β-catenin function is determined co-translationally by its partners

Overview & Synopsis

Many proteins participate in biochemically distinct complexes that have unrelated functions. A well-known example is β-catenin, which can either bridge E-cadherin to the actin cytoskeleton at adherent junctions, or act as the main transcription factor of the Wnt pathway. In the absence of Wnt signaling, the β-catenin protein localized at adherent junctions is stable whereas the cytoplasmic one is rapidly bound and degraded by the co-called destruction complex, composed of APC, Axin1 and GSK3β. In presence of Wnt however, the destruction complex is inactivated and the cytoplasmic β-catenin is stabilized and can enter the nucleus to activate transcription of target genes. We performed a large-scale smFISH screen and it revealed that β-catenin mRNAs have an unusual cytoplasmic localization: they accumulate in foci that are not P-bodies or stress granules. These mRNA foci contain the destruction complex and nascent β-catenin protein, and they dissolve upon activation of the Wnt pathway. They thus function as regulated translation factories that promote the co-translational degradation of β-catenin. Immunoprecipitation of APC and Axin1 followed by RNA sequencing confirmed that the destruction complex binds β-catenin mRNAs. These interactions were disrupted by translation inhibition, indicating that APC and Axin1 bind nascent β-catenin protein rather than directly its mRNA. In situ, β-catenin mRNA foci were dissolved by translation inhibition. This suggests that they form by polysome condensation, induced by the multivalent interactions that APC makes with nascent β-catenin. The previous data demonstrate that β-catenin degradation is determined while it is synthesized by the ribosome, but they do not explain how a pool of β-catenin escapes degradation to reach E-cadherin at the plasma membrane. We hypothesized that cells might have a second pool of β-catenin mRNA that would be specialized in the synthesis of plasma membrane β-catenin. Indeed, immunoprecipitation of E-cadherin followed by RNA sequencing revealed an association with β-catenin mRNA, which was also translation dependent. Moreover, labeling experiments of β-catenin mRNA with APC and E-cadherin showed that some mRNAs co-localized with APC in the cytosol, while others localized at the plasma membrane with E-cadherin. These data show that by virtue of distinct co-translational interactions, β-catenin mRNA generates two types of polysomes that segregate spatially in the cell, to synthesize identical proteins that will have different fates and functions. We refer to this novel mechanism as polysome sorting. Our genome-wide experiments suggest that several mRNAs could be subjected to polysome sorting, providing an original mechanism to regulate multifunctional proteins.