Systemic sclerosis dermal fibroblast-derived exosomes trigger a type i interferon response in keratinocytes, through tank binding kinase 1

Abstract Background/Aims Activation of Type I IFN response has been shown to be present in blood and tissue in systemic sclerosis (SSc) and correlates with disease pathogenesis and therapy response. It is currently unknown whether the tissue-specific Type I IFN activation is a consequence of the response observed in blood or rather its source. SSc fibroblasts share many cellular and molecular features with cancer-associated fibroblasts (CAF), and there is a wealth of evidence supporting the role of CAF in inducing Type I IFN activation in epithelial cancer cells. Here, we aimed to determine whether keratinocytes in SSc show evidence of Type I IFN activation and the potential role of SSc in the process. Methods Skin biopsies were obtained from healthy and SSc patients’ forearms and processed for immunohistochemistry and subculture of primary fibroblasts and keratinocytes. The human skin keratinocyte cell line (HaCats) were stimulated with conditioned media from healthy (n = 3) and SSc (n = 3) fibroblasts. Exosomes were isolated from healthy and SSc patient fibroblast cultured supernatant by a combination of ultracentrifugation and polymer-based precipitation and used to stimulate HaCats. Cellular exosomes were investigated by immunoblotting and electron microscopy. TANK-binding kinase (TBK) was inhibited using a small molecule inhibitor (GSK8612). RT-PCR based IFN array and western blot for pSTAT1 (Tyr701) were employed to assess Type I IFN activation in these cells. Results Consistent with previous studies we found high levels of pSTAT1 in keratinocytes from SSc patient biopsies. Contrary to observations in dermal fibroblasts, primary keratinocytes isolated from SSc skin biopsies lost IFN activation once cultured in-vitro. Conversely, conditioned media from SSc patient fibroblasts triggered ISG induction; MX1 (2.7-fold, p < 0.05), CXCL10 (2.8-fold, p < 0.01) and CXCL11 (1.9-fold, p < 0.05). Media fractionation by ultracentrifugation indicated that the effect on keratinocytes was driven mainly by pelleted micro-vesicles, suggesting a potential role for cellular exosomes. Fibroblasts from SSc patients shed cellular exosomes of similar quantities and biophysical qualities, compared to healthy. SSc exosomes induced a 2.2-fold increase in phosphorylation of STAT1 (p < 0.05), as well as an induction in ISGs; MX1 (1.45-fold, p < 0.05), CXCL10 (1.84-fold, p < 0.01), CXCL11 (2.11-fold, p < 0.01) and OAS (1.24-fold, p < 0.05). Suppression of TBK activity by 10μM of GSK8612 suppressed the induction of pSTAT1. Consistent with these findings IFN array of HaCats stimulated with SSc exosomes showed significant (p < 0.05) upregulation of 16 ISGs by at least 1.2-fold compared to the effect of healthy control exosomes. Conclusion We have shown for the first time that scleroderma keratinocytes ex-vivo IFN activation is lost in-vivo and can be regained following exposure to fibroblast-derived exosomes though a TBK-dependent mechanism. Our data indicates that exosomes from SSc fibroblasts may carry the “signal zero” of local Type I IFN activation, potentially though activation of toll-like receptors and/or pattern recognition receptors. Disclosure J.A. Bryon: None. C. Wasson: None. R.L. Ross: None. E. Zeqiraj: None. F. Del Galdo: Consultancies; FDG received consultancies and research support from Abbvie, AstraZeneca, Boheringer-Ingelheim, Capella, Chemomab, Ergomed, Janssen, Kymab, Mitsubishi-Tanabe.. Grants/research support; FDG received consultancies and research support from Abbvie, AstraZeneca, Boheringer-Ingelheim, Capella, Chemomab, Ergomed, Janssen, Kymab, Mitsubishi-Tanabe.