Logo
DeutschClear Cookie - decide language by browser settings
Polier, Sigrun (2009): Structural Basis for the Cooperation of Hsp110 and Hsp70 Molecular Chaperones in Protein Folding. Dissertation, LMU München: Faculty of Chemistry and Pharmacy
[img]
Preview
PDF
Polier_Sigrun.pdf

6Mb

Abstract

Protein folding is a crucial process for cell survival. Only natively structured proteins can perform their essential biological functions. Although all structure-relevant information is principally encoded in the amino acid sequence of a protein, the efficient folding of many larger proteins depends on the assistance of molecular chaperones. These proteins bind reversibly to exposed hydrophobic sequences in folding intermediates, thereby preventing aggregation and supporting effective folding. The Hsp70 family proteins constitute key components of the cellular chaperone network in eukaryotes and bacteria. They are involved in diverse protein processing reactions, reaching from folding and assembly of nascent polypeptides to protein transport across membranes. Regular Hsp70s consist of an N-terminal nucleotide binding domain (NBD) and an allosterically coupled C-terminal substrate binding domain, which is further divided into a beta-sandwich domain and a three helix bundle domain (3HBD). Hsp70s perform their cellular functions through ATP-driven cycles of substrate binding and release: In the ATP state, peptide binding is dynamic. ATP hydrolysis results in a dramatic structural rearrangement, leading to a conformation in which hydrophobic peptide segments are locked between 3HBD and beta-sandwich domain. Thus, substrate proteins are stably bound in the ADP and apo state. This Hsp70 folding cycle is tightly controlled by a large complement of cochaperones. Whereas J-domain proteins recruit substrates and trigger ATP hydrolysis, nucleotide exchange factors (NEFs) accelerate ADP release. In eukaryotes, four evolutionarily unrelated classes of Hsp70 NEFs have been identified, among which Hsp110 homologs are most abundant. As judged by their conserved domain composition, Hsp110s derive from canonical Hsp70s, but have evolved into NEFs, preserving the ability to stabilize misfolded proteins in solution. In the present study, the cooperation of Hsp70 and Hsp110 molecular chaperones in protein folding was investigated. First, the crystal structure of a functional complex between the yeast Hsp110 homolog Sse1p and the NBD of human Hsp70 was determined. The structure was solved by selenium multiple wavelength anomalous diffraction and refined at 2.3 Å resolution to a crystallographic R-factor of 19.7 %. The structure of Sse1p is characterized by extended domain-domain interactions. Beta-sandwich domain and 3HBD are arranged along the NBD and point into opposite directions. Importantly, Sse1p has ATP bound, a prerequisite for efficient complex formation with Hsp70. In the complex, the NBD and 3HBD of Sse1p embrace the Hsp70 NBD, thereby opening the nucleotide binding cleft of Hsp70 and releasing ADP. In a subsequent mutational analysis, key features of the chaperone complex were targeted. Specifically, amino acid exchanges were introduced (i) in the areas of close surface contacts between Sse1p and Hsp70N, (ii) at the interfaces between individual Sse1p domains, (iii) in the nucleotide binding pocket of Sse1p, and (iv) at the putative substrate binding site in the beta-sandwich domain of Sse1p. Sse1p mutations affecting the interaction between the Hsp70 NBD and the Sse1p 3HBD strongly impaired Hsp70 complex formation and nucleotide exchange. Consistently, these Sse1p variants were less effective than wildtype (wt) Sse1p in supporting substrate release from Hsp70 and Hsp70-mediated refolding of thermally denatured firefly luciferase. In vivo, the respective sse1 mutations caused severe stress and a pronounced growth defect, likely because of reduced substrate flux through the Hsp70 machinery. Taken together, these results define nucleotide exchange on Hsp70 as the main function of Sse1p. Furthermore, they highlight the importance of the interaction between the Hsp70 NBD and the Sse1p 3HBD for the nucleotide exchange activity of Sse1p. No evidence was found for an ATP-driven, Hsp70-like conformational cycle of Sse1p. Mutations targeting the Sse1p inter-domain communication and the ATPase activity did not impair Sse1p function in vitro and in vivo. Mutations at the putative substrate binding cleft of Sse1p aggravated the functional defect of partially NEF-deficient Sse1p mutants in vitro and in vivo. Thus, direct substrate interactions mediated by the beta-sandwich domain may support Hsp70-assisted protein folding in addition to the nucleotide exchange function of Sse1p. According to our mutational analysis, Hsp110s contribute to Hsp70-assisted protein folding in two ways: Their major function is the acceleration of ADP dissociation from Hsp70 by stabilizing the Hsp70 NBD in an open conformation with low affinity for the nucleotide. Consequently, Hsp110s support the rapid conformational cycling of Hsp70 and thus efficient substrate binding and release. In addition, Hsp110s may directly interact with the unfolded substrate upon binding to Hsp70. By holding the misfolded protein chain, Hsp110s might cooperatively aid Hsp70 in remodeling the substrate via large-scale thermal motions of the Hsp70 PBD.