Dipolar recoupling NMR of biomolecular self-assemblies: determining inter- and intrastrand distances in fibrilized...

We demonstrate a new method for investigating the structure of self-associating biopolymers using dipolar recoupling NMR techniques. This approach was applied to the study of fibrillar beta-amyloid (Abeta) peptides (the primary component of the plaques of Alzheimer's disease) containing only a single isotopic spin label (13C), by employing the DRAWS (dipolar recoupling with a windowless sequence) technique to measure 13C-13C distances. The 'single-label' approach simplified analysis of DRAWS data, since only interstrand contacts are present, without the possibility of any intrastrand contacts. As previously reported [T.L.S. Benzinger, D.M. Gregory, T.S. Burkoth, H. Miller-Auer, D.G. Lynn, R.E. Botto, S.C. Meredith, Proc. Natl. Acad. Sci. 95 (1998) 13407.], contacts of approximately 5 A were observed at all residues studied, consistent with an extended parallel beta-sheet structure with each amino acid in exact register. Here, we propose that our strategy is completely generalizable, and provides a new approach for characterizing any iterative, self-associating biopolymer. Towards the end of generalizing and refining our approach, in this paper we evaluate several issues raised by our previous analyses. First, we consider the effects of double-quantum (DQ) transverse relaxation processes. Next, we discuss the effects of various multiple-spin geometries on modeling of DRAWS data. Several practical issues are also discussed: these include (1) the use of DQ filtering experiments, either to corroborate DRAWS data, or as a rapid screening assessment of the proper placement of isotopic spin labels; and (2) the comparison of solid samples prepared by either lyophilization or freezing. Finally, data obtained from the use of single labels is compared with that obtained in doubly 13C-labeled model compounds of known crystal structure. It is shown that such data are obtainable in far more complex peptide molecules. These data,taken together, refine the DRAWS method, and demonstrate its precision and utility in obtaining high resolution structural data in complex biomolecular aggregates such as Abeta.