STAT4 is required for IL-23 responsiveness in Th17 memory cells and NKT cells

Abstract

STAT4 is a critical mediator of inflammatory immunity and is required for all known IL-12 biological responses, including the induction of IFN-γ and development of Th1 cells. We demonstrate that IL-23, an IL-12-related cytokine, also requires STAT4 for optimal IL-17 secretion from memory T helper cells and NKT cells. Although IL-23 stimulation had modest effects on STAT4 activation, STAT4-deficiency results in reduced Il23r expression. These data demonstrate a restricted requirement for STAT4 in innate and adaptive IL-17-secreting T cell responses that might contribute to inflammatory immunity.

Abbreviations

Alum

aluminum hydroxide

IBD

inflammatory bowel disease

LTi-like

lymphoid tissue inducer-like

NKT

natural killer T

OVA

ovalbumin

RA

rheumatoid arthritis

SLE

systemic lupus erythematosus

Th

T helper

Introduction

The Signal Transducer and Activator of Transcription molecule, STAT4, is critical in responses to the cytokine IL-12 in human and mouse T cells.^1-3 STAT4 is required for the induction of IFN-γ and promotes the differentiation of T helper 1 cells. Mice that are deficient in STAT4 have impaired inflammatory immunity becoming resistant to the development of T cell-mediated autoimmunity, but susceptible to infection with intracellular pathogens.⁴ Thus, the IL-12-STAT4 signaling pathway regulates specific components of immune responses.

The role of STAT4 in immunity is not restricted to IL-12 signaling. Type I IFNs activate STAT4 when expression of STAT1 decreases following initial viral infection.⁵ IL-23 binds to a heterodimeric receptor composed of the IL-12Rβ1 and IL-23R subunits and activates STAT4 though not as efficiently as STAT3.^6-8 We previously demonstrated that IL-23-induced IL-17 production was diminished in STAT4-deficient T cells.⁹ In those experiments, performed early in the examination of IL-17-producing T cells, STAT4-dependency was observed using IL-23 stimulation of total CD4⁺ T cells. Thus, it was not clear if specific cell populations were responding. In this report, we define populations of T cells that are STAT4-dependent in responses to IL-23, and present evidence for potential roles of STAT4 in mediating IL-23 responsiveness.

Results

We previously demonstrated that STAT4-deficient CD4⁺ T cells cultured in vitro with IL-23 display reduced IL-17 production compared to wild type cells, in response to re-stimulation with IL-23 and IL-18 or IL-1β, cytokines known to synergize in the induction of IL-17.⁹ To define the STAT4-dependent IL-17-secreting T cell populations, we assessed the development and function of sorted populations from wild type and Stat4^−/− spleen. We first determined if STAT4 is important for the development of a subset of known IL-17-secreting lymphocytes by assessing cell numbers in spleen. Splenic γδ T cell and natural killer T (NKT) cell numbers were similar between control and STAT4-deficient mice, however STAT4-deficient mice displayed significantly increased splenic memory CD4⁺ T cell numbers, with a trend toward reduced splenic memory CCR6⁺CD4⁺ T cell numbers in comparison to wild type mice (Fig. 1A-D). We further found a similar frequency of proliferative and apoptotic in vivo splenic memory CD4⁺ T cells and memory CCR6⁺CD4⁺ T cells between wild type and STAT4-deficient mice (data not shown).

Figure 1.

Reduced IL-23-stimulated IL-17 production in the absence of STAT4. (A-D), Splenocytes from WT and Stat4^−/− mice were stained with antibodies and cell populations were identified as follows: memory Th cells (CD4⁺CD62L⁻CD44^hi), CCR6⁺ memory Th cells (CCR6⁺CD4⁺CD62L⁻CD44^hi), γδ T cells (γδ TCR⁺), NKT cells (TCR-β⁺CD1d-Tetramer⁺) and NK1.1⁻ NKT cells (TCR-β⁺CD1d-Tetramer⁺NK1.1⁻). Representative flow cytometry dot plots are shown (A-C). The cell number of the indicated lymphocyte populations was determined in the spleens of WT and Stat4^−/− mice (D). (E), Total memory Th cells or memory Th cells based on CCR6 expression were sorted by flow cytometry from the spleens of WT and Stat4^−/− mice, stimulated with IL-23 and IL-1β for 48 h, cell-free supernatants were collected, and IL-17A production was determined by ELISA. (F), γδ T cells were sorted from the spleens of WT and Stat4^−/− mice and stimulated and analyzed as in E. (G), NKT cells were sorted based on NK1.1 expression from the spleens of WT and Stat4^−/− mice and stimulated and analyzed as in (E). The data are the mean ± SEM of 3–4 mice (D, E- first bar graph, F), the mean ± SD of replicates from an individual assay performed with cells pooled from 3–4 mice (E- second bar graph) or mean ± SEM of 3 independent experiments with IL-17A secretion normalized to WT NK1.1⁻ levels (G). Statistical analysis was performed using the Student's t test. *p < 0.05, compared with WT samples.

