Th1-derived IFN-γ targets naive T cells and inhibits Th17 development. However, Th1, Th17, and memory but not naive T cells are colocalized in an inflammatory environment. To demonstrate the kinetic relationship between these T cell subsets, we investigated the role of IFN-γ in regulating the development and balance between Th17 and Th1 in humans. We show that IFN-γ stimulates B7-H1 expression on APC subsets and abates their Th1 polarization capacity in a B7-H1-dependent manner. Interestingly, IFN-γ triggers APCs to produce IL-1 and IL-23 and enables them to induce memory Th17 expansion via IL-1 and IL-23 in a B7-H1-independent manner. We propose a novel dynamic between Th1 and Th17 in the course of inflammation as follows: Th1-mediated inflammation is attenuated by IFN-γ-induced B7-H1 on APCs and is evolved toward Th17-mediated chronic inflammation by IFN-γ-induced, APC-derived IL-1 and IL-23. Our study challenges the dogma that IFN-γ suppresses Th17 and enhances Th1 development.
Inflammation is an active component in autoimmune diseases, infectious diseases, and tumors. Acute inflammation may be evolved toward chronic inflammation in a variety of pathological scenarios. However, the underlying cellular and molecular mechanisms controlling the evolution of inflammation remain poorly understood. IL-17+CD4+ T cells (Th17) are largely found in chronic inflammatory environments and are thought to play an important role in inflammation (1, 2, 3, 4, 5, 6, 7, 8, 9). Th1 cells target naive T cells and inhibit Th17 development through IFN-γ (10, 11, 12). However, memory but not naive T cells predominantly accumulate in the inflammatory environment. Further, Th1 and Th17 cells are often colocalized in pathological environments (1, 2, 3, 4, 5, 6, 7, 8, 9). Altogether, this suggests that kinetic interaction between Th1 and Th17 cells may possibly be involved in regulating the evolution of inflammation in pathological environments.
Materials and Methods
Human immune cell isolation and culture
Treatment of APCs
APCs including MDDCs and CD14+ cells (0.5–1 × 106/ml) were incubated for 0–3 days with IFN-γ (0–100 ng/ml), IL-6 (5 ng/ml), IL-10 (5 ng/ml), and M-CSF (10 ng/ml) (R&D Systems). B7-H1 expression was detected by FACS (13, 14). After incubation with cytokines, APCs were stimulated for 12 h with LPS (1 μg/ml) and then washed and used to stimulate T cell responses. In some cases, APCs were stimulated with LPS for 24 h. Quantification of gene expression in the cultured APCs was performed as we described previously (19, 20). The supernatants were collected for detection of cytokines by ELISA (all from R&D Systems, except IL-23p19 from eBioscience).
T cell culture system
The treated APCs were cultured with T cell subsets (0.5 × 106 T cells/ml) in ratios of 1:1 to 1:10 for 5 days in the presence of anti-CD3 (5 μg/ml) and anti-CD28 (2.5 μg/ml) mAbs (BD Biosciences). In some cases, T cells were labeled with CFSE (22). IL-1β (2.5 ng/ml), IL-23 (10 ng/ml), anti-IL-1 (1 μg/ml anti-IL-1R plus 1 μg/ml anti-IL-1α) (all from R&D Systems), and anti-human B7-H1 (13, 14) were used as indicated. In some cases, before culturing with T cells the APCs were exposed to IFN-γ (100 ng/ml) for 72 h and/or were transfected with plasmids containing human IL-23-specific small interfering RNA (siRNA) (23). T cell phenotype and cytokine profiles were determined by FACS, ELISA, or real-time PCR (19, 20).
The Wilcoxon rank-sum test was used to determine pairwise differences and the χ2 test was used to determine differences between groups. p < 0.05 was considered significant (StatSoft).
