Article Text

Extended report
Role of IL-17 in the Th1 systemic defects in rheumatoid arthritis through selective IL-12Rβ2 inhibition
  1. Myew-Ling Toh,
  2. Masanori Kawashima,
  3. Arnaud Hot,
  4. Philippe Miossec,
  5. Pierre Miossec
  1. Department of Immunology and Rheumatology, Immunogenomics and Inflammation Research Unit EA 4130, University of Lyon, Edouard Herriot Hospital, Lyon, France
  1. Correspondence to Professor Pierre Miossec, Clinical Immunology Unit, Department of Immunology and Rheumatology, Edouard Herriot Hospital, 5 Place d'Arsonval, 69437 Lyon Cedex 03, France; miossec{at}univ-lyon1.fr

Abstract

Background Patients with rheumatoid arthritis (RA) have a systemic Th1 defect associated with inflammation.

Objective To examine the hypothesis that interleukin 17 (IL-17) contributes to this defect.

Methods IL-17 effects on Th1 markers were examined on T-bet and interferon γ (IFNγ) expression in peripheral blood mononuclear cells (PBMCs) from patients with RA or healthy controls (HC). Receptor specificities were determined by analysis of the Th1-specific IL-12 receptor β2 (IL-12Rβ2), Th17-specific IL-23R and the common IL-12Rβ1 chain expression. Effects of IL-17 or IFNγ on IL-6, IL-1, IL-8, matrix metalloproteinase-8 (MMP-8) were measured by real-time RT-PCR in RA synovial cells.

Results RA PBMCs were less responsive to IL-12-induced IFNγ than HC PBMCs. IL-12 hyporesponsiveness was increased by IL-17 treatment associated with a selective reduction in IL-12Rβ2, but not IL-23R, IL-12Rβ1 or T-bet, which was reversed with IL-17R inhibition. IL-17 inhibited IL-12Rβ2 expression in developing Th1 cells. In RA synovial cells, IL-17 induced IL-6, IL-1, IL-8 and MMP-8, whereas IFNγ had minimal or inhibitory effects.

Conclusion In RA, IL-12 hyporesponsiveness is associated with IL-17R-mediated downregulation of IL-12Rβ2 expression. IL-17 may reinforce Th17 lineage commitment and proinflammatory and destructive effects through Th1 inhibition and positive feedback effects in RA synovial cells. Anti-inflammatory effects of IL-17/IL-17R antagonism may include the restoration of protective Th1 responses.

View Full Text

Statistics from Altmetric.com

Th1 defects have been reported in rheumatoid arthritis (RA) associated with increased susceptibility to cell-mediated infections including tuberculosis.1 The defect is characterised by reduced responsiveness to interleukin 12 (IL-12) and IL-18 and reduced interferon γ (IFNγ) production by blood cells2 and has been attributed to chronic inflammation. Because of the contribution of IL-17 to inflammation,3 we considered here the possible contribution of IL-17 to this systemic defect.

Th1, Th2 and Th17 differentiation is polarised and dependent on positive autocrine and negative counterregulatory effects on key receptors and signature cytokines.4 Th1 development requires T-bet activation,5 IL-12 and IL-18 stimulation leading to IL-12 receptor β2 (IL-12Rβ2) upregulation, and IFNγ production. Th17 differentiation is dependent on transforming growth factor β (TGFβ), IL-6, IL-1, IL-21, and IL-23 stimulation leading to IL-23R upregulation and IL-17 production.3 In Th1 development, IL-12 mediates its effects via binding to the IL-12Rβ1 and IL-12Rβ2 subunits of the IL-12R.3 IL-12Rβ2 is crucial and highly regulated in Th1 development, whereas the IL-12Rβ1 subunit is constitutively expressed and shared with the functional IL-23R.6 IL-12 or IFNγ reinforce Th1 commitment by upregulation of IL-12Rβ2, and simultaneously inhibit Th17 differentiation. IL-4 promotes Th2 development, partly by suppression of IL-12Rβ2, and thus Th1 differentiation.

