dsRNA activation of endothelin-1 and markers of vascular activation in endothelial cells and fibroblasts

Abstract

Background: In patients with systemic sclerosis (SSc), the relationship between innate immune activation, represented by increased expression of interferon (IFN)-regulated genes, and vascular injury/activation, manifest by increased endothelin-1 (ET-1), endothelin converting enzyme-1 (ECE1) and intercellular adhesion molecule-1, is uncertain.

Objective: To investigate the potential roles of innate immune ligands in both these pathogenic pathways.

Methods: The effect of known Toll-like receptor (TLR) ligands was tested in vitro on dermal microvascular and pulmonary arterial endothelial cells, and on dermal fibroblasts cultured from healthy controls and patients with SSc. To test the effect of double-stranded RNA (dsRNA) on vascular activation/injury in vivo, polyinosinic/polycytidylic acid (poly(I:C)) was administered continuously over 7 days by subcutaneous osmotic pump.

Results: dsRNA/poly(I:C), but not other TLR ligands, highly stimulated ET-1 protein and mRNA (EDN1), as well as intercellular adhesion molecule-1 (ICAM-1) and IFN-regulated MX2, by endothelial cells and dermal fibroblasts. Poly(I:C) induced EDN1, ECE1, and ICAM-1 mRNA expression in poly(I:C) treated skin. Poly(I:C)-induced EDN1, ECE1 and MX2 was not blocked in mice with the type I IFN receptor deleted. However, poly(I:C)-induced EDN1 and ECE1, but not poly(I:C)-induced ICAM-1 expression was blocked in mice with the TLR3 signalling protein TRIF/TICAM-1 deleted.

Conclusion: Together these data show that the dsRNA can regulate genes associated with vascular activation, as seen in SSc, that type I IFNs do not mediate these effects, and that EDN1 and ECE1 but not ICAM-1 activation is mediated by TLR3.

Figure 1. Regulation of Endothelin 1 and ICAM1 expression in skin from patients with dcSSc and by TLR ligands on human dermal microvascular endothelial cells

Panels a and b: mRNA expression of EDN1 (n=25) and MX1 (n=36) in lesional skin from dcSSc (SSc) patients and in control skin (n=6). Fold-changes shown on the graph are normalized to mRNA expression by one of the healthy controls. The average fold change of EDN1 and MX1 in SSc skin (13.05 ± SE 2.56 and 2.89 ± SE 0.33, respectively) compared to the average fold-change in control, healthy skin (2.62 ± SE 1.67 and 0.98 ± SE 0.29, respectively) was increased for EDN1 (4.98-fold increase, p<0.05) and MX1 (2.95-fold increase, p<0.05). Panels c-e: Human dermal microvascular endothelial cells (HDMEC) treated with the TLR ligands: poly(I:C), LPS, sspolyU or CpGA in duplicate wells, as described in the methods, were analyzed for EDN1 (panel c), MX2 (panel d) and ICAM-1 (panel e) expression. Expression in each case was normalized to one of the control, untreated wells. Panel f: HDMECs were treated with bafilomycin (30 nM), poly(I:C) or both and analyzed for EDN1 expression.

Figure 2. Regulation of Endothelin-1 and ICAM1 expression by TLR ligands in human pulmonary artery endothelial cells

Human pulmonary endothelial cells treated with the TLR ligands: poly(I:C), LPS, and CpGC or CpGA as indicated in duplicate wells as described in the methods were analyzed for EDN1 (panel a) and ICAM-1 (panel b) expression. Results shown are from two independent experiments. Expression in each case was normalized to one of the control, untreated wells, * indicate p<0.05 compared to media control.

Figure 3. Induction of Endothelin-1 by TLR ligands in dermal fibroblasts from patients with dcSSc and healthy controls

Panel a: Dermal fibroblasts from healthy controls (△) or SSc patients ( ) were treated with TLR ligands, TGFβ, IFN α, IFN β or IFNγ as described in the methods and END1 mRNA analyzed by qRT-PCR. Fold-change shown is normalized to mRNA expression by the corresponding unstimulated cells. The average fold-change is represented by a horizontal line ± SE. Compared to untreated cells, poly(I:C) (TLR3 ligand) stimulated 34.3-fold increase in EDN1 expression, (p<0.0001). TGFβ, and IFNγ also induced, respectively, 7.8 (p<0.01) and 10.4-fold (p<0.05) increase in EDN1 expression. Panel b: Bioactive 21-aa ET-1 peptide in normal (△) and in SSc ( ) dermal fibroblasts by Poly(I:C) (p<0.0001) stimulation for 24h treatment. ET-1 protein was measured by ELISA in the supernatants from SSc and normal fibroblast cultures. The average protein concentration for each group is represented as a bar ± SE.

Figure 4. In vivo effect of poly(I:C) on EDN1, ECE1 and ICAM-1 expression

Expression of EDN1 (a) and ECE1 (b), and ICAM-1 (c) by RT-PCR analysis of skin mRNA from C57BI/6 WT (n=10), C57BI/6/IFNAR−/− (n=8) and C57BI/6 TICAM−/− (n=10) mice one week after subcutaneous insertion of osmotic pumps containing poly(I:C) as described in methods. Fold-change shown in the graphs is normalized to mRNA expression by one of the control mice. Results presented are means ± SE and are representative of four independent experiments; * p<0.05; ** p<0.01.