The ETS transcription factor ERG is constitutively expressed in endothelial cells (EC) and acts as a master-regulator of endothelial function. ERG drives expression of genes which determine endothelial lineage and control homeostasis, such as VE-Cadherin, ICAM-2 and HDAC-6. ERG also represses expression of pro-inflammatory genes such as ICAM-1 and IL-8, through inhibition of NF-kB activity (1), thus acting as a gatekeeper for endothelial activation. Interestingly, ERG expression is down-regulated in EC by pro-inflammatory agents such as TNF-α, suggesting that down-regulation of ERG during inflammation is necessary to allow full NF-kB activation. This is in line with the decrease of ERG expression in the endothelium overlaying the “inflamed” regions of human atherosclerotic plaques (1). In vitro, ERG deletion in EC results in enhanced leukocyte adhesion, whilst over-expression of ERG by adenovirus inhibits TNF-induced leukocyte adhesion in vitro and acute TNF-induced inflammation in mouse (1). To confirm the role of endothelial ERG in controlling vascular inflammation in vivo, we used Tie2-Cre/Ergfl/+ mice (Birdsey et al, under review) in a zymosan-induced peritonitis model. Zymosan injection resulted in enhanced leukocyte infiltration in Tie2-Cre/Ergfl/+ mice compared to littermate controls (Figure 2), confirming that endothelial ERG can negatively control acute inflammation. Thus targetting ERG to promote its anti-inflammatory effects could be beneficial against vascular inflammation.
A novel small molecule ETS inhibitor, YK-4–279, has recently been shown to bind to ERG and disrupt protein-protein interactions (2). By in silico molecular dynamic modelling, we confirmed YK-4–279 binds the putative protein-binding site within the pointed domain of ERG. In vitro, analysis of protein and gene expression following YK-4–279 treatment of HUVEC resulted in upregulation of ICAM-1 and IL-8 expression in an NF-kB dependent manner, in line with ERG siRNA (Figure 1). However, YK-4–279 did not affect the expression of genes transactivated by ERG, ICAM-2 and HCAD-6, suggesting that ERG’s transactivation and repression functions can be separated. In vivo , i.p. injection of YK-4–279 caused exacerbated leukocyte infiltration into the peritoneum under basal conditions and after Zymosan challenge.
Together, these findings confirm that ERG acts to prevent vascular inflammation in vitro and in vivo, and that compounds can be developed which specifically target ERG’s anti-inflammatory activity. Therefore, the development of ERG mimetic molecules to restore ERG’s anti-inflammatory activity may be a novel therapeutic approach to reducing vascular inflammation.