Interferon α2 and interferon γ induce the degranulation independent production of VEGF-A and IL-1 receptor antagonist and other mediators from human mast cells

Abstract

Background: Mast cells are resident immune effector cells, often studied in the context of allergic disease. Found in substantial numbers at sites of potential infection they are increased at sites of angiogenesis and can be pivotal for the sensing and clearance of a variety of pathogens. Interferons (IFNs) are cytokines that are critical for host defence against intracellular pathogens. Increased levels of IFNs are observed during viral infection and in autoimmune diseases. IFNs are also widely used therapeutically and have been examined in the therapy of severe asthma.

Objective: To define the selective human mast cell cytokine and chemokine response following activation with type I or type II IFN's.

Methods: The ability of both IFNα2 and IFNγ to induce cytokine production by human cord blood-derived mast cells was examined in vitro. Cytokine and chemokine production at 6 and 24 h was assessed by multiplex protein analysis. Degranulation was assessed by β-hexosaminidase release. Mast cells were also treated with reovirus or respiratory syncytial virus and their production of CXCL10, IL-1 receptor antagonist (IL-1Ra), and vascular endothelial growth factor (VEGF) examined after 24 h.

Results: In addition to increased expression of classical IFN response genes, such as CXCL10, small but significant increases in CCL5 and IL-17 production were observed following IFN activation. Notably, human mast cells produced both VEGF and IL-1Ra in a dose dependent manner. These responses occurred in the absence of mast cell degranulation by a mechanism consistent with classical IFN signaling. Both reovirus and respiratory syncytial virus infection of mast cells, were also associated with IFN-dependent IL-1Ra expression.

Conclusion and clinical relevance: Our findings demonstrate that IFNs have profound impact on cytokine and chemokine expression by human mast cells, alone or in the context of viral infection. Mast cell VEGF and IL-1Ra responses to IFNs could impact the regulation of local inflammatory responses and subsequent tissue remodeling.

Keywords: CXCL10; IL-1Ra; VEGF; interferons; mast cells; reovirus; respiratory syncytial virus.

Figure 1

CBMC respond to type I and type II interferons. IFN response gene expression was assessed following 24 h treatment of CBMC (10⁶/ml) with A) 10 ng/ml (100 IU/ml) IFNγ (n = 10) or B) 5 ng/ml (100 manufacturer U/ml) IFNα2 (n = 5) for 24 h. Data are depicted as fold change over media control. *p < 0.05, **p < 0.01,****p < 0.0001, based on paired t‐test of relative normalized expression.

Figure 2

CBMC secrete a distinct cytokine response following activation with type I or type II IFN. The profile of cytokine production by CBMC (10⁶/ml) following 24 h activation with 100 IU/ml IFNγ (n = 17 for IL‐1Ra; n = 13 for all others) or 100 manufacturer U/ml IFNα2 (n = 13 for IL‐1Ra; n = 9 for all others) was determined using immunoassay. Each line depicts an individual CBMC sample. The limit of detection is depicted by hatched line. For statistical analysis, those samples with undetectable levels were assigned the limit of detection value. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, Wilcoxon matched pairs signed‐rank test.

Figure 3

CBMC secrete a distinct chemokine and growth factor response following activation with type I or type II IFN. The profile of chemokine and growth factor production by CBMC (10⁶/ml) following 24 h activation with 100 U/ml IFNγ (n = 17 for CXCL10, and VEGF; n = 13 for all others) or IFNα2 (n = 13 for CXCL10 and VEGF; n = 9 for all others) was determined using immunoassay. Each line depicts an individual CBMC sample. The limit of detection is depicted by hatched line. For statistical analysis, those samples with undetectable levels were assigned the limit of detection value. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, Wilcoxon matched pairs signed‐rank test.

Figure 4

Type I and type II interferons do not induce short term CBMC degranulation. CBMC (0.5 × 10⁶/ml) in modified HEPES‐Tyrode's buffer were treated with increasing doses of IFNα2 or IFNγ or with the calcium ionophore A23187 (0.5 µM) for 20 min. The level of degranulation was assessed via β‐hexosaminidase release. Data depict the mean ± SEM for three independent experiments using three different CBMC cultures activated in triplicate. ***p < 0.001, paired t‐test.

Figure 5

Dose and kinetic response of CXCL10, VEGF, IL‐1Ra and CXCL9 production by IFN‐activated CBMC. CBMC (10⁶/ml) were treated with increasing doses of A) IFNα2 or B) IFNγ. CXCL10, VEGF and IL‐1Ra production was assayed at 6 h and 24 h by multiplex immunoassay (n = 6) and CXCL9 production was assayed by ELISA (n = 8). *p < 0.05, **p < 0.01, ***p < 0.001, paired t‐test (6 hour) or Dunn's multiple comparisons post test compared to media control (24 h).

Figure 6

Virus‐activated human mast cells secrete IL‐1Ra and VEGF. CBMC (10⁶/ml) were incubated with (A) Live reovirus, UV‐inactivated reovirus and medium control or with (B) Live RSV, UV‐inactivated RSV and medium control, washed and further incubated for 24 h. IL‐1Ra and VEGF production was measured in cell‐free supernatants by multiplex immunoassay. *p < 0.05, **p < 0.01, Dunn's post‐hoc multiple comparisons test. N = 7 independent CBMC cultures for reovirus experiments, N = 6 independent CBMC cultures for RSV experiments.

Figure 7

Effects of type I IFN blockade on RSV induced VEGF‐A and IL‐1Ra production by human CBMC. ^aThe profile of mediator production by CBMC (10⁶/ml) incubated with MOI 3–4 live (RSV) or UV‐inactivated (UV‐RSV) respiratory syncytial virus or with medium control for 24 h in the presence of media control or 5 μg/ml anti‐IFNAR or IgG2a isotype control was determined using immunoassay (N = 6). Differences between groups were assessed by Friedman's test with Dunn's post‐hoc test. *p < 0.05, **p < 0.01 compared to media control.

Figure 8

IFNα2 induction of CXCL10, IL‐1Ra and VEGF‐A is dependent on classical JAK/STAT signalling. CBMC (10⁶/ml) were pre‐treated for 1 h with 2 μM of the PI3 K inhibitor LY 294002, the p38‐MAPK inhibitor SB 203580 or with the JAK/STAT inhibitor JAK inhibitor 1, then incubated for 12 h with 5 ng/ml (equivalent to 100 manufacturer U/ml) IFNα2 or media control. CXCL10, IL‐1Ra and VEGF mRNA production was determined by qRT‐PCR. Data are depicted as percent decrease in fold change over media + vehicle control from IFN treated CBMC. ***p < 0.001, **p < 0.01, *p < 0.05, paired t‐test of relative normalized expression of IFN‐treated CBMC versus inhibitor‐treated CBMC (n = 3).