Characterization of Chemical Exchange between Soluble and Aggregated States of -Amyloid by Solution-State NMR upon Variation of Salt Conditions ² Saravanakumar Narayanan ‡,§ and Bernd Reif* ,‡,| Forschungsinstitut fu ¨r Molekulare Pharmakologie (FMP), Robert-Ro ¨ssle-Strasse 10, D-13125 Berlin, Germany, and Charite UniVersita ¨tsmedizin, D-10117 Berlin, Germany ReceiVed August 12, 2004; ReVised Manuscript ReceiVed NoVember 13, 2004 ABSTRACT: Alzheimer’s disease (AD) is characterized by the accumulation of insoluble fibrillar aggregates of -amyloid peptides (A), a 39-42 residue peptide, in the brain of AD patients. It is hypothesized that the disease causing form is not the fibrillar species but an oligomeric A molecule, which is often referred to as the “critical oligomer” of A. We show in this paper that A 1-40 undergoes chemical exchange between a monomeric, soluble state and an oligomeric, aggregated state under physiological conditions. In circular dichroism spectroscopy, we observe for this intermediate an R-helical structure. The oligomer is assigned a molecular weight of >100 kDa by diffusion-ordered spectroscopy-solution-state NMR spectroscopy (NMR). We can show by saturation transfer difference NMR experiments that the oligomer is related to monomeric A. This experiment also allows us to identify the chemical groups that are involved in interactions between mono- and oligomeric A molecules. Variation of the anionic strength in the buffer induces a shift of equilibrium between mono- and oligomeric states and possibly allows for the stabilization of these intermediate structures. Alzheimer’s disease (AD) 1 is the most abundant age- related neurodegenerative disease. The -amyloid peptide (A), which exists in different lengths (39-43 amino acids), is generated after processing of a transmembrane protein, APP (1, 2). The disease is characterized by two fundamental events, the accumulation of insoluble fibrillar aggregates of A and the degeneration and death of neurons in the brain regions that are concerned with learning and memory processes. Abnormal protein deposition is also a shared characteristic of other age-related neurodegenerative dis- esases, such as Parkinson’s disease, Huntington’s disease, and the prion diseases. There is increasing evidence that the mechanism of this aggregation may be similar in each of these diseases (3). At the same time, it is found that probably not the fibrillar state but a protofibrillar state is responsible for neurotoxicity (4-6). Most interestingly, it was found recently that antibodies can specifically recognize these soluble oligomers (7). So far, the molecular basis for the neurotoxicity of A could not be identified. It is speculated that A 1-40 forms a pore in the membrane and leads to a unregulated flux of Ca 2+ in and out of the cell (6). Alternatively, A 1-40 may generate radicals that are neuro- toxic themselves (8-10). However, it is not understood why only certain aggregation states of A 1-40 should be associated with neurotoxicity given the identical primary structure of A 1-40 . So far, NMR structural studies of A 1-40 show that the peptide is mostly unstructured in aqueous solution (11, 12). It was shown as well that A 1-40 adopts a helix-turn- helix structure upon addition of tetrafluoroethylene (13, 14) or sodium dodecyl sulfate (SDS) (15, 16). Recently, also solid-state NMR structural data is available for A 1-40 in the fibrillized state (17). Interestingly, a similar loop at approximately the same position in the primary structure was found around residues S 26 NKG, which was already observed in the SDS solution-state structure (15). In the solid-state structure, the only charged side chains in the core are those of D23 and K28, which form a salt bridge (17). The C terminus of the peptide folds back onto the hydrophobic core, so that the hydrophobic core and the aliphatic side chains of the C terminus come in close contact. The goal of the proposed NMR experiments is to find solution conditions that possibly allow for the study of different intermediates in the A aggregation process. In addition, we aimed to characterize the chemical groups of A 1-40 , which are involved in peptide-peptide interactions during aggregation. Knowledge of these interactions might possibly show avenues to prevent accumulation of a neuro- toxic, oligomeric intermediate. We want to answer this question by performing a series of saturation transfer difference (STD) experiments. STD experiments have been introduced recently to screen compound mixtures for binding ² This research was supported by a Grant of the Deutsche Forschungsgemeinschaft (DFG), Re1435/2. * To whom correspondence should be addressed. Telephone: +49(30)94793-191. Fax: +49(30)94793-199. E-mail: reif@ fmp-berlin.de. ‡ Forschungsinstitut fu ¨r Molekulare Pharmakologie (FMP). § Institut fu ¨r Organische Chemie und Biochemie II, TU Mu ¨nchen, Lichtenbergstrasse 4, D-85747 Garching, Germany. | Charite ´ Universita ¨tsmedizin. 1 Abbreviations: A, human -amyloid peptide; AD, Alzheimer’s disease; APP, amyloid precursor protein; CD, circular dichroism; DOSY, diffusion-ordered spectroscopy; LMW, low molecular weight; NMR, solution-state NMR spectroscopy; PBS, phosphate-buffered saline; STD, saturation transfer difference. 1444 Biochemistry 2005, 44, 1444-1452 10.1021/bi048264b CCC: $30.25 © 2005 American Chemical Society Published on Web 01/08/2005