Differential effects of interferon-ß1b on cytokine patterns of CD4+ and CD8+ T cells derived from RRMS and PPMS patients

Introduction

Interferon-β (IFN-β) treatment exerts a proven beneficial effect in relapsing–remitting MS (RRMS) but not primary progressive MS (PPMS).^1–2 IFN-β is a pleiotropic cytokine altering the regulation of several hundred genes in peripheral blood mononuclear cells (PBMCs). ³ Functional in vitro and ex vivo studies have demonstrated that IFN-β interferes with T-cell activation and migration while data on alterations in cytokine patterns remain contradictory.^4–6

We designed this study to elucidate whether the effects of IFN-β on cytokine profiles are dependent on lymphocyte subtype and disease course. Therefore, cytokine secretion patterns of CD4+ and CD8+ T cells derived from different donor groups (healthy controls, RRMS, and PPMS) were analyzed in a highly standardized assay.

Materials and methods

Nineteen RRMS (age 37 [19–54], Expanded Disability Status Scale [EDSS] 1.5 [0.0–4.0]), nine PPMS patients (age 43 [32–61], EDSS 4.0 [3.5–8.0]) and 10 healthy controls (age 33 [24–56]) (median [range]) were recruited. Patients were clinically stable and naïve to immunomodulatory and immunosuppressive treatments, and had not received steroids during the 30 days prior to blood sampling. The study was approved by the local ethics committee. Participants provided written informed consent.

PBMCs were isolated by standard Ficoll gradient centrifugation. CD4+ or CD8+ T-lymphocytes were isolated using the Dynabead and Detach-a-bead kit (Dynal, Oslo, Norway). The purity of T-lymphocyte subsets was >80%, as assessed by Fluorescence-activated cell sorter (FACS).

Cell activation was induced via the T cell receptor (TCR) by CD3/CD28 Dynabeads (Dynal, Oslo, Norway) in the absence or presence of increasing concentrations (100 IU/ml to 10,000 IU/ml) of IFN-β1b (Betaferon®, Bayer Schering Pharma, Berlin, Germany). In some experiments, Phytohemagglutinin (PHA) (1 µg/ml) was used instead of CD3/CD28 Dynabeads. For each condition, a control experiment was performed with non-stimulated cells.

Proliferation was determined by a standard 72-h ³H-thymidine incorporation assay.

Interleukin (IL)-4, IL-5, IL-10 and tumor necrosis factor (TNF)-α were determined by human TH1/TH2 BD^TM CBA (Becton Dickinson, San Diego, CA) on a Becton Dickinson FACScan. IFN-γ and IL-17A were measured by enzyme-linked immunosorbent assay (ELISA) (Biosource Europe S.A., Nivelles, Belgium and Natu Tec, eBiosience, Frankfurt/Main, Germany). Cytokine concentrations were calculated as [cytokine concentration ^{(with stimulants)}] − [cytokine concentration ^{(without stimulants)}].

Statistical analysis was performed for paired observations using Graph Pad Prism 5.0 (Graph Pad Software, La Jolla, CA, USA). Friedmann tests with Dunn’s post tests were applied to analyze the data since the majority did not show a Gaussian distribution.

Results

Proliferation was inhibited by IFN-β1b. In TCR-stimulated cells, a mild dose-dependent effect was observed. Proliferation in PHA-stimulated cells was already impaired at the lowest IFN-β1b concentration. To minimize toxic effects of IFN-β1b on cytokine patterns, all subsequent analyses were performed with the more physiological anti-CD3/ anti-CD28 stimulation.

In the absence of IFN-β1b, cytokine secretion by PPMS-derived cells differed from controls. PBMC from PPMS secreted more IL-4 (67 [17.5–126.7] vs. 0 [0–5] pg/ml; p < 0.001), more IL-5 (304 [122–580] vs. 40 [28–84] pg/ml; p < 0.001) and more IL-10 (54 [34–79] vs. 0 [0–7] pg/ml; p < 0.001). CD8+ T-cells released less IFN-γ (0 [0–95] vs. 180 [97–396] pg/ml; p < 0.05: (median [interquartile range]). No differences were observed between healthy controls and RRMS-derived cells (data not shown).

IFN-β1b inhibited TNF-α secretion in PBMCs from all donor groups (healthy p = 0.0002; RRMS p < 0.0001; PPMS p = 0.0014; Figure 1a–c). TNF-α production by isolated T cells was detectable but not altered by IFN-β1b (data not shown).

OPEN IN VIEWER

Figure 1.

Effect of IFN-β1b on TNF-α, IFN-γ and IL-10 production.

Starting at a concentration of 100 U/ml, IFN-β1b inhibited IFN-γ, the prototypical TH1 type cytokine, in CD4+ T cells derived from RRMS patients (p = 0.0007, Figure 1e), but not PPMS patients (p > 0.05, Figure 1f). In CD4+ T cells from healthy controls, IFN-β1b induced a trend towards lower IFN-γ secretion (p = 0.0734, Figure 1d). In CD8+ T cells from healthy donors, a dose-dependent increase of IFN-γ was observed (p = 0.0123). CD8+ T cells derived from RRMS (p = 0.0947) and PPMS (p = 0.1285) patients exhibited a similar trend. In contrast, IL-17A remained unchanged (data not shown).

A dose-dependent induction of IL-10 by IFN-γ was observed in all cell types from RRMS patients (PBMCs p < 0.0001; CD4+ T cells p < 0.0001; CD8+ T cells p < 0.0001) and in CD4+ T cells from healthy controls (p = 0.0006), but not in PPMS-derived cells (Figure 1j–r).

IFN-β did not induce significant changes in IL-4, the prototypical TH2 type cytokine, with the exception that 10,000 IU/ml of IFN-β1b induced IL-4 in CD4+ T cells derived from RRMS patients (p = 0.0005; Figure 2a–c).

OPEN IN VIEWER

Figure 2.

Effect of IFN-β1b on IL-4 and IL-5 production.

One of the most consistent findings was the inhibition of IL-5 by IFN-β; this was most pronounced in PBMC but also seen in all other cell types investigated (Figure 2d–l).

Discussion

The current study, which applied a highly standardized in vitro assay, demonstrates distinct IFN-β1b effects on cytokine induction, depending on immune cell type and donor group. This is in line with recent reports demonstrating cell type-specific activation of signal cascades by IFN-β and activation pathway-dependent responses to IFN-β treatment.^7–8

In general, CD8+ T cells appeared less susceptible than CD4+ T cells to IFN-β1b. Also, TNF-α release from PBMC, but not isolated T cells, was inhibited by IFN-β1b. Together with observations in PBMCs of IFN-β treated MS patients, ⁹ our results suggest that the inhibitory effect of IFN-β on TNF-γ release is not directly exerted on T cells, but requires other cells included in PBMCs.

The inhibition of IL-5 release was unexpected because IL-5 is considered a TH2-type cytokine. However, enhanced IL-5 production was seen in T cells from patients with active MRI lesions, suggesting a role for IL-5 in the immunopathogenesis of MS. ¹⁰ Future studies are required to clarify the role of IL-5 in autoimmunity.

For some cytokines critical in MS, the alterations induced by IFN-β were related to the MS type. In CD4+ T cells, IL-10, a cytokine considered beneficial in MS, was induced in RRMS and controls, but not in PPMS. The inhibition of IFN-γ, a cytokine considered detrimental in MS, ⁶ was solely observed in RRMS-derived CD4+ T cells. These observations are intriguing as they correlate with clinical observations that IFN-β has beneficial effects on the disease course in RRMS but not PPMS.