Key publications

Ferrari L, Stucchi R, Konstantoulea K, van der Kamp G, Kos R, Geerts WJC, van Bezouwen LS, Förster FG, Altelaar M, Hoogenraad CC, Rüdiger SGD

Arginine π-stacking drives binding to fibrils of the Alzheimer protein Tau

Nat Commun. 2020 11:571 Link

This paper describes a new molecular concept for aberrant interaction of neurotoxic fibrils. We identify pi-stacking as molecular mechanism for fibrils to specifically attract a specific subset of disordered protein. This offers a molecular reasoning for potential toxicity of fibrils in diseases such as Alzheimer.


Morán Luengo T, Kityk R, Mayer MP, Rüdiger SGD

Hsp90 breaks the deadlock of the Hsp70 chaperone system.

Mol Cell. 2018 70:545-552 (cover story). Link

This paper describes how chaperones can fold proteins. It shows that, contrary to expectation, Hsp70 at molecular level blocks folding, by constant and fast rebinding, effectively preventing protein folding at physiological concentrations. It is Hsp90 that breaks this vicious circle by offering a less hydrophobic binding site, which allows the protein to reach the native state. Based on biochemical folding assays, we provide a new paradigm how these chaperones ensure high folding yields without affecting the folding path. Thus, our results provide a consistent link between the chaperone paradigm and the Anfinsen concept that the primary structure determines the three-dimensional fold of a protein.
This paper was selected for the front cover.

Figure 4: General Model for Chaperoned Protein Folding (A) Three-dimensional scheme for protein folding. Energy landscape shows the effect of chaperones in early stages. Ub, a misfolded intermediate, is boosted by chaperones to a state that allows self-evolution following a slow path toward the native state. (B) Time scheme indicating the action of the Hsp70/Hsp90 cascade early on the folding path, far ahead of the transition state. (C) Chaperone-assisted folding mechanism. Unfolded protein directly evolves into two states, folding-competent Ua and aggregation/misfolding-prone Ub, which binds to Hsp70. After release from Hsp70, the protein can further evolve to Ua or Ub. However, high Hsp70 levels inhibit the return to the Ua/Ub junction. Hsp90 removes the Hsp70-inflicted folding block, promoting progression to the native state. Roman numbers indicate the three modes proposed for protein folding (I, spontaneous folding; II, Hsp70 cycling; III, Hsp70-Hsp90 cascade).

Anvarian Z, Nojima H, van Kappel EC, Madl T, Spit M, Viertler M, Jordens I, Low TY, van Scherpenzeel R, Kuper I, Richter K, Heck AJR, Boelens R, Vincent JP, Rüdiger SGD, Maurice MM.

Axin cancer mutants form nano-aggregates to rewire the Wnt signaling network.

Nature Struct Mol Biol. 2016 23:324-32. Link

This publication describes the key result of the joint “High Potential” project with Madelon Maurice. Our findings established a paradigm for misregulation of signaling in cancer and show that targeting aggregation-prone stretches in mutated scaffolds holds attractive potential for cancer treatment. This study nailed an important biological question by using a multidisciplinary set of techniques, ranging from atomics resolution to living animal, including spectroscopy (fluorescence, CD and NMR), SAXS, mass spectrometry, cell culture, signalling assays, microscopy and drosophila genetics. The mutant, destabilised Axin, gained novel functions by forming non-amyloid nanometer-scale aggregates, which rewire the Axin interactome. Importantly, tumour suppressor activity of the Axin cancer mutant is rescued by preventing aggregation of a single, non-conserved segment.

Figure 7: Model for the mechanism of action of Axin RGS cancer variants. Axin WT forms a complex with partner proteins (blue ovals) and mediates tumor-suppressor activity. A single cancer point mutation endows Axin with new properties through formation of an oligomeric core with disordered tentacles. The altered conformation perturbs the associated subproteome (blue and red ovals) through both loss and gain of binding partners. The combined events drive Wnt-pathway activation and tumor growth. Tumorigenic behavior of the mutant protein is corrected by interference with aggregon-mediated oligomer formation.

Karagöz GE, Duarte AMS, Akoury E, Ippel H, Biernat J, Morán Luengo T, Radli M, Didenko T, Nordhues BA, Veprintsev DB, Dickey CA, Mandelkow E, Zweckstetter M, Boelens R, Madl T, Rüdiger SGD.

