Abstract
NFAT (Nuclear factor of activated T cells) transcription factors play important physiological roles in the development and function of many organs, notably in the immune system and nervous system. As a consequence, their dysregulation has been implicated in various human diseases such as cancer, neurodegenerative diseases, and auto-immune diseases. The regulation of NFAT activity by calcium-dependent nuclear-cytoplasmic shuttling has been extensively studied. In contrast, the regulation of NFAT protein level by the ubiquitin-proteasome system is still poorly understood. However, NFATs are short-lived proteins and regulation of their stability is critical for controlling their activity.In a previous study, my group has shown that the E3 ubiquitin-ligase Trim17 binds NFATc3 but does not promote its ubiquitination and rather stabilizes it. Preliminary results suggested that Trim39, a partner of Trim17, might be an E3 ubiquitin-ligase for NFATc3 and that SUMOylation of NFATc3 might modulate its stability. Therefore, the goal of my PhD was to understand the mechanisms through which Trim39, Trim17, and SUMO regulate the stability of NFATc3.During my PhD, I have characterized Trim39 as an E3 ubiquitin-ligase of NFATc3. Indeed, my results indicate that overexpression of Trim39, but not its inactive mutant, induces the ubiquitination of NFATc3 in cells. In contrast, silencing of endogenous Trim39 decreases the ubiquitination level of NFATc3. Recombinant Trim39 directly induces the ubiquitination of NFATc3 in vitro. Moreover, overexpression of Trim39 decreases the protein levels of NFATc3 whereas the silencing of Trim39 increases it. I have also shown that Trim17, which can bind Trim39, inhibits Trim39-mediated ubiquitination of NFATc3, both in cells and in vitro. Trim17 acts by both reducing the intrinsic E3 ubiquitin-ligase activity of Trim39 and by preventing the interaction between NFATc3 and Trim39. Furthermore, I found that a SUMOylation-deficient mutant of NFATc3 is less ubiquitinated and more stable than the wild type NFATc3, suggesting that SUMOylation of NFATc3 is important for its ubiquitination and degradation. Importantly, I identified one SUMO interacting motif (SIM) in the sequence of Trim39 through which Trim39 binds SUMO2 polymers via one of these SIMs. Mutation of this SIM in Trim39 or SUMOylation consensus sites in NFATc3 decreased the interaction between Trim39 and NFATc3, and the ubiquitination of NFATc3 mediated by Trim39. These results strongly suggest that Trim39 binds and ubiquitinates preferentially the SUMOylated forms of NFATc3 and therefore acts as a SUMO-targeted E3 ubiquitin-ligase (STUbL) for NFATc3. Finally, we have measured the impact of these mechanisms on the physiological function of NFATc3. I first found that Trim39 decreases the transcriptional activity of NFATc3. Furthermore, using primary cultures of cerebellar granule neurons as a model, we have shown that the mutation of the SUMOylation sites of NFATc3 and silencing of endogenous Trim39 enhances neuronal apoptosis, probably by stabilizing the NFATc3 protein. Taken together, these data indicate that Trim39 modulates neuronal apoptosis by acting as a STUbL for NFATc3 and by controlling its stability.