Our group is interested in the study of epithelial to mesenchymal transition (EMT), a process that occurs during embryo development as well as in tumour cells. EMT is characterized by the loss of epithelial markers such as E-cadherin and by the gaining of mesenchymal markers such as fibronectin. Previous studies of our group have demonstrated that the expression of the Snail1 transcription factor is sufficient to trigger EMT during tumour progression.
Snail1 is a highly unstable protein, due to rapid degradation by the proteasome after ubiquitination. Generally, ubiquitination requires the participation of various enzymes: E1 or ubiquitin-activating enzymes, E2 or ubiquitin-conjugating enzymes and E3 or ubiquitin ligase enzymes. Our group has characterised the FBXL14 F-Box protein as an essential part of an E3 ubiquitination complex. FBXL14 has a cytoplasmic localisation and is capable of polyubiquitinating Snail1 in lysines 98, 137 and 146. The mRNA levels of FBXL14 decrease during hypoxia both in cellular models and in colon adenocarcinomas, causing a stabilisation of Snail1 and the induction of EMT in these systems (Viñas-Castells et al, 2010).
Our most recent investigation indicates that an elevated proportion of Snail1 is also ubiquitinated in the nucleus. Using a screening with a library of shRNAs we have characterised FBXL5 as the E3 ligase that ubiquitinates Snail1 in the nucleus. Snail1 and FBXL5 bind through the C-terminal region of Snail1, facilitating its degradation. FBXL5 is found in the nucleus and is capable of polyubiquitinating Snail1 in lysines K85, K146 and K234. It is interesting to note that the K234 of Snail1 is found in its C-terminal, a region that contains domains with zinc fingers necessary for binding to DNA. A new unexpected function for FBXL5 is its capacity to modulate the levels of Snail1 bound to chromatin. The ubiquitination of Snail1's C-terminal probably decreases its capacity to bind to DNA. Curiously, the degradation of Snail1 by FBXL5 continues to occur in the cytoplasm, although ubiquitination takes place in the nucleus. It is therefore probable that ubiquitination by FBXL5 facilitates the export of Snail1 and its subsequent degradation in the cytoplasm. The protein FBXL5 binds iron in order to be functional and stable. The elimination of the iron or treatment with gamma radiation decreases the levels of FBXL5, stabilising Snail1 in these conditions (Viñas-Castells et al, 2014).
Our group is currently attempting to characterise two new mechanisms of stabilisation of Snail1. The first model includes the participation of enzymes that eliminate Snail1's polyubiquitin chains (deubiquitinases). These enzymes are an interesting therapeutic target in tumours that overexpress Snail1 as their pharmacological inhibition would decrease the levels of Snail1 and they would therefore block EMT. The second model that we studied was phosphorylation by GSK3β as a potent inducer of its degradation. Previous results indicated an important role of AKT2 in EMT (Villagrasa P; 2012). Our research into Notch-activated endothelial cells indicates that AKT2 plays a prominent role in the inactivation of GSK3β, resulting in an increase of Snail1 levels.