To define the requirement for STAT4 in IL-23-induced IL-17 production from memory T helper cells, total memory CD4⁺ T cells were sorted from the spleen of wild type and STAT4-deficient mice and stimulated ex vivo for 48 hours in the presence of IL-23 and IL-1β before IL-17 production was analyzed. STAT4-deficient memory T helper cells displayed significantly less cytokine-induced IL-17 production in comparison to control cells (Fig. 1E). As we observed a trend toward lower memory CCR6⁺ T helper cell numbers in the spleen of STAT4-deficient mice in comparison to wild type mice (2.5×10⁵ ± 0.3 memory CCR6⁺ T helper cells in the spleen of STAT4-deficient mice versus 3.5×10⁵ ± 0.4 from WT mice), it is possible that decreased cytokine-induced IL-17 production from total STAT4-deficient memory T helper cells might be due to reduced memory CCR6⁺ T helper cells within the total T helper memory cell pool. However, sorted CCR6⁺ memory T helper cells stimulated with IL-23 and IL-1β displayed reduced IL-17 production in the absence of STAT4, similar to that observed from total memory T helper cells (Fig. 1E). Thus, STAT4 contributes to IL-23-stimulated IL-17 production from memory T helper cells.

To determine if STAT4 is required for IL-23-dependent IL-17 production from innate IL-17-producing lymphocytes, we sorted γδ T cells, lymphoid tissue inducer-like (LTi-like) cells (CD4⁺CD3⁻CD11c⁻B220⁻), and NKT cells from the spleen of wild type and STAT4-deficient mice and stimulated them ex vivo. Similar levels of IL-17 were produced from wild type and STAT4-deficient γδ T cells and LTi-like cells in response to stimulation with IL-23 and IL-1β (Fig. 1F and data not shown). As NK1.1⁻ NKT cells are the dominant IL-17-producing NKT cell population,^10,11 we further sorted NKT cells based on NK1.1 expression. Upon stimulation with IL-23 and IL-1β, STAT4-deficient NKT cells secreted lower levels of IL-17 in comparison to control cells (Fig. 1G). Thus, STAT4 is required for optimal IL-23-dependent IL-17 production from NKT cells.

To begin to define the mechanism for STAT4 function in IL-17-producing cells, we tested whether IL-23 activated STAT4 in memory T helper cell populations. We have routinely observed that less than 5 percent of total memory T helper cells from the spleen of naïve wild type mice produce IL-17 (data not shown). To increase the memory T helper cell pool, we sensitized wild type mice with ovalbumin (OVA) adsorbed to aluminum hydroxide (alum). Seven weeks after the second OVA and alum sensitization, CD4⁺ T cells were isolated from the spleen, stimulated with cytokines and pSTAT activation was analyzed in memory T helper cells. As expected, IL-23 induced STAT3 activation in IL-17A⁺ memory T helper cells (Fig. 2A). However, while we observed IL-12-induced STAT4 activation in the IL-17A⁻ memory T helper population, IL-23 did not activate STAT4 in IL-17A⁺ memory T helper cells (Fig. 2A). Furthermore, NKT cells sorted from naïve wild type mice displayed IL-12-induced but not IL-23-induced STAT4 activation (Fig. 2B). To explore this further, and to try to increase the sensitivity for detecting STAT4 activation, we stimulated in vitro-derived Th17 cells in the presence or absence of IL-12 or IL-23, and performed immunoblots on nuclear extracts from these cells. Using this approach, we detected an increase of pSTAT4 following IL-23 stimulation, albeit far less than observed following IL-12 stimulation (Fig. 2C). We also observed non-phosphorylated STAT4 in the nucleus, even in the absence of stimulation, although this was not altered by IL-23 stimulation (Fig. 2C). Taken together, these data suggest that although we observe STAT4-dependent IL-23-induced IL-17 production from memory T helper cells and NKT cells, IL-23 has very modest effects on STAT4 activation in these cell populations.

Figure 2.

