Guanine nucleotide exchange factor regulation of adherens junction biogenesis and maturation

Abstract

Dynamic regulation of cell-cell contact formation during epithelial tissue biogenesis is critical for many developmental processes, which, when recapitulated in adult tissue, can disrupt cell-cell adhesion and contribute to tumourigenesis. During de novo adherens junction assembly, Rho small GTPases drive a multitude of diverse cellular processes to build strong, functional cell-cell contacts. How E-cadherin signalling is transduced to specify Rho protein signalling is not well characterised. It is thought to involve spatio-temporal activation of Rho family members by their GEFs, but which Rho GEFs contribute to the initial stages of epithelial biogenesis remains to be elucidated. Here I identify activators of Rac and RhoA GTPases that contribute to assembly and stabilisation of cadherin contacts. I find that junction induction modulates EGFR phosphorylation in a profile distinct from the canonical ligand-dependent pattern to activate Rac. Functionally, EGFR but not the GEFs DOCK180 or Trio, is required for junction-induced actin reorganisation, a Rac1-dependent cellular process. RNAi screening shows that multiple Rho GEFs regulate the distribution of E-cadherin and/or F-actin at nascent contacts by two modes of action: negative E-cadherin regulation is correlated with, but positive regulation is independent of, F-actin re-distribution. Furthermore, I find that junction formation induces an early transient wave of RhoA activation, which is attenuated by depletion of the GEFs, ARHGEF10 or Trio. ARHGEF10 and Trio negatively regulate the distinct cellular processes of E-cadherin availability and stability. Finally, I find that, loss of ARHGEF10 leads to increased incorporation of unstable E-cadherin into junctions resulting in global destabilisation of newly-formed contacts. This suggests a fine balance between the amount of cadherin present at junctions and their ability to engage with the cortical cytoskeleton. These results are surprising and suggest that the amount of cadherin at junctions may not directly correlate with stability and they may be regulated by distinct pathways.