REST (repressor-element-1 silencing transcription factor) is a transcriptional repressor that is associated with the maintainenance of neuron stem cell self-renewal and has been characterized as an oncogene in neuronal stem cells and, paradoxically, as a tumour suppressor in epithelial tissues. How REST accomplishes these activities and how it is incorporated into cell signalling pathways is unclear. Two papers published in Nature, from Westbrook and colleagues, and Guardavaccaro and colleagues, report a novel pathway linking REST to the ubiquitin-proteasome system and cell-cycle regulation with implications for this dual activity of REST.

REST expression decreases as stem cells differentiate to become neural stem and progenitor cells, and this was previously shown to result from a threefold decrease in the half-life of REST, indicating that it is post-translationally regulated. Consequently, Westbrook and colleagues found that REST is K48 polyubiquitinated (which leads to proteasomal degradation) and this was dependent on cullin 1, a component of the SCF ubiquitin ligase complex. SCF ubiquitin ligases are targeted to specific groups of substrates by F-box proteins and Westbrook and colleagues and Guardavaccaro and colleagues both identified βTRCP1 (also known as BTRC) and βTRCP2 (also known as FBXW11) — collectively known as βTRCP owing to their similarity — as the F-box proteins that were responsible for REST polyubiquitylation and subsequent proteasomal degradation. Interestingly, Guardavaccaro and colleagues also showed that REST exhibits cell-cycle-dependent expression, wherein it is degraded in G2 and mitosis in a βTRCP-proteasome-dependent manner.

βTRCP, like other F-box proteins, recognizes and binds its targets through serine/threonine phosphorylation of a specific sequence (the phosphodegron). Westbrook and colleagues identified a phosphodegron towards the carboxyl terminus of REST and this was required for βTRCP binding and degradation. They also showed that a REST frame shift (REST-FS) mutant from a colon cancer cell line no longer expressed the phosphodegron and thus was no longer bound by βTRCP, which led to a more stable protein that was expressed at higher levels.

So, what is the relevance of REST turnover? Westbrook and colleagues showed that overexpression of βTRCP decreased REST expression and increased anchorage-independent growth of breast epithelial cells. Furthermore, they showed that βTRCP-mediated transformation was impaired when REST expression was restored, demonstrating that REST can function as a tumour suppressor and is targeted by the oncogenic activity of βTRCP. Guardavaccaro and colleagues investigated the oscillatory expression of REST through the cell cycle. They found that MAD2 (also known as MAD2L1) expression increased as REST was degraded in G2 and M phases and that REST binds to a consensus sequence in the MAD2 promoter. MAD2 is a crucial component of the spindle checkpoint machinery that serves to prevent progression through mitosis until sister chromatids are bound by kinetochore microtubules of the mitotic spindle. Consistently, Guardavaccaro and colleagues showed that ectopic expression of a REST phosphodegron mutant (which cannot be degraded) leads to increased chromosome instability, characterized by lagging chromosomes, chromosome bridges and premature separation of sister chromatids. This was also the case for the REST-FS colon cancer cells, indicating that REST also exhibits oncogenic activity, which leads to chromosome instability.

This cell-cycle-dependent tumorigenicity is not unique: βTRCP has numerous targets, which have either oncogenic activity (for example, CDC25A, β-catenin and EMI1) or tumour suppressor activity (for example, IκB, PDCD4 and WEE1), although the outcome appears to be tissue-type-specific. Therefore, these data raise the question of how we might effectively target such double-faced proteins for therapeutic intervention.