Interleukin-17 and interferon-gamma are produced concomitantly by human coronary artery-infiltrating T cells and act synergistically on vascular smooth muscle cells

Abstract

Background: Atherosclerosis is an inflammatory disease in which interferon (IFN)-gamma, the signature cytokine of Th1 cells, plays a central role. We investigated whether interleukin (IL)-17, the signature cytokine of Th17 cells, is also associated with human coronary atherosclerosis.

Methods and results: Circulating IL-17 and IFN-gamma were detected in a subset of patients with coronary atherosclerosis and in referent outpatients of similar age without cardiac disease but not in young healthy individuals. IL-17 plasma levels correlated closely with those of the IL-12/IFN-gamma/CXCL10 cytokine axis but not with known Th17 inducers such as IL-1beta, IL-6, and IL-23. Both IL-17 and IFN-gamma were produced at higher levels by T cells within cultured atherosclerotic coronary arteries after polyclonal activation than within nondiseased vessels. Combinations of proinflammatory cytokines induced IFN-gamma but not IL-17 secretion. Blockade of IFN-gamma signaling increased IL-17 synthesis, whereas neutralization of IL-17 responses decreased IFN-gamma synthesis; production of both cytokines was inhibited by transforming growth factor-beta1. Approximately 10-fold fewer coronary artery-infiltrating T helper cells were IL-17 producers than IFN-gamma producers, and unexpectedly, IL-17/IFN-gamma double producers were readily detectable within the artery wall. Although IL-17 did not modulate the growth or survival of cultured vascular smooth muscle cells, IL-17 interacted cooperatively with IFN-gamma to enhance IL-6, CXCL8, and CXCL10 secretion.

Conclusions: Our findings demonstrate that IL-17 is produced concomitantly with IFN-gamma by coronary artery-infiltrating T cells and that these cytokines act synergistically to induce proinflammatory responses in vascular smooth muscle cells.

Figure 1

Serologic evidence for Th17 and Th1 immune responses in patients with coronary atherosclerosis. (A) IL-17 and (B) IFN-γ plasma levels were measured in patients with symptomatic coronary atherosclerosis (n=108), in referent outpatients without a diagnosis of coronary atherosclerosis (n=59), and in healthy control subjects (n=18). (C) Circulating IL-17 levels were further compared in coronary atherosclerosis patients with lower disease extent scores between 1 and 22 (n=60) vs. higher disease extent scores between 23 and 45 (n=48) or (D) with presenting symptoms of chronic stable angina (n=87) vs. acute coronary syndrome (n=21). (E) Circulating IL-17 levels were also compared in referent outpatients with no current medical illnesses (n=12) vs. current non-cardiac medical illnesses (n=48). Individual data are shown. The proportion of patients with detectable levels of IL-17 and IFN-γ (≥7.8 pg/ml) were compared between groups using Fisher's exact test.

Figure 2

Circulating IL-17 correlated with the systemic IFN-γ cytokine axis in patients with coronary atherosclerosis. (A) IFN-γ, (B) IL-12, (C) CXCL10, (D) IL-1β, (E) IL-6, and (F) IL-23 plasma levels were correlated to those of IL-17 in patients with coronary atherosclerosis (n=108) using the Spearman method (r coefficients and P values). The relationships are also depicted using linear regression (solid line) with 95% confidence bands (interrupted lines).

