Peripheral and islet interleukin-17 pathway activation characterizes human autoimmune diabetes and promotes cytokine-mediated β-cell death

Abstract

Objective: CD4 T-cells secreting interleukin (IL)-17 are implicated in several human autoimmune diseases, but their role in type 1 diabetes has not been defined. To address the relevance of such cells, we examined IL-17 secretion in response to β-cell autoantigens, IL-17A gene expression in islets, and the potential functional consequences of IL-17 release for β-cells.

Research design and methods: Peripheral blood CD4 T-cell responses to β-cell autoantigens (proinsulin, insulinoma-associated protein, and GAD65 peptides) were measured by IL-17 enzyme-linked immunospot assay in patients with new-onset type 1 diabetes (n = 50). mRNA expression of IL-17A and IFNG pathway genes was studied by qRT-PCR using islets obtained from subjects who died 5 days and 10 years after diagnosis of disease, respectively, and from matched control subjects. IL-17 effects on the function of human islets, rat β-cells, and the rat insulinoma cell line INS-1E were examined.

Results: A total of 27 patients (54%) showed IL-17 reactivity to one or more β-cell peptides versus 3 of 30 (10%) control subjects (P = 0.0001). In a single case examined close to diagnosis, islet expression of IL17A, RORC, and IL22 was detected. It is noteworthy that we show that IL-17 mediates significant and reproducible enhancement of IL-1β/interferon (IFN)-γ-induced and tumor necrosis factor (TNF)-α/IFN-γ-induced apoptosis in human islets, rat β-cells, and INS-1E cells, in association with significant upregulation of β-cell IL17RA expression via activation of the transcription factors STAT1 and nuclear factor (NF)-κB.

Conclusions: Circulating IL-17(+) β-cell-specific autoreactive CD4 T-cells are a feature of type 1 diabetes diagnosis. We disclose a novel pathway to β-cell death involving IL-17 and STAT1 and NF-κB, rendering this cytokine a novel disease biomarker and potential therapeutic target.

FIG. 1.

β-Cell autoantigen-specific IL-17 reactivity. A: IL-17 responses to β-cell autoantigens in patients with type 1 diabetes (■; n = 50) are significantly more prevalent than in healthy control subjects (□; n = 30), whereas the prevalence of reactivity to recall antigens is similar. B: For comparison, IFN-γ responses to β-cell autoantigens in patients with type 1 diabetes are also significantly more prevalent than in healthy control subjects, and the prevalence of reactivity to recall antigens is similar (***P < 0.0001). C: Graph shows number of peptides provoking IL-17 production in five patients tested in the original sampling near diagnosis and again a median of 12 months (range 9–14) later. There is a nonsignificant trend for the number of peptides testing positive to decline over time (paired t test; P = 0.09). D: Frequency of responses to individual β-cell autoantigenic peptides among type 1 diabetic patients as measured by production of IL-17 (■) and IFN-γ (□). E: Photomicrographs of wells in which cells have been cultured with negative control (peptide diluent), β-cell autoantigen, and recall stimulus (Pediacel vaccine) for IL-17 (upper row) and IFN-γ (lower row). Spots indicate responder antigen-specific CD4 T-cells. It is noteworthy that the frequency of IFN-γ responder cells for recall antigens is severalfold higher than for IL-17⁺ cells.

FIG. 2.

Real-time quantitative PCR analysis of expression of IL17 (A), IFNG (B), RORC (C), TBX21 (D), and IL22 (E) in islets of Langerhans obtained from two patients with type 1 diabetes (T1D1, who died within 5 days of diagnosis, and T1D2, who died 10 years after diagnosis). Mean (SEM) values (technical replicates) are also shown for three control subjects (organ donors). Levels of target gene mRNA transcripts are normalized to the housekeeping gene HPRT and compared with an in-house calibrator sample as described. HC, healthy controls.

FIG. 3.

