|Ph.D Student||Gurevich Sylvia|
|Subject||Rub1, a Ubiquitin-Like Modifier, is also a Ubiquitin|
|Department||Department of Biology||Supervisors||Professor Michael Glickman|
|Dr. Gunnar Dittmar|
|Full Thesis text|
Ubiquitin is the best characterized member of a family of Ubiquitin-like proteins that covalently modify cellular proteins to alter their fate. Modification of target proteins with ubiquitin has been reported to function in essential cellular processes such as protein degradation, endocytosis and DNA repair.
Of all ubiquitin-like family members, Rub1 (Related to Ubiquitin 1) is the closest kin of ubiquitin, but despite these profound similarities, Rub1 and ubiquitin are different proteins and signal for different outcomes. In contrast to ubiquitin that targets thousands of proteins, Rub1 only has a handle of substrates, members of the Cullin ubiquitin ligases. Moreover, in contrast to ubiquitin that can polymerize, Rub1 is found only as a monomeric modifier on the Cullins. Surprisingly, most of the features understood in ubiquitin are conserved in Rub1: the hydrophobic patch, by which most of the ubiquitin binding partners recognize ubiquitin, as well as most of the Lysines, with which ubiquitin is polymerized. Their 3D structures are even superimposable.
Both ubiquitin and Rub1 are conjugated to substrates by an ensemble of enzymes including their own activating, conjugating and ligating enzymes.
The strong similarity between Rub1 and ubiquitin in term of structure, contrasting with their marked differences in physiological/cellular outcome, is perplexing.
To understand which properties of Rub1 differentiate it from ubiquitin, we engineered Rub1 to modify targets that are normally ubiquitinated, and ubiquitin to conjugate to Rub1 targets. The resulting perturbations served as an innovative tool to understand the manner by which each of these small protein modifiers is recognized, discriminated or handled, and how they signal for their respective unique outcomes.
In this thesis, I report how a single residue at position 72 out of their 76 amino acids (Thr in Rub1 versus Arg in ubiquitin) is critical for discrimination in-vivo between Rub1 and ubiquitin by their respective activating enzymes. A single site mutation at position 72 transforms ubiquitin to conjugate to Cullins replacing Rub1. However, this ubiquitin-modified form of Cullin1/Cdc53 is turned-over by the proteasome. Blocking polymerization of this 'rubylised' ubiquitin restores the typical "Rub1" modification pattern of Cullin1/Cdc53 and rescues Δrub1 phenotypes, so we conclude that Rub1 does not polymerize on Cullins and can be replaced by a non-polymerisable ubiquitin.
Moreover the comparable ubiquitinized form of Rub1 is able to enter the ubiquitinome independently of the rubylation enzymes. Mixing of Rub1 into polyubiquitin modifications perturbed, comparably to a non-polymerisable ubiquitin, pathways driven by Lysine63-linked polyubiquitin chains such as endocytosis and protein sorting processes. Based on these observations it seems that Rub1 has the properties of a monomeric ubiquitin.
These findings suggest that the ubiquitination and rubylation enzymatic pathways are highly specific for their cognate protein, and are the key to their discrimination, but once Rub1 or ubiquitin are channeled into their cognate pathways, they largely overlap in their associations and mode of recognition.
In my PhD I showed for the first time that Rub1 is not only a ubiquitin-like modifier, but is also a ubiquitin modifier, entering in the ubiquitinome and capping preferentially the K63 ubiquitin chains.