Structural characterization of reaction intermediates of HECT-mediated ubiquitin transfer

Abstract

Summary The ubiquitination reaction is a post-translational modification in which the small protein ubiquitin (Ub) is attached to a substrate protein. The attachment of Ub onto a substrate generates a signal for the cell e.g. leading to the degradation of the substrate by the proteasome. Ub fulfils therefore a wide array of cellular functions and regulatory circuits. The Ub reaction is mediated by an enzymatic cascade involving three enzymes called E1, E2 and E3. HECT (Homologues to E6-AP) -type-ubiquitin E3 ligases define the specificity for the substrate protein as well as determine the Ub chain linkage. Hence, HECT ligases control the output signal of the ubiquitination reaction. The HECT domains consist of two lobes. The C-lobe contains the catalytic cysteine (Cys) which forms an intermediate Ub-thioester during the Ub reaction. However, the N-lobe possesses a ubiquitin binding surface (UBS) which is involved in the Ub-reaction by binding the Ub non-covalently. The C-lobe can adopt various orientations with respect to the N-lobe. This flexibility of the C-lobe is essential during the Ub transfer reaction and is also involved in HECT domain inhibition. Mutations in HECT E3s are linked to various human diseases i.e. cancer where HECT E3s either act as oncoproteins or tumor suppressors. Therefore, it is important to understand the mechanistic details and regulation of the Ub transfer mediated by the HECT E3 ligases. Here, I present a method to form disulfides mimicking intermediates of the ubiquitination reaction of HECT domains utilizing chemically activated ubiquitin. With this method it is possible to bypass the highly instable thioester between the catalytic cysteine and the C-terminus of the Ub. This enables the study of otherwise unstable and inaccessible intermediates of the ubiquitination transfer reaction on a structural level by biophysical methods e.g. NMR spectroscopy. With this method in hand, I was able to study the first step of the HECT mediated Ub transfer in which the Ub is bound to the catalytic cysteine of the HECT domain. This enabled the discovery of a beta-sheet augmentation during Ub transfer between the HECT C-lobe and the C-terminus of Ub. Of note, I was able to show for the first time that this beta-sheet augmentation is conserved among the different HECT E3 families. In addition, I could identify a novel auto-inhibition mechanism of the Huwe1 HECT domain in which the non-covalent Ub binding surface (UBS) of the Huwe1 HECT domain is blocked by the Huwe1 HECT domain C-lobe. Moreover, I was able to show that during the thioester formation a large rearrangement in the HECT domain occurs which subsequently allows the binding of Ub to the UBS of Huwe1. Hence, the thioester intermediate adopts an open conformation. Furthermore, I identified residues on the C- and N-lobe interface responsible for the auto-inhibition. Mutations of those residues resolve the auto-inhibition and are able to accelerate the ubiquitination reaction. In contrast, mutations in the UBS impair Huwe1 activity thereby providing further evidence of the importance of the UBS during ubiquitination. The data presented here enable new insights into the mechanism of the Ub-transfer mediated by the HECT E3 ligases and furthermore lead to a better understanding of the Ub reaction.