Although IFNγ has counter-regulatory effects on Th17 development,4 7 the inverse situation is not well known in humans.8 In this study, we examined the hypothesis that the proinflammatory cytokine IL-17 contributes to the Th1 defect in RA through negative regulation of Th1 differentiation.

Materials and methods

Cells

Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll centrifugation from 16 patients with RA or 10 age- and sex-matched healthy controls (HC). The characteristics of patients with RA are described in online supplementary file, part 1. CD14 monocytes were depleted by CD14 microbeads (Miltenyi Biotec, Paris, France). Cells were cultured in serum-free medium (Cambrex, Paris, France) for 24 h with IL-12 10 ng/ml or IL-17 10–100 ng/ml (R&D Systems, Lille, France). IL-17 antagonism was achieved using an anti-IL-17R monoclonal antibody or control goat IgG 2 µg/ml (R&D Systems). IFNγ was measured in supernatants by ELISA (R&D Systems).

RA synovium tissue and cells were obtained from patients with RA undergoing joint surgery who fulfilled the American College of Rheumatology criteria. Synovial tissue from 11 patients with RA was snap frozen and stored at −80°C for subsequent RNA extraction. A detailed description is included in the online supplementary file, part 2.

Real time RT-PCR

RNA was extracted from cells or synovial tissue using TRIzol (Gibco BRL, Cergy Pontoise, France). RNA (1 µg) was reverse transcribed using ThermoScript RT-PCR System (Invitrogen, Carlsbad, California, USA), and PCR amplification performed on a LightCycler instrument (Roche) using a LightCycler DNA Master SYBR Green I kit (Roche, Meylan, France). Primers used for IL-17, IL-12Rβ1 and IL-12Rβ2 are described in online supplementary files, part 3. Primers for IFNγ, TGFβ, tumour necrosis factor α (TNFα), IL-1β, T-bet, GATA-3, IL-10, IL-4, IL-6, matrix metalloproteinase-8 (MMP-8), PPIB and β-actin were purchased from Search-LC (Heidelberg, Germany). mRNA was normalised to PPIB expression in peripheral blood cells or β-actin in synovial tissue or cells, and results presented as the fold induction of mRNA expression relative to control samples.9

Statistical analysis

Results are expressed as the mean±SEM. Analysis was performed using Student t test and p values <0.05 were considered statistically significant. All data are the result of separate experiments from at least three separate RA or HC donors.

Results

IL-17 inhibits IFNγ in RA PBMCs and selectively down regulates IL-12Rβ2

IL-12 increased IFNγ mRNA expression in HC PBMCs with a smaller effect in RA PBMCs (fourfold vs ninefold increase, respectively, figure 1A) and this was confirmed at the protein level. IL-17 more potently decreased IL-12-induced IFNγ mRNA expression in RA PBMCs than in HC (figure 1A). The suppressive effects of IL-17 on IFNγ expression was associated with a 100% decrease in IL-12-induced expression in RA PBMCs, compared with 50% in HC PBMCs (figure 1A). To exclude autocrine effects of IL-17 on its own expression, we examined IL-17 mRNA expression in the same conditions. IL-17 did not increase its own expression and IL-12 reduced by 50% IL-17 expression in RA PBMCs (figure 1B).

Figure 1

(IL-17 inhibits IFNγ expression in PBMCs from patients with RA. (A) RA or HC PBMCs were treated with IL-12 and/or IL-17 10 ng/ml for 24 h, and IFNγ mRNA measured by real-time PCR. (B) IL-17 mRNA expression was measured in RA PBMCs. The results based on a ratio of target gene mRNA/PPIB mRNA amplification are presented as fold induction in target gene mRNA expression relative to the untreated control (mean±SEM, n=6). *p<0.05 compared with untreated control; †p<0.05 compared with IL-12 treatment alone. HC, healthy controls; IFNγ, interferon γ; IL-17, interleukin 17; PBMCs, peripheral blood mononuclear cells; RA, rheumatoid arthritis.

