Two-dimensional conventional magnitude COSY, phase-sensitive pure absorption DQF-COSY, and HOHAHA experiments have been successfully applied to paramagnetic low-spin ferric sperm whale metcyanomyoglobin to identify the spin systems of the side chains of the residues located in the heme cavity. The assignments from the present bond correlation spectroscopy for the residues of lie FG5/99, Phe CD1/43, His F8/93, Val El 1/68, and heme itself confirm the assignments made previously by interpreting nuclear Overhauser effect, NOE, data on the basis of the X-ray crystal structure. The spin system of a Leu side chain with strongly hyperfine shifted resonances has been unambiguously identified with a combination of COSY and HOHAHA spectra and assigned to the Leu F4/89 residue located in the heme pocket on the proximal side, for which the initial partial assignments are shown to be invalid. The incorrect earlier assignments resulted from a difference between the solution and crystal orientation of the side chain. Molecular modeling and dipolar shift calculations upon rotating the Leu F4/89 side chain about the C,g-Cr bond by ~120° lead to an orientation consistent with both the observed dipolar shifts and NOEs to the heme. The methods of bond correlation spectroscopy are qualitatively evaluated and shown to be able to identify cross peaks from pairs of protons experiencing very rapid paramagnetic relaxation rates with T,s as short as ~20 ms and line widths as large as ~ 100 Hz. Preliminary 2D bond correlation spectroscopic data on the cyano complex of horseradish peroxidase suggest that J cross peaks can identify residues in even larger paramagnetic heme proteins. The various methods considered, however, differ significantly in their information content, with the magnitude COSY map providing the optimal information on the strongly relaxed protons. In particular, the double quantum filter is found to strongly discriminate against rapidly relaxed, broad, and spin-coupled resonances, making DQF-COSY experiments virtually useless for the protons closest to the iron. The variable-temperature HOHAHA and COSY maps serve to define complete spin systems. Therefore, it is clear that 2D bond correlation spectroscopy can be profitably applied to paramagnetic macromolecules and that, along with the 2D nuclear Overhauser effect spectroscopy (NOESY) [Emerson, S. D.; La Mar, G. N. Biochemistry 1990, 29, 1545-1556], the solution structures of paramagnetic molecules, at least low-spin ferric molecules, should be determinable with 2D NMR techniques in a fashion now attainable for diamagnetic molecules.
ASJC Scopus subject areas
- Colloid and Surface Chemistry