Inhibitors of endocytosis prevent Wnt/Wingless signalling by reducing the level of basal β-catenin/Armadillo.
Bottom Line: Here, we use specific inhibitors (Dynasore and Dyngo-4a) to confirm the essential role of endocytosis in Wnt/Wingless signalling in human and Drosophila cells.Moreover, we show that activation of signalling through chemical blockade of GSK3β is prevented by endocytosis inhibitors, suggesting that endocytosis impacts on Wnt/Wingless signalling downstream of the ligand-receptor complex.We propose that, through an unknown mechanism, endocytosis boosts the resting pool of β-catenin upon which GSK3β normally acts.
Affiliation: MRC's National Institute for Medical Research, The Ridgeway, Mill Hill, London NW71AA, UK.Show MeSH
Mentions: Wingless signalling is activated by preventing GSK3β from phosphorylating β-catenin. Indeed, chemical inhibitors of GSK3β trigger signalling, even in the absence of formation of a Wnt–receptor complex. For example, LiCl, which has been known for a while to block GSK3β activity (Stambolic et al., 1996), causes marked accumulation of β-catenin in L cells, and this is suppressed by concomitant treatment with monodansylcadaverine, chlorpromazine or hypertonic sucrose (Blitzer and Nusse, 2006). This provided an early indication that endocytosis could modulate signalling at or below the level of GSK3β, i.e. downstream of the ligand–receptor complex. This finding is at odds with the sequestration hypothesis; therefore, we sought to confirm it with more specific inhibitors. As an inhibitor of GSK3β, we chose SB-216763, which activates TOPFlash in the absence of exogenous Wnt in human cells (Coghlan et al., 2000). We found that SB-216763 activates TOPFlash in Drosophila S2R+ cells in a similar manner and therefore used this compound for subsequent investigation [the effect of other potent GSK3 inhibitors, such as CHIR98014 (Naujok et al., 2014; Ring et al., 2003) was not investigated]. We found earlier that TOPFlash activation with SB-216763 was prevented by concomitant treatment with Dynasore or Dyngo-4a (Fig. 1A). We conclude that inhibitors of endocytosis are unlikely to block Wingless signalling through preventing internalisation of the ligand receptor complex. Instead, it appears that endocytosis impacts on a more downstream event. The most immediate downstream sign of Wnt/Wingless signalling is the accumulation of β-catenin/Armadillo. We therefore investigated the effects of Dynasore or Dyngo-4a on the level of Armadillo in S2R+ cells. As shown in Fig. 4A, a 30 minute pre-treatment with either drug prevented the levels of Armadillo from rising in response to Wingless-conditioned medium or to SB-216763. In fact, it appeared that both drugs caused the level of Armadillo to dip below that seen in otherwise unstimulated cells, whereas the amounts of Actin and Syntaxin remained unchanged. This suggests that endocytosis could be required to maintain a sufficient steady-state level of Armadillo in resting cells. Indeed, addition of Dyngo-4a to S2R+ cells without any treatment to stimulate signalling led to a marked decrease in Armadillo (Fig. 4B). Interestingly, this effect was reversible – in cells that had been treated for 2 hours with Dyngo-4a and then allowed to recover in normal culture medium, Armadillo levels rose back to normal (Fig. 4B), suggesting that the effect of Dyngo-4a is transient and that cell viability was not adversely affected.
Affiliation: MRC's National Institute for Medical Research, The Ridgeway, Mill Hill, London NW71AA, UK.