-AM (cells were stained after 72 h with 1 μM calcein-AM; Molecular Probes, Breda, The Netherlands), respectively, as described previously 6.

Flow cytometric analysis

Intracellular expression of IFN-γ and IL-17 was measured in CD4+ T cells cultured in medium alone or in the presence of IL-12 (200 pg/mL) or IL-23 (100 ng/mL). CD4+ T cells were isolated from PBMC by CD4 MicroBead magnetic cell sorting (Miltenyi Biotech) per the manufacturer's instructions. After 5 days of culture, cells were activated with PMA/ionomycin for 5 h in the presence of 5 mg/mL Brefeldin A. Cells were washed, permeabilized and fixed using Fixation and Permeabilization Kit for Flow Cytometry according to the protocol issued by the manufacturer (Dako). Cells were stained for 30 min on ice with the following mAb: PE-conjugated mouse anti-IFN-γ (BD Pharmingen, 10 μg/mL final concentration) or PE-coupled biotinylated rabbit anti-human IL-17A (Peprotech, 2.5 μg/mL final concentration) and FITC-conjugated anti-human CD4 (BD Pharmingen). After washing, cells were fixed in 4% paraformaldehyde and analyzed using a FACSCalibur cytometer and CellQuest software (Becton Dickinson, San Jose, CA).

Statistics

Statistical significance of differences between groups was calculated with the Student's t-test using Prism Software (GraphPad). Results differing with a value p<0.05 were considered significant and are referred to as such in the manuscript.

Acknowledgements

This work was supported by grants from The Netherlands Organization for Scientific Research (NWO) and The Netherlands Leprosy Foundation (NLR). We thank Drs. A.G. van Halteren and F. Koning for critically reading the manuscript.

Footnotes

  1. 1

    WILEY-VCH

  2. 2

    WILEY-VCH

  3. 3

    WILEY-VCH

  4. 4

    WILEY-VCH

  5. 5

    WILEY-VCH

Ancillary