For IL-12R chain mRNA expression, there were no differences in baseline IL-12Rβ2 or IL-12Rβ1 mRNA expression between RA and HC PBMCs (figure 2A), even in larger cohorts (n=25 patients, data not shown). In contrast, IL-12-induced IL-12Rβ2 mRNA expression was decreased in RA compared with HC PBMCs (2.4-fold vs 4.2-fold increase, respectively, figure 2A). Similarly, RA PBMCs were less responsive to IL-12 induction of IL-12Rβ1 compared with HC (figure 2B). Unfortunately, we could not confirm these results at the protein level because of the lack of adequate reagents for IL-12Rβ2 fluorescence-activated cell sorting staining.

Figure 2

IL-17 inhibits IL-12Rβ2 expression in RA PBMCs. (A–C) PBMCs from patients with RA or HC were treated with IL-12 and/or IL-17 10 ng/ml for 24 h and IL-12Rβ2, IL-12Rβ1, IL-23R mRNA expression measured or (D) the ratio of the fold induction of IL-12Rβ2/IL-23R measured by real-time RT-PCR. Effect of IL-17R inhibition. (E) RA PBMCs were treated with IL-12±anti-IL-17R monoclonal antibody or control goat IgG 2 µg/ml for 24 h and IFNγ, IL-12Rβ2 mRNA expression analysed by real-time RT-PCR. *p<0.05 compared with untreated control; †p<0.05 compared with IL-12 treatment alone. HC, healthy controls; IFNγ, interferon γ; IL-12Rβ2, IL-12 receptor β2; IL-17, interleukin 17; PBMCs, peripheral blood mononuclear cells; RA, rheumatoid arthritis.

We next examined if IL-17 may modulate IL-12Rβ2 mRNA expression. IL-17 led to a 50% decrease in IL-12-induced IL-12Rβ2 expression in HC PBMCs, compared with a massive inhibition in RA PBMCs (figure 2A, black bars). In contrast, IL-17 did not reduce, but rather increased, IL-12-induced IL-12Rβ1 or IL-23R expression (figure 2B,C). IL-17 reduced IL-12-induced IL-12Rβ2/IL-23R ratios by approximately 2.7-fold in RA PBMCs (figure 2D). To confirm the downregulation of IL-12Rβ2 by IL-17, we examined the effect of IL-17R inhibition. A neutralising anti-IL-17R monoclonal antibody increased endogenous or IL-12-induced IFNγ and IL-12Rβ2 mRNA expression in RA PBMCs to HC levels (figure 2E).

IL-17 does not suppress T-bet

To determine if the IL-17-mediated increase in IL-12 hyporesponsiveness occurred at the level of T-bet or downstream, we examined modulation of IL-12-induced T-bet by IL-17. There were no differences in baseline T-bet expression between RA and HC. The trend to reduced IL-12-induced T-bet expression with IL-17 was not statistically significant (figure 3A). This indicated that the effect of IL-17 on IL-12 responsiveness occurred downstream of T-bet.

Figure 3

Regulatory effects of IL-17 on developing Th1 cells require antigen-presenting cell interactions. Effect of IL-17 on T-bet expression in RA PBMCs. (A) PBMCs from patients with RA or HC were treated with IL-12 and/or IL-17 10 ng/ml for 24 h and T-bet mRNA expression measured by real-time RT-PCR. Effect of monocyte depletion. (B) CD14 monocytes (Mo) were depleted from RA PBMCs, treated with IL-12 and/or IL-17 10 ng/ml for 24 h and IFNγ, IL-12Rβ2 mRNA expression measured by real-time RT-PCR. Effect of IL-17 on IL-4, IL-10 and TGFβ mRNA expression. (C) RA PBMCs were treated with IL-12 and/or IL-17 10 ng/ml for 24 h and IL-4, IL-10 or TGFβ mRNA expression measured by real-time RT-PCR. *p<0.05 compared with untreated control; †p<0.05 compared with IL-12 treatment alone. HC, healthy controls; IFNγ, interferon γ; IL-12Rβ2, IL-12 receptor β2; IL-17, interleukin 17; PBMCs, peripheral blood mononuclear cells; RA, rheumatoid arthritis; TGFβ, transforming growth factor β.