IL-23-induced STAT4 activation in memory T helper cells and NKT cells. (A), WT mice were sensitized i.p. with OVA and alum on days 0 and 7 and were sacrificed 7 weeks after the last sensitization. CD4⁺ T cells were sorted by MACS bead selection from the spleens of allergen-exposed mice and were stimulated with PMA and ionomycin for 5 h. GolgiPlug was added after 2 h of stimulation with or without the addition of IL-23 (50 ng/ml) or IL-12 (5 ng/ml). IL-17A-secreting cells were labeled and cells were fixed and permeabilized. Cells were stained for CD4, CD44 and pSTAT3 or pSTAT4. pSTAT was detected in the IL-17A⁺ or IL-17A⁻ memory Th cell population (CD4⁺CD44⁺). (B), Total NKT cells (TCR-β⁺CD1d-Tetramer⁺) were sorted by flow cytometry from the spleens of naïve WT mice, left unstimulated or stimulated with IL-23 (50 ng/ml) or IL-12 (5 ng/ml) for 3 h, and stained for pSTAT4. (C), In vitro-derived Th17 cells from WT mice were stimulated with IL-12 or IL-23 for 1 h before nuclear extracts were used for immunoblotting of the indicated proteins. IRF-4 was used as a nuclear loading control and was not detected in corresponding cytoplasmic extracts. Numbers indicate the ratio of pSTAT4/total STAT4, normalized to IRF-4 expression. The data are representative of 3–5 mice (A, C) or are representative experiments (B).

To define how STAT4-deficiency results in diminished cellular responses to IL-23, we analyzed the expression of IL-23R and IL-12Rβ1 in sorted memory T helper cells and NKT cells isolated from the spleen of wild type and STAT4-deficient mice. Whereas memory T helper cells from wild type and STAT4-deficient mice displayed similar levels of Il12rb1 expression, significantly reduced ll23r expression was observed in memory T helper cells from STAT4-deficient mice. Memory T helper cells from STAT4-deficient mice also expressed lower levels of Rorc and Ccr6 in comparison to control cells (Fig. 3A). As STAT4-deficient memory T helper cells displayed reduced expression of Th17-associated genes, we next wanted to determine if these cells had adopted the phenotype of other T helper subsets in the absence of STAT4. We observed similar expression of the Th2-associated genes, Gata3 and Ccr8, along with the Th1-associated genes, Tbx21 and Cxcr3, in control and STAT4-deficient memory T helper cells, suggesting that in the absence of STAT4, memory T helper cells do not adopt a Th1 or Th2 phenotype in naïve mice (Fig. 3A). We then performed a similar analysis with NKT cells. Whereas wild type and STAT4-deficient NKT cells expressed similar levels of Rorc, Tbx21 and Il12rb1, STAT4-deficient NKT cells displayed significantly reduced levels of Il23r in comparison to control cells (Fig. 3B). Taken together, these data illustrate that in the absence of STAT4, memory T helper cells and NKT cells have reduced expression of IL-23R.

Figure 3.

Reduced expression of Th17-associated genes in the absence of STAT4. (A-B), Memory Th cells (CD4⁺CD62L⁻CD44^hi) (A) or NKT cells (TCR-β⁺CD1d-Tetramer⁺) (B) were sorted from the spleens of WT and Stat4^−/− mice. RNA was isolated from unstimulated cells. Expression of the indicated genes was measured using quantitative PCR; samples were normalized to the expression of β2-microglobulin mRNA and are relative to levels in WT cells. The data are the mean ± SEM of 3 mice (A), or the mean ± SEM of 3 independent experiments (B). Statistical analysis was performed using the Student's t test. *p < 0.05, compared with WT samples.

Given diminished Il23r expression in NKT cells in the absence of STAT4, we hypothesized that IL-12-STAT4 signaling could lead to the induction of IL-23R expression. To examine this, we sorted NKT cells from the spleen of wild type mice and left them unstimulated or stimulated them with IL-23 or IL-12 for 12 hours. Whereas IL-23 stimulation resulted in the induction of Il23r expression, IL-12 did not induce Il23r expression in NKT cells, but did induce Il18r1 (Fig. 4). These data suggest that the reduced IL-23R expression and IL-23-stimulated IL-17 production observed in memory T helper cells and NKT cells from STAT4-deficient mice is likely an indirect effect of STAT4 in another cell type that ultimately leads to regulation of the Il23r gene.

Figure 4.

