PROTEASOME-ADAPTORS · Adapting proteasomes to Proteotoxicity

Adapting proteasomes to Proteotoxicity

Project context and objectives The project's objectives were defined as the following: 1. the role of P97/valosin-containing protein (VCP) interaction with arsenite-inducible regulatory particle-associated protein like (AIRAPL); 2. identifying the poly-ubiquitinated proteins associated with AIRAPL; 3. establishing the role of AIRAPL in endoplasmic-reticulum-associated protein degradation (ERAD). Project results 1. Initial attempts to reveal the role of AIRAPL as an adaptor for P97 binding to proteasomes under protein misfolding conditions revealed that this interaction is AIRAPL independent, as knock-down of AIRAPL did not impair P97 binding to proteasomes. Nevertheless, as this discovery is a novel detection of a new ATPase bound to proteasomes during protein misfolding conditions, we have further analysed this interaction. In a set of experiments we found that P97 binds the 19S particle and not the 20S although a similar HbYX motif resides in the C-terminus of P97 as do the 19S ATPases. Using peptide stimulation of 20S particle activity combined with deletion analysis indicated the lack of P97 C-terminus dependency for proteasome binding. Cross-linking experiments followed by double affinity purification of P97 proteasome particles indicate a direct binding of P97 to the molecular weight corresponding to 26S ATPase subunits. In vivo and in vitro binding experiments of AIRAPL to P97 have mapped the binding sites in both interactors and have placed AIRAPL binding domain in amino acids 140-170 and the N-terminal domain (NTD) of P97 as being the binding site to AIRAPL. In agreement with these results, in vitro competition experiments revealed displacement of AIRAPL using various other known P97 co-factors that bind to the NTD site (Npl4-Ufd1 and P47). The results summarising the P97 proteasome interaction have been recently published in the Journal of Biological Chemistry (see attached manuscript) and the European Research Council (ERC) contribution has been acknowledged. 2. In an initial attempt to characterise the binding of AIRAPL to poly-ubiquitin, the minimal domain of AIRAPL that binds poly-ubiquitin was defined as amino acids 160-240. Mass spectrometry identified a unique feature of AIRAPL that enables it to specifically bind K48 poly-ubiquitin chains. As only two proteins in the human proteome contain a double tandem ubiquitin-interacting motif (UIM), we are currently analysing the properties that enable such specificity. Our current understanding using point mutations and extending the distances between the UIMs is that the double-sided UIM possesses intrinsic properties that specify K48 poly ubiquitin chains. Further evaluations of other double-sided UIM-containing proteins will be required to validate the hypothesis that double-sided UIM specify K48 poly-ubiquitin chain binding. In collaboration with Prof. Soichi Wakatsuki we have solved the 3D structure of AIRAPL bound to tri-ubiquitin K48. The revealed structure clearly confirms our in vivo data and explains at atomic resolution how ubiquitin specificity is achieved by double UIM domains. 3. To investigate the role of AIRAPL in ERAD processes we have carried out a mass spectrometry screen for AIRAPL interactors. We have identified in addition to P97 also UFDX8 and Npl4-Ufd1 as AIRAPL binding partners. Since these P97 co-factors have been reported to specify P97 to ERAD processes, these findings further strengthened our hypothesis that AIRAPL is involved in ERAD processes. To validate this notion we have found that AIRAPL can be co-precipitated with Derlin-1, a specific bona fide ERAD component. All in all our results indicate AIRAPL as part of a multi-protein complex that is composed of many of the ERAD components. We are currently attempting to screen through various known mammalian ERAD substrates and see if half-lives of these substrates are affected by over expression, knock-down or dominant negative expressed AIRAPL. Our obtained results indicate impairment of a subset of luminal ERAD substrates as compared in wild type and AIRAPL knock-down cells. Our results are in line with the knowledge that impairment of ERAD can contribute to protein misfolding diseases. The collective data of our results clearly indicates that AIRAPL may indeed be such a molecular contributor to such phenotypes.

Data: CORDIS, © European Union