IL-17 inhibition of IL-12Rβ2 does not require T-cell–monocyte interactions

Monocyte depletion of RA PBMCs increased IL-12 responsiveness and IL-12Rβ2 mRNA expression to levels in HC, which were then subsequently reduced by IL-17 (figure 3B). This suggested that the inhibitory effects of IL-17 on IL-12Rβ2 in PBMCs did not require T-cell interactions with antigen-presenting cell (APC) monocytes. This also excluded possible indirect suppressive effects of IL-10 mostly produced by monocytes.10 To confirm a lack of indirect effects by IL-10 and exclude indirect inhibition of IFNγ by IL-4 or TGFβ, we examined mRNA levels of these cytokines in RA PBMCs but there were no significant changes in the presence of IL-12 and/or IL-17 treatment (figure 3C). However, IL-17 treatment alone decreased IL-4 mRNA expression.

IL-17 has more potent regulatory effects on proinflammatory and destructive mediators than IFNγ in RA synovium

To extend our results in RA PBMCs to joints, we examined baseline IL-17, IFNγ, IL-23R and IL-12Rβ2 mRNA expression in RA synovium cells. We detected increased synovial IL-17 compared with IFNγ mRNA expression, which was barely detectable (p<0.05, figure 4A, left panel) and increased IL-23R compared with IL-12Rβ2 mRNA expression (p<0.05, figure 4A, right panel).

Figure 4

Effect of IL-17 and IFNγ in RA synovial cells and RA FLS. (A) IFNγ, IL-17, IL-23R, IL-12Rβ2 mRNA expression levels were measured in RA synovial tissue samples by real-time PCR. The results are based on the ratio of target gene/β-actin mRNA amplification (mean±SEM, n=11). Median values are indicated by a horizontal line. *p<0.05, IL-17 or IL-23R mRNA compared with IFNγ or IL-12Rβ2 mRNA, respectively. (B) RA synovial cells were cultured for 24 h with IL-17 or IFNγ (0.1–100 ng/ml) and IL-17, IFNγ, IL-6 or MMP-8 mRNA levels determined by real-time RT-PCR (mean±SEM, n = 3). (C) RA FLS were cultured for 24 h with IL-17 or IFNγ (0.1–100 ng/ml) and IL-6 or MMP-8 mRNA levels determined by real-time RT-PCR (mean±SEM, n=3). Results are based on a ratio of target gene mRNA/β-actin mRNA amplification, and presented as the fold induction in target gene mRNA expression relative to the untreated control. *p<0.05 compared with untreated control. FLS, fibroblast-like synoviocytes; HC, healthy controls; IFNγ, interferon γ; IL-12Rβ2, IL-12 receptor β2; IL-17, interleukin 17; MMP-8, matrix metalloproteinase-8; PBMCs, peripheral blood mononuclear cells; RA, rheumatoid arthritis; TGFβ, transforming growth factor β.

We examined the effects of IL-17 on IFNγ or vice versa in RA synovial cells. IFNγ further increased IFNγ mRNA expression (figure 4B, left panel). IL-17 inhibited IFNγ mRNA expression in RA synovial cells only at 100 ng/ml. Conversely, IFNγ mildly suppressed IL-17 mRNA expression but this was not statistically significant (figure 4B, right panel). IL-17 had no effect on IL-17 mRNA expression in RA synovial cells.

In RA synovial cells, IL-17 significantly induced IL-6, IL-1 and MMP-8 mRNA expression (figure 4B, lower panels) and IFNγ inhibited or had no significant effect.