Cytokine-induced IL-23R expression in NKT cells. NKT cells (TCR-β⁺CD1d-Tetramer⁺) were sorted by flow cytometry from the spleens of WT mice and left unstimulated or stimulated with IL-23 (10 ng/ml) or IL-12 (5 ng/ml) for 18 h. RNA was isolated and the indicated genes were measured using quantitative PCR; samples were normalized to the expression of β2-microglobulin mRNA and are relative to levels in unstimulated cells. The data are the mean ± SD of replicates from an individual assay performed with cells pooled from 3–4 mice and are representative experiments.

Discussion

STAT4 is required for the development of inflammatory immunity, although how it contributes to this process is still not entirely clear. Th1-dependent immunity depends upon STAT4, as do most IL-12-stimulated responses. We previously demonstrated that IL-23-stimulated IL-17 production was also compromised in the absence of STAT4, though it was unclear if this was a direct or indirect effect on specific cell populations. In this report, we define memory T helper cells and NKT cells as the populations that require STAT4 for cytokine-induced IL-17 production.

IL-23 is important for the maintenance of Th17 cells and is critical for the development of a pathogenic Th17 phenotype.^12-14 It is also able to induce rapid cytokine production, particularly in combination with IL-1 family cytokines.^9,15-17 IL-23R is not expressed on naïve CD4⁺ T cells, but is induced by the STAT3-activating cytokines, IL-6, IL-21 and IL-23, during Th17 cell differentiation.^8,18,19 In contrast, IL-23R is constitutively expressed on IL-17-producing innate immune cells, enabling rapid IL-17 production from these cell populations in response to IL-23 stimulation.^17,20 In our studies, memory T helper cells and NKT cells displayed diminished IL-23-induced IL-17 production in the absence of STAT4, and this correlated with decreased expression of Il23r. There are several possible mechanisms for the observed phenotypes. First, there could be direct activation of STAT4 by IL-23 leading to increased IL-17 and/or IL-23R expression. However, we did not detect phospho-STAT4 by intracellular staining in NKT cells or Th17 memory cells following IL-23 stimulation, and only modest activation of STAT4 by IL-23 of in vitro-derived Th17 cells was detected by immunoblot. Second, there could be indirect effects of other STAT4-activating cytokines on IL-23R expression. We hypothesized that IL-12 or type l IFN-STAT4 signals may be required for optimal IL-23R expression and IL-23-dependent IL-17 production from these populations. Consistent with this hypothesis, NKT cells isolated from mice deficient in the p35 subunit of IL-12, which is not shared with IL-23, displayed reduced Il23r and IL-23-induced IL-17 production (data not shown), similar to NKT cells from STAT4-deficient mice. However, neither IL-12 nor IFN-α (data not shown) could directly induce Il23r from wild type NKT cells, suggesting the possibility of a T cell-extrinsic role for STAT4 in mediating IL-23 responsiveness. Finally, it is possible that STAT4 has a non-canonical function in T cells.²¹ Non-phosphorylated STAT proteins have important functions in chromatin regulation and it is possible, even in the absence of STAT4 phosphorylation by relevant cytokines, that STAT4 could be regulating gene expression. Indeed, even in unstimulated Th17 cells, we observed significant STAT4 in the nucleus. These possibilities are not mutually exclusive and further work will help to distinguish the relative roles of these distinct mechanisms.

STAT4 and the IL-23-Th17 pathway have also been implicated in the development of mouse and human autoimmune and inflammatory diseases, such as inflammatory bowel disease (IBD), psoriasis, rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), allergy and asthma.^4,22 Genome-wide association studies have implicated STAT4 in diseases such as asthma, IBD, RA and SLE ^23-25 and IL23R in IBD, psoriasis and RA.^26-29 Our data provide a link between a T cell-extrinsic role for STAT4 in mediating IL-23 responsiveness to memory T helper and NKT cells, and IL-17 production. It will be important to define the specific contributions of these T cell subsets to inflammation in these diseases to more fully understand the contribution of STAT4.

Materials and Methods

Mice

The generation of Stat4^−/− mice was described previously.¹ Mice were used on a C57BL/6 or BALB/c background for analysis of innate immune cells or memory T helper cells, respectively. C57BL/6 and BALB/c mice were purchased from Harlan Laboratories. Mice were kept in pathogen-free conditions and all studies were approved by the Indiana University School of Medicine Animal Care and Use Committee.

