Prof. Aaron Ciechanover

A system and structural approach to the analysis of the ubiquitome and the ubiquitin signal.

Introduction:  Covalent modification of proteins by ubiquitin (Ub) and ubiquitin-like (Ubl) proteins serves numerous functions and regulates even more cellular processes.  The best studied function is Ub-mediated protein degradation that regulates, among many processes - cell cycle and division, the level and hence the activity of basically all transcription factors, the activity of cell membrane proteins - among them ion channels and receptors, the generation of peptides for presentation of MHC class I complex, and the removal of denatures/misfolded  proteins - thus maintaining the cellular quality control.  Degradation of proteins by the system is a two-step mechanism.  Initially Ub is conjugated to the target substrate in a reaction catalyzed by the concerted activity of three enzymes, a single Ub-activating enzyme (E1), several Ub-carrier proteins (E2s), and almost 1,000 Ub ligases (E3s), that recognize specifically the myriad substrates of the system (each recognizes obviously a small subset) and tag them with Ub.  The ubiquitinated protein is then degraded to short peptides by the 26S proteasome complex that recognizes almost exclusively ubiquitinated proteins.  The ‘canonical’ proteasomal signal has been a polyUb chain where the Ub moieties are linked to one another via internal lysine-48 of the previous moiety.  However, chains where the other six internal lysine residues of Ub are involved have been described.  Furthermore, monoubiquitated as well as multiply monubiquitinated substrates have also been reported to be degraded by the proteasome, as well as substrates conjugated by mixed Ubl-Ub chains.  Needless to point that almost all these various modes of ubiquitination and modification buy Ubl’s along with novel ones, have been described for the non-proteolytic functions the system.  Thus, decoding of the signal is of fundamental importance in understanding the multitude functions of the system and their aberrations in a broad array of diseases.  Here we shall use a system approach to shed light on some of the aspects of the complexity of the signal.

Experimental plan:  Typically, researchers study the fate of a single ubiquitin system substrate, identify its ligase, characterize its mode recognition, ubiquitination and fate, and study its regulation under varying pathophysiological conditions.  Prime examples for proteins studied this way are p53 and β-catenin.  The assembly in the I-CORE of a team of experts in proteomics, informatics, and structural biology provides us with a unique opportunity to take an integrated view of the system from a distance.  First, in collaboration with Arie Admon and Haim Wolfson we shall use a proteomic approach and an informatic analysis to identify the entire ubiquitome (by identifying proteins that have the GG C-terminal signature of ubiquitin attached to them).  Identification of the GG signature attached to ubiquitin-derived peptides, will allow also quantification of the different ubiquitin chain types generated in the cell.  We shall then ask which of these proteins are degraded following ubiquitination (by analyzing quantitatively the reduction in the level of their peptides during chase of the heavy isotope-labeled amino acid used in the SILAC labeling) and which are stable and ubiquitinated for other purposes.  We shall then monitor the effect of stress, initially starvation and induction of autophagy on the ubiquitome, mostly the degrading ubiquitome.  Informatic analysis will reveal whether there are programmed waves of degradation (or stabilization) of specific proteins following induction of stress, i.e. whether the stability of certain regulators is  first changed which then induces the degradation of the bulk of proteins.  In collaboration with Gideon Schreiber and Arie Admon we shall study the changes that occur in the ubiquitome and proteome following stimulation of cells with IFN, known to induce specific subunits in the proteasome (immunoproteasome) that generate more efficiently peptides that bind to MHC class I complex.  Systematic siRNA-mediated inactivation of the different E3 will enable to map for the first time the repertoire of substrates of the different ligases.  We are aware that the mapping will initially be only partial, that because of possible redundancy among ligases, and also because that different substrates are degraded only under certain conditions/following specific stimuli.  However, this initial mapping will allow, in the future, a more complex manipulation of the E3s, and attainment of a more accurate map of the ligases along with the network of their cognate substrates.

One major hurdle in studies of the ubiquitin system is the lack of well-defined conjugates with ubiquitin or a chain attached in a defined site.  Biologically synthesized conjugates (either in cell free reconstituted systems or in cells) are heterogeneous: they are of different chain length, with ubiquitin attached to several sites along the substrate, and most importantly in “carrier free” amounts.  Recent advances in synthetic organic chemistry methodologies allow us now to synthesize large amounts of well-defined conjugates of proteins of several hundred amino acids (thus far up to 300) to which a chain of ubiquitin (up to four moieties and made of moieties linked to one another via each of the seven internal lysines of ubiquitin) is attached at any point of choice.  The invaluable tool will allow, in the first time, to start and decode the ubiquitin signal.  Why the proteasome preferentially binds lysine-48 based chains and no other internal linkages-based chains? how it recognizes monoubiquitinated substrates? how di-, tri, and tetra ubiquitin bind to the proteasome and to which subunits in the 19S regulatory particle? how multiple monoubiquitinated proteins bind to the proteasome? do they have multiple binding points?  We shall synthesize (in collaboration with Professor Ashraf Brik from Ben Gurion University) a set of elected ubiquitin-protein conjugates (initially α-globin, α-synuclein and p19INK4a) to which ubiquitin will be attached in several modes (mono, multiple mono, and a single chain of di-, tri, and tetraubiquitin, initially of the lysine-48 based chain).  We try to co-crystallize them with Rpn10 (and later with Rpn13) the known proteasomal ubiquitin receptors) and in collaboration with Deborah Fass determine their structure.  Along with Lucio Frydman, and using ultrafast NMR, we shall determine their structure in solution, now using however more physiological conditions, where the binder will be the entire 19S regulatory complex (that will be difficult to use in X-ray crystallography).  Based on the two structural approaches, and along with Haim Wolfson, we shall try to predict the exact structure of the proteasomal binding domains, and whether they have any similarity to known ubiquitin-associated domains (UBAs) or ubiquitin-interacting motifs (UIMs).

Taken together, our more general, system level approach to the ubiquitin pathway, that has become possible due to the assembly of a unique multidisciplinary team in the proposed I-CORE, will allow us to shed light on the dynamic ubiquitome under various physiological  conditions and to decode, for the first time, the complexity and diversity of the ubiquitin signal at both the conjugation and degradation (proteasome) levels.

Names of PhD and Post-Doctoral students in the lab, that are funded by the center:

  1. Ori Braten