Abstract:
Environmental influences demand a constant adaptation of the cell proteome. A key player in this process is the proteasome, a self-compartmentalizing complex that is responsible for the majority of targeted protein degradation via the ubiquitin (Ub) tagging system. This pathway was long believed to have originated in eukaryotes, while prokaryotes were seen to use a multitude of other less characterized systems instead. However, with the increasing number of sequenced genomes and more sophisticated bioinformatic analyses, evolutionary links and common concepts in prokaryotic and eukaryotic systems emerged. In this work, we illuminate these relationships by presenting a characterization of two novel bacterial proteasome homologs, Anbu and BPH, and of the first archaeal (de-) ubiquitination system.
Anbu (ancestral beta subunit) sequences are widely distributed amongst phylogenetically distant organisms and therefore its discovery meant that the broadly accepted concept, according to which the proteasome evolved from its only other known homolog, HslV, needed to be reviewed. In chapter 3.1, we unravel these relationships via cluster analysis and the first biophysical characterization of Anbu. Anbu forms a dodecameric complex with a unique architecture that was only accessible through the combination of X-ray crystallography and small-angle X-ray scattering. While forming continuous helices in crystals and electron microscopy preparations, refinement of sections from the crystal structure against the scattering data revealed a helical open-ring structure in solution, contrasting the ring-shaped structures of proteasome and HslV. Based on this primordial architecture and exhaustive sequence comparisons, we propose that Anbu represents an ancestral precursor at the origin of self-compartmentalization.
In this evolutionary scenario, BPH (beta-proteobacterial proteasome homolog) constitutes a comparably recent innovation that evolved by duplication of HslV. In chapter 3.2, we present the first analysis of BPH and reveal crystal structures of two representatives from of Thiobacillus denitrificans and Cupriavidus metallidurans. Remarkably, both proteins assemble in two stacked heptameric rings, unlike the hexameric rings seen in HslV. This unprecedented assembly is supported by SAXS measurements, analytical ultracentrifugation and averaging of EM images. Its proteolytic activity is evidenced by mass spectrometry and a crystal structure with the proteasome specific inhibitor epoxomicin bound to the BPH active site. In contrast to HslV, BPH appears to act independent of AAA+ unfoldases and thus might pursue a function distinct from known bacterial self-compartmentalizing proteases, possibly in analogy to the proteasome's ATP-independent functions.
Finally (chapter 3.3), we study the first sequenced archaeal Ub tagging operon from Caldiarchaeum subterraneum, showing that protein ubiquitination is not restricted to eukaryotes. Via cluster analysis, we retrace the evolutionary history of this operon, revealing that its step-wise assembly was accompanied by co-evolution of increasingly Rpn11-like deubiquitinases, similar to those found in the eukaryotic proteasome. The C. subterraneum representative, CsRpn11, marks the apex of this process. CsRpn11 displays bimodal activity towards ubiquitin (CsUb) precursor (KD = 14.6 µM; KM = 24.2 µM) and isopeptide-linked ubiquitinated proteins. Our CsUb-CsRpn11 crystal structure reveals the first catalytically active deubiquitinase-Ub complex with the catalytic water oriented for cleavage. A comparison with the Ub-free structure shows a Ub-induced change between two conformations, that were previously associated with distinct binding modes in eukaryotic Rpn11 and its homolog AMSH. Consequently, the CsRpn11-CsUb interaction presents a useful model for homologous eukaryotic systems and supports an archaeal origin of protein ubiquitination.
Print-on-Demand