Surface staining and cell purification

For surface staining, splenocytes were stained for CD4 (BD Biosciences, 552775), CD44 (BD Biosciences, 553134), CD62L (eBioscience, 11-0621-85), CCR6 (BioLegend, 129814), γδ TCR (BD Biosciences, 553178), TCRβ (BD Biosciences, 553171), PBS57-loaded CD1d-Tetramer and/or NK1.1 (BD Biosciences, 557391) for 30 min at 4°C, following incubation with FC Block (BD Biosciences, 553142) for 10 min. Cells were incubated with 7-AAD viability die (eBioscience, 00-6993-50) for 10 min on ice and live memory Th cells (CD4⁺CD62L⁻CD44^hi), CCR6⁺ memory Th cells (CCR6⁺CD4⁺CD62L⁻CD44^hi), γδ T cells (γδ TCR⁺), NKT cells (TCR-β⁺CD1-d-Tetramer⁺) and NK1.1⁻ NKT cells (TCR-β⁺CD1-d-Tetramer⁺NK1.1⁻) were analyzed by flow cytometry with a LSR ll (Becton Dickinson) or sorted using a Reflection sorter (iCyt). Flow cytometry results were analyzed with FlowJo. The PBS-57 (α-galactosylceramide analog)-loaded CD1d-tetramer was obtained from the NIH Tetramer Core Facilities (Emory University).

ELISA

Sorted cells were stimulated with IL-23 (20 ng/ml; R&D Systems, 1887-ML) and IL-1β (25 ng/ml; eBioscience, 14-8012) for 48 h at 37°C. Cell supernatants were collected and cytokines were analyzed by ELISA (BD Bioscience, 555068 and 555067).

Quantitative RT-PCR

Quantitative PCR was performed with sorted cells using TaqMan assays as previously described.³⁰

pSTAT Analysis

For the memory Th cell analysis, wild type mice were administered OVA (20 μg; Sigma-Aldrich, A5503) and alum (2 mg; Sigma-Aldrich, A8222) via intraperitoneal (i.p.) injection on days 0 and 7 of the protocol and mice were sacrificed 7 weeks after the last sensitization. Splenic CD4⁺ T cells were sorted via MACS bead selection (Miltenyi Biotec, 130-095-248), stimulated for 5 h with PMA (50 ng/ml; Sigma-Aldrich, P8139) and ionomycin (500 ng/ml; Sigma-Aldrich, I3909) at 37°C. After 2 h of stimulation, GolgiPlug (BD Bioscience, 555029) was added with or without the addition of IL-23 (50 ng/ml) or IL-12 (5 ng/ml, PeproTech, 210-12) for the remaining 3 h of stimulation. IL-17A-secreting cells were labeled using the mouse IL-17 secretion assay cell enrichment and detection kit (Miltenyi Biotec, 130-094-213). Cells were fixed for 10 min with 2% formaldehyde at room temperature and permeabilized for 15 min at 4°C with 100% methanol. Cells were then stained for CD4 (BD Biosciences, 552775, CD44 (BD Biosciences, 560568) and pSTAT3 (BD Biosciences, 557815) or pSTAT4 (BD Biosciences, 558137) for 30 min at room temperature. pSTAT expression was detected in the IL-17A⁺ or IL-17A⁻ CD4⁺CD44^hi cell population. For the NKT cell analysis, NKT cells were sorted from the spleens of naïve WT mice and left unstimulated or stimulated with IL-23 (50 ng/ml) or IL-12 (5 ng/ml) for 3 h at 37°C. Cells were fixed and permeabilized as mentioned above and stained for pSTAT4 for 30 min at room temperature. Cells were analyzed by flow cytometry with a LSR ll and results were analyzed with FlowJo. For the in vitro-derived Th17 cell immunoblot analysis, splenic naive CD4⁺ T cells were sorted via MACS bead selection (Miltenyi Biotec, 130-093-227) and cultured under Th17 conditions for 5 d as previously described.³¹ Th17 cells were left unstimulated or stimulated with IL-23 (50 ng/ml) or IL-12 (5 ng/ml) for 1 h at 37°C. Cytoplasmic and nuclear lysates were extracted from Th17 cells according to the manufacturer's instructions (Thermo Scientific, 78833) and immunoblotted with pSTAT4 (Cell Signaling Technology, 5267), STAT4 (BD Biosciences, 610926) or IRF-4 (Santa Cruz Biotechnology, sc-6059) as a control.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Funding

This work was supported by Public Health Service grants AI045515 to MHK. NLG-B was supported by PHS T32HL007910. Support provided by the HB Wells Center was in part from the Riley Children's Foundation.