Analysis of local conformation of membrane-bound and polycrystalline peptides by two-dimensional slow-spinning rotor-synchronized MAS exchange spectroscopy.

2D slow-spinning, rotor-synchronized MAS exchange spectroscopy (SSRS-MASE) was applied to study local secondary structure of three structurally different peptides, two of which were membrane-bound. Each peptide was (13)C carbonyl labeled at two adjacent residues in the peptide backbone. In general, this methodology is attractive for membrane-bound peptides because of its lenient spinning, decoupling, and RF homogeneity requirements. For a single set of raw SSRS-MASE data, two linearly independent methods exist for obtaining a 2D spectrum and each spectrum can be fit to obtain conformational constraints. An approach is described for combining the results of these two fits and this method is shown to work for spectra with both resolved and unresolved labeled site resonances. A spectrum is often fit well to a few different conformations which have somewhat different values of the fitting parameter chi(2). A simple statistical theory is developed which relates the deltachi(2) difference between a local minimum and the global minimum chi(2) to the likelihood that the local minimum conformation is the correct structure. Because uncertainty in the simulated data can also contribute to the overall fitting uncertainty, an empirical method is described for incorporating the simulation uncertainty into the deltachi(2) analysis. These data analysis methods were tested on polycrystalline Ala-Gly-Gly and then applied to the membrane-bound melittin and HIV-1 fusion peptides. Melittin gave a best-fit alpha helical structure at Ala-4 while the fusion peptide gave a good-fit beta strand structure at Phe-8. The melittin analysis is in agreement with the known overall structure of this peptide.