In principle, the relaxation of any proton ( 1 H ) , carbon carbon ( 13 C) relaxation in a 13 C, 15 N-doubly enriched sample ( 13 C ) , or nitrogen ( 15 N ) spin in a protein will contain of the thermostable Sso7d protein have been investigated. Pulse information about dynamics. The most common applicasequences for measurements of longitudinal and transverse 13 C tion is the use of 15 N relaxation measurements to characrelaxation rates were implemented, and these rates were measured terize the dynamics in the polypeptide backbone ( 2 -4 ) at magnetic fields of 11.7 and 14.1 T. The field dependence in as well as that of certain amino acid side chains [ e.g., measured rates is small and consistent with a predominant contriarginine ( 5) ] . The success of 15 N relaxation methods to bution from chemical-shift anisotropy (CSA) to 13 C relaxation.
study protein motions relies on the fact that the relaxation
A pulse sequence for measurement of { 1 H} -13 C cross-relaxation pathways of this nucleus in backbone amides are relatively rates (steady-state NOEs) was also developed. This experiment reveals a significant NOE between protons and all 13 C, indicating well characterized and that 15 N enrichment in most cases that dipolar interactions between these nuclei contribute to 13 C is easily carried out. More recently also a-carbon ( 13 C a ) relaxation. Experiments designed to suppress cross correlation berelaxation has been used to extract information about tween CSA relaxation and dipole-dipole (DD) relaxation due to backbone motions ( 6, 7 ) . Additional complications assoneighboring 13 C a indicate that this effect is negligible. A more ciated with this nucleus, compared to backbone 15 N, are quantitative treatment is also presented, in which backbone dythe poor resolution for 13 C a in a 1 H -13 C HSQC spectrum namics parameters are fitted to average 13 C relaxation rates using and poor sensitivity ( for 13 C at natural abundance ) or ho-Lipari-Szabo expressions for the spectral density. This fit, which monuclear J couplings that affect transverse relaxation reproduces well expected backbone dynamics parameters for a rates ( in the case of 13 C-enriched samples ) .
folded protein, is used to estimate the relative contributions of various mechanisms to 13 C relaxation. It is found that both longi-Here, we investigate the predominant relaxation mechatudinal and transverse relaxation rates are dominated by CSA nisms for backbone carbonyl carbon ( 13 C ) magnetization relaxation and contain significant contributions due to DD relaxin a 13 C, 15 N-enriched protein. The investigation was motiation induced by nearby protons. Contributions from DD relaxvated by the need to characterize backbone motions in a ation due to covalently bound 13 C a and 15 N are comparably small. protein in which many of the amide protons exchange The predominant effects of CSA and 1 H-13 C DD interactions, rapidly with the solvent under relevant pH conditions ( 8 ) .
for which physical and geometrical parameters are uncertain, com-In this case, the 13 C magnetization would have been explicate the use of 13 C relaxation as a sequence-specific probe for cited and detected via the a protons. However, in order protein backbone dynamics. ᭧ 1997 Academic Press to evaluate the relaxation pathways for 13 C in proteins and the use of these rates for characterization of dynamics in 13 C, 15 N-enriched proteins, we have performed the mea-