Mechanisms through which IL-17 promotes cytokine-induced β-cell apoptosis. A: Human islets were treated with IL-1β (50 units/mL), TNF-α, and IFN-γ (both at 1,000 units/mL) in the presence of 20 ng/mL rIL-17A for 48 h. Apoptosis was evaluated using HO/PI staining. Results are means ± SEM of six to nine independent experiments (***P < 0.001 vs. untreated cells; §§§P < 0.001 vs. IL-1β + IFN-γ–treated cells; §§§§P < 0.001 vs. TNF-α and IFN-γ–treated cells) (ANOVA with Bonferroni correction). B: Human islets were treated with IL-1β (50 units/mL) and IFN-γ (1,000 units/mL) in the presence of 20 ng/mL rIL-17A for 48 h with or without the addition of L-NMA. Apoptosis was evaluated using HO/PI staining. Results are means ± SEM of three independent experiments (***P < 0.001 vs. untreated cells; §§P < 0.01 vs. IL-1β + IFN-γ–treated cells; §§§P < 0.05 vs. IL-1β + IFN-γ + L-NMA–treated cells) (ANOVA with Bonferroni correction). C–F: Panels show representative images of cell death in whole human islets without cytokine (C), islets cultured with IL-17A (20 ng/mL) (D), islets cultured with IL-1β (50 units/mL) + IFN-γ (1,000 units/mL) (E), and islets stimulated with IL-1β (50 units/mL) + IFN-γ (1,000 units/mL) + IL-17A (20 ng/mL) (F). G: Human islets were treated with cytokines as described above for 48 h in the presence of L-NMA, and supernatants were assayed for nitrite content. Nitrite is significantly raised in the presence of IL-1β (50 units/mL) + IFN-γ (1,000 units/mL) and further increased when supplemented with 20 ng/mL rIL-17A. In the presence of L-NMA, nitrite is significantly decreased. Results are means ± SEM of three independent experiments (*P < 0.05 vs. untreated cells; §P < 0.05 vs. IL-1β + IFN-γ + IL-17A–treated cells) (ANOVA with Bonferroni correction). H: The combination of cytokines IL-1β and IFN-γ upregulate IL-17RA mRNA expression in human islets after 24 h. Human islets were treated with IL-1β (50 units/mL) and IFN-γ (1,000 units/mL) for 24 h; IL-17RA mRNA expression was assayed by RT-PCR and normalized for the housekeeping gene β-actin (ACTB); *P < 0.05 vs. untreated (Student t test). Results are means ± SEM of six independent experiments. Results are expressed as fold variation compared with untreated control. I: In contrast, under the same conditions as in H, IL-1β and IFN-γ do not increase IL-17RC mRNA expression in human islets after 24 h. Results are means ± SEM of six independent experiments. Results are expressed as fold variation as compared with untreated control. (A high-quality color representation of this figure is available in the online issue.)

FIG. 4.

IL-17RA expressed in rat insulin-producing INS-1E cells is upregulated by cytokines via STAT1 and NF-κB activation. INS-1E cells were transfected with 30 nmol/L of either a control siRNA (□) or a siRNA targeting STAT1 (■). After 1 day of recovery, cells were left untreated or were treated with IL-1β (10 units/mL) + IFN-γ (100 units/mL) for 24 h. IL-17RA mRNA expression was assayed by RT-PCR and normalized for the housekeeping gene GAPDH. Cytokine treatment significantly upregulates IL-17RA (**P < 0.01 vs. untreated cells; Student t test). siRNA targeting of STAT1 significantly reduces IL-17RA induction (§P < 0.05 vs. control siRNA). Results are means ± SEM of five independent experiments. In another series of experiments, INS-1E cells were infected with an adenovirus encoding a super-repressor IκBα to inhibit NF-κB activity (■) or with adenovirus encoding luciferase as control (□). Cells were left untreated or treated with IL-1β (10 units/mL) + IFN-γ (100 units/mL) for 24 h. Cytokine treatment significantly upregulates IL-17RA (***P < 0.001 and *P < 0.05 vs. relevant untreated cells). Adenovirus-mediated inhibition of NF-κB significantly reduces IL-17RA induction. §P < 0.05 vs. control. Results are means ± SEM of four independent experiments.

FIG. 5.

Schematic representation of possible synergistic effects of cytokines in mediating β-cell apoptosis. We propose that β-cell death could have two phases. In the first, T_H1 cells and macrophages in the local milieu secrete IFN-γ and IL-1β, respectively, which bind to specific receptors leading to the activation of STAT1 and NF-κB transcription factors. Both STAT1 and NF-κB lead to the expression of IL-17R through an as yet unknown mechanism. In the second phase, the inflammatory milieu includes T_H17 cells, which secrete IL-17A. Triggering through IL17RA/RC exacerbates apoptosis initiated by IL-1β and IFN-γ, leading to increased β-cell death.