IFN-γ abates the Th1 polarization capacity of APCs
Th1 cell-derived IFN-γ targets T cells and NK cells to further enhance Th1 responses (24). In this study we tested the effects of IFN-γ on APC-mediated T cell polarization. We incubated MDDCs or CD14+ cells with IFN-γ and used these IFN-γ-conditioned APCs to stimulate total T cell responses. We observed that T cell IFN-γ was significantly reduced by IFN-γ-conditioned MDDCs, as compared with medium-conditioned MDDCs (Fig. 1,A). Similar results were obtained with IFN-γ-conditioned CD14+ cells. We further sorted naive and memory T cells. We observed that IFN-γ-conditioned MDDCs mediated reduced IFN-γ-expressing T cells from both naive and memory T cells (not shown). This indicates that IFN-γ abates the Th1 polarization capacity of APCs. IFN-γ stimulates the inhibitory B7 family member B7-H1 on human monocytes and tumor cells (13, 14, 15). We showed that B7-H1 expression was dramatically increased on MDDCs and macrophages following IFN-γ treatment (Fig. 1,B). The effects were IFN-γ dose dependent (Fig. 1,C). Anti-human B7-H1 recovered T cell IFN-γ expression stimulated by IFN-γ-conditioned APCs (Fig. 1 D). The data indicate that IFN-γ stimulates B7-H1 expression on APCs and abates the Th1 polarization capacity of APCs through B7-H1.
IFN-γ programs APCs to stimulate Th17 cell development
Exogenous cytokines, including TGFβ and IL-6, induce Th17 cell development (2, 4, 5, 6, 7, 8, 9). To mimic physiologic and pathological environments in vivo, we initially examined the roles of APCs in inducing Th17 cells in the absence of exogenous cytokines. We showed that both macrophages and MDDCs were capable of inducing high levels of T cell IL-17 production, whereas in the absence of APCs the levels of Th17 cells were negligible (Fig. 2,A). Interestingly, macrophages were significantly superior to MDDCs in inducing Th17 cell responses (Fig. 2 A). This indicates that APCs contribute to Th17 cell polarization in humans.
IFN-γ abates the Th1 polarization capacity of APCs through B7-H1 (Fig. 1,D). Because Th1 and Th17 cells are often colocalized in the inflammatory environment (1, 2, 3, 4, 5, 6, 7, 8, 9), we next investigated whether IFN-γ affected the capacity of APCs to induce Th17 cells. To test this possibility, normal blood CD14+ cells were conditioned with IFN-γ and tested for their capacity to induce Th17 cells. IFN-γ profoundly increased the capacity of CD14+ cells to elicit IL-17+ T cells (Fig. 2,B). To determine whether this is a selective feature for IFN-γ, we also treated CD14+ cells with IL-6, IL-10, and M-CSF. We showed that treatment of IL-6, IL-10, and M-CSF had minimal effects on CD14+ cell-mediated Th17 cell development (Fig. 2,B). IFN-γ-treated macrophages or IFN-γ-treated MDDCs resulted in increased secretion of T cell IL-17 (Fig. 2,C). Further, we showed that IFN-γ-treated APCs induced T cell IL-17 production in a dose-dependent manner (Fig. 2 D). The data indicate that in contrast to the effects on Th1 polarization, IFN-γ can selectively condition APCs to promote Th17 cell development.
APCs stimulate memory IL-17+ T cell expansion
APCs induce Th17 cell development (Fig. 2,A). We sorted naive and memory T cells and examined whether APCs stimulate memory Th17 cell expansion or Th17 cell differentiation from naive T cells. Macrophages stimulated minimal numbers of Th17 cells from naive T cells during 5 days of culture (Fig. 3,A), but both MDDCs and macrophages induced Th17 cells and elicited robust production of IL-17 by memory T cells (Fig. 3, A and B). LPS stimulation moderately increased this T cell IL-17 production (Fig. 3 B).
To determine whether APCs induce memory IL-17+ T cell proliferation, we labeled memory T cells with CFSE and stimulated them with macrophages. We observed that the majority of IL-17+ T cells, but not IL-17− T cells, experienced six divisions (Fig. 3,C). Similar to unconditioned APCs (Fig. 3), IFN-γ-conditioned APCs also stimulated memory IL-17+ T cell proliferation rather than Th17 cell differentiation from naive T cells (not shown). Taken together, these results suggest that APCs may maintain a memory Th17 cell pool in the inflammatory environment in humans.
APC-induced memory Th17 cell expansion is IL-1 and IL-23 dependent
We further investigated the mechanism by which APCs induce Th17 cell expansion. We (20) and others have demonstrated that IL-1 and IL-23 are involved in the induction of Th17 cells (1, 2, 3, 4, 5, 6, 7, 8, 9). We confirmed that in the presence of APCs, T cell IL-17 production was enhanced by IL-1 and IL-23 (Fig. 4 A).
