Reconstitution of retrotranslocation by the Hrd1 ubiquitin ligase with purified components
by Vedran Vasic
Date of Examination:2019-06-27
Date of issue:2020-03-11
Advisor:Dr. Alexander Stein
Referee:Prof. Dr. Peter Rehling
Referee:Prof. Dr. Claudia Steinem
Referee:Prof. Dr. Michael Meinecke
Referee:Dr. Alexis Caspar Faesen
Referee:Prof. Dr. Henning Urlaub
Files in this item
Name:Vedran_Vasic_Dissertation_for_publication.pdf
Size:11.7Mb
Format:PDF
Abstract
English
In eukaryotic organisms, a large fraction of newly-synthesized proteins are destined for the endomembrane system or for secretion. The entry gate for these proteins is the endoplasmic reticulum (ER). Proteins are imported into the ER in an unfolded state, where they fold and assemble into their native conformations before being exported to the Golgi apparatus. Proteins which fail to fold or assemble correctly have a propensity to aggregate in the ER, which is toxic for the cell. Accordingly, a quality control pathway termed ER-associated protein degradation (ERAD) recognizes misfolded proteins and retrotranslocates them across the ER membrane into the cytosol, where they are ubiquitinated and degraded by the proteasome. One of the biggest questions about ERAD is how substrates are retrotranslocated across the ER membrane. Recent evidence implicates one of the central components of ERAD, the Hrd1 ubiquitin ligase, in forming a retrotranslocon. Despite this, it is still unknown if Hrd1 is sufficient for retrotranslocation of luminal ERAD substrates. I created a novel reconstituted system to study retrotranslocation, in which luminal substrates are encapsulated into liposomes and delivered to the luminal side of Hrd1 by SNARE-mediated fusion. The encapsulation was efficient and fusion was shown to mix membrane proteins and luminal contents between liposomes, a technique that has broad applications in membrane protein research. However, retrotranslocation by Hrd1 could not be detected after fusion, which was most likely because of substrate aggregation. In another set of reconstitutions, Hrd1 reconstituted in nanodiscs bound misfolded proteins on its luminal side, while Hrd1 reconstituted in liposomes bound misfolded proteins with high affinity on its cytosolic side upon autoubiquitination. This was dependent on autoubiquitination in the RING domain. Misfolded proteins could be partially released from the cytoplasmic binding site by deubiquitination. Hrd1 was also incorporated into planar lipid bilayers, and was shown to have channel activity and to conduct ions in a voltage-dependent manner. This was dependent on autoubiquitination in its RING domain. Substrate addition stimulated channel gating and opened the pore to diameters sufficient to fit multiple alpha helices. Remarkably, deubiquitination of Hrd1 closed the channel. Overall, this thesis provides strong evidence that Hrd1 forms a protein-conducting channel in the ER membrane. A model is proposed, whereby an affinity gradient between luminal and cytoplasmic binding sites provides the driving force during retrotranslocation.
Keywords: Protein quality control; Endoplasmic reticulum; Ubiquitination; Protein translocation; ERAD; Membrane proteins; Protein folding