Discussion

IL-17 has an established proarthritic role in RA.11 12 In contrast, IFNγ is rather protective, although its role is complex and poorly understood.13 14 A systemic Th1 defect has been reported in patients with RA characterised by decreased IL-12 responsiveness associated with systemic inflammation; however, the role of IL-17 has not been elucidated in RA.2 8 In this study, we demonstrate that IL-17 may contribute to this systemic Th1 defect. IL-17-mediated inhibition of IL-12Rβ2 enhanced IL-12 hyporesponsiveness in Th1 cells, which was restored by IL-17R antagonism. In the local RA synovial milieu, IL-17 had more important proinflammatory and destructive effects than IFNγ and increased the key cytokines which amplify Th17 differentiation.

IL-17 selectively inhibited IL-12Rβ2 but not IL-12Rβ or IL-23R expression. Removal of monocyte IL-10 did not affect IL-17 inhibition of IL-12Rβ2. Furthermore, IL-17R antagonism restored IL-12Rβ2 expression to HC levels. The more potent effects of IL-17R antagonism compared with modest effects of exogenous IL-17 are consistent with data in vitro and in vivo indicating synergistic interactions between IL-17 and other inflammatory cytokines.11 In addition, IL-17 inhibition of IL-12Rβ2 expression did not require T-cell–monocyte interactions. However, IL-17 inhibited IL-12Rβ2 in developing Th1 cells in the presence of APCs. We did not exclude the regulatory effects of IL-17 on APCs other than monocytes such as natural killer (NK) cells, NK T cells or dendritic cells, which secrete abundant amounts of IFNγ and express IL-12Rβ2.4 15

Inhibitory effects of IL-17 on Th1 development may help explain systemic Th1 defects observed in patients with RA. The impaired Th1 response has been linked to increased disease activity and is reversed with anti-TNF therapy.16 In this study, either IL-17R antagonism or removal of IL-10 almost completely restored IL-12 responsiveness in vitro. Whereas inhibitory interactions between IL-10 and IFNγ have been reported,10 direct inhibition of IFNγ by IL-17 was found in this study. Taken together, anti-inflammatory treatment such as IL-17/IL-17R antagonism may increase IFNγ and reduce TNF-related cytokine networks with a net increase in Th1 function in patients with RA.

In the local synovial tissue environment, we observed limited reciprocal regulatory effects between IFNγ and IL-17. There was a relative increase in synovial IL-17 and IL-23R compared with IFNγ and IL-12Rβ2. There was a trend towards suppressive effects of IFNγ on IL-17 expression, whereas IL-17 did not upregulate its own expression. IL-17 was a more potent mediator of joint inflammation and destruction than IFNγ. These results are in line with the inhibitory effect of IFNγ on mouse arthritis though inhibition of IL-17.14 RA synovium contains abundant amounts of TGFβ, IL-6 and IL-1, three cytokines implicated in Th17 differentiation and simultaneously in suppression of TGFβ-mediated regulatory T-cell development.17 18 IFNγ either had no effect or inhibited synovial cell IL-6, IL-1, IL-8 or MMP-8. RA synovial tissue IL-17 correlated with increased joint damage, and IFNγ expression with reduced joint damage.12 IL-17 increased and IFNγ inhibited osteoclastogenesis.19 20 These data suggest an important local amplifying role of IL-17 on Th17 development in the interplay between T cells and stromal cells.

In conclusion, IL-17 inhibited IL-12Rβ2 expression in developing Th1 cells, which may contribute to systemic Th1 defects in RA. Locally, more potent immunoregulatory effects of IL-17 compared with IFNγ on stromal and accessory cells had positive feedback effects on Th17 development. Finally, these data suggest an additional important anti-inflammatory effect of IL-17/ IL-17R antagonism through restoration of protective IFNγ expression in RA.

Acknowledgments

We thank Gaelle Gonzales for expert technical help.

References

View Abstract

Supplementary materials

  • Web Only Data ard.2009.111757

    Files in this Data Supplement:

Footnotes

  • Funding This work was supported in part by grants from the Hospices Civils of Lyon and the Region Rhône-Alpes. M-LT and MK were supported by a fellowship from the Region Rhône-Alpes. AH was supported by a grant from the Société Française de Médecine Interne.

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the Hospital of Lyon.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Patient consent Obtained.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.