Dissecting the Mechanism of Protein Unfolding by the Unfoldase PAN using Structural Biology

Abstract

The proteasome as the major cellular protease plays a central role in regulated protein deg-radation in all kingdoms of life. Besides the proteolytic core particle, the proteasome com-prises an ATP-dependent unfoldase subcomplex from the family of AAA+ ATPases. Numer-ous structural, biochemical and biophysical studies have provided insights into the mecha-nism of ATP-dependent protein unfolding by hexameric AAA+ unfoldase from bacteria, ar-chaea and eukaryotes, resulting in the suggestion of a common mechanism for how all AAA+ unfoldases couple ATP-hydrolysis to protein unfolding. However, conflicting biochemical, biophysical and structural data on unfoldases from various organisms challenge the universal validity of this mechanism, calling for further studies on unfoldases from all kingdoms of life. The proteasomal ATPase PAN from archaea is considered to be an evolutionary precursor to its eukaryotic homolog, the 19S regulatory particle, and its much simpler architecture makes it an excellent system for structural and mechanistic studies in solution. In this PhD thesis I employed solution-state NMR spectroscopy to study the conformational states of PAN and gain insights into the unfolding reaction of the model substrate GFP-ssrA by PAN. After assigning NMR resonances of all subdomains and of the full complex, I could confirm a functional asymmetry in hexameric PAN particles. Asymmetry has also been found in most of the recent cryo-EM models from various AAA+ ATPases; however, it has been located mostly to the hexameric ring formed by the nucleotide-binding domain of the ATPases. By contrast, my studies show conformational heterogeneity and asymmetry in the so-called OB-ring which is located above the nucleotide-binding domain and is thought to allosterically link the nucleotide state of PAN to substrate recruitment. Further, studies on the binding of PAN to an ssrA-tagged model substrate demonstrate the absence of strong, stable interactions be-tween the PAN N-terminal domains and its substrates, in disagreement with what has been previously proposed. Further, I used time-resolved NMR experiments to study the dynamic process of unfolding of the model substrate GFP-ssrA. Detecting selectively the NMR-signals of isotopically labeled GFP-ssrA, I was able to monitor the states experienced by the substrate during the unfolding and degradation reaction by PAN at amino acid resolution. The disappearance of natively folded GFP signals and the concomitant build-up of unstructured proteolysis-products in the NMR experiments with similar time-constants confirm a tight coupling of the two processes and are in good agreement with findings of a collaborative study using time-resolved small angle neutron scattering.1 These studies demonstrate the power of NMR spectroscopy in helping to understand dynam-ic processes like protein unfolding and emphasize its complementarity to other solution-based techniques like small-angle scattering which probe biological samples in a more native-like environment and thus allow dynamic processes to take place and be monitored.