Triple-negative breast cancer (TNBC) is an aggressive disease subtype with no current targeted systemic therapies due to a lack of actionable molecular targets. Efforts to identify therapeutic targets in TNBC have revealed that Fn14, a cytokine receptor, is over-expressed in most invasive breast cancers (1, 2) but not in healthy breast tissue. Fn14 has been shown to activate canonical NF-κB signalling (3), which in turn, transactivates the Fn14 promoter (3) leading to an autoregulatory loop that may drive invasive capacity (2) in breast cancer. We hypothesised that repression of the Fn14/NF-κB feedback loop may reduce Fn14 expression and invasive capacity in breast cancer.
As corticosteroids are known to attenuate NF-κB activity through glucocorticoid receptor alpha (GR⍺), we postulated that corticosteroids might impact Fn14 expression. We investigated this likelihood in three TNBC cell lines positive for GR⍺ and Fn14 (BT-549, MDA-MB-436 and SUM149PT). Following 24h treatment with the corticosteroid, dexamethasone (DEX), we observed a decrease in Fn14 protein expression. In parallel, DEX reduced invasive capacity in vitro by an average of 40% with the most significant and consistent response in BT-549 cells.
The DEX-mediated reduction in Fn14 expression and invasive capacity are likely due to a repressive relationship between GR⍺ and NF-κB. However, the potential mechanisms of repression are complex and poorly understood in breast cancer. To investigate this mechanism, we first used immunocytochemistry to visualise the spatial location of GR⍺ and NF-κB (p65) in BT-549 cells after 24h of DEX. Our results showed rapid nuclear translocation of GR⍺ and subtle nuclear exclusion of p65. We quantified this observation using western blotting over a 24h time course with DEX using subcellular fractionation of BT-549 cells. These data suggest that a repressive physical interaction between GR⍺ and p65 is unlikely, due to the opposite movement of transcription factors between cellular compartments. Interestingly, IκB⍺, an inhibitor of NF-κB, exhibited a subtle DEX-mediated increase in protein expression over 24h, corresponding with the cytoplasmic retention of p65. Our data, therefore, supports a model that repression of NF-κB signalling may be achieved through de novo NF-κB inhibition.
Additionally, BT-549 cells transfected with an NF-κB-response element/luciferase reporter plasmid showed an intriguing increase in luciferase activity following 24h of DEX. It was later shown that this may have been caused by DEX-induced glucocorticoid resistance due to the downregulation of GR⍺ after DEX treatment.
These results demonstrate a novel mechanism for Fn14-mediated invasion. However, investigation in additional cell lines is required. The induction of DEX-mediated glucocorticoid resistance also questions the clinical significance of glucocorticoid based therapy in the treatment of TNBC.
