Old and new concepts in Ubiquitin and NEDD8 recognition

Abstract

Post-translational modifications by ubiquitin and ubiquitin-like proteins (Ubls) have known roles in a myriad of cellular processes. Ubiquitin- and Ubl-binding domains transmit the information conferred by these post-translational modifications by recognizing functional surfaces and, when present, different chain structures. Numerous domains binding to ubiquitin have been characterized and their structures solved. Analogously, motifs selectively interacting with SUMO (small ubiquitin-like modifier) have been identified in several proteins and their role in SUMOdependent processes investigated. On the other hand, proteins that specifically recognize other Ubl modifications are known only in a few cases. The high sequence identity between NEDD8 and ubiquitin has made the identification of specific NEDD8-binding domains further complicated due to the promiscuity in the recognition by several ubiquitin-binding domains. Two evolutionarily related domains, called CUBAN (cullin-binding domain associating with NEDD8) and CoCUN (cousin of CUBAN), have been recently described. The CUBAN binds monomeric NEDD8 and neddylated cullins, but it also interacts with di-ubiquitin chains. Conversely, the CoCUN domain only binds ubiquitin. CUBAN and CoCUN provide an intriguing example of how nature solved the issue of promiscuity versus selectivity in the recognition of these two highly related molecules. The structural information available to date suggests that the ancestor of CUBAN and CoCUN was a three-helix bundle domain that diversified in KHNYN (KH and NYN domain-containing) and N4BP1 (NEDD4-binding protein-1) by acquiring different features. Indeed, these domains diverged towards two recognition modes, that recall respectively the electrostatic interaction utilized by the E3-ligase RBX1/2 in the interaction with NEDD8, and the hydrophobic features described in the recognition of ubiquitin by CUE (coupling ubiquitin conjugation to ER degradation) domains. Intriguingly, CUBAN and CoCUN domains are only found in KHNYN and N4BP1, respectively, both proteins belonging to the PRORP family whose members are characterized by the combination of protein modules involved in RNA metabolism with domains mediating ubiquitin/NEDD8 recognition. This review recapitulates the current knowledge and recent findings of CUBAN and CoCUN domains and the proteins containing them.