Protein Structures Very Important Paper DOI: 10.1002/anie.201406357 A Hexameric Peptide Barrel as Building Block of Amyloid-b Protofibrils** Christofer Lendel, Morten Bjerring, Anatoly Dubnovitsky, RobertT. Kelly, Andrei Filippov, OlegN. Antzutkin, Niels Chr. Nielsen, and Torleif Härd* Abstract: Oligomeric and protofibrillar aggregates formed by the amyloid-b peptide (Ab) are believed to be involved in the pathology of Alzheimers disease. Central to Alzheimer pathology is also the fact that the longer Ab 42 peptide is more prone to aggregation than the more prevalent Ab 40 . Detailed structural studies of Ab oligomers and protofibrils have been impeded by aggregate heterogeneity and instability. We pre- viously engineered a variant of Ab that forms stable proto- fibrils and here we use solid-state NMR spectroscopy and molecular modeling to derive a structural model of these. NMR data are consistent with packing of residues 16 to 42 of Ab protomers into hexameric barrel-like oligomers within the protofibril. The core of the oligomers consists of all residues of the central and C-terminal hydrophobic regions of Ab, and hairpin loops extend from the core. The model accounts for why Ab 42 forms oligomers and protofibrils more easily than Ab 40 . Alzheimers disease (AD) is linked to the aggregation and deposition of fibrous amyloid-b (Ab) peptide into senile plaques in the brain. However, accumulating evidence suggests that the neurodegeneration in AD is associated with soluble prefibrillar oligomers and protofibrils. [1] The relative ratio of the 42-residue Ab 42 to the shorter and less aggregation prone Ab 40 is also important for the disease mechanism. For instance, an increased Ab 42 to Ab 40 ratio is associated with familial AD [2] and the formation of neurotoxic aggregates in vitro is critically dependent on this ratio. [3] The Ab 42 :Ab 40 ratio dependent neurotoxicity can be attributed to distinct biochemical properties and aggregation pathways of the two peptides. Ab 42 aggregates more rapidly than Ab 40 and can, in particular, more easily form pentameric or hexameric building blocks (paranuclei) that are believed to be the constituents of protofibrils. [4] There is also direct biological evidence for the neurotoxicity of protofibrils. For instance, Ab 42 carrying the Arctic mutation (Glu22Gly), which is associated with familial early onset AD, forms protofibrils at a much higher rate than wild-type Ab 42 . [5] Furthermore, disulfide-linked dimers of Ab 40 that carry an Ser26Cys mutation and, unlike wild-type Ab 40 , very rapidly form protofibrils, are potent inhibitors of hippocampal long- term potentiation, whereas fresh non-protofibrillar prepara- tions of this peptide are not. [6] Observations of protofibrils of Ab by electron microsco- py [7] and atomic force microscopy [8] were first reported almost two decades ago. Their appearance is polymorphic and depend on aggregation conditions and surfaces. [9, 10] Still, typical protofibrils are 5 to 6 nm wide when observed by transmission electron microscopy with negative staining [6, 7] and 3 to 5 nm wide when observed by atomic force microscopy (AFM) on mica surfaces. [8, 10, 11] They are, in general, shorter and more flexible than mature amyloid fibrils, and they frequently appear with curly and beaded features. Protofibrils of wild-type Ab are inherently unstable and ultimately form amyloid fibrils. This property has impeded detailed structural studies. We have addressed the issue of protofibril instability by engineering a variant of Ab (called Abcc) that forms oligomers and protofibrils, but cannot form amyloid fibrils. Briefly, in Abcc, alanine residues at positions 21 and 30 are replaced by cysteines, so that a disulfide bond locks the peptide in a conformation that is incompatible with fibril formation and aggregation and is therefore arrested at the protofibril state. [11, 12] Protofibrils formed by Ab 42 cc are [*] Dr. C. Lendel, Dr. A. Dubnovitsky, Prof.Dr. T. Härd Department of Chemistry and Biotechnology Swedish University of Agricultural Sciences (SLU) Box 7015, SE-750 07 Uppsala (Sweden) E-mail: Torleif.Hard@slu.se Dr. M. Bjerring, Prof. Dr. N. C. Nielsen Center for Insoluble Protein Structures (inSPIN) Interdisciplinary Nanoscience Center (iNANO) and Department of Chemistry, Aarhus University Aarhus (Denmark) R. T. Kelly, Prof. Dr. O. N. Antzutkin Department of Physics, Warwick University Coventry (United Kingdom) Dr. A. Filippov, Prof. Dr. O. N. Antzutkin Chemistry of Interfaces, Luleå University of Technology Luleå (Sweden) Dr. A. Dubnovitsky Present address: Department of Clinical Neuroscience, Center for Molecular Medicine, Karolinska Institutet Stockholm (Sweden) [**] This work was supported by grants from the Swedish Research Council (VR 621-2011-5812), the Danish National Research Foun- dation (DNRF59), and the Swedish Alzheimer Foundation (Nr. 10- 03-100 and Nr. 12-03-119). Molecular modeling was performed using resources provided by the Swedish National Infrastructure for Computing (SNIC) at NSC, Linkçping, Sweden. NMR analyses of synthetic peptides were performed using resources provided by the UK 850MHz solid-state NMR facility at Warwick, UK. We thank Dr. Ingemar AndrØ (Lund University), Dr. Evalena Anderson (SLU), Dr. Joakim Bergstrçm (Uppsala University), and Dr. Sergij Afonin for their assistance. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/anie.201406357. . Angewandte Communications 12756  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2014, 53, 12756 –12760