Hsp90-Tau complex reveals molecular basis for specificity in chaperone action.

Cell. 2014 156:963-974. Link

This is a key publication that turned the direction of my lab towards proteostasis control in neurobiology and neurodegenerative diseases by linking Hsp90 and Tau. It is the result of the previous 9 years of work on the Hsp90 research line in my group. This paper provided several important conceptional advances: (i) We localised the substrate binding region of Hsp90; (ii) we found that the binding principle of Hsp90 relies on a spreading of numerous contacts over a large surface; (iii) we characterised the disordered Tau protein as bona fide Hsp90 client; (iv) we identified the principles that allow intrinsically disordered proteins to become Hsp90 clients; (iv) we proposed a model that explains the timing of chaperone action in the Cell.

Figure 6: Partitioning of Chaperone Action (A) Predicted Hsp70-binding sites (side chains of leucine, isoleucine, valine, phenylalanine, and tyrosine shown as yellow spheres) form the nucleus of folding intermediates (four examples: dihydrolipoyllysine-residue acetyltransferase [1W4G], pre-mRNA-processing factor 40 homolog A [2KZG], thermonuclease [2KQ3], and Flavodoxin [2KQU]), leaving scattered hydrophobic residues in the periphery (LIVFYW residues; red spheres). (B) The Tau ensemble, color coded as in (A). The repeat regions are indicated as in Figure 1E. (C) Model for partitioning of chaperone action between the Hsp70 and Hsp90 systems. Hsp70 binds early when its binding sites are accessible. During folding, the Hsp70-binding sites disappear, leaving scattered hydrophobics that are recognized by Hsp90. The folded protein will bury also those residues.

Karagöz GE, Duarte AMS, Ippel H, Uetrecht C, Sinnige T, van Rosmalen M, Hausmann J, Heck AJR, Boelens R, Rüdiger SGD.

N-terminal domain of human Hsp90 triggers binding to the cochaperone p23.

Proc Natl Acad Sci U S A. 2011 108:580-5. Link

This study established the technology that provided me with a competitive advantage in the field, the ability to monitor human Hsp90 by NMR. Strikingly, we found that the complex of Hsp90 with its co-chaperone p23 becomes asymmetric despite symmetric stoichiometry. This has interesting implication for understanding the molecular mechanism.

Figure 5 The N-terminal domain of Hsp90 triggers p23 binding. Both Apo and ATP-bound human Hsp90 (Hsp90-N, red; Hsp90-M, green; Hsp90-C, blue) are predominantly open under equilibrium conditions (10). p23 (orange) cannot bind to Apo Hsp90, indicating that the interactions with Hsp90-M are not sufficient for binding. p23 can bind to Hsp90 after dimerization of the N-terminal domains are achieved, allowing for strengthening of the interaction by contacts with Hsp90-M. Afterward, a second p23 molecule binds, resulting in a Hsp902p232 complex (boxed). The steps of the Hsp90-p23 association are indicated by numbers 1–5, as described in the text.

Rüdiger SGD, Germeroth L, Schneider-Mergener J, Bukau B.

Substrate specificity of the DnaK chaperone determined by screening cellulose bound peptide libraries.

EMBO J. 1997 16:1501-7. Link

This paper is the most significant from my time with Bernd Bukau. It is my most cited paper (>600 citations). This paper defined the molecular basis for substrate recognition of the Hsp70 chaperone family. I established an algorithm to predict binding sites of a major chaperone class in substrate proteins. In the course of this study, I invented a novel detection method to use peptide libraries, which is still widely used.

Figure 1: Localization of DnaK binding sites in native protein structures (partial figure). (B) Ribbon and space filling representations (INSIGHT II, Biosym) of the structures of the corresponding native proteins (the DNA binding fragment and tetramerization domain in the case of p53, mature forms in the case of AP and insulin, conformation at pH 7 in the case of HA) are shown (Wilson et al., 1981; Hua et al., 1991; Kim and Wyckoff, 1991; Schultz et al., 1991; Cho et al., 1994; Jeffrey et al., 1995; Conti et al., 1996). Red and pink segments indicate good DnaK binding sites and DnaK binding sites with weaker affinity respectively; backbone atoms are yellow. DnaK binding sites are in most cases completely buried, in a few cases some binding site residues are exposed.