Department of Biochemistry and Cell Biology
Kevin R. MacKenzie
This experiment exploits the strong (~92 Hz) J coupling between the amide hydrogen and its directly bonded amide nitrogen in this 15N-enriched sample. The spectrum contains one crosspeak for every amide proton/ nitrogen pair. Peaks corresponding to the amides of residues involved in the mutant GpA dimer interface are identified. Intense peaks belong to residues located in the disordered, floppy ends of the 40 residue peptide.
The 2D 1H-15N HSQC usually provides the best dispersion of any 2D shift correlation
spectrum of a protein, and backbone resonance assignment strategies are designed to exploit this ability to resolve
individual amide proton-nitrogen correlations in two dimensions. A 2D HSQC
like the one above, where a resolved peak is observed for each of the 40 residues in the peptide,
generally means that obtaining backbone assignments will be straightforward. The preliminary
stages of a structure determination project usually involve optimizing conditions to improve
the quality of the 2D 1H-15N HSQC. Improved resolution can be obtained
by moving to higher magnetic field strength, by labeling the protein with 2H, by using pulse
sequences that provide narrower lines, or by enhancing the tumbling of the molecule by changing the
size of the peptide-micelle complex. Since this critically affects the amount of information
that can be extracted from the spectra, the detergent and buffer conditions of the sample need to
be thoroughly explored to obtain a well-behaved sample.The 3D 1H-15N NOESY-HSQC contains 1H-1H NOE correlations between amide hydrogens and other hydrogens located nearby to them in space. The amide protons are detected in the direct dimension of the experiment, while the amide nitrogen chemical shifts and the chemical shifts of the hydrogens nearby in space are detected indirectly.
These 3D spectra are conveniently displayed as narrow 2D slices or strips (below). Each (vertical) strip provides the proton chemical shift of the hydrogens nearby in space to a given amide hydrogen if the 2D HSQC fully resolves the amide hydrogen/amide nitrogen correlations. Examination of the slice for A82 shows how poorly resolved amides give rise to a strip containing a mixture of information from different amide hydrogens that may prove difficult to decipher.
Tertiary folds are hard to 'spot' in such spectra, but the polypeptide secondary structure can be readily determined. For instance, a-helices have strong NOEs between consecutive amide hydrogens. Well-resolved NOEs are observed between each amide and both the previous and the following amide. (NOEs between Y80-M81 and I85-S86 are poorly resolved, because these pairs have nearly identical amide hydrogen resonance frequencies.) Each amide hydrogen makes a strong NOE to its own Ha and a weak NOE to the Ha behind it, but in a helix the amide will also make an NOE to the Ha 3 or 4 residues behind it in the sequence. These NOEs establish that this peptide is an a-helix; this is confirmed by chemical shift analysis and J couplings.
Slices from another 3D spectrum available here.