Role of Prion Disease-Linked Mutations in the Intrinsically Disordered N‑Terminal Domain of the Prion Protein Xiaojing Cong, †,‡,⊥,○ Nicola Casiraghi, §,‡,⊥,○ Giulia Rossetti,* ,#,∥,⊥,‡,○ Sandipan Mohanty, ∥ Gabriele Giachin, † Giuseppe Legname, †,▽ and Paolo Carloni* ,‡,⊥ † Laboratory of Prion Biology, Department of Neuroscience, Scuola Internazionale Superiore di Studi Avanzati (SISSA), via Bonomea 265, 34136 Trieste, Italy ‡ Laboratory for Computational Biophysics, German Research School for Simulation Sciences (GRS), Forschungszentrum Jü lich−RWTH Aachen, 52425 Jü lich, Germany § Department of Biology, University of Bologna, via Selmi 3, 40126 Bologna, Italy ∥ Jü lich Supercomputing Centre, Forschungszentrum Jü lich, 52425 Jü lich, Germany ⊥ Computational Biomedicine Section (IAS-5), Institute of Advanced Simulation (IAS), 52425 Jü lich, Germany # Institute for Research in Biomedicine and Barcelona Supercomputing Center Joint Research Program on Computational Biology, Barcelona Science Park, Baldiri I Reixac 10, 08028 Barcelona, Spain ▽ ELETTRA Laboratory, Sincrotrone Trieste S.C.p.A., 34149 Basovizza, Trieste, Italy * S Supporting Information ABSTRACT: Prion diseases are fatal neurodegenerative disorders in mammals and other animal species. In humans, about 15% of these maladies are caused by pathogenic mutations (PMs) in the gene encoding for the prion protein (PrP C ). Seven PMs are located in the naturally unfolded PrP C N-terminal domain, which constitutes about half of the protein. Intriguingly and in sharp contrast to other PMs clustered in the folded domain, N-terminal PMs barely affect the conversion to the pathogenic (scrapie, or PrP Sc ) isoform of PrP C . Here, we hypothesize that the neurotoxicity of these PMs arises from changes in structural determinants of the N- terminal domain, affecting the protein binding with its cellular partners and/or the cotranslational translocation during the PrP C biosynthesis. We test this idea by predicting the conformational ensemble of the wild-type (WT) and mutated mouse PrP C N-terminal domain, whose sequence is almost identical to that of the human one and for which the largest number of in vivo data is available. The conformational properties of the WT are consistent with those inferred experimentally. Importantly, the PMs turn out to affect in a subtle manner the intramolecular contacts in the putative N-terminal domain binding sites for Cu 2+ ions, sulphated glycosaminoglycans, and other known PrP C cellular partners. The PMs also alter the local structural features of the transmembrane domain and adjacent stop transfer effector, which act together to regulate the protein topology. These results corroborate the hypothesis that N-terminal PMs affect the PrP C binding to functional interactors and/or the translocation. ■ INTRODUCTION Familial forms of human (Hu) transmissible spongiform encephalopathies (TSEs) or prion diseases are fatal and incurable neurodegenerative disorders. The key event in these maladies is the post-translational conversion of the ubiquitously expressed cellular form of the Hu prion protein (PrP C ) into the misfolded pathogenic isoform, PrP Sc , without any detectable covalent modifications. 1 Prion diseases are associated with 58 missense or insertional/ deletional mutations identified so far in the gene coding for HuPrP C (PRNP). Missense mutations include 44 nonsynon- ymous codon substitutions, or point mutations (PMs), and five nonsense (or “stop”) mutations. 2−4 These mutations lead to neurodegeneration and give rise to abnormal forms of HuPrP in the brain. 2 The PMs are located all over the protein, from the disordered N-terminal domain (N-term_HuPrP C hereafter, residues 23−124) to the folded C-terminal globular domain (GD, residues 125−230; Figure 1A). Most PMs are located in the HuPrP C GD. They may affect GD secondary structure (SS) elements and structure flexibility 5−9 and accelerate the misfolding process in vitro. 2,10−12 Received: June 24, 2013 Published: October 16, 2013 Article pubs.acs.org/JCTC © 2013 American Chemical Society 5158 dx.doi.org/10.1021/ct400534k | J. Chem. Theory Comput. 2013, 9, 5158−5167