We analyzed the expression of IL-1 and IL-23 in macrophages treated with IFN-γ. The expression of IL-12p35 and IL-12p40 transcripts was increased by 2- to 3-fold, and IL-23p19 expression was increased by 50-fold at 72 h (Fig. 4,B). IFN-γ also enhanced the expression of IL-1α and IL-1β by macrophages (Fig. 4,C). Although IFN-γ increased the protein levels of IL-23p19, IL-12p40, and IL-12p70, the levels of IL-23p19 were significantly higher than those of IL-12p70 (Fig. 4 D).
We next examined the role of macrophage-derived IL-1 and IL-23 in the induction of Th17 cell expansion. We used neutralizing Abs blocking the IL-1 signal and siRNA sequences that specifically block human IL-23 (not shown). Separate blockade of IL-1 and IL-23 each reduced T cell IL-17 production induced by macrophages and IFN-γ-conditioned macrophages. Simultaneous blockade of both cytokines abrogated the IL-17 response (Fig. 4,E). Interestingly, blocking B7-H1 with anti-B7-H1 increased IFN-γ+IL-17− T cells and had no significant effect on IFN-γ−IL-17+ T cells and IFN-γ+IL-17+ T cells (Fig. 4 F). Taken together, our data indicate that APCs expand memory Th17 cells through IL-1 and IL-23 in a B7-H1-independent manner, and this effect is further amplified by IFN-γ treatment.
In this report we have explored the effects of IFN-γ on APCs to induce Th1 and Th17 cell development in humans.
We first show that IFN-γ potently stimulates B7-H1 expression on APCs. IFN-γ-conditioned APCs exhibit reduced capacity for Th1 polarization through B7-H1. This data indicates that as an effector cytokine, IFN-γ can down-regulate active immune responses through induction of B7-H1 on APCs. IFN-γ has been shown to stimulate the expression of IDO and arginase on APCs (25, 26). Such APCs could suppress antitumor immune responses. IFN-γ was also found to mediate CD4+ T cell loss and impair secondary antitumor immune responses (27). Indeed, the physiological functions of inhibitory B7-family members are to limit and attenuate T cell responses, by which they prevent T cell hyperactivation and avoid tissue and organ damage during immune responses (16, 17, 18). The data suggest that by inducing suppressive molecules on APCs, IFN-γ may participate in terminating acute inflammation.
In the absence of exogenous cytokines, APCs are capable of inducing human Th17 cells, basically from memory Th17 cell expansion. This effect is dramatically amplified by IFN-γ treatment. This observation may explain why IL-17+ T cells exhibit a memory phenotype and are often found in inflammatory tissues and organs. We demonstrate the mechanisms by which APCs and IFN-γ-conditioned APCs induce memory Th17 cell expansion. APCs induce memory Th17 expansion through IL-1 and IL-23, and IFN-γ further triggers APCs to produce high levels of IL-1 and IL-23, which in turn induce memory Th17 expansion. This effect is B7-H1 independent. Therefore, IFN-γ may play dual roles in regulating the IL-17+ T cell pool as follows: IFN-γ targets APCs to initiate and promote Th17 polarization of memory T cells (this study), whereas it targets naive T cells and suppresses Th17 polarization (1, 2, 3, 5, 28).
In summary, we show that IFN-γ stimulates B7-H1 expression on APCs and abates the Th1 polarization capacity of APCs through B7-H1, whereas IFN-γ can condition APCs to induce Th17 cell expansion from memory T cells through IL-1 and IL-23 in a B7-H1-independent manner. We propose a novel kinetic development and interaction between Th1 and Th17 cells in the course of inflammation as follows: Th1-mediated inflammation is attenuated by IFN-γ-induced B7-H1 and subsequently evolves toward Th17-mediated chronic inflammation by IFN-γ-induced IL-1 and IL-23. Therefore, our study reveals a novel regulatory role for IFN-γ in controlling the evolution of inflammation and challenges the dogma that IFN-γ suppresses Th17 and enhances Th1 development.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was partially supported by National Cancer Institute, National Institutes of Health Grants CA123088 and CA099985 (to W.Z.) and by the Marsha Rivkin Center for Ovarian Cancer Research (to I.K.).
Abbreviations used in this paper: MDDC, monocyte-differentiated dendritic cell; siRNA, small interfering RNA.