Figure 3

Coronary artery-infiltrating T cells produced both IL-17 and IFN-γ. Equal length segments of atherosclerotic or macroscopically normal coronary arteries in organ culture were treated with agonistic antibodies (α) to CD3 (0.1 μg/mL) and CD28 (1 μg/mL) for 24-72 hr or no antibodies for 72 hr (control) and supernatant levels of IL-17 (A) and IFN-γ (B) were measured by ELISA. Alternatively, atherosclerotic coronary artery rings were treated with cytokines either alone or in various combinations, including IL-1α (1 ng/mL), IL-6 (30 ng/mL), IL-15 (30 ng/mL), IL-21 (30 ng/ml), IL-23 (30 ng/mL), and TGF-β1 (10 ng/mL) or with the well characterized T cell activators PMA (1 μg/mL) and ionomycin (1 μM) for 48 hr and IL-17 supernatant levels were measured by ELISA (C). The data are means ± SEM (n=6-12) and are pooled from independent experiments using arteries from six different donors. The dotted line represents the assay's lower limit of detection and statistical analyses were not performed for responses to inflammatory cytokines resulting in IL-17 concentrations below this level. *P<0.05 and **P<0.001, treated vs. control and ^#P<0.05 atherosclerotic vs. non-diseased (Repeated Measures ANOVA).

Figure 4

Distinct regulation of coronary artery-infiltrating Th17 and Th1 cells. (A) T cells infiltrating atherosclerotic coronary arteries were stimulated with agonistic antibodies (α) to CD3 (0.1 μg/mL) and CD28 (1 μg/mL) for 48 hr in the presence of neutralizing antibodies to IL-17RA or IFN-γR1 vs. control IgG (10 μg/mL). Additionally, the effects of (B) TGF-β1 (10 ng/mL) or (C) IL-12 (20 ng/mL) plus IL-18 (100 ng/mL) were tested on atherosclerotic coronary arteries in the presence or absence of α-CD3/CD28 for 48 hr. Cytokines levels in the supernatant were measured by ELISA. Untreated coronary artery rings did not produce detectable cytokines (<7.8 pg/mL) and these results are not shown. The data are means±SEM (n=6-12) and are pooled from three independent experiments using arteries from different donors. *P<0.05 and **P<0.001, other treated groups vs. α-CD3/CD28 alone (ANOVA).

Figure 5

A subset of coronary artery-infiltrating T helper cells were IL-17 and IFN-γ double producers. CD4⁺ T cells from atherosclerotic coronary arteries or from peripheral blood of healthy subjects were isolated using antibody-conjugated magnetic beads (A). The cells were stimulated with PMA (1 μg/mL) and ionomycin (1 μM) for 6 hr in the presence of brefeldin A, fixed and permeabilized, labeled with anti-IL-17-PE, anti-IFN-γ-FITC, or fluorescently-labeled, isotype-matched control antibodies, and analyzed by flow cytometry. Composite results (B) and representative histograms of intracellular cytokine staining including that of isotype-matched antibody controls are shown for artery-infiltrating (C) and circulating (D) T helper cells. The data are means±SEM (n=3-4) from independent donors with % IL-17⁺ T cells representing those detected in the upper left quadrant, % IFN-γ⁺ T cells representing those detected in the lower right quadrant, and % IL-17+IFN-γ⁺ T cells representing those detected in the upper right quadrant of the flow cytometry histograms. *P<0.05, blood vs. artery-infiltrating CD4⁺ T cells (t test).

Figure 6

IL-17 and IFN-γ synergistically induced pro-inflammatory cytokine and chemokine production by VSMCs. (A) IL-6, (B) CXCL8, and (C) CXCL10 levels were measured in the supernatant of VSMCs treated with IL-17±IFN-γ at 5, 10, or 20 ng/mL for 48 hr. The data are means±SEM (n=6) and are representative of three independent experiments. The effects of IL-17 and IFN-γ are synergistic as the responses of combined treatment exceed those of either treatment alone at twice the dose.

Figure 7

Interactions between IL-17 and IFN-γ signaling. VSMCs were treated with IL-17±IFN-γ for various times and at different doses. Signaling activity was compared to that of TNF-α at 10 ng/mL for 15 min. Cell lysates were prepared and the expression of IκBα, phospho-p38 MAPK, and phospho-STAT1 were analyzed by immunoblotting and compared to that of β-actin to control for protein loading. The results are representative of